University of Southern California Dissertations and Theses

Novel nanomaterials for electronics, optoelectronics and sensing applications

(USC Thesis Other)

Novel nanomaterials for electronics, optoelectronics and sensing applications

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1 Novel Nanomaterials for Electronics, Optoelectronics and Sensing Applications by Ahmad Abbas A Thesis Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (ELECTRICAL ENGINEERING) December 2016 Copyright 2016 Ahmad Abbas 2 Table of Content 1. Introduction 1.1 Graphene Nanoribbons………………………………………….… 4 1.1.1 Introduction to Graphene Nanoribbons………………………….................... 4 1.1.2 Producing Graphene Nanoribbons………………………………..………...… 5 1.1.3 Graphene Synthesis…………………………………………………………… 6 1.2 Black Phosphorus…………………………………………….……. 7 1.2.1 Introduction to Black Phosphorus and Black Arsenic-Phosphorus……….…7 1.2.2 Black Phosphorus Synthesis………………………………….……...……….. 8 1.2.3 Black Arsenic-Phosphorus Synthesis………………………….……...………..9 1.3 References……………………………………………….…..…… 10 2. Patterning, Characterization and Chemical Sensing Applications of Graphene Nanoribbon Arrays Down to 5 nm Using Helium Ion Beam Lithography 2.1 Introduction………………………………………………………. 15 2.2 Results and Discussion………………………………………….... 16 2.3 Conclusion………………………………………………………... 31 2.4 Methods…………………………………………………………... 32 2.5 References………………………………………………………... 34 3. Deposition, Characterization, and Thin-Film-Based Chemical Sensing of Ultra-Long Chemically Synthesized Graphene Nanoribbons 3.1 Introduction………………………………………………………. 36 3.2 Results and Discussion…………………………………………… 37 3.3 Conclusion………………………………………………………... 49 3.4 Methods…………………………………………………………... 50 3.5 References………………………………………………………... 52 3 4. Vapor-Phase Transport Deposition, Characterization, and Applications of Large Nanographene 4.1 Introduction………………………………………………………. 54 4.2 Results and Discussion…………………………………………… 57 4.3 Conclusion………………………………………………………... 76 4.4 Methods…………………………………………………………... 76 4.5 References………………………………………………………... 79 5. Black Phosphorus Gas Sensors 5.1 Introduction………………………………………………………. 81 5.2 Results and Discussion…………………………………………… 84 5.3 Conclusion………………………………………………………... 95 5.4 Methods…………………………………………………………... 95 5.5 References………………………………………………………... 97 6. Room Temperature Black Arsenic-Phosphorus Photodetectors at 5 µm 6.1 Introduction…………………………………….…………….…. 100 6.2 Results and Discussion……………………………….………..…103 6.3 Conclusion………………………………………………………..110 6.4 References………………………………………………………..110 4 1. Introduction: In the past decades, nanomaterials have attracted strong attention due to their versatile applications. Specifically, nanomaterials gained special focus in the fields of electronics and sending due to their flexibly adjustable electronic properties enabled by engineering their size, structure, chemical identity, and other material characteristics. In recent years, materials such as carbon nanotube, 1 nanowires, 2-4 graphene, 5 transition metal dichalcogenides (TMDCs), 6 and black phosphorus (BP) 7-9 were utilized in a wide range of electronic applications such as digital, 10, 11 radio frequency (RF), 12-14 bio-sensing, 15, 16 chemical sensing, 17-19 and optical devices 20-23 . In this letter, I will present our recent efforts to use graphene nanoribbons (GNRs), BP, and black arsenic-phosphorus (BAsP) in various applications. 1.1. Graphene Nanoribbons (GNRs): 1.1.1. Introduction to Graphene Nanoribbons (GNRs): Since the rise of graphene, it has attracted strong attention due to its exceptional electronic, 5 thermal, 24, 25 mechanical, 26, 27 and optical properties. 22, 28, 29 Due to the lack of an energy bandgap in graphene, electronic applications of graphene have been limited to functionalities where switching off the device is not necessary. Accordingly, technologies such as graphene radio frequency transistors have been drawing lots of attention because it is easily implementable in real life applications. 13, 14, 30 On the contrary, digital electronics applications require a sufficient bandgap for the channel material in order to switch off the device for proper operation. As a result, extensive research was done to use graphene as a digital switch by employing novel device structures that made use of the high tunability of graphene Fermi level. 