A study of transparent p type semiconducting oxide cu al o

P-type CuAlO2 thin film, with a delafossite structure and an optical band gap of 3.5 eV, was produced by Kawazoe et al.. Effect of substrate temperature during deposition is the main par

Chapter 1

Introduction

Transparent conducting oxides (TCOs) or transparent oxide semiconductors (TOS) are widely used as coating films for IR reflection and transparent electrodes in flat panel displays, functional glass, solar cells, touch panels [1-3] and other optoelectronic applications, with the purpose of producing “invisible circuits” [4] TCOs always possess wide-gap (> 3 eV) properties and show optical transparency in visible wavelength (400 to 800 nm) region New materials with high electrical conductivity and optical transparency are still being explored Most of the TCOs show n-type electrical conductivity N-type TCOs, for example, zinc oxide (ZnO), and tin oxide (SnO2) have promising electrical and optical properties with and without dopants However, p-type TCOs are needed for the fabrications of devices, especially p-n junction The availability of p-type TCOs is still limited with the absence of optimised electrical and optical properties Therefore, fabrications of p-type TCOs have been the focus of researchers recently

To date, TCO with delafossite structure has been identified to be an important candidate for p-type TCO P-type CuAlO2 thin film, with a delafossite structure and an

optical band gap of 3.5 eV, was produced by Kawazoe et al in 1997 using pulsed laser

deposition (PLD) [5] Preparations of CuAlO2 films by sputtering [6], chemical vapour deposition (CVD) [7, 8], and sol-gel [9] have also been tried However, sputtering deposition of CuAlO2 was not successful Although CuAlO2 was successfully deposited by PLD, small area coverage has been the disadvantage of PLD for

industrial production Hence, the current research work concentrates on depositing Al-O films with optimised optoelectronic properties by using sputtering technique with large area coverage Effect of substrate temperature during deposition is the main parameter to affect the properties of Cu-Al-O system This work will therefore, specifically address the effect of substrate temperature on the Cu-Al-O thin films deposited by PLD and sputtering, the effect of sputtering modes and parameters on the properties of as-deposited films, and the effect of post-deposition annealing (PDA) on the physical and electrical properties, including phase transformation, crystallinity, and electrical conductivity of the Cu-Al-O films

Cu-In order to give a comprehensive overview of the Cu-Al-O system, the effect of temperature was studied through the solid state reaction of CuO and Al2O3 powders In addition, a review of current work and literature in the TCO area is included in this chapter Subsequently, a brief introduction of PLD and sputtering will also be presented

1.1 P-type Transparent Conducting Oxides (TCOs)

1.1.1 The Need for P-type TCOs

For the fabrications of transparent devices, especially p-n junction, p-type and n-type TCOs with optimised electrical and optical properties are required Most of the useful oxide-based materials are n-type conductors that ideally have a wide band gap (> 3 eV), and the ability to be doped to degeneracy through the introduction of native or substitutional dopants, and a conduction band shape (dictating electron effective mass) that ensures that the plasmon-absorption edge lies in the infrared range [10] The most popular n-type TCOs are tin oxide (SnO2), indium tin oxide (ITO), zinc oxide (ZnO),

doped ZnO (dopants: Al, Ga, In, etc.), and cadmium stannate (Cd2SnO4) Among these n-type TCOs, ITO has been widely employed as electrodes in device fabrications [10] The high electrical conductivity of the n-type TCOs results mainly from stoichiometric deviation, where excess metal ions or oxygen vacancies supply the conduction electrons Considerable interests exist in developing p-type TCOs With the aim of fabricating transparent electronics, focus has been placed in exploring new p-type TCOs with comparable electrical and optical properties as existing n-type TCOs

1.1.2 Candidates for P-type TCOs

Currently, many experimental efforts are underway to search for new p-type TCOs

Sato et al reported the NiO films possessed p-type characteristic in 1993 with a

resistivity of 1.4×10-1 Ω-cm [11] Doping method was also employed in producing type TCOs P-type ZnO, a potential candidate for the fabrication of p-n homojuction, was produced by simultaneous addition of NH3 in carrier gas and excess Zn in the source ZnO [12] The reported resistivity was approximately 100 Ω-cm, which was too high for application in devices

p-Another p-type TCO was reported in 1997 by Kawazoe et al [5], CuAlO2, a ternary oxide with delafossite structure The p-type CuAlO2 film was prepared by laser ablation with a resistivity of 10.5 Ω-cm The discovery of CuAlO2 has placed the focus

in producing new p-type TCOs with delafossite structure from other elements (through substituting or doping)

The difficulty in fabricating p-type TCO is due to the electronic structure of oxides The valence band in oxide materials is mainly constituted by the oxygen 2p energy band However, the 2p energy levels of oxygen ions are generally lying much lower

than the valence orbitals of metallic atoms [13] As a result, a positive hole, if it is successfully introduced by doping, localizes on a single oxygen ion and cannot migrate within the crystal lattice, even under an applied electric field In other words, the positive hole constitutes a deep acceptor level, which could result in the low mobility

of holes In order to delocalize these positive holes and reduce the strong Coulomb force by oxygen ions, the valence band edge of oxide should be modulated by the covalency in the metal-oxygen bonding to induce the formation of an extended valence-band structure This can be achieved by mixing orbital of appropriate cations that have similar energy filled levels with O 2p [14]

Fig 1.1 Schematic illustration of the chemical bond between an oxygen ion and a

cation that has a closed shell electronic configuration (adapted from reference 14)

Fig 1.1 shows a schematic illustration of the necessary electronic configuration of the cationic species The cation is expected to have a closed shell electronic configuration

in order to avoid coloration Transition-metal cations with an open d shell (partially occupied) are not appropriate because of the intra-atomic excitations (d-d transitions) [14] If the energy level of the uppermost closed shell on the metallic cation is almost equivalent to those of the 2p levels of the oxygen ions, chemical bonds with considerable covalency can be formed between the metallic cations and the oxygen

ions Both of the atomic orbitals are occupied by electron pairs, and the resulting antibonding level becomes the highest occupied level, which is the valence band edge

The closed shell electronic configuration of d10s0 is reported to be a successful candidate for the construction of p-type TCOs There are numerous monovalent cations that fulfil d10s0 configuration but Cu+ and Ag+ cations are reported to be the species that satisfy the desired electronic structure Furthermore, the energy of the d10closed shell electrons for Cu+ and Ag+ cations is the highest They are expected to have higher probability to overlap with the 2p electrons on oxygen ions The overlapping can lead to the formation of covalent bond, which will give rise to a large dispersion in the valence band or reduce the localization of positive holes As for the trivalent cation

