Synthesis and Applications of Heterostructuresof Two-Dimensional Materials

Synthesis and Applications of Heterostructures of Two-Dimensional Materials

アザ, スクマ, アジ

https://doi.org/10.15017/1931943

出版情報：九州大学, 2017, 博士（工学）, 課程博士 バージョン：

権利関係：

Synthesis and Applications of

Heterostructures of Two-Dimensional Materials

A Dissertation Presented to

Interdisciplinary Graduate School of Engineering Science of Kyushu University

Adha Sukma Aji

2018

Abstract

The potential of atomically thin layered materials, such as zero band gap semiconductor graphene, semiconducting transition metal dichalcogenides (TMDCs), semiconducting metal chalcogenides (MCs), and insulating hexagonal boron nitride (h- BN) are promising for future building blocks of electronic devices. The heterostructures of these ultimate thin materials offer new solutions for the problem in Si-based device miniaturization. For example, graphene with its high conductivity could be utilized as the future interconnection or electrical contact for semiconducting 2D materials. WS

is a prospective choice for a semiconducting channel owing to predicted high electron mobility that is comparable with carrier mobility of Si. However, the poor electrical contact between single-layer WS

and conventional metal electrodes retards the carrier mobility of WS

. In this thesis, the improvement of single-layer WS

carrier mobility (50 cm

/Vs) with chemical vapor deposition (CVD)-grown multi-layer graphene (MLG) electrodes is demonstrated. The improved contact is explained by the tunable work function of MLG electrodes that can overcome the high Schottky barrier (SB) height usually formed with conventional metal electrodes. The high mechanical robustness of MLG-WS

heterostructures allowed the demonstration of flexible fast (response time of 2 ms), and sensitive (responsivity of 4500 A/W) photodetector.

Next, due to its unipolarity of n-type WS

, it is difficult to form a p-n junction or complementary field-effect transistors (FETs) with solely WS

. The heterostructures of two different materials with different polarities could solve this problem. Therefore, this thesis also presents the novel SnS-WS

p-n junctions realized by using two-step CVD method. The SnS (a member of MCs) was found to have a p-type polarity with the mobility of 15 cm

/Vs. Then, the formation of SnS-WS

p-n junction was confirmed based on its rectifying behavior. The p-n junction also exhibited a tunable rectification ratio and good ideality factor because of clean interface resulting from two-step CVD method.

Moreover, the presence of SnS could enhance the photodetector performance due to high optical absorption of SnS.

This thesis demonstrates the advantages of two-dimensional-based

heterostructures which can be applied to cutting-edge electronic and optoelectronic

applications in the near future

Table of Contents

Abstract ... i

Table of Contents ... ii

Chapter 1 – Introduction ... 1

1.1. 2D materials beyond graphene ... 1

1.1.1. Graphene and its zero band-gap nature ... 1

1.1.2. Transition metal dichalcogenides and metal dichalcgonides... 4

1.1.3. Heterostructure of 2D materials ... 6

1.2. Motivation ... 8

1.3. Outline of thesis ... 10

Reference ... 11

Chapter 2 - Structure, Electrical Properties, and Preparation Methods of WS

and SnS ... 15

2.1. Structures of WS

and SnS... 15

2.1.1. Structures of WS

... 15

2.1.2. Structutres of SnS ... 16

2.1.3. Phonon modes of WS

and SnS ... 17

2.1.3.1. Phonon modes of WS

... 18

2.1.3.2. Phonon modes of SnS ... 19

2.2. Electrical properties of WS

and SnS ... 19

2.2.1. Band structures of WS

and SnS ... 20

2.2.2. Expected carrier mobilities of WS

and SnS ... 22

2.3. Material preparation: CVD growth of TMDCs, MCs, and 2D materials-based

heterostructures ... 24

References ... 28

Chapter 3 – 2D Materials-Based Field-Effect Transistors ... 33

3.1. Field-effect transisotr operation ... 33

3.2. WS

- and SnS-based FET ... 35

3.3. 2D heterostructures-based p-n Junctions ... 38

3.4. Metal contact to 2D materials-based semiconductor ... 40

3.4.1. Problems on interface between metal and 2D semiconductor ... 40

3.4.2. Schottky barrier extraction: thermionic emission... 41

References ... 44

Chapter 4 – High Mobility WS

Transistors Realized by Multi-Layer Grapehene Electrodes and Application to High Responsivity Flexible Photodetector ... 49

4.1. Introduction ... 49

4.2. Experimental method ... 52

4.2.1. CVD growth of single-layer WS

and MLG ... 52

4.2.2. Characterizations ... 53

4.2.3. Device fabrication ... 54

4.2.4. Device measurement ... 54

4.3. Results and discussion ... 55

4.3.1. Fabrication and characterization of WS

-MLG heterostacked devices . 55 4.3.1.1. Fabrication of WS

-MLG devices ... 55

4.3.1.2. Characterization of WS

-MLG devices ... 57

4.3.1.3. Transport properties of WS

-MLG FET ... 58

4.3.2. Investigation of electrical contact of MLG to single-layer WS

... 60

4.3.3. Flexible optoelectronic WS

devices with MLG electrodes ... 66

4.3.3.1. Fabrication and transport properties of flexible devices ... 66

4.3.3.1. Optical response ... 68

4.4. Conclusions ... 71

References ... 73

Chapter 5 – Two-step Synthesis and Characterizations of Vertically Stacked SnS-WS

and SnS-MoS

p-n Heterojunctions ... 79

5.1. Introduction ... 79

5.2. Experimental setup ... 81

5.2.1. Two-step CVD growth ... 81

5.2.2. Characterizations ... 82

5.2.3. Device fabrication and measurements ... 82

5.3. Results and discussion ... 83

5.4. Conclusions ... 102

Chapter 6 – Conclusion and Future Outlook ... 109

6.1. Conclusion ... 109

6.2. Future outlook ... 110

References ... 113

Acknowledgements ... 115

Chapter 1

Introduction

1.1. 2D materials beyond graphene

1.1.1. Graphene and its zero band-gap nature

Discovery of exceptional properties of isolated single-layer graphene in 2004 has

triggered many researchers to explore other possibilities of new physical phenomena in

2D structure.

The isolation to atomic thickness introduces so-called 2D quantum

confinement effect. This effect offers many interesting new properties of graphene, for

example, massless charge carrier in graphene which theoretically predicts that the

carriers in graphene can drift at the speed of light.