31-35 Although these devices perform fairly well, their structures are complex and lack the use of excellent charge transport properties of graphene. On the other hand, numerous efforts to induce a bandgap in 5 graphene by chemical modification, 36-38 use of multilayer graphene with dual gates, 39-41 and other methods 42 were devoted. Nevertheless, these methods are either difficult to control or devices fail to perform well. The narrowing of graphene into stripes with widths ~ 10 nm was studied both theoretically 43 and experimentally 44 to create an energy bandgap in the electronic structure of graphene due to quantum confinement. However, the theoretical predictions and the experimental data did not match due to inability to control edge structures and edge states of graphene which contribute to the creation of a transport gap. 45 Consequently, several methods were used to create GNRs including top down etching of graphene into GNR, 44, 46, 47 unzipping of carbon nanotubes, 48-52 chemically producing GNR, 53-55 bottom-up chemical synthesis, 56-58 and various methods for producing arrays of GNR. 52, 59-64 Additionally, several interesting transport properties were experimentally tested for several types of GNRs. 65, 66 Nevertheless, in order to create a reliable technology that uses GNR as a platform, better control of GNR should be accomplished. And issues like GNR alignment, width control, aspect ratios, ribbon to ribbon variations, density of GNR and the quality of the materials have to be precisely and reproducibly controlled. These features are essential to control the electronic characteristics of the devices fabricated and proper operation of complex circuits. 1.1.2. Producing Graphene Nanoribbons: GNRs can be synthesized with both bottom-up 56 and top-down approaches. 17, 44 Figure 1 shows the difference between the two approaches. Top-down GNRs are prepared using a continuous graphene film while bottom-up GNRs are prepared via assembling molecular monomers to form a GNR through a chemical process. 6 Figure 1. Comparative schemes for top-down and bottom-up approached for preparation of GNRs. 1.1.3. Graphene Synthesis: In this letter, whenever graphene is used, it is grown using a low pressure chemical-vapor- deposition (CVD) process on a 25 µm thick copper foil (99.98% in purity). The copper foil was annealed at 1000 ᴼC in hydrogen for 20 minutes. For graphene growth, temperature was maintained at 1000 ᴼC for 30 minutes and a mixture of hydrogen, argon and methane with flow rates of 4, 46 and 7 SCCM respectively was flowed at a pressure of 500 mTorr. Transfer was done by spin coating (2000 rpm, 1 min) two layers of methyl methacrylate (MMA) as a polymer scaffold on graphene and subsequently baking at 180 ᴼC for 1 min to solidify the film and improve the adhesion. Afterwards, copper was etched using a water diluted ferric chloride solution, and the MMA/graphene film was rinsed with water and cleaned in a water diluted hydrochloric solution. The film was then transferred to the target substrate and MMA was removed using acetone and hydrogen annealing (350 Cº and 2 hours) subsequently. Figure 2 shows a scanning electron microscope (SEM) image of a single CVD graphene flake on copper. Top-down Fabrication Bottom-up Synthesis b c d e f 30 nm 30 nm 30 nm 100 nm 50 nm graphene GNR Ti/Au Electrodes Oxygen RIE Helium Ion Cutting graphene a 7 Figure 2. SEM image of a graphene CVD flake. 1.2. Black Phosphorus: 1.2.1. Introduction to Black Phosphorus and Black Arsenic-Phosphorus: Recently, the rediscovery of black phosphorus (BP) 7-9 as a new single-element two- dimensional (2D) layered semiconducting material has sparked the interest of scientists in various fields. BP is a layered material with a strong in-plane covalent bonds and weak van der waals interaction between individual layers, which makes the mechanical exfoliation of this material possible. Phosphorus atoms in a single layer are arranged in a puckered honeycomb structure (Figure 3). Electronic and optical properties showed great promise for using BP in numerous applications. The field-effect transistor (FET) of few-layer BP exhibited high charge mobility, anisotropic transport behavior, high operating frequencies, and relatively high current on/off ratios, making BP a potential candidate for future electronics. 7-9, 12, 67, 68 The recently reported device optimization techniques of BP FETs have yielded transistors with even better performance (e.g. higher mobility and lower contact resistance). 20, 69-71 In Addition, optical applications including photovoltaics (PV), photodetectors, and imaging devices were created 8 using BP FETs with different device structures. 20, 21, 72-74 Moreover, passivation and stability of black phosphorus has also been studied. 