the university of jammu for the award of the degree of doctor of...
TRANSCRIPT
0
METAL-SEMICONDUCTOR INTERFACE
FOR SCHOTTKY BARRIER DEVICES AND OHMIC CONTACTS
Thesis submitted to
The University of Jammu for the award of the degree of
Doctor of Philosophy (PhD) in
Physics
by
Narinder Kumar
Supervised by
Prof. Naresh Padha
Department of Physics & Electronics University of Jammu
Jammu – 180006
( March - 2011 )
1
I
DECLARATION
I, hereby, declare that the thesis entitled “METAL – SEMICONDUCTOR
INTERFACE FOR SCHOTTKY BARRIER DEVICES AND OHMIC CONTACTS”, submitted for
the award of the degree of Doctor of Philosophy (Ph D) is a record of the
research work carried out by me in the Department of Physics and Electronics,
University of Jammu, under the guidance of Prof. Naresh Padha. I also
ascertain that no part of this thesis is presented elsewhere for the award of
any degree or diploma of any University or Institution.
Place: Jammu Narinder Kumar
Date: 31-03-2011 (Research Scholar)
II
POST GRADUATE DEPARTMENT OF PHYSICS AND ELECTRONICS,
UNIVERSITY OF JAMMU, JAMMU
CERTIFICATE It is certify that Mr. Narinder Kumar worked under my supervision and the work done by him is
worthy of consideration for the award of Ph.D degree in Physics. I further certify that:
(i) The dissertation embodies the work of the candidate itself.
(ii) The candidate has worked under me for the period required under statutes.
(iii) The candidate has put in required attendance in the Department of Physics and Electronics
during the period of research.
(iv) The conduct of the candidate remained satisfactory during the period of his research.
(Prof. Naresh Padha)
Supervisor
Department of Physics & Electronics
University of Jammu
Jammu
Countersigned by
(Prof. Rajnikant)
Head
Department of Physics & Electronics
University of Jammu.
Jammu
III
ACKNOWLEDGEMENT
“The more intensely we feel about an idea or a goal, the more assuredly the idea, buried deep in our subconscious, will direct us along the path to its fulfillment.”
First of all, I would like to thank GOD for being there all the times when I needed
His blessings. At the very outset, I extend my heartfelt indebtedness to my Ph. D.
supervisor Prof. Naresh Padha for sailing me through this work with keen interest. His
dedication, commitment and professional expertise, which helped me throughout my
research, will remain a leading light for me forever. He has been excellent mentor and very
supportive throughout the work.
I express my deep sense of gratitude to Prof. Rajnikant, Head of Department of
Physics and Electronics for his generosity and all possible help that he bestowed upon me,
whenever required. I owe my deepest regards to Dr. C. J. Panchal (M. S. University of
Baroda, Gujarat) for all possible help during the course of my research work.
I am grateful to Prof. Mushahid Husain (JMI, New Delhi), Prof. R. M. Mehra
(Sharda University, Greater Noida, New Delhi), Dr. Shiv Kumar (Scientist „F”, SSPL,
New Delhi) and Prof. Ajay Gupta (Centre director UGC-DAE Consortium, Indore) for
providing me facilities for the characterization of my samples.
I also extend my thanks to all my friends and my research colleague from the
department Mr. Vishal Sharma, Mr. Ramesh Sachdeva, Meena Gupta, Monika Sharma,
Munesh sagar, Mrs.Jyoti Dubey, Mrs. Jyoti Rokhri, Ms. Ritu Sihotra, Ms. Anjali Devi &
Mr. Rakesh Sharma. I am also thankful to Mr. Vishal Pathania, Navneet Sharma, Rakesh
Kumar, Ms. Shivanu Suri, Ms. Jyoti Dubey, Ms. Shaveta Kohli, Ms. Monika Sharma, &
Mr. Daya Ram and all my friends for their valuable help and moral support.
IV
I am also thankful to Dr. Rakesh, Dr. Dinesh, Dr Sanjay, Dr.C.K Bhatt, Dr.
Vipul,Dr. Sanjeev Sharma,Sh. Darshan Abrol,Sh. Bedi, Keyur, Gopal, Jaymin, Praveen
and Satya who were the sources of moral support for me.
I Cherished friendship and company of my colleagues, Inder, Devinder, ShivNadan,
Pawan, Pajesh, Suram, Rajneesh, Sanjay, Kulbushan (IAF), Kulbushan, Ashok,
Parveen, Jatinder, Summit and Deepti.
I do not find words to express heartfelt gratitude to my research colleague Usha,
Meenakshi,Rakesh, Jatt Sahab and Ranjeet singh.
With great pleasure, I take this opportunity to acknowledge all those who have
directly or indirectly helped me during my Ph.D work. My sincere thanks to all other
teaching and non teaching staff of department of Physics, for their cooperation and help,
especially “Rattan Chachu ji”.
This stage of my academic career could only be reached with the blessings, love and
support of my parents, brother (Dr. D. Kumar), sisters (Anupam and Meena), brother in
laws (Sh. Surinder Khajuria and Sh. Vinay Kumar) & my wife (Pooja Sharma). I owe my
thanks to Deepak Sharma, Naveen Sangra, Bhanu, Chetan, Ravinder, Nishant, Shakshi,
Garima, Sachit and Kamakshi. I dedicate this piece of work to them as a token of love and
affection.
I expect to be pardoned if I have missed the name of any body inadvertently who
helped me to accomplish this work.
Dated: (Narinder Kumar)
V
Contents Declaration I
Certificate II
Acknowledgement III
Contents V
Abstract X
List of Figures and Tables XVII
List of Symbols and Physical Constants XXIV
List of Publications XXVI
Chapter 1
Semiconductor Material Fundamentals and Schottky Contacts 1- 29
1.1 Introduction to Semiconductor Materials 1
1.2 Classification of Semiconductors 2
1.2.1 Elemental semiconductors 2
1.2.2 Compound semiconductors 4
1.3 IV-VI Group Compound Semiconductor 7
1.4 P-T x Phase Diagram of the System Sn-Se 9
1.5 Introduction to Thin Film Technology 11
1.5.1 Thin film growth process 13
1.6 Metal -Semiconductor Contacts 17
1.7 Current Transport Mechanism in Metal-Semiconductor Contacts 18
1.7.1 Thermionic Emission (TE) theory 19
1.7.2 The Diffusion Theory 22
1.7.3 Thermionic-Emission-Diffusion (TED) Theory 24
1.8 Recombination Processes 25
1.8.1 Recombination in the depletion region 25
1.8.2 Recombination in the neutral region 27
References 28
VI
Chapter 2
____________________________________________________________
Measuring Equipments and Experimental Techniques 30-55
____________________________________________________________
2.1 Introduction 30
2.2 Vacuum Technology 30
2.2.1 Vacuum coating unit 32
2.2.1.1Vacuum pumps 32
2.2.1.2Pressure measurement gauges 33
2.2.2Selection of substrate 34
2.2.3Cleaning of substrates 35
2.2.4 Masks for generation of pattern in the deposited films 36
2.2.5 Substrate heating 37
2.2.6 Various Characterization Techniques 37
2.2.6.1 Structural and Morphological characterization 37
2.2.6.1.1 X-ray diffraction (XRD) 37
2.2.6.1.2Scanning Electron Microscope (SEM) 39
2.2.6.1.3Atomic force microscope (AFM) 40
2.2.6.1.4Energy-dispersive X-ray spectroscopy 42
2.2.6.2 Electrical characterization 44
2.2.6.2.1Hot-probe experiment 44
2.2.6.2.2 Hall experiment 45
2.2.6.2.3 Two-probe method 47
2.2.6.2.4 Four probe method 49
2.2.6.3 Optical Characterization 51
2.2.6.4 Temperature and Frequency dependent I-V
and C-V Measurements 52
References 54
VII
Chapter 3
__________________________________________________________________________
Impact of Substrate Temperature on Crystallite size and
Properties of SnSe thin films 56-89 __________________________________________________________________________
3.1 Introduction` 56
3.2 Preparation of Tin Selenide (SnSe) Thin films 56
3.3 Results and Discussion 57
3.3.1 X-ray diffractogram of SnSe thin films 57
3.3.1.1 Grain size measurement 58
3.3.1.2 Strain and Dislocation density 60
3.3.1.3 Dislocation density of SnSe thin films 63
3.3.2 Compositional analysis 63
3.3.3 Optical Properties 65
3.3.4 Morphological study of SnSe thin films by Scanning
Electron Microscope (SEM) 69
3.3.5 Electrical Studies 72
3.3.5.1 Resistivity Measurement 72
3.3.5.2 Activation energy 73
3. 4.1 Structural properties 78
3.4.1.1Surface Morphological Studies 81
3.4.1.2 Optical properties of SnSe thin films 84
3.4.2 Electrical Characterization 87
References 91
Chapter 4 _________________________________________________________________________
Impact of the Film on the Grain Size and Properties
of SnSe Thin Films. 93-116
_______________________________________________________________
4.1 Introduction 93
4.2 Preparation Details of Tin Selenide Thin Film 94
VIII
4.3 Results and Discussion 94
4.3.1 X-ray diffractogram of SnSe thin films 94
4.3.2 Atomic Force Microscopy (AFM) Studies 99
4.3.3 Optical Studies 101
4.3.4 Electrical Studies 110
4.4 Study of the effect of varying film thickness of Polycrystalline
SnSe Thin Films deposited at 575K Substrate Temperature. 113
4.4.1 Structural characterization 113
4.4.2 Electrical Properties of SnSe thn Film 114
References 116
Chapter 5
__________________________________________________________________________
Metal-Interface of SnSe Polycrystalline Thin Films for
Schottky Barriers and Ohmic Contacts
118-140 __________________________________________________________________________
5.1 Introduction 118
5.2 Experimental Details 119
5.2.1 Ohmic contact formation 119
5.2.2 Schottky Diode Fabrication 121
5.3 Results and discussion 123
5.3.1 Current-Voltage (I-V) Characteristics 123
5.4 Impact of Geometrical Shapes on the I-V
Characteristics of Schottky diodes 126
5.5 Influence of Different Areas 128
5.6 Capacitance -Voltage Characteristics 136
References 139
IX
Chapter 6
__________________________________________________________________________
Temperature dependent Current Voltage (I-V)
Characteristics of Ag/p-SnSe Schottky diodes 141-166 __________________________________________________________________________
6.1 Introduction 141
6.2 Results and discussion 142
6.2.1 Forward-bias I-V characteristics 142
6.2.2 Richardson Plots 147
6.2.3 The analysis of barrier height inhomogenities 149
6.2.4 Temperature dependent current-voltage characteristics
of Schottky diode(300K to 220K) by Cheung’ method. 154
6.2.5 Reverse bias Current-voltage characteristics 158
References 164
Chapter 7 ________________________________________________________________________
Conclusion of the work 167-170
________________________________________________________________________
X
Abstract
Tin Selenide compound semiconductor (a group IV-VI material) attracts considerable
scientific attention due to its potential applications in the field of photovoltaic solar cells, sensors,
semiconductor Lasers, polarizer and as thermoelectric cooling materials. In general, group IV-VI
semiconductors can be classified into three distinct class of materials: rhombohedral (GeTe); cubic
(PbS, PbSe, etc) and orthorhombic (SnS, SnSe, etc) compounds. The temperature coefficient of the
energy bandgap is positive for these compounds except for SnSe which shows a negative
temperature coefficient. Further, in lead salt materials, energy bandgap does not vary
monotonically in PbSxSe1-x with increasing value of x. Moreover, the static dielectric constants are
unusually very different in these compound semiconductors (i.e εs = 218 εo for PbSe, εs = 400 εofor
PbTe, εs = 17εo for SnSe). SnSe has numerous applications in memory switching devices, in
holographic recording systems or as an anode material to improve Lithium diffusivity.
Tin Selenide semiconducting compound crystallizes in orthorhombic crystallographic
structure (space group D2h16) whose atomic arrangement within the crystal resembles a severely
distorted NaCl structure. The stacking of Sn and Se atoms along the crystallographic c-axis leads to
a highly pronounced layer type structure joined with weak vander Waals bonding between layers.
Thin film technology has been the fastest growing market because of its being less
expensive and better employed for various device applications. Therefore, it is becoming the thrust
area of research being explored by the researcher’s world wide. Smaller and faster is the
technological imperative of our times and so there is a need for suitable materials and processing
techniques. Thin film plays an important role in fulfilling this need. Beside this, these have been
used for device purposes over the past 45 years and have replaced corresponding bulk materials
and their materials at cheaper costs. In light of above mention facts, an attempt has been made to
study SnSe in its thin film forms.
XI
In better understanding of the electrical properties of any semiconductor material, it is of
great importance that the materials be known for its contact behaviour with several metals.
Further, the metal-semiconductor contacts are becoming an essential part of all discrete electronic
devices and integrated circuits. Therefore, one must know how to fabricate reliable and efficient
metal-semiconductor contacts which have high yield and stability. The investigations of the metal-
semiconductor interface structures are important research tool for the current transport behaviour
of new semiconductor materials. The knowledge gained from the study of these devices can be
used for the development of future electronic devices. At the same time, these structures plays a
crucial role in the fabrication of useful devices like Schottky Diodes, MOSFET’s, Solar Cells and
MESFET’s etc.
In this thesis, an attempt has been made to fabricate Ag/p-SnSe Schottky Diodes and
investigate their current transport behaviour. Attempts have also been made to optimize the
thermal deposited SnSe thin films on the basis of their structural, morphological, optical and
electrical properties. Ag/p-SnSe Schottky Diodes have been fabricated on Al coated glass plates
with different Schottky interface areas. The diodes were undertaken for current-voltage (I-V)
measurements at room temperature as well as temperatures lower than and higher to room
temperature (220K to 350K). Studies have been focused to identify the possible factors which cause
deviations from the experimental data. The purpose has been to develop a complete picture of the
current transport mechanism across the metal-semiconductor interface of SnSe semiconductor
materials.
The work undertaken in the thesis has been divided into six chapters as per the details
given as under:
XII
Chapter 1- Semiconductor Material Fundamentals and Schottky Contacts:
This chapter presents a brief introduction of the compound semiconductors and their
types. It also presents the usefulness for extending thin film technology to the semiconductor
device applications. A brief literature survey of MS contacts formation and their current transport
mechanism has been presented. Comprehensive discussions on the importance of Tin Selnide
(SnSe) films for the fabrications of different optoelectronics devices have also been outlined. A brief
review of research activities on SnSe compound semiconductor as well as the techniques used for
their preparation has been presented in this chapter.
Chapter 2- Measuring Equipments and Experimental Techniques:
This chapter deals with the details of SnSe thin film preparations by thermal evaporation
method using vacuum coating unit. The techniques used for the structural, electrical and optical
properties have been discussed. The various experimental setups used for this purpose have also
been described viz. the transmittance and absorption spectra have been recorded using UV- Visible
Spectrophotometer, structural details were determined on the basis of diffraction data using X-ray
Diffractometer (Rigaku D-Max-III), morphology related information were obtained using Scanning
Electron Microscope (Model JEOL 5600), as well as Atomic Force Microscope while the
compositional analysis was established from Energy Dispersive X-ray Analysis. The experimental
arrangements used for the measurement of electrical resistivity of the films comprises Hot Probe
method as well as Two and Four Probe methods. The current-voltage (I-V) as well as capacitance-
voltage (C-V) measurements of the Ag/p-SnSe Schottky Diodes were measured using a computer
interfaced setup comprising a programmable Keithley source meter (model-2400), Closed Cycle
Liquid Helium Cryogenic setup (CTI-Cryotronics model 22C) equipped with temperature controller
(Lake Shore-model-321) and Precision programmable LCR meter (Aglient make 4284A). Interface of
XIII
I-V and C-V measurement equipments was achieved by using LabVIEW software by National
Instruments (U.S.A).
Chapter 3- Impact of substrate temperature on the grain size and properties of Tin Selenide
(SnSe) thin films:
The behaviour of SnSe thin films depends on the methods used for their preparation and
deposition conditions. The substrate temperature (Ts) plays a crucial role in controlling those
particular properties of SnSe films which are useful in electronic device applications. In this chapter,
the results of the structural, surface morphological, compositional, optical and electrical properties
of SnSe thin films obtained at different substrate temperatures have been presented in detail. The
XRD spectra and the SEM analysis suggest that the polycrystalline thin films of SnSe possess
uniform distribution of grains along the (111) diffraction planes. The intensity of the diffraction
peaks increased with the increase in substrate temperature and well-resolved peaks were seen at
523 K. The compositional analysis was done by EDAX which confirmed that the atomic percentage
of Sn and Se achieved the stoichiometric ratio within acceptable limits. The stoichiometric ratio
between atomic masses of Sn and Se is nearly 1:1 for the films deposited at 523K. The analysis of
the data obtained from the optical transmission spectra suggests that the films possess energy
bandgap in the range of 1.38-1.18 eV while Hall-effect measurements revealed that the resistivity
of films falls in the range 112-20 Ω-cm. Investigations have also been made to study the effect of
substrate temperature (Ts) of 550K, 575K and 600K on 200nm thick SnSe films on the basis of above
referred characterization methods. This attempt has been initiated to achieve better crystallinity as
well as improved material parameters in order of achieving optimized current transport behavior of
the deposited SnSe thin films.
XIV
Chapter 4- Impact of the film thickness on the grain size and other properties of SnSe thin films:
A set of SnSe films with varying thickness were prepared on the glass substrate at the
room temperature. Atomic Force Microscopy (AFM) and X-ray diffraction (XRD) analyses were used
to investigate the effects of thickness variation on the surface morphology and crystallinity. From
the XRD data, the various parameters like grain size, FWHM, strain and dislocation density were
calculated and it was observed that the crystallite sizes and strain increases with increase in
thickness. AFM images depicted the compact surface morphology at higher thickness and surface
roughness behaviour revealed a linear increase with thickness. It was observed that the crystalline
quality, electrical and optical behaviour of the films changed with the film thickness. The
deductions are made to obtain the optical parameters such as the optical bandgap energy,
absorption coefficient, extinction coefficient and Urbach’s energy. The analysis of the optical
transmission spectra suggests that the bandgap of films deceases (from 1.71 to 1.13 eV) with
increase in film thickness (from 150 to 500nm). A close examination of the investigations reveals
that the absorption coefficient (α) decreases with increase in thickness in the high energy region of
the plot. The predictions of the optical studies were equally supported by the Urbach’s energy
analyses results based on the x-ray diffraction data.
Attempts have also been made to study thickness variation effects of the films at higher
substrate temperature of 575K. Consequently, another set of films with thicknesses 150nm, 200nm,
250nm and 500nm have been obtained at substrate temperature of 575K. These films were then
analyzed for the structural, optical and electrical characterizations. The results, thus obtained were
found to be better than those of the studies made at room temperature.
XV
Chapter 5-Metal-interface of SnSe thin films for the formation of Schottky barriers and ohmic
contacts:
In this chapter, an attempt has been made to identify the metals which make
reasonably acceptable ohmic contacts to SnSe films. The various metals tried for this purpose are
Al, In and Ag. Out of these, Al contacts were found to be better ohmic with an ease for working at
comparatively higher temperatures. Further, reasonably good SnSe Schottky diodes were
fabricated with Ag metal and the corresponding Schottky structure of Ag/p-SnSe was formed on the
Al coated glass substrates. Circular Schottky diodes of the areas of 6x10-3cm2, 9x10-3cm2 and 9x10-
2cm2 were fabricated which was tested for the I-V behaviour at room temperature. Out of these
only good quality Schottky Diodes with better breakdown voltage were undertaken for further
analysis. These undertaken diodes were tested for the room temperature Current Voltage (I-V) and
Capacitance Voltage (C-V) measurements. The data, thus, obtained were used for evaluating the
diode parameters such as ideality factor (η), barrier height (фb), series resistance (Rs) & breakdown
voltages (VBR). The data was analyzed to study the effect of change in metal semiconductor
interface area on the current transport phenomena of the Schottky diodes.
Chapter 6-Temperature dependent current voltage (I-V) characteristics of Ag/p-SnSe Schottky
diodes:
This chapter deals with the current transport in Ag/p-SnSe Schottky diodes on the basis of
temperature dependent current voltage (I-V) characteristics. Separate investigations were
undertaken to study the current transport behaviour at temperature higher and lowered than the
room temperature. From the analysis, it was observed that zero bias barrier height (Фbo) decreases
while ideality factor (η) increases with decrease in temperature. Efforts have been made to
establish the effect of SnSe semiconductor surface on the current transport behaviour of the
undertaken Schottky diodes. An attempt has also been made to analyze the factors which
XVI
contribute in conduction mechanism and also to understand deviations of the experimental data
from theoretical prediction. The non ideal I-V behaviour of the Schottky diodes has been attributed
to the change in barrier height due to interface states defects, dislocations, surface states, series
resistance etc.
Chapter 7–Conclusion of the work :
A brief summary of the results drawn from all undertaken investigations were
presented. The conclusions of the study of investigations and future scope were highlighted.
XVII
List of Figures
Fig. 1.1 Tetrahedral atomic configuration 3
Fig. 1.2 Structure of Diamond Lattice. 4
Fig. 1.3 Schematic representation of Zincblend structure 6
Fig. 1.4 Schematic arrangement of atoms in Tin Selenide (SnSe). 8
Fig. 1.5 Bulk form of Tin Selenide (SnSe). 9
Fig. 1.6 Phase diagram of the system SnSe. 10
Fig. 1.7 Simple model for thin film deposition. 13
Fig. 1.8 (a, b, c) Layer by layer growth of thin films 16
Fig. 1.9 Structure of a Schottky Diode 17
Fig 1.10 Transport processes in a forward-biased Schottky barrier 19
Fig. 2.1 a) Block diagram. (b) Vacuum coating unit for
thermal evaporation
34
Fig. 2.2 Schematic diagrams of different types of films deposited
on glass substrate
37
Fig. 2.3 X-ray diffractometer (Rigaku D-Max-III) 39
Fig. 2.4 Modern Version of Scanning Electron Microscope 40
Fig.2.5 Atomic Force Microscopy 42
Fig. 2.6 EDX Experiment set up (JEOL) 43
Fig. 2.7 Experimental set-up of the "hot-probe" experiment 45
Fig. 2.8 Configuration for measuring Hall Effect 45
Fig. 2.9 A photograph of a probe with a Ge sample, with contact numbers
labeled. The sample is on the front surface of the plastic substrate
in this photo.
47
Fig. 2.10 Low temperature resistivity measurements by Two-probe method 48
Fig. 2.11 Four Probe Set-up (a) Schematic view (b) Experiment set-up 50
Fig. 2.12 Geometrical patterns of thin films used for four probe set up 50
Fig. 2.13 Transmission spectrum of glass substrate used for optical studies 51
Fig. 2.14 Optical Set Up for the transmission measurement 51
Fig. 2.15 Automated Temperature and Frequency Dependent I-V
Measurement Setup
52
Fig. 3.1 XRD spectra of SnSe thin film of the thickness100nm deposited
at substrate temperature323K,373K,423K,473K and 523K.
59
XVIII
Fig. 3.2 Variation of the grain size(nm) and FWHM(degrees) of SnSe
thin films with substrate temperature(K)
60
Fig. 3.3 Plot of βcosθ vs sinθ for the SnSe thin film deposited at
the substrate temperature (a)323K (b)523K
61
Fig. 3.4 Variation of Dislocation Density of 100nm SnSe thin films
deposited at different substrate temperatures
63
Fig. 3.5 The EDAX spectrum giving the compositional
information of SnSe thin film deposited at substrate temperature of
(a) 323K (b) 523K.
64
Fig.3.6 (a) Plots of (αhν) 2 versus (hν) for SnSe thin films deposited
at Ts of 323k
66
Fig.3.6 (b) Plots of (αhν) 2 versus.hν for SnSe thin films deposited
at Ts of 373k
66
Fig.3.6 (c) Plots of (αhν) 2 versus hν for SnSe thin films deposited
at Ts of 423K
67
Fig.3.6 (d) Plots of (αhν) 2 versus hν for SnSe thin films deposited
at Ts of 473K
67
Fig.3.6 (e) Plots of (αhν)2 versus hν for SnSe thin films deposited
at Ts of 523K
68
Fig. 3.7 SEM images of SnSe thin films of 100nm deposited on
glass substrate at substrate temperature (a) 323K (b) 373K
(c) 423K (d) 473K and(e)523K
71
Fig. 3.8 The variation of the electrical resistivity of SnSe thin films
measured at different substrate temperatures.