At the early stage of graphene’s

finding, the graphene flakes were fabricated by using an mechanical exfoliation method,

where a starting graphite was separated by using adhesive tape repeatedly. Eventually,

due to the low van der Walls interaction in each layer of graphene, the single-layer

graphene could be separated and placed onto the insulating substrate (i.e. SiO

). The isolation of single-layer graphene allows A. Geim and K. Novoselov to examine electrical properties of graphene by fabricating the field-effect transistors (FETs) with graphene as the channel of a transistor. The report on graphene FET later gave these two researchers a Nobel Prize in 2010.

For electronic device applications, graphene is very desirable, because it exhibits very high carrier mobility of 2×10

cm

V

s

.

The exceptional conductivity of graphene was exploited as the electrical contact to the 2D semiconducting material in this thesis.

However, one major drawback of graphene is that graphene does not have a band gap which hinders its application to the logic devices. Two approaches have been proposed to open the band gap of graphene; confining graphene sheet to one-

Figure 1-1. Classification of 2D materials based on their band gaps. Adapted from Ref. 8.

dimensional nanoribbons structure or applying a strong perpendicular electric field to double-layer graphene. For nanoribbon case, the width of graphene is needed to be scaled down to below 10 nm to achieve on-off switching at room temperature.

In the double-layer graphene, the controlled stacking order (called AB stacking) is required to break the graphene’s symmetry, and the vertical electric field is needed to be applied to open band gap.

However, the complicated material preparation and device design make these approaches unfavorable for large-scale device production.

To solve this issue, many researchers have shifted or widenend their focus from graphene to other 2D materials to find alternatives. More than 600 layered materials are predicted to be able to be thinned to single-layer form, including 2D material with semiconducting properties.

Figure 1-1 shows a classification of well-known 2D materials based on their structures which demonstrate the broad choices of 2D materials with different band gaps.

This figure shows a large number of material selections with different band gaps which cover the applications that requir broad optical response from ultraviolet to far infrared. Additionally, various properties can be found in these 2D

Figure 1-2. Schematic view of (a) graphene, (b) WS2, and (c) SnS.

SnS WS

Graphene

a) b) c)

structures, for example, the semiconducting properties in layered MoS

, superconducting properties in layered NbSe

, insulating properties in layered h-BN, and so on. In the last several years, MoS

, WS

, and other members categorized in transition metal dichalcogenides (TMDCs) have been gaining the wide interest beside graphene, because these materials have finite band gaps that are required for the logic transistor. Another class of 2D materials called metal chalcogenides (MCs), such as SnS and SnSe, have recently emerged due to its remarkable properties coming from their puckered structures.

Figure 1-2 shows the structure of graphene, MoS

(TMDCs), and SnS (MCs), where MCs has rhombohedral symmetry that is different from hexagonal symmetry of graphene and TMDCs. Since the TMDCs and MCs are the main topics in this thesis, the brief overview of the two classes of 2D materials will be described in the next following two sections.

Figure 1-3. a) Schematic view and b) transfer curve of single-layer MoS2 transistor. Inset is the output curves of MoS² under different gate voltages. Reproduced from Ref. 11.

a) b)

1.1.2. Transition metal dichalcogenides and metal chalcogenides

The simple method of graphene preparation triggered a curiosity on how other layered materials behave when they are thinned down to atomic thickness, especially for TMDCs. TMDCs have been extensively studied recently due to its semiconducting properties which cannot be found in graphene. Figures 1-2a,b show schematic view of graphene and TMDC which share similar hexagonal structure. One single cell of TMDCs (MoS

, WS

, MoSe

, WSe

, and so on) consists of a transition metal atom (Mo, W, and so on) which covalently bonds other three neighboring chalcogenide atoms (S, Se, and Te).

The early report on electrical properties of atomically thin MoS

was reported by the

same team who reported electrical properties of graphene.

However, the

semiconducting properties of MoS

were not prominent enough at that time. A. Kis’s team

demonstrated mobility of single-layer MoS

of 200 cm

/Vs with on-off ratios of 10

under

the influence of high electric fields from a thin high-K dielectric gate (Figure 1-3b).

The

exceptional carrier mobility of single-layer MoS

that can be comparable with graphene

nanoribbon or even thin strained Si film promises the further application of TMDCs in

electronic devices.

Then, this finding became a starting point of the exploration of

other 2D TMDCs, such as WS

, which is the focus of this thesis. It is theoretically predicted

that WS

has several advantages compared to the well-known MoS

. Having a higher

carrier mobility than MoS

, WS

is considered as one contender for future post-Si

electrical devices.

The strong photoexcitation in single-layer WS

also makes it suitable

for optical devices.

Moreover, due to the large difference of atomic numbers of W and

S, it is also theoretically predicted and experimentally confirmed that WS

has a large

band splitting which is promising for next-generation spintronic applications.

Another class of 2D material, called metal chalcogenides (MC), is also gaining a great deal of interest nowadays due to their unique structures. SnS, SnSe, PbS, and PbSe, are several examples of 2D materials that belong to the MC family. Different from hexagonal graphene and TMDCs, the MCs have a rhombohedral structure, as shown in Figure 1-2c. This unique structure gives different properties from TMDCs and graphene.

In the cases of SnS and SnSe, this rhombohedral structure generates interesting properties, such as anisotropic electrical properties and large piezoelectric coefficients, which cannot be seen in hexagonal-shaped TMDCs.

Moreover, considering very high light absorption (10

cm

) of SnS, this earth abundant material was greatly studied in the past for application in the solar cell.

SnS is a semiconductor with a small bandgap of ~1 eV that can be exploited to the near infrared detector applications.

Moreover, SnS is also predicted to have a comparable mobility with other semiconducting 2D materials, making them interesting to be studied as well.

Considering these advantages, SnS was chosen to be another focus complementing aforementioned WS

.

Figure 1-4. a) Schematic view of heterostructures of 2D materials. b) Cross-sectional TEM image of h-BN and graphene superlattices. Figures a and b are reproduced from ref. 23 and 24

a) b)

Graphene h-BN MoS₂ Graphene

WSe2 Graphene

h-BN Graphene h-BN Graphene

1.1.3. Heterostructure of 2D materials

Combination of two or more different materials is commonly called as

“heterostructure.” The heterostructure of different materials could extend the properties of materials. For examples, a heterostructure-based solar-cell can widen the light spectrum absorption which leads to photovoltaic efficiency enhancement.

