Fabrication and properties of semiconducting germanium clathrate film( 本文(Fulltext) )

(1)

Title Fabrication and properties of semiconducting germanium clathrate film( 本文(Fulltext) )

Author(s) RAHUL KUMAR

Report No.(Doctoral

Degree) 博士(工学) 工博甲第611号

Issue Date 2021-12-31

Type 博士論文

Version ETD

URL http://hdl.handle.net/20.500.12099/87497

※この資料の著作権は、各資料の著者・学協会・出版社等に帰属します。

(2)

Fabrication and Properties of

Semiconducting Germanium Clathrate Film

RAHUL KUMAR

A THESIS SUBMITTED FOR THE AWARD OF DOCTOR OF PHILOSOPHY

Division of Environmental and Renewable Energy Engineering Systems, Graduate School of Engineering

Gifu University, Japan 501-1193

December 2021

(3)

i

LIST OF ABBREVIATIONS AND SYMBOLS

a-Ge Amorphous Germanium

α-Ge Cubic diamond structured Germanium

RF Radio frequency

DC Direct current

pVEAS Portable vacuum annealing and evaporation system UV-vis-NIR Ultraviolet-visible near-infrared

FTIR Fourier transform infrared Ge(cF136) Type-II Ge clathrate

GIXRD Grazing angle incidence X-ray diffraction FESEM Field emission scanning electron microscopy

μm Micron

Å Angstrom

nm Nanometer

°C Degree Celsius

Eg Bandgap energy

a Lattice parameter

rpm Rotation per minute

Pa Pascal

h Hour

α Absorption coefficient

h Plank’s constant

ρ Density

d Film thickness

RT Room temperature

(4)

ii

LIST OF FIGURES

Fig 1.1 (a) Crystal structure of the NaxSi46 clathrate with ܲ݉͵ത݊ symmetry.

(b) Crystal structure of the NaxSi136 clathrate with ܨ݀͵ത݉ cubic symmetry. ...5 Fig 2.1 Image of the RF sputtering system. ... 12 Fig 2.2 Schematic diagram of the setup used for two-step clathrate synthesis process. ... 13 Fig 2.3: The growth mechanism of the clathrate film in the two-step annealing method. ... 13 Fig 2.4 Schematic diagram of the “Portable Vacuum Evaporation and Annealing System (pVEAS). ... 15 Fig 2.5 Image of the pVEAS setup. ... 16 Fig 2.6 A schematic diagram of apparatus used for Na removal from the clathrate film by applying electric field. ... 18 Fig 2.7 (a) X-ray scattering by atoms at O and A [70]. (b) Schematic diagram of Bragg’s law. ... 19 Fig 2.8 X-ray diffraction geometry for the asymmetrical diffraction [71]. .... 20 Fig 2.9 X-ray diffractometer with CuKα radiation (Rigaku, Ultima VI). ... 21 Fig 2.10 NR-2100G (JASCO) with triple monochromator Raman system. ... 23 Fig 2.11 Electron-sample interaction: the different types of signals generated [82]. ... 26 Fig 2.12 The setup of scanning electron microscopy [83]. ... 27 Fig 2.13 Lorentz force (F) and Hall voltage (VH) for a bar-shaped n-type semiconductor, subjected to the electric current (I) and magnetic field (B) [84]. ... 28 Fig 2.14 Schematic diagram of a rectangular van der Pauw configuration [84]. ... 29 Fig 2.15 Schematic diagram of the Hall measurement in van der Pauw technique [84]. ...30 Fig 3.1 Photographic image of the as-prepared Ge(cF136) film on the sapphire substrate. ... 33

(8)

vi

Fig 3.2 GIXRD reflections of the as-prepared Ge(cF136) film obtained at the grazing angle of 0.1° and Rietveld analysis fitting. ... 34 Fig 3.3 Raman spectra of the as-prepared type-II Ge clathrate film with the laser incident from the front and back sides of the film sample. ... 35 Fig 3.4 GIXRD and Rietveld refinement results of the almost guest free Ge136

film. ... 37 Fig 3.5 Optical absorption spectra of the as-prepared type-II Ge clathrate film and after successive Na removal processing. The plot for the estimation of bandgap energy is shown in the inset. ...39 Fig 3.6 I-V curve obtained from Na0.38Ge136 film by van der Pauw method and four electrode system. Measurement was performed in 1-3:2-4 mode (diagonally opposite) of deposited electrode pairs (1 × 1 mm²). Current was applied through the 1-3 (black square) and 2-4 (red circle) electrode pairs alternatively, while voltage was measured from the remaining pair, respectively. ... 40 Fig 3.7 Conductivity (left side) and absorbance at 0.2 eV (right side) as a function of Na content x. Linear dependences were observed in both cases.

... 41 Fig 4.1 Synthesis method (Method-I and Method-II) with respect to the temperature of a-Ge film during deposition of Na layer. ... 44 Fig 4.2 Photographic image of type-II Ge clathrate film on the sapphire substrate. ... 45 Fig 4.3 GIXRD pattern of the type-II Ge clathrate film obtained at an angle of incidence (ω = 2°) and calculated pattern after Rietveld refinement (olive dashed line) with their difference plotted below (red line). ... 47 Fig 4.4 Raman spectra obtained from the top view as well as backside (substrate side) incidence of the laser. ... 48 Fig 4.5 (a) and (b) Plan-view SEM micrograph of type-II Ge clathrate film surface, (c) and (d) cross-sectional SEM micrograph of the film sample... 49 Fig 4.6 The optical absorption spectrum of type-II Ge clathrate film on the sapphire substrate. The periodic fluctuation found in the spectrum is due to interference. ... 50

(9)

vii

Fig 4.7 Predicted film growth process: (a) vacuum deposition of Na on a-Ge film, (b) onset of NaGe phase formation accompanied with re-evaporation of Na from the sample surface during IR lamp heating, (c) NaGe phase transformation and further re-evaporation of Na, and (d) formation of the type-II Ge clathrate film. ... 52 Fig 4.8 Chamber pressure (left stack, red) and temperature profile (right stack, olive) of nearby to film sample inside pVEAS. ... 53 Fig 4.9 GIXRD reflections (ω = 2°) obtained from the film sample after IR annealing was stopped just after the starting of unique chamber pressure increase (240 °C, 1.18 h, red curve) and at the middle of the downward slope of decreasing chamber pressure profile (240 °C, 1.8 h, black curve) of unique pressure change region. ... 54 Fig 4.10 Photographic image of the film sample prepared by pVEAS (Method-II). ... 56 Fig 4.11 Photograph of the film prepared by pVEAS (Method-II) with initial film thickness, (a) dini = 110 nm and (b) dini = 250 nm. ... 57 Fig 4.12 GIXRD reflections (ω = 2°) obtained from as-prepared Na20.18Ge136

film sample on sapphire substrate and prepared by pVEAS (Method-II) and Rietveld refinement analysis. ... 58 Fig 4.13 GIXRD reflections (ω = 2°) and Rietveld refinement of the Ge(cF136) film sample after additional vacuum annealing (300 °C, 10 minutes). ... 59 Fig 4.14 Raman spectra obtained from Na20.18Ge136 film sample prepared on sapphire substrate by pVEAS (Method-II). ... 61 Fig 4.15 (a) and (b) Plan-view SEM micrograph of Na20.18Ge136 film sample surface. ... 62 Fig 4.16 Plan-view SEM micrograph of Na2.98Ge136 film sample (240 °C, 20 h, WNa = 0.1 gm, dini = 110 nm). ... 62 Fig 4.17 The optical absorption spectrum of as-synthesized (Na20.18Ge136) film (indicated by black colour) prepared by Method-II using pVEAS and the film sample (Na0.8Ge136) after vacuum annealing at 300 °C for 10 minutes (red). ...63

(10)

viii

Fig 4.18 Optical absorption coefficient spectra of type-II Ge clathrate with film thickness 250 nm (left side) and 413 nm (right side). ... 64 Fig 4.19 Optical absorption spectra of type-II Ge clathrate film obtained from the reliable region of spectra of samples with various film thickness, and compared with a-Ge [97] and crystalline Ge [98]. ... 66 Fig 4.20 Room temperature I-V curve of Na20.18Ge136 and Na0.82Ge136 film sample prepared by pVEAS (Method-II). ... 67

(11)

ix

LIST OF TABLES

Table 3.1 Rietveld refinement results of as prepared type-II Ge clathrate film.

