A strategy for fabricating wafer-scale platinum disulfide - ACS Applied

Subscriber access provided by UNIV OF BARCELONA

Functional Inorganic Materials and Devices

A strategy for fabricating wafer-scale platinum disulfide Hongjun Xu, Hsin-Pan Huang, HaiFeng Fei, Jiafeng Feng, Huei-Ru Fuh, Jiung Cho, Miri Choi, Yanhui Chen, Lei Zhang, Dengyun Chen, Duan Zhang, Cormac Ó Coileáin, Xiufeng Han, Ching-Ray Chang, and Han-Chun Wu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b19218 • Publication Date (Web): 07 Feb 2019 Downloaded from http://pubs.acs.org on February 7, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

A strategy for fabricating wafer-scale platinum disulfide Hongjun Xu,†, ‡ Hsin-Pan Huang, ╪ HaiFeng Fei,† Jiafeng Feng, ‖ Huei-Ru Fuh, ┴ Jiung Cho,§ Miri Choi,# Yanhui Chen,Δ Lei Zhang, □ Dengyun Chen,† Duan Zhang,† Cormac Ó Coileáin,†, × Xiufeng Han,‖ Ching-Ray Chang, ╪, ┬ Han-Chun Wu†,* †School

of Physics, Beijing Institute of Technology, Beijing, 100081, People’s Republic of China,

╪Graduate

Institute of Applied Physics, National Taiwan University, Taipei 106, Taiwan, ‖Beijing

National Laboratory of Condensed Matter Physics, Institute of Physics, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing 100190, P. R. China, ┴Department of Chemical Engineering & Materials Science, Yuan Ze University, Taoyuan City 320, Taiwan, §Western

Seoul Center, Korea Basic Science Institute, Seoul 03579, Republic of Korea,

#Chuncheon ΔInstitute

Center, Korea Basic Science Institute, Chuncheon 24341, Republic of Korea,

of Microstructure and Property of Advanced Materials, Beijing University of

Technology, Beijing, 100124, China, □School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, P. R. China, хSchool of Chemistry, AMBER and CRANN, Trinity College Dublin, Dublin 2, Ireland, ┬ Department of Physics, National Taiwan University, Taipei 106, Taiwan ABSTRACT

1 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

PtS2 is a newly developed group 10 2D layered material with high carrier mobility, wide bandgap tunability, strongly bound excitons, symmetrical metallic and magnetic edge states, and ambient stability, making it attractive in nanoelectronic, optoelectronic and spintronics fields. To the aim of application, a large-scale synthesis is necessary. For transition metal dichalcogenide (TMD) compounds, a thermally assisted conversion method has been widely used to fabricate wafer-scale thin films. However, PtS2 cannot be easily synthesized using the method, as the tetragonal PtS phase is more stable. Here, we use a specified quartz part to locally increase the vapor pressure of sulfur in a CVD furnace and successfully extend this method for the synthesis of PtS2 thin films in a scalable and controllable manner. Moreover, the PtS and PtS2 phases can be interchangeable converted through a proposed strategy. FET characterization and photocurrent measurements suggest that PtS2 is a ambipolar semiconductor with a narrow bandgap. Moreover, PtS2 also shows excellent gas sensing performance with a detection limit of ~0.4 ppb (parts per billion) for NO2. Our work presents a relative simple way of synthesizing PtS2 thin films and demonstrates their promise for high performance ultra-sensitive gas sensing, broadband optoelectronics and nanoelectronics in a scalable manner. Furthermore, the proposed strategy is applicable for making other PtX2 compounds and TMDs which are compatible with modern silicon technologies. KEYWORDS: platinum disulfide, 2D materials, ambipolar semiconductor, gas sensing, photoelectronic INTRODUCTION Platinum and its compounds have wide-ranging applications in industry and fundamental research, such as catalysts, counter electrodes in electrochemical cells, thermocouples, infrared photodetectors, and spin-charge convertors in spin-electronics. As a transition metal, its twodimensional (2D) dichalcogenide compounds (PtS2, PtSe2 and PtTe2) have recently received