Also, the heterostructure of two different materials with different charge polarity (n-type and p- type) can be built into a p-n junction or complementary metal oxide semiconductor field- effect transistor (CMOSFET) which is the cores of digital circuits.

Hence, due to its usefulness, the researches of heterostructure has been going strong than ever.

Compared with 3D materials, the heterostructure of 2D materials is preferable

due to several following advantages. First, in the 2D regime, any different 2D materials

with different properties can be stacked each other as we like due to their van der Waals

interaction.

Thanks to the recent development of mechanical exfoliation method and

dry transfer techniques, the fabrication of 2D materials-based heterostructures has

become simple and cheap. The first 2D material is firstly exfoliated from its bulk and

stacked to the desired substrate. Then, the second 2D material is stacked onto the pre-

existing 2D material similarly. This method can be repeated to achieve a 2D

heterostructure made from more than two kinds of materials. Figure 1-4a shows the

cartoon of fabrication of 2D heterostructure with more than two different materials. On

the contrary, the selection of materials that can be made into heterostructure in 3D

regime is limited due to their lattice mismatch which requires high temperature or

specialized synthesis apparatus. On the contrary, the absence of dangling bonds in 2D

materials can also contribute to the clean interfaces at the junction of 2D materials-based

heterostructures. Cross-sectional transmission electron microscopy (TEM) image of graphene and h-BN superlattice shown in Figure 1-4b reveals clean interface at each layer of 2D materials that was made by manual stacking.

Moreover, owing to clean interfaces, the strong inter-layer coupling is formed, showing very fast and efficient charge transfer even though neighboring layers are not covalently bonded.

This fast and efficient charge transfer is indeed desirable for optoelectronics applications.

1.2. Motivation

In the previous subsection, the general introduction of TMDCs, MCs, and the 2D materials-based heterostructures is described. In particular, 2D materials, especially TMDCs, offer a large potential for future optoelectronic applications. However, to apply the TMDCs in electronic devices with current Si dominance, great amount works are still needed. This is because the experimental results of electronic properties of devices based on TMDCs do not reach the predicted values. For instance, even though the carrier mobility of WS

is expected to be able to compete with Si, the real carrier mobility in the lab scale is still notably inferior.

Furthermore, the unipolar transport properties (n- type) of WS

also limits its application to CMOS or p-n diode. Again, the predicted ambipolar behavior of WS

is still difficult to be obtained empirically due to crystal defect.

Thus, the problems seen in WS

will slow down its practice in the real world.

This thesis demonstrates possible solutions to these problems by exploring its

heterostructures. First, the improvement of WS

electrical properties was realized by

using multilayer graphene (MLG) and WS

heterostructure. Second, the p-n

heterojunction consisting of p-type SnS and n-type WS

is described.

Recently, researchers focus their attention to improve the electrical properties of WS

. Different from its kin, MoS

, the reports on the improvement of electrical properties of WS

, particularly on the carrier mobility improvement, are still scarce. Besides the low quality of crystal and influence from substrates surface, the most limiting factors of WS

mobility is its electrical contact.

To solve the poor electrical contact to WS

, contact engineering is highly needed. Several reports have been published to improve the electrical contact to WS

. The utilization of chemical doping, graphene electrodes, and dielectric shield was reported to improve the electrical contact to WS

.

However, these methods are still not enough to achieve a perfect contact to WS

. In this thesis, the (MLG) was used to form a better contact to WS

. The Ohmic contact between MLG and WS

was found to be a key factor to achieve significant improvement on carrier mobility.

Next, as discussed earlier, the unipolar transport property of WS

limits its applications to p-n junction, which is desired for further implementation in solar cells and light emitting diodes. Thanks to van der Waals heterostructure, different polarities of TMDCs could be built into p-n junctions with ease. For examples, the light emitting diodes and the solars cell based on MoS

-WSe

, where MoS

and WSe

were fabricated act as n- and p-type semiconductor, respectively.

In this thesis, the new heterostructure of MC and TMDCs is presented. The formation of p-n junctions is realized by combining n-type WS

and p-type SnS.

Another important issue to be addressed is the scalability and reproducibility of

materials. Almost all of the reports above on TMDCs-based electronic devices were

realized by conventional exfoliation method which produces only several grains with

small lateral size. Even though this method is fast and cheap, it is not suited for large-

scale fabrication due to the size limitation. In this thesis, the chemical vapor deposition

(CVD) syntheses are performed to prepare all the materials. The CVD method proposed in this thesis overcomes the size limitation, initiating the possibility to move further into large-scale implementation.

1.3. Outline of thesis

This thesis is covering two main parts, the improvement of WS

electrical properties by using MLG as the electrodes and the formation of SnS and WS

p-n heterojunctions. Before discussing the experimental results, the electrical and optical properties of WS

and SnS are briefly introduced in Chapter 2. Next, Chapter 3 presents the example and explanation of WS

and SnS based devices which will give a better understanding of the next main parts. In Chapter 4, the heterostructure of MLG and WS

is presented. The carrier mobility improvement of WS

and its applications to flexible

photodetector are also described in details. In Chapter 5, the synthesis and

optoelectronic properties of WS

-SnS and MoS

-SnS are presented. Finally, this thesis is

closed with the conclusion and suggested future works which are given in Chapter 6.

References

1. Novoselov, K. S. et al. Electric Field Effect in Atomically Thin Carbon Films. Science 306, 666–669 (2004).

2. Kane, C. L. Erasing Electron Mass: Materials Science. Nature 438, 168–170 (2005).

3. Novoselov, K. S. et al. Two-Dimensional Gas of Massless Dirac Fermions in Graphene.

Nature 438, 197–200 (2005).

4. Bolotin, K. I. et al. Ultrahigh Electron Mobility in Suspended Graphene. Solid State Commun. 146, 351–355 (2008).

5. Son, Y.-W., Cohen, M. L. & Louie, S. G. Energy Gaps in Graphene Nanoribbons. Phys.

Rev. Lett. 97, 216803 (2006).

6. Castro, E. V. et al. Biased Bilayer Graphene: Semiconductor with a Gap Tunable by the Electric Field Effect. Phys. Rev. Lett. 99, 216802 (2007).

7. Ashton, M., Paul, J., Sinnott, S. B. & Hennig, R. G. Topology-Scaling Identification of Layered Solids and Stable Exfoliated 2D Materials. Phys. Rev. Lett. 118, 106101 (2017).

