Page 1
1
CHAPTER 1
GALLIUM NITRIDE (GaN) : PROPERTIES, GROWTH
AND APPLICATIONS – A REVIEW
1.1 INTRODUCTION
In recent years, III nitride semiconductors have several advantages
over other wide bandgap semiconductors such as SiC and diamond. They can
be doped either as p- or as n – type. They can be grown epitaxially over large
number of substrates and monocrystalline layers. This has arisen from the
need for electronic devices capable of operation at high-temperatures,
especially emitters, which are active in the blue and ultraviolet (UV)
wavelengths. Aluminium nitride (AlN), Gallium nitride (GaN), Indium
nitride (InN) and their alloys, have direct band gaps that cover the entire
visible range from 1.1 eV for InN, 3.39 eV for GaN and 6.28 eV for AlN
(Strite and Morkoc 1992), therefore covering a part of the electromagnetic
spectrum that is not covered by conventional semiconductor technology.
Among group III nitrides, GaN is a promising material because of
its band gap of 3.4 eV which makes it the best candidate for devices operating
in the blue or UV part of the spectrum. GaN has a high chemical stability at
elevated temperatures (up to about 1173 K) and is suitable for caustic
environments, but also presents a technological challenge in device
manufacturing due to its high stability. The achievement of p-type doping in
GaN has led to excellent p-n junction LEDs and by alloying with AlN and
Page 2
2
InN, light in the violet, blue, green and yellow bands of the visible spectrum
has been obtained.
Numerous review papers (Strite et al 1993; Monemar 1998; Pearton
et al 1999; Jain et al 2000) are available concerning III-nitride growth,
properties and their applications. Table 1.1 shows the properties of group III
nitrides.
Table 1.1 Properties of group III nitrides
Properties AlN GaN InN
Band gap energy
Eg (eV) at 300 K 6.2 3.4 1.89
Eg (eV) at 5 K 6.28 3.5 --------
Temperature coefficient dE/dT (eV K-1)
-----
-6.010-4
-1.810-4
Pressure coefficient dE/dP (eVK-1)
------ 4.210-3 --------
Lattice constants
a (Å) 3.112 3.189 3.548
c (Å) 4.982 5.185 5.760
Thermal
expansion
∆a/a (K-1) 4.210-6 3.17 10-6 --------
∆c/c (K-1) 5.310-6 5.59 10-6 ---------
Thermal conductivity (W cm-1K-1) 2.85 1.3 ----------
Index of refraction η 2.15 2.33–2.67 2.80 – 3.05
Dielectric constant (εr) 8.5 ~9 --------
This chapter briefly describes about crystal structure,
physical/chemical properties and growth aspects of GaN and its important
applications as well.
Page 3
3
1.1.1 Crystal Structure of GaN
The chemical bond of group III elements with N is predominantly
covalent, that is, the constituents develop four tetrahedral bonds for each
atom. Because of the large differences in electronegativity of the two
constituents, there is a significant ionic contribution to the bond, which
determines the stability of the respective phase. GaN is known to occur in the
wurtzite as well as the zinc blende crystal structure. Such polytypism is
common in wide bandgap semiconductors. The equilibrium structure is
thought to be the wurtzite structure based on experimental results. Both
crystal structures are shown in Figure 1.1.
Figure 1.1 Crystal structures of III-nitrides (a) wurtzite and
(b) zincblende
(i) Wurtzite structure
The III-nitrides crystallise mainly in wurtzite structure, which is the
thermodynamically stable structure at ambient conditions. The wurtzite
Ga or N N or Ga (a) (b)
Page 4
4
structure has a hexagonal unit cell and thus two lattice constants, a and c. It
contains six atoms of each type. The space group for the wurtzite structure is
P63mc (C46V). This structure consists of two interpenetrating hexagonal close
packed (HCP) sub lattices, each with one type of atom, offset along the c axis
by 5/8 of the cell height (5/8 c).
(ii) Zincblende Structure
The zincblende structure has a cubic unit cell, containing four
group III atoms and four nitrogen atoms. The space group for this structure is
T2d (F43M). The position of the atoms within the unit cell is identical to the
diamond crystal structure. Both structures consist of two interpenetrating
face-centered cubic sublattices, offset by one quarter of the distance along a
body diagonal. Each atom in the structure may be viewed as positioned at the
center of a tetrahedron, with its four nearest neighbours defining the four
corners of the tetrahedron.
