p-type doping of MoS2 thin films using NbMasihhur R. Laskar, Digbijoy N. Nath, Lu Ma, Edwin W. Lee II, Choong Hee Lee, Thomas Kent, Zihao Yang,

Rohan Mishra, Manuel A. Roldan, Juan-Carlos Idrobo, Sokrates T. Pantelides, Stephen J. Pennycook, Roberto

C. Myers, Yiying Wu, and Siddharth Rajan

Citation: Applied Physics Letters 104, 092104 (2014); doi: 10.1063/1.4867197 View online: http://dx.doi.org/10.1063/1.4867197 View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/104/9?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Enhancing the photocurrent and photoluminescence of single crystal monolayer MoS2 with resonant plasmonicnanoshells Appl. Phys. Lett. 104, 031112 (2014); 10.1063/1.4862745 Controlled growth of vertically aligned MoO3 nanoflakes by plasma assisted paste sublimation process J. Appl. Phys. 114, 184310 (2013); 10.1063/1.4830278 Electrical performance of monolayer MoS2 field-effect transistors prepared by chemical vapor deposition Appl. Phys. Lett. 102, 193107 (2013); 10.1063/1.4804546 Physical properties of Cu2ZnSnS4 thin films deposited by spray pyrolysis technique J. Renewable Sustainable Energy 5, 023113 (2013); 10.1063/1.4795399 Effect of Li doping in NiO thin films on its transparent and conducting properties and its application inheteroepitaxial p-n junctions J. Appl. Phys. 108, 083715 (2010); 10.1063/1.3499276

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 81.106.1.98

On: Sat, 26 Apr 2014 11:59:21

p-type doping of MoS2 thin films using Nb

Masihhur R. Laskar,1,a) Digbijoy N. Nath,1,a) Lu Ma,2,a) Edwin W. Lee II,1,a)

Choong Hee Lee,1 Thomas Kent,3 Zihao Yang,1 Rohan Mishra,4,5 Manuel A. Roldan,5,6

Juan-Carlos Idrobo,7 Sokrates T. Pantelides,4,5,8 Stephen J. Pennycook,9

Roberto C. Myers,1,3 Yiying Wu,2 and Siddharth Rajan1,3,b)

1Department of Electrical and Computer Engineering, The Ohio State University, Columbus, Ohio 43210, USA2Department of Chemistry, The Ohio State University, Columbus, Ohio 43210, USA3Department of Material Science and Engineering, The Ohio State University, Columbus, Ohio 43210, USA4Department of Physics and Astronomy, Vanderbilt University, Nashville, Tennessee 37235, USA5Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831,USA6Department Fisica Aplicada III, Universidad Complutense de Madrid, Madrid 28040, Spain7Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831,USA8Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville,Tennessee 37235, USA9Department of Materials Science and Engineering, University of Tennessee, Knoxville, Tennessee 37996, USA

(Received 21 December 2013; accepted 9 February 2014; published online 4 March 2014)

We report on the first demonstration of p-type doping in large area few-layer films of (0001)-oriented

chemical vapor deposited MoS2. Niobium was found to act as an efficient acceptor up to relatively

high density in MoS2 films. For a hole density of 3.1� 1020 cm�3, Hall mobility of 8.5 cm2 V�1 s�1

was determined, which matches well with the theoretically expected values. X-ray diffraction scans

and Raman characterization indicated that the film had good out-of-plane crystalline quality.

Absorption measurements showed that the doped sample had similar characteristics to high-quality

undoped samples, with a clear absorption edge at 1.8 eV. Scanning transmission electron microscope

imaging showed ordered crystalline nature of the Nb-doped MoS2 layers stacked in the [0001]

direction. This demonstration of substitutional p-doping in large area epitaxial MoS2 could help in

realizing a wide variety of electrical and opto-electronic devices based on layered metal

dichalcogenides. VC 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4867197]

Transition metal dichalcogenides (TMDs), such as

MoS2, WS2, WSe2, etc., have recently attracted widespread

attention for a variety of next-generation electrical1,2 and

optoelectronic3,4 device applications including low cost,

flexible,5 and transparent electronics.6,7 These layered mate-

rials provide ultra-high confinement and are intrinsically

two-dimensional (2D) in nature, which is therefore promis-

ing for highly scaled vertical transistor topologies.8 Besides,

through van der Waals epitaxy, they circumvent limitations

such as lattice mismatch in heterostructure growths of con-

ventional semiconductors. Devices including field effect

transistors (FETs) with excellent on/off ratio and high cur-

rent densities,1,9 as well as integrated circuits10 have been

reported using flakes of MoS2 mechanically exfoliated from

bulk geological samples. More recently, large area (0001)

oriented MoS2 with excellent crystalline and structural qual-

ities grown by chemical vapor deposited (CVD) on (0001)

sapphire was reported.11 Such CVD-grown MoS2 eliminates

the limitations associated with the commonly used exfoliated

approach, such as control on thickness and area, and are

therefore viable for large scale device fabrication.