Page 2 of 23

Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


considerable attention due to high carrier mobility,1-2 layer-number-dependent band gap,3-4 typeII Dirac Fermion properties5-6 and ultrastability.1, 7 These materials have demonstrated potential applications in field effect transistors (FETs)1-2 and photoelectronic devices.8-10 Furthermore, Pt dichalcogenides have effective activity for hydrogen evolution reactions.11 In particular, PtS2 is a newly developed group 10 2D layered material with relatively high charge carrier mobility, 3, 12 wide bandgap tunability,13 and good stability. The indirect bandgap of PtS2 can be tuned from 1.6 eV (monolayer) to 0.25 eV (bulk), and an indirect-to-direct gap transition can occur under strain.14 Moreover, monolayer PtS2 has strongly bound excitons15 and symmetrical metallic and magnetic edge states,16 making PtS2 attractive for the fields of nanoelectronics, optoelectronic17 and spintronics.18 So far most PtX2 compounds have been prepared by exfoliating the bulk phases. Although physical exfoliation can produce high-quality PtS2, it is not suitable for large scale applications.13 Liquid exfoliation provides a huge yield but lacks intrinsic quality and homogeneity.11 Direct sulfurizing/selenizing of transition-metal thin films can produce TMDs thin films with high throughput and quality, making it compatible with conventional planar technologies.9,

19-21

Compared with the conventional CVD process,22 this thermally assisted

conversion method shows better controllability and reproducibility.23-25 However, sulfurization of Pt can result in both PtS and PtS2 phases and the tetragonal PtS with its 3-dimensional crystal structure is more stable than its hexagonal 2D counterpart, PtS2.26 Here, we for the first time synthesized thin-film PtS2 through sulfurization of Pt in a scalable and controllable manner. A modified quartz tube was adopted to confine the sources and locally increase the vapor pressure of S. Moreover, the PtS and PtS2 phases can be reversibly converted through the proposed strategy. High performance FETs and broadband photodetectors are demonstrated. PtS2 also presents excellent gas sensing performance, with a detection limit of ~0.4 ppb for NO2. What’s more, an



Page 4 of 23

interesting reversal of resistance response to NO2 was found and a quantitative analysis was demonstrated to identify the roles of different adsorption sites in energy band of PtS2 which relate to its ambipolar-semiconductor property.

RESULTS AND DISCUSSION Figure 1a shows a schematic drawing of the thermally assisted conversion arrangement. To locally increase the vapor pressure of S, a small quartz tube with a small vent hole was used, the hole was left open on the downstream end of the tube. S powder and substrates with sputtered Pt thin films were placed in cold (~200 oC) and hot zones (400-500 oC) respectively. Figure 1c shows a HRTEM image of a typical PtS2 thin film, indicating the high quality of the grown PtS2 thin film. It exhibits a hexagonal lattice with lattice spacing of 3.09 Å, corresponding to the (100) plane of 1T PtS2 (Figure 1b).13 Raman spectroscopy was used to further confirm the PtS2 phase (Figure 1d). Three prominent Raman modes are present, i.e. Eg (~302 cm-1), A1g1 (~331 cm-1) and A1g2 (~343 cm-1), which are consistent with results for single crystal of PtS2.13,

27

Comparing the Raman peak

positions and relative intensities with layer-dependent Raman spectra of PtS213, we found the grown film is tri-layer-PtS2-like. X-ray photoelectron spectrometry (XPS) confirms the majority composition of PtS2 although a minority of PtS2-x or PtS may exist at the grain boundaries (see Fig. S1 a&b). Raman mapping and atomic force microscopy (AFM) reveal the high uniformity of PtS2 thin film with a root-means-square roughness of 1-2 nm (see Figure S1 & 2). Remarkably, the PtS and PtS2 phases can be reversibly converted. Annealing PtS2 in vacuum can change it into the PtS phase; while annealing PtS with sufficient S vapor pressure can convert it into PtS2. 𝐴𝑁 𝑖𝑛 𝑆 𝑣𝑎𝑝𝑜𝑟

𝑃𝑡𝑆

𝐴𝑁 𝑤𝑖𝑡ℎ𝑜𝑢𝑡 𝑆

𝑃𝑡𝑆2.