M, almost every trivalent cation (Al, Sc, Cr, Fe, Co, Ga, In, La, Y, Rh, Pr, Nd, Sm, Eu, and Tl) [15] is suitable for the fabrication of p-type AMO2 In this project, delafossite CuAlO2 will be the focus of discussion as it represents the prototype of p-type TCOs

Furthermore, an appropriate crystal structure will improve the covalent bonding between the cations and oxygen ions Tetrahedral coordination of oxygen ions leads to

no antibonding level on an oxygen ion and reduces the localization behaviour of 2p electrons on oxygen ions Cu2O is a p-type semiconducting oxide, with the oxygen ions exhibit the tetrahedral coordination The tetrahedral coordination of oxygen ions

in the structure is advantageous for p-type conductivity as the localization of the valence band edge can be reduced The valence state of the oxygen ions is sp3 in this conformation and hence the eight electrons (including 2s2) on an oxygen ion contribute

to the formation of four σ bonds with the four coordinating Cu+

cations

However, the three-dimensional interactions between 3d10 electrons of neighbouring

Cu+ ions may lead to the band gap narrowing effect This is the reason why Cu2O has a small band gap value of 2.1eV [16] and it is not transparent in visible range The band gap can be widened by lowering the three-dimensional cross-linking of the Cu+network (in the form of dumbbell units) in Cu2O to two-dimensional in delafossite structure of CuMO2 where M is a trivalent cation

Delafossite crystal structure of CuAlO2 consists of a hexagonal and layered structure The layers of Cu cations and AlO2 are stacked alternately, perpendicular to the c-axis There is no oxygen within the Cu cation layer and only two oxygen ions are linearly coordinated to each Cu cation in axial positions The AlO2 layers consist of AlO6

octahedral, sharing edges Each oxygen ion is in pseudo-tetrahedral coordination, as

Al3CuO Another expression of the crystal structure is CuAlO2, consisting of stacked layer of -O-Al-O-Cu-O- along the c-axis built by three different structural units: octahedral AlO6, linear O-Cu-O units, and hexagonal Cu layers Local symmetries around Cu+ and O2- ions in CuAlO2 phase are the same as in Cu2O, except that the nearest neighbouring cations of the oxygen ions in this structure are one Cu+ and three

Al3+

Four Cu cations coordinate with each O anion in Cu2O, whereas only one Cu cation coordinates with each O anion in CuAlO2 As a result, it is considered that mixing and interaction between Cu 3d and O 2p orbital in Cu2O is greater than in CuAlO2 However, the repulsive interactions between the Cu cations in Cu2O could reduce the band gap energy As abovementioned, the larger dispersion of the valence band results

in large mobility Lower dimensional structure of Cu cations is found in CuAlO2 Consequently, CuAlO2 has an appropriate dispersion in the valence band and a wide

band gap, which gives rise to the optical transparency in the visible light region The delafossite crystal structure of CuAlO2 is presented in Fig 1.2

Fig 1.2 The delafossite crystal structure of CuAlO2 where Cu+ cation is in two-fold linear coordination to oxygen and the Al3+ cation is in octahedral coordination (adapted from reference 17)

1.1.3 P-type Conduction Mechanism

The major charge carriers in p-type conventional semiconductors (Si and GaAs) are the positive holes introduced by doping process There is no exception for p-type TCOs The most probable origin of the positive holes is the excess oxygen There are two possible models for this excess oxygen: cation vacancy and interstitial oxygen Porat and Riess [18] proposed that the origins of the positive holes in Cu2-yO were possibly due to the ionized Cu vacancies and ionized interstitial oxygen The admixed state of Cu 3d and O 2p primarily constitutes the upper valence band, which controls the transport of positively charged holes For delafossite CuAlO2, the possible origins

of positive holes of this p-type TCO are also Cu vacancies and interstitial oxygen ions

as proposed by Kawazoe et al [14]

1.1.4 Recent Developments on P-type TCOs

Fabrication of transparent devices is a major focus in recent years N-type TCOs with excellent electrical and optical properties have been extensively achieved However, due to the lack of compatible p-type TCOs, research of new p-type transparent semiconducting oxide materials has been the focus in recent years In 1997, Kawazoe

et al [5] reported the discovery of CuAlO2 thin film, which has better optical

transmittance than NiO discovered by Sato et al [11] The films of 500 nm thick

showed 80% transmission and a conductivity of 0.95 Scm-1 Furthermore, p-type Al-O films prepared by CVD showed better electrical conductivity [7, 8] Since then, p-type delafossite TCOs such as CuGaO2 [19], CuFeO2, CuCrO2, CuScO2+x and bipolarity of CuInO2 [20] have been extensively developed Doping in p-type TCO has been tried by several research groups but there is no satisfactory result for doping of CuAlO2 and CuGaO2

Cu-Furthermore, new p-type TCOs with new crystal structure such as LaCuOS [21] and ScCu2O2 [22] have been developed with promising electrical and optical properties All these developments show a promising future of TCO materials Good quality of transparent optoelectronic devices can be produced in future

Although p-type CuAlO2 has been produced successfully by PLD [5] and Cu-Al-O has been studied by CVD [7, 8], Cu-Al-O system by PLD and sputtering techniques is not well studied Furthermore, the effects of substrate temperature and the sputtering parameters on the growth behaviour of the Cu-Al-O films are not understood The effect of post-deposition annealing on the properties of Cu-Al-O films is also not well-studied This work will therefore, specifically address these effects on the Cu-Al-O films produced by PLD and sputtering

1.2 Deposition Techniques

Thin films are thin layers of material, usually less than 1 µm thickness and sometimes

as thin as 1 nm [23] Due to their extreme thinness, thin films possess unique properties that are different from their bulk counterparts Properties such as electrical conductivity and density may show significant differences from the bulk materials Furthermore, thin films might even be regarded as ‘all surface’ as the surface properties of thin films are also significant and unique They can be produced by thinning down from the bulk materials (aluminum foil) However, it has become customary to refer to specimens prepared by thinning down specimens as thin foils and

to reserve the name thin films for specimens made by building up the material molecule by molecule Throughout the building up process, thin films are generally formed on and supported by a more substantial base or substrate since they are so thin and fragile