75, 76 Black-Arsenic phosphorus is an alloy of arsenic and phosphorus atoms in a crystal structure similar to figure 3. As the arsenic (As) percentage increases, the bandgap of the material is reduced. The largest arsenic percentage compound we used was b-As 0.83 P 0.17 (83% As) which has a bandgap of ~0.15 eV. Figure 3. a) Atomic structure of four layers of black phosphorus showing a puckered structure in the in- plane direction. b) Top view of the Layered black phosphorus showing honeycomb stacking of phosphorus atoms. 1.2.2. Black Phosphorus Synthesis: We synthesized BP samples from red phosphorus (Chempur, 99.999+ %) and tin/tin (IV) iodide (Sn/SnI 4 = 10/5 mg per 250 mg batch) in evacuated (p < 10 -3 mbar) silica ampoules a b 9 according to literature procedures. 77 Subsequently, the temperature of the starting materials was raised to 650 °C in a period of 8 hours and that temperature was held for 5 hours. Then, the oven chamber was cooled down to 550 °C in a period of 7.5 hours and was kept at that temperature for 6 hours. Eventually, the mixture was cooled to room temperature. Exfoliation of BP was done using a commercial tape. 1.2.3. Black Arsenic-Phosphorus Synthesis: The b-As 0.83 P 0.17 alloyes were synthesized from gray arsenic (Chempur, 99.9999%) and red phosphorus (Chempur, 99.999 + %), with a molar ratio of 83:17. The mineralizing agent used was lead (II) iodide (PbI 2 , 12 mg per 625 mg batch). All chemicals were enclosed in vacuum in silica glass ampoules during the reaction. The reaction chamber was in a Nabertherm furnace (L3/11/P330), while (i) heating up to 550 °C within 8 h and holding for 84 h at this temperature and (ii) cooling down to room temperature within 20 h. 10 1.3. References: 1. Iijima, S., Helical microtubules of graphitic carbon. Nature 1991, 354 (6348), 56-58. 2. Morales, A. M.; Lieber, C. M., A Laser Ablation Method for the Synthesis of Crystalline Semiconductor Nanowires. Science 1998, 279 (5348), 208-211. 3. Cui, Y.; Wei, Q.; Park, H.; Lieber, C. M., Nanowire Nanosensors for Highly Sensitive and Selective Detection of Biological and Chemical Species. Science 2001, 293 (5533), 1289-1292. 4. Zhang, D.; Liu, Z.; Li, C.; Tang, T.; Liu, X.; Han, S.; Lei, B.; Zhou, C., Detection of NO2 down to ppb Levels Using Individual and Multiple In2O3 Nanowire Devices. Nano Letters 2004, 4 (10), 1919-1924. 5. Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A., Electric Field Effect in Atomically Thin Carbon Films. Science 2004, 306 (5696), 666-669. 6. Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, V.; Kis, A., Single-layer MoS2 transistors. Nat Nano 2011, 6 (3), 147-150. 7. Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D.; Chen, X. H.; Zhang, Y., Black phosphorus field-effect transistors. Nat Nano 2014, 9 (5), 372-377. 8. Xia, F.; Wang, H.; Jia, Y., Rediscovering black phosphorus as an anisotropic layered material for optoelectronics and electronics. Nat Commun 2014, 5. 9. Liu, H.; Neal, A. T.; Zhu, Z.; Luo, Z.; Xu, X.; Tomànek, D.; Ye, P. D., Phosphorene: An Unexplored 2D Semiconductor with a High Hole Mobility. ACS Nano 2014, 8 (4), 4033-4041. 10. Chen, H.; Cao, Y.; Zhang, J.; Zhou, C., Large-scale complementary macroelectronics using hybrid integration of carbon nanotubes and IGZO thin-film transistors. Nat Commun 2014, 5. 11. Wang, H.; Yu, L.; Lee, Y.-H.; Shi, Y.; Hsu, A.; Chin, M. L.; Li, L.-J.; Dubey, M.; Kong, J.; Palacios, T., Integrated Circuits Based on Bilayer MoS2 Transistors. Nano Letters 2012, 12 (9), 4674-4680. 12. Wang, H.; Wang, X.; Xia, F.; Wang, L.; Jiang, H.; Xia, Q.; Chin, M. L.; Dubey, M.; Han, S.-j., Black Phosphorus Radio-Frequency Transistors. Nano Letters 2014, 14 (11), 6424-6429. 13. Lin, Y.-M.; Dimitrakopoulos, C.; Jenkins, K. A.; Farmer, D. B.; Chiu, H.-Y.; Grill, A.; Avouris, P., 100- GHz Transistors from Wafer-Scale Epitaxial Graphene. Science 2010, 327 (5966), 662. 14. Badmaev, A.; Che, Y.; Li, Z.; Wang, C.; Zhou, C., Self-Aligned Fabrication of Graphene RF Transistors with T-Shaped Gate. ACS Nano 2012, 6 (4), 3371-3376. 15. Chang, H.-K.; Ishikawa, F. N.; Zhang, R.; Datar, R.; Cote, R. J.; Thompson, M. E.; Zhou, C., Rapid, Label-Free, Electrical Whole Blood Bioassay Based on Nanobiosensor Systems. ACS Nano 2011, 5 (12), 9883-9891. 16. Aroonyadet, N.; Wang, X.; Song, Y.; Chen, H.; Cote, R. J.; Thompson, M. E.; Datar, R. H.; Zhou, C., Highly Scalable, Uniform, and Sensitive Biosensors Based on Top-Down Indium Oxide Nanoribbons and Electronic Enzyme-Linked Immunosorbent Assay. Nano Letters 2015, 15 (3), 1943-1951. 17. Abbas, A. N.; Liu, G.; Liu, B.; Zhang, L.; Liu, H.; Ohlberg, D.; Wu, W.; Zhou, C., Patterning, Characterization, and Chemical Sensing Applications of Graphene Nanoribbon Arrays Down to 5 nm Using Helium Ion Beam Lithography. ACS Nano 2014. 18. Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng, S.; Cho, K.; Dai, H., Nanotube Molecular Wires as Chemical Sensors. Science 2000, 287 (5453), 622-625. 19. Liu, B.; Chen, L.; Liu, G.; Abbas, A. N.; Fathi, M.; Zhou, C., High-Performance Chemical Sensing Using Schottky-Contacted Chemical Vapor Deposition Grown Monolayer MoS2 Transistors. ACS Nano 2014, 8 (5), 5304-5314. 20. Buscema, M.; Groenendijk, D. J.; Steele, G. A.; van der Zant, H. S. J.; Castellanos-Gomez, A., Photovoltaic effect in few-layer black phosphorus PN junctions defined by local electrostatic gating. Nat Commun 2014, 5. 11 21. Engel, M.; Steiner, M.; Avouris, P., Black Phosphorus Photodetector for Multispectral, High- Resolution Imaging. Nano Letters 2014, 14 (11), 6414-6417. 22. Emani, N. K.; Chung, T.-F.; Ni, X.; Kildishev, A. V.; Chen, Y. P.; Boltasseva, A., Electrically Tunable Damping of Plasmonic Resonances with Graphene. Nano Letters 2012, 12 (10), 5202-5206. 23. Mueller, M. L.; Yan, X.; McGuire, J. A.; Li, L.-s., Triplet States and Electronic Relaxation in Photoexcited Graphene Quantum Dots. Nano Letters 2010, 10 (7), 2679-2682. 24. Balandin, A. A., Thermal properties of graphene and nanostructured carbon materials. Nat Mater 2011, 10 (8), 569-581. 25. Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan, D.; Miao, F.; Lau, C. N., Superior Thermal Conductivity of Single-Layer Graphene. Nano Letters 2008, 8 (3), 902-907. 26. Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H., Large-scale pattern growth of graphene films for stretchable transparent electrodes. Nature 2009, 457 (7230), 706-710. 27. Lee, C.; Wei, X.; Kysar, J. W.; Hone, J., Measurement of the Elastic Properties and Intrinsic Strength of Monolayer Graphene. Science 2008, 321 (5887), 385-388. 28. Yan, H.; Low, T.; Zhu, W.; Wu, Y.; Freitag, M.; Li, X.; Guinea, F.; Avouris, P.; Xia, F., Damping pathways of mid-infrared plasmons in graphene nanostructures. Nat Photon 2013, 7 (5), 394-399. 29. GuT; PetroneN; McMillan, J. F.; van der Zande, A.; YuM; Lo, G. Q.; Kwong, D. L.; HoneJ; Wong, C. W., Regenerative oscillation and four-wave mixing in graphene optoelectronics. Nat Photon 2012, 6 (8), 554-559. 30. Liao, L.; Lin, Y.-C.; Bao, M.; Cheng, R.; Bai, J.; Liu, Y.; Qu, Y.; Wang, K. L.; Huang, Y.; Duan, X., High- speed graphene transistors with a self-aligned nanowire gate. Nature 2010, 467 (7313), 305-308. 31. Britnell, L.; Gorbachev, R. V.; Jalil, R.; Belle, B. D.; Schedin, F.; Mishchenko, A.; Georgiou, T.; Katsnelson, M. I.; Eaves, L.; Morozov, S. V.; Peres, N. M. R.; Leist, J.; Geim, A. K.; Novoselov, K. S.; Ponomarenko, L. A., Field-Effect Tunneling Transistor Based on Vertical Graphene Heterostructures. Science 2012, 335 (6071), 947-950. 32. Yang, H.; Heo, J.; Park, S.; Song, H. J.; Seo, D. H.; Byun, K.-E.; Kim, P.; Yoo, I.; Chung, H.-J.; Kim, K., Graphene Barristor, a Triode Device with a Gate-Controlled Schottky Barrier. Science 2012, 336 (6085), 1140-1143. 33. Georgiou, T.; Jalil, R.; Belle, B. D.; Britnell, L.; Gorbachev, R. V.; Morozov, S. V.; Kim, Y.-J.; Gholinia, A.; Haigh, S. J.; Makarovsky, O.; Eaves, L.; Ponomarenko, L. A.; Geim, A. K.; Novoselov, K. S.; Mishchenko, A., Vertical field-effect transistor based on graphene-WS2 heterostructures for flexible and transparent electronics. Nat Nano 2012, 8 (2), 100-103. 34. Zeng, C.; Song, E. B.; Wang, M.; Lee, S.; Torres, C. M.; Tang, J.; Weiller, B. H.; Wang, K. L., Vertical Graphene-Base Hot-Electron Transistor. Nano Letters 2013, 13 (6), 2370-2375. 35. Yu, W. J.; Li, Z.; Zhou, H.; Chen, Y.; Wang, Y.; Huang, Y.; Duan, X., Vertically stacked multi- heterostructures of layered materials for logic transistors and complementary inverters. Nat Mater 2012, 12 (3), 246-252. 36. Elias, D. C.; Nair, R. R.; Mohiuddin, T. M. G.; Morozov, S. V.; Blake, P.; Halsall, M. P.; Ferrari, A. C.; Boukhvalov, D. W.; Katsnelson, M. I.; Geim, A. K.; Novoselov, K. S., Control of Graphene's Properties by Reversible Hydrogenation: Evidence for Graphane. Science 2009, 323 (5914), 610-613. 37. Balog, R.; Jorgensen, B.; Nilsson, L.; Andersen, M.; Rienks, E.; Bianchi, M.; Fanetti, M.; Laegsgaard, E.; Baraldi, A.; Lizzit, S.; Sljivancanin, Z.; Besenbacher, F.; Hammer, B.; Pedersen, T. G.; Hofmann, P.; Hornekaer, L., Bandgap opening in graphene induced by patterned hydrogen adsorption. Nat Mater 2010, 9 (4), 315-319. 38. Chang, C.-K.; Kataria, S.; Kuo, C.-C.; Ganguly, A.; Wang, B.-Y.; Hwang, J.-Y.; Huang, K.-J.; Yang, W.- H.; Wang, S.-B.; Chuang, C.-H.; Chen, M.; Huang, C.-I.; Pong, W.-F.; Song, K.-J.; Chang, S.-J.; Guo, J.-H.; 12 Tai, Y.; Tsujimoto, M.; Isoda, S.; Chen, C.-W.; Chen, L.