The use of Niobium (Nb) as a substitutional impurity on

the metal site to get p-type conductivity was suggested sev-

eral decades ago12–14 for bulk MoS2 and WSe2. Hole

mobility of 10 cm2/Vs in Nb-doped MoS2 flakes obtained by

transport reaction had been reported15 earlier, but such MoS2

flakes were few microns to few hundred microns thick.

Besides, no detailed investigations into the material quality

of such thick Nb-doped MoS2 were reported. P-type conduc-

tivity has also been electrostatically achieved16,17 on MoS2

mechanically exfoliated from geological samples using

back-gating and liquid-gating approaches. However, no

report exists for in situ p-doping using an acceptor dopant in

epitaxial (and even mechanically exfoliated) MoS2 thin

films, which is necessary for demonstration of MoS2-based

bipolar devices such as heterojunction bipolar transistors

(HBT), light emitting diodes (LEDs), and photodetectors.

In this work, we show that as predicted by density func-

tional theory (DFT) based calculations,18 Nb behaves as an

acceptor in MoS2. Raman spectroscopy, X-ray diffraction

(XRD) scans, and aberration-corrected scanning transmis-

sion electron microscopy (STEM) imaging indicated that the

crystalline nature of MoS2 was preserved after Nb-doping,

and optical absorbance measurements showed an absorption

edge at 1.8 eV indicative of a direct band-to-band transition

in the semiconductor. Finally, electrical measurements dem-

onstrated p-type conductivity in these films with a mobility

of 8.5 cm2/Vs at room temperature and a low contact resist-

ance of 0.6 X-mm.

We followed a vapor deposition method for the growth

of MoS219 that was demonstrated to lead to high crystalline

quality MoS2 in previous work 11. A series of three (0001)

a)M. Laskar, D. Nath, L. Ma, and E. Lee contributed equally to this work.b)Author to whom correspondence should be addressed. Electronic mail:

rajan.21@osu.edu

0003-6951/2014/104(9)/092104/4/$30.00 VC 2014 AIP Publishing LLC104, 092104-1

APPLIED PHYSICS LETTERS 104, 092104 (2014)

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 81.106.1.98

On: Sat, 26 Apr 2014 11:59:21

sapphire samples were prepared for MoS2 growth by e-beam

evaporation of Mo (2.5 nm)/Nb/Mo (2.5 nm) (schematic in

Fig. 1(a)) layers with thickness of Nb varied as 0.3 nm (sam-

ple A), 0.2 nm (sample B), and 0 nm (control sample C,

undoped). The typical dimensions of the samples were 1–

2 cm� 0.7 cm, with the breadth of the samples being limited

by the size of the quartz tube used for the deposition. The

sample was then subjected to sulfurization in the CVD cham-

ber at 900 �C for 10 min. Further details of the CVD growth

of undoped MoS2 on sapphire are reported elsewhere.11 The

conditions optimum for growing large area undoped MoS2

with excellent structural and surface qualities as reported in

Ref. 11 were used in this study to grow all the samples.

Using high-resolution STEM techniques, the thickness of

MoS2 samples grown under these conditions was found to be

10 nm and 13 nm for the undoped and Nb-doped samples,

respectively. The difference in thickness is attributed to ex-

perimental variation, rather than a systematic effect of Nb

doping.

A Ni (30 nm)/Au (50 nm)/Ni (30 nm) metal stack was

deposited by e-beam evaporation to form Ohmic contacts to

Nb-doped MoS2 (samples A and B). Devices were isolated

by etching MoS2 using BCl3/Ar-based inductively coupled

plasma/reactive ion (ICP-RIE) etch chemistry. Hall measure-

ments were performed using standard van der Pauw pads,

and both samples A and B had positive Hall coefficients indi-

cating hole transport. Temperature dependent Hall measure-

ments for sample A showed no carrier freeze-out even at

20 K (Fig. 1(b)), indicating that degenerate p-type doping

had been achieved. Negligible dependence of hole mobility

on temperature was observed for sample A. The room tem-

perature hole mobility measured for sample A was 0.5

cm2/Vs with a hole density of NA¼ 1.5� 1021 cm�3.