There are numerous methods of depositing thin films Basically, the techniques can be classified into two categories, chemical vapour deposition (CVD) and physical vapour deposition (PVD) CVD has been widely employed for the thin film deposition in the semiconductor industry Thin film deposition by CVD techniques involves the molecular species in the gas phase chemically reacting at a film surface, resulting in the formation of a condensed film as well as the emission of volatile by-products For PVD, there are a lot of techniques such as thermal evaporation, sputtering, and electro-deposition Recently, pulsed laser deposition (PLD) has emerged as a novel PVD technique These techniques are generally atomic in nature, in which the films are deposited from single atoms or small clusters Any reaction (such as oxidation or nitridization) that occurs on the film surface is independent of the source process

Since the p-type Cu-Al-O films were prepared by pulsed laser deposition and sputtering, more information of these two techniques will be introduced in the following sections

1.2.1 Pulsed Laser Deposition (PLD)

Pulsed laser deposition (PLD) belongs to the physical vapour deposition (PVD) category as it fulfils the three basic requirements of PVD: a source material, a substrate, and an energy supply to transport material from the source to the substrate during film deposition The form of energy and its associated mass transfer mechanism are the unique characteristics for a particular technique For PLD, as the name suggests, a high-energy pulsed laser is used as an external power source to ablate the source or target material [24] The interaction is short but intensive, and introduces ablation via a cascade of complex events

In a PLD system, a laser beam is focused onto the target surface The type of laser used for PLD has been evolving with time Ruby and Nd: Glass lasers were used in early studies [24] They were replaced by the more reliable Nd: YAG and TEA-CO2 lasers

In recent development, short-wavelength lasers have been chosen for PLD application

to produce ablation plumes with higher plasma temperature and thin films with better crystallinity UV excimer lasers such as 248 nm KrF excimer laser and 193 nm ArF excimer laser have been employed in most of the PLD systems

In order to avoid deep surface pitting caused by repetitive ablation, the focused laser beam changes location constantly This can be done by target rotation, or beam rastering, or a combination of both The most common configuration uses rotating target holder attached to a motor that provides constant rotation speed Furthermore, a

special substrate holder that can withstand high operating temperature (up to 850°C) in

an oxidizing environment is required for the deposition of oxide films

Target-substrate geometry is a key parameter for PLD deposition and significantly affects the film quality There are several configurations such as 1) face-to-face configuration; 2) off-center configuration; 3) off-axis configuration; 4) reverse face configuration; and 5) shadow-masked configuration [24]

Material ablation induced by high-power laser radiation is a very complex process and can be described as a series of heterogeneous events The sequential steps are: (a) photon absorption; (b) molten surface layer formation; (c) vaporization and plasma formation; (d) plasma heating and recoil force induced splashing; and (e) plume expansion and molten layer cooling [24]

Congruent vaporization is difficult to be achieved due to the different volatilities (or different vapour pressures) of elements in a multiple-component target In PLD, the short thermal cycle induced by laser pulse provides a solution to produce the films that maintain the target stoichiometry Furthermore, the film crystallinity can be enhanced

in PLD Reactive deposition and multilayer growth are applicable in PLD PLD is also known for its fast turnaround time for growing a thin film of a new material starting from its powder form However, the difficulty in achieving large-area uniformity has always been PLD’s main drawback This is due to the narrow angular profile of the plume and the narrow separation between the target and substrate Both limitations pose severe problems in scaling up to large area Another disadvantage is the rough surface morphology of films produced by PLD due to “splashing” Splashing phenomenon in PLD causes the presence of micro-size particles on deposited films There are three major causes of splashing: (1) exfoliation; (2) subsurface boiling; and (3) shock-wave-induced droplet expulsion

1.2.2 Sputtering Techniques

When a solid surface is bombarded by heavy particles (usually fast ions) with sufficient energy, atoms of the surface are ejected into all directions This process, known as sputtering, is the result of a momentum transfer from the bombarding particles and has the following characteristics [25]:

(1) The angular distribution of sputtered particles depends on the direction of impinging particles

(2) Particles sputtered from single crystal targets show preferred directions

(3) Sputtering yields depend on both the particle energy and the mass

(4) Sputtered particles have higher mean velocity than the thermally evaporated particles

After being ejected by the impinging ions, the sputtered atoms must typically travel for some distance (cms or more) before they impact a sample surface to form a deposited film [26] The operating pressure for most sputtering applications ranges from 10-5 to

10-1 Torr, over which the mean free path for gas atoms varies from 500 cm down to 5

mm

The sputtering process involves the generation of a gas plasma and can be classified into four categories: 1) DC, 2) RF, 3) magnetron, and 4) reactive However, there are important variants within each category and even hybrids among the categories such as reactive RF magnetron sputtering

No matter of the mode of sputtering, plasma generation is the most important process during the deposition Plasma generation always involves an ion beam source used to produce ions that traverse the workspace to strike the target and cause sputtering The ions produced must reach the target without being diverted by collisions with other gas

atoms Therefore, the workspace is evacuated so as to avoid the collisions Inside the ion beam sources, ions are normally produced by collisions between electrons and atoms Practically Ar gas is always used as the source for plasma generation An atom consists of a nucleus containing positively charged protons and is surrounded by an equal number of negatively charged electrons which make the atom electrically neutral

If an electron collides with this atom and is travelling fast enough, there is a possibility that an electron will be torn out of the atom by the electron impact The atom will be left with a net charge and become an ion The threshold for this ionization process is usually around 10 to 15 electron volts In the ionization process, a second electron has been produced and this can be used to generate more ions and electrons Plasma, a gas possessing a substantial and equal number of electrons and positive ions, has begun to form This plasma usually exists within ion beam sources too, but it is more convenient and usual to generate the plasma within the main vacuum chamber so that the ions can

be produced close to the material to be sputtered

RF (radio frequency) sputtering is carried out in a RF diode, in which alternating voltage is applied By applying alternating voltage, the ion current density can be increased and concerns with the charging of insulating cathodes are eliminated Typical RF frequencies used range from 5 to 30 MHz and the most commonly applied frequency is 13.56 MHz, which has been reserved for plasma processing by the Federal Communications Commission This is why this technology is often known as

RF diode sputtering [24] Several advantages of RF sputtering are discussed as follows