73
Fig. 3.9 Plots of the resistance (R) as a function of temperature (T)
for SnSe thin films deposited at (a) Ts = 323K (b) Ts = 373K
(c) Ts = 423K (d) Ts = 473K and (e) Ts = 523K
74
Fig. 3.10 Plot of ln(R/R300K) versus 1/T for a SnSe thin film grown at
(a)Ts =323K (b Ts =373K(c) Ts =423K (d Ts =473K and (e) Ts = 523K
76
Fig. 3.11 The variation of the electrical Activation Energy of SnSe
thin films deposited at different substrate temperatures.
77
Fig. 3.12 XRD patterns of SnSe thin films deposited at various
substrate temperatures: (a) 550K, (b) 575K and (c) 600K
79
Fig. 3.13 AFM images(2D and 3D) of Tin Selenide thin films deposited
at different substrate temperature (a)550K, (b) 575K and (c)600K
83
Fig. 3.14 Plot of (αhν)2 versus hν for SnSe thin films thermally deposited
at 575K substrate temperature.
84
XIX
Fig. 3.15 Plot of ln(αhν) versus ln(hν-Eg) of SnSe film deposited at
575K substrate temperature
85
Fig. 3.16 Plot of h versus ln(α) of SnSe thin films at 575K Substrate
temperature
86
Fig. 3.17 Variation of extinction coefficient with incident photon energy 86
Fig. 3.18 Plots of the resisivity as a function of temperature (T) forSnSe
thin films deposited at (a)Ts =550K (b)Ts =575 & (c)Ts =600K
87
Fig. 3.19 Plot of ln (σ) versus 1000/T for a SnSe thin film deposited at
different substrate temperatures(a)550K (b)575K and (c)600K.
89
Fig. 4.1 XRD Spectra of SnSe thin films of different thicknesses
(a) 150nm (b) 200nm (c) 250nm and (d) 500nm deposited at room temperature
95
Fig. 4.2 (a) & (b) A plot of β cosθ versus sinθ for the SnSe thin films
deposited at different thicknes (a) = 500nm and ( b) = 150nm.
97
Fig. 4.3 Plot shows the effect of thickness on the grain size and strain 99
Fig. 4.4 (a) to (c) AFM image of SnSe thin film of thickness150 nm,300nm
and 500nm deposited at glass substrate at room temperature
101
Fig. 4.5 Plots of the transmission coefficient versus wavelength of
incident photons of SnSe films having thicknesses of
(a) 150 nm (b)200 nm (c)250 nm and (d)500 nm.
102
Fig. 4.6 Absorption coefficient versus photon energy plots for the
SnSe film thicknesses (a) 150nm (b) 200nm((c) 250nm and (d) 500nm
103
Fig. 4.7 Plots of h versus (h) 2 for films deposited with the
thicknesses(a)150 nm(b)250 nm (c)350 nm and (d)500 nm.
105
Fig. 4.8 Plot of Energy Bandgap versus Thickness for the SnSe thin films 106
Fig. 4.9 Variation of ln(αhν) versus ln(hν-Eg) for different thicknesses
(a) 150nm (c)250nm and (c)500nm SnSe thin films.
108
Fig. 4.10 Variation of extinction coefficient with incident photon energy 109
Fig. 4.11 Variation of urbach’s energy with photon energy for
different thicknesses.
110
Fig. 4.12 Temperature depended Electrical Resistivity measurement of
SnSe thin film using four probe method in the temperature
range 325K to 450K
111
Fig. 4.13 Plot of ln() versus 1000/T for a SnSe thin film of thickness of
500nm grown at room temperature
112
XX
Fig. 4.14 Plot of ln(σ) versus 1000/T for a SnSe thin film of different
thickness deposited at 575K substrate temperature
115
Fig. 5.1 (a)Pictorial representation of Indium (In) ohmic contacts
to SnSe semiconductor layer and (b) the current-voltage
characteristic of In/p-SnSe/In structure.
120
Fig. 5.2 (a) Pictorial representation of (Al) ohmic contacts to SnSe
semiconductor layer and (b) the current-voltage characteristic
of Al/p-SnSe/Al structure.
121
Fig. 5.3 (a) Structure of the Ag/p-SnSe schottky diode fabricated on
Aluminum coated glass substrates (b) animated elevated view
and (c) top images of the different areas of Ag/p-SnSe Schottky diodes.
122
Fig. 5.4 The Current-Voltage characteristics of Ag/p-SnSe structure. 123
Fig. 5.5 Equivalent electrical circuit of Schottky Diode. 124
Fig. 5.6 The forward I-V characteristics of Ag/p-SnSe Schottky diode
having square and circular shapes each of anarea =1.6x10-1cm2
the Curves have been fitted with the TED equation 5.5 using
Is, η and Rs adjustable parameters.
127
Fig. 5.7 Forward current-voltage characteristics of Ag/p-SnSe Schottky
diode with different diode areas.
129
Fig.5.8 Plots of d(V)/d(lnJ) vs. J and H(J) vs. J of Ag/p-SnSe Schottky
diodes of different areas (a) 6x10-3cm2 , (b)
9x10-3cm2
and (c) 9x10-2cm2
133
Fig. 5.9 Reverse biased I-V Characteristics of Ag/p-SnSe schottky diodes
with different diode areas measured at Room Temperature.
135
Fig. 5.10 The reverse breakdown voltage VBR (V) versus area (cm2) for
Ag/p-SnSe Schottky diode.
135
Fig. 5.11 Plot of 1/C2 versus V of Ag/p-SnSe Schottky Diode at different
frequencies
137
Fig.6.1 Plot of temperature dependent current(A) versus voltage(V)
characteristics of Ag/p-SnSe Schottky diode measured from
(a) room temperature down to 220K (b) 303to 328K
144
Fig. 6.2 The ideality factor (η) versus temperature and the zero-bias barrier
height (фbo) versus temperature; the solid curves represent
simulated the data generated by using an analytical potential
fluctuation model.
145
Fig. 6.3 Richardson plot of ln(Jo/T2) versus 103/T for the Ag/p-SnSe
Schottky diode.
147
XXI
Fig 6.4 Zero-bias apparent barrier height versus ideality factor of
Ag/p-SnSe Schottky diode at various temperatures. 148
Fig. 6.5 (a) & (b) Zero-bias apparent barrier height and ideality factor
versus(2KT)-1 plot of Ag/p-SnSe Schottky diode as per to Gaussian
distribution of the barrier heights.
151
Fig. 6.6 The Modified Richardson ln(I0/T2)-(q2σ2s/2k2T2) versus 103/T
plot for the Ag/p- SnSe Schottky diode generated on the
basis of Gaussian distribution of the barrier heights
154
Fig. 6.7 Plot of dV/d[In(I)] versus Js for Ag/p-SnSe Schottky diode at
temperatures (a) 220K (b) 290K
156
Fig.6.8 Plot of temperature dependent reverse bias current
(A) versus voltage(V) characteristics of Ag/p-SnSe Schottky
diode in the temperature range (30 8to 338K)
158
Fig 6.9 Plot of Breakdown voltage versus temperature of Ag/p-SnSe
Schottly Diode
159
Fig. 6.10 Plot of ln(Js/T) versus 1/T of Ag/p-SnSe Schottky Diode. 160
Fig. 6.11 Variation of reverse current( log IR) with Veff1/4 at
temperature of (310,320 and330K)
162
XXII
List of Tables
Table 1.1
Section of the Periodic Table with elements from atomic
groups IIb to VIa.
6
Table 1.2
Energy band gaps of various semiconductors 8
Table 3.1 Structural parameters of SnSe thin films deposited at a
thickness of 100 nm on glass substrate at different substrate
temperatures with preferred orientation along (111) plane
60
Table 3.2 Micro-structural parameters of SnSe thin films deposited
on the glass substrateat different substrate temperature for
the most prominent (111) planes
62
Table 3.3 Structural parameter of SnSe film deposited on glass substrate at
temperature of (a)525K (b)575K and(c)600K.
80
Table 3.4 Average Surface Roughness. 83
Table 3.5
Resistivity values of SnSe thin films grown at different substrat
temperatures at (a) Ts =550K (b) Ts =575 and, (c) Ts =600K.
88
Table 4.1 Structural parameters of SnSe films deposited on glass substrate
at room temperature with preferred orientation along (111) planes.
96
Table 4.2 Micro-structural parameters of SnSe films deposited on glass
substrate with preferred orientation along (111) plane.
98
Table 4.3
Root Mean Square Roughness at different thicknesses of SnSe
thin film.
100
Table 4.4 Structural parameters of SnSe thin films of different thicknesses
deposited at 575K substrate temperatures.
113
Table 4.5
Micro-structural parameters of SnSe thin films deposited
on the glass substrate at different substrate temperatures.
114
Table 5.1 Ideality factor, Barrier heights and Series Resistance obtained from the
undertaken Schottky Diodes.
128
Table 5.2 Schottky Diode parameters for different areas. 130
Table 5.3 Comparison of diodes parameters extracted from the linear
fitting of ln(I) vs. V plots using thermionic emission equation
and those extracted from the Cheung’s method.
134
Table 6.1 Parameter of SnSe Schottky diode extracted from the
temperature dependent current voltage ( I-V) data.
146
XXIII
Table 6.2 Parameters of Ag/p-SnSe Schottky diodes extracted from
the Cheung’s Method.
157
Table 6.3
Schottky diode parameters for reverse bias at different
Temperature.
161
Table 6.4 Carrier concentrations(Na) obtained from reverse I-V
characteristics in the temperature range 310 to330K for
Ag/p-SnSe Schottky Diode.
163
XXIV
List of Symbols and Physical Constants
List of Symbols Symbol Description
A** Effective Richardson Constant(Am-2K-2)
Ec Bottom of Conduction Band (V)
Ef Fermi Energy Level (V)
Eg Energy Band gap (V)
Ev Top of Valence Band (V)
h Planck’s constant (J-sec)
I Current (A)
Is Saturation Current (A)
J Current Density (A/cm2)
α Temperature Coefficient of Zero bias BH
β Temperature Coefficient of flat-band BH
JS Current Density from Metal to semiconductor
kT Thermal Energy
m0 Electron rest mass (Kg)
ni Intrinsic carrier concentration
Nc Effective density of states in conduction band (cm-3)
ND Donor Impurity density (cm-3)
ni Intrinsic electron concentration (cm-3)
Nv Effective Valence band density (cm-3)
A Area of Contact (cm2)
T Absolute Temperature (K)
V Applied bias (V)
Vbi Built in potential (V)
VF Forward bias (V)
VBR Reverse bias (V)
ε0 Permittivity in Vacuum (F/cm)
εs Semiconductor Permittivity (F/cm)
εi Permittivity of interfacial layer (F/cm)
εmax Electric-field at the metal semiconductor interface
(V/cm)
Ideality factor
Rs Diode series resistance
фbo Zero biased barrier height (V)
ф m Work function of metal (V)
ф s Work function of semiconductor (V)
τ Electron effective lifetime (sec)
χ Electron affinity of semiconductor (V)
so Standard Deviation
XXV
List of Physical Constants
S Area of contact (6x10-3 cm2, 9x10-3 cm2, 9 x10-2 ,1.6x10-1 cm2)
q Elementary charge (1.6 x 10-19 C)
k Boltzmann constant ( 1.38 x 10-23 J/K)
A** Effective Richardson constant (18 Am-2K-2)
εo Absolute Permittivity free space (8.85x10-14 Fcm-1 )
ε s (17 εo) Relative Permittivity
1
Chapter 1 Semiconductor Material Fundamentals and Schottky Contacts __________________________________________________________________________
1.1 Introduction to Semiconductor Materials
There are numerous semiconductor materials and a wide variety of electronic and
optical properties of these materials provide great flexibility in the design of devices for
electronic and optoelectronic functions. The elemental semiconductor Germanium (Ge) was
widely used in the early days of semiconductor development for transistors and diodes.
Silicon dominated majority of rectifiers, transistors and integrated circuits for the last several
decades. However, the compound semiconductors are widely used in high speed-devices and
devices requiring the emission and absorption of light. The two-element (binary) III-V
compound semiconductors such as GaAs and GaP are common in light emitting diodes
(LEDs). Three element (ternary) compound semiconductors such as GaAsP and four
element (quaternary) compound semiconductors such as InGaAsP can be grown to provide
added flexibility in choosing material properties.
Fluorescent materials such as those used in television screens are usually II-VI
compound semiconductors such as ZnS. Light detectors are commonly made with InSb,
CdSe, or other compounds such as PbTe and HgCdTe. Silicon and Germanium are also
widely used as infrared and nuclear radiation detectors. An important microwave device, the
2
Gunn diode, is usually made of GaAs or InP. Semiconductor lasers are made by using GaAs,
AlGaAs, and other ternary or quaternary compounds.
One of the most important characteristics of a semiconductor, which distinguishes it
from metals and insulators, is its energy bandgap. This property determines the wavelength
of light that can be absorbed or emitted by the semiconductor. For example, the bandgap of
GaAs is about 1.43 eV, which corresponds to light wavelengths in near infrared region. In
contrast, GaP has a bandgap of about 2.3 eV, corresponding to wavelength in the green
portion of the spectrum. As a result of the wide variety of semiconductors bandgap, light
emitting diodes and lasers can be constructed with wavelengths over a broad range of the
infrared and visible portions of the spectrum.
The electronic and optical properties of semiconductor materials are strongly
affected by impurities, which may be added in precisely controlled amounts (doping). Such
impurities are used to vary the conductivities of semiconductors over wide ranges and even
to alter the nature of current transport process from conduction by negative charge carriers
to positive charge carriers e.g. an impurity concentration of one part per million can cause a
sample of Si from a poor conductor to a good conductor of electricity.
1.2 Classification of Semiconductors
1.2.1 Elemental semiconductors
The group IV semiconductors are called elemental semiconductors, such as
Germanium and Silicon, because they are single species of atoms. All the group IV elements
have two electrons in their outermost subshell and crystallize in a diamond shape structure
in which the neighbouring atoms are bounded by homopolar cohesive forces. These forces
3
determine the two important properties of these materials, that is, the melting point and the
energy bandgap. With an increase in atomic number, the cohesive forces decreases and the
size of the atom increases. As a result, the melting point and energy bandgap decreases.
Thus, there is a gradual transition from strongly insulating to essentially metallic behaviour
[1]. Silicon is the prime example of elemental semiconductor. It is located in column IV of
the periodic table. In a silicon crystal, an atom forms four covalent bonds with four other
atoms (four nearest neighbours) and shares two valence electrons with all four nearest
neighbours. In other words, in a silicon crystal, each atom is tetrahedrally coordinated in
order to share eight electrons (two electron per bond), corresponding to complete p and s
valence subshells as shown in Fig.1.1. The valence configuration of silicon has two s
electrons and two p electrons. In Si crystal, tetrahedron band configuration is repeated,
forming a same crystal structure like diamond as shown in Fig.1.2.
Fig. 1.1 Tetrahedral atomic configuration of
(a)Silicon and (b) Gallium Arsenide
4
Fig. 1.2 Structure of Diamond lattice.
1.2.2 Compound Semiconductors
Compound semiconductor materials can be realized by the formation of "solid
solutions" of two or more starting materials. The “solid solutions” occur when atoms of
different elements are able to substitute a given constituent of a material without altering its
crystal structure. The ability to do so by the new atom is referred to as its miscibility. In
order that atoms can form solid solutions over large ranges of miscibility, they must satisfy
the Hume Rothery rules:
They must belong to the same group of the periodic table.
They must have comparable atomic diameters allowing substitution without
large mechanical distortion.
Their iconicity must not be very different so as not to affect the tendency to
attract repel electrons from the site by a large amount.
The crystal structure of each constituent must be the same.
5
The semiconductor compounds can be formed by combining different elements as
shown in the section of the periodic Table (refer Table 1.1) e.g. if we combine Ga and As
into GaAs compound, then each atom on an average will have two s electrons and two p
electrons, like in Si. In a similar way, we can form many semiconductor compounds,
combining other elements of column III of the periodic table (two s electrons and one p
electron) with elements from column V (two s electrons, three p electron) .
Most of the III-V semiconductors have the Zincblende crystal structure as shown in
Fig.1.3. Eight valence electrons are shared between a pair of nearest atoms, and on an
average, each atom has four valence electrons. This suggests that the bonding has a covalent
character, and to a first approximation, the cohesion between the atoms is homopolar.
Therefore, we would expect the properties of these compounds to be similar to those of
corresponding group IV elements. However, since the elements of group III are more
electropositive, and those of group V are more electronegative than the group IV elements,
the bonding in III-V compounds has a „partial ionic character‟ as well. Therefore, the
cohesive force between atoms represents the cohesive force of the covalent bonding plus an
additional term because of ionic contribution. As a result, the cohesive force and the strength
with which the valence electrons are bound to the atoms are higher for these crystals than
those for the corresponding group IV semiconductors.
6
Fig. 1.3 Schematic representation of Zincblende structure.
Table 1.1 Section of the Periodic Table with elements from atomic groups
IIb to VIa.
7
1.3 IV-VI Group Compound Semiconductors
The IV-VI group semiconducting narrow energy bandgap compounds attract
considerable scientific attention due to their potential applications in the field of solar energy
conversion strategies, sensors, laser materials, thin film polarizers and thermoelectric
cooling materials. Tin Selenide (SnSe) is IV-VI semiconducting compound that crystallizes
in orthorhombic crystallographic structure (space group D2h 16) whose atomic arrangement
within the crystal resembles a severely distorted NaCl-type of structure. The existence of the
tightly bound double layers of tin and selenium atoms stacked along the crystallographic c-
axis suggests the bonding between the layers being of the weak Vander Waals type, which
leads to a highly pronounced layered type structure. Such a structural arrangement leads to a
pronounced anisotropy for the physical properties of this compound, which makes it
particularly appealing for the fabrication of solar cells, because of their higher chemical
stability without passivity than other semiconductors, such as Si, GaAs, InP, and CdSe,
which need special passivity procedures in order to avoid photocorrosion. The bandgap of
some important elemental and compound semiconductors along with corresponding
wavelengths are given in Table 1.2.
From Table 1.2, we can interpret that this compound semiconductor has lower
bandgap as compared to other semiconductors. The schematic arrangement of atoms of Tin
Selenide (SnSe) is shown in Fig. 1.4.
8
Material Symbol Bandgap (eV)
@300K
Longest
Wavelength (μm)
Silicon Si 1.11 1.10
Germanium Ge 0.67 1.85
Silicon Carbide SiC 2.86 0.43
Aluminum Nitride AlN 6.3 0.19
Diamond C 5.5 0.22
Gallium(III) Arsenide GaAs 1.43 0.95
Gallium(III) Nitride GaN 3.4 0.36
Indium(III) Arsenide InAs 0.36 3.44
Zinc Selenide ZnSe 2.7 0.46
Zinc Telluride ZnTe 2.25 0.55
Cadmium Sulfide CdS 2.42 0.51
Cadmium Selenide CdSe 1.73 0.72
Tin Selenide SnSe 1.26 0.98
Tin Diselnide SnSe2 1.15 1.08
Table 1.2 Energy band gaps of various semiconductors.
Fig. 1.4 Schematic arrangement of atoms in Tin Selenide (SnSe).
9
SnSe has numerous applications in memory switching devices, in holographic
recording systems or as an anode material to improve lithium diffusivity. Owing to this,
SnSe has been studied in the form of both single crystals and thin films. Researchers
investigated a number of methods to prepare SnSe thin films viz. atomic layer deposition,
chemical bath deposition, vacuum evaporation, chemical vapor deposition, spray pyrolysis,
electro deposition, etc. Thermal evaporation of pulverized SnSe is a simple technique,
whose attractive features are low temperature growth, producing large-area devices and film
thickness controlled by readily adjusting the electrical parameters. Bulk form of Tin
Selenide (SnSe) is shown in Fig. 1.5.
Fig. 1.5 Bulk form of Tin Selenide (SnSe).
1.4 P-T-x phase diagram of the System Sn-Se
The P-T-x phase diagram of the system Sn-Se is given in Fig.1.6 according to the
data from the papers [2-6]. In this system there exist two chemical compounds: SnSe and
SnSe2. The above discussed supposition on the existence in the system SnSe of the third
compound, Sn2S3, based on the DTA data, has not been confirmed by XRD, NMR and
10
microstructural investigations of this system [7, 8, 5]. The NMR spectrum of Sn2S3
represents a superposition of the spectra of SnSe and SnSe2.
Fig. 1.6 Phase diagram of the system SnSe.
SnSe melts congruently at 1153 ± 5 K [5]. The melting heat of SnSe is 32.63 ± 3.7
kJ/mol[4]. There are known two polymorphous modifications. The low temperature α
modification (type B16) at 807 K transforms into high temperature rhombic β-modification
of type TII (structural type B33) [9]. The phase transitions in SnSe are second order
transitions [10, 11]. The transition α→β of the compound SnSe is of λ - type and occurs in
an extended temperature interval, e.g. 200K above the temperature of the phase
transformation. At these temperatures take place the shift of the atoms in the α-form only
along the a-axis in the interval of coordinate 0 ≤ x ≤ 0.12 and 0.48 ≤ x ≤ 0.50 for selenium
11
atoms. The structure of β-modification is derived from the NaCl type structure: the
neighbouring octahedral layers with NaCl structure are shifted one to another by a/2. During
the transition from α-form to β-form the coordination of all the atoms changes from three to
two. Such phase transition is known as chemical reaction of the type Sn2 [9].
The homogeneity domain of SnSe is situated in the range of selenium excess and its
extension is 10-8
–10-4
% of Se [12]. From the results of Hall Effect measurements in
polycrystalline samples, annealed for various partial vapour pressure of Se in the
temperature range 823-963 K, it has been concluded that the defects responsible for the
deviation from stoichiometry for high temperatures are the doubly ionized tin vacancies
[VSn2+
]. The ionization energy of these defects changes as a function of nonstoichiometry of
the composition, in the range 0.012 - 0.20 eV. For low temperatures become essential the
processes of association of the neutral vacancies [(VSn)*]: for the temperatures 663-713K
occur vacancy pairs [(VSn)2*] and below 663K one forms aggregates from four vacancies
[(VSn)4*]. The association energies of these complexes are 1.9 and 1.15 eV respectively.
1.5 Introduction to Thin Film Technology
Thin film technology is the fastest growing technology in the market because it is
less costly to manufacturer and better employed for various applications. Thin films are the
basic building blocks for solid-state devices and their simple geometry makes them ideal
models for fundamental scientific studies. Thin film technology is simultaneously one of the
oldest arts and one of the newest sciences. Involvement with thin films dates to the metal
ages of antiquity. Consider the ancient craft of gold beating, which has been practiced
continuously for at least four millennia. A thin film may be defined as two-dimensional
material born of atom-by-atom or molecule-by-molecule or ion-by-ion condensation
12
process. A thin film is geometry of material having one of its dimensions about 1μm. It
should be emphasized here that it is not simply small thickness, which endows thin films
with special and distinct properties, but rather the microstructure resulting from the unique
way of their coming into progressive addition of the basic building blocks one by one, which
is more important. Typical applications of thin films include microelectronics, magnetic
sensors, gas sensors, tailored materials, optics-anti-reflection coating and corrosion
protection and wear resistance etc.
In thin films, deviation from the properties of the corresponding bulk materials arise
because of their small thickness, large surface to volume ratio and unique physical structure
which is a direct consequence of the growth process. Some of the phenomenon arising as a
natural consequence of small thickness are optical interference, electronic tunneling through
an insulating layer, high resistivity and low temperature coefficient of resistance, increase in
critical magnetic field and critical temperature of superconductor. The high surface to
volume ratio of thin films due to their small thickness and microstructure can influence a
number of phenomena such as gas adsorption, diffusion and catalytic activity.
Amongst the various types of thin films, semiconductors have been widely
investigated. This is quite understandable as metals and ceramics or insulators have very
limited applications. They either have very high (metals) or very low (ceramics)
conductivities. With proper doping, semiconductors can behave as semi-metals to semi-
insulators. These materials show very promising optical properties which is not the case of
metals or ceramics. It is the semiconductors that have been useful in making all kinds of
electronic devices and have unlimited applications such as solar cells, optoelectronics laser
applications and memory devices. In view of this, the study of semiconductors has become
13
very important area of research. Previously majority of work has been done on micro-
structured semiconductors. Study of materials in nanometer scale is an emerging area these
days because in nanometer scale structures, finite size gives rise to novel electronic,
magnetic, optical and structural properties. Thus, there is a tremendous scope to design new
materials with unusual properties. The drive towards miniaturization of electronic
components and integration to accommodate huge number of these components in small
volume has been there for decades.
