photophysics of carbon nanotubes interfaced with organic and inorganic materials

Photophysics of Carbon Nanotubes Interfaced with Organic and Inorganic Materials

Igor A. Levitsky · William B. Euler Victor A. Karachevtsev

1 3

Photophysics of Carbon Nanotubes Interfaced with Organic and Inorganic Materials

Igor A. LevitskyEmitech, Inc. Fall River MA USA

and

Department of Chemistry University of Rhode Island Kingston USA

© Springer-Verlag London 2012This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law.The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

ISBN 978-1-4471-4825-8 ISBN 978-1-4471-4826-5 (eBook)DOI 10.1007/978-1-4471-4826-5Springer London Heidelberg New York Dordrecht

Library of Congress Control Number: 2012952684

William B. EulerDepartment of Chemistry University of Rhode Island Kingston USA

Victor A. KarachevtsevB. I. Verkin Institute of Low Temperature Physics and Engineering National Academy of Sciences of Ukraine Kharkov Ukraine

v

Preface

Since Iijima’s discovery of carbon nanotubes (CNTs) in 1991, these unique nanoobjects have been the focus of enormous research in physics, chemistry, and material science.

It is hard to overestimate the contribution of CNT research for the past two dec-ades in understanding the fundamental science of carbon nanostructures and their applications ranging from renewable energy to nanobiology and nanomedcine.

One of the exciting fields of CNT science is a light interaction with carbon nanotubes revealing principally new features in light absorption, luminescence, and photoconductivity associated with their quasi-one dimensional nature such as nanotube chirality, diameter, aspect ratio, etc. Photophysics of CNTs is rich and full of remarkable phenomena existing only in CNT structures, which do not have their bulk analogs distinct, for example, from quantum dots. Because of great interest in CNT optical spectroscopy and optoelectronics, a large number of books, book chapters, and reviews appeared in recent years considering not only the fun-damental principles of CNT optics, but also various applications in the field of photovoltaics, IR detectors and imaging, transparent conductive coating, nonlinear optics, photo-mechanical actuators, LEDs, and optochemical/bio sensing.

If photophysical properties of pristine CNTs are studied relatively thoughtfully, a much less explored area is light interaction with nanotubes interfaced with other materials (e.g., organic, inorganic, bulk or nanoscale structures forming physical or chemical bonding with nanotubes). An addition of another compound to CNT and creation of CNT-based nanohybrid open new opportunities for research-ers; first of all, because of much versatility of CNT composites and existence of interface between CNT and its counterpart, which is not possible for pristine nanotubes. Interfacial region in such hybrids plays a critical role being responsi-ble for various photoinduced mechanisms such as charge transfer and recombina-tion, energy transfer, photo-mechanical elastic response, thermal effect, spectral changes in Raman, absorption, and photoluminescence. This aspect, photophysics of carbon nanotubes interfaced with other materials, is the main focus of the pre-sented monograph covering three areas: (i) light harvesting and energy conversion, photoinduced charge transfer, polarization and charge separation in CNT-based nanohybrids (I. A. Levitsky); (ii) the use of CNT composite for photo-mechanical actuators (W. B. Euler); and (iii) CNT/DNA hybrid optical spectroscopy, structure,

Prefacevi

and MD simulations and related applications in biosensing and biomedicine (V. A. Karachevtsev).

The first chapter primarily describes the recent advances and new achievements in fundamental and applied sciences shedding light on the nature of photoconver-sion mechanisms in CNT nanohybrids with a short background on previous stud-ies in the field of photoinduced charge transfer, hybrid photovoltaics, photodecting devices, and bolometers.

The second chapter is dedicated to CNTs and mostly CNT composites employed in photo-mechanical actuators with large photo-elastic response associ-ated with charge accumulation and interface polarization. This is a relatively new discipline, existing for less than a decade; however, with impressive promises for future applications in light to mechanical energy conversion.

The third chapter presents a review of recent works in the field of photophys-ics of CNT/DNA hybrids which continue to be an active research area. In spite of essential differences between DNA and nanotube structures, properties of these two nanoobjects supplement each other forming a hybrid with specific physical and optical features. Here, the major focus is done on absorption and lumines-cence spectroscopy with conjunction of molecular dynamics simulation of CNT/DNA hybrids as well as possible applications in biosensing and drug delivery.

We hope that this monograph will be of interest for physicists, chemists, and material scientists working on carbon nanotube composites in fundamental and applied fields.

vii

1 Light Energy Conversion at Carbon Nanotubes - Organic and Inorganic Interfaces: Photovoltaics, Photodetectors and Bolometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Introduction: Light Absorption and Charge Separation

in Carbon Nanotubes Interfaced with Other Materials . . . . . . . . . . . 11.2 CNT/Organic Based Photovoltaics and Photodetectors . . . . . . . . . . . 4

1.2.1 CNTs Interfaced with Small Molecules . . . . . . . . . . . . . . . . . 41.2.2 Role Carbon Nanotubes in Light Absorption

and Photocarrier Generation: CNT/Fullerene Solar Cells and Photodetectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.2.3 CNT/Polymer Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221.2.4 Photocarrier Separation and Multiplication

at p–n CNT Junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271.3 CNT/Quantum Dots Photoinduced Charge Transfer and Related

Photovoltaic Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301.4 CNT/Semiconductor Based Photovoltaics and Photodetectors . . . . . 35

1.4.1 CNT/Si Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351.4.2 CNT/Si Mid-IR Photodetectors . . . . . . . . . . . . . . . . . . . . . . . 391.4.3 Carbon Nanotubes Interfaced with Other Semiconductors,

Nanostructured and Amorphous Si and Perspective of CNT/Semiconductor Hybrid Photovoltaics . . . . . . . . . . . . 41

1.5 CNT/Polymer Based Bolometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461.5.1 Bolometric Response of Pristine CNT Films . . . . . . . . . . . . . 471.5.2 CNT/Polymer Bolometers . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

1.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

2 Use of Carbon Nanotubes in Photoactuating Composites . . . . . . . . . . 692.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692.2 Carbon Nanotube Bundles and Freestanding Films . . . . . . . . . . . . . . 70

2.2.1 Freestanding Bundles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 702.2.2 Freestanding Films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Contents

http://dx.doi.org/10.1007/978-1-4471-4826-5_1

http://dx.doi.org/10.1007/978-1-4471-4826-5_1

http://dx.doi.org/10.1007/978-1-4471-4826-5_1

http://dx.doi.org/10.1007/978-1-4471-4826-5_1#Sec1






















http://dx.doi.org/10.1007/978-1-4471-4826-5_1#Bib1

http://dx.doi.org/10.1007/978-1-4471-4826-5_2





Contentsviii

2.3 Carbon Nanotubes in Mixed Composites . . . . . . . . . . . . . . . . . . . . . . 732.3.1 Rubbery Polymer Host Materials . . . . . . . . . . . . . . . . . . . . . . 732.3.2 Hydrogel Host Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

2.4 Carbon Nanotube Layered Composites . . . . . . . . . . . . . . . . . . . . . . . 772.4.1 Carbon Nanotube/Acrylic Elastomer/Poly

(vinylchloride) Trilayer Composites . . . . . . . . . . . . . . . . . . . 772.4.2 Carbon Nanotube/Photoresist Bilayer Composites . . . . . . . . 782.4.3 Carbon Nanotube/Silicon Nitride Bilayer Composites . . . . . 802.4.4 Carbon Nanotube/Nafion Bilayer Composites . . . . . . . . . . . . 80

2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

3 Photophysical Properties of SWNT Interfaced with DNA . . . . . . . . . . 893.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 893.2 SWNT:DNA Hybrid: Structures and Energy Interaction . . . . . . . . . . 91

3.2.1 DNA Helix on SWNT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 913.2.2 Nucleic Bases on SWNT: Ab initio Calculation . . . . . . . . . . . 943.2.3 Calculation of Nucleoside Binding to SWNT . . . . . . . . . . . . 973.2.4 Structures of Oligonucleotides Adsorbed on SWNT

and Energy Interaction Between Them: Molecular Dynamics Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

3.2.5 Wrapping of Relatively Long DNA Around SWNT . . . . . . . 1033.2.6 Influence of Adsorbed Biopolymer Structure on Optical

Properties of SWNT: Double-Stranded DNA Adsorbed on the Nanotube Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

3.3 Absorption Spectroscopy of SWNT Interfaced with DNA . . . . . . . . 1083.3.1 Absorption Spectroscopy of SWNTs . . . . . . . . . . . . . . . . . . . 1083.3.2 Absorption Spectra Analyses of SWNT Composition . . . . . . 1113.3.3 Comparison of Absorption Spectra of SWNTs Covered

with DNA or Surfactants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1153.3.4 Peculiarities of SWNT and DNA Interaction Revealed

in Absorption Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1183.3.5 The Effect of ss-DNA Helical Negative Potential

on the SWNT Electronic Spectrum . . . . . . . . . . . . . . . . . . . . 1233.4 Photoluminescence of Semiconducting SWNTs: The Influence

of Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1243.4.1 Emission Properties of Semiconducting SWNTs . . . . . . . . . . 1243.4.2 Quantum Yield of Semiconducting SWNT Emission:

The Role of DNA Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . 1323.4.3 Influence of Environment on SWNT Photoluminescence

Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1363.4.4 Comparison of Protection Properties of SDS, SDBS

and DNA Covering of SWNTs Against pH Influence Using Luminescence and Absorption Spectroscopy . . . . . . . 144

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149












http://dx.doi.org/10.1007/978-1-4471-4826-5_3

































1

1.1 Introduction: Light Absorption and Charge Separation in Carbon Nanotubes Interfaced with Other Materials

This chapter is devoted to the recent exploration of light energy conversion at the interface between carbon nanotubes (CNTs) and other materials, ranging from small molecules and quantum dots to bulk and nanostructured semiconductors. In this context, photoconversion processes comprises light absorption, photoinduced charge transfer (PICT) or exciton dissociation, photocarrier transport and heating effect (as a result of absorbed light energy) which basic photophysical principles provide functionality of optoelectronic devices such as solar cells, photodetectors and bolometers.

The first reports about CNT photoconductivity [1–3] motivated a growing inter-est of light energy conversion employing unique optical, electrical, thermal and mechanical properties of carbon nanotubes. Carbon nanotubes as organic, quasi 1D, nanoscaled objects [4, 5] outperform their organic counterparts in many aspects, making CNTs favorite candidate for various optoelectronic applications. For example, semiconducting CNTs have a high light absorbance in visible and near infra-red (NIR) spectrum (absorbance coefficient is in the range of 104–105 cm−1), with a band gap depending on their diameter, while the most organic compounds are not capable of absorbing NIR light. Another advantage of CNTs is very high charge mobility (up to 105 cm2/V s for individual nanotubes [6] and ~60 cm2/V s for CNT films [7]) as compared with conductive organic materials. In addition, carbon nanotubes exhibit an exceptional environmental stability and resistance to photobleaching which is one of the major drawbacks of organic optoelectronics. Device fabrication is simple and cost effective as CNTs can be easily incorporated in the device’s architecture by wet processing (coating, spraying, and printing). Finally, CNTs can be simply doped [8–10] or functionalized by many covalent and non-covalent routes [11, 12], forming nano-assemblies with other molecules and polymers to provide an efficient PICT or tuning Fermi level to the favorable posi-tion at heterojunction with semiconductors.

Chapter 1Light Energy Conversion at Carbon Nanotubes - Organic and Inorganic Interfaces: Photovoltaics, Photodetectors and Bolometers

I. A. Levitsky et al., Photophysics of Carbon Nanotubes Interfaced with Organic and Inorganic Materials, DOI: 10.1007/978-1-4471-4826-5_1, © Springer-Verlag London 2012

2 1 Light Energy Conversion at Carbon Nanotubes

CNT light absorption is an initial step leading to a generation of bound excitons [13–15]. In order to convert light energy into an electrical signal, excitons should be separated on free charge carriers (electrons and holes) by an external or inter-nal built-in electric field, before they relax to the ground state. Finally, the resulting carriers should be transported to the external electrodes minimizing the recom-bination and trapping processes. Such a scenario is realized for photovoltaics (PVs) and photodetectors when the internal built-in field is required at the inter-face between carbon nanotubes and other materials. Without interface, only the external electric field can separate the carriers. However, this process cannot pro-vide substantial light-to electricity conversion. Nevertheless, the photoconduc-tivity of pristine CNTs (individual nanotubes and their network) attracted lots of attention during the past 10 years [1–3, 16–29] as understanding of fundamental principals of exciton generation, their dissociation and charge transport was criti-cal for the further investigation of CNT hybrid nanostructures. Especially for individual semiconducting CNTs, substantial progress has been achieved in the investigation of excitonic nature of photoexcitations and charge separation through observation of direct photoconductivity employing photocurrent spectroscopy and photovoltage imaging [3, 16–19]. The electrons and holes were separated by the external field [3], either by internal field at Schottky barriers with a metal electrode [16, 17], or at CNT p–n junction [18, 19]. Distinct from individual nanotubes, in CNT films, the directed photoconductivity is mostly masked by a more pronounced bolometric response [30] or it exists at a very short time scale [1]. The heating effect occurs when the energy of absorbed light is transferred to the CNT film through non-radiative decay owing to strong electron–phonon coupling. CNT film comprises a lot of nanotube–nanotube junctions which are responsible for the fast non-radiative relaxation. Because the semiconducting nanotube network is usually characterized by the negative temperature coefficient of resistance (TCR), the heat-ing induces a current increase similar to the effect of direct photoconductivity.

For hybrid nanostructures where the critical feature is the interface between CNTs and other photoactive components, the situation is quite different. In such a system, the major source of the photoresponse is free carriers generated and separated at the interface due to the internal built-in electrical field or PICT. Besides, light can be absorbed not only by CNT but also its counterpart. Distinct from pristine CNTs, interface related photoconversion processes are more complex but at the same time very intriguing, rich in novel phenomena and are extremely attractive for many opto-electronics applications. For instance, very recently, a surprisingly high photocon-version efficiency (PCE) of ~14 % for CNT/Si hybrid cells has been reported [31], exceeding any PCE for organic and hybrid photovoltaics (PV). Noteworthy, research in the field of CNT/semiconductor PV is very new (just past 5 years) and limited by a few groups [31–34] as compared with substantial efforts and time (about 20 years) spent by the PV community in other directions such as polymers based and dye-synthesized solar cells where the best PCE is still in the range of 10–12 % [35–37].

Figure 1.1 demonstrates a variety of photoactive CNT/X structures (where X is the material interfaced with CNT) including significant diversity in the CNT morphology (network and individual CNTs), their structure (SWNTs and MWNTs) and electronic properties (semiconducting and metallic). The choice of CNT counterpart can also be

3

very different including small molecules, oligomers, polymers, quantum dots and bulk semiconductors. CNT/X hybrid can absorb light through one component (CNT or X) or simultaneously through CNTs and X. The most studied structures exhibiting efficient PICT and utilized in solar cell architecture are CNTs/small molecules and CNTs/poly-mers, where CNTs act as electron acceptor (with some exceptions) and light is absorbed through the X component. Interestingly, in majority PV studies of CNT hybrids, the role of CNTs in light harvesting was underestimated in the photoconversion process. Recent reports demonstrated that CNTs can be involved not only in charge separation and transport processes, but also in efficient light absorption [33, 38].

This chapter is organized in the following way. The second section is dedicated to the photoconversion processes at interface between CNTs and other organic com-pounds (e.g. small molecules, fullerenes and polymers). We will start out with a brief review of PICT between SWNTs and small molecules and how this phenomenon can be employed in the design of novel organic and hybrid solar cells (electrochemi-cal and solid thin film). We will focus on distinctive features between PICT in liq-uid medium and PICT in the solid film, especially at the field effect conditions. The recent progress of CNT/polymer solar cells will be reviewed, emphasizing an impor-tance of CNT morphology and electronic properties for PV performance. We will then describe the SWNT p–n photodiode and effect of the multicarrier generation in carbon nanotubes. In the third section we will investigate PICT between CNTs and semiconducting quantum dots (QD) and outline some important aspect of such double nanostructures in terms of size, shape, and binding route to create a sharp junction interface. The integration of CNT/QD in solar cell architecture will also be considered. The fourth section will cover photophysics of charge generation and sep-aration at CNT/semiconductor heterojunction with major attention paid to bulk and nanostructured Si as the most common material in the solar cell industry. We will review the current status of CNT/Si solar cells, point out their features and excel-lent PV performance and discuss the perspectives in future research and develop-ment of these promising photoactive nanohybrids. Finally, in the fifth section, we will

Fig. 1.1 Schematic presentation of various CNT/X photoactive hybrid materials; inset shows structure of single walled carbon nanotube (SWNT) and multi walled carbon nanotube (MWNT); CNT*—carbon nanotubes with different electronic properties than CNT

1.1 Introduction: Light Absorption and Charge Separation


describe CNT based bolometers and discuss more specifically bolometric response of CNT/polymer composite. We will provide a comparative analysis of pristine CNT and CNT/polymer bolometers emphasizing a critical role of polymer matrix in the achievement of high responsivity and temperature coefficient of resistance.

We assume that the reader is familiar with the basis of CNT physics and optics in terms of band structure, chirality, band gap dependence on diameter, semicon-ducting and metallic properties whose definitions and related theoretical models can be found elsewhere [4, 5].

1.2 CNT/Organic Based Photovoltaics and Photodetectors

1.2.1 CNTs Interfaced with Small Molecules

In this section we will discuss the major directions in the research of nanoscale pho-toactive composites consisting of light harvesting small molecules (dyes) assembled with carbon nanotubes (covalently or by supramolecular interaction). Such hybrid sys-tems could be utilized as efficient building blocks in the design of organic optoelec-tronic devices because of remarkable photoconversion properties originating from the nature of nanohybrid components: a very high coefficient of absorption of many dyes in the broad spectral range; favorable position of HOMO/LUMO levels between CNTs and dyes for photoinduced charge transfer (PICT); and exceptional carrier mobility in CNTs. In the future, this may lead to the construction of nanoconjugated assemblies with great promise for the solar energy conversion and organic photodetectors. Distinct from CNT/polymer photoactive composite (see Sect. 1.2.3), the photoinduced charge transfer between several well known dyes (porphyrins, phthalocyanines, ferrocenes, etc.) and CNTs has been studied very intensively during the past decade [11, 39], due to a relatively simple hybrid structure allowing for an unambiguously interpretation of experimental data obtained from time-resolved transient absorption (TA) and lumines-cence spectroscopy (important tool for study of PICT process). In addition, many efforts were undertaken to create a variety of CNT/dye nanohybrids with established synthetic routes to understand how the specific structure and assembly method (covalent binding, noncovalent interaction) affect the PICT process. In the beginning of this section, we will briefly describe the main synthetic approaches allowing today, create a great vari-ety of dye/CNT donor–acceptor hybrids followed by a more specific discussion of their PICT properties and finally turn the attention to possible applications in electrochemical and solid film organic solar cells. Also, we will describe some interesting aspects of the photoconversion process at conditions of the electric field effect.

1.2.1.1 Photoinduced Charge Transfer Between Small Molecules and CNTs

There are several excellent recent reviews [11, 12, 39, 40] presenting many syn-thetic routes to create CNT/dye complexes exhibiting a strong charge transfer

5

between components (where CNTs act mostly as an electron acceptor). Here, we just summarize these results focusing on some PICT features between donor and acceptor as well as the efficiency of the energy conversion for PV applications. To date, two general approaches are employed for assembling small molecules with CNTs: covalent attachment of molecule to the open end or side wall of CNT and noncovalent interaction of molecule core with conjugated system of nanotube [11].

Covalent binding provides better complex stability (Fig. 1.2, right); however, this also induces additional CNTs defects, as a result of chemical functionalization. This is not a desirable effect because the surface defects on nanotube walls can significantly reduce the carrier transport along the nanotube conjugated system. CNT oxidation in strong acids (as a precursor formation for covalent binding) may irreversibly disrupt an intrinsic electronic structure of CNTs [41, 42]. In addition, most works on CNT functionalization, except a few recent ones [43, 44], were carried out with SWNTs of different chirality in mixture of metallic (m-SWNT) and semiconducting (s-SWNT) tubes. In general, m-SWNTs are more reactive than s-SWNTs, and their reactivity also depends on the tube’s diameter. Therefore, covalent binding at such conditions may lead to sufficient diversity in the photochemical and photophysical characteristics of the donor–acceptor ensembles, preventing a clear interpretation of the experimen-tal results. Examples of the covalently formed CNT/dye nanohybrids include metal and free based porphyrins [11, 45–49] (Fig. 1.2, left); phthalocyanines [39, 50–54]; ferrocenes [55, 56]; tetrathiafulvalene [57, 58], naphtalimide, fluorecein [59]; ruthenium (II) bipyridine complex [60] and other light harvesting molecules [11].

Noncovalent functionalization of CNTs (Fig. 1.2, right) is the more favorable approach in the context of PV applications, as carbon nanotubes preserve their

Fig. 1.2 Examples of covalent binding of porphyrin to SWNT trough ester bond (left, adapted from [46]); and non-covalent binding of Zn porphyrine functionalized with four pyrenes to (6,5)-SWNT (right): a ZnP-pyrene structure; b donor–acceptor nanohybrid as a result of π–π inter-action [43]. Reprinted with permission from Journal of American Chemical Society, 2005, 127 (19), pp 6916–6917, Copyright © 2005 American Chemical Society; Adapted with permission from Journal of American Chemical Society, 2010, 132 (23), pp 8158–8164, Copyright © 2010 American Chemical Society



pristine conductive properties which is important for an efficient carrier transport and charge collection at the electrodes. A common route for such supramolecu-lar association is the use of van-der Vaals or/and electrostatic interaction between CNTs and electron donor molecules [11, 12]. For example, one of the first studies [61] reported that the noncovalent attachment of porphyrine molecule to the side wall of SWNT through π–π interaction and demonstrated electron transfer from the porphyrine core to nanotube conjugated system. In another study [62], the pos-itively charged groups of cationic pyren (Pyr) attached to SWNT by π–π stack-ing to form an assembly with porphyrin (P) bearing negatively charged groups by electrostatic forces. Thus, photoactive complex SWNT/Pyr+/P− has been created without chemical functionalization of SWNTs. The complex formation was con-firmed by spectroscopic studies: the red shift of Sorret band in absorption spec-tra and porphyrine fluorescence quenching due to PICT between porphyrine and SWNT. The direct evidence of the electron transfer from porphyrine core to nano-tube with a transfer rate of ~0.2 ns was obtained through time-resolved transient absorption spectroscopy [62].

An alternative opportunity to outer-shell CNT fictionalization is the encapsula-tion of small molecules inside the inner nanotube shell. Such an approach could be considered as the next step towards completely preserving CNT electronic structure as compared to outer-shell noncovalent and covalent functionalization. Although most of the inner-shell fictionalization techniques utilized metals or molecules with low absorbance coefficients, several studies reported about photosynthesizing of SWNT by the encapsulation of conjugated dyes such as carotene [63, 64] and squarylium [65]. Photoluminescence spectra revealed energy transfer from the trapped molecules to the SWNTs. Note that in these studies, the only energy transfer from the small molecules to SWNTs was observed (at least no direct evidence of PICT was presented).

Recently, with significant progress in CNT sorting and separation, several studies were focused on PICT in SWNT/dye hybrids incorporated nanotubes of different chirality and diameter. D’Souza’s group, investigated PICT between Zn-Porphyrine (ZnP) and Zn-Phtalocyanine (ZnPc) and SWNT of different diameters corresponding (6,5) and (7,6) chiralities [43, 44]. In one report [44], SWNTs were noncovalently modified by pyrene cation and subsequently assem-bled through electrostatic interaction with ZnP and ZnPc decorated by benzo-crown macrocycles. Utilizing time-resolved fluorescence and transient absorption spectroscopy, it was revealed that charge separation and charge recombination processes in SWNT/Pyr/ZnP (or ZnPc) hybrids depend on SWNT chirality. The rate of the charge separation process for ZnP and ZnPc hybrids with (7,6) SWNTs (~2–4 × 109 s−1) exceeded that for the same hybrids formed with (5,6) SWNTs (~1.7–3 × 109 s−1) (Fig. 1.3).

The rate of charge recombination has been determined utilizing transient absorption experiment from the temporal decay of the intensity of bands assigned to ion radical pairs (direct indication of charge transfer and recombination pro-cesses), where a similar trend was observed: prevailing recombination rate for (7,6) SWNT hybrid over (6,5) SWNT hybrids. Finally, the ratio of charge separate

7

rate to charge recombination rate (important parameter for possible PV applica-tions) was estimated, which again, depended critically on SWNT diameter and chirality. This ratio (can be defined as a figure of merit in terms of charge stabi-lization) is generally better for hybrids from (7,6) nanotubes compared to hybrids comprised of (6,5) nanotubes. A similar trend was observed for diameter-sorted SWNTs assembled with porphyrins modified by covalently attached pyrene mol-ecules [43]. To interpret PICT dependence on nanotube diameters, the authors applied Markus model [66, 67]. However, a clear understanding of this inter-esting phenomenon requires further experimental and modeling efforts. In the case of SWNT/C60 nanohybrid, electron transfer from SWNT (nanotubes here act as electron donor) depends on the nanotube diameter as well [38, 68] and can be explained by a small mismatch between SWNT and C60 LUMO energy levels [38] (see more in the Sect. 1.2.2); however, this is not appropriate for SWNT/porphyrine (phthalocyanine) hybrids [43, 44, 69].

1.2.1.2 Electrochemical and Thin Solid Solar Cells Based on CNT/Dye Nanohybrids

In parallel with the synthesis of various photoactive CNT/dye nanohybrids, sub-stantial efforts were undertaken to integrate these structures as building optoelec-tronic blocks in the architecture of electrochemical and thin film solar cells. The incorporation of carbon nanotubes in dye synthesizing solar cells (DSSCs), a spe-cial type of electrochemical solar cells with relatively high PCE, will be considered in the next section. For proper functioning of photoelectrochemical cell (PELC), two main conditions should be fulfilled: (i) photosensing material should be depos-ited on the transparent conductive electrode (e.g. ITO) and (ii) a presence of liq-uid electrolyte supplying electrons (through mobile anions) for the regeneration

Fig. 1.3 Schematic representation of the energy-level diagram for photoinduced charge-separa-tion (CS) processes of SWNT(n,m)/PyrNH3+:MP (MP = ZnP an ZnPc). CS-1 and CS-2 pro-cesses produce radical ion pairs SWNT−:MP+ and SWNT+:MP− respectively [44]. Reprinted with permission from ChemPhysChem, 2011, 12, 2266–2273, Copyright © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim



(reduction) of the photoactive dye. The major parameter characterizing PV perfor-mance is power conversion efficiency (PCE), ηp,, which is defined as: ηp = FF * JSC * VOC / P where VOC is the open circuit voltage, JSC is the short circuit current, FF is the fill factor, and P is the incident power. Another important parameter is external quantum efficiency (EQE) or incident photon-to current efficiency (IPCE), which defines a percentage of photocarrriers per incident photon and can be calcu-lated through the expression EQE(λ) = 100 % * 1240 * JSC / (P * λ), where λ is the wavelength in nanometers, current in amperes and power in watts.

One of the first reports about PELC based on SWNT/dye hybrid [70] describes the formation of photosynthesizing film on ITO electrode of layer by layer dep-osition. Initially, as a based layer polyelectrolyte (such as PSSn−) was deposited on ITO electrode followed by deposition of SWNT/Pyr+ and finally oppositely charged ZnP8− resulting in stack SWNT/Pyr+/ZnP8−. Maximal EQE at 420 nm wavelength was reported as 4.2 %, which exceed by a factor of four the EQE of the control sample (without SWNT/Pyr+). The electrochemical nature of photo-conversion allows to utilize biasing of the working electrode (with respect to the reference electrode) to affect the photocurrent magnitude. For example, applying positive bias 0.2 mV, a sixfold photocurrent increase was observed [70]. In the study [71], EQE of 9.9 % has been recorded for similar structure (SWNT/ZnP) when bias 0.5 V was applied.

EQE of photocells constructed from SWNT/Pyr/ZnP (or ZnPc) nanohybrid with nanotube of a different diameter and chirality demonstrated a good correla-tion with efficiency of photoinduced charge separation and recombination pro-cesses [44]. The working electrode was prepared by drop coating of nanohybrids on FTO/SnO2 slide, using Pt counter electrode and I−/I3

− in acetonitrile as a redox electrolyte. The highest EQE ~12 % was observed for SWNT(7,6)/Pyr/ZnP mod-ified electrode that is consistent with the figure of merit for charge stabilization (see previous Sect. 1.2.1.1).

Umeyama et al. [72] reported the correlation of EQE with the degree of SWNT bundle functionalization by alkyl chains and porphyrins. It was found that the degree of SWNT functionalization exhibited selectivity to the diameter of SWNT bundles, so that PELC with the lowest bundle diameter and dense functionaliza-tion demonstrated the enhanced EQE (4.9 %, at 400 nm).

Another strategy to build the photoactive material on the transparent elec-trode is electrophoretic deposition [73, 74]. In particular, in the electric field of 200 V applied between two electrodes, SWNT/protonated porphyrin hybrids were assembled on nanostructured SnO2 electrode [74]. The photoelectrochemical cell was constructed of transparent SnO2 electrode with deposited clusters of SWNT/porphyrin hybrid immersed in the electrolyte solution (0.5 M NaI and 0.01 M I2 in acetonitrile) and Pt counter electrode. The photocurrent spectra demonstrated a band in the spectral range of the porphyrin Soret band with EQE = 13 % at an applied bias of 0.2 V versus saturated calomel electrode. However, PCE was very small (0.012 %) even at low light intensity (12.4 mW/cm2). Interestingly, the majority of studies related to photolectrochemical cells composed of CNT/dye hybrids [43, 44, 70–72, 75–77] (except DSSCs and CNT fiber based PELC [78])

9

rarely report an attained PCE at AM1.5 solar spectrum, and only confining itself with EQE data. A probable explanation for this is a low PCE, less than 0.1 %, that cannot compete with DSCC [36, 37] or thin film organic solid solar cells [35].

Special attention should be paid to the recent report [78] describing PELC fabricated from CNT fiber infiltrated with N719 dye (bis(tetrabutylammonium) cis-diisothiocyanato-bis(2,2′-bipyridyl-4,4′ dicarboxylato)ruthenium(II)) and dem-onstrating substantial PCE in the range of 2.1–2.6 %. Highly aligned micro fibers of CNT (diameter from 6 to 20 μm) have been produced by spinning from high quality nanotube array grown by the CVD technique [79, 80]. Authors believe that high PV performance is the result of the efficient electron transport (after charge separation at CNT/N719 interface) through aligned nanotube network within the fiber. Another key issue is the contact area between the fiber and the conductive substrate which critically depends on the fiber diameter [78].

It is worth mentioning the first report discusses photoelectrochemical cells that are capable of autonomously regenerating their photoactive functions [77]. Photoelectrochemical devices can suffer from continuous light exposure damaging dye molecules. This results in molecular photodegradation lowering the efficiency of photoconversion process. In nature, such a problem is solved by removing pho-todamaged fragments and replacing them with new species. Ham et al. [77] tried to reproduce such a self-repair process exploiting the chemical signal (adding/removing surfactant) to trigger dissembling/assembling the photoactive complex (Fig. 1.4). The natural photosynthetic reaction centers were non-covalently assem-bled on functionalized SWNTs. Upon the addition of surfactant (chemical signal), these photo-damaged complexes were desorbed from SWNTs and removed dur-ing dialysis with following replacement by “healthy” complexes and other com-ponents before next assembling with nanotubes (surfactant removal). Only in assembled state, the complexes exhibit photoconversion activity, where SWNTs

Fig. 1.4 Schematic of self-assembled photoelectrochemical complexes. The self-assembly pro-cess involves carbon nanotubes and photosynthetic reaction centres and occurs upon surfactant (sodium cholate) removal. Membrane dialysis induces spontaneous self-assembly of DMPC and membrane scaffold proteins to form nanodiscs, which reconstitute the reaction centres while sus-pending nanotubes in aqueous solution. The resulting, highly ordered complex is shown in the right-hand panel. Addition of sodium cholate completely decomposes the complexes back into the individual components in the initial condition (left-hand panel) [77]. Reprinted with permis-sion from Nature Chemistry, 2010, 2, 929–936, Copyright © 2010, Rights Managed by Nature Publishing Group



serve as hole conductors. Light absorption by the photosynthetic reaction center, consisting of photoactive proteins, results in charge separation with the following hole transfer to SWNT noncovalently assembled with reaction center through a lipid bilayer nanodisc. Despite low EQE (8 × 10−5 %) at 785 nm laser illumi-nation, an attraction of this study in prolongated system lifetime by reversible assembling and disassembling that allow the complete functional regeneration (after each regeneration cycle the photocurrent was restored to the previous maxi-mum during 168 h [77]). Such a remarkable feature provides a new platform in the building of PV mimicking the natural photosynthetic process.

Alternatively, electrochemically solar cells, thin film solid PVs composed of dye sensitizer and carbon nanotubes were investigated as well [81–86]. Note that all these structures included conjugated polymers or fullerenes forming planar or bulk heterojunction with CNT/dye nanocomposite to amplify photovoltaic charac-teristics, which could be negligible without polymer and fullerene additives. This is in striking contrast to fullerene/dye (e.g. phthalocyanine derivatives) thin film solid cells demonstrating a relatively high PCE ~4–5 % [39, p. 6805]. Kumakis et al. [81] proposed utilizing naphthalocyanine (NaPc) dye blended with SWNTs and poly(3-octylthiophene) (P3OT), expecting to enhance photoabsorbtion in the red spectral range and enhance the charge separation and charge transport through SWNT network. Naphthalocyanine has absorption coefficient of 105 cm−1, which by the factor of ten higher than that for SWNT:P3OT at ~700 nm and its HOMO/LUMO energy levels are favorable to donate electrons to SWNTs and holes to P3OT. Despite such a promising design, the conversion efficiency was low (<0.01 %) and did not demonstrate any enhancement as compared with the control sample (without NaPc). Low photocurrent was explained by the recombination process induced by trap sites created after dye introduction.

Silva’s group reported improved PV characteristics employing tetrosulfonate copper phthalocyanine (TS-CuPc) blended with MWNTs and interfaced with

Fig. 1.5 Left pictorial representation of cofacially extended TS-CuPc aggregates adsorbed onto the outer wall of a surface oxidized MWNT (o-MWNT); Right schematic energy level diagram for the bulk heterojunction organic solar cell: ITO/o-MWNT:TS-CuPc/P3HT:PCBM/LiF/Al [84]. Reprinted with permission from Langmuir, 2007, 23, 6424–6430, Copyright © 2007 American Chemical Society

11

PCBM layer [83]. Photoconversion efficiency of 0.45 % was determined at 1 sun, almost two times higher than the control sample without dye. Such PCE enhance-ment was a result of the VOC increase (ISC and FF were practically unchanged) that is consistent with an increase of the work function of oxidized MWNT (4.1 eV) with respect to ITO. Later, the structure of TS-CuPC/MWNT hybrid had been determined [84] from spectroscopic and morphological experiments reveal-ing the MWNT scaffold is decorated by adsorbed stacks of TS-CuPc molecules (Fig. 1.5, left). The solar cell has been constructed in a similar study [83]; how-ever, the PCBM layer was replaced by a PCBM:P3HT blend to improve charge collection efficiency. The proposed charge separation and transport scheme suggested charge separation at PCBM:P3HT bulk heterojunction followed by holes and electrons transporting through MWNT and PCBM network respec-tively (Fig. 1.5, right). It was found that PCE = 1.25 % of the composite device exceeded ~25 % PCE of control samples (without MWNT or TS-CuPc).

Other studies [85, 86] also demonstrated an improvement of PV performance of CNT/dye based thin film solar cells with respect to controlled devices (with-out dyes) owing to dye sensitizing (perylene derivatives [85] and oligothiophenes [86]). However, PCE was still lower than 0.1 %. Thus, an incorporation of sen-sitizing dye in CNT based solid cells is indicative of possible benefits for the future development of thin film OPV despite an insufficient conversion efficiency attained so far.

1.2.1.3 Incorporation of CNTs in Dye Synthesized Solar Cells

Dye synthesized solar cells (DSSCs) are a special class of photoelectrochemical solar cells comprising nanostructured oxide semiconductors (TiO2), Ruthenium based photosynthesizing dye (Ru(II) complex) and iodide electrolyte. Since Grätzel and O’Regan first reported this new type of solar cell in 1991, [87] DSSCs have attracted great interest due to their simple and cost effective fabrication and relatively higher photoconversion efficiency than other types of electrochemical and solid organic PVs [88]. Despite the fact that DSSCs recently overcame the PCE threshold of 11 %, [36, 37] further improvement in the performance of these solar cells is still required. Several major challenges facing the future development of DSSC are mostly related to the leakage of liquid electrolyte, dye photodegra-dation, suppression of photoelectron recombination and improvement of electron transport trough TiO2 network. Two later challenges (prevention of charge recom-bination and improvement of charge transport) inspired intensive research attempt-ing to incorporate the carbon nanotubes in TiO2:Ru(II) composite interfacing with liquid electrolyte [60, 89–100].

Although the employment of CNTs as an additive to electrolyte and as replace-ment of counter Pt electrode are also beneficial for DSSC performance [98], in this section we will consider only photoanode structure: transparent conducting electrode/CNT:TiO2:Ru(II) complex, where the addition of CNT improves elec-tron transport and suppress charge recombination.



Generally, the incorporation of carbon nanotubes in nanostructured photoanode of DSSC can be performed by several ways: (i) mixing CNTs and TiO2 followed by deposition on the transparent conductive electrode and soaking in solution with Ru(II) complex [60, 89, 94]; (ii) covalent attachment of Ru(II) complex to CNTs and their deposition on TiO2/electrode [60]; (iii) electrophoretic deposition of CNTs on electrode followed by spreading of a TiO2 colloid on top, annealing and soaking in Ru(II) complex solution [91]. In the first reports [60, 89], two approaches of SWNT insertion were employed resulting in higher short current value (approach (i)) and additional increase in open circuit voltage (approach (ii)). If an increase of ISC could be anticipated owing to the improvement of electron transport through SWNT/TiO2 network, a VOC enhancement is not so evident and is probably associated with a neg-ative shift of flat band potential of TiO2 (apparent Fermi level) because of the basic-ity increase of TiO2 surface in the presence of NH group of Ru(II) complex [60].

Interesting results highlighting the SWNT’s role in the DSSC functioning have been obtained in the study [91]. Carbon nanotubes were incorporated in TiO2 net-work by electrophoretic deposition to build the conductive scaffold facilitating the carrier transport in mesoscopic semiconductor films. The transient absorbtion and emission time-resolved spectroscopy was applied to estimate the rate of electron transfer from Ru(II) complex to TiO2, which was unchanged after the addition of carbon nanotubes. Meanwhile, the rate of the back electron transfer (recombination rate) was decreased in the presence of SWNTs. Overall, DSSC with incorporated SWNTs demonstrated an enhancement in the short current generation (as a result of improving of electron transport and suppressing of recombination), but this ben-eficial trend was compensated by the reduction of the open circuit voltage as the apparent Fermi level of the TiO2 in the presence of SWNTs becomes more positive than that of pristine TiO2 [90, 91] (Fig. 1.6). Note an opposite trend in VOC values in report [60], where the negative shift of the apparent Fermi lever was observed.

Fig. 1.6 Energy diagram illustrating the charge injection from excited sensitizer (S*) into TiO2 and transport of photoinjected electrons to the electrode surface without (a) and with (b) SWNT network (The Fermi level of TiO2 (Ef

′) shifts to more positive potentials (Ef″) as it equilibrates

with SWNT) [91]. Reprinted with permission from Journal Physical Chemistry C, 2008, 112 (12), pp 4776–4782, Copyright © 2008 American Chemical Society

13

There was also a reported improvement of DSSC performance (with natural dye extracted from the plant) by incorporation of MWNTs in photoanode struc-ture by a modified acid-catalyzed sol–gel procedure [95]. The enhancement of the short circuit current was associated with the increased adsorption area of thin films and the improved interconnectivity between TiO2 particles. However, at higher levels of loading, MWNT absorbance and scattering result in the screening effect for dye absorbance and consequently lowering the device efficiency. An optimal loading of 0.3 wt% MWNTs was proposed [95], resulting in PCE enhancement by 30 % (from 0.278 to 0.359 %).

Substantial progress in the performance of DSSC photoanodes fabricated from CNT/TiO2 nanohybrids has been observed for the past couple of years [93, 94, 99, 100] that is associated mostly with more elaborate processing techniques and the employ-ment of graphene as an additive to carbon nanotubes and titanium dioxide. Oxygen plasma treated SWNTs (PT-SWNTs) were incorporated in TiO2 matrix exhibiting an advantage over chemically oxidized SWNTs and pristine TiO2 alone. It was shown that plasma oxidation provides more uniform distribution of nanotubes in TiO2 matrix compared with the traditional treatment of nitric or sulfuric acids to disperse SWNTs. Such uniform dispersion increases the photoelectron transport, minimizing the charge recombination/trapping at TiO2 grain boundaries (the major source of the photocur-rent losses). As a result, the PT-SWNT/TiO2 photoanode demonstrated PCE = 6.34 %, which is over 75 % greater than the PCE of the cell with pristine TiO2 electrode [93]. Also, Yu et al. [94] reported a PCE enhancement (4.71 %) at optimal MWNT loading (<0.1 wt%) as compared with control device without MWNTs (3.94 %).

Growing interest in fabric based photovoltaics stimulated recent research in the field of the fiber DSSCs where CNTs were employed to wrap the photoactive cylindrical layer [101] or to work as coaxial working and counter electrodes [102].

Fig. 1.7 Schematic illustration of a wire-shaped DSSC fabricated from two CNT fibers. a Two CNT fibers twined into a cell. b Top view of a cell. c Working mechanism [102]. Reprinted with permission from Nano Letters, 2012, 12 (5), pp 2568–2572, Copyright © 2012 American Chemical Society



Zhang et al. [101] proposed utilizing the CNT network as a transparent anode by wrapping CNT film around TiO2/dye array deposited on core Ti wire. Such single-wire DSSC demonstrated a PCE of 1.6 % under 1 sun (AM 1.5) which was further improved up to 2.6 % by the incorporation of a second conventional metal wire. Figure 1.7 shows DSSC design [102], where CNT fibers serve as coaxial elec-trodes (anode covered by the layer of TiO2/dye and counter electrode) immersed in the electrolyte solution. CNT wires were produced by the same technique [79, 80] applied to fabricate fiber shaped PECL [78] (see Sect. 1.2.1.2). The power conver-sion efficiency was reported as high as 2.94 % (1 sun, AM 1.5), which can be rated as an excellent result for such unique device architecture. Note that this is a first fiber shaped DSSC without metal electrodes which usually suffer from corrosion in fabric PVs. In addition, such flexible fiber solar cells may be easily scaled up, as PCE is not dependant on the cell length.

From a general point of view, the replacement of CNTs by 2D graphene may benefit from the better organization of the conducting scaffold in TiO2 matrix, because two dimensional graphene not only has a high electron mobility but also a strong interaction with titanium dioxide owing to electrostatic binding and charge transfer interaction [103]. Introduction of graphene in small loading (0.6 %) resulted in a faster electron transport and a lower recombination, together with a higher light scattering. Overall, PCE was 6.97 %, which was increased by 39 %, compared to the nanocrystalline titanium dioxide photoanode [104]. On the other hand, the study [100] did not reveal an advantage of graphene over carbon nanotubes, and only their blend exhibited better PV performance. The photoconversion efficiencies of DSCC photoanode prepared from pristine TiO2, TiO2-graphene, TiO2-CNT, and TiO2-graphene-CNTs were 4.54, 5.35, 5.50 and 6.11 % respectively.

Clearly, an incorporation of carbon nanotubes in mesoscopic titanium dioxide is beneficial for DSSC performance because of higher electron transport, suppressed recombination, and enhanced light scattering. Almost all studies [60, 89–100] report an increase of PCE of CNT/TiO2 based cells in the range of 30–50 % over devices with pristine TiO2. A rough estimate (taking conservative 30 % improve-ment and highest today DSSC conversion efficiency of 12 %) with the addition of some imagination could result in amazing conversion efficiency as high as 16 %.

1.2.1.4 Dye/CNT Phototransduction in Solid State: Phototransistors and Opto-Electrical Molecular Switches

In previous sections (1.2.1.1–1.2.1.3) photoinduced charge transfer and related pho-tovoltaic devices (PELC, DSSC) have been considered for various dye/CNT nano-hybrids mostly in the liquid medium (with one exception regarding a short review of thin film solid PVs [81–86]). Here we will present the studies related to opto-electri-cal conversion mechanisms for dye/CNT assemblies in solid state. The importance of the photo transduction for solid phase is directly related to various applications of these nanohybrids as nanoscale modules in phototransistors, photodetectors, optical switches and thin solid film photovoltaics.

15

The simplest design for studying light modulated conductivity is the use of the lateral geometry of functionalized carbon nanotubes (individual or network) deposited on the insulating substrate between two electrodes. One of the first observations of such light-sensitive devices has been reported for film of SWNTs covalently bonded with ruthenium bypyrideine complex (Ru(bpy)3

2+) [105], dye which is extensively utilized in dye synthesized solar cells. Under illumination in the absorption band of Ru(bpy)3

2+ (440–550 nm), an increase of SWNT film con-ductivity has been observed. Because major carriers in semiconducting SWNT are holes, the conductivity enhancement was explained by the electron withdrawing from SWNT by photoexcited dye molecule (electron transfer from SWNT to dye) according to the scheme:

The dynamic of this process was characterized by the relatively fast conductiv-ity enhancement (~1 s) and slow relaxation time in the darkness (~70 s) associated with back electron transfer from Ru (bpy)+3 to SWNT+ [105]. However, these val-ues cannot be assigned to actual rates of charge transfer between dye and SWNT (distinct from the case of isolated hybrids when time-resolved luminescence and photoinduced absorbance can be employed [11]) because of substantial contribu-tion in photoconductivity dynamics the photocarrier trapping, recombination and interaction with oxygen in dye/SWNT film resulting in reduction of total rate constants.

Field effect transistor (FET) design provides more detailed information about photoconductivity mechanisms (Fig. 1.8) than a simple two terminal resistor. The main idea of FET functioning is associated with control of the ISD current by gate voltage VG changing the concentration of charge carriers in source-drain chan-nel (Fig. 1.8). Interaction of various molecules with SWNTs using field effect has been intensively studied for the past two decades. Molecules attached to SWNTs can affect the conductivity in the Source-Drain (SD) channel in three major ways (i) changing the concentration of major carriers in SWNT due to charge transfer;

Ru (bpy)2+3

/

SWNThv−→ Ru (bpy)

+3

/

SWNT+

Fig. 1.8 a Schematic view of the FET with network of functionalized SWNTs; b typical response on light of ISD(VG) curve



(ii) changing the local electric field; and (iii) serving as scattering centers or traps reducing the carrier mobility. As it was established in many SWNT studies, FET allows to distinguish between these three mechanisms: the change of the carrier concentration and local electric field results in the shift of the gate voltage thresh-old VG

T, while a change in mobility alters the tilt of ISD (VG) curves (Fig. 1.8) [106]. Mechanism (ii) is rare enough and mostly related to molecules with a large dipole moment changing under the light illumination (photoisomers) and can be also be discriminated from mechanism (i) as we will show further.

Hecht et al. investigated the photoinduced charge transfer (PICT) between SWNT and zinc-porphyrin at field effect conditions [106]. It was found that under illumination in the Soret band of porphyrin (420 nm), the threshold of gate volt-age exhibits the positive shift which is indicative of electron transfer from SWNT to photoexcited zinc-porphyrin. The quantitative parameter of electron transfer was estimated as 0.37 electrons per porphyrin molecule at light intensity of 100 W/m2. Surprisingly, the direction of electron transfer in this case was opposite to electron transfer studied in similar hybrids, where photoexcited porphyrin was an electron donor [11, 45–49]. Authors explain this discrepancy by the fact that initially, elec-tron transfer from porphyrin to nanotubes occurred in the ground state followed by partial recovery (back electron transfer) in the excited state. Nevertheless, such inter-pretation cannot be in full compliance with results of the studies [11, 45–49], where electron transfer from photoexcited porphyrins to nanotubes was confirmed by time-resolved luminescence and photoinduced absorption spectroscopy. Probably, the existence of the porphyrin/SWNT hybrid as an isolated species (in solution) and in the solid film is so different that it could affect the direction of electron transfer (e.g. orientation of molecules on nanotubes, existence of SWNT bundles, proximity to junction in solid film, etc.). This point is supported by the results when SWNT also acts as an electron donor in PICT between SWNT and Ru(bpy)3

2+ in solid film [105], although electron transfer from photoexcited Ru(bpy)3

2+ to SWNT has been indirectly established for this nanohybrid incorporated in dye synthesizing solar cell [60]. It should be noted, that there are scarce numbers of CNT/small molecules sys-tems where CNTs work as an electron donor [11].

An elegant study of PICT between SWNT and Ru(bpy)32+ in FET geometry has

been carried out by Zhao et al. [107] demonstrating a similar trend: electron trans-fer in the excited state from SWNT to the photosensing molecule. Here, adamantly-modified Ru(bpy)3

2+ complex was immobilized as a guest inside cyclodexrtrine cavity non-covalently attached to SWNT through pyrene tether. Upon exposure to light at 490 nm (maximum of Ru(bpy)3

2+ absorption), the positive shift of the threshold voltage of ISD (VG) curves has been observed indicating electron transfer (SWNT hole doping) from SWNT-cyclodextrine to adamantly-Ru(bpy)3

2+ complex.Photochromic dye spiropyran, self-assembled with individual SWNT was inves-

tigated as opto-conductive switch in FET geometry [108]. Under exposure of UV light spiropyran switched from colorless neutral form to a zwitterionic, colored form while reverse switching to original neutral form is induced by the visible light. The SWNT conductivity correlated well with optical switching demonstrating reduction of ISD and its recovery under UV and visible light, respectively.

17

Another interesting direction where the field effect is intensively employed is the study of the opto-electrical transduction of semiconducting SWNTs func-tionalized with molecules exhibiting photoisomerization in the excited state [109–111]. One of the most investigated group of organic photoisomers are azobenzen-based chromophores demonstrating the transfer from stable trans isomer in the ground state to meta stable cis isomer under the light excitation. Such photoisomerization is accompanied with a significant change in the dipole moment (~several Debays) changing the electrostatic potential and consequently modulating SWNT conductance. Such a mechanism of electrostatic switching is principally different from the above mentioned PICT effect utilized in SWNT/dye phototransistors [106–108], because no charge transfer occurred between chromophore and SWNT and only change of dipole moment affects the local electrostatic environment. This mechanism is more general than PICT as it does not rely on specific matching electronic states between donor and accep-tor required for PICT (e.g. SWNT and organic molecules) and can be applied to other semiconducting nanostructures. Also, such electrostatically switched nano-hybrids show no indication of photobleaching and are capable of functioning for a long period of time [109].

The first opto-electrostatically switched SWNT hybrid FET has been reported by Simmons et al. [109] for azobenzene based chromophore (Disperse Red 1 (DR1) dye was non-covalently attached to individual semiconducting SWNT through antracene tether). Upon the UV illumination in the band of DR1 absorption, the positive threshold shift of ~0.7 V was observed, which can be associated with PICT or change in the local electrostatic environment. The mechanism of PICT switching was almost ruled out because of the long spacer (between dye and nanotube) and control experiment, while modeling of quan-tum transport of FET characteristics in the presence of azobenzene dipoles (9D in trans and 6D in cis conformations) demonstrated excellent agreement with experimental data.

Recently, the electrostatic FET switching was employed for nanoscale color detection, where three azobenzene based chromophores (with absorption windows in blue, green and red spectral range) served as photoabsorbers and individual nanotubes as electronic red-out [110]. All three dyes were noncovalently attached to nanotubes through pyrene tether. It was shown that the spectral maximum of the threshold voltage for each dye correlates well with its absorption band. From ab initio calculations of these SWNT/dye nanohybrids the dipole moments in trans and cis configurations were calculated, followed by the estimation of the electro-static potential change (ΔV) under light excitation. Finally, shift of the gate volt-age ΔVG was determined as ΔVG ≈ 20 ΔV through the ratio of capacitance of nanotube to the total capacitance. Thus, a relatively small change in electrostatic potential (~0.1 V) is amplified to large gate voltage shift (~2 V) being in good agreement with experimental observations [110]. Note that the above model (as well as the model in [109]) is valid for the direction of dipole moment normal to the SWNT axis. Therefore, special efforts (spectroscopic analysis and ab initio calculations) were undertaken to gain information about structural characteristics



of DR1 dye attached to SWNT trough pyrene tether (DR1P) [112]. According to UV, Raman and XPS data DR1P should exhibit some parallel orientations of the dipole moment to nanotube axis; nevertheless, the second harmonic genera-tion (SGH) experiment confirmed a substantial contribution of normally directed dipoles in SHG signal [112].

More generally (not only in FET geometry), azobenzene based dyes (AZO) can be employed as nanoscale opto-electrical switches when interfaced with car-bon nanotubes. Theoretical study of charge transport through SWNT-AZO-SWNT junction revealed the importance of contact topology (cis/trans isomers, spacer length and flexibility) and nanotube chirality [113]. The theory predictions [113] were recently confirmed in experimental work, where a light driven optical switch prepared from AZO linked to CNT (few-walled nanotube) through two differ-ent spacers has been investigated in face to face geometry [114]. The resistance of CNT-AZO network critically depends on tunneling barriers between nanotubes determined by the AZO cis or trans conformation. Compared with short and rigid spacer, the flexible and long spacer between CNT and AZO facilitate dye isomeri-zation under UV light due to an increase of the free volume and results in more efficient switching effect.

A significant feature of light driven dye/nanotube FET is low light intensities for optical modulation of transistors (~10 mW/cm2 for PICT mechanism [106] and ~100 μW/cm2 for electrostatic mechanism [109, 110]) in contrast to intrinsic SWNT photoconductivity which typically require 1 kW/cm2 light intensities [3].

1.2.2 Role Carbon Nanotubes in Light Absorption and Photocarrier Generation: CNT/Fullerene Solar Cells and Photodetectors

It appears, somewhat ironically, that until recent time, most investigations of CNT composite for organic photovoltaics paid the most attention to the photophysical properties of CNT counterparts (e.g. polymers, quantum dots, small molecules) rather than carbon nanotubes themselves [115–117]. CNTs in these studies played a secondary role functioning as external electrode or material for photo carriers accepting and transporting, where the primary source of photocarriers was a light absorption in the organic counterpart.

Meanwhile, carbon nanotubes posses efficient absorption in Vis–NIR spectral range being the darkest material in the world [118]. For example, semiconducting SWNTs (s-SWNTs) have absorption coefficient ≥105 cm−1 at their NIR bandgap providing a complete light absorption for film of 100 nm thickness [119]. Because the substantial part of solar spectrum covers the NIR range, and there are almost no organic compounds absorbing NIR light, SWNTs are highly demanded for organic PV as NIR efficient absorbers. In addition, the optical band gap of s-SWNTs can be tuned to the desirable part of optical spectrum by varying the carbon nanotube diameter in the broadband range from visible/NIR to mid-IR [4, 5].

19

The first study of NIR photoresponse of semiconducting SWNT (s-SWNT) in a blend with polymers has been reported in 2005 [120]. It was demonstrated that photocurrent spectra in NIR spectral range correlates well with absorption spec-tra of s-SWNTs. An achievement of composite photoresponse as a result of NIR absorption and charge separation clearly indicated to a new pathway in the con-struction of composite solar cells and photodetectors with the predominant role of CNTs as NIR absorbers.

A primary source of electron–hole pairs photogenerated in s-SWNT is exci-tons [13–15]. However, without external or internal electric fields, excitons can-not dissociate as the exciton binding energy EB ~0.2 eV exceeds thermal energy kBT at room temperatures. In the case of composite SWNT/X structure (X—could be polymers, small molecules, fullerenes, quantum dots, semiconductors, see Fig. 1.1), exciton dissociation occurs through the internal built-in field at the inter-face between SWNT and X component (type-II heterojunction), where energy offset at the interface exceeds EB [121]. Energy offset is defined as a difference between electron affinities (ΔEA) or ionization potentials (ΔEIP) of s-SWNT and X component. After exciton dissociation at the interface, separated photo carri-ers diffuse through both mediums to external electrodes providing photovoltaic effect either photocurrent for photodetecting device under biasing. Thus, a criti-cal issue for effective photocarrier separation at the s-SWNT/X interface is the proper band energies of both components satisfying condition ΔEA, ΔEIP ≥ EB. Recently Bindl et al. [121] reported an investigation of photoinduced charge sepa-ration at the interface between s-SWNTs and several organic compounds including polymers and fullerene derivatives. Two distinctive features in this study shed light upon the process of SWNT exciton dissociation at different interfaces: (i) the spec-trum of the incidents light was in the NIR range (900–1400 nm) matching the E11 electronic transitions of s-SWNTs of several different chiralities, where absorption of the X component is negligible; (ii) photosensitive capacitor measurements were employed avoiding the influence of the photothermal effect on the photocurrent passing through the interface. Table 1.1 shows the summary of this investigation with indication of average photoresponsivity averaged in the 900–1400 nm range with expected energy offsets for s-SWNT and each X component.

As followed from these results, the strongest exciton dissociation occurs at the s-SWNT/fullerene interface with electron transfer from nanotube to fullerene (Fig. 1.9b). This is consistent with observations of photovoltaic effect and pho-todetecting capability of SWNT/fullerene hybrid devices [38, 122] and previous studies of SWNT/C60 nanohybrids where carbon nanotube exhibited the electron donating properties [123, 124]. A moderate responsivity is observed for P3HT and P3OT components; however, in this case s-SWNT works as a hole donors. SWNT electron/hole donating properties from Table 1.1 is in agreement with the ratio between EB and ΔEA, ΔEIP values. In the case of fullerene, component excitons generated in SWNT are dissociated with the electron transfer to C60 (ΔEA ≥ EB); while for polymers, the more favorable is exciton dissociation with hole trans-fer to polymeric component (ΔEIP ≥ EB). Despite some simplification regard-ing correlation between photoresponsivity and offset/exciton binding energies,



this approach interprets qualitatively well the magnitude and sign of the charge transfer at interface of SWNT/X hybrids. Note, as a result of visible light absorp-tion, excitons generated in P3HT and P3OT are dissociated with electron trans-fer to s-SWNTs. Such process was considered to date as a major photosensitive mechanism for SWNT/polymer solar cells [115] (see next section). However, according to data from [121], a sizable contribution to charge transfer from NIR excitons generated in SWNTs (hole extracting from nanotubes) can be expected. With recent advances in SWNT sorting and separation (selection of s-SWNT fraction) and nanotube debundling, SWNT light harvesting in the NIR spectral range could significantly improve the performance of SWNT-polymer composite photovoltaics.

Another recent study by Arnold’s group [38] provides detailed information about electron transfer at SWNT/C60 interface depending on nanotube band gap and chirality. Here, an internal quantum efficiency (IQE) has been estimated for each E11 band in the range of 900–1500 nm corresponding s-SWNT chiralities (7,5), (7,6), (8,6), (8,7) and (9,7). Structure of the hybrid PV device and schematic of charge transfer at s-SWNT/C60 interface are shown in Fig. 1.9a, b. It was found that IQE values of ~0.8–1.0 for (7,5), (7,6), (8,6) chiralities were much larger than IQE ~0.2–0.4 for (8,7) (9,7) chiralities that correlate well with the dependence of the energy offset ΔEA (driving energy) on nanotube chirality/diameter and corre-sponding bad gap energy (Fig. 1.9c).