The wurtzite and zincblende structures are similar. In both cases,
each group III atom is coordinated by four nitrogen atoms. Conversely, each
nitrogen atom is coordinated by four group III atoms. The main difference
between these two structures is in the stacking sequence of closest packed
diatomic planes. For the wurtzite structure, the stacking sequence of (0001)
planes is ABABAB in the <0001> direction. For the zincblende structure, the
stacking sequence of (111) plane is ABCABC in the <111> direction. It is
expected that the properties of cubic GaN would be very similar to those of
wurtzite GaN, except that cubic GaN would have lower phonon scattering
(due to its being isotropic) and different bandgap as well as impurity levels.
Page 5
5
1.2 PROPERTIES OF GaN
The majority of GaN researchers are currently interested in
semiconductor device applications. Conventional wet etching techniques used
in the semiconductor processing are not successful to a great extent for GaN
device technology development. To overcome this problem, well-established
chemical etching processes are essential. Promising possibilities are various
dry etching processes, which have recently been developed and reviewed by
Pearton et al (1999) and by Mohammad et al (2000). The basic properties of
gallium nitride are given in the Table 1.2.
Table 1.2 Basic Properties of GaN
Properties Wurtzite
GaN Zincblende
GaN
Density g cm-3 6.15
Band gap energy
Eg (eV) at 300 K 3.4 3.2-3.3
Eg (eV) at 5 K 3.5
Temperature coefficient dE/dT (eV K-1) -6.0x10-4
Pressure coefficient dE/dP (eV kbar-1) 4.2x10-3
Lattice constants
a (Å) 3.189 4.503
c (Å) 5.185
Thermal expansion
a/a (K-1) 5.59x10-6
c/c (K-1) 3.17x10-6
Thermal conductivity (W cm-1 K-1) 1.3
Index of refraction (n) 2.33 (1 eV)
2.67 (3.38 eV) 2.5 (3 eV)
Dielectric constant (r) ~ 9
Optical phonon energy (meV) 91.8 91.9
Page 6
6
The lattice parameters of wurtzite GaN are a = 3.189 Å. and
c = 5.185 Å. However, for zincblende, the calculated lattice constant based on
measured bond distance in wurtzite GaN is a = 4.503 Å. Regarding structural
properties, a group theory approach predicts the optical phonon modes of
WZ-GaN contain one A1 mode, one E1 mode, two E2 modes, and B1 modes
whose wave numbers have been measured by Raman Scattering.
1.3 GROWTH OF GALLIUM NITRIDE
1.3.1 Bulk growth
Several attempts were made to synthesize GaN crystals during the
period 1930-1960 but good quality crystals of reasonable size could not be
grown. It is very difficult to employ the well-known growth methods such as
Bridgman or Czochralski for nitride crystal growth on account of the extremely
high melting temperatures and nitrogen vapour pressures required. The two
techniques available so far for bulk GaN growth are high-temperature, high
pressure solution growth (Porowski and Grzegory 1997) and sublimation. Early
attempts at growing bulk GaN single crystals met with limited success and
small crystal size, typically of the order of a few millimeters. Later, Zetterstrom
(1970) was able to increase the crystal size to 1-2 mm platelets by heating pre-
synthesized GaN crystals in an ammonia ambient at temperatures between
1423–1453 K. Karpinski et al (1984a) have grown bulk GaN single crystals
using high nitrogen ambient pressures and reported that a nitrogen pressure of
20 kbar at the temperature of 1873 K are the best conditions for GaN
crystallization from the melt. Recently, novel techniques have been developed
to grow relatively large single crystals of GaN at temperatures below the
melting point of GaN (Porowski et al 1997). Given the unusually high
temperatures and pressures for bulk GaN crystal growth, many groups have
pursued the heteroepitaxial thin film approach to obtain high quality single
crystals. From then onwards, many research groups have demonstrated the
Page 7
7
growth of GaN powders through various synthetic methods (Balkas and Davis
1996). GaN powders were synthesized by injecting NH3 into molten Ga under
atmospheric pressure in the temperature range of 1173–1253 K (Shibata
1999). Senthil Kumar et al (2000) reported the X-ray diffraction spectra
(XRD), Raman studies and X–ray photoelectron spectroscopy studies of
synthesized GaN material. The study of bulk high-quality freestanding GaN
single crystals has been slowed due to extremely high processing
temperatures and pressures involved in fabricating such crystals. Growth of
GaN epitaxial layer using hydride VPE (HVPE) was reported and
subsequently, thick layers have been grown using this method. With the
increased study of the nitrides, researchers have been able to develop the
growth processes to produce high-quality GaN thin films for device
applications. The fabrication of large-area bulk crystals has also been refined
over the past two years that motivated a call for obtaining GaN substrate.