1.5.1 Thin film growth process
Any thin film deposition process involves three main steps:
(a) Emission of particles from source.
(b) Transport of particles from source to substrate.
(c) Condensation of particles on substrate.
Fig. 1.7 Simple model for thin film deposition.
14
Various steps involved in thin film formation are:
(a) Thermal accommodation (b) Binding (c) Surface diffusion (d) Nucleation (e) Island
growth (f) Coalescence and (g) Continued growth.
We will examine each of these steps in turn. The general picture of step-by-step growth
process emerging out of the various experimental and theoretical studies can be presented as
follows:
The unit species, on impinging the substrate lose their velocity component normal to
the substrate and are physically adsorbed on the substrate surface.
The adsorbed species are not in thermal equilibrium with the substrate initially and
move over the substrate surface. In this process, they interact among themselves
forming bigger clusters.
The cluster or the nuclei as they are called, thermodynamically unstable and tend to
desorbs in a time depending on the deposition parameters. If the deposition
parameters are such that a cluster collides with other adsorbed species before getting
desorbed it starts growing in the size. After a certain critical size is reached the
cluster becomes thermodynamically stable and nucleation barrier is said to have been
overcome. This step involving the formation of stable, chemisorbed, critical sized
nuclei called as the nucleation stage.
The critical nuclei grow in number as well as in size until a saturation nucleation
density is reached. The nucleation density and the average nucleus size depend on a
number of parameters such as the energy of the impinging species, the rate of
impingement, the activation energies of adsorption, desorption and thermal diffusion,
15
the temperature, topography and chemical nature of the substrate. A nucleus can
grow both parallel to the substrate by surface diffusion of the adsorbed species as
well as perpendicular to it by direct impingement of the incident species. In general,
however, the rate of lateral growth at this stage is much higher than the perpendicular
growth. The grown nuclei are called islands.
The next stage in the process of film formation is the coalescence stage, in which the
small islands start coalescing with each other in attempt to reduce the surface area.
This tendency to form bigger islands is termed agglomeration and is enhanced by
increasing the surface mobility of the adsorbed species, as for example, by increasing
the substrate temperature in some cases formation of new nuclei may occur on the
areas freshly exposed as consequence of coalescence.
Larger islands grow together leaving channels and holes of uncovered substrate. The
structure of the films at this stage changes from discontinuous island type to porous
network type. Filling of the channels and holes forms a completely continuous film.
The growth process may be summarized as consisting of a statistical process of
nucleation surface-diffusion controlled growth of the three dimensional nuclei and
formation of a network structure and its subsequent filling to give a continuous film.
Depending on the thermodynamic parameters of the deposit and the substrate
surface, the initial nucleation and growth stages may be described as: Layer type,
Island type and Mixed type called Stranski-Krastanov type.
16
The details of thin film growth process have been illustrated in the Fig. 1.8. In almost all
practical cases, the growth takes place by island fabrication.
a) Island growth (Volmer-Weber)
Island growth forms three dimensional islands.
Source: film atoms more strongly bound to each other than to substrate and/or slow
diffusion.
b) Layer by layer growth (Frank - van der Merwe)
Layer by layer growth generally have highest crystalline quality.
Source: film atoms more strongly bound to substrate than to each other and/or fast diffusion.
c) Mixed growth (Stranski - Krastanov)
Initially layer by layer growth followed by the formation of three dimensional islands.
.
Fig. 1.8 (a, b, c) Layer by layer growth of thin films
17
Structure of a Schottky Diode
1.6 Metal -Semiconductor Contacts
All semiconductor devices have contacts. These contacts may be metal-
semiconductor or semiconductor-semiconductor type. The junction between metal and
semiconductor may be rectifying (Schottky) or non-rectifying (Ohmic) depending upon the
type and level of semiconductor doping and the relative work function of metal used.
A Schottky barrier is a junction between a metal and a semiconductor, which
exhibits rectifying characteristics of the junction and is known as Schottky diode. Schottky
diodes have low forward voltage drops and display an extremely fast switching action. They
have forward voltage drops of about 0.3 volts, as compared to 0.7 volts in Silicon p-n
junction diodes, which use adjacent p-type (positive) and n-type (negative) semiconductors.
Due to this, greater current density is present in Schottky barrier diodes as compared to p-n
junction diodes. Schottky devices also possess fast switching, because only one
semiconductor is used and no time is lost in recombination of majority and minority carriers
when conduction is to be halted (switched off).
Fig. 1.9 Structure of a Schottky Diode.
18
This also results in smaller devices, making these diodes useful in switch-mode
power converters operating at high frequencies. The structure of Schottky diode is shown in
Fig 1.9. It consists of a metal contacting a piece of semiconductor. An ideal ohmic contact, a
contact where no potential exists between the metal and the semiconductor, is made to the
other side of the semiconductor.
1.7 Current Transport Mechanism in Metal-Semiconductor Contacts
The junction between metal and semiconductor may be rectifying (Schottky) or non-
rectifying (ohmic) which depends on the type and extent of semiconductor doping and the
work function of the metal used. The knowledge of conduction mechanisms across a
schottky barrier is essential in order to calculate the schottky barrier parameters and explain
the observed effects.
The current transport in metal-semiconductor contacts is mainly due to majority
carriers in contrast to p-n junctions, where minority carriers are responsible. The various
ways in which electrons can be transported across a metal-semiconductor junction under
forward bias for n-type semiconductor are shown schematically in Fig. 1.10. The various
mechanisms are:-
(a) Emission of electrons from the semiconductor over the top of the barrier into
the metal.
(b) Quantum mechanical tunneling through the barrier.
(c) Recombination in the space charge region.
(d) Recombination in the neutral region.
19
Fig 1.10 Transport processes in a forward-biased Schottky barrier
It is possible to make practical Schottky-barrier diodes in which (a) is the most
important and such diodes are generally referred to as „nearly ideal‟. Processes (b), (c) and
(d) causes departures from this ideal behaviour [13].
1.7.1 Thermionic Emission (TE) Theory
The thermionic emission theory of the current transport phenomenon of metal
semiconductor schottky barrier has been derived by Bethe [14] from the assumptions as:
1. The barrier height qbn is much larger than kT.
2. Thermal equilibrium is established at the plane that determines emission.
3. The existence of a net current flow does not affect this equilibrium, so that
one can superimpose two current fluxes-one from metal to semiconductor, the
other from semiconductor to metal, each with a different image force.
20
Because of these assumptions, the slope of the barrier profile is immaterial and
current flow depends solely on the barrier height. The current density JSM from the
semiconductor to the metal is given by the concentration of electrons with energies
sufficient to overcome the potential barrier and traversing the x-direction.
JSM =
bF qE
xq
dn.. (1.1)
where EF + qb is the minimum energy required for thermionic emission into the metal and
ν x is the carrier velocity in the direction of the transport.
The electron density in an incremental energy range is given by
dvvkT
vm
kT
qV
h
mdn 2
23
*
42
*expexp2
(1.2)
Equation 1.2 gives the number of electrons per unit volume that have speeds between ν and
ν+dν distributed over all directions. If the speed is resolved into its components along the
axis with the x-axis parallel to the transport directions, we have
ν 2
= ν x2 + ν y
2 + ν z
2 (1.3)
with the transformation dv24 =d ν x.d ν y.d ν z
Substituting equation 1.2 and 1.3 in equation 1.1, we get
kT
vm
kT
qVT
h
kqmJ oxn
SM2
expexp4
2*
2
3
2* (1.4)
21
The velocity ν ox is the minimum velocity required in the x-direction to surmount the barrier
and is given by
VVqm biox 2*
2
1 (1.5)
Where Vbi is the built-in potential at zero-bias.
Substituting equation 1.5 into equation 1.4, it yields
kT
qV
kT
VVqT
h
kqmJ bin
SM expexp4 2
3
2* (1.6)
kT
qV
kT
qTAJ b
SM expexp2** (1.7)
where b is the barrier height and equals to the sum of Vn and Vbi and
3
2** *4
h
kqmA
(1.8)
is called the effective Richardson constant for thermionic emission, neglecting effects of
optical phonon scattering and quantum mechanical reflection.
Since the barrier height for electrons moving from metal into semiconductor remains
the same, the current flowing into the semiconductor is thus unaffected by the applied
voltage. It must, therefore, be equal to the current flowing from the semiconductor into the
metal when thermal equilibrium prevails (i.e. when V=0). The corresponding current density
is obtained from equation 1.7 by setting V=0.
22
kT
qTAJ nb
MS
exp2** (1.9)
The total current density is given by the sum of equations 1.7 and 1.9
1expexp2**
kT
qV
kT
qTAJ bn
n
(1.10)
1exp
kT
qVJJ STn (1.11)
where
kT
qTAJ bn
ST
exp
2**
1.7.2 The Diffusion Theory
To derive the current/voltage characteristic according to diffusion theory, we
consider the following assumptions [15]
1. Barrier height is much larger than kT.
2. The effect of electron collisions within the depletion region is included.
3. Carrier concentrations at x = 0 and at x = W are unaffected by the
current flow ( i.e. they have their equilibrium values).
4. The impurity concentration of the semiconductor is non-degenerate.
The current density in the depletion region in a usual way is given as
x
nDxnqJJ nnx )(
(1.12)
23
Where n(x) is the concentration of electrons in n-type semiconductor, μ their mobility, Dn
their diffusion constant, –q is the charge on an
electron [15]. For a semiconductor case, we assume the current to split up into drift and
diffusion components which are independent to each other. We now assume the quasi-fermi
level for electrons ζ n defined by
kTEqNn ncc /exp
(1.13)
where Nc is the effective density of states in the conduction band, Ec is the energy of the
bottom of the conduction band. Making use of Einstein‟s relationship μ/Dn = q/kT it is
possible to write equation 1.12 in the form:
dx
dnqJ n
(1.14)
Which shows the gradient of ζ n supplies “driving force” for electrons [15].
Combining equations 1.12 and 1.13 and integrating in the limits x = 0 to x = W we get
W
n
W
nkT
xqVxnqDdx
kT
xqVJ
00
)(exp)(
)(exp
(1.15)
Applying boundary conditions (1) to (4) as mentioned above and on simplification equation
1.15 yields
1expexp
22/1
2
kT
qV
kT
qNVVq
kT
NDqJ bn
s
Dbicn
n
(1.16)
24
1exp
kT
qVJJ SDn
(1.17)
where JSD is the saturation current density. The current density expressions of diffusion and
thermionic emission theories are similar. However, the “saturation current density” JSD for
diffusion theory varies more rapidly with voltage but is less sensitive to temperature
compared with the saturation current density JSD of thermionic emission theory [15].
1.7.3 Thermionic-Emission-Diffusion (TED) Theory
Many authors [16, 17, 18] have combined the thermionic-emission and diffusion
theories by considering the two mechanisms to be in series and effectively finding the
position of the quasi-fermi level at the interface which equalizes the current flowing through
each of them. Crowell and Sze [18] gave a most fully developed theory by incorporating the
boundary conditions of the thermionic recombination velocity R near metal-semiconductor
interface. Thus, the complete expression of J-V characteristics developed according to the
TED theory is given as [18]
1/ kTqV
s eJJ (1.18)
where
kT
qTAJ bn
s
exp2**
(1.19)
and DRQP
QP
ff
AffA
/1
*
**
(1.20)
25
Here A** is the effective Richardson‟s constant, A* the Richardson constant for thermionic
emission, R the thermionic recombination velocity, D effective diffusion velocity, fP
probability of electron emission over the potential maximum and fQ the ratio of total current
flow considering the quantum-mechanical tunneling and reflection to the current flow
neglecting these effects depends strongly on the electric field.
1.8 Recombination Processes
1.8.1 Recombination in the depletion region
The process of recombination normally takes place via localized states, and
according to recombination theory [19, 20] the most effective centers are those with energies
lying near the centre of the forbidden bandgap. The current density due to such
recombination centers in Schottky diodes for low forward bias is given by [13]
kT
qV
kT
qVJJ ror exp1
2exp
(1.21)
where Jro = qniw/2r, ni is the intrinsic electron concentration which is proportional to exp
(-qEg/2kT), w is the thickness of the depletion region, and r is the lifetime within the
depletion region.
The simple result embodies several drastic assumptions, namely that the energy
levels of the centers coincide with the intrinsic level, that the capture cross-sections
for electrons and holes are equal, and the centers are distributed in a spatially
uniform manner. None of these assumptions is likely to be true in practice,
especially the equality of electron and hole capture cross-sections and depend upon the
26
ratio of capture cross-section, the value for recombination may be between 1 and 2.
The total current density is given by
J = Jte + Jr
= Jto {exp(qV/kT)-1}+Jroexp(qV/2kT){1-exp(-qV/kT)}
= Jtoexp(qV/kT) + Jroexp(qV/2kT)}{1-exp(-qV/kT)} (1.22)
where, assuming the thermionic-emission theory, Jto=A** T 2
exp(-qb/kT).The ratio of the
thermionic to the recombination current is proportional to
T2r
exp(-q(Eg+V-2b)/2kT)
This ratio increases with r, V, and Eg, and decreases with b. Also, since
Eg+V-2b is usually negative for n-type semiconductor and small values of V, the
ratio increases with T. Thus the recombination component is likely to be relatively
more important in high barriers, in material of low lifetime, at low temperatures, and
at low forward-bias voltage. It is much more important in GaAs than in Si. When
recombination current is important, the temperature variation of the forward current shows
two activations energies. At high temperature the activation energy tends to the value b -V,
characteristic to the thermionic-emission component and at low temperature it approaches
the value (Eg-V)/2, characteristic of the recombination component. Recombination current
may therefore cause apparent deviations of from unity and the pre-exponential term from
the ideal value A**
T2exp(-qb/kT).
27
1.8.2 Recombination in the Neutral Region
If the height of a Schottky barrier on n-type material is greater than half the bandgap,
as is often the case, the region of the semiconductor adjacent to the metal becomes p-type
and contains a high density of holes. It might be expected that some of these holes diffuse
into the neutral region of the semiconductor under forward bias, thus giving rise to the
injection of holes. Hole injection at metal contacts was extensively studied in the early days
of semiconductors, and has been summarized by Henisch [21].
28
References -
[1] M S Tyagi, “Introduction to Semiconductor devices and materials”, John
Willy and Sons, Asia Pt. Ltd.1991
[2] E. A. Aleshina, V. P. Zlomanov, A. V. Novoselova,”Research on the P-T-x
phase diagram of the system Sn-Se”, Izv. Akad. Nauk SSSR, Neorg.Mater.
,18, 1982, 913.
[3] E. A. Kuliuhina, V. P. Zlomanov, A. V. Novoselova, P-T projection of the
phase diagram of the system SnS-Se, Izv. Akad. Nauk SSSR, Neorg. Mater.
13, 1977, 237.
[4] A. S. Pashinkin, A. S. Malkova, V. A. Surkova, T. V. Zotova,” Vapor
pressure at the surface of liquid SnSe”, Izv. Akad. Nauk SSSR, Neorg. Mater.
17, 1981, 169.
[5] M. I. Karahanova, A. S. Pashinkin, A. V. Novoselova, “On the melting
diagram of the system Sn-Se”, Izv. Akad. Nauk SSSR, Neorg. Mater. 2, 1966,
1186.
[6] A. M. Gasikov, V. P. Zlomanov, Iu. A. Sapojnikov, A. V. Novoselova, Study
of the phase diagram of the system Sn-Se, Vestnik Mosk. Univ., Ser. Himia,
No. 3, 1968, 48.
[7] B. I. Boltaks, K. V. Perepeci, P. P. Sereghin, V. T. Shipatov, Research on the
compounds of tin with the element of the sixth group by NMR, Izv. Akad.
Nauk SSSR, Neorg. Mat.6, 1970, 818.
[8] G. M. Bartenev, A. D. Tsiganov, S. A. Dembovskii, V. I. Michailov, Study of
the system Sn-Sand SnSe by Mossbauer effect., Izv. Akad. Nauk SSSR, Norg.
Mater, No. 7, 1971, 1442.
[9] H. G. Schnering, H. Wiedemeyer, The high temperature structure of β-SnS
and β-SnSe and the B16 to B33 type lambda transition path, Z. Kristallogr.
156, 1981, 143.
29
[10] V.V. Jdanova, “Second order phase transition in SnSe”, Fiz. Tverd. Tela
(russ.), 3, 1961, 1619.
[11] S. A. dembovskii, V. N. Egorov, A. S. Pashinkin, Iu. Ia. Poliakov, On the
problem of second order phase transition in SnSe, J. Neorg. Him. (russ.) 8,
1963, 1025.
[12] A. Dumon, A. Lichanot, S. Gromb, Propriétés électroniques du séléniure
d‟étain SnSe fritte:domaine d‟existence, J. Phys. Chem. Solids 38, 1977,
279.
[13] E.H Rhoderick and R.H. Williams,” Metal-Semiconductor Contacts”. 2nd
Ed. (Claredon, Oxford 1988).
[14] H.A Bethe.,”Theory of the boundary layer of crystal rectifier” MIT
Radiat.Lab. Rep.43 1942.
[15] W. Schottky , “Halbleitertheorie der Sperrschicht,” Naturwissenschaften,
1938, 26.
[16] W.Schultz , Z. Phys., 1954, 138.
[17] B.R. Gossick , Solid-St. Electron, 1963, 6.
[18] C.R Crowell and S.M Sze, “Current Transport in M-S Barriers”, Solid State
Electron, 1966, 9.
[19] W Shockley and W.T Read, Phys. Rev 1952, 87.
[20] R.N. Hall , Phys. Rev ,1952, 87.
[21] H. K. Henisch,”Rectifying Semiconductor Contacts” (Clarendon Press,
Oxford 1957).
30
Chapter 2 Measuring Equipments and Experimental Techniques _________________________________________________________________________________
2.1 Introduction
In this chapter, details of sample preparation, vacuum evaporation techniques and
experimental techniques used to study different structural, morphological, compositional,
optical and electrical properties have been discussed. Automated temperature and frequency
dependent setup used to measure current-voltage (I-V) as well as capacitance-voltage(C-V)
characteristics of the Schottky diodes have also been presented in this chapter.
2.2 Vacuum Technology
Virtually all thin-film deposition and processing methods as well as techniques
employed to characterize and measure the properties of films require a vacuum or some sort
of reduced-pressure environment. For this reason the relevant aspects of vacuum science and
technology are discussed at this point. „Vacuum‟– the word stands for pressure levels much
lower than the atmospheric pressure level. General units of vacuum are Torr, Pascal etc.
(1Torr = 1mm of mercury displacement). A true vacuum is a space containing no matter at
all. It is however, impossible to create a true vacuum. Pumping air out of a container can
produce an almost total vacuum, but some air molecules will always remain inside. The
open space between the stars is the closest thing we know to be a true vacuum.
31
2... 2 RP
KTmfp
All thin film deposition methods require vacuum because it improves the “mean free
path” of atoms of material being deposited. The “mean free path” is an average distance of
travel between subsequent collisions. It is most valuable concept since it gives a measure of
how readily particles will travel through gas. Now, during the travel from source to
substrate, the phenomena of atomic or molecular scattering and randomization occur. This
scattering is due to the collision of atoms or molecules of all kinds of vapour species and
residual gas molecules in the chamber. The scattering is related to density, pressure of atoms
and molecules in the gas phase and it defines the mean free path. From the kinetic theory of
gas this mean free path is calculated as:
(2.1)
where, K is Boltzmann constant at absolute temperature, R is molecular diameter and P is
pressure in Pascal. Thus mean free path is directly proportional to temperature of gas and
inversely proportional to pressure and square of the molecular diameter. At room
temperature (300K), for the typical diameter of 3A =1.455/P, the scattering probability is
given as fraction N/No of molecules that are scattered in distance “d” during their travel
through gas.
d
eN
N
10
(2.2)
where, No is total number of molecules that suffer collisions, d is the distance between
source and substrate and is the mean free path. So at pressure of 10-4
Pascal, only 0.3%
molecules will suffer collisions i.e. during evaporation molecular motion is non-randomized.
Here, we have seen that to have greater λ, pressure should be reduced and a high λ gives a
32
least scattering probability and thus the film deposition rate will be also high because of the
less scattering most molecules get migrated from source to substrate.
2.2.1 Vacuum coating unit
2.2.1.1 Vacuum pumps
The vacuum systems employed to deposit and characterize thin films contain an
assortment of pumps, tubing, valves, and gauges to establish and measure the required
reduced pressures [1]. Vacuum pumps (more correctly called air pumps) create a near
vacuum by removing the gas molecules from a vessel. Vacuum pumps may be divided into
two broad categories: gas transfer pumps and entrapment pumps. Gas transfer pumps
remove gas molecules from the pumped volume and convey them to the ambient in one or
more stages of compression. Entrapment pumps condense or chemically bind molecules at
walls situated within the chamber being pumped. In contrast to gas transfer pumps, which
remove gas permanently, some entrapment pumps are reversible and release trapped
(condensed) gas back into the system upon warm-up. Gas transfer pumps may be further
subdivided into positive-displacement and kinetic vacuum pumps. Rotary mechanical and
Roots pumps are important examples of the positive-displacement variety. Diffusion and
turbomolecular pumps are the outstanding examples of kinetic vacuum pumps. Diffusion
pumps allow a substance to diffuse into the pump intake, and then remove it. Modern
diffusion pumps can reduce the pressure inside a container to approximately 10-4
Pascal i.e.
about 10-9
times normal atmospheric pressure. Among the entrapment pumps commonly
employed are the adsorption, sputter-ion and cryogenic pumps. Each pump is used singly or
in a combination of variety of pumping system configurations. Ion pumps are used for very
33
low pressure gases. They work by ionizing a gas and then absorbing the ions on the surface
of a cathode inside the vessel. Turbo pumps use a rotating turbine to extract the gas.
Confusingly a high vacuum is one in which the pressure is very low, and a low vacuum is
one in which it is not quite so low.
Thus, the vacuum can be obtained with the help of following different types of pumps:
Rotary Pump
Diffusion Pump
The Rotary Pump is used to obtain rough vacuum. The Diffusion pump is used to
obtain high vacuum.
2.2.1.2 Pressure measurement gauges
Just as different pumping schemes must be used in the viscous and molecular flow
regimes, different methods of measuring the pressure must be used in different ranges as
well. To measure the vacuum there are two pressure gauges employed in this system. They
are: a) Pirani Gauge b) Penning Gauge.
(a)
34
(b)
Fig. 2.1 (a) Block diagram. (b) Vacuum coating unit for thermal evaporation
2.2.2 Selection of substrate
For deposition of thin films of a suitable supporting material known as substrate is
required. The surface of the substrate plays a major role in the nucleation and growth
process of the film and thereby influences the thin film properties considerably. An ideal
substrate should have the following requirements [3, 4].
(a) The surface should be flat and smooth.
(b) High mechanical strength to enable the substrate to withstand strain during
processing and monitoring.
(c) High resistivity.
(d) High thermal conductivity.
35
(e) Nearly same coefficient of thermal expansion with that of the film to
minimize thermal monitoring.
(f) Zero porosity to minimize out gassing and to ensure film uniformity.
(g) Low cost.
It has been observed that there is no material that would satisfy all these
requirements. Glass is most widely used as substrate material for deposition of
polycrystalline films. In the present study microscopic glass slides of the dimension
(75x25x1mm2) (Super deluxe glass slide, India) were used as substrate material due to their
fulfilment of most of the requirements as good substrates.
2.2.3 Cleaning of substrates
The quality of the substrate, prior to the growth of the thin film, is a crucial factor,
which influences the material properties of the deposited thin films. The substrates should be
highly cleaned and uncontaminated for proper adhesion of the films and for enhancing film
properties. Usually the fine dust particles due to packaging, fingerprints and sticking of
different impurity atoms are the common contaminants. The removal of these contaminants
by different cleaning techniques depends upon the nature of the substrate and the type of
contaminants. The chemical reagents such as acids, alcohol or alkalis with proper
concentrations remove the contaminants by breaking the bonds between the contaminants
molecules as well as between the contaminants and substrate. Acid converts oxide layer if
any and greases into water-soluble compounds. The ultrasonic cleaning is another
recommended process for removing gross contaminants such as greasy particles and
36
fingerprints. This procedure enhances the dissolution of residues sticking on the substrate by
the intense local stirring action by the shock waves created in the solvent. In order to obtain
glass substrates with a high degree of chemical cleanliness, the following procedure of
organic cleaning was used: 1) the glass substrates were rinsed in hydrogen peroxide to
remove contaminants; 2) substrates were then cleaned, in turn, under vapours of acetone,
trichloroethylene and methanol, respectively. Afterwards these were again rinsed with de-
ionised water. The drying of cleaned wet substrates is also critical because of probable
recontamination due to adsorption of gaseous particles and dust. So, necessary precautions
were taken in drying the substrate in an atmosphere free from any air borne contaminants.
