Black Phosphorus Flexible Thin Film Transistors at GighertzFrequenciesWeinan Zhu, Saungeun Park, Maruthi N. Yogeesh, Kyle M. McNicholas, Seth R. Bank,and Deji Akinwande*

Microelectronics Research Center, Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin,Texas 78758, United States

*S Supporting Information

ABSTRACT: Black phosphorus (BP) has attracted rapidly growingattention for high speed and low power nanoelectronics owing to itscompelling combination of tunable bandgap (0.3 to 2 eV) and high carriermobility (up to ∼1000 cm2/V·s) at room temperature. In this work, wereport the first radio frequency (RF) flexible top-gated (TG) BP thin-filmtransistors on highly bendable polyimide substrate for GHz nanoelectronicapplications. Enhanced p-type charge transport with low-field mobility ∼233cm2/V·s and current density of ∼100 μA/μm at VDS = −2 V were obtainedfrom flexible BP transistor at a channel length L = 0.5 μm. Importantly, withoptimized dielectric coating for air-stability during microfabrication, flexibleBP RF transistors afforded intrinsic maximum oscillation frequency fMAX ∼ 14.5 GHz and unity current gain cutoff frequency f T∼ 17.5 GHz at a channel length of 0.5 μm. Notably, the experimental f T achieved here is at least 45% higher than prior results onrigid substrate, which is attributed to the improved air-stability of fabricated BP devices. In addition, the high-frequencyperformance was investigated through mechanical bending test up to ∼1.5% tensile strain, which is ultimately limited by theinorganic dielectric film rather than the 2D material. Comparison of BP RF devices to other 2D semiconductors clearly indicatesthat BP offers the highest saturation velocity, an important metric for high-speed and RF flexible nanosystems.

KEYWORDS: Black phosphorus, phosphorene, flexible nanoelectronics, two-dimensional semiconductors, thin-film transistors,radio frequency

Two dimensional (2D) atomic layered semiconductorshave been widely studied as promising candidates for

flexible nanosystems with functionalities ranging from sensingto wireless communication owing to their compelling electrical,optical, and mechanical properties, including ultimate thicknessscalability down to one atomic layer, ideal electrostatic control,and superior mechanical flexibility with strain limit >20%.1−10

As the most studied 2D layered nanomaterial, graphene withhigh carrier mobilities (∼10 000 cm2/(V·s) at room temper-ature) is most promising for ultrahigh-frequency analognanosystems, such as terahertz detectors.10,11 The zero bandgapsemimetal nature of graphene, however, results in low field-effect modulation and high OFF state current limiting itsapplications for logic and low power nanoelectronics. On theother hand, semiconducting transition metal dichalcoginides(TMDs) offer sizable band gaps with large ON/OFF currentratio (∼107) that is favorable for low power electronics anddigital circuits. However, their relatively low room temperaturecarrier mobilities (10−100 cm2/(V·s)) limit their prospects forhigh-frequency applications.12−15 With the emerging needs forboth high speed and low energy consumption in flexiblenanosystems, few-layer black phosphorus (BP), phosphorene asits monolayer counterpart, has been rediscovered as apromising candidate to combine the fast transport of grapheneand the high band gap of TMDs such as high room-

temperature carrier mobility ∼1000 cm2/(V·s)16 and highlysensitive thickness-dependent direct bandgap from 0.3 eV(bulk) to 2 eV (monolayer).17−20

In this work, we present the first realization of flexible top-gated (TG) BP transistors with intrinsic current gain frequency,f T ∼ 17.5 GHz and power gain frequency, fMAX = 14.5 GHzachieved at a channel length L= 0.5 μm. Another device withthe same channel length that afforded intrinsic f T ∼ 20 GHzprovides an indication of the reproducibility of our flexible BPtransistors. Despite the many challenges experienced, such aslack of precise control of size, thickness and orientation of BPfilms, and the difficulties of device integration on nonatomicallysmooth low-thermal budget soft substrates, the compellingperformance obtained from the first investigation of BP-basedflexible RF transistors demonstrates the potential for thisemerging 2D layered semiconductor for flexible smart nano-systems. Bending test was conducted to investigate the high-frequency performance of the flexible TG BP transistors withextrinsic f T measured under tensile strain up to 1.5%, invariablylimited by the relatively stiff high-κ atomic layer deposited(ALD) dielectrics.12