The space group was assumed as ܨ݀͵ത݉. ... 34 Table 3.2 Rietveld refinement results of the almost guest free type-II Ge clathrate obtained after 120 h vacuum annealing followed by 15 h electric field treatment. ... 37 Table 4.1 Rietveld refinement results of type-II Ge clathrate film prepared by pVEAS. The space group was assumed as ܨ݀͵ത݉. ... 47 Table 4.2 Rietveld refinement analysis parameters of Na20.18Ge136 film sample. ... 58 Table 4. 3 Rietveld refinement analysis parameters of Na0.82Ge136 film sample prepared by pVEAS (Method-II) and vacuum annealed for Na reduction. ... 60 Table 4.4 Hall measurement results (RT, 0.5 T) of type-II Ge clathrate film prepared by Method-I (pVEAS) & two-step method and transport properties of conventional semiconductors [101]. (*Reliability range of carrier type ~ 30%) ... 67

(12)

1

Chapter 1: Introduction

Clathrates represent the crystalline compounds whose structures are made up of three-dimensional framework of the cage like polyhedrons of one chemical species known as host lattice that encloses the atom or molecule of other chemical species, the guest lattice. The word “clathrate” originated from the Latin word clatratus meaning enclosed or protected by the crossed bars of a grating. Clathrates are also termed as guest-host complexes and these clathrates are sometimes indexed as inclusion compounds. Several families of the clathrates are well studied, having host lattices of organic or inorganic nature. The most important of them are clathrate hydrates by the number of their members and the variety of encountered structures and inside of these clathrate hydrates, face-sharing cages of solid water enclose gaseous or liquid or both types of guest species. These hydrates have been known for two centuries; however, a new series of isostructural compounds has been evidenced recently, in which elements of group IV, Si, Ge and Sn forms the host lattice and alkali-metals, alkaline-earth metals or less common metals serves as guest species. The main reason for the surprising similarities between group IV clathrates and hydrates comes from the fact that the structure of solid H2O, as well as those of C, Si, Ge and α-Sn, are tetrahedrally bonded at atmospheric pressure and this surrounding is mainly preserved in clathrate structures. This new series of clathrates are termed also termed as inorganic clathrates [1].

1.1 Group IV Inorganic Clathrates

Group IV elements possess outstanding importance in modern technology and Si serves as the backbone of the semiconductor industry. Si is widely used in the semiconductor industry and Si-based devices constitute almost 95% of the world production in comparison to other semiconductor devices. Moreover, Si is also one of the most abundant materials on the earth.

During the early days of semiconductor electronics, discrete transistors were mainly made from semiconducting materials of germanium. The lower

(13)

2

melting point of germanium enabled easy growth of a single crystal; however, impurity doping posed the major challenge. The only available option at that stage was changing the melt, with different dopants as per the requirement, and also the impurity type to achieve the desired doping concentrations.

However, this trend has changed in modern technology and lighter, cheaper, mechanically stronger and more abundant silicon has replaced germanium in semiconductor device applications. By the late 1960s, silicon was favoured for two primary reasons. Silicon possesses a larger band gap (1.1 eV) in comparison to germanium (0.66 eV), and this enables a device operating temperature above 100 °C. The upper limit of operating temperature of germanium devices is 85 °C, which is much lower than that of silicon (150 °C).

The second advantage includes the stable oxide of silicon (silicon dioxide).

This oxide serves as the perfect insulator which is chemically quite stable with high dielectric strength. Germanium does not offer any such oxides with insulating nature and oxides formed are not stable. Oxygen reacts with germanium to form a compound (GeO2), which is less stable and soluble in water. However, germanium exhibit better carrier mobility compared to silicon which is a required property for high-speed device applications [2].

The diamond structured germanium plays a key role in the thermoelectric as well as optoelectronic application to name a few. However, the lower band gap value of diamond structure germanium limits its photovoltaic application.

Furthermore, the most stable diamond cubic structured silicon is an indirect band gap semiconductor, which has limitations for light emission and absorption device applications.

The group IV materials with a different structure than the diamond structure have provided attractive research subjects for a long time.

Clathrates are one of the most investigated materials with different structures [3], [4]. Research interest in the group IV inorganic clathrate was mainly fuelled by the discovery of Si clathrate (Na8Si46 and NaxSi136, 3 ≤ x ≤11) by Kasper et al. [5] in 1965 and their structure identified by comparison with those of liquid and gas hydrates. Thereafter, clathrates containing other alkali-metals guests and group IV host lattice were studied with the first

(14)

3

physical properties investigation. At the same time, the unique ability of clathrate host lattice framework to be slightly non-stoichiometric or modified by partial substitution of group 14 elements with neighbouring ones (group 13 or 15) was demonstrated, leading to the either positively or negatively charged host lattices framework compensated by ionized guest species. This gives rise to a large variety of semi-conducting clathrates. Despite the interesting results of these studies, but on the whole, relatively little attention was paid to these clathrates compounds by the scientific community which could be considered due to the lack of any obvious potential application at that time. In the mid-80s of the last century, newly discovered fullerene forms of carbon and high transition temperature (Tc) superconducting properties of Ba-La cuprates attracted the attention of a great number of researchers on these two new and exciting research fields. Superconductivity was found in alkali-metal fullerides X3C60 (X = K, Rb, Cs) with Tc values ranging from 19 to 32 K after few years. Hence, the search of other kinds of structures which are able to exhibit similar properties and given the obvious similarities with fullerides, cage-like inorganic clathrates began to attract significantly increasing attention.

The inorganic group IV clathrates (Si, Ge or Sn) which are structurally analogous to the ice clathrate structures, exhibit unique physical and electrical properties in comparison to its diamond cubic structure counterpart. A considerable amount of research activity has been performed on the semiconductor clathrates in the last few years due to several new findings. First, (Na, Ba)xSi46 Si clathrate (type-I) with guest species (metals) inside the cages were found to be superconducting [6]. Although its transition temperature (Tc) was only 4 K, it was the first superconductor with a covalent sp³ network. Secondly, theoretical work suggested that the band gap of the pure Si clathrates is increased to the energy of photon in the optical red region of the spectrum [7]–[10]. Finally, their tunable thermoelectric and electric properties, along with low thermal conductivities presumably owing to the presence of trapped guest species inside the cages, make them a potential candidate for thermoelectric applications [11], [12]. The

(15)

4

concentration of the guest metals strongly affects the electrical conductivity and Seebeck coefficient leading to tunable behaviour. The thermal and electrical conductivities, along with the Seebeck coefficient serve as the essential parameters to determine the figure of merit of a thermoelectric material [13].