1 RF sputtering can provide higher ionization as electrons in the plasma pick up additional energy from the oscillation at the same frequency of applied RF voltage

to the cathode The higher ionization results in higher ion current at the same applied power than in a DC diode

2 The cathode does not receive net current from the plasma The incident ions from one part of the RF cycle are compensated by the incident electrons from the other part of the cycle The cathode and anode switch places once in every RF cycle, which result in no net current or charging

3 Since the electrons in a plasma move much more rapidly than the ions, the electron bombardment rate of the cathode during the positive part of the RF cycle can greatly exceed the ion flux during the negative half-cycle If the cathode is capacitively coupled to the power supply, the net negative charge will start to look like a net negative potential on the cathode, in which the cathode (target) behaves like a DC target This phenomenon is known as negative target bias [27], which further enhances the RF sputtering efficiency

With the aid of magnetron, the plasma is confined to a region next to the cathode surface by employing permanent or electromagnets [25] The magnetic filed is parallel

to the target surface and orthogonal to the electric field This enclosure of the plasma increases the efficiency of a sputter system as the number of ions available for bombarding the target is greatly increased Furthermore, the cathode heating effect is substantially reduced and the intensified plasma leads to higher deposition rates

Furthermore, reactive sputtering is a process that the atoms sputtered from a metallic target combine with the background reactive gas molecules on the substrate surface to form thin film of compounds The reactive gas is always mixed with the inert gas such

as Ar [27] The resultant film is either a solid solution alloy of the target doped with the reactive element, or a compound of the two

1.3 Thesis Outline

In this chapter, a review of the challenges and the requirements associated with the type TCO was provided A brief introduction of the PLD and the sputtering was also presented The experimental methods in Chapter 2 describe the deposition and characterization techniques used in this work as well as the conditions used for post-deposition annealing treatments of the Cu-Al-O films Four separated chapters are devoted to present the results and discussions Chapter 3 presents the results of temperature effects on the formation and properties of Cu-Al-O by using solid state reaction The growth behaviour of Cu-Al-O thin films at different substrate temperatures prepared by pulsed laser deposition will be discussed in Chapter 4 Chapter 5 covers the discussions of physical and electrical properties of Cu-Al-O thin films prepared by sputtering of single oxide target of CuAlO2 and 2 metallic targets of

p-Cu and Al (also known as reactive sputtering) and different deposition parameters Effects of post-deposition annealing on the properties of the films prepared by PLD and sputtering are presented in Chapter 6 Finally, Chapter 7 concludes the thesis together with some suggestions for future work

[7] H Gong, Y Wang, and Y Luo, Appl Phys Lett 76 (2000) 3959

[8] Y Wang, and H Gong, Chem Vapor Depos 6 (2000) 285

[9] M Ohashi, Y Iida, and H Morikawa, J Am Ceram Soc 85 (2002) 270

[10] H L Hartnagel, A L Dawar, A K Jain, and C Jagadish, Semiconducting Transparent Thin Films (IOP Publishing, Bristol, 1995), pp 219-354

[11] H Sato, T Minami, S Takata, and T Yamada, Thin Solid Films 236 (1993) 27

[12] K Minegishi, Y Koiwai, Y Kikuchi, K Yano, and A Shimizu, J Appl Phys

36 (1997) L1453

[13] S Fraga, K M S Saxena, and J Karwowski, Handbook of Atomic Data (Elsevier Scientific Publishing Company, Amsterdam, 1976), pp 49-188

[14] H Kawazoe, H Yanagi, K Ueda, and H Hosono, MRS Bull 25 (2000) 28

[15] A Buljan, P Alemany, and E Ruiz, J Phys Chem B 103 (1999) 8060

[16] A Buljan, M Llunell, E Ruiz, and P Alemany, Chem Mater 13 (2001) 338 [17] R Nagarajan, N Duan, M K Jayaraj, J Tate, and A W Sleight, Int J Inorg

Mater 3 (2001) 265

[18] O Porat, and I Riess, Solid State Ionics 81 (1995) 29

[19] K Ueda, T Hase, K Yanagi, H Kawazoe, and M Hirano, J Appl Phys 89

(2001) 1790

[20] H Yanagi, T Hase, S Ibuki, K Ueda, and H Hosono, Appl Phys Lett 78

(2001) 1583

[21] H Hiramatsu, M Orita, M Hirano, K Ueda, and H Hosono, J Appl Phys 91

[27] M Ohring, The Materials Science of Thin Films (Academic Press, Boston, 1992),

pp 79-306

Chapter 2

Experimental Detai s

In this project, various experimental techniques have been employed They are discussed in the following sections

2.1 Synthesis of CuAlO 2 Powder and PVD Targets

2.1.1 Comminution Process of CuO and Al 2 O 3 Powder

During the ceramic processing, the powder materials must possess fine particles with uniform size that are packed densely and homogeneously in order to produce a final sintering product with high density Comminution can be described as a technique for reducing large-sized solids into smaller one (grains, particles) by mechanical forces such as crushing, grinding, and milling [1] Crushing can produce 1 mm (or fractions thereof) sized particles whereas grinding and milling are able to produce micron- to submicron-sized particles

Raw powder materials, which possess non-uniform sizes, are segregated chemically or physically Ball milling of CuO and Al2O3 mixed powder reduces the average particle size of the powder, produces a uniform size distribution, and provides effective mixing This leads to the uniform composition of the final product The tumbling media in a rotating mill produces a grinding action on the powder materials by impacting and shearing the particles on their surfaces Experimental variables include the size of the feed material, angular velocity of the mill, the size of the media relative to the size of the feed material, the loading of the mill, the relative volumes of the media to the feed

material, and, in wet milling, the viscosity of the slurry during the milling process For consistency, these variables were kept constant throughout the experiment

During the experiment, 40 ml of ethanol solution, 200 gram of zirconia balls with various sizes and shapes as the wear resistant (grinding) media, and 100 gram of the feed materials, i.e CuO and Al2O3 powder with desired molar ratio, were charged into the a mill The sum of the feedstock and the solution occupied about 50% of the mill The weight ratio of the feed materials, grinding media, and ethanol solution was kept constant throughout the project The milling of CuO and Al2O3 powder was run for 24 hours For the powder obtained after calcinations, the milling process was performed for 48 hours since the granules size was larger even after the mechanical grinding process Since ethanol was used as the solution, the slurry after milling was heated at 80°C for 5 hours in order to evaporate the ethanol (boiling point of ethanol is 78°C) Dry powder was then obtained and used for further processing and characterizations