Taking out the substrates from the deionised water, the substrates were put vertically in a
clean pettry dish so that there would be no water stick mark on the substrate. These were
then put inside a cleaned closed stainless still oven for drying for an hour at 373K.
2.2.4 Masks for generation of pattern in the deposited films
The desired shapes or patterns of films were generally obtained by masking
substrates during deposition so that only desired areas receive the vapour atoms. A mask
should be stable over the temperature range encountered during deposition and chemically
inactive with the vapour atoms. In the present case, freshly cleaved good quality of mica
sheets were used for making different masks according to the requirement. They were cut to
different geometrical shapes by shaving blades, micro punch square and rectangular. In
some cases aluminium and molybdenum foils were also used. The masks were thoroughly
cleaned by detergent and finally washed in acetone. They were dried properly by hot air
blower.
37
Fig. 2.2 Schematic diagrams of different types of films deposited on glass substrate.
2.2.5 Substrate heating
The radiation heater fitted above the substrate holder assembly was used to raise the
temperature of the substrate during deposition to any desired value (called substrate heater).
The power to the radiation heater was supplied and controlled from outside through an
autotransformer and a thermostat. The temperature was measured with the help of a copper-
constantan thermocouple which was connected to a digital micro voltmeter. By this method
the substrate temperature could be raised up to 635K.
2.2.6 Various Characterization Techniques
2.2.6.1 Structural and Morphological Characterization
2.2.6.1.1 X-ray diffraction (XRD)
38
X-ray diffraction (XRD) is one of the most important characterization technique used
in material science. X-ray scattering techniques are a family of non-destructive analytical
techniques which reveal information about the crystallographic structure, chemical
composition, and physical properties of materials and thin films. X-ray diffractometer is a
very useful analytical instrument for determination of the whole range of detail information
viz. crystal structure, orientation, crystalline sizes, lattice constants, defects, stresses and
strains developed in the samples. These techniques are based on observing the scattered
intensity of an X-ray beam hitting a sample as a function of incident and scattered angle,
polarization and wavelength or energy. When X-ray is incident on the surface of any
specimen, it reflects from different atomic position in different lattice planes. Interference of
the reflected ray from successive planes is in accordance with the Bragg‟s condition
2dsinθ = nλ, where n is order of diffraction analogous to a ruled grating, λ is wavelength of
the X-ray, d is the distance between the collimated incident X-ray beam from lattice plane in
the crystal. Photograph of X-ray diffractometer (Rigaku D-Max-III) used in this work is
shown in Fig 2.3. Here the specimen is mounted at the centre of the diffractometer and
rotated by an angle „‟ around an axis in the film plane. The counter is attracted to an arm
rotating around the same axis by angles twice as those of the specimen rotation. Only (h k l)
planes parallel to the film plane contribute to the diffraction intensity. The effective
thickness of the film in the thickness t is given by (t/sin) which decreases with increasing
diffraction angle. Therefore, the effective thickness of a film in the thickness range of
100nm is sufficient to excite measurable diffracted radiation at small angles but the intensity
falls of rapidly for higher index reflections. X-ray diffraction analysis gives a whole range of
information about the crystallographic aspects of a thin film.
39
Fig. 2.3 X-ray diffractometer (Rigaku D-Max-III).
2.2.6.1.2 Scanning Electron Microscope (SEM)
The Scanning Electron Microscope (SEM) is a type of electron microscope that
images the sample surface by scanning it with a high-energy beam of electrons in a raster
scan pattern. It finds its application to examine surface morphology of the material. In other
words we can say, SEM is a microscope that uses electrons rather than light to form an
image. The SEM is designed for direct studying of the surfaces of solid objects. The
electrons interact with the atoms that make up the sample producing signals that contain
information about the samples surface topography, composition and other properties such as
electrical conductivity. By scanning with an electron beam that has been generated and
focused by the operation of the microscope, an image is formed in much the same way as a
TV. The SEM allows a greater depth of focus than the optical microscope. For this reason
the SEM can produce an image that is a good representation of the three-dimensional
sample.
40
There are many advantages to use the SEM instead of a light microscope. The SEM
has a large depth of field, which allows a large amount of the sample to be in focus at one
time. A wide range of magnifications is possible, from about x 25 (about equivalent to that
of a powerful hand-lens) to about x 250,000, about 250 times the magnification limit of the
best microscopes. Preparation of the samples is relatively easy since most SEMs only
require the sample to be conductive. The combination of higher magnification, larger depth
of focus, greater resolution, and ease of sample observation makes the SEM one of the most
heavily used instruments in research areas today. Fig. 2.4 shows photograph of SEM Model:
JEOL JSM 5600 which was used in this work.
Fig. 2.4 Modern Version of Scanning Electron Microscope.
2.2.6.1.3 Atomic force microscope (AFM)
The Atomic Force Microscope (AFM) or scanning force microscope (SFM) is a very
high-resolution type of scanning probe microscope, with demonstrated resolution of
41
fractions of a nanometer, more than 1000 times better than the optical diffraction limit. The
precursor to the AFM, the scanning tunneling microscope, was developed by Gerd Binnig
and Heinrich Rohrer in the early 1980s, a development that earned them the Nobel Prize for
Physics in 1986. Binnig, Quate and Gerber invented the first AFM in 1986. The AFM is one
of the foremost tools for imaging, measuring and manipulating matter at the nanoscale. The
information is gathered by "feeling" the surface with a mechanical probe. Piezoelectric
elements that facilitate tiny but accurate and precise movements on (electronic) command
enable the very precise scanning.
The AFM has several advantages over the scanning electron microscope (SEM).
Unlike the electron microscope which provides a two-dimensional projection or a two-
dimensional image of a sample, the AFM provides a true three-dimensional surface profile.
Additionally, samples viewed by AFM do not require any special treatments (such as
metal/carbon coatings) that would irreversibly change or damage the sample. While an
electron microscope needs an expensive vacuum environment for proper operation. In
principle, AFM can provide higher resolution than SEM. It has been shown to give true
atomic resolution in ultra-high vacuum (UHV) and, more recently, in liquid environments.
High resolution AFM is comparable in resolution to Scanning Tunneling Microscopy and
Transmission Electron Microscopy. The disadvantage of AFM compared with the scanning
electron microscope (SEM) is the image size. The SEM can image an area on the order of
millimetres by millimetres with a depth of field on the order of millimetres. The AFM can
only image a maximum height on the order of micrometres and a maximum scanning area of
around 150 by 150 micrometres [6].
42
Fig.2.5 Atomic Force Microscopy.
2.2.6.1.4 Energy-dispersive X-ray spectroscopy
Energy dispersive X-ray spectroscopy (EDS, EDX or EDXRF) is an analytical
technique used for the elemental analysis or chemical characterization of a sample. It is one
of the variants of XRF. As a type of spectroscopy, it relies on the investigation of a sample
through interactions between electromagnetic radiation and matter, analyzing x-rays emitted
by the matter in response to being hit with charged particles. Its characterization capabilities
are due in large part to the fundamental principle that each element has a unique atomic
structure allowing x-rays that are characteristic of an element's atomic structure to be
identified uniquely from each other.
To stimulate the emission of characteristic X-rays from a specimen, a high energy
beam of charged particles such as electrons or protons or a beam of X-rays is focused into
the sample being studied. At rest, an atom within the sample contains ground state (or
43
unexcited) electrons in discrete energy levels or electron shells bound to the nucleus. The
incident beam may excite an electron in an inner shell, ejecting it from the shell while
creating an electron hole where the electron was. An electron from an outer, higher-energy
shell then fills the hole, and the difference in energy between the higher-energy shell and the
lower energy shell may be released in the form of an X-ray. The number and energy of the
X-rays emitted from a specimen can be measured by an energy dispersive spectrometer. As
the energy of the X-rays are characteristic of the difference in energy between the two
shells, and of the atomic structure of the element from which they were emitted, this allows
the elemental composition of the specimen to be measured [7]. Experimental setup of EDX
(JEOL) is shown in Fig. 2.6.
Fig. 2.6 EDX Experiment set up (JEOL).
44
2.2.6.2 Electrical characterization
The electrical properties of SnSe thin films are required for the optimization of the
preparation conditions to use it as an absorber layer in solar cell. Electrical characterization
of SnSe thin film was carried out by using silver-paste (ohmic contact). The type of
electrical conduction in SnSe thin film is verified using the hot-probe method while the
resistivity measurements were carried out using the standard Hall Effect setup. Two probe
low temperature resistivity measurement and four probes with high resistivity temperature
measurement was used for temperature dependent resistivity measurement. The referred set
ups has been described as under:
2.2.6.2.1 Hot-probe experiment
The "hot-probe" experiment provides a very simple way to distinguish between n-
type and p-type semiconductors using a soldering iron and a standard multimeter. The
experiment is performed by contacting a semiconductor wafer with a "hot" probe such as a
heated soldering iron and a "cold" probe. Both probes are wired to a sensitive current meter.
The hot probe is connected to the positive terminal of the meter while the cold probe is
connected to the negative terminal. The experimental set-up is shown in the Fig.2.7. When
applying the probes to n-type material one obtains a positive current reading on the meter,
while p-type material yields a negative current.
45
Fig. 2.7 Experimental set-up of the "hot-probe" experiment.
A simple explanation for this experiment is that the carriers move within the
semiconductor from the hot probe to the cold probe. While diffusion seems to be a plausible
mechanism to cause the carrier flow but it is actually not the most important mechanism
since the material is uniformly doped.
2.2.6.2.2 Hall experiment
The Hall Effect was discovered by E.H. Hall in 1879 during an investigation on the
force acting on a current carrying conductor in a magnetic field. The Hall effect
measurements provide information on the carrier type, the carrier concentration and the
mobility of the carriers at a given temperature.
Fig. 2.8 Configuration for measuring Hall Effect.
46
When a magnetic field is applied perpendicular to the direction of current flowing in
the z-direction as shown in Fig 2.8, the carriers will be deflected due to then Lorentz force
and an electric field is built up along the y-direction resulting from the accumulated carriers
at y=0 surface of the semiconductor. The electric field produced by the deflected carriers is
called the Hall field. The direction of this field depends on the type of carriers responsible
for the current flow. Since there is no net current along the y-direction in the steady state, the
magnetic field force will be exactly balanced by the induced electric field force which can
be expressed as [11]
0)( BvEqE
(2.3)
The current density in terms of drift velocity is defined as,
vnqJ
(2.4)
Thus from eq.2.3 and eq.2.4, the hall field can be obtained for the configuration considered,
as the following
nq
BJE zx
y (2.5)
Hall field is proportional to the product of current density and magnetic field. The
proportionality constant is defined as the Hall coefficient and in general given by,
pq
r
nq
r
BJ
ER
zx
y
H , (2.6)
where r is the Hall factor which depends on the scattering mechanism in the semiconductor.
In the high magnetic field limit, r is of the order of unity. The positive sign is for the case
when the free carriers are holes and negative sign for the case when free carriers are
47
electrons. Therefore, Hall coefficient leads to a determination of the carrier type as well as
the carrier concentration. If Ohm‟s law obeyed, the conductivity must be independent of the
applied electric field. Thus the conductivity is defined as ζ = nqμ ,where μ is the electron
drift velocity per unit electric field (or equal to the Hall mobility for free electrons). Then,
the Hall mobility is given by the following expression,
RHH (2.7)
and measured quantity hall voltage (VH=Eyw) can be derived from equation (2.8) as
t
BIRV
H
H
(2.8)
where I is the current passing through the sample and t is the thickness along the direction of
the magnetic field.
Fig. 2.9 A photograph of a probe with a Ge sample, with contact numbers labelled.
The sample is on the front surface of the plastic substrate in this photo.
2.2.6.2.3 Two-probe method
Low temperature resistivity is done in a liquid nitrogen bath in the temperature range
80-330K using Keithley Model 6517A programmable electrometer. Using the Keithley
Model 6521 scanner card, measurements of resistivity of ten samples simultaneously can be
done. Well-shielded standard triax cables are used to obtain accurate resistance values as the
48
measurements being done require for very low currents and low noise. Shielding and
grounding are also done for obtaining reliable data. Depending on the type of the sample
conductivity, a voltage limit is adjusted to obtain reliable data. In addition, longer waiting
time (4-5m) for measurements of each data point for highly resistive samples is also
observed. The data collected is normally repeated for reproducibility check. A Lakeshore
model 340-temperature controller is used for controlling and measuring the temperature (T).
After stabilizing to the desired temperature, the resistance Values are normally recorded
three times and their mean is noted. Once the dimensional factors are determined for each
sample, the resistivity, either surface or volume, are obtained. The resistivity thus obtained
has an estimated error within 5%. The master control in the data acquisition is done through
personal computer using a visual basic program.
Fig. 2.10 Low temperature resistivity measurements by Two-probe method.
49
2.2.6.2.4 Four probe method
The Four Probe Method is one of the standard and most widely used method for the
measurement of resistivity of semiconductors. The experimental arrangement is illustrated in
Fig.2.11. In its useful form, the four probes are collinear. The error due to contact resistance,
which is specially serious in the electrical measurement on semiconductors, is avoided by
the use of two extra contacts (probes) between the current contacts. In this arrangement the
contact resistance may all be high compared to the sample resistance, but as long as the
resistance of the sample and contact resistances are small compared with the effective
resistance of the voltage measuring device (potentiometer, electrometer or electronic
voltmeter),the measured value will remain unaffected. Because of pressure contacts, the
arrangement is also specially useful for quick measurement on different samples or sampling
different parts of the same sample.
(a)
50
(b)
Fig. 2.11 Four Probe Set-up (a) Schematic view (b) Experiment set-up.
Sample Geometry:
It is preferable to fabricate samples from thin plates of the semiconductor material
and to adopt a suitable geometry, as illustrated in Fig. 2.12.
Fig. 2.12 Geometrical patterns of thin films used for four probe set up.
51
2.2.6.3 Optical Characterisation
Optical characterization of thin films is done by using J.Tauc‟s method used for
materials having direct transition [8, 9]. The glass substrates have been used as reference for
optical studies. They have high transparency in the required wavelength range. Fig 2.13
shows the transmission spectrum of substrate used for optical measurements.
Fig. 2.13 Transmission spectrum of glass substrate used for optical studies.
400 600 800 1000 1200 1400 1600 1800 2000
99.6
99.8
100.0
100.2
100.4
100.6
100.8
101.0
Tra
nsm
issio
n
Intensity
__ Transmission Value
Chopper
Controller
Optical
chopper
PC
Tungsten
Halogen
Lamp
Lock-in
amplifier
Mono-
chromator
Detector
Sample
Fig. 2.14 Optical Set Up for the transmission measurement.
52
Fig. 2.14 shows the transmission measurement setup for optical studies. Here, a
Tungsten-Halogen lamp is used as a polychromatic light source. The light from the lamp is
focused on the monochromator input slit using a convex lens. We have used 1/8m
monochromator (CVI-CM110). The output beam from the monochromator is chopped using
a mechanical chopper. The higher order wavelengths coming out of the monochromator are
removed using an optical high-pass filter as shown in the Fig.2.14. This chopped beam is
then incident on the sample near-normal geometry and the transmittance beam is directed to
the photo-detector. The detector measures the intensity of the transmission beam with the
help of lock-in amplifier (SR-530) and the transmittance of the sample is measured. The
monochromator and the lock-in amplifier have been interfaced with the computer using
COM port and GPIB, respectively. The experiment is automated using LabVIEW.
2.2.6.4 Temperature and Frequency dependent I-V & C-V measurements
Fig. 2.15 Automated Temperature and Frequency Dependent I-V Measurement Setup.
53
The current-voltage (I-V) as well as capacitance-voltage (C-V) measurements of the
Ag/p-SnSe Schottky Diodes were measured using a computer interfaced setup comprising
a programmable Keithley source meter (model-2400), Closed Cycle Liquid Helium
Cryogenic setup (CTI-Cryotronics model 22C) equipped with temperature controller (Lake
Shore-model-321) and Precision programmable LCR meter (Aglient make 4284A).
Interfacing of I-V and C-V measurement equipments was achieved by using LabVIEW
software by National Instruments (U.S.A).
54
References -
[1] “Vacuum Technology: Its Foundations, Formulae and Tables,” in Product
and Vacuum Technology Reference Book, Leybold-Heraeus, SanJose, CA,
1986.
[2] Vacuum Coating Unit, “operation and maintenance manual”, HINDHIVAC
(12A4D).
[3] B. Tareev, Physics of Dielectric Materials, Mir publisher, Moscow, 1975, 46.
[4] L. Holland, Vocuum Deposition of Thin Films, Chapmann and Hall Ltd.,
London, 258, 1963, 100.
[5]. S. Tolansky, Multiple beam Interferomerty of surfaces and films, Oxford
Uni.Press, London, N Y, 1948.
[6] M. S. Tyagi, “Introduction to Semiconductor Materials and Devices.” John
Wiley & Sons, New York , 1991.
[7] D. K. Rao, J. J. B. Prasad, D. Sridevi, K. V. Reddy, J. Sobhnadri, Phys. Stat.
sol.(a), 153 , 1986, 94
[8] J. Tauc ,A. Menth, J. Non-Cryst. Sol., 8, 1972, 569.
[9] I. Chakraborty , S.P. Moulik, J. Nanopart. Res. ,7, 2005, 237.
[10] Agilent, "Agilent 4284A Prescision LCR Meter Operation Manual," Japan,
1999.
55
[11] E. H. Putley “The Hall Effect and Semiconductor Physics”, Dover
Publications, New York, 1960.
56
Chapter3 Impact of Substrate Temperature on the Grain Size and Other Properties of SnSe Thin Films _____________________________________
3.2 Introduction
It has been observed that the structural, compositional, morphological, optical and
electrical properties of thin film compound semiconductors vary with the change in the
substrate temperature. Thus, substrate temperature plays a crucial role in the improvement of
grain size as well as to control the properties of thin film of any polycrystalline
semiconducting materials. The improvement in the grain size and other parameters of the
SnSe thin films may results in their use for better device applications.
3.2 Preparation of Tin Selenide (SnSe) Thin films
The SnSe thin films were all grown on organically cleaned soda lime glass substrates
at different substrate temperatures (in the range of 350-550 K) from fine-grained pulverized
Tin Selenide (SnSe) powder (99.99 %) obtained from Alfa Aesar (USA). The deposition
parameters like rate of deposition, thickness and distance between source to substrate were
kept constant. The deposition rate was kept 0.3 nm/s which was continuously monitored
during the deposition using a quartz crystal thickness monitor DTM -101(Hind Hi Vac.,
India) with base pressure maintained at 10-5
Torr.The thickness of deposited films has been
57
fixed around 100nm.The substrate temperature was monitored using chromel-alumel
thermocouple, which was kept in direct contact with the substrate.
Another set of films of fixed thickness of 200nm was deposited at the substrate
temperatures of 550K, 575K, and 600K.This attempt has been made to achieve better
crystallinity as well as improved material parameters with a purpose of better current
transport phenomenon in SnSe polycrystalline films for use in electronics devices.
3.3 Results and Discussion
3.3.1 X-ray diffractogram of SnSe thin films
The structural properties of SnSe films (as-deposited) were studied by X-ray
diffraction method. The XRD profile of SnSe thin films deposited on glass substrate at the
substrate temperatures (Ts) are 323K, 373K, 423K, 473K and 523K shown in Fig.3.1.The
prominent Bragg reflection is occurring at or around 2θ = 30° corresponding to (111)
diffraction plane, along with three other very weak diffraction peaks viz. (113), (020), (203)
which confirms the polycrystalline nature of the film. A similar preferred orientation of
grains along the (111) plane in SnSe film was observed in the evaporated SnSe thin films by
Bhatt et al. [1] and by Dang Tran Quan [2]. On the other hand, H. Chandra et al. [3] had
observed (400) diffraction plane for films grown by flash evaporation technique and Teghil
et al. [4] had reported orientation of grains along (011) and (200) crystallographic planes in
the SnSe thin film prepared by Laser Ablation method. The various preferred orientation of
grains reported for SnSe films deposited using different techniques indicate that the mode of
deposition plays a decisive role on the growth structure of the films. The analysis of the
58
cos
94.0D
diffraction patterns also suggest that the undertaken SnSe thin-films possess orthorhombic
crystal system with lattice parameters a=b=0.429nm and c=0.523nm belonging to the D2h16
space group while the d-value corresponding to the (111) prominent peak determined to be
0.292 nm at 523K. The obtained d-values of film matches well with Joint Council of Powder
Diffraction Standards (JCPDS) data card [5]. Furthermore, it is observed that as Ts increases
the intensity of the diffraction peaks increases and we get well-resolved peaks at 523K
substrate temperature. This could be linked with the grain-growth with increase in substrate
temperature. This point requires further investigation of the film‟s microstructure.
3.3.1.1 Grain size measurement
The inter-planar spacing dhkl was calculated for the (111) plane by using X-ray
diffraction data using the Bragg‟s relation [6]:
sin2
ndhkl (3.1)
Where λ represents the wavelength of the X-ray used, d; the lattice spacing, n; the order
number and θ; the Bragg‟s angle. The factor d is related to (hkl) indices of the planes and the
dimension of the unit cells. The Full Width at Half Maximum (FWHM) for most preferred
(prominent diffraction peak) planes (111) of all the films prepared at different substrate
temperatures were measured. FWHM has been found to decrease markedly with increase in
substrate temperature. The grain size (D) of SnSe films was calculated from the value of
FWHM (β) of (111) peaks expressed in radian by using the Debye-Scherrer‟s relation [6]
(3.2)
59
Where D represents the grain Size and β the FWHM calculated from the (111)
plane. Table 3.1 represents the grain size for different planes of the SnSe thin films of
different substrate temperatures while Fig. 3.2 shows deviation of FWHM as well as grain
size with change in substrate temperature.The grain size is found to increase with increase of
substrate temperature.
Fig. 3.1 XRD spectra of SnSe thin film of the thickness100nm deposited at
substrate temperature323K,373K,423K,473K and 523K.
60
Substrate Temperature 0C
Interplanar Spacing(d )A0
FWHM (Degree)
Grain Size (nm)
323K 2.918 0.88 9.78
373K 2.918 0.75 11.48
423K 2.918 0.39 22.07
473K 2.918 0.38 22.65
523K 2.879 0.32 26.93
Table 3.1 Structural parameters of SnSe thin films deposited at a thickness of
100 nm on glass substrate at different substrate temperatures with preferred
orientation along (111) plane.
300 350 400 450 500 550
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Substrate Temperature (K)
Grain
Size (n
m)
FW
HM
(2)
Substrate Temperature (K)
300 350 400 450 500 550
8
10
12
14
16
18
20
22
24
26
28SnSe Thin Film
Fig. 3.2 Variation of the grain size(nm) and FWHM(degrees) of SnSe thin films with
substrate temperature(K).
3.3.1.2 Strain and Dislocation density
The Strain (η), particle size (D) and Dislocation density (δ) have been calculated
using the Williamson and Smallman relation [6]
61
sincos
D (3.3)
Where λ is the wavelength of the radiation used (0.15418nm), is the full width at half
maximum, and θ the angle of diffraction.
Fig. 3.3 Plot of βcosθ vs sinθ for the SnSe thin film deposited at the substrate temperature
(a)323K (b)523K.
62
Fig.3.3 (a) & (b) represents the Williamson and Smallman plots for the SnSe thin
films prepared at different substrate temperatures. In this study, the corrected value of full
width half maximum (β) of each peak was measured by subtracting instrumental broadening
from the observed peak width. The average grain size (D) and average strain (η) of the films
were calculated from the intercept and slope of these linear plots respectively. The
calculated average grain sizes and average strain of different films are tabulated in Table
3.2. The average grain size of the thin films prepared at 323K is found to be 9.6nm which is
increased to 26.6nm for the film prepared at 523K.
Table 3.2 Micro-structural parameters of SnSe thin films deposited on the glass substrate
at different substrate temperatures for the most prominent (111) planes.