Sample B with reduced Nb thickness (0.2 nm) was found to

exhibit a room temperature hole mobility of 8.5 cm2/Vs with

a corresponding p-type charge density of 3.1� 1020 cm�3.

From transfer length method (TLM) measurements, a low

contact resistance of 0.6 X mm was extracted for sample B

(Fig. 1(c)). The sheet resistance extracted from TLM was

1.8 kX/� which was found to match that obtained from Hall

measurement. The significant improvement in hole mobility

(8.5 cm2/Vs from 0.5 cm2/Vs) with a reduction in p-doping

density indicates that the mobility is limited mainly by ion-

ized impurity scattering at such high degenerate doping

densities.

A simple estimate was made for impurity scattering lim-

ited hole mobility in bulk MoS2 using three-dimensional

(3D) formalism, given by lh¼ q seff/m*, where lh is the

hole mobility in MoS2 while seff and m* are effective hole

scattering time and effective hole mass, respectively. An

in-plane effective hole mass of 0.43 is used for the mobility

estimate based on DFT calculations for unstrained bulk

MoS2.20 The momentum scattering time sm(NA,EF) at Fermi

level EF was calculated using21

1

smðNA;EFÞ¼ p NA

�h

q2LTF2

e0es

� �2

gcðEFÞ: (1)

Here, �h is reduced Planck constant, es is the dielectric con-

stant9 of MoS2 (¼7.6), and T is temperature. gc(E) in Eq. (1)

is the 3D density of states for holes and LTF is the

Thomas-Fermi screening length for a degenerate gas given

by LTF ¼ffiffiffiffiffiffiffiffiffiffiffiffiffi

es

q2gcðEFÞ

q. Fig. 1(d) shows hole mobility as a func-

tion of acceptor density (NA) in p-MoS2, as calculated by

Eq. (1). Given the large uncertainty in various parameters

used in the model, the agreement with the experimental data

is satisfactory.

The material and structural quality of the MoS2 samples

were assessed using Raman spectroscopy, high resolution

XRD scans, aberration-corrected STEM imaging, and optical

absorbance measurements. The undoped MoS2 (sample C),

as well as Nb-doped p-type samples A and B were found to

FIG. 1. (a) Schematic of Nb-doping in

CVD grown MoS2. (b) Temperature-

dependent Hall measurement on

degenerately doped sample A (p¼ 1.5

� 1021 cm�3) showing no carrier

freeze-out at 20 K. (c) TLM fitting to

extract sheet and contact resistance on

sample B using Ni/Au/Ni metal con-

tacts. (d) Ionized impurity scattering

limited hole mobility in p-MoS2, theo-

retically estimated using 3D formalism,

compared with measured data.

092104-2 Laskar et al. Appl. Phys. Lett. 104, 092104 (2014)

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 81.106.1.98

On: Sat, 26 Apr 2014 11:59:21

be crystalline in nature exhibiting good structural and mate-

rial qualities as evident from the measurements using these

various characterization techniques. Fig. 2 shows the on-axis

XRD scans for the samples A, B, and C. The (002) peak of

MoS2 was observed in all the three scans although the inten-

sity for sample A was lower, indicating degradation of mate-

rial quality under such high doping levels (1.5� 1021 cm�3).

Sample B exhibited high intensity (002), (004), and (006)

peaks, indicative of good crystalline quality of MoS2. Fig.

3(a) shows the Raman spectra of the samples taken with a

Renishaw, 514 nm laser (with 60mW power). Characteristic

in-plane (E12g) and out-of-plane (A1g) vibrational modes

were observed at 381 and 407 cm�1, respectively,22,23 for all

the three samples. This indicates that the phonon spectrum of

MoS2 remains unchanged despite the addition of Nb.

Optical absorbance measurements were performed on

all the samples using a broad UV-VIS-NIR deuterium-tung-

sten-halogen white light source. The absorbance spectra was

taken for a reference piece of sapphire in order to determine

a reference intensity I0(k). The MoS2 sample was measured

and the absorbance was determined using A ¼ IðkÞI0ðkÞ, where

I(k) is the intensity collected by the monochromator after the

light is transmitted through the MoS2 grown on sapphire.

The normalized absorbance spectra (Fig. 3(b)) show that all

three samples A, B, and C exhibit an absorption edge at

1.8 eV indicative of a direct band-to-band transition in these

semiconductors. DFT calculations in Ref. 24 show that

although the lowest energy transition in bulk MoS2 is

approximately 1.2 eV, the direct transition of approximately

1.8 eV still exists at the K point. This is verified25 by the fact

that both PL as well as absorbance measurements on bulk

MoS2 have been found to exhibit a prominent transition or

peak around 1.8–1.9 eV which has been attributed to

direct-gap transitions between split valence band maxima

(v1 and v2) and conduction band minima all located at Kpoint of Brillouin zone. For the Nb-doped samples A and B,

the 1.8 eV direct gap is found to exhibit clear band tails,

which is a good indicator of the shallow Nb states. The mul-

tiple blue/UV peaks are possibly due to transitions involving

higher bands.