2.1.2 Calcination of Mixed CuO and Al 2 O 3 Powder

Copper aluminium oxide (CuAlO2) was produced by calcining the mixture of CuO and

Al2O3 powder of 2-to-1 molar ratio The completeness of the reaction and the uniformity of the product depend on the particle sizes, size distribution, mixedness of reactants, duration, temperature, atmosphere, and their uniformity during the calcination The reaction temperature was determined from both the phase diagram and the DTA measurements In this experiment, the calcinations were performed at 750°C, 900°C, 1000°C, 1100°C, and 1200°C The duration for every calcination was set for 10 hours The desired reaction temperature leading to the formation of CuAlO2 was found

at 1200°C with the total time for completion of reaction was predicted to be less than

20 hours The powders after calcination were characterized by various methods and will be further discussed in later sections

During the calcination process at high temperature, shrinkage and densification of the powder were observed This could be by virtue of the atomic transport processes, which will induce changes in grain size and shape, as well as pore size and shape As a result, the originally porous powder was consolidated with the powder particles joining together into an aggregate with higher strength This was the reason accountable for the requirement of longer time for mechanical grinding and ball milling of the powder after calcinations than the initial batch of powder in order to produce a well-mixed powder with finer particle size and uniform size distribution

2.1.3 Die Pressing

Pressing is the simultaneous compaction and shaping of a powder or granular material confined in a rigid die or a flexible mould [2] In this project, the powder was filled into rigid dies with 1-inch and 2-inch diameters for PLD and sputtering targets, respectively The powder was pressed with the load of 10kN Steps in dry pressing include the filling of the die, compaction and shaping, and ejection When the powder

is compacted, it first reorganizes into a close packed structure It undergoes plastic deformation of the agglomerates to further increase the packing density when the applied pressure is increased At higher pressure, comminution products fill up the remaining pores Pressed parts must be strong enough to survive from the ejection and the subsequent handling Pressing problems are generally minimal when the bulk density of the feed in the die, called the filled density, is high [3] Dense, nearly spherical particles or granules with smooth, non-sticky surfaces that are coarser than about 40 µm, are preferred for pressing because of their good flow behaviour

2.1.4 Firing Processes

In order to develop the desired microstructure and properties, the compacted powder is heat-treated at high temperature in a kiln or furnace This process, called firing, proceeds in three stages: (1) reactions preliminary to sintering, which include binder burnout and the elimination of gaseous products of decomposition and oxidation; (2) sintering; and (3) cooling, which may include thermal and chemical annealing [1] Sintering is the term used to describe the consolidation of the product during firing Consolidation implies that particles have joined together into an aggregate that has strength as a result of elimination of pores within the product Sintering of ceramic powder particles is the most important step during powder processing It does not commonly begin until the temperature in the product exceeds one-half to two-third of the melting temperature, which is sufficient to cause significant atomic diffusion for solid-state sintering Sintering rate is approximately proportional to the inverse of the particle size Hence, good control of the preparation and compaction processes is very important [4] Sintering temperature of CuAlO2 was 1200°C with heating and cooling rate of 5°C per minute The sintering duration was set for 10 hours The cooling process is important for developing the proper oxidation states and annealing differential strain in pressed products If the cooling rate is too fast, the sintered product could crack easily

Finally, all the preparation processes can be summarized in the flow diagram shown in next page

CuO + Al2O3 powder 1) Ball Milling Slurry of mixed powder 2) Heat at 80°C

Dry powder mixture

3) Characterizations 4) Calcination in an alumina dish

Bulk coagulated powder

5) Mechanical grinding 6) Characterizations

Non-CuAlO2powder CuAlO2 powder

7) Second ball milling for 48 hours

Back to Step 1) CuAlO2 with finer particle size

8) Pressing 9) Firing

PVD Target

2.2 Differential Thermal Analysis (DTA)

Differential thermal analysis (DTA) is a method that the temperature difference between the sample and the reference (thermally inert material which exhibits no phase change over the temperature range of the experiment) is measured against time or temperature while the temperature of the sample in a specific atmosphere is programmed [5] Temperature changes in the sample are due to the endothermic or exothermic enthalpic transitions or reactions such as those caused by phase changes, fusion, crystalline structure inversions, boiling, melting, sublimation, vaporization,

dehydration reactions, dissociation or decomposition reactions, oxidation and reduction reactions, destruction of crystalline lattice structure, and other chemical reactions [6] Any phenomenon that produces an enthalpic change or a change in heat capacity can be detected by DTA provided that the instrument has the required sensitivity The heat conduction mechanism in DTA is by radiation of heat, hence high temperature region (up to 1400°C) can be investigated The position (on the temperature or x-axis), shape, and number of peaks in a DTA curve are used for qualitative identification of a substance The areas of the peaks, which are related to the enthalpy of the reaction, are used for quantitative estimation of the reactive substance [6] The DTA curves are always reproducible However, DTA curve of a sample can be affected by various factors (instrumental and sample-wise), hence the peak temperature and the shape of the peak are rather empirical

Before the experiment, alumina pans, which were used as reference and sample holders, were cleaned by using concentrated nitric acid (70%) and distilled water The pans were also burnt by using Bunsen burner for about 20 seconds in order to enhance the cleaning process The reference material, Al2O3, and the proportional-mixed sample were placed separately in the alumina pans The weight of the sample and the reference should be the same The two thermocouples, which measure the temperature differences, must not be in contact with each other as well as with the wall of the tube

so as to produce accurate measurement When the sample holder assembly is heated at

a programmed rate, the temperatures of both the sample and reference increase uniformly

The furnace temperature is recorded as a function of time If the sample undergoes a phase transformation, energy is absorbed or emitted, and a temperature difference between the sample and the reference (∆T) is detected The minimum temperature

difference that can be measured by DTA is 0.01 K In this project, CuO and Al2O3

powder with a molar ratio of 2-to-1 and the reference material (Al2O3), both with a weight of 18.5 mg, were placed in the alumina pans, respectively The temperature range was set from room temperature to 1400°C (maximum temperature of the DTA instrument in our department) with ramping rate of 10°C/min The DTA curves obtained from our instrument plot the temperature difference as a function of temperature since a scanning mode is used rather than an isothermal mode