8. Castellanos-Gomez, A. Why All the Fuss about 2D Semiconductors? Nat. Photonics 10, 202–204 (2016).

9. Zhou, X. et al. Booming Development of Group IV-VI Semiconductors: Fresh Blood of 2D Family. Adv. Sci. 3, 1600177 (2016).

10. Novoselov, K. S. et al. Two-dimensional Atomic Crystals. Proc. Natl. Acad. Sci. 102, 10451–10453 (2005).

11. Radisavljevic, B., Radenovic, A., Brivio, J., Giacometti, V. & Kis, A. Single-Layer MoS

Transistors. Nat. Nanotechnol. 6, 147–150 (2011).

12. Li, X., Wang, X., Zhang, L., Lee, S. & Dai, H. Chemically Derived, Ultrasmooth Graphene Nanoribbon Semiconductors. Science 319, 1229–1232 (2008).

13. Gomez, L., Aberg, I. & Hoyt, J. L. Electron Transport in Strained-Silicon Directly on Insulator Ultrathin-Body n-MOSFETs With Body Thickness Ranging From 2 to 25 nm. IEEE Electron Device Lett. 28, 285–287 (2007).

14. Liu, L., Kumar, S. B., Ouyang, Y. & Guo, J. Performance Limits of Monolayer Transition Metal Dichalcogenide Transistors. IEEE Trans. Electron Devices 58, 3042–3047 (2011).

15. Gutiérrez, H. R. et al. Extraordinary Room-Temperature Photoluminescence in Triangular WS

Monolayers. Nano Lett. 13, 3447–3454 (2013).

16. Kormányos, A., Zólyomi, V., Drummond, N. D. & Burkard, G. Spin-Orbit Coupling, Quantum Dots, and Qubits in Monolayer Transition Metal Dichalcogenides. Phys.

Rev. X 4, 011034 (2014).

17. Fei, R., Li, W., Li, J. & Yang, L. Giant Piezoelectricity of Monolayer Group IV

Monochalcogenides: SnSe, SnS, GeSe, and GeS. Appl. Phys. Lett. 107, 173104 (2015).

18. Ramakrishna Reddy, K. T., Koteswara Reddy, N. & Miles, R. W. Photovoltaic Properties of SnS Based Solar Cells. Sol. Energy Mater. Sol. Cells 90, 3041–3046 (2006).

19. Mahdi, M. S. et al. A Highly Sensitive Flexible SnS Thin Film Photodetector in The Ultraviolet to Near Infrared Prepared by Chemical Bath Deposition. RSC Adv. 6, 114980–114988 (2016).

20. Mitzi, D. B., Kosbar, L. L., Murray, C. E., Copel, M. & Afzali, A. High-mobility Ultrathin

Semiconducting Films Prepared by Spin Coating. Nature 428, 299 (2004).

21. Gur, I., Fromer, N. A., Geier, M. L. & Alivisatos, A. P. Air-Stable All-Inorganic Nanocrystal Solar Cells Processed from Solution. Science 310, 462–465 (2005).

22. Sze, S. M. & Ng, K. K. Physics of semiconductor devices. (Wiley-Interscience, 2007).

23. Geim, A. K. & Grigorieva, I. V. Van der Waals Heterostructures. Nature 499, 419–425 (2013).

24. Haigh, S. J. et al. Cross-sectional Imaging of Individual Layers and Buried Interfaces of Graphene-based Heterostructures and Superlattices. Nat. Mater. 11, 764 (2012).

25. Hong, X. et al. Ultrafast Charge Transfer in Atomically Thin MoS

/WS

Heterostructures. Nat. Nanotechnol. 9, 682–686 (2014).

26. Yu, Y. et al. Equally Efficient Interlayer Exciton Relaxation and Improved Absorption in Epitaxial and Nonepitaxial MoS

/WS

Heterostructures. Nano Lett. 15, 486–491 (2015).

27. Zhang, W., Huang, Z., Zhang, W. & Li, Y. Two-dimensional Semiconductors with Possible High Room Temperature Mobility. Nano Res. 7, 1731–1737 (2014).

28. Zhang, Y. et al. Controlled Growth of High-Quality Monolayer WS

Layers on Sapphire and Imaging Its Grain Boundary. ACS Nano 7, 8963–8971 (2013).

29. Salehi, S. & Saffarzadeh, A. Atomic Defect States in Monolayers of MoS

and WS

. Surf. Sci. 651, 215–221 (2016).

30. Allain, A., Kang, J., Banerjee, K. & Kis, A. Electrical Contacts to Two-dimensional Semiconductors. Nat. Mater. 14, 1195–1205 (2015).

31. Khalil, H. M. W., Khan, M. F., Eom, J. & Noh, H. Highly Stable and Tunable Chemical Doping of Multilayer WS

Field Effect Transistor: Reduction in Contact Resistance.

ACS Appl. Mater. Interfaces 7, 23589–23596 (2015).

32. Lee, Y. T. et al. Graphene Versus Ohmic Metal as Source-Drain Electrode for MoS

Nanosheet Transistor Channel. Small 10, 2356–2361 (2014).

33. Furchi, M. M. et al. Photovoltaics in Van der Waals Heterostructures. IEEE J. Sel. Top.

Quantum Electron. 23, 106–116 (2017).

Chapter 2

Structure, Electrical Properties, and Preparation Methods of WS 2 and SnS

2.1. Structures of WS2 and SnS

2.1.1. Structures of WS2

WS

belongs to space group symmetry of P63/mmc. WS

has two polytypes which

are 2H-WS

(hexagonal) and 1T-WS

(distorted). These two polytypes have completely

different electrical properties. Semiconducting properties can be found in 2H-WS

and

metallic behavior is observed in 1T-WS

. Interestingly, the semiconducting 2H-WS

is the

most stable phase of WS

that can be found in nature.

In this thesis, the discussion is

limited to 2H-WS

as it is the main focus. In bulk structure, each layer of WS

is stacked

each other by van der Waals interaction. Figures 2-1a,b illustrate the top and side views

of two-layer WS

stacked by the van der Waals interaction. Due to this van der Waals

interaction, the coupling between WS

layers, and quantum confinement, the different

number of layer of WS

results in different electrical and optical properties which will be

discussed in later. Figure 2-1c presents a unit cell of WS

with the lattice parameters of a

= b = 3.154 Å and c = 12.364 Å.

