100 - shodhgangashodhganga.inflibnet.ac.in/bitstream/10603/25230/9/09... · 2018-07-09 · 100 3.1...
TRANSCRIPT
99
100
3.1 INTRODUCTION
Scanning Tunneling Microscope was developed first in 1982 by Binning, Rohrer,
Gerber and Weibel at IBM in Zurich Switzerland and they won Noble prize for
this invention in 1986. AFM (atomic force microscope) is an improvement of
STM which was first invented by Gerd Binnig, Quate and Christoph Gerber in
1986. Scanning Probe microscopy (SPM) consists of a family of microscopy
where a sharp probe is scanned across a surface and the tip-sample interactions
are monitored. SPM consists of many forms like scanning tunneling microscope
(STM), atomic force microscope (AFM), magnetic force microscope (MFM),
electric force microscope (EFM) etc.
In present work, AFM has been used to measure surface morphology. The
essential property of AFM is the interaction force between the tip and the
sample, which depends on their distance. At close contact the force is repulsive
while at a larger separation the force is attractive. This results in different
operation modes which should be chosen according to the characteristics of the
sample, since each mode has different advantages.
For microanalysis the scanning electron microscope (SEM) has been used.
Microanalysis is the analysis of “very small” samples—by whatever technique is
available. Historically, however, the term has had a much narrower meaning.
When electrons of appropriate energy impinge on a sample, they cause the
emission of x-rays whose energies and relative abundance depend upon the
composition of the sample. Using this phenomenon to analyze the elemental
content of microvolumes (roughly one to several hundred cubic micrometers) is
what we commonly mean by microanalysis.
By scanning the beam in a television-like raster and displaying the intensity of a
selected X-ray line, element distribution images or 'maps' can be produced. Also,
images produced by electrons collected from the sample reveal surface
topography or mean atomic number differences according to the mode selected.
The scanning electron microscope (SEM), which is closely related to the electron
probe, is designed primarily for producing electron images, but can also be used
for elements mapping, and even point analysis, if an X-ray spectrometer is added.
101
There is thus a considerable overlap in the functions of these instruments.
Structure, microstructure, and defect geometry, as well as chemical composition
and spatial distribution are important parameters determining the behavior of
materials and practical applications.
3.2 SCANNING ELECTRON MICROSCOPE (SEM)
3.2.1 Introduction
The scanning electron microscope (SEM) is a type of electron microscope
capable of producing high resolution images of a sample surface. Due to the
manner in which the image is created, SEM images have a characteristic three
dimensional appearance and are useful for observing the surface structure of the
sample. In scanning electron microscopy (SEM), a fine probe of electrons with
energies typically up to 40keV is focused on a specimen, and scanned along a
pattern of parallel lines. Various signals are generated as a result of the impact of
the incident electrons, which are collected to form an image or to analyse the
sample surface. These are mainly secondary electrons, with energies of a few
tens of eV, high-energy electrons back-scattered from the primary beam and
characteristic X-rays. Such rich physical interaction in a practical tool, making
the SEM the powerful instrument it is today in materials and life-science.
The history of electron microscopy began with the development of electron
optics. In 1926, Busch studied the trajectories of charged particles in axially
symmetric electric and magnetic fields, and showed that such fields could act as
particle lenses, laying the foundations of geometrical electron optics [1]. Nearly at
the same time, the French physicist de Broglie introduced the concept of
corpuscule waves. A frequency and hence a wavelength was associated with
charged particles: wave electron optics began [2]. Following these two discoveries
in electron optics, the idea of an electron microscope began to take shape.
In 1931, independently of the ‘‘material wave’’ hypothesis put forward by de
Broglie several years earlier (1925), Ruska and his research group in Berlin,
were working on electron microscopy. They were disappointed learning that
102
even with electrons a wavelength would limit the resolution. But they found
using de Broglie equation that electron wavelengths were almost five orders of
magnitude smaller than the wavelength of light used in optical microscopy. It
was thus considered that electron microscopes still could prove a better
resolution than light instruments, and no reason existed to abandon this aim. In
1932, Knoll and Ruska tried to estimate the resolution limit of the electron
microscope. Assuming the resolution limit formula of the light microscope was
still valid for material waves, they replaced the light wavelength by the electrons
wavelength at an accelerating voltage of 75kV. A theoretical limit of 0.22nm
resulted, a value experimentally reached only 40 years later. Although these
calculations proved it was possible to reach a better-than-light-microscope
resolution when working at high magnifications [3], many technical limits had to
be to overcome. Ruska and Knoll tried to implement Busch’s lens formula
experimentally. Their work resulted in the construction of the first transmission
electron microscope (TEM) in 1931, with a magnification of 16 [4].
Knoll built a first ‘‘scanning microscope’’ in 1935. However, as he was not using
demagnifying lenses to produce a fine probe, the resolution limit was around
100mm because of the diameter of the focused beam on the specimen. In 1938,
von Ardenne clearly expressed the theoretical principles underlying the scanning
microscope, as we know it today. Because it was difficult to compete with TEM in
resolution achieved for thin specimens, the scanning microscopy development
was oriented more toward observing the surface of samples. The first true SEM
was described and developed in 1942 by Zworykin, who showed that secondary
electrons provided topographic contrast by biasing the collector positively
relative to the specimen. One of his main improvements was using an electron
multiplier tube as a preamplifier of the secondary electrons emission current. He
reached a resolution of 50 nm, which was still considered low in comparison
with the performance obtainable in TEM [5].
In 1948, Oatley began to build an SEM based on Zworykin’s microscope.
Following this development, Smith (1956) shown that signal processing could be
used to improve micrographs. He introduced nonlinear signal amplification, and
103
improved the scanning system. Besides, he was also the first to insert a stigmator
in the SEM to correct lens cylindrical imperfections.
In 1960, Everhart and Thornley greatly improved the secondary electron
detection. A new detector was created with a positively biased grid to collect
electrons, a scintillator to convert them to light, and a light-pipe to transfer the
light directly to a photomultiplier tube [5].
In 1963, Pease and Nixon combined all of these improvements in one
instrument: the SEMV with three magnetic lenses and an Everhart–Thornley
detector (ETD). This was the prototype for the first commercial SEM, developed
in 1965—the Cambridge Scientific Instruments Mark I ‘‘Stereoscan’ [6]. The SEM
we are using today are not very different from this first instrument.
3.2.2 Principle of SEM
Scanning electron microscope (SEM) consists of an electron gun and
electromagnetic lens system to study the surface structure and morphology
of solids. A well-defined electron beam impinges on the specimen and leads
to generation of secondary electrons, back scattered electrons, absorbed
electrons, characteristic X-rays etc (figure 3.1). These electrons can be detected
by suitable detectors and give information about the surface structure and
morphology of the specimens. The characteristic X-rays generated are used for
Figure 3.1 Example of some of the different types of signals produced when
high-energy electron impinge on a material.
104
the identification and estimation of different elements present in the specimen
by energy dispersive spectrometer (EDS) and wavelength dispersive
spectrometer (WDS). The depth of focus is much larger than the optical
microscope image even at lowest magnifications is one of the major advantages
of SEM.
3.2.3 Working of an SEM instrument
Electrons from a thermionic or field-emission cathode are accelerated by a
voltage of 1-50kV between cathode and anode. The smallest beam cross section
at the gun (the crossover) with a diameter of the order of 10-50µm for
thermionic and 10-100nm for field-emission guns, is demagnified by two or
three stage electromagnetic lens system, so that an electron probe of diameter 1-
10nm carrying current of 10-10 to 10-12 Å is formed at the specimen surface.
A deflection coil system in front of the last lens scans the electron probe in a
raster across the specimen and in synchronism with the electron beam of a
separate cathode ray tube (CRT). The intensity of the CRT is modulated by one of
the signals recorded to form an image. The magnification can be increased
simply by decreasing the scan-coil current and keeping image size on CRT
constant. Figure 3.2 shows the general schematic diagram of SEM.
An advantage of SEM is the wide variety of electron-specimen interactions that
can be used to form an image to give qualitative and quantitative information.
The large depth of focus, excellent contrast and the straight forward preparation
of solid specimen are advantages of SEM.
Here an electron beam scans the object (the specimen) and because of
synchronized scans of electron beam and the CRT screen (nowadays, monitor),
there is one-to-one correspondence between the spot on the specimen and the
spot on the screen. Unlike Optical microscopy, SEM requires the vacuum
environment and specimen surface to be electrically conductive.
The electron beam is produced by hair-pin shaped tungsten (W) filament by
thermionic emission. The acceleration voltage of 5 to 50kV can be applied
between anode and cathode and hence we get the electron beam of such energy.
105
Figure 3.2 Schematic diagram of Scanning Electron Microscope (SEM).
This diverged electron beam passes through a pair of electro-magnetic lenses
(coils) and finely passes through probe forming lens, which makes the beam in a
form of a fine probe (~10nm diameter). This fine electron probe scans on the
specimen area (marked as lines in figure 3.2) in a linear manner. Another
electron beam is in synchronization with this beam which scans on the CRT (or
106
monitor) with the help of same scan generator. In SEM, the formation of image
takes place because of electron beam-specimen interaction [7].
3.2.4 Sample preparation
Since the electron probe analyses only to a shallow depth, specimens should be
well polished so that surface roughness does not affect the results. Sample
preparation is essentially as for reflected light microscopy, with the provision
that only vacuum compatible materials must be used. Opaque samples may be
embedded in epoxy resin blocks.
In principle, specimens of any size and shape (within reasonable limits) can be
analyzed. Holders are commonly provided for 25mm (1") diameter round
specimens and for rectangular glass slides. Standards are either mounted
individually in small mounts or in batches in normal-sized mounts.