1.2 Structural Properties of Clathrates

The group IV elements can crystallize in various kinds of potential structures due to the flexibility of sp³ bonding [14]–[16]. The valency of silicon requires the formation of its four bonds and lower energy structures mainly involve four-connected networks. Thus, there exists a potential analogy between Si and SiO2 structures. SiO2 exhibits hundreds of different polymorphs and zeolites represent a class of SiO2 with four connected phases that are compared to their α-quartz ground state. The variety of zeolite phases is attributed to many ways of connecting the tetrahedra with the lower energy required for bending of bonds compared to stretching them. These structures incorporate channels and cages leading to the open regions which can accommodate guest impurities. This property is widely exploited in numerous applications of zeolites [13].

Similarly, the clathrate structure possesses the low-density sp³ framework in which face-shared polyhedral cages provide open spaces and the clathrate structure of Ge, Si and ice demonstrate the increase in the number of atoms per primitive unit cell as the diamond lattice having two atoms per cell expands into multi-atom clathrates. The variety of the combination of polyhedrons leads to the various types of structure and the polyhedral are covalently bonded to each other by shared faces. The characteristic cages which form the clathrate structure by edge-sharing pentagonal and hexagonal faces are related to the fullerene forms of carbon, the first element of group IV. The structure of fullerene is based on geometrical rules and the C60 buckyball is built upon by 12 pentagonal faces sharing edge with 20 hexagonal ones (5¹²6²⁰). The clathrate structure could be considered as the smallest possible fullerene clusters and form 3-D

(16)

5

frameworks to neutralise the strains arising due to dangling bonds, whereas all the valence electrons are engaged in single or double bonds for a C60

buckyball, which enables it to form isolated clusters [1]. The clathrate material can also be regarded as the expanded forms of germanium, silicon and tin and owing to their cage-like structure, they can be considered as nanoporous crystalline solids. The volume per framework atom for clathrate structure is as much as 15 to 20% larger compared to the diamond structure.

However, the average bond length (E-E) for the guest free Si136 (2.34 Å) [16]

and Ge136 (2.45 Å) [17] are almost similar to the ideal bond lengths of the corresponding diamond structures. Additionally, the bond angles of the type- II clathrate average close to the ideal bond angle for tetrahedral coordination (109.47°). The incorporation of guest species into the structure results in a small but significant expansion of the structure [18]–[21]. The type-II clathrate allotropes (e.g., Si136, Ge136) are energetically metastable phases with respect to the diamond structured phases. However, many studies [7], [22], [23]

suggest that the energy difference is quite low and there exist significant energy and/or kinetic barrier allowing the clathrate structure to endure [24].

Moreover, the Si136 framework has shown stable behaviour under pressure up to 11 GPa [25], [26].

Fig 1.1 (a) Crystal structure of the NaxSi46 clathrate with ܲ݉͵ത݊ symmetry. (b) Crystal structure of the NaxSi136 clathrate with ܨ݀͵ത݉ cubic symmetry.

(17)

6

The most common forms of clathrates are type-I and type-II. The type- I structure (space group: ܲ݉͵ത݊) contains two E20 cages (E = Si, Ge or Sn) and six E24 cages in the unit cell (Fig 1.1a), represented as M8E46, if the cages are fully occupied by the guest atoms (M). The type-I clathrate phase contains 46 atoms per unit cell and is a simple cubic phase. The type-II clathrate structure consists of 16 E20 (20 membered, point symmetry Ih) and 8 E28 (28 membered, point symmetry Td) cages per conventional unit cell (Fig 1.1b) [12], [27]–[34].

The open cages can accommodate guest atoms of alkali or alkaline earth metals and usually, the clathrate structure is stabilised with the aid of these guest species in the cages [4]. These guest atoms do not enter the host clathrate lattice substitutionally but are interstitial which resides inside the oversize cage formed by the framework. There exist three crystallographic sites (space group: ܨ݀͵ത݉) for the framework: 96g, 32e and 8a (Wyckoff notation) whereas guest species reside at the 8b and 16c sites, inside the E28 and E20 cages, respectively. Hence, the resulting structure is face-centered cubic structure with 136 atoms per unit cell which contains just 34 atoms in its primitive cell. The general chemical formula of type-II clathrates can be written as A8B16E136 (A: guest in E28, B: guest in E20).

1.3 Physical and Electrical Properties of Clathrates

Since the first synthesis of Si clathrates (Na8Si46 and NaxSi136) in 1965 by Kasper et al. [5], various clathrate materials have been synthesized and extensively investigated [16], [20], [21], [35]–[37]. The guest species strongly affects the properties of the clathrates and binary clathrates such as M8E46

and M24E136 show metallic properties due to electrons contributed by the guest species to the clathrate framework. The clathrate with guest atoms has been investigated due to its guest-host interaction, giving rise to interesting physical properties such as lower thermal conductivity [12], [32], superconductivity [38]–[41], phonon glass electron crystal (PGEC) [42]–[44]

and so on. Semiconductor clathrates can be realised by partially substituting the group IV atoms with group III atoms [45]–[47]. For example, in a ternary type-I clathrate Na8Ga8Si38 [45], [46], all the electrons contributed by the

(18)

7

guest species to the host clathrate framework was compensated by Ga atoms (Group III), leading to the semiconductor properties. Such semiconductor clathrates including their guest atoms have been considered mainly as thermoelectric materials based on PGEC concept. The occupancy of guest species inside the cages of type-II clathrate structure could be potentially controlled which gives rise to the wide range properties whereas there are no such reports for type-I structure. The semiconducting nature has been observed with a reduction in the occupancy of guest atoms. Guest-free clathrates are expected to show a wide bandgap energy (Eg). Theoretical studies suggested that the guest-free Si clathrates of type-I (Si46) and type-II (Si136) both have the bandgap energy ܧ_௚ ൌ ͳǤͻ in the visible wavelength region [9], [13], [48]. For the Ge clathrates of type-I (Ge46) and type-II (Ge136), 1.9 eV and 1.45 eV have been predicted, respectively [13]. Experimental studies reported the bandgap values, ܧ_௚ ൌ ͳǤͻ for Si136 [16] and ܧ_௚ ൌ ͲǤ͸ for Ge136 [17] (There have been no report on the synthesis of guest free type-I clathrate of Si or Ge). The bandgap tuning by alloyed clathrates (SixGe1-x)136

has also been suggested on the basis of theoretical calculations [49] and the bandgap range was estimated as 1.2 eV – 1.9 eV. However, considering the experimental reports of bandgap of Ge136 and Si136, the tunable range is estimated to be 0.6 eV – 1.9 eV. Theoretical calculations suggested the direct (or quasi-direct) bandgap for type-II clathrates. Although, the value of Eg has not been well established, direct bandgap corresponding to the visible wavelength serves as attractive feature for the photoelectric materials.