Substrate Temperature
(K)
FWHM (Degrees)
Grain Size (nm)
Dislocation Density
(1010line/cm2) Strain
323K 0.88 9.679 0.10225 0.24175
373K 0.75 11.352 0.08711 0.20732
423K 0.39 21.821 0.04531 0.1072
473K 0.38 22.432 0.04415 0.10334
523K 0.32 26.612 0.03713 0.0823
63
3.3.1.3 Dislocation density of SnSe thin films
The dislocation density along preferred orientation [111] calculated with the help of
grain size by using equation (3.4)
2
1
D
(3.4)
Fig. 3.4 Variation of Dislocation Density of 100nm SnSe thin films deposited at different
substrate temperatures.
It is seen that the dislocation density decreases significantly with increase of substrate
temperature up to 423K and then decreases slowly beyond this.
3.3.2 Compositional analysis
Energy dispersive X-ray spectra of SnSe thin films revealed that the Sn and Se
contents depend critically on the substrate temperature. Figure 3.5 (a) & (b) show the
64
EDAX pattern of the SnSe thin films grown at different substrate temperatures.
Fig. 3.5 The EDAX spectrum giving the compositional information of SnSe thin film
deposited at substrate temperature of (a) 323K (b) 523K.
From the figure 3.5 (a) and (b), Sn and Se peaks are observed at all substrate
temperatures and some other peaks are also observed which corresponds to Si, Na, Ca and
O that can be attributed to the glass substrate used [7]. It is also observed that as the
substrate temperature increases, the change in atomic mass percentage of Sn and Se has
been observed. The atomic mass percentages of Sn and Se of the films grown at 323K
temperature have been found to be 31.78 and 33.18 respectively. This shows that the film
(a)
(b)
65
grown at 323K is rich in Selenium. As substrate temperature increases, the atomic mass
percentages of Sn and Se approach to stoichiometric ratio. In order to deposit SnSe films,
the stoichiometric ratio between atomic mass percentages should be 1:1 (SnSe). The
stoichiometric ratio between atomic masses is nearly 1:1 (27.45:27.12) for the SnSe film
deposited at 523K Fig.3.5 (b). Therefore, the films deposited at substrate temperature 523K
are stoichiometric in nature.
3.3.3 Optical Properties
The transmission spectra of SnSe thin films, which were deposited at different
substrate temperatures (Ts), were recorded in the wavelength range of 500 to 1500 nm, at
normal incidence. From these spectral data, the optical absorption coefficient „α‟ was
calculated using Lambert‟s law
dAI
IIn 303.20
(3.5)
where, A represents the optical absorbance, d the film thickness, Io and I are the intensities
of the incident and the transmitted light, respectively.
66
Fig. 3.6(a) Plots of (αhν) 2
versus (hν) for SnSe thin films deposited at Ts of 323k.
Fig.3.6 (b) Plots of (αhν) 2
versus.hν for SnSe thin films deposited at Ts of 373k.
67
Fig.3.6 (c) Plots of (αhν) 2
versus hν for SnSe thin films deposited at Ts of 423K.
Fig. 3.6 (d) Plots of (αhν) 2
versus hν for SnSe thin films deposited at Ts of 473K.
68
Fig.3.6 (e) Plots of (αhν)2
versus hν for SnSe thin films deposited at Ts of 523K.
The absorption coefficient (α) was found to follow the relation
2)( gEhvBhv (3.6)
Where B is a constant, and Eg is the bandgap energy. Plots of (αhν) 2 versus the photon
energy (hν), for films deposited at different substrate temperature (Ts) viz.
323K,373K,423K,473K and 523K are shown in Fig.3.6 (a) to (e). The linearity of the above
plots near the absorption edge indicates that the material is direct bandgap. Hence, the
energy-axis intercept of the linear part yields the energy bandgap of SnSe thin films [8-9].
The energy bandgap of films deposited was in the range of 1.59-1.19 eV. These values are in
good agreement with the bandgap values as reported by other workers [8-12]. It is clear that
as Ts increases the energy bandgap decreases. The decrease in the direct bandgap energy
with increase in Ts can be on the basis of the fact that the crystallinity of the deposited
polycrystalline SnSe films improves with increasing substrate temperature.
69
3.3.4 Morphological study of SnSe thin films by Scanning Electron Microscope
(SEM)
The microstructure of the SnSe thin films, deposited at different substrate
temperature range from 323K to 523K were investigated using Scanning Electron
Microscopy (SEM) to observe its surface topography. Fig. 3.7 (a) to (e) shows SEM images
of the synthesized SnSe thin films deposited at different substrate temperatures. The SEM
micrograph shows that the grains are distributed to cover the surface of the substrate
completely. No such microscopic defects like void, pinholes, peeling and cracks could be
observed for the undertaken samples. It is also observed that with the increase in substrate
temperature the grain size increases, while the density of the grains decreases. It may be due
to coalescence of small grains into effectively large grain. The increase in grain size with
substrate temperature indicates the increase in the crystallinity of the film. These results of
SEM analysis corroborate with the results obtained from the XRD data.
70
71
Fig. 3.7 SEM images of SnSe thin films of 100nm deposited on glass substrate at substrate
temperature (a) 323K (b) 373K(c) 423K (d) 473K and(e)523K.
72
3.3.5 Electrical Studies
3.3.5.1 Resistivity Measurement
For all electrical measurements performed in this study, non-rectifying (ohmic)
contacts of the investigated films were achieved using silver-paste electrodes. The type of
electrical conduction (p-type) in SnSe thin films was verified using the hot-probe method
while the resistivity measurements were carried out using the Standard Hall-Effect
measurement setup. Fig. 3.8 shows the variation of electrical resistivity with substrate
temperature. The decrease in resistivity with the increase in Ts can be explained using the
Petritz barrier model, which predicts that the crystallites do not grow sufficiently large at
low temperatures and the larger inter-crystalline regions offer high resistance for the
movement of the charge carriers. At high substrate temperature, the formation of fewer
nucleation centre results in large crystallite size, which may ultimately decrease the inter-
crystalline barriers, hence decreasing the electrical resistivity. The resistivity values of SnSe
films, deposited at different substrate temperatures varied between 112- 15 Ω-cm and was
found strongly influenced by substrate temperature. The resistivity value for the thin films
deposited at 523K is found in close approximation to the value reported by H. Chandra et al.
[3].
73
Fig. 3.8 The variation of the electrical resistivity of SnSe thin films measured at different
substrate temperatures.
3.3.5.2 Activation energy
The electrical conductivity of a polycrystalline thin film sample is a complex
phenomenon, involving charge-carriers transport through both the “bulk-like” part of the
semiconductor crystals and through the inter-crystalline (grain) boundaries.
In the literature [13] the temperature dependence of the semiconductor material‟s
conductivity is expressed by the equation.
KT
Eaexp0 (3.7)
Where 0 represents the pre-exponential factor, Ea the activation energy for this thermally
activated process and k the Boltzmann constant.
74
Fig. 3.9 Plots of the resistance (R) as a function of temperature (T) for SnSe thin films
deposited at (a) Ts =323K (b) Ts = 373K(c) Ts = 423K (d) Ts = 473K and (e)
Ts = 523K.
75
76
Fig. 3.10 Plot of ln(R/R300K) versus 1/T for a SnSe thin film grown at (a) Ts =323K (b) Ts =
373K(c) Ts = 423K (d) Ts = 473K and (e) Ts = 523K.
77
Fig. 3.11 The variation of the electrical Activation Energy of SnSe thin films deposited at
different substrate temperatures.
Clearly, a plot of ln(R/R300K) versus 1/T indicates a straight line, from the slope of
which, the activation energy can be calculated. Thus, to measure the conductivity it is
enough to measure the electrical resistance R since we are interested in the slope of the
linear-least square fit only. So, the temperature dependence of resistance of SnSe thin films
have been studied by measuring the resistance in the temperature range 80-333 K.
Depending on the sample conductivity, a voltage limit was adjusted to obtain reliable data.
The data collected was normally repeated for reproducibility check. After stabilizing to the
desired temperature, the resistance values were normally recorded three times and their
mean was noted. Once the dimensional factors were determined for each sample, the
resistivity values were calculated. The resistivity thus obtained had an estimated error within
5%.
78
Figure 3.9 (a) to (e) shows the plots of resistance versus temperature of SnSe thin
films deposited at different substrate temperature. The decrease in resistance with increase in
temperature indicates semiconducting behaviour of the thin films. The activation energy
values calculated from the linear-least square fit of the plots for SnSe thin films, deposited at
different Ts, were in the range 0.14-0.28 eV which is shown in the Fig. 3.11.The values of
the activation energy for electrical conduction closely correspond with the measurements
performed by another group [14].
3. 4.1 Structural properties
Alternative studies of deposited 200nm thick film were investigated at the substrate
temperature of 550K, 575K and 600K. This attempt was initiated to achieve better
crystallinity as well as improved material parameters to achieve an optimized current
transport phenomenon in SnSe films for use in electronic devices. X-ray diffraction patterns
recorded for the deposited SnSe films deposited on glass plate at different substrate
temperatures ( 550K, 575K and 600K) are shown in Figure 3.12 (a) to (c). The XRD studies
revealed that the are polycrystalline in nature with orthorhombic structure. The prominent
Bragg reflection is occurring at around 2θ = 300 corresponding to (111) diffraction planes,
along with two other very weak peaks of (221) and (420) planes. The different peaks in the
diffractogram were indexed and the corresponding values of interplanar spacing „d‟ were
calculated and compared with the standard values [5].
79
Fig. 3.12 XRD patterns of SnSe thin films deposited at various substrate temperatures
(a) 550K, (b) 575K and (c) 600K.
The main features of the diffraction patterns of the films prepared at different
substrate temperatures were the same but only change occurs in width and intensity of the
peaks. It has been found that films deposited at substrate temperature 575K led to the
formation of well -crystallized films. The width of (111) peak in X-ray diffraction pattern
has shown sharp peaks and small FWHM data which contribute in the enhancement of
crystallite size.
80
The FWHM (Full Width at Half Maximum) for most preferred (prominent
diffraction peak) (111) plane of the films prepared at different substrate temperatures has
been found to decrease from 550K to 575K thereafter it sharply increases at 600K.
The values of Interplanar spacing (d), Grain size (D) and Dislocation density ( )
were calculated from eq. (3.1), (3.2) and (3.4) are given in Table 3.4 respectively.
Substrate Temperature(K)
Hkl d-Spacing FWHM(2θ) Grain Size
(nm)
Dislocation density (1010
line\m2)
550K
1 111 2.88636 0.9446 9.12 0.01202
2 420 1.4418 0.768 12.79 0.0061
575K
1 111 2.88179 0.25 34.48 0.00081
2 221 1.83748 0.9446 9.68 0.0106
3 420 1.44346 1.152 8.52 0.0137
600K
1 111 2.8865 0.9446 9.12 0.01202
2 420 1.4388 0.768 12.80 0.0061
Table 3.3 Structural parameter of SnSe film deposited on glass substrate at temperature of
(a)525K (b)575K and(c)600K..
It is observed from the table that the crystallite size increases with increase in
substrate temperature and films deposited at 575K were found to have maximum value of
crystallite size It is also observed that the dislocation density decrease with increase in
temperature from 550K to 575K, thereafter it slightly increases. The minimum value of
dislocation density was found at substrate temperature of 575K.
81
3.4.1.3 Surface Morphological Studies
In this technique, we look at surface contours, measuring average roughness Ra and
root-mean-square roughness Rq of the surface of some selected area of SnSe thin films. The
average surface roughness Ra, the most frequently used roughness parameter is the defined
as
n
i
ia ZR14
1
(3.8)
where Zi is the height or depth of the ith highest or lowest deviation and n is the number of
discrete profile deviation. The root-mean-square surface roughness Rq, which is defined as
the root-mean-square of the deviations in the height from the profile mean which may be
defined as follows
2
1
1
n
i
iZn
Rq (3.9)
where roughness parameters Ra and Rq are often used as quantitative parameters. The
morphological analysis of the films, grown at different substrate temperatures were carried
out using Atomic force Microscope (AFM). Figure 3.13 (a) to (c) shows two and three
dimensional surface of AFM images of the film grown at different Ts. The SnSe thin films
prepared at 550K Ts indicate that the growth of small grains distributed across the surface of
the substrate as seen in Figure 3.13(a). The size of the grains is rather different from each
other indicating irregular growth rate of the grain. At substrate temperature 575K, from
micrographs one can see the uniform distribution of grain size over total coverage of the
substrate with a compact and fine grained morphology. There is an increase in nucleation
82
over growth and the film surface is covered with uniform grain without pinholes as seen in
Figure 3.13(b). At substrate temperature of 600K uniform distribution of grain size of SnSe
thin films strongly effect and shows the reduction in grain size as seen in Figure 3.13(c).
(a)
(b)
83
(c)
Fig. 3.13 AFM images(2D and 3D) of Tin Selenide thin films deposited at different substrate
temperature (a)550K, (b) 575K and (c)600K
On the other hand the roughness of the deposited SnSe thin films was measured
using AFM technique. The corresponding values of surface roughness which were
calculated are shown in Table 3.5 respectively. Root Mean Square (RMS) surface roughness
defined as the standard deviation of the surface height profile from the average height is the
most commonly reported measurement of the surface roughness [15]. The surface roughness
is unavoidable since the grains are grown with different thicknesses. It is observed from
table 3.4 that the root mean- square (RMS) surface roughness increases from 11(550K)to
19.20 nm (575K) with increasing substrate temperature and at TS 600K slightly decrease to
17.89nm.
Substrate Temperature(K)
Average Roughness(nm)
550K 11
575K 19.20
600K 17.89
Table 3.4 Average Surface Roughness
84
3.4.1.4 Optical properties of SnSe thin films
The optical bandgap has been calculated by using equation (3.6). Plot of (αhν) 2
versus the photon energy (hν), for films deposited at substrate temperature Ts = 575K is
shown in Figure 3.14. The linearity of the above plots near the absorption edge indicates that
the material is of the direct bandgap. Hence, the energy-axis intercept of the linear part
yields the energy bandgap of SnSe thin films [8]. The energy bandgap of deposited film was
1.18 eV.
Fig. 3.14 Plot of (αhν)2 versus hν for SnSe thin films thermally deposited at 575K substrate
temperature.
The plot of ln (αhν) vs. ln (hν-Eg) gives straight line, the slope gives the value of n
which is found to be 0.80 (close to 0.5) indicating that the transition is direct [8-12].
85
Fig. 3.15 Plot of ln(αhν) versus ln(hν-Eg) of SnSe film deposited at 575K substrate
temperature.
The values of the Urbach‟s energy (Eu) was calculated by using relation
uE
Eexp0
(3.10)
where EU is the Urbach energy, αo is a constant. Thus, a plot of ln(α) versus hν should be
linear whose slope gives Urbach energy. The Urbach plots of the films are shown in Figure
3.16. Urbach energy was calculated from the reciprocal gradient of the linear portion of
these Eu energy values change inversely with the optical band gap.
86
Fig. 3.16 Plot of h versus ln(α) of SnSe thin films at 575K Substrate temperature
The Extinction coefficient K of the thin films is also calculated using the formula:
/)4(a
(3.11)
Fig. 3.17 Variation of extinction coefficient with incident photon energy
87
The variation of extinction coefficient as a function of photon energy is shown in
Figure 3.17. The rise and fall of the extinction coefficient in the forbidden gap region is
directly related to the absorption of light. In the case of polycrystalline films, extra
absorption of light occurs at the grain boundaries. This leads to the non-zero value of
extinction coefficient (K) for photon energies smaller than the fundamental absorption edge.
3.4.2 Electrical Characterization
Electrical resistivity measurements of a semiconducting SnSe thin films deposited at
550K, 575K and 600K wer performed by Four-probe method in the temperature range from
320K to 450K. Figure 3.18 shows the variation of electrical resistivity with change in
temperature. The decrease in resistivity as temperature increase shows the semiconducting
nature of the film [16]
Fig. 3.18 Plots of the resisivity as a function of temperature (T) for SnSe thin films
deposited at (a) Ts =550K (b) Ts =575 and (c) Ts =600K
88
Temperature(K) Resistivity (ohm-cm) Ts =550K
Resistivity (ohm-cm) Ts =575K
Resistivity (ohm-cm) Ts =600K
300 4250.03322 56.33223 291.04983
310 3962.59136 50.07309 253.49502
320 3572.49169 45.90033 227.41528
330 3510.89701 41.72757 208.63787
340 3449.30233 38.59801 189.86047
350 3079.73422 35.46844 170.03987
360 2853.88704 32.33887 154.39203
370 2730.69767 29.2093 143.96013
380 2361.12957 27.12292 125.18272
390 1950.49834 23.99336 105.36213
400 1827.30897 21.90698 93.88704
410 1560.39867 19.8206 82.41196
Table 3.5 Resistivity values of SnSe thin films grown at different substrate temperatures at
(a) Ts =550K (b) Ts =575 and, (c) Ts =600K
The electrical resistivity of a polycrystalline thin film sample is a complex
phenomenon, involving charge-carriers transport through both the “bulk-like” part of the
semiconductor crystals and through the inter-crystalline (grain) boundaries. In the literature
[14] the temperature dependence of the semiconductor material‟s resistivity is expressed by
the equation (3.12) as shown below
be
b
e kTk
Eglog
2log (3.12)
where ρ is the resistivity (Ω cm), Eg is the energy bandgap, kb is the Boltzmann constant and
T is the corresponding temperature. The variation of the electrical resistivity of SnSe films
with substrate temperature (Ts) is shown in Figure 3.18. It is observed that the resistivity
decreases with increase in substrate temperature up to 575K. This shows that the crystallites
do not grow sufficiently large at low substrate temperature, the intercrystalline regions are
89
wide offering a high resistance to the movement of charge carriers. For substrate
temperature 575K, the formation of fewer nucleation centre results in larger crystallite size
which ultimately decrease the intercrystalline barrier size. The charge carriers therefore
cross narrow intercrystalline barriers and this may be responsible for the decrease in
resistivity. The SnSe films grown above 575K are observed to have a higher resistivity result
due to change in composition of SnSe thin films.
The corresponding value of ρ is given by
S
WG7
0 (3.13)
where ρ0 = (V/I)*2πS and G7(W/S) = (2S/W) loge2 is the geometrical correction factor.
SI
V 20
and 2log
27 e
W
SG
Fig. 3.19 Plot of ln (σ) versus 1000/T for a SnSe thin film deposited at different substrate
temperatures(a)550K (b)575K and (c)600K.
90
Clearly, a plot of ln(ζ) versus 1000/T is a straight line, the slope of which provides
the activation energy (Ea). So, the temperature dependence of resistivity of SnSe thin films
have been studied in the temperature range from 320K to 440K.The films deposited at 575K
substrate temperature show high conductivity as compare to other substrate temperature.
91
References -
[1] V. P. Bhatt, K. Gireesan, C. F. Desai, Cryst. Res. Technol., 24, 1989, 187.
[2] D. T. Quan, J. Phys. Status Solid, A 86 , 1984, 421.
[3] G. H. Chandra, J. Naveen Kumar, N. Madusudhana Rao, S. Uthanna, J. Cryst.
Growth 68, 2007, 306.
[4] R. Teghil, A. Santagata, V. Marotta, S. Orlando, G. Pizzella, A. Giardini-Guidoni, A.
Mele, Appl. Surf. Sci. ,90, 1995, 505.
[5] Powder Diffraction File, Joint Committee on Powder Diffraction Standards(JCPDS),
ASTM , 1998, (Card no. 32-1392).
[6] B. D. Cullity, “ Elements of X-ray Diffraction”, Addison-Wesley, Reading, M.A,
1972.
[7] I. Lefebvre, M. A. Szymanski, J. Olivier-Fourcade, J. C. Jumas, J.Phys. Rev.,
B 58, 1998, 1896.
[8] S. Prabahar, M. Dhanam, J. Cryst. Growth, 41, 2005, 285.
[9] D. P. Padiyan, A. Marikani, K. R. Murali, J.Cryst.Res.Technol. 35, 2000, 949.
[10] Z. Zainal, N. Saravanan, K. Anuar, M. Z. Hussein,W. M. M. Yunus, Mater. Sci. Eng.
, B 107, 2004, 181.
[11] V. E. Drozd, I. O. Nikiforova, V. B. Bogevolnov, A. M. Yafyasov, E. O. Filatova,
and D. Papazoglou, J. Phys.Rev. D 42, 2009, 125-306.
92
[12] N. D. Boscher, C. J. Carmalt, R. G. Palgrave, I. P. Parkin, Thin Solid Films 516,
2008, 4750.
[13] N.Kumar V. Sharma, N. Padha, N. M. Shah, M. S. Desai, C. J. Panchal and I. Yu.
Protsenko, J.Cryst. Res. Technol. 45,2010, 53 – 58.
[14] A. J. Moulson, Electroceramics Wiley, USA, 1990, 26.
[15] K. Chung, D. Wamwangi, M. Woda, M. Wuttig, and W. Bensch, J. Appl. Phys. 103,
2008, 083523.
[16] S. M. Sze, “Physics of Semiconductor Devices”, John Wiley & Sons 2nd ed.
1981, 30.
93
Chapter 4 Impact of the Film Thickness on the
Crystallite Size and Properties of SnSe Thin
Films. _________________________________________________________________________________
4.2 Introduction
In this Chapter, investigations have been made for the study of usefulness of semiconductor
layer thickness of the films deposited at room temperature on the structural, morphological, optical
and electrical properties of the SnSe thin films. They were investigated on the basis of XRD spectra,
AFM images, UV-Visible Spectrophotometer and Four Probe method respectively. It has been
observed that crystalline quality, electrical and optical properties of the films depend on the film
thickness and parameter improved with increasing film thickness. The conductivity of the films
increased with thickness [1]. This has been attributed to the fact that the films with thickness less
than 500 nm contain severe misfit strain and the structure improved with the film thickness [2].
Further an attempt has also been made to optimize thickness of the films prepared at the
substrate temperature of 573K. Consequently, a set of films with varying thickness ranging from
150-500 nm was prepared at the substrate temperature of 573K. These films were undertaken for
structural as well as electrical characterizations. The XRD analysis confirmed that with this increase
the crystallite size improve from 34 to 43 nm.
94
4.2 Preparation Details of Tin Selenide Thin Film
A set of SnSe thin films with varying thicknesses were deposited on a glass substrate at
temperatures (300K and 573K) by using thermal evaporation method. To study the influence of
thickness of SnSe films on their behaviour, they were readilys undertaken for structural,
morphological, optical and electrical studies. The rate of deposition was kept 0.3 nm/s and typical
thicknesses of the films were 150 nm, 200 nm, 250 nm and 500 nm. The detail of the experimental
setup has already been discussed in the chapter 3.
4.3 Results and Discussion
4.3.1 X-ray diffractogram of SnSe thin films
The X-ray spectrum exhibits polycrystalline nature of SnSe thin films with sharp peaks at 2θ
values and corresponding d (inter planar distance) values along with relative intensities which
matches well with the data cards of Joint Council of Powder Diffraction Standards (JCPDS). The
thickness variation factor has demonstrated a pronounced effect on the X-ray diffraction spectra as
shown in the Fig.4.1 (a) to (d). A comparison between the spectra of these films in the given figures
shows that there is an improvement in the crystallization and more orientations in case of films
having higher thickness. The prominent Bragg reflection is occurring at or around 2θ = 30°
corresponding to (111) diffraction planes, along with three other very weak diffraction peaks of
(011), (311), (411) at the film thickness of 500 nm confirmed the polycrystalline nature of the films.
A similar preferred orientation of (111) plane in SnSe film was observed by Bhatt et. al. [3]
and Dang Tran Quan et. al. [4] in the thin films grown by the Vacuum Evaporation Technique and
by Singh
95
20 25 30 35 40 45 50 55
0
100
200
300
400
500
(a)
Inte
nsis
ty(coun
ts/s
ec)
0
100
200
300
400
500
(111)
(111)
(311)
(011)
(311)
(111)
(111)
(011)
0
100
200
300
400
(411)
(011)
(d)
(c)
(b)
0
100
200
300
400
500
Fig. 4.1 XRD Spectra of SnSe thin films of different thicknesses (a) 150nm (b) 200nm (c) 250nm and
(d) 500nm deposited at room temperature.
& Bedi et. al. [5] prepared by Hot Wall Epitaxy method. Whereas Teghil et. al. [6] observed
preferred orientation in the (011) & (200) crystallographic planes in the SnSe thin film prepared by
Laser Ablation Method and John et. al [7] repeated (400) plane for films grown by Reactive
Evaporation. The various preferred orientations reported for SnSe films indicate that the deposition
96
technique plays an important role for the orientation of SnSe thin films. The XRD data is also found
useful for establishing the Inter-planar spacing dhkl, Crystallite size (D), Strain (ε) and Dislocation
density ( ) which were calculated for (111) plane by using the equations (3.1, 3.2 and 3.4) already
discussed in chapter (3) and are given in Table 4.1.