3.2.5 Electron-specimen interaction
Electron beam-specimen interaction gives various signals which can be used to
form images and also can be used for other important information.
In inelastic scattering, the trajectory of the incident electron is only slightly
perturbed, but energy is lost through interactions with the orbital electrons of
the atoms in the specimen. ‘Loss’ of kinetic energy (of the primary electron) is
mainly caused by the interaction with the electrons of atoms of the specimen.
The concept of interaction volume of the primary beam electrons and the
sampling volume of the emitted secondary radiation are important both in
interpretation of SEM images and in the proper application of quantitative X-ray
microanalysis. The image details and resolution in the SEM are determined not
by the size of the electron probe by itself but rather by the size and
characteristics of the interaction volume.
When the accelerated beam electrons strike a specimen they penetrate inside it
to depths of about 1µm and interact both elastically and inelastically with the
solid, forming a limiting interaction volume from which various types of
107
radiation emerge, including BSE, SE, characteristic and bremsstrahlung x-rays,
and cathodoluminescence in some materials (figure 3.3).
The combined effect of elastic and inelastic scattering controls the penetration of
the electron beam into the solid. The resulting region over which the incident
electrons interact with the sample is known as interaction volume. The
interaction volume has several important characteristics, which determine the
nature of imaging in the SEM.
The energy deposition rate varies rapidly throughout the interaction volume,
being greatest near the beam impact point. The interaction volume has a distinct
shape (figure 3.3). For low-atomic-number target it has distinct pear shape. For
intermediate and high-atomic number materials the shape is in the form of hemi-
sphere. The interaction volume increases with increasing incident beam energy
and decreases with increasing average atomic number of the specimen. For
secondary electrons the sampling depth is from 10 to 100nm and diameter
equals the diameter of the area emitting backscattered electrons. BSE are
emitted from much larger depths compared to SE. Ultimately the resolution in
Figure 3.3 The interaction volume.
108
the SEM is controlled by the size of the interaction volume.
3.2.6 Magnification, brightness and contrast of the SEM images
Electron images are obtained by rastering the electron beam across the
specimen surface using the deflection coils inside the objective lens and
synchronously rastering the output signal of the detector on a cathode-ray tube
(CRT). The ratio of the area rastered on the specimen to that of the CRT gives the
magnification. For example, a rastered area of 200µm2 (200 x 10-6 m2) displayed
on CRT with an area of 20cm (200,000 x 10-6 m2) yields a magnification of 1000x.
This is a very different process than the production of an image by an optical
microscope. Electronic images are sequentially "constructed" during the
rastering of the beam, whereas in optical systems all parts of the sample are
imaged simultaneously. Brightness can be increased/decreased by amplifying
the signal from the sample and it can be affected by topography, composition,
electrical conductivity, and other properties of the sample. Contrast reflects the
variation in the signal from point to point. Contrast can also be enhanced by
electronically increasing the difference between small and large signals. Working
distance (WD), probe current and probe diameter, incorrect electron gun
alignment, astigmatism, diffused scattering of electrons from the edges (edge
effect), electrostatic charging of the sample, external disturbances(magnetic
fields, flooring, improper grounding) are factors affecting the quality of the
image. The SEM image looks three dimensional because of SEM's ability to focus
up to certain depth and hence the images can be easily interpreted.
3.3 IMAGE FORMATION AND ELEMENT ANALYSIS WITH A SEM-EDAX
SYSTEM
On the page about the principle of a scanning electron microscope (SEM), we saw
that the sample in this instrument is bombarded by an electron beam in order to
obtain a detailed topographical image of the surface of the sample from the
ejected electrons (secondary electrons, see figure 3.4 A here below). Besides,
there are scanning electron microscopes which are equipped with EDS (Energy
109
Dispersed Spectroscopy) or EDAX (Energy-Dispersed Analysis of X-rays)
detectors that capture the emitted X-ray (figure 3.4 C, 3.4 D and 3.4 E). With such
instruments, like the one shown here below (figure 3.18), it is possible to
determine which elements are present in the surface layer of the sample (at a
depth in the micrometer range) and where these elements are present
("mapping technique"). This particular microscope also allows one to capture
directly reflected electrons, the so-called back scattered electrons (figure 3.4 B),
from which one can obtain a global appreciation whether one or several
elements are present in the surface layer of the sample. Also the so-called Auger
electrons, which are emitted just under the surface (figure 3.4 F, 3.4 G, 3.4 H),
provide information about the nature of the atoms in the sample.
Figure 3.4 A: The bombarding electrons (=primary electrons) can penetrate in
Figure 3.4 (A, B, C, D, E, F, G, H) Secondary emission in Scanning Electron
Microscopy.
110
the electron shells of the atoms composing the surface of the sample. The energy
(negative charge, mass, velocity) of these incident electrons can be converted to
eject local electrons, so-called secondary electrons, from the shells of the atoms
in the surface of the specimen. This information can be utilized to reconstruct a
detailed topographical image of the sample (SEI = Secondary Electrons Imaging).
The final image looks like a shadow-cast photograph of the surface of the sample.
This record of the morphology is the best-known application of a scanning
electron microscope.
Figure 3.4 B: Primary electron can also be reflected by atoms at about 10-100
nanometer depth at the surface. These so-called "back-scatter" conserve their
energy at incidence, but their direction of propagation has been modified upon
interaction. One can obtain a rough representation whether the surface of the
sample is constituted of a single or multiple elements.
Figure 3.4 C, D, E: At the surface of the sample electrons in the deeper electron
shells (shell K in 3.4 C) can be ejected by primary electrons (Pe- indicated in red),
resulting in an electron hole. When this lower-shell position is filled by an
electron from higher shell (green arrow in 3.4 D) energy is released. This can be
as light (photons; the phenomenon is also called Cathodoluminescence) or as X-
ray. Because each element emits an own characteristic energy value, the
elements present in the micrometer range depth of the sample can be
determined.
Figure 3.4 F, G, H: Another phenomenon is that the energy released upon filling a
hole in the K shell by an electron from the L shell is used to expulse an electron
from the external M shell: a so-called Auger electron. The released energy is
characteristic for the type of atom. Auger electrons are produced in the
outermost surface layer (at nanometer depth) of the sample [8].
3.4 RESULTS AND DISCUSSION
Surface morphological studies were carried out using the scanning electron
microscopy. The SEM images of as-deposited and annealed samples of ZnTe films
111
are shown in figure 3.5 (A, B, C, D & E). From these images it is clear that, the
surface morphology is different.
Figure 3.5 (A, B, C, D & E) SEM photographs of as-grown ZnTe thin film and
annealed at 323K, 373K, 423K and 473K respectively.
68 nm
112
3.4.1 For different annealing temperature
The SEM of the as-deposited film (68nm) and thermally annealed films at 323K,
373K, 423K, 473K are displayed in figures 3.5 (A, B, C, D & E) respectively. It was
found that as deposited film at 300K (figure 3.5 A) exhibited a rough surface and
there is no smooth, feature surface that can easily be observed.
However when the films are annealed at higher temperatures, we get the grass
root like structures seen in figure 3.5 (B, C). These films exhibit non-uniform
surfaces with voids. On annealing, voids and porosity of the films are found
to decrease with increase in annealing temperature. Annealing resulted in
the transformation of the as deposited films to one with a morphology
consisting of almost uniform surface. The SEM micrographs also show an
improvement in the structural homogeneity with heat treatment.
The surface morphology of the thin films annealed at 423K and 473K is
presented in figure 3.5 (D, E). For annealed films at 423K and 473K, the SEM
images show uniform surfaces with reduced intergranular spacing. The observed
surface is uniform and randomly distributed small rods are observed over the
surface. This is most probably referred to the foreign particles during annealing
process. It is concluded that the film surfaces become dense, uniform and
compact in nature after post-annealing [9, 10].
3.4.2 For different thickness
The surfaces of the films with different thicknesses were examined by SEM.
Figure 3.6 shows the SEM images of ZnTe films of different thicknesses. All films
are deposited as well as analyzed at room temperature. The SEM morphology of
the film deposited with 10 SILAR cycles having thickness 68nm has discussed in
previous section as figure 3.5 (A). Figures 3.5(A) and 3.6(B) show that the film
compactness was high; free of pinholes. Figure 3.6 (B) shows morphology of
ZnTe thin film for thickness 103nm with initial notches. Initial notches observed
for 103nm is now converted into cracks for thickness 156nm shown in figure 3.6
(C). These cracks increase with increase in film thicknesses due to effects of
evaporation of absorbed water and reorganization of the grain as shown in
figure 3.6 (D, E) for 210nm and 278nm thicknesses.
113
Figure 3.6 (B, C, D & E) SEM photographs of ZnTe thin film having
thicknesses 103nm, 156nm, 210nm & 278nm respectively.
One possible explanation for this unexpected behavior could be that the ‘‘growth
stresses’’ in these films are of opposite sign leading to compressive strains in the
plane of the substrate and hence causing the band gap to shrink (chapter 5,
section 5.6.1.8 (table 5.2)) rather than expand. However, this is contrary to the
microstructural evidence (chapter 4, section 4.8.2.1(table 4.8)), albeit indirect,
that these films are in less tension: thicker films undergo mud cracking,
indicating that the films are indeed under biaxial tension [11-14].