2.3 Deposition Techniques

2.3.1 Substrate Cleaning

The substrates are thoroughly cleaned to remove all traces of grease or oil, which can have disastrous effect on thin film surfaces The substrates were cleaned by using the ultrasonic tank before being placed in the deposition chamber The ultrasonic cleaner uses transducers mounted to the bottom of the tank to create high frequency sound wave in the tank’s liquid The sound wave then causes tiny vacuum cavities to form in the liquid (called “cavitations”) When these cavities collapse or “implode”, large amount of energy are released Particles of dirt, stains, and other debris are removed from the surfaces immersed in the liquid

The substrates used in this project were glass and quartz The substrates were immersed in a beaker containing a particular solvent The beaker itself was placed in the ultrasonic tank filled with the de-ionized water The substrates were cleaned by using acetone, alcohol and de-ionized water for 15 minutes for each solution Thereafter, the cleaned substrates were dried with a jet of pure N2 gas The substrates were placed in the deposition chamber immediately after drying

2.3.2 Pulsed Laser Deposition (PLD) System

Cu-Al-O thin films were deposited by using a pulsed laser deposition (PLD) system located at Centre for Superconducting and Magnetic Materials (CSMM) in Department

of Physics A single-target of CuAlO2 phase was employed for the deposition The depositions were carried out in partial oxygen gas pressure environment A schematic illustration of the PLD system is given in Fig 2.1

Fig 2.1 A schematic illustration of the PLD system

The system used a non-rotatable substrate holder that was facing the target holder directly, which is known as face-to-face configuration The target holder was produced with the same size of the target, which was about 1-inch in diameter A KrF excimer laser source (Lambda Physik, 248 nm wavelength, 30 ns pulse width, known as a UV laser) was employed to ablate the CuAlO2 target The incident angle of the KrF excimer laser with respect to the target surface was around 45°

The deposition parameters such as substrate temperature, oxygen gas partial pressure, and target-to-substrate distance were monitored Furthermore, the repetition rate of the laser pulse was varied during the deposition The deposition conditions for each sample are given in Chapter 4

2.3.3 Sputtering Equipment

In this project, Cu-Al-O (CAO) thin films were deposited by using a MSS3A Sputter System by RF magnetron sputtering Metallic targets of copper (Cu) and aluminium (Al) of 99.99% purity were employed for reactive co-sputtering deposition in the environment of Ar/O2 mixture Furthermore, a single oxide target of CuAlO2 phase was fabricated and employed for the single-source sputtering deposition A schematic illustration of the main features of the sputtering system and the sputtering chamber configuration are given in Fig 2.2

Fig 2.2 A schematic illustration of the MSS3A sputtering system and the side view of

the machine chamber (adapted from manual)

In order to control the properties of the resultant films, the following parameters were closely monitored

• Discharge voltage, current, and power

The sputtering system includes three RF (0 – 600 Watts) generators and two DC (0 – 1 kW) power supplies The RF generator has an output frequency of 13.56 MHz (± 0.005%) Automatic impedance matching networks are fitted for the RF power source and are automatically selected to any of the magnetron targets or the substrate for cleaning or reverse bias etching

The three planar magnetrons are made up of cathodes (2-inch diameter) with permanent magnets placed directly behind and are water-cooled A controllable pneumatic shutter is located in front of each magnetron to block undesirable flux of atoms

• Base and working pressure

The base pressure is defined as the pressure of the sputter chamber prior to the introduction of the sputtering ambience It is important for limiting the level of undesirable impurities in the resultant film The base pressure required for the deposition was achieved through the use of a mechanical rotary pump and a turbo molecular pump Together with N2 cold trap fill and bake-out, the maximum vacuum achievable for the sputter chamber is ~10-8 Torr The pressure within the sputter chamber was monitored with a gauge controller

The working pressure is defined as the pressure of the sputter chamber prior to the initiation of the glow discharge The gas flow through the system was maintained and controlled using a Four Channel Mass Flow Controller Power Supply/Readout (Model: MKS Type 247D by MKS INSTRUMENTS) so that the flow of all gases could be controlled at one time The system allows one to monitor and provide set

point levels for the mass flow controllers (MFCs) and to provide ratio-ed set points for multiple gas control The gas ratio of the ambience inside the sputtering system was calculated by adjusting the relative flow rate of each type of gas:

Flow Rate of Gas 1 + Flow Rate of Gas 2

• Substrate temperature

The substrates were placed on a rotating stage (0 – 40 rpm) with diameter of 160

mm that acts as the anode in the sputtering system A graphite heater underneath the stage was used to vary the temperature (50 – 600°C) and was connected externally to a temperature controller (Model: 2408 Pid Controllers by EUROTHERM CONTROLS LIMITED)

• Film thickness

Based on the geometrical configurations of the magnetrons relative to the substrate stage, it can be easily deduced that the thickness of the film grown will differ depending on their placement on the stage This is due to the difference in the target-to-substrate distance at each position This is further complicated when two

or more targets are used for film growth, as it is expected that the relative compositions will also change The use of a rotatable substrate holder [7] will only solve the uniformity of the films partially

Details of the sputter conditions employed for each series of samples are given in the respective chapter

2.4 Characterizations of Thin Films

2.4.1 Thickness Measurement - Stylus Profiler and Filmetric System

Stylus Profiler

Stylus profiling is a form of contact measurements for film thickness Unlike the point probe measurement system (another form of contact measurement) which requires knowledge of the film resistivity and needs substantial contact pressure to make accurate measurements, stylus profilometers use less pressure, but require an abrupt step in height in order to measure a film thickness Stylus forces are in the order

four-of the mg, which produces little permanent damage on sample surface Although the instrument approaches Angstrom type sensitivity, it cannot measure layers of this thickness, as the convolution of the vertical step with the tip radius produces a very gradual slope, which is extremely difficult to detect

The stylus profiler (Model: Alpha –Step 500 Surface Profiler) is located at the Surface Science Lab in Department of Physics This system characterizes the surface by scanning it with a diamond stylus An inductive sensor will register the vertical motion

of the stylus This system has a 50 Å – height limit and a force range of 1.0 – 99.9 mg with resolution of 0.1 mg In addition, the data signal is digitized hence allowing easy and precise quantification of the results