There is a lattice mismatch as well as thermal mismatch between
the III-nitrides and the substrates on which they are grown. Due to lattice
mismatch, good epilayers could not be obtained. The interest in the growth of
III-nitride epilayers was revived by Amano et al (1988) and Dam et al (2006).
Metal organic vapour phase epitaxy (MOVPE) growth of GaN epilayers on
sapphire substrate is facilitated by first depositing a thin AlN buffer layer at
low temperature (Akasaki and Amano 1994).
1.3.2 Epitaxial Growth Techniques
The growth of epitaxial layers of III nitrides and alloys using either
MOCVD, MBE and HVPE has been reported by several authors (Walker et al
1997; Minsky et al 1998; Nasser et al 2001; Lee and Harris 1996; Topf et al
(1998), Attolini et al (2000) have grown GaN by HVPE process.
Page 8
8
Hydride Vapour Phase Epitaxy (HVPE) (Molnar et al 1997; Fornari
et al 2001), Metal Organic Chemical Vapour Disposition (MOCVD) (Amano
et al 1988; Nakamura et al 1991; Hemmingsson et al 2007) and Molecular
Beam Epitaxy (MBE) (Markus et al 1997; Ng et al 1998; Lang et al 2005) are
some of the methods used for GaN epitaxial growth. GaN native substrates
are not available in high quality and large quantity for homoepitaxial growth
and hence, heteroepitaxial growth is widely adopted for GaN. As the melting
point of GaN is very high, the growth of GaN crystals from a liquid melt is
extremely difficult and therefore, GaN is normally grown from vapour phase
containing an equilibrium mixture of nitrogen and Ga containing gas.
1.3.2.1 Hydride Vapour Phase Epitaxy (HVPE)
HVPE is one of the techniques used for GaN growth as it offers
high growth rate and bulk GaN films. It is a very attractive technique for the
production of high quality, large diameter and thick GaN layers (> 200 m in
thickness), which are used as free standing substrates. So far free standing
GaN layers with thickness of 230 m and area up to 10 mm2 have been
achieved. In a typical HVPE system designed for deposition, the precursors
are ammonia and gallium chloride (GaCl), while the carrier gas can be either
nitrogen or hydrogen. In order to provide GaCl, pure HCl in gaseous mixture
with H2 is injected separately into the reactor and put in contact for a
sufficiently long period with liquid gallium at around with 1123 K. At this
temperature, within a few seconds, the acid reacts completely with Ga and
gives GaCl plus a negligible amount of GaCl3. At this point, ammonia and
GaCl, still separated, are transported in the deposition zone of the reactor
where the substrate is kept in the range of 1173–1373 K temperature. The
schematic diagram of the HVPE system is shown in Figure 1.2.
Page 9
9
Figure 1.2 Schematic representation of HVPE technique
The type of chemical reaction which occurs in the HVPE growth is
as follows:
GaCl + NH3 GaN+ HCl+ H2 (1.1)
From this reaction it is clear that, since one of the products of the
reaction is H2, the growth rate will be lower when hydrogen is used as carrier
gas instead of nitrogen. The high growth rate is indeed one of the most
important features of this technique as it may allow the preparation of very
thick layers, which can be employed as self-standing substrates for
subsequent re-growth.
1.3.2.2 Chloride Vapour Phase Epitaxy (Cl-VPE)
Figure 1.3 shows the schematic diagram of the Cl-VPE system
(Varadarajan et al 2004). The reaction chamber is made up of a quartz tube
and is kept inside a single zone resistively heated furnace. The GaCl3 cell
Page 10
10
NH3 gas
N2 Gas
Single Zone Furnace Reactor Tube
Substrate holder
Exhaust
SV
SV
SV
MFC
MFC NH3 Purifier
GaCl3 Cell
Assembly
assembly is made up of a quartz tube and the crystalline GaCl3 in the cell is
melted in a liquid paraffin bath. The GaCl3 vapor was transported to the
reactor by the nitrogen carrier gas. The purifier system consists of molecular
sieves to remove moisture, carbon monoxide and carbon dioxide from the
ammonia gas. The outlet of the reactor is connected to dilute H2SO4 to
neutralize the unreacted ammonia gas species. After loading the substrate, the
reactor was purged with nitrogen gas.