These results are consistent with the rule of ΔEA ≥ EB ~0.2 eV required for effi-cient electron transfer from SWNT to C60: for higher chiralities, this condition can-not be fulfilled because of the reduction of ΔEA and band gap energies. In addition, the dependence of IQE on the SWNT film thickness allowed for the estimation of exciton diffusion lengths in SWNT (~3 nm), which is usually affected by the qual-ity of individual SWNTs, their lengths, barriers at junctions, degree of bundling and film morphology. Obviously, the longer diffusion length should provide the better

Table 1.1 Comparison of measured photoresponsivity averaged over the range 900–1400 nm with expected energy offsets [121]

Material Averaged NIR Photoresponsivity (μA/W-1)

Carrier extracted from SWNT

Ionization potential (eV)

Electron affinity (eV)

s-SWNT 4.9–5.3 3.7–4.1C60 58 ± 27 e− 6.2 4[C61]-PCBM 93 ± 45 e− 6.1 3.8p3HT 12 ± 3 h+ 4.7 2.1rrP3HT 5 ± 2 h+ 5 2.2P3OT 11 ± 6 h+ 5 2.2MDMO-PPV 3 ± 2 5.3 2.8PVP <1 N/A N/APolycarbonate <1 N/A N/APFO <1 5.8 2.2

Adapted with permission from ACS Nano, 2010, 4 (10), pp 5657–5664, Copyright © 2010 American Chemical Society

21

photoconversion efficiency being in the same time commensurate with light absorp-tion length for efficient harvesting of solar energy. In this study, the light absorption length was determined as 25 nm for SWNT film [38]. This means that only very thin nanotube layer (~3–4 nm) are capable of transfering all absorbed light energy to electricity, while thicker films result in energy harvesting losses. This problem is typical for thin film organic PV when an exciton diffusion length is smaller than light absorption lengths.

The similar SWNT/C60 heterojunction has been exploited for design of the broadband Vis–NIR photodetector [122]. Previous efforts in the field of CNT based photodetectors were focused on individual CNTs [3, 16–19], CNT films [1, 2, 20–29] or CNT/polymer composites [125, 126] designed mostly in lateral geometry, and just a few studies [127, 128] reported about the fabrication of planar structures where photosensitive bulk or planar heterojunction is formed in the face to face geometry. Arnold et al. [122], demonstrated first, that planar SWNT/C60 heterojunction can be utilized for photodetector application and characterized the device in terms of spectrally resolved responsivity, IQE, detectivity and response time under zero and non-zero bias. It was found that the spectrum of responsivity in NIR range coincides with absorption bands of SWNTs assigned to semicon-ducting and metallic nanotubes of different chiralities (as distinct from study [38], where only s-SWNT were employed). The determined response time was short enough, of 7.16 ns (bandwidth of ~31 MHz at 3 dB), to be comparable with com-mercial photodetectors for high speed imaging applications. The device detectivity D* was more than 1010 cm Hz1/2W−1 from 400 to 1400 nm. Authors claimed that it was a first report “of organic photodetector with D* extending beyond 1000 nm, and with appreciated responsivity at λ > 1200 nm”.

In this section, we specially focused on reports [38, 121], because these studies shed light on the mechanism of exciton generation in s-SWNTs, their dissocia-tion at interface and direction of charge transfer. This analysis clearly indicates an underestimated role of s-SWNT in the past as an active photosensitive material in composite structures and provides the basis for their effective exploiting in the future (first of all as photoabsorbers) for IR detectors and solar cells.

Fig. 1.9 a Device architecture. b Schematic depicting charge transfer at nanotube/C60 interface. c Internal QE versus diameter and chirality (circles) compared with the expected energetic driving force for exciton dissociation at the s-SWNT/C60 interface (triangles) [38]. Reprinted with permission from Nano Letters, 2011, 11 (2), pp 455–460, Copyright © 2010 American Chemical Society



1.2.3 CNT/Polymer Solar Cells

For the past decade, organic photovoltaics (OPVs) have attracted a growing inter-est as an alternative approach to semiconductor based photovoltaics because of their low cost fabrication process, flexibility, light weight and large variety of organic materials. Today, the most promising among various OPVs and demon-strating the highest PCE are plastic solar cells comprised of various conjugated polymers with different organic additives required for formation of bulk or flat heterojunction [35, 129]. However, despite substantial research efforts in this field, the best PCEs for plastic cells still stay in the range of 6–8 % (published in scientific literature [130–132]) either 9–10 % (reported by several companies: Mitsubishi, Konarka and Polyera [133–135]).

The history of OPV research counts numerous studies aimed to maximize the conversion efficiency utilizing a various combinations of polymers, small mole-cules and inorganic nanostructures such as nanorods and quantum dots [35, 129]. The best OPV devices fabricated to date are based on bulk heterojunction (BHJ) between conjugated polymers (e.g. polythiophene derivatives) and C60 functional-ized fullerenes [130–132, 135]. The excitons generated in the polymer effectively dissociate at polymer/C60 interface followed by the electron transfer to C60 with holes remaining in polymer system. Holes and electrons are collected at external electrodes due to carrier diffusion from interface to the bulk. However, the elec-trons’ transport and collection at the external electrode depends strongly on the hopping mechanism (which cannot provide high charge mobility) and C60 aggre-gation required for the formation of conductive clusters. Therefore, it was initially anticipated that the replacement of C60 by carbon nanotubes possessing a high car-rier mobility and ballistic mechanism of the charge transport should significantly improve the OPV performance.

However, first studies of CNT/polymer OPVs reported poor PCE, which was less than 0.1 % [136–141]. Despite substantial efforts for almost a decade, the maximum reported PCEs for CNT/polymer devices stayed in the range of only 0.2–0.5 % [142, 143]. Therefore, special attention must be paid to recent studies [144, 145] focused on exciton dissociation, role semiconducting and metallic car-bon nanotubes and the morphology of SWNT networks.

Generally, the majority of works in the field of CNT/polymer solar cells dealt with insufficient dispersion of carbon nanotubes in polymer matrix (resulting in bundle formation) and the presence of metallic SWNT (statistically grown SWNTs compose of 1/3 metallic and 2/3 semiconducting nanotubes [4]). It was suggested that SWNT aggregation into bundles and the existence of metallic SWNTs are the main obstacles for efficient exciton dissociation, charge separa-tion and transport. Another reason related to the limited charge transport through SWNT network could be associated with carbon nanotube junctions (individual or bundles). It was shown [146] that the junction between metallic and semicon-ducting tubes creates a Schottky barrier with height about one half of the band gap of a s-SWNT. Blau et al. [147] tried to investigate how PV performance

23

depends on the type of CNT (SWNT, DWNT and MWNT were employed), their length and weight ratio in the blend with P3HT. However, obtained results were not conclusive; at least no correlation has been found between CNT type and PCE, suggesting that other dominating factors (e.g. CNT dispersion, impurities, metallic SWNT/DWNT fraction) are critical in the design of composite photoac-tive material. Such an outcome is understandable in the context of recent reports [144, 145], where the ratio of semiconducting/metallic SWNT [144] and SWNT morphology (different set of individual semiconducting SWNTs) [145] were well controlled.

Holt et al. [144] using the time-resolved microwave conductivity (TMRC) con-clusively demonstrated that semiconducting SWNTs (s-SWNTs) indeed play a critical role in charge separation and maintenance of long-lived free photocarri-ers, while metallic SWNTs (m-SWNTs) reduced the charge separation in SWNT/P3HM composite. TRMS pump-probe technique allows detecting only free carri-ers, product of exciton dissociation, at SWNT/P3HT bulk interface. This observa-tion is in good agreement with theoretical results by Kanai et al. [148] considering m-SWNTs as inhibitors of charge separation owing to redistribution of the charge density as a result of ground state interaction between P3HT and m-SWNTs. In contrast, following the same model, s-SWNT and P3HT form a type-II hetero-junction leading to enhanced population of free carriers (electrons in s-SWNTs and holes in P3HT) after exciton dissociation. Also, SWNT/P3HT type-II hetero-junction was confirmed experimentally in the study [149] utilizing highly ordered CNT/polymer structures. In addition to theoretically predicted mechanism of the negative impact of m-SWNT on charge separation [148], several other potential pathways for exciton and free carrier recombination at m-SWNT/P3HT interface were considered [142, 144]: (i) excitons can be effectively quenched due to the lack of an electronic band gap for m-SWNT; (ii) free photoelectron in P3HT at LUMO level can be captured by an empty midgap state of m-SWNT; (iii) photo-hole in P3HT can be quenched by electron transfer from m-SWNT filled midgap state to P3HT HOMO level. Thus, for an effective charge separation in the design of SWNT composite solar cells, the metallic SWNTs should be excluded as much as possible.

Other important aspects in the s-SWNT composite architecture for photovol-taic effect, such as SWNT bundling and positioning of SWNT Fermi level have been investigated by Strano’s group [145]. Semiconducting enriched SWNTs were grown by CVD method on Si/SiO2 substrate followed by the deposition of P3HT layer forming planar heterojunction between individual nanotubes and the polymer. The PCE per one SWNT has been estimated as 3 % and was based on the assumption that the actual device area is defined as A = W * L * N, where W = 2(R + Ld), R is the nanotube radius (~1 nm), Ld is the effective diffusion length of excitons in P3HT (~8.5 nm [150]), L is the nanotube length (~1 mm) and N is the number of individual nanotubes in the device architecture. Comparison between bulk heterojunction (1 wt% of SWNT in P3HT [142]) and flat heterojunc-tion [145] in the same units of actual area A per cm2 of surface area (2.25 cm2/cm2 and 0.00038 cm2/cm2, respectively) implies that PCE per nanotube in the device



with individual SWNT/P3HT planar heterojunction exhibit much higher photovol-taic efficiency than traditional polymer/SWNT solar cells with bulk heterojunc-tion. Authors believe that major reasons of such enhanced efficiency are associated with SWNT fully debundling (existing only individual SWNTs); the absence of the junction between SWNTs due to their alignment; nanotube high quality (D band in Raman spectra was not observed) and their semiconducting nature. The junctions in aggregated bundles of carbon nanotubes can be a source of exciton recombination [148, 149] and junctions between individual nanotubes (semicon-ducting and metallic) can considerably reduce the free carrier transport owing to the formation of barriers [146].

Also, any defects and impurities increase the carrier trapping and recombina-tion processes in carbon nanotubes. The proposed design of the device in study [145] (Fig. 1.10a) is almost free from the above deficiencies which are typical for most BHJ CNT/polymer solar cells. Indeed, as soon as electrons are injected in SWNT from P3HT, they are able to freely travel to an external electrode with-out loss related to trapping or barriers. Also, study [145] demonstrated the pos-sibility to convert SWNT from p-type semiconductor to n-type semiconductor by polyethylene imine (PEI) doping (Fig. 1.10b). It is known that PEI has strong electron donating groups and is often used to convert naturally p-type SWNTs to n-type SWNTs. PEI doping resulted in a sizable increase of the open circuit voltage (VOC), from 2.5 V for p-SWNT to ~0.5 V for n-SWNT, with little reduc-tion of the short circuit current (ISC). A VOC increase is consistent with the model developed for the plastic solar cells [151], where VOC value correlates to the dif-ference between LUMO energy level of the electron acceptor and HOMO level of the electron donor. Then VOC change can be explained taking into account that SWNT n-doping moves up SWNT Fermi level (Fig. 1.10b) almost through the whole bandgap above the P3HT HOMO level (Fermi energy in semiconduc-tors can be roughly interpreted as a LUMO level in molecular systems). This

Fig. 1.10 a Schematic of planar heterojunction solar cell based on a P3HT film on laterally aligned SWNTs. b Schematic energy level diagram of the P3HT/SWNT heterojunction solar cell. For s-SWNTs, the work function and bandgap are 4.73 and 0.5 eV, respectively. The Fermi level of the SWNTs is elevated by PEI doping into the SWNTs, resulting in n-type conversion of the SWNTs [145]. Reprinted with permission from ACS Nano, 2010, 4 (10), pp 6251–6259, Copy-right © 2010 American Chemical Society

25

observation demonstrates a promising approach allowing improved conversion efficiency employing a simple, wet processed doping of semiconducting carbon nanotubes.

Photoinduced charge transfer at SWNT/polymer interface is a complex process requiring deep understanding of nanoscale assembling of polymer chains with SWNTs and their electronic behavior at the junctions. In this regard, recent theo-retical calculations and molecular dynamics (MD) simulations shed light on intrin-sic electronic and structural features of SWNT–P3HT interaction [148, 152] which could be applied to optimize photovoltaic characteristics.

Kanai and Grossman [148]applied the density functional theory (quantum mechanical ab initio calculation) to evaluate such parameters as charge transfer in the ground state, density of state (DOS) in the interfacial region, energy alignment and built-in potential. Semiconducting (10,2) and metallic (12,3) SWNTs with the fragment of P3HT in periodically repeating super cell were taken for calcula-tions. The results showed the striking difference between s-SWNT and m-SWNT interfaced with P3HT: semiconducting tubes demonstrated the minimal interac-tion with polymer (charge transfer from P3HT to s-SWNT in the ground state was <0.02 electron) resulting in the ideal type-II heterojunction. Therefore, after exci-ton dissociation, an electron can be easily injected to s-SWNT conducting band leaving a hole at P3HT HOMO level. However, for metallic tubes, it was found that there was a strong interaction with the polymer (charge transfer was ~0.3 elec-tron), making P3HT attractive for the negative charge at interface. In this case, charge separation should be very limited at the moment of exciton dissociation. Also, as a result of such interaction, the built-in potential is quite small (0.06 eV) making difficult the diffusion of free photocarries from the interface to the bulk (assuming that charge separation is happened despite of its low probability). Thus, semiconducting SWNTs should be a critical component (as well as the absence of metallic tubes) in the SWNT/polymer photovoltaic device.

Complementary to ab initio calculation [148], molecular dynamic (MD) simu-lation has been employed to evaluate the nanoscale self-assembly of P3HT around SWNT and how the interface organization affects the rate of the charge transfer [152]. Bernardi et al. [152] revealed not only the different conformations of P3HT wrapped individual SWNT, but also the interaction of several SWNTs with a set of polymer chains for a more realistic situation. In addition, it was discovered that the tubes lead to an increase of P3HT conjugation length and more effective charge transfer process. Figure 1.11 illustrates the helical and more disordered (bundled) conformation found in MD simulation of 20 and 50 m P3HT assembling on indi-vidual SWNT. The bundled shape is in good agreement with recent experimental data obtained by high-resolution TEM imaging [153] Also, it was found that nano-tube concentration had an influence on average conjugation lengths of the poly-mer. These results display that the one-dimensional structure of nanotubes acts as a rigid template for adsorbed polymer reducing a number of the “conjugation-breaking” torsion angles between thiophene rings and consequently increasing the polymer conjugation length. Furthermore, using a Bardeen’s tunneling the-ory [154], they could estimate how an increase of conjugation length affects the



electron transfer rate from P3HT to nanotube. Based on these results, an important conclusion has been drawn for the composite PV device: higher SWNT concentra-tion should provide a more effective charge transfer from P3HT to nanotubes(e.g. increase SWNT concentration from 2 to ~30–40 wt% should increase the charge transfer rate by the factor of 2–20). Note, that most of the experimental works [137–144] studying SWNT/P3HT photovoltaic effect, deal with SWNT concentra-tion in the range of 2–5 wt%, because of the poor nanotube dispersion at higher weight loading. Nevertheless, permanent progress in nanotube dispersion allows anticipating a successful preparation of homogeneous polymer composites with considerably higher SWNT weight fraction. Another benefit of the longer conju-gation length is the reduction of exciton binding energy in P3HT facilitating the electron–hole separation followed by the electron transfer to SWNT.

In the context of CNT high charge mobility, it is worth mentioning the series of studies where carbon nanotubes were introduced in polymer/C60 composite [155–161]. Because, so far the highest PCE among OPVs has been reached for pol-ymer/C60 composites [130–132, 135], and direct replacement C60 by CNTs only made the PV performance worse [136–143], it was expected that ternary compos-ite (polymer/C60/CNT) could be an advantage, combining efficient exciton disso-ciation at polymer/C60 interface and fast electron transport through the nanotube network. Nevertheless, to date, the best conversion efficiency for ternary compos-ite (PCE = 4.9 % [155]), is still lower than for polymer/C60 photoactive material. The reason for that is probably again associated with insufficient SWNT enrich-ment by semiconducting species and poor nanotube debundling. Thus, the gain due to improved charge transport is compensated by losses from exciton quenching by m-SWNTs, charge trapping and the existence of barriers at nanotube junctions; as was previously discussed in this section.

Notwithstanding, there is a good chance that optimized composition of the ter-nary nanohybrid with careful accounting of all photophysical, chemical and opto-electronic factors related to polymer, C60 and CNTs could outperform existing polymer/C60 organic photovoltaics.

Fig. 1.11 Locally stable conformations found during the simulation runs and also observed experimentally. a Helices form on (15,0) and (10,4) SWNTs during the folding of P3HT 20 m with orthogonal initialization. The chirality may affect the pitch distance to some extent, but is not crucial in determining the overall time evolution or the simulation runs. b A P3HT 50 m is kinetically trapped in a bundled, disordered shape [152], Reprinted with permission from ACS Nano, 2010, 4 (11), pp 6599–6606, Copyright © 2010 American Chemical Society

27

1.2.4 Photocarrier Separation and Multiplication at p–n CNT Junction

In this section we will consider photoconversion at p–n interface comprising only carbon nanotubes which is principally a different from previously discussed inter-faces between CNTs and photoactive material of different nature. Similar to silicon, the creation of p–n junction in carbon nanotubes requires the existence of two intrin-sic regions interfaced one to the other with different levels of doping. The doping of semiconducting SWNTs can be performed electrostatically using a field effect transistor (FET) with dual-gate geometry either by applying special chemical treat-ment to inject holes or electrons by the charge transfer from the dopant. The first nano-solar cell consisting of single SWNT has been reported by Lee [18] utilizing electrostatic doping, where different bias polarities at dual gates electrostatically form regions with electron and hole doping (n and p type) along a single SWNT (Fig. 1.12). The device shown I–V characteristics of an ideal p–n diode because of the suspension of nanotube central part between two terminal (as it was reported pre-viously [162], when SWNT rests completely on SiO2, the diode characteristics were not ideal due to the interaction of nanotube with silica surface). Upon light exposure, the typical photovoltaic behavior has been observed resulting in low limit of PCE of ~0.2 %. Because of the small diameter of SWNT and circularly polarized light, it was difficult to calculate the actual power of the incident light. The uniqueness of this elegant study is a demonstration of the PV effect in single nanotube acting as a light absorber, carrier separator (due to built-in electric field at p–n junction) and transporter to external electrodes, all of it within a one-dimensional nanostructure.

Fig. 1.12 Left the inset shows the split gate device where VG1 and VG2 are biased with opposite polarities (VG1 = −VG2 = +10 V) to form an ideal p–n junction diode along a SWNT. Data are typical dark current–voltage I–V curve at T = 300 K with a fit to ideal diode equation. Right I–V characteristics under increased light intensity showing a progressive shift into the fourth quadrant (PV) where the diode generates power. The inset shows the expected linear increase in the current measured at VDS = 0 (Isc) with illuminated power [18]. Reprinted with permission from Applied Physics Letters, 2005, 87, pp 073101, Copyright © 2005, American Institute of Physics



Chemical doping as an alternative approach to electrostatically doping looks quite attractive at first sight, as it does not require sophisticated device fabrica-tion with two additional electrical terminals, and just relies on rational choice of the dopant species supplying holes and electrons to opposite part of nanotube to form p–n junction. However, nanotube chemical treatment as a rule is not very stable, resulting in gradual device degradation for a long period of time. Several studies [163–165] reported the effect of chemical doping of single SWNT convert-ing semiconducting nanotube in p–n one-dimensional diode. Photovoltaics action in such chemically gated system has been investigated by the simple patterning of two different polymers applied to opposite halves of nanotube conductive channel [165]. Spatial doping has been achieved by UV lithography of PMMA contain-ing TCNQ (p-type dopant) followed by spin casting polyetylemin (n-type dopant). Photovoltaic response (short circuit current) of the device was clearly detected when the scanning laser beam was centered on the depletion region of nanotube (p–n interface). It was not possible to determine PCE because of the saturation limit induced by high laser intensity, and only maximum power of ~0.14 nW was estimated from I–V characteristics upon light exposure.

Quite recently, PV response of single SWNT p–n photodiode was investigated at low temperatures resulting in the discovery of carrier multiplication (CM) phe-nomenon [166]. The effect of carrier multiplication and a similar effect called multiple exciton generation (MEG) can be utilized to exceed the PCE theoretical limit of 33 % (for single-junction solar cell) which originated from fundamental losses such as termalization of carriers or excitons; incomplete absorption of the solar spectrum; radioactive recombination and thermodynamic losses [167–169].

Until Gabor et al. report [166], CM and MEG have been widely investigated in colloidal quantum dots (QDs) whose properties are very different from those of their bulk counterparts, allowing to observe the MEG process at reasonably low intensities and wavelengths matching UV–Vis part of solar spectrum [170–174]. However, there were some controversial reports regarding MEG efficiency deter-mined from transient absorption spectroscopy [170–173]. MEG has been char-acterized in a number of spectroscopic studies of colloidal solutions of QDs, and some studies have suggested that MEG-like processes occur in films of electron-ically-coupled QDs. Previous measurements of MEG in colloidal solutions were challenged because either the measurement was performed at high photon fluen-cies, or the quantum dots were not electronically coupled to each other. If MEG had an intensity dependence, it would be impossible to improve solar cell efficiency at typical intensities of the solar spectrum. Similarly, it was unclear if electronic cou-pling of the QDs would remove the quantization effects that produce the enhanced Coulomb coupling that drives MEG in QDs. Just very recently, it was reported that PCE enhancement of electronically-coupled QDs is due to CM effect [174].

MEG and CM in semiconducting carbon nanotubes should be distinctive from MEG/CM in quantum dots for at least two major reasons: SWNTs are one-dimen-sional systems as compared with zero-dimensional QDs, and carbon nanotubes do not have their bulk analog which makes them truly nanoscale objects (in bulk semiconductors CM, known as impact ionization, is a well studied process [173];

29

however, without any potential for PV application owing to much higher threshold energies as compared with QDs). Two-dimensional quantum confinement of elec-trons and holes around the SWNT circumference leads to a distinctive density of states and well-defined selection rules for optically allowed transitions. Like semi-conductor QDs, SWNTs exhibit a strong Coulomb interaction between electrons and holes, which suggests that SWNTs could also exhibit MEG and CM effect [166, 175–178]. Thus, SWNTs could provide an important alternative system for MEG studies, since the important question of MEG efficiency relative to the “bulk” is irrelevant, as semiconducting SWNTs have no direct bulk analog.

MEG process in (6,5) SWNT suspension has been observed [177, 178] by ana-lyzing the dependence of the temporal profile of the transient absorption signal on light intensity and excitation energy. Ueda et al. [177] reported MEG quantum yield as 100 and 130 % for probe energies of 400 and 275 nm, which were close to the results of Wang et al. [178] with quantum yields of 110 and 130 % for 400 and 335 nm. These results indicate that MEG threshold can be close to 2Eg (Eg is the bandgap for semiconducting nanotube). An observation of MEG and faster Auger recombination [178] than for QD nanostructures is in agreement with the physi-cal picture that electrons and holes in SWNTs exhibit stronger confinement when compared to QDs.

So far, MC generation in SWNTs has been observed only for SWNT photodi-ode in split-gate geometry [166], resulting in formation of p–n junction, similar to Lee’s study [18]. The device demonstrated typical I–V diode characteristics in the dark and under the light for photon energies below 2Eg [166]. When photon ener-gies were above 2Eg several steps in the reverse photocurrent were observed. The origin of these steps has been explained by MC generation of electron–hole pairs by impact ionization from εe2–εh2 bands. It was suggested that photocarriers are accelerated by the internal diode field gaining the energy in εe2–εh2 bands and gen-erating multiple carriers. The importance of this study is the clear demonstration that multiple e–h pairs can be not only generated, but also collected in p–n nano-tube junction, while for QD structures such an observation still remains controver-sial [173]. Later in theoretical model [176], a detailed mechanism based on carrier dynamics and impact excitation (Fig. 1.13) has been proposed for MC generation as observed in study [166].

Fig. 1.13 Schematic representation of the four step carrier multiplication by impact excita-tion mechanism in a CNT photodiode. (1) Photoexcitation into a quasi-bound state. (2) Exciton breaks into an electron–hole pair in the εe3–εh1 bands. (3) Energy gain of the carriers by inter-play of diode field and phonon emission. (4) Carrier multiplication by impact excitation [176]. Reprinted with permission from Nano Letters, 2010, 10 (9), pp 3277–3282, Copyright © 2010 American Chemical Society



Several conclusions were drawn from this model: (i) excitation in εe2–εh2 bands cannot result in CM by impact excitation, but can be realized only for electrons and holes excited in εe3 and εh3 band and higher; (ii) MC threshold energy is slightly higher of 2Eg owing to exciton breaking in e–h pairs in the εe3–εh1 bands; (iii) the mechanism predicts an onset temperature for MC observation and depends on SWNT length as L−2/3, which means that such an effect can be observed at room temperature for SWNT shorter than 500 nm. This is in agreement with Gabor et al. results [166], observing an onset temperature of 100 K for ~2 μm nanotube.

1.3 CNT/Quantum Dots Photoinduced Charge Transfer and Related Photovoltaic Effect

Colloidal quantum dots (QDs) (or colloidal nanocrystals) present a large and impor-tant class of nanomaterial being nanometer sized fragments of the corresponding bulk crystals which are typically synthesized and processed as solution species [179]. Their optical (e.g. absorbance and fluorescence bands) and electrical prop-erties are to be size dependent when QD intrinsic physical size is smaller than the critical characteristic length called the exciton Bohr radius. The large surface-to-vol-ume atom ratio of QDs alters the chemical potential of the structural units in com-parison to that for the corresponding bulk crystals. This variety of size dependent properties coupled with solution based processability make colloidal QDs attractive for a number of applications, including nanocomposite photovoltaics [180].

At the beginning of the 2000s it was reported the use of quantum dots/quantum rods in the blends with polymers as photosensitive materials for solution proces-sible thin film solar cells [181, 182]. These studies demonstrated that excitons gen-erated in the polymer are dissociated at the interface with QD followed by hole transport through polymer medium and presumable electron transport through QD network to external electrodes. Also, the possible mechanism of exciton generation in QDs and their dissociation has been considered. Because the exciton binding energy in colloidal QDs is small (<kT), generated excitons readily dissociate on free carriers within QD structure. Thus, as distinct from polymeric and small mol-ecules OPV, QD based solar cells do not require interface for exciton dissociation [183]. The favorable electron affinity (LUMO level) and work function (HOMO level) of many QDs of appropriate size with respect to corresponding energies of P3HT and P3OT polymers facilitates the charge separation at the interface: electron transfer from polymer to QD and hole transfer from QD to polymers (Fig. 1.14).

However, light absorption efficiency of QDs is smaller than absorption of many the organic dyes and polymers. Another drawback is the quite low carrier mobility due to organic insulating shell required for QD stabilization, which is comparable in magnitude with organic amorphous and organic layers [184]. As a result, QD network in the bulk heterojunction device (or QD film) is not capable of provid-ing sufficient electron mobility for effective charge transport and collection at the

31

electrode. Therefore, substantial efforts were undertaken to compensate QD low charge mobility by introducing carbon nanotubes in the QD/polymer blends, due to excellent carrier transport properties of CNTs and additional source of charge separation at CNT/QD interface [185–191, 194, 195]. Moreover, in some studies, the polymer component was completely excluded [191–193], as QD electron affin-ity suitable for CNT/QD charge transfer can be tailored by a selection of semicon-ducting material, and their band gap is controlled by the QD size. In addition, the binary CNT/QD hybrid is less affected by interfacial defects and charge recombi-nation than ternary hybrid comprising CNT/QD/polymer composite.

To date, sufficient research work have been conducted to investigate PV charac-teristics of CNT/QD complexes themselves or (more often) their incorporation in a film of photoactive, hole transporting polymers like P3HT, P3OT, PPV derivatives, PVK, etc [185–191, 195].

In these reports, the major types of colloidal QDs were composed from ZnO, ZnSe, CdS, CdSe, CdTe, PbSe, Pbs compounds while the CNT/QD complex for-mation was varied from simple physical blending to various strategies of covalent binding and electrostatic interaction.

Fig. 1.14 a Potential energy level diagram adjusted in relation to the vacuum level for the P3OT polymer, CdSe QDs, semiconducting SWNTs (S-SWNT), and metallic SWNTs (M-SWNT). b Schematic which illustrates a QD/SWNT/Polymer solar cell equilibrated at the Fermi energy. The corresponding electronic transitions from optical absorption, exciton dissociation, and carrier transport are shown for each of the components [185]. Adapted with permission Solar Energy Materials and Solar Cells, 2005, 87, (1–4), pp 733–746, Copyright © 2005, Elsevier

1.3 CNT/Quantum Dots Photoinduced Charge Transfer


Raffaelle et al. first reported an investigation of PV properties of ternary hybrid SWNT/QD/polymer [185]. Colloidal CdSe quantum dots with amino ethanethiol (AET) ligand were covalently attached to carboxylic acid-functionalized SWNTs. Solar cell has been constructed as a device with a photoactive composite, a bulk heterojunction between P3OT and SWNT/QD complex, cupped by ITO and Al electrodes. Despite a high expectation based on cascade charge transfer between appropriate interfaces (Fig. 1.14) and fast carrier transport trough SWNT network, the resulting photoconversion efficiency was very small (less than 0.0001 %) under illumination of AM0 solar spectrum. An analysis of I–V curves suggested a high concentration of the interfacial defects resulting in a leakage current in the reverse bias direction. Cho et al. [186] tried to utilize advantages of PbSe QDs as potential IR absorbers due to narrow band gap (0.28 eV [195]) and large Bohr radius (~46 nm [196]). Photoactive thin film comprised PVK polymer blended with covalently attached PbSe QD to SWNT. Although the photocurrent for PbSe QD/SWNT/PVK composite was substantially higher than the control sample of PbSe QD/PVK, photovoltaic action was not observed. The same team recently described a similar photoactive composite incorporating SWNT network [187] without covalent binding of PbSe QDs to nanotubes with the hopes of avoid-ing SWNT defects induced by chemical bonds and impeding an efficient carrier transport. Nevertheless, no significant improvement was found; moreover, the addition of SWNTs resulted in zero open circuit voltage (similar to report [186]) presumably due to device shortening.

Just recently, PCE = 3.03 % at AM1.5 solar spectrum has been attained for similar ternary hybrid including CNTs, PbS colloidal QDs and polymer [189]. Here PbS QD/MWNT nanostructure was prepared by simple mixing of high qual-ity PbS QDs with functionalized MWNTs followed by hybrid incorporation in P3HT hole-conducting polymer. The reference sample of P3HT/PCBM (common photoactive material for polymeric OPV [35, 129]) showed lower PCE (~2.5 %) and the lack of photocurrent in the NIR range, where PbS QDs absorb light.

Li et al. studied PV effect of MWNTs covalently bound with ZnO QDs [192] and CdSe:ZnSe core–shell quantum dots [193]. Here, the primary attention was focused on binary MWNT/QD nanohybrid forming bulk heterojunction, and only poly(N-vinylcarbazole) (PVK) polymer was introduced for comparison [192, 193]. Although authors demonstrated efficient charge transfer from CdSe:ZnSe QDs to MWNTs observing QD luminescence quenching [193], the resulting PCE was less than 1 % at UV light of lower intensity, which probably should be fur-ther reduced at AM1.5 solar spectrum. In the case of MWNT/ZnO nanohybrid, PV action was also inefficient (PCE < 1 % under UV light), reaching better perfor-mance after the addition of PVK polymer [192].

More variety of CNT structures for photovoltaic applications is related to Pt [197], Si [190] and CuS [198] nanoparticles decorated carbon nanotubes in com-bination with photoactive, hole transporting polymers. Another type of nano-structured PV composite with CNT incorporation is related to Si nanowire array—polymers composite [199]. A promising trend in all these studies was an improvement of PCE of the CNT/nanoparticle/polymer photoactive material as

33

compared with the pristine CNT/polymer or nanoparticle/polymer structures although PCE magnitude was less than 1 %. For Si QD/MWNT/P3HT compos-ite, PCE = 0.04 % [190] and EQE spectra were similar with MWNT/P3HT con-trol sample indicating that there was no contribution from light absorption of Si QD. In contrast, Wang et al. reported PCE ~3 % [189], where the spectral pho-toresponse of Pb QDs in the NIR range was clearly observed. Somani et al. [197] found that the presence of metal (Pt) nanoparticles attached to MWNTs improve the PV device performance owing to plasmonic effect (PCE ~0.7 % at AM1.5). They interpreted the obtained results by an existence of a strong built-in field at the interface between Pt QDs with MWNTs and P3OT polymer leading to effi-cient exciton dissociation and charge separation between P3OT and MWNT in the vicinity of Pt QDs. In addition, owing to the increase in optical electrical field inside the photoactive layer, the inclusion of Pt nanoparticles should enhance an optical absorption and consequently increase photoconversion efficiency. Overall, insufficient light absorption of QDs, charge recombination and trapping at inter-face with CNTs and polymer as a rule results in poor photovoltaic performance. In this context, one of the sources for PCE enhancement could be the involvement of NIR absorption of enriched semiconducting SWNTs, as it was demonstrated by Bindl et al. [38]. However, so far, most of the QD/CNT/polymer composite for photovoltaics utilize MWNTs or blends of semiconducting and metallic SWNTs where carbon nanotubes NIR absorbance cannot be realized in its full capacity.

In parallel with the investigation of CNT/QD/polymer solid nanocomposite, QD/CNT based photoelectrochemical cells were the subject of active research for the past decade [200–208]. Despite a low conversion efficiency, many inter-esting strategies of QDs assembly with CNTs were proposed followed by a detailed examination of photoinduced charge transfer using time-resolved tran-sient absorption and luminescent spectroscopy. Guldi et al. [200] integrated CdTe quantum dots with SWNTs functionalized with cationic pyrene molecules. Trimethylammonium groups in pyrene interact electrostatically with nega-tively charged groups of thioglycolic acid (TGA)passivated CdTe quantum dots (Fig. 1.15). As a results CdTe QD/SWNT nanohybrid has been assembled owing

Fig. 1.15 Partial structure of the SWNT/Pyrene+/CdTe used in the study [200]. Reprinted with permission from Journal of American Chemical Society, 2006, 128 (7), pp 2315–2323, Copyright © 2006 American Chemical Society



to electrostatic (between end groups of pyrene and QD) and π–π interaction (between SWNT and pyrene). Bilayers made from SWNT/pyrene and TGA sta-bilized CdTe QDs were deposited on ITO modified electrodes by a layer-by-layer assembly technique followed by the immersion of ITO/SWNT/CdTe QD anode in the electrolyte solution with the Pt cathode. Short circuit current between ITO and Pt electrodes upon the visible light indicated to the photoactive action in the phote-lectrochemical cell [200]. The proposed mechanism of the observed photocurrent was associated with electron injection from CdTe QD to SWNTs (or MWNTs) and then to the ITO layer. Thus, CNTs act as a charge transporting channel pro-viding the photoelectron transport from quantum dot to ITO electrode. A novel approach to link CdTe QD to SWNTs was recently implemented by Guldi’s group [201]. Distinct from the earlier method [200], here, the pyrene was covalently attached to CdTe QD followed by the van der Waals interaction of the pyrene core with SWNT. The ultrafast charge transfer, recombination and relaxation processes in CdTe QD/SWNT complex were examined with fentosecond time resolved tran-sient absorption technique. The Fig. 1.16 illustrates the major channels of deacti-vation and charge transfer of the quantum dot excited state based on the analysis of the transient absorption data from the study [201].

Similar to photosynthesizing dyes [91], electrophoresis technique was employed to assemble QDs with CNTs as building blocks for light harvesting in photoelec-trochemical cells [202, 203]. CdS QDs have been directly deposited onto SWNTs (in suspension as well as in ITO electrode) under applied DC current. [202]. The resulting CdS QD/SWNT complexes were probed by time resolve transient absorp-tion and luminescent spectroscopy revealing the moderate electron transfer from excited CdS QD to SWNTs with a rate of 4 × 108 s−1 The low external quan-tum efficiency (EQE < 1 %) can be associated with fast carrier recombination and interfacial trapping. Significantly better performance of the light harvesting

Fig. 1.16 Energy diagram that illustrates the deactivation pathways and related transitions with corresponding lifetimes of CdTe QD-pyrene/SWNT on ITO upon photoexcitation at 420 nm. V.B. valence band; C.B. conduction band; I excitonic state of electron; II/III surface trapping states; IV charge-separated state; V electron injection into ITO; CS charge separation; CR charge recombi-nation [201]. Reprinted with permission from Angewandte Chemie International Edition, 2010, 49, 6425–6429, Copyright © 2010 John Wiley and Sons, Inc

35

has been demonstrated for a CdSe QD/stacked-cup CNTs (SCCNT) complex prepared by similar electrophoretic deposition on SnO2/ITO electrode [203]. SCCNTs present stacking carbon nanofibers cupped by the conical graphene layers. These structures exhibit a large interfacial area with open ends favorable for chemi-cal functionalization. The electron transfer rates for CdSe QD/SCCNTs complexes with different size of QDs were determined from time resolved transient absorption experiment: 9.51 × 109 and 7.04 × 1010 s−1 for quantum dots with diameters of 4.5 and 3 nm, respectively. The results were consistent with data from report [209], where a similar correlation between size of QDs and electron transfer rates was observed. An average EQE value of ~3 % in the visible range exceeds more than two orders of magnitude the EQE for CdS QD/SWNT complex [202] and can be readily explained by the higher electron transfer rate (~1010 vs. 4 × 108 s−1).

1.4 CNT/Semiconductor Based Photovoltaics and Photodetectors

1.4.1 CNT/Si Solar Cells

Silicon remains the dominant material in photovoltaic industry due to excellent optoelectronic properties, stability and its sufficient supply in the environment. Today, silicon based solar cells occupy more than 80 % of the commercial pho-tovoltaic market. However, the further broad dissemination of Si PVs is partially limited by the high price of high quality crystalline Si wafers as well as the com-plicated fabrication process. The use of nanotechnology and employment of novel organic materials in combination with inexpensive polycrystalline, amorphous Si and nanostructured Si such as porous Si (PSi) or Si nanowires (SiNW) opens an opportunity to develop a new generation of hybrid organic/Si solar cells with a conversion efficiency higher than 10 % and a fabrication cost lower than for con-ventional Si PV technologies. For the past several years, there was a surge of research dealing with the investigation of photosensitive hybrid materials com-prising Si nanostructures (SiNW array [209–217], free standing SiNW [218–222], porous Si [223–226]) interfaced with polymers, oligomers and small molecules. Also, there have been substantial research efforts devoted to the investigation of organic/crystalline Si heterojunctions [227]. Very recently, a record PCE of 11.1 % has been reported for PEDOT:PSS/Si nanocone array hybrid solar cells [209].

Nevertheless, in terms of PV performance, conjugated polymers have serious drawback as compared to their silicon counterpart due to gradual degradation, photobleaching and low charge mobility making further progress challenging, primarily because of poor long term stability.

In this context, carbon nanotubes present an almost ideal organic compound to be interfaced with Si to create an efficient light harvesting nanohybrid structure. We already mentioned in the introduction that CNTs posses excellent environmental



stability, high charge mobility, and size dependent band gap [4, 5]. Specifically for CNT/Si hybrid solar cells, additional advantages over other organic materials may include: efficient absorbance in NIR spectral range for semiconducting SWNT; simple doping by acid treatment allowing significantly reduced film resistance [8–10, 31]; thin CNT film can simultaneously work not only as a photoactive mate-rial, but also as a transparent conductive electrode, replacing conventional metal grids or metal oxide coatings [228–232] (for other organic materials a deposition of the top metal/metal oxide electrode is required); Fermi level of s-SWNTs can be shifted through either the electrical or chemical doping and functionalization resulting in a favorable position to increase built-in electric field at the interface with silicon.

Recently a photoconversion efficiency record of 13.8 % has been reported for SWNT/Si solar cells [31] exceeding the best PCEs for any organic (8–10 % [130–135]) and DSSC (10–12 % [36, 37]) devices. Such remarkable results were achieved for a relatively short term (about five years since first report in 2007 with PCE of ~1 % for similar device [233]) suggesting further progress in the near future. According to study [31], such high efficiency can be associated with sev-eral factors. First, the use of high quality p-type semiconducting SWNT forming p–n heterojunction with n-type Si. It means that s-SWNTs contribute to not only charge separation and transport, but also to NIR light absorbance (see next discus-sion in this section); Second, SWNT doping by diluted HNO3 reduces SWNT film resistance (and consequently device serial resistance) resulting in an increase of fill factor and short circuit current; Third, SWNT infiltration with the acid results in formation of Si-acid-CNT micro-units working as photoelectrochemical solar cells with acid as an aqueous electrolyte. Thus, two photosensitive mechanisms act in parallel: electrochemical and charge separation at p–n heterojunction. Upon exposure to air, PCE was reduced to 8.9 % when the nitric acid has completely vaporized. An improved design with relatively stable efficiency of ~10 % has been reported for the similar heterojunction; however, a thin oxide layer was introduced between SWNT and Si to suppress dark saturation current and reduce charge recombination at interface [234]. Also, SWNT film was encapsulated by polydi-methylsiloxane (PMDS) to provide anti-reflection properties.

The doping of SWNT film by thionyl chloride also enhanced conversion effi-ciency more than 50 % (PCE ~ 4 % was achieved after doping) [32]. Using Hall-effect measurements, the change of the 2D carrier density and effective mobilities have been determined for undoped and doped SWNT film: 3.1 × 1015 versus 4.6 × 1017 cm−2, and 2.1 versus 17.2 cm2/Vs, respectively. If increase of the car-rier density after doping is expectable (an increase of hole concentration owing to oxidation), the mobility enhancement is not a trivial outcome and can be related to resistivity reduction at junctions between nanotube bundles and the lowering of barriers between s-SWNTs and m-SWNTs [32]. An additional benefit of the doping for PV performance is the adjusting of Fermi level of s-SWNT (moving down toward the top of the valence band) resulting in an increase of the open cir-cuit voltage. Recently, PV properties of boron doped SWNT/n-Si solar cells has been investigated, exhibiting PCE less than 1 %. [235]. Another report by the same authors [236] demonstrated an interesting trend in nanotube doping strategy.

37

Here, naturally p-type semiconducting SWNTs were functionalized by polymer (polyethylene imine (PEI)) with an electron donating groups that converted SWNTs in n-type semiconductors (method similar to study [145]). Then a rectify-ing behavior has been observed for heterojunction formed between PEI-SWNTs and p-type Si, while no rectification was detected for pristine SWNT/p-Si. Although conversion efficiency was less than 1 %, such an approach indicates an opportunity to invert the doping type of SWNT/Si junction counterparts.

In spite of impressive progress in the study of optoelectronic properties of inter-face between carbon nanotubes and silicon, a lot of fundamental aspects of photo-conversion at CNT/Si heterojunction require further investigation. For example, what is the role of the interface in dissociation of excitons generated in carbon nanotubes and photocarrier recombination process?; What junction type (Schottky diode or p–n diode) is formed between Si and SWNT film, where 1/3 of nanotubes are metallic and other 2/3 are semiconducting?; How should the Anderson model be adjusted to quasi-1D nanotubes and their inhomogeneous density in the contact area with silicon (because of the high film porosity); and What is the mechanism of the hole transport trough the nanotube bundles and their junctions in the depletion region?

Ural et al. [237] fabricated CNT/Si/Me planar structure to investigate the nature of contact between carbon nanotubes and silicon. It was found that thin CNT film forms Shottky contact with Si (both n- and p-type) and thermionic emission is a dominant transport mechanism above 240 K, while for lower temperatures the major mechanism is tunneling. These results were consistent with metallic behav-ior of CNT film obtained from temperature dependency of resistivity, that could be associated with prevailing metallic nanotubes over semiconducting in the CNT network. While, other reports [31–33, 238] indicate p–n heterojunction between semiconducting SWNTs and n-type Si (see further discussion). An effect of elec-trical gating on photoconversion process of SWNT/n-Si device has been exploited by Rinzler’s group in recent reports [34, 239]. The gating of SWNT film was per-formed in lateral geometry through electrolyte and SWNT film (gate electrode) insulated from the other SWNT film forming the Schottky contact with n-type Si. Upon negative bias of −0.75 V, PCE = 10.9 % has been attained that demon-strated significant improvement compared to zero bias (PCE = 8.4 %). The authors believe that such gate voltage-induced behavior can be explained by several mechanisms including (i) the enhancement of built-in potential owing to the shift of Fermi level, (ii) reduction of SWNT film resistivity; (iii) existence of the inter-face dipoles; and (iv) an appearance of additional electric field across the depletion layer in the n-Si [34]. A similar solar cell, however, with a grid of SWNTs cover-ing only a fraction of the n-Si, demonstrated PCE = 12 % at negative bias [239]. A higher efficiency than in Ref. [34] is related to electrolyte-induced depletion layer in the n-Si across large gaps between SWNT grid lines, where photogenerated holes are trapped within this layer and diffuse toward SWNT strips.

So far, in all aforementioned studies, CNT/n-Si interface was considered as a Scottky photodiode [34, 237, 239] or as a p–n heterojunction [31–33, 236, 238]; however, without direct evidence of SWNTs (or DWNTs) contribution in light absorption, especially in the NIR range. Recent reports [33, 238] clearly indicated



the importance of semiconducting SWNTs as NIR light absorbing material with energies below Si bang gap (1.1 eV). Ong et al. [33] observed matching of the S11 band (corresponding to the first interband transition for s-SWNTs with 7,6 and 8,6 chirality) with the photocurrent band located at ~1150 nm (Fig. 1.17, left). Thus, semiconducting SWNT network contributes to the photo conversion process not only as a charge separator/transporter/collector but also as a light absorber. This is an important fact, distinguishing between a p–n heterojunction solar cell with two active light absorbing components and a Schottky cell, where the metal com-ponent is not capable of absorbing photons. It was suggested that photo-generated excitons in SWNT dissociate to holes and electrons at the heterojunction followed by hole transport and collection through the SWNT network (Fig. 1.17, right, a). In parallel, photoelectrons generated at the Si side diffuse from the depletion region to the external electrode. Also, photoholes from Si and photoelectrons from SWNT can be involved in the charge separation process; however, with lower effi-ciency [33]. Note that the energy band diagram (Fig. 1.17, right, a) was based on the standard Anderson model [240] without corrections on quasi-1D geometry of SWNTs and film porosity.

400 600 800 1000 1200 1400 1600 1800

0.01

0.1

1

Nor

mal

ized

Pho

tocu

rren

t (a.

u.)

Wavelength (nm)

Sediment Supernatant

Si + SWNT

SWNT

400 600 800 1000 1200 1400 1600 18000.00

0.04

0.08

0.12

0.16

8,6

7,6

8,6

Abs

orba

nce

(opt

ical

den

sity

)

Sediment Supernatant

7,6M11 S22 S11

(a)

(b)

Fig. 1.17 Left a UV–Vis–NIR spectrum of sediment and supernatant fraction from centrifu-gation process of SWNT films on glass. M11, S22, and S11 represents the band-gap transitions in metallic and semiconducting SWNTs; b normalized photocurrent spectra of the SWNT/n-Si solar cell devices (supernatant and sediment) showing a current band matching the S11 absorb-ance band. Right a schematic energy band diagram of SWNT/n-Si heterojunction based on the Anderson model. Electron affinity, χ, and conduction band offset, ΔEc, for SWNT and n-Si are shown in the energy band diagram; b SEM images of cross-sectional view of SWNT/n-Si inter-face [33]. Reprinted with permission from Nanotechnology, 2010, 21, 105203, Copyright © 2010 IOP Publishing

39

Hatakeyama et al. [238] also demonstrated a critical role of SWNTs as NIR photon absorbers for the enhancement of PV performance in the infrared spec-tral range. They compared conversion efficiency of SWNT/Si cell with the con-trol sample (Ag/Si Schottky cell) observing a significantly higher PCE of SWNT based device for wavelengths longer than 850 nm. Additionally, they found that the encapsulation of C60 inside nanotubes improves the device PV performance through adjusting the Fermi level of SWNTs. A similar contribution of carbon nanotubes in external conversion efficiency, however, for UV spectral range owing to π–π plasmon band has been observed in study [241].

It is noteworthy, that these studies [33, 238] are in good agreement with a report [38], where a similar trend has been investigated for SWNT/C60 solar cells (see Sect. 1.2.2) with the major focus on the importance of SWNT infrared absorbance for hybrid photovoltaics.

When this chapter was completed, an interesting review of CNT/Si photovoltaics was published [242], where the major reports discussed here were also critically reviewed, however with a somewhat different interpretation in several parts.

1.4.2 CNT/Si Mid-IR Photodetectors

One of the most interesting physical properties of semiconducting carbon nano-tubes is the dependence of the band gap on nanotube size [4, 5]. As it was estab-lished for SWNTs [4, 5] and MWNTs [243], the energy of the optical band gap is inversely proportional to their diameters. This means that carbon nanotubes, as an IR photosensitive material, could cover a broad spectral range from Near-IR (0.7–5 μm) to Mid-IR (5–25 μm) and further extend photosensitivity into the Far-IR (25–200 μm) by simple variation of CNT size. This is in stark contrast to conventional bulk semiconductors (e.g. Si, Ge, GaAs, CdS, etc.), in which a deter-mined band gap limits the ability to tune a spectral response. As we mentioned in the introduction the photoconductivity of carbon nanotubes as a rule is observed for individual CNTs [3, 16–19] or at very short times for CNT network [1]. To absorb sufficient light energy carbon nanotubes should be prepared as a film, how-ever for CNT films bolometric response [30] usually dominates over direct photo-conductivity. Also, major reports related to CNT based photodetectors described the lateral device geometry making such structures inefficient as compared with planar/bulk heterojunction where CNTs are interfaced with other photoactive materials On the other hand, an investigation of p–n heterojunction devices were mostly focused on either visible or NIR light energy owing a great interest in CNT contribution to the improvement of hybrid photovoltaics. In particular, almost all studies of CNT/Si p–n heterojunction structures discussed above [31–33, 233–236, 238] were devoted to solar cell performance with the spectral response below 1.5–2 μm, and only a few groups investigated CNT/Si photoconversion in mid-IR range [244–246]. Meanwhile, CNT with a tunable optical band gap from sub-microns to tens of microns can be a novel light sensitive nanostructure replacing traditional low band gap semiconductors utilized in IR photodetectors.



Tzolov et al. investigated IR sensing in the range of 1–25 μm of the MWNT array (diameter of ~50 nm, Fig. 1.18, left)) grown with the assistance of the ordered alumina template on n-type Si [244].

In photocurrent spectra it was observed a broad band (centered at 2.8 μm, Fig. 1.18, right), which was interpreted as the interband transitions for semiconducting MWNTs being in agreement with nanotube diameter. Authors tried to rule out a pure bolometric response speculating on different behavior of temperature dependencies of photocurrent intensity and time constant; however, the absolute value of the response time was too slow (~1 s [245]) to completely exclude the heating effect. In addition, the presence of alumina matrix could result in several undesirable effects in photodetector performance, such as a barrier layer between nanotubes and Si or surface traps in alumina.

Fig. 1.18 Left SEM image of one of the many nanotube array samples fabricated and tested; Right normalized photoresponse spectrum at room temperature. The inset is a linear plot of the same spectrum [244]. Reprinted with permission from Journal of Physical Chemistry C, 2007, 111 (15), pp 5800–5804, Copyright © 2007 American Chemical Society

-0.010 -0.005 0.000 0.005 0.010

10-6

10-5

10-4

10-3

Voltage (V)

J (m

A/c

m2 )

IR OFFIR ON

-0.6 -0.3 0.0 0.3 0.6-0.1

0.0

0.1

0.2

0.3

0.4

(a) (b)

Fig. 1.19 a Current–voltage (I–V) plot of MWNT/n-Si devices showing the photovoltaic response under low intensity IR light (15 mW/cm2). Inset dark I–V characteristics on a linear scale: x-axis is V and y-axis is mA/cm2; b energy band diagram of MWNT/n-Si heterojunction. Electron affinity, χ, and conduction band offset, ΔEc, for MWNT and n-Si are shown in the energy band diagram [246]. Reprinted with permission from Applied Physics Letters, 2010, 96 (3), 033106, Copyright © 2010 American Institute of Physics

41

A more recent study [246] reported mid-IR response of MWNT/n-Si diode prepared by a simple spray deposition technique and working in nonbiased mode (Fig. 1.19a). As distinct from Refs. [244, 245], such an approach provides direct contact of MWNTs with n-type Si, allowing one to study control samples (SWNT/n-Si) where no IR response in the same spectral range was detected as com-pared with MWNT/n-Si structure.

The mechanism of photocurrent (broad band with maximum at 8.3 μm) was associated with mid-IR absorption of semiconducting MWNTs (diameter of 15–30 nm), exciton generation and their dissociation at the nanotube/Si inter-face, followed by charge transport and collection at the external electrodes. It was suggested that the separated photoholes at the MWNT/n-Si heterojunction dif-fuse from the depleted region to the external electrode through the nanotube net-work, while photoelectrons accumulate at the interfacial area (see band diagram, Fig. 1.19b) polarizing the interface and partially tunneling through the barrier to the Si component. The response time of ~16 ms was much shorter than that reported in study [245] and heating effect could be neglected owing to a vanishing small temperature change upon low intensity IR illumination (~10−8 K).

Mid-IR sensing by CNT/Si photodiodes is currently in the inception phase, nevertheless these first reports [244–246] already indicate that the tunable spec-tral response, high carrier mobility, simple wet deposition, and scalability make such hybrid heterostructures highly attractive for room temperature infra red detection.

1.4.3 Carbon Nanotubes Interfaced with Other Semiconductors, Nanostructured and Amorphous Si and Perspective of CNT/Semiconductor Hybrid Photovoltaics

A recent success in achieving PCE exceeding 10 % for CNT/Si hybrid solar cells [31, 34, 234, 239] stimulated an intensive search of different semiconducting sub-strates aiming to further increase conversion efficiency (e.g. GaAs [247–249]), replace crystalline Si by less expensive amorphous Si without dramatic PCE reduction [250–252], and utilize nanostructured Si such as Si nanowire (SiNW) array for the improvement of charge separation and light trapping [253, 254].

Similar to n-type Si, it was found that n-type GaAs forms p–n heterojunc-tion with SWNT [247, 248] and photovoltaic action has been demonstrated for individual SWNT interfaced with GaAs surface [247]. Note, that GaAs selection is motivated by better electrical properties as compared with Si, such as higher electron mobility and lower noise making GaAs attractive material for high speed transistors and multi-junction solar cells. Liang and Roth [247] demonstrated a strong rectifying behavior SWNT/n-GaAs junction, while ohmic-like contact formed in the SWNT/p-GaAs junction. Temperature dependence of I–V curves revealed that the dominant transporting mechanism is the tunneling; however,



with different height of the barrier at low and high biases. Upon illumination by the green laser, the PV response was detected with PCE = 3.8 %. Such photoac-tive reaction of a single nanotube contacting with GaAs gave good promise for further fabrication and investigation of the scalable hybrid solar cells comprising SWNT film contacting with GaAs.

Also, CdSe semiconducting nanostructures were exploited to build a pho-toactive interface with carbon nanotubes [255, 256]. In particular, a heterojunc-tion solar cell was fabricated by overlapping a CdSe nanobelt with a SWNT, or DWNT or multiple CNT, where CdSe nanobelt works as a light absorption mate-rial (bad gap of 1.74 eV) and CNT is the hole acceptor and transporter. It was demonstrated that several CNTs can be integrated in parallel with a single CdSe nanobelt to construct an array of cross-junction cells simultaneously [256]. Such unique double-nanostructure showed PCE = 1.87 % indicating the potential for further improvement through dense parallel connection and large scale fabrication.

Another recent trend in the development of CNT/semiconductor light harvest-ing devices is the employment of amorphous Si [250–252] and nanostructured Si [253, 254]. Hydrogenated amorphous silicon (a-Si) has advantageous over crys-talline Si because of the lower cost and opportunity to fabricate thin film flexible structures comprising p–i–n junction; however, the major drawback of a-Si is the low charge mobility. Nevertheless, today a-Si solars attract growing interest as a light weigh, thin film, flexible and cost effective alternative to crystalline silicon photovoltaics. The research of CNT/a-Si hybrids is now in the beginning stage; however, recent studies already indicated a potential for future PV applications. Schriver et al. [250] examined the photoactive properties of MWNT networks in junction with a-Si undoped thin film. The heterojunction demonstrated rectify-ing behavior and PV action with small short circuit current (~0.25 mA/cm2) that can be associated with the use of MWNTs (instead of s-SWNTs) and undoped a-Si (instead of n-doped Si). Interestingly, a similar device (with graphene layer instead of MWNTs) results in worse PV performance. A sophisticated design of coaxial solar cells has been reported by Zhou et al. [252] where heterojunction was constructed from a metallic inner core (MWNTs) contacting with the outer shell (a-Si). This strategy was based on coating vertically aligned MWNT nanow-ire (NW) array grown on a flat substrate with amorphous silicon shells followed by ITO deposition as a top contact (Fig. 1.20).

The short circuit current of coaxial MWNT/a-Si array showed a 25 % increase in short circuit current without a noticeable change of the open circuit as com-pared with control sample - a similar device with planar MWNT/a-Si hetero-junction. These results suggested that coaxial solar cell works as a manifold of individual freestanding photovoltaic nanowires (Schottky photodiodes) connected in parallel. An enhancement in PV performance has been explained not only by an increase of the interfacial area through the use MWNT NW array but also by the light trapping which is typical for many photovoltaics with nano/microstructured surfaces. Because MWNTs exhibit a low Vis–NIR extinction coefficient, the primary absorber was only a-Si, and thus potential of carbon nanotubes as Vis/NIR absorbers (such as s-SWNTs) was not utilized in this design [252].

43

The above circumstance gives a hint to employ semiconducting SWNTs interfaced with nanostructured silicon to further enhance light harvesting for such hybrid structures.

When s-SWNTs serve as effective light absorbers, an existence of nanostruc-tured heterojunction with Si becomes critically important. Generally, the concept of nanostructured heterojunction is very beneficial for organic/inorganic interfaces with photoactive organic material, first of all, because of the relatively short exciton diffusion lengths in the organic component (e.g. 100–300 nm for SWNT [257, 258] and ~10 nm for many conjugated polymers [259]).

Figure 1.21 presents the schematic demonstrating potential advantages of nano-structured organic/inorganic interface over planar heterojunction. In fact, it can be considered as a variant of bulk heterojunction often employed in the design of organic photovoltaics. In terms of the efficiency of the organic layer, in the case of a planar heterojunction, the figure of merit can be formulated as F ~ Lex/Lab, where Lex is the exciton diffusion length and Lab = 1/k is the light penetration length (k is the absorption coefficient). Thus, a thin organic layer with a thickness of d ~ Lex and d < Lab, provides efficient charge separations (all excitons reach the interface) but low absorbance (Fig. 1.20a). Contrarily, a relatively thick organic layer (Lab ~d > Lex) absorbs more light energy, but only a fraction of the excitons are capable of reaching the interface and become dissociated (Fig. 1.20b). In the case of nanostructured heterojunction, the characteristic length (average distance between pillars or pore diameter) can be comparable with the exciton diffusion length, while the nanostructured layer can be thick enough to absorb substantial light energy (Fig. 1.20c). Taking k = 105–104 cm−1 for SWNTs [62]and average exciton diffusion length ~200 nm, an optimal design of nanostructured interface

Fig. 1.20 Schematics of device fabrication and SEM images of coaxial MWNT/aSi solar cell arrays. a Device-fabrication processes: I growth of patterned MWNTs on tungsten-coated sili-con wafer; II deposition of amorphous silicon onto MWNTs arrays; III deposition of transparent conductive top electrode; IV schematic of a line-pattern array of coaxial MWNTs/aSi nanowires; b tilted SEM images of MWNTs arrays in I dot-pattern and II line-pattern configurations as grown on tungsten; III and IV MWNTs arrays shown in I and II, respectively, after aSi coating. Scale bar is 2 μm [252]. Reprinted with permission from Advanced Materials, 2009, 21, 3919–3923, Copyright © 2009 John Wiley & Sons, Inc



should comprise of axial nanoobjects with a diameter of 200 nm and length of 100–1000 nm or longer. The structure of such nanoobjects may have differ-ent morphology including SWNT NW array coated by amorphous Si (similar to MWNT NW array/a-Si [252]), or SiNW array infiltrated with SWNTs [255, 256], either Si nanopores infiltrated with SWNTs (similar nanoporous Si/phthalocyanine solar cells [223]).