1.4 Synthesis of Type-II Clathrates

The type-II Ge and Si clathrates can be synthesized by the conventional method from the Zintl alloys of Na and Ge/Si in powder form [5], [20], [50]–[54]. This method involves a separate synthesis of intermediate Zintl phase of NaGe (P21/b) or NaSi (C2/c) by annealing of Na and Ge/Si in an inert atmosphere (Ar or N2) for Ge/Si clathrates, respectively. A mixture of Na and Ge/Si with a molar ratio, slightly greater than unity (Na:Ge/Si), sealed in a closed vessel inside a glove box filled with dry Ar or N2. For type-II

(19)

8

clathrate of Si, this vessel is heated at 650 °C for 48 h in an electric furnace which results in the formation of NaSi. The Zintl phase of NaSi is subsequently transferred into NaxSi136 by thermal annealing at 400-450 °C for 3-5 h under dynamic vacuum (≤ 10^-4 Pa) [53], [54]. Higher temperature, as well as longer duration of annealing, leads to a reduction in Na contents (x) in the synthesized NaxSi136. Similarly, for the preparation of type-II Ge clathrate powder, the precursor NaGe was prepared by annealing the mixture of Na and Ge in an inert atmosphere (400-500 °C) and thereafter annealed at

~300 °C in dynamic vacuum [36]. However, this method leads to the formation of the powder sample which usually consists of some impurity phases (e.g., d-Si and Na8Si46 for type-II Si clathrate) and Horie et al. [50]

suggested that the vacuum annealing condition strongly affects the formation of impurity phases.

Several other methods have been explored by researchers for the synthesis of type-II clathrate in powder form. A novel synthesis method by a chemical reaction route was also investigated [17], [55]–[58]. Guloy et al. [17]

reported the synthesis of guest-free type-II Ge clathrate by chemical reaction of the intermediate Zintl phase with ionic liquid (dodecyl trimethyl ammonium chloride and aluminium trichloride in 1:1 molar ratio) at 300 °C in inert condition. Beekman et al. [59] have synthesized the guest rich Na24Si136 by spark plasma sintering (SPS) at high pressure which was further modified by Stefanoski et al. [60] and optimal condition for the synthesis of single-crystalline Na8Si46 and Na24Si136 was reported as 200-500 μm in size.

High-pressure synthesis (at gigapascals) lead to the formation of Na30.5Si136 in which two sodium atoms occupy the same silicon cages [61].

The films of type-II silicon clathrates have been synthesized by many researchers [4], [62]–[65]. Kume et al. [4], [62] and Ohashi et al. [65] have reported the synthesis of NaxSi136 on silicon substrate by the conventional method. It involves thermal annealing of the silicon wafer with sodium in an inert atmosphere (580-600 °C, 24-48 h) using muffle furnace and subsequent annealing under dynamic vacuum (< 10^-2 Pa) in another setup for 3 h at 400 °C.

The as-prepared sample contained sodium which was removed by annealing

(20)

9

inside the vacuum-sealed glass tube containing iodine powder. Martinez et al. [63] have reported the synthesis of Na3Si136 film of about 1 μm thickness on the sapphire (Al2O3) substrate. Recently, Fix et al. [66] reported a new approach that removes the necessity of glove box for the synthesis of sodium doped silicon clathrate film on the silicon substrate. The intermediate Zintl phase of NaSi was prepared by annealing of silicon wafer with sodium inside Ar-flushed stainless-steel tube and Ar pressure of 1.6 × 10⁵ Pa was maintained inside the tube by closing the valves on each side. Subsequently, sample was annealed at 400 °C for 3 h under dynamic pressure (< 10^-5 Pa). The prepared silicon clathrate film was pressure annealed for denser film and reduction of surface roughness.

The film fabrication of type-II Ge clathrate was reported by Kume et al. [67] on a germanium substrate. A Ge(111) substrate was used as the starting material and was prepared by the two-step synthesis method. Firstly, NaGe precursor film was prepared by reaction of Na vapour with germanium substrate at 400 °C for 2 h inside a tightly sealed container and using a muffle furnace. The NaGe film thickness could be controlled by annealing duration.

In the second step, prepared NaGe film was annealed inside a quartz tube under dynamic pressure (< 10^-2Pa) at 300 °C for 12 h, which resulted in the synthesis of NaxGe136 film on a germanium substrate.

1.5 Motivation and Novelty

There have been many reports available for type-II Ge clathrates.

However, previous studies have been achieved mainly in powder form [17], [68], [69]. There are few reports of the film sample of type-II Ge clathrate, the films contained a lot of crystal defects such as voids or cracks [67] and there is no available report for the guest-free type-II Ge clathrate film to the best of our knowledge. A homogenous film sample on transparent substrate would facilitate the precise and extensive studies of the optical properties of type-II Ge clathrate. Furthermore, film sample on insulating substrate would enable the in-depth investigation of electric properties of the type-II Ge clathrate

(21)

10

and would provide the pathway for the successful integration in real devices and investigation of its performance.

Hence, the aim of this thesis work mainly includes the development of a synthesis method for uniformly grown type-II Ge clathrate film on the transparent and insulating substrate. Thereafter, the investigation of structural, optical and electrical properties of type-II Ge clathrate with respect to the concentration of guest species and the variation in film thickness. However, defects present in the film could be detrimental for the electrical studies as well as device applications such as photovoltaic, the defect density optimisation serves as the imperative.

(22)

11

Chapter 2: Synthesis Method and Characterization

Synthesis of clathrate in film form possesses major challenges. For proper investigation as well as device application, film quality should be optimised to facilitate uniform film growth impression. The film samples were investigated with various characterisation techniques. This section provides the details of the clathrate film synthesis method and the characterisation technique.

2.1 Sample Preparation

For the preparation of clathrate film sample of Si/Ge on the substrate like sapphire, silica etc; Si/Ge film had been deposited on the substrate initially. Thereafter, the deposited film was reacted with sodium (Na) and subsequent/concurrent annealing led to the formation of intermediate precursor film of NaGe. Further annealing results in the formation of the clathrate film. The step-wise explanation has been presented in the following section.

2.1.1 Ge Film Deposition by RF Sputtering

Sapphire/silica substrates were cut in the size of 20 × 10 mm² from the substrate disk supplied by the manufacturer. These substrates were washed before loading inside the radio frequency (RF) sputtering chamber. The washing was performed by ultrasonication in the solution of acetone firstly, followed by Semicoclean 56 and finally in ethanol for 10 minutes each. Each substrate was taken out from the ethanol solution slowly and during heating of the solution (≤ 50 °C) by a hotplate heater, to avoid any remaining droplet or impurity on the substrate surface. Subsequently, the substrates were transferred inside the sputtering chamber. The RF sputtering (LS-430/1.3, Biemtron) chamber (Fig 2.1) was fitted with a rotary pump (RP) and turbo molecular pump (TMP) to maintain the high pressure inside (≤ 10^-5 Pa). Ar (99.9%) gas was used as the sputtering gas for plasma creation. The flow rate of Ar gas inside the chamber was controlled by mass flow controller (MFC) and the flow rate fixed to 50 sccm. The gas pressure was maintained at 1.0 Pa

(23)

12

and pre-sputtering was performed before the sputtering on the substrate to remove any oxide layer of the target (Ge) with the aid of shutters. Targets were fitted in the bottom part of the chamber facing upward

Fig 2.1 Image of the RF sputtering system.

and substrate holder was fitted above the target facing downwards. The substrate temperature was controlled by the fitted temperature controllers and substrates were kept rotating at the constant rotation speed (~ 4-5 rpm) during deposition.