3.5 ENERGY DISPERSIVE SPECTROSCOPY (EDS/EDAX) ON THE SCANNING
ELECTRON MICROSCOPY (SEM)
The brief introduction and working of energy dispersive spectroscopy
(EDS/EDAX) on the scanning electron microscopy (SEM) are described as follow:
114
3.5.1 Introduction
In this cursory treatment of the subject, we can divide our task into three major
parts. First, we consider the processes that follow the excitation of the sample by
an electron beam. We are most interested in the process by which x-rays are
emitted, but our efforts will be repaid if we also look at some of the other
interactions that occur. Next, we are interested in the means by which the
emitted x-rays are collected, sorted, and counted. That is, we want to know how
the energetic emissions of an electron-excited sample get translated into
analyzable data. Finally, we look at the analysis techniques themselves [15, 16].
Initial EDS/EDAX analysis usually involves the generation of an X-ray spectrum
from the entire scan area of the SEM. SEM photograph is a secondary electron
image of a specimen and the corresponding X-ray spectra was generated from
the entire scan area. The Y-axis shows the counts (number of X-rays received
and processed by the detector) and the X-axis shows the energy level of those
counts.
3.5.2 Principle of EDAX
The process of x-ray emission is shown schematically in figure 3.7. First, an
electron from, say, a scanning electron microscope, ejects an electron from an
Figure 3.7 X-ray microanalysis is based on electronic transitions between
inner atomic shells. An energetic electron from an electron column
dislodges an orbital electron from a shell of low energy (E1). An electron
from a shell of higher energy subsequently fills the vacancy, losing energy
in the process. The lost energy appears as emitted radiation of energy E2 -
E1.
115
inner shell of a sample atom. The resulting vacancy is then filled by an electron
from a higher energy shell in the atom. In “dropping” to a state of lower energy,
this vacancy-filling electron must give up some of its energy, which appears in
the form of electromagnetic radiation. The energy of the emitted radiation, then,
is exactly equal to the energy difference between the two electronic levels
involved. Since this energy difference is fairly large for inner shells, the radiation
appears as x-rays.
To complicate matters a bit, there are many energy levels-therefore many
potential vacancy-filling mechanisms-within every atom. As a consequence, even
a sample of pure iron will emit x-rays at many energies.
When excited by electrons of sufficient energy, every element in a sample will
emit a unique and characteristic pattern of x-rays. Furthermore, under given
analysis conditions, the number of x-rays emitted by each element bears a more
or less direct relationship to the concentration of that element.
Converting these x-ray emissions to analyzable data is the job of a series of
electronic components (see figure 3.8), which, in the end, produce a digital
spectrum of the emitted radiation.
Figure 3.8 In energy dispersive microanalysis, each emitted x-ray produces
a charge pulse in a semiconductor detector. This tiny and short-lived
current is converted first into a voltage pulse, then into a digital signal
reflecting the energy of the original x-ray. The digital signal, in turn, adds a
single count to the appropriate channel of a multi-channel analyzer (MCA).
116
The x-ray photon first creates a charge pulse in a semiconductor detector; the
charge pulse is then converted into a voltage pulse whose amplitude reflects the
signal, which causes one count to be added to the corresponding channel of a
multichannel analyzer. After a time, the accumulated counts from a sample
produce an x-ray spectrum [15, 16].
EDAX makes use of the X-ray spectrum emitted by a solid sample bombarded
with a focused beam of electrons to obtain a localized chemical analysis. All
elements from atomic number 4 (Be) to 92 (U) can be detected in principle, inner
shell of a sample atom. The resulting vacancy is then filled by an electron though
not all instruments are equipped for 'light' elements (Z < 10). Qualitative analysis
involves the identification of the lines in the spectrum and is fairly
straightforward owing to the simplicity of X-ray spectra. Quantitative analysis
(determination of the concentrations of the elements present) entails measuring
line intensities for each element in the sample and for the same elements in
calibration standards of known composition.
3.5.3 Accuracy and sensitivity
X-ray intensities are measured by counting photons and the precision obtainable
is limited by statistical error. For major elements it is usually not difficult to
obtain a precision (defined as 2σ) of better than ± 1% (relative), but the overall
analytical accuracy is commonly nearer ± 2%, owing to other factors such as
uncertainties in the compositions of the standards and errors in the various
corrections which need to be applied to the raw data. As well as producing
characteristic X-ray lines, the bombarding electrons also give rise to a continuous
X-ray spectrum, which limits the detachability of small peaks, owing to the
presence of 'background'.
Using routine procedures, detection limits are typically about 1000 ppm (by
weight) but can be reduced by using long counting times.
3.5.4 Spatial resolution
Spatial resolution is governed by the penetration and spreading of the electron
beam in the specimen (figure 3.9). Since the electrons penetrate an
117
approximately constant mass, spatial resolution is a function of density. In the
case of silicates (density about 3 g cm-3), the nominal resolution is about 2 µm
under typical conditions, but for quantitative analysis a minimum grain size of
several micrometers is desirable.
Figure 3.9 Simulated trajectories of electrons (energy 20 keV) in Si
(rectangle = 1x2µm).
Better spatial resolution is obtainable with ultra-thin (~100 nm) specimens, in
which the beam does not have the opportunity to spread out so much. Such
specimens can be analyzed in a transmission electron microscope (TEM) with an
X-ray spectrometer attached, also known as an analytical electron microscope, or
AEM.
3.6 X-RAY SPECTROSCOPY
Soon after x-rays were discovered in 1895, it became apparent that x-ray
energies are intimately related to the atomic structure of the substances that
emit them. And since the atomic structure of each chemical element is different,
it follows that each element-when stimulated to do so-emits a different pattern
of x-rays. By the 1920s, these characteristic patterns had been recorded for most
of the elements. Until the late forties, however, analyzing substances by
stimulating and recording their x-ray emissions remained the province of the
research scientist.
118
3.6.1 Atomic structure
According to Rutherford-Bohr model of the atom, electrons orbit around the
positive nucleus. In the normal state the number of orbital electrons equals the
number of protons in the nucleus (given by the atomic number, Z). Only certain
orbital states with specific energies exist and these are defined by quantum
numbers. With increasing Z, orbits are occupied on the basis of minimum energy,
those nearest the nucleus, and therefore the most tightly bound, being filled first.
Orbital energy is determined mainly by the principal quantum number (n). The
shell closest to the nucleus (n = 1) is known as the K shell; the next is the L shell
(n = 2), then the M shell (n = 3), etc. The L shell is split into three subshells
designated L1, L2 and L3, which have different quantum configurations and
slightly different energies (whereas the K shell is unitary). Similarly, the M shell
has five subshells. This model of the inner structure of the atom is illustrated in
figure 3.10.
The populations of the inner shells are governed by the Pauli Exclusion Principle,
which states that only one electron may possess a given set of quantum numbers.
The maximum population of a shell is thus equal to the number of possible states
possessing the relevant principal quantum number. In the case of the K shell this
is 2, for the L shell 8, and for the M shell 18. Thus for Z ≥ 2 the K shell is full, and
for Z ≥10 the L shell is full.
Electrons occupying outer orbits are usually not directly involved in the
Figure 3.10 Schematic diagram of inner atomic electron shells.
119
production of X-ray spectra, which are therefore largely unaffected by chemical
bonding etc.
3.6.2 Origin of Characteristic X-rays
'Characteristic' X-rays result from electron transitions between inner orbits,
which are normally full. An electron must first be removed in order to create a
vacancy into which another can 'fall' from an orbit further out. In electron probe
analysis vacancies are produced by electron bombardment, which also applies to
X-ray analysis in the TEM.
X-ray lines are identified by a capital Roman letter indicating the shell containing
the inner vacancy (K, L or M), a Greek letter specifying the group to which the
line belongs in order of decreasing importance α, β, etc.), and a number denoting
the intensity of the line within the group in descending order (1, 2, etc.). Thus the
most intense K line is Kα1 (The less intense Kα2 line is usually not resolved, and
the combined line is designated Kα1, 2 or just Kα). The most intense L line is Lα1.
Because of the splitting of the L shell into three subshells, the L spectrum is more
complicated than the K spectrum and contains at least 12 lines, though many of
these are weak.
Characteristic spectra may be understood by reference to the energy level
diagram (figure 3.11), in which horizontal lines represent the energy of the atom
with an electron removed from the shell (or subshell) concerned. An electron
transition associated with X-ray emission can be considered as the transfer of a
vacancy from one shell to another, the energy of the X-ray photon being equal to
the energy difference between the levels concerned. For example, the Kα line
results from a K-L3 transition (figure 3.11). Energies are measured in electron
volts (eV), 1 eV being the energy corresponding to a change of 1V in the potential
of an electron (= 1.602 x10-19J). This unit is applicable to both X-rays and
electrons. X-ray energies of interest in electron probe analysis are mostly in the
range 1-10keV.
The 'critical excitation energy' (Ec) is the minimum energy which bombarding
electrons (or other particles) must possess in order to create an initial vacancy.
120
Figure 3.11 Energy level diagram for Ag showing transitions responsible
for main K and L emission lines (arrows show direction of vacancy
movement), energy of emission line indicated in brackets.
Figure 3.12 shows the dependence of Ec on Z for the principal shells. In electron
probe analysis the incident electron energy (E0) must exceed Ec and should
preferably be at least twice Ec to give reasonably high excitation efficiency. For
atomic numbers above about 35 it is usual to change from K to L lines to avoid
the need for excessively high electron beam energy (which has undesirable
implications with respect to the penetration of the electrons in the sample, and in
any case may exceed the maximum available accelerating voltage).
3.6.3 Wavelengths, energies and intensities of X-ray lines
The preceding discussion treated X-rays as photons possessing a specific energy
(E). Sometimes it is more appropriate to describe X-rays by their wavelength (λ),
which is related to energy by the expression:
E.λ=12396
121
Figure 3.12 Energies of principal characteristic lines (_) and their
excitation energies (----).
where E is in electron volts and λ is in Å, where 1Å = 10-10m.