-C.; Chen, K.-H., Band Gap Engineering of Chemical Vapor Deposited Graphene by in Situ BN Doping. ACS Nano 2013, 7 (2), 1333-1341. 39. Lui, C. H.; Li, Z.; Mak, K. F.; Cappelluti, E.; Heinz, T. F., Observation of an electrically tunable band gap in trilayer graphene. Nat Phys 2011, 7 (12), 944-947. 40. Velasco, J.; Jing, L.; Bao, W.; Lee, Y.; Kratz, P.; Aji, V.; Bockrath, M.; Lau, C. N.; Varma, C.; Stillwell, R.; Smirnov, D.; Zhang, F.; Jung, J.; MacDonald, A. H., Transport spectroscopy of symmetry-broken insulating states in bilayer graphene. Nat Nano 2012, 7 (3), 156-160. 41. Zou, K.; Zhang, F.; Clapp, C.; MacDonald, A. H.; Zhu, J., Transport Studies of Dual-Gated ABC and ABA Trilayer Graphene: Band Gap Opening and Band Structure Tuning in Very Large Perpendicular Electric Fields. Nano Letters 2013, 13 (2), 369-373. 42. Nakaharai, S.; Iijima, T.; Ogawa, S.; Suzuki, S.; Li, S.-L.; Tsukagoshi, K.; Sato, S.; Yokoyama, N., Conduction Tuning of Graphene Based on Defect-Induced Localization. ACS Nano 2013, 7 (7), 5694-5700. 43. Son, Y.-W.; Cohen, M. L.; Louie, S. G., Energy Gaps in Graphene Nanoribbons. Physical Review Letters 2006, 97 (21), 216803. 44. Han, M. Y.; Özyilmaz, B.; Zhang, Y.; Kim, P., Energy Band-Gap Engineering of Graphene Nanoribbons. Physical Review Letters 2007, 98 (20), 206805. 45. Pan, M.; GirÃ£o, E. C.; Jia, X.; Bhaviripudi, S.; Li, Q.; Kong, J.; Meunier, V.; Dresselhaus, M. S., Topographic and Spectroscopic Characterization of Electronic Edge States in CVD Grown Graphene Nanoribbons. Nano Letters 2012, 12 (4), 1928-1933. 46. Bai, J.; Duan, X.; Huang, Y., Rational Fabrication of Graphene Nanoribbons Using a Nanowire Etch Mask. Nano Letters 2009, 9 (5), 2083-2087. 47. Abramova, V.; Slesarev, A. S.; Tour, J. M., Meniscus-Mask Lithography for Narrow Graphene Nanoribbons. ACS Nano 2013, 7 (8), 6894-6898. 48. Jiao, L.; Zhang, L.; Wang, X.; Diankov, G.; Dai, H., Narrow graphene nanoribbons from carbon nanotubes. Nature 2009, 458 (7240), 877-880. 49. Kosynkin, D. V.; Higginbotham, A. L.; Sinitskii, A.; Lomeda, J. R.; Dimiev, A.; Price, B. K.; Tour, J. M., Longitudinal unzipping of carbon nanotubes to form graphene nanoribbons. Nature 2009, 458 (7240), 872-876. 50. Wei, D.; Xie, L.; Lee, K. K.; Hu, Z.; Tan, S.; Chen, W.; Sow, C. H.; Chen, K.; Liu, Y.; Wee, A. T. S., Controllable unzipping for intramolecular junctions of graphene nanoribbons and single-walled carbon nanotubes. Nat Commun 2013, 4, 1374. 51. Xie, L.; Wang, H.; Jin, C.; Wang, X.; Jiao, L.; Suenaga, K.; Dai, H., Graphene Nanoribbons from Unzipped Carbon Nanotubes: Atomic Structures, Raman Spectroscopy, and Electrical Properties. Journal of the American Chemical Society 2011, 133 (27), 10394-10397. 52. Jiao, L.; Zhang, L.; Ding, L.; Liu, J.; Dai, H., Aligned graphene nanoribbons and crossbars from unzipped carbon nanotubes. Nano Research 2010, 3 (6), 387-394. 53. Wang, X.; Dai, H., Etching and narrowing of graphene from the edges. Nat Chem 2010, 2 (8), 661-665. 54. Wang, X.; Ouyang, Y.; Li, X.; Wang, H.; Guo, J.; Dai, H., Room-Temperature All-Semiconducting Sub-10-nm Graphene Nanoribbon Field-Effect Transistors. Physical Review Letters 2008, 100 (20), 206803. 55. Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H., Chemically Derived, Ultrasmooth Graphene Nanoribbon Semiconductors. Science 2008, 319 (5867), 1229-1232. 56. Cai, J.; Ruffieux, P.; Jaafar, R.; Bieri, M.; Braun, T.; Blankenburg, S.; Muoth, M.; Seitsonen, A. P.; Saleh, M.; Feng, X.; Mullen, K.; Fasel, R., Atomically precise bottom-up fabrication of graphene nanoribbons. Nature 2010, 466 (7305), 470-473. 13 57. Ruffieux, P.; Cai, J.; Plumb, N. C.; Patthey, L.; Prezzi, D.; Ferretti, A.; Molinari, E.; Feng, X.; Mullen, K.; Pignedoli, C. A.; Fasel, R., Electronic Structure of Atomically Precise Graphene Nanoribbons. ACS Nano 2012, 6 (8), 6930-6935. 58. Schwab, M. G.; Narita, A.; Hernandez, Y.; Balandina, T.; Mali, K. S.; De Feyter, S.; Feng, X.; Mullen, K., Structurally Defined Graphene Nanoribbons with High Lateral Extension. Journal of the American Chemical Society 2012, 134 (44), 18169-18172. 59. Liang, X.; Wi, S., Transport Characteristics of Multichannel Transistors Made from Densely Aligned Sub-10 nm Half-Pitch Graphene Nanoribbons. ACS Nano 2012, 6 (11), 9700-9710. 60. Kato, T.; Hatakeyama, R., Site- and alignment-controlled growth of graphene nanoribbons from nickel nanobars. Nat Nano 2012, 7 (10), 651-656. 61. Sokolov, A. N.; Yap, F. L.; Liu, N.; Kim, K.; Ci, L.; Johnson, O. B.; Wang, H.; Vosgueritchian, M.; Koh, A. L.; Chen, J.; Park, J.; Bao, Z., Direct growth of aligned graphitic nanoribbons from a DNA template by chemical vapour deposition. Nat Commun 2013, 4. 