2.1.2 Synthesis of Ge Clathrate Film on Sapphire Substrate by Two-step Annealing Method

In the two-step annealing method for the Ge clathrate film synthesis, precursor film of intermediate Zintl phase (NaGe) and the clathrate film had been prepared in separate steps. In the first step (Ar annealing step), Ge deposited film (a-Ge on sapphire) was placed with Na in a tightly sealed stainless-steel container. Deposition of Ge on sapphire was not considered as a step as it involves just transfer of Ge from Ge target in RF sputtering system

(24)

13

to sapphire substrate (section 2.1.1). Slices of Na (>99.95%, Nippon Soda, Japan) were put inside an open Ta box of height 15 mm with as sputtered Ge film placed on the top of the box and the film facing down. Sample handling has been done inside the Ar filled glove box (Ar- 99.999%, H2O < 10 ppm) to avoid oxidation/contamination. Subsequently, the tightly sealed container was transferred to the muffle furnace and annealed at 400 °C for 3 h, leading to the synthesis of precursor NaGe film and this step is termed as the Ar- annealing step due to annealing in the Ar environment. In the second step (vacuum annealing step), NaGe film was transferred in a quartz tube inside the glove box and annealed in a high vacuum (≤ 10^-4 Pa) at 300 °C for 12 h by using a tubular furnace. This vacuum annealing of precursor film leads to the formation of type-II Ge clathrate film. The schematic diagram of the setup has been shown in Fig 2.2 and the film growth mechanism has been illustrated in Fig 2.3. The experimental results of the samples prepared in this two-step method will be described in Chapter 3.

Fig 2.2 Schematic diagram of the setup used for two-step clathrate synthesis process.

Fig 2.3: The growth mechanism of the clathrate film in the two-step annealing method.

(25)

14

2.1.3 Synthesis of Ge Clathrate Film on Sapphire Substrate by pVEAS

A new setup namely “Portable Vacuum Evaporation and Annealing System (pVEAS)” has been developed to improve the synthesised clathrate film quality. The pVEAS enables vacuum evaporation of Na and infrared (IR) lamp heating of the film in the same setup. It contains two main chambers namely the IR bulb chamber and sample chamber. These chambers are detachable, and the sample chamber could be transferred to the glove box without exposure to air and a schematic diagram is shown in Fig 2.4 whereas the image of the setup has been shown in Fig 2.5. IR bulb chamber contains an IR lamp that acts as the heater and radiation is focussed on the sample via parabolic reflectors and passes through fitted transparent fused silica windows into the sample chamber. The advantage of the IR lamp heating over the resistive heating is the downsizing and simplification of the main chamber, which was achieved by separating the heating system. The simplified and downsized chamber allows us the easy handling of the sample inside the glove box. The sample chamber is connected to a vacuum system (rotary pump and turbomolecular pump) from one side and a pair of feedthroughs is connected to the other side which is used to supply power from the external power source to the tungsten basket employing resistive heating for thermal evaporation of Na. Slices of Na, cut from Na lumps are placed in the tungsten basket and a rectangular sample holder rests above this tungsten basket (~50 mm above).

(26)

15

Fig 2.4 Schematic diagram of the “Portable Vacuum Evaporation and Annealing System (pVEAS).

The sample holder is placed on a rectangular window (20 × 30 mm²), kept at the centre of a silica disk supported by three metal rods. A carbon sheet with 0.5 mm in thickness is placed on the top of the film sample to absorb the heat from the IR lamp and subsequent homogeneous distribution of heat across the sample during IR bulb heating. The temperature of the heated sample is controlled and estimated by the output power of the IR lamp with a maximum rating of 1 kW. The relationship between the temperature and the output power was calibrated in advance by the preliminary experiment using a thermocouple attached to the substrate which has been for the film sample. Thus, the pVEAS allows the vacuum evaporation of Na, immediately followed by (or simultaneously) the thermal annealing. The whole setup is equipped with the water flow to remove excess heat generated inside and the water chiller (Eyela, CA-1116AS) is connected to the water flow as a heat sink.

In this method, deposited Ge film on sapphire is used as the starting material. Slices of Na are loaded inside the tungsten basket inside Ar filled

(27)

16

glove box and a film sample (20 × 10 mm²) is placed on the sample holder.

Thereafter, the sample chamber is transferred outside and attached to the IR bulb chamber. The sample chamber is evacuated with help of the fitted vacuum system. After achieving a high vacuum level (≤ 10^-4 Pa), Na is evaporated by applying a voltage across the tungsten basket through connected feedthroughs by the external power source. The substrate temperature of the film sample varied (room temperature to 300 °C) depending upon the requirement. Thereafter, the film sample is annealed in a high vacuum to obtain the type-II Ge clathrate film sample.

Fig 2.5 Image of the pVEAS setup.

2.1.4 Clathrate Film Sample Washing in Deionized Water

As synthesized clathrate film samples were washed in ultra-pure deionized (DI) water to remove any remaining impurity on the surface. The

(28)

17

clathrate film sample was dipped in DI water for 30 seconds with gentle stirring. The remaining water from the film surface was removed by a nitrogen gun. Subsequently, the film samples were placed inside the quartz tube and annealed at 150 °C for 3 h in a high vacuum by using a tubular furnace (Fig 2.2) for drying.

2.1.5 Na Reduction of Clathrate Film Sample

The as-prepared type-II Ge clathrate films generally contain Na as the guest species. The concentration of Na varies depending upon the preparation method/parameters. Depending upon the concentration of Na, clathrate shows a significant change in its properties. Hence, different methods for Na reduction from the clathrate film sample has been discussed in the following section.

2.1.5.1 Na Reduction by Vacuum Annealing

In this method, the type-II Ge clathrate film sample was placed inside the quartz tube and annealed inside the high vacuum (≤ 10^-4 Pa) at 280 °C using a tubular furnace (Fig 2.2). This results in the reduction of Na inside the clathrate film sample. However, this method becomes ineffective at lower Na concentration of type-II Ge clathrate and prolonged duration of annealing does not affect the Na concentration (x ~ 1.0 or below in NaxGe136). Hence, we were unable to obtain guest free type-II Ge clathrate using this method.

2.1.5.2 Na Reduction by Application of Electric Field

For the synthesis of guest free type-II Ge clathrate film, a new method of Na reduction by application of electric field has been employed since the vacuum annealing method becomes inefficient at lower Na concentration inside clathrate. In this method, the electric field is applied normally to the NaxGe136 film surface and a schematic diagram of the setup has been shown in Fig 2.6. A direct current (DC) voltage of 80 V was applied to the connected electrodes with the film sample sandwiched inside. The whole setup was sealed inside a stainless-steel container with dry Ar filled inside and annealed at 280 °C during electric field application. The annealing was performed by muffle furnace (As One, HPM-1G). The Na ions were expected to diffuse

(29)

18

towards the cathode due to the applied electric field and available thermal energy. The carbon sheet was used to absorb the Na diffusing out from the film sample.

Fig 2.6 A schematic diagram of apparatus used for Na removal from the clathrate film by applying electric field.

2.2 Characterisation Methods

2.2.1 X-Ray Diffraction (XRD) and Rietveld Refinement Analysis X-Ray diffraction (XRD) corresponds to the non-destructive characterisation technique based upon the diffraction principle of the X-Ray from the sample. Measurement of X-rays diffracted from the surface of lattice planes in the crystal serves as a powerful tool to analyse the crystal structure.