(1)


Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure S3 shows the Raman spectra of the same PtS sample before and after multi interchangeable conversions. Before conversion, only a relatively weak peak around 330 cm-1 is observed from the Raman spectrum.26 After conversion, three strong peaks around 302 cm-1, 331 cm-1 and 343 cm-1 show up, indicating the PtS has been converted into PtS2. To probe the electronic properties of PtS2, several FET devices were fabricated by transferring thin films onto a p-doped silicon wafer with a 300 nm thick SiO2. Details of the transfer process can be found in the supporting information (Figure S4). A clear gating effect was observed at room temperature with n-type semiconducting behavior for different thickness (Figure 2 & S5). From the transfer curves (Ids-Vg), the mobility of the PtS2 devices can be calculated using the relationship μe = (dIds/dVg)(Ld/WVdsε0εr), where ε0 and εr are the permittivity of vacuum and SiO2, respectively. The calculated mobility at room temperature are around 10-3 to 4×10-2 cm2/V.S, which is comparable to that of naoncrystalline-PtSe2 devices,22 but it is 1-2 orders of magnitude lower than the value reported for single-crystal monolayer PtS2.13 The reduced mobility is mainly because of grain boundary scattering, which could be improved by increasing the size of the single-crystal domains of PtS2 or reducing the gap between the electrodes down to the dimensions of a single domain.28 The higher on-off ratio in our PtS2, in comparison with PtSe2, is related to its wider band gap.29 The lower on/off ratio but higher charge carrier mobility for thicker PtS2 films compared with thinner films is consistent with the results from single-crystal PtS2 devices (see Fig. S5).13 At lower temperature, the transfer curve further shows a clear ambipolar-semiconductor behavior. Figure 2c shows the Ids-Vds curves under various gate voltages. Linear behavior was observed, indicating Ohmic contact between electrodes and PtS2. Temperature dependent resistivity measurement further confirms the semiconductor-behavior of the PtS2 (see Figure 2d), from 30 K to 300 K the R-T curve can be nicely fitted by the 2D variable range hopping model, lnR ~ T-1/3.



Figure 3 shows the representative optoelectronic properties of the same PtS2 thin films. Clear photo responses are observed for illumination by lasers of different wavelengths, from the ultra-violet to infrared range (1550 nm), indicating PtS2 can be used as a wide spectrum photodetector. It also suggests that trilayer PtS2 has a bandgap less than 0.8 eV. Density functional theory simulations suggest that monolayer PtS2 has a bandgap of 1.74 eV. The bandgap decreases significantly to 0.56 eV for bilayer PtS2 and 0.31 eV for trilayer PtS2 (Figure S6). Interestingly, the response time depends on the laser wavelength, with increasing the wavelength the response time becomes longer. The typical response times for the 532 nm, 635nm, 785nm and 1064nm lasers are 4.6 s, 12.7 s, 15.6 s and 23 s respectively. Figure 3c shows photocurrent (IPC) as a function of laser power. The photocurrent increases almost linearly with laser power, indicating the mechanism of photocurrent generation is due to photoconductive effects. Note, the device shows an ultrasensitive response under illumination by the 635 nm laser. A clear photocurrent response can be detected even for low powers which cannot be detected by our commercial laser power meter as shown in Figure 3d. The performance of these optoelectronics devices can be further improved by reducing the spacing between the electrodes. When the gap between electrodes is reduced from ~1 mm to 40 μm, the on/off ratio doubled and photocurrent increased by an order of magnitude (see Figure S7). These results demonstrate that PtS2 is a promising candidate for the wide spectrum photo detection (from UV to infrared regions) with a good linear response proportional to the power for different light sources. PtS2 is also a good candidate for high performance gas sensors. Figure 4a & 4b shows the dynamic sensing response of a PtS2 thin film exposed to NO2 and NH3 respectively at room temperature. The sensitivity S is defined as S = ΔR/R0 = (Rg-R0)/R0*100%, where Rg and R0 are the resistance under the sensing conditions with and without analyte. A clear signal with a sensitivity of 0.15%