Received: November 22, 2015Revised: March 14, 2016Published: March 15, 2016

Letter

pubs.acs.org/NanoLett

© 2016 American Chemical Society 2301 DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

Results. In this work, mechanically exfoliated blackphosphorus flakes with typical thickness of 10−25 nm wereselected as the semiconducting channels. Optical microscopyimage of a typical BP sample is shown in Figure 1a with

uniform thickness ∼13 nm verified by atomic force microscopy(AFM). The crystalline nature of BP was verified by Ramanspectroscopy with incident laser (λ = 532 nm) polarized alongboth armchair (AC) and zigzag (ZZ) directions, as shown inFigure 1b, where the ratio among characteristic peaks of Ag

1,B2g, and Ag

2 clearly indicates the in-plane orientations.18,21 Forexfoliated BP flakes with thickness more than 10 nm, thetheoretical bandgap is ∼0.3 eV,19 which is confirmed in Figure1c by Fourier transform infrared (FTIR) spectroscopy that isrepresented as a Tauc plot ((αhν)2 = hν − Eg),

22,23 an

established method for determining the optical band gap. hν isthe photon energy, Eg is the optical bandgap, and α is theabsorption coefficient. As shown in Figure 1d, highly bendableKapton polyimide (PI) sheets ∼125 μm thick were adopted asthe substrate for the flexible BP transistors with double sidedsolution-based PI thin film coating and curing process appliedto smoothen the substrate to about 1 nm surface rough-ness.10,12,24 A 25 nm Al2O3 dielectric thin film was thendeposited using atomic layer deposition (ALD) at 250 °C ontop of the PI substrate to provide (i) a high-k environment forenhancing carrier mobility by suppressing charge impurityscattering,12,25 and (ii) better adhesion between the BP channelmaterial and the substrate.To fabricate the flexible TG BP transistors as illustrated in

Figure 2a, the source and drain metal contacts were defined byelectron beam lithography (EBL) followed by electron beamevaporation of Ti/Au = 1.5 nm/50 nm with Au selected toenhance the p-type transport due to its relatively large workfunction.26 Subsequently, 25 nm of Al2O3 was deposited againat 250 °C to serve as the high-κ top-gate dielectric. TGelectrodes were formed by EBL followed by e-beamevaporation of Ti/Au = 1.5 nm/50 nm. Here, a dual-fingerTG structure with an underlap configuration was adopted toenhance the current transport as well as to reduce parasiticcapacitance between TG and S/D electrodes.10,12,27 In order topreserve the pristine electrical properties of the flexible TG BPtransistors, ambient exposure time of hygroscopic BPchannel28,29 was minimized during the aforementioned nano-fabrication process by (i) immediate poly(methyl methacrylate)(PMMA) coating after the exfoliation of the BP flakes onto theflexible PI substrate, where PMMA layer functions as both aneffective encapsulation layer for BP characterization and the e-beam resist for subsequent device fabrication, and (ii)immediate ALD deposition of 25 nm Al2O3 after S/D metalliftoff process, where the Al2O3 layer functions as both TGdielectric and the encapsulation layer.26,28,29

Enhanced p-type current transport was obtained from theDC characteristics of the flexible TG BP transistors as shown inFigure 2b,c. The transport measurements were conducted usinga Cascade probe station and an Agilent 4156 semiconductorparameter analyzer. Low-field hole mobility μp ∼ 233 cm2/V·swas extracted using the Y-function method30 from the transfercharacteristics, as shown in Figure 2b, which is comparable withprior works of bottom-gated (BG) flexible BP transistors26 andTG BP transistors on rigid Si substrates.27 The contact

Figure 1. (a) Optical image of multilayer BP flake with uniformthickness of 13 nm as determined by AFM shown as the red line. (b)Raman spectrum of a typical BP flake with incident laser polarizationwithin a close affinity to AC and ZZ directions, respectively. A 532 nmgreen laser was used to obtain the Raman spectrum. (c) Arepresentative Tauc plot for estimating the optical bandgap, whichconfirms a gap of ∼0.3 eV in the bulk limit. Insert is the absorptioncoefficient of BP derived from Fourier transform infrared (FTIR)spectrum. (d) Picture of fabricated flexible TG BP devices on highlybendable PI substrate.