The basic formula underlying the XRD is the Lau formula (eqs. 2.1-2.3) and which is further simplified as Bragg’s formula (eq. 2.4) [70]. Bragg’s formula is commonly used due to its simplified representation of the diffraction condition.

Laue formula is given by the following equation:

ܽ_ଵή ሺܵ െ ܵ_଴ሻ ൌ ݄ߣ (2.1)

ܽ_ଶή ሺܵ െ ܵ_଴ሻ ൌ ݄ߣ (2.2)

(30)

19

ܽ_ଷή ሺܵ െ ܵ_଴ሻ ൌ ݄ߣ (2.3) Where a1, a2 and a3 are reciprocal vectors of the crystal lattice. S0 and S are the incident and the diffracted beam, respectively. his the plank’s constant and λ is the wavelength of the incident X-rays (Fig 2.7a).

Bragg’s formula is given by the equation:

݊ߣ ൌ ʹ݀ݏ݅݊ߠ (2.4)

Where λ is the wavelength of the incident X-rays, n is an integer, d is the interplanar spacing of the lattice crystal and ߠ is the diffraction angle (Fig 2.7b).

Fig 2.7 (a) X-ray scattering by atoms at O and A [70]. (b) Schematic diagram of Bragg’s law.

Depending upon the nature of X-ray radiation and scanning method various methods has been developed e.g., Laue method, rotating-crystal method, powder method and so on. Since the thickness of the film sample in the present thesis work was in a few microns or less, the investigation by the commonly used powder diffractometer (symmetrical diffraction geometry) technique proves difficult owing to weak diffraction peak intensity from the film and high background intensity caused by diffraction and scattering from the substrate [71]. Therefore, the thin film method (symmetric geometry method) has been used for the collection of diffraction data from the film sample (Fig 2.8). In this Grazing Incidence (GIXRD) geometry, the X-ray beam is incident at a small angle (αi) with respect to the film surface and the

(31)

20

detector scans and records the diffraction signals from the lattice planes that are inclined to the film sample surface. In the present study, XRD measurement has been performed by SmartLab (Rigaku, Ultima VI) with CuKα radiation (Fig 2.9).

Fig 2.8 X-ray diffraction geometry for the asymmetrical diffraction [71].

(32)

21

Fig 2.9 X-ray diffractometer with CuKα radiation (Rigaku, Ultima VI).

Rietveld refinement of XRD data proves a powerful tool for structural analysis of crystalline materials. This method was proposed by Hugo Rietveld [72], [73] which is based upon the comparison of the measured X-ray diffraction pattern to the calculated diffraction pattern. The accuracy of the refinement is examined by the various fitting parameters e.g. goodness of fit, χ² and R factors [74], [75]. The type-II Ge clathrate consists of five different atomic positions owing to the symmetry of the crystal and the main fitting parameters include XYZ coordinates of the atoms, occupancy of atoms at each site, the lattice parameter of the crystal and the temperature factor. In the current work, the XRD pattern has been analyzed for Rietveld refinement using PDXL (Rigaku) software package.

(33)

22 2.2.2 Raman Scattering Spectroscopy

Raman scattering is the inelastic scattering of photons from the sample material and Raman scattering spectroscopy serves as a powerful tool to investigate the microstructure of the molecules and crystals. Incident light (photon) on the semiconductor surface results in reflection (scattering), absorption or transmission for most of the photons i.e., the incident photon and scattered photons have the same frequency. This phenomenon occurs due to first-order elastic interactions of photons with phonons, electrons or impurities of materials, commonly termed as Rayleigh scattering. However, a small fraction of photons (~ 10^-5of the incident beam) interacts inelastically with the phonon modes of the material leading to the shift in the scattered photon frequency. This process is known as “Raman scattering” and the magnitude of the Raman shift is independent of the excitation frequency. The Raman scattered photons shift either to a longer wavelength (Stokes shift) by absorbing a phonon or shorter wavelength (anti-Stokes shift) by emitting a phonon according to the energy and momentum conservation rules (eqs. 2.5- 2.6). In Raman spectroscopy, the sample material is irradiated by intense laser beams in the UV-Vis region where the excitation, as well as Raman lines, appear and the scattered light is mainly observed in the direction, normal to the incident beam. According to quantum mechanics, a vibration is Raman- active if the polarizability is changed during the vibration. Although Raman spectra are generally observed for rotational and vibrational transitions of the sample material, it is possible to observe the Raman spectra of electronic transitions between ground states and lower energy excited states [76], [77].

߱_௦ ൌ ߱_௜ േ ȳ (2.5)

ݍ_௦ ൌ ݍ_௜ േ ܭ (2.6)

Where ωi and ωs represent the frequency of the incoming and scattered photon respectively, qi & qs are wave vectors of the incoming and scattered photon respectively. Ω and K are the frequency and wave vector of phonon respectively. In the present thesis work, JASCO NRS-2100G Raman system (Fig 2.10) is used for recording Raman scattering spectra in backscattering geometry and only Stokes shifted photons are recorded. A solid laser (YVO4)

(34)

23

was used as the excitation source to record the spectra with λ = 532 nm. The resolution and the recorded spectral range of Raman shifts are 1 and 60-1000 cm^-1 respectively. While recording the Raman shift, the excitation laser radiation is tightly focused on the surface of the film sample to a spot size of

≤ 1 μm.

Fig 2.10 NR-2100G (JASCO) with triple monochromator Raman system.

2.2.3 Fourier Transform Infrared Spectroscopy

Fourier transform infrared (FTIR) spectroscopy is a non-destructive and efficient technique that is used to obtain the infrared spectrum of emission, absorption and photoconductivity of the materials. FTIR spectroscopy technique can be used to collect high spectral resolution data over a wide range, usually between 5000 and 400 cm^-1 for the mid-infrared region and between 10,000 & 4000 cm^-1 for near-infrared region wavelength [78]. The basic principle of this method is based upon the fact that each molecule exhibits its own distinct quantized vibrational levels whose frequency lie in the mid-infrared region (200-400 cm^-1) and when the frequency of incident radiation matches with the frequency of a molecular vibration mode, the molecules absorb the radiation. This frequency of

(35)

24

absorption serves as a tool to identify the vibrational modes of the molecules.

The number of absorption peaks is functional to the degree of vibrational freedom of the molecule. The intensity of absorption peaks corresponds to the change in dipole moment and the possibility of the transition in vibrational energy levels. Hence, analysis of the infrared spectrum could provide the structural information of a molecule. Most molecules are infrared active except for a few homo-polar diatomic molecules e.g., O2, N2 etc. which have zero net dipole moment since vibration is IR-active if the dipole moment is changed during the vibration according to quantum mechanics [76], [77], [79]. A typical FTIR spectrometer consists of an IR source, Michelson interferometer and IR detector. Usually, a heated quartz tungsten halogen lamp is used as the IR source. The source generates the radiation which passes the sample through the interferometer and reaches the detector. Thereafter, the signal is amplified and converted to a digital signal. Eventually, the signal is processed in a computer system where Fourier transform is carried out and generates the IR spectrum. The main advantage of this technique includes the simultaneous measurement in each frequency with high resolution. In this research work, Spectrum 100 (Perkin Elmer) spectrometer was used for recording the IR spectrum. From the FTIR transmission spectra, the absorbance has been calculated by the formula -log(T) and absorption coefficient, α, of the films using Beer-Lambert law (eq. 2.8)

ܶ ൌ ܶ_଴ሺെߙ ή ݀ሻ (2.8) Where T and T0 are the transmittance of the film and the substrate respectively; d is the film thickness.

