Literature DB >> 25852359

Influence of post-annealing on the off current of MoS2 field-effect transistors.

Seok Daniel Namgung1, Suk Yang1, Kyung Park2, Ah-Jin Cho1, Hojoong Kim1, Jang-Yeon Kwon1.   

Abstract

Two-dimensional materials have recently been spotlighted, due to their unique properties in comparison with conventional bulk and thin-film materials. Among those materials, MoS2 is one of the promising candidates for the active layer of electronic devices because it shows high electron mobility and pristine band gap. In this paper, we focus on the evolution of the electrical property of the MoS2 field-effect transistor (FET) as a function of post-annealing temperature. The results indicate that the off current drastically decreased at 200°C and increased at 400°C while other factors, such as the mobility and threshold voltage, show little variation. We consider that the decreasing off current comes from the rearrangement of the MoS2 film and the elimination of the surface residue. Then, the increasing off current was caused by the change of the material's composition and adsorption of H2O and O2.

Entities:  

Keywords:  Field-effect mobility; Field-effect transistors; MoS2; Molybdenum disulfide; On/off current ratio

Year:  2015        PMID: 25852359      PMCID: PMC4385010          DOI: 10.1186/s11671-015-0773-y

Source DB:  PubMed          Journal:  Nanoscale Res Lett        ISSN: 1556-276X            Impact factor:   4.703


Background

Two-dimensional (2D) materials, such as graphene and transition metal dichalcogenides (MoS2, MoSe2, WS2, etc.), are widely used recently for fabricating next-generation nanoelectronics [1-10]. This is because of the high electron mobility of 2D materials, compared with the original bulk material. Typically, graphene shows over 5,000 cm2/Vs of electron mobility [11], and this feature is valuable for applications such as sensors [12] and photovoltaic cells [13]. However, graphene has a fundamental disadvantage for electronic devices, which is the lack of an intrinsic band gap. This has resulted in several reports of insufficient on/off current ratio of field-effect transistors (FETs) [14-17]. Though engineering a band gap of graphene can be an answer for this technical issue, it increases the number of fabrication steps [18,19] and reduces the electron mobility of graphene [20]. As an alternative, MoS2 has an intrinsic band gap, which leads to reduced off current. For example, MoS2 FETs have in general recorded an on/off current ratio of 105 ~ 1010 [21-28], and some MoS2 FETs with high-k dielectrics have recorded an electron mobility of 200 cm2/Vs, which is higher than that of band gap-engineered graphene [21]. Many reports have announced that the annealing process is dispensable for improving the electrical property of various FETs using original IV semiconductors [29], oxide semiconductors [30,31], layered semiconductors [32-34], etc. In the case of 4H-SiC included in the original IV, the annealing process created a passivation layer at the interface, and device parameters were improved, such as the electron mobility and subthreshold swing (SS). In the case of InGaZnO included in oxide semiconductors, the annealing process rearranged defects, and all the device parameters improved, such as Vth, SS, mobility, hysteresis, and the on/off current ratio. For graphene included in a layered material, the annealing process eliminated the resist residue on the surface and increased conductance. For MoS2, a few results have been reported from the viewpoint of the post-annealing process [21,23,26]. One paper showed variation in the optical property, by observing the change of the photoluminescence (PL) peak of single-layer MoS2 with respect to post-annealing [35]. Although it did not evaluate the electrical property of FETs, it reported that the annealing process induced structural rearrangement, and this could also affect the electrical properties of MoS2. Another paper investigated the influence of vacuum annealing on MoS2 FET during measurement of the electrical property [22]. It announced a drastic improvement of electrical performance by annealing, especially in the conductance of the device. However, it focused on the electrical characteristics caused by movement of carriers at elevated temperature, which consequently present the thermally activated characteristics of MoS2 FET. Here, we summarize the evolution of the electrical performance of MoS2 FET at room temperature, which is the conventional operating temperature, with various post-annealing temperatures.

Methods

MoS2 flakes were prepared using a scotch-tape micromechanical cleavage technique, from bulk MoS2 crystal (429ML-AB, SPI Supplies, Inc., West Chester, PA, USA), and were transferred to highly doped silicon substrates covered with 300-nm-thick SiO2. Source and drain (S/D) were patterned by photolithography, and 50-nm-thick Ti was deposited by an e-beam evaporator. Then, a conventional lift-off process was accomplished for the patterning of the S/D electrode. The fabricated MoS2 FET was annealed in a nitrogen environment for 2 h at various temperatures. The electrical characteristic was measured under atmospheric pressure at room temperature. Furthermore, the thicknesses of the MoS2 flakes were measured using atomic force microscopy (AFM; XE-100, Park Systems, Suwon, South Korea).