Figure 2. Flexible top-gate BP transistor structure and DC performance at 300 K. (a) Illustration of flexible TG BP transistor. Dual-finger TGstructure was adopted with Ti/Au = 1.5 nm/50 nm as both the source/drain and gate electrode stack. Twenty-five nanometer Al2O3 was employedas the TG dielectric as well as the protection layer for air stability. (b) Transfer characteristics of fabricated device featuring high-mobility p-typecharge transport. VDS = −0.1 V. The extracted low-field hole mobility ∼233 cm2/V·s. (c) Output characteristics of the same device showing softcurrent saturation. The W/L = 10.8 μm/0.5 μm, and flake thickness is 13 nm.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

2302

resistance, Rc ∼ 4.5−6.7 kΩ·μm was also extracted simulta-neously via Y-function method,30 which is ∼1/3 of the totalchannel resistance Rtotal ∼ 16−17 kΩ·μm, and as such it is notthe dominant factor in charge transport. The currentmodulation is within 102 and agrees well with our prior studiesof BP transistors fabricated from flakes with thickness around13 to 15 nm.26,28 The relatively low current modulation isattributed to fixed surface charge induced by the ALD top gatedielectric deposition process31 and the small band gap (∼0.3eV) of the relatively thick 13 nm BP flake.32 The trans-conductance (gm) of the same device under drain bias of VDS =−0.1 V is shown in the Supporting Information Figure S1 withpeak value gm ∼ 56.7 μS obtained at VGS = −0.72 V. The highmobility value and transconductance obtained from this firstrealization of flexible top-gate BP transistors indicate strongpotential for higher frequency applications beyond TMDs.The output characteristics of the same flexible TG BP

transistor are shown in Figure 2c. The highest current densityobtained from the output characteristics is ∼100 μA/μm underbias conditions of VDS = −2 V and VGS = −2.3 V. The outputcurrent saturation, however, was not as strong as our priorstudy of flexible BG BP transistors,26 which we attributeprimarily to contact resistance Rc resulting from the ungatedaccess region in this underlap TG configuration.10,27,33 This hasbeen previously shown to have the effect of obscuring currentsaturation in 2D devices.34,35 Therefore, with optimized self-aligned TG structure the access region will be reduced andenhanced current saturation can be expected.The high-frequency performance of flexible TG BP

transistors was characterized using standard S-parametertechnique with Cascade electrical probe station and an AgilentVector Network Analyzer (VNA-E8361C) up to 30 GHz. GSGprobing pads were adopted in the design of flexible TG BPtransistors to ensure high-frequency signal fidelity. Gate anddrain were selected as the pair of transmission electrodes,whereas the source functioned as the ground electrode. Tocalibrate-out the response of the GSG radio frequency probesand cables the standard processes of short, open, load, andthrough were followed strictly before moving to the flexible BPtransistors.Unity current gain (|h21|) cutoff frequency, f T, is one of the

most important figure-of-merit to evaluate the high-frequencyprospects of transistor devices.10,27,36 In this work, flexible TGBP transistors with channel length of L = 1, 0.5, 0.25 μm wereinvestigated with the DC characteristics for typical devices of L= 1 μm and L = 0.25 μm shown in the Supporting InformationFigures S2 and S3, respectively. Here in this work, the physicalgated length was taken to be the electrical channel length,which is reasonably justified because Schottky barrier widths atthe contacts, typically on the order of nanometers, are muchsmaller than our fabricated device dimensions.37 With muchshorter Schottky barrier width than the physical (gated)channel length expected in our underlap flexible BP RF devicestructure, the transistor performance is governed by classicaldrift-diffusive transport and is not a Schottky-barrier FET.38

Analysis of the Schottky barrier width of the flexible BPtransistor is presented in the Supporting Information. Strongcurrent saturation is observed in the output characteristics ofthe device with L = 1 μm. The statistics of extrinsic f Tmeasured for the flexible TG BP transistors with all threechannel lengths are shown in Figure 3a. Here, we attribute theobserved extrinsic f T variations within devices of the samechannel length to be due to both extrinsic factors, such as