Figure 1.3 Experimental setup of Cl-VPE system for the growth of GaN
The furnace was heated to the desired temperature of 1263 K and
GaCl3 was supplied to Thu Vj3 a[[;oed the reactor using nitrogen carrier gas.
The reaction of the growth processes is given by the following equation
GaCl3 + NH3 → GaN + 3HCl (1.2)
Page 11
11
The typical N2 flow rate for GaCl3 transport was 100 sccm. During
the processes 3.0 slm of NH3 and 2.0 slm of N2 were introduced in the reactor.
1.3.2.3 Metal Organic Chemical Vapour Deposition (MOCVD)
MOCVD is a process for the deposition of a material that utilizes
volatile metalorganic compounds to transport metallic atoms that are
relatively nonvolatile at the convenient deposition temperature. The
organometallic compounds are usually mixed with other source materials
“Hydrides” that react to form compound semiconductor film.
Growth of GaN can be performed by introducing trimethyl gallium
(TMGa) and ammonia (NH3) simultaneously into the reaction chamber loaded
with a substrate, such as sapphire and heated to elevated temperatures (usually
1073 –1273 K)
(CH3)3Ga (g)+ NH3(g) GaN (s) + 3 (CH3)H (1.3)
Ternary compounds such as AlGaN and InGaN can be obtained by
combining TMAl or TMIn simultaneously with TMGa as described in
equation (1.3). Adjusting the gas phase composition of the TMAl and TMGa
or the TMIn and TMGa controls the solid composition.
The organometallics are transported to the heated substrate by
passing the carrier gas, usually H2 or N2, through the organometallic which is
kept in a bubbler vessel at a controlled temperature to keep the compound in
molten state. The growth temperature is the important parameter to decide the
chemical reaction between the substrate and the reaction zone. Schematic
representation of the MOCVD instrument is shown in Figure 1.4.
Page 12
12
Figure 1.4 Schematic representation of MOCVD technique
1.3.2.4 Molecular beam epitaxy (MBE)
Producing high quality layers with sharp abrupt interfaces and good
control of thickness, doping and different compositions are effectively
realised by the MBE technique. Conventional MBE cannot be used to grow
GaN, because N2 cannot be dissociated by using conventional effusion cells.
Plasma source (RF or electron cyclotron resonance plasma) is used to activate
N2. Ga vapour beam from an effusion cell and the activated nitrogen are
directed towards the heated substrate forming GaN film. In addition to Ga, a
MBE chamber can be equipped with Si and Mg cells for n-and p-type doping
and In and Al for InGaN and AlGaN layer growth. Ammonia or hydrazine is
another source of activated nitrogen that can be cracked into atomic nitrogen
in high temperature cracking cell. Hydrazine can be decomposed at relatively
lower temperatures than ammonia but its extreme reactivity makes it
dangerous to handle so that it cannot replace ammonia as a source for
nitrogen. The schematic picture of the MBE system is shown in Figure 1.5.
Table 1.2 compares the advantages and disadvantages of MOCVD, MBE and
HVPE techniques.
Page 13
13
Figure 1.5 A typical MBE system configured with ammonia injector
Table 1.3 Comparison between MOCVD, MBE and HVPE
Growth Technology Advantage
Disadvantage
MOCVD
1. Atomically sharp interfaces
2. In-situ thickness monitoring
3. Very high quality film 4. High throughput
1. Lack of in–situ characterisation
2. Large quantities of NH3 are needed
3. P-type doping associated with Mg-H complex that need post growth process to activate doping process
MBE
1. Atomically sharp interfaces
2. In-situ characterisation 3. High purity growth 4. Hydrogen free
environment
1. Need ultra-high vacuum 2. Low growth rate
(1-1.5 um/h) 3. Low throughput 4. Very expensive
HVPE
1. Simple growth technique 2. Very high growth rates 3. Reasonably good quality
film 4. Bulk GaN
1. No sharp interfaces 2. Work in Hydrogen
environment 3. Extreme temperature
condition
Page 14
14
1.4 SUBSTRATES
Single crystalline III nitride films are grown heteroepitaxially on a
number of closely lattice-matched substrates.