Results and discussion

Figure 1a is a schematic diagram of the MoS2 FET, and Figure 1b is an AFM profile that corresponds to the red line of the MoS2 image from the inset. The thickness of the MoS2 channel measured by AFM was 11 nm. While there has been controversy over whether using a single-layer MoS2 channel is a requirement for getting higher device performance, some papers proved that a multilayer MoS2 channel was also able to attain comparable device performance, such as a high electron mobility over 100 cm2/Vs and a high on/off current ratio of over 106 [23,36]. Therefore, it is thought that the performance of the multilayer MoS2 is sufficient to study the post-annealing effect.
Figure 1

Schematic representation of MoS FET and AFM image. (a) Schematic representation of MoS2 FET with highly doped silicon as the back gate and (b) atomic force microscopy (AFM) height profile of multilayer MoS2 that has a thickness of 11 nm. The inset is the corresponding AFM image.

Schematic representation of MoS FET and AFM image. (a) Schematic representation of MoS2 FET with highly doped silicon as the back gate and (b) atomic force microscopy (AFM) height profile of multilayer MoS2 that has a thickness of 11 nm. The inset is the corresponding AFM image. Figure 2 shows the representative Id-Vg characteristics under constant Vd = 10 V, with respect to the post-annealing temperature among the many multilayer MoS2 FETs shown in Additional file 1: Figure S1. This representative flake has a channel length of 10 μm and a width of 20 μm. This represents the n-type nature of the MoS2 channel that makes the accumulation layer of electrons at positive gate biases, and it is observed as increasing the drain current at positive gate biases. Theoretically, the drain current is supposed to be below 10−9 A at high negative gate biases, due to a depletion layer; however, a drain current of over 10−5 A was observed with various gate biases at room temperature (black line) and at 400°C (pink line). The drain current at the high negative gate biases drastically decreased by approximately 106, compared to that of the device under room temperature, and it seemed that MoS2 has a depletion layer at both 200°C (red line) and 300°C (blue line).
Figure 2

Transfer curves of the back-gated MoS transistor according to the post-annealing temperature. Transfer curves at room temperature (black), 200°C (red), 300°C (blue), and 400°C (pink) of MoS2 FET under various annealing temperatures, at V d = 10 V.

Transfer curves of the back-gated MoS transistor according to the post-annealing temperature. Transfer curves at room temperature (black), 200°C (red), 300°C (blue), and 400°C (pink) of MoS2 FET under various annealing temperatures, at V d = 10 V. In Figure 3a, the aforementioned transfer curves are arranged in terms of on and off current, with respect to post-annealing temperatures. The on current was defined as the highest drain current measured at high positive gate biases, and the off current was defined as the lowest drain current recorded at low negative gate biases. Figure 3a shows that the on current consistently decreases as the post-annealing temperature increases, while the off current decreases up to 200°C and increases with further increase of temperature. The lowest value of the off current was observed as approximately 10−11 A for the 200°C-annealed device, and this trend is in line with the transfer curve characteristics.
Figure 3

Trends of on current, off current, and field-effect mobility. (a) Details of I on and I off and (b) trends of field-effect mobility in terms of annealing temperature.

Trends of on current, off current, and field-effect mobility. (a) Details of I on and I off and (b) trends of field-effect mobility in terms of annealing temperature. Figure 3b elaborates the field-effect mobility, which increased as the temperature rose and reached a high value of approximately 20.7 cm2/Vs at 200°C and 300°C. The field-effect mobility with respect to the post-annealing temperature is also in accordance with the trend of the off current. Table 1 summarizes the details of the FET device performance parameters as annealing temperature.
Table 1

Device performance summary

Temperature On/off current ratio On current (A) Off current (A) Field-effect mobility (cm 2 /Vs) Subthreshold swing [V/dec]
Room temperature3.5 × 1001 6.38 × 10−04 1.80 × 10−05 8.7536.20
200°C1.7 × 1007 4.06 × 10−04 2.34 × 10−11 21.190.91
300°C8.7 × 1006 3.87 × 10−04 4.43 × 10−11 20.211.43
400°C3.2 × 1000 8.39 × 10−05 2.66 × 10−05 4.3477.51

Exact values of the on/off current ratio, on current, off current, subthreshold swing, and field-effect mobility, at different temperatures.