Recently, photoelectrochemical solar cells constructed from SiNW array infil-trated with DWNTs [253] and CNTs (blend of SWNTs and DWNTs) [254] were investigated. Here, SiNWs were prepared from crystalline Si by Ag-assisted etching. Importantly, such a fabrication method retains sufficiently high car-rier mobility inherited from crystalline Si as distinct from a-Si pillars [252]. In the study [253] DWNT film was transferred on top of SiNW array followed by soaking in redox electrolyte. Solar cell exhibited conversion efficiency of 1.29 %. Later Shu et al. [254] investigated the effect of the distance between CNT film and top of the SiNW array allowing to separate electrochemical mechanism from p–n heterojunction mechanism (when CNT film was in the contact with SiNW array). A maximal PCE ~6 % (with optimal balance between resistiv-ity and transparency) has been attained for CNT film deposited right on the top of SiNW array and soaked with electrolytes, while electrolyte removal for the same configuration reduced efficiency down to 0.17 % [254]. This low conver-sion efficiency is surprising for such a promising design as we discussed above, especially in comparison with SWNT/Si planar heterojunction [31, 34, 234, 239] and recent impressive results achieved for polymer/SiNW array based solar cells (PCE = 11.1 %, [209]). As it follows from the report [209], a critical issue for PV performance is the optimal ratio between SiNW diameter, their lengths and sepa-ration distance in the array in conjunction with dense polymer infiltration. Then the modest performance of CNT/SiNW “dry” solar cell [254] could be explained by (i) poor contact between nanotubes and silicon nanowires (indeed, CNT film

Fig. 1.21 Exciton generation (excitons depicted as balls) in the organic (light gray)/inorganic (dark grey) hybrid structures with planar (a, b) and nanostructured heterojunction (c)

45

was just transferred on the top of SiNWs without dense infiltration through entire array structure) and (ii) non-optimized structure of SiNW array.

In spite of high expectations, CNT/SiNW array structure probably cannot sat-isfy the growing demands of the PV community for thin film, flexibility and light weight features because crystalline Si is required for SiNW array preparation. In this context, it could be very beneficial to create a thin film, flexible solar device comprised of amorphous Si forming nanostructured heterojunction (Fig. 1. 22) with semiconducting SWNTs. Such a solution can be realized through the direct growth of s-SWNTs inside the nanopores of etched a-Si film. Figure 1.22 shows the design of such a hypothetical solar cell that has a good chance to outperform existing thin film photovoltaics based on amorphous Si. Here, nanostructured het-erojunction between p-type s-SWNTs and intrinsic nanoporous a-Si should pro-vide important advantages over traditional p-i-n a-Si device: (i) The replacement of p-type a-Si by s-SWNTs will extend the device absorption range up to 2 μm (instead of 0.73 μm, limited by a-Si band gap) owing to s-SWNT absorbance of NIR light; (ii) Enhanced interfacial area between nanoporous intrinsic a-Si and s-SWNTs (200–800 m2/cm3) compared to the planar junction between p-type, intrinsic, and n-type a-Si; (iii) Higher carrier mobility for quasi-aligned nano-tubes (up to 105 cm2/V s for individual nanotubes [6] and ~60 cm2/V s for SWNT films [7]) as compared with a-Si (hole mobility is ~0.2 cm2/V s [260] and electron mobility is ~2 cm2/V s [261, 262]. Furthermore, because of controllable s-SWNT growth hole mobility can be even enhanced as compared with disordered nanotube network; (iv) Possibility to increase charge concentration and Fermi level tuning for s-SWNT by simple acidic doping [8–10, 31] or various functionalization [11, 12]; (v) A special graded index antireflection (GIAR) porous structure can be inte-grated with the porous layer to minimize the reflection from the surface. The con-cept of GIAR coating is very attractive because such structures should provide a broadband (from UV to NIR) diminishing reflection of 1–5 % for angles up to 60°–80°, as followed from theoretical models [263, 264]; (vi) Because s-SWNT will be utilized instead of p-type a-Si layer it should reduce the total device degradation upon light exposure. This process is called the Staebler–Wronski effect [265] when the defect density of a-Si increases with light exposure, causing an increase in the recombination current and consequently reducing the conversion efficiency.

Fig. 1.22 Schematic of s-SWNT/porous a-Si thin film solar cell. SWNTs are grown directly inside nanoporous a-Si layer with GIAR structure. GIAR structure is fabricated by electrochemi-cal etching with gradually changing porosity from high (top) to low (bottom)



It is known that similar to crystalline Si, amorphous Si can be electrochemi-cally etched (either to be treated by simple wet etching) to produce straight nano-pores or micropores with lengths in the range from submicron to several tens micron [266–268]. Then the semiconducting enriched SWNTs can be reliably grown inside pores (individual or bundles depending on pore diameter) applying CVD methods [269, 270]. Today, CNT growth by CVD is a highly controlled pro-cess when desired characteristic of CNTs are determined by CVD parameters and can be even optimized with computer simulations and predictive modeling [271].

The current level of stabilized PCE for conventional thin film a-Si solar cells is in the range of 7–9 % [272]; therefore, there is a strong demand to increase their efficiency maintaining the same low production cost. The permanent progress in nanostructuring of Si and other group IV, III–V, IV–VI semiconductors in con-junction with novel techniques for CNT growth, sorting, deposition and interface optimization promise further advances in the near future to develop a new genera-tion of hybrid CNT/semiconductor thin, flexible, robust and inexpensive photovol-taics with conversion efficiency exceeding 15 %.

1.5 CNT/Polymer Based Bolometers

In all previous sections, we discussed photophysical phenomena related to light har-vesting and charge separation at the interface between carbon nanotubes and their counterparts, which are the basis of direct photoconductivity and photovoltaic effect. Alternatively, thermal heating, caused by absorbed photons, can perform the conver-sion of light energy to electrical signal in CNT composites. In this case, incident IR irradiation heat the thermistor material and increase its temperature which is measured by a temperature dependent mechanism such as resistance (bolometers), thermoelec-tric voltage (thermopile detectors) or pyroelectric voltage (pyroelectric detectors). The most commonly used in practice and widely investigated for decades has been bolom-eters; thermal detectors based on the resistance decrease upon IR illumination. With respect to photon detectors, bolometers are relatively less sensitive and slower, but simpler to fabricate and provide more opportunities to operate at room temperatures. The major bolometric characteristics, responsivity, detectivity and response, time are determined by the absorption coefficient, thermal coefficient of resistance (TCR), ther-mal conductivity and thermal capacity [273]. Advances in nanotechnology, especially the synthesis of new nanoscale materials, such as CNTs, provide new approaches for reducing the thermal capacity, enhancing IR absorption in the broad spectral range, and increasing TCR by material processing. As a result, novel thermo sensing materi-als based on CNTs and CNT composites have emerged during the past several years as an alternative to traditional metal oxides (e.g. VOx) employed for sensitive elements in conventional bolometers. So far, CNT based bolometric nanostructures and related phenomena are in the research stage; however, their rapid progress indicates that they could be soon transitioned into technologies for various applications in the field of thermal sensing and IR imaging.

47

In the first part of this section, we will present a short review of CNT films for thermal sensing followed by a more detailed analysis in the second part of CNT/polymer composites, focusing on their distinction from pristine carbon nanotubes.

1.5.1 Bolometric Response of Pristine CNT Films

As we mentioned in the introduction, the bolometric response in pristine CNT film usually dominates over direct photoconductivity for steady-state Vis/IR illumination [27, 30, 274–279]. The pioneering work of Huddon’s group [30] presented convincing evidence that photocurrent in SWNT film has a bolomet-ric nature. It was demonstrated that SWNT film responsivity was significantly enhanced (by nearly five orders of magnitude) when film was suspended in vacuum, which is typical for thermal response due to insulation from the envi-ronment (reduction of thermal conductivity to the heat sinks and atmosphere). Also, the direction of photocurrent change correlated with the TCR sign (current increased for negative TCR and decreased for positive values), indicating a heat-ing effect. The response time of ~50 ms was slow enough to be attributed to the direct photoconductivity mechanism, involving exciton generation and their dis-sociation on photoholes and photoelectrons. The film annealing resulted in TCR enhancement (up to 1 %/K at room temperature) because of the elimination of metallic nanotubes, which contributed to non desirable R(T) dependency with a positive TCR sign. The maximum responsivity RV at room temperature was about 30 V/W (Table 1.2) [30].

Table 1.2 Bolometric characteristics of CNT and CNT/polymer composites reported in literature

References Material Ad, device area (mm2)

τ, response time

α, TCR (%/K−1) at ~300 K

Max. RV, responsivity (V/W) at ~300 K

D*, detectivity (cm Hz1/2/W)

Itkis et al. [30] SWNT 1.75 50 ms 1 ~30 N/ALu et al. [274] SWNT 0.073 40–50 ms 0.17 250 3.5 × 105

Lu et al. [276] MWNT 0.073 1.0–2.5 ms 0.07 N/A 3.3 × 106

Xiao et al. [277] MWNT 10 4.4 ms 0.144 30 N/ARao et al. [27] SWNT S* 0.5 ms 0.25 30 N/AGohier et al. [278] MWNT 1.4–0.01 1–3 s 0.19 1160 N/AGustavo et al. [283] SWNT/

polymer76 0.94 ms 2.94 230 1.22 × 108

Glamazda et al. [284]

SWNT/polymer

6–28 150–200 ms 0.5–0.8 500 N/A

Aliev [282] SWNT/polymer

1.48 ≤1 s 0.3 150 N/A

*Area covers 1 mm long part of the 8–20 aligned SWNTs, including the suspended part across the trench with the width of 2 or 35 μm



Wu’s group investigated suspended SWNT [274–276] and MWNT [276] thin films over microchannels patterned on a Si substrate using electron-beam lithog-raphy (Fig. 1.23b). They observed an improved responsivity (up to 250 V/W in vacuum) for suspended SWNT film as compared with unsuspended samples. This result was expected, due to the reduction of thermal conductance according to the expression for bolometric voltage responsivity [273]:

where I is the current, R is the sample resistance, α is the thermal coefficient of resistance (TCR), ε is the light absorption efficiency, G is the thermal conduct-ance, C is the thermal capacitance, and ω is the frequency of light modulation. Note that Eq. (1.1) is valid for a relatively low current I when the Joule heating can be omitted; for higher currents, Joule heating is a limiting factor control-ling the maximum responsivity. Thus, crucial factors for highly sensitive bolom-eters are high TCR values, efficient absorption, low thermal capacitance (thermal mass), and excellent insulation (low thermal conductance). On the other hand, the reduction of thermal conductance should increase the response time as τ ~ C/G. Therefore, a trade off must be considered to provide an optimal balance between responsivity and response time. Keeping this in mind, a strategy for optimization of CNT bolometric material can be formulated as follows: (i) CNT film should be suspended to reduce thermal conductance G; (ii) CNT film should be relatively thin to minimize the thermal capacitance C (in order to decrease response time) but thick enough to maintain sufficient light absorption; (iii) TCR values should be as high as possible. Another important bolometric characteristic is the detectivity D*, which is defined as D* = RV (Ad Δf) 1/2/Vn, where Ad is the detector area, Δf is the unit bandwidth, and Vn is the noise level in volts. The larger detector area, Ad results in higher detectivity; but responsivity is reduced due to the increase

(1.1)RV =I Rαε

√G2 + ω2C2

Fig. 1.23 SEM images of a horizontally aligned and suspended SWNTs across the trench [27] and b SWNT film on top of EBL patterned substrates [274]. Adapted with permission from Nanotechnology, 2009, 20, p 055501, Copyright © 2009 IOP Publishing; Adapted with permis-sion from Applied Physics Letters, 2009, pp 94, 163110 Copyright © 2009, American Institute of Physics

49

of thermal conductance and capacitance. Table 1.2 shows the major bolometric parameters for CNT and CNT composite bolometers reported in literature.

Lu et al. [276] carried out a comparative analysis of the bolometric perfor-mance of SWNT and MWNT suspended/unsuspended films over similar micro-channels/flat surface as in study [274]. They found that sensitivity of MWNT films (defined as relative change of resistance ΔR/R upon incident IR) is higher then for SWNT film, with a much shorter response time (2 ms) with respect to SWNT (50 ms). Authors did not report on MWNT film responsivity, however, its value should be low, considering that the best ΔR/R value for suspended MWNT was only ~0.2 %. Furthermore, from the modeling based on response time deduced from the heat balance equation and in comparison with the experiment they con-cluded, that an increase of the response time (from unsuspended to suspended MWNT film) is consistent with bolometric model, while the decrease in the response time of suspended SWNT film with respect to unsuspended films, cannot be explained by the bolometric effect (indeed, a rough estimate from expression τ ~ C/G indicates an increase of response time due to G reduction). No alternative mechanism has been proposed; however, the direct photoconductivity model is probably ruled out since a diffusion time of separated photocarriers cannot depend on the film suspension.

It is worth noting that there is sufficient speculation in the literature [27, 280, 281] regarding coexisting bolometric effect and direct photoconductivity, when CNT film photoresponse is partially determined by heating, while the remaining portion of the signal change is caused by exciton generation and charge separation (direct photoconductivity mechanism). One simple solution to resolve this problem could be to monitor responsivity changes in the chamber with an inert gas (e.g. Ar) during the pressure reduction from elevated levels to mid/high vacuum (the use of atmosphere should be excluded as atmospheric oxygen might affect the direct pho-toconductivity response [2]).The ratio of responsivity at elevated pressure to the vacuum responsivity could quantitatively estimate the contribution of each mech-anism in the total photoresponse, because the direct photoconductivity is inde-pendent of thermal conductance (which correlates with the vapor concentration). Another approach to separate these mechanisms could be a response detection for light energies lower than bandgap of semiconducting nanotubes; as was done in study [284] for s-SWNT/polymer composite (see more in the next section).

A very short response time of 0.5 ms has been reported for aligned SWNTs directly grown across trenches fabricated in a SiO2/Si wafer [27] (Fig. 1.23a). Authors suggested that such a fast response is the sign of the direct photoconduc-tivity mechanism and ruled out the bolometric effect. However, a short response time could be also explained by the low thermal capacitance C due to extremely low nanotube loading (individual nanotubes/bundles were separated from each other) and fast thermal conductivity (high thermal conductance) because of the absence of intertube junctions. The highest reported responsivity (1200 V/W) has been reached using a MWNT film deposited on a flexible polyimide substrate in an attempt to replace suspended architecture by flexible polymeric substrate with a low thermal conductivity [278]. Here, the high responsivity was mostly associated



with a very small sensing area (~0.01 mm2) as compared with other studies (see Table 1.2), leading to significant reduction of thermal conductance and capaci-tance (Eq. (1.1)). Accordingly, the detectivity D* (not reported) should be low due to the same size dependence, as was mentioned earlier.

Photothermoelectric effect in suspended SWNT films was investigated in stud-ies [286, 287] revisiting previous results [22, 25, 26] where the photoconductivity position effect was interpreted as a diffusion of photoexcited carriers around the Schottky barriers at nanotubes electrode junctions. St-Antoine et al. [286] demon-strated that the photoresponse dependence on laser position is explained by a ther-mal mechanism that is independent of the nanotube-metal barrier. They developed a model that relates the photovoltage profile to the spatial variations of thermoelec-tric power. Although direct photoconductivity through the Schottky barrier might have occurred, this work showed that it is negligible as compared to the photother-mal effect. Similar results were later reported by Omari and Kouklin [287].

1.5.2 CNT/Polymer Bolometers

When carbon nanotubes are embedded in a polymer matrix, bolometric response of such composite material can be drastically changed as compared with pristine CNT film because of a significant increase of TCR and responsivity [280–285]. This is a definite advantage over pristine CNT networks, making composite films a promising material for applications in thermal IR sensing. Other advantages may include thin film flexibility, easy fabrication and scalability, high elasticity and durability as distinct from the fragility of extra thin pristine CNT film [30]. Indeed, to reach an acceptable TCR pristine SWNT film should be very thin (sev-eral tens nanometers) and annealed at high temperature [30], which makes device fabrication complicated and technologically problematic. This is in stark contrast to CNT/polymer composite films, which can be readily prepared by simple wet deposition techniques (casting, printing) over large areas with a good reproducibil-ity and controlled thickness in the range of several microns. It is important to note that relatively thick composite film can provide TCR values equal or greater [283, 284] than ultra thin SWNT films [30, 274, 276] (Table 1.2). Furthermore, because polymer matrix has a critical impact on bolometric characteristics of the compos-ite film, optimal properties of the thermal sensor can be adjusted by varying many polymer parameters such as its formula/functionalization, molecular weight, ther-mal conductivity, CNT alignment using thermal pooling, etc.

The primary cause of enhanced TCR value for CNT/polymer composite is likely related to fluctuation-induced tunneling (FIT) conductivity in nanotube-pol-ymer-nanotube junctions based on the model developed by Sheng et al. [288, 289] for carbon polymer composite. According to FIT model, the composite resistivity ρ is expressed as follows:

(1.2)ρ = ρ0e

T1T +T0

51

where kT1 is the activation energy, and T0 is the temperature determining the thermal activated mechanism. As was previously shown [281], T0 value is small (~1–5 °K) and can be ignored at room temperatures. Thus, the critical parameter characterizing the slope of the ρ(t) dependence is T1, which can be expressed as [288]:

where A is the area of the junction between two nanotube bundles, separated by the polymer at distance ω, V0 is the barrier potential in the center of the junction, and e is the electron charge. Higher T1 magnitude provides a steeper slope for the ρ(t) dependence and consequently a higher TCR value. The fluctuation-induced tunneling conductivity model was proposed for disordered heterogeneous sys-tems as opposed to systems with hopping charge carrier transport between local-ized sites. These systems (e.g., conductor–insulator composites) consist of large conductive segments (where electrons are delocalized and free to move over very large distances as compared to the atomic dimensions) separated by small insulat-ing gaps. The FIT model has been applied to pristine CNT films and fibers with some restrictions [290, 291], since the carrier transport in disordered nanotube networks can also be determined by charge tunneling through junctions between nanotubes or their bundles.

From a general point of view and according to Eq. (1.3) the T1 value for a CNT/polymer composite should exceed that for a pristine CNT network due to higher barrier potential (V0) which amplitude is critically enhanced in the pres-ence of the insulating polymer as distinct from direct nanotube–nanotube junction.

In studies of SWNT/polycarbonate bolometric materials [281, 284] experimen-tal R(T) dependencies were in agreement with the FIT model allowing to estimate T1 values for different types of SWNTs [281] and TCR values [284] for disordered

(1.3)T1 =8AV

20

πωe2k

ω1 ω2

(a) (b)

Fig. 1.24 Schematic demonstrating a characteristic parameters of junction between SWNT bun-dles according to Eq. (1.3) for isotropic (A1, ω1) and aligned (A2, ω2) SWNT-polymer composite (A2 > A1, ω2 < ω1); and b a smaller critical depth ZC

⊥ (normal to the alignment direction) as compared to critical depth ZC

|| (parallel to the alignment direction) [284] Reprinted with permis-sion from Advanced Functional Materials, 2012, 22, pp 2177–2186, Copyright © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim



and quasi-aligned SWNTs. Pradhan et al. [280, 281] reported that pristine SWNT films exhibited a bolometric response, while CNT/polymer composites displayed a substantial contribution from direct photoconductivity. However, the proposed method of mechanisms separation (bolometric vs. direct photoconductivity) based on direct temperature change upon light illumination followed by the estimate from modeling cannot be accurate enough to result in such quantitative outcome. Also, an increased of sensitivity of CNT/polymer composite could be readily explained by the same bolometric response with the reduced thermal conductance (G) due to polymer low thermal conductivity and enhanced TCR (Eq. (1.1)) when compared with pristine SWNT film, as previously discussed.

Glamazda et al. [284] demonstrated that enhanced TCR in SWNT/polystyrene composite (as compared with pristine SWNT) is not only attained for isotropic composite films, but can be further increased for quasi-aligned SWNT bundles produced by the thermal stretching (Fig. 1.25b).

Keeping in mind the Eq. (1.3) we can evaluate how the stretching process affects the T1 parameter: as stretching results in SWNT bundles with substantial alignment, the junction area A should increase with simultaneous reduction of the distance ω between bundles (Fig. 1.24a). Thus, the film stretching should enhance the tunneling activation energy, kT1 leading to a higher TCR value, compared with an isotropic film. Furthermore, the alignment effect can explain why the TCR value measured normal to the stretching direction was larger than the TCR measured along the stretching direction, which arises from the gradient character of SWNT coating [284]. A distinctive feature of this study was a gradient SWNT

0

2

4

6

8

10

12Gradient,aligned,cut normalSWNT-polymer

Gradient,aligned,cut parallelSWNT-polymer

Gradient,isotropicSWNT-polymer

Uniform SWNT-polymer

SWNT film

F, a

.u

(a) (b)

(c)

Fig. 1.25 a Histogram demonstrating the responsivity figure of merit (F) for SWNT film and different types of SWNT-polymer composite; b SEM image of SWNT/polystyrene film stretched by thermal poling [284]; c SEM image of horizontally aligned SWNT forming dense network [283]. Reprinted with permission from Advanced Functional Materials, 2012, 22, pp 2177–2186, Copyright © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim; Adapted with permis-sion from ACS Appl. Mater. Interfaces, 2011, 3 (8), pp 3200–3204, Copyright © 2011 American Chemical Society

53

distribution into polystyrene matrix when nanotube concentration is characterized by a gradient along the film depth from the surface to the inner areas (in all previ-ous reports [280–283, 285] CNTs were uniformly distributed over polymer film). The strategy to improve bolometric responsivity has been deduced: from pristine SWNT film, to isotropic SWNT/polymer film, next to anisotropic composite and finally to anisotropic film with axial direction normal to nanotube bundles align-ment. Figure 1.25a demonstrates this trend. As a result the highest responsivity (500 V/W) among CNT/polymer bolometric materials reported in literature has been reached (Table 1.2).

The importance of SWNT alignment in a polymer matrix for improved bolo-metric performance has been also demonstrated by Gustavo et al. [283]. Here an exceptionally high TCR of 2.94 %/K (comparable with commercial VOx bolom-eters [273]) was obtained for horizontally aligned SWNT large bundles embedded in surfactant polymer (Table 1.2). An alignment of the SWNTs occurred during the deposition process and, as authors believe, might be associated with spontane-ous self-orientation in the thick bundles consisting of many aggregated individ-ual tubes (Fig. 1.25c). It is important to note that control samples without bundle alignment exhibited lower TCR, responsivity and detectivity than highly align-ment samples. Presumably such high TCR is associated with an enhancement of the junction area A (Eq. (1.3)) for aligned bundles according FIT model which is consistent with results of the study [284].

1.6 Conclusion

As we considered in this chapter, CNTs interfaced with various materials dem-onstrate remarkable photophysical effects ranging from PICT in relatively simple CNT/dye assemblies to charge separation at CNT/semiconductor heterojunction. This area is less investigated than the photophysics of pristine nanotubes, just because plenty of organic and inorganic compounds which, in conjunction with nanotubes, create new opportunities for light interaction with such nanostructured hybrids.

The purpose of this chapter was not a detailed review of all CNT based pho-toactive hybrid systems, (e.g. reader can find excellent reviews about PICT in CNT/small molecules hybrids [11, 12], or light conversion in CNT/polymer solar cells [35, 129]) but rather provide the latest results updating and revising previ-ous achievements in this field opening new horizons for the future investigations. In this context, it is worth mentioning the studies of Strano group shedding light on the problem of low efficiency CNT/polymer solar cells [145], or recent reports [33, 38] demonstrating the importance of nanotubes as photoabsorbers in light energy conversion.

Another exciting direction is the observation of carrier multiplication in split CNT FET system and related theoretical models predicting possible observation of this effect at room temperature. These findings give promise to employ CNT



to boost conversion efficiency beyond Shockley–Queisser limit in real solar cells, when split FET geometry could be replaced by heterojunction with other materials.

One more example of a rapid progress in CNT hybrid photoconversion is an emerging class of novel CNT/Si solar cells with outstanding PCE of ~14 % [31] exceeding any others among organic and hybrid PVs. It seems that such nano-structures being modified to thin film flexible architecture could be highly compet-itive with conventional amorphous Si solar cells. Also, it is very attractive for the thermal IR sensing as CNT/polymer composites are demonstrating several advan-tages over pristine CNT based bolometers in terms of responsivity, detectivity, mechanical and robustness. Importantly, the polymeric matrix provides not only higher TCR at room temperatures but also makes processing easier as compared with pristine CNT bolometers.

There is the opinion that carbon nanotubes, to some extent, have exhausted their potential as an innovative research nano-object, with the focus shifting to the graphene concept. However, as we have tried to demonstrate in this review, CNTs and especially CNT hybrids still remain unique photophysical objects full of intrinsic merit and promising breakthroughs in fundamental and applied science.

References

1. Fujiwara, A., Matsuoka, Y., Suematsu, H., Ogawa, N., Miyano, K., Kataura, H., Maniwa, Y., Suzuki, S., Achiba, Y.: Photoconductivity in semiconducting single-walled carbon nanotubes. Jpn. J. Appl. Phys. 40, L1229–L1231 (2001)

2. Levitsky, I.A., Euler, W.B.: Photoconductivity of single wall carbon nanotubes under CW NIR illumination. Appl. Phys. Lett. 83, 1857–1859 (2003)

3. Freitag, M., Martin, Y., Misewich, J.A., Martel, R., Avouris, P.: Photoconductivity of single carbon nanotubes. Nano Lett. 3, 1067–1071 (2003)

4. Saito, R., Dresselhaus, G., Dresselhaus, S.M.: Physical Properties of Carbon Nanotubes. Imperial College Press, London (1998)

5. Dresselhaus, M.S., Dresselhaus, G., Avouris, P.: Carbon Nanotubes: Synthesis, Structure, Properties and Applications. Springer, New York (2000)

6. Durkop, T., Getty, S.A., Cobas, E., Fuhrer, M.S.: Extraordinary mobility in semiconducting carbon nanotubes. Nano Lett. 4, 35–39 (2009)

7. Xiao, K., Liu, Y., Hu, P., Yu, G., Wang, X., Zhu, D.: High-mobility thin-film transistors based on aligned carbon nanotubes. Appl. Phys. Lett. 83, 150–152 (2003)

8. Sankapal, B.R., Setyowati, K., Chen, J., Liu, H.: Electrical properties of air-stable, iodine-doped carbon-nanotube-polymer composites. Appl. Phys. Lett. 91, 173103 (2007)

9. Lee, R.S., Kim, H.J., Fischer, J.E., Thess, A., Smalley, R.E.: Conductivity enhancement in single-walled carbon nanotube bundles doped with K and Br. Nature 388, 255–257 (1997)

10. Chen, J., Hamon, M.A., Hu, H., Chen, Y., Rao, A.M., Eklund, P.C., Haddon, R.C.: Solution properties of single-walled carbon nanotubes. Science 282, 95–98 (1998)

11. Sgobba, V., Guldi, D.M.: Ground and excited state charge transfer and its implications. In: Guldi, D.M., Martin, M. (eds.) Carbon Nanotubes and Related Structures, pp. 233–289. Wiley-VCH, Weinheim (2010)

12. Karousis, N., Tagmatarchis, N.: Current progress on the chemical modification of carbon nanotubes. Chem. Rev. 110, 5366–5397 (2010)

13. Ando, T.: Excitons in carbon nanotubes. J. Phys. Soc. Jpn. 66, 1066–1073 (1997)

55

14. Perebeinos, V., Tersoff, J., Avouris, P.: Scaling of excitons in carbon nanotubes. Phys. Rev. Lett. 92, 257402 (2004)

15. Wang, F., Dukovic, G., Brus, L.E., Heinz, T.F.: The optical resonances in carbon nanotubes arise from excitons. Science 308, 838–841 (2005)

16. Balasubramanian, K., Friedrich, M., Fan, Y., Wannek, U., Mews, A., Burghard, M., Kern, K.: Photoelectronic transport imaging of individual semiconducting carbon nanotubes. Appl. Phys. Lett. 84, 2400–2402 (2004)

17. Freitag, M., Tsang, J.C., Bol, A., Yuan, D., Liu, J., Avouris, P.: Imaging of the Schottky barri-ers and charge depletion in carbon nanotube transistors. Nano Lett. 7, 2037–2042 (2007)

18. Lee, J.U.: Photovoltaic effect in ideal carbon nanotube diodes. Appl. Phys. Lett. 87, 073101 (2005)

19. Lee, J.U., Codella, P.J., Pietrzykowski, M.: Direct probe of excitonic and continuum transi-tions in the photocurrent spectroscopy of individual carbon nanotube p–n diodes. Appl. Phys. Lett. 90, 053103 (2007)

20. Lien, D.H., Hsu, W.K., Zan, H.W., Tai, N.H., Tsai, C.H.: Photocurrent amplification at car-bon nanotube—metal contacts. Adv. Mater. 18, 98–103 (2006)

21. Mohite, A., Chakraborty, S., Gopinath, P., Sumanasekera, G.U., Alphenaar, B.W.: Displacement current detection of photoconduction in carbon nanotubes. Appl. Phys. Lett. 86, 061114 (2005)

22. Merchant, C.A., Markovica, N.: Effects of diffusion on photocurrent generation in single-walled carbon nanotube films. Appl. Phys. Lett. 92, 243510 (2008)

23. Fujiwara, A., Matsuoka, Y., Suematsu, H., Ogawa, N., Miyano, K., Kataura, H., Maniwa, Y., Suzuki, S., Achiba, Y.: Photoconductivity of single-wall carbon nanotube films. Carbon 42, 919–922 (2004)

24. Sarker, B.K., Arif, M., Stokes, P., Khondaker, S.I.: Diffusion mediated photoconduction in multiwalled carbon nanotube films. J. Appl. Phys. 106, 074307-1–074307-4 (2009)

25. Lu, S., Panchapakesan, B.: Photoconductivity in single wall carbon nanotube sheets. Nanotechnology 17, 1843–1850 (2006)

26. Liu, Y., Lu, S., Panchapakesan, B.: Alignment enhanced photoconductivity inside single wall carbon nanotube films. Nanotechnology 20, 035203 (2009)

27. Rao, F., Liu, X., Li, T., Zhou, Y., Wang, Y.: The synthesis and fabrication of horizontally aligned single-walled carbon nanotubes suspended across wide trenches for infrared detect-ing application. Nanotechnology 20, 055501 (2009)

28. Sun, J.-L., Xu, J., Zhu, Z.-L., Li, B.: Disordered multi-walled carbon nanotube mat for light spot position detecting. Appl. Phys. A 91, 229–233 (2008)

29. Wei, J., Sun, J.-L., Zhu, J.-L., Wang, K., Wang, Z., Luo, J., Wu, D., Cao, A.: Carbon nano-tube macrobundles for light sensing. Small 2, 988–993 (2006)

30. Itkis, M.E., Borondics, F., Yu, A., Haddon, R.C.: Bolometric infrared photoresponse of sus-pended single-walled carbon nanotube films. Science 312, 413–416 (2006)

31. Jia, Y., Cao, A., Bai, X., Li, Z., Zhang, L., Guo, N., Wei, J., Wang, K., Zhu, H., Wu, D., Ajayan, P.M.: Achieving high efficiency silicon-carbon nanotube heterojunction solar cells by acid doping. Nano Lett. 11, 1901–1905 (2011)

32. Li, Z., Kunets, V.P., Saini, V., Xu, Y., Dervishi, E., Salamo, G.J., Biris, A.R., Biris, A.S.: Light-harvesting using high density p-type single wall carbon nanotube/n-type silicon hetero-junctions. ACS Nano 3, 1407 (2009)

33. Ong, P.-L., Euler, W.B., Levitsky, I.A.: Hybrid solar cells based on single-walled carbon nanotubes/Si heterojunction. Nanotechnology 21, 105203 (2010)

34. Wadhwa, P., Liu, B., McCarthy, M.A., Wu, Z., Rinzler, A.G.: Electronic junction control in a nanotube-semiconductor Schottky junction solar cell. Nano Lett. 10, 5001–5005 (2010)

35. Dennler, G., Brabec, C.J.: Organic Photovoltaics: Materials, Device Physics, and Manufacturing Technologies. Wiley-VCH, Weinheim (2008)

36. Chen, C.-Y., Wang, M., Li, J.-Y., Pootrakulchote, N., Alibabaei, L., Ngoc-le, C., Decoppet, J.D., Tsai, J., Gratzel, C., Wu, C., Zakeeruddin, S.M., Gratzel, M.: Highly efficient light-harvesting ruthenium sensitizer for thin-film dye-sensitized solar cells. ACS Nano 3, 3103–3109 (2009)

References


37. Yu, Q., Wang, Y., Yi, Z., Zu, N., Zhang, J., Zhang, M., Wang, P.: High-efficiency dye-sensi-tized solar cells: the influence of lithium ions on exciton dissociation, charge recombination, and surface states. ACS Nano 4, 6032–6038 (2010)

38. Bindl, D.J., Wu, M.-Y., Prehn, F.C., Arnold, M.S.: Efficiently harvesting excitons from elec-tronic type-controlled semiconducting carbon nanotube films. Nano Lett. 11, 455–460 (2011)

39. Bottari, G., De la Torre, G., Guldi, D.M., Torres, T.: Covalent and noncovalent phthalocya-nine carbon nanostructure systems: synthesis, photoinduced electron transfer, and application to molecular photovoltaics. Chem. Rev. 110, 6768–6816 (2010)

40. Imahori, H., Umeyama, T.: Donor–acceptor nanoarchitecture on semiconducting electrodes for solar energy conversion. J. Phys. Chem. C 113, 9029–9039 (2009)

41. Strano, M.S., Dyke, C.A., Usrey, M.L., Barone, P.W., Allen, M.J., Shan, S.W., Kittrell, C., Hauge, R.H., Tour, J.M., Smalley, E.R.: Electronic structure control of single-walled carbon nanotubes. Science 301, 1519–1522 (2003)

42. Itkis, M.E., Niyogi, S., Meng, M.E., Hamon, M.A., Hu, H., Haddon, R.C.: Spectroscopic study of the fermi level electronic structure of single-walled carbon nanotubes. Nano Lett. 2, 155 (2002)

43. Maligaspe, E., Sandanayaka, A.S.D., Hasobe, T., Ito, O., D’Souza, F.: Sensitive efficiency of photoinduced electron transfer to band gaps of semiconductive single-walled carbon nano-tubes with supramolecularly attached zinc porphyrin bearing pyrene glues. J. Am. Chem. Soc. 132, 8158–8164 (2010)

44. Sandanayaka, A.S.D., Subbaiyan, N.K., Das, S.K., Chitta, R., Maligaspe, E., Hasobe, T., Ito, O., D’Souza, F.: Diameter sorted SWNT-porphyrin and SWNT-phthalocyanine conjugates for light-energy harvesting. ChemPhysChem 12, 2266–2273 (2011)

45. D’Souza, F., Smith, P.M., Zandler, M.E., McCarty, A.L., Itou, M., Araki, Y., Ito, O.: Energy transfer followed by electron transfer in a supramolecular triad composed of boron dipyrrin, zinc porphyrin and fullerene: a model for the photosynthetic antenna-reaction center com-plex. J. Am. Chem. Soc. 126, 7898–7907 (2004)

46. Baskaran, D., Mays, J.W., Zhang, X.P., Bratcher, M.S.: Carbon Nanotubes with covalently linked porphyrin antenna: photoinduced electron transfer. J. Am. Chem. Soc. 127, 6916 (2005)

47. Alvaro, M., Atienzar, P., De la Cruz, P., Delgado, J.L., Troiani, V., Garcia, H., Langa, F., Palkar, A., Echegoyen, L.: Synthesis, photochemistry, and electrochemistry of single-walled carbon nanotubes with pendent pyridyl groups and of their metal complexes with zinc porphyrin. Comparison with pridyl bearing fullerenes. J. Am. Chem. Soc. 128, 6626–6635 (2006)

48. Campidelli, S., Sooambar, C., Diz, E.L., Ehli, C., Guldi, D.M., Prato, M.: Dendrimer func-tionalized single-walled carbon nanotubes: synthesis, characterization, and photoinduced electron transfer. J. Am. Chem. Soc. 128, 12544 (2006)

49. Yu, J., Mathew, S., Flavel, B.S., Johnston, M.R., Shapter, J.G.: Ruthenium porphyrin func-tionalized single-walled carbon nanotube arrays. A step toward light harvesting antenna and multibit information storage. J. Am. Chem. Soc. 130, 6788–8796 (2008)

50. De la Torre, G., Blau, W., Torres, T.: A survey on the functionalized single-walled nanotubes. The chemical attachment of phthalocyanine moieties. Nanotechnology 14, 765 (2003)

51. Xu, H.-B., Chen, H.-Z., Shi, M.-M., Bai, R., Wang, M.: A novel donor–acceptor hetero-junction from single-walled carbon nanotubes functionalized by erbium bisphthalocyanine. Mater. Chem. Phys. 94, 342–346 (2005)

52. Yang, Z., Pu, H., Yuan, J., Wan, D., Liu, Y.: Phthalocyanines-MWCNT hybrid materials: fab-rication, aggregation and photoconductivity properties improvement. Chem. Phys. Lett. 465, 73–77 (2008)

53. Dyke, C.A., Tour, J.M.: Covalent functionalization of single-walled carbon nanotubes for materials applications. J. Phys. Chem. A 108, 11151–11159 (2004)

54. Ballesteros, B., de la Torre, G., Ehli, C., Rahman, G.M.A., Agullo′-Rueda, F., Guldi, D.M., Torres, T.: Single-wall carbon nanotubes bearing covalently linked phthalocyanines—pho-toinduced electron transfer. J. Am. Chem. Soc. 129, 5061–5068 (2007)

55. Guldi, D.M., Marcaccio, M., Paolucci, D., Paolucci, F., Tagmatarchis, N., Tasis, D., Va′zquez, E., Prato, M.: Single-wall carbon nanotube-ferrocene nanohybrids: observing intramolecular electron transfer in functionalized SWNTs. Angew. Chem. Int. Ed. 42, 4206–4209 (2003)

57

56. Yang, X., Lu, Y., Ma, Y., Li, Y., Du, F., Chen, Y.: Noncovalent nanohybrid of ferrocene with single-walled carbon nanotubes and its enhanced electrochemical property. Chem. Phys. Lett. 420, 416–420 (2006)

57. Segura, J.L., Martin, N.: New concepts in tetrathiofulvalene chemistry. Angew. Chem. Int. Ed. 40, 1372–1409 (2001)

58. Martin, N., Sanchez, L., Herranz, M.A., Illescas, B., Guldi, D.M.: Electronic communication in tetrathiofulvalene (TTF)/C60 systems: toward molecular solar energy conservation materi-als? Acc. Chem. Res. 40, 1015–1024 (2007)

59. Kam, N.W.S., Jessop, T.C., Wender, P.A., Dai, H.: Nanotube molecular transporters: inter-nalization of carbon nanotubes-protein conjugates into mammalian cells. J. Am. Chem. Soc. 126, 6850–6851 (2004)

60. Jang, S.R., Vittal, R., Kim, K.J.: Incorporation of single-walled carbon nanotubes in dye-sen-sitized TiO2 solar cells. Langmuir 20, 9807–9810 (2004)

61. Nakashima, N., Tomonari, Y., Murakami, H.: Water-soluble single-walled carbon nanotubes via noncovalent sidewall-functionalization. Chem. Lett. 31, 638 (2002)

62. Guldi, D.M., Rahman, G.M.A., Jux, N., Tagmatarchis, N., Prato, M.: Integrating single-wall carbon nanotubes into donor–acceptor nanohybrids. Angew. Chem. Int. Ed. 43, 5526 (2004)

63. Yanagi, K., Iakoubovskii, K., Kazaoui, S., Minami, N., Maniwa, Y., Miyata, Y., Kataura, H.: Light-harvesting function of β-carotene inside carbon nanotubes. Phys. Rev. B 74, 155420 (2006)

64. Yanagi, K., Miyata, Y., Kataura, H.: Highly stabilized β-carotene in carbon nanotubes. Adv. Mater. 18, 437 (2006)

65. Yanagi, K., Iakoubovskii, K., Matsui, H., Matsuzaki, H., Okamoto, H., Miyata, Y., Maniwa, Y., Kazaoui, S., Minami, N., Kataura, H.: Photosensitive function of encapsulated dye in car-bon nanotubes. J. Am. Chem. Soc. 129, 4992 (2007)

66. Marcus, R.A., Sutin, N.: Electron transfers in chemistry and biology. Biochim. Biophys. Acta 811, 265–322 (1985)

67. Marcus, R.A.: Electron transfer reactions in chemistry: theory and experiment. Angew. Chem. Int. Ed. 32, 1111–1121 (1993)

68. Sandanayaka, A.S.D., Maligaspe, E., Hasobe, T., Ito, O., D’Souza, F.: Band gap dependent electron transfer in supramolecular nanohybrids of (6,5)- or (7,5)-enriched semiconducting SWNT as donors and fullerenes as acceptor. Chem. Commun. 46, 8749 (2010)

69. Das, S.K., Subbaiyan, N.K., D’Souza, F., Sandanayaka, A.S.D., Hasobe, T., Ito, O.: Photoinduced processes of the supramolecularly functionalized semi-conductive SWNTs with porphyrins via ion-pairing interactions. Energy Environ. Sci. 4, 707–716 (2011)

70. Guldi, D.M., Rahman, G.M.A., Prato, M., Jux, N., Qin, S., Ford, W.T.: Single-wall carbon nanotubes as integrative building blocks for solar-energy conversion. Angew. Chem. Int. Ed. 44, 2015–2018 (2005)

71. Rahman, G.M.A., Troeger, A., Sgobba, V., Guldi, D.M., Jux, N., Tchoul, M.N., Ford, W.T., Mateo-Alonso, A., Prato, M.: Improving photocurrent generation: supramolecularly and covalently functionalized single-wall carbon nanotubes-polymer/porphyrin donor–acceptor nanohybrids. Chem. Eur. J. 14, 8837–8846 (2008)

72. Umeyama, T., Fujita, M., Tezuka, N., Kadota, N., Matano, Y., Imahori, H.: Electrophoretic deposition of single-walled carbon nanotubes covalently modified with bulky porphyrins on nanostructured SnO2 electrodes for photoelectrochemical devices. J. Phys. Chem. C 111, 11484 (2007)

73. Hasobe, T., Fukuzumi, S., Kamat, P.V.: Stacked-cup carbon nanotubes for photoelectrochem-ical solar cells. Angew. Chem. Int. Ed. 45, 755–759 (2006)

74. Hasobe, T., Fukuzumi, S., Kamat, P.V.: Organized assemblies of single-wall carbon nano-tubes and porphyrin for photochemical solar cells: charge injection from excited porphyrin into single-walled carbon nanotubes. J. Phys. Chem. B 110, 25477–25484 (2006)

75. Rahman, G.M.A., Guldi, D.M., Cagnoli, R., Mucci, A., Schenetti, L., Vaccari, L., Prato, M.: Combining single wall carbon nanotubes and photoactive polymers for photoconversion. J. Am. Chem. Soc. 127, 10051–10057 (2005)

References


76. Guldi, D.M., Rahman, G.M.A., Sgobba, V., Jux, N., Campidelli, S., Prato, M.: Supramolecular assemblies of different carbon nanotubes for photoconversion processes. Adv. Mater. 18, 2264–2269 (2006)

77. Ham, M.-H., Choi, J.H., Boghossian, A.A., Jeng, E.S., Graff, R.A., Heller, D.A., Chang, A.C., Mattis, A., Bayburt, T.H., Grinkova, Y.V., Zeiger, A.S., Van Vliet, K.J.E., Hobbie, E.K., Sligar, S.G., Wraight, C.A., Strano, M.S.: Photoelectrochemical complexes for solar energy conversion that chemically and autonomously regenerate. Nat. Chem. 2, 929–936 (2010)

78. Chen, T., Wang, S., Yang, Z., Feng, Q., Sun, X., Li, L., ShengWang, Z., Peng, H.: Flexible, light-weight, ultrastrong, and semiconductive carbon nanotube fibers for a highly efficient solar cell. Angew. Chem. Int. Ed. 50, 1815–1819 (2011)

79. Peng, H., Jain, M., Li, Q., Peterson, D.E., Zhu, Y., Jia, Q.: Vertically aligned pearl-like carbon nanotube arrays for fiber spinning. J. Am. Chem. Soc. 130, 1130–1131 (2008)

80. Peng, H., Jain, M., Peterson, D.E., Zhu, Y., Jia, Q.: Composite carbon nanotubes/silica fibers with improved mechanical strengths and electrical conductivities. Small 4, 1964–1967 (2008)

81. Kymakis, E., Amaratunga, G.A.J.: Photovoltaic cells based on dye-sensitisation of single-wall carbon nanotubes in a polymer matrix. Sol. Energy Mater. Sol. Cells 80, 465–472 (2003)

82. Bhattacharyya, S., Kymakis, E., Amaratunga, G.A.J.: Photovoltaic properties of dye func-tionalized single-wall carbon nanotube/conjugated polymer devices. Chem. Mater. 16, 4819–4823 (2004)

83. Hatton, R.A., Blanchard, N.P., Miller, A.J., Silva, S.R.P.: A multi-wall carbon nanotube-molec-ular semiconductor composite for bi-layer organic solar cells. Physica E 27, 124–127 (2007)

84. Hatton, R.A., Blanchard, N.P., Stolojan, V., Miller, A.J., Silva, S.R.P.: Nanostructured copper phthalocyanine-sensitized multiwall carbon nanotube films. Langmuir 23, 6424–6430 (2007)

85. Feng, W., Fujii, A., Ozaki, M., Yoshino, K.: Perylene derivative sensitized multi-walled car-bon nanotube thin film. Carbon 43, 2501–2507 (2005)

86. Klare, J.E., Murray, I.P., Goldberger, J., Stupp, S.I.: Assembling p-type molecules on single wall carbon nanotubes for photovoltaic devices. Chem. Commun. 25, 3705–3707 (2009)

87. Oregan, B., Grätzel, M.: A low cost, high efficiency solar cell based on dye-sensitized colloi-dal TiO2 films. Nature 353, 737–740 (1991)

88. Grätzel, M.: Recent advances in sensitized mesoscopic solar cells. Acc. Chem. Res. 42(11), 1788–1798 (2009)

89. Jung, K.-H., Hong, J.S., Vittal, R., Kim, K.-J.: Enhanced photocurrent of dye-sensitized solar cells by modification of TiO2 with carbon nanotubes. Chem. Lett. 31, 864–865 (2002)

90. Kongkanand, A., Dominguez, R.M., Kamat, P.V.: Single wall carbon nanotube scaffolds for photoelectrochemical solar cells. Capture and transport of photogenerated electrons. Nano Lett. 7, 676–680 (2007)

91. Brown, P., Takechi, K., Kamat, P.V.: Single-walled carbon nanotube scaffolds for dye-sensi-tized solar cells. J. Phys. Chem. C 112, 4776–4782 (2008)

92. Lee, W., Lee, J., Lee, S., Yi, W., Han, S.H., Cho, B.W.: Enhanced charge collection and reduced recombination of CdS/TiO2 quantum dots sensitized solar cells in the presence of single-walled carbon nanotubes. Appl. Phys. Lett. 92, 153510 (2008)

93. Zhang, S., Niu, H., Lan, Y., Cheng, C., Xu, J., Wang, X.: Synthesis of TiO2 nanoparticles on plasma-treated carbon nanotubes and its application in photoanodes of dye-sensitized solar cells. J. Phys. Chem. C 115, 22025–22034 (2011)

94. Yu, J., Fan, J., Cheng, B.: Dye-sensitized solar cells based on anatase TiO2 hallow spheres/carbon nanotube composite films. J. Power Sources 196, 7891–7898 (2011)

95. Yen, C.Y., Lin, Y.F., Hung, C.H., Tseng, Y.H., Ma, C.C.M., Chang, M.C., Shao, H.: The effects of synthesis procedures ob the morphology and photocatalytic activity of multi-walled carbon nanotubes/TiO2 nanocomposites. Nanotechnology 19, 045604 (2008)

96. Miller, A.J., Hatton, R.A., Silva, S.R.P.: Interpenetrating multiwall carbon nanotube elec-trodes for organic solar cells. Appl. Phys. Lett. 89, 133117 (2006)

97. Lee, T.Y., Alegaonkar, P.S., Yoo, J.B.: Fabrication of dye-sensitized solar cell using TiO2 coated carbon nanotubes. Thin Solid Films 515(12), 5131–5135 (2007)

59

98. Brennan, L.J., Byrne, M.T., Bari, M., Gun’ko, Y.K.: Carbon nanomaterials for dye-sensitized solar cell applications: a bright future. Adv. Energy Mater. 1, 472–485 (2011)

99. Kyaw, A.K.K., Tantang, H., Wu, T., Ke, L., Peh, C., Huang, Z.H., Zeng, X.T., Demir, H.V., Zhang, Q., Sun, X.W.: Dye-sensitized solar cell with a titanium-oxide-modified carbon nano-tube transparent electrode. Appl. Phys. Lett. 99, 021107 (2011)

100. Yen, M.Y., Hsiao, M.C., Liao, S.H., Liu, P.-I., Tsai, H.M., Ma, C.C.M., Pu, N.W., Ger, M.D.: Preparation of graphene/carbon multi-walled carbon nanotube hybrid and its use as photoanodes of dye-sensitized solar cells. Carbon 49, 3597–3606 (2011)

101. Zhang, S., Ji, C., Bian, Z., Liu, R., Xia, X., Yun, D., Zhang, L., Huang, C., Cao, A.: Single-wire dye-sensitized solar cells wrapped by carbon nanotube film electrodes. Nano Lett. 11, 3383–3387 (2011)

102. Chen, T., Qiu, L., Cai, Z., Gong, F., Yang, Z., Wang, Z., Peng, H.: Intertwined aligned car-bon nanotube fiber based dye-sensitized solar cells. Nano Lett. 12(5), 2568–2572 (2012)

103. Williams, G., Seger, B., Kamat, P.V.: TiO2-graphene nanocomposites. UV-assisted photo-catalytic reduction of graphene oxide. ACS Nano 2, 1487–1491 (2008)

104. Yang, N., Zhai, J., Wang, D., Chen, Y.S., Jiang, L.: Two-dimensional graphene bridges enhanced photoinduced charge transport in dye-sensitized solar cells. ACS Nano 4, 887–894 (2010)

105. Khairoutdinov, R.F., Doubova, L.V., Haddon, R.C., Saraf, L.: Persistent photoconductivity in chemically modified single-wall carbon nanotubes. J. Phys. Chem. B 108, 19976–19981 (2004)

106. Hecht, D.S., Ramirez, R.J.A., Briman, M., Artukovic, E., Chichak, K.S., Stoddart, J.F., Gruner, G.: Bioinspired detection of light using a porphyrin-sensitized single-wall nanotube field effect transistor. Nano Lett. 6, 2031–2036 (2006)

107. Zhao, Y.-L., Hu, L., Gruner, G., Stoddart, J.F.: A tunable photosensor. J. Am. Chem. Soc. 130, 16996–17003 (2008)

108. Guo, X., Huang, L., O’Brien, S., Kim, P., Nuckolls, C.: Directing and sensing changes in molecular confirmation on individual carbon nanotube field effect transistors. J. Am. Chem. Soc. 127, 15045–15047 (2005)

109. Simmons, J.M., In, I., Campbell, V.E., Mark, T.J., Léonard, F., Gopalan, P., Eriksson, M.A.: Optically modulated conduction in chromophore functionalized single-wall carbon nano-tubes. Phys. Rev. Lett. 98, 086802 (2007)

110. Zhou, X., Zifer, T., Wong, B.M., Krafcik, K.L., Léonard, F., Vance, A.L.: Color detection using chromophore-nanotube hybrid devices. Nano Lett. 9, 1028–1033 (2009)

111. Vijayakumar, C., Balan, B., Kim, M.J., Takeuchi, M.: Noncovalent functionalization of SWNTs with azobenzene-containing polymers: solubility, stability, and enhancement of photoresponsive properties. J. Phys. Chem. C 115, 4533–4539 (2011)

112. Huang, C., Wang, R.K., Wong, B.M., McGee, D.J., Leonard, F., Kim, Y.J., Johnson, K.F., Arnold, M.S., Eriksson, M.A., Gopalan, P.: Spectroscopic properties of nanotube-chromo-phore hybrids. ACS Nano 5(10), 7767–7774 (2011)

113. Del Valle, M., Gutierrez, R., Tejedor, C., Cuniberti, G.: Tuning the conductance of a molec-ular switch. Nat. Nanotechnol. 2, 176–179 (2007)

114. Feng, Y., Zhang, X., Ding, X., Feng, W.: A light-driven reversible conductance switch based on a few-walled carbon nanotube/azobenzene hybrid linked by a flexible spacer. Carbon 48, 3091–3096 (2010)

115. Kymakis, E.: Photovoltaic devices based on carbon nanotubes and related structures. In: Guldi, D.M., Martin, M. (eds.) Carbon Nanotubes and Related Structures, pp. 291–303. Wiley-VCH, Weinheim (2010)

116. Peng, X., Chen, J., Misewich, J.A., Wong, S.S.: Carbon nanotube–nanocrystal heterostruc-tures. Chem. Soc. Rev. 38, 1076–1098 (2009)

117. Sgobba, V., Guildi, D.M.: Carbon nanotube-electronic/electrochemical properties and appli-cations for nanoelectronics and photonics. Chem. Soc. Rev. 38, 165–184 (2009)

118. Yang, Z.P., Ci, L., Bur, J.A., Lin, S.Y., Ajayan, P.M.: Experimental observation of an extremely dark material made by a low-density nanotube array. Nano Lett. 8, 446–451 (2008)

References


119. Tsyboulski, D.A., Rocha, J.D.R., Bachilo, S.M., Cognet, L., Weisman, R.B.: Structure-dependent fluorescence efficiencies of individual single-walled carbon nanotubes. Nano Lett. 7, 3080–3085 (2007)

120. Kazaoui, S., Minami, N., Nalini, B., Kim, Y., Hara, K.: Near-infrared Photoconductive and Photovoltaic devices using Single-Wall carbon Nanotubes in Conductive Polymer films. J. Appl. Phys. 98, 084314 (2005)

121. Bindl, D.J., Safron, N.S., Arnold, M.S.: Dissociating excitons photogenerated in semicon-ducting carbon nanotubes at polymeric photovoltaic heterojunction interfaces. ACS Nano 4, 5657–5664 (2010)

122. Arnold, M.S., Zimmerman, Z.D., Renshaw, C.K., Xu, X., Lunt, R.R., Austin, C.M., Forrest, S.R.: Broad spectral response using carbon nanotube/organic semiconductor/C60 photode-tectors. Nano Lett. 9, 3354–3358 (2009)

123. D’Souza, F., Chitta, R., Sandanayaka, A.S.D., Subbaiyan, N.K., D’Souza, L., Araki, Y., Ito, O.: Supramolecular carbon nanotubefullerene donor–acceptor hybrids for photoinduced electron transfer. J. Am. Chem. Soc. 129, 15865–15871 (2007)

124. Shen, Y., Reparaz, J.S., Wagner, M.R., Hoffmann, A., Thomsen, C., Lee, J.-O., Heeg, S., Hatting, B., Reich, S., Saeki, A., Seki, S., Yoshida, K., Babu, S.S., Möhwald, H., Nakanishi, T.: Assembly of carbon nanotubes and alkylated fullerenes: nanocarbon hybrid towards pho-tovoltaic applications. Chem. Sci. 2, 2243–2250 (2011)

125. Pradhan, B., Setyowati, K., Liu, H., Waldeck, D.H., Chen, J.: Carbon nanotube-polymer nanocomposite infrared sensor. Nano Lett. 8, 1142–1146 (2008)

126. Aliev, A.E.: Bolometric detector on the basis of single-wall carbon nanotubes/polymer com-posite. Infrared Phys. Technol. 51, 541–545 (2008)

127. Tzolov, M.B., Kuo, T.-F., Straus, D.A., Yin, A., Xu, J.: Carbon nanotube silicon heterojunc-tion arrays and infrared photocurrent responses. J. Phys. Chem. C 111, 5800–5804 (2007)

128. Ong, P.-L., Euler, W.B., Levitsky, I.A.: Carbon nanotube-Si diode as a detector of mid-infra-red illumination. Appl. Phys. Lett. 96, 033106 (2010)

129. Kumar, P., Chand, S.: Recent progress and future aspects of organic solar cells. Prog. Photovolt. Res. Appl. 20, 377–415 (2012)

130. Kim, J.Y., Lee, K., Coates, N.E., Moses, D., Nguyen, T.Q., Dante, M., Heeger, A.J.: Efficient tandem polymer solar cells fabricated by all-solution processing. Science 317, 222–225 (2007)

131. Liang, Y., Feng, D., Wu, Y., Tsai, S.T., Li, G., Ray, C., Yu, L.: Highly efficient solar cell pol-ymers developed via fine-tuning of structural and electronic properties. J. Am. Chem. Soc. 131, 7792–7799 (2009)

132. Liang, Y., Xu, Z., Xia, J., Tsai, S.T., Wu, Y., Li, G., Ray, C., Yu, L.: For the bright future-bulk heterojunction polymer solar cells with power conversion efficiency of 7.4%. Adv. Mater. 22, E135–E138 (2010)

133. http://www.m-kagaku.co.jp/english/aboutmcc/RC/special/feature1.html 134. http://www.osa-direct.com/osad-news/667.html 135. http://www.polyera.com/newsflash/polyera-achieves-world-record-organic-solar-cell-

performance 136. Ago, H., Petritsch, K., Shaffer, M.S.P., Windle, A.H., Friend, R.H.: Composites of carbon

nanotubes and conjugated polymers for photovoltaic devices. Adv. Mater. 11(15), 1281–1285 (1999)

137. Kymakis, E., Amaratunga, G.A.J.: Single-wall carbon nanotube/conjugated polymer photo-voltaic devices. Appl. Phys. Lett. 80(1), 112–114 (2002)

138. Landi, B.J., Raffaelle, R.P., Castro, S.L., Bailey, S.G.: Single-wall carbon nanotube-polymer solar cells. Prog. Photovolt. Res. Appl. 13(2), 165–172 (2005)

139. Kazaoui, S., Minami, N., Nalini, B., Kim, Y., Hara, K.: Near-infrared photoconductive and photovoltaic devices using single-wall carbon anotubes in conductive polymer films. J. Appl. Phys. 98(8), 084314 (2005)

140. Xu, Z., Wu, Y., Hu, B., Ivanov, I.N., Geohegan, D.B.: Carbon nanotube effects on electrolumi-nescence and photovoltaic response in conjugated polymers. Appl. Phys. Lett. 87, 263118 (2005)

http://www.m-kagaku.co.jp/english/aboutmcc/RC/special/feature1.html

http://www.osa-direct.com/osad-news/667.html

http://www.polyera.com/newsflash/polyera-achieves-world-record-organic-solar-cell-performance

http://www.polyera.com/newsflash/polyera-achieves-world-record-organic-solar-cell-performance