1.4.1 Sapphire (Al2O3) and Silicon carbide (6H- SiC)
Sapphire is the most extensively used substrate material for the
growth of GaN. It is transparent, stable at high temperature and the
technology of growth of the nitrides on sapphire is quite mature. The lattice
mismatch between sapphire and GaN is ~13%. Though sapphire possesses a
substantially different lattice constant and thermal expansion coefficient from
the nitrides, it is most commonly used because of its good thermal stability
and hexagonal crystal structure. Good optical quality, availability of large size
and low cost also makes sapphire as a suitable substrate material. Nitridation
of sapphire substrate prior to AlN buffer layer growth provides the best
quality GaN layer.
Currently, Silicon Carbide (SiC) materials system is challenging the
GaN/sapphire system in both the optoelectronic and electronic arena.
(Matsushita et al 1995 and Sriram et al 1996). SiC offers a higher electrical
and thermal conductivity compared to sapphire and is available in the
hexagonal crystal structure. Its lattice mismatch with GaN is only 3.5 %, and
with AlN, the mismatch is very small. Despite these advantages, SiC suffers
from being substantially more expensive compared to sapphire. The
prohibitive cost of using SiC has limited its usefulness and availability to only
a small number of research groups.
Page 15
15
1.4.2 Other Substrate Materials
In spite of rapid improvements in the growth and materials quality
of GaN on sapphire, a variety of other substrate materials have also been
investigated. In the early stages, many researchers explored the growth of
GaN on GaAs substrates in order to obtain the metastable zincblende phase of
GaN (Okumura et al 1991; Kikuchi et al 1994). Other substrates, which were
used in order to achieve the zincblende GaN phase include MgAl2O3 (Powel
et al 1990) and Si (Lei et al 1991).
1.4.3 Buffer Layers
The heteroepitaxial growth of GaN thin films typically uses various
techniques to minimize the generated defects during deposition. These
approaches focus primarily on optimizing a buffer layer between the main
GaN epilayer and sapphire substrate. AlN buffer layers, (Amano et al 1986)
and later low-temperature GaN buffer layers (Nakamura 1991) of thickness
ranging between 50-100 nm, were recognized as a way to relieve the stress
and associated defects in GaN thin films grown on large lattice-mismatched
substrates. The buffer layer acted as a template to supply nucleation sites for
growth of the GaN epilayer.
This buffer relieves some of the large lattice mismatch that is
detrimental to achieving high-quality films. The layer also effectively absorbs
the stress created by the lattice and thermal expansion coefficient mismatch
through the generation of extended defects such as dislocations and stacking
faults. The creation of these defects helps in eliminating the significant stress
build up within the main GaN epilayer.
Page 16
16
1.5 ALLOYS AND HETEROSTRUCTURES OF GaN
By alloying AlN, GaN and InN, which is critical for making high
efficiency visible light sources and detectors. InGaAlN semiconductor alloy is
the promising material for the active layer of double heterostructure LED’s
and LDs. AlGaN can have bandgap ranging from 3.4 eV to 6.2 eV. The
compositional dependence of the lattice constant, the direct energy gap,
electrical and luminescence properties of the AlGaN alloys have been studied
(Yoshida et al 1982). Most semiconductor devices are optimised by
heterojunctions, which are commonly achieved through the use of alloys.
Many of the properties such as band gap energy, effective mass of electrons
and holes and dielectric constant are dependent on alloy compositions.
Of the lattice constant, the direct energy gap, electrical and
luminescence properties of the AlGaN alloys have been studied. The lattice
mismatch between AlN and GaN is 2.4%. InGaN can have a direct bandgap
ranging between 0.8 eV and 3.43 eV. The ternary alloy InxGa1-xN grown on
GaN is an important semiconductor heterostructure having wide applications
in the fabrication of lasers and optoelectronic devices. InxGa1-xN epitaxially
grown films with x < 0.1 have been well studied due to their potential
applications as blue emitters. The quaternary InGaAlN system is a wide direct
banad gap semiconductor and having wurtzite structure. By varying the In,
Ga, and Al mole fractions, this material allows the construction of a double
heterostructure (DH structure) even when each layer is lattice-matched
(Matsuoka et al 1992). This structure is also needed for semiconductor laser
diodes in shorter than blue wavelength regions, so this material system is
promising for high performance devices.