Device performance summary Exact values of the on/off current ratio, on current, off current, subthreshold swing, and field-effect mobility, at different temperatures. Under those trends, the status of the device can be categorized into two regions. The first region, here termed region I, is that in which the device performance improves from room temperature to 200°C with decreasing off current and increasing field-effect mobility. The second region (region II) is that in which the device performance degrades from 200°C to 400°C with increasing off current and decreasing field-effect mobility. In region I, the decrease of off current is thought to be caused by the atomic arrangement of MoS2 atoms in local sites due to thermal energy. This kind of internal structural modification ends up with the release of a native point defect at the interface between the insulator and the channel material [30]. The interface properties between the MoS2 and SiO2 seemed to be improved, in that the subthreshold swing decreased from 36.20 to 0.91 [V/dec], as the post-annealing temperature increased to 200°C. Also, it is thought that the resist residue included during the fabrication process might be eliminated by the post-annealing process. The photoresist and organic materials from the 3M tape (3M, St. Paul, MN, USA) are one of the plausible candidates to be eliminated, and specifically, elimination of the photoresist residue of the graphene FET was observed with improvement of the device performance during the post-annealing process [33]. In region II, as mentioned, an increase of the off current by 5 or 6 orders was measured. First, it is thought that such huge increase is caused by the change of the channel material itself. This is supported by the case of oxide semiconductors, such as InGaZnO4 where desorption of Zn and O atoms over 700°C annealing and degradation in device performance were observed [30]. Similarly, the results of the X-ray photoelectron spectroscopy (XPS) proved that the S to Mo composition ratio significantly increased after annealing at 400°C in N2 (Table 2). Furthermore, time-of-flight secondary ion mass spectroscopy (TOF-SIMS) depth profiles in Additional file 1: Figure S2 show that Mo decreased after annealing at 400°C in N2, which correlated with the XPS data.
Table 2

MoS composition ratio change based on XPS data

Sample condition Atom Atomic % Simplified ratio (let Mo be 1)
Non-annealedS63.141.712
Mo36.861
400°C-annealedS69.512.280
Mo30.491

Change of composition ratio between molybdenum and sulfur with respect to post-annealing.

MoS composition ratio change based on XPS data Change of composition ratio between molybdenum and sulfur with respect to post-annealing. From Figure 4, Mo 3d5/2 and S 2p3/2 peaks were shifted in a higher energy by 0.6 and 0.5 eV, respectively, after annealing at 400°C. The molybdenum peak shift means that Mo4+ (228.98 eV) was changed into Mo5+ (230.3 eV) [37], and the S 2p3/2 peak shift toward a high binding energy (over 161.88 eV) has been ascribed to polysulfide or thiomolybdate species [38]. That is, one of the strong candidates for explaining the increase of the off current is the phase transformation of MoS2 into Mo2S5 [39] by thermal energy. Furthermore, previous literature [40] provided evidence for this changed form to have high off current in terms of resistivity.
Figure 4

XPS data of the non-annealed and 400°C-annealed MoS . (a) Molybdenum peak and (b) sulfur peak of the non-annealed (black) and 400°C-annealed (red) MoS2.

XPS data of the non-annealed and 400°C-annealed MoS . (a) Molybdenum peak and (b) sulfur peak of the non-annealed (black) and 400°C-annealed (red) MoS2. From a different point of view, adsorption of H2O and O2 on MoS2 can also be one of the reasons for the increase of the off current. Under vacuum conditions, the off current actually decreased by average 102 level and this change is elaborated in Additional file 1: Figure S3. Therefore, it is guessed that adsorption was carried out after the high-temperature annealing process for the measurement of electrical characteristics at an atmosphere environment, and it was also supported by the case of graphene [41].

Conclusions

The evolution of off current for MoS2 FET due to annealing temperature was systematically analyzed. As a result, the off current decreased up to 200°C annealing and increased for higher temperature annealing. Plausible explanations for the decrease in off current are the rearrangement of MoS2 atoms and the elimination of the surface residue. Possible explanations for the increase in off current are the changes of the material's composition ratio and adsorption of H2O and O2. This research is meaningful in that the off current was controlled by the post-annealing temperature.
  23 in total

1.  High-performance single layered WSe₂ p-FETs with chemically doped contacts.

Authors:  Hui Fang; Steven Chuang; Ting Chia Chang; Kuniharu Takei; Toshitake Takahashi; Ali Javey
Journal:  Nano Lett       Date:  2012-06-19       Impact factor: 11.189

2.  Graphene-on-silicon Schottky junction solar cells.

Authors:  Xinming Li; Hongwei Zhu; Kunlin Wang; Anyuan Cao; Jinquan Wei; Chunyan Li; Yi Jia; Zhen Li; Xiao Li; Dehai Wu
Journal:  Adv Mater       Date:  2010-07-06       Impact factor: 30.849