2.1.2. Structure of SnS

Tin sulfide is a compound that can exist in different polymorphs. The different oxidation number of Sn makes tin sulfide exists in three different polymorphs: SnS, SnS

, and Sn

S

. Moreover, SnS itself has three different polytypes that belong to different space group symmetry: Pnma, Fm3m, and Cmcm. SnS with Pnma structure is the most stable phase, and other phases can be obtained at higher deposition temperatures of 700

°C and 1000 °C.

The Pnma is orthorhombic symmetry which shows large anisotropic behavior between the zig-zag and the armchair edge, where the top and side views of SnS

Figure 2-1. Atomic models of top and side view of (a,b) WS² and (d,e) SnS. Schematic view of a unit cell of (c) WS2 and (f) SnS.

a b

c

Sulf ur Tungsten Sulf ur v an der Waals gap

a c

b

Side view

Unit cell

Unit cell

a b

c

f

Tin Sulf ur

v an der Waals gap

Side view Top view

d

e

SnS

WS

are given in Figures 2-1d,e.

Moreover, the symmetry breaking of this phase and flexible structure of SnS generate a giant piezoelectric coefficient in SnS.

Having similar layered structured with WS

, each layer of SnS is also stacked together by van der Waals interactions. However, the van der Waals interaction of each layer of metal chalcogenides (i.e., SnS and SnSs) is very strong and makes it difficult to form single-layer structure of SnS.

Until now; there is no report on properties of single-layer SnS, because the isolation of single-layer SnS is still difficult to be obtained. Figure 2-1f shows one unit cell of SnS with the lattice parameters of a = 4.334, b = 3.987, and c = 11.199.

2.1.3. Phonon modes of WS

and SnS

In order to analyze atomically thin materials, advanced characterization apparatuses with nanometer-scale resolutions are needed. TEM has the ability to observe

Figure 2-2. a) Schematic view of Raman modes of WS². Typical Raman spectra of single-layer WS² collected with excitation wavelength of (b) 514 and (c) 488 nm. d) Peak frequencies of A1g, E12g, and 2LA vibrational modes as a function of number of WS2 layers. Figures b,c are reproduced from Ref. 11

E²2g E1g E¹2g A1g

a

b

d c

each atom arrangement in 2D materials. Furthermore, the TEM apparatus equipped with energy-dispersive X-ray spectroscopy (EDS) module can determine what kind of atoms are existing in the 2D materials.

However, these methods are time-consuming and destructive to the sample. A fast and non-destructive characterization is needed to confirm the presence of 2D materials. Phonon modes studies could give insight into the presence of 2D materials, especially Raman spectroscopy. Since the different 2D material has unique Raman fingerprint, Raman spectroscopy becomes a powerful tool in 2D material characterization. In this thesis, the main characterization to confirm the presences of 2D materials is based on Raman spectroscopy. Therefore, the short introduction to the phonon modes of WS

and SnS is given in this section.

2.1.3.1. Phonon modes of WS2

2H-WS

has four vibrational modes that can be detected by Raman spectroscopy, which assigned to A

and E

, E

, and E

modes.

These vibrational modes are

Figure 2-3. a) Schematic view of Raman vibrations of SnS. b) Typical Raman spectra of SnS with different thickness. Figure b is reproduced from Ref. 13.

A

(40 cm

) A

(96 cm

) A

(192 cm

)

A

(219 cm

) B

(49 cm

) B

(164 cm

)

a b

depicted in Figure 2-2a. A

mode is out-plane atomic displacement that can be detected at Raman shift of 417 cm

. Next, at Raman shift of 356 cm

, the E

, in-plane atomic displacement, also can be detected.

Since these two peaks gives strong Raman intensity, these two modes are usually used to confirm the existence of WS

among other Raman peaks. Another peak called 2LA is originated from the double resonant scattering, which is only detected by green excitation source.

Another interesting point is that the peak positions of A

and E

strongly depend on the number of WS

layers, as shown in Figure 2-2d. The frequency separation of A

and E

is often used for determination of the number of layers of WS

.

2.1.3.2. Phonon modes of SnS

There exist 21 vibrational modes in SnS.

Among these vibrational modes, only two vibrational modes show intense peaks strong enough to be detected by the confocal Raman spectroscopy.

These Raman modes are known as A

and B

which are illustrated in Figure 2-3a. Moreover, the Raman intensity of SnS also behaves similarly to WS

where the larger number of layers of SnS gives more intense Raman intensity. Figure 2-3b shows an example of Raman spectra of SnS with different thickness.

2.2. Electrical properties of WS

and SnS

In this section, the electrical properties of WS

and SnS from theoritical point of

view will be described. This introductory is expected to give some insight how the

electronic devices based on WS

or SnS perform. Then, the summaries of previous work

on empirical electrical properties of WS

and SnS will be given in Chapter 3.

2.2.1. Band structures of WS

and SnS

The band structure of single-layer WS

and those of larger number of layers are shown in Figure 2-4a. First, the discussion of the band gap evolution of WS

will be described. Interestingly, it can be seen that the indirect band gap of WS

becomes wider with the decreasing the number of WS

layers.

On the other hand, the direct band gap of WS

remains unchanged with different thickness, as can be seen in Figure 2-4b.

Eventually, at the single-layer form, the indirect band gap of WS

becomes larger than its direct band gap and the natural band gap of single-layer WS

changes from indirect to direct. This evolution is very beneficial, because the direct bandgap semiconductor is preferable in optoelectronic applications. Theoretical result is consistent with result

Figure 2-4. a) Band structures of bulk, quad-, double-, and single-layer WS2 calculated by DFT/PBE simulation.

b) Calculated indirect and direct band gap of different number (N) of WS² layers. Circles and triangles data sets show the result with different calculation methods. Reproduced from Ref. 14.

E

(h a rt re e )

bulk WS

quad-layer WS

double-layer WS

single-layer WS

a

b

obtained by photoluminescence spectroscopy. Since the direct band semiconductor emits strong photoluminescence, one can distinguish the presence of single- and double- layer WS

just by examining their PL intensities.

The photoluminescence of double- layer WS

is much lower than its single-layer form, originating from the indirect band gap of double-layer WS

.

Next, the band structure of SnS with different number of layers is summarized.

The band gap evolution of SnS also shows a similar trend with WS

, where the SnS band gap increases as the number of layer decreases, as shown in Figures 2-5a-d.

In bulk SnS, the band gap is predicted to be ~1.2 eV. The band gap gradually increases to 1.9 eV for the single-layer case.