Page 6 of 23

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


is observed for 2 ppb NO2. From Figure 4a, the detection limit (DL) can be calculated by using DL ~ 3*rmsnoise/slope, where rmsnoise ~ 0.0046 is the root-mean-square (rms) of the noise and slope ~ 0.0355 ppb-1 is the slope of the sensitivity versus concentration at the initial linear region.30 The calculated DLs for NO2 and NH3 in darkness are ~ 0.4 ppb and 945 ppb respectively. Similar work was performed on CO and H2S obtaining the DLs of ~13 and 51 ppb (Fig. S9). The performance of detecting noxious gases can be improved by photo illumination (Fig. 4c & Ss9) excepting the case of detecting much diluted NO2. The DLs of 475, 5.5 and 26 ppb were achieved for NH3, CO and H2S respectively. To the best of our knowledge, such a DL for NO2 is among the lowest values reported in gas-sensing devices based on TMDs and other 2D materials (see Table 1).31 What’s more, the obvious difference between the dynamic responses to NO2 and to other noxious gases in the concentration of parts per million (ppm) region makes for selectively detecting of NO2 (Fig. 4c). In contrast to the behaviors of other gases above the sensitivity of low-concentration NO2 is decreased under photo illumination (DL ~1 ppb), which might imply a different adsorption process for NO2. Figure 4d shows the dynamic sensing response of PtS2 to higher concentrations of NO2. Interestingly, there is an inflection point in the gas sensing response, the sensitivity changes from positive to negative when the concentration of NO2 increases from the ppb to ppm range. A clear reversal is observed to happen at ~150 ppb and ~ 60 ppb for measurements in the dark and under violet light (405 nm) respectively. Resistance is first seen to increase and then sharply decrease. Many samples were investigated and such a reversal process was observed for all PtS2 samples when the concentration of NO2 was increased beyond ultra-low levels to a relatively high range, and violet light was noted to decrease the required concentration for this transition. It is also worth



Page 8 of 23

noting that in the higher NO2 concentration region the device’s sensitivity is enhanced by violet light, which is consistent with its behavior for the others gases and the results in literature.30, 32-33 According to the Langmuir isotherm model, the gas-sensing response can be roughly described by an exponential function, G(t) ∝ G∞(1 - e-t/τ), where G(t) is the conductance at time t, G∞ is the final conductance under the equilibrium coverage of an analyte and τ is the transient response time.9 The conventional Langmuir isotherm model considers only one adsorption site or process, which is clearly not suitable for our case, (e.g. 90-176 ppb in Figure 5a). To explain the reversal process, multiple adsorptions sites or processes must be taken into account34 and the corresponding gas-sensing response can be represented in I-t or dI/dt-t as: 𝑖

𝐼(𝑡) ∝ ∑𝑖∆𝐼𝑖∞(1 ― 𝑒 ―𝑡/𝜏 ) or 𝑑𝐼/𝑑𝑡 ∝ ∑𝑖(∆𝐼𝑖∞/𝜏𝑖)𝑒 ―𝑡/𝜏

𝑖

.

(2)

These parameters, ∆𝐼𝑖∞ and 𝜏𝑖, stand for the equilibrium current change from adsorption of NO2 molecules onto a site i and the corresponding response time respectively. Figure 5b shows the gas sensing response for 90 ppb NO2. A two-component electrical current response can be clearly identified from I(t) plot, indicating at least two distinct adsorption sites with negative ∆𝐼1∞― and positive ∆𝐼2∞+ . Multiple exponential fitting gives (see the dashed lines in Figure 5b): ∆𝐼1∞― ~ -1.85 nA, τ1 ~ 83 s; ∆𝐼2∞+ ~ +2.94 nA, τ2 ~ 213 s. Figure 5c further shows the gas sensing response for the introduction of a higher concentration of NO2 (~606 ppb). A three-component response curve is first identified in the dI/dt plot (see the blue data in Figure 5c). Best fitting with three exponential functions for both dI/dt and I(t) plots gives ∆𝐼1∞+ ~ +6.25 nA, τ1 ~ 14 s, ∆𝐼2∞― ~ -18.8 nA, τ2 ~ 29 s and ∆𝐼3∞+ ~ +26.77 nA, τ3 ~ 85 s. It is worth mentioning that although the negative response is not clearly evident in the I(t) curve, its action simultaneously affects the total change of current together with the positive responses. Similar analysis processes were performed for others NO2


Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


concentrations and for the case of sensing NO2 with a violet light (Figure S10). Figure 5d-f summarize the fitting results. Evidently, three different adsorption sites are identified by this method, which is not easily recognized by other methods.34 ∆𝐼∞― dominates the current response at relatively low concentrations of NO2. As the NO2 becomes more concentrated, both ∆𝐼∞+ and ∆𝐼∞― increased but the latter exceeds the former after a certain medium concentration and dominates the current change. Moreover, a new source of positive current change starts to manifest itself at 300600 ppb NO2 with the fastest response (Figure 5d). As PtS2 is a n-type semiconductor with a narrow bandgap at room temperature and ambipolar semiconductor at low temperature, the above interesting response reversal can be understood from different perspectives. One could expect the PtS2 itself changed from an n-type to p-type semiconductor caused by the increasing exposure to NO235-37 or a change in the doping effects from NO2, i.e. p-type doping (NO2 accept electron) for dilute NO2 and n-type doping (NO2 donate electron) at relatively concentrated NO2.38 For the first case, the majority charge carrier in PtS2 should then be changed from electrons to holes, imitating a field effect transistor such that the continuous charge transfer between the NO2 molecules and the PtS2 thin film greatly shifts the Fermi level of the PtS2. For the second case, one needs to assume a change of electron affinity in PtS2, which is related to its Fermi level as well. A key point in our previous analysis is that in each NO2-sensing cycle although the current changes (negative/positive) are observed in succession, the component factors are occurring in parallel and are responsible for to the total current response of PtS2 device. This excludes the above overall change in carrier type and electron affinity scenarios because the Fermi level can be changed only in one direction and it cannot explain the sequence of positive-negative-positive change under more concentrated NO2. On the other hand, it seems the small amount of current change during the response-reversing process is unlikely to



change the majority carrier in PtS2. Taking the FET in Fig. 2a for example, even a current change ∆I/I0 ~ 7.5% (between Vg = 0 & -50V) is not enough to change the n-type PtS2 into p-type, while the current change were mere ~ 0.2% when the response to NO2 reversing (see Fig. 4d & 5a) . An alternative explanation is considering two distinct types of absorptions sites in PtS2: the lower absorption-energy barrier one with a lower density (of states) results in higher resistance and the higher absorption-energy barrier sites with a higher density (of states) leads to lower resistance. In this case, the shift of the Fermi level in PtS2 is not necessary large when exposed to NO2. As we pointed out before, at least three adsorption sites were identified, one negative-response site and two positive-response sites (in terms of conductance). When a NO2 molecule, typically an electron acceptor, is adsorbed on PtS2 electrons will normally be transferred from the PtS2 to the NO2, thus forming NO2. The FET test on our PtS2 thin films (Figure 2b) implies two types of charge carriers are present, although the n-type conceal the p-type carriers at room temperature. Therefore, the electrical current response depends on the adsorption sites in the energy band: when adsorbing at the sites near the conduction band, it reduces the electron density and thus the current; while adsorbing at the sites near the valence band, it increases the hole density and increase the current.37 By shedding light on PtS2, photons excite electrons from valence band to conduction band and generate electron-hole pairs in the PtS2. The photo-generated electrons can be captured by the NO2 molecules at the adsorption sites before recombination with photo-generated holes occurs. Note that the excited electrons can relax onto donor/acceptor sites with deep energy levels, which means more active sites are available even for the sites with a higher adsorption energy barrier. A schematic illustration is shown in Figure 6 a-b based on our explanations. With the assistance of light more NO2 molecules tend to adsorb at the sites near the valence band which accounts for the current reversal. This in turn reflects the extra energy required for such absorption, while


Page 10 of 23

Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


adsorption occurring at lower NO2 concentrations indicates the preferential lower adsorption energy barrier (e.g. on the vacancy sites of sulfur). Note that apart from excitation by light, a higher concentration of analyte may also lower the adsorption energy barrier, as would thermodynamic driving forces for the otherwise higher adsorption energy barrier sites.39 As a typical 2D semiconducting TMDs, MoS2 was reported that its point defects (e.g. sulfur vacancies and antisite defects) can induce a high density of gap states both near the conduction band and valance band edges.40 Similarly, we expect many of mid-gaps states near both conduction band and valance band of PtS2, which may provide distinct adsorption sites for NO2 molecules. Moreover, for MoS2 it was reported that chemisorption of oxygen on surface points defects at lower partial oxygen pressures (< 3×10-7 mbar) induces an n-doping effect38 while physisorption of oxygen at higher partial oxygen pressures produce a p-doping effect.41-42 The underlying picture in PtS2 may be similar and further investigation is required. CONCLUSIONS In summary, we have successfully synthesized PtS2 thin films by directly sulfurizing Pt thin films in a modified quartz tube. Our strategy is promising for making devices in an up-scalable manner at relatively low temperature, applicable for other TMDs and compatible with modern planar semiconductor industry as well as the flexible electronics. Our PtS2 devices present ambipolar semiconductor behavior and wide-range spectrum photo detection has been achieved. Besides, a superior gas-sensing performance for NO2 was realized with an interesting resistance reversal. We have demonstrated a method to identify the contributions from different adsorption sites on PtS2. Experimental Section Sample preparation and characterization: PtS2 thin films were synthesized in a CVD furnace. S source and sputtered Pt films were loaded into the modified quartz-tube as shown in Fig. 1a. Before