contact resistance and parasitic capacitances, and intrinsicfactors such as thickness variations17,18,20 and the strong in-plane anisotropic electrical transport of BP, where thesaturation velocity vsat is theoretically projected to vary by afactor of 2 between the ZZ and AC crystallographicorientations.39 The latter is a unique property of BP andhence it is worthwhile to further elaborate on its impact on f T.Because f T ∝ vsat for RF devices biased under high lateralelectrical field condition, the orientation dependent intrinsic f Tcan be expressed as f T = vsat,AC(1 + cos2 θ)/4πL, where vsat,AC isthe effective saturation velocity along the AC direction, and θ isthe in-plane angle between the transport direction and the ACdirection. As such, the orientation dependence is expected toresult in a two-fold intrinsic f T device-to-device variation inrandomly oriented BP RF devices. Further research on devicestatistics will be most beneficial in elucidating the variationimposed by this unique in-plane anisotropy present in BP andminimizing both the extrinsic and intrinsic sources of devicevariability. Additional high-frequency results of device withchannel length of L = 0.25 μm are available in the SupportingInformation Figure S4.In addition to the high-frequency performance, the

mechanically robustness of the flexible BP transistors wasinvestigated by ex situ f T measurements under tensile strain (ε)up to 1.5%. The tensile strain is computed by ε = t/(2r), wheret is the overall thickness of PI substrate, and r is the bendingradius.10,26 For our flexible BP devices, t = 150 μm and thebending radius was reduced from 9 mm down to 5 mm, whichcorresponds to tensile strain up to 1.5%. A typical flexible BPtransistor with channel length L = 0.5 μm was chosen for thisstudy. As is demonstrated in Figure 3b, the extrinsic f Tnormalized to the nonstrained value was investigated as tensilestrain increased up to 1.5%. Strong mechanical robustness wasverified with slight performance degradation less than 7% withε applied up to 1.2%, and the trend is consistent with ourprevious study on the strain effect on carrier mobility and ONstate current of flexible bottom gated BP transistors.26 Withstrain increased to 1.5%, much severe performance degradationwas observed mainly due to the fracture toughness limitation of

Figure 3. High-frequency performance of flexible top-gate BP RFtransistors. (a) Statistics of extrinsic cutoff frequency of flexible BP RFtransistors with channel lengths of 0.25, 0.5, and 1 μm. The sources ofdevice-to-device variation include extrinsic contributions (e.g., contactresistances and parasitic capacitances) and intrinsic contributions suchas the anisotropic orientation dependence that is unique to BP.Further research is warranted to minimize device to device variationsin emerging 2D materials. (b) Analysis of mechanical robustness offlexible TG BP transistor with normalized extrinsic f T measured underuniaxial tensile strain up to 1.5%. All RF measurement was done ex situat 300 K. Blue dash line is a visual guide.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

2303

high-κ dielectric.12 This result agrees well with our recent studyon BG flexible BP transistors, where 25 nm Al2O3 was adoptedfor both bottom gate dielectric and BP channel encapsulationlayer.26 Similarly in this work, ALD Al2O3 was adopted for TGdielectric, with critical strain that is ∼2%.10,12,26 In order tofurther enhance the mechanical robustness of our flexible BPtransistors, optimization of device structure by substitutinghigh-κ dielectric with h-BN1 or nanoscale polymer dielectric40

are potential solutions.To further evaluate the potential of high-frequency flexible

TG BP transistors, the intrinsic high-frequency performances ofthe BP active region were determined through standard openand short de-embedding process.10,27 It is worth noting that thechannel semiconductor was absent in the open and shortstructure. As a result, the metal−BP interface and ungatedaccess regions that comprise the contact resistance are not de-embedded, hence, the extracted intrinsic high-frequencymetrics are a conservative or lower estimate of the BPsemiconductor performance on flexible substrate.36 We presentthe high-frequency performance obtained from a typical flexibleTG BP transistor with channel length of L = 0.5 μm, the DCperformance of which is shown in Figure 2b,c. To achievemaximum carrier transport, VDS = −1.8 V was applied to inducehigh lateral electrical field. The flexible TG BP transistor wasbiased near its peak transconductance (gm,max) point where VGS= −3.7 V. As presented in Figure 4a, the extrinsic f T value was