2.2.4 UV-Vis-NIR Transmission Spectroscopy

Ultraviolet-visible near-infrared (UV-vis-NIR) spectroscopy was performed in transmission mode for the films deposited on a sapphire substrate. The absorbance calculated using the formula -log(T) and the absorption coefficient α was calculated using Beer-Lambert law (eq. 2.8) [76], [77], [79]. The measured spectrum indicates the fraction of light transmitted by the films and the absorption owing to the electronic transitions from the

(36)

25

ground state to an excited state. The interference fringes were observed in some cases (a-Ge, Ge(cF136) film sample) due to interference of light. These interference fringes were used to determine the thickness and optical constants of the film by the Swanepoel method [80]. For the semiconductor material, the absorption coefficient (α) due to interband transition near the bandgap is described by the Tauc’s relation (eq. 2.9) [81].

ߙ݄߭ ൌ ܥ൫݄߭ െ ܧ_௚൯^௣ (2.9) Where hν is the energy of the incident light, Eg is the energy band gap and C is a constant independent of ν. The values of exponent p are 2 and ½ for the indirect gap and direct gap transition, respectively. By using equation (2.9), the optical band gap Eg of the film was calculated from (αhν)² vs hν curve. In the present research work, UV-vis-NIR transmission data was collected using V-670 (Jasco) and Lamda 950 (Perkin Elmer) dual beam spectrophotometers, depending upon the availability of the equipment. As a source of light in this spectrophotometer, a tungsten lamp has been used in the near-infrared and visible region (wavelength 340~2600 nm) and a deuterium lamp is used in the ultraviolet region (wavelength 150~340 nm).

2.2.5 Scanning Electron Microscopy

Scanning electron microscopy (SEM) corresponds to the electron microscopy in which the sample is scanned using a focussed beam of electrons by raster scanning. A schematic diagram of the interaction between incident electron beams and the sample is shown in Fig 2.11 [82]. Hence, SEM serves as a powerful tool to investigate the sample surface topography and composition. Fig 2.12 shows the schematic of a typical SEM device configuration.

An electron gun is used as the electron beam source (thermo-ionic emission) and the electron (~ 50 μm diameter) is focused on the sample with the spot size of about 0.1 μm using a condenser and objective lenses to analyze the sample in the sub-micron range. The probe is used to scan the sample surface in raster scanning and by the scanning coil. The secondary electrons generated from the sample surface are converted into electric signals and

(37)

26

amplified using a scintillator-optical pipe-photomultiplier detection system.

The amplified signal is further sent to the grid of a cathode ray tube (CRT) which scans in synchronization with the electron beam flux and the beam intensity is modulated to the CRT. In the present research work, Hitachi- 4800 device has been used for the characterization of the film sample at the accelerating voltage of 5.0 kV.

Fig 2.11 Electron-sample interaction: the different types of signals generated [82].

(38)

27

Fig 2.12 The setup of scanning electron microscopy [83].

2.2.6 Laser Microscopy

A laser microscope utilises a He-Ne gas laser or semiconductor laser as a light source and serves as one of the most powerful tools to investigate a wide variety of materials including thin films. Among them, the microscopes that employ a special optical system called the confocal optical system are generally known as confocal microscopes which are commonly used for various observations, measurements and analyses such as surface roughness, three-dimensional shape and so on.

In this study, a 3D Laser scanning microscope (Keyence, VK-X200K) was used which was fitted with a laser light source (408 nm, maximum output- 0.95 mW) and uses AI feature for image processing and automatic upper-lower bound estimation (Height resolution- 0.5 nm, spatial resolution- 1 nm).

(39)

28 2.2.7 Hall Measurement

The main objective of the Hall measurement in the van der Pauw technique includes the determination of sheet carrier density ns by measuring the Hall voltage VH. The basic principle underlying the Hall effect phenomenon is the Lorentz force which refers to a combination of two separate forces: electric and magnetic force. When an electron moves along the direction of electric field (E) and perpendicular to an applied magnetic field (B), it experiences a magnetic force െݍݒ ൈ ܤacting normal to both directions; where q and v refer to the charge and velocity of the electron respectively and B is the applied electric field (Fig 2.13). The direction of this magnetic force can be determined by the right-hand thumb rule and the resulting Lorentz force F is therefore equal to െݍሺܧ ൅ ݒ ൈ ܤሻ.

Fig 2.13 Lorentz force (F) and Hall voltage (VH) for a bar-shaped n-type semiconductor, subjected to the electric current (I) and magnetic field (B) [84].

For an n-type, bar-shaped semiconductor (Fig 2.13), the carriers are predominantly electrons with current density n. Assuming a constant current flow I along the x-axis from left to right in the presence of magnetic field (z- direction), electrons subjected to the Lorentz force initially drift away from the current direction towards the negative y-axis and resulting in an excess

(40)

29

negative surface electrical charge on this side of the sample. This charge leads to the presence of Hall voltage, a potential drop across the two sides of the sample. This transverse voltage is termed as the Hall voltage VH and its magnitude is equal to _௤௡ௗ^ூ஻, where I is the current passing through the sample, B is the applied magnetic field, d is the sample thickness and q (1.6 × 10^-19 C) is the elementary charge. In some cases, it is convenient to use sheet or layer density (݊_௦ ൌ ݊Ǥ ݀) instead of bulk density and following equation is obtained:

Fig 2.14 Schematic diagram of a rectangular van der Pauw configuration [84].

݊_௦ ൌ ^ூ஻

௤ȁ௏ಹȁ. Hence, by measuring the Hall voltage (VH) and from the known values of q, B and I, sheet density (ns) of the charge carrier can be determined.

The sheet resistance Rs of the semiconductor can be determined by the van der Pauw resistivity measurement technique, which in turn will facilitate the determination of Hall mobility.

The van der Pauw technique is widely used in the semiconductor industry to determine the resistivity of uniform samples due to its convenient application [85], [86]. For the determination of both mobility (μ) and the sheet density ns, a combination of a resistivity measurement and a Hall measurement is required. This van der Pauw method was originally devised by van der Pauw and uses an arbitrarily shaped (but with continuous connection i.e., no holes or nonconducting islands or inclusions), a thin-plate

(41)

30

sample containing four very small ohmic contacts placed on the periphery of the plate. Fig 2.14 shows the schematic diagram of the van der Pauw configuration. The objective of the resistivity measurement is to determine the sheet resistance Rs. Van der Pauw demonstrated that there exist two characteristic resistance RA and RB, associated with the corresponding terminals as shown in Fig 2.14. RA and RB are related to the sheet resistance through the following equation (Eq. 2.10):

ቀെ^஠ோ^ఽ

ோೄ ቁ ൅ ሺെ^గோ^ಳ

ோೄ ሻ ൌ ͳ (2.10)

Fig 2.15 Schematic diagram of the Hall measurement in van der Pauw technique [84].

Which can be solved numerically for RS. The bulk electrical resistivity ρ can be calculated by using the formula ߩ ൌ ܴ_ௌ݀. In order to obtain the two characteristic resistances, a dc current (I) is applied into contact 1 and out of contact 2 while measuring the voltage V43 from contact 4 to contact 3 (Fig 2.14). Thereafter, current I is applied into contact 2 and out of contact 3 while measuring the voltage V14 from contact 1 to contact 4. RA and RB are calculated by using the formula: ܴ_஺ ൌ ^௏^రయ

ூ_భమ and ܴ_஻ൌ ^௏^భర

ூ_మయ. The main objective of the Hall measurement in the van der Pauw technique is to determine the sheet carrier density ns by measuring the Hall voltage VH. The measurement of Hall voltage consists of a series of voltage measurements with a constant current I and a constant magnetic field B applied perpendicularly to the plane of the sample.