However, the indirect-to-direct band gap transition, which is observed in single-layer WS

, is absent in the SnS case. SnS is indirect band gap irrespective of the number of layers. Moreover, the most interesting point is that the band gap evolution of SnS is also influenced by the odd/even number of layer of SnS (Figure 2-

Figure 2-5. a-d) Band structures of (a) bulk, (b) triple-, (c) double-, and (d) single-layer SnS calculated by DFT simulation. b) Calculated band gap of odd-even number of SnS layers with different approximation methods.

Reproduced from Ref. 14.

a b

d c

e

5e)

. It is predicted that the even-numbered layer of SnS has a larger band gap compared with the odd-numbered layer. Moreover, due to high asymmetric structures of odd- numbered layer of SnS, the SnS with an odd number of layer will be having a giant band- splitting of more than 80 meV (see Figures 2-4b,d).

This value is very high compared with other MCs and TMDCs, makiing SnS as a promising material for spintronic applications.

2.2.2. Expected carrier mobility of WS

and SnS

For predicting the field effect carrier mobility of 2D materials, the Takagi model promises a better approximation because it evaluates from 2-dimensional electron gas point of view than other carrier mobility simulations.

At first, this method was used to calculate the carrier mobility of Si with different crystal orientations, and it agreed well with the experimental data. However, recently, it becomes popular to approximate the carrier mobility in 2D materials such as antimonene, fluorinated graphene, and phosphorene.

Thus, the expected carrier mobility of WS

and SnS by using Takagi model are summarized in this section. The Takagi model used for calculating carrier mobility of 2D material (μ

) can be written as:

𝜇

=

𝑘_b𝑇𝑚_e^∗𝑚_d(𝐸_l^𝑖)²

(2-1)

where 𝑚

is the effective carrier (electron or hole) mass along the transport direction and

𝑚

= √𝑚

𝑚

is the average density-of-state mass. The deformation potential and elastic

modulus of the longitudinal acoustic mode are defined as 𝐸

and C

, respectively. The e,

k

, T and ℏ are elementary charge, Boltzmann’s constant, temperature, and reduced

Plank’s constant, respectively.

Equation 2-1 shows that the carrier mobility of 2D materials is inversely proportional to the effective carrier mass of the materials, which is similar to the mobility and effective carrier mass relation in 3D materials. Table 2-1 shows the comparison of effective carrier mass of WS

and SnS and their carrier mobilities.

Furthermore, the calculation result of well-known MoS

(for WS

comparison) and SnSe (for SnS comparison) are also given.

The mobility of WS

is much higher than the mobility of MoS

, because the effective mass of electrons in WS

is lighter than MoS

. In SnS and SnSe case, SnS is predicted to have higher carrier mobility than SnSe. Even though the effective mass of SnS is heavier than one in SnSe, the deformation potential of SnS (9.2 eV) is much smaller than that of SnSe (3.9 eV).

Thus, considering the factor of deformation potential ( 𝐸

) of equation 2-1, smaller deformation potential likely promotes higher carrier mobility, because the deformation potential is inversely proportional to the carrier mobility.

However, the calculated mobility values here are overestimated than the values that are observed experimentally. There are several factors that needed to be considered.

Table 2-1. Expected carrier mobility of WS², MoS², SnS, and SnSe. The data were obtained from ref. 16, 21, and 22.

Material

Effective electron mass me (m0)

Effective hole mass mh (m0)

Electron mobility

μe (cm²/Vs) Hole mobility μh (cm²/Vs)

WS2 0.24 1,103

MoS2 0.45 340

SnS 0.22 10,000

SnSe 0.13 723

First, the intrinsic factors such as impurities, defects, and phonon scattering that can be described by Matthiessen rule, (

𝜇

=

𝜇impurity

+

𝜇_lattices

+

𝜇_defects

+ ⋯ ) , are needed to be considered.

Next, the extrinsic factors might also alter the mobility. For example, the rough substrates or humidity could degrade also the carrier mobility of 2D materials.

Therefore, the review of mobility values and device performance based on experimental data will be described further in Chapter 3.

2.3. Material preparation: CVD growth of TMDCs, MCs, and 2D materials-based heterostructures

CVD is the best method to synthesize high-quality 2D materials in large area up to

now. The development of CVD method of TMDCs is very decent, resulting a lot of

publications in last several years. The simplest way to obtain good TMDCs is to use

transition metal oxide (WO

and MoO

) and solid chalcogenide (S, Se) as the precursors,

because the solid precursors are easy to handle. The first demonstration of CVD growth

of transition metal dichalcogenides (TMDCs) was done by Lee et al. in 2012.

They

heated up the MoO

and sulfur powder at an elevated temperature of 650 °C and the

resulted MoS

grains with lateral size of ~50 μm were deposited on SiO

substrates, as

shown in Figures 2-6a,b. The first report on CVD growth of WS

was recorded in 2013 by

Zhang et al.

The method to grow WS

is almost similar to MoS

growth. The main

difference is that WO

precursor is needed to be heated up to 900 °C due to its high

melting point. The lateral size of WS

made by using this method was around ~50 μm, as

shown in scanning electron microscopy (SEM) image in Figures 2-6c,d.

The studies on metal chalcogenides (MCs) growth, especially SnS, are gaining interest just recently. The first report of SnS growth by CVD was published in 2015.

Again, the method is almost similar to TMDCs growth, which SnO

and solid sulfur precursor were used as feedstock. As shown in Figure 2-7a, the SnO

and sulfur

Figure 2-6. a) Schematic view of MoS2 CVD setup and (b) optical image MoS2 grains on SiO2 substrate. c) Schematic view of WS2 CVD setup and (d) SEM image of a WS2 grain on sapphire substrate. Figures a and b are reproduced from Ref. 25. Figures c and d are reproduced from Ref. 26

a ^MoS

c ^WS

MoS

MoS

b d

Figure 2-7. a) Schematic view of CVD setup for SnS growth and (b) optical image SnS grain on SiO2 substrate.

Reproduced from ref. 27

SnS

a b

precursors were heated inside a quartz tube to 640 °C and 120 °C independently. The resulted SnS grain shown n Figure 2-7b has square shape with lateral size of ~50 μm, reflecting its rhombohedral structure.

Various kinds of 2D materials-based heterostructure can be also synthesized by CVD method, opening the opportunity in large-scale production. The first report on 2D materials-based heterostructures can be found in 2014.