heating, the system was evacuated (base pressure ≤ 3 mtorr) and then protected by flowing N2/H2 (9:1) with a rate of 100 sccm. The pressure of system during sulfurization was in the range of 50100 Torr and a temperature 400-500 oC. The typical sulfurization time is 1h. Raman analysis was carried out using a Bruker Raman microscope with an excitation wavelength of 532 nm. STEM samples were prepared by transferring the as-grown films onto TEM grids. The STEM studies were performed in a JEOL JEM-2100 F with a probe size under 0.5 nm, 200 kV. Device fabrication and measurement: Two types of devices were made. For gas sensing and photoresponse, the devices were made directly on the Al2O3/PtS2 wafers by evaporating Ti/Au (10/25 nm) onto the thin films through a shadow mask. The distance between electrodes was between ~40 μm to ~1 mm. For making FET, the PtS2 thin films were first transferred onto a high-doped SiO2/Si wafer.43-44 Subsequently, Ti/Au electrodes were deposited using a shadow mask. The PtS2 gas sensor was tested in a custom-made chamber with remote-controllable mass-flow controllers. All of the sensors were measured at atmospheric pressure and room temperature. NO2 (100 ppb or 1.3 ppm), CO (1 ppm), H2S (4.6 ppm) and NH3 (193 ppm) diluted with pure nitrogen were used as these analyte respectively. The change in resistance upon different gas exposures was monitored by Keithley 2400A source-meters. DFT simulation: Electronic band structures of monolayer and bilayer PtS2 were calculated using the density functional theory (DFT) implemented in the VASP package considering the effect of spin polarization and van der Waals force.45 Projector augmented wave (PAW) method with the Perdew-Burke-Ernzerhof (PBE) generalized gradient approximation (GGA) were used. The energy convergence criteria is to be 10-5 eV per unit cell and a 12x12x1 k-grid mesh is used to sample the Brillouin zone.


Page 12 of 23

Page 13 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FIGURES

Figure 1. (a) schematic layout for sulfurization of Pt thin films in a CVD furnace, where a small ‘isolated’ quartz tube is sited in the middle bottom. (b) Schematic drawing of atomic structure of PtS2 in 3D, top- and side-views, where Pt and S are marked by the deep blue and yellow and brown symbols. (c) HRTEM image of PtS2 thin film. It is zoomed in as shown in the inset, where the atomic arrangements of Pt and S are marked by the overlaid deep blue and yellow symbols. (d) Raman spectrum of typical PtS2 thin film decomposed by three Lorentz peaks. (e) PtS2 thin film on Si/SiO2 after transferring from its original wafer.



(b)

(a)

1.65μ

300 K

250n

Isd (A)

Isd (A)

100 K

300n

1.60μ

1.55μ

200n

1.50μ 150n -50

(c)

-25

0

25

50

-80

-60

-40

-20

125

64

Gate voltage (V)

(d)

300n 200n

100 K

Vg = -50 V Vg = +50 V Vg = 0 V

1E11

300

0

Vg (V) 37

20

40

60

23.3 15.6

80 11.0

T (K)

1E10

R ()

100n

Isd (A)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

0

-100n

1E8 1E7

-200n -300n-10

1E9

-5

0

Vsd (V)

5

10

1000000

0.15

0.20

0.25

0.30

T

-1/3

0.35

0.40

0.45

Figure 2. FET properties of PtS2 thin films. (a) and (b) Presentative transfer curves at room temperature and 100 K respectively Inset of (a) show schematic of the PtS2 FET device. (c) I-V curves under different gate voltages. (d) Resistance as a function of temperature was fitted by a 2D VRH model (at ~ 30-300 K).