62. Solís-Fernández, P.; Yoshida, K.; Ogawa, Y.; Tsuji, M.; Ago, H., Dense Arrays of Highly Aligned Graphene Nanoribbons Produced by Substrate-Controlled Metal-Assisted Etching of Graphene. Advanced Materials 2013. 63. Son, J. G.; Son, M.; Moon, K.-J.; Lee, B. H.; Myoung, J.-M.; Strano, M. S.; Ham, M.-H.; Ross, C. A., Sub-10 nm Graphene Nanoribbon Array Field-Effect Transistors Fabricated by Block Copolymer Lithography. Advanced Materials 2013, 25 (34), 4723-4728. 64. Pan, Z.; Liu, N.; Fu, L.; Liu, Z., Wrinkle Engineering: A New Approach to Massive Graphene Nanoribbon Arrays. Journal of the American Chemical Society 133 (44), 17578-17581. 65. Wang, X.; Ouyang, Y.; Jiao, L.; Wang, H.; Xie, L.; Wu, J.; Guo, J.; Dai, H., Graphene nanoribbons with smooth edges behave as quantum wires. Nat Nano 6 (9), 563-567. 66. Koch, M.; Ample, F.; Joachim, C.; Grill, L., Voltage-dependent conductance of a single graphene nanoribbon. Nat Nano 2012, 7 (11), 713-717. 67. Koenig, S. P.; Doganov, R. A.; Schmidt, H.; Castro Neto, A. H.; Özyilmaz, B., Electric field effect in ultrathin black phosphorus. Applied Physics Letters 2014, 104 (10), 103106. 68. Zhang, X.; Xie, H.; Liu, Z.; Tan, C.; Luo, Z.; Li, H.; Lin, J.; Sun, L.; Chen, W.; Xu, Z.; Xie, L.; Huang, W.; Zhang, H., Black Phosphorus Quantum Dots. Angewandte Chemie International Edition, DOI: 10.1002/anie.201409400. 69. Kamalakar, M. V.; Madhushankar, B. N.; Dankert, A.; Dash, S. P., Low Schottky Barrier Black Phosphorus Field-Effect Devices with Ferromagnetic Tunnel Contacts. Small, DOI: 10.1002/smll.201402900. 70. Li, L.; Ye, G. J.; Tran, V.; Fei, R.; Chen, G.; Wang, H.; Wang, J.; Watanabe, K.; Taniguchi, T.; Yang, L.; Chen, X. H.; Zhang, Y., Quantum Oscillations in Black Phosphorus Two-dimensional Electron Gas. arXiv 2014, arXiv:1411.6572. 71. Du, Y.; Liu, H.; Deng, Y.; Ye, P. D., Device Perspective for Black Phosphorus Field-Effect Transistors: Contact Resistance, Ambipolar Behavior, and Scaling. ACS Nano 2014, 8 (10), 10035-10042. 72. Buscema, M.; Groenendijk, D. J.; Blanter, S. I.; Steele, G. A.; van der Zant, H. S. J.; Castellanos- Gomez, A., Fast and Broadband Photoresponse of Few-Layer Black Phosphorus Field-Effect Transistors. Nano Letters 2014, 14 (6), 3347-3352. 73. Youngblood, N.; Chen, C.; Koester, S. J.; Li, M., Waveguide-integrated black phosphorus photodetector with high responsivity and low dark current. Nat Photon 2015, doi:10.1038/nphoton.2015.23. 74. Yuan, H.; Liu, X.; Afshinmanesh, F.; Li, W.; Xu, G.; Sun, J.; Lian, B.; Ye, G.; Hikita, Y.; Shen, Z.; Zhang, S.-C.; Chen, X.; Brongersma, M.; Hwang, H. Y.; Cui, Y., Broadband Linear-Dichroic Photodetector in a Black Phosphorus Vertical p-n Junction. arXiv 2014, arXiv:1409.4729. 14 75. 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., Effective Passivation of Exfoliated Black Phosphorus Transistors against Ambient Degradation. Nano Letters 2014, 14 (12), 6964-6970. 76. Ziletti, A.; Carvalho, A.; Campbell, D. â. K.; Coker, D. â. F.; Castro Neto, A. â. H., Oxygen Defects in Phosphorene. Physical Review Letters 2015, 114 (4), 046801. 77. Köpf, M.; Eckstein, N.; Pfister, D.; Grotz, C.; Krüger, I.; Greiwe, M.; Hansen, T.; Kohlmann, H.; Nilges T., Access and in situ growth of phosphorene-precursor black phosphorus. 2014; 405, 6-10. 15 2. Patterning, Characterization and Chemical Sensing Applications of Graphene Nanoribbon Arrays Down to 5 nm Using Helium Ion Beam Lithography 2.1 Introduction: Helium ion beam lithography (HIBL) 1, 2 and helium ion microscopy 3 have recently demonstrated high resolutions mainly due to their small beam spot size and relatively small scattering length. 4 Additionally, it has been recently shown that HIBL can cut and pattern graphene with features down to ~10 nm. 5, 6 Furthermore, HIBL was used to create random defects in a graphene lattice in order to open a bandgap. 7 Nevertheless, those techniques didn’t take advantage of the extremely high resolution and control HIBL can offer. Here we report the use of HIBL to create GNR field-effect transistors (GNR FETs) with highly dense GNR arrays with one of the smallest half-pitch GNR reported to date. In addition, we can control the aspect ratio of the GNR up to 400 (i.e. the widths and lengths of GNRs are 5 nm and 2000 nm, respectively). Raman spectra of HIBL fabricated GNRs show relatively large G-band to D-band intensity ratios (i.e. I G /I D ) which indicate the quality of these GNRs compared to lithographically etched GNRs. 8 Subsequently, we have performed low temperature electrical measurements of the 5 nm GNR array device and observed thermal activation behavior of carriers in the graphene/GNR junction. Finally, we measured the sensitivity of GNR arrays to NO 2 gas and observed ppb level sensitivity showing the potential for using HIBL GNR arrays for sensing applications. 