The vertical and in-plane heterostructures of MoS

and WS

were achieved by changing the growth temperatures, as shown in Figures 2-8a,b. The sharp interface between MoS

and WS

, shown in TEM image in Figure 2-8c, is the main advantage over the heterostructure fabricated by exfoliation method. Another approach to achieved different heterostructure is by using two-step CVD. For example, the heterostructure of graphene-MoS

and PbS-MoS

have been reported by the two-step CVD.

This approach needs two times CVD processes, where the first material was synthesized beforehand. In the graphene-MoS

case, the

Figure 2-8. Schematic views and optical micrograph images of (a) lateral and (b) in-plane of MoS2 and WS2

heterostructures resulted from different growth temperature. c) TEM images of interface of in-plane MoS²- WS2 heterostructure. Reproduced from Ref. 28

a b

c

graphene was grown firstly.

Then, the synthesized graphene was advanced into different CVD furnace system to growth the MoS

, resulting in graphene-MoS

heterostructure. This two-step CVD approach can avoid the formation of transition metal alloy of TMDCs (Mo

W

S

) or contamination which can occur when both MoO

and WO

precursors are used in one CVD batch.

References

1. Leong, S. X. et al. 2H → 1T Phase Change in Direct Synthesis of WS

Nanosheets via Solution-Based Electrochemical Exfoliation and Their Catalytic Properties. ACS Appl.

Mater. Interfaces 9, 26350–26356 (2017).

2. Wilson, J. A. & Yoffe, A. D. The Transition Metal Dichalcogenides Discussion and Interpretation of The Observed Optical, Electrical and Structural Properties. Adv.

Phys. 18, 193–335 (1969).

3. Burton, L. A. & Walsh, A. Phase Stability of the Earth-Abundant Tin Sulfides SnS, SnS

, and Sn

S

. J. Phys. Chem. C 116, 24262–24267 (2012).

4. Xia, J. et al. Physical Vapor Deposition Synthesis of Two-Dimensional Orthorhombic SnS Flakes with Strong Angle/temperature-dependent Raman responses. Nanoscale 8, 2063–2070 (2016).

5. Fei, R., Li, W., Li, J. & Yang, L. Giant Piezoelectricity of Monolayer Group IV

Monochalcogenides: SnSe, SnS, GeSe, and GeS. Appl. Phys. Lett. 107, 173104 (2015).

6. Qin, G. et al. Diverse Anisotropy of Phonon Transport in Two-dimensional Group IV–

VI Compounds: A Comparative Study. Nanoscale 8, 11306–11319 (2016).

7. Zhang, L.-C. et al. Tinselenidene: A Two-dimensional Auxetic Material with Ultralow Lattice Thermal Conductivity and Ultrahigh Hole Mobility. Sci. Rep. 6, 19830 (2016).

8. Ehm, L. et al. Pressure-induced Structural Phase Transition in The IV–VI Semiconductor SnS. J. Phys. Condens. Matter 16, 3545 (2004).

9. Haigh, S. J. et al. Cross-sectional Imaging of Individual Layers and Buried Interfaces

of Graphene-based Heterostructures and Superlattices. Nat. Mater. 11, 764 (2012).

10. Zhang, X. et al. Phonon and Raman Scattering of Two-dimensional Transition Metal Dichalcogenides from Monolayer, Multilayer to Bulk Material. Chem. Soc. Rev. 44, 2757–2785 (2015).

11. Berkdemir, A. et al. Identification of Individual and Few Layers of WS

using Raman Spectroscopy. Sci. Rep. 3, 1755 (2013).

12. Chandrasekhar, H. R., Humphreys, R. G., Zwick, U. & Cardona, M. Infrared and Raman Spectra of The IV-VI Compounds SnS and SnSe. Phys. Rev. B 15, 2177–2183 (1977).

13. Li, M. et al. Revealing Anisotropy and Thickness Dependence of Raman Spectra for SnS Flakes. RSC Adv. 7, 48759–48765 (2017).

14. Kuc, A., Zibouche, N. & Heine, T. Influence of Quantum Confinement on The Electronic Structure of The Transition Metal Sulfide TS

. Phys. Rev. B 83, 245213 (2011).

15. Gutiérrez, H. R. et al. Extraordinary Room-Temperature Photoluminescence in Triangular WS

Monolayers. Nano Lett. 13, 3447–3454 (2013).

16. Xin, C. et al. Few-Layer Tin Sulfide: A New Black-Phosphorus-Analogue 2D Material with a Sizeable Band Gap, Odd–Even Quantum Confinement Effect, and High Carrier Mobility. J. Phys. Chem. C 120, 22663–22669 (2016).

17. Takagi, S., Toriumi, A., Iwase, M. & Tango, H. On The Universality of Inversion Layer Mobility in Si MOSFET’s: Part II-effects of Surface Orientation. IEEE Trans. Electron Devices 41, 2363–2368 (1994).

18. Pizzi, G. et al. Performance of Arsenene and Antimonene Double-gate MOSFETs from

First Principles. Nat. Commun. 7, 12585 (2016).

19. Bruzzone, S. & Fiori, G. Ab-initio Simulations of Deformation Potentials and Electron Mobility in Chemically Modified Graphene and Two-Dimensional Hexagonal Boron- nitride. Appl. Phys. Lett. 99, 222108 (2011).

20. Qiao, J., Kong, X., Hu, Z.-X., Yang, F. & Ji, W. High-Mobility Transport Anisotropy and Linear Dichroism in Few-layer Black Phosphorus. Nat. Commun. 5, 4475 (2014).

21. Zhang, W., Huang, Z., Zhang, W. & Li, Y. Two-Dimensional Semiconductors with Possible High Room Temperature Mobility. Nano Res. 7, 1731–1737 (2014).

22. Zhou, M., Chen, X., Li, M. & Du, A. Widely Tunable and Anisotropic Charge Carrier Mobility in Monolayer Tin(ii) Selenide Using Biaxial Strain: A First-principles Study.

J. Mater. Chem. C 5, 1247–1254 (2017).

23. Sze, S. M. & Ng, K. K. Physics of Semiconductor Devices. (Wiley-Interscience, 2007).

24. Chhowalla, M., Jena, D. & Zhang, H. Two-Dimensional Semiconductors for Transistors. Nat. Rev. Mater. 1, natrevmats201652 (2016).

25. Lee, Y.-H. et al. Synthesis of Large-Area MoS

Atomic Layers with Chemical Vapor Deposition. Adv. Mater. 24, 2320–2325 (2012).

26. Zhang, Y. et al. Controlled Growth of High-Quality Monolayer WS

Layers on Sapphire and Imaging Its Grain Boundary. ACS Nano 7, 8963–8971 (2013).