Filmetric System

Unlike the stylus method mentioned previously, this method does not require a step to

be generated prior to the measurement The sample is illuminated by a halogen white light source, which possesses the wavelength ranging from 400 nm to

tungsten-3000 nm However, the effective measurement range is between 400 nm and 1013 nm The light is directed via a fibre optic cable from the light source output port The light

reflected by the sample is captured and fed back to the Filmetric unit via another cable

to the spectrometer input port The system compares the reflectivity spectrum of the sample with its internal mathematical reflectivity spectrum The calculated curve is then fitted to the measured curve and the corresponding thickness of the calculated sample is obtained The accuracy of the thickness measurement is dependent on the fitting of both curves In practical, fitting error less than 5% is considered accurate

The Filmetric system (Model: Filmetric F20) was employed A baseline was first obtained with a bare substrate of the sample By replacing the substrate with the sample, the thickness of the sample can be obtained by entering the refractive index of the sample The main advantage of this system is that the thickness of the films may be determined easily and creation of a step is not required Both stylus profiler and Filmetric system were employed in the characterizations of the films

where d is the interplanar spacing, θ is the Bragg’s angle and λ is the wavelength of the incident X-ray beam Bragg’s law illustrates that using X-ray of known wavelength,

λ, and measuring the diffraction angle, θ; the spacing of various planes in a crystal can

be determined Furthermore, the average crystallite size may also be determined from the XRD spectrum by using Scherrer equation [8],

θβ

λ

k

where t is the crystal size (nm), λ is the wavelength of the X-ray (1.54056 Å), θ is the diffraction angle, β is the integral breadth or full-width-half-maximum (FWHM) of

particular peak, and k is a constant However, this equation does not take into account

of the contribution of instrumental and strain broadening

In this project, the X-ray diffraction spectra of both powder and thin films were obtained by using a Philips X’pert-MPD (multi-purpose diffraction) X-ray diffractometer The system employed Cu as the anode material with Cu Kα radiation (λ

= 1.54056 Å) and nickel (Ni) as the filter for Cu Kβ radiation The optimum generator tension and current used for the studies were 45 kV and 40 mA, respectively In order

to obtain accurate results, the system was first calibrated using a piece of pure silicon (111) with a peak at 28.443° and then applying the appropriate offsets When scanning the thin film sample, a grazing angle of 1.5° was used so that the major intensity of the spectrum obtained would be a result of the thin film on top of the substrate, rather than that of the substrate The sample (with the film side facing upwards) was placed on top

of a glass slide and two bare substrates were placed at each side of the sample and on top of the glass slide at the clamping positions in order to ensure that the x-ray would glance off the film surface rather than the substrate Table 2.1 summarizes the experimental parameters used for powder samples and thin films

Table 2.1 Experimental parameters of X-ray diffraction scanning for both powder

samples and thin films

Parameters Powder Samples Thin Films

Scan Range 10 to 80° 20 to 80°

Detector Upper Lower

2.4.3 Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Analyzer

(EDX)

The surface topology, microstructure characterization and compositions of both powder and thin films were investigated by using a field emission gun scanning electron microscope (Model: XL30 FEG by PHILIPS), which is integrated with an energy dispersive X-ray analyzer (EDX)

For surface topology and microstructure analysis, powder samples were prepared by placing the powder on top of an adhesive carbon tape so that the powder would not spread all over the chamber For thin film samples, the deposited-substrates were placed flat in the chamber with the deposited side facing upwards The SE detector uses secondary electrons generated by the interactions between the primary electron beam and weakly bound conduction-band electrons in the atoms of the sample to produce an image [9] The characterizations of non-conductive materials can be done

by gold coating on the sample surface with gold coating system (Model: Baltec SCD 005) If non-conductive materials are subjected to an electron beam, a fraction of the current will be emitted in the opposite direction as secondary or backscattered electron current [9] The net current between primary current and SE must be earthed in order

to avoid the sample from being negatively charged up, which will act as a capacitor The change in the sample’s potential can influence the scanning beam and affect both the resolution and the quality of the image The reason is that the electrons in non-conductive materials such as polymers or ceramics are tightly bounded and unable to leave the surface

Energy dispersive X-ray (EDX) analyzer is a useful technique for x-ray analysis The fact that the spectrum of interest from 0.1 keV to the beam energy (e.g 20 keV) can be obtained in a short time (10 – 100 s), allows rapid evaluation of the sample However,

elements with low atomic number (Z < 11) are hardly detected due to the beryllium

(Be) window, which is used to isolate the cooled detector from the vacuum system

2.4.4 Atomic Force Microscopy (AFM)

Atomic force microscope is a mechano-optical instrument, which detects atomic level force (~ nN) through optical measurement of the movement of highly sensitive cantilever with a very fine tip moving along the sample surface The forces applied depend on the nature of the sample, the probe geometry, the distance between the probe and the sample, and the sample surface condition When the tip approaches the sample surface, the cantilever bends due to the atomic force between the cantilever and sample A schematic illustration of an AFM system is shown in Fig 2.3

Fig 2.3 A schematic illustration of an AFM system (adapted from manual)

There are various forces that can be sensed by AFM In non-contact mode (with the tip-sample distance greater than 10Å), van der Waals, electrostatic, magnetic, or even capillary forces are sensed by the tip through oscillations causing the reflected laser beam to oscillate and produce signals

In this project, the surface morphology of the thin film samples was studied by using

an atomic force microscope (Model: SPM-Nanoscope IIIa by DI Veeco Metrology Group) Tapping mode was employed throughout the studies of morphology The tapping mode measures the topography by tapping the surface with an oscillating tip This eliminates shear forces that will damage the surface of samples Before commencing the experiment, the cantilever on a tip holder was placed on top of the sample with the tip facing the surface of thin film The laser spot was focused on the cantilever, which was then tuned automatically The scan area for the thin film samples was 1 × 1 µm Other parameters such as scan rate, feedback sensitivity, proportional and integral gain, were adjusted accordingly during the experiment in order to obtain the best resolution for the images

2.4.5 Hall Effect Measurement

The Hall effect measurement technique has been widely employed for characterization

of electrical properties of semiconductor materials It provides the information of conductivity type, resistivity, charge carrier concentration, and carrier mobility of semiconductor materials The Hall effect is based upon the deflection of moving charge carriers [10] Consider a sample in the form of a rectangular bar as shown in

Fig 2.4(a) An electric field E is applied in the x-direction while a magnetic field B is

applied along the z-direction According to Lorentz’s law, the force acting on the

charged particle will then be given by F = q (v × B), where v is the velocity of the

particle and q is its charge Electrons and holes will be deflected in opposite directions,

resulting in the separation of charge across the two opposite sample surfaces

perpendicular to the y-axis [11] An electric field, E H, called the Hall voltage, is thus created across these sample surfaces