61

141. Nakayama, K.-I., Asakura, Y., Yokoyama, M.: Photovoltaic device using composite films of poly-mer and carbon nanotubes cut by acid treatment. Mol. Cryst. Liq. Cryst. 424, 217–224 (2004)

142. Kymakis, E., Koudoumas, E., Franghiadakis, I., Amaratunga, G.A.J.: Post-fabrication annealing effects in polymer-nanotube photovoltaic cells. J. Phys. D Appl. Phys. 39, 1058–1062 (2006)

143. Lanzi, M., Paganin, L., Caretti, D.: New photoactive oligo- and Poly-alkylthiophenes. Polymer 49, 4942–4948 (2008)

144. Holt, J.M., Ferguson, A.J., Kopidakis, N., Larsen, B.A., Bult, J., Rumbles, G., Blackburn, J.L.: Prolonging charge separation in P3HT-SWNT composites using highly enriched semi-conducting nanotubes. Nano Lett. 10, 4627–4633 (2010)

145. Ham, M.-H., Paulus, G.L.C., Lee, C.Y., Song, C., Kalantar-zadeh, K., Choi, W., Han, J.-H., Strano, M.S.: Evidence for high-efficiency exciton dissociation at polymer/single-walled carbon nanotube interfaces in planar nano-heterojunction photovoltaics. ACS Nano 4, 6251–6259 (2010)

146. Fuhrer, M.S., Nugard, J., Shih, L., Forero, M., Yoon, Y.G., Mazzoni, M.S., Choi, H.J., Ihm, I., Louie, S.G., Zettl, A., McEuen, P.L.: Crossed nanotube junctions. Science 288, 494–497 (2000)

147. Arranz-Andre, J., Blau, W.J.: Enhanced device performance using different carbon nanotube types in polymer photovoltaic devices. Carbon 46, 2067–2075 (2008)

148. Kanai, Y., Grossman, J.C.: Role of semiconducting and metallic tubes in P3HT/carbon nanotube photovoltaic heterojunction: density functional theory calculations. Nano Lett. 8, 908–912 (2008)

149. Schuettfort, T., Nich, A., Nicholas, R.G.: Observation of type II heterojunction in a highly ordered polymer-carbon nanotube nanohybrid structure. Nano Lett. 9, 3871–4948 (2009)

150. Shaw, P.E., Ruseckas, A., Samuel, I.D.W.: Exciton diffusion measurements in poly(3-hex-ylthiophene). Adv. Mater. 20, 3516–3520 (2008)

151. Brabec, C.J., Cravino, A., Meisssner, D., Sariciftci, N.S., Fromherz, T., Rispen, M.T., Sanchez, L., Hummelen, J.C.: Origin of the open circuit voltage of plastic solar cells. Adv. Funct. Mater. 11, 374–380 (2001)

152. Bernardi, M., Giulilianini, M., Grossman, J.C.: Self-assembly and its impact on interface charge transfer in carbon nanotube/P3HT solar cells. ACS Nano 4, 6599–6606 (2010)

153. Giulianini, M., Waclawik, E., Bell, J.M., Scarselli, M., Castrucci, P., De Crescenzi, M., Motta, N.: Evidence of multiwall carbon nanotube deformation caused bypoly(3-hexylthio-phene) adhesion. J. Phys. Chem. C 115(14), 6324–6330 (2011)

154. Bardeen, J.: Tunneling from a many-particle point of view. Phys. Rev. Lett. 6, 57–59 (1961) 155. Chaudhary, S., Lu, H., Muller, A.M., Bardeen, C.J., Ozkan, M.: Hierarchical placement and

associated optoelectronic impact of carbon nanotubes in polymer-fullerene solar cells. Nano Lett. 7(7), 1973–1979 (2007)

156. Berson, S., De Bettignies, R., Bailly, S., Guillerez, S., Jousselme, B.: Eloboration of P3HT/CNT/PCBM composites for organic photovoltaic cells. Adv. Funct. Mater. 17(16), 3363–3370 (2007)

157. Li, C., Chen, Y., Wang, Y., Iqbal, Z., Chhowalla, M., Mitra, S.: A fullerene-single wall car-bon nanotube complex for polymer bulk heterojunction photovoltaic cells. J. Mater. Chem. 17(23), 2406–2411 (2007)

158. Kymakis, E., Kornilios, N., Koudoumas, E.: Carbon nanotube doping of P3HT: PCBM pho-tovoltaic devices. J. Phys. D Appl. Phys. 41, 165110 (2008)

159. Li, C., Chen, Y., Ntim, S.A., Mitra, S.: Fullerene multiwalled carbon nanotube complexes for bulk heterojunction photovoltaic cells. Appl. Phys. Lett. 96, 143303 (2010)

160. Alley, N.J., Liao, K.-S., Andreoli, E., Dias, S., Dillon, E.P., Orbaek, A.W., Barron, A.R., Byrne, H.J., Curran, S.A.: Effect of carbon nanotube-fullerene hybrid additive on P3HT:PCBM bulk-heterojunction organic photovoltaics. Synth. Met. 162, 95–101 (2012)

161. Nismy, N.A., Adikaari, A.A.D.T., Silva, S.R.P.: Functionalized multiwall carbon nanotubes incorporated polymer/fullerene hybrid photovoltaics. Appl. Phys. Lett. 97, 033105 (2010)

162. Lee, J.U., Gipp, P.P., Heller, C.M.: Carbon nanotube p–n junction diodes. Appl. Phys. Lett. 85, 145 (2004)

References


163. Kong, J., Dai, H.: Full and modulated chemical gating of individual carbon nanotubes by organic amine compounds. J. Phys. Chem. B 105, 2890–2893 (2001)

164. Li, Y.F., Hatakeyama, R., Shishido, J., Kato, T., Kaneko, T.: Air-stable p–n junction diodes based on single-walled carbon nanotubes encapsulating Fe nanoparticles. Appl. Phys. Lett. 90, 173127-1–173127-3 (2007)

165. Abdula, D., Shim, M.: Performance and Photovoltaic response of polymer doped carbon nanotube p–n diodes. ACS Nano 2(10), 2154–2159 (2008)

166. Gabor, N.M., Zhong, Z., Bosnick, K., Park, J., McEuen, P.L.: Extremely efficient multiple elec-tron hole-pair generation in carbon nanotube photodiodes. Science 325, 1367–1371 (2009)

167. Shockley, W., Queisser, H.J.: Detailed balance limit of efficiency of p–n junction solar cells. J. Appl. Phys. 32, 510-1–510-10 (1961)

168. Hanna, M.C., Nozik, A.J.: Solar energy conversion of photovoltaic and photoelctrolysis cells with carrier multiplication absorbers. J. Appl. Phys. 100, 074510-1–074510-8 (2006)

169. Green, M.A., Emery, K., Hishikawa, Y., Warta, W.: Solar cell efficiency tables (version 37). Prog. Photovolt. 19, 84–92 (2011)

170. Nair, G., Chang, L.Y., Geyer, S.M., Bawendi, M.G.: Perspective on the prospects of a carrier multiplication nanocrystal solar cell. Nano Lett. 11, 2145–2151 (2011)

171. McGuire, J.A., Joo, J., Pietryga, J.M., Schaller, R.D., Klimov, V.I.: New aspects of carrier multiplication in semiconductor nanocrystals. Acc. Chem. Res. 41, 1810–1819 (2008)

172. Midgett, A.G., Hillhouse, H.W., Huges, B.K., Nozik, A.J., Beard, M.C.: Flowing versus static conditions for measuring multiple exciton generation in PbSe quantum dots. J. Phys. Chem. C 114, 17486–17500 (2010)

173. Nozik, A.J., Beard, M.C., Luther, J.M., Law, M., Ellingson, R.J., Johnson, J.C.: Semiconductor quantum dots and quantum dot arrays and applications of multiple exciton generation to third generation photovoltaic solar cells. Chem. Rev. 110, 6873–6890 (2010)

174. Semonin, O.E., Luther, J.M., Choi, S., Chen, H.-Y., Gao, J., Nozik, A.J., Beard, M.C.: Peak external photocurrent quantum efficiency exceeding 100 % via quantum dot solar cell. Science 334, 1530–1533 (2011)

175. Perebeinos, V., Avouris, P.: Impact excitation by hot carriers in carbon nanotubes. Phys. Rev. B. 74, 121410 (2006)

176. Baer, R., Rabani, E.: Can impact excitation explain efficient carrier multiplication in carbon nanotube photodiodes? Nano Lett. 10, 3277–3282 (2010)

177. Ueda, A., Matsuda, K., Tayagaki, T., Kanemitsu, Y.: Carrier multiplication in carbon nano-tubes studied by femto-second pump-probe spectroscopy. Appl. Phys. Lett 92, 233105-1–233105-3 (2008)

178. Wang, S., Khafizov, M., Tu, X., Zheng, M., Krauss, T.D.: Multiple exciton generation in single-walled carbon nanotubes. Nano Lett. 10, 2381–2386 (2010)

179. Murray, C.B., Kagan, C.R., Bawendi, M.G.: Synthesis and characterization of monodisperse nanocrystals and close-packed nanocrystal assemblies. Annu. Rev. Mater. Sci. 30, 545–610 (2000)

180. Nelson, J.: The Physics of Solar Cells. Imperial College Press, London (2003) 181. Huynh, W.U., Dittmer, J.J., Alivisatos, A.P.: Hybrid nanorod-polymer solar cells. Science

295, 2425–2427 (2002) 182. Sun, B., Marx, E., Greenham, N.C.: Photovoltaic devices using blends of branched CdSe

nanoparticles and conjugated polymers. Nano Lett. 3, 961–963 (2003) 183. Choi, J.J., Luria, J., Hyun, B.R., Bartnik, A.C., Sun, L.F., Lim, Y.F., Marohn, J.A., Wise,

F.W., Hanrath, T.: Photogenerated exciton dissociation in highly coupled lead salt nanocrys-tal assemblies. Nano Lett. 10, 1805–1811 (2010)

184. Barkhouse, D.A.R., Pattantyus-Abraham, A.G., Levina, L., Sargent, E.H.: Thiols passivate recombinant centers in colloidal quantum dots leading to enhanced photovoltaic device effi-ciency. ACS Nano 2, 2356–2362 (2008)

185. Landi, B.J., Castro, S.L., Ruf, H.J., Evans, C.M., Bailey, S.G., Raffaelle, R.P.: CdSe quan-tum dot-single wall carbon nanotube complexes for polymeric solar cells. Solar Energy Mater. Solar Cells 87, 733–746 (2005)

63

186. Cho, N., Choudhury, K., Thapa, R., Sahoo, Y., Ohulchanskyy, T., Cartwright, A., Lee, K., Prasad, P.N.: Efficient photodetection at IR wavelengths by incorporation of PbSe-carbon nanotube conjugates in a polymeric nanocomposite. Adv. Mater. 19, 232–236 (2007)

187. Shukla, S., Ohulchanskyy, T.Y., Sahoo, Y., Samoc, M., Thapa, R., Cartwright, A.N., Prasad, P.N.: Polymeric nanocomposites involving a physical blend of ir sensitive quantum dots and carbon nanotubes for photodetection. J. Phys. Chem. C 114, 3180–3184 (2010)

188. Leventis, H.C., King, S.P., Sudlow, A., Hill, M.S., Molloy, K.C., Haque, S.A.: Nanostructured hybrid polymer-inorganic solar cell active layers formed by controllable in situ growth of semiconducting sulfide networks. Nano Lett. 10, 1253–1258 (2010)

189. Wang, D., Baral, J.K., Zhao, H., Gonfa, B.A., Truong, V.-V., Khakani, M.A.E., Izquierdo, R., Ma, D.: Controlled fabrication of PbS quantum-dot/carbon-nanotube nanoarchitecture and its significant contribution to near-infrared photon-to current conversion. Adv. Funct. Mater. 21, 4010–4018 (2011)

190. Chen, L., Pan, X., Zheng, D., Gao, Y., Jiang, X., Xu, M., Chen, H.: Hybrid solar cells based on P3HT and Si@MWCNT nanocomposite. Nanotechnology 21, 345201-1–345201-9 (2010)

191. Fushan, L., Son, D.I., Sung Hwan, C., Won Tae, K., Tae Whan, K.: Flexible photo-voltaic cells fabricated utilizing ZnO quantum dot/carbon nanotube heterojunctions. Nanotechnology 20(15), 155202-1–155202-4 (2009)

192. Li, F., Cho, S.H., Son, D.I., Tae Whan, K., Lee, S.-K., Cho, Y.H., Jin, S.: UV photovoltaic cells based on conjugated ZnO quantum dot/multiwalled carbon nanotube heterostructures. Appl. Phys. Lett. 94, 111906-1–111906-3 (2009)

193. Li, F., Son, D.I., Tae Whan, K., Ryu, E., Kim, S.W., Lee, S.K., Cho, Y.H.: Photovoltaic cells fabricated utilizing core-shell CdSe/ZnSe quantum dot/multiwalled carbon nanotube hetero-structures. Appl. Phys. Lett. 95, 061911-1–061911-3 (2009)

194. Feng, Y., Yun, D., Zhang, X., Feng, W.: Solution-processed bulk heterojunction photovol-taic devices based on poly(2-methoxy,5-octoxy)-1,4-phenylenevinylene-multiwalled carbon nanotubes/PbSe quantum dots bilayer. Appl. Phys. Lett. 96, 093301-1–093301-3 (2010)

195. Luther, J.M., Law, M., Song, Q., Perkins, C.L., Beard, M.C., Nozik, A.J.: Structural, opti-cal and electrical properties of self-assembled films of PbSe nanocrystals treated with 1,2-Ethanedithiol. ACS Nano 2, 271–280 (2008)

196. Wise, F.W.: Lead salt quantum dots: the limit of strong confinement. Acc. Chem. Res. 33, 773–780 (2000)

197. Somani, P.R., Somani, P.S., Umeno, M.: Application of metal nanoparticles decorated car-bon nanotubes in photovoltaics. Appl. Phys. Lett. 93, 033315-1–033315-3 (2008)

198. Zhan, Z., Liu, C., Zheng, L., Sun, G., Li, B., Zhang, Q.: Photoresponse of multi-walled carbon nanotube-copper sulfide (MWNT-CuS) hybrid nanostructures. Phys. Chem. Chem. Phys. 13, 20471–20475 (2011)

199. Kalita, G., Adhikari, S., Aryal, H.R., Afre, R., Soga, T., Sharon, M., Koichi, W., Umeno, M.: Silicon nanowire array/polymer hybrid solar cell incorporating carbon nanotubes. J. Phys. D Appl. Phys. 42, 115104 (2009)

200. Guldi, D.M., Rahman, G.M.A., Sgobba, V., Kotov, N.A., Bonifazi, D., Prato, M.: CNT CdTe versatile donor–acceptor nanohybrids. J. Am. Chem. Soc. 128, 2315–2323 (2006)

201. Schultz-Drost, C., Sgobba, V., Gerhard, C., Leubner, S., Calderon, R.M.K., Ruland, A., Guldi, D.M.: Innovative inorganic-organic nanohybrid materials: coupling quantum dots to carbon nanotubes. Angew. Chem. Int. Ed. 49, 6425–6429 (2010)

202. Robel, I., Bunker, B.A., Kamat, P.V.: Single-walled carbon nanotube-CdS nanocomposites as light harvesting assemblies: photoinduced charge transfer interactions. Adv. Mater. 17, 2458–3263 (2005)

203. Farrow, B., Kamat, P.V.: CdSe quantum dot sensitized solar cells. Shuttling electrons through stacked carbon nanocups. J. Am. Chem. Soc. 131, 11124–11131 (2009)

204. Lee, W., Lee, J., Lee, S., Yi, W., Han, S.H., Cho, B.W.: Enhanced charge collection and reduced recombination of CdS/TiO2 quantum-dots sensitized solar cellsin the presence of single-walled carbon nanotubes. Appl. Phys. Lett. 92, 153510-1–153510-3 (2008)

References


205. Sheeney-Haj-Ichia, L., Basnar, B., Willner, I.: Efficient generation of photocurrents by using CdS/carbon nanotube assemblies on electrodes. Angew. Chem. Int. Ed. 44, 78–83 (2005)

206. Banerjee, S., Wong, S.S.: Synthesis and characterization of carbon nanotube–nanocrystal heterostructures. Nano Lett. 2, 195–200 (2002)

207. Lu, C., Akey, A., Wang, W., Herman, I.P.: Versatile formation of CdS nanoparticle-single walled carbon nanotube hybrid structures. J. Am. Chem. Soc. 131, 3446–3447 (2009)

208. Robel, I., Kuno, M., Kamat, P.V.: Size-dependent electron injection from excited CdSe quantum dots into TiO2 nanoparticles. J. Am. Chem. Soc. 129, 4136–4137 (2007)

209. Jeong, S., Garnett, E.C., Wang, S., Yu, Z., Fan, S., Brongersma, M.L., McGehee, M.D., Cui, Y.: Hybrid silicon nanocone-polymer solar cells. Nano Lett. 12, 2971–2976 (2012)

210. Gowrishankar, V., Scully, S.R., Chan, A.T., McGehee, M.D., Wang, Q., Branz, H.M.: Exciton harvesting, charge transfer, and charge carrier transport in amorphous-silicon nano-pillar/polymer hybrid solar cells. J. Appl. Phys. 103, 064511 (2008)

211. Peng, K.-Q., Wang, X., Wu, X.-L., Lee, S.-T.: Platinum nanoparticle decorated silicon nanowires for efficient solar energy conversion. Nano Lett. 9, 3704–3709 (2009)

212. Shiu, S.-C., Chao, J.J., Hung, S.-C., Yeh, C.-L., Lin, C.-F.: Morphology dependence of sili-con nanowire/Poly (3,4 ethylenedioxy thiophene): poly(styrenesulfonate) heterojunction solar cells. Chem. Mater. 22, 3108–3113 (2010)

213. Ozdemir, B., Kulakci, M., Turan, R., Unalan, H.E.: Silicon Nanowire- poly (3,4 ethylen-edioxy thiophene)- poly(styrenesulfonate) heterojunction solar cells. Appl. Phys. Lett. 99, 113510 (2011)

214. Lu, W., Wang, C., Yue, W., Chen, L.: Si/PEDOT: PSS core/shell nanowire arrays for effi-cient hybrid solar cells. Nanoscale 3, 3631–3634 (2011)

215. Kalita, G., Adhikari, S., Aryal, H.R., Afre, R., Soga, T., Sharon, M., Koichi, W., Umeno, M.: Silicon nanowire array/polymer hybrid solar cell incorporating carbon nanotubes. J. Phys. D Appl. Phys. 42, 115104 (2009)

216. He, L., Jiang, C., Rusli, Lai, D., Wang, H.: Highly efficient Si-nanorods/organic hybrid core sheath heterojunction solar cells. Appl. Phys. Lett. 99, 021104 (2011)

217. Shen, X., Sun, B., Liu, D., Lee, S.-T.: Hybrid heterojunction solar cell based on organic–inorganic silicon nanowire array architecture. J. Am. Chem. Soc. 133, 19408–19415 (2011)

218. Svrcek, V., Mariotti, D., Shibata, Y., Kondo, M.: A hybrid heterojunction based on fullerenes and surfactant-free, self-assembled, closely packed silicon nanocrystals. J. Phys. D Appl. Phys. 43, 415402 (2010)

219. Dietmueller, R., Stegner, A.R., Lechner, R., Niesar, S., Pereira, R.N., Brandt, M.S., Ebbers, A., Trocha, M., Wiggers, H., Stutzmann, M.: Light-induced charge transfer in hybrid composites of organic semiconductors and silicon nanocrystals. Appl. Phys. Lett. 94, 113301 (2009)

220. Liu, C.-Y., Holman, Z.C., Kortshagen, U.R.: Optimization of Si NC/P3HT hybrid solar cells. Adv. Funct. Mater. 20, 2157–2164 (2010)

221. Chen, L., Pan, X., Zheng, D., Gao, Y., Jiang, X., Xu, M., Chen, H.: Hybrid solar cells based on P3HT and Si@MWCNT nanocomposite. Nanotechnology 21, 345201 (2010)

222. Eisenhawer, B., Sensfuss, S., Sivakov, V., Pietsch, M., Andrae, G., Falk, F.: Increasing the efficiency of polymer solar cells by silicon nanowires. Nanotechnology 22, 315401 (2011)

223. Levitsky, I.A., Euler, W.B., Tokranova, N., Xu, B., Castracane, J.: Hybrid solar cells based on porous Si and copper phthalocyanine derivatives. Appl. Phys. Lett. 85, 6245–6247 (2004)

224. Tokranova, N., Levitsky, I.A., Xu, B., Castracane, J., Euler, W.B.: Hybrid solar cells based on organic material embedded into porous silicon. In: Proceedings of the SPIE: Organic Photonic Materials and Devices VII, San Jose, 2005, pp. 183–190

225. Prabakaran, R., Fortunato, E., Martins, R., Ferreira, I.: Fabrication and characterization of hybrid solar cells based on copper phthalocyanine/porous silicon. J. Non-Cryst. Solids 354, 2892–2896 (2008)

226. Nahor, A., Berger, O., Bardavid, Y., Toker, G., Tamar, Y., Reiss, L., Asscher, M., Yitzchaik, S., Sa’ar, A.: Hybrid structures of porous silicon and conjugated polymers for photovoltaic applications Phys. Stat. Sol. C 8, 1908–1912 (2011)

65

227. Ong, P.-L., Levitsky, I.A.: Organic/IV, III–V semiconductor hybrid solar cells. Energies 3, 313–334 (2010)

228. van de Lagemaat, J., Bames, T.M., Rumbles, G., Shaheen, S.E., Coutts, T.J., Weeks, C., Levitsky, I.A., Peltola, J., Glatkowski, P.: Organic solar cells with carbon nanotubes replac-ing In2O3:Sn as the transparent electrode. Appl. Phys. Lett. 88, 233503 (2006)

229. Xu, H., Anlage, S.M., Hu, L., Gruner, G.: Microwave shielding of transparent and conduct-ing single-walled carbon nanotube films. Appl. Phys. Lett. 90, 183119 (2007)

230. Li, Z., Kandel, H.R., Dervishi, E., Saini, V., Xu, Y., Biris, A.R., Lupu, D., Salamo, G.J., Biris, A.S.: Compararive study on different carbon nanotube materials in terms of transpar-ent conductive coatings. Langmuir 24, 2655–2662 (2008)

231. De, S., Lyons, P.E., Sorel, S., Doherty, E.M., King, P.J., Blau, W.J., Nirmalraj, P.N., Boland, J.J., Scardaci, V., Joimel, J., Coleman, J.N.: Transperent, flexible, and highly conductive thinfilms based on polymer-nanotube composites. ACS Nano 3, 714–720 (2009)

232. Parekh, B.B., Fanchini, G., Eda, G., Chhowalla, M.: Improved conductivity of transparent single-wall carbon nanotube thin films via stable post deposition functionalization. Appl. Phys. Lett. 90, 121913 (2007)

233. Wei, J., Jia, Y., Shu, Q., Gu, Z., Wang, K., Zhuang, D., Zhang, G., Wang, Z., Luo, J., Cao, A., Wu, D.: Double-walled carbon nanotube solar cells. Nano Lett. 7, 2317–2321 (2007)

234. Jia, Y., Li, P., Gui, X., Wei, J., Wang, K., Zhu, H., Wu, D., Zhang, L., Cao, A., Xu, Y.: Encapsulated carbon nanotube-oxide-silicon solar cells with stable 10 % efficiency. Appl. Phys. Lett. 98, 133115 (2011)

235. Saini, V., Li, Z., Bourdo, S., Kunets, V.P., Trigwell, S., Couraud, A., Rioux, J., Boyer, C., Nteziyaremye, V., Dervishi, E., Biris, A.R., Salamo, G.J., Viswanathan, T., Biris, A.S.: Photovoltaic devices based on high density boron doped single-walled carbon nanotube/n-Si heterojunctions. J. Appl. Phys. 109, 014321 (2011)

236. Li, Z., Saini, V., Dervishi, E., Kunets, V.P., Zhang, J., Xu, Y., Biris, A.R., Salamo, G.J., Biris, A.S.: Polymer functionalized n-type single wall carbon nanotube photovoltaic devices. Appl. Phys. Lett. 96, 033110 (2010)

237. Behnam, A., Johnson, J.L., Choi, Y., Ertosun, M.G., Okyay, A.K., Kapur, P., Saraswat, K.C., Ural, A.: Experimental characterization of single-walled carbon nanotube film-Si Schottky contacts using metal-semiconductor-metal structures. Appl. Phys. Lett. 92, 243116 (2008)

238. Hatakeyama, R., Li, Y.F., Kata, T.Y., Kaneko, T.: Infrared photovoltaic solar cells based on C60 fullerene encapsulated single-walled carbon nanotubes. Appl. Phys. Lett. 97, 013104 (2010)

239. Wadhwa, P., Seol, G., Petterson, M.K., Guo, J., Rinzler, A.G.: Electrolyte-induced inversion layer Schottky junction solar cells. Nano Lett. 11, 2419–2423 (2011)

240. Sze, S.M.: Physics of Semiconductor Devices, 2nd edn. Wiley, New York (1981) 241. Borgne, V.L., Castrucci, P., Gobbo, S.D., Scarselli, M., Crescenzi, M.D., Mohamedi, M.,

Khakani, M.A.E.: Enhanced photocurrent generation from UV-laser synthesized-single-wall-carbon-nanotubes/n-silicon hybrid planar devices. Appl. Phys. Lett. 97, 193105 (2010)

242. Tune, D.D., Flavel, B.S., Krupke, R., Shapter, J.G.: Carbon nanotube-silicon solar cells. Adv. Energy Mater. (2012). doi:10.1002/aenm.201200249

243. Rakitin, A., Papadopoulos, C., Xu, J.M.: Electronic properties of amorphous carbon nano-tubes. Phys. Rev. B 61, 5793–5796 (2000)

244. Tzolov, M.B., Juo, T.-F., Yin, A., Straus, D.A., Xu, J.M.: Carbon nanotube-silicon het-erojunction arrays and infrared photocurrent responses. J. Phys. Chem. C 111, 5800–5804 (2007)

245. Kuo, T.F., Straus, D.A., Tzolov, M., Xu, J.: Carbon nanotube—silicon heterojunction hyper-spectral photocurrent response. Mater. Res. Soc. Symp. Proc. 963, 0963–Q22-01 (2007)

246. Ong, P.-L., Euler, W.B., Levitsky, I.A.: Carbon nanotube-Si diode as a detector of mid-infra-red illumination. Appl. Phys. Lett. 96, 033106 (2010)

247. Liang, C.-W., Roth, S.: Electrical and optical transport of GaAs/carbon nanotube hetero-junctions. Nano Lett. 8, 1809–1812 (2008)

References

http://dx.doi.org/10.1002/aenm.201200249


248. Li, H., Loke, W.K., Zhang, Q., Yoon, S.F.: Physical device modeling of carbon nanotube/GaAs photovoltaic cells. Appl. Phys. Lett. 96, 043501 (2010)

249. Behnam, A., Johnson, J.L., Choi, Y., Noriega, L., Ertosun, M.G., Wu, Z., Rinzler, A.G., Kapur, P., Saraswat, K.C., Ural, A.: Metal-semiconductor-metal photodetectors based on single-walled carbon nanotube film GaAs Schottky contacts. J. Appl. Phys. 103, 114315-1–114315-6 (2008)

250. Schriver, M., Reganb, W., Loster, M., Zettl, A.: Carbon nanostructure a Si:H photovoltaic cells with high open-circuit voltage fabricated without dopants. Solid State Commun. 150, 561–563 (2010)

251. Del Gobbo, S., Castrucci, P., Scarselli, M., Camilli, L., De Crescenzi, M., Mariucci, L., Valletta, A., Minotti, A., Fortunato, G.: Carbon nanotube semitransparent electrodes for amorphous silicon based photovoltaic devices. Appl. Phys. Lett. 98, 183113 (2011)

252. Zhou, H., Colli, A., Ahnood, A., Yang, Y., Rupesinghe, N., Butler, T., Haneef, I., Hiralal, P., Nathan, A., Amaratunga, G.A.J.: Arrays of parallel connected coaxial multiwall-carbon nanotube-amorphous-silicon solar cells. Adv. Mater. 21, 3919–3923 (2009)

253. Shu, Q., Wei, J., Wang, K., Zhu, H., Li, Z., Jia, Y., Gui, X., Guo, N., Li, X., Ma, C., Wu, D.: Hybrid heterojunction and photoelectrochemistry solar cell based on silicon nanowires and double-walled carbon nanotubes. Nano Lett. 9, 4338–4342 (2009)

254. Shu, Q., Wei, J., Wang, K., Song, S., Guo, N., Jia, Y., Li, Z., Xu, Y., Cao, A., Zhu, H., Wu, D.: Efficient energy conversion of nanotube/nanowire based solar cells. Chem. Commun. 46, 5533–5535 (2010)

255. Zhang, L., Jia, Y., Wang, S., Li, Z., Ji, C., Wei, J., Zhu, H., Wang, K., Wu, D., Shi, E., Fang, Y., Cao, A.: Carbon nanotube and CdSe nanobelt Schottky junction solar cells. Nano Lett. 10, 3583–3589 (2010)

256. Shi, E., Nie, J., Qin, X., Li, Z., Zhang, L., Li, Z., Li, P., Jia, Y., Ji, C., Wei, J., Wang, K., Zhu, H., Wu, D., Li, Y., Fang, Y., Qian, W., Wei, F., Cao, A.: Nanobelt-carbon nanotube cross-junction solar cells. Energy Environ. Sci. 5, 6119–6125 (2012)

257. Xie, J., Inaba, T., Sugiyama, R., Homma, Y.: Intrinsic diffusion length of excitons in long single-walled carbon nanotubes from photoluminescence spectra. Phys. Rev. B 85, 085434/1–085434/6 (2012)

258. Liu, T., Xiao, Z.: Exact and closed form solutions for the quantum yield, exciton diffusion length, and lifetime to reveal the universal behaviors of the photoluminescence of defective single walled carbon nanotubes. J. Phys. Chem. C 115, 16920–16927 (2011)

259. Pei, Q., Yu, G., Zhang, C., Yang, Y., Heeger, A.J.: Polymer light-emitting electrochemical cells. Science 269, 1086–1088 (1995)

260. Gu, Q., Wang, Q., Schiff, E.A., Li, Y.M., Malone, C.T.: Hole drift measurements in amor-phous silicon-carbon alloys. J. Appl. Phys. 76, 2310/1–2310/6 (1994)

261. Schiff, E.A., Devlen, R.I., Grahn, H.T., Tauc, J., Guha, S.: Picosecond electron drift mobility measurements in hydrogenated amorphous silicon. Appl. Phys. Lett. 54, 1911/1–1911/3 (1989)

262. Juska, G., Arlauskas, K., Kocka, J., Hoheisel, M., Chabloz, P.: Hot electrons in amorphous silicon. Phys. Rev. Lett. 75, 2984–2987 (1995)

263. Minot, M.J.: Single layer, Gradient refractive index antireflection films effective from 0.35 to 2.5 μ. J. Opt. Soc. Am. 66, 515–519 (1976)

264. Chen, M., Chang, H.-C., Chang, A.S.P., Lin, S.Y., Xi, J.Q., Schubert, E.F.: Design of optical path for a wide-angle gradient-index antireflection coatings. Appl. Opt. 46, 6533–6538 (2007)

265. Staebler, D.L., Wronski, C.R.: Optically induced conductivity changes in discharge pro-duced hydrogenated amorphous silicon. J. Appl. Phys. 51, 3262/1–3262/7 (1980)

266. Bustarret, E., Ligeon, M., Ortega, L.: Visible light emission at room temperature from par-tially oxidized amorphous silicon. Solid State Commun. 83, 461–464 (1992)

267. Bustarret, E., Ligeon, M., Rosenbauer, M.: Anodized amorphous silicon: present status. Phys. Status Solidi B 190, 111–116 (1995)

268. Estes, M.J., Hirsch, L.R., Wichart, S., Moddel, G., Williamson, D.L.: Visible photolumines-cence from porous a-Si:H and porous a-Si:C:H thin films. J. Appl. Phys. 82, 1832/1–1832/9 (1997)

67

269. He, M., Chernov, A.I., Fedotov, P.V., Obraztsova, E.D., Sainio, J., Rikkinen, E., Jiang, H., Zhu, Z., Tian, Y., Kauppinen, E.I., Niemelä, M., Krause, A.O.I.: Predominant(6,5) single walled carbon nanotube growth on a copper-promoted iron catalyst. J. Am. Chem. Soc. 132, 13994–13996 (2010)

270. Kato, T., Hatakeyama, R.: Direct growth of short single-walled carbon nanotubes with nar-row chirality distribution by time-programmed plasma chemical vapor deposition. ACS Nano 4, 7395–7400 (2010)

271. Vasenkov, A.V.: Modeling nanotube synthesis: the multiscale approach. In: Nalwa, H.S. (ed.) Encyclopedia of Nanoscience and Nanotechnology, 2nd edn, pp. 281–302. American Scientific Publishers, Valencia (2011)

272. Razykov, T.M., Ferekides, C.S., Morel, D., Stefanakos, E., Ullal, H.S., Upadhyaya, H.M.: Solar photovoltaic electricity: current status and future prospects. Sol. Energy 85, 1580–1608 (2011)

273. Rogalski, A.: Infrared Detectors, Electrocomponent Science Monographs. Gordon and Breach Science Publishers, Amsterdam (2000)

274. Lu, R., Li, Z., Xu, G., Wu, J.Z.: Suspending single wall carbon nanotube thin film infrared bolometers on microchannels. Appl. Phys. Lett. 94, 163110 (2009)

275. Lu, R., Xu, G., Wu, J.Z.: Effects of thermal annealing on noise property and temperature coefficient of resistance of single-walled carbon nanotube films. Appl. Phys. Lett. 93, 213101 (2008)

276. Lu, R., Shi, J.J., Javier Baca, F., Wu, J.Z.: High performance multiwall carbon nanotube bolometers. J. Appl. Phys. 108, 084305/1–084305/5 (2010)

277. Xiao, L., Zhang, Y., Wang, Y., Liu, K., Wang, Z., Li, T., Jiang, Z., Shi, J., Liu, L., Li, Q.Q., Zhao, Y., Feng, Z., Fan, S., Jiang, K.: A polarized infrared thermal detector made from super-aligned multiwalled carbon nanotube films. Nanotechnology 22, 025502 (2011)

278. Gohier, A., Dhar, A., Gorintin, L., Bondavalli, P., Bonnassieux, Y., Cojocaru, C.S.: All-printed infrared sensor based on multiwalled carbon nanotubes. Appl. Phys. Lett. 98, 063103/1–063103/3 (2011)

279. Tarasov, M., Svensson, J., Kuzmin, L., Campbell, E.E.B.: Carbon nanotube bolometers. Appl. Phys. Lett. 90, 163503 (2007)

280. Pradhan, B., Setyowati, K., Liu, H.Y., Waldeck, D.H., Chen, J.: Carbon nanotube-polymer nanocomposite infrared sensor. Nano Lett. 8, 1142–1146 (2008)

281. Pradhan, B., Kohlmeyera, R.R., Setyowatia, K., Owen, H.A., Chen, J.: Advanced carbon nanotube/polymer composite infrared sensors. Carbon 47, 1686–1692 (2009)

282. Aliev, A.E.: Bolometric detector on the basis of single-wall carbon nanotube/polymer com-posite. Infrared Phys. Technol. 51, 541–545 (2008)

283. Gustavo, V.-R., Trevor, J.S., Mariela, B.-S., Vidal, M.A., Hugo, N.-C., Gonzalez, F.J.: High-sensitivity bolometers from self-oriented single-walled carbon nanotube composites. ACS Appl. Mater. Interf. 3, 3200–3204 (2011)

284. Glamazda, A.Y., Karachevtsev, V.A., Euler, W.B., Levitsky, I.A.: Achieving high mid-IR bolometric responsivity for anisotropic composite materials from carbon nanotubes and pol-ymers. Adv. Funct. Mater. 22, 2177–2186 (2012)

285. Bang, D., Lee, J., Park, J., Choi, J., Chang, Y.W., Yoo, K.-H., Huh, Y.-M., Haam, S.: Effectively enhanced sensitivity of a polyaniline-carbon nanotube composite thin film bolo-metric near-infrared sensor. J. Mater. Chem. 22, 3215–3219 (2012)

286. St-Antoine, B.C., Menard, D., Martel, R.: Position sensitive photothermoelectric effect in suspended single-walled carbon nanotube films. Nano Lett. 9, 3503–3508 (2009)

287. Omari, M., Kouklin, N.A.: Photothermovoltaic effect in carbon nanotubes: En route toward junctionless infrared photocells and light sensors. Appl. Phys. Lett. 98, 243113–243116 (2011)

288. Sheng, P., Sichel, E.K., Gittleman, J.L.: Fluctuation-induced tunneling conduction in car-bon-poly(vinyl chloride) composite. Phys. Rev. Lett. 40, 1197–1200 (1978)

289. Sichel, E.K., Gittleman, J.L., Sheng, P.: Transport properties of the composite material car-bon-polyvinylchloride. Phys. Rev. B 18, 5712–5716 (1978)

References


290. Zhou, W., Vavro, J., Guthy, C., Winey, K.I., Fischer, J.E., Ericson, L.M., Ramesh, S., Saini, R., Davis, V.A., Kittrel, C., Pasqualli, M., Hauge, R.H., Smalley, R.E.: Single wall carbon nanotube fibers extruded from super-acid suspensions: preferred orientation, electrical, and thermal transport. J. Appl. Phys. 95, 649–655 (2004)

291. Kim, G.T., Jhang, S.H., Park, J.G., Park, Y.W., Roth, S.: Non-ohmic current-voltage charac-teristics in single-wall carbon nanotube networks. Synth. Met. 117, 123–126 (2001)

69

2.1 Introduction

The direct conversion of light energy into to mechanical motion, the optom-echanical effect, is also termed photostriction or photoactuation. Photostrictive materials are scarce but the technology is of significant interest with applica-tions ranging from civil infrastructures to medicine to the aerospace industry to telecommunications. These types of smart materials are expected to be able to produce a variety of devices and are envisioned to be used as light-driven motors, for energy scavenging, artificial muscles, and tactile displays. The development goals for photoactuators include large range of motion, high preci-sion, high speed, high strain energy density, and low fatigue rate. Ideally, these types of materials could be used in both microelectromechanical and nanoelec-tromechanical systems.

Carbon nanotubes (CNTs) have been use to construct electrically driven actuators using a variety of structures [1–7]. Generally, in the electrically driven systems the charge separation at or near the electrodes effect the biggest component of the volume change. There has been a flurry of theoretical activ-ity to understand the mechanical properties of CNTs [8–19]. However, none of these reports considered the effect of light on the stress/strain behavior of CNTs.

CNTs can be observed to photoactuate under a variety of motifs, including freestanding bundles [20, 21], pure CNT films [22], mixed into polymer hosts [23–27], and as part of layered composites [28–33]. While the magnitude of the volume response under exposure to light is remarkably large in all cases, the exact composition and preparation method used to make the photoactuator appears to have an influence on the underlying mechanism.

Chapter 2Use of Carbon Nanotubes in Photoactuating Composites


70 2 Use of Carbon Nanotubes in Photoactuating Composites

2.2 Carbon Nanotube Bundles and Freestanding Films

2.2.1 Freestanding Bundles

Zhang and Iijima [20] were the first to report photoactuation in carbon nanotubes in 1999. Laser ablation of a graphite target (impregnated with metal catalysts) gave single-walled carbon nanotubes deposited onto Kovar electrodes. The SWNTs deposited as bundles consisting of filaments with lengths up to 1 mm and diame-ters of about 10 μm. The filaments extended between the two electrodes and could be affixed to one electrode or to both electrodes, depending upon the experiment.

Using a white light (halogen lamp) with an intensity of about 20 mW/cm2 the tip displacement of a ~1 mm long, 10 μm diameter filament could be as high as 170 μm with the time frame for full response being on the order of 100 ms. For thicker filaments, 50–100 μm, the bending displacement reached as high as 1.3 mm. The magnitude of the displacement was found to depend upon the inten-sity of the light in the visible region but not upon the wavelength. This latter con-clusion was based upon a comparison between the white light of the halogen source with different color filters and a single laser wavelength, 633 nm from a He–Ne laser. No investigation of using light in the near infrared region was reported.

Figure 2.1 shows the behavior of one of these filaments under several differ-ent illumination conditions. Under high intensity illumination, Fig. 2.1a (white light, ~30 mW/cm2) and c (633 nm light, ~800 mW/cm2), the tip of the SWNT filament formed a loop, as denoted by the arrows, and the mechanical motion was independent of the color of the light. When the intensity of the light is reduced to ~5 mW/cm2, Fig. 2.1b (white light), a loop no longer forms. Rather, there is simply bending of the tip of the filament. For comparison purposes, Fig. 2.1d, the same filament was exposed to an electric field in the dark and a loop similar to those found under strong illumination was found.

Zhang and Iijima attributed the photoactuation in the SWNT bundles to an elec-trostatic effect. They arrived at this conclusion primarily because of the similar mechanical deformation observed under light or an electric field. A photon pres-sure effect was eliminated since the bending did not always match the orientation of the light source. Thermal expansion was thought to be unlikely because of the large magnitude of the displacements observed. The proposed mechanism was that the light induced a photocurrent in the nanotubes. Since the filaments are com-posed of nonuniform bundles of SWNTs, a local charge accumulation could result, especially at bends in the filaments. The local charge then creates an electrostatic field that leads to repulsion between different nanotubes within the filament. Depending upon the diameter of the filament and the amount of light absorbed, different directions of bending could be observed.

Cronin et al. [21] measured the Raman spectra of individual SWNTs and were able to deduce certain characteristics of the change in shape of the nanotubes under irradiation. SWNTs were grown suspended across a trench and the Raman spectrum measured with a tunable laser at ~1 mW power. The Stokes/anti-Stokes

71

ratio indicated that the laser caused minimal temperature change in each sample. The results implied a strong electron–phonon coupling and that radial expansion, not axial elongation, was responsible for the observed results. Although no pho-toactuation was directly observed, the results imply that the radial expansion of the SWNTs can be significant.

2.2.2 Freestanding Films

Ahir et al. [22] prepared free standing films of both MWNTs and SWNTs. The nanotubes were dispersed in isopropyl alcohol, ~0.1 mg/mL, using ultrasonica-tion but no surfactant. The dispersion was vacuum filtered through a filter com-posed of a mixed cellulose ester. The thickness of the SWNT layer was controlled

Fig. 2.1 A SWNT filament under different illumination conditions: a 30 mW/cm2 white light source (halogen lamp); b 5 mW/cm2 white light c 800 mW/cm2 He–Ne laser (633 nm); d no illumination but in an electric field. Reprinted Fig. 2.3 with permission from Ref. [20]. Copyright 1999 by the American Physical Society



by varying the concentration of the nanotubes in the dispersion. The filter mem-brane containing the nanotube layer was rinsed and dried and the resulting films could be peeled off of the filter membrane. The photoactuation experiments were done using a LCD light source with maximum output at 675 nm and a breadth ranging from ~500 to ~800 nm. The sample was uniformly illuminated and the incident intensity was ~1.5 mW/cm2 at 675 nm. The samples were clamped to supports placed into a thermally controlled housing. A dynamometer was used to exert force on the sample and the length of the sample could be measured to 1 μm precision.

The photoactuation of a MWNT film is shown in Fig. 2.2. When the light is turned on the stress drops, indicating that the sample is expanding, i.e., experi-encing tension. The time constant for the photoactuation is on the order of 10 s of minutes. When the light is turned off, the sample returns to its original state, exhibiting a fully reversible response. However, at very short times the initial pho-toactuation is an increase in stress, i.e. compression. This initial event is complete in seconds. The same two directional response is observed when the light is turned off (not shown in the figure): an initial drop in stress for a few seconds and then an increase in stress until the initial state is recovered.

The response of the SWNT film is completely different, as shown in Fig. 2.3. Upon exposure to the light, the SWNT film shows an increase in stress, i.e., com-pression. The time constant is fast, on the order of seconds. The SWNT films exhibit a slight fade in the stress after reaching the maximum value while the light is on. When the light is turned off, the film rapidly relaxes to its initial state, again with a time constant on the order of seconds. The SWNT films do not show the small initial feature that was observed for the MWNT films. Independent experi-ments using a standard heating source showed smaller stress changes, indicating that the observed response is not a simple heating effect.

Fig. 2.2 Stress response of a MWNT film under 675 nm light. The inset shows an expansion of the first few seconds of the measurement in the light-on regime. Reprinted Fig. 2.6 with permission from Ref. [22]. Copyright 2007 by the American Physical Society

73

The authors explained the differences between the MWNT and SWMT free-standing films based on the network behavior of each type of nanotube. The MWNT were envisioned as a granular network. In contrast, the SWNT was described to be more like a cross-linked polymer network.

2.3 Carbon Nanotubes in Mixed Composites

2.3.1 Rubbery Polymer Host Materials

In a series of papers, Terentjev et al. [23–26] examined the photoactuation of car-bon nanotubes mixed into rubbery polymers. Most of their research was done using poly (dimethylsiloxane) (PDMS) but they also reported studies in poly (sty-rene–isoprene-styrene) triblock polymer (SIS), and monodomain and polydomain liquid crystal elastomers. The nanotubes were multiwalled and were thoroughly mixed into the rubber host prior to polymerization. Great care was taken to insure that the MWNTs were completely dispersed in the host system. Loadings of the MWNTs ranged from 0 to 7 wt%. Samples loaded with 3 wt% carbon black were also prepared for comparison. Fibers were drawn from the cured polymer and samples were cut into cantilevers of dimensions 30 mm long × 1.5 mm wide × 0.2 mm thick. Samples were pre-stretched to a determined amount of strain prior to measuring the photoactuation, with pre-strain values ranging from 2 to 40 % (ε = 0.02 − 0.40).

The photoactuation experiments were done using a LCD light source with max-imum output at 675 nm and a breadth ranging from ~500 to ~800 nm. The sample was uniformly illuminated and the incident intensity was ~1.5 mW/cm2 at 675 nm.

Fig. 2.3 Stress response of a SWNT film under 675 nm light. The inset shows an expansion of the first few seconds of the measurement in the light-on regime. Reprinted Fig. 2.7 with permission from Ref. [22]. Copyright 2007 by the American Physical Society



The samples were clamped to supports placed into a thermally controlled housing. A dynamometer was used to exert force on the sample and the length of the sam-ple could be measured to 1 μm precision. Absorption measurements of the host material showed that ~30 % of the incident light was absorbed or scattered but even low loadings of MWNTs (0.3 wt%) caused more than 97 % of the light to be absorbed. The spectrum of the MWNTs in the host polymers was featureless, only displaying a slight rise from 1000 to 300 nm. The composites were pre-strained in the apparatus to a determined value, allowed to relax for 10 min, and then stress measurements were taken. Thermocouples were placed on both sides of the sam-ple to determine the temperature change across the sample during illumination.

Figure 2.4 shows the results of the measured stress of a PDMS sample as a function of time when the light is turned on. There is a strong dependence of the

Fig. 2.4 Stress response of a PDMS sample containing 3 wt% MWNTs. The different curves show the photoactuation with different pre-strain levels. The light is turned on at t = 60 s. Reprinted Fig. 2.8 with permission from Ref. [25]. Copyright 2006 by the American Physical Society

Fig. 2.5 A comparison of the normalized stress response (dashed line, left axis) and the normalized temperature response (solid line, right axis) for a PDMS sample containing 3 wt% MWNTs and a pre-strain of 20 % upon exposure to light. Reprinted Fig. 2.9 with permission from Ref. [25]. Copyright 2006 by the American Physical Society

75

photoresponse on the amount of pre-strain induced in the sample. At low pre-strain the stress a negative stress is measured, indicative of tension in the sample. At higher pre-strains the stress reverses to a positive value, meaning that the sample experiences compression during photoactuation. This type of behavior was observed for all hosts and the crossover point from tension to compression was found to be for pre-strains of 6–10 %. The time to reach maximum strain is 10–15 s for all samples. The time dependence was fit to a compressed exponential of the form 1 − exp(–(t/τ)β) where the time constant, τ, is ~5 s and the exponent, β, is ~2 were constant at sufficiently high loadings of MWNTs. The response was slower when the concentration of MWNTs was below the percolation threshold with τ ~10 s and β ~1. The decay of the photoinduced stress (not shown) follows a simple exponential function, exp(–t/τ), with the relaxation time, τ, ~5 s.

To understand the relationship between the photoactuation and the temperature change, a direct measurement of the temperature change across the sample was made during exposure to light. While the temperature change depended upon the loading of the MWNTs, typical values were ΔT ~15 °C. A plot of the normalized temperature change and the normalized stress change as a function of exposure time is shown in Fig. 2.5. The difference in the response of the two measurements is striking. As discussed above, upon light exposure the stress reaches is maximum value in 10–15 s. In contrast, the thermal response is ~10× slower, taking 100–150 s to reach the equilibrium temperature. This indicates that the stress response is not primarily caused by simple heating effects. To confirm this, the authors measured the stress when samples were raised to the same temperature by a mica-insulated heater. Under these conditions the stress response was 10 times smaller than when exposed to light.

Figure 2.6 shows the change in stress as a function of loading and applied pre-strain for PDMS as the host. Remarkably, the transition from tension to compression is essentially constant for all MWNT loadings. Figure 2.6 also shows the strain calculated using the observed stress and Young’s modulus.

Fig. 2.6 A plot of the maximum change in stress under illumination for different loading so MWNTs in PDMS as a function of applied pre-strain. The left axis shows the photoactuation stress and the right axis shows the calculated change in length of the sample found from the stress change and the Young’s modulus. Reprinted Fig. 2.10 with permission from Ref. [25]. Copyright 2006 by the American Physical Society



The photoinduced length change is large, reaching values as high as 10 %. Control experiments using no nanotubes or substituting carbon black for the nanotubes showed no photoactuation. This demonstrates that the primary contributor to the effect arises from the properties of the nanotubes.

A model was developed to explain the magnitude of the photoactuation and the effect of the pre-strain. The distribution of the orientations of the MWNTs in the host polymer were described by an orientation parameter that depends upon the amount of pre-strain. A strain factor, Δ = Rz(hν)/Rz(0), is defined where R is the effective length of the nanotube in the host. Based on the experimental observa-tion that the photoactuation is compressive at high loadings, Δ < 1, and Δ should be proportional to the light intensity (although this was not tested experimentally). A strain tensor can be written in terms of Δ, with terms as shown in Fig. 2.7, relative to the average strain, 〈λz〉, shown along the macroscopic z axis. After accounting for the orientational averaging, the calculated strain was found to be

Setting the strain to 0 allows determination of the crossover pre-strain

�λz� ≈1

3

(

∆ +2

√∆

)

−2

5ε

(

1√

∆− ∆

)

ε∗ ≈

5

(

2 −√

∆ − ∆

)

6

(

1 +√

∆ + ∆

)

Fig. 2.7 Model of a carbon nanotube in the host matrix and the physical parameters used in the model: Δ, the linear contraction from the photoactuation, and λz, the local strain projected on the macroscopic axis. Reprinted Fig. 2.13 with permission from Ref. [25]. Copyright 2006 by the American Physical Society

77

At the observed crossover point of ε* ~0.1, Δ is found to be 0.8. Calculation of the length change as a function of pre-strain using the value found for ε* (which is used in the orientational averaging) matched the experimental data well, confirming the essential physics of the model.

2.3.2 Hydrogel Host Materials

A recent paper published by Zhang et al. [27] used poly(N-isopropylacrylamide) (pNI-PAM) as the host for SWNTs. In contrast to the rubbers used by the Terentjev group, pNIPAM forms a hydrogel that is thermally responsive. In these hydrogels there is a phase transition between hydrophilic and hydrophobic states that is accompanied by a significant volume change. At lower temperatures pNIPAM is hydrophilic and contains water in the structure. Upon heating to above the transition temperature, ~32–33 °C, the conversion to the hydrophobic state drives the water from the hydrogel, which is responsible for the volume change. Addition of carbon nanotubes allows modification of the hydrogel properties and a chromophore to absorb light.

Samples were prepared by dispersing SWNT, various concentrations up to 1 mg/mL, into water using sodium deoxycholate (chosen to be compatible with pNI-PAM) as a surfactant. These were ultrasonicated for an hour, which lead to SWNTs primarily with open ends. The SWNT dispersion was mixed with pNIPAM monomer and polymerized with UV light. For the photoactuation experiments, stripes were laid out that were ~35 μm wide and were illuminated by a laser with a ~20 μm spot size (785 nm, ~30 mW). Exposure to the light caused the pNIPAM to undergo its phase transition, as observed by the sample changing from optically transparent to optically dense and opaque. Although the light was focused on a single stripe, after ~8 s the composite had equilibrated to a black spot of ~150 μm. The process was reversible and fast: after turning the laser off the original state was recovered in ~0.3 s.

The authors attributed the observed effect solely to heating. The SWNT absorb the light and the heat generated by the radiationless decay is transferred to the surrounding medium. Once the phase transition temperature has been attained, the dense, hydropho-bic state forms and is accompanied by the volume change. No control experiments were reported to confirm this hypothesis, such as mixing the hydrogel with graphite to see if the same temporal and photoactuation response would have been observed.

2.4 Carbon Nanotube Layered Composites

2.4.1 Carbon Nanotube/Acrylic Elastomer/Poly (vinylchloride) Trilayer Composites

Lu and Panchapakesan [28] developed photoactuators based on bilayers of SWNTs deposited onto polymers. The polymer base was on poly (vinylchloride) (PVC) coated with an acrylic elastomer, i.e. an adhesive tape. SWNTs were dispersed in



isopropyl alcohol at a concentration of 0.16 mg/mL. This suspension was filtered through a poly (tetrafluoroethylene) filter and then rinsed and dried. The SWNT sheet could then be peeled off of the filter to give a free-standing film with a thickness of 30–40 μm. The SWNT film was then directly attached to the acrylic elastomer side of the polymer film. The dimensions of the sample were 30 mm long × 2 mm wide × ~110–120 μm thick. The layers were composed of 50 μm PVC, 30 μm elastomer, and 30–40 μm SWNTs.

To measure the photoactuation, one end of the sample was fixed to a base and the amount of bending determined upon exposure to light. The light source was a halo-gen lamp and for typical experiments was operated at an intensity of 60 mW/cm2. The light was illuminated onto the SWNT layer and the bending occurred towards the PVC layer, away from the SWNT layer. The bending was reversible and repeat-able for multiple cycles. To measure the strain generated by the photoactuation, the cantilever was clamped on both ends and the strain measured with exposure to light. In all cases the strain was positive, indicating that the sample experienced tension. Rise times of the strain or maximum bending was ~20 s and the return to rest condi-tions with the light off was about the same.

The strain was measured as a function of light intensity using the white light source and it was found that as the light intensity increased the strain increased linearly to about 40 mW/cm2 and then nonlinearly at higher light intensities. At the maximum power used, 120 mW/cm2, the strain reached ~0.3 %. The effect of water on the pho-toactuation was also measured at 80 mW/cm2: in a dry environment the strain was found to be ~0.25 % but in water this value was reduced significantly to about ~0.06 %. Finally, using a series of semiconductor lasers, the strain was measured as a function of wavelength (635–1550 nm) at constant, low, power (15 mW/cm2). While all wave-lengths induced photoactuation, the magnitude increased considerably at higher pho-ton energies. At 0.8 eV the strain was ~0.02 %, which was nearly constant up 1.3 eV but then rose to ~0.04 % at 1.9 eV. No comparison to the absorption spectrum of the SWNTs used in the experiment was given but the wavelength response of the photoac-tuator strain did not appear to follow a typical SWNT spectrum.

The photoactuation was attributed primarily to thermal effects amplified by coupling to the elastomer layer. The absorption of light was thought to heat the sample and introduce electrostatic repulsion in the SWNT bundles. The electro-static contribution was supported by measuring the photocurrent in the samples. No temperatures change during the photoactuation experiments was reported, so the relative importance of the electrostatic and thermal effects was not determined. The reduction in the measured strain in the aqueous environment could, in part be, explained by heat dissipation into the surrounding environment but some of the reduction is also due to absorption of NIR light by water.

2.4.2 Carbon Nanotube/Photoresist Bilayer Composites

In a follow-up to their initial report, Lu and Panchapakesan [29] developed a photo-lithography protocol that could be used to make arrays of photoactuator cantilevers.

79

SWNTs were dispersed in isopropyl alcohol, ~0.1 mg/mL, using ultrasonication but no surfactant. The dispersion was vacuum filtered through a filter composed of a mixed cellulose ester. The thickness of the SWNT layer, ranging from ~40 to ~780 nm, could be controlled by varying the concentration of the nanotubes in the dispersion. The filter membrane containing the nanotube layer was transferred onto a silicon substrate and the cellulose ester dissolved with acetone, leaving a layer of SWNTs with controlled thickness. Photoresist was added to the top of the SWNT layer and standard photolithography was used to form patterns. O2 etching was used to removed nanotubes in exposed areas and XeF2 etching was used to remove Si beneath the nanotube layer. For photoactuation experiments, the photoresist (SU8) was not removed. Typical cantilevers were ~300 μm long × 30 μm wide × 7 μm thick—the thickness includes both the SWNT layer and the photoresist.

Photoactuation measurements were done with 808 nm laser light using vari-ous powers. The laser was focused an area of ~0.5 × 2 mm2, indicating that the entire cantilever surface was illuminated. When the cantilever was exposed to light, bending towards the SWNT layer was observed and the tip displacement was linear with respect to laser intensity, as shown in Fig. 2.8. The maximum dis-placement of the cantilever was ~23 μm under 170 mW of laser power. As in their earlier work, the photoactuation response was attributed to a combination of elec-trostatic, optical, and thermal effects.

It is interesting to compare the different cantilevers used by Lu and Panchapakesan. In both studies the nanotube layer was primarily composed of SWNTs, although of significantly different thicknesses. In the earlier report the polymer portion was a bilayer—PVC and an acrylic elastomer while in the lat-ter report the polymer was photoresist. The resulting photoactuation is similar in either case but with at least one distinguishable contrast: the observed strain response is nonlinear in the PVC case at laser powers above ~40 mW/cm2 while when the substrate is photoresist the strain is linear to at least 170 mW/cm2. No explanation for this has been given.

Fig. 2.8 Displacement of a photoresist/SWNT cantilever as a function of laser (808 nm) intensity. The line is drawn as a guide for the eye. The inset shows a cross-section of the cantilever under illumination. Reprinted with permission from Ref. [29]. Copyright 2006, American Institute of Physics



2.4.3 Carbon Nanotube/Silicon Nitride Bilayer Composites

More recently, Flannigan and Zewail [30] used 4D electron microscopy to exam-ine the photoactuation of multiwalled carbon nanotubes deposited onto an amor-phous silicon nitride grid as the substrate. The MWNTs were deposited onto the Si3N4 from an aqueous dispersion made by sonicating the MWNTs with sodium dodecyl sulfate (SDS), followed by centrifiguation. After deposition, the sample was dried at 80 °C under flowing argon for 30 min, cooled, washed with water, and then dried again for 1 h. The amount of residual SDS, if any, was not reported.

Upon exposure to light (776 nm) via a train of femtosecond pulses onto a dense mat of MWNTs, the absorption of light by the MWNTs induced thermal crystalli-zation of the silicon nitride substrate. The crystallization was observed as the sud-den appearance of Bragg diffraction spots by the concurrent electron beam. In the absence of light no crystallization was observed, demonstrating that the electron beam was not responsible for the phase change. This implied that the laser pulses were heating the MWNTs to temperatures on the order of 1000 °C and that there is strong coupling between the MWNT mat and the Si3N4 substrate. The research-ers took advantage of this strong coupling to observe the photoactuation of the MWNTs: the diffraction pattern of the underlying α-Si3N4 was used to monitor the photoactuation of the adjacent nanotube network. To monitor the photoactua-tion, single laser pulses (120 fs) with a low fluence, 6.4 μJ/cm2 were used. Under these conditions, a radial expansion of each concentric tubule of the MWNTs of 8 ± 0.3 pm was observed, which corresponds to an expansion of the entire nano-tube of about 400 pm. No expansion or contraction along the length of the nano-tubes were observed.