An atomic force micrograph (AFM) of sample B shown

in Fig. 4 (rms roughness �1.3 nm, scans size: 2 lm� 2 lm)

indicated complete coverage of surface by MoS2 and a rela-

tively smooth surface.

STEM imaging was performed on an aberration-

corrected Nion UltraSTEM-100TM electron microscope oper-

ating at an acceleration voltage of 60 kV.26 Fig. 5 shows an

atomic resolution Z-contrast STEM image of the Nb-doped

sample B which exhibited the highest mobility of 8.5 cm2/Vs.

The Z-contrast image reveals that the ordered crystalline na-

ture of the MoS2 layers stacked in the [0001] direction on a

sapphire substrate is preserved even with degenerate p-type

doping. Based on larger HRTEM images, the grain sizes are

estimated to be larger than 100 nm. Recent MoS2 samples

(unrelated to this present experiment) grown under slightly

different conditions have large area epitaxial ordering (as

shown by off-axis XRD spectra with 6-fold symmetric

peaks). However, the AFM morphology was quite similar to

FIG. 2. High-resolution XRD scans for samples A (p¼ 1.5 � 1021 cm�3), B

(p¼ 3.1 � 1020 cm�3), and C (undoped).

FIG. 3. (a) Raman spectra on samples

A (p¼ 1.5 � 1021 cm�3), B (p¼ 3.1

� 1020 cm�3), and C (undoped). (b)

Absorbance measurements for samples

A (p¼ 1.5 � 1021 cm�3), B (p¼ 3.1

� 1020 cm�3), and C (undoped).

092104-3 Laskar et al. Appl. Phys. Lett. 104, 092104 (2014)

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 81.106.1.98

On: Sat, 26 Apr 2014 11:59:21

the samples discussed in this paper. Therefore, the AFM mor-

phology is believed to be probably indicative of adatom dif-

fusion dynamics rather than grain size. More investigations

are currently carried out to estimate the grain size.

In conclusion, we show that Nb can act as an efficient

acceptor in MoS2 leading to high hole density and relatively

high mobility. For a hole concentration of 3.1� 1020 cm�3, a

hole mobility of 8.5 cm2/Vs was measured at room tempera-

ture and was found to be limited by ionized impurity limited

scattering. The use of Nb substitutional impurity for p-type

doping demonstrated here for MoS2 could be extended to

other dichalcogenides and could therefore have wider appli-

cations. Furthermore, the simple deposition scheme used

here could be employed to directly pattern areas with p-type

MoS2, thus providing flexibility for device design. This first

demonstration of substitutional p-type doping in large area

thin film CVD-grown MoS2 is expected to enable several

high-performance electrical and opto-electronic devices.

S.R. acknowledges funding from the NSF NSEC

(CANPD) Program (EEC0914790) and NSF Grant ECCS-

0925529. L.M. and Y.W. acknowledge the support from NSF

(CAREER, DMR-0955471). M.L. was supported by the OSU

NSF MRSEC CEM Seed Program. R.C.M. acknowledges

funding from NSF (CAREER, DMR-1055164) and (CBET-

1133589). This research was partially supported by the Office

of Basic Energy Sciences, Materials Sciences and Engineering

Division, U.S. Department of Energy (RM, MAR, STP, SJP),

the Center for Nanophase Materials Sciences User Facility

(J.C.I.), which is sponsored at Oak Ridge National Laboratory

(ORNL) by the Scientific User Facility Division, Office of

Basic Energy Sciences, U.S. Department of Energy, the

European Research Council Starting Investigator Award

STEMOX #239739 (MAR).

1B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti, and A. Kis, Nat.

Nanotechnol. 6, 147 (2011).2B. Radisavljevic, M. B. Whitwick, and A. Kis, ACS Nano 5, 9934 (2011).3R. S. Sundaram, M. Engel, A. Lombardo, R. Krupke, A. C. Ferrari, Ph.