9-11 16 2.2 Results and Discussion: 3. Figure 1. a) Scheme of GNR arrays fabricated by HIBL. b, c, d) Helium ion microscope images of 5nm, 6nm and 7.5 nm half-pitch arrays . e) Helium ion microscope image of high b c d e f 30 nm 30 nm 30 nm 100 nm 50 nm graphene GNR Ti/Au Electrodes Oxygen RIE Helium Ion Cutting graphene a 17 aspect ratio GNRs (width × length is 5 nm × 1200 nm). f) Helium ion microscope image shows smooth interface between graphene and patterned GNRs. For All images, bright lines represent graphene. Figure 1a shows the fabrication steps for the HIBL GNR array transistors. First, chemical vapor deposition (CVD) graphene was transferred onto a P ++ Si/SiO 2 wafer using common polymer-mediated transfer techniques 12-14 . Then, Ti/Au electrodes were patterned on top of graphene to form contacts for source and drain. Afterwards, a channel definition took place using oxygen reactive ion etch (RIE) to remove unwanted graphene. Finally, the whole device was loaded into a HIBL machine to create the desired GNR patterns. Figure 1b-c shows helium ion microscope images of several GNR half-pitches (i.e. 5, 6 and 7.5 nm respectively). These images show the precise control of the widths, spacing, as well as the alignment of GNRs we achieved for GNR arrays (i.e. the bright lines in the helium ion microscope image is graphene). Additionally, Figure 1e shows the high aspect ratios and density we can achieve using HIBL. For the employed device structure, the Ti/Au electrodes contact graphene which makes graphene the actual source and drain in contact with the GNR array channel (Figure 1f). Helium ion microscope was used to capture all the images of the GNR arrays, which cannot otherwise be imaged by scanning electron microscope (SEM), because helium ion microscope not only has better resolution but also better signal to image graphene patterns. The interaction volume of helium ions is much smaller than that of electrons, and thus the image carries more information from surface properties (i.e. the graphene layer). For all the devices created using HIBL, we have used monolayer graphene CVD (Figure 2). 18 Figure 2. Raman spectrum of monolayer graphene used in this study. Laser wavelength is 514 nm. In order to assess the quality of HIBL GNRs, Raman spectra of the GNR arrays were performed. Figure 3a show the Raman spectra of GNR arrays with 5, 6, 8, 10, and 15 nm half- pitch. For all the Raman spectra measurements, a 532 nm laser was used. It can be observed that the G-band (~1589 cm -1 ) and the 2D-band (~2670 cm -1 ) broaden as the widths of GNR decrease from 15 nm to 5 nm. For GNR with widths of 5 nm and 6 nm, the 2D-band cannot be distinguished due to further broadening. This can be explained by softening of the phonons because of the dominance of defects in the GNR lattice. 15 Simultaneously, since for narrower GNR the defects represent a higher percentage of the Raman signal, the heterogeneous nature near the edge causes the broadening of the peaks. 8 Moreover, it has been predicted that the G- band upshifts in frequency with decreasing GNR widths which can be explained by the fact that with higher defect densities higher frequency phonons are allowed and an additional peak 19 merges with the G-band peak which results in an apparent up shift in frequency. 15 However, the G-band shift we observed is random and not monotonic (figure 4). The G-band shift can also be explained by quantum confinement effects in GNR. 16 Figure 3b shows the dependence of I G /I D on the inverse of GNR array half-pitch (i.e. GNR width). This plot shows the effect of increasing the percentage of edge defects on both G- and D-bands. Below a certain size, the number of ordered carbon rings compared to defects decrease and I D decreases accordingly. On the other hand, I G is related to the bond stretching of sp2 bond and accordingly will not be affected by GNR narrowing. 15 Consequently, the I G /I D will increase as GNRs get smaller which further confirms S. Ryu et al. 8 observation on Raman relaxation length of D-mode phonons. The measured I G /I D data points in figure 3b can be fitted into a line with equation: I G I D ≈ 8.39 W ; where W is the width of GNRs in nm. This equation does not apply to wider GNR (i.e. W ≥ 15 nm) because the mechanisms that govern I D are different. 