∼7 GHz. After de-embedding the effect of parasiticcapacitances and resistances from the device structure, theintrinsic f T value was extracted ∼17.5 GHz. As shown in Table1, this intrinsic f T obtained from the flexible TG BP transistorsis 45% higher than the prior works by Wang et al.,27 althoughthe prior work was on rigid substrate. To exclude channellength dependence for comparison and circuit design purposes,a channel length normalized metric f TL was selected to evaluatethe high-frequency prospects as listed in Table 1. For RFdevices operating in high lateral field region, f T is limited bysaturation velocity, which is directly proportional to f TL, f TL =vsat/2π. In this work, vsat is defined as the effective saturationvelocity due to the nonuniform lateral field distributed alongthe channel.41,42 We consider the channel length as the physicalgated length for the estimation of the effective saturationvelocity. For the best flexible TG BP device reported in thiswork, an intrinsic f TL= 8.75 GHz·μm was obtained, which ismore than two times higher than that reported for BP RFtransistor on a Si substrate.27 A similar improvement wasobtained for the vsat ∼ 6 × 106 cm/s. We attribute thisimprovement to the optimized fabrication process includingminimized ambient exposure time as well as hydrophobicencapsulation, which effectively preserves the electricalperformance of flexible BP thin film transistors.Another key metric for evaluating the high-frequency

performance of transistors is the maximum oscillationfrequency, fMAX, which is defined as the frequency of unitypower gain. While f T is closely related to the intrinsic speedperformance of mixed-signal and analog transistor circuits, fMAXis regarded as an upper limit for RF circuits where input andoutput are often investigated in terms of signal power.36 Forthis purpose, we investigated both the extrinsic and intrinsicfMAX extracted from the unilateral power gain, U, as shown inFigure 4b. From Figure 4b, the extrinsic and intrinsic fMAX wereextracted to be 10.3 and 14.5 GHz, respectively, which validatesthe potential for BP-based flexible RF nanosystems. Thereproducibility of our flexible TG BP transistors is discussed inthe Supporting Information Figure S5 and S6 where anotherdevice with channel length of L = 0.5 μm affords intrinsic f T ∼20 GHz and fMAX ∼ 11 GHz.An insightful comparison of the high-frequency performance

metrics of f T, fMAX, f TL and vsat among widely studied thin filmnanomaterials including BP, few-layer MoS2, indium galliumzinc oxide (IGZO), and poly-Si are listed in Table 1. Flexibletop-gate BP transistor demonstrated the highest intrinsic f Tvalue of ∼17.5 GHz with channel length of L = 0.5 μm. The

Figure 4. Radio-frequency performance of the same flexible BP RFtransistor shown in Figure 2 with GHz f T. (a) Short-circuit currentgain |h21| before (extrinsic) and after (intrinsic) standard de-embedding. Extrinsic and intrinsic cutoff frequency f T were extractedas 7 and 17.5 GHz, respectively. (b) Unilateral power gain, U,featuring extrinsic and intrinsic fMAX ∼ 10.3 and 14.5 GHz,respectively. Channel length is 0.5 μm.

Table 1. High-Frequency Performance Comparison of Widely Studied 2D and Thin Film Transistors in Ambient Condition at300 K

f T (GHz) fMAX (GHz)

material sub L (μm) ext int ext int vsata (106 cm/s) f T, Int·L (GHz·μm)

this work BP PI 0.5 7 17.5 10.3 14.5 5.5 8.75ref 27 BP Si 0.3 8 12 12 20 2.3 3.6ref 43 MoS2 PI 0.068 4.7 13.5 5.4 10.5 0.58 0.92ref 44 MoS2 PI 0.5 2.7 5.6 2.1 3.3 0.85 1.35ref 45 IGZO glass 1.5 0.384 1.06 0.38 0.58ref 46 Poly-Si Si 0.2 1.1 1.6 0.2 0.32ref 47 graphene glass 0.14 6 95 8.36 13.3ref 48 graphene PET 0.2 32 64 20 34 8.05 12.8ref 49 InAs PI 0.075 9.38 105 10.5 22.9 4.95 7.87

aExperimental vsat in this table is the effective saturation velocity or highest achieved velocity.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