Page 17
17
1.6 GALLIUM NITRIDE NANOSTRUCTURES
Nanostructured semiconducting materials have had a monumental
impact on our society. They are the foundation of modern technology and the
basis for the entire information revolution. In recent years, group III–V
compound semiconductors are the choice for many optoelectronic and high
speed electronic applications mainly due to their direct band gap, high
saturation velocity and high electron mobility. GaN nanostructures have
attracted much attention because of their potential for near visible and UV
optoelectronic applications. Synthesis of GaN nanowires has been achieved
by various techniques such as laser ablation (Duan and Lieber 2000), carbon–
nanotube – confined reaction (Han et al 1997) etc. In particular, a large
number of crystalline GaN nanowires have been obtained by a catalytic
synthesis based on a vapour-liquid-solid (VLS) growth mechanism.
Srivastava et al (2005) have reported on the XRD and AFM results of GaN
nanowires grown by chemical vapour transport technique.
1.7 DOPING ISSUES OF GaN
Unintentionally doped GaN is the foremost problem in growing
device quality material. Most material grown so far has been n-type with
carrier concentrations typically in the range from 1016 to 1019 cm-3. Since no
impurities have so far been observed to account for such a large background
carrier concentrations it is thought to be due to nitrogen vacancies. To
overcome this problem, many groups tried to compensate with shallow p-type
dopants. Most potential dopants that were tried produced a highly resistive
material, but none seemed to generate p-type material. Nevertheless, doping
GaN p-type was finally accomplished by Amano et al (1989) using Mg to
grow compensated GaN and converting it to p-type material with low energy
Page 18
18
electron beam irradiation (LEEBI). Later, Nakamura et al (1992) improved
those results and were the first to report on the doping mechanism.
1.7.1 Transition metals doping in GaN
The possibility of using electrons' spins in addition to their charge
in information technology has created much enthusiasm for a new field of
electronics popularly known as "spintronics." An intensely studied approach
to obtaining spin-polarized carriers for data-storage devices is the use of
diluted magnetic semiconductors created by doping ions like Mn, Fe, or Co
having a net spin into a semiconducting host such as GaAs, ZnO, or GaN. The
interaction among these spins leads to ferromagnetic order at low
temperatures, which is necessary to create spin-polarized carriers (Figure 1.7).
The formation of magnetic semiconductors is of wide spread
interest due to the possibility of making highly efficient spin injectors. Many
research groups have demonstrated RT ferromagnetism in a variety of
wide band gap doped semiconductors, such as TiO2, ZnO and AlN
(Matsumoto et al 2001). Reed et al (2001) and Sato et al (2001) reported various
magnetic properties ranging from spin-glass-like to ferromagnetic for GaN
incorporating various concentrations of Cr, Co, Fe, Mn and Ni based on a
local spin-density approximation which assumed that Ga atoms were
randomly substituted with the magnetic atoms and did not take into account
any additional carrier doping effects.
Page 19
19
Figure 1.6 Types of semiconductors: (a) nonmagnetic, (b) Diluted
magnetic semiconductor and (c) DMS with ferromagnetic
order mediated by charge carriers
Thaler et al (2002) and Kane et al (2005) reported the magnetic and
optical properties of Ga1-xMnxN grown by metal organic chemical vapour
deposition (MOCVD) technique. Manganese doped GaN nanowires have
been of great interest recently because it was predicted that GaMnN solid
solutions might be magnetic semiconductors with a curie temperature above
RT. Recently Choi et al (2005) reported the single crystalline magnetic Mn
doped GaN nanowires have been grown without the extrinsic phase formation
in a controlled manner via a unique chloride – based chemical vapour
transport method.
1.7.2 Rare earth doping in GaN
The doping of nitrides with rare earth ion may serve as an
alternative to transition metal doping with the possibility of room temperature
ferromagnetism mediated either by carriers or by conduction through bands
induced by the rare earth ions. Rare-earth (RE) element doping of the wide
band gap semiconductors has recently attracted great attention, as light
emission extending from the infrared to the blue emission. Recently
Page 20
20
ferromagnetism with Tc above 300 K has been observed in very dilute
Gadolinium (Gd) doped GaN (Dhar et al 2006).
1.8 APPLICATIONS OF GaN
The successful development of a method for the production of
p-type GaN has resulted in the fabrication of blue light emitting diodes and
various microelectronics devices. With the fabrication of efficient blue LED's
it is now feasible to work on full color LED flat panel displays as well. The
main industries profiting from such a display are TV and computer
manufacturers. Flat panel displays could replace bulky CRT tubes and permit
large light-weight displays with superior resolution. This becomes especially
important with the current development efforts of high resolution television.