Based on the temporal response of the photoactuation and the low laser flu-ence, the authors considered a thermal mechanism unlikely. Using estimates of the absorption coefficient (6 × 104 cm−1), the specific heat capacity (2 J/g•K), and the density (2 g/cm3), the temperature rise could be no more than 1 °C in the MWNTs. Further, since the time constant of the experiments was found to be 120 ps, only an electronic process was thought to be reasonable to account for the observations. Charge separated polarons were suggested to be formed that could induce an electrostatic repulsion, that was responsible for the mechanical motion. Given the nonuniform nature of the MWNT layer, gradient absorption of the light could also contribute to the photoactuation.

2.4.4 Carbon Nanotube/Nafion Bilayer Composites

Levitsky and Euler [31–33] used SWNTs deposited onto a Nafion substrate to form photoactuating cantilevers. Nafion has a number of interesting properties that led to its choice as a substrate. Nafion contains sulfonic acid group that is effective for hydrogen bonding so Nafion films are always hydrated. This leads

81

to a pore structure with channels of a few nm that allows water transport through the material. SWNTs were dispersed in chloroform (~0.8–1 mg/mL) by sonica-tion and then air-brushed onto a warm (50 °C) Nafion film (50 μm thick). By repeatedly applying the SWNT suspension with the airbrush, the thickness of the nanotube layer could be controlled. Typical thicknesses for the SWNT layer were 10–20 μm. In the photoactuation experiments, the cantilevers used were 13 mm long × 1 mm wide × 60–70 μm thick. The light source used was a tungsten halo-gen lamp, a broad band source, with average intensity of 75 mW/cm2.

When exposed to light the Nafion/SWNT cantilevers exhibited bending that was always has a net response in the direction of the Nafion layer, independent of the direction of the light source. After the light is turned off the cantilever returns to its initial position, demonstrating the reversibility of the system. Two control experiments were done. When SWNTs are airbrushed onto a polyethylene sub-strate, no photoactuation is observed. Also, when graphite was applied to Nafion, using the same protocols, there again was no photoactuation. These observations suggest that in this system, both the substrate Nafion and the SWNTs play a role in the photoactuation.

Figure 2.9 shows the tip displacement of a cantilever exposed to light as a func-tion of time. When the light is turned on, the initial response is a small bend in the direction of the nanotube layer but quickly (less than a few seconds) reverses to bend towards the Nafion layer. The equilibrium displacement is reached in ~100 s. When the light is turned off, the reverse process takes place: the initial displacement is increased bending towards the Nafion layer for a few seconds and then reverses to relax to the original position in ~100 s. The fast process was only observed under low humidity conditions. At higher humidity levels, only a single exponential rise and decay were observed when the light was turned on and off, respectively.

The photoactuation response is linear with light intensity at low powers, as demonstrated in Fig. 2.10. The slopes of the lines in Fig. 2.10 are similar, suggest-ing that conversion from light energy to mechanical energy is similar across the

Fig. 2.9 Tip displacement of a Nafion/SWNT cantilever upon exposure to white light as a function of time. The solid lines are fits to exponential functions. The insets show the first few seconds after the light is turned on (lower left) and the light is turned off (upper right). Reprinted with permission from Ref. [33]. Copyright 2010 American Chemical Society



spectrum. Nafion is transparent in the visible and NIR regions, implying that the light absorption is solely due to the SWNTs. This was confirmed by measuring the photoactuation as a function of wavelength, as shown in Fig. 2.11. The photoactu-ation spectrum was normalized to a constant light intensity and compared with the absorption spectrum of the SWNTs. The two spectra match well. This implies that even with a mixture of chiralities in the SWNTs, each type of nanotube invokes a similar photoactuation response.

To investigate the role of the Nafion, absorption measurements in the IR region were taken to observe changes in the Nafion structure during photoactua-tion. Upon exposure to light, with the cantilever anchored to prevent mechanical motion, the intensity of the 2889 nm (3461 cm−1) peak decreased. When the light was turned off, the intensity of this peak returned. This peak arises from the OH stretching of hydrogen-bonded water molecules, which indicates that the internal structure of the Nafion changed during the photoactuation process.

Fig. 2.10 Observed tip displacement of a Nafion/SWNT cantilever as a function of light intensity at 1000 nm (filled circles) and 1300 nm (open squares). Reprinted with permission from Ref. [32]. Copyright 2006 American Chemical Society

Fig. 2.11 The tip displacement of a Nafion/SWNT cantilever normalized to light intensity as a function of wavelength. For comparison purposes, the solid line shows the absorption spectrum of the SWNTs in the same wavelength region. Reprinted with permission from Ref. [32]. Copyright 2006 American Chemical Society

83

Rate constants for the light-on and light-off processes were determined as a function of the thickness of the SWNT layer. Simple exponential functions were used to model the temporal response:

where zon and zoff are the cantilever tip displacement under light-on and light-off conditions, respectively, z∞ is the equilibrium position of the tip displacement with the light on, and kon and koff are the light-on and light-off rate constants, respectively. (A second exponential term was used to model the fast response, but not enough data could be collected to obtain reliable fitting parameters.) Each parameter depends upon the SWNT thickness. The equilibrium displacement, z∞, increases with increasing SWNT thickness, consistent with more light being absorbed driving more photoactuation, but are the same for a given on–off cycle, indicating the reversibility of the photoactuation process. The rate constants kon and koff also vary with the thickness of the SWNT layer, but are not the same. The light-on process has an increasing rate constant as a function of nanotube thick-ness, indicating that amount of light absorbed influences both the total displace-ment and the rate of bending. In contrast, the dark relaxation has a decreasing rate constant as function of increasing SWNT thickness. This effect arises because of the mechanical stiffness of the cantilever. As the nanotube thickness increases the modulus of the bilayer composite also increases, which reduces the rate of the recovery to the dark equilibrium state.The rate constants for the absorption change of the OH stretch used monitor the Nafion were also determined using the same exponential equations. Remarkably, kon and koff were found to match the photoactuation rate constants. This implied that the reorganization of the water molecules trapped in the Nafion pores contrib-uted to the rate limiting process.

zon = z∞(

1 − e−kont)

zof f = z∞e−kof f t

Fig. 2.12 Absorption at 2889 nm (3461 cm−1) of a Nafion/SWNT cantilever as a function of time with light on and light off. The wavelength was chosen to monitor the OH stretching frequency of water trapped in the Nafion pores. The cantilever was anchored to prevent movement. Reprinted with permission from Ref. [33]. Copyright 2010 American Chemical Society



The effect of a load added to the cantilevers was also measured. As the load increased, the maximum tip displacement decreased. For cantilever with a mass of 1.3 mg, a measurable displacement was found up to a load of ~100 mg, i.e. the photoactuator was able to generate enough force to move more than 75 times its own mass. The rate constants for both light-on and light-off were unchanged as a function of the added mass (Fig. 2.12).

To account for the observations, a model that involved polarization of the Nafion/SWNT interface was described and is shown in Fig. 2.13. The proposed mechanism envisions that charge separates in the SWNT excited state, in part due to the interfacial electric field. The surface charge on the SWNTs became negative and the bulk charge positive, which amplifies the interfacial field. Then, at the interface, water entrapped in the Nafion pores migrates towards the SWNT/Nafion boundary. This induces swelling of the Nafion at the inter-face and depletion of water in the bulk of the Nafion, causing the cantilever to bend towards the Nafion layer. The proposed mechanism accounts for the major response, but does not account for the short time photoactuation in the reverse direction. It was suggested that the dimensional changes that occur in the SWNTs as a result of the excitation and initial charge separation was responsible for the fast effect.

The proposed model is consistent with the observations of the direction of bending and the influence of humidity on the rate constants. The observation of

Fig. 2.13 Model used to explain the mechanism of photoactuation in the Nafion/SWNT photoac-tuation. Reprinted with permission from Ref. [33]. Copyright 2010 American Chemical Society

85

simple exponential functions to describe the temporal response requires that the diffusion component of the model occur in a thin film region, a few tens of nanom-eters from the interface.

2.5 Conclusions

Both single walled and multiwalled carbon nanotubes can be used to prepare materials capable of substantial, macroscopic photoactuation. Amplification of the optomechanical response occurs in composite materials and appears to be an inter-facial effect in nearly all cases. In all systems the carbon nanotubes absorb the incident light and then transform the energy into mechanical motion. Generally, it is thought that electrostatic effects dominate the mechanism of photoactuation but the details vary with the structure of the system. Many questions about the nature of photoactuation are still unanswered and there is no unifying understanding, yet.

When small bundles of pure carbon nanotubes are exposed to light there is sufficient charge separation between different individual nanotubes to induce electrostatic repulsion between fibers, which provides enough repulsion to cause deformations on the micron scale. When the bundles are assembled into freestand-ing films, the photoactuation is amplified to a macroscopic length scale, on the order of millimeters.

When carbon nanotubes are dispersed in a host matrix, the absorption of light by the nanotubes also causes macroscopic deformations. Whether the host is a rub-bery polymer or a hydrogel, the energy absorbed by the carbon nanotubes causes a structural change in the surrounding medium. For rubbery polymers, the volume change of the individual carbon nanotubes induces a local strain in the host mate-rial, which is amplified throughout the bulk. In the case of hydrogels, the excited state energy of the carbon nanotubes induces a phase change from a hydrophilic to hydrophobic state. Since the hydrophobic state has a higher density, a significant volume change ensues.

Layered composites also lead to macroscopically detectable photoactuation. The best characterized layered systems use polymers as the substrate for the car-bon nanotube layer, but the types of polymer vary considerably, including acrylic elastomers, photoresist, and Nafion. In each case the experimental evidence sug-gests that the interface between the carbon nanotube layer and the substrate layer plays an important role in the optomechanical response. The electrostatic interac-tion between nanotubes and the substrate is generally attributed as the key feature.

References

1. Baughman, R.H., Cui, C., Zakhidov, A.A., Iqbal, Z., Barisci, J.N., Spinks, G.M., Wallace, G.G., Mazzoldi, A., DeRossi, D., Rinzler, A.G., Jaschinski, O., Roth, S., Kertesz, M.: Carbon nanotube actuators. Science 284, 1340–1344 (1999)



2. Minett, A., Fràyasse, J., Gang, G., Kim, G.-T., Roth, S.: Nanotube actuators for nanomechanics. Curr. Appl. Phys. 2, 61–64 (2002)

3. Gartstein, Yu.N, Zakhidov, A.A., Baughman, R.H.: Charge-induced anisotropic distortions of semiconducting and metallic carbon nanotubes. Phys. Rev. Lett. 89, 045503/1–045503/4 (2002)

4. Sun, G., Kürti, J., Kertesz, M., Baughman, R.H.: Dimensional changes as a function of charge injection in single-walled carbon nanotubes. J. Am. Chem. Soc. 124, 15076–15080 (2002)

5. Landi, B.J., Raffaelle, R.P., Heben, M.J., Alleman, J.L., VanDerveer, W., Gennett, T.: Single wall carbon nanotube–nafion composite actuators. Nano Lett. 2, 1329–1332 (2002)

6. Tahhan, M., Truong, V.-T., Spinks, G.M., Wallace, G.G.: Carbon nanotube and polyaniline composite actuators. Smart Mater. Struct. 12, 626–632 (2003)

7. Levitsky, I.A., Kanelos, P.T., Euler, W.B.: Novel actuating system based on a composite of single-walled carbon nanotubes and an ionomeric polymer. Mater. Res. Soc. Symp. Proc. 785, D9.1.1–D9.1.6 (2004)

8. Levitsky, I.A., Kanelos, P., Euler, W.B.: Electromechanical actuation of composite material from carbon nanotubes and ionomeric polymer. J. Chem. Phys. 121, 1058–1065 (2004)

9. Minett, A., Fràysse, J., Gang, G., Kim, G.-T., Roth, S.: Nanotube actuators for nanomechanics. Curr. Appl. Phys. 2, 61–64 (2002)

10. Gartstein, Yu.N, Zakhidov, A.A., Baughman, R.H.: Charge-induced anisotropic distortions of semiconducting and metallic carbon nanotubes. Phys. Rev. Lett. 89, 045503/1–045503/4 (2002)

11. Liu, J.Z., Zheng, Q., Jiang, Q.: Effect of bending instabilities on the measurements of mechanical properties of multiwalled carbon nanotubes. Phys. Rev. B. 67, 075414/1–075414/8 (2003)

12. Li, C., Chou, T.-W.: Single-walled carbon nanotubes as ultrahigh frequency nanomechanical resonators. Phys. Rev. B 68, 073405/1–073405/3 (2003)

13. Bozovic, D., Bockrath, M., Hafner, J.H., Lieber, C.M., Park, H., Tinkham, M.: Plastic deformations in mechanically strained single-walled carbon nanotubes. Phys. Rev. B 67, 033407/1–033407/4 (2003)

14. Verissimo-Alves, M., Koiller, B., Chacham, H., Capaz, R.B.: Electromechanical effects in carbon nanotubes: ab initio and analytical tight-binding calculations. Phys. Rev. B 67, 161401/1–161401/4 (2003)

15. Cao, J., Wang, Q., Dai, H.: Electromechanical properties of metallic, quasimetallic, and semi-conducting carbon nanotubes under stretching. Phys. Rev. Lett. 90, 157601/1–157601/4 (2003)

16. Farajian, A.A., Yakobson, B.I., Mizuseki, H., Kawazoe, Y.: Electronic transport through bent carbon nanotubes: nanoelectromechanical sensors and switches. Phys. Rev. B 67, 205423/1–205423/6 (2003)

17. Sapmaz, S., Blanter, Ya.M., Gurevich, L., van der Zant, H.S.J.: Carbon nanotubes as nano-electromechanical systems. Phys. Rev. B. 67, 235414/1–235414/7 (2003)

18. Minot, E.D., Yaish, Y., Sazonova, V., Park, J.-Y., Brink, M., McEuen, P.L.: Tuning carbon nanotube band gaps with strain. Phys. Rev. Lett. 90, 156401/1–156401/4 (2003)

19. Pastewka, L., Kosinen, P., Elsasser, C., Moseler, M.: Understanding the microscopic pro-cesses that govern the charge-induced deformation of carbone nanotubes. Phys. Rev. B Cond. Matt. Phys. 180, 155428/1–155428/16 (2009)

20. Zhang, Y., Iijima, S.: Elastic response of carbon nanotube bundles to visible light. Phys. Rev. Lett. 82, 3472–3475 (1999)

21. Cronin, S.B., Yin, Y., Walsh, A., Capaz, R.B., Stolyrov, A., Tangney, P., Cohern, M.L., Louie, S.G., Swan, A.K., Ünlü, M.S., Goldberg, B.B., Tinkham, M.: Temperature dependence of the optical transition energies of carbon nanotubes: the role of electron-phonon coupling and thermal expansion. Phys. Rev. Lett. 96, 127403/1–127402/4 (2006)

22. Ahir, S.V., Terentjev, E.M., Lu, S.X., Panchapakesan, B.: Thermal fluctuations, stress relaxation, and actuation in carbon nanotube networks. Phys. Rev. B Cond. Matt. Phys. 76, 165437/1–165437/6 (2007)

23. Ahir, S.V., Terentjev, E.M.: Photomechanical actuation in polymer-nanotube composites. Nat. Mater. 4, 491–495 (2005)

87

24. Ahir, S.V., Terentjev, E.M.: Fast relaxation of carbon nanotubes in polymer composite actua-tors. Phys. Rev. Lett. 96, 133902/1–133902/4 (2006)

25. Ahir, S.V., Squire, A.M., Tajbakhsh, A.R., Terentjev, E.M.: Infrared actuation in aligned poly-mer-nanotube composites. Phys. Rev. B Cond. Matt. Phys. 73, 085420/1–0854201/2 (2006)

26. Ahir, S., Huang, Y.Y., Terentjev, E.M.: Polymers with aligned carbon nanotubes: active com-posite materials. Polymer 49, 3841–3854 (2008)

27. Zhang, X., Pint, C.L., Lee, M.H., Schubert, B.E., Jamshidi, A., Takei, K., Ko, H., Gillies, A., Bardhan, R., Urban, J.J., Wu, M., Fearing, R., Javey, A.: Optically- and thermally-respon-sive programmable materials based on carbon nanotube-hydrogel polymer composites. NanoLetters 11, 3239–3244 (2011)

28. Lu, S.X., Panchapakesan, B.: Optically driven nanotube actuators. Nanotechnology 16, 2548–2554 (2005)

29. Lu, S.X., Panchapakesan, B.: Nanotube micro-optomechanical actuators. Appl. Phys. Lett. 88, 253107/1–253107/3 (2006)

30. Flannigan, D.J., Zewail, A.H.: Optomechanical and crystallization phenomena visualized with 4D electron microscopy: interfacial carbon nanotubes on silicon nitride. NanoLetters 10, 1892–1899 (2010)

31. Levitsky, I.A., Kanelos, P.T., Viola, E.A., Euler, W.B.: Photoactuation in nafion-carbon nano-tube bilayer composites. Proc. SPIE Nanosens. Mater. Device II(6008), 600802/1–600802/6 (2005)

32. Levitsky, I.A., Kanelos, P.T., Woodbury, D.S., Euler, W.B.: Photoactuation from a carbon nanotube–nafion bilayer composite. J. Phys. Chem. B 110, 9421–9425 (2006)

33. Viola, E.A., Levitsky, I.A., Euler, W.B.: Kinetics of photoactuation in single wall carbon nanotube–nafion bilayer composite. J. Phys. Chem. C 114, 20258–20266 (2010)

References

89

3.1 Introduction

Single-walled carbon nanotubes (SWNTs) due to their unique mechanical, electronic, and optical properties are promising materials for different nanotech-nological applications including electronics, new composite materials, energy conversion devices, and sensors, as well as for various biological applications [1–5]. SWNTs can be chemically modified and used as nanosized building blocks. Nanotube modification can be performed by covalent or non-covalent function-alization. A drawback of the covalent chemical modification of the nanotube conjugated backbone is a possible loss of some unique properties including the alteration of its electronic structure and optical distinguishing characteristics which pristine nanotubes possess. To avoid this loss, the non-covalent modifica-tion of the nanotube is usually employed. Some success has been achieved in this direction using an organic or biological molecule/polymer wrapped around nano-tubes. Among them, single-stranded DNA (ss-DNA) is the most perspective poly-mer. An important feature has been revealed that the interaction between ss-DNA and the nanotube can alter optical and electrical properties of SWNTs.

Biomolecules interacting with inorganic nanostructures form different nanohy-brids with unusual multifunctional properties, which have the wide spectrum of applications in nanomedicine, biosensoring, nanoelectronics, environmental safety and so on. DNA is the main molecule of life as it is a carrier of genetic informa-tion in all living matters. This biopolymer has strong molecular recognizable capa-bilities and the ideal molecular structure to fabricate self-assembling nanodevices. Interaction between DNA and SWNT in water environment leads to nanohybrids creation [6, 7], which is a subject of intensive current interest. SWNTs have a set of unique physical/chemical properties, for example, unusual photophysical properties, mechanical stiffness and so on. Besides, nanotubes exhibit a versatile electronic nature which leads to modern molecular electronics (see, for exam-ple, book [3]). In spite of essential difference in ss-DNA and nanotube structures (ss-DNA is a flexible, amphiphilic biopolymer whereas SWNTs are stiff

Chapter 3Photophysical Properties of SWNT Interfaced with DNA


90 3 Photophysical Properties of SWNT Interfaced with DNA

hydrophobic nanorods), they form a stable hybrid in water. It is obvious that prop-erties of these two nanostructures supplement each other, and, as a result, a hybrid with specific structural features is formed. Due to its helical structure, ss-DNA can wrap tightly around the nanotube in water spontaneously. As ss-DNA in aqueous solution carries a negative charge in the chain and, as well, contains hydrophobic components, the stable hybrid with the tube is created when hydrophobic nitrogen bases are adsorbed to the nanotube surface via π–π stacking, while the hydrophilic sugar-phosphate backbone is directed to water [7–9]. Varying the sequence of four bases which form ss-DNA involves a set of polymers with slightly different physi-cal properties (rigidity, thermostability and so on). It is possible to suppose that each polymer binds most effectively only to the nanotube of the certain chirality or diameter. Thus, the separation of nanotubes of certain chirality or diameter from the bulk material can be relied on choosing an appropriate DNA sequence [7, 10].

Besides nanotube solubilization and separation with the help of DNA, this polymer provides carbon nanotube biocompatibility and facilitates biofunction-alization of this nanomaterial. For last purpose, this biopolymer can be used as a molecular interface between nanotubes and the recognition element of the sen-sor or can be connected with a drug because DNA biochemistry gives a useful knowledge of covalent or noncovalent attachment of different biochemical mol-ecules to DNA that can simplify elaboration of biological sensors and promote drug delivery. DNA molecule attached to carbon nanotubes can be also applied as a functional element of a nanosized genosensor. A growing body of researches on the development of DNA-hybridized biosensors is motivated by the wide scale of genetic testing, clinical diagnostics of genetic diseases, drug discovery, envi-ronmental control and fast detection of biological weapon. In such a genosensor SWNT plays, first, a role of a template to hold DNA in the stretched form and, second, provides sensing of DNA hybridization with different electrical measure-ments, for example, with scanning tunneling microscopy (STM) [11] or a carbon nanotube field-effect transistor (FET) [12]. As well, the development of a genosen-sor with optical detection, using SWNTs for this purpose, is a very perspective approach [13] too.

Integration of nucleic acids possessing high recognition capabilities with car-bon nanotubes has a considerable potential for creation of SWNT-based devices through a self-assembly mechanism. It is well-known that DNA is an ideal mate-rial for the assembly of metallic (Au, Pt, Ag) nanoparticles mainly due to the fact that DNA chains can create various functional self-assembly structures through the sequence-specific pairing interactions [14, 15]. In contrast to flexible DNA mol-ecules, carbon nanotubes have a very high native rigidity which should make them well suited as templates for organizing nanoscaled objects. So far, engineered DNA-wrapped carbon nanotubes have been successfully used as scaffolds to organize metallic nanotubes into micrometer-long linear conductive arrays or other desired molecular architecture.

Carbon nanotubes represent unique pores for DNA sequencing using electric sig-nal detection when DNA is translocated through a carbon nanotube with some volt-age applied to it [16]. For this purpose an atomic force microscope (AFM) can be

91

exploited to pull an ss-DNA oligomer from a carbon nanotube pore [17]. Molecular dynamics simulations (MD) showed that DNA could be spontaneously inserted into the carbon nanotube in a water environment [18]. Later, the biopolymer inserted inside the nanotube was observed experimentally by high-resolution transmission microscopy (HRTEM) [19]. Possibility of DNA sequencing employing carbon nanotubes as nanopores was demonstrated by the MD modeling [20, 21].

SWNTs have unique photophysical characteristics among which photolumi-nescence (PL) properties such as photoblanching stability, emission in the near-IR region in which living matter is transparent, relatively high quantum yield (QY) used in different biological and biomedical applications, including fluorescence imaging, optical diagnostics of in vivo biological processes, and detection of path-ogens in environmental and clinical samples. DNA providing the biocompatibility of carbon nanotubes facilitates these applications.

Thus, fundamental understanding of mechanisms providing self-assembly of SWNT:DNA hybrids, the determination of their structures, evaluation of the inter-action energy between hybrid components, as well as the photophysical characteri-zation of these DNA-functionalized carbon nanotubes are necessary now before their usage in different applications.

Since 2002 year, when near-infrared (NIR) PL of semiconducting carbon nanotubes was discovered [22], significant progress has been made in the funda-mental research of SWNT photophysics. This phenomenon was described in sev-eral reviews from different experimental and theoretical points of view, which appeared recently [23–31]. We are not aimed at touching all the aspects of SWNTs photophysics in the framework of one review, however, we try to concentrate on the role of SWNT:DNA hybrids in the study of photophysical processes in the car-bon nanotube, as well as on peculiarities of SWNT photophysics, caused by DNA binding to them.

3.2 SWNT:DNA Hybrid: Structures and Energy Interaction

3.2.1 DNA Helix on SWNT

A stumbling block on the way of practical realization of many SWNT applications is a formation of self-organized crystalline bundles (ropes) because of extensive van der Waals (vdW) interactions among nanotubes through their electron-rich surfaces. Thus, beginning from the nanotube discovering to present days, a devel-opment of the reliable method for preparing individual nanotube dispersions with large concentrations of individual nanotubes is extremely required. Individual SWNTs can be obtained by sonication of as-prepared SWNT bundles using cer-tain organic solvents [32–34] or in water solution of surfactants [35–39] or poly-mers [40–42]. Intense sonication overcomes the large intertube attraction splitting

3.1 Introduction


the bundles and free surfactant molecules, which are adsorbed to SWNT surfaces creating micelles around the individual tube, prevents SWNTs reaggregation. Such surfactants as sodium dodecyl sulfate (SDS) [22, 35, 37], sodium dodecyl benzene sulfonate (SDBS) [36], or sodium cholate (SC) [39] were often used to gener-ate stable nanotube suspensions. Sonication of an aqueous nanotube suspension containing a polymer can also lead to SWNT debundling. Among them, DNA has been used successfully to disperse SWNTs in water [6–9]. DNA strands have been proven particularly effective in solubilizing SWNTs producing particularly high yields of solubilized individual nanotubes and small bundles [41]. Nanotube solu-bility in water is provided with the negative charge of phosphate groups, whereby hydrophobic DNA bases π-stack on the nanotube surface. The idea of ss-DNA wrapping helically around a nanotube has been provided firstly by Smalley with coworkers [8] and then was supported by experiments and force-field calculations of SWNTs wrapped by ss-DNA [7, 9]. In the following researches this hybrid manifested itself as an extremely promising material for different medical applica-tions [43–46], in which DNA assured a biocompatibility of SWNTs. Furthermore, DNA-wrapped SWNTs can be separated (see, review [47] and References therein) according to type/chirality by ion-exchange chromatography [9, 48], or density gradient ultracentrifugation [49–51]. Size-exclusion chromatography has been used to separate SWNT:DNA dispersions into different lengths and remove impu-rities [52–54]. Separation of nanotubes by the length, diameter and conductivity was also achieved by other techniques such as flow-field flow fractionation [55], dielectrophoresis [56], and agarose gel electrophoresis [57].

Many of the existing experimental studies on solubilization of SWNTs with DNA have focused primarily on short synthetic homooligonucleotides or oligonu-cleotides (short strands of DNA) of the definite nucleotide sequence with the length less than 30–40 nucleotides (see, for example, review [58]). Among them, the poly-mer with alternating dG and dT nucleotides with a length between 10 and 20 bases is best suited for solubilization and separation of different nanotubes [7, 9, 59, 60].

Adsorption of ss-DNA with the helix structure takes place upon polymer wrap-ping around the tube during the ultrasonic treatment [7, 8] or at replacement of surfactants [13]. In a high-resolution AFM image a wire-like structure obtained after deposition of the drop of SWNT:DNA aqueous suspension on the mica sub-strate corresponds to an individual carbon nanotube [61]. Bulges stretched along the nanotube have a symmetrical form relatively to the tube. They are caused by ss-DNA wrapping around SWNT. This image shows clearly the helical ss-DNA structure upon a single SWNT with some periodicity along its axis. So, d(GT)30:SWNT (produced by HiPCO method) hybrids have a much more uniform periodic structure with a regular pitch of ~18 nm and height of ~2 nm [9]. The similar structure parameter (~14-nm pitch) was obtained by Campbell et al. [61]. They observed helical turns of wrapped DNA strands that are closely arranged end-to-end in a single layer along SWNT. They found also that the obtained value of pitches is independent of the length and sequence of wrapping DNA. Heights of the pattern determined above the nanotube (~0.4 nm) were attributed to a single layer of ss-DNA. However, including the mean diameter of HiPCO SWNT that is

93

~0.9 nm and an average height of ~2 nm hybrids obtained earlier [9], the polymer height can be estimated as ~1 nm. This height value of the wrapping polymer is consistent with that one obtained recently [62, 63] for HiPCO SWNT.

However, employing STM to reveal the structure of carbon nanotube:DNA complexes (GA)20 wrapping around (6,5) nanotubes showed that a coiling period is 3.3 nm [11]. The smaller values of pitches (close to 2.2 nm) for ss-DNA cover-ing the nanotube were also determined employing HRTEM images [64].

TEM images of SWNT:DNA demonstrate helical structures oriented along the individual nanotube axes [65]. TEM image showing helical wrapping of d(GT)15 around an individual SWNT in detail is presented in Fig. 3.1. Note that this image showed a partial covering of nanotube with d(GT)15 pitches of which had irregu-lar character. Thus, the microscopy observation demonstrates that parameters of the polymer wrapping around the nanotube surface are different. We believe that it depends on the polymer type and nanotube species as well as the method of hybrid preparation.

As the interaction between ss-DNA and the nanotube alters the optical and electrical properties of SWNTs [66–69], the detail experimental and theoretical analysis of the structures and estimation of binding energies between hybrid com-ponents are much needed.

Models of hybrids formed by DNA and the nanotube with different structure parameters were simulated by MD modeling [7, 70–83]. These simulations indi-cate that the ss-DNA binding to SWNTs is very strong. This modeling confirms that π–π stacking between nucleic acid bases (NAB) and the nanotube surface is the basic mechanism of SWNT:DNA interaction and the major reason of the hybrid stability. To understand better the mechanism of DNA binding to the nanotube surface, it is desirable to evaluate the relative strength of SWNT:NAB interaction. An application of accurate quantum-chemical methods to describe these interac-tions provides a detailed and quantitative characterization of formation processes and structures of nanotube:DNA hybrids.

Fig. 3.1 TEM image showing helical wrapping of d(GT)15 around an individual SWNT (HiPCO). Reprinted from [65], with kind permission from © Elsevier 2007



3.2.2 Nucleic Bases on SWNT: Ab initio Calculation

Based on Ab initio calculations of the SWNT:NAB systems are associated with some difficulty, the main reason of it is the large size of the carbon surface which should be used in the calculation for correct representation of SWNT. Moreover, the stacked complexes are stabilized by vdW forces which require the applica-tion of the computationally expensive quantum chemical methods for their correct description. Thus, different approximations or approaches for decreasing a size of the calculated system are taken mainly into account in this calculation. In the last case nanotubes of the small diameter or the fragment of the nanotube surface are usually considered [84–92].

The interaction between NABs and carbon nanotubes was investigated using the Hartree–Fock method [84], local density approximation (LDA) [85, 86] within the density functional theory (DFT) or DFT with modern functionals and MP2 [87–92] (second order Møller-Plesset perturbation theory) methods. MP2 is a very large resource-intensive method; therefore, it is employed only for control or for optimized structures. As one may expect, the Hartree–Fock method which does not account for the dispersion interaction was unable to reproduce reliably the stacked structure of the SWNT:NAB hybrids. The Hartree–Fock geometry optimizations of the complexes did not converge to stacked structures and this method leads to an almost perpendicular structure of the base with the tube with the interaction energy of −6.7 kJ/mol, which is much smaller than the interac-tion energy predicted by MP2 or DFT methods for the stacked orientation of the complex. The interaction between NABs and small-diameter (5,0) metallic car-bon nanotube was studied using the LDA approximation [85]. It was found that molecular polarizability of base molecules plays the dominant role in the interac-tion strength of base molecules with SWNT. The base molecules exhibited sig-nificantly different interaction strengths and the calculated binding energies follow Guanine (G) > Adenine (A) > Thymine (T) > Cytosine (C) > Uracil (U) hierarchy which appears to be independent of the nanotube curvature. Shtogun and cowork-ers [86] used LDA within DFT to study the interactions between adenine and thy-mine and their radicals with the nanotube surface. Their studies suggest that both DNA bases and their radicals can be easily adsorbed on nanotube surfaces due to the noncovalent interaction between the delocalized π orbitals from DNA and nanotubes. However, as was demonstrated recently [92], LDA method within DFT gives underestimated binding energies for π–π stacking molecule arrangements.

Wang and Bu [87] studied the interaction between cytosine and a small frag-ment of the nanotube surface (C24H12), using the novel DFT (with PW91LYP and MPWB1K functionals) and MP2 methods. However, the fragment of the nanotube surface ((5,5) nanotube chirality) used in this calculation was too small to allow cytosine [89] or thymine molecules [88] to be arranged in an optimal position on the surface through the efficient interaction of nucleobases with terminated hydro-gens. Employing the larger nanotube surface (C36H18) (fragment of zigzag (10,0) nanotube) calculation performed at the MP2/6-31++G(d,p) level of theory, with

95

accounting for the basis set superposition error during optimization, orientation of the cytosine [89] and other bases [90] with respect to the nanotube surface was found and the stability row of SWNT:bases systems was established: G (−67.1 kJ/mol) > A (−59.0 kJ/mol) > C (−50.3 kJ/mol) ≈ T (−50.2 kJ/mol) > U (−44.2 kJ/mol). The geometry optimizations of the SWNT:NABs complexes performed with DFT and MP2 methods produced three stable conformations for each of adenine-nanotube, guanine-nanotube, cytosine-nanotube and uracil-nanotube complexes and two stable conformations for the thymine-nanotube complex. One of the sta-ble conformations for each base is shown in Fig. 3.2. The order demonstrates that there is a correlation between the size of the bases molecule and the interaction energy with the fragment of the nanotube surface. Because of the additional five-member imidazole ring in the structure of the purine bases, the π stacking inter-actions between adenine/guanine with SWNT are stronger than cytosine/thymine ones. It is seen from optimized structures that nucleobases are located above the surface of SWNT at a distance of around 3.2–3.3 Å.

The nanotube diameter increasing leads to flattening of the nanotube surface and, as a result, to an increase of the contact between NABs and the nanotube sur-face and, therefore, is accompanied with an increase of the interaction energy. The dependency of the interaction energy between cytosine and the nanotube surface for two hybrid conformations on the nanotube diameter is presented in Fig. 3.3. As expected intuitively, the interaction energy shows an increasing trend while nanotube diameter grows. With the nanotube diameter rise, the interaction energy approaches the results obtained for the cytosine:graphene complex which was equal to −59.1 kJ/mol. The difference between three stable conformations lays in the orientation of the cytosine side groups with respect to the nanotube surface (A conformation was shown in Fig. 3.2, B conformation differs from A by rota-tion of cytosine by about 60°). There is an interesting feature in the interaction energy changing behaviour calculated for different conformations and nanotube diameters. Interaction energy difference between conformations decreases with the nanotube diameter increasing (Fig. 3.3). In complexes with the planar graphite surface the differences in the interaction energies between conformations disap-pear as the hybrids become identical. It is clear that planar graphene may serve as an upper limit for interactions with carbon surfaces.

Fig. 3.2 The stable conformation for adenine, guanine, cytosine, thymine and uracil adsorbed on the nanotube fragment. Reprinted with permission from [90]. © American Chemical Society (2009)



To compare the influence of different side groups of NABs on the interaction energy, adenine and guanine hybrids with the nanotube were analyzed, including also purine, 2-aminopurine and 6-oxopurine (hypoxanthine) molecules [90]. The lowest interaction energy was found for non-substituted purine:nanotube hybrids: −52.0 kJ/mol, the adenine (6-aminopurine):nanotube complex is by 5.2 kJ/mol−1 more stable. At the same time the hybrid of 2-aminopurine with the nano-tube fragment is more stable than purine:nanotube dimers by 6.0 kJ/mol−1. It was demonstrated that 2-aminopurine-nanotube dimers are more stable than the 6-ami-nopurine (adenine):nanotube hybrids. In adenine the NH2 group is located perpen-dicular to the longest dimension of the molecule but 2-aminopurine molecule has the extended shape. In the latter case the contact between 2-aminopurine and the nanotube surface is larger than in the case of adenine, and this fact results in the stronger interaction. Based on this calculation, a conclusion was made that the con-tribution of side group of bases into the interaction energy is about 5–6 kJ/mol−1.

Recently, the interaction energy of bases with the full nanotube surface but small diameter (about 5.5 Å, zigzag nanotube (7,0)) was calculated employ-ing DFT level of the theory with functional M05-2X [91]. The stability raw (G > A > C > T > U) was confirmed. For the raw of the bases obtained the amount of Mulliken charges was determined which varies from 0.04 till 0.02 e. It is evi-dent that only small amount of the electronic charge transfers from bases to SWNT in the complexes.

Novel M06-2X functional within the dispersion-corrected DFT has been also applied recently for calculation of structures and the binding energy of SWNTs of small diameters (3,3), (4,4), (5,5) with five nitrogen nucleobases [92]. Computational studies showed that the nucleobases bind to these nanotubes in G > T ~ A > C > U order, and for graphene the order is G > A > T > C > U. Nucleus-independent chemi-cal shift calculations pointed to the substantial enhancement of aromaticity for all

Fig. 3.3 The dependency of the interaction energy in SWNT:cytosine conformations A and B on the nanotube diameter. With the increasing nanotube diameter the interaction energy approaches the horizontal line which corresponds to the energy (−59.1 kJ/mol) of cytosine binding to graphene. Reprinted from [89], with permission from © Elsevier 2012

97

nucleobases upon their binding to SWNT and graphene. Calculations showed also that there was no appreciable change in the HOMO–LUMO gap value upon the complex formation in all the complexes considered.

3.2.3 Calculation of Nucleoside Binding to SWNT

Structures of the single nucleoside and SWNT and interaction energy between them were calculated by Kaxiras et al. [93]. They focused their study on the semi-conducting (10,0) nanotube (the diameter is 7.9 Å). The potential energy surface of biomolecules is extremely complicated and currently precludes direct explora-tion with ab initio methods. The search gave about 1000 distinct potential energy minima for each SWNT:nucleoside system. Despite the numerous configurations, the authors found that very few of them are dominant, with significant room-tem-perature populations. For instance, there are three most stable configurations for adenosine with populations 28.4, 27.6, and 10.1 %. Therefore, they focused only on the dominant configurations in the evaluation of SWNT:nucleoside interac-tions. The nucleoside binds on carbon nanotubes through the base unit located in 3.3 Å away from the SWNT wall. Whereas the base unit remains planar without significant bending, the sugar residue is more flexible. It was demonstrated that all four of the most stable configurations involving nucleoside adsorption on SWNT are with the sugar-base direction pointed perpendicular to the tube axis or slightly tilted. The structures obtained from the force-field calculations were further opti-mized using LDA within DFT. As a result, the authors received significantly underestimated interaction energies which for these nucleosides were from 42 till 44 kJ/mol [93].

Using calculations based on the first-principles pseudopotentials within DFT, adsorption of adenine dinucleoside on SWNTs with various diameters and chiral angles was reported [94]. The calculations indicated that, in addition to nonco-valent ππ-interactions between adenine and SWNT, the hydrogen bond interac-tions due to the hydrogen bond between H atom of dinucleoside and π-orbital of the nanotube, are also supplemented. It was also found that the adsorption energy depends strongly on SWNT type and diameter.

3.2.4 Structures of Oligonucleotides Adsorbed on SWNT and Energy Interaction Between Them: Molecular Dynamics Simulation

A dynamic behavior of SWNT interacting with DNA in aqueous environment can be observed with MD simulations. Currently MD is intensively exploited to clarify self-assembly mechanisms characterizing SWNT:DNA to determine the hybrid



structure and to evaluate the interaction energy [7, 69–83, 95]. This powerful com-putational method gives an opportunity to take into account such very important experimental parameters as temperature, the water environment, and the ionic strength. This gives a certain advantage to MD over ab initio calculations which can characterize only small molecules or systems in vacuum. MD simulations car-ried out by Gao et al. [18, 70] showed strong association of short SWNT and ss-DNA octamers in water, which are not only adsorbed to the outer tube surface but can insert into the nanotube. Lu et al. [71] modeled a periodic array of SWNTs in contact with DNA, emphasizing on the electron transport in both components, and, a result, this nanosystem was proposed as a very sensitive nanoscale elec-tronic device for ultrafast DNA sequencing.

Using MD simulations, Manohar et al. [73] demonstrated that the free energy upon the hybrid formation is contributed by adhesion between DNA bases and SWNT, entropy of the DNA backbone, and electrostatic interactions between back-bone charges. It was also revealed that the ionic strength of solution has a strong influence on SWNT:DNA structure. The influence of the salt (NaCl) concentration on the binding energy between (6,0) SWNT and different nucleotide monophos-phates (NMPs) in aqueous solution was also simulated [75]. The largest salt effect occurs for adenylic (A) (the binding energy decreases) and uridylic (U) (the binding energy increases) nucleotides, with a weaker possible effect for thymidylic (T) nucle-otides. Binding energies of cytidylic (C), and guanidylic (G) nucleotides are the same with or without salt within the statistical uncertainty. This was due to differences in the association of sodium ions with phosphate groups and also to differences in NMP conformations for A and U in salt, compared to the simulations without salt.

Martin et al. [77] studied association of several ss-DNA decamers (d(T)10, d(G)10, d(GT)10) with SWNTs of different chiralities in the aqueous environ-ment. They found that, after the fast adsorption onto the nanotube surface, oligo-nucleotides undergo a slow structural rearrangement. It was shown that DNA in the hybrid acquires a number of distinct backbone geometries which depend both on DNA sequence and the nanotube diameter. Exploring self-assembly mecha-nisms, Johnson et al. [78] performed MD simulations to determine the structure and energetic properties of SWNT:(14-base oligonucleotide) hybrid. They found that in aqueous solution short ss-DNA near SWNT undergoes a conformational change via the π–π stacking interaction between nitrogen bases and the nano-tube surface. This structural conformation enables the biopolymer to wrap spon-taneously around SWNT into compact right- or left-handed helices. Driving forces which provide the polymer helical wrapping are electrostatic and torsional interac-tions within the sugar-phosphate backbone. In the recent publication, these authors showed the entire ensemble of oligonucleotide conformations in SWNT:(GT)7 hybrid [79]. They calculated the free energy landscape and found the global min-imum corresponding to a nonhelical loop structure of the polymer. SWNT:NAB binding is dominated by vdW forces between the base and nanotube sidewall, while solvation and entropic effects play a relatively minor role .

A spontaneous adsorption of relatively long homooligonucleotides dC25, dT25, dG25, dA25 on the surface of the carbon nanotube (16,0) (the nanotube diameter is

99

1.24 nm) was studied by the MD method too [82] (Fig. 3.4). In the initial step of this modeling one deoxyribooligonucleotide selected from dC25, dT25, dG25, dA25 raw in the self-ordering helical B-form was located near the carbon nanotube sur-face (Fig. 3.4).

Figure 3.5 presents dependences of the nanotube:oligonucleotide interaction energy on simulation time for d(C)25, d(T)25, d(A)25 and d(G)25. The gradual increase of the binding energy within 40 ns is common for all the dependences. Note that the very strong energy (~300 kcal/mol (1260 kJ/mol)) between compo-nents of SWNT:dC25 hybrid was reached during the first 10 ns (107 steps) of simu-lation. In spite of the fastest rate of the binding energy increase, dC25 did not make a pitch around the nanotube after first 10 ns.

The second oligonucleotide which demonstrates the high rate of achieving the stable conformation on the tube surface was dT25 with 13 thymines stacked (a nitrogen base is considered as stacked if more than half of pyrimidine or purine ring atoms are in vdW contact with the nanotube surface) with the tube after 10 ns. The interaction energy of the purine oligonucleotides with the tube increases with significantly lower rate. After 10 ns only 8 adenines and 7 guanines were stacked with the nanotube surface.

By 30 ns the interaction energy between SWNT and dC25 reached the maxi-mum with 20 cytosines stacked with the tube, and during the following 20 ns this energy does not change practically (Fig. 3.5). After 30 ns, 18 thymines of dT25

Fig. 3.4 Snapshot of d(C)25 structure and SWNT (16,0) in the initial simulation step (upper). The sugar-phosphate backbone is depicted by solid curve. Snapshots of hybrids formed by nano-tube (16,0) with oligonucleotides d(C)25 and d(G)25 (lower) after 50 ns simulation. Water mol-ecules and Na+ counterions were removed for better visualization. Reprinted with permission from [82]. © American Chemical Society (2011)



were in stacking with the tube surface and the polymer was wrapped, there was no loop, though 2 dimers can be observed. The situation with purine oligonucleo-tides is different: the number of bases stacked with the tube increased with time, oligonucleotides are not wrapped around the nanotube but form the loop, which is characteristic feature for them. The number of bases stacked with the tube was 16 adenines and 17 guanines after 50 ns. Nevertheless, after 50 ns, each of these polymers did not form the pitch. Up to 45 ns dT25 reached the energy favora-ble conformation on the nanotube and the number of thymines stacked with the tube runs up to 22 (more than for dC25). This is accompanied with the energy increase the value of which for SWNT:dT25 was by ~25 kcal/mol (105 kJ/mol) higher than that for SWNT:dC25. Up to 50 ns interaction energies between d(A)25 or d(G)25 and the tube become even equal which is turned out by ~20 kcal/mol (84 kJ/mol) lower than for SWNT:dC25. A low rate of increase in the energy of purine oligonucleotide binding to the tube surface and their unwrapping of SWNT are caused by a higher self-stacking energy than that in pyrimidine ones [96]. A sta-ble ordered self-stacking structure of the polymer prevents from its structural re-ori-entation which needs to take the energy favorable conformation on the tube surface.

Estimations obtained from modeling allowed to establish the oligonucleo-tide row which demonstrates decreasing in interaction energies between oligonu-cleotides and the carbon nanotube: d(T)25 > d(C)25 > d(A)25 ≈ d(G)25. Recently Hughes et al. [97], upon comparison of homooligonucleotides (dA15, dG15, dC15, and dT15) to disperse and exfoliate SWNTs in water employing absorption and PL spectroscopies, concluded that nanotubes were more temporally stable in dC15 and dT15 suspension. The order of the dispersion efficiencies was found to be d(T)25 > d(C)25 > d(A)25 ≈ d(G)25, where thymine oligonucleotide produced the most intense nanotube photoluminescence and absorption signals.

On the basis of the phenomenological model of SWNT:DNA complexes sug-gested, Enyashin et al. [98] concluded also that homopolymeric ss-DNAs based on the pyrimidine nucleotides are more effective in SWNT wrapping. They also found that densities of states of the SWNT:DNA complexes are close to

Fig. 3.5 Dependence of interaction energy between SWNT and d(C)25, d(T)25, d(A)25, d(G)25 oligonucleotides on simulation time (with 1 fs time step). Reprinted with permission from [82]. © American Chemical Society (2011)

101

the superposition of those of the unbound components with some additional states below the Fermi level. The band gap in a hybrid SWNT:DNA system is determined by the competition between the Fermi levels of the ‘free’ DNA and SWNT. The authors of this work indicated that, in some specific cases, such as for (8,2) and (7,4) metallic tubes, a considerable charge transfer (as large as 0.2–0.4 e where e is the electron charge) from DNA to SWNT was observed. In these cases an essential gain in the SWNT:DNA formation energy has been obtained. However, in some other cases this charge transfer was no more than 0.05 e. This calculation showed that even in the case of SWNT weakly bonded with DNA an essential change in optical and conductive properties of SWNTs of specific species could be observed.

As follows from MD simulation [77, 78, 82], the mean energy of the interaction between one nucleotide and the nanotube is about 13–15 kcal/mol (54.6–63 kJ/mol). The nucleotide-nanotube binding energy is equal to the difference of two sums [80]. The first of them is determined by the stacking energy of the nucle-otide-nanotube (vdW interaction) and water–water interactions which appeared because of decreasing the hydrophobic surface after binding of the hydrophobic base of the nucleotide to the hydrophobic nanotube surface. The second one is a result of water-nanotube and water-nucleotide interactions. Fulfilled recently, the estimation of the binding energy of the nitrogen base with the nanotube showed that the value of this energy is determined as the decreasing value of the stack-ing energy by about 3 kcal/mol (12.6 kJ/mol) through the solvation effect [80]. The evaluation of the interaction energy of one nucleotide and its base with the nanotube (vdW interaction) showed that the base-nanotube stacking energy gives ~60 % contribution to the interaction energy of one nucleotide with the nanotube. As other components of the nucleotide (ribose and the phosphate group) are more hydrophilic than hydrophobic and as they do not contact with the nanotube sur-face directly, it was suggested that the solvation effect for these components is not essential. Thus, the binding energy between the polynucleotide and SWNT was mainly determined with vdW energy a contribution of which is decreased by about 15 % through the solvation effect [80].

The loops formed during oligonucleotides adsorption on the tube surface hinder their wrapping around the tube, and this limitation is manifested greatly in the case of purine oligonucleotides. Bases contacted with the tube surface at the beginning/end of the loop play a significant role in keeping this conformation because they serve as anchors for this loop [82]. Although the energy barrier for base moving along nanotube surface is low [77], the polymer elasticity restrains this hopping over barriers. Whereas the energy of purine base self-stacking [96] as the energy of their adsorption onto the nanotube [90] is stronger than that for pyrimidine one, therefore, such a loop is especially stable in the case of purine oligonucleotides hindering these polymers to wrap around the nanotube.

The temperature growth can increase the rate of oligonucleotides to reach the maximum binding energy mainly due to the destruction of nitrogen base self-stacking. The temperature influence on spontaneous adsorption of oligonucleo-tides on the SWNT surface is manifested in two different effects. On the one hand,



the temperature rise breaks the base self-stacking, the polymer becomes more flexible and disposed to reorientation of the strand structure. Thus, the oligonu-cleotide increases the rate of achieving the favorable energy conformation on the nanotube to obtain the maximum magnitude of the binding energy with the tube. Temperature provides the energetically more favorable position of the polymer on the nanotube in a shorter time when more bases are stacked with the nanotube sur-face. At the same time the temperature rise makes easier the barrier overcoming between neighboring hexagons. This promotes the movement of the base along the nanotube surface and permits the polymer to occupy an energetically more favored contact with the SWNT surface. Simulation showed that higher temperature makes the process of the oligomer wrapping around the nanotube easier. On the other hand, it is necessary to take into account the base desorption from the nanotube surface, which increases with the temperature rise too. However, moderate tem-perature heating (in the temperature range up to 100 °C) can be used to increase the rate of carbon nanotube solubilization in water.

Evidently, temperatures used in the simulation of SWNT:r(C)25 hybrid forma-tion (293, 343 and 363 K) are not enough to provide the effective base desorp-tion from the nanotube surface (Fig. 3.6). However, under these temperatures the polymer order → disorder transition takes place. It should be noted that noticea-ble desorption of poly(rA) from the nanotube was not observed in UV-absorption spectroscopy study of SWNT:poly(rA) suspension [76, 99] when it was heated from room temperature up to 363 K. Figure 3.6 presents dependence of the inter-action energy in SWNT:r(C)25 hybrid as a function of time at three temperatures. As Fig. 3.6 shows, temperature has an essential influence on the rate of achieving the energetically more favored conformation on the nanotube, which is reached for 20 ns at 363 K and for twice as long at 343 K. At 293 K this rate is very slow: so, for 50 ns only 16 cytosines are stacked with the tube surface, oligonu-cleotide does not achieve the stable conformation on the tube surface even for 50 ns [82].

Fig. 3.6 Dependence of interaction energy between SWNT and r(C)25 on simulation time at 293, 343 and 363 K. Reprinted with permission from [82]. © American Chemical Society (2011)

103

Recently Tu et al. [10] have shown that highly sequence-specific short oligo-nucleotides (10–20 nucleotides) can select certain SWNT species (chiralities) from synthetic mixtures. They have identified more than 20 short DNA sequences which recognize 12 nanotubes of certain chirality and then the chromatographic purification of selected nanotube species occurs. To explain the observed recogni-tion ability of selected oligonucleotides, a model was proposed by Jagota group, in which ordered ss-DNAs forms β-barrel structures [10, 81]. In these structures both the backbone and bases are arranged helically on an imaginary cylinder which per-mits the insertion of a nanotube of a specific diameter. The barrels are composed of two or more strands of ss-DNA wrapped helically and stabilized by interstrand hydrogen bonding between bases of different strands. In the following studies they considered another ordered structure (a consistent motif) as a possible model in which the ss-DNA strand forms a right-handed helical wrap around SWNT, stabi-lized by hydrogen bonding between distant bases of this strand (self-stitched struc-tures) [83, 100]. It was also shown that the motif stability increases for sequences with the ability to connect the ends of adjoining strands through hydrogen bonds. The presence or lack of such hydrogen bonding at the ends of the strands is signif-icant for the strong sequence specificity. Addition or subtraction of one base from a recognition sequence affects strongly the relative SWNT:DNA binding strength. For example, it was shown experimentally that (TAT)4 sequence which recognizes (6,5) SWNT binds about 20 times stronger than either (TAT)4T or (TAT)3TA (short DNA strands of this family) [100]. This conclusion was made on the basis of the brightest photoluminescence of nanotubes wrapped with (TAT)4 among three samples indicating the highest concentration of nanotubes present in the sample. This means that this recognition sequence has the better nanotubes dispersion effi-ciency, which is critical for effective separation. Thus, in addition to π–π stacking interaction between bases and nanotube surface, hydrogen bonding between nitro-gen bases plays essential role in stability of SWNT:ss-DNA hybrid, and additional experiments as well as simulation are needed to understand better the nature of SWNT:ss-DNA interactions as well as the recognition ability of sequence-specific structures.

3.2.5 Wrapping of Relatively Long DNA Around SWNT

As it has been already mentioned, in the most of articles devoted to SWNT:DNA the researchers described hybrids formed by nanotubes with short synthetic oli-gonucleotides with the length less than 30 nucleotides. However, a relatively high cost of short oligonucleotides restrains their wide practical application. The price of d(GT)20 is typically $25000/g but cost-effective genomic DNAs ($1000/g) can be also employed as SWNT dispersing reagent [101]. First experiments [66, 102–104] with a relatively long ss-DNA demonstrated that genomic ss-DNA (>100 bases) of a completely random sequence of bases can efficiently solubilize SWNTs in water, forming complexes with nanotubes by wrapping around them as tight helices.



For some particular applications, for example, to fabricate SWNT networking for a field-effect transistor [12, 105, 106], relative long ss-DNA can be suitable because such DNA may provide the localization and interconnection of nanotubes properly or to construct larger-scale microelectronics by means of DNA hybridiza-tion or local arrangement on the surface. ss-DNA can be synthesized but its length does not exceed 100–120 nucleotides. Longer ss-DNAs can be obtained from the genomic DNA [66, 72, 103] which is applicable to wrap and disperse SWNTs, but the separation of the single-stranded genomic DNA of the precise sequence is a rather difficult task. Long ss-DNA can be also synthesized by a biochemical tech-nique known as “rolling circle amplification” [104] or by asymmetric polymerase chain reaction [107].

Knowledge of the adsorption scheme/model of a longer polymer on the nanotube surface can be essential for understanding an operation of future SWNT-based genosensors. Such sensor will contain a probe single-stranded oli-gonucleotide of the defined sequence being complementary to the characteristic fragment of DNA diagnosed. The inexpensive extraction of this fragment from the target can be performed by employing ultrasonic fragmentation. However, in most cases this method does not provide DNA fragmentation in the length less than 100 base pairs. It should be added that adsorption of a longer polymer on the nanotube can differ from that of a short DNA which can adsorb to SWNT in folded sections on one tube side rather than in wrapping [108].

A detailed analysis of AFM images of SWNT hybrids with ss-DNA obtained from a genomic polymer (or with a synthetic polymer, for example, polyC) reveals that in some cases polymer fragments are able to wrap in several layers around the nanotube, forming strain-like spindles [62]. Section analysis of the hybrid height along the nanotube demonstrates a step increase of the polymer thickness around the tube. The height of the first polymer layer (above the nano-tube with 0.9 nm diameter) is 0.8 nm. However, the height of the upper polymer layer was close to 0.5–0.6 nm. This lower height of the polymer in the upper layer was explained by different packaging of the polynucleotide strand above the lower layer. So, if the polymer is located in a groove between two neighbor-ing helices, the height of this layer becomes smaller. The thickness of the bulge (i.e. the number of polymer layers) is likely dependent on SWNT:DNA concen-tration ratio [62].

MD simulations of single-, double- or triple-stranded biopolymers wrapping around the nanotube have shown that such multilayer structures are stable [62]. As follows from simulations, the multilayer polymer coating is mainly a result of interactions between the bases of neighboring polymer strands or between the base of one strand and charged phosphate groups of other one. Interaction energies cal-culation at the MP2 level of theory [62] showed that the energy value between phosphate group and cytosine is ~92.4 kJ/mol and between two cytosines ranges from 42 till 67.2 kJ/mol for different structures.

It should be noted that oligonunucleotide d(GT)20 can also wrap around SWNT (with diameter ~1.4 nm) in several layers as was also found by Toita et al. [63] analyzing the diameter of SWNT:DNA hybrids by AFM.

105

3.2.6 Influence of Adsorbed Biopolymer Structure on Optical Properties of SWNT: Double-Stranded DNA Adsorbed on the Nanotube Surface

The model of double-stranded DNA (ds-DNA) adsorption on the nanotube surface is not so intuitively understandable as in the case of ss-DNA the helical structure of which assumes polymer wrapping. Hydrophobic bases in the double-stranded polymer are located inside the double helix bound not only by π-stacking interac-tion inside the each strand and effective cross-stacking with the neighbor strand but also with H-bonds formed between strands. Thus, an ascertainment of the adsorption mechanism of this sufficiently rigid polymer on the nanotube surface needs the detailed research. However, as follows from experimental observations [6, 9, 67, 72, 101, 109–113], nanotubes form stable hybrids with ds-DNA in aque-ous suspension. As SWNTs can be readily dispersed by long salmon genomic ds-DNA, this gives a hope that the nanotube solubilization will not be very expensive [101]. It is surprisingly but nanotube debundling occurs spontaneously when the concentration of solution with ds-DNA is reduced, indicating that SWNT:DNA hybrids exist in water as a solution rather than as a dispersion [109].

Fragmented ds-DNA is less efficient in nanotube solubilization than ss-DNA [72]. This conclusion was made from comparison of emission intensities of semi-conducting SWNTs covered with fragmented ds-DNA or ss-DNA. Note that the observation of emission from semiconducting SWNTs covered with ds-DNA indicates effective stabilization of individual carbon nanotubes with ds-DNA in water.

A possible model of ds-DNA adsorption on the nanotube assumes that the for-mation of SWNT:ds-DNA hybrids starts due to the interaction between the nano-tube and untwisted ss-DNA formed mainly at polymer ends [72]. These untwisted regions are always presented in the polymer at room temperatures and are also formed after sonication (the common method used for the nanotube hybrid preparation). The comparative analysis of IR spectra obtained for the native, ss-DNA and DNA fragmented by sonication points to the presence of both ds- and ss-regions in polymer fragments [72]. Studies on the ultrasonic action on DNA in water solutions showed that DNA is fragmented and, as a result, the biopolymer changes its size and structure [114]. However, the small power of ultrasonica-tion (less than 85 W) [115] produced DNA fragments with the double-stranded structure without polymer denaturation. For example, under using the tip-actuated method (frequency 44 kHz) with ~1 W power, the length of native DNA fragments was 200–500 base pairs after 30 min sonication (typical time for the hybrid prepa-ration) [72]. Polymer adsorption on the nanotube surface starts from wrapping of single strands around the tube, which serve as an “anchor” for the whole polymer (Fig. 3.7). Our simulation of ds-DNA fragment (consisting of d(AU)15) adsorption on the nanotube surface showed that untwisted tails of d(AU)15 are really adsorbed to the nanotube surface through π-stacking of nucleotide bases and hold the whole polymer fragment near the tube, as can be seen in Fig. 3.7.