2304

extracted saturation velocity is ∼6 × 106 cm/s, which is morethan 1 order of magnitude higher than the effective velocitiesachieved for MoS2, IGZO, and poly-Si.43−46 As for themaximum oscillation frequency, the flexible TG BP transistordemonstrated fMAX ∼ 14.5 GHz, which is much higher thanthose reported for TMDs (e.g., MoS2), IGZO, and poly-Siwhich are among the most studied thin film materials forflexible electronics. In addition, graphene and InAs-basedflexible transistors are also included in Table 1 forcomprehensive benchmarking of the state-of-the-art flexiblehigh-frequency transistors. However, graphene is rather limitedin terms of low power electronic applications due to its lack of abandgap, and InAs III−V thin-film is intrinsically of limitedmechanical flexibility owing to its low strain limit.1,47−49 Fromthis comparison, BP is an outstanding nanomaterial for flexibleRF nanosystems and applications.Conclusion. In this work, we present flexible top-gated BP

transistors with high-frequency performance including recordintrinsic f T ∼ 17.5−20 GHz, intrinsic fMAX ∼ 14.5 GHz, andhigh experimental saturation velocity ∼6 × 106 cm/s. Theeffects of the native in-plane anisotropy in BP was considered inthe understanding of the channel length dependence suggestingan intrinsic 2-fold device-to-device variation in the effectivesaturation velocity and the intrinsic f T. The mechanicalrobustness of our flexible BP RF transistors was investigatedthrough ex situ f T measurements under tensile strain up to1.5%. Our results indicate that BP offers the highest chargevelocities and higher-frequency performance compared to other2D layered semiconductors with a natural band gap making itarguably the most promising layered material for flexible RFnanosystems.

■ ASSOCIATED CONTENT*S Supporting InformationThese materials are available free of charge via the Internet atThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.nano-lett.5b04768.

Transconductance derived from transfer characteristicsshown in Figure 2b, transfer and output characteristics oftypical flexible BP RF devices with channel length L = 1and 0.25 μm, high-frequency performance of flexible BPRF devices with channel length L = 0.25 μm showingextrinsic and intrinsic f T and fMAX, DC and high-frequency performance of flexible TG BP transistorshowing intrinsic f T ∼ 20 GHz, AFM analysis of lengthof access region and Schottky barrier width analysis arepresented. (PDF)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: deji@ece.utexas.edu.Author ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work is supported in part by the Office of Naval Research(ONR) under contract N00014-1110190, the NSF National

Nanotechnology Coordinated Infrastructure (NNCI), and theNSF-NASCENT Engineering Research Center under Cooper-ative Agreement No. EEC-1160494. D.A acknowledges the TI/Jack Kilby Faculty Fellowship. S.B acknowledges the support ofAFOSR.