The practicability of today's laptop computers is mainly limited by their
operating time on a battery charge. Tremendous efforts are made by laptop
manufacturers to minimize the power consumption. Current full color
displays are the main user of the precious energy and with the introduction of
efficient LED screens the battery lifetime can be extended dramatically.
The great achievement as result of high brightness blue and green
LEDs is the fabrication of white LEDs. Traffic lights may prove to be a fertile
application for the blue/green GaN LEDs. Traffic lights using InGaN/AlGaN
blue-green LEDs promise to save vast amounts of energy. By combining high
power and high brightness blue,green and red LDs many kinds of applications
such as LED full coulour displays and LED white lamp are now possible with
the benefits of high reliability, high durability and low energy consumption.
Blue/violet LDs with lifetime of over 10,000 h were successfully
demonstrated (Nakamura et al 1997) and consequently commercialized.
Implementation of blue LDs into CD/DVD read/write systems in computers
Page 21
21
had increased the data storage capacity by five times. This enhancement will
become ever larger when using UV LDs instead.
Several high performance electronic devices such as Schottky
diodes, JFETS, HFETS, MESFETS and MOSFETS with promising results
have been demonstrated based on GaN and its alloys (Asif Khan et al 1993;
Carrano et al 1998). UV devices could also take advantage of the properties of
the earth's ozone layer, which nearly completely absorbs radiation in
the 240-280 nm (~4.75 eV) band. This would permit space to space
communication undetectable from earth and surveillance of objects leaving
the atmosphere on the dark side of the earth. Like most wide band gap
semiconductors, the nitrides are expected to exhibit superior radiation
hardness compared to GaAs and Si, which also makes them attractive for
space applications.
1.9 ION IRRADIATION: A REVIEW
Most of the interesting properties of matter in the solid state are
related to the presence of defects and impurities. The imperfections in solids
differentiate the real solids from an ideal crystalline structure. In
semiconductors the defects are introduced due to either thermodynamic
considerations or the presence of impurities during the crystal growth process.
The defects can also be introduced in the crystalline semiconductors by ion
irradiation. Ion beam processing of semiconductors is an integral part of
device fabrication process in the semiconductor industry. The defects in
perfect solids may be considered as a perturbation in the local symmetry
around the atoms. In general, defects in crystalline semiconductors can be
characterized as point defects or extended defects. They can be due to
Page 22
22
chemical impurities, interstitials, vacancies, vacany-impurity complex or due
to internal surfaces, glide planes, dislocations and grain boundaries.
1.9.1 Energy loss and damage
After penetrating inside materials, ions lose their kinetic energy
through collisions with (1) nuclei, (2) bound electrons of the target atoms and
(3) free electrons inside the target material. Ions penetrate some distance
inside the material until they stop. In case of high energy ions, the ions slow
down mainly by electronic energy loss in the beginning of the slowing down
process and they move atoms in a straight path. When the ions are slowed
down sufficiently, the collision with nuclei (the nuclear stopping) become
more and more probable, and eventually ions are stopped by nuclear
scattering. When atoms of the solid receive significant recoil energies they are
removed from their lattice positions and produce a cascade of further
collisions in the material.
The calculation of range requires the knowledge of the rate of
energy loss of ions. According to the classical scattering theory, the
interaction of the moving ions with the target atoms is described assuming
two separate processes, collisions with nuclei and collision with electrons.
The former is due to the coulomb repulsion between the ion and the target
nuclei. The nuclear stopping component is usually considered separately
because the heavy recoiling target nucleus can be assumed to be unconnected
from its lattice during the passage of the ion. The elastic recoil energy
transferred to it can be treated simply as the elastic scattering to two heavily
screened particles. Excitations or ionization of electrons are only a source of
energy loss and do not influence the collision geometry. This is justified if the
energy transferred to the electrons is small compared to the exchange of
Page 23
23
kinetic energy between the atoms, a condition usually fulfilled in ion
implantation. The ion is thus deflected by nuclear encounters and
continuously loses energy to the electrons.
1.9.2 Ion beam and Semiconductors
Ion beam irradiation can play an important role in the study of
defects related to semiconductors, as controlled amount of defects can be
introduced by selecting suitable irradiation parameters. Defects like point
defects or extended defects can also be selectively introduced by proper
choice of ion mass and energy of the irradiation ions.
Ion beams can be classified into three broad groups on the basis of
energy and mass of the irradiated ions.