3.  Chemically derived, ultrasmooth graphene nanoribbon semiconductors.

Authors:  Xiaolin Li; Xinran Wang; Li Zhang; Sangwon Lee; Hongjie Dai
Journal:  Science       Date:  2008-01-24       Impact factor: 47.728

4.  High-sensitivity photodetectors based on multilayer GaTe flakes.

Authors:  Fucai Liu; Hidekazu Shimotani; Hui Shang; Thangavel Kanagasekaran; Viktor Zólyomi; Neil Drummond; Vladimir I Fal'ko; Katsumi Tanigaki
Journal:  ACS Nano       Date:  2014-01-06       Impact factor: 15.881

5.  How good can monolayer MoS₂ transistors be?

Authors:  Youngki Yoon; Kartik Ganapathi; Sayeef Salahuddin
Journal:  Nano Lett       Date:  2011-08-02       Impact factor: 11.189

6.  MoS₂ P-type transistors and diodes enabled by high work function MoOx contacts.

Authors:  Steven Chuang; Corsin Battaglia; Angelica Azcatl; Stephen McDonnell; Jeong Seuk Kang; Xingtian Yin; Mahmut Tosun; Rehan Kapadia; Hui Fang; Robert M Wallace; Ali Javey
Journal:  Nano Lett       Date:  2014-02-27       Impact factor: 11.189

7.  Channel length scaling of MoS2 MOSFETs.

Authors:  Han Liu; Adam T Neal; Peide D Ye
Journal:  ACS Nano       Date:  2012-09-12       Impact factor: 15.881

8.  Vertical field-effect transistor based on graphene-WS2 heterostructures for flexible and transparent electronics.

Authors:  Thanasis Georgiou; Rashid Jalil; Branson D Belle; Liam Britnell; Roman V Gorbachev; Sergey V Morozov; Yong-Jin Kim; Ali Gholinia; Sarah J Haigh; Oleg Makarovsky; Laurence Eaves; Leonid A Ponomarenko; Andre K Geim; Kostya S Novoselov; Artem Mishchenko
Journal:  Nat Nanotechnol       Date:  2012-12-23       Impact factor: 39.213

9.  High performance multilayer MoS2 transistors with scandium contacts.

Authors:  Saptarshi Das; Hong-Yan Chen; Ashish Verma Penumatcha; Joerg Appenzeller
Journal:  Nano Lett       Date:  2012-12-19       Impact factor: 11.189

10.  Toward intrinsic graphene surfaces: a systematic study on thermal annealing and wet-chemical treatment of SiO2-supported graphene devices.

Authors:  Zengguang Cheng; Qiaoyu Zhou; Chenxuan Wang; Qiang Li; Chen Wang; Ying Fang
Journal:  Nano Lett       Date:  2011-01-10       Impact factor: 11.189

View more
  4 in total

1.  Polarity control in WSe2 double-gate transistors.

Authors:  Giovanni V Resta; Surajit Sutar; Yashwanth Balaji; Dennis Lin; Praveen Raghavan; Iuliana Radu; Francky Catthoor; Aaron Thean; Pierre-Emmanuel Gaillardon; Giovanni de Micheli
Journal:  Sci Rep       Date:  2016-07-08       Impact factor: 4.379

2.  The Two-Dimensional Nanocomposite of Molybdenum Disulfide and Nitrogen-Doped Graphene Oxide for Efficient Counter Electrode of Dye-Sensitized Solar Cells.

Authors:  Chao-Kuang Cheng; Che-Hsien Lin; Hsuan-Chung Wu; Chen-Chi M Ma; Tsung-Kuang Yeh; Huei-Yu Chou; Chuen-Horng Tsai; Chien-Kuo Hsieh
Journal:  Nanoscale Res Lett       Date:  2016-02-29       Impact factor: 4.703

3.  The Influence of Hafnium Doping on Density of States in Zinc Oxide Thin-Film Transistors Deposited via Atomic Layer Deposition.

Authors:  Xingwei Ding; Cunping Qin; Jiantao Song; Jianhua Zhang; Xueyin Jiang; Zhilin Zhang
Journal:  Nanoscale Res Lett       Date:  2017-01-23       Impact factor: 4.703

4.  Effect of Doping on Hydrogen Evolution Reaction of Vanadium Disulfide Monolayer.

Authors:  Yuanju Qu; Hui Pan; Chi Tat Kwok; Zisheng Wang
Journal:  Nanoscale Res Lett       Date:  2015-12-10       Impact factor: 4.703

  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.