Page 15 of 23

(a)

(c)10.5

532 nm~ 11 mW 635 nm~ 23.6 mW

Ipc (nA)

Ipc (nA)

405 nm

8.5

120 80 785 nm ~ 34 mW 40 0 0

60

120

60 40

42 W

20 0 2

4

400

68 W

0.5

0

300

1.0

Ipc (nA)

80

200

Time (s) 135 W

532 nm 635 nm 785 nm 1064 nm

100

1550 nm (Ipc*4) 100

(d)

140 120

4.5

0.5 0

180

Time (s)

(b)

6.5

2.5

1064 nm ~ 23 mW

Ipc (nA)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


6

8

power density (mW/mm2)

10

0

20 W 60

120

180

240

Time (s)

Figure 3. Photo response of PtS2 thin films. (a), Photocurrent under different laser illumination. (b), photocurrent under diodes with light wavelengths of 405 nm and 1550nm. (c) Power density dependent photocurrent for different lasers. (d) Photocurrent under 635nm laser with relatively low power.



(b) 50

0.3

40 30

0.2

20

0.1

1.5

30k

dark 405 nm 1.0

20k

R/R0 (%)

0.4

NH3 (ppb)

R/R0 (%)

(a)

NO2 (ppb)

0.5

10k

10 0.0

(c)

150

Time (min) 2 0

R/R (%)

100

-2 -4 -6

100

Time (min)

(d) 33M

1.2 ppm NO2 1.2 ppm H2S

50

1 ppm CO

3.9 ppm NH3

dark light

150

0 600

32M

R ()

50

0.0

0 200

400

NO2 (ppb)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

31M

200 30M

-8 -10

29M 50

100

Time (min)

150

0 200

Figure 4. Gas sensing performance of PtS2 devices, where ΔR/R0 is sensitivity and all devices generated from 1 nm Pt, where the grey shadow regions indicate the time exposed to analytes and their corresponding concentrations (see right axes). Sensing NO2 (a) and NH3 (b) with PtS2 thin films under dark (black) and 405 nm light (violet) luminescence. (c) Representative gas-sensing sensitivities of NO2, H2S, CO and NH3 with and without light illumination. (d) The reversal of the resistance response of sensing NO2 with its concentration increased from ~60 ppb to 460 ppb. The insets zoom in the reverse response processes.


(a)

(b)

320n

0

90 ppb

50p

310n

-50p 8000

50n

10000

-200p 0

5750

30n

Time (s)

-6p 6500

6250

405 405dark+ dark-

600

10n 0

400

12000

12100

Time (s)

12200

12300

dark 405 nm

30n

20n

10n

200

-10n

11900

(f) +

 (s)

20n

6000

(e)

dark+_slow dark+_fast 405+_slow 405-_fast

40n

0 310n

12000

Time (s)

40p

315n

-3p

-400p

I (A)

6000

(d)

3p

I (A)

0

80p

606 ppb

dI/dt (A/s)

315n

I (A)

6p

dI/dt (A/s)

I (A)

320n

305n

(c)

100p

325n

dI/dt (A/s)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


I (A)

Page 17 of 23

-

-20n 0

dark 405100 200

0

0 300

400

500

600

0

NO2 (ppb)

100

200

300

400

500

600

NO2 (ppb)

0

100

200

300

400

500

600

NO2 (ppb)

Figure 5. (a) I-t and dI/dt-t plots for dynamic response to NO2 in dark from low (90 ppb) to high concentration (660 ppb). Two (b) and three (c) exponential functions fitting for I-t and dI/dt-t plots under 90 ppb and 660 ppb NO2 respectively which are from the dashed oval regions in (a). Note that a linear background I(t) was removed in (b) and the initial times when absolute values of dI/dt increased were skipped in the fitting. The contributions to current change (d) and response time (e) for positive and negative components respectively under dark (black points) and 405 nm light (violet points) are plotted, where the solid lines are guiding lines. (f) Total current changes after sum of ∆𝐼𝑖∞, which may be different with the observed values in experiment as they are the equilibrium values after enough exposure.



Page 18 of 23

Figure 6. (a) The adsorption sites in energy band for NO2 on PtS2 when NO2 is diluted. (b) adsorption sites for NO2 when light is on and NO2 is more concentrated.

Table 1. A comparison of PtS2-based gas sensor with others based on 2D materials 2D materials

Condition

Sensitivity

Reponses/recovery time

Detection limit

Reference

GR thin film

RT

65%@2.5ppm

1400/1500 s

0.387 ppb

46

Ozone-treated graphene

RT

1% @1ppm

~80 s

1.3 ppb

47

Trilayer graphene

73.4 oC

12.5%

89/579s

6.87 ppb

48

MoS2

RT

-80%@0.4ppm

5-9 min

A strategy for fabricating wafer-scale platinum disulfide - ACS Applied

Recommend Documents