1.9.2.1 Irradiation with low energy ions
Low energy ions in the range of 50 keV to 500 keV are extensively
used for doping in semiconductors. Low energy light ions produce point
defects and heavy ions produce extended defects in materials but range of
both light and heavy ions is very low as compared to that in case of high
energy ions.
For the last several decades, low energy ion beams have been used
extensively for the purpose of ion implantation. Ion implantation is an
essential process for the production of modern devices and integrated circuits
(IC) based on Si and compound semiconductors. In the case of III-V
compound semiconductors, there are two important applications. The first is
the implantation of dopants to create n-or p-type III-V semiconductors. The
Page 24
24
second is the implantation of ions like H, He, B, O, etc to convert a doped
layer to a highly resistive one through defect engineering (Ahmed et al 2001).
Ion beam interaction in semiconductors with ions having energy, such that
during slowing down, the energy of the ion is dominantly lost through nuclear
energy loss process is well understood. During slowing down, an ion interacts
inelastically with electrons and elastically with other target atoms. If the
kinetic energy, E1, transferred to the host atom is higher than the displacement
threshold energy Ed (~8eV for Ga and ~6eV for As), the knock-on atom
leaves its lattice site and depending on the residual kinetic energy, E1-Ed, they
can move for a certain path length. These atoms recoil and collide with other
atoms giving rise to further generation of collisions which produce many low-
energy recoils and induce small displacements in nearly random directions.
This sequence of collisions and displaced atom multiplication is often called
collision cascade and it lasts for ~10-13 seconds, that is the ion range divided
by the average ion velocity.
This fast process is followed by a redistribution of the energy into
surrounding material by both lattice and electron conduction, and lasts for an
additional time interval of 10-11 seconds, resulting in local thermal
equilibrium. In the following 10-9 seconds, the unstable disorder relaxes and
some ordering occurs by a local diffusion process and the system attains total
thermal equilibrium.
1.9.2.2 Irradiation with swift heavy ions (SHI)
1.9.2.2.1 Irradiation with high energy light ions (HELI)
These ions are most suitable for defect engineering, because the
point defects produced by these ions are almost uniformly distributed within
deep inside the sample and the ions get implanted at a depth of more than
Page 25
25
100 µm. This excludes the possibility of any interferences from implanted
ions in modifying the material properties. In some cases, the samples can be
made thinner than the range of ions where the ions pass through the samples
or samples may be grown on substrates such that the range of the ions is more
than the film thickness. In contrary to interaction of swift heavy ion (SHI)
where electronic energy loss above threshold causes track formation, the
damage accumulation by high energy light ions (HELI) is largely due to the
nuclear energy loss (Kamarou et al 2005). The HELI irradiation produces
point defects due to nuclear energy loss in the samples and can be estimated
using Stopping Range of Ions in Matter (SRIM) calculation. Moreover, the
electronic energy loss of HELI is very high compared to the nuclear energy
loss but much less than the threshold energy for track formation, which can be
uniquely utilized for defect engineering and material modification through
ionization of native defects.
1.9.2.2.2 Irradiation with high energy heavy ions (HEHI)
During slowing down process, low energy ions lose energy through
nuclear energy loss process. But in the case of high energy ions the electronic
energy loss dominates over nuclear energy loss. Heavy ions lead to extremely
strong electronic excitations inside a narrow cylinder around each ion path.
The initial interaction processes of the energy transfer from a high energy
heavy ion to electrons bound to inner shells take only 10-19 to 10-17 s and
slightly longer for collective electronic excitations like formation of plasmons
(Schiwietz 2004). Hence, just after the passage of the SHI, the narrow
cylindrical target zone coaxial with the ion path consists of a two component
plasma of cold lattice atom and hot electrons. Such a narrow region is often
called ionization spike.
Page 26
26
1.10 SCOPE OF THE PRESENT INVESTIGATIONS
In the last two (2006-2007) years GaN nanostructures and related
compounds have emerged as the key materials for applications in LEDs, LDs,
UV detectors and high temperature - high power electronics. There has been a
substantial worldwide interest in the development of GaN based low noise
circuit elements, particularly transistors with superior characteristics. While
the research on the current issue moves towards the realisation of better
quality layers, the fundamental material studies are still in progress with many
issues not fully resolved or understood. The aim of the present investigations
is to study GaN nanostructures and Mn doped GaN nanostructures and
irradiation effects on MOCVD grown GaN epilayers.