Since X-ray lines originate in transitions between inner shells, the energy of a
particular line shows a smooth dependence on atomic number, varying
approximately as Z2 (Moseley's law). The energies of the Kα1, Lα1 and Mα1 lines
are plotted against Z in figure 3.12.
The total X-ray intensity for a particular shell is divided between several lines. In
the case of the K shell, more than 80% of the total intensity is contained in the
combined Kα1, 2 line (figure 3.13). The relative intensity of the Kβ line decreases
with decreasing atomic number, in accordance with the electron occupancies of
the relevant energy levels.
3.6.4 The continuous spectrum
Electron bombardment not only produces characteristic X-ray lines resulting
from electron transitions between inner atomic shells but also a continuous X-
ray spectrum or continuum, covering all energies from zero to E0 (the incident
electron energy). This continuum arises from interactions between incident
electrons and atomic nuclei. The intensity of the continuum decreases
monotonically with increasing X-ray energy, and is approximately proportional
to Z.
122
Figure 3.13 Typical K spectra.
The main significance of the continuum in the present context is that it
contributes the 'backgrounds' upon which characteristic elemental lines are
superimposed.
3.7 ENERGY-DISPERSIVE SPECTROMETER
Energy-dispersive spectrometers (EDSs) employ pulse height analysis, a detector
giving output pulses proportional in height to the X-ray photon energy is used in
conjunction with a pulse height analyzer (in this case a multichannel type). A
solid state detector is used because of its better energy resolution. Incident X-ray
photons cause ionization in the detector, producing an electrical charge, which is
amplified by a sensitive preamplifier located close to the detector. Both detector
and preamplifier are cooled with liquid nitrogen to minimize electronic noise. Si
(Li) or Si drift detectors (SDD) are commonly in use.
3.7.1 Energy resolution
The ED spectrum is displayed in digitized form with the x-axis representing X-
ray energy (usually in channels 10 or 20 eV wide) and the y-axis representing the
number of counts per channel (figure 3.14). An X-ray line (consisting of
effectively mono-energetic photons) is broadened by the response of the system,
producing a Gaussian profile. Energy resolution is defined as the full width of the
peak at half maximum height (FWHM). Conventionally, this is specified for the
Mn Kα peak at 5.89keV. For Si(Li) and SDD detectors, values of 130-150eV are
123
typical (Ge detectors can achieve 115eV). The resolution of an EDS is about an
order of magnitude worse than that of a WDS, but is good enough to separate the
K lines of neighboring elements (figure 3.14).
3.7.2 Dead time and throughput
In processing the pulses from a solid state detector prior to pulse-height analysis,
it is necessary to use certain integrating time to minimize noise. The system
consequently has a specific 'dead time', or period after the arrival of an X-ray
photon during which the system is unresponsive to further photons. This limits
the rate at which pulses can be processed and added to the recorded spectrum.
‘Throughput' passes through a maximum above which it decreases with further
increases in input count rate. The maximum throughput rate is a function of the
integration time and the design of the system.
Energy resolution is determined partly by the statistics of the detection process
and partly by noise fluctuations in the baseline upon which the pulses are
superimposed. The longer the integration time, more the noise is smoothed out,
and the better the energy resolution. There is thus a 'tradeoff' between
resolution and throughput. Maximum throughput rates have been typically in the
region of 20 000 counts s-1 for Si (Li) and 100 000 counts s-1 and above for SDD.
Figure 3.14 ED spectrum of jadeite (part), showing K peaks of Na, Al and Si.
124
3.8 QUALITATIVE ANALYSIS
Qualitative analysis is the process of identifying which elements are present in a
sample. As suggested in our discussion of minimum detection limits, qualitative
analysis has as its goal a statement of the form, “Elements X, Y, and Z are
definitely present in the sample; if other elements are present, they must be
present at concentrations less than the minimum detection limit (MDL).” MDLs
must always be kept in mind during qualitative analysis.
3.8.1 Line identification
The object of qualitative analysis is to find what elements are present in an
'unknown' specimen by identifying the lines in the X-ray spectrum using tables of
energies or wavelengths. Ambiguities are rare and can invariably be resolved by
taking into account additional lines as well as the main one.
3.8.2 Qualitative ED analysis
The ED spectrometer is especially useful for qualitative analysis because a
complete spectrum can be obtained very quickly. Aids to identification are
provided, such as facilities for superimposing the positions of the lines of a given
element for comparison with the recorded spectrum (figure 3.15). Owing to the
relatively poor resolution, there are cases where identification may not be
immediately obvious. An example showing unresolved S K and Pb M lines is
given in figure 3.16.
Figure 3.15 Line markers for S K and Ba L lines in the ED spectrum of barite.
125
3.9 QUANTITATIVE ANALYSIS – EXPERIMENTAL
In its simplest form, qualitative analysis proceeds by determining the energies of
peaks present in the spectrum and comparing them with a chart listing the
known energies of x-ray emissions. Modern analyzers have automated this
process to varying degrees, and most provide markers that can be called to the
video display by atomic number or symbol.
In highly automated versions, software routines detect the location of spectral
peaks, compare them with tabulated energy values, check for inconsistencies (for
example, an apparent Kβ peak but no corresponding Kα), then print out a list of
the elements present.
In general, however, routines of this type are not intended to make sophisticated
judgments, but rather to limit the number of judgments required of the user.
3.9.1 Counting statistics
X-ray intensities are measured by counting pulses generated in the detector by
X-ray photons, which are emitted randomly from the sample. If the mean number
of counts recorded in a given time is n, then the numbers recorded in a series of
discrete measurements form a Gaussian distribution with a standard deviation
(σ) of n1/2/n. A suitable measure of the statistical error in a single measurement
is ±2σ. It follows that 40 000 counts must be collected to obtain a 2σ precision of
± 1% (relative). Such statistical considerations thus dictate the time required to
measure intensities for quantitative analysis.
Figure 3.16 Unresolved peaks in an ED spectrum (PbS sample).
126
3.9.2 Choice of conditions
The optimum choice of accelerating voltage is determined by the elements
present in the specimen. The accelerating voltage (in kV) should be not less than
twice the highest excitation energy Ec (in keV) of any element presents, in order
to obtain atomic number is commonly Fe, which also has the highest excitation
energy (7.11 keV), hence the accelerating voltage should be at least 15 kV. Line
intensities increase with accelerating voltage, but so does electron penetration,
making spatial resolution worse and increasing the absorption suffered by the
emerging X-rays.
The other important variable selected by the user is beam current. The higher
the current the higher the X-ray intensity, but there are practical limitations.
Some samples are prone to beam damage, which necessitates the use of a low
current. In the case of ED analysis, the limited throughput capability of the
system has to be considered and a current as low as a few nA may be
appropriate.
3.10 QUANTITATIVE ANALYSIS - DATA REDUCTION
Quantitative analysis seeks to establish not only the identities of the elements
present in a sample, but also their concentrations, together with an indication of
the confidence that can be placed in the computed results. Assuming that a
qualitative analysis has been concluded, the quantitative analysis must proceed
through several phases: background removal, deconvolution of overlapped
peaks, and calculation of elemental concentration.
3.10.1 Castaing's approximation
As shown by Castaing (1951), the relative intensity of an X-ray line is
approximately proportional to the mass concentration of the element concerned.
This relationship is due to the fact that the mass of the sample penetrated by the
incident electrons is approximately constant regardless of composition. (The
electrons are decelerated by interactions with bound electrons and the number
of these per atom is equal to the atomic number, which, in turn, is approximately
127
proportional to the atomic weight). Given this approximation, an 'apparent
concentration' (C') can be derived using the following relationship:
(
) 3.1
where Isp and Ist are the intensities measured for specimen and standard
respectively, and Cst is the concentration of the element concerned in the
standard. To obtain the true concentration, certain corrections are required.
3.10.2 Matrix corrections
'Apparent concentrations' require various corrections which are dependent on
the composition of the matrix within which the element concerned exists. The
first of these corrections arises because the mass penetrated is not strictly
constant, but is affected by differences in the stopping power of different
elements. This is the 'stopping power correction'. The second allows for the loss
of incident electrons, which escape from the surface of the sample after being
defected by target nuclei and thus can make no further contribution to X-ray
production, this is the 'backscattering correction'. The third takes account of
attenuation of the X-rays as they emerge from a finite depth in the sample
('absorption corrections'), and the fourth allows for enhancement in X-ray
intensity, arising from fluorescence by other X-rays generated in the sample
('fluorescence corrections').
The absorption correction is generally the most important. It is dependent on the
angle between the surface of the specimen and the X-ray path to the
spectrometer, or 'X-ray take-off angle (ψ)' (figure 3.17), which is about 40° for
current models of electron probes, although earlier instruments used different
angles. The angle of incidence of the beam is also relevant, standard correction
methods assume normal incidence and modifications are required for non-
normal incidence (as sometimes applies in SEMs).
3.10.3 ZAF corrections
The acronym 'ZAF' describes a procedure in which corrections for atomic
number effects (Z), absorption (A) and fluorescence (F) are calculated separately
128
Figure 3.17 X-ray source region, with path of X-rays to the spectrometer ψ=
take-off angle).
from suitable physical models. (The atomic number correction encompasses
both the stopping power and backscattering factors). Certain specific methods of
calculating these corrections, developed in the 1960s, constitute the standard
form of the ZAF correction, which is still used quite successfully. Its main
drawback is that the absorption correction is inadequate when the correction is
large; alternative absorption correction procedures are therefore preferable.