27. Ahn, J.-H. et al. Deterministic Two-Dimensional Polymorphism Growth of Hexagonal n-Type SnS

and Orthorhombic p-Type SnS Crystals. Nano Lett. 15, 3703–3708 (2015).

28. Gong, Y. et al. Vertical and In-plane Heterostructures from WS

/MoS

Monolayers.

Nat. Mater. 13, 1135–1142 (2014).

29. Roy, K. et al. Graphene-MoS

Hybrid Structures for Multifunctional Photoresponsive

Memory Devices. Nat. Nanotechnol. 8, 826–830 (2013).

30. Wen, Y. et al. Integrated High-Performance Infrared Phototransistor Arrays

Composed of Nonlayered PbS–MoS

Heterostructures with Edge Contacts. Nano Lett.

16, 6437–6444 (2016).

This page is intentionally left blank.

Chapter 3

2D Materials-Based Field-Effect Transistors

3.1. Field-effect transistor operation

Complementary metal oxide semiconductor (CMOS) is the heart of electronic

devices. CMOS consists of two different polarities (p- and n-) field-effect transistors

(FETs) that can be switched on and off.

The active materials of FETs are semiconductors

whose conductivity can be tuned by using external electric field. Figures 3-1a shows a

schematic view of 2D-based FETs, where WS

is chosen as semiconductor for the

definitive purpose. The FET operation in 2D materials is more or less similar to that seen

in 3D materials. This FET consists of 4 main components: source, drain, channel, and

dielectric gate.

Source and drain are connected by 2D semiconductor that acts as a

channel. The conductance of the semiconducting channel that can be altered by using

and electric field induced by the gate. The higher conductance of semiconducting channel leads to higher drain-source current value (I

), while lower conductance will give a lower I

value.

Now, the on-off operation of n-type FET is described.

Figure 3-1b shows typical I

-V

curves in linear and logarithmic scale (transfer curve) of n-type FET. When the gate is applied with a positive voltage, the FET is considered at the on-state. The I

value is high due to high conductance of the channel, because the Fermi-level of semiconductor moves to its conduction band edge. The Fermi-level transition is originated in the gate- induced electric-field, which is applied by a positive voltage. When the gate voltage is not applied, the FET is at off-state, because the drain-source current cannot flow in the 2D channel due to low conductance of the channel. The low conductance of the channel is caused by the position Fermi-level of semiconducting channel located far below from its conduction band edge. The on-off operation of p-type FET is vice versa with n-type FET

Figure 3-1. (a) Schematic view of 2D material-based FET device. (b) Typical transfer curve of a conventional n-type FET device. Red and blue curves show logarithmic and linear scales.

Source Drain

Dielectric Gate 2D channel

a

b

-40 -20 0 20 40

10^-13 10^-11 10^-9 10^-7 10^-5

Drain cur re nt (A)

Gate voltage (V)

0 5 10 15 20

Drain cur re nt (  A)

operation where the applied negative voltage on gate makes conductance of p-type semiconducting channel higher.

To evaluate the FET performance, three factors need to be considered. First, the on-off ratio, a ratio between drain current at on- and off- states, must be between 10

and 10

in order to supply the demand in real life applications.

Second, the high carrier mobility is preferable to achieve high performance FET.

The carrier mobility (µ) of FET can be extracted from the I

-V

curve shown in Figure 3-1b by using the following expression:

𝜇 =

𝑊_ch𝐶_g𝑉_DS

(3-1)

where g

=dI

/dV

is transconductance (can be derived from I

-V

curve linear scale), L

is the channel length, W

is the channel width, C

is the gate capacitance, and V

is the drain voltage. Third, the smallest subthreshold swing parameter (the sensitivity of drain current by the gate voltage), S = dV

/d(logI

), is required to obtain the low power operation.

3.2. WS

-and SnS-based FET

Here the example from ref. 6 is chosen, because it is one of the earliest report on

WS

-based FET. Figure 3-2a shows a schematic view of single-layer WS

FET. The

transfer curves given in Figure 3-2b show that the on-off ratios of WS

reach 10

in a

vacuum environment and after annealing treatment. The high on-off ratio of WS

is

originated from its large band gap of ~2 eV. It is worthy to note that the drain current is

low when the measurement was done in low vacuum or in ambient environment.

Generally, the WS

FET tends to show n-type behavior. Even though it was predicted as ambipolar semiconductor, the high amount of sulfur vacancies makes WS

behave as an n-type semiconductor.

The drain current degradation suggests that the WS

is very sensitive to the measurement environment. The increased carrier mobility of WS

at the lower temperature shown in Figure 3-2c indicates that the phonon scattering limits the mobility in WS

. This mobility is very low compared to its prediction shown in Chapter 2. Several attempts have been studied to improve the carrier mobility in WS

by using contact engineering or dielectric encapsulation.

For example, the WS

mobility increased from 163 cm

/Vs to 214 cm

/Vs after encapsulation by h-BN layers.

The

Figure 3-2. (a) Optical micrograph of single-layer WS² FET device. (b) Transfer curves of the single-layer WS² under different environment. (c) Field-effect mobility as a function of temperature. Reproduced from Ref. 6.

a

5 μm

b

c

subthreshold swing parameter (S) could be also improved by the ionic liquid which produce a large electric field. Jo et al. demonstrated the very low subthreshold swing parameter of <60 meV by using gating with one of ionic liquid (Diethylmethyl (2- methoxyethyl) ammonium bis (trifluoromethylsulfonyl) imide).

In Chapter 4, the improvement of WS

carrier mobility by the contact engineering will be described.

Different from WS

, the report on SnS FET is quite scarce. Until now, there are only three reports that have reported SnS FET so far.

The reason might be because the SnS does not show good gate dependence, indicating that the SnS FET device cannot be turned off at room temperature. There are two possible reasons why the SnS FET cannot be turned off. First, the high degree of tin vacancy in SnS generates a large amount of hole concentration (10

cm

), resulting in metallic-like behavior.

Second, the large number of SnS layers screen the electric field applied from the gate, resulting in small gate dependency.

Figures 3-3a shows p-type transfer curves of thin SnS FET (~12 nm) measured at different temperatures. This thin SnS was turned off at low temperature.

Figure 3-3. (a) Optical micrograph of single-layer FET device. (b) Transfer curves of single-layer WS² under different environment. (c) Field-effect mobility as function of temperature. Reproduced from Ref. 10.

a b