Zhao and Johnson [74] reported on MD simulation of the dodecamer ds-DNA (12 base pairs) binding to SWNT in water. It was found that ds-DNA does not wrap around the tube but rather attaches to the surface via its hydrophobic end groups.

With time, ds-DNA adsorbed to the SWNT surface can unzip and then sin-gle-stranded polymers wrap spontaneously around the nanotube. This process is facilitated by dangling ends on the ds-DNA interacting with the nanotube sur-face. Dissociation of the double polymer located on the nanotube with time was recently observed experimentally [64]. To overcome the nearest-neighbor interac-tions within the duplex, essential energy consumption is necessary, therefore, as the first step, the denaturing process involves dissociation of a small part of the polymer and, as a result, the unbound bases may contact closely with the nano-tube surface. In the next step, bases can move freely along the nanotube surface to occupy the energetically favorable position to increase the binding energy. Thus, the single-stranded polymer wraps around the nanotube spontaneously. With time DNA moves to an ordered form, and the strands become more compact. This is accompanied by elimination of the hydrophobic interface between nanotubes and the water that also leads to a significant entropy increase. As this process has an entropic cost, it proceeds very slowly. As follows from the experimental observa-tion, it takes more than 1 month [64]. Ordering of the DNA structure on the nano-tube surface was controlled by absorption and PL spectroscopes, and their spectra improvements coincide with the completion of DNA monolayer covering of the carbon nanotube. The spectra were characterized by the transformation with time from poorly defined absorption peaks to a set of sharp, well-resolved peaks of a higher intensity. The intensity of the absorption spectra increased during 35 days, while the increase of the PL intensity reached its maximum intensity by 49th day, when the nanotubes were coated fully with the polymer monolayer [64]. The authors of Ref. [64] concluded that PL spectra are more sensitive to the degree of DNA coverage than absorption spectra. Optical transition energies of SWNTs depend on dielectric properties of their local surroundings, and, as a result, the

Fig. 3.7 Snapshot of d(AU)15 adsorbed to the nanotube surface through π-stacking nucleo-tide bases of the oligomer end. Simulation time was 15 ns with 1 fs step, water molecules were extracted for better visualization

107

degree of the polymer covering and the value of the interaction between nanotubes and the surrounding water influence the absorption and PL band positions.

It is known that UV absorption of nitrogen bases is sensitive to changes in the nucleic acids structure [116]. When we employed the differential absorption spec-troscopy, an opportunity to observe conformational changes in nucleic acids appears, in particular, the helix ↔ coil transition. Upon the polymer solution heating, the base π-stacking is destroyed, resulting in the optical absorption increase (Fig. 3.8).

This DNA absorption intensity increase is called as hyperchromicity which is a common effect of stacked organic π-systems [117] (we discuss this effect in detail in the next paragraph).

A temperature dependence of the change in optical absorption of ds-DNA (ΔA) has S-like form which is characteristic of the helix-coil transition in dou-ble-stranded polymers. This temperature dependence of changes in the optical density of ds-DNA is called as the DNA melting curve. The melting curve is char-acterized by the temperature (Tm) at which ΔA reaches up 50 % of its total value (or 50 % of DNA base pairs become denatured) and the width of the transition interval (ΔT) [116]. As shown in Fig. 3.8, free ds-DNA (curve 1) has a very nar-row ΔT (4–5 °C). With ds-DNA fragmentation by sonication, the polymer length decreases and single-strand sections appear which is accompanied with low-ering Tm and broadening the temperature range ΔT of the helix-coil transition (Fig. 3.8, curve 2) [72].

Comparison of ΔA(T) dependences for fragmented ds-DNA free and adsorbed to SWNT showed that the interaction between this polymer and the nanotube sur-face shifts the melting curve to a higher temperature by 4–6 °C (the value of this temperature increasing depends on the ionic force of solution), i.e. the polymer thermal stability increased (Fig. 3.8, curve 3). The polymer thermostabilization occurs due to the interaction of DNA strands with the tube surface, which restrains the ds-DNA unzipping with the temperature growth.

fds-DNA attached to the ends of different nanotubes due to DNA sequence-specific pairing interactions or wrapping of one polymer fragment around two nanotubes simultaneously can lead to the formation of the branched structures of

Fig. 3.8 Temperature dependences of the relative change in DNA light absorption in solutions without (1, 2) and with (3) nanotubes. 1 native ds-DNA; 2 fragmented ds-DNA; 3 fragmented ds-DNA:SWNT. Absorption was detected in the maximum of DNA band absorption (at 268 nm). Reprinted from [72], with permission from © Elsevier 2006



SWNTs which was observed in AFM images [62]. In addition, compact globular structures formed by fds-DNA near the nanotube with the height up to 25 nm were also detected, which appear because of using the high weight nanotube:polymer ratio [62].

Thus, the polymer structures as well as the polymer ordering on the nanotube or the degree of the tube surface covered with the polymer have an essential influence on the photophysical properties of carbon nanotubes hybridized with the polymer.

3.3 Absorption Spectroscopy of SWNT Interfaced with DNA

3.3.1 Absorption Spectroscopy of SWNTs

Optical absorbance spectroscopy is a common and informative method for charac-terizing SWNTs with the inexpensive instrumentation exploited. Through analy-sis of the characteristic nanotube absorbance peaks, very useful information about various properties of nanotubes, such as the nanotube chirality/diameter, the con-ductivity type, excitonic transition energy, evaluation of purity nanotubes and so on, can be obtained. This method provides a simple and rapid measure of the rela-tive dispersion state, the presence of bundles or individual SWNT in suspension. Optical absorption bands in the nanotube spectrum are related to allowed transi-tions between van Hove singularities in the valence (vi) and conductive (ci) bands of the nanotube electronic density of states (DOS). These diameter-dependent sin-gularities appear due to 1-D character of the nanotube electronic structure [118]. Semiconducting SWNTs possess an energy gap at the Fermi level (Ef) while metallic species has no zero DOS at this level. DOS diagrams of both types of SWNTs are illustrated in Fig. 3.9.

Based on the tight-binding theory [118], the average nanotube diameter could be calculated from the spectrum, using the following equations: E11

S = 2αβ/d, E22

S = 4αβ/d and E11M = 6αβ/d where α is C–C bond length (0.141 nm), β is the

transfer or resonance integral between ππ-orbitals (2.9 eV) and d is the diameter of nanotubes. Tight-binding formulas suggest that the ratios of the excitation ener-gies should be E11

S: E22S: E11

M = 1:2:3 whereas both experiment and detailed calculation give: E11

S: E22S: E11

M = 1:1.7:2.4 [119]. A set of empirical equations describing how E11 and E22 frequencies correlate to semiconducting nanotube diameters can be found in Ref. [120] and for metallic nanotubes in Ref. [121].

Optical absorption of carbon nanotubes strongly depends on the incident light polarization [122, 123], for light polarized parallel to the SWNT axis the prefer-ential absorption at the band gaps was observed [124, 125]. Absorption for light polarized perpendicular to the tube axis is significantly weaker because in this case the optical transition is strongly suppressed by the depolarization effect, in which the induced charges in SWNT weaken essentially the electric field of light.

109

For semiconducting SWNTs with diameters close to 1 nm, the first transition will appear in the NIR region, and the second one is observed in the visible–NIR field. Metallic SWNTs of similar diameters will have their lowest energy optical transitions at visible wavelengths in the range between E22 and E33 features of semiconducting nanotubes.

Figure 3.10 shows absorption spectra of samples prepared using HiPCO (high-pressure carbon monoxide) SWNTs [126] and CoMoCAT (cobalt–molybdenum catalyst) tubes [127]. These nanotubes produced by chemical vapor deposition (CVD) methods are commercially available and are the most popular among researchers. HiPCO nanotubes exhibit more peaks [120, 128] than CoMoCAT SWNTs [129] because they have the larger distribution of (n,m) structures in the starting material. HiPCO SWNTs have a diameter distribution of 0.6–1.3 nm and an average diameter of 0.9 nm [120]. In contrast, CoMoCAT SWNTs have a much smaller diameter distribution (0.6–1.05 nm) and a lower average diameter of 0.8 nm [130]. Upon growth CoMoCAT nanotubes are obtained with prevailing content of (6,5) and (7,5) SWNTs with a characteristic absorption peak appearing at 989 and 1050 nm (Fig. 3.10), respectively [129].

Electronic transitions E11S in the semiconducting nanotube (the band gap tran-

sition) is observed in NIR region 800–1700 nm of the absorption spectrum of HiPCO SWNT. The earlier observed absorption spectra [22, 120] showed sepa-rate peaks corresponding to individual nanotubes of the certain chirality predomi-nant in the sample. Light absorption at 550–900 nm corresponds to electronic transitions between the second pair of van-Hoff singularities (E22

S transition) of the semiconducting SWNTs, and E11

m transition in metallic nanotubes can be observed at 400–600 nm. The width of the absorption line at the half maximum of the intensity reaches 25 meV [22, 131, 132]. Thus, UV–Vis spectroscopy can be used as an efficient spectroscopic tool for determining the average diameter of nanotube samples.

Fig. 3.9 Schematic presentation of DOS of semiconducting (a) and metallic (b) SWNTs. Arrows indicate the optical transitions between pair of van Hove singularities in valence and con-duction bands. Electronic transitions (Eii) are allowed between bands of the same index



In the range of shorter wavelengths (less than 500 nm) absorption peaks of metallic nanotubes are superimposed over these absorption peaks of semiconduct-ing nanotubes, induced by electronic transitions between third or firth pairs of van Hove singularities. As follows from Fig. 3.10, HiPCO sample contains many different chiralities, and the spectrum is a superposition of absorption spectra of different (n,m) nanotubes.

Resonant transition peaks in the absorption nanotube spectrum are arranged onto a slope of the broad and nearly featureless background. The origin of this background is associated with a number of reasons among which near-ultravio-let π-plasmon resonances associated with collective excitations of π-electrons of nanotubes [133–136], and carbonaceous impurities (amorphous carbon, graphitic particles, etc.) give the most contribution. A broad band of π-plasmon resonances is located at around 275–206 nm (4.5–6.0 eV). Reed and Sarikaya performed elec-tron energy loss spectroscopy (EELS) measurements on purified SWNTs [134] and explained peaks at 4.2–4.5 eV and 5.2 eV as surface and bulk π-plasmon exci-tations, respectively.

Murakami et al. [135, 136] studied anisotropic optical absorption properties of SWNTs and determined from a vertically aligned SWNT film that absorption peaks at 4.5 and 5.25 eV exhibit remarkable polarization dependence and can be related to optical properties of graphite. Peaks observed at 4.5 (276 nm) and

Fig. 3.10 Absorption spectra of HiPCO (in bundles) and CoMoCAT (solubilized with DNA as surfactant) SWNTs. Both prominent peaks in CoMoCAT spectrum can be assigned to E11

S (989 nm) and E22

S (575 nm) transitions of (6,5)-nanotubes, respectively. S1 and S2 mark area of absorption band of nanotubes with maximum at 575 nm and area of background, respectively

111

5.25 eV (236 nm) in spectra of aligned nanotubes have the maxima when polari-zation of the incident light is parallel to SWNT axis or perpendicular to the SWNT axis, respectively.

Recently, employing UV–Vis spectroscopy, Rance et al. [137] studied π-plasmon absorbance of a series of SWNTs. Using complementary experimen-tal and theoretical approaches, they deduced that π-plasmon absorbance correlated with the nanotube diameter (dNT). The relationship between the energy (Ep) and the nanotube diameter was described as Ep = 4.80 + 0.70/(dNT)2.

3.3.2 Absorption Spectra Analyses of SWNT Composition

Absorption spectroscopy from the UV–Visible region to the NIR one (UV–Vis–NIR) can be used to obtain quantitative information on SWNT composition. An analysis of highly resolved spectral features in mixed samples allows to assign each band in the spectrum to the electron transition of certain SWNT. However, this assignment based on the absorption spectrum presents some difficulties, one of them is the influence of the nanotubes environment on the spectral position of narrow peaks and their width. The first and striking example of the environ-ment influence is nanotube bundling. SWNTs aggregates are formed at nanotube growth due to the strong vdW interactions between nanotube sidewalls contacts (~500 eV/μm) [22]. Bundling obscures the SWNT resolved absorption spectrum which exhibits severe inhomogeneous broadening which can be attributed to the energy states mixing of different nanotube structures [133]. Aggregation is accom-panied by the red-shift of resonant absorption peaks (~70–150 meV, depending on nanotube chirality) [22, 138–141]. As a result, nanotubes are soluble weakly in all organic solvents and insoluble in water at all. To solubilize SWNTs in aqueous environments, surfactant is added to solution and then the ultrasonic dispersion process is used to induce bundle splitting, allowing surfactant adsorption onto the nanotube sidewall. However, we should remember that the extensive tip sonica-tion of nanotube solutions with the surfactant can induce the appearance of the background in the absorption spectrum. This additional background rises strongly toward shorter wavelengths.

To provide an appropriate interface between SWNT and solvent or to remove impurities, the nanotube sidewall can be functionalized partly, however, in this process nanotube unique electronic properties vanish. Thus, covalent functionali-zation eliminates resonant absorption peaks and leaves a broad featureless spec-trum, so, for optical studies SWNT non-covalent functionalization is the preferred solubilization method.

An unknown part of carbon contaminations in some nanotube samples intro-duces errors in the quantitative evaluation of nanotube species. As was shown by Itkis et al. [142], a solution-phase visible–NIR absorption spectroscopy can be used to provide a rapid, quantitative procedure for evaluation of the carbonaceous purity of bulk quantities of as-prepared SWNT soot. The procedure starts with



preparation of nanotube aqueous suspension with surfactant and evaluation of the purity by utilizing the region of the second transition (E22

S) for semiconducting SWNTs.

This spectral method of the nanotube purification control is based on estimation of the ratio γ = S1/S2 where S1 and S2 correspond to the area of E22

S band and to the area of the background in the absorption spectrum, respectively (Fig. 3.10). It was showed that a higher ratio corresponds to higher nanotube purity [142]. As an example, we demonstrate now how γ quotient changed for spectra of aqueous suspensions of CoMoCat nanotubes with DNA after sonication and centrifugation (18000 g, 21 min), analyzing supernatant and precipitated parts. Absorption spec-tra of three nanotube suspensions are shown in Fig. 3.11. Quotient γ = 0.033 was obtained for the nanotube band at 575 nm after sonication, but after centrifuga-tion this value increased till 0.051 in the supernatant. However, in the spectrum of the precipitated sample this quotient was 0.014 that indicated efficiency of simple and quick SWNT purification by the sonication/centrifugation method [143]. To obtain purified SWNTs, supernatant suspension can be deposited onto the filter and washed out by distilled water to remove the rest of the surfactant.

The correct SWNT spectrum analysis in the visible–NIR ranges is complicated because of broad and nearly featureless background that is more intense than the bands to be quantified. Thus, a right correction for the background absorption is a critical step in the determination of relative magnitudes of specific (n,m) peaks. The background increases toward shorter wavelength and is caused by extrinsic and intrinsic contributions. The extrinsic factors include backgrounds induced by ultra-sonication of SWNTs, carbonaceous impurities, contributions from chemical func-tionalization, and spectral broadening from nanotube bundling. The intrinsic factors are mainly caused by spectral congestion and metallic SWNT contributions [144].

Fig. 3.11 Absorption spectra fragments of CoMoCat nanotubes aqueous suspension with DNA. Solid line corresponds to the spectrum observed after sonication, dashed and dotted lines correspond to spectra obtained after centrifugation for supernatant and precipitate, respectively. For comparison, the spectra of nanotubes in supernatant and of precipitate were scaled by 1.15 and 1.3, respectively

113

For the background absorbance at the certain wavelength, Abkg(λ), due to col-loidal graphite and π-plasmon absorption, the following functional form was assumed Abkg(λ) = k/λb, where b is a real number and k is an empirical fit param-eter [145, 146]. Recently Naumov et al. [144] suggested the three-parameter expo-nential form A(λ) = a(y0 + exp(−bλ)) where the amplitude parameter a depends linearly on the particle concentration with Beer’s Law proportionality constants. The parameter b depends on the form of amorphous carbon which, however, can be fixed for the certain type.

Typical covalent functionalization of SWNTs induces the conversation of car-bon atoms from sp2 to sp3 hybridization that removes electrons from the delo-calized π-system and introduces localized electronic perturbations. As a result, resonant absorption peaks become wider that leads to a broad featureless spectrum [147]. Such an absorption spectrum transformation is determined by a degree of extensive covalent functionalization.

Extensive tip sonication of surfactant solutions can cause the formation of small particles presumably through chemical reactions of surfactant during sol-vent cavitation. Although the sonication-induced background can be suppressed by centrifugation, nevertheless, the additional background appears strongly toward shorter wavelengths. Small red shifts of E22 and E11 peaks (∼3 nm) induced by the processing were also observed [144].

Metallic nanotube content in the sample increases significantly the absorption background. In contrast, well-dispersed SWNT samples enriched with semicon-ducting (n,m) species demonstrate nearly background-free absorption spectra [40]. Possibly, broad absorption background in metallic nanotubes is caused by the long tail of the π-plasmon resonance which may be more intense in metallic SWNTs than in semiconducting ones [134]. Another possible reason is E01

M transitions which would appear at longer wavelengths than E11

M in metallic nanotubes [148]. Such symmetry-forbidden transitions can be observable in chiral metallic SWNTs in which the axial symmetry might lead to subbands of mixed angular momen-tum. Note that these transitions have no counterpart in semiconducting nanotubes. Naumov et al. [144] assessed the backgrounds caused by metallic nanotubes, using the exponential function, A = aexp(−bλ) with fixed b = 0.00155 nm−1. It was observed that the amplitude parameter a had a nearly linear correlation with the concentration of metallic SWNTs. For variety of well-known SWNT sources, metallic concentration has been determined [149], and thus, if the total SWNT mass concentration of a sample is known, the background contribution of metallic nanotubes can be estimated.

Nanotube aggregation into bundles is also accompanied with the increase of the absorption background, mainly, due to broadening and red-shift of resonant absorption peaks, which leads to the increased spectral congestion. SWNT bun-dling in typical polydisperse samples can cause a factor of 2 decreases in peak-to-valley absorbance ratios [144]. Aggregation complicates deconvolution of the absorption spectrum of SWNTs into distinct (n,m) species, especially it concerns nanotube samples containing relatively broad (n,m) distributions because transi-tion energies are closely spaced.



An algorithm that performs deconvolution for the entire HiPCO SWNT absorp-tion spectrum using Voigt line shapes was suggested by Strano and coworkers [146]. The simulations rely on the knowledge of peak locations from the spectral assignment of metallic and semiconducting nanotubes obtained earlier [120, 150]. Line widths for metallic (ΔE11

M = 93.42 meV) and two semiconducting regions (ΔE22

S = 57.96 meV and ΔE11S = 29.86 meV) were obtained from the absorp-

tion spectra of DNA-wrapped SWNT fractionated by ion-exchange chromatogra-phy [146]. For SWNT:SDS spectra a value of 25 meV of the line width for ΔE11

S was obtained too. It is supposed that this difference in the line width is caused by the surfactant effect which might affect E11

S more than E22S or E11

M. Similar val-ues of the line width for nanotubes covered with this surfactant were observed ear-lier [22, 132, 151]. Reasonable fitting of CoMoCAT SWNT absorption spectrum gave ΔE11

S = 60 meV [146]. This much higher value could be due to the pres-ence of inhomogeneous nanotube bundles [146]. Note that, to provide a reasonable spectrum deconvolution, the E11

S region should be selected because it contains more well-defined, intensive peaks located in a larger energy range in comparison with E22

S or E11M regions.

Another quantitative methodology for unknown (n,m) abundance SWNT samples was suggested, based on NIR absorption and photoluminescence data obtained in aqueous suspensions of individually dispersed SWNTs [152]. The obtained results indicated that diameter distributions of CVD-grown nanotube samples obey log-normal distribution using of which E11

S absorption spectrum was reconstructed with two adjustable parameters in conjunction with the theoreti-cally derived (n,m)-dependent extinction coefficients.

UV–Vis–NIR spectroscopy is a simple, efficient and accurate method which can be used to evaluate quantitatively the ratio of metallic to semiconducting nanotubes (M/S ratio) in bulk samples. Knowledge and control of these ratios are particularly important for many SWNTs electronic applications which demand highly refined samples of metallic or semiconducting nanotubes. For example, elaboration of field effect transistors needs semiconducting SWNTs [105, 106, 153, 154], and, on the contrary, metallic SWNTs can be used as wires in nanoscale circuits [155] or for production of low cost flexible transparent conductive films (see, for example, Ref. [156] in which DNA was used to prepare such a film). Unfortunately, current methods of SWNTs syntheses afford only a mixture of metallic and semiconducting nanotubes, and often these two types must be sepa-rated before they can be applied.

To identify metallic and semiconducting SWNT structures present in bulk soot suspensions [157] and purified SWNT solvent dispersions [158–161], UV–Vis–NIR absorption spectroscopy was often used. Recently, a quantitative method to evaluate the metal-to-semiconductor ratio in bulk SWNTs using optical absorption spectroscopy has been described [157]. Obtained SWNT film absorption spec-trum was reproduced using weighted sum of the spectra corresponding to high purity metallic and semiconducting SWNTs which were obtained from bulk sam-ple through density centrifugation process. After such reproduction M/S ratio in SWNT film was determined. Another efficient method for generalized M/S ratio

115

evaluation of separated SWNTs has been proposed recently [160], measuring only the UV–Vis–NIR spectra of mixed solutions with different ratios of sepa-rated metallic-rich and semiconducting-rich SWNT samples. A procedure of the M/S ratio determination in as-produced nanotubes was suggested, which involves acquisition of UV–Vis–NIR absorption spectra of solvent-suspended sam-ples [161]. The key aspect is the subtraction of the π-plasmon absorbance from the overall absorbance. The total π–plasmon absorbance was calculated using π-plasmon peak parameters and the corresponding equation was suggested. The proposed procedure can be also applied to soot samples that possess non-overlap-ping E22

S or E11M spectral features.

3.3.3 Comparison of Absorption Spectra of SWNTs Covered with DNA or Surfactants

Using CoMoCAT SWNT sample with a narrow distribution of (n,m) structures, Tan and Resasco [162] developed a method for quantifying dispersability of nanotube samples from their optical absorption spectra in terms of two ratios: the “resonance ratio” and the “normalized width.” The resonance ratio is defined as γ quotient introduced before (as the ratio of the resonant band area and its back-ground) (Fig. 3.10). The normalized width is determined as the ratio of the width of the resonance band at half-height to the peak height on a spectrum. The authors used several surfactants with high debundling ability, and after spectra analyz-ing they concluded that the presence of ring structures in the hydrophobic tail as well as the charge of the hydrophilic head play important roles in the dispersion process.

Recently Haggenmueller et al. [59] analyzing UV–Vis–NIR absorption behav-ior of CoMoCAT SWNT solutions compared the abilities of different surfactants including DNA to suspend nanotubes. To quantify the surfactant efficiency, they compared the original SWNT concentration used for suspensions (0.5 mg/mL), divided by the concentration after sonication and centrifugation for each system. Such surfactants as sodium deoxy cholate (SDOCO) and carboxymethyl cellulose (CBMC) reached the highest solubilization efficiencies of the surfactant systems studied (~60 %). SDS and SDBS demonstrate reduced efficiencies as compared to SDOCO and CBMC (~50 %). Sodium cholate (SCO) and chitosan have ~45 % efficiency. The ss-DNA samples with GT, AC, and cytosine (C) bases had compa-rable efficiencies of ~30 %, independent of the chain length. SWNT:DNA suspen-sion with thymine (T) base demonstrates a decrease in efficiency with the polymer strand length decreasing (T30, T20, T10 were studied). This result is in contrast with observation of Vogel et al. [110] which found a maximum amount of solubi-lized HiPCO nanotubes when a mixture of two hexamers d(AC)3 and d(GT)3 was used. Earlier Zheng et al. [7] also concluded that, among homopolymers, poly(T) had the highest dispersion efficiency. They also found that T30 gave the highest yield of solubilized nanotubes in comparison with oligomers of different lengths



(20-, 30-, 20-, and 15-m). In our opinion, a possible reason of this discrepancy lies in different conditions of nanotube suspension preparation: first of all, the sonica-tion duration, power applied, tip or bath processing.

The spectral resolution of optical absorption spectra provides an estimate of the quality of SWNT suspensions too. The change in the peak-to-valley ratio reflects a change in nanotube aggregation [59, 142, 144]. Thus, well-resolved peaks in the absorption spectra indicate better SWNT exfoliation while a poor peak resolution certifies SWNT bundling. Remember that debundling reduces electronic intertube coupling which obscures the fine structure in the spectra. High-resolved peaks assigned to the band gap absorption (E11

S) of the semiconducting nanotube are present for all ss-DNA but the polymer with GT sequence demonstrated the most sharp and separated peaks [59]. Chain length decreasing reduces slightly the peak quality. Peaks in nanotube spectra with A30 and G10 are distinctively weaker and broader than with GT or AC. SDOCO shows peaks resolved better than those for (GT)15 while peaks for SCO suspensions are weaker and broader than those for (GT)15. Chitosan and SDBS show somewhat broader peaks which are still sepa-rated. Poor peak resolution was observed for suspension with SDS [59].

The ratio between the areas of the absorbance for E11 of the semiconducting nanotube and the baseline absorbance where absorbance is minimal (e.q. for CoMoCat nanotubes it can be E11 of (6,5) SWNT and the baseline absorb-ance at ~905 nm) may be also a quantitative indicator of suspension quality. Haggenmueller et al. [59] showed that SDOCO, (GT)15, and (GT)10 have the highest ratio of ≈2.45 indicating the best SWNT flotation in these suspensions (Fig. 3.12a). ss-DNA with base sequence AC or with C has a ratio of ≈2.2, T shows a decreasing ratio with a decreasing polymer length. This ratio is higher for A30 than those for A20 and A10. SDBS, SCO, CBMC, and chitosan have com-parable ratios of ≈2.0, while among them SDS has the lowest ratio (~1.6). For comparison, PL intensities of (6,5) and (7,5) nanotubes covered with various sur-factant systems are presented in Fig. 3.12b. SDOCO dominates in this surfactant comparison of PL intensities, which differs from comparison of absorption intensi-ties where this surfactant has a like intensity value with biopolymers (Fig. 3.12a).

The real quantitative evaluation of the nanotube concentration is hampered because of several reasons some of which are of an intrinsic feature but other ones are caused with extrinsic factors [144]. We have already discussed extrinsic nano-tube effects such as sonication power effect, possible chemical functionalization, bundling and amorphous carbon impurities which complicate quantitative analysis of the absorption spectrum. Moreover, absorption matrix elements of nanotubes depend on (n,m) nanotube chirality that also introduces an additional error into the quantitative evaluation [163–165]. We note also that the data for extinction coefficients of different SWNTs (20–50 mL/mg−1 cm−1) presented in literature [59, 159, 166–169] differ essentially, and better measurements on the extinction coefficient of different (n,m) nanotubes are still necessary. Besides, it should be added that extinction coefficients of metallic SWNTs differ from those of semi-conducting nanotubes, so M/S ratio of these extinction coefficients was estimated as 0.352 [170] although Miyata et al. [157] found that this ratio is close to 1.

117

Recently, properties of various SWNTs produced by arc-discharge, CVD, and HiPCO methods without purification after their dispersion in aqueous solution assisted by DNA (oligomer (dT)30 was selected) were compared [42], employing UV–Vis–NIR absorbance measurement and AFM imaging. Based on systematic studies on optimizing dispersing conditions to maximize SWNT solubility and remove insoluble materials, low power and short sonication time were recom-mended to disperse SWNT with keeping their average lengths. In addition, AFM images showed that SWNT produced by arc-discharge method demonstrated the highest nanotube purity (better than HiPCO SWNTs). It was also revealed that even at the highest sonication power used (the sonic bath cleaner with 90 W power

Fig. 3.12 Comparison of absorption intensity ratio of (6,5) SWNT peak versus absorbance intensity of the baseline at ~905 nm (a) and of PL intensity (b) (left/right bar corresponds to (6,5)/(7,5) nanotubes, respectively) for various surfactant systems. Reprinted with permission from [59]. © American Chemical Society (2008)



setting) (dT)30 was not fragmented. Although SWNT raw materials are very inho-mogeneous, upon dispersion into aqueous solution, they form uniform solutions which permit dilution of the dispersed SWNT in the wide concentration range.

Optical transition energies for nanotubes wrapped with DNA are red-shifted by 10–20 meV [48, 59, 66, 171] in comparison with the optical transition ener-gies for nanotubes covered with such surfactants as SDS or SDBS. If the sur-factant concentration is above the critical micelle concentration, the surfactant molecules form micelle in water around the nanotube, covering the most of the nanotube surface. At the same time the DNA coverage depends on the polymer ability to adsorb on a nanotube in a stable helical wrapping conformation which is determined by SWNT chirality too. All this can lead to the incomplete coverage of the nanotube surface with the oligonucleotide comparing to surfactants, which provides an accessibility of the SWNT surface to water molecules. AFM images of SWNT:DNA hybrids demonstrate the nanotube surface free of polymer that confirms this explanation. The SWNT contact with water increases the effective dielectric constant of the medium surrounding the nanotube [147, 172], this leads to a decrease in the SWNT electron transition energy. The red-shift of the electron transition may also appear as a result of stronger polymer binding to the nanotube surface as compared to surfactant molecules.

3.3.4 Peculiarities of SWNT and DNA Interaction Revealed in Absorption Spectra

π–π interaction between the nanotube sidewall and organic molecules or poly-mers possessing the π-aromatic polycyclic moiety plays an important role in the development of SWNT applications ranging from nanoelectronics to biomedical devices. π–π interactions are caused by intermolecular overlapping of p-orbitals in π-conjugated systems [116]. In aqueous solution the stacking binding of flat π-conjugated molecules is also influenced by an additional hydrophobic inter-action the driving force of which is the entropic factor. The stacking interaction influences the absorption spectrum of binding molecules the intensity of which decreases because of changes in electronic interactions between compounds. In literature this effect is known as hypochromism (the opposite effect is called as hyperchromism) [116, 117]. Theory considers hypochromism as a result of weak dipole–dipole interactions between stacking chromophores modified by the light wave. The well-known example is DNA, UV absorption of which decreases when the duplex structure is formed by two single strands and increases under the helix → coil transition [117].

Hughes et al. [173] have recently studied the UV–Visible absorption spec-tra of thirty-base-long homooligonucleotides wrapped around SWNT in aqueous suspension. Absorption spectra of different homopolymers in the UV range from 200 to 300 nm were altered significantly after adsorption on the nanotube surface. The problem of this spectrum analysis consists in the superimposition of spectra

119

of hybrid components in this spectral range. The polymer absorption spectrum is mainly determined by absorption of nitrogen bases (see for example [116]). To establish the contribution of attached DNA homopolymers to the overall absorb-ance, the absorbance due to the unbound SWNT was subtracted from the spectra of the hybrids. The differences between absorption spectra of the free homopoly-mers ((dA) 30, (dC)30, (dG)30, and (dT)30) and those interacting with SWNTs are shown in Fig. 3.13. It should be noted that there are significant differences both in the intensity magnitude and peak position of optical dipole transitions when oligomers are free and bound. The shape of the absorbance spectrum for (dA)30 looks similar both in the bound and free states, however, the absorbance peak was red-shifted by about 10 nm in the last case.

At the same time (dC)30 showed disappearing of absorbance in 220–240 nm region when oligomer is bound to SWNTs. When (dG)30 interacts with SWNTs, redistribution in the intensity between peaks is observed in the polymer spectrum. On the contrary, it turned out that (dT)30 is mostly unchanged when coupled with SWNTs. Differences in absorption spectra of these four samples were explained with anisotropic hypochromicity of transitions in the oligomer bases. For homoo-ligomer transitions which induced dipole moments align with the nanotube axis, a

Fig. 3.13 Absorbance spectra for each bound homopolymer (bold) and each free one (thin). Insets show the direction and wavelengths of optical dipole transitions. Vertical lines indicate the position and relative intensities of base optical dipole transitions. Reprinted with permission from [173]. © American Chemical Society (2007)



strong hypochromicity was observed while, for those transitions that align perpen-dicular to the nanotube axis, the hypochromicity was suppressed (Fig. 3.13) [173].

The π–π stacking interaction between the nanotube and π-conjugated nitro-gen bases manifested directly not only in DNA absorption spectrum but can be also observed in the absorption spectrum of polymer-wrapped SWNTs [174]. Figure 3.14 shows a fragment of the absorption spectrum of SWNT:poly(rC) hybrid in 200–600 nm region. For comparison, the absorption spectrum of SWNTs aqueous suspensions with SDS was presented in Fig. 3.14 too. Absorption spectra of the samples studied are similar in the range of 400–1100 nm but differ in the intensity which is conditioned with different nanotube concentrations in aqueous suspensions. As SDS absorption in UV region begins at the wave length less than 200 nm, therefore, the spectrum of SWNT:SDS aqueous suspension observed in 200–300 nm range is caused by nanotubes absorption. On the contrary, the absorp-tion spectrum of SWNT:polymer suspension in this range is a result of superim-position of nanotubes and polymer absorption spectra. To compare the spectra obtained, the spectrum intensity of the polymer-wrapped nanotubes was scaled to their spectrum intensity in SDS environment (Fig. 3.14) using the multiplier (wavelength independent). In this case, a small shift of nanotube spectral peaks, induced by various environments, can be neglected.

It is seen from Fig. 3.14 that the scaled optical density of SWNT:poly(rC) sample in 300–400 nm range is somewhat lower than for SWNT:SDS ones. It should be noted that the value of the spectra discrepancy increases monotoni-cally with the wavelength decreasing. The spectrum of nanotubes interacting with the polymer is obtained by subtracting the poly(rC) spectrum from that of SWNT:poly(rC). This subtraction can be described by the following expression:

Fig. 3.14 UV-Visible absorption spectra of aqueous suspensions: SWNT:SDS (dash) and SWNT:poly(rC) (bold), absorption spectrum of poly(rC) (dotted), differential spectra (ΔA) of SWNT:poly(rC)-poly(rC) (thin) and SWNT:poly(rC)-SWNT:SDS (thick). Intensity of SWNT:poly(rC) spectrum was normalized to SWNT:SDS one in 500–800 nm spectral range. Reprinted from [174], with permission from © Elsevier 2012

121

ANTi = APNT(λ) − AP(λ) where ANTi(λ) is the spectrum of nanotubes interacting with the polymer, APNT is the spectrum of the nanotubes:polymer sample and, AP is that of the unbound polymer absorption. The differential spectrum is shown in Fig. 3.14 too. At the wavelength less than 400 nm the differential spectrum has a weaker absorption intensity than the SWNT:SDS one. The absorption decrease is mainly caused by hypocromism of nanotubes induced by their interaction with the polymer or, indicating precisely, with nitrogen bases (with cytosine in the given sample).

For HiPCO nanotubes 22 bands are located in 300–400 nm range [120], induced by E33 and E44 electronic transitions in semiconducting nanotubes of different chiralities and diameters (from 0.757 to 1.201 nm) (Fig. 3.15). As well, in 300–350 nm range absorption bands should appear, caused by E22 transition in metallic nanotubes. But most likely, the contribution of these tubes into the hypochromic effect observed can be weak because of possible quick dissipation of the induced dipole moment because of the collectivized nature of electrons in metallic nanotubes. For comparison, the differential spectrum (SWNT:poly(rC)—SWNT:SDS) is shown in Fig. 3.14 too, which is the spectrum of the polymer inter-acting with the tube. However, in this spectrum a negative fragment was observed for the wavelength less than 300 nm which was caused by a weaker absorption intensity of nanotubes interacting with the polymer, in comparison with the sub-tracted spectrum of nanotubes (with SDS). Note that in these measurements a red-shift (6 nm) of the band maximum of poly(rC) bound to SWNT (Fig. 3.14) relative to the band maximum of the unbound polymer (at 273 nm) was observed too.

Quantitatively, the value of SWNT hypochromic effect can be described by the hypochromic coefficient determined as K(λ) = (ANTi(λ) − ANT(λ))/ANT(λ) = (APNT(λ) − AP(λ) − ANT(λ))/ANT(λ) (where ANT(λ) is the spectrum of “bare”

Fig. 3.15 Scheme of energy levels and electron transitions of cytosine and SWNT. Si denotes singlet levels of cytosine, cytosine and Eii

S and EiiS and Eii

M represent electron levels of semi-conducting and metallic carbon nanotubes, respectively. Curve H(λ) shows growing (schematic) of the hypochromic effect as energy levels of nanotubes are approached towards cytosine singlet levels. Reprinted from [174], with permission from Elsevier



nanotubes (in SDS environment)). However, a reliable determination of K(λ) value can be deduced at wave lengths longer than 300 nm where absorption spectra of nanotubes and the polymer are not overlapped. As follows from Fig. 3.16, the coef-ficient K(λ) begins to grow noticeably from 500 nm and reaches ~(–0.1) value up to 300 nm. Such a monotonous growth of the K(λ) absolute value upon the wavelength decrease can be explained by the interaction increasing between electronic levels of nanotubes and cytosine as the levels approach each other. Comparison of energy lev-els of nanotubes and cytosine is shown in Fig. 3.15. K(λ) value depends on the type of the polymer binding to the nanotube, spectral dependences of this coefficient for nanotubes covered with poly(rG), ss-DNA and ds-DNA are presented in Fig. 3.16. The least hypochromic coefficient was found for nanotubes coated with poly(rG), and the greatest K(λ) was observed for SWNT:ss-DNA hybrid. K(λ) value of nano-tubes covered with poly(rG) is by about 30–40 % lower than that for SWNT:poly(rC) suspension. Thus, in spite of the fact that the energy of purine bases interactions with the nanotube surface is higher than that of pyrimidines [89–91], the small K(λ) value was obtained for poly(rG). It is explained by a lower number of guanines stacked with the nanotube surface comparing to cytosines (see Sect. 3.2.4 part). Pyrimidine polymers possess higher flexibility of the strand due to a weaker self-stacking inter-action [82] that enables them to arrange on the nanotube surface in such a confor-mation which provides more effective stacking interaction. Most likely, just the alternating sequence of purine and pyrimidine bases proves more effective interac-tion of the polymer with nanotubes. This suggestion is confirmed with the observa-tion of the best nanotube separation with a repeating sequence of alternating G and T when an anion exchange column was used [9]. The aforesaid was supported with spectral studies of SWNT:ss-DNA suspensions. The hypochromic coefficient value for this sample was about 40 % higher than for SWNT:poly(rC) suspension. At the

Fig. 3.16 Spectral dependences of the hypochromic coefficient determined for nanotubes covered with poly(rC) (2), poly(rG) (4), ss-DNA (1) and ds-DNA (3). Shadow region indicates the spectral range in which reliable determination of K(λ) is hampered because of absorption spectra of nanotubes and the polymer overlapping. Reprinted from [174], with permission from Elsevier

123

same time K(λ) value for SWNT covered with ds-DNA was about twice lower than that with ss-DNA and its curve K(λ) is placed between curves corresponding to K(λ) for poly(rC)- and poly(rG)-covered nanotubes. A comparatively small K(λ) value for nanotubes with ds-DNA was conditioned with the base arrangement in the polymer at which they are hidden into the double helix, and the formation of base stacking with the nanotube surface is mainly possible at untwisted polymer ends. Note that the hypochromic effect observed is the averaged characterization of biopolymer π–π interaction with all types of SWNTs which differ in lengths and chiralities.

3.3.5 The Effect of ss-DNA Helical Negative Potential on the SWNT Electronic Spectrum

Theoretical modeling predicts [175–178] that the Coulomb potential created with ss-DNA regular helical wrapping around the nanotube causes changes in the band structure of nanotubes, which should be taken into account upon interpreting experimental data.

In the framework of the continuum approximation Michalski and Mele [177] showed that changes in the band structure of nanotubes depend on the strength of this potential and on the dimensionless geometrical parameter, P, which is the ratio of the nanotube circumference to the pitch of the helix. They found that the minimum band gap of a semiconducting nanotube is reduced by a helical poten-tial, for each one there exists an optimal P that produces the biggest change in the band gap. This potential reduces Fermi velocity in metallic nanotubes, and in the case of strong fields two small gaps appear at Fermi surface in addition to the gapless Dirac point. However, the authors noted that the predicted effects of the helical potential are small (estimations gave about 0.01 meV difference between the original and perturbed band gaps) to be detected under typical conditions, and several methods for increasing the size of these effects were suggested.

Employing the empirical tight-binding theory to investigate the symmetry breaking and modulation of SWNT electronic structure in the field of an ion-ized DNA, Rotkin et al. [175, 176, 178] computed the polarization component of hybrid’s energy of cohesion, which was estimated as 0.5 eV for DNA base. They showed that Coulomb potential of the regular helix can lower the symmetry of the SWNT band structure. This causes a little change in the SWNT absorption spec-trum obtained in the parallel polarization, whereas in the cross-polarized absorp-tion of SWNT:ss-DNA hybrid a new peak appears at a frequency lower than that of E12 transition in the bare nanotube. Thus, upon DNA hybridization with non-chiral SWNTs, changes in absorption spectra or in optical circular dichroism spectra of SWNTs may be observed when perpendicular (or circular) polarization of the incident light with respect to the tube axis will be applied. This effect was explained by nanotube’s electrons (or holes) polarization, induced by a permanent dipole directed across the nanotube which was created by the transverse electric field of DNA phosphate groups. This dipole may be excited by the incident light the electric field of which has the perpendicular direction to the nanotube axis.



As a result, the (negative) electron density shifts to the opposite side of SWNT, whereas the (positive) hole density moves towards the DNA backbone. Another important consequence of breaking SWNT symmetry, induced by helical per-turbation is opening of gaps in the energy spectrum of metallic armchair carbon nanotubes and shifting of Fermi points, which depend on the nanotube and helical parameters.

Recently, upon field-effect transistor measurements [68] in which SWNT: ss-DNA hybrids were deposited by dielectrophoresis across pairs of electrodes, a transition from a metallic one to a p-type semiconductor after helical DNA wrapping was observed. It was found that water molecules play the key role in the activation of the transition. A reversible transition between the metallic and semiconducting behavior of the transistor was demonstrated through repeated hydration and dehydration of the polymer. The authors expected that the proper-ties of SWNT:ss-DNA hybrids are influenced by the interaction between DNA and the surrounding water molecules. They believe that a band gap can open up in ss-DNA-wrapped metallic SWNTs in the presence of water molecules due to the charge transfer between SWNT and ss-DNA as well as a helical perturbation resulting from ss-DNA wrapping. According to their first-principles calculations, the energy band gap opens up by ∼30 meV in the metallic (6,6) SWNT [68].

Recently Bobadilla and Seminario [69] performed the classical MD simulations of the SWNT:DNA wrapping process with following analysis of structural con-formations by first-principles electronic structure methods studying small diameter nanotubes (zigzag (4,0)). A reversible semiconductor-metallic behavior was found in SWNT:DNA hybrid structures. It was shown that a DNA base is able to polarize the carbon nanotube electronic density in the region lying near the DNA base with respect to the nanotube axis and in water conditions electronic density induced in carbon nanotube surface by the DNA base is higher than in vacuum ones. They revealed the shift of HOMO and LUMO toward lower energy levels induced by DNA wrapping and due to existence of highly polarizable electrons in SWNT. Changes in the electronic structure of SWNT occur due to the electronic symmetry breaking of nanotube after DNA wrapping. The authors expect a stronger DNA transistor-like gating mechanism effect for SWNTs with higher polarizability.

3.4 Photoluminescence of Semiconducting SWNTs: The Influence of Environment

3.4.1 Emission Properties of Semiconducting SWNTs

It is well-known that the electron transitions between valence and conductive bands in semiconductors occur not only with photon absorption but with light emission too. For more than 10 years from SWNTs discovery, emission from them could not be detected because of the nanotube bundling. In bundles which

125

contain both metallic and semiconducting species the rapid excitation transfer from semiconducting to metallic nanotubes originates. Through non-radiative recombination, metallic SWNTs act as non-radiative channels for luminescence of semiconducting tubes. Discovered by O’Connell et al. [22] in 2002, band-gap fluorescence was observed in aqueous micelle-like suspension of SWNTs after of nanotube bundles splitting using sonication treatment with surfactants (SDS) with following ultracentrifugation. Micelles formed by surfactant molecules around nanotubes prevented the tubes from regenerating bundles. The decanted nanotube supernatant was homogenous suspension which is stable for many months.

An electronic band structure for a typical semiconducting nanotube is shown in Fig. 3.9a, in which electron transition with photon emission is presented too. As the molecular luminescence originates from the lowest-lying electronic state (in accordance with Kasha’s Rule [179]), SWNT emission is observed exclusively for E11

S transition between pairs of peaks corresponding to van Hove singularities in conduction and valence bands. Light is initially absorbed at higher energies (e.g. at E22 as shown in Fig. 3.9a, v2 → c2 transition) to promote electron excitation. In such a case fast nonradiative electron relaxation through phonons to the lowest electronic level takes place with the following radiative transition across the semi-conducting band gap. Just the same scheme is often employed to excite emission from semiconducting SWNTs.

Emission from semiconducting SWNTs is observed in the NIR range. To study emission from dispersed semiconducting nanotubes, solid-state diode lasers with the fixed-wavelength in visible or red regions with power not above ~50 mW can be exploited.

PL from isolated nanotubes is richly structured and is similar to the absorption spectrum in the range of 900–1500 nm [22]. Peak positions and their intensities are characteristic of the distinct (n,m) nanotube species. To investigate emission from semiconducting nanotubes, deuterium water is often used for their aqueous suspension preparation because absorption of usual H20 in 1350 nm field caused by stretching overtone of O–H group can result in emission photon reabsorption. Isotopic replacement in D2O is accompanied by the red shift of this vibration band to 1900 nm, and nanotube emission without any interference can be observed [22].

Figure 3.17 shows the emission spectrum of semiconducting DNA-wrapped CoMoCat nanotubes with prevailing content of (6,5) SWNT, which is charac-terized by PL peak at 994 nm. Structured emission spectrum has a set of peaks which is nearly coincident with peaks in the absorption one (Fig. 3.17). Bands in the emission spectrum are red-shifted by about 5–8 meV relatively to those in the absorption spectrum that indicated a very small Stokes shift [22, 163, 180]. Due to the wider nanotube diameter distribution, PL from HiPCO nanotubes (Fig. 3.17) demonstrates more peaks than CoMoCAT SWNTs. The band shape of the nanotube PL from individual SWNT species was often described with nearly Lorentzian function but in bulk samples the PL spectrum is better mod-eled by Voigt profiles [180], which allowed estimating the relative contributions of Lorentzian and Gaussian components using the parameter μ that ranges from 0 for pure Gaussian to 1 for pure Lorentzian. Recently, Rocha et al. [181] have



found values of the full width at half maximum (fwhm) of PL band between 150 and 180 cm−1 and μ near 0.6 for major emission components. These values were obtained using numerous experimental data approximations for SWNTs dispersed in aqueous suspension with SDS or SDBS.

The large spectral interval between excitation (E22S) and emission (E11

S) facilitates the registration and analysis of the nanotube spectrum. On the con-trary, detection of emission upon direct E11 excitation is very difficult because it is not simple to suppress the exciting laser light which should be applied close to the emission because of the small Stokes-shift. Each laser provides the fluores-cence spectrum which is different from others as the particular excitation wave-length is in resonance only with some nanotubes of certain chiralities which give intensive bands in this spectrum. It was demonstrated that three properly chosen fixed excitation wavelengths are sufficient to provide detection of all semicon-ducting SWNTs in the sample because each growth method gives quite nar-row distribution of diameters [181]. However, the full depiction of fluorescence as well as complete identification of SWNT species present in the sample can be made when an ensemble of PL spectra is obtained using the set of differ-ent excitation wavelengths. In this case the excitation source wavelength (light of halogen or krypton lamps after wavelength selection with monochromator is often used) is scanned over the range of E22

S transitions with a small step

Fig. 3.17 Emission spectrum of semiconducting ss-DNA-wrapped HiPCO and (6,5) CoMoCat nanotubes aqueous suspension. Emission was excited by laser with line generation at 532 nm. For comparison, the absorption spectrum of (6,5) CoMoCat nanotubes (dash spectrum) is shown too

127

(10–20 nm). As a result, 3D picture or 2D contour map with the intensity as a function of the excitation and emission wavelength can be plotted (Fig. 3.18). Each distinct peak in 3D landscape originates from specific (n,m) species of the semiconducting nanotubes. In this 2D plot the intensity is showed with differ-ent colors (heights). This plot is known as a photoluminescence-excitation (PLE) map [128]. A well-resolved point in the space of excitation/emission wave-lengths corresponds to certain nanotube chirality. Figure 3.18 shows the typi-cal PL map for ss-DNA-wrapped SWNTs (synthesized by CoMoCat method) in aqueous suspension. A major peak in the PL map corresponds to the excitation transition energy of the second subband (E22

S) and to the emission energy of the first subband (E11

S) for (6,5) SWNTs. Near the dominant peak additional less intensive peaks can be observed and they can be assigned to nanotube species with (8,3), (6,4), (7,5), (9,1) chiralities [129, 130]. PLE spectrum from specific (n,m) SWNTs can be obtained as a cross section of the PL map at the energy corresponding to emission of relevant semiconducting nanotubes. Thus, PL map reflects (n,m) distribution of certain SWNT species. In addition, an emission spectrum at the fixed excitation wavelength can be obtained at the cross section of 2D map.

The fluorescence observation confirms the semiconducting character of SWNT and also gives researchers a powerful tool for investigating the nanotube band structure and dynamics of excitations, and allows identifying SWNT species pre-sent in the sample. In addition, the nanotube fluorescence has optimistic perspec-tives in many applications, first of all, in the field of fluorescence-based imaging and sensing. Fluorescence methods are highly effective for detecting SWNTs in

Fig. 3.18 3D and 2D contour map representing PL intensity of individual polyC-wrapped SWNTs as a function of the excitation wavelength. The intensity is showed with different gray gradations. The resulting excitation-emission profile map illustrates the existence of local maxima in emission intensity, and each maximum location corresponds to particular semiconducting nanotube chirality



the complex environmental or biological medium because this method has high sensitivity and selectivity. We also note such an important emission advantage as the absence of the interfering background.

The plotted dependence of the emission energy on the excitation one (E22S)

showed the branch like patterns in transition energies for various (n,m) species [128]. All tubes with 2n + m = constant value belong to a specific family [182, 183]. The family pattern can be separated into two groups: for type I the value of (2n + m mod 3) is equal to 1 while in the other group (type II) it is equal to 2. SWNTs of the certain family have similar diameters and, thus, similar transition energies as well. Small differences in energy are caused by the trigonal warping effect originating in deviation from the circular symmetry of equal-energy contours around the K point in Brillouin Zone (BZ) of graphene which has a triangular shape. For the SWNT type I the closest to the K point cutting line is on one side of this point, while for the SWNT type II this line is on the other side of K point. This is the main reason for the appearance of the family pattern in the plot of optical transition energies.

Although the single-electron tight-binding model (TBM) [118] has been very successful in describing qualitative phenomena in SWNTs, including fluores-cence, it does not describe entirely the more complex nature of photophysics in carbon nanotubes. For a quantitative analysis and even for the correct qualita-tive understanding of the nanotube physics, electron–electron interactions should be considered, especially this consideration is important to clarify the nature of SWNT excited states. The importance of electron–electron interactions in nano-tube excited states was early predicted in the theoretical work by Ando [182] in 1997. Within the static screened Hartree–Fock approximation, he indicated that the electron–electron interaction in semiconducting SWNTs leads to a wider band gap. However, the photoexcited electron–hole pair bound by Coulomb interaction leads to the formation of excitons, which partially compensates the excita-tion energy growth. The manifestation of excitonic electron–hole attraction and Coulomb electron–electron repulsion in SWNTs was first revealed experimentally [128] in the context of so-called ratio problem [183]. It was shown that E22

S/E11S

ratios for individual tubes deviate from 2 as predicted by TBM, and it depends on the chiral angle and the tube type. This ratio is closer to 1.8 when extrapolated to large SWNT diameters.

Subsequently, experimental verification of the exciton picture was reported in 2005, based on two-photon absorption experiments employing very high-power laser pulses [184–186]. From these experiments, the binding energy of the exciton at the lowest energy state was measured which is ~0.3–0.4 eV for semiconduct-ing nanotubes with diameters between 0.68 and 0.9 nm [184, 185]. The significant amount of this exciton binding energy, which is comparable to a semiconducting nanotube energy gap (E11

S = 0.5–1 eV for nanotubes with 2–0.5 nm diameters) suggests that excitonic effects may be relevant in all the aspects of SWNTs optical properties.

A number of theoretical studies incorporating electron–electron interactions, which appeared in 2004, presented the detailed description of excitons in semi-conducting SWNTs [187–192] that could explain many optical observations and

129

predict new experiments. Theoretical calculations showed that exciton bind-ing energies are anomalously large in carbon nanotubes, evidencing the impor-tance of many-body effects in this quasi-1D system. In general, these theoretical approaches agree well with finer estimations including ab initio [187, 189] or den-sity matrix [193] calculations. They all predicted exciton binding energies to be about few hundreds of meV for nanotubes with diameters d < 1 nm and increasing the band-gap for narrower SWNTs like ~1/d. Note that in typical 3D semiconduc-tors the Coulomb interaction is decreased by the dielectric constant of surrounding medium (so called dielectric screening). The value of the exciton binding energy in 3D materials is of ~10 meV order, and, as a result, usually excitonic effects appeared only at low temperatures. However, in 1D system (in our case, in semi-conducting nanotubes) the motion of particles is restricted to one dimension, and the electric field generated by the electron–hole pair is largely outside of the nano-tube so the screening effect is weakened. As a result, the electron–hole has a rela-tively large binding energy (~0.5 eV) so excitons can be observed even at room temperature.

Energy of the lowest optically allowed state EOptical is described by EOptical = ESP + EBGR − EBind where ESP denotes the single particle energy inter-action, EBGR (band-gap renormalization) is the electron–electron Coulomb repul-sion energy and EBind is the excitonic binding energy caused by electron–hole attraction [182]. Theoretical studies [188, 194] indicated that EBGR > EBind and, therefore, many-particles interactions increase the excitation energy, EOptical, above its ESP value.

The size of exciton in SWNTs has been estimated to be ~2 nm, the value being larger than the nanotube diameter [194, 195]. Thus, in SWNT the electron–hole distance is much greater than the lattice constant, therefore excitons in SWNT belong to Mott-Wannier type which is typical for semiconductors.

As follows from theoretical studies of exciton structures [148, 182, 183, 187–193, 196], there are 16 exciton states in semiconducting SWNTs. This quantity of states follows from the presence of spin in the electron and from the exist-ence of two nonequivalent K and K′ points in BZ in which π states are crossed. Strong electron–electron interactions remove the degeneracy of fourfold degen-erated electron–hole pair excitations due to the doubly degenerated valence and conduction single-particle bands. Besides, due to the exchange interaction, these excitations split into singlet (four) and triplet (twelve) excitons. Among these states, 1KK (which indicates that the spin is singlet and both electron and hole exist in K valley) and 1K′K′ exciton states play important roles in optical processes of carbon nanotubes. According to the calculation by Ando [148], a short-range part of the Coulomb interaction gives rise to an electron scatter-ing between K and K′ points, excitons |KK > and |K′K′ > are coupled to form a bonding state |KK-K′K′(+) > with a lower energy and an antibonding one |KK-K′K′(−) > with a higher energy. Among the excitons only one singlet exci-ton in the bonding state (odd-parity) is optically allowed (bright exciton) and all others (even-parity dipole-forbidden singlet excitons as well as triplet excitons) do not contribute to absorption and emission (dark excitons). It is very important



that some of the nonemissive states (singlet and triplet) are located lower in energy than the lowest bright state and form nonradiative relaxation channels. Thus, the complex of excitonic states has an essential influence on the PL quantum yield of SWNTs and exciton decay rates.

A bright-dark splitting energy about 29 meV was calculated for (10,0) tube in vacuum employing ab initio calculation [187]. The close value was obtained by the TB approach [197]. In the last energy calculation for (10,0) tube the splitting energy is about 14 meV without π-electron screening, but inclusion of this screen-ing gave 25 meV increase of the splitting energy.

A direct observation of the dark state is complicated because of broad line widths in ensemble-averaged spectra of PL that show multiple inhomogeneously broadened peaks with about 30 meV line width. The observation of dark excitons can be pos-sible only on single nanotubes at low temperatures at which the line width becomes smaller than the dark-bright splitting or when the magnetic field is applied.

The magnetic field directed parallel to the nanotube axis affects the phase of the exciton wave function due to Aharonov–Bohm (AB) effect and lifts the degen-eracy of exciton states [148]. The wave function mixing between K and K′ val-leys diminishes and KK and K′K′ excitons gradually become independent of each other for a sufficiently large magnetic field (at B > 25 T [148]) when AB splitting, ΔAB, exceeds the splitting energy Δbd between the bright and dark exciton states at the zero-magnetic field. Consequently, the dark exciton state gradually becomes the optically allowed state with the increasing magnetic field. Thus, in high mag-netic fields two optical-active exciton states appear in carbon nanotubes.

Dark excitons brightening in the magnetic flux was observed in recent photolu-minescence experiments (see reviews [198, 199] and Reference therein, [200–203]). Values of bright-dark exchange splitting Δbd determined in magneto-optical experi-ments with aligned SWNTs varied from 5 till 10 meV. The exciton splitting in mag-netic fields depends on nanotube diameters and is proportional to 1/d or to 1/d2 predicted by tight-binding [197] or the first-principles [187] theories, respectively. Note that in magneto-photoluminescence experiments DNA-wrapped CoMoCAT SWNTs samples dispersed in a polyacrylic acid matrix were exploited [202] too. These samples aligned by the stretch method had strong PL signal and sharp absorption features.

Excitons populate almost the lower energy dark state at low temperatures when the thermal energy is much smaller than the splitting energy of the bright and dark exciton states [204–209]. With temperature increasing, the population of the lower dark exciton state decreases drastically because of the thermal distribution between the bright and dark exciton states. The presence of the optically forbid-den transition is also confirmed by the non-monotonic temperature dependence of the radiative decay rate, with the maximum at a certain temperature, as predicted by Spataru et al. [187] and Perebeinos et al. [196] and observed experimentally [204–207, 210].

Note that spectroscopic experiments with two photon photoluminescence exci-tations [184–186] also can be regarded as a direct experimental proof of the exist-ence of dark excitonic states in SWNTs.

131

Time-resolved spectroscopy reveals multi-exponential behavior of the radiative PL decay at room [208, 209, 211, 212] and low [206–208] temperatures, demon-strating at least two different characteristic decay times. The first, fast, component can be associated with the rapid decay of bright exciton into the lower-lying dark state. The long time dynamics can be explained by the trapping of a dark exciton that does not have a radiative recombination path. Exciton dynamics in enriched (6,5) nanotube-DNA suspensions was investigated by femtosecond time-resolved pump-probe spectroscopy [212] which demonstrated a rapid decay of excitons into lower lying states such as the lowest lying dark singlet state with 6 ps time constant.

Interactions with the environment, edge effects, defects and impurities in the nanotube lattice further split and mix energies of excitonic bands [213]. Mixing between different excitons, because of interactions and deviations from the ideal nanotube structure, makes some optically forbidden states weakly active (semi-dark) rather than being strictly dark [203, 214–217]. All these experimental investigations of dark excitons were focused on the zero-momentum dark singlet exciton which can be brightened by application of an external magnetic field or due to lattice defects induced by intense pulsed laser excitation.

However, beyond the zero-momentum dark singlet exciton, the theory pre-dicted other two time reversal degenerate dark singlet excitons with center-of- mass momentum near K and K′ points of graphene BZ. These K-momentum dark singlet excitons located above the bright exciton cannot be directly photoexcited because of momentum conservation. However, according to the theoretical pre-diction by Perebeinos et al. [218], optical phonons near K point of the graphene BZ (in-plane TO phonon [219, 220]) have strong exciton-phonon coupling, and this dark exciton can be excited in combination with K-momentum phonon. As a result, the momentum conservation rule is fulfilled.