3.10.4 Accuracy
For major elements, it is usually possible to obtain a statistical precision of ±1%
relative (2σ). However, various factors apply which limit the accuracy obtainable
in the final result. Instrumental instabilities should be less than 1%, and
uncertainty in standard compositions may be similarly small (though not
always). The largest uncertainties are generally in the matrix corrections,
especially the absorption correction when absorption is severe. Generalization is
difficult, but ±1% is attainable in favorable cases.
3.10.5 Detections limits
With decreasing concentration, statistical errors and uncertainties in
background corrections become dominant. For a concentration in the region of
100 ppm the intensity measured on the peak consists mainly of background. The
smallest detectable peak may be defined as three times the standard deviation of
the background count [17-24].
129
3.11 EXPERIMENTAL SET UP OF ENERGY DISPERSIVE ANALYSIS OF X-RAY
(EDAX)
As growth of ZnTe thin films depends on various growth conditions, it is
essential to know about the chemical content of the grown thin films. For device
applications of these thin films their chemical content is an important parameter
to look for its influence on electrical properties such as resistivity, carrier
concentration etc. Energy Dispersive Analysis of X-ray (EDAX), a most suitable
technique for this purpose is employed in the present work and hence it is
discussed in brief along with the analysis and conclusions of investigations made
for the material in thin form used in the present thesis.
Energy Dispersive Analysis of X-ray spectroscopy (EDAX) is a chemical analysis
technique used in conjunction with scanning electron microscopy (SEM). The
EDAX technique detects X-rays emitted from the sample during bombardment by
an electron beam to characterize the elemental composition of the analyzed
volume. This technique generates a spectrum in which the peaks correspond to
specific X-ray lines and the elements can be easily identified. Quantitative data
can also be obtained by comparing peak heights or areas in the unknown with a
standard material. Data collection and analysis with EDAX is a relatively quick
and simple process because the complete spectrum of energies is acquired
simultaneously.
The essential feature of analysis is the localized excitation of a small area at the
sample surface by a finely focused electron beam, or probe giving a resolution of
about 1 µm. The energy of the beam is typically in the range 10-20keV. This
causes X-ray to be emitted from the point of the material. The energy of the X-
rays emitted depends on the material under examination. The X-rays are
generated in a region of about 2 microns in depth. By moving the electron beam
across the material an image of each element in the sample can be acquired. In
principle, the concentration of an element could be determined by comparison of
the intensity of a particular characteristic line from the sample to that of a known
standard, usually the pure element, under identical experimental conditions.
Analysis at a point could be carried out, or the specimen could be moved
130
continuously in one direction while the X-ray output is recorded to give the
distribution of the element.
Figure 3.18 shows an experimental arrangement of EDAX attached to a SEM
(Model: JSM-5610LV Make: JEOL, Japan) used in experiment study. An EDAX
system comprises four basic components that must be designed to work together
to achieve optimum results: the beam source, the X-ray detector, the pulse
processor and the analyzer. X-ray detector measures the relative abundance of
emitted X-rays versus their energy. The detector is typically lithium-drifted
silicon, a solid state surface barrier device. When an incident X-ray strikes the
detector, it creates a charge pulse that is proportional to the energy of X-ray. The
charge pulse is converted to a voltage pulse (which remains proportional to the
X-ray energy) by a charge-sensitive preamplifier. The signal is then sent to a
multichannel analyzer where the pulses are sorted by voltage. The energy, as
determined from the voltage measurement, for each incident X-ray is sent to a
computer for display and further data evaluation. The spectrum of X-ray energy
versus counts is evaluated to determine the elemental composition of the
sampled volume.
Elements of low atomic number are difficult to detect by EDAX. The Si(Li)
detector is often protected by a Beryllium window. The absorption of the
soft X-rays by the Be preclude the detection of elements below an atomic number
of 11 (Na). In windowless systems, elements with as low atomic number as 4
(Be) have been detected, but the problems involved get progressively worse
as the atomic number is reduced [25, 26].
Brief Description:
Scanning Electron Microscope XL 30 ESEM with EDAX: Resolution: upto 20A;
Acc. voltage: 30 kV; Magnification: upto 2, 50, 000x.
131
Figure 3.18 Experimental set up of Energy Dispersive analysis of X-rays
(ESEM).
Specifications:
Emission current:0 to 200 µA
Accelerating Voltage: 0.2 to 30 kV
Resolution: With LaB6 filament 2nm at 30 kV, With W filament 3.5nm at
30 kV
Magnification:10x to 400000x or higher
Automatic scaled micron marker
Eucentric goniometer stage
Specimen movement: X=50mm,Y=50mm
Rotation: n x 360 degrees
Z movement: 25mm internal & external
Secondary and back scattered electron detectors
3.12 RESULTS AND DISCUSSION
Figure 3.19 and 3.20 show EDAX spectra of ZnTe thin films deposited by SILAR
method with different thicknesses and annealing temperatures respectively. The
weight percentage of the constituent elements obtained from the EDAX of grown
132
thin films with different thicknesses and annealing temperatures is shown in
table 3.1.
3.12.1 For different thickness
The elemental composition of the ZnTe films deposited at different thicknesses
68nm, 117nm, 155nm, 186nm and 212nm at room temperature (300K) were
evaluated by EDAX technique. The percentage of Zn increases with thickness
while the amount of detected Te changes randomly.
The peak corresponding to silicon, sodium and oxygen shows a transition from L
to the K-shell which can be termed as a K-Alpha peak (Kα). The peaks
corresponding to Zn and Te show three types of transitions from L family i.e.; L-
Alpha (Lα), L-beta (Lβ), L-Gamma (Lγ). These all transitions are shown in the
EDAX patterns of ZnTe thin films of different thickness.
3.13.2 For different annealing temperature
Figure 3.20 shows the EDAX patterns of ZnTe thin films of thickness 68nm at
room temperature (300K) and annealed at 323K, 373K, 423K and 473K
temperature. The peak of Zn decrease with annealing temperature 323K and
373K while increase with 423K and 473K though the amount of detected Te
decreases. This observation is also evidenced by the improvement of hexagonal
ZnTe phase which is revealed from x-ray diffraction analysis [27]. This result is
consistent with x-ray diffraction analysis of the samples with phase correspond
to ZnTe. The peak corresponding to silicon, sodium and oxygen shows a
transition from L to the K-shell which can be termed as a K-Alpha peak (Kα). The
peaks corresponding to Zn and Te show three types of transitions – L-Alpha (Lα),
L-beta (Lβ), L-Gamma (Lγ). All these transitions are shown in the EDAX patterns
of ZnTe thin films.
The results obtained in this research for films deposited at different thicknesses
as well as for different annealing temperatures show that the films are non-
stoichiometric. Generally elements of low atomic number are difficult to detect
by EDAX. So, in the cases of ZnTe films hydrogen is difficult to detect. The
presence of Silicon and Oxygen peaks are due to the glass substrates [28] while
133
Figure 3.19 EDAX spectra of ZnTe thin films with different thickness grown
by SILAR method.
134
Figure 3.20 EDAX spectra of ZnTe thin films with different annealing
temperature grown by SILAR method.
135
the presence of Na is due to the starting material Na2Te. The O2 peak detected
from the EDAX spectrum is unavoidable in any chemically deposited thin film.
This has previously been reported [29].
Table 3.1 Chemical compositions (wt. %) of ZnTe thin films grown by SILAR
method.
Elem
ents
Wt% of elements obtained from EDAX Wt%
o
f elem
ents
calculated
Different thickness Different annealing temperature
68nm 117nm 155nm 186nm 212nm 300K 323K 373K 423K 473K
Zn 51.57 52.06 51.64 51.28 52.85 51.57 50.6 50.62 53.11 56.33 33.89
Te 43.03 42.09 43.35 43.36 42.01 43.03 44.26 44.03 42.77 39.29 66.11
O 1.87 1.12 1.08 1.17 1.14 1.87 0.97 1.02 0.67 0.71
Na 1.24 1.67 1.36 1.33 1.29 1.24 1.03 1.07 0.83 1.08
Si 2.29 3.06 2.57 2.86 2.71 2.29 3.14 3.26 2.62 2.59
3.13 ATOMIC FORCE MICROSCOPY
Atomic force microscopy is currently applied to various environments (air,
liquid, vacuum) and types of materials such as metal semiconductors, soft
biological samples, conductive and non-conductive materials. With this
technique size measurements or even manipulations of nano-objects may be
performed.
3.13.1 Introduction
Atomic Force Microscope (AFM) is one type of scanning probe microscope,
which has an ability to create three dimensional micrograph of sample surface
with resolution down to the nano-meter and angstrom scale. The AFM is one of
the foremost tools for imaging, measuring and manipulating matter at the
nanoscale. AFM is also known as scanning force microscope (SFM) because by
using AFM one can image the surface with atomic resolution and at the same
time one can also measure the force at nano-scale. This force is between the tip
and the sample surface like Vander Waal force with resolution in the range of
few nano-newtons.
The AFM is being increasingly used in solving processing and materials problems
136
in a wide range of technologies being used in electronics, telecommunications,
biological, chemical, automotive, aerospace and energy industries. The materials
being investigated include thin and thick film coatings, ceramics, composites,
glasses, synthetic and biological membranes, metals, polymers and
semiconductors. The AFM is also being applied to the studies of phenomena such
as abrasion, adhesion, cleaning, corrosion, etching, friction, lubrication, plating
and polishing.
3.13.2 Principle of AFM
Atomic Force Microscopy (AFM) measures the interaction force between the tip
and the surface. The tip may be dragged across the surface, or may vibrate as it
moves.