■ REFERENCES(1) Akinwande, D.; Petrone, N.; Hone, J. Nat. Commun. 2014, 5,5678.(2) Bae, S.; Kim, H.; Lee, Y.; Xu, X.; Park, J. S.; Zheng, Y.;Balakrishnan, J.; Lei, T.; Kim, H. R.; Song, Y. I.; Kim, Y. J.; Kim, K. S.;Ozyilmaz, B.; Ahn, J. H.; Hong, B. H.; Iijima, S. Nat. Nanotechnol.2010, 5 (8), 574−8.(3) Jin, Z. H.; Li, X. D.; Mullen, J. T.; Kim, K. W. Phys. Rev. B:Condens. Matter Mater. Phys. 2014, 90 (4), 045422.(4) Kim, S.; Konar, A.; Hwang, W. S.; Lee, J. H.; Lee, J.; Yang, J.;Jung, C.; Kim, H.; Yoo, J. B.; Choi, J. Y.; Jin, Y. W.; Lee, S. Y.; Jena, D.;Choi, W.; Kim, K. Nat. Commun. 2012, 3, 1011.(5) Kobayashi, T.; Bando, M.; Kimura, N.; Shimizu, K.; Kadono, K.;Umezu, N.; Miyahara, K.; Hayazaki, S.; Nagai, S.; Mizuguchi, Y.;Murakami, Y.; Hobara, D. Appl. Phys. Lett. 2013, 102 (2), 023112−4.(6) Li, Y. V.; Ramirez, J. I.; Sun, K. G.; Jackson, T. N. IEEE ElectronDevice Lett. 2013, 34 (7), 891−893.(7) Nathan, A.; Ahnood, A.; Cole, M. T.; Lee, S.; Suzuki, Y.; Hiralal,P.; Bonaccorso, F.; Hasan, T.; Garcia-Gancedo, L.; Dyadyusha, A.;Haque, S.; Andrew, P.; Hofmann, S.; Moultrie, J.; Chu, D. P.; Flewitt,A. J.; Ferrari, A. C.; Kelly, M. J.; Robertson, J.; Amaratunga, G. A. J.;Milne, W. I. Proc. IEEE 2012, 100 (Special Centennial Issue), 1486−1517.(8) Rahimi, S.; Tao, L.; Chowdhury, S. F.; Park, S.; Jouvray, A.;Buttress, S.; Rupesinghe, N.; Teo, K.; Akinwande, D. ACS Nano 2014,8 (10), 10471−9.(9) Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano,M. S. Nat. Nanotechnol. 2012, 7 (11), 699−712.(10) Lee, J.; Ha, T. J.; Li, H.; Parrish, K. N.; Holt, M.; Dodabalapur,A.; Ruoff, R. S.; Akinwande, D. ACS Nano 2013, 7 (9), 7744−50.(11) Moon, J. S.; Seo, H. C.; Son, K. A.; Yang, B.; Le, D.; Fung, H.;Schmitz, A. In Zero-bias THz detection using graphene transistors.Microwave Symposium (IMS), 2015 IEEE MTT-S International, May17−22, 2015; pp 1−4.(12) Chang, H. Y.; Yang, S.; Lee, J.; Tao, L.; Hwang, W. S.; Jena, D.;Lu, N.; Akinwande, D. ACS Nano 2013, 7 (6), 5446−52.(13) Wang, H.; Yu, L.; Lee, Y. H.; Shi, Y.; Hsu, A.; Chin, M. L.; Li, L.J.; Dubey, M.; Kong, J.; Palacios, T. Nano Lett. 2012, 12 (9), 4674−80.(14) Sanne, A.; Ghosh, R.; Rai, A.; Yogeesh, M. N.; Shin, S. H.;Sharma, A.; Jarvis, K.; Mathew, L.; Rao, R.; Akinwande, D.; Banerjee,S. Nano Lett. 2015, 15 (8), 5039−5045.(15) Das, S.; Gulotty, R.; Sumant, A. V.; Roelofs, A. Nano Lett. 2014,14 (5), 2861−6.(16) Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D.; Chen,X. H.; Zhang, Y. Nat. Nanotechnol. 2014, 9 (5), 372−7.(17) Das, S.; Zhang, W.; Demarteau, M.; Hoffmann, A.; Dubey, M.;Roelofs, A. Nano Lett. 2014, 14 (10), 5733−5739.(18) Zhang, S.; Yang, J.; Xu, R.; Wang, F.; Li, W.; Ghufran, M.;Zhang, Y. W.; Yu, Z.; Zhang, G.; Qin, Q.; Lu, Y. ACS Nano 2014, 8(9), 9590−6.(19) Tran, V.; Soklaski, R.; Liang, Y. F.; Yang, L. Phys. Rev. B:Condens. Matter Mater. Phys. 2014, 89 (23), 235319.(20) Castellanos-Gomez, A.; Vicarelli, L.; Prada, E.; Island, J. O.;Narasimha-Acharya, K. L.; Blanter, S. I.; Groenendijk, D. J.; Buscema,M.; Steele, G. A.; Alvarez, J. V.; Zandbergen, H. W.; Palacios, J. J.; vander Zant, H. S. J. 2D Mater. 2014, 1 (2), 025001.(21) Xia, F.; Wang, H.; Jia, Y. Nat. Commun. 2014, 5, 4458.(22) Woomer, A. H.; Farnsworth, T. W.; Hu, J.; Wells, R. A.; Donley,C. L.; Warren, S. C. Phosphorene: Synthesis, Scale-up, andQuantitative Optical Spectroscopy, 2015, arXiv:1505.04663.(23) Tauc, J. Mater. Res. Bull. 1968, 3 (1), 37−46.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