Table 4.1 Structural parameters of SnSe films deposited on glass substrate at room temperature with
preferred orientation along (111) planes.
The Strain (η), Particle size (D) and Dislocation density ( ) are also calculated by using the
Williamson and Smallman relation [8]
sincos
D (3.3)
where λ denotes the wavelength of the radiation used (0.15418nm), β
the full width half
maximum(FWHM) and θ the angle of diffraction.
Thickness (nm)
Interplanar Spacing
(d ) (A0)
FWHM (Degrees)
Grain Size(D) (nm)
Dislocation density (1010 line\m2)
150 2.934 0.56 15.36 0.00423
200 2.934 0.51 16.87 0.00351
250 2.933 0.45 19.12 0.00273
500 2.924 0.33 25.69 0.00151
97
0.256 0.258 0.260 0.262 0.264 0.266 0.268 0.2700.3226
0.3228
0.3230
0.3232
0.3234
0.3236
0.3238 SnSe thin Film
Thickness=500nm
(a)
A 0.34717
B -0.09113
cos
sin
(a)
0.256 0.258 0.260 0.262 0.264 0.266 0.268 0.2700.5392
0.5394
0.5396
0.5398
0.5400
0.5402
0.5404
0.5406
0.5408
0.5410
0.5412
SnSe thin film
thickness=150nm
(b)
A 0.58045
B -0.15275
cos
sin
(b)
Fig. 4.2 (a) & (b) A plot of β cosθ versus sinθ for the SnSe thin films deposited at different
thicknesses (a) = 500nm and ( b) = 150nm.
98
According to equation (1) the slope of graph between βcosθ and sinθ provides strain (η)
while particle size (D) is determined from the intercept as shown in Fig. 4.2(a) & (b). The value of
grain size (D) and dislocation density ( ) obtained from this method are same as that obtained using
Scherer‟s Method. The values of Strain (η), Grain size (D), and Dislocation density ( ) calculated
are given in Table 4.2 respectively.
The influence of thickness variation on the grain size and strain on the thermally evaporated
SnSe thin films are shown in Fig. 4.3. It is observed that the grain size increases from 15.36 to 25.69
nm, while strain increases 1.6 times with the increase of film thickness from 150 to 500nm. It was
further observed that dislocation density and FWHM decreases with increase in thickness of the film.
Table 4.2 Micro-structural parameters of SnSe films deposited on glass substrate with preferred
orientation along (111) plane.
Thickness
(nm)
Average internal
Strain
FWHM
(Degrees)
Average Grain
Size (nm)
Dislocation
density
(1010
line\m2)
150 0.09113 0.56 15.20 0.00432
200 0.12282 0.51 16.70 0.00358
250 1.13911 0.45 18.92 0.00279
500 0.15275 0.33 25.45 0.00154
99
Fig. 4.3 Plots show the effect of thickness on the grain size and strain.
4.3.2 Atomic Force Microscopy (AFM) Studies
The surface morphology of the deposited SnSe thin films was characterized using Atomic
Force Microscopy (AFM) technique. Fig. 4.4(a) to (c) shows the 2-D as well as 3-D representation of
SnSe thin films deposited at the film thickness of 150 nm, 300 nm and 500 nm. The SnSe thin films
prepared at lower thicknesses indicate that the growth of small grains has been distributed across the
surface of the substrate. The size of the grains is rather different from each other indicating irregular
growth rate of grains. The granules are made up of different sizes. However, the sizes of the grains
have been noticed to increase with the increase in the thickness. The films deposited at higher
thickness of 500nm shows compact morphology. Based on the AFM image (Fig. 4.4(c)), the grain
density is reduced indicating the smaller grains agglomerate together to form larger grains of SnSe.
The Root Mean Square (RMS) surface roughness is defined as the standard deviation of the
surface height profile from the average height. It is the most commonly reported measurement of the
100
surface roughness [9-11]. The surface roughness is unavoidable since the grains are grown with
different sizes. RMS roughness of the deposited SnSe thin films were measured using AFM technique
for the film thicknesses of 150, 300 and 500 nm. The corresponding values of surface roughness are
shown in Table 4.3. It has been observed that thickness plays a vital role on the behaviour of SnSe thin
films. The AFM images related to the different thicknesses of SnSe films are shown in Fig. 4.4 (a) to
(c).
Table 4 .3. Root Mean Square Roughness at different thicknesses of SnSe thin films
(a)150nm
Thickness
(nm)
RMS Roughness
(nm)
150 15
300 30.18
500 72.19
101
(b)300nm
(c)500nm
Fig. 4.4 (a) to (c) AFM image of SnSe thin film of thickness150 nm,300nm and 500nm
deposited at glass substrate at room temperature.
4.3.3 Optical Studies
The optical transmission spectra of the studied SnSe thin films with varying thickness are shown
in Fig. 4.5 respectively. The higher transmittance indicates a fairly smooth surface and relatively good
102
homogeneity of the thinner films, their results are consistent with the results of AFM measurements
[12]. The appearance of the maxima and minima, were due to the interference effect from the
substrate-film and the film-air interferences. The transmittance of the films was found to decrease with
increase in thickness. This is because of the reason that in case of thicker films more atoms are present
in the film, thus, make more states available for the photons for getting absorbed [13]. However, with
the increase of film thickness, the scattering of the light also increase thus causing a loss to the
coherence between the primary light beam and the beam reflected between the film boundaries and
results in the disappearance of the interference as well as reduction of the transmittance [14].
400 500 600 700 800 900 1000 1100 1200-10
0
10
20
30
40
50
60
70
80
90
100
SnSe Thin films
(a)
(c)(b)
(d)
Tra
nsm
itta
nce,T
(%)
Wavelength,(nm)
Fig. 4.5 Plots of the transmission coefficient versus wavelength of incident photons of SnSe films
having thicknesses of (a 150 nm (b)200 nm (c)250 nm and (d)500 nm.
The absorption coefficients (α) of these films at different energies (h) are shown in Fig. 4.6. A
close examination of Fig. 4.6 revealed that the thick films have lower α values in the band to band
absorption region. This effect may be explained by proposing that thicker films have bigger crystallites
103
(grains), so they are closer to bulk crystalline SnSe, however bigger grain sizes results in larger
unfilled inter-granular volume so the absorption per unit thickness is reduced.
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4-20000
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
220000
240000
260000
280000
300000
320000
340000
360000
SnSe Thin films
(d)
(c)
(b)
(a)
(c
m)-1
h (eV)
Fig. 4.6 Absorption coefficient versus photon energy plots for the SnSe film thicknesses (a) 150nm
(b) 200nm((c) 250nm and (d) 500nm.
The optical bandgap Eg is estimated by using the following relation [14]
g
mEhh (3.6)
where „A‟ denotes a characteristic parameter independent of photon energy, „h‟ the incident photon
energy and „m‟ a constant which depends on the nature of the transition between the top of the
valance band and bottom of conduction band. The lowest bandgap energy in semiconducting
materials is referred to as the fundamental absorption edge nature of interband transition and is
characterized by m. For the allowed indirect transition, m=1/2 and for the allowed direct transition
we have m=2. By plotting (αh) m
versus the incident photon energy (h) and extrapolating the
straight-line portion of the plots towards low energies, the optical band gap can be obtained as shown
in Fig. 4.7
104
1.0 1.2 1.4 1.6 1.8 2.0 2.2
0.00E+000
1.00E+011
2.00E+011
3.00E+011
4.00E+011
5.00E+011 SnSe Thin film
Thickness=150nm
Band Gap=1.71eV
(a)
(h)2
(h)
(a)
1.0 1.2 1.4 1.6 1.8 2.0 2.2
0.00E+000
5.00E+010
1.00E+011
1.50E+011
2.00E+011
2.50E+011
3.00E+011
SnSe thin film
Thickness=250nm
Band Gap=1.59eV
(b)
(h)2
(h)
(b)
105
1.0 1.2 1.4 1.6 1.8 2.0 2.2
-2.00E+010
0.00E+000
2.00E+010
4.00E+010
6.00E+010
8.00E+010
1.00E+011
1.20E+011
1.40E+011
1.60E+011SnSe Thin film
Thickness=350nm
Band Gap=1.21eV
(h)2
(h)
(c)
1.0 1.2 1.4 1.6 1.8 2.0 2.2
0.00E+000
1.00E+010
2.00E+010
3.00E+010
4.00E+010
5.00E+010
6.00E+010SnSe thin Film
Thickness=500nm
Band Gap=1.13eV
(h)
(h)
(d)
Fig. 4.7 Plots of h versus (h) 2
for films deposited with the thicknesses(a)150 nm(b)250 nm
(c)350 nm and (d)500 nm.
106
These plots indicate that the wider linear regions are observed for the allowed direct transition
(m=2) .
The value of optical bandgap energy for increasing film thickness is found to be decreased in the
range of 1.13-1.71eV. On the basis of experimental results, it is concluded that bandgap of SnSe thin
films alter with increase in thickness. Fig. 4.8 shows the variation of the bandgap for increasing
thickness of thin films.
150 200 250 300 350 400 450 500
1.2
1.3
1.4
1.5
1.6
1.7
1.8
SnSe Thin film
Energ
y b
and g
ap (
eV
)
Thickness (nm)
Fig. 4.8 Plot of Energy Bandgap versus Thickness for the SnSe thin films
Taking the plot of ln(αhν) versus ln(hν-Eg), we get straight line graphs as shown in Fig. 4.9,
the slope of these graphs gives η ranges from 0.81 to 0.73 respectively and is close to 0.5 indicating
that the transition is direct [13].
107
-2.6 -2.4 -2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.011.0
11.2
11.4
11.6
11.8
12.0
12.2SnSe Thin Film
(n=0.81)
ln(
h
)
ln(h-Eg)
(a)
-2.4 -2.2 -2.0 -1.8 -1.6
10.7
10.8
10.9
11.0
11.1
11.2
11.3
11.4(b)
(n=0.75)
ln(
h)
ln(h-Eg)
(b)
108
-2.2 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6
11.4
11.6
11.8
12.0
12.2
12.4
12.6
(c)
(n=0.73)
(h)
ln(h-Eg)
Fig. 4.9 Variation of ln(αhν) versus ln(hν-Eg) for different thicknesses (a) 150nm (c)250nm and
(c)500nm, SnSe thin films.
The extinction coefficient K of the thin films is also calculated using the formula
/)4(a (3.11)
Variations of extinction coefficient as a function of photon energy are shown in Fig.4.10. The variation
of the extinction coefficient in the forbidden gap region is directly related to the absorption of light. In
the case of polycrystalline films, extra absorption of light occurs at the grain boundaries. This leads to
the non-zero value of K for photon energies smaller than the fundamental absorption edge.
109
1.0 1.2 1.4 1.6
0
1000000
2000000
3000000
4000000
5000000
6000000
7000000
8000000
9000000
10000000
(2500A0)
(1500A0)
(2000A0)
(5000A0)
Extinction
coe
ffic
ien
t(k)
Photon Energy(h)
Fig. 4.10 Variation of extinction coefficient with incident photon energy
The values of the Urbach‟s energy (Eu) were calculated by using relation [14]:-
uE
Eexp0
(3.10)
Where EU is the Urbach‟s energy, which corresponds to width of the band tail, αo is a constant. Thus,
a plot of ln(α) versus hν should be linear whose slope gives Urbach energy. The Urbach plots of the
films are shown in Fig. 4.11. Urbach‟s energy was calculated from the reciprocal gradient of the
linear portion of these Eu energy values change inversely with the optical band gap. The Urbach‟s
energy is found to increase with the increase in film thickness. This suggests that the crystalline
nature of the thin films increase with the increase in the film thickness and further supports the
crystalline nature as implied from the enhanced peak intensity of the XRD peaks for large film
thickness.
110
1000 1500 2000 2500 3000 3500 4000 4500 50000.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.20 SnSe Thin Film
Urb
ach E
nerg
y
Thickness(A0)
Fig. 4.11 Variation of urbach’s energy with photon energy for different thicknesses.
4.3.4 Electrical Studies
One of the most common methods of measuring a material‟s surface resistivity is by using
either the Two or the Four-point probe methods [15]. This method uses probes aligned linearly or in
a square pattern that make contacts with the surface of the test material. Many conventional methods
used for measuring resistivity of semiconductor devices are not satisfactory because metal-
semiconductor contacts are rectifying at several cases and also there is generally minority carriers
injection by one of the current carrying contacts. An excess concentration of minority carriers will
affect the potential of other contacts and modulate the resistance of the material.
The method described here overcomes the difficulties mentioned above and also offers
several other advantages. It permits measurements of resistivity in samples having a wide variety of
shapes, including the resistivity of small volumes within bigger pieces of semiconductor. This
method of measurement is applicable both in single crystal as well as thin film semiconductors.
111
Temperature dependent electrical resistivity of semiconducting SnSe thin films measured
by Four-probe method in the temperature range of (320K to 450K) has been illustrated in Fig. 4.12.
The decrease in resistivity with increase in temperature shows the semiconducting nature of the film
[16].
300 320 340 360 380 400 420 440 460
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
SnSe Thin film
Thickness=500nm
Resis
ivity
(ohm
-cm
)
Temperature(K)
Fig. 4.12 Temperature depended Electrical Resistivity measurement of SnSe thin film using four
probe method in the temperature range 325K to 450K.
In general, the electrical resistivity of a polycrystalline thin film sample is a complex
phenomenon, involving charge-carriers transport through both the “bulk-like” part of the
semiconductor crystals and through the “inter-crystalline” grain boundaries. In the literature [16], the
temperature dependence of the semiconductor materials resistivity is expressed by the equation
(3.11)
be
b
e kTk
Eglog
2log
(3.11)
112
Where ρ is the resistivity (Ω cm), Eg is the energy band gap, kb is the Boltzmann constant and T is
the corresponding temperature.
The corresponding value of ρ is given by
S
WG7
0 (3.12)
Where SI
V 2
and 2log
27 e
W
S
S
WG
is the geometrical correction
factor.
2.2 2.4 2.6 2.8 3.0 3.2
2.8
2.9
3.0
3.1
3.2
3.3
3.4
SnSe Thin Film
Thickness=500nm
Activation Energy=0.49224
log(
)
Temperature K (1000/T)
Fig. 4.13 Plot of ln() versus 1000/T for a SnSe thin film of thickness of 500nm grown at room
temperature.
Fig.4.13 represents a plot between ln() versus 1000/T in accordance to the equation 3.11. The
slope of plots provides activation energy of value 0.49eV. The P-type behaviour of the deposited
SnSe thin films was also verified for all thicknesses of the SnSe thin films by Hot–Probe Method.
113
4.4 Study of the effect of varying film thickness of polycrystalline SnSe
thin films deposited at 575K substrate temperature
In order to get the optimum combination of thickness and substrate temperature of SnSe thin
films deposition, these films with thickness ranging from150nm to 500nm were also deposited at Ts
= 575K.The prepared samples were then undertaken for structural as well as electrical
characterization.
4.4.1 Structural characterization
The structural properties of the prepared samples was carried by XRD technique already
discussed in this chapter. The XRD data was then analyzed for the determination of crystallite size,
FWHM and dislocation density whose values have been given in the Table 4.4.
Table 4.4 Structural parameters of SnSe thin films of different thicknesses deposited at 575K
substrate temperatures.
The Strain(η), Particle size (D) and Dislocation density (δ) are also calculated using the
Williamson and Smallman relation. The strain (η) was calculated from the slope of βcosθ versus sinθ
plot of thin films of different thicknesses deposited at 575K substrate temperature.
Thickness Thin
films
FWHM(2θ) Grain Size(D)
(nm)
Dislocation density (δ)
(1010
line\m2)
200nm 0.25 34.48 0.00084
300nm 0.23 37.47 0.00071
400nm 0.21 41.03 0.00059
500nm 1 0.21 41.04 0.00059
2 0.20 43.03 0.00054
114
Thickness
Thin films
FWHM(2θ) Grain Size
(nm)
Strain Dislocation
density(δ)
(1010
line\m2)
200nm 0.25 33.98 0.06871 0.0294
300nm 0.23 36.92 0.06377 0.0270
400nm 0.21 40.42 0.05821 0.0247
500nm 20 42.44 0.05553 .0235
Table 4.5 Micro-structural parameters of SnSe thin films deposited on the glass substrate at
different substrate temperatures.
From the above table, a significant increase in the grain size observed while there occurs a
gradual decrease in dislocation density and FWHM.
4.4.2 Electrical Properties of SnSe thin Film
Temperature depended electrical resistivity measurement of semiconducting SnSe thin film
was done by Four-probe method in the temperature ranges from 320K up to 450K. The decrease in
resistivity as temperature increase shows the semiconducting nature of the film [16].The temperature
dependence conductivity of the deposited SnSe thin films has been analyzed using the following
equation
= 0 exp (-Ea/kT) (4.9)
where „k‟ represents the Boltzmann constant and „0‟ the pre-exponential factor.
115
2.2 2.4 2.6 2.8 3.0 3.2 3.4-8.5
-8.0
-7.5
-7.0
-6.5
-6.0
-5.5
-5.0
-4.5
-4.0
-3.5
-3.0
-2.5
(500nm)
(300nm)
(200nm)
ln(o
hm
-1cm
-1)
103/T(K
-1)
Fig. 4.14 Plot of ln(σ) versus 1000/T for a SnSe thin film of different thickness deposited at 575K
substrate temperature
It is clear from the Fig. 4.14 that the plots of ln(ζ) versus 1000/T for a SnSe thin film of
different thickness deposited at 575K substrate temperature are straight lines, indicating that the
conduction occurs in these films through an activated process having single activation energy. The
value of conductivity is found to increase with increasing film thickness.
116
References -
[1]. Jae-Min Myoun, Wook- Hi Yoon, Dong Hi Lee, Iigu Yun, Sang- Hyuck
BAE, Sang- Yeol Lee, Jpn. J.Appl. Phys., 41, 2002, 28.
[2] Shadia J. Ikhmayies, Riyad N. Ahmad Bitar, A. J. Appl.. Sc., 5, 2008, 1141.
[3] V.P. Bhatt, K. Girreesan, C.F. Desai, Crystal Res. Technol ., 24, 1989, 187.
[4] D. T. Quan, Phys. Status solid, 86, 1984, 421.
[5] J. P. Singh, R. K. Bedi, J Applied Physics, 68, 1990, 2776.
[6] R. Teghil, A. Santagata, V. Marotta, S. Orlando, G. Pizzella, A. Giardini-
Guidoni, A. Mele, J. Appl. Surf. Sci., 90, 1995, 505.
[7] K J John, B. Pardeep, E. Mathal, J. Matter Sci., 29, 1994, 1581.
[8] I. Lefebvre, M. A. Szymanski, J. Olivier-Fourcade, J. C. Jumas, J. Phys. Rev.
,58, 1998, 1896.
[9] N. Tigau, V. Ciupina, G. Porodan, G. I. Rusu, E. Vasile, J.Cryst. Growth, 269, 2004,
392.
[10] N. Tigau, V. Ciupina, G. Porodan, G. I. Rusu, C. Gheorghies, E. Vasile,
J.Optoelectron.Adv. Mater, 6, 2004, 211.
[11] T. Hall Jiang, Morin, “Qualitative analysis of electrodeposited tin films
morphologies by atomic force microscopy”, Thin Solid Films, 417, 2005,
76.
117
[12] Opt.1Y. Gao, Y. Masuda, K. Koumoto, J. Korean Ceramic Society, 40,
2003, 213.
[13] N. Kumar, V. Sharma, N. Padha, N. M. Shah, M. S. Desai, C. J. Panchal, I.
Yu. Protsenko, Cryst. Res. Technol. 45, 2010, 53.
[14] A. J. Moulson, Electroceramics, Wiley, USA, 1990, 26.
[15] M.Y. Nadeem, W. Ahmed, Turk .J. Phys. 24, 2000, 651.
[16] S. M. Sze, “Physics of Semiconductor Devices”, John Wiley & Sons 2nd ed.
198, 30.
118
Chapter 5
Metal-Interface of SnSe Polycrystalline Thin Films for Schottky Barriers and Ohmic Contacts
_______________________________
5.1 Introduction
The current transport phenomenon across the Schottky Barrier Diode has been
widely studied so far and various attempts have been made to understand its behaviour. But
the complete description of the conduction mechanism is still a challenging problem.
Several factors such as the surface state charges, interface states, defects & dislocations on
the semiconductor surface, image force lowering, barrier inhomogenities, field emission and
presence of interfacial layer between metal and the semiconductor have been found
contributing in the current transport behaviour.
In this chapter, an attempt has been made to identify the metals which make
reasonably acceptable ohmic contacts to P-type SnSe films. The various metals tried for this
purpose includes Al, In and Ag. Out of which Al contact was found to be the most suitable
for working at comparatively higher deposition temperatures. Further, reasonably good SnSe
Schottky Barrier diodes were fabricated with Ag metal and the corresponding Schottky
structure of Ag/p-SnSe was formed on the Al coated glass substrates. Circular as well as
119
Square shaped Schottky diodes of the different areas (6x10-3
cm2, 9x10
-3cm
2 and 9x10
-2cm
2)
have also been fabricated. These diodes were tested for Current-Voltage (I-V) behaviour at
room temperature out of which only those display good schottky behaviour were selected for
the further analysis. The various parameters of the diode such as η, фb,and Rs for the
undertaken diodes were extracted from the forward I-V measurements while the reverse
breakdown phenomenon of these diodes was also investigated. Transport analysis has been
made and results were drawn from the forward and reverse biased current voltage (I-V) as
well as room temperature capacitance voltage (C-V) data. Emphasis of the study has been to
find out the effect of change in the diode areas, on the characteristics of Ag/p-SnSe Schottky
diodes.
5.2 Experimental Details
5.2.1 Ohmic contact formation
In general, p-type semiconductors usually make non-rectifying junctions with metals
whose work functions are greater than that of the semiconductors. However, due to the
complex band structure of p-SnSe thin films it is not always the case experimentally. This is
due to the self-compensating nature of p-type semiconductor by which deep donors are
created. Thus, p-SnSe films can have non-rectifying structures with both low and high work
function metals. For the case of low work function metals, semiconducting thin films make
ohmic contacts with a negative space charge region, whereas for the high work function
metals the ohmic contact is created by ionized deep donors [1]. The Current-Voltage (I-V)
measurements in this study have shown that p-type SnSe thin films make the best rectifying
contact with Ag metal junctions while the In and Al metals show ohmic behaviour.
120
Fig. 5.1 and Fig. 5.2 represent the plots of current versus voltage in forward and
reverse biased conditions. The current changes linearly with increasing voltage. Thus, it can
be concluded that Indium (In) and Aluminium (Al) make ohmic contacts to p-SnSe thin
films.
(a)
(b)
Fig. 5.1 (a) Pictorial representation of Indium (In) ohmic contacts to SnSe semiconductor
layer and (b) the current-voltage characteristic of In/p-SnSe/In structure.
121
Fig. 5.2 (a) Pictorial representation of (Al) ohmic contacts to SnSe semiconductor layer and
(b) the current-voltage characteristic of Al/p-SnSe/Al structure.
5.2.2 Schottky Diode Fabrication
SnSe thin films of the thickness of about 500nm were deposited at 575K substrate
temperature by thermal evaporation technique on Aluminium (Al) coated glass plate which
serves as back ohmic contact. The thickness of Al layer was kept 150nm which was
thermally deposited on organically cleaned Glass substrate. For the formation of Schottky
Diodes (rectifying contact), a thin layer of Silver (Ag) of thickness 200nm was deposited on
the SnSe films. Different areas of Ag/p-SnSe Schottky Diodes (6x10-3
cm2, 9x10
-3cm
2 and
122
9x10-2
cm2) were obtained by using suitable masks. The room temperature current-voltage (I-
V) as well as capacitance-voltage (C-V) measurements of the Ag/p-SnSe Schottky Diodes
was measured using a computer interfaced setup comprising a programmable Keithley
Source Meter (model-2400) and Precision programmable LCR meter (Aglient make 4284A).
Interfacing of I-V and C-V measurement equipments were achieved by using LabVIEW
software by National Instruments (U.S.A).