The interaction force will depend on the nature of the sample, the probe tip and
the distance between them. AFM provides a 3D profile of the surface on a
nanoscale, by measuring forces between a sharp probe (<10 nm) and surface at
very short distance (0.2-10 nm probe-sample separation). The probe is
supported on a flexible cantilever. The AFM tip “gently” touches the surface and
records the small force between the probe and the surface.
3.14.3 Working of an AFM instrument
The probe is placed on the end of a cantilever (which one can think of as a spring
as shown in figure 3.21). The amount of force between the probe and sample is
dependent on the spring constant (stiffness of the cantilever and the distance
Figure 3.21 Spring depiction of cantilever and SEM image of cantilever with probe (tip).
137
between the probe and the sample surface. This force can be described using
Hooke’s Law:
3.2
Where, F = Force; k = spring constant; x = cantilever deflection.
If the spring constant of cantilever (typically ~ 0.1-1 N/m) is less than surface,
the cantilever bends and the deflection is monitored. This typically results in
forces ranging from nN (10-9 ) to µN (10-6) in the open air.
The motion of the probe across the surface is controlled similarly to the STM
using feedback loop and piezoelectronic scanners (figure 3.22(A, B)). The
primary difference in instrumentation design is how the forces between the
probe and sample surface are monitored. The deflection of the probe is typically
measure by a “beam bounce” method. The beam-bounce method is now widely
used as a result of the excellent work by Alexander and colleagues [30]. In this
system an optical beam is reflected from the mirrored surface on the back side of
the cantilever onto a position-sensitive photodetector. In this arrangement a
small deflection of the cantilever will tilt the reflected beam and change the
position of beam on the photodetector. A third optical system introduced by
Figure 3.22 (A, B) Schematic of AFM instrument showing “beam bounce”
method of detection using a laser and position sensitive photodiode
detector.
A
B
138
Sarid [31] uses the cantilever as one of the mirrors in the cavity of a diode laser.
Motion of the cantilever has a strong effect on the laser output, and this is
exploited as a motion detector. A semiconductor laser diode is bounced off the
back of the cantilever onto a position sensitive photodiode detector. This
detector measures the bending of cantilever during the tip is scanned over the
sample. The measured cantilever deflections are used to generate a map of the
surface topography.
3.14 MEASUREMENT OF FORCES
The dominant interactions at short probe-sample distances in the AFM are Van
der Waals (VdW) interactions. However long-range interactions (i.e. capillary,
electrostatic, magnetic) are significant further away from the surface. These are
important in other SPM methods of analysis.
Several forces typically contribute to the deflection of an AFM cantilever which
are intermolecular microscopic forces. They are described below:
3.14.1 Intermolecular microscopic interaction
In general, the attractive interaction between two microscopic or macroscopic
bodies is
( )
3.3
where r is the distance between two bodies, C is a constant and n is integer.
Some examples of microscopic interactions are:
3.14.1.1 Coulomb interaction
Coulomb interaction exists between two isolated charged bodies. If the two
charges are q1 and q2 and distance between them is r then the interaction energy
is given by the expression
( )
3.4
This interaction is very strong.
139
3.14.1.2 Ionic interaction
Ionic interaction results from the interaction between ions. If we add or remove
the electron in a atom then what results is a ion. Interaction between ions with
charge +/- is +/-
. This interaction is attractive between ions of opposite
charge and repulsive between ions of same charge.
3.14.2 Van der Waals interaction
Van der Waals interaction includes following:
3.14.2.1 Van der Waals attractive interaction
Consider two identical inert gas atoms at a separation r in comparison with the
radius of the atoms. Then there is no electrostatic potential so we have no
electrostatic interaction, but atoms have some dipole moment and the atoms
induce dipole moments in each other and the induced moments cause an
attractive interaction between the atoms which may be given as
( )
3.5
This is the Van der Waal’s interaction or the induced dipole – dipole interaction.
It is the principle attractive interaction in crystals of inert gases.
3.14.2.2 Van der Waals repulsive interaction
As the two atoms are brought together their charge distributions gradually
overlap, thereby changing the electrostatic energy of the system. At sufficiently
closed separation, the overlap energy is repulsive in large part because of the
Pauli Exclusion Principle. The statement of this principle is that two electrons
can not have their entire quantum numbers same. When the charge distribution
of two atoms overlaps, there is a tendency of electrons from atom B to occupy in
part state of atom A, which is already occupied by electron of atom A and vice
versa. The Pauli’s principle prevents multiple occupancy and electron
distribution of atoms with closed shell can overlap only if accompanied by the
partial promotion of electrons to the unoccupied high-energy states of the atoms.
Thus the electron overlap increases the total energy of the system and gives a
140
repulsive contribution to the interaction. This repulsive potential is of the form
of
( )
3.6
where b is a constant. This repulsive interaction is called the Van der Waals
repulsive interaction.
The total potential energy of two atoms at separation r is given by
( ) [(
)
(
)
] 3.7
Where ε and σ are new parameters appear which can be expressed as 4εσ6=a,
4εσ12=b.
This potential is called the Lennard – Jones potential.
3.14.2.3 Ion – dipole interaction
The interaction between ion and dipole is given as
( )
3.8
where q is the charge of the ion, µ is the dipole moment, Ѳ is the angle between
the direction of the dipole moment and the ion.
3.14.2.4 Dipole – dipole interaction
The interaction between the dipoles are given by the relation
( ) [ ]
3.9
The force commonly used with AFM is a Van der Waals force. Force verses
distance (tip to sample) curve is shown in the figure 3.23.
3.15 PRIMARY MODES OF IMAGING
Primary modes of imaging (1) contact mode (2) Intermittent mode and (3) non-
contact mode are described below:
141
Figure 3.23 Force Vs distance (tip to sample) curve.
3.15.1 Contact Mode AFM (repulsive VdW)
When the spring constant of cantilever is less than surface, cantilever bends. The
force on the tip is repulsive. By maintaining a constant cantilever deflection
(using the feedback loops) the force between the probe and the sample remains
constant and an image of the surface is obtained.
Advantages: fast scanning, good for rough samples, used in friction analysis.
Disadvantages: at time forces can damage/deform soft samples (however
imaging in liquids often resolves this issue).
3.15.2 Intermittent mode (Tapping)
The imaging is similar to contact. However, in this mode the cantilever is
oscillated at its resonant frequency. The probe lightly “taps” on the sample
surface during scanning, contacting the surface at the bottom of its swing. By
maintaining constant oscillation amplitude a constant tip-sample interaction is
maintained and an image of the surface is obtained.
Advantages: allows high resolution of samples that are easily damaged and/or
loosely held to a surface; good for biological samples.
Disadvantages: more challenging to image in liquids, slower scan speeds needed.
142
3.15.3 Non-contact mode (attractive VdW)
The probe does not contact the sample surface, but oscillates above the adsorbed
fluid layer on the surface during scanning (all samples unless in a controlled UHV
or environmental chamber have some liquid adsorbed on the surface). Using a
feedback loop to monitor changes in the amplitude due to attractive VdW forces
the surface topography can be measured.
Advantages: Very low force exerted on the sample (10-12 N), extended probe
lifetime.
Disadvantages: generally lower resolution; contaminant layer on surface can
interfere with oscillation; usually need ultra-high vacuum (UHV) to have best
imaging.
3.16 FORCE CURVES
Force curves measure the amount of force felt by the cantilever as the probe tip
is brought close to - and even indented into - a sample surface and then pulled
away. In a force curve analysis the probe is repeatedly brought towards the
surface and then retracted, figure 3.24. Force curve analyses can be used to
determine chemical and mechanical properties such as adhesion, elasticity,
Figure 3.24 A typical force curve showing the various probe-sample
interactions.
143
hardness and rupture bond lengths. The slope of the deflection (C) provides
information on the hardness of a sample. The adhesion (D) provides information
on the interaction between the probe and sample surface as the probe is trying
to break free. Direct measurements of the interactions between molecules and
molecular assemblies can be achieved by functionalizing probes with molecules
of interest.
3.17 SURFACE ROUGHNESS PARAMETERS AND STATISTICAL MEASURES
WITH AFM
Surface roughness can be quantified using a variety of different parameters [32]
that can be divided into two groups. (1) Single values to describe the roughness
and (2) statistical measures of the roughness. Two common single value
parameters for roughness are the arithmetic roughness (Ra) and the rms
roughness (Rrms). Unfortunately, these two parameters only describe the vertical
roughness of a surface. Two surfaces with identical vertical roughness can have
very different surface morphology [35, 36], and a measure of the horizontal
roughness is needed to accurately specify their roughness distinctly [33].
Statistical measures of surface roughness provide a more complete description
of the surface than single valued parameters [32]. Two commonly used statistical
measures of surface roughness are the bearing ratio and the power spectral
density (PSD). The bearing ratio is a useful parameter for characterizing abrasive
wear of the surfaces [32] and the PSD is the power spectrum of the Fourier
transform of the surface profile [32] which provides useful information on the
geometrical structure of the surface. In particular, it is useful for detecting
periodic structure in the profile.
The meanings of all such parameters are discussed below.