Conveniently, the same sample can also be used for the Hall measurement as depicted in Fig 2.15. In order to measure the Hall voltage (VH), a current I is

(42)

31

forced through the opposing pair of contacts 1 & 3 and the Hall voltage VH (=

V24) is measured across the remaining pair of contacts 2 and 4. Once the Hall voltage VH is measured, the sheet carrier density ns can be calculated by using the formula: ݊_௦ ൌ ^ூ஻

௤ȁ௏_ಹȁ from known values of I, B and q.

There exist some practical aspects which must be considered during Hall and resistivity measurements. Primary concerns include (I) ohmic contact quality and size, (II) sample uniformity and accurate thickness determination, (III) thermomagnetic effects due to nonuniform temperature and (IV) photoconductive and photovoltaic effects which can be minimized by measuring in a dark environment. In this thesis work, electrical measurement in van der Pauw technique has been performed by Resitest 8400 (Tokyo Corporation) at room temperature in air. In order to facilitate electrical measurement of Ge(cF136) film sample, Ag (99.99%, Nilaco) films were deposited on the Ge(cF136) film using thermal evaporation under high vacuum (≤ 10^-4 Pa) as circular electrodes with a diameter of 0.5 mm at the corners of the 5 × 5 mm² area.

(43)

32

Chapter 3: Synthesis and Characterization of Guest Free Type-II Ge Clathrate by Two-step Synthesis

Method

A synthesis method for the growth of Ge(cF136) thin films on Ge(111) substrates by thermal annealing of NaGe precursor has been already developed [4], [67], [87]–[89]. The synthesis method was further developed in this work to prepare NaxGe136 films on a transparent and insulating substrate by changing the starting material with the aim of optical and electrical studies.

3.1 Synthesis and Methodology

The amorphous Ge (a-Ge) films were deposited on the sapphire substrate (20mm × 10 mm) using RF sputtering system (Section 2.1.1) and the sputtering was performed for 3 h at the substrate temperature of 300 °C. The as-deposited a-Ge film on the sapphire substrate was placed at the height of 15 mm from the Na lumps (>99.95%, Nippon Soda, Japan) (0.15 gm) inside the Ta box and the sample handling was done in an Ar filled glove box to avoid any contamination. The sample was prepared by a two-step method involving two steps namely (1) Ar-annealing step for the preparation of intermediate NaGe film and (2) vacuum annealing step for the type-II Ge clathrate film (Section 2.1.2). In the present work, UV-vis-NIR study was done by Lamda 950 (Perkin Elmer) dual-beam spectrophotometer.

3.2 Results and Discussion

The a-Ge film deposited by RF sputtering on the sapphire substrate, upon Ar-annealing with Na lumps inside a tightly sealed container in an inert atmosphere (dry Ar) leads to the formation of the precursor films. Thereafter, the Na rich precursor phase was annealed under dynamic vacuum (≤ 10^-4Pa) by using vacuum annealing apparatus (Fig 2.2). The sample was rinsed in the DI water to remove any impurity (remaining Na or oxides etc.) from the film surface and subsequently dried in high vacuum annealing (section 2.1.4). Fig

(44)

33

3.1 shows the photograph of the as-prepared Ge clathrate grown on the sapphire substrate and a greyish film with good uniformity (i.e., no visible pinholes, surface defect, change in colour contrast) was observed.

3.2.1 XRD Results of the Film Sample

The GIXRD pattern obtained at a grazing angle of 0.1° was shown together with the Rietveld refinement fitted curve in Fig 3.2 and Rietveld refinement results shown in Table 3.1. All reflections were attributed to the polycrystalline NaxGe136 (cF136) films with random crystal orientation. This suggests that the precursor film was successfully transformed into the type- II Ge clathrate structure without the inclusion of any impurity phase such as cubic diamond structured Ge (α-Ge). The observed random crystal orientation lies in contrast to that of the previously reported NaxGe136 film grown on the α-Ge(111) substrate which showed epitaxial growth [67] and can be attributed to the difference in the substrate material. The observed polycrystalline nature is believed to be originating from the large lattice mismatch between NaxGe136 and sapphire (α-Al2O3: hR30, a = 0.48 nm, c = 1.299 nm) [90]. A broad band around 25° is due to the glass holder plate used in the XRD measurement. According to the Rietveld refinement analysis (Rwp

= 4.3%, S = 1.31), the Na contents x in the NaxGe136 was determined as 1.29(6) and the lattice constant a was estimated as 1.5212(1) nm which lies in good agreement with the previously reported values [17], [68], [69].

Fig 3.1 Photographic image of the as-prepared Ge(cF136) film on the sapphire substrate.

(45)

34

Fig 3.2 GIXRD reflections of the as-prepared Ge(cF136) film obtained at the grazing angle of 0.1° and Rietveld analysis fitting.

Table 3.1 Rietveld refinement results of as prepared type-II Ge clathrate film.

The space group was assumed as ܨ݀͵ത݉.

3.2.2 Raman Spectroscopy Results

Fig 3.3 shows the Raman spectra of the as-prepared Na1.29Ge136 film measured with the laser incident from the front and back sides of the film.

Upon considering the penetration depth of the laser wavelength (532 nm),

(46)

35

Raman signals originate mainly from the vicinity of air/sample surface or sample/substrate interface. All the Raman peaks in both cases of front and back side incidence were assigned to NaxGe136, indicating the throughout synthesis of the clathrate film across its thickness. The broad peaks around 70 and 280 cm^-1 seem to be superimposed in the spectrum for front side incidence. This could be attributed to the presence of layer of the amorphous Ge (a-Ge) phase at the surface which has been probably formed during the washing process. Furthermore, the close inspection in the low-frequency region revealed that the peak at 67 cm^-1 is very low in intensity as compared to the peak at 76 cm^-1. This feature is interpreted by the lower Na contents, according to the previous work on NaxSi136 [91].

Fig 3.3 Raman spectra of the as-prepared type-II Ge clathrate film with the laser incident from the front and back sides of the film sample.

Fabrication and properties of semiconducting germanium clathrate film( 本文(Fulltext) )

Fabrication and Properties of

Semiconducting Germanium Clathrate Film

Division of Environmental and Renewable Energy Engineering Systems, Graduate School of Engineering

Gifu University, Japan 501-1193

December 2021

LIST OF ABBREVIATIONS AND SYMBOLS

TABLE OF CONTENTS

LIST OF FIGURES

LIST OF TABLES

Chapter 1: Introduction

1.1 Group IV Inorganic Clathrates

1.2 Structural Properties of Clathrates

1.3 Physical and Electrical Properties of Clathrates

1.4 Synthesis of Type-II Clathrates

1.5 Motivation and Novelty

Chapter 2: Synthesis Method and Characterization

2.1 Sample Preparation

2.2 Characterisation Methods

Chapter 3: Synthesis and Characterization of Guest Free Type-II Ge Clathrate by Two-step Synthesis

Method

3.1 Synthesis and Methodology

3.2 Results and Discussion