2305

(24) Tao, L.; Lee, J.; Akinwande, D. J. Vac Sci. Technol. B 2011, 29,06FG07.(25) Jena, D.; Konar, A. Phys. Rev. Lett. 2007, 98 (13), 136805.(26) Zhu, W.; Yogeesh, M. N.; Yang, S.; Aldave, S. H.; Kim, J. S.;Sonde, S.; Tao, L.; Lu, N.; Akinwande, D. Nano Lett. 2015, 15 (3),1883−90.(27) Wang, H.; Wang, X.; Xia, F.; Wang, L.; Jiang, H.; Xia, Q.; Chin,M. L.; Dubey, M.; Han, S.-j. Nano Lett. 2014, 14 (11), 6424−6429.(28) Kim, J.-S.; Liu, Y.; Zhu, W.; Kim, S.; Wu, D.; Tao, L.;Dodabalapur, A.; Lai, K.; Akinwande, D. Sci. Rep. 2015, 5, 8989.(29) Wood, J. D.; Wells, S. A.; Jariwala, D.; Chen, K. S.; Cho, E.;Sangwan, V. K.; Liu, X.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C.Nano Lett. 2014, 14, 6964.(30) Chang, H.-Y.; Zhu, W.; Akinwande, D. Appl. Phys. Lett. 2014,104 (11), 113504.(31) Han, L.; Neal, A. T.; Mengwei, S.; Yuchen, D.; Ye, P. D. IEEEElectron Device Lett. 2014, 35 (7), 795−797.(32) Liu, H.; Neal, A. T.; Zhu, Z.; Luo, Z.; Xu, X.; Tomanek, D.; Ye,P. D. ACS Nano 2014, 8 (4), 4033−41.(33) Petrone, N.; Meric, I.; Chari, T.; Shepard, K. L.; Hone, J. IEEE J.Electron Devices Soc. 2015, 3 (1), 44−48.(34) Meric, I.; Han, M. Y.; Young, A. F.; Ozyilmaz, B.; Kim, P.;Shepard, K. L. Nat. Nanotechnol. 2008, 3 (11), 654−659.(35) Parrish, K. N.; Akinwande, D. Appl. Phys. Lett. 2011, 98 (18),183505.(36) Schwierz, F. Proc. IEEE 2013, 101 (7), 1567−1584.(37) Pierret, R. F. Semiconductor Device Fundamentals; PearsonEducation India: Chennai, India, 1996.(38) Heinze, S.; Tersoff, J.; Martel, R.; Derycke, V.; Appenzeller, J.;Avouris, P. Phys. Rev. Lett. 2002, 89 (10), 106801.(39) Lam, K.-T.; Dong, Z.; Guo, J. IEEE Electron Device Lett. 2014,35 (9), 963−965.(40) Park, S.; Chang, C. Y.; Rahimi, S.; Lee, A.; Akinwande, D. InFlexible 2D electronics using nanoscale transparent polyimide gatedielectric; Device Research Conference (DRC), 2015 73rd Annual,June 21−24 2015; pp 193−194.(41) Sze, S. M.; Ng, K. K. Physics of Semiconductor Devices; JohnWiley & Sons: New York, 2006.(42) Shahidi, G. G.; Antoniadis, D. A.; Smith, H. I. IEEE ElectronDevice Lett. 1988, 9 (2), 94−96.(43) Cheng, R.; Jiang, S.; Chen, Y.; Liu, Y.; Weiss, N.; Cheng, H. C.;Wu, H.; Huang, Y.; Duan, X. Nat. Commun. 2014, 5, 5143.(44) Chang, H.-Y.; Yogeesh, M. N.; Ghosh, R.; Rai, A.; Sanne, A.;Yang, S.; Lu, N.; Banerjee, S. K.; Akinwande, D. Adv. Mater. 2016, 28(9), 1818−1823.(45) Su, L.-Y.; Huang, J. Solid-State Electron. 2015, 104, 122−125.(46) Hu, H.; Huang, H. IEEE Electron Device Lett. 2015, 36 (4),345−347.(47) Park, S.; Shin, S. H.; Yogeesh, M.; Lee, A.; Rahimi, S.;Akinwande, D. IEEE Electron Device Lett. 2016, No. 99, 1−1.(48) Yeh, C.-H.; Lain, Y.-W.; Chiu, Y.-C.; Liao, C.-H.; Moyano, D.R.; Hsu, S. S. H.; Chiu, P.-W. ACS Nano 2014, 8 (8), 7663−7670.(49) Wang, C.; Chien, J.-C.; Fang, H.; Takei, K.; Nah, J.; Plis, E.;Krishna, S.; Niknejad, A. M.; Javey, A. Nano Lett. 2012, 12 (8), 4140−4145.

Nano Letters Letter

DOI: 10.1021/acs.nanolett.5b04768Nano Lett. 2016, 16, 2301−2306

2306