Fig. 5.3 (a) Structure of the Ag/p-SnSe schottky diode fabricated on Aluminum coated
glass substrates (b) animated elevated view and (c) top images of the different
areas of Ag/p-SnSe Schottky diodes.
123
5.3 Results and discussion
5.3.1 Current-Voltage (I-V) Characteristics
The current-voltage (I-V) and capacitance-voltage (C-V) measurements of the
fabricated Ag/p-SnSe schottky diodes measured at room temperature were undertaken for
further analysis. The device is forward biased when the Al side is made positive with respect
to Ag electrode. The forward current increases exponentially with increase in voltage
whereas reverse current increases with voltage slowly up to the breakdown voltage after
which a rapid change in current with voltage has been observed which causes breakdown
phenomena at the schottky interface. A typical current-voltage characteristic of Ag/p-SnSe
structure is shown Fig. 5.4.
Fig. 5.4 The Current-Voltage characteristics of Ag/p-SnSe structure.
124
The wide non-linear forward I-V behaviour of Ag/p-SnSe Schottky Diode is due to
high series resistance associated with the diode [14].
The behaviour of real Schottky diode can be modeled by equivalent electrical circuit
as shown in Fig. 5.5. A series resistance Rs is associated with the semiconductor layer and
the back ohmic contact Gp is the parallel conductance, which may account for leakage
current. They are both independent for the applied voltage drop (Vd) across the junction and
is usually given by the following equation
VGkT
qVII ps
1exp
(5.1)
Fig. 5.5 Equivalent electrical circuit of Schottky Diode.
The effect of parallel conductance (Gp) is more important for diode with high barrier
height on the reverse bias characteristics. Moreover, Werner [2] Showed that correction of
the forward current I for the shunt current does not influence the determination of the
different parameters of diodes with the Schottky barrier as high as 0.830eV. Therefore Gp
will be neglected.
125
The four mechanisms for carrier transport over Schottky barriers are thermionic
emission, carrier tunneling, carrier recombination and generation in the depletion region,
which are due to the minority-carrier injection. In these mechanisms, the dominant modes of
carrier flow in m-s contacts are thermionic emission and carrier tunneling [3]. In Ag/p-SnSe
Schottky diodes, the experimental data has been considered on the basis of thermionic
emission (TE) theory. According to which, the current flow over a uniform metal–
semiconductor interface at a forward bias V 3kT/q can be expressed as [4]
1exp
KT
VqII s
(5.2)
Where Is is the saturation current and Js is the saturation current density, defined as
kT
qTA
AI
J boss
exp2**
(5.3)
The quantities, A; is the diode area, A**
; the effective Richardson constant for p-type
SnSe (18 A/cm2) [5], T; the measurement temperature in Kelvin, k; Boltzmann‟s constant
(1.38 ×10-23
J/K), q; the electron charge (1.6 × 10-19
C), V; the forward applied voltage, Φbo
and RS are the zero-bias barrier height and series resistance of the diode respectively. The
ideality factor η has been introduced to describe the deviation of the experimental I-V data
from the ideal TED mechanism using the definition as given below [4].
)/ln( sIId
dV
kT
qn (5.4)
126
The zero bias barrier heights (Φb0) of the Al/p-SnSe Schottky diodes have been
determined using the equation (5.3) by assuming thermionic emission and diffusion model
This work presents the detailed analysis of forward current-voltage characteristics of Ag/p-
SnSe Schottky diodes measured at room temperature. The non-linear least square fitting of
the experimental data has been performed by using equation (5.3) with Microcal Origin
software [6] with Is, η and Rs as adjustable parameters as illustrated in Fig. 5.6 (a) and (b).
The computer programme is initially run by assuming ideality factor (η) as unity and series
resistance (Rs) as zero, to obtain an approximate value of Is that fit the experimental data
well and computer programme is run again to determine the value of Rs, iterations continue
until one finds a set of Is, η and Rs value that fit the experimental data. When the applied
voltage is sufficiently large, [7] the voltage across the diode can be expressed in terms of the
total voltage drop across the diode and the series resistance Rs. This is accounted by
replacing the voltage V by (V-IRs) in equation (5.2), therefore, equation (5.2) becomes
1exp
KT
IRVqII S
s
(5.5)
5.4 Impact of Geometrical Shapes on the I-V Characteristics of Schottky
diodes
Fig. 5.6 (a) and (b) shows the forward I-V characteristics of Ag/p-SnSe Schottky
diode having Square and Circular contact areas. The fitted curves of square and circular
shaped schottky diodes of same area are shown with solid lines. The electrical parameters
such as zero bias barrier height (Φbo) and ideality factor (η) were obtained from the linear
fitting of the forward ln(I) vs V characteristics. The values of η and Φbo were determined
127
from the slope and intercept of the straight line of the ln(I)-V characteristics of the I-V data
have been listed in the Table 5.1.
Fig. 5.6 The forward I-V characteristics of Ag/p-SnSe Schottky diode having square and
circular shapes each of an area =1.6x10-1
cm2 the Curves have been fitted with
the TED equation 5.5 using Is, η and Rs adjustable parameters.
128
Table 5.1 Ideality factor, Barrier heights and Series Resistance obtained from the
undertaken Schottky Diodes.
From the Table 5.1, it has been observed that Ag/p-SnSe Schottky Diodes of circular area
showed better results as compared to Square types as the value of ideality factor and barrier
height for the circular shaped Schottky diodes are comparatively near to reported values of
6.33 and 0.51eV respectively [5]. Thus, the circular junctions showed better results than the
square shaped in submicron region. Further, circular junction would be important when
minimum junction size reduce to submicron range as compared to square junction.
5.5 Influence of Different Areas
Fig. 5.7 shows the forward I-V characteristics of circular shaped Schottky diodes at
three different areas. It has been observed that slopes of the currents increase and shift
toward, higher voltage side with decrease in area i.e. the larger the Schottky contact area
higher the current passing through the sample [6]. The saturation current was obtained by
Diode
Parameter
Square
Diode
Circular
Diode
Is 2.1389E-7 3.3E-6
n 7.72 7.3
Rs 168.82852 570
Фb 0.6840 0.5641
129
extrapolating the linear region of the semilog forward I-V curves to the zero applied voltage
and barrier height (Φbo) values were calculated from the equation (5.3).
Fig. 5.7 Forward current-voltage characteristics of Ag/p-SnSe Schottky diode with different
diode areas.
From the comparative look of the parameters viz, η and Φbo from the Table 5.2, it is
observed that ideality factor increases and barrier height decreases with increase in area of
the schottky diodes. It is attributed to the fact that as device area increases, effects of the
defects on the surface and other factors at the interface would increase and cause deviations
in the current transport behaviour.
130
Diode Parameter Diode area
6x10-3
cm2
Diode area
9x10-3
cm2
Diode area
9x10-2
cm2
Η 3.45 5.021 5.147
Js(A) 0.0013 0.0032 0.0074
Фb(eV) 0.54 0.51 0.49
Rs() 15x103
8x103
198
VBR(V) -1.32 -1.22 -0.55
Table 5.2 Schottky Diode parameters for different areas.
A method used to extract the series resistance Rs of ideal Schottky diodes (i.e η=1)
was first proposed by Norde [15]. In this method, equation (5.3) is used when the Schottky
diode is assumed to be ideal, namely with η = 1 for which Norde has proposed a new
technique based on an auxiliary function
RsIV
ATA
I
q
kTVIVF b
2ln
2),(
2* (5.6)
By plotting F vs V and F vs I, one finds a minimum F(Vo, Io), which is the point of
interest. From the value of F (Vo, Io) and the corresponding current Io, at the minimum, the
barrier height and the series resistance can be obtained
,s
oqR
kTI (5.7)
131
kTqVIVqF ooob ),( (5.8)
The differential conductance of an ideal diode Gd=dI/dVd =qI/kT can be defined for
each point of the I-V curve. It is to be noted that Gd(Io) =dI/dVd = l/Rs at the minimum point
of F(V) The disadvantages of this method are that:
(1) The ideality factor η is assumed to be unity, which is not always true for a real diode;
(2) A single point (V, Io), corresponding to the minimum, is used to calculate the barrier
height. Sato and Yasumura [16] had modified Norde‟s approach for η>1 cases, to extract the
values of η, Фb and Rs from I-V measurements. The approach requires that for a given
Schottky Diode, two experimental I-V measurements conducted at two different
temperatures and the determination of the corresponding minimum to the modified Norde‟s
function. An another method was given by S. Cheung‟ who presented an alternate approach
to determine the value of of η, Фb and Rs from a single I-V measurement, this proposed
technique was applied to characterize Ag/p-SnSe Schottky diodes subjected to various areas
and different temperature ranges. Thus,
2**ln
TA
JRAJV B
(5.9)
Where
kT
q (5.10)
Differentiating equation (5.9) with respect to J and rearranging terms, we obtain
RAJ
Jd
Vd
)(ln
)( (5.11)
132
Thus, a plot of d(V)/d(lnJ) vs J will give RAeff as the slope and η/β as the y axis intercept. To
evaluate Фb, we can define a function H (J)
2**ln)(
TA
JVJH
(5.12)
For equation (5.9) we can deduce
BRAJjH )( (5.13)
Using the η value determined from equation (5.11), a plot of H (J) vs J will also give a
straight line with y-axis intercept equal to ηФB. The slope of this plot provides a second
determination of Rs which can be used to check the consistency of this approach. Thus,
performing two different plots equations (5.11) and (5.13) of the J-V data obtained from one
measurement can determine all the three key diode parameters η, ФB and Rs. We have
applied this proposed procedure to characterize Ag/p-SnSe Schottky diodes of different
areas 6x10-3
cm2,
9x10-3
cm2 and 9x10
-2cm
2.
133
Fig.5.8 Plots of d(V)/d(lnJ) vs. J and H(J) vs. J of Ag/p-SnSe Schottky diodes of different
areas (a) 6x10-3
cm2
, (b) 9x10
-3cm
2 and (c) 9x10
-2cm
2 .
134
Fig. 5.8 shows d(V)/d(lnJ) versus J and H(J) versus J plots for the Schottky diodes
with threes different areas. The parameter extracted from the linear fitting of ln(I) versus V
plots using thermionic emission equation as well as those extracted from the Cheung‟s
method are summarized in Table 5.3.
Table 5.3 Comparison of diodes parameters extracted from the linear fitting of ln(I) versus
V plots using thermionic emission equation and those extracted from the
Cheung’s method.
However, it can be seen that there is relatively small difference in the values of the η,
Фo from the down curvature region of forward bias I-V characteristics and linear portion of
the same characteristics. The reason for this difference can be attributed to the existence of
effects such as the series resistance, bias dependence of barrier heights (bo) according to
voltage drop across the interfacial layer and change of the interface states with bias in the
concave region of I-V plot[17,18]. There is also contribution of the interface states and the
interfacial layers being effective in the downward curvature region to the value of the series
resistance calculated from the data in this region of the forward I-V characteristics.
Diode
Area(cm2
)
Linear Fitting using TE equation Cheung’s Method
Ideality Factor
(η)
Barrier Height
(bo)eV
Ideality Factor
(η)
Barrier Height
(bo)eV
6x10-3
3.45 0.54 3.36 0.58
9x10-3
5.021 0.51 4.03 0.55
9x10-2
5.147 0.49 4.22 0.76
135
Fig. 5.9 Reverse biased I-V Characteristics of Ag/p-SnSe schottky diodes with different
diode areas measured at Room Temperature.
Fig. 5.10 The reverse breakdown voltage VBR (V) versus area (cm2) for Ag/p-SnSe Schottky
diode.
It is observed from Fig. 5.9 that current increases slowly with voltage for higher
interface areas. The knee point is not well defined and the process is known as soft
136
breakdown phenomenon. The breakdown voltage has been defined as a voltage
corresponding to a reverse current of few micro-amperes. Further referring to the same Fig.
5.9, it has been observed that the breakdown voltage (VBR) is hard in small area schottky
diodes (Area=0.006 cm2) in the sense that it has a well defined knee voltage above which the
current increases very rapidly with change in area. Further, it is also observed from the Fig.
5.10 that at room temperature the breakdown voltage (VBR) increases with increase in area
of diode.
5.6 Capacitance -Voltage Characteristics
In order to access the doping concentration and barrier height, C-2
versus VR plots
were obtained from the C–V data. The C–V relationship is applicable to intimate MS
Schottky barriers on uniformly doped materials and can be written as [12]
22
21
ANq
VV
C As
oR
(5.14)
where VR is the reverse bias voltage, Vo is the built-in potential or diffusion potential, which
is usually measured by extrapolating the C−2
-V plot to the V-axis. The zero-bias barrier
height from the C–V measurement is defined by
ndbo VV (5.15)
Where Vd is the voltage axis intercept of the above plot,
d
cn
N
N
q
kTV ln (5.16)
137
is the energy difference between the Fermi level and the bottom of the conduction band edge
in SnSe and
2/3
2
22
h
kTmN e
c
(5.17)
is the effective density of states in the conduction band of SnSe, where me is the effective
mass of SnSe = 0.15 eV [13]
22
2
dc
dv
AKqN
so
A
(5.18)
is the acceptor density of SnSe, Ks = 18 is the dielectric constant of SnSe, εo = 8.85 × 10−14
F
cm–1
is the permittivity of the free space, A = 6 × 10−3
cm2 is the area of the Schottky diode.
Fig. 5.11 Plot of 1/C2versus V of Ag/p-SnSe Schottky Diode at different frequencies.
138
The variation of C-V curves with frequency is illustrated in Fig. 5.11, where 1/C2 is
plotted V at two different frequencies. It is seen that, the magnitude of slope decreases as
measuring frequency decreases. The slope variation suggests a large density of slow traps or
interface states for the Schottky devices.
Also, the acceptor concentration NA and the zero-bias barrier height Φbo at the two
frequencies for the Ag/p-SnSe Schottky diode have been calculated from the experimental
C-2
-V characteristics at two different frequencies of 5Hz and 10Hz. The value of acceptor
concentration is found to be varying from 8.2x1018
to 6.6x1018
while the value of barrier
height is found to be equal to 0.54 eV. The Schottky barrier height deduced from the room
temperature I-V analysis is less than that obtained from the C-V characteristics at the two
different frequencies. This is may be due to the reason that I-V analysis includes both the
image force lowering and dipole lowering effects and is also reduced by the tunneling and
leakage currents. However, the capacitance–voltage measurements can be used to directly
measure the barrier height. Nevertheless, it has to be noted that the measured capacitance
may be considerably influenced by carrier trapping if the lifetime of the trapping levels in
the semiconductor is of the same order as the period of the ac signal applied during the
capacitance measurement.
139
References -
[1] S.M. Sze, “Physics of Semiconductor Devices”, John Wiley and Sons, New
York, 1981.
[2] R. Hackam, P.Harrop, “Electrical properties of nickel-low-doped n-type
gallium arsenide Schottky barrier diodes”, IEEE. Trans. Electron. Devices
ED-19, 1231, 1972.
[3] Zs.J.Harvoth; “Evaluation of Schottky current-voltage characteristics for
thermionic field emission”, Proc. Of Int. Conf., 263, 1996.
[4] J. Martinez-Pastor, A. Segura, J.L. Valdes, A. Chevy, J. Appl. Phys., 62,
1987. 539.
[5] J.H Werner, J. Appl. Phys. ,A 47, , 1991, 291.
[6] M.S. Tyagi, “Introduction to Semiconductor Materials and Devices” John
Wiley & Sons, New York, 1991.
[7] M.S. Tyagi, “Introduction to Semiconductor Materials and Devices” John
Wiley & Sons, New York, 2008.
[8] Origin ® version 6.0; Microcal software, Inc.,Northampton, MA USA.
[9] Soraj Bala, Dissertation, NIT Hamirpur, H.P.
[10] Haluk Safak, Mehmet Sahin, Omer Faruk Yuksel, “Solid State Electronics”.
[11] M. Daraee , M.Hajian, M.Rastgoo, L. Lavasanpour Adv./ Studies theory
Phys., 2, 2008, .957.
140
[12] E H Rhoderick, R H Williams, Metal–Semiconductor Contacts, Oxford:
Clarendon, 1988, 20.
[13] D L Polla , A K Sood , J. Appl. Phys., 51, 1980,4908.
[14] A. Turut, M. Saglam, H. Efeoglu, N. Yalcin, M. Yildirim, B. Abay, Physica
B 205 (1995) 41).
[15] E.H Rhoderick, R.H Williams, Metal-Semiconductor contacts 2nd ed.
Oxford, Clarendon, 1988.
[16] A. Deneuville, M.H, J.Appl. Phys. 50, 1977, 1414.
[17] E.H. Rod,.P. Cova , A. Singh, Solid State Electron 33, 1990,11.
[18] A.Turut, M.Saglam,H.Efeoglu,N.Yalcic, M.Yildirim, and B.Abay, Physica
B, 205, 1995,41.
141
Chapter 6 Temperature Dependent Current Voltage
(I-V) Characteristics of Ag/p-SnSe Schottky
Diodes
6.1 Introduction
It has been observed that the Schottky Diodes with low barrier heights have found
application in devices operating at cryogenic temperatures such as infrared detectors and
sensors in thermal imaging [1-15]. Thus, information about their electrical characteristics of
Schottky diodes at low temperatures is vital for its better understanding which will enable us
to tailor the devices to particular requirement.
The current-voltage (I-V) characteristics of the Schottky barrier, measured only at
room temperature, does not give satisfactory information about the conduction process and
the nature of barrier formation at the metal-semiconductor (M-S) interface. However, the
temperature dependent I-V characteristics allow us to understand to different aspects of
conduction mechanisms. It is therefore essential to know the more information of the
conduction processes to deduce the junction parameters, namely barrier height and the
ideality factor. Although Thermionic emission (TE) theory is widely used to extract the
Schottky barrier diode parameters [2-17] yet there have been several reports of certain
anomalies [1-5] at low and temperatures.
142
It is observed that the ideality factor(η) increases, while the Schottky barrier height
(Φbo) decreases with decrease in temperature. The More recently, some authors [5-8]
stressed the importance of barrier height inhomogenities over the M-S contact area.
Nowadays, the nature and origin of the temperature dependence of barrier heights and
ideality factors has been successfully explained on the bias of a TE mechanism barrier in
homogenities explained on the basis of analytical potential fluctuation model using
Gaussian distribution function of barrier heights and standard deviation [4-20].The ballistic
electron emission microscopy (BEEM) studies have also supported the existence of
Gaussian distribution of barrier heights in Schottky diodes [4-25]. Simulation studies on I-V
characteristics of inhomogeneous diodes with a Gaussian distribution have also yielded
results similar to those observed in experimental data [26-27].
In present, an attempt has been made to analyze the factors which contribute in the
conduction mechanism of the undertaken Schottky Diodes and also to understand deviations
of the experimental data from theoretical predictions reported in literature. The non ideal I-V
behaviour of these Schottky diodes has been attributed on the basis of deviations caused
inhomogeneous interfaces and barriers heights.
6.2 Results and discussion
6.2.1 Forward-bias I-V characteristics
The electrical current voltage (I-V) and capacitance voltage (C-V) measurements of
the Ag/p-SnSe Schottky diodes were measured in the temperature range 220-300K(Low
Temperatures) as well as 303K-328K (High Temperatures) .The current through a uniform
143
metal semiconductor Schottky barriers due to thermionic emission (TE) can be expressed as
[1].
KT
qV
KT
qVII o exp1exp
(6.1)
Where Js is the saturation current density derived from the state line intercept of lnI at V=0
and is given by
kT
qTA
A
IJ bs
s
exp2**
(6.2)
Φbo is effect barrier height at zero bias height which is determined from the extrapolated Is in
the usual analysis of the experiment data on Schottky contacts. A**
is the effect Richardson
constant equal to 18Acm-2
K-2
for n-SnSe. A is the diode area and (η) is ideality factor and is
a measure of conformity of the diode to pure thermionic emission, determined from the
slope of the straight line regain of the forward bias ln(Is) versus V characteristics.
(a)
144
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
10-4
10-3
10-2
(Ag/p-SnSe schottky Diode)
(Na = 6.6X10
18 cm
-3)
Forw
ard
Curr
ent (A
mpere
)
Forward Voltage(Volts)
303K
308K
313K
318K
323K
328K
(b)
Fig.6.1 Plot of temperature dependent current(A) versus voltage(V) characteristics of Ag/p-
SnSe Schottky diode measured from (a) room temperature down to 220K (b) 303to 328K
The effect ideality factor and barrier heights are given by
)(ln Id
dv
kT
q (6.3)
And
o
bI
ATAkTq
2**ln (6.4)
Fig. 6.1(a) and (b) represents the current-voltage (I-V) characteristics of Ag/p-SnSe
Schottky Diodes in the temperature range 328K down to 220K.The I-V plots shift towards
higher bias side with decrease in temperature. We have performed least-square fits of
equation (1) to the linear part of the measured I-V plots (Fig.1). From these fits, the
145
experimental values of ideality factor (η) and barrier height (фb) were determined from
intercepts and slopes of the forward bias ln I versus V plot at each temperature, respectively.
Once Is is known, the zero bias barrier height can be computed with the help of equation
(6.4). The experimental ideality factors (η) and barrier height (фb) were found to be a strong
function of temperature. The ideality factor (η) was found to increase, while the фb
decreased with decreasing temperature which has been illustrated in Fig. 6.2.
Fig. 6.2 The ideality factor (η) versus temperature and the zero-bias barrier height (фbo) versus
temperature; the solid curves represent simulated the data generated by using an
analytical potential fluctuation model.
The experimental values (фb) and (η) for the device range from 0.63eV and 1.78 (at 328 K)
to 0.40eV and 3.60 (220K), respectively. As explained in [1-6] since current transport across the
metal/semiconductor (MS) interface is a temperature activated process; electrons at low
temperatures are able to surmount the lower barriers and therefore current transport will be
146
dominated by current flowing through the patches of lower Schottky barrier heights and a large
ideality factors. As the temperature increases, more and more electrons have sufficient energy to
surmount the barrier height. As a result, the dominant barrier height will increase with the
temperature and bias voltage. An apparent increase in the ideality factor and a decrease in
barrier height at low temperatures are caused possibly by other effects such as inhomogenities
of thickness and non-uniformity of the interfacial charges. This gives rise to an extra current
such that the overall characteristics still remain consistent with the TE process [6]. This result is
attributed to inhomogeneous interface and barrier heights because of a linear relationship
between the barrier height and ideality factor as obtained in fig 6.4.
Temperature
(K)
Barrier Height
(фbo)
Ideality Factor
(η)
Current Density
(Js)
220 0.409 3.61 2.15E-6
230 0.427 3.05 2.44E-6
240 0.448 2.92 2.311E-6
250 0.475 2.43 1.77E-6
270 0.504 2.54 2.449E-6
300 0.56 2.18 2.44E-6
308 0.609 2.01 1.07E-6
318 0.624 1.97 1.377E-6
328 0.632 1.78 2.239E-6
Table 6.1 Parameter of SnSe Schottky diode extracted from the temperature dependent
current voltage ( I-V) data
147
6.2.2 Richardson Plots
The barrier height can also be determined from ln(Js/T2) versus (1000/T) plot
obtained by re-arranging equation 6.2 in the manner given below :
KT
qA
T
Js bo
**
2lnln (6.5)
Fig. 6.3 Richardson plot of ln(Jo/T2) versus 10
3/T for the Ag/p-SnSe Schottky diode.
Fig. 6.3 shows the conventional energy variation of ln(Js/T2) against (1000/T). The
dependence of ln(Js/T2) on (1000/T) is found to be non-linear in the temperature measured.
The nonlinearity of the conventional ln(Js/T2) versus (1000/T) is caused by the temperature
dependence of the barrier height and ideality factor. Similar results have also found by
several authors.[16,20,21].The experimental data are fit asymptotically with a straight line
at higher temperature only, yielding a Richardson constant(A**
) of 3.076X10-3
Acm-2
K-2
and barrier height value of 0.38eV. Which is much lower than the known value of
148
18 Acm-2
K-2
for the holes in p-SnSe [28]. The deviations in the Richardson plots may be due
to the spatial inhomogeneous barrier heights and potential fluctuations at the interface that
consists of low and high barrier areas [5, 9,19]. In other words, the current of the diode will
flow preferentially through the lower barriers in the potential distribution. As was explained
by Horvath [28], the A**
value obtained from the temperature dependence of the I-V
characteristics may be affected by the lateral inhomogeneity of the barrier, and the fact that
it is different from the theoretical value which may be connected to a value of the real
effective mass that is different from the calculated one.