Fig 2.4 Sample geometries for Hall measurements (a) Bar-shaped specimen, (b) thin

film sample used in the van der Pauw method, and (c) square shaped sample (adapted from reference 11 and manual)

For the measurements of thin film samples, the van der Pauw method is applied Two common geometries for the van der Pauw method of Hall effect measurement for a thin sample are shown in Fig 2.4(b), and (c) Current is fed through contacts 3 and 4 while the voltage is measured across contacts 1 and 2 The clover-leaf and square geometries are commonly used Furthermore, the small contacts are located on the circumference of the specimen with uniform thickness in order to improve the ohmic contacts If the contacts are rectifying, the measurement will not be reliable and will cause signal distortion and minority carrier injection

Van der Pauw showed that the Hall coefficient is given by [11]:

B I

d B V B V B

I

d V B V

R H

34

12 12

34

12 12

2

)]

()([)]

0()(

=

−

where d is the thickness of the film, B is the magnetic field and I 34 is the current

flowing from contact 3 to contact 4 (see Fig 2.4(b) and (c))

The sample resistivity ρ can also be measured with the van der Pauw method In this

case two adjacent contacts such as 2 and 3 (I 23) are used as current contacts while the

two remaining contacts are used for measuring the voltage drop (V 41) The resultant

resistance is defined as R 41,23:

23 41 23 ,

Another measurement is then made in which the current is sent through contacts 1 and

3 (I13) instead while the voltage is measured across contacts 2 and 4 (V24) From the resulting resistance R24,13, together with the previously obtained R41,23, ρ can be calculated with the expression:

2ln2

)(R24,13 R41,23 f

=π

where f is a factor that depends on the ratio R24,13/R41,23; f is equal to 1 when the ratio is

exactly 1 and decreases to 0.7 when the ratio is 10 A large value for this ratio is undesirable and suggests that either the contacts are bad or the sample is inhomogeneously doped [11]

In this project, the electrical properties of Cu-Al-O thin films were characterized by using Hall effect measurement system (Model: HL 5500 PC by BIO-RAD Semiconductor Division) with van der Pauw square shaped geometry The four contacts (indium) were placed at the circumferences of the samples in the square geometry in order to minimize the misalignment voltages Furthermore, in order to ensure that the results obtained are reproducible, every measurement was repeated for

at least three times especially for the results of Hall coefficient, carrier mobility, and carrier concentration

2.4.6 Seebeck Measurement

Seebeck measurement, also known as thermoelectric probe method, is used to determine the conductivity type of semiconductor materials by generating a temperature gradient over a sample surface from the hot to the cold probes Thermal gradients produce currents in a semiconductor; the majority carrier currents for n- and p-type materials are [12]:

dx

dT P qn

dx

dT P qp

probe positive with respect to the cold probe Analogous reasoning leads to the opposite potential for p-type samples

Seebeck measurement is effective over the 10-3 to 103 ohm-cm range The voltmeter tends to indicate n-type for high resistivity material even if the sample is weakly p-type

because the method actually determines the nµn or the pµp product With µn > µp, the

intrinsic or high resistivity material is measured n-type if n~p

2.4.7 UV-Visible Spectrophotometry

Spectrophotometry involves the measurement of light absorbed by a sample that is related to the transitions between electronic energy levels This transition generally occurs between a lone pair orbital or bonding orbital and an anti-bonding or unfilled non-bonding orbital The energy level separation of the orbitals can be related to the wavelength of the absorbed light, which has the relationship with the photon energies

In the discussion of absorption, it is assumed that the absorption is due to the presence

of absorption centers such as transition metal ions or small metal particles As a result, the degree of absorption is a function of the concentration of absorption centers, which

is taken into account in the Beer-Lambert’s Law [13]:

lc I

I0 =ε10

The optical absorption coefficient (α) can be determined from the value of absorbance

(A) and the film thickness (d) using the formula [13]:

E hv A

hv= ( − )

where A is a constant, and Eg is the optical band gap energy, m can have the value of

½ and 2 A plot of against hv near the fundamental absorption edge allows the direct energy band gap (E

hv

In this project, transmittance and absorbance of the samples were determined by

UV-visible spectrophotometer (Model: UV 1601 by SHIMADZU Asia PACIFIC PTE LTD) It is equipped with a monochromatic beam, which splits into two paths that pass through both the reference material (bare substrate) and the thin film sample Photon will be absorbed from the test sample beam when the photon energy matches a transition of the test sample The remaining electromagnetic energy then reaches a detector so that the amount of energy absorbed by the test sample is obtained by

comparing it with the reference beam The spectrum is plotted as transmittance (T) or absorbance (A) against the wavelength

2.4.8 X-ray Photoelectron Spectroscopy (XPS)

Kai Siegbahn and his research group at the University of Uppsala, Sweden, developed X-ray photoelectron spectroscopy (XPS) in mid-1960s [16] The technique was first known by its acronym ESCA (Electron Spectroscopy for Chemical Analysis) Surface analysis by XPS involves irradiating a sample in vacuum with monoenergetic soft x-rays and analyzing the energies of the emitted photoelectrons The source of primary radiation is generally Al Kα (1486.6 eV) or Mg Kα (1253.6 eV) x-rays An electron energy analyzer (of the cylindrical or hemispherical type) is used for the analysis of the electron energies, and the spectrum is obtained as a plot of intensity (or counts per second) versus the kinetic or binding energy of the electrons [17]

While the kinetic energy (KE) of the electron is the experimental quantity measured by the spectrometer, it is dependent on the energy of the x-ray source employed and is not

an intrinsic material parameter The binding energy of the electron, (BE), is the parameter that identifies the electron specifically, both in terms of its parent element and atomic energy level The binding energy (BE) of the electron is given by:

BE = hν - KE - φs (2.11) where hν is the photon energy, KE the kinetic energy of the electron, and φs the spectrometer work function

The physical basis of the XPS technique is illustrated in Fig 2.6 An energetic x-ray photon incident on the target atom excites it to a higher energy state, from which the target atom relaxes by emitting a photoelectron [18] Photoelectrons are emitted from all energy levels of the target atom, and the resultant electron energy spectrum is characteristic of the emitting atom type and may be thought as its XPS fingerprint