Avouris, and M. Steiner, Nano Lett. 13(4), 1416–1421 (2013).4W. Choi, M. Y. Cho, A. Konar, J. H. Lee, G.-B. Cha, S. C. Hong, S. Kim,

J. Kim, D. Jena, J. Joo, and S. Kim, Adv. Mater. 24(43), 5832–5836

(2012).5H.-Y. Chang, S. Yang, J. Lee, L. Tao, W.-S. Hwang, D. Jena, N. Lu, and

D. Akinwande, ACS Nano 7(6), 5446–5452 (2013).6T. Georgiou, R. Jalil, B. D. Belle, L. Britnell, R. V. Gorbachev, S. V.

Morozov, Y.-J. Kim, A. Gholinia, S. J. Haigh, O. Makarovsky, L. Eaves,

L. A. Ponomarenko, A. K. Geim, K. S. Novoselov, and A. Mishchenko,

Nat. Nanotechnol. 8(2), 100–103 (2012).7G.-H. Lee, Y.-J. Yu, X. Cui, N. Petrone, C.-H. Lee, M. S. Choi, D.-Y. Lee,

C. Lee, W. J. Yoo, K. Watanabe, T. Taniguchi, C. Nuckolls, P. Kim, and

J. Hone, ACS Nano 7(9), 7931–7936 (2013).8Y. Yoon, K. Ganapathi, and S. Salahuddin, Nano Lett. 11, 3768 (2011).9S. Kim, A. Konar, W.-S. Hwang, J. H. Lee, J. Lee, J. Yang, C. Jung, H.

Kim, J.-B. Yoo, J.-Y. Choi, Y. W. Jin, S. Y. Lee, D. Jena, W. Choi, and K.

Kim, Nat. Commun. 3, 1011 (2012).10H. Wang, L. Yu, Y.-H. Lee, Y. Shi, A. Hsu, M. L. Chin, L.-J. Li,

M. Dubey, J. Kong, and T. Palacios, Nano Lett. 12(9), 4674–4680

(2012).11M. Laskar, L. Ma, S. Kannappan, P. S. Park, S. Krishnamoorthy, D. N.

Nath, W. Lu, Y. Wu, and S. Rajan, Appl. Phys. Lett. 102(2), 252108

(2013).12J. A. Wilson and A. D. Yoffe, Adv. Phys. 18(73), 193–335 (1969).13R. S. Title and M. W. Shafer, Phys. Rev. Lett. 28(13), 808–810 (1972).14J. V. Acrivos, W. Y. Liang, J. A. Wilson, and A. D. Yoffe, J. Phys. C:

Solid State Phys. 4(1), L18–L20 (1971).15R. Fivaz and E. Mooser, Phys. Rev. 163(3), 743–755 (1967).16A. Pospischil, M. M. Furuchi, and T. Mueller, “Solar energy conversion

and light emission in an atomic monolayer p-n diode,” arXiv:1309.7492

(2013).17Y. J. Zhang, J. T. Ye, Y. Yomogida, T. Takenobu, and Y. Iwasa, Nano

Lett. 13(7), 3023–3028 (2013).18K. Dolui, I. Rungger, C. D. Pemmaraju, and S. Sanvito, Phys. Rev. B

88(7), 075420 (2013).19Y. Zhan, Z. Liu, S. Najmaei, P. M. Ajayan, and J. Lou, Small 8(7),

966–971 (2012).20H. Peelaers and C. G. Van de Walle, Phys. Rev. B 86(24), 241401 (2012).21M. Lundstrom, Fundamentals of Carrier Transport (Cambridge

University Press, Digitally printed edition, 2009).22C. Lee, H. Uan, L. E. Brus, T. F. Heinz, J. Hone, and S. Ryu, ACS Nano

4(5), 2695 (2010).23A. Molina-Sanchez and L. Wirtz, Phys. Rev. B 84(15), 155413 (2011).24A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.-Y. Chim, G. Galli, and

F. Wang, Nano Lett. 10(4), 1271–1275 (2010).25K. F. Mak, C. Lee, J. Hone, J. Shan, and T. F. Heinz, Phys. Rev Lett. 105,

136805 (2010).26O. L. Krivanek, G. J. Corbin, N. Delby, B. F. Elston, R. J. Keyse, M. F.

Murfitt, C. S. Own, Z. S. Szilagyi, and J. W. Woodruff, Ultramicroscopy

108, 179–195 (2008).

FIG. 4. AFM of sample B (2 lm� 2 lm scan), data scale: 8 nm; RMS

roughness: 1.3 nm.

FIG. 5. Z-contrast STEM image of Nb-MoS2 film (sample B) on Al2O3 sub-

strate showing the stacking of MoS2 layers in the [0001] direction.

092104-4 Laskar et al. Appl. Phys. Lett. 104, 092104 (2014)

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 81.106.1.98

On: Sat, 26 Apr 2014 11:59:21