Various groups have studied exciton-phonon interactions in SWNTs and their studies have revealed the existence of a phonon sideband in PLE spectrum approx-imately 200 meV above the energy level of the singlet bright exciton E11

S [163, 221–223]. However, this energy was larger than predicted energy of K-momentum phonon [218]. In earlier articles the absorption sideband located ~200 meV above E11

S has been interpreted as an exciton-phonon band with participation of a longi-tudinal optical mode (LO-phonon at Γ-point in BZ) with approximately the same energy [163, 221–223]. However, the theory predicted that although LO phon-ons have the prominent signal in SWNT resonance Raman scattering spectra (G band) (see, for example [140]), they do not produce a significant phonon side-band of the E11

S bright exciton [218, 219, 224]. In accordance with these theories, K-momentum phonon dominates over LO phonon by more than an order of mag-nitude for all chiralities.

The origin of the absorption sideband located by ~200 meV above E11S became

understandable after detailed photoluminescence spectroscopy studies of dif-ferent SWNT types were performed, in which the existence of emission side-bands at approximately 140 meV below E11

S level was unambiguously shown for all SWNT types. Based on the experimental measurement of DNA-wrapped



SWNTs in the aqueous dispersion sorted by ion-exchange chromatography to enrich (6,5) species, Torrens et al. [219] proposed that this sideband originates from dipole-forbidden dark excitons coupled with K-point phonons. Murakami et al. [225] performed detailed PL spectroscopy investigations of three different SWNT types (samples were prepared in toluene with polymer (PFO) and con-tained essentially SWNTs with chiralities (6,5) (7,5) and (10,5)). Their studies confirmed the proposed model to describe sideband of PLE and PL. According to with this model, 200 meV energy of exciton-phonon band observed in PLE spectrum above the main peak E11

S is the sum of the K-point phonon energy which is about 170 and ~40 meV energy which corresponds to the energy differ-ence between K-momentum dark and bright exciton states. The energy value of PL subband (130 meV lower than the main peak E11

S) corresponds to the energy difference between the above phonon and the dark-bright energy separation. This conclusion was also confirmed by decreasing of this peak intensity at low tem-peratures, observed by Matsunaga et al. [226]. Note that the authors of these arti-cles indicated that the energy separation from the main peak is almost independent on the tube diameter. Nevertheless, using a collection of 12 semiconducting car-bon nanotube samples selected by short nucleotide sequences to enrich with nano-tubes only a single chirality, Vora et al. [224] revealed the chirality dependence of K-momentum dark singlet exciton, employing phonon sideband optical spec-troscopy. They found out that 2n + m family and chiral index exhibit systematic dependencies on the nanotube diameter. In contrast to the above studies which did not reveal such a dependence, they examined simultaneously both phonon sub-bands in PLE and PE from nanotubes of single chirality. This allowed them to observe chiral-family behavior of the dark exciton energy (Edark) which cannot be determined from PL only.

Recently Matsunaga et al. [226] found an additional low-energy PL peak appeared under intense pulsed-laser irradiation. Peak intensity increased with decreasing temperature, and the energy separation depends strongly on the tube diameter. The appearance of the additional peak was explained with the laser-induced brightening of triplet dark exciton states.

3.4.2 Quantum Yield of Semiconducting SWNT Emission: The Role of DNA Coverage

PL efficiency of semiconducting SWNTs has fundamental sense and gives the knowledge about the excitonic nature of the optical process in nanotubes. High fluorescence QY is desirable for different applications such as fluorescent markers and the development of biosensors or opto-electronic devices. The PL efficiency is known to be dependent on intrinsic factors such as the nanotube diameter and chirality [227, 228] and also on extrinsic influences such as the chemical environ-ment [229], nanotube aggregation, pH of aqueous solution [230], finite-length effects [231], defects introduced with purification and sonication processes.

133

The preparation method, surfactants used to form the suspensions, method of car-bon nanotube growth are also relevant to the emission QY.

Low PL efficiency (<0.1 %) for surfactant-micellarized nanotubes in aque-ous suspension was initially attributed to the presence of nonemissive metal-lic nanotubes and residual bundles among isolated SWNTs [22]. Even with the recent advances in sample preparation methods (see for example review [232]) which produce nearly single chirality samples and these undesired small bundles are significantly removed, the fluorescence QY of SWNTs in aqueous suspension remains poor (~1 %) [233].

Individual SWNTs (produced by HiPCO method) on the glass slide obtained from the aqueous suspension with SDBS have fluorescent QY of ~8 % for the brightest species [227]. Comparing the measured absorption cross sections and the fluorescence intensities from individual (CoMoCAT-manufactured) SWNTs, dis-persed with SC in D2O and then spun cast onto a quartz cover slip, and quantum dots (QDs), the average QY of individual isolated SWNTs was determined to be ~3 % [228]. The absolute QY for single nanotubes suspended across pillars does not provide high value too, which was estimated to be ~7 % [234].

SWNTs dispersed in organic solvents with polymer also have shown low quan-tum yields, i.e. 1.5 % for poly-9,9-di-n-octyl-fluorenyl-2,7-diyl (PFO)–wrapped SWNTs [40]. From the point of photophysics view, QY decreases through vari-ous nonradiative pathways in nanotubes, among them exciton quenching through the nanotube hole doping via oxygen adsorption [230, 235], or direct protona-tion [215] are most possible. In the acid or neutral environment [235], oxygen can quench photoluminescence through hole doping and subsequent nonradiative Auger recombination [230, 236]. Most SWNT surfactants do not obstruct oxy-gen penetration to the nanotube surface, which interacts and dopes these nano-tubes. Therefore, the most perspective way to enhance fluorescence QY lies in the limitation or exception of the oxygen sorption onto the nanotube surface. For this purpose, a tight coverage of nanotubes by special molecules can be applied as an efficient trend. Aliphatic (dodecyl) analog of flavin mononucleotide, FC12, has been examined recently as such possible molecules [237]. It turned out that the surface self-organization of FC12 around the nanotube is sufficiently tight to exclude oxygen penetration into the toluene environment, which led to quantum yields as high as 20 %.

Although SWNT fluorescence in certain organic solvents with polymer has suf-ficiently high QY, the increase of the nanotube fluorescence efficiency in water remains the most important issue because many applications including nanomedi-cine, biosensing etc. imply aqueous environments. The effect of the surfactant type (SDS, NaDDBS, NaC and DNA(GT)10 oligomers), the surfactant/nanotube con-centration influence on the light absorption and emission efficiency for the indi-vidualized nanotubes of different chiralities were studied by Fantini et al. [171]. The ratio between PL and optical absorption intensities was applied to compare the efficiency of different surfactants used for HiPCO nanotube suspension prep-aration. This ratio can be associated with the relative amount of isolated nano-tubes in suspension, since nanotube aggregation reduces the PL efficiency. It was



shown that NaDDBS and NaC gave higher PL intensities comparing to SDS or DNA((GT)10). Their measurements demonstrated that for DNA-wrapped nano-tubes suspension PL intensity is approximately twice smaller than for nanotubes suspension with SDBS. Surfactant molecules cover the nanotube surface entirely while DNA covers SWNT partly. This portion of the polymer coverage depends on many factors such as the DNA helicity, the polymer sequence, the nanotube chirality/diameter [67], solution pH and so on. The higher exposure of the nano-tube surface to water molecules is responsible for the increase in the effective dielectric constant in SWNT:DNA system and, consequently, is accompanied with the red-shift in the optical transition energies due to excitonic effects [172].

In Sect. 3.3.3 we have already discussed the comparative study results concern-ing the quality of aqueous dispersions of SWNTs covered with DNA or surfactants, using their absorption spectra obtained recently by Haggenmueller et al. [59]. They have also compared fluorescence properties of nanotubes in different surroundings including oligonucleotides, peptides, chitosan, and cellulose and such surfactants as cholates, ionic liquids, and organosulfates to solubilize CoMoCAT SWNTs in water. They have revealed that fluorescence intensities of various SWNT surfactant systems are dependent on SWNT concentration in suspension. Thus, the fluores-cence intensity increased from low concentrations to the intensity maximum at 0.01 mg/mL SWNT but at higher concentrations the intensity decreases. It turned out that this behavior is independent on the surfactant system. They explained this independence with the reabsorption effect reported earlier by Rickard et al. [238]. To eliminate variations of the fluorescence intensity as a function of concentration they performed measurements at 0.01 mg/mL SWNT concentration. To compare fluorescence of various systems, excitation fluorescence maps and line scans at the main excitation wavelength were measured. ss-DNA with 30 nucleotides ((GT)15, (AC)15, T30, C30,) sequences) demonstrated the best fluorescence efficiency among ss-DNA samples and showed the best fluorescence stability with time [59]. Fluorescence was measured for samples that remained suspended sufficiently long to allow fluorescence measurements (>3 days). It was revealed that SWNT: (short-chain ss-DNA) samples showed agglomeration after ~3 weeks, while SWNT: (longer chain ss-DNA) samples were stable over a longer period. The stability of the SWNT suspensions was the best (several months) for high molecular weight biomolecules and for short surfactants (SDS, SDBS, SDOCO). Peptide agglom-erated soon after centrifugation (~3 days). 3-D excitation-emission maps showed clearly that the relative fluorescence intensity of SWNTs with SDOCO is the strongest. Fluorescence of the (6,5) SWNTs with (GT)15 and SCO is compara-ble but showed reduced fluorescence than with SDOCO (Fig. 3.12b). SDBS and SDS showed emission peaks with lower intensities than with SCO. The compari-son of emission peak position for (6,5) nanotubes in different environments showed that this peak with (GT)15 is red-shifted relatively to PL bands of nanotubes with such popular surfactants as SDS (8 nm (10 meV)), SDBS (15 nm (20 meV)) and SCO/SDCO (10 nm (13 meV)). The red shift of optical transition energies for nanotubes wrapped with DNA was observed by other researchers too [102, 171]. A possible explanation for this shift is based on the change of the effective dielectric

135

constant of the nanotube surrounding medium, which may be associated with less effective nanotube coverage with DNA comparing to surfactants. This shift demon-strates the important influence of the environment on SWNT fluorescence.

Some knowledge about the electron deletion/reducing process on the nanotube surface can be obtained from the recent research of self-ordering of DNA adsorbed onto the nanotube, which occurs slowly in time [64]. The ordered domains are char-acterized with close contact between bases and the nanotube, and this changed the degree of nanotube surface coating effects on the semiconducting nanotubes PL. Cathcart et al. [64] observed an increase in the SWNT PL intensity by a factor of 50 and a considerable sharpening of van Hove absorption peaks after 20–50 days of the sample preparation. HRTEM images showed the progressive DNA covering of the nanotubes walls over this time period. They suggested that the initial quenching of NIR photoluminescence and absorption and luminescence peak broadening are related with the presence of protonated surface oxides on the nanotubes. In this case the ordered DNA coating on the nanotube can be accompanied by deprotonation and removal of surface oxides. They supposed that oxides are displaced from the nanotube with com-petitively binding to DNA and the most probable mechanism for this is one in which the endoperoxides react with nucleotide bases on DNA.

The coating material influence on the SWNT fluorescence efficiency was stud-ied by Tsyboulski et al. [239] with NIR fluorescence microscopy applied to observe PL from individual HiPCO SWNTs in dilute samples that differ only in their coat-ing material. It was demonstrated that emission of ss-DNA-coated SWNTs as well as using the common biocompatible coating material Pluronic F127 (synthetic nonionic surfactants [240]) or bovine serum albumin [241] is significantly (by an order of magnitude) weaker than that from nanotubes with ionic surfactant SDBS. However, when special peptides engineered to give higher SWNT emission were exploited, they revealed intensive PL which was by ~40 % more intensive com-paring to PL in SDBS surrounding [239]. They suggested that these peptides use the SWNT surface as a template for self-assembly to form a stable coating which provides an uniform dielectric environment for the nanotube. It was concluded that SDBS insulate SWNTs from the environment better in comparison with DNA irregular wrapping which increases water accessibility to the SWNT surface and can possibly quench fluorescence. Self-assembling properties of these peptides were found to correlate strongly with their ability to suspend nanotubes and pre-serve their emission efficiency. We note that recently these authors also suggested other short multidomain peptides as biocompatible solubilizing agents providing for SWNTs the NIR fluorescence intensity comparable to that in Pluronics [242].

To increase PL nanotube intensity, some researchers have developed processes to control the surfactant structure around SWNTs. These attempts include in situ polymerization of polyvinylpyrrolidone (PVP) [243], which enhances the pro-tection against such external action as solution pH. Mixing surfactant–stabilized SWNTs suspensions with organic solvents increased the intensity of some large diameter SWNT (n,m) types by more than 175 % [244], passing the SWNT flow through microchannels, which effects the surfactant structure. After such flowing a significant increase in PL intensity was observed [245].



An unexpected fluorescence brightening and QY increase up to an order of mag-nitude were reported recently for individual DNA-wrapped SWNTs in aqueous suspension [246]. The enhancement was observed upon addition of reducing agents to SWNT suspensions such as dithiothreitol (DTT), Trolox, and β-mercaptoethanol (BME). As a result, QY for ~20 individual nanotubes ranged between 15 and 40 %. Special experiments showed that water displacement from SWNT by the reducing molecules cannot completely explain the effect of QY enhancement. It was con-cluded that reductants bind to exposed SWNTs defects and that the reduction effect is responsible for the SWNT fluorescence enhancement. This surprising discovery is believed to arise from passivation of the defective SWNT surface with reduct-ant molecules that donate electrons to trap sites along the nanotube surface, which is predominantly doped with holes [215, 230, 235, 237]. At this the interaction between nanotubes and these molecules is supposed to be noncovalent. This obser-vation indicates that the intrinsic PL QY of semiconducting SWNTs is much higher than it was previously thought and that their poor emission arises from defective nanotubes. Using identical experimental conditions, the fluorescence intensity of individual DNA-wrapped SWNTs (after reductant addition) was compared to that of individual SWNT:SDBS deposited onto the cover slip. It turns out that DNA-wrapped SWNTs were four times brighter than SWNTs covered with SDBS. In these experiments the cross section of the E22

S transition of (6,5) SWNTs was also estimated as σ = 4.2 × 10−14 cm2 [246].

3.4.3 Influence of Environment on SWNT Photoluminescence Properties

The surrounding exhibits a significant influence upon the SWNTs optical proper-ties (see for example review [31] and ces therein). Such external factors as sol-vents, the type of surfactants or polymers wrapped around the nanotube have an influence on Eii value of a particular (n,m) SWNT, which can be shifted by a large amount. Environmental effects on SWNTs optical properties were observed in the earliest spectroscopic measurements by several groups [35, 124, 247, 248]. In par-ticular, it was shown that the observed spectral shift of the optical transition energy varied, depending on the surfactants type used to solubilize SWNTs [35].

Lefebvre et al. [124] observed that the peak position of PL band from SWNTs freely suspended in air is blue-shifted relatively to the peak position of the corre-sponding nanotube species obtained in aqueous suspension being encapsulated in micelle [128]. It was shown that these shifts are 28 and 16 meV for E11

S and E22S

transitions, respectively. Okazaki et al. [248] also observed ~50 meV blue-shifts of PL bands from as-grown SWNT relatively to the peak position of the nanotube in aqueous suspension with SDS.

Ohno et al. [249] investigated the environmental effect on emission of 20 SWNT species by comparing PL maps of air-suspended and SDS-covered SWNTs. Blue shifts of E11

S for air-suspended nanotubes varied from 6 till 30 meV

137

relatively to nanotubes in aqueous solution. They showed also that the energy dif-ferences depend on (n,m), specifically on the chiral angle (θ) and on the family of SWNT species (type-I or type-II). With the chiral angle growth, ΔE11 increased in SWNT type I and decreased in SWNT type II, however, ΔE22 demonstrated the opposite dependence. The (n,m) dependence of the environmental effect was explained by the (n,m) dependence of the excitation effective mass. Later these authors [250] investigated optical band gaps of the single SWNTs bridging over trenches in air by immersing SWNTs in various organic solvents with dielectric constants (εenv) ranging from 1.9 to 37, using PL and the excitation spectroscopy. With εenv increasing both E11

S and E22S showed 33–49 meV and 26–30 meV red

shifts, respectively and a tendency to saturate at εenv ~ 5 without a significant (n,m) dependence was observed. Red shifts were explained by the dielectric screening of the electron–electron repulsion energy. The εenv dependence of E11

S and E22S

were expressed by a simple empirical equation with the power law of εenv. The equivalent εenv of SDS-covered SWNTs was estimated to be ~2. In addition to the red shift with increasing εenv, they noted the PL spectrum broadening in liquids. For example, the PL linewidth of (9,7) SWNT increased from 23 meV in air to 40 meV in acetonitrile. This linewidth broadening was explained by an inhomo-geneous broadening due to εenv local fluctuation in nanodimension such as the exciton diameter or the size of solvent molecules. Depending on the organic mol-ecule orientation and number, the local εenv would fluctuate near the nanotube that would result in inhomogeneous broadening of PL spectrum.

Kiowski et al. [251] applied PL laser microscopy with the scanning laser exci-tation wavelength to determine optical transition energies E11

S and E22S of indi-

vidual semiconducting SWNTs suspended on the top of MWCNT “forests,” grown by CVD on silicon substrates. It was found that blue shifts of PL peaks amount to 40–55 meV and 24–48 meV for E11

S and E22S, respectively, for 19 different

(n,m) nanotube species suspended in air or vacuum relatively to SWNTs in water-surfactant dispersion. Within the experimental error, they found no systematic correlation between the nanotube (n,m) structure and variations in energy shifts. CVD-grown SWNTs embedded in paraffin oil and 1-methylnaphthalene show nearly the same PL peak positions as SWNTs in aqueous dispersion, indicating similar dielectric screening of excitons in SWNTs in these media.

Choi and Strano [252] studied solvatochromic shifts, analyzing SWNT PL energies in various dielectric media including DNA. They have developed a scal-ing model to predict solvatochromic shifts for SWNTs in various dielectric envi-ronments, and they deduced the linear scaling between optical (Eii) and structural (diameter) parameters (Eii

2 ΔEii vs d−4) according to their studies.Photophysical properties of nanotubes encapsulated with microenvironments of

the nonpolar organic solvent were recently studied by Silvera-Batista et al. [253]. PL and absorbance spectra of HiPCO SWNTs in 16 nonpolar solvents showed solvatochromic shifts (in the range from 25 to 100 meV), which are proportional to the solvent induction polarization. It was concluded that the PL intensity of SWNTs is very sensitive to the solvent polarity changes. So, a change of the dielec-tric constant (εenv) from 2 to 5 could result in more than 50 % drop in PL intensity.



It could have significant implications for PL intensity of poorly coated SWNTs in aqueous environments. The most significant effects of the environment were observed on the peak position of the smallest diameter SWNTs and on the inten-sity of nanotubes with the largest diameters. It was found that the average solva-tochromic shift for each (n,m) type in all solvents varies linearly with the nanotube inverse diameter (1/d). The spectral shift dependence on the (n–m)mod3 value was not observed. The simple expression describing average solvatochromic shifts as a function of the inverse diameter of SWNTs (ΔE11(eV) = 0.076–0.119/dt(nm)) was suggested after experimental data approximation. The authors concluded that this simple expression excludes specific solvent effects but provides an estimate of the solvatochromic shift for SWNTs in low dielectric media.

Influence of DNA coverage on SWNTs optical properties was permanently stud-ied during this entire time period, starting with an earlier observation of environ-ment effects. Chou et al. [254] observed in resonance Raman scattering (RRS) and PL measurements that both E11

S and E22S transitions for DNA-wrapped CoMoCAT

nanotubes were red-shifted by 30 meV in average, relatively to SDS-isolated HiPCO nanotubes with the same (n,m) assignment. They suggested that shifts in Eii values between SDS- and DNA-wrapped nanotubes are observed because these wrapping agents perturb nanotubes electronic structures in different ways.

Peak position red shifts (11–19 meV) for PL bands of ss-DNA wrapped SWNT (HiPCO) relatively to those of SWNT:SDS suspension were observed too [66, 102]. The similar value of red shifts (~20 meV) was also found by Fantini et al. [171] for the optical transition energies for HiPCO nanotubes wrapped with (GT)10 oligomers relatively to three surfactants (SDS, NaC and SDBS).

Influences of different polymers and surfactants on E11 peak position of (6,5) CoMoCAT SWNTs were analyzed by Haggenmueller et al. [59]. Thus, PL peak of this nanotube species covered with (GT)15 oligomers in aqueous suspension was red-shifted by 10 and 19 meV relatively to that with SDS and SDBS coverage, respectively.

Strano and coworkers demonstrated influence of divalent ion-induced conforma-tional changes in ds-DNA-encapsulated SWNTs on the optical transition in nano-tubes [67, 255]. PL from 30-nucleotide oligomer-wrapped SWNTs manifested a red shift when nanotubes were exposed to counter ions that screen the charged backbone. So, the peak energy of the (7,5) nanotube was shifted by 10 meV with increasing con-centrations of HgCl2 added to SWNT:ds-DNA suspension. The PL modulation had ion selectivity identical to B to Z form transition for the corresponding free ds-DNA.

Environmental effects were theoretically discussed by several groups [172, 188, 190, 197, 256–262]. The environmental effect on Eii can be understood when the excitonic dielectric screening effect is taken into account. It was showed that the potential of the electron–hole interaction which forms exciton should be substan-tially weakened by the dielectric environment of the nanotube. Perebeinos et al. [188] demonstrated the exciton binding energy EBind dependence on the dielectric constant of the environment εenv: EBind ~ εenv

−α with α = 1.4, which is not similar to that in the 3D case (α = 2). It was pointed out that this dependence is accurate only for nanotubes surrounded with highly permittive media (εenv ≥ 4). Therefore, this treatment can be readily applied to relatively narrow tubes embedded in SiO2

139

(ε ~ 4) or higher-ε materials. However, for isolated nanotubes in air (ε = 1) or tubes in the low-ε media, for example, in the hydrocarbon environment of SDS micelles with corresponding empirical εenv = 2 − 2.5, this treatment was not very accurate.

To model the effect of the external dielectric function changing on particle inter-action energies, Walsh et al. [256, 257] modified the simple single-particle picture. They showed that electron–electron and electron–hole interaction energies depend strongly on the screening effect. However, these energies largely counteract each other, resulting in small changes in the optical transition energy. As was mentioned above, the optical transition energy is largely tuned by the excitonic binding energy (EBind) and the electron–electron Coulomb repulsion energy (EBGR). These two terms are inversely proportional to the dielectric function of the medium (εinv), varying EOptical and, consequently, the fluorescence peak position (the screening effect). Both energies depend on εm as EBGR ≈ 1/εinv [192] for small electron wave vectors near the zone center. EBind was predicted to scale with εinv

−1.2 for small εinv [256] and with εinv

−1.4 for εinv > 4 [188]. The overall shift of EPL is negative because EBGR is slightly larger than EBind. As an example, authors of Ref. [256] considered the case with εinv = 1, in which EBGR and EBind are equal to 730 and 580 meV, respectively. In aqueous environment with surfactant micelle (εinv = 1.78) these energies strongly decrease to 410 and 290 meV, respectively. Thus, reductions in the exciton binding energy and EBGR are ~320 and ~290 meV, respectively. The difference between these two values gives 30 meV spectral red shift of the PL band.

Calculations carried out by Jiang et al. [197] within TB approximation found 2n + m = const family behavior in the exciton wave function length, excitation energy, binding energy, and the environmental shift. This family behavior was explained by the trigonal warping effect around K point of graphene BZ and cur-vature effects. It turns out that within the same 2n + m family, the value of ΔE11 rises with increasing the chiral angle θ for SWNTs type I while it decreases with θ for SWNTs type II. In contrast, the value of ΔE22 decreases with θ for SWNTs type I within the same 2n + m family while it increases with θ for SWNTs type II.

In parallel, Miyauchi et al. [172] have proposed a simple model to describe the dielectric material around SWNTs through the static dielectric constant ε in the calculation, using 1/ε = Ctube/εtube + Cenv/εenv where εtube is the dielectric con-stant within the nanotube except for π bands, and Ctube and Cenv are coefficients describing the nanotube itself and outer environment, respectively (Ctube and Cenv are diameter-dependent coefficients). Based on the model, they suggested the for-mula for the energy dependence of the semiconducting SWNT transition on the dielectric constant of various surrounding materials and for various dt and θ.

Using the extended-TB model with many-body corrections plus the diame-ter-dependent dielectric constant εinv, Araujo et al. [260] theoretically described within experimental accuracy Eii vs (n,m) values. It was shown that εinv repro-duces well the measured Eii values for (1.2–2.7 eV) energy region and for the tube diameter from 0.7 till 3.8 nm. Their treatment for εinv gave them an opportunity to assign both 2n + m family numbers and (n,m) of SWNTs belonging to each family for different SWNT samples. Recently, Nugraha et al. [261] found that εinv



depends not only on dt and the subband index but also on the exciton size lk in the reciprocal space. As well, they established an empirical formula to calculate unknown Eii for different sample environments.

Within the elementary potential model in which an exciton was represented as a bound state of two oppositely charged quasi-particles confined to the nano-tube surface, Smyrnov and Adamyan [258, 259] studied properties of excitons in semiconducting SWNTs isolated in vacuum or in a medium as well as their con-tributions to the nanotube optical spectra. The obtained binding energies EBind of excitons in the ground state and the differences between the ground and first excited exciton energy levels in nanotubes surrounded with medium with εenv ≥ 4 are in good accordance with the corresponding experimental data. Also, in the range of εenv (4–16) the ground-state exciton binding energies EBind obey the relation E ~ εenv

−α where α = 1.4, the similar result was obtained in Ref. [188]. However, if εenv is in (1–1.75) and (2–4.5) ranges these binding energies are smaller with values of α = 1.121 and 1.258, respectively. These results are very close to those obtained in Ref. [256].

Recently Ando [262] studied effects of environmental dielectric screening on excitons in carbon nanotubes within kp scheme and the continuum model. It was shown that dielectric screening is sensitive to the effective distance between the nanotube and the dielectric medium, and the binding energy of excited exciton states disappears rapidly with the dielectric constant increasing. It was pointed that, for dielectric material inside the nanotube, dielectric screening are much weaker and excited exciton states remain as bound states even for very large ε (~100). As well, Ando studied effect of environment on the SWNT family and found out that for each family excitation energies are red shifted almost equally without special dependence on (n,m) values.

In addition to the dielectric screening effect, other environmental factors such as substrate should be considered. Kiowski et al. [263] applied PL laser micros-copy with scanning laser excitation wavelength for imaging and characteriz-ing individual SWNTs grown by CVD method on structured Si/SiO2 substrates. 3.5-μm-wide and 500-nm-deep parallel trenches separated one from another by 300 μm were lithographically etched on the substrate before the tube growth. Such configuration of trenches allowed to compare directly PL properties of the same semiconducting nanotubes air-suspended parts and those contacted with SiO2. For the nanotube arranged on SiO2 segments, they found ∼10- to 20-fold decrease in PL intensity and red shifts of emission and excitation transitions by 7–27 meV (E11

S) and 5–24 meV (E22S), respectively, in comparison with air-sus-

pended regions of the same SWNTs. These shifts can be explained by the different extents of external dielectric screening which influences excitons in SWNTs.

Steiner et al. [264] also studied optical properties of a single, semiconducting SWNT that is partially suspended across a trench and partially supported with SiO2-substrate. By comparing the E33

S resonance spectra measured by RRS and PL excitation spectroscopies in the suspended SWNT segment, they observed that the peak energy in the PL excitation spectrum is by 40 meV red-shifted comparing to RRS excitation spectrum. This (substrate-induced) red shift was associated with

141

the energy difference between the localized exciton observed in the PL excitation spectrum of the SiO2-supported SWNT segments and the free exciton giving rise to the Raman excitation spectrum which was detected from the suspended SWNT segment. The appearance of additional peaks in the strongly broadened G band of high-resolution Raman spectra was interpreted as the substrate-induced symmetry breaking. The symmetry is broken by perpendicular fields induced by the interac-tion with a substrate, giving rise to similar spectral effects as those observed upon changing the laser excitation light polarization.

As is seen from NIR PL microscopy study of single semiconducting SWNTs, the emission energy of these nanotubes can be modulated by the dielectric con-stant which is non-uniform along nanotubes. We have already discussed red shifts about several tens of meV in the PL energy upon DNA-wrapping of nanotubes compared to the values obtained for micelle-covered SWNTs in aqueous suspen-sion [13, 52, 66, 102, 130, 171, 254], which was explained by the dielectric con-stant increasing [252]. Because of the finite length of DNA, the surface coverage with polymer segments is non-uniform along the nanotubes, resulting in a non-uniform dielectric environment. Unfortunately, the diffraction limit hinders the imaging and investigation of processes on length scales below the wavelength of light, so the confocal microscopy has about 300 nm spatial resolution. However, high resolution tip-enhanced near-field optical microscopy (TENOM) application overcomes the diffraction limit, and a possibility to study the optical properties of single SWNTs with nanoscale spatial resolution (~15 nm) appears [265]. In this microscopy, a sharp gold tip is used as an optical antenna that strongly increases the excitation and emission rates in the nanoscale volume, and this results in the diffraction limit overcoming. In addition, optical response from the nanoobject enhances significantly that provides a very high sensitivity of this method for detection of PL and RRS signal. Recently Hartschuh with coworkers [265] used TENOM to resolve PL variations along DNA-wrapped (6,5) and (6,4) nanotubes. They revealed two distinct emission bands which were identified with emission from DNA-wrapped and unwrapped segments. The determined energy shifts were 18 meV for (6,5) and 30 meV for (6,4) nanotubes, respectively. When they used confocal spectroscopy for this purpose, a red shift of PL band was found between 7 and 17 meV for (6,5) and (8,3) nanotube, respectively. Based on this observa-tion, they concluded that confocal measurements underestimate (by about 2) the energy stabilization induced by DNA wrapping.

Near-field scanning along the DNA-wrapped nanotube yields different optical responses when the tip is moved along the nanotube. In Fig. 3.19 three different positions marked with numbers 1, 2 and 3 are shown. At position 1, the nanotube is locally excited with a DNA-wrapped segment and exciton emission occurs at low energy as shown in the spectrum in Fig. 3.19. At position 2, the tip probes leads to the high emission energy of the bare nanotube (Fig. 3.19d). Confocal and near-field spectra were detected at the locations along the nanotube. Both confocal spectra (a, b) have fwhm of 40 meV, while the near-field spectra (c, d) are sig-nificantly sharper with fwhm of 27 and 25 meV, respectively. When the tip probing is located on the top of both DNA-wrapped and unwrapped segments (depicted



Fig. 3.19 Confocal (a, b) and near-field (c–e) PL spectra of a DNA-wrapped CoMoCAT SWNT on mica substrate. Gaussian and Lorentzian curves shown in a, b and c–e, respectively, were obtained as a result of fitting to experimental spectra. In the near-field spectrum (e) two distinct peaks are clearly resolved and can be attributed to DNA-wrapped and unwrapped sections of the nanotube. f Schematic illustration of the exited state energy PL landscape (thick line) along the DNA-wrapped nanotube including two wrapped and one unwrapped parts. The PL spectra (numbers 1, 2, 3) are shown on top of each position. ε denotes the dielectric constant of the local environment. Reprinted with permission from [265]. Copyright (2008) American Chemical Society

143

at position 3), double peaks including both low and high emission energies are observed in the spectrum (Fig. 3.19e).

The simultaneous observation of two distinct emission peaks as in Fig. 3.19e demonstrates that the transition between the two emissive energy levels occurs rapidly. Figure 3.19f shows a schematic illustration of the exited state energy PL landscape along the DNA-wrapped nanotube including two wrapped and one unwrapped parts indicated by the nanotube below. Corresponding PL spectra are presented above the probe schematic depiction for each position (numbers 1, 2, 3).

In the following study they observed exciton localization in SWNTs at room temperature [266]. This effect was revealed for DNA-wrapped SWNTs on the mica surface and leads to highly confined and bright PL emission spots on the nanotube. Localization results from the narrow exciton energy minima with depths more than 15 meV. These spots can not be attributed to local quenching sites only because their confinement requires such small distances among quenching sites. The authors believed that bright emissive spots result from the increased exciton density at exciton energy minima, due to the trapping-like process. The energy variations were attributed to inhomogeneous DNA-wrapping of nanotubes, which locally reduces the energy.

To exploit unique NIR PL properties of SWNTs for future photonic applica-tions, isolated and fluorescent semiconducting SWNTs should be encapsulated in different media, first of all, in polymers. SWNT composite material development opens the door to the realization of unique nanotube properties in many applica-tions, to novel technologies being able to create complex multifunction photonic circuits. SWNT composite materials are easily manipulated with low-cost manu-facturing methods (see reviews [267, 268]).

However, SWNTs packing in the solid form may be accompanied by the exciton energy transfer (EET) between nanotubes, which weakens the luminescence effi-ciency of SWNTs. Thus, the amount of nanotube bundling in fluorescent composites is a key parameter. In bundles excited states relaxate faster comparing to isolated SWNTs leading to PL quenching. In addition, large bundles with sizes comparable to the wavelength cause undesirable light scattering. PL and PLE spectroscopy may be used to identify bundles by monitoring EET. When small bundles composed of semiconducting nanotubes are formed, the red-shift in E11 emission wavelengths is observed which can be attributed to EET from large-bandgap SWNTs (donors) to small-bandgap SWNTs (acceptors) [269–275]. Thus, EET can be used as an effec-tive tool to detect the presence of small bundles in different media. It should be noted that EET is independent of perturbations induced by dielectric environments whereas both SWNTs are in the same bundle.

Polymers such as gelatin [205, 276–278], agarose [279], poly(vinyl alcohol) (PVA) [280], PVP [281], PFO [282], composite gels [283], carboxymethylcellu-lose (CMC) [284] have been successfully used to make fluorescent SWNT:polymer composites. The mechanical stretching of the polymer film induces considerable uniaxial alignment of nanotubes as was demonstrated by highly polarized absorp-tion, photoluminescence and Raman spectroscopy of SWNTs [285]. The highly aligned and luminescent SWNT thin films facilitate the development of novel opto-electronic materials based on SWNTs.



Due to helical wrapping of polymer around SWNT sidewalls with the strong binding through π-stacking, DNA is an effective SWNT dispersant in aqueous environments. Thus, DNA may be used simultaneously as an effective surfactant giving high amounts of isolated SWNTs and as the host polymer in the compos-ite. PL of DNA-wrapped SWNTs was also revealed in the film which was pre-pared by dropping SWNTs suspension with DNA onto Si substrate and drying in warm air [66, 102]. PL band peaks of nanotubes in the film are red shifted by 4–15 meV and broadened relatively to their aqueous suspension (from the initial fwhm of 25–30 meV order up to 40–50 meV in DNA film). As PL was observed from nanotubes in the film, it indicates the presence of individual tubes in the film or small bundles formed by polymer separated tubes. DNA wrapped around SWNT precludes the full nanotubes aggregation in the film. The red shift corresponds to an increase of the average dielectric constant of the nanotube environment. The polymer pressure onto the nanotube can also be a reason of a red shift. Such an assumption is reasonable as the polymer wraps tightly around the nanotube and this adsorption is very strong both in water and in film but the interaction between the nanotube and DNA in the solid state is stronger and it induces mechanical strain on the nanotubes [286]. As a result, nanotube band gaps can be shifted to the low energy [287]. The line broadening can be of inhomogeneous nature, namely, due to local variations of the dielectric constant of the environment near the nanotube which can lead to deviation of spectral lines from the mean position. Another type of the broadening mechanism can be related to environment-induced dephasing pro-cesses that are especially exhibited in nanostructures. These processes may involve more efficient coupling to the phonon modes of the solid matrix (with respect to a liquid surrounding) or be associated with local surface charges fluctuations in the dielectric environment of the nanotube, such fluctuations may be induced with photoexcitation. In addition, the nanotube:polymer interaction with the substrate can also increase the emission bandwidth. It should be noted that a stable nano-tube suspension was obtained after short time (3–5 min) sonication of SWNT:DNA film in water, in contrast to a relatively long sonication in the initial step to prepare SWNT:DNA stable aqueous suspension. This experiment indicated that, due to the polymer interlayer between nanotubes, bundles in SWNT:DNA films are different from usual SWNT bundles [102, 288]. Note that transparent conductive films were prepared recently from aqueous dispersion of SWNTs covered with DNA [156].

3.4.4 Comparison of Protection Properties of SDS, SDBS and DNA Covering of SWNTs Against pH Influence Using Luminescence and Absorption Spectroscopy

Optical properties of surfactant-wrapped SWNTs are extremely sensitive to envi-ronments, for example, slight changes in solution pH affect the absorption and luminescence of the nanotubes. Especially, these spectral properties of nanotubes

145

are changed with the acidity increase, at low pH the suspension can be destabi-lized and even flocculation can be noticed. Due to promising SWNT applications, especially in biological systems, including drug delivery, exploiting nanotube-based biosensors working in a biological environment, the question is actual how the sensors respond to biological conditions such as pH, e.g. how their spectral properties are transformed in such an environment.

pH influence on SWNT electronic properties has been studied earlier by O’Connell et al. [22] who observed pH dependence of luminescence and absorp-tion spectra of nanotubes in aqueous suspension with SDS. Absorption spectra of nanotubes in acidic solutions with pH < 3 had broad and weakened absorption in the E11

S region. However, upon neutralizing the solution to pH > 7, the intensity of the nanotube spectrum recovered. Emission spectra of nanotubes demonstrated a similar behavior with pH changing. The researchers supposed that this pH sen-sitivity reflects nanotube protonation, which was of reversible character. It was noted that semiconducting tubes with a larger diameter (a smaller band gap) were protonated first as pH decreased.

Later Strano et al. [235] also studied acidification of SDS-dispersed SWNTs in water, which was monitored by weakening of the absorption intensity of the E11

S transition, with a reduction of RRS spectra, and quenching of PL from the nano-tube. They supposed that pre-adsorption of molecular oxygen plays a critical role in this process; ambient O2 catalyzed H+ interaction with the nanotube sidewall either by lowering the energetic barrier for reaction or by participating in the com-plex directly.

Then, Dukovic et al. [230] supposed that dissolved oxygen actually forms 1,4-endoperoxide across the aromatic ring in the honeycomb structure of the car-bon nanotube. When pH is lowered to acidic conditions, 1,4-endoperoxide can be protonated that results in the ring opening up and hydroperoxide carbocation that injects a hole into SWNT π-electron valence band. They claimed also that this positive delocalized hole is responsible for luminescence quenching through a non-radiative Auger recombination process.

The choice of the surfactant can enhance or weaken pH effect on SWNT spec-tral properties [289–292]. As was mentioned above, anionic surfactants are widely employed to suspend SWNTs in water as well as biopolymer (DNA) which also showed sufficiently high efficiency in nanotube solubilization. Zhao and cowork-ers [293] have reported firstly that optical transitions of ds-DNA-wrapped semi-conducting nanotubes are dependent on solution pH. Then they examined redox chemistry of ds-DNA-covered SWNTs under variable pH conditions, using hydro-gen peroxide [294]. To observe the pH influence, RRS and absorption spectros-copy were employed. The reversibility of the NIR spectral intensity was observed after NaOH addition, pH solution increases to 11, and the NIR band intensities restore totally. On the contrary, after HCl addition into the recovered solution the spectrum features were smoothed, although with small hysteresis.

Han et al. [295] showed that ds-DNA-wrapped carbon nanotubes can serve well as rigid templates for the assembly of gold nanoparticles, and variations in pH were used to control aggregation of SWNT:DNA hybrids. At low pH values,



SWNTs:ds-DNA dispersion aggregated whereas at higher pH values the aggre-gates disappeared.

Nakashima and coworkers [111] described band gap modulation of individually solubilized SWNTs (HiPCO and CoMoCAT) covered with ds-DNA, by changing pH of aqueous solutions. They found that a very small pH change (from pH 5.8 to pH 6.4 in NIR absorption spectroscopy and from pH 6.4 to pH 7.4 in PL) caused a dramatic spectral transformation. The difference in the observed pH break-points for NIR absorption bleaching and PL quenching was explained with the difference in numbers of holes generated on SWNTs stimulating changes in absorption or PL properties. It was shown that for 400-nm length SWNTs more than 250 holes are needed for absorption bleaching whereas about 10 holes are sufficient for PL quenching [230].

Recently, protection properties of SDS, SDBS and ss-DNA coverings of CoMoCat SWNTs against pH influence were compared, employing lumines-cence and absorption spectroscopy in visible and NIR ranges [292]. To compare changes in emission of SWNTs covered with different surfactants and the poly-mer, the intensity dependence of the most intensive band (chirality (6,5)) on pH was controlled (Fig. 3.20a). As follows from Figure, with pH decreasing the emis-sion intensity of semiconducting nanotubes with chirality (6,5) (prevailed in the sample) in aqueous suspension with SDBS increased in the range of 11–9 pH up to 17 %, and then PL is quenched weakly at 9–4 pH. Below pH4 the intensity reduces more quickly but this decrease does not exceed more than 35 % of the maximum value. As well, a small spectral shift (up to 2 meV) of nanotube peak bands to a low energy at pH decrease was observed.

SWNTs luminescence intensity in suspension with SDS (Fig. 3.20a) dem-onstrated more dramatic reducing than with SDBS. So, emission quench occurs immediately when pH decreasing starts from 11 and at low pH (3.1) the intensity drops up to 5 % of the maximum intensity value. A small shift (not above 1 meV) of peak bands to the high energy of the first electronic transition at pH decrease was observed too. The main reason for emission quenching of nanotubes covered with surfactants was explained by the micelle destruction and so it becomes pos-sible for water molecules (and oxygen too) to access the tube surface and, as a result, this surface is protonated. SDS molecules form a layer on SWNT surface in which the alkyl chain as well as the surfactant charge group interact with the nanotube surface [296]. In contrast to this type of micelle, the coupling of SDBS monomer to SWNT surface (due to π–π stacking between the benzene ring and the carbon nanotube surface) is more tight [36] which allows them to protect sur-face from the environmental influence more effectively than SDS molecules.

As for above surfactants, a similar PL intensity behavior of SWNTs covered with DNA was demonstrated at pH lowering (Fig. 3.20a). So, SWNT emission is quenched at pH3 up to 30 % of the maximum value with about 4 meV downshift of the most intensive band as suspension pH decreases from 11 to 3.

The emission intensity changes under the influence of external factors can be caused by changes in the population of excitation states and/or by appear-ing quenching centers. Information on the population of excitation states can

147

be obtained from absorption spectroscopy. As for luminescence, depend-ences of the absorption intensity of the most intensive band at 1.25 eV on pH (Fig. 3.20b) were plotted for three SWNT aqueous suspensions. It follows from Fig. 3.20 that the intensity of SWNT:SDBS and SWNT:DNA bands assigned to E11

S transition is less than by 20 % reduced with pH decrease while the SWNT:SDS band intensity drops up to 25 % of the maximal value. The fact that emission intensity of SWNT:SDS quenches up to 5 % of the maximal value indicates that this strong quenching was caused not only by the excitation popu-lation decrease but rather by the appearance of quenching centers on the nano-tube surface. This observation confirms also a conclusion that low pH destroys micelle formed by SDS molecules around nanotubes, however, SDBS micelle demonstrates higher protection properties against pH influence. The polymer wrapping around the nanotube manifested high stability and resistance to pH influence. Spectral changes observed in this suspension can be explained by the partial covering of the nanotube surface with the polymer, and, as a result, the polymer-free nanotube surface contacts with water and protonation effects take place.

Fig. 3.20 Dependence of normalized emission (a) and absorption (b) intensities on aqueous suspension pH for semiconducting nanotubes with SDBS (--▪--), SDS (—●—) and ss-DNA (···▲···) surrounding. Emission and absorption intensities of each hybrid were normalized to their maximal values. Reprinted from [292], with permission from John Wiley and Sons



Strano et al. [235] revealed that PVP addition (0.1 %) into aqueous suspension of nanotubes with SDS leads to a partial restoration of the absorption and fluores-cence spectra of suspended carbon nanotubes after complete protonation at pH2. This phenomenon was attributed to a strong interaction between SDS micelles and PVP molecules.

Recently Pasquali with coworkers [243] reported a simple strategy to obtain stable and highly luminescent suspended individual nanotubes at pH values rang-ing from 1 to 11. This strategy relies on mixing SWNT:SDBS aqueous suspen-sion with biocompatible PVP (0.1 %) which can be polymerized in situ to entrap SWNT:SDBS micelles. They demonstrated that, upon acidification, the PL intensity of nanotubes was enhanced approximately by a factor of 2 and this enhancement was accompanied by narrowing and blue-shift (~25 nm) of PL peaks. The proposed model assumes the strong adsorption of PVP or its monomer (VP) to the external surface of SDBS micelle, due to the charge transfer between the SDBS sulfate group and nitrogen of PVP and/or VP. In acidic conditions, conformational changes of PVP and VP polymerization provide their efficient wrapping around SWNT:SDBS, with-out displacing SDBS micelle. The resulting surfactant-polymer complex protects the nanotube luminescence properties, providing a stable barrier between nanotubes and their local environment which would benefit various biomedical applications.

It was found out that the PL intensity of aqueous SWNT:SDS suspension after flowing through microchannels was less sensitive to quenching effects attributed to the acid medium [245]. So, the PL intensity of this SWNT suspension at low pH (2.5) was approximately an order of magnitude higher comparing to the initial suspension.

Fagan et al. [297] suggested to use composites of DNA-wrapped SWNTs with polyacrylic acid in aqueous suspension for evaluating of nanotube dispersion by various methods which are frequently used to assess SWNT dispersion. The prepared composite demonstrated the intensive SWNT emission in acidic envi-ronment (at pH2). Upon varying pH, SWNT aggregation in the dispersion was evaluated by scattering and optical spectroscopy. The researchers concluded that small-angle neutron scattering provides the most direct measure of dispersion. Among optical methods, fluorescence spectroscopy was reported as a sensitive method to control SWNT dispersion since nanotube bundling quenches the NIR fluorescence. In contrast, optical absorption spectroscopy and RRS have limited sensitivity to dispersions without appreciable bundling.

SWNTs with other aqueous soluble polymer, poly(l-lysine), manifested some distinctive pH response of absorption spectra as was reported by Wang and Chen [290]. They revealed that SWNTs sonicating in poly(l-lysine) solutions with pH values greater than 9 does not lead to stable dispersions. In addition, AFM images showed that SWNTs were individually dispersed in acidic and neutral environ-ments but aggregated into large bundles at pH 9. SWNT emission is partially quenched because of aggregation at pH 9.7 but remains largely unchanged under other pH conditions. It was supposed that at high pH the polymer changes its con-formation resulting in fewer contacts of hydrocarbon linkers with SWNT surface.

149

References

1. Kumar, C. (ed.): Nanomaterials for Biosensors. Wiley-VCH Verlag GmbH&Co./KgaA, Weinheim (2007)

2. Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.): Carbon Nanotubes: Advanced Topics in the Synthesis, Structure, Properties and Applications. Springer, Berlin (2008)

3. Javey, A., Kong, J. (eds.): Carbon Nanotube Electronics. Springer, New York (2008) 4. Léonard, F.: The Physics of Carbon Nanotube Devices. William Andrew, Norwich (2009) 5. Guldi, D.M., Martın, N. (eds.): Carbon Nanotubes and Related Structures: Synthesis,

Characterization, Functionalization, and Applications. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim (2010)

6. Nakashima, N., Okuzono, S., Murakami, H., Nakai, T., Yoshikawa, K.: DNA dissolves sin-gle-walled carbon nanotubes in water. Chem. Lett. 32, 456–457 (2003)

7. Zheng, M., Jagota, A., Semke, E., Diner, B., Mclean, R., Lustig, S., Richardson, R., Tassi, N.: DNA-assisted dispersion and separation of carbon nanotubes. Nat. Mater. 2, 338–342 (2003)

8. O’Connell, M.J., Boul, P., Ericson, L.M., Huffman, Ch., Wang, Y., Haros, E., Kuper, C., Tour, J., Ausman, K.D., Smalley, R.E.: Reversible water solubilization of single-walled car-bon nanotubes by polymer wrapping. Chem. Phys. Lett. 342, 265–271 (2001)

9. Zheng, M., Jagota, A., Strano, M., Santos, A., Barone, P., Chou, S., Diner, B., Dresselhaus, M., Mclean, R., Onoa, G., Samsonidze, G., Semke, E., Usrey, M., Walls, D.: Structure-based carbon nanotube sorting by sequence-dependent DNA assembly. Science 302, 1545–1548 (2003)

10. Tu, X., Manohar, S., Jagota, A., Zheng, M.: DNA sequence motifs for structure-specific recognition and separation of carbon nanotubes. Nature 460, 250–253 (2009)

11. Yarotski, D.A., Kilina, S.V., Talin, A.A., Tretiak, S., Prezhdo, O.V., Balatsky, A.V., Taylor, A.J.: Scanning tunneling microscopy of DNA-wrapped carbon nanotubes. Nano Lett. 9, 12–17 (2009)

12. Star, A., Tu, E., Niemann, J., Gabriel, J.P., Joiner, C.S., Valcke, C.: Label-free detection of DNA hybridization using carbon nanotube network field-effect transistors. Proc. Natl. Acad. Sci. U.S.A. 103, 921–926 (2006)

13. Jeng, E.S., Moll, A.E., Roy, A.C., Gastala, J.B., Strano, M.S.: Detection of DNA hybridi-zation using the near-infrared band-gap fluorescence of single-walled carbon nanotubes. Nano Lett. 6, 371–375 (2006)

14. Hazani, M., Hennrich, F., Kappes, M., Naaman, R., Peled, D., Sidorov, V., Shvarts, D.: DNA-mediated self-assembly of carbon nanotube-based electronic devices. Chem. Phys. Lett. 391, 389–392 (2004)

15. Li, S., He, P., Dong, J., Guo, Z., Dai, L.: DNA-directed self-assembling of carbon nano-tubes. J. Am. Chem. Soc. 127, 14–15 (2005)

16. Liu, H., He, J., Tang, J., Liu, H., Pang, P., Cao, D., Krstic, P., Joseph, S., Lindsay, S., Nuckolls, C.: Translocation of single-stranded DNA through single-walled carbon nano-tubes. Science 327, 64–67 (2010)

17. Lulevich, V., Kim, S., Grigoropoulos, C.P., Noy, A.: Frictionless sliding of single-stranded DNA in a carbon nanotube pore observed by single molecule force spectroscopy. Nano Lett. 11, 1171–1176 (2011)

18. Gao, H., Kong, Y., Cui, D.: Spontaneous insertion of DNA oligonucleotides into carbon nanotubes. Nano Lett. 3, 471–473 (2003)

19. Okada, T., Kaneko, T., Hatakeyama, R., Tohji, K.: Electrically triggered insertion of sin-gle-stranded DNA into single-walled carbon nanotubes. Chem. Phys. Lett. 417, 288–292 (2006)

20. Yeh, I.C., Hummer, G.: Nucleic acid transport through carbon nanotube membranes. Proc. Natl. Acad. Sci. U.S.A. 101, 12177–12182 (2004)

References


21. Xie, Y., Kong, Y., Soh, A.K., Gao, H.: Electric field-induced translocation of single-stranded DNA through a polarized carbon nanotube membrane. J. Chem. Phys. 127, 225101–225108 (2007)

22. O’Connell, M.J., Bachilo, S.M., Huffman, C.B., Moore, V.C., Strano, M.S., Haroz, E.H., Rialon, K.L., Boul, P.J., Noon, W.H., Kitrell, C., Ma, J., Hauge, R.H., Weisman, R.B., Smalley, R.E.: Band gap fluorescence from individual single-walled carbon nanotubes. Science 297, 593–596 (2002)

23. Dresselhaus, M.S., Dresselhaus, G., Saito, R., Jorio, A.: Exciton photophysics of carbon nanotubes. Annu. Rev. Phys. Chem. 58, 719–747 (2007)

24. Carlson, L.J., Krauss, T.D.: Photophysics of individual single-walled carbon nanotubes. Acc. Chem. Res. 41, 235–243 (2008)

25. Ma, Y.-Z., Hertel, T., Vardeny, Z.V., Fleming, G.R., Valkunas, L.: Ultrafast spectroscopy of carbon nanotubes. In: Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 321–352. Springer, Heidelberg (2008)

26. Lefebvre, J., Maruyama, S., Finnie, P.: Photoluminescence: science and applications. In: Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 287–319. Springer, Heidelberg (2008)

27. Spataru, C.D., Ismail-Beigi, S., Capaz, R.B., Louie, S.G.: Quasiparticle and excitonic effects in the optical response of nanotubes and nanoribbons. In: Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 195–227. Springer, Heidelberg (2008)

28. Hertel, T.: Photophysics in carbon nanotubes and related structures. In: Guldi, D., Martin, N. (eds.) Carbon Nanotubes and Related Structures, pp. 77–102. Wiley-VCH Verlag GmbH, Weinheim (2010)

29. Avouris, P., Freitag, M., Perebeinos, V.: Carbon-nanotube optoelectronics. In: Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 423–454. Springer, Heidelberg (2008)

30. Matsuda, K.: Exciton dephasing in a single carbon nanotube studied by photoluminescence spectroscopy. In: Marulanda, J.M. (ed.) Carbon Nanotubes Electronic Properties of Carbon Nanotubes, pp. 353–368. InTech, Vukovar (2011)

31. Ohno, Y., Maruyama, S., Mizutani, T.: Environmental effects on photoluminescence of sin-gle-walled carbon nanotubes. In: Marulanda, J.M. (ed.) Carbon Nanotubes, pp. 109–121. InTech, Vukovar (2010)

32. Bergin, S.D., Sun, Z., Streich, P., Hamilton, J., Coleman, J.N.: New solvents for nanotubes: approaching the dispersibility of surfactants. J. Phys. Chem. C 114, 231–237 (2010)

33. Cheng, Q., Debnath, S., O’Neill, L., Hedderman, T.G., Gregan, E., Byrne, H.J.: Systematic study of the dispersion of SWNTs in organic solvents. J. Phys. Chem. C 114, 4857–4863 (2010)

34. Forney, M.W., Poler, J.C.: Significantly enhanced single-walled carbon nanotube dispersion stability in mixed solvent systems. J. Phys. Chem. C 115, 10531–10536 (2011)

35. Moore, V.C., Strano, M.S., Haroz, E.H., Hauge, R.H., Smalley, R.E., Schmidt, J., Talmon, Y.: Individually suspended single-walled carbon nanotubes in various surfactants. Nano Lett. 3, 1379–1382 (2003)

36. Islam, M.F., Rojas, E., Bergey, D.M., Johnson, A.T., Yodh, A.G.: High weight fraction surfactant solubilization of single-wall carbon nanotubes in water. Nano Lett. 3, 269–273 (2003)

37. Nish, A., Nicholas, R.J.: Temperature induced restoration of fluorescence from oxidised single walled carbon nanotubes in aqueous sodium dodecylsulfate solution. Phys. Chem. Chem. Phys. 8, 3547–3551 (2006)

38. Blanch, A.J., Lenehan, C.E., Quinton, J.S.: Optimizing surfactant concentrations for disper-sion of single-walled carbon nanotubes in aqueous solution. J. Phys. Chem. B 114, 9805–9811 (2010)

39. Nair, N., Kim, W.-J., Braatz, R.D., Strano, M.S.: Dynamics of surfactant-suspended single-walled carbon nanotubes in a centrifugal field. Langmuir 24, 1790–1795 (2008)

151

40. Nish, A., Hwang, J.-Y., Doig, J., Nicholas, R.J.: Highly selective dispersion of single-walled carbon nanotubes using aromatic polymers. Nat. Nanotechnol. 2, 640–646 (2007)

41. Sánchez-Pomales, G., Pagán-Miranda, C., Santiago-Rodríguez, L., Cabrera, C.R.: DNA-wrapped carbon nanotubes: from synthesis to applications. In: Marulanda, J.M. (ed.) Carbon Nanotubes, pp. 721–748. InTeh, Vukovar (2010)

42. Koh, B., Park, J.B., Hou, X., Cheng, W.: Comparative dispersion studies of single-walled carbon nanotubes in aqueous solution. J. Phys. Chem. B 115, 2627–2633 (2011)

43. Kam, N.W.S., Liu, Z., Dai, H.: Carbon nanotubes as intracellular transporters for proteins and DNA: an investigation of the uptake mechanism and pathway. Angew. Chem. Int. Ed. 45, 577–581 (2006)

44. Singh, R., Pantarotto, D., McCarthy, D.: Binding and condensation of plasmid DNA onto functionalized carbon nanotubes: toward the construction of nanotube based gene delivery vectors. J. Am. Chem. Soc. 127, 4388–4396 (2005)

45. Taghdisi, S.M., Lavaee, P., Ramezani, M., Abnous, K.: Reversible targeting and controlled release delivery of daunorubicin to cancer cells by aptamer-wrapped carbon nanotubes. Eur. J. Pharm. Biopharm. 77, 200–206 (2011)

46. Heller, D.A., Jin, H., Martinez, B.M., Patel, D., Miller, B.M., Yeung, T.-K., Jena, P.V., Hobartner, C., Ha, T., Silverman, S.K., Strano, M.S.: Multimodal optical sensing and ana-lyte specificity using single-walled carbon nanotubes. Nat. Nanotechnol. 4, 114–120 (2009)

47. Tu, X., Zheng, M.: DNA-based approach to the carbon nanotube sorting problem. Nano Res. 1, 185–194 (2008)

48. Strano, M.S., Zheng, M., Jagota, A., Onoa, G.B., Heller, D.A., Barone, P.W., Usrey, M.L.: Understanding the nature of the DNA-assisted separation of single-walled carbon nano-tubes using fluorescence and Raman spectroscopy. Nano Lett. 4, 543–550 (2004)

49. Arnold, M.S., Stupp, S.I., Hersam, M.C.: Enrichment of single-walled carbon nanotubes by diameter in density gradients. Nano Lett. 5, 713–718 (2005)

50. Zhang, L., Zaric, S., Tu, X., Wang, X., Zhao, W., Dai, H.: Assessment of chemically sepa-rated carbon nanotubes for nanoelectronics. J. Am. Chem. Soc. 130, 2686–2691 (2008)

51. Lustig, S.R., Jagota, A., Khripin, C., Zheng, M.: Structure-based carbon nanotube separa-tions by ion-surface interactions. Mater. Res. Symp. Proc. 923, 1–6 (2006)

52. Huang, X., McLean, R.S., Zheng, M.: Huang high-resolution length sorting and purifica-tion of DNA-wrapped carbon nanotubes by size-exclusion chromatography. Anal. Chem. 77, 6225–6228 (2005)

53. Bauer, B.J., Becker, M.L., Bajpai, V., Fagan, J.A., Hobbie, E.K., Migler, K., Guttman, C.M., Blair, W.R.: Measurements of single-wall nanotube dispersion by size exclusion chromatography. J. Phys. Chem. C 111, 17914–17918 (2007)

54. Asada, Y., Sugai, T., Kitaura, R., Shinohara, H.: Chromatographic length separation and photoluminescence study on DNA-wrapped single-wall and double-wall carbon nanotubes. J. Nanomaterials 2009, 1–9 (2009)

55. Chun, J., Fagan, J.A., Hobbie, E.K., Bauer, B.J.: Size separation of single-wall carbon nanotubes by flow-field flow fractionation. Anal. Chem. 80, 2514–2523 (2008)

56. Sickert, D., Taeger, S., Neumann, A., Jost, O., Eckstein, G., Mertig, M., Pompe, W.: Separation and assembly of DNA-dispersed carbon nanotubes by dielectrophoresis. AIP Conf. Proc. 786, 271–274 (2005)