3.17.1 Roughness average /arithmetic roughness
In general, the roughness average can be considered as the arithmetical mean
deviation of all the points or it can be defined as the average deviation of
roughness of all points in the profile from a mean line over the evaluation length
[37]. The arithmetic roughness Ra of a surface can be expressed as
144
∑ | |
3.10
where N is the number of data points of the profile and
3.11
Where are the data points that describe the relative vertical height of the
surface and is the mean height of the surface given by the equation
∑ | |
3.12
3.17.2 Root mean square roughness
Root mean square roughness is defined as “The average of the measured height
deviations taken within the evaluation length and length measured from the
mean line”[38]. It can be mathematically expressed as
√
∑
3.13
3.17.3 Maximum Profile Peak Height
It is defined as “the height of the maximum/highest high intensity peaks in the
roughness profile over the evaluation length” [37]. The maximum profile peak
height is represented mathematically as
| | 3.14
3.17.4 Maximum Profile Valley Depth or
It is defined as “the depth of the deepest valley in the roughness profile over the
evaluation length” [37] which may be given mathematically as
| | 3.15
3.17.5 Maximum Height of the Profile or
Maximum height of the profile basically is a measure of the maximum peak to
valley height in the roughness profile. The exact definition of it may be given as
“The absolute value between the highest and lowest peaks” [38]. The
145
mathematical expression used to calculate the parameter is
| | | | 3.16
3.17.6 The Amplitude Distribution Function (ADF)
The amplitude distribution function is a probability function that gives the
probability that a profile of the surface has a certain height z at any position x [39].
3.17.7 The Bearing Ratio Curve (BRC)
The Bearing Ratio Curve is related to the ADF. It is the corresponding cumulative
probability distribution. The bearing ratio curve is always integral (from the top
down) of the ADF [54]. The bearing ratio ( ) is defined as the length of the profile
above a horizontal line through the distribution [32] and is typically shown as a
graph in which the ordinate is the height below the highest peak in the profile.
This makes the bearing ratio curve more accurate reflection of the vertical
roughness of the surface than arithmetic or RMS roughness. The length of the
surface is a measure of the vertical roughness of the surface (of the film) whereas
the shape of the bearing ratio curve is an indication of the topography of the
surface [40]. It finds much greater use in evaluating surface finish.
3.17.8 Power Spectrum
The most important property of rough surfaces is the surface roughness power
spectrum. It determines the contact area between two solids and can provide
both lateral and longitudinal information. Thus it is more informative
measurement than all the statistical quantities.
Surface roughness is often understood as a departure of the roughness
parameters from the planarity. A convenient way of describing surface
roughness is to represent it in terms of profile height z(x, y). For typical digitized
AFM scans, the values of x and y are quantized. To determine PSD, one needs to
transform 2D AFM images from real space to reciprocal space. Two types of
transformations are used by various authors: one is the Fast Hartley transform
(FHT) [35] and the other is the Fast Fourier transform (FFT) [41-47].
Even though the 2D FFT and FHT give us a transformed version of the reciprocal
146
space, it is still difficult to make use of this 2D information. The standard method
to solve this problem is to extract a 1D magnitude of the 2D transform; that is, a
1D power spectral density is plotted against the spatial frequency. For a typical
2D AFM Scan, it consists of 250 000 (i.e. 512 × 512) data points, but the 1D PSD
requires 512 pieces of information to describe the surface roughness. This
method was proposed independently by Dumas et al. [42] and Strausser et al. [43,
48].
In AFM, one and two-dimensional power spectral density (PSD) are used to
characterize the structure of the surfaces. In the present work, (1D) PSD is
considered, as the simulations used are line scans, not two dimensional images.
The 1D PSD is given by the relation [48],
(
∫ ( ) ( ) )
3.17
where L is the length of the profile and y(x) is the profile.
The power spectral density is advantageous because it allows comparison of the
roughness data taken over various spatial frequency regions. Such methodology
also offers a convenient representation of the direct space periodicity and
amplitude of the roughness.
3.18 ATOMIC FORCE MICROSCOPE
The instrument used to obtain data/images for this work was the Veeco CP II,
which is shown in figure 3.25. The CP-II Scanning Probe Microscope (SPM)
provides high performance and value for material and physical science research.
Brief Description:
The CP-II head, designed to accept an on-axis microscope, utilizes patented Scan
Master closed-loop can linearization to image areas of up to 100 x 100 µm. The
system is designed for ease-of-use with integrated high-magnification color
premounted tips, and switching samples fast and easy. The system’s complete
library of both standard and advanced image modes, and a host of accessories
and hardware options, guarantee application flexibility and make the CP-II the
best performer of any scientific SPM.
147
Figure 3.25 Scanning probe microscope (AFM).
Specifications:
Available Modes (Contact & Non-Contact AFM, STM, LFM, EFM, MFM,
Nanolithography and Nano-manipulation)
100 µm Scanner- Z-range: 7.5 µm
DAC XY-Resolution: 0.25 Å & Z- Resolution: 0.025Å
5 µm Scanner- Z-range: 2.5 µm;
DAC XY-Resolution: 0.0013 Å & Z- Resolution: 0.009 Å with associated
image analysis software.
3.19 RESULTS AND DISCUSSION
The as-grown thin film of ZnTe was viewed under the non-contact AFM mode.
Silicon nitride tip was used to obtain topographic images of surface of the thin
film. One can also obtain various line analyses spectrums from real images by the
digital Instruments 2SPM controller using di CP-II ProScan 1.9 software.
Figure 3.26 (A, B) shows the 2D and 3D AFM images covering an area of 1µm ×
1µm of as prepared ZnTe thin film of thickness 68 nm by SILAR method at room
temperature (300 K). The images are computer generated and the original data
can be manipulated so that the surface of interest can be viewed from different
directions. Distance scales in the x, y and z directions are marked on the 3D
148
images and the z scale can be exaggerated as required.
The 3D images (maximum 256 × 256 points) do facilitate over all visualization of
the film surface. An image in a 3D format gives a rendition of what the surface
topography actually looks like. That is, the data is displayed in the x, y and z axis.
Often the scale between the x, y and z axis are not equal. The surface features are
very small with relationship to the x and y dimensions; however, in 3D image
they look large. This figure reflects that the surface possess hilly (also valley)
regions and is not absolutely flat surface [49]. The two dimensional image shows
the x and y axis and color is used to depict the height of image. An AFM image
displayed in the 2D format looks much like an image obtained from a traditional
microscope such as an optical microscope.
To know the surface morphology of ZnTe thin film, the line analysis is carried out
in detail. The line profile is a two dimensional profile or cross section extracted
from an AFM image. This may be taken horizontally, vertically or at obtuse
angles. The results of line analysis for ZnTe thin film is given in table 3.1. A line
Figure 3.26 AFM images of the as-deposited ZnTe film: (A) 2D and (B) 3D
(C) histogram (D) height profile and (E) power spectrum.
149
trace of the surface profile along the chosen traverse length can be printed out
together with the height profile, histogram, power spectrum, bearing ratio along
with various statistical parameters are mentioned in table 3.1. The line – 1, line –
2, line – 3, line – 4 and line - 5 (shown in fig. 3.15(A)) had drawn on the surface of
the film show indirectly the typical arrangement of the growth features in those
regions. The height profiles taken along the horizontal line – 1 to line - 5 of the
AFM image of ZnTe thin film is shown in table 3.1. This reveals the fact that the
difference between the peak p and the valley z or v values (Rp-v) are around120
nm (maximum) and 84.79 nm (minimum) of line – 4 and line – 3 respectively
indicating film at location of line 4 has maximum peak to valley difference than
at location of line – 3. The rms (Rq) and average (Ra) roughness values
(maximum) for line – 2 are found to be 20.15 nm and 16.69 nm, whereas for line
– 3 these values (minimum) are 13.93 nm and 10.81 nm respectively, indicating
that there is significant difference in average and rms values of peaks across
these lines while for lines 1, 4 and 5, the rms (Rq) and average (Ra) roughness
values are average around 17.70 nm and 13.96 nm respectively, comparable
with lines 2 and 3. It may be noted that these parameters indicate roughness in
vertical direction. Furthermore, the arrangement of atoms as revealed in 3D
image shows peaks and valleys. This type of arrangement of atoms is uniform
from 84.79 nm to 120 nm and is depicted in 3D image of the same film. However
these two single value parameters (i.e. Rrms and Ra), though simple and reliable,
make no distinction between peaks and valleys and do not account for the lateral
distribution of surface features.
Table 3.1 Line analyses of the AFM profiles of as deposited ZnTe film.
Line 1 Line 2 Line 3 Line 4 Line 5 Rp-v, nm 94.01 95.10 84.79 120.0 107.5 Rms Rough (Rq), nm 17.74 20.15 13.39 17.45 17.92 Ave Rough (Ra), nm 13.79 16.69 10.81 12.97 15.13 Mean Ht, nm 50.42 50.46 50.40 50.42 50.46 Median Ht, nm 46.56 46.98 50.71 46.58 47.93 Arc length, µm 5.575 5.500 5.517 5.567 5.298 Bearing [email protected]%, nm 57.24 59.71 56.71 54.83 60.48 Bearing [email protected]%, nm 36.62 33.39 38.34 38.21 33.01 Peak (Rp), nm 57.04 51.47 55.14 79.99 79.23 Valley (Rv), nm -36.98 -43.63 -29.65 -39.98 -28.24
150
Bearing ratio for line – 3 is more than line – 4 is more than line – 1 is more than
line – 2 is more than line – 5 suggesting horizontal roughness along for line – 3 is
more than line – 4 is more than line – 1 is more than line – 2 is more than line – 5.
The height distributions (Histograms) obtained from 2D AFM image of ZnTe film
is shown in figure 3.26 (c). The histogram is a continuous bar diagram in which
each column represents a height range. The height of each column represents
the number of image pixels which have a height value in a particular range. The
histogram of height distributions in figure 3.26 (c) shows that most of the
corrugation heights on film surface lie between 9nm to 10 nm above the lowest
point within the given area. The histogram indicates that the height distribution
is asymmetric.