Fig 6.4 Zero-bias apparent barrier height versus ideality factor of Ag/p-SnSe Schottky diode
at various temperatures.
According to Refs. [6-9,13,14], the ideality factor of Schottky barrier diode with a
distribution of low Schottky barrier heights may increase with a decrease in temperature.
Schmitsdorf et al. [13] and Monch [14] used Tung‟s theoretical approach and they found a
149
linear correlation between the experimental zero-bias Schottky barrier heights and ideality
factors. We prepared a plot of the experimental barrier height versus the ideality factor
(Fig.6.4). The straight line in Fig.6.4 is the least-square fit to the experimental data. As can
be seen from Fig. 6.4, there is a linear relationship between the experimental effective
barrier heights and the ideality factors of the Schottky contact. The extrapolation of the
experimental barrier heights versus ideality factor plot to η=1 has given a homogeneous
barrier height of approximately 0.72 eV.
6.2.3 The analysis of barrier height inhomogenities
On the other hand, the modified thermionic emission (TE) model that accounts in
homogeneous Schottky barriers has explained differences from the conventional TE theory.
Although conventional models treat the interface between the metal and semiconductor as
atomically flat and spatially homogeneous, it is now established by ballistic electron
emission microscopy (BEEM) that Schottky barrier having inhomogeneous area consists of
patches of relativity lower or higher barriers with respect to a mean Bh (фb0). The total
current is the sum of the currents through all of these patches and the whole junction area.
In this model, It is assumed that there are a number of parallel diodes having SBH
and every SBD can independently make contribution to the current so that the total current
across the SBD with barrier in homogeneities is composed of those flowing in all the
individual patches with its own area and SBH as expressed in equation 6.6 but with an
apparent barrier height and ideality factor, both of which are temperature dependent.
1exp
kT
qvII d
o
(6.6)
150
The above discussed unusual contact behaviour can be explained by using an analytical
potential fluctuation model based on spatially inhomogeneous barrier heights at the interface
[4, 5, 10-15]. Suppose that the distribution of the barrier heights is Gaussian character with a
mean value b and a standard deviation ζs, which can be given as [4,5]
2
22
exp1
o
bb
o
bP
(6.7)
where
2
1
o
is the normalization constant of the Gaussian barrier-height distribution. The
total current for any forward bias V is then given by
dPVIVI bb )(),()( (6.8)
(6-9)
with
KT
qTAAI
ap
o
exp2**
(6.10)
where nap and Φap are the apparent ideality factor and barrier height at zero bias,
respectively, and the latter is given by
KT
qT o
boap2
)0(
2
(6.11)
In the ideal case (η = 1), the expression is obtained as
(6.12)
KT
Vq
KT
Vq
KT
q
KT
qTAVI ss
b
)(exp1
)(exp
2exp**)(
22
KT
q
ap
3211
151
(a)
(b)
Fig. 6.5.(a)&(b) Zero-bias apparent barrier height and ideality factor versus(2KT)-1
plot of Ag/p-SnSe Schottky diode as per to Gaussian distribution of the barrier heights.
152
The temperature dependence of ζs is usually small and thus can be neglected [22].
However, it is assumed that ζs and b are linearly bias dependent on Gaussian parameters
such that b=bo +2V and ζso = ζso + ρ3 V, where ρ2 and ρ3 are the voltage coefficients
that may depend on temperature and they quantify the voltage deformation of the barrier
height distribution [10,22]. It is obvious that the decrease of zero-bias barrier height is
caused by the existence of the Gaussian distribution and the extent of influence is
determined by the standard deviation itself. Also, the effect is particularly significant at low
temperatures. On the other hand, the abnormal increase of ideality factor occurs due to the
variation of mean barrier height and standard deviation with bias, i.e., terms involving
voltage coefficients ρ2 and ρ3.
Since equations (6.2) and (6.10) have the similar form, the fitting of the experimental
data to equation (6.2) gives Φap and ηap, which should in turn obey equations (6.11) and
(6.12). Thus, the plot of Φap versus 1000/T shown in figure 6.5(a) should be a straight line
giving () and (ζso ) intercept and slope as 1.09eV and 0.163V respectively. Furthermore, as
can be seen from figure 6.5(a and b) the experimental results of Φap are in good agreement
with equation (6.11) with (b) =1.09eV, ζso = 0.163 V values. The solid line curves in
figure 6.2 represent data estimated from these parameters by means of equation (6.11). By
comparing the (bo) and (ζso) parameters, it is seen that the standard deviation which is a
measure of the barrier homogeneity is ≈15% of the mean barrier height. Since the lower
value of ζso corresponds to a more homogeneous barrier height, this result indicates that
Ag/p-SnSe device has larger inhomogeneities at the interface. According to these results, it
can be deduced that barrier inhomogeneities can occur due to inhomogeneities in the
153
composition of the interfacial oxide layer, non uniformity of interfacial charges and
interfacial oxide layer thickness, as was discussed in [4,5,6,22].
The temperature dependence of the ideality factor can be understood on the basis of
equation 6.12. The fitted ideality factor n plot shown in figure 6.5(b) is a straight line that
gives voltage coefficients ρ2 and ρ3 from the intercept and slope as ρ2 = - 0.0284 V and ρ3
= 0.02814 V. Furthermore, as can be seen from figure 6.5(b), the experimental results of n
show good agreement with equation 6.12 for the same parameters. The linear plot of (1/nap-
1) versus q/2kT confirms that the ideality factor does indeed denote the voltage deformation
of the Gaussian distribution of the barrier height. The continuous line in figure 6.2
represents data estimated with these parameters.
The Richardson plot is now modified by combining equations 6.10 and 6.11,
KT
qA
Tk
q
T
Js boOS
**
22
22
2ln
2ln (6.13)
The modified ln(Js/T2) - (q
2ζ
2so/2k
2T
2) versus 1000/T plot, given in figure 6.6,
should also be a straight line with the slope and the intercept at the ordinate yielding the
mean barrier height(b) and A**, respectively. The modified Richardson plot has quite a
good linearity over the whole temperature range corresponding to single activation energy
around(bo). By the least-squares linear fitting of the data, bo = 1.14eV and A**
= 13.29 A
cm - 2
K - 2
are obtained. It can be seen that the value of bo is in close agreement with the
value of b = 1.09eV obtained from the plot of Φap versus 1000/T in figure 6.6, while
modified Richardson constant A** = 13.29 A cm - 2
K - 2
is in a closer agreement with the
154
theoretical value of A** = 18 A cm - 2
K - 2
than the values of the A** obtained in the
previous section.
Fig. 6.6 The Modified Richardson ln(I0/T2)-(q
2σ
2s/2k
2T
2) versus 10
3/T plot for the Ag/p-SnSe
Schottky diode generated on the basis of Gaussian distribution of the barrier heights.
6.2.4 Temperature dependent current-voltage characteristics of Schottky
diode(300K to 220K) by Cheung’ method.
The series resistance Rs is an important parameter on the electrical characteristics of
Schottky barrier diodes. This parameter is significant in the downward curvature (nonlinear
region) of the forwards bias I-V characteristics, but the other two parameters (η and фb) are
significant in both the linear and nonlinear regions of the I-V characteristics. An efficient
technique to evaluated Rs , η and фb have been proposed by S.K. Cheung and N. W. Cheung
[31]. This technique has been applied to Ag/SnSe Schottky barrier diodes and the value Rs,
155
η and фb calculated as a function of temperature. The diode parameters such as Rs, η and фb,
were obtained from the functions of cheung, for this from equation 6.14.
kT
IRVq
kT
qTAAI sb
(expexp2**
(6.14)
the following function can be written as:
ss
s
JARJd
Vd
)(ln
)( (5.11)
where kT
q .
As already discussed in chapter 5 and a plot of d(V)/d(lnJ) vs J will give RA as the slope and
η/β as the y axis intercept. To evaluate Фb, we can define a function H(J):
2**ln)(
TA
JVJH s
s
(5.12)
For equation 5.12, we can deduce
Bss JRsAJH )( (5.13)
A plot of dV/d(ln(Js)) and H(Js) versus Js is shown in Fig. 6.7 (a) & (b). The values of
various parameters so obtained are presented in Table 6.2.
156
(a)
(b)
Fig. 6.7 Plot of dV/d[In(I)]versus Js for Ag/p-SnSe Schottky diode at temperatures
(a) 220K (b) 290K
157
dV/d(lnJ) vs J
H(J) vs J
Temperature Ideality
Factor(η)
Rs(KΩ) Rs(KΩ) Barrier
height(фbo)
290 3.796 3.438 3.347 0.5495
280 4.116 3.024 3.096 0.5450
270 4.583 3.024 3.047 0.5330
260 4.822 3.050 3.136 0.5329
240 6.1290 2.433 2.4 0.5249
220 6.18352 2.867 2.980 0.52444
Table 6.2 Parameters of Ag/p-SnSe Schottky diodes extracted from the Cheung’s Method.
As shown in Table 6.2, the barrier height and ideality factor determined from the
Cheung‟s method were found to be a strong function of temperature. The barrier height
found to increase, while the ideality factor decreases with increase in temperature. It is also
clear from the table, the value of ideality factor and barrier height obtained from the
downward curvature region which results from the effect of series resistance and interface
states are greater than ideality factor values obtained from the linear region of the same
characteristics in effect of only interface state.
As can be seen in Table 6.2, the Rs calculated from the Cheung function showed an
unusual behaviour that it increases with increase of temperature. In generally, such
158
temperature dependence is an obvious disagreement with the reported negative temperature
coefficient of the series resistance. Such behaviour was attributed to the lack of free charges
at low temperature and in the temperature region where there is no carrier freezing out
which is non-negligible only at low temperature [32]. At higher temperatures, the contact
resistance and the resistance of outer connections are probably the prevalent source of the
Rs. Similar temperature dependence was obtained both experimentally and theoretically
[33,34].
6.2.5 Reverse bias Current-voltage characteristics
The temperature dependent reverse bias current voltage (I-V) characteristics of Ag/p-
SnSe Schottky diode as shown in Fig. 6.8.
Fig.6.8 Plot of temperature dependent reverse bias current (A) versus voltage (V)
characteristics of Ag/p-SnSe Schottky diode in the temperature range (30 8to
338K)
159
Fig 6.9 Plot of breakdown voltage versus temperature of Ag/p-SnSe Schottky Diode.
It has observed that the breakdown voltage decreases with increase in temperature.
So, the diode show tunneling phenomena as observed in Zener diode.
Using the general diode equation,
kT
qVJI sR
exp (6.14)
where η is the diode ideality factor and Js is the reverse saturation current density of the
junction and is defined as
KT
qV
K
TVqAJ bibi
s exp*
(6.15)
160
These parameters were calculated from the slope and intercept of the linear region of
In(IR) vs V plots for a typical junction at different temperatures.
From the estimated values of Js for different temperatures, ln(Js/T) vs T-1
graph is
plotted for the junction (Figure 6.10). From the figure, it is seen that the plot is a straight
line, which indicates that the current transport is controlled by the thermionic emission
process and follows the relation (6.2). From the slope of the plot, the built-in potential Vbi
the junction was estimated. The built-in potential of a typical Ag/p-SnSe Schottky measured
from current-voltage characteristics is given in Table 6.3. The built-in potential of the
junction calculated to be 0.90eV has been found independent of temperature.
Fig. 6.10 Plot of ln(Js/T) versus 1/T of Ag/p-SnSe Schottky Diode.
161
Temperature(K) Ideality
factor(η)
Saturation
Current
Density(Jo)
Built in
potential(eV)
305 4.46 7.07E-8 0.90
310 4.41 1.31E-7
315 4.35 2E-7
320 4.21 3.81E-7
325 3.30 5.9E-7
Table 6.3 Schottky diode parameters for reverse bias at different temperature.
In the case of V+Vbi>KT/q, reverse current expression might be reduced to
4/1)(exp bisR VVJI
Here parameter is defined as follows [35].
4/12/1
2
4
s
A
s
qNq
KT
q
(6.16)
Where εs is the dielectric constant (εs=17εo for p-SnSe) [36-38].
NA is the acceptor concentration in p-type semiconductor and Vbi the built in potential.
Since, Vbi, the built in potential, for any contact can be determined by means of the variation
of In(I) with the inverse temperature, 1/T. Therefore, an effective potential, Veff= V+Vbi, can
be introduced and the reverse bias current density be represented as
162
4/1exp effsR VJI
Thus, the parameter in the above equation and hence NA carrier densities can be estimated
by plotting the ln(IR)-Veff1/4
graph , which is given in fig. 6.11
Fig. 6.11 Variation of reverse current( log IR) with Veff1/4
at temperature of (310,320 and
330K)
The parameters found by slopes of curves in Fig 6.11, for each temperature. The
values of acceptor concentration (NA) calculated by using Equation
4/1)(exp bisR VVJI (6.17)
are listed in Table 6.4 for Ag/p-SnSe Schottky barrier diodes.
163
Temperature (K) Carrier concentration
310 12X1020
320 9X1020
330 6X1020
Table 6.4 Carrier concentrations(Na) obtained from reverse I-V characteristics in the
temperature range 310 to330K for Ag/p-SnSe Schottky Diode.
164
References -
[1] E.H. Rhoderick and R.H. Williams 1988 Metal-Semiconductor Contacts 2nd
edn (Oxford: Claredon)
[2] F.A Padovani and G Summer, Appl. Phys., A 36, 1965, 3744.
[3] Zs Horvath, Solid State-Electron, 39, 1996, 176.
[4] Y P Song, R L Van Meirhaeghe, W F Laflere and F Cardon, Solid-State
Electron, 29, 1986, 633.
[5] J H Werner and H H Guttler, J. Appl. Phys., 69, 1991, 1522.
[6] H H Guttler and J H Werner, Appl. Phys. Lett., 56, 1990, 1113.
[7] J P Sullivan, R T Tung, M R Pinto, W R Graham, J. Appl. Phys.,70,
1991,7403.
[8] R T Tung, Appl. Phys., Lett,. 58, 1991, 2821.
R T Tung, Phys.Rev,. B 45, 1992, 13509.
[9] R F Schimistsdorf, T U Kampen and W Moonch , Surf. Sci., 324, 1995, 249.
[10] W Moonch, J.Vac.Sci.Technol., 17, 1999, 1867.
[11] A. Gumus A. Turut and N Yalcin, J. Appl. Phys. 91, 2002, 245.
[12] M. K.Hudait, P. Venkateshwarlu and S. B. Krupanidhi, Solid-State Electron,
45, 2001, 133.
165
[13] B. Abay, G.Cankaya, H. S. Guder, H. Efeogh and Y. K. Yogurtcu, Semicond.
Sci. Technol. 18, 2003, 75.
[14] S .Chand and J. Kumar, J.Appl.Phys. 80, 1996, 288.
[15] S. Chand and J. Kumar J.Appl.Phys A.63, 1996, 171.
[16] C. Coskun, S. Aydogan and H. Efeoglu , Semicond. Sci. Technol. 19, 2004,
242.
[17] S. Chand and J. Kumar , Semicond. Sci. Technol. 10, 1995, 1680.
[18] M. O. Aboelfotoh Solid State-Electron, 34, 1991, 51.
[19] F.A. Padovani and G.G. Sumner, J. Appl. Phys. 38, 391 196.
[20] Zs.J. Harvoth, Mater. Res. Soc. Proc. 260, 1992, 367 .
[21] Zs.J. Harvoth; Proc. Of Int. Conf, 1996 263.
[22] W. Monch J. Vac. Sci. Technol.B 17, 1999, 1867.
[23] S.Y.Zhu, R .L. Van Meirhaeghe, Detavenier C, Cardon F, Ru G P, Q X P and
Li B Z, Solid-State Electron., 44, 2000, 663.
[24] H. Safak, M. Sahin and Yuksel O F, Solid-State Electron, 46, 2002, 49.
[25] G.M Vanalme Goubert L, R.L Van Meirhaeghe Cardon F and Daele P.V
Semicond. Sci. Technol. 14, 1999, 871.
[26] H. Palm, M. Arbes and M. Schulz Phys. Rev.Lett. 71, 1993, 2224.
[27] S. Chand and J. Kumar Semicond., Sci. Technol 12 899.
166
[28] E. Dobrocka and J. Osvald ,Semicondor Materials (New York: CRC Press) p
F.A. Padovani and G.G. Sumner, J. Appl. Phys. 38, 1997, 391, 196.
[29] Zs. Horvath Solid-State Electron. 39, 1996, 176.
[30] F.D.Auret and P.M. Mooney, J.Appl. Phy. 55, 1984, 984; 55, 1984, 988.
[31] M. Biber, Phisica B, 325, 2003, 138.
[32] S.K. Cheung, N.W. Cheung, appl. Phys. Letts., 49, 1986, 85.
[33] S. M. Sze, Physics of Semiconductor Devices, 2nd Edn. Willey, New York
1981.
[34] B. Cvikl, D. Korosak and Zs. Horvath, Vacuum, 50, 1998, 385.
[35] J. Osvald, Zs.J. Horvath, Appl. Surf. Sci., 234, 2004,349.
[36] H.Safak, M.Sahin, O.F Yuksel, Solid State Electron., 46, 2000, 49.
[37] M. Merdan,Ph.D thesis, Nottingham University, U.K, 1977.
[38] H.Safak,Ph.D thesis, The University of Selcuk, Konya,2002, 49.
[39] H.Safak, M.Sahin, O.F Yuksel, Solid State Electron., 46, 2000, 49.
167
Chapter 7
Conclusion of the work __________________________________________________________________________
The work presented in the thesis deals with the deposition of SnSe thin films, their
ohmic and schottky contact formation and the analysis of current transport phenomena in the
Ag/p-SnSe Schottky barriers and the study of their ohmic contact formation. In this study,
an emphasis has been laid down on the effect of deposition parameters like substrate
temperature (Ts) and film thickness on the „crystallite size‟ and other parameters of the SnSe
thin films. The focus of the study has been to improve the crystallite size so as to make the
semiconductor more useful for device applications. It is well known that this semiconductor
till date could not successfully been used for the formation of reliable p-n junctions. This is
despite of the fact that SnSe semiconductor is a potential material with its wide applications
in solar cells and optoelectronic devices.
Some of the results drawn out of the chapters 3 to 6 have been presented below while
the
Chapter 1: deals with the literature survey, basic semiconductor fundamentals, MS
Schottky barriers and Ohmic contacts.
Chapter 2: refers to the deposition and characterization techniques and the equipments used
for this purpose.
168
Chapter 3: The studies undertaken to describe the role of substrate temperature (Ts) in
controlling the properties of SnSe semiconductor. The energy bandgap found decreasing
with increase in substrate temperature with a typical value Eg ~ 1.19 eV at 523K which is
found close to the corresponding energy bandgap (Eg) value of SnSe reported in the
literature. The investigations made has revealed that the grain size increases with substrate
temperature upto 575K and beyond this temperature the material properties aggravates and
show deterioration, thus SnSe composition was adjudged best at 575K at which we observe
the crystallite size and resistivity displaying their optimum values of 34.5 nm and 56.33 Ω-
cm. These parameters start degrading beyond this temperature which may be due to the
decomposition of SnSe compound at higher substrate temperatures.
Chapter 4: Tin Selenide (SnSe) thin films with varying film thickness from 150nm to
500nm deposited by thermal evaporation method, at room temperature have been
undertaken for investigations. The crystallite size increased and energy bandgap decreased
with increase in thickness. Also, the effect of strains and dislocations were found to be
varying with change in thickness. The films deposited at high substrate temperature showed
better response with increase in thickness than those deposited at room temperature. Further
good polycrystalline quality SnSe films were obtained at room temperature by thermal
evaporation process; it is in contrast to the previous report of SnSe Films which showed
their amorphous nature when deposited at room temperature. The crystalline nature of the
SnSe films obtained at room temperature could possibly be due to the precise control of the
deposition rate as well as optimized source to substrate distance of the present deposition.
169
Chapter 5: Ag/p-SnSe Schottky diodes were studied for their various shapes and sizes. The
circular diodes shows improved rectifying behaviour than the square shaped schottky
diodes. In case of the Schottky diodes of different area, those with lower diameters found to
be of better quality than those with larger area. The value of barrier heights increased and
the ideality factors decreased with decrease in area. It has been observed that the slope as
well as the shift towards higher voltage side increase with decrease in area. The possible
reason for this deviation is attributed to the fact that as the device area increases, the effect
of surface defects and other factors at the interface would add up causing deviations in the
current transport behaviour. Further, there has been relatively small difference in the η and
фbo values measured from the linear region as well as downward curvature region of the
forward bias I-V characteristics of same diode. It is due to the existence of effects such as
series resistance, bias dependence of barrier height and voltage drop across the interfacial
layer and change in the interface states with bias in the concave region of I-V plot. The
reverse breakdown voltage of the Schottky diode was found to increase with decrease in
area. The C-V characteristics of the undertaken Schottky diode were used to calculate the
acceptor concentration and barrier height of the Schottky diodes. The values thus obtained
were found to be inconformity with those obtained from the current voltage characteristics.
Chapter 6: The temperature dependent forward biased current voltage (I-V) characteristics
were measured in the temperature range 220-230K. The barrier height (Φbo) decreased and
ideality factor (η) increased with decrease in temperature. The deviations were explained on
the basis of barrier height inhomogenities at the interface which were found to produce an
additional current component such that I-V characteristics continue to remain consistent
170
with the thermionic emission (TE) process. The activation energy plot (Richardson Plot) was
got modified incorporating the effect of barrier inhomogenities in the current component, the
resultant values of A**
and фbo thus obtained, were found in close agreement with the
reported ones.
To summarize, the thesis presents a detailed analysis of the current transport
phenomena of Ag/p-SnSe Schottky diodes having Aluminium (Al) back ohmic contacts.
The polycrystalline form of the SnSe material under the optimized conditions of substrate
temperature and thickness found to provide a good crystalline quality comparable with
single crystalline thin films of SnSe to be used for device applications. Thus, it may create a
scope of economical viability of the use of SnSe polycrystalline materials in various device
applications.
171
List of Publications
1 Influence of the substrate temperature on the structural, optical, and electrical properties
of Tin Selenide thin films deposited by thermal evaporation method.
N. Kumar, V. Sharma, N. Padha, N. M. Shah, M. S. Desai, C. J. Panchal, I.Yu. Protsenko,
Cryst. Res. Technol. 45,(2010),53.
2. Structural, optical and electrical characterization of Tin Selenide thin films deposited at
room temperature by thermal evaporation method.
N. Kumar, V. Sharma, U. Parihar, R. Sachdeva N. Padha, C. J. Panchal, J.Nano. Electron.
Phys. 1(2011),110.
3. Impact of annealing on CuInSe2 thin films and its Schottky interface.
U. Parihar, J. R. Ray, N. Kumar, R. Sachdeva, C.J Panchal , N. Padha, J.Nano. Electron.
Phys.(In Press).
4. Growth, Structural and Optical Properties of the Thermally Evaporated Tin Diselenide
(SnSe2) Thin Films.
R. Sachdeva, Meenakshi Sharma, Anjali Devi, N. Kumar, Usha Parihar, N. Padha and C. J.
Panchal, J.Nano. Electron. Phys.(In Press).
Paper Presentation in International Conference/ Symposium
1 International Conference on “Advances in Condensed & Nano
materials”(23-26 Feb. 2011) organized by Punjab University Chandigarh.
2 International Symposium on Semiconductor Materials and Devices(ISSMD-2011) (28-30 January 2011) organized by Applied Physics Department,
Faculty of Technology and Engineering, The M.S. University of Baroda, Vadodara.
International Conference/Symposium/School Attended
1 International School on Optoelectronic Materials and Devices ( July 27 – August 2, 2008)
held at HBSCE, Tata Institute of Fundamental Research, Mumbai
2 Indian-Japan Workshop on Zno Materials and Devices ( 18 – 20 December 2006) organized
by Department of Electronic Science, University of Delhi South Campus, New Delhi.
172
National Conference/Symposium/Workshop Attended
1. Awareness Workshop on “The Facilities of UGC-DAE Consortium for Scientific Research” (4-5 December 2006) organized by Department of Physics, Kurukshetra University, Kurukshetra
2. Workshop on “Use of Diffraction Methods in Condensed Matter” (7-9 March 2007) organized by Department of Physics and Electronics, University of Jammu and UGC-DAE Consortium for Scientific Research, Mumbai centre
3. National Workshop on “ Recent Trends In Optoelectronic Materials And Devices” (RTOMAD-2005) (3-4 October 2005) organized by Department of Electronics, Govt. Degree College, Bemina Srinagar