57. Vetcher, A.A., Srinivasan, S., Vetcher, I.A., Abramov, S.M., Kozlov, M., Baughman, R.H., Levene, S.D.: Fractionation of SWNT/nucleic acid complexes by agarose gel electrophore-sis. Nanotechnology 17, 4263–4269 (2006)

58. Müller, K., Richert, C.: The unlikely surfactant: DNA as a ligand for single-walled carbon nanotubes. In: Marulanda, J.M. (ed.) Carbon Nanotubes, pp. 749–766. InTeh, Vukovar (2010)

59. Haggenmueller, R., Rahatekar, S.S., Fagan, J.A., Chun, J.H., Becker, M.L., Naik, R.R., Krauss, T., Carlson, L., Kadla, J.F., Trulove, P.C., Fox, D.F., DeLong, H.C., Fang, Z.C., Kelley, S.O., Gilman, J.W.: Comparison of the quality of aqueous dispersions of single wall carbon nanotubes using surfactants and Biomolecules. Langmuir 24, 507050–507078 (2008)

References


60. Simpson, J.R., Fagan, J.A., Becker, M.L., Hobbie, E.K., Hight Walker, A.R.: The effect of dispersant on defects in length-separated single-wall carbon nanotubes measured by Raman spectroscopy. Carbon 47, 3238–3241 (2009)

61. Campbell, J.F., Tessmer, I., Thorp, H.H., Erie, D.A.: Atomic force microscopy studies of DNA-wrapped carbon nanotube structure and binding to quantum dots. J. Am. Chem. Soc. 130, 10648–10655 (2008)

62. Karachevtsev, M.V., Lytvyn, O.S., Stepanian, S.G., Leontiev, V.S., Adamowicz, L., Karachevtsev, V.A.: SWNT-DNA and SWNT-polyC hybrids: AFM study and computer modeling. J. Nanosci. Nanotechnol. 8, 1473–1480 (2008)

63. Toita, S., Kang, D., Kobayashi, K., Kawamoto, H., Kojima, K., Tachibana, M.: Atomic force microscopic study on DNA-wrapping for different diameter single-wall carbon nano-tubes. Diam. Relat. Mat. 17, 1389–1393 (2008)

64. Cathcart, H., Nicolosi, V., Hughes, J.M., Blau, W.J., Kelly, J.M., Quinn, S.J., Coleman, J.N.: Ordered DNA wrapping switches on luminescence in single-walled nanotube disper-sions. J. Am. Chem. Soc. 130, 12734–12744 (2008)

65. Malik, S., Vogel, S., Rosner, H., Arnold, K., Hennrich, F., Kohler, A.-K., Richert, C., Kappes, M.M.: Physical chemical characterization of DNA–SWNT suspensions and asso-ciated composites. Compos. Sci. Technol. 67, 916–921 (2007)

66. Karachevtsev, V.A., Glamazda, A. Yu., Leontiev, V.S., Mateichenko, P.V., Dettlaff-Weglikowska, U.: SWNTs with DNA in aqueous solution and films. AIP Conf. Proceed. 786, 257–262 (2005)

67. Heller, D.A., Jeng, E.S., Yeung, T.-K., Martinez, B.M., Moll, A.E., Gastala, J.B., Strano, M.S.: Optical detection of DNA conformational polymorphism on single-walled carbon nanotubes. Science 311, 508–511 (2006)

68. Cha, M., Jung, S., Cha, M.-H., Kim, G., Ihm, J., Lee, J.: Reversible metal-semiconductor transition of ssDNA-decorated single-walled carbon nanotubes. Nano Lett. 9, 1345–1349 (2009)

69. Bobadilla, A.D., Seminario, J.M.: DNA-CNT Interactions and Gating Mechanism Using MD and DFT. J. Phys. Chem. C 115, 3466–3474 (2011)

70. Gao, H., Kong, Y.: Simulation of DNA-nanotube interactions. Annu. Rev. Mater. Res. 34, 123–150 (2004)

71. Lu, G.P., Maragakis, P., Kaxiras, E.: Carbon nanotube interaction with DNA. Nano Lett. 5, 897–900 (2005)

72. Gladchenko, G.O., Karachevtsev, M.V., Leontiev, V.S., Valeev, V.A., Glamazda, A. Yu., Plokhotnichenko, A.M., Stepanian, S.G.: Hybrids of fragmented double-stranded DNA and carbon nanotubes in aqueous solution. Mol. Phys. 104, 3193–3201 (2006)

73. Manohar, S., Tang, T., Jagota, A.: Structure of homopolymer DNA-CNT hybrids. J. Phys. Chem. C 111, 17835–17845 (2007)

74. Zhao, X., Johnson, J.K.: Simulation of adsorption of DNA on carbon nanotubes. J. Am. Chem. Soc. 129, 10438–10445 (2007)

75. Frischknecht, A.L., Martin, M.G.: Simulation of the adsorption of nucleotide monophos-phates on carbon nanotubes in aqueous solution. J. Phys. Chem. C 112, 6271–6278 (2008)

76. Karachevtsev, V.A., Gladchenko, G.O., Karachevtsev, M.V., Valeev, V.A., Leontiev, V.S., Lytvyn, O.S.: Adsorption of poly(rA) on the carbon nanotube surface and its hybridization with poly(rU). Chem. Phys. Phys. Chem. 9, 2872–2881 (2008)

77. Martin, W., Zhu, W., Krilov, G.: Simulation study of noncovalent hybridization of carbon nanotubes by single-stranded DNA in water. J. Phys. Chem. B 112, 16076–16089 (2008)

78. Johnson, R.R., Johnson, A.T.C., Klein, M.L.: Probing the structure of DNA-carbon nano-tube hybrids with molecular dynamics. Nano Lett. 8, 69–75 (2008)

79. Johnson, R.R., Kohlmeyer, A., Johnson, A.T.C., Klein, M.L.: Free energy landscape of a DNA-carbon nanotube hybrid using replica exchange molecular dynamics. Nano Lett. 9, 537–541 (2009)

80. Johnson, R.R., Johnson, A.T.C., Klein, M.L.: The nature of DNA-base–carbon-nanotube interactions. Small 6, 31–34 (2010)

153

81. Roxbury, D., Manohar, S., Jagota, A.: Sequence-specific self-stitching motif of short single-stranded DNA on a single-walled carbon nanotube. J. Phys. Chem. C 114, 13267–13276 (2010)

82. Karachevtsev, M.V., Karachevtsev, V.A.: Peculiarities of homooligonucleotides wrapping around carbon nanotubes: molecular dynamics modeling. J. Phys. Chem. B 115, 9271–9279 (2011)

83. Roxbury, D., Jagota, A., Mittal, J.: Sequence-specific self-stitching motif of short single-stranded DNA on a single-walled carbon nanotube. J. Am. Chem. Soc. 133, 13545–13550 (2011)

84. Das, A., Sood, A.K., Maiti, P.K., Das, M., Varadarajan, R., Rao, C.N.R.: Binding of nucle-obases with single-walled carbon nanotubes: theory and experiment. Chem. Phys. Lett. 453, 266–273 (2008)

85. Gowtham, S., Scheicher, R.H., Pandey, R., Karna, S.P., Ahuja, R.: First-principles study of physisorption of nucleic acid bases on small-diameter carbon nanotubes. Nanotechnology 19, 125701–125718 (2008)

86. Shtogun, Y.V., Woods, L.M., Dovbeshko, G.I.: Adsorption of adenine and thymine and their radicals on single-wall carbon nanotubes. J. Phys. Chem. C 111, 18174–18181 (2007)

87. Wang, Y., Bu, Y.: Noncovalent Interactions between cytosine and SWNT: curvature dependence of complexes via.ππ. Stacking and cooperative CH…. π/NH… π. J. Phys. Chem. B 111, 6520–6526 (2007)

88. Wang, Y.: Theoretical evidence for the stronger ability of thymine to disperse SWNT than cytosine and adenine: self-stacking of DNA bases vs their cross-stacking with SWNT. J. Phys. Chem. C 112, 14297–14303 (2008)

89. Stepanian, S.G., Karachevtsev, M.V., Glamazda, A. Yu., Karachevtsev, V.A., Adamowicz, L.: Stacking interaction of cytosine with carbon nanotubes: MP2, DFT and Raman spec-troscopy study. Chem. Phys. Lett. 459, 153–158 (2008)

90. Stepanian, S.G., Karachevtsev, M.V., Glamazda, A. Yu., Karachevtsev, V.A., Adamowicz, L.: Raman spectroscopy study and first-principles calculations of the interaction between nucleic acid bases and carbon nanotubes. J. Phys. Chem. A 113, 3621–3629 (2009)

91. Shukla, M.K., Dubey, M., Zakar, E., Namburu, R., Czyznikowska, Z., Leszczynski, J.: Interaction of nucleic acid bases with single-walled carbon nanotube. Chem. Phys. Lett. 480, 269–272 (2009)

92. Umadevi, D., Sastry, G.N.: Quantum mechanical study of physisorption of nucleobases on car-bon materials: graphene versus carbon nanotubes. J. Phys. Chem. Lett. 2, 1572–1576 (2011)

93. Meng, S., Maragakis, P., Papaloukas, C., Kaxiras, E.: DNA nucleoside interaction and identification with carbon nanotubes. Nano Lett. 7, 45–50 (2007)

94. Wang, H., Ceulemans, A.: Physisorption of adenine DNA nucleosides on zigzag and arm-chair single-walled carbon nanotubes: a first-principles study. Phys. Rev. B 79, 195419–195426 (2009)

95. Xiao, Z., Wang, X., Xu, X., Zhang, H., Li, Y., Wang, Y.: Base- and structure-dependent DNA dinucleotide–carbon nanotube interactions: molecular dynamics simulations and thermodynamic analysis. J. Phys. Chem. C 115, 21546–21558 (2011)

96. Sponer, J., Rileya, K.E., Hobza, P.: Nature and magnitude of aromatic stacking of nucleic acid bases. Phys. Chem. Chem. Phys. 10, 2595–2610 (2008)

97. Hughes, J.M., Cathcart, H., Coleman, J.N.: From dispersion and exfoliation of nanotubes with synthetic oligonucleotides: variation of dispersion efficiency and oligo-nanotube inter-action with base type. J. Phys. Chem. C 114, 11741–11747 (2010)

98. Enyashin, A.N., Gemming, S., Seifert, G.: DNA-wrapped carbon nanotubes. Nanotechnology 18, 245702-1-10 (2007)

99. Gladchenko, G.O., Lytvyn, O.S., Karachevtsev, M.V., Valeev, V.A., Leontiev, V.S., Karachevtsev, V.A.: Binding of polynucleotides with single-walled carbon nanotubes: effect of temperature. Materialwissenschaft und Werkstofftechnik 42, 92–97 (2011)

100. Roxbury, D., Tu, X., Zheng, M., Jagota, A.: Recognition Ability of DNA for carbon nano-tubes correlates with their binding affinity. Langmuir 27, 8282–8293 (2011)

References


101. Kim, S.N., Kuang, Z., Grote, J.G., Farmer, B.L., Naik, R.R.: Enrichment of (6,5) single wall carbon nanotubes using genomic DNA. Nano Lett. 8, 4415–4420 (2008)

102. Karachevtsev, V.A., Glamazda, A. Yu., Dettlaff-Weglikowska, U., Leontiev, V.S., Mateichenko, P.V., Roth, S., Rao, A.M.: Spectroscopic and SEM studies of SWNTs: poly-mer solutions and films. Carbon 44, 1292–1297 (2006)

103. Gigliotti, B., Sakizzie, B., Bethune, D.S., Shelby, R.M., Cha, J.N.: Sequence- independent helical wrapping of single-walled carbon nanotubes by long genomic DNA. Nano Lett. 6, 159–164 (2006)

104. Zhao, W., Gao, Y., Brook, M.A., Li, Y.: Wrapping single-walled carbon nanotubes with long single-stranded DNA molecules produced by rolling circle amplification. Chem. Commun. 1, 3582–3584 (2006)

105. Keren, K., Berman, R.S., Buchstab, E., Sivan, U., Braun, E.: DNA-templated carbon nano-tube field-effect transistor. Science 302, 1380–1382 (2003)

106. Fu, D., Okimoto, H., Lee, C.W., Takenobu, T., Iwasa, Y., Kataura, H., Li, L.-J.: Ultrasensitive detection of DNA molecules with high on/off single-walled carbon nanotube network. Adv. Mater. 22, 4867–4871 (2010)

107. Liang, Z., Lao, R., Wang, J., Liu, Y., Wang, L., Huang, Q., Song, S., Li, G., Fan, C.: Solubilization of single-walled carbon nanotubes with single-stranded DNA generated from asymmetric PCR. Int. J. Mol. Sci. 8, 705–713 (2007)

108. Jeng, E.S., Barone, P.W., Nelson, J.D., Strano, M.S.: Hybridization kinetics and thermody-namics of DNA adsorbed to individually dispersed single-walled carbon nanotubes. Small 3, 1602–1609 (2007)

109. Cathcart, H., Quinn, S., Nicolosi, V., Kelly, J.M., Blau, W.J., Coleman, J.N.: Spontaneous debundling of single-walled carbon nanotubes in DNA-based dispersions. J. Phys. Chem. C 111, 66–74 (2007)

110. Vogel, S.R., Kappes, M.M., Hennrich, F., Richert, C.: An unexpected new optimum in the structure space of DNA solubilizing single-walled Carbon nanotubes. Chem. Eur. J. 13, 1815–1820 (2007)

111. Noguchi, Y., Fujigaya, T., Niidome, Y., Nakashima, N.: Regulation of the near-IR spectral properties of individually dissolved single-walled carbon nanotubes in aqueous solutions of dsDNA. Chem. Eur. J. 14, 5966–5973 (2008)

112. Noguchi, Y., Fujigaya, T., Niidome, Y., Nakashima, N.: Single-walled carbon nanotubes/DNA hybrids in water are highly stable. Chem. Phys. Lett. 455, 249–251 (2008)

113. Yamamoto, Y., Fujigaya, T., Niidome, Y., Nakashima, N.: Fundamental properties of oligo double-stranded DNA/single-walled carbon nanotube nanobiohybrids. Nanoscale 2, 1767–1772 (2010)

114. Elsner, H.I., Lindblad, E.B.: Ultrasonic degradation of DNA. J. Mol. Cell. Biol. 8, 697–701 (1989)

115. Mann, T.L., Krull, U.J.: The application of ultrasound as a rapid method to provide DNA fragments suitable for detection by DNA biosensors. Biosens. Bioelectron. 20, 945–955 (2004)

116. Cantor, C.R., Schimmel, P.R.: Biophysical Chemistry. Part II. W.H. Freeman & Company, San Francisco (1980)

117. Tinoco, I.: Hypochromism in polynucleotides. J. Am. Chem. Soc. 82, 4785–4790 (1960) 118. Dresselhaus, M.S., Dresselhaus, G., Eklund, P.C: Science of Fullerenes and Carbon

Nanotubes. Academic Press, San Diego (1996) 119. Hamon, M.A., Itkis, M.E., Niyogi, S., Alvaraes, T., Kuper, C., Menon, M., Haddon, R.C.:

Effect of rehybridization on the electronic structure of single-walled carbon nanotubes. J. Am. Chem. Soc. 123, 11292–11293 (2001)

120. Weisman, R.B., Bachilo, S.M.: Dependence of optical transition energies on structure for single-walled carbon nanotubes in aqueous suspension: an empirical Kataura plot. Nano Lett. 3, 1235–1239 (2003)

121. Tu, X., Walker, A.R.H., Khripin, C.Y., Zheng, M.: Evolution of DNA sequences toward recognition of metallic armchair carbon nanotubes. J. Am. Chem. Soc. 133, 12998–13001 (2011)

155

122. Ajiki, H., Ando, T.: Ahronov-Bohm effect in carbon nanotubes. Phys. B 201, 349–352 (1994)

123. Bozovic, I., Bozovic, N., Damnjanovic, M.: Optical dichroism in nanotubes. Phys. Rev. B 62, 6971–6974 (2000)

124. Lefebvre, J., Fraser, J.M., Finnie, P., Homma, Y.: Photoluminescence from an individual single-walled carbon nanotube. Phys. Rev. B 69, 075403-1-075403-5 (2004)

125. Miyauchi, Y., Oba, M., Maruyama, S.: Cross-polarized optical absorption of single-walled nanotubes by polarized photoluminescence excitation spectroscopy. Phys. Rev. B 74, 205440-1-205440-6 (2006)

126. Nikolaev, P., Bronikowski, M.J., Bradley, R.K., Rohmund, F., Colbert, D.T., Smith, K.A., Smalley, R.E.: Gas-phase catalytic growth of single-walled carbon nanotubes from carbon monoxide. Chem. Phys. Lett. 313, 91–97 (1999)

127. Alvarez, W.E., Pompeo, F., Herrera, J.E., Balzano, L., Resasco, D.E.: Characterization of single-walled carbon nanotubes (SWNTs) produced by CO disproportionation on Co–Mo catalysts. Chem. Mater. 14, 1853–1858 (2002)

128. Bachilo, S.M., Strano, M.S., Kittrell, C., Hauge, R.H., Smalley, R.E., Weisman, R.B.: Structure-assigned optical spectra of single-walled carbon nanotubes. Science 298, 2361–2366 (2002)

129. Bachilo, S.M., Balzano, L., Herrera, J.E., Pompeo, F., Resasco, D.E., Weisman, R.B.: Narrow (n, m)-distribution of single-walled carbon nanotubes grown using a solid sup-ported catalyst. J. Am. Chem. Soc. 125, 11186–11187 (2003)

130. Fantini, C., Jorio, A., Santos, A.P., Peressinotto, V.S.T., Pimenta, M.A.: Characterization of DNA-wrapped carbon nanotubes by resonance Raman and optical absorption spectrosco-pies. Chem. Phys. Lett. 439, 138–142 (2007)

131. Hagen, A., Hertel, T.: Quantitative analysis of optical spectra from individual single-wall carbon nanotubes. Nano Lett. 3, 383–388 (2003)

132. Hartschuh, A., Pedrosa, H.N., Peterson, J., Huang, L., Anger, P., Qian, H., Meixner, A.J., Steiner, M., Novotny, L., Krauss, T.D.: Single carbon nanotube optical spectroscopy. Chem. Phys. Chem. 6, 577–582 (2005)

133. Pichler, T., Knupfer, M., Golden, M.S., Fink, J., Rinzler, A., Smalley, R.E.: Localized and delocalized electronic states in single-wall carbon nanotubes. Phys. Rev. Lett. 80, 4729–4732 (1998)

134. Reed, B.W., Sarikaya, M.: Electronic properties of carbon nanotubes by transmission elec-tron energy-loss spectroscopy. Phys. Rev. B 64, 195404–1–195404–13 (2001)

135. Murakami, Y., Einarsson, E., Edamura, T., Maruyama, S.: Polarization dependence of the optical absorption of single-walled carbon nanotubes. Phys. Rev. Lett. 94, 087402–1–087402–4 (2005)

136. Murakami, Y., Maruyama, S.: Coupling between localized resonance and excitation of sur-face waves in metal hole arrays. Phys. Rev. B 79, 155445–1–155445–4 (2009)

137. Rance, G.A., Marsh, D.H., Nicholas, R.J., Khlobystov, A.N.: UV–vis absorption spec-troscopy of carbon nanotubes: relationship between the p-electron plasmon and nanotube diameter. Chem. Phys. Lett. 493, 19–23 (2010)

138. Reich, S., Thomsen, C., Ordejon, P.: Electronic band structure of isolated and bundled car-bon nanotubes. Phys. Rev. B 65, 155411–1–155411–11 (2002)

139. Karachevtsev, V.A., Glamazda, A. Yu., Dettlaff-Weglikowska, U., Leontiev, V.S., Plokhotnichenko, A.M., Roth, S.: Spectroscopy study of SWNT in aqueous solution with different surfactants. AIP Conf. Proc. 685, 202–207 (2003)

140. Doorn, S.K.: Raman studies of new carbon nanotube sample types. J. Nanosci. Nanotechnol. 5, 1023–1034 (2005)

141. Park, J.S., Oyama, Y., Saito, R., Izumida, W., Jiang, J., Sato, K., Fantini, C., Jorio, A., Dresselhaus, G., Dresselhaus, M.S.: Raman resonance window of single-wall carbon nano-tubes. Phys. Rev. B 74, 165414–1–165414–6 (2006)

142. Itkis, M.E., Perea, D.E., Niyogi, S., Rickard, S.M., Hamon, M.A., Hu, H., Zhao, B., Haddon, R.C.: Purity evaluation of As-prepared single-walled carbon nanotube soot by use of solution-phase near-IR spectroscopy. Nano Lett. 3, 309–314 (2003)

References


143. Nepal, D., Kim, D.S., Geckeler, K.E.: A facile and rapid purification method for single-walled carbon nanotubes. Carbon 43, 660–662 (2005)

144. Naumov, A.V., Ghosh, S., Tsyboulski, D.A., Bachilo, S.M., Weisman, R.B.: Analyzing absorption backgrounds in single-walled carbon nanotube spectra. ACS Nano 5, 1639–1648 (2011)

145. Ryabenko, A.G., Dorofeeva, T.V., Zvereva, G.I.: UV–VIS–NIR spectroscopy study of sensitivity of single-wall carbon nanotubes to chemical processing and Van-der-Waals SWNT/SWNT interaction. Verification of the SWNT content measurements by absorption spectroscopy. Carbon 42, 1523–1535 (2004)

146. Nair, N., Usrey, M.L., Kim, W.J., Braatz, R.D., Strano, M.S.: Estimation of the (n, m) con-centration distribution of single-walled carbon nanotubes from photoabsorption spectra. Anal. Chem. 78, 7689–7696 (2006)

147. Dyke, C.A., Tour, J.M.: Unbundled and highly functionalized carbon nanotubes from aque-ous reactions. Nano Lett. 3, 1215–1218 (2003)

148. Ando, T.: Role of the Aharonov–Bohm phase in the optical properties of carbon nano-tubes. In: Jorio, A., Dresselhaus, G., Dresselhaus M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 229–250. Springer, Heidelberg (2008)

149. Naumov, A.V., Kuznetsov, O.A., Harutyunyan, A.R., Green, A.A., Hersam, M.C., Resasco, D.E., Nikolaev, P.N., Weisman, R.B.: Quantifying the semiconducting fraction in single-walled carbon nanotube samples through comparative atomic force and photoluminescence microscopies. Nano Lett. 9, 3203–3208 (2009)

150. Strano, M.S., Doorn, S.K., Haroz, E.H., Kittrell, C., Hauge, R.H., Smalley, R.E.: Assignment of (n, m) Raman and optical features of metallic single-walled carbon nano-tubes. Nano Lett. 3, 1091–1096 (2003)

151. Hartschuh, A., Pedrosa, H.N., Novotny, L., Krauss, T.D.: Simultaneous fluorescence and Raman scattering from single carbon nanotubes. Science 301, 1354–1356 (2003)

152. Luo, Z., Pfefferle, L.D., Haller, G.L., Papadimitrakopoulos, F.: (n, m) Abundance evalu-ation of single-walled carbon nanotubes by fluorescence and absorption spectroscopy. J. Am. Chem. Soc. 128, 15511–15516 (2006)

153. Topinka, M.A., Rowell, M.W., Goldhaber-Gordon, D., McGehee, M.D., Hecht, D.S., Gruner, G.: Charge transport in interpenetrating networks of semiconducting and metallic carbon nanotubes. Nano Lett. 9, 1866–1871 (2009)

154. Asada, Y., Miyata, Y., Shiozawa, K., Ohno, Y., Kitaura, R., Mizutani, T., Shinohara, H.: Thin-film transistors with length-sorted DNA-wrapped single-wall carbon nanotubes. J. Phys. Chem. C 115, 270–273 (2011)

155. Yao, Z., Kane, C.L., Dekker, C.: High-field electrical transport in single-wall carbon nano-tubes. Phys. Rev. Lett. 84, 2941–2944 (2000)

156. Wang, R., Sun, J., Gao, L., Zhang, J.: Dispersion of single-walled carbon nanotubes by DNA for preparing transparent conductive films. J. Mater. Chem. 20, 6903–6909 (2010)

157. Miyata, Y., Yanagi, K., Maniwa, Y., Kataura, H.: Optical evaluation of the metal-to-sem-iconductor ratio of single-wall carbon nanotubes. J. Phys. Chem. C 112, 13187–13191 (2008)

158. Attal, S., Thiruvengadathan, R., Regev, O.: Determination of the concentration of single-walled carbon nanotubes in aqueous dispersions using UV–visible absorption spectroscopy. Anal. Chem. 78, 8098–8104 (2006)

159. Jeong, S.H., Kim, K.K., Jeong, S.J., An, K.H., Lee, S.H., Lee, Y.H.: Optical absorption spectroscopy for determining carbon nanotube concentration in solution. Synth. Met. 157, 570–574 (2007)

160. Huang, L., Zhang, H., Wu, B., Liu, Y., Wei, D., Chen, J., Xue, Y., Yu, G., Kajiura, H., Li, Y.: A generalized method for evaluating the metallic-to-semiconducting ratio of separated single-walled carbon nanotubes by UV-vis-NIR characterization. J. Phys. Chem. C 114, 12095–12098 (2010)

161. Jacobsen, N.S., Pantano, P.: Determining the percentages of semi-conducting and metallic single-walled carbon nanotubes in bulk soot. Carbon 49, 1998–2006 (2011)

157

162. Tan, Y., Resasco, D.E.: Dispersion of single-walled carbon nanotubes of narrow diameter distribution. J. Phys. Chem. B 109, 14454–14460 (2005)

163. Berciaud, S., Cognet, L., Poulin, P., Weisman, R.B., Lounis, B.: Absorption spectroscopy of individual single-walled carbon nanotubes. Nano Lett. 7, 1203–1207 (2007)

164. Islam, M.F., Milkie, D.E., Kane, C.L., Yodh, A.G., Kikkawa, J.M.: Direct measurement of the polarized absorption cross section of single-wall carbon nanotubes. Phys. Rev. Lett. 93, 037404-1-037404-4 (2004)

165. Jiang, J., Saito, R., Gruneis, A., Dresselhaus, G., Dresselhaus, M.S.: Optical absorption matrix element in single-wall carbon nanotubes. Carbon 42, 3169–3176 (2004)

166. Kam, N.W.S., O’Connell, M., Wisdom, J.A., Dai, H.: Carbon nanotubes as multifunctional biological transporters and near-infrared agents for selective cancer cell destruction. Proc. Natl. Acad. Sci. U.S.A. 102, 11600–11605 (2005)

167. Schoppler, F., Mann, C., Hain, T.C., Neubauer, F.M., Privitera, G., Bonaccorso, F., Chu, D.P., Ferrari, A.C., Hertel, T.: Molar extinction coefficient of single-wall carbon nanotubes. J. Phys. Chem. C 115, 14682–14686 (2011)

168. Zhao, B., Itkis, M.E., Niyogi, S., Hu, H., Zhang, J., Haddon, R.C.: Study of the extinction coefficients of single-walled carbon nanotubes and related carbon materials. J. Phys. Chem. B 108, 8136–8141 (2004)

169. Landi, B.J., Ruf, H.J., Evans, C.M., Cress, C.D., Raffaelle, R.P.: Thermal oxidation profil-ing of single-walled carbon nanotubes. J. Phys. Chem. B 109, 9952–9965 (2005)

170. Kim, W.-J., Lee, C.Y., O’brien, K.P., Plombon, J.J., Blackwell, J.M., Strano, M.S.: Connecting single molecule electrical measurements to ensemble spectroscopic proper-ties for quantification of single-walled carbon nanotube separation. J. Am. Chem. Soc. 131, 3128–3129 (2009)

171. Fantini, C., Cassimiro, J., Peressinotto, V.S.T., Plentz, F., Souza Filho, A.G., Furtado, C.A., Santos, A.P.: Investigation of the light emission efficiency of single-wall carbon nanotubes wrapped with different surfactants. Chem. Phys. Lett. 473, 96–101 (2009)

172. Miyauchi, Y., Saito, R., Sato, K., Ohno, Y., Iwasaki, S., Mizutani, T., Jiang, J., Maruyama, S.: Dependence of exciton transition energy of single-walled carbon nanotubes on sur-rounding dielectric materials. Chem. Phys. Lett. 442, 394–399 (2007)

173. Hughes, M.E., Brandin, E., Golovchenko, J.A.: Optical absorption of DNA–carbon nano-tube structures. Nano Lett. 7, 1191–1194 (2007)

174. Karachevtsev, V.A., Plokhotnichenko, A.M., Karachevtsev, M.V., Leontiev, V.S.: Decrease of carbon nanotube UV light absorption induced by π-stacking interaction with nucleotide bases. Carbon 48, 3682–3691 (2010)

175. Snyder, S.E., Rotkin, S.V.: Optical identification of a DNA-wrapped carbon nanotube: signs of helically broken symmetry. Small 4, 1284–1286 (2008)

176. Puller, V., Rotkin, S.V.: Helicity and broken symmetry of DNA-nanotube hybrids. Europhys. Lett. 77, 27006 (2007)

177. Michalski, P.J., Mele, E.J.: Carbon nanotubes in helically modulated potentials. Phys. Rev. B 77, 085429-1-085429-11 (2008)

178. Rotkin, S.V.: Electronic properties of nonideal nanotube materials: helical symmetry break-ing in DNA hybrids. Annu. Rev. Phys. Chem. 61, 241–261 (2010)

179. Kasha, M.: Characterization of electronic transitions in complex molecules. Discuss. Faraday Soc. 9, 14–19 (1950)

180. Jones, M., Engtrakul, C., Metzger, W.K., Ellingson, R.J., Nozik, A.J., Heben, M.J., Rumbles, G.: Analysis of photoluminescence from solubilized single-walled carbon nano-tubes. Phys. Rev. B 71, 115426-1-115426-9 (2005)

181. Rocha, J.-D.R., Bachilo, S.M., Ghosh, S., Arepalli, S., Weisman, R.B.: Efficient spectro-fluorimetric analysis of single-walled carbon nanotube samples. Anal. Chem. 83, 7431–7437 (2011)

182. Ando, T.: Excitons in carbon nanotubes. J. Phys. Soc. Jpn. 66, 1066–1073 (1997) 183. Kane, C.L., Mele, E.J.: Ratio problem in single nanotube fluorescence spectroscopy. Phys.

Rev. Lett. 90, 207401-1-207401-4 (2003)

References


184. Wang, F., Dukovic, G., Brus, L.E., Heinz, T.F.: The optical resonances in carbon nanotubes arise from excitons. Science 308, 838–841 (2005)

185. Maultzsch, J., Pomraenke, R., Reich, S., Chang, E., Prezzi, D., Ruini, A., Molinari, E., Strano, M.S., Thomsen, C., Lienau, C.: Exciton binding energies in carbon nanotubes from two-photon photoluminescence. Phys. Rev. B 72, 241402–1–241402–4 (2005)

186. Dukovic, G., Wang, F., Song, D., Sfeir, M.Y., Heinz, T.F., Brus, L.E.: Structural depend-ence of excitonic optical transitions and band-gap energies in carbon nanotubes. Nano Lett. 5, 2314–2318 (2005)

187. Spataru, C.D., Ismail-Beigi, S., Benedict, L.X., Louie, S.G.: Excitonic effects and optical spectra of single-walled carbon nanotubes. Phys. Rev. Lett. 93, 077402-1- 077402-4 (2004)

188. Perebeinos, V., Tersoff, J., Avouris, P.: Scaling of excitons in carbon nanotubes. Phys. Rev. Lett. 92, 257402–1–257402–4 (2004)

189. Chang, E., Bussi, G., Ruini, A., Molinari, E.: Excitons in carbon nanotubes: an ab initio symmetry-based approach. Phys. Rev. Lett. 92, 196401–1–196401–4 (2004)

190. Pedersen, T.G.: Variational approach to excitons in carbon nanotubes. Phys. Rev. B 67, 073401–1–073401–4 (2003)

191. Zhao, H., Mazumdar, S.: Electron–electron interaction effects on the optical excitations of semiconducting single-walled carbon nanotubes. Phys. Rev. Lett. 93, 157402–1–157402–4 (2004)

192. Kane, C.L., Mele, E.J.: Electron interactions and scaling relations for optical excitations in carbon nanotubes. Phys. Rev. Lett. 93, 197402–1–197402–4 (2004)

193. Malic, E., Hirtschulz, M., Reich, S., Knorr, A.: Excitonic absorption spectra and ultrafast dephasing dynamics in arbitrary carbon nanotubes. Phys. Status Solidi Rapid. Res. Lett. 3, 196–198 (2009)

194. Lüer, L., Hoseinkhani, S., Polli, D., Crochet, J., Hertel, T., Lanzani, G.: Size and mobility of excitons in (6,5) carbon nanotubes. Nat. Phys. 5, 54–58 (2009)

195. Tretiak, S., Kilina, S., Piryatinski, A., Saxena, A., Martin, R.L., Bishop, A.R.: Excitons and peierls distortion in conjugated carbon nanotubes. Nano Lett. 7, 86–92 (2007)

196. Perebeinos, V., Tersoff, J., Avouris, Ph: Radiative lifetime of excitons in carbon nanotubes. Nano Lett. 5, 2495–2499 (2005)

197. Jiang, J., Saito, R., Samsonidze, G.G., Jorio, A., Chou, S.G., Dresselhaus, G., Dresselhaus, M.S.: Chirality dependence of exciton effects in single-wall carbon nanotubes: tight-bind-ing model. Phys. Rev. B 75, 035407–1–035407–13 (2007)

198. Kono, J., Nicholas, R.J., Roche, S.: High magnetic field phenomena. In: Jorio, A., Dresselhaus, G., Dresselhaus, M.S. (eds.) Carbon Nanotubes, Topics in Applied Physics, vol. 111, pp. 393–422. Springer, Heidelberg (2008)

199. Shaver, J., Kono, J.: Temperature-dependent magneto-photoluminescence spectroscopy of carbon nanotubes: evidence for dark excitons. Laser Photon. Rev. 1, 260–274 (2007)

200. Matsunaga, R., Matsuda, K., Kanemitsu, Y.: Evidence for dark excitons in a single carbon nanotube due to the Aharonov-Bohm effect. Phys. Rev. Lett. 101, 147404–1–147404–4 (2008)

201. Srivastava, A., Htoon, H., Klimov, V.I., Kono, J.: Direct observation of dark excitons in individual carbon nanotubes: inhomogeneity in the exchange splitting. Phys. Rev. Lett. 101, 087402–1–087402–4 (2008)

202. Shaver, J., Crooker, S.A., Fagan, J.A., Hobbie, E.K., Ubrig, N., Portugall, O., Perebeinos, V., Avouris, Ph., Kono, J.: Magneto-optical spectroscopy of highly aligned carbon nano-tubes: identifying the role of threading magnetic flux. Phys. Rev. B 78, 081402(R) (2008)

203. Matsunaga, R., Matsuda, K., Kanemitsu, Y.: Direct observation of dark exciton states in single carbon nanotubes. J. Lumin. 129, 1702–1705 (2009)

204. Mortimer, I.B., Nicholas, R.J.: Role of bright and dark excitons in the temperature-depend-ent photoluminescence of carbon nanotubes. Phys. Rev. Lett. 98, 027404–1–027404–4 (2007)

205. Berger, S., Voisin, C., Cassabois, G., Delalande, C., Roussignol, P.: Temperature depend-ence of exciton recombination in semiconducting single-wall carbon nanotubes. Nano Lett. 7, 398–402 (2007)

159

206. Scholes, G.D., Tretiak, S., McDonald, T.J., Metzger, W.K., Engtrakul, C., Rumbles, G., Heben, M.J.: Low-lying exciton states determine the photophysics of semiconducting sin-gle wall carbon nanotubes. J. Phys. Chem. C 111, 11139–11149 (2007)

207. Metzger, W.K., McDonald, T.J., Engtrakul, C., Blackburn, J.L., Scholes, G.D., Rumbles, G., Heben, M.J.: Temperature-dependent excitonic decay and multiple states in single-wall carbon nanotubes. J. Phys. Chem. C 111, 3601–3606 (2007)

208. Ma, Y.Z., Valkunas, L., Bachilo, S.M., Fleming, G.R.: Temperature effects on femtosec-ond transient absorption kinetics of semiconducting single-walled carbon nanotubes. Phys. Chem. Chem. Phys. 8, 5689–5693 (2006)

209. Hagen, A., Steiner, M., Raschke, M.B., Lienau, C., Hertel, T., Qian, H., Meixner, A.J., Hartschuh, A.: Exciton dynamics in individual single–walled carbon nanotubes. Phys. Rev. Lett. 95, 197401–1–197401–4 (2005)

210. Takeyama, S., Suzuki, H., Yokoi, H., Murakami, Y., Maruyama, Sh.: Aharonov-Bohm exci-ton absorption splitting in chiral specific single-walled carbon nanotubes in magnetic fields of up to 78 T. Phys. Rev. B 83, 235405–1–235405–4 (2011)

211. Jones, M., Metzger, W.K., McDonald, T.J., Engtrakul, C., Ellingson, R.J., Rumbles, G., Heben, M.J.: Extrinsic and intrinsic effects on the excited-state kinetics of single-walled carbon nanotubes. Nano Lett. 7, 300–306 (2007)

212. Zhu, Z., Crochet, J., Arnold, M.S., Hersam, M.C., Ulbricht, H., Resasco, D., Hertel, T.: Pump-probe spectroscopy of exciton dynamics in (6,5) carbon nanotubes. J. Phys. Chem. C 111, 3831–3835 (2007)

213. Habenicht, B.F., Prezhdo, O.V.: Nonradiative quenching of fluorescence in a semicon-ducting carbon nanotube: a time-domain ab initio study. Phys. Rev. Lett. 100, 197402–1–197402–4 (2008)

214. Kiowski, O., Arnold, K., Lebedkin, S., Hennrich, F., Kappes, M.M.: Direct observation of deep excitonic states in the photoluminescence spectra of single-walled carbon nanotubes. Phys. Rev. Lett. 99, 237402–1–237402–4 (2007)

215. Blackburn, J.L., McDonald, T.J., Metzger, W.K., Engtrakul, C., Rumbles, G., Heben, M.J.: Protonation effects on the branching ratio in photoexcited single-walled carbon nanotube dispersions. Nano Lett. 8, 1047–1054 (2008)

216. Kishida, H., Nagasawa, Y., Imamura, S., Nakamura, A.: Direct observation of dark exci-tons in micelle-wrapped single-wall carbon nanotubes. Phys. Rev. Lett. 100, 097401–1–097401–4 (2008)

217. Harutyunyan, H., Gokus, T., Green, A.A., Hersam, M.C., Allegrini, M., Hartschuh, A.: Defect-Induced photoluminescence from dark excitonic states in individual single-walled carbon nanotubes. Nano Lett. 9, 2010–2014 (2009)

218. Perebeinos, V., Tersoff, J., Avouris, P.: Effect of exciton-phonon coupling in the calculated optical absorption of carbon nanotubes. Phys. Rev. Lett. 94, 027402-1-4 (2005)

219. Torrens, O.N., Zheng, M., Kikkawa, J.M.: Energy of K-momentum dark excitons in carbon nanotubes by optical spectroscopy. Phys. Rev. Lett. 101, 157401-1-4 (2008)

220. Saito, R., Jorio, A., Souza Filho, A.G., Dresselhaus, G., Dresselhaus, M.S., Pimenta, M.A.: Probing phonon dispersion relations of graphite by double resonance Raman scattering. Phys. Rev. Lett. 88, 027401-1-4 (2002)

221. Miyauchi, Y., Maruyama, S.: Identification of an excitonic phonon sideband by photolumi-nescence spectroscopy of single-walled carbon-13 nanotubes. Phys. Rev. B 74, 035415-1-9 (2006)

222. Plentz, F., Ribeiro, H.B., Jorio, A., Strano, M.S., Pimenta, M.A.: Direct experimental evi-dence of exciton-phonon bound states in carbon nanotubes. Phys. Rev. Lett. 95, 247401-1-4 (2005)

223. Htoon, H., O’Connell, M.J., Doorn, S.K., Klimov, V.I.: Single carbon nanotubes probed by photoluminescence excitation spectroscopy: the role of phonon-assisted transitions. Phys. Rev. Lett. 94, 127403-1-4 (2005)

224. Vora, P.M., Tu, X., Mele, E.J., Zheng, M., Kikkawa, J.M.: Chirality dependence of the K-momentum dark excitons in carbon nanotubes. Phys. Rev. B 81, 155123-1-9 (2010)

References


225. Murakami, Y., Lu, B., Kazaoui, S., Minami, N., Okubo, T., Maruyama, S.: Photoluminescence sidebands of carbon nanotubes below the bright singlet excitonic lev-els. Phys. Rev. B 79, 195407-1-5 (2009)

226. Matsunaga, R., Matsuda, K., Kanemitsu, Y.: Origin of low-energy photoluminescence peaks in single carbon nanotubes: K -momentum dark excitons and triplet dark excitons. Phys. Rev. B 81, 033401-1-4 (2010)

227. Tsyboulski, D.A., Rocha, J.-D.R., Bachilo, S.M., Cognet, L., Weisman, R.B.: Structure-dependent fluorescence efficiencies of individual single-walled carbon nanotubes. Nano Lett. 7, 3080–3085 (2007)

228. Carlson, L.J., Maccagnano, S.E., Zheng, M., Silcox, J., Krauss, T.D.: Fluorescence effi-ciency of individual carbon nanotubes. Nano Lett. 7, 3698–3703 (2007)

229. Berge, J.J., Gallaway, C., Barron, A.R.: Fluorescence quenching of single-walled carbon nanotubes in SDBS surfactant suspension by metal ions: quenching efficiency as a function of metal and nanotube identity. J. Phys. Chem. C 111, 17812–17820 (2007)

230. Dukovic, G., White, B.E., Zhou, Z., Wang, F., Jockusch, S., Steigerwald, M.L., Heinz, T.F., Friesner, R.A., Turro, N.J., Brus, L.E.: Reversible surface oxidation and efficient lumines-cence quenching in semiconductor single-wall carbon nanotubes. J. Am. Chem. Soc. 126, 15269–15276 (2004)

231. Rajan, A., Strano, M.S., Heller, D.A., Hertel, T., Schulten, K.: Length-dependent optical effects in single-walled carbon nanotubes. Phys. Chem. B 112, 6211–6213 (2008)

232. Howard, W.: Dispersing carbon nanotubes using surfactants. Curr. Opin. Colloid. 14, 364–371 (2009)

233. Crochet, J., Clemens, M., Hertel, T.: Quantum yield heterogeneities of aqueous single-wall carbon nanotube suspensions. J. Am. Chem. Soc. 129, 8058–8059 (2007)

234. Lefebvre, J., Austing, D.G., Bond, J., Finnie, P.: Photoluminescence imaging of suspended singlewalled carbon nanotubes. Nano Lett. 6, 1603–1608 (2006)

235. Strano, M.S., Huffman, C.B., Moore, V.C., O’Connell, M.J., Haroz, E.H., Hubbard, J., Miller, M., Rialon, K., Kittrell, C., Ramesh, S., Hauge, R.H., Smalley, R.E.: Reversible, band-gap selective protonation of single-walled carbon nanotubes in solution. J. Phys. Chem. B 107, 6979–6985 (2003)

236. Wang, F., Dukovic, G., Knoesel, E., Brus, L.E., Heinz, T.F.: Observation of rapid Auger recombination in optically excited semiconducting carbon nanotubes. Phys. Rev. B 70, 241403-1-4 (2004)

237. Ju, S.-Y., Kopcha, W.P., Papadimitrokopoulos, F.: Brightly fluorescent single-walled car-bon nanotubes via an oxygen-excluding surfactant organization. Science 323, 1319–1323 (2009)

238. Rickard, D., Giordani, S., Blau, W.J., Coleman, J.N.: Quantifying the contributions of inner-filter, re-absorption and aggregation effects in the photoluminescence of high-concen-tration conjugated polymer solutions. J. Lumin. 128, 31–40 (2007)

239. Tsyboulski, D.A., Bakota, E.L., Witus, L.S., Rocha, J.-D.R., Hartgerink, J.D., Weisman, R.B.: Self-assembling peptide coatings designed for highly luminescent suspension of sin-gle-walled carbon nanotubes. J. Am. Chem. Soc. 130, 17134–17140 (2008)

240. Cherukuri, P., Bachilo, S.M., Litovsky, S.H., Weisman, R.B.: Near-infrared fluorescence microscopy of single-walled carbon nanotubes in phagocytic cells. J. Am. Chem. Soc. 126, 15638–15639 (2004)

241. Matsuura, K., Saito, T., Okazaki, T., Ohshima, S., Yumura, M., Iijima, S.: Selectivity of water-soluble proteins in single-walled carbon nanotube dispersions. Chem. Phys. Lett. 429, 497–502 (2006)

242. Bakota, E.L., Aulisa, L., Tsyboulski, D.A., Weisman, R.B., Hartgerink, J.D.: Multidomain peptides as single-walled carbon nanotube surfactants in cell culture. Biomacromolecules 10, 2201–2206 (2009)

243. Duque, J.G., Cognet, L., Parra-Vasquez, A.N.G., Nicholas, N., Schmidt, H.K., Pasquali, M.: Stable luminescence from individual carbon nanotubes in acidic, basic, and biological environments. J. Am. Chem. Soc. 130, 2626–2633 (2008)

161

244. Wang, R.K., Chen, W.-C., Campos, D.K., Ziegler, K.J.: Swelling the micelle core surround-ing single-walled carbon nanotubes with water-immiscible organic solvents. J. Am. Chem. Soc. 130, 16330–16337 (2008)

245. Silvera-Batista, C.A., Weinberg, Ph, Butler, J.E., Ziegler, K.J.: Long-term improvements to photoluminescence and dispersion stability by flowing SDS-SWNT suspensions through microfluidic channels. J. Am. Chem. Soc. 131, 12721–12728 (2009)

246. Lee, A.J., Wang, X., Carlson, L.J., Smyder, J.A., Loesch, B., Tu, X., Zheng, M., Krauss, T.D.: Bright fluorescence from individual single-walled carbon nanotubes. Nano Lett. 11, 1636–1640 (2011)

247. Fantini, C., Jorio, A., Souza, M., Strano, M.S., Dresselhaus, M.S., Pimenta, M.A.: Optical transition energies for carbon nanotubes from resonant Raman spectroscopy: environment and temperature effects. Phys. Rev. Lett. 93, 147406-1-4 (2004)

248. Okazaki, T., Saito, T., Matsuura, K., Ohshima, S., Yumura, M., Iijima, S.: Photoluminescence mapping of “as-grown” single-walled carbon nanotubes: a comparison with micelle-encapsulated nanotube solutions. Nano Lett. 5, 2618–2624 (2005)

249. Ohno, Y., Iwasaki, S., Murakami, Y., Kishimoto, S., Maruyama, S., Mizutani, T.: Chirality-dependent environmental effects in photoluminescence of single-walled carbon nanotubes. Phys. Rev. B 73, 235427-1-5 (2006)

250. Ohno, Y., Iwasaki, S., Murakami, Y., Kishimoto, S., Maruyama, S., Mizutani, T.: Environmental dielectric screening effect on exciton transition energies in single-walled carbon nanotubes. Phys. Status Solidi B 244, 4002–4005 (2007)

251. Kiowski, O., Lebedkin, S., Hennrich, F., Malik, S., Rosner, H., Arnold, K., Surgers, C., Kappes, M.M.: Photoluminescence microscopy of carbon nanotubes grown by chemical vapor deposition: influence of external dielectric screening on optical transition energies. Phys. Rev. B 75, 075421-1-7 (2007)

252. Choi, J.H., Strano, M.S.: Solvatochromism in single-walled carbon nanotubes. Appl Phys Lett 90, 223114-1-3 (2007)

253. Silvera-Batista, C.A., Wang, R.K., Weinberg, P., Ziegler, K.J.: Solvatochromic shifts of single-walled carbon nanotubes in nonpolar microenvironments. Phys. Chem. Chem. Phys. 12, 6990–6998 (2010)

254. Chou, S.G., Ribeiro, H.B., Barros, E.B., Santos, A.P., Nezich, D., Samsonidze, Ge.G., Fantini, C., Pimenta, M.A., Jorio, A., Filho, F.P., Dresselhaus, M.S., Dresselhaus, G., Saito, R., Zheng, M., Onoa, G.B., Semke, E.D., Swan, A.K., Unlu, M.S., Goldberg, B.B.: Optical characterization of DNA-wrapped carbon nanotube hybrids. Chem. Phys. Lett. 397, 296–301 (2004)

255. Jin, H., Jeng, E.S., Heller, D.A., Jena, P.V., Kirmse, R., Langowski, J., Strano, M.S.: Divalent ion and thermally induced DNA conformational polymorphism on single-walled carbon nanotubes. Macromolecules 40, 6731–6739 (2007)

256. Walsh, A.G., Vamivakas, A.N., Yin, Y., Cronin, S.B., Unlu, M.S., Goldberg, B.B., Swan, A.K.: Screening of excitons in single, suspended carbon nanotubes. Nano Lett. 7, 1485–1489 (2007)

257. Walsh, A.G., Vamivakas, A.N., Yin, Y., Cronin, S.B., Unlu, M.S., Goldberg, B.B., Swan, A.K.: Scaling of exciton binding energy with external dielectric function in carbon nano-tubes. Physica E 40, 2375–2379 (2007)

258. Smyrnov, O.A.: Excitons in single-walled carbon nanotubes: environmental effect. Ukr. J. Phys. 55, 1217–1224 (2010)

259. Adamyan, V.M., Smyrnov, O.A., Tishchenko, S.V.: Effects of environmental and exciton screening in single-walled carbon nanotubes. J. Phys: Conf. Ser. 129, 012012 (2008)

260. Araujo, P.T., Jorio, A., Dresselhaus, M.S., Sato, K., Saito, R.: Diameter dependence of the dielectric constant for the excitonic transition energy of single-wall carbon nanotubes. Phys. Rev. Lett. 103, 146802-1-3 (2009)

261. Nugraha, A.R.T., Saito, R., Sato, K., Araujo, P.T., Jorio, A., Dresselhaus, M.S.: Dielectric constant model for environmental effects on the exciton energies of single wall carbon nanotubes. Appl. Phys. Lett. 97, 091905-1-3 (2010)

References


262. Ando, T.: Environment effects on excitons in semiconducting carbon nanotubes. J. Phys. Soc. Jpn. 79, 024706-1-10 (2010)

263. Kiowski, O., Jester, S.-S., Lebedkin, S., Jin, Z., Li, Y., Kappes, M.M.: Photoluminescence spectral imaging of ultralong single-walled carbon nanotubes: micromanipulation-induced strain, rupture, and determination of handedness. Phys. Rev. B 80, 075426-1-7 (2009)

264. Steiner, M., Freitag, M., Tsang, J.C., Perebeinos, V., Bol, A.A., Failla, A.V., Avouris, Ph: How does the substrate affect the Raman and excited state spectra of a carbon nanotube? Appl. Phys. A 96, 271–282 (2009)

265. Qian, H., Araujo, P.T., Georgi, C., Gokus, T., Hartmann, N., Green, A.A., Jorio, A., Hersam, M.C., Novotny, L., Hartschuh, A.: Visualizing the local optical response of semi-conducting carbon nanotubes to DNA-wrapping. Nano Lett. 8, 2706–2711 (2008)

266. Georgi, C., Green, A.A., Hersam, M.C., Hartschuh, A.: Probing exciton localization in single-walled carbon nanotubes using high-resolution near-field microscopy. ACS Nano 4, 5914–5920 (2010)

267. Hasan, T., Scardaci, V., Tan, P.H., Bonaccorso, F., Rozhin, A.G., Sun, Z., Ferrari, A.C.: Nanotube and graphene polymer composites for photonics and optoelectronics. In: Hayden, O., Nielsch, K. (eds.) Molecular- and Nano-Tubes, pp. 279–354. Springer, New York (2011)

268. Hasan, T., Sun, Z., Wang, F., Bonaccorso, F., Tan, P.H., Rozhin, A.G., Ferrari, A.C.: Nanotube–polymer composites for ultrafast photonics. Adv. Mater. 21, 3874–3899 (2009)

269. Tan, P.H., Rozhin, A.G., Hasan, T., Hu, P., Scardaci, V., Milne, W.I., Ferrari, A.C.: Photoluminescence spectroscopy of carbon nanotube bundles: evidence for exciton energy transfer. Phys. Rev. Lett. 99, 137402-1-4 (2007)

270. Tan, P.H., Hasan, T., Bonaccorso, F., Scardaci, V., Rozhin, A.G., Milne, W.I., Ferrari, A.C.: Optical properties of nanotube bundles by photoluminescence excitation and absorption spectroscopy. Physica E 40, 2352–2359 (2008)

271. Hasan, T., Tan, P.H., Bonaccorso, F., Rozhin, A., Scardaci, V., Milne, W., Ferrari, A.C.: Polymer-assisted isolation of single wall carbon nanotubes in organic solvents for optical-quality nanotube–polymer composites. J. Phys. Chem. C 112, 20227–20232 (2008)

272. Wang, F., Sfeir, M.Y., Huang, L., Huang, X.M.H., Wu, Y., Kim, J., Hone, J., O’Brien, S., Brus, L.E., Heinz, T.F.: Interactions between individual carbon nanotubes studied by Rayleigh scattering spectroscopy. Phys. Rev. Lett. 96, 167401-1-4 (2006)

273. Qian, H., Georgi, C., Anderson, N., Green, A.A., Hersam, M.C., Novotny, L., Hartschuh, A.: Exciton energy transfer in pairs of single-walled carbon nanotubes. Nano Lett. 8, 1363–1367 (2008)

274. Koyama, T., Miyata, Y., Asada, Y., Shinohara, H., Kataura, H., Nakamura, A.: Bright lumi-nescence and exciton energy transfer in polymer-wrapped single-walled carbon nanotube bundles. J. Phys. Chem. Lett. 1, 3243–3248 (2010)

275. Lefebvre, J., Finnie, P.: Photoluminescence and Forster resonance energy transfer in ele-mental bundles of single- walled carbon nanotubes. J. Phys. Chem. C 113, 7536–7540 (2009)

276. Kim, Y., Minami, N., Kazaoui, S.: Highly polarized absorption and photoluminescence of stretch-aligned single-wall carbon nanotubes dispersed in gelatin films. Appl. Phys. Lett. 86, 73103–73106 (2005)

277. Hirori, H., Matsuda, K., Miyauchi, Y., Maruyama, S., Kanemitsu, Y.: Exciton localization of single-walled carbon nanotubes revealed by femtosecond excitation correlation spectros-copy. Phys. Rev. Lett. 97, 257401-1-4 (2006)

278. Berger, S., Iglesias, F., Bonnet, P., Voisin, C., Cassabois, G., Lauret, J-S., Delalande, C., Roussignol, P.: Optical properties of carbon nanotubes in a composite material: the role of dielectric screening and thermal expansion. J. Appl. Phys. 105, 094323 (094323-5) (2009)

279. Cognet, L., Tsyboulski, D.A., Rocha, J.-D.R., Doyle, C.D., Tour, J.M., Weisman, R.B.: Stepwise quenching of exciton fluorescence in carbon nanotubes by single-molecule reactions. Science 316, 1465–1468 (2007)

163

280. Barone, P.W., Yoon, H., Ortiz-García, R., Zhang, J., Ahn, J.-H., Kim, J.-H., Strano, M.S.: Modulation of single-walled carbon nanotube photoluminescence by hydrogel swelling. ACS Nano 3, 3869–3877 (2009)

281. Tsyboulski, D.A., Bachilo, S.M., Weisman, R.B.: Versatile visualization of individual single-walled carbon nanotubes with near-infrared fluorescence microscopy. Nano Lett. 5, 975–979 (2005)

282. Gaufres, E., Izard, N., Vivien, L., Kazaoui, S., Marris-Morini, D., Cassan, E.: Enhancement of semiconducting single-wall carbon nanotubes photoluminescence. Opt. Lett. 34, 3845–3847 (2009)

283. Zamora-Ledezma, C., Anez, L., Primera, J., Silva, P., Etienne-Calas, S., Anglaret, E.: Photoluminescent single wall carbon nanotubesilica composite gels. Carbon 46, 1253–1255 (2008)

284. Chernov, A.I., Obraztsova, E.D.: Photoluminescence of single-wall carbon nanotube films. Phys. Status Solidi B 247, 2805–2809 (2010)

285. Levshov, D.I., Yuzyuk, YuI, Michel, T., Voisin, C., Alvarez, L., Berger, S., Roussignol, P., Sauvajol, J.-L.: Raman probing of uniaxial strain in individual single-wall carbon nano-tubes in a composite material. J. Phys. Chem. C 114, 16210–16214 (2010)

286. Stranks, S.D., Sprafke, J.K., Anderson, H.L., Nicholas, R.J.: Electronic and mechanical modification of single-walled carbon nanotubes by binding to porphyrin oligomers. ACS Nano 5, 2307–2315 (2011)

287. Leeuw, T.K., Tsyboulski, D.A., Nikolaev, P.N., Bachilo, S.M., Arepalli, S., Weisman, R.B.: Strain measurements on individual single-walled carbon nanotubes in a polymer host: structure-dependent spectral shifts and load transfer. Nano Lett. 8, 826–831 (2008)

288. Glamazda, A. Yu., Leontiev, V.S., Linnik, A.S., Karachevtsev, V.A.: Luminescence study on hybrids of carbon nanotubes with DNA in water suspension and film at 5–290 K. Low. Temp. Phys. 34, 1313–1318 (2008)

289. Nepal, D., Geckeler, K.E.: pH-sensitive dispersion and debundling of single- walled carbon nanotubes: lysozyme as a tool. Small 2, 406–412 (2006)

290. Wang, D., Chen, L.: Temperature and pH-responsive single-walled carbon nanotube disper-sions. Nano Lett. 7, 1480–1484 (2007)

291. Ikeda, A., Totsuka, Y., Nobusawa, K., Kikuchi, J.: Reversible solubilisation and precipi-tation of carbon nanotubes by temperature and pH control in water. J. Mater. Chem. 19, 5785–5789 (2009)

292. Karachevtsev, V.A., Glamazda, A.Yu., Plokhotnichenko, A.M., Leontiev, V.S., Linnik, A.S.: Comparative study on protection properties of anionic surfactants (SDS, SDBS) and DNA covering of single-walled carbon nanotubes against pH influence: luminescence and absorption spectroscopy study. Materialwiss. Werkst. 42, 41–46 (2011)

293. Kelley, K., Pehrsson, P.E., Ericson, L.M., Zhao, W.: Optical pH response of DNA wrapped HiPco carbon nanotubes. J. Nanosci. Nanotechnol. 5, 1041–1044 (2005)

294. Xu, Y., Pehrsson, P.E., Chen, L., Zhang, R., Zhao, W.: Double-stranded DNA single-walled carbon nanotube hybrids for optical hydrogen peroxide and glucose sensing. J. Phys. Chem. C 111, 8638–8643 (2007)

295. Han, X., Li, Y., Wu, S., Deng, Z.: General strategy toward pH-controlled aggregation–dispersion of gold nanoparticles and single-walled carbon nanotubes. Small 4, 326–329 (2008)

296. Yurekli, K., Mitchell, C.A., Krishnamoorti, R.: Small-angle neutron scattering from sur-factant-assisted aqueous dispersions of carbon nanotubes. J. Am. Chem. Soc. 126, 9902–9903 (2004)

297. Fagan, J.A., Landi, B.J., Mandelbaum, I., Simpson, J.R., Bajpai, V., Bauer, B.J., Migler, K., Hight Walker, A.R., Raffaelle, R., Hobbie, E.K.: Comparative measures of single-wall carbon nanotube dispersion. J. Phys. Chem. B 110, 23801–23805 (2006)

References

photophysics of carbon nanotubes interfaced with organic and inorganic materials

Documents