The Mean height indicates the central value in the roughness profile over the
evaluation length and is found to be 50.42 nm, while median height is a midpoint
on the roughness profile over the evaluation length such that half of the data fall
above it and half below it which turns out to be 46.58 nm. The horizontal
measurement of roughness n(0) can be studied from height profile (figure 3.26
(D)), which indicates the number of intersections that the profile makes with the
mean line of the profile over a specified distance.
The power spectral density curves corresponding to line – 4 is shown in figure
3.26 (E). The curve indicates that the overall surface of ZnTe film is non-
homogeneous. At higher spatial frequencies the amplitude of the PSD curve
gradually decreases. This PSD curve also provide information across the entire
frequency range of the scanned surface (1µm × 1µm) including waviness (mid
frequency) and form (lower frequency). The peaks in the PSD indicate the
periodicity of the surface. The frequency of each peak gives the length that
defines this periodic surface and the spread in the peak indicates the magnitude
of the deviation from the average value. The presence of pin holes in studied
films is not observed. The surfaces of the films show crack free areas in the
explored section.
151
3.20 CONCLUSIONS
SEM micrographs of ZnTe thin films show the film to be dense and composed of
largely distributed petite rods on uniform background that was free of pits or
pinholes after annealing treatment are seen in the SEM micrographs. EDAX
suggests that the elemental compositions of the all samples are non-
stoichiometric and impurities of O, Si and Na have been detected. The peak of Zn
decrease with annealing temperature 323K and 373K while increase with 423K
and 473K though the amount of detected Te decreases. The percentage of Zn
increases with thickness while the amount of detected Te changes randomly.
Atomic force microscopy (AFM) measurements were performed to obtain the
surface roughness of as-deposited ZnTe thin film. It is observed clearly that the
film surface is not homogeneous; the RMS roughness varies between 13.39 nm to
20.15 nm. The surface morphology of the as-deposited ZnTe films appears to
change significantly as a function of the thermal annealing temperature. The
microstructural analysis of the film shows that SEM micrographs illustrate the
arising and broadening in cracks with film thickness indicating that the films are
indeed under biaxial tension. Finally, it is concluded that the overall surface of as
prepared ZnTe film investigated in present case is substantiated.
REFERENCES
1. Oatley, C.W., 1982, the early history of the scanning electron
microscope, J. Appl. Phys. 53 (2).
2. Hawkes, P., 2004, Recent advances in electron optics and electron
microscopy. Ann. Fondation Louis de Broglie 29-1.
3. Ruska, E., 1986, the development of the electron microscope and of
electron microscopy. Nobel Lecture.
4. Haguenau, F., Hawkes, P.W., Hutchison, J.L., Satiat-Jeunemaıˆtre, B., Simon,
G.T., Williams, D.B., 2003, key events in the history of electron
microscopy. Microsc. Microanal. 9, 96–138.
5. Goldstein, J., et al., 2003, Scanning Electron Microscopy and X-ray
Microanalysis, 3rd ed. Plenum Press, New York.
152
6. Breton, P.J., 1999, from microns to nanometers: early landmarks in the
science of SEM imaging, Scanning Microsc. 13-1, 1–6.
7. J. I. Goldstein, H. Yakowitz, 1977, Practical Scanning Electron Microscopy,
Plenum Press.
8. http://www.vcbio.science.ru.nl/en/fesem/eds/print/
9. Atoyan, D., 2006, Characterization of Thin-Film Zinc Telluride on Glass
Prepared by Low-temperature Nanosecond Pused-Laser Deposition,
Master Thesis.
10. Rouleau, C. M., Lowndes, D. H., McCamy, J. W., Budai, J. D., Poker, D. B.,
Geohegan, D. B., Puretzky, A. A., Zhu, S., 1995, Growth of Highly Doped p-
type ZnTe Films by Pulsed Laser Ablation in Molecular Nitrogen Applied
Physics Letter 67: pp. 2545 – 2547.
11. Z. C. Jin, I. Hamberg, and C. G. Granqvist, 1988, J. Appl. Phys. 64, 5117.
12. V. Srikant, V. Sergo, and D. R. Clarke, 1995, J. Am. Ceram. Soc. 78, 1931and
references there in.
13. V. Srikant, V. Sergo, and D. R. Clarke, 1995, J. Am. Ceram. Soc. 78, 1935.
14. V. Srikant, 1995, Ph.D. thesis, University of California, Santa Barbara.
15. J. I. Goldstein, D. E. Newbury, P. Echlin, D. C. Joy, C. Fiori, and E. Lifshin,
1981, Scanning Electron Microscopy and X-Ray Microanalysis (Plenum,
New York).
16. K.F. J. Heinrich, 1981, Electron Beam X-Ray Microanalysis (Van Nostrand
Reinhold, New York).
17. Agarwal B.K., 1991, X-ray Spectroscopy, 2nd edn, Springer-verlag, Berlin.
18. Goldstein, J. I., et al., 2003, Scanning Electron Microscopy and X-ray
Micronalysis, 3rd ed Plenum Press, New York.
19. Heinrich, K.F.J., 1981, Electron Beam X-ray Microanalysis, Van Nostrand
Rheinhold, New York.
20. Heinrich, K.F.J. and Newbury, D .E. (eds), 1991, Electron Probe
Quantitation, Plenum Press, New York.
21. Reed, S.J.B., 1993, Electron Microprobe Analysis, 2nd edn. Cambridge
University Press, Cambridge.
22. Reimer, L., 1985, Scanning Electron Microscopy, Springer-Verlag, Berlin.
153
23. Russ, J. C., 1984, Fundamentals of Energy Dispersive X-ray Analysis,
Butterworths. London.
24. Scott, V. D. and Love, G., 1994, Quantitative Electron Probe Microanalysis,
2nd edn. Ellis Horwood, Chichester.
25. J. P. Sibilia, 1998, a Guide to Materials Characterization and Chemical
Analysis, VCH Publishers.
26. D.Brondon and W.D.Kaplan, 1999, Micro structural Characterization of
Materials, John-Wileyand Sons.
27. Q.J.Feng, D.Z.Shen, J.Y.Zhang, C.X.Shan, Y.M.Lu, Y.C.Liu, X.W.Fan, 2004,
Journal of Magnetism and Magnetic Materials Vol. 279, p. 435.
28. Prabahar S. and Dhanam M., 2005, CdS thin films from two different
chemical baths-structural and optical analysis‟, J. Crys. Growth, Vol. 285,
pp. 41- 48.
29. C.D.Lokhande, B.R.Sankapal, R.S.Mane, H.M. Pathan, M.uller, M.Giersig and
V. Ganesan, 2002, Appl. Sur. Sci., 193, 1.
30. Martin A. Hubbe, Joel J. Pawlak and Alexander A. Koukoulas, 2008,
“Paper’s appearance: A review,” BioResources 3(2), 627-665.
31. Dror Sarid, Dorian Kermisch, 1979, a practical integrated optics device,
Proc. SPIE 0176, Guided Wave Optical Systems and Devices II, 41.
32. B. Nowicki, 1985, Multiparameter representation of surface
roughness, Wear 102, pp. 161-176.
33. J. M. Bennett and J. H. Dancy, 1981, Stylus profiling instrument for
measuring statistical properties of smooth optical surfaces, Applied
Optics,Vol. 20, Issue 10, pp. 1785-1802 .
34. J. C. Stover, 1990, Optical Scattering, Measurements and Analysis
(McGraw- Hill, New York.
35. S. D. Halepete, H. C. Lin, S. J. Fang and C. R. Helms, 1995, Mater. Res. Soc.
Symp. Proc.,386, 383.
36. S. J. Fang, W. Chen, T. Yamanaka and C. R. Helms, 1996, Appl. Phys. Lett.,
68, 2837.
37. Standards: ASME B46.1-1995, ASME B46.1-1985, ISO 4287-1997, ISO
4287/1-1997.
38. Standards: ASME B46.1-1995, ISO 4287-1997.
154
39. Standards: ISO 4287-1997.
40. K. L. Westra, D. J. Thomson, 1995, J. Vac. Sci. Technol. B13(2), 344.
41. C. R. Helms, 1996, In Semiconductors Characterization: Present Status
and Future Needs, edited by W. M. Bullis, D.G. Seiler, A.C. Diebold, p.110.
42. Ph. Dumas, B. Bouffakhreddine, C. Amra, O. Vatel,, E. Andre, R. Galindo and
F. Salvan, 1993, Europhys. Lett., 22, 717.
43. Y. E. Strausser, B. Doris, A. C. Diebold and H.R. Huff, 1994, Extended
Abstract of 185th Electrochem. Soc. Meeting, No., 284, 461.
44. R. M. Feenstra M.A. Lutz, F. stern, K. Ismail, P.M. Mooney, F. K. LeGoues, C.
Stanis, J. O. Chu and B. S. Meyerson, 1995, J. Vac. Sci. Technol. B13, 1608.
45. T. Yoshinobu, A. Iwamoto, Iwasaki, 1994, Jpn. J. Appl. Phys. Part. 1(33).
46. W. M. Bulliks, 1996, Micro., 14, 47.
47. D. Courboin A. Grouillet and E. Andre, 1995, Surf. Sci., 342, L1111.
48. Y. E. Strausser, 1994, Industrial Applications of Scanned Probe
Microscopy, National Institute of Standards and Technology, Gaitherburg,
MD, P.25.
49. S. E. Shaheen, C. J. Brabec, N. Serdar Sariciftci , F Pandinger, T. Fromherz
and J. C. Hummelen, 2001, Appl. Phys. Lett. 78(6), 841.