8, 15, 17 Furthermore, comparing I G /I D for the same width can give an indication about the quality of the patterned GNRs. In Figure 3c we compare our I G /I D to other recent publications in order to highlight the relative quality of HIBL GNR patterning. Since I G originates from the doubly degenerate zone center mode which corresponds to the carbon atoms (sp2 bond) phonons in graphene 17 and I D corresponds to various defects in graphene (e.g. vacancies, stone-wales, foreign adatoms and edge defects) 15, 18, 19 , the ratio of I G /I D for the same GNR width can be a measure for material quality. For example, our 15 nm wide GNRs have I G /I D of 0.52 compared to 0.2 for electron beam lithography (EBL) patterned GNR 8 . This highlights the quality of GNRs patterned with HIBL compared to other lithography based techniques. 20 Figure 3.Raman spectra of HIBL patterned GNR arrays. a) 5 nm, 6 nm, 8 nm, 10 nm and 15 nm half-pitch. b) G-band (I G ) to D-band (I D ) intensity ratios versus the inverse of the GNR array half-pitch and the corresponding fitted curve. c) G-band (I G ) to D-band (I D ) intensity ratios of this work compared to some recent publications. Figure 4. a) D-band and b) G-band peak positions vs. the widths of GNR arrays. 1000 1500 2000 2500 3000 Intensity (a.u.) Raman Shift (cm -1 ) W = 5 nm W = 6 nm W = 8 nm W = 10 nm W = 15 nm 0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.0 0.4 0.8 1.2 1.6 2.0 Measurement Fitted Curve I G /I D 1/Graphene nanoribbon width (nm -1 ) 0 5 10 15 20 25 0.0 0.4 0.8 1.2 1.6 2.0 This Work Ref 54 Ref 35 Ref 45 Ref 31 Ref 28 Ref 43 I G /I D Graphene nanoribbon width (nm) a b c D G 2D G+D 4 6 8 10 12 14 16 1588 1590 1592 1594 1596 1598 1600 1602 1604 G-band Raman Shift (cm -1 ) GNR width (nm) 4 6 8 10 12 14 16 1338 1340 1342 1344 1346 1348 1350 D-Band Raman shift (cm -1 ) GNR width (nm) a b 21 Electrical measurements were carried out for the GNR array devices using P ++ Si as a back gate. We performed the measurements on a 6 nm half-pitch GNR array device with a long channel length (i.e. 2 µm). Figure 5a shows the graphene device before patterning the GNR with a Dirac point close to zero gate voltage. Figure 5b and 5c show the device performance after patterning the GNR array devices at 300 K and 77K respectively. In order to compare the ON/OFF ratios of the GNR array device at different conditions, we define the ON/OFF ratio to be the current at gate voltage of (V g = -15V) divided by the minimum current at the charge neutrality point. It can be observed that the ON/OFF ratio increases from 2 to 2.25 after patterning the GNR at 300K and reaches a value of 4.75 at 77K. Additionally, the minimum conductance decreased for the device from 300 K to 77 K which is a typical semiconducting behavior. Figure 5d shows all the measured curves under log scale. It should be noted that the ON current level dropped dramatically after patterning the GNR array. Moreover, it should be noted that the gate voltage sweeping range in the GNR array device is smaller than that of the graphene device before patterning the GNR. This is due to the increased leakage through the gate oxide after patterning. Additionally, we also measured HIBL GNR with width of 15 nm, and the transport characteristics (Figure 6a) showed ON current value of 4 μA at drain voltage of 0.1 V and gate voltage of -25 V, which would convert to a reduction in current density (i.e. current normalized by width) by a factor of ~2 when compared to graphene before HIBL. Comparatively, a reduction in current density of ~5000 was observed for 6 nm HIBL GNR which may be attributed to: (i) GNR edge scattering which is inversely proportional to GNR width 20 ,(ii) increased impurity scattering after HIBL due to helium ions breaking Si-O bonds in SiO 2 dielectric and creating surface defect states, (iii) increased electron and hole masses for narrow GNR 21 and (iv) reduced effective transconductance of HIBL GNR devices because of 22 charge traps created in SiO 2 which contribute to effects such as weak Fermi level pinning and the screening of gate electric field which would also negatively affect the ON/OFF ratio . 22, 23 Figure 5.Transfer characteristics of a) graphene device before patterning, b) device after patterning 6 nm GNR arrays at room temperature and c) 6 nm GNR array device measured at (77K). d) Transfer characteristics in log scale of a-c. V d was 0.1 V for all the measurements. -15 -10 -5 0 5 10 15 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 T = 77 K Drain Current (nA) Gate Voltage (V) -15 -10 -5 0 5 10 15 0.6 0.8 1.0 1.2 1.4 T = 300 K Drain Current (nA) Gate Voltage (V) -30 -20 -10 0 10 20 30 4 6 8 10 12 14 T= 300 K Drain Current ( A) Gate Voltage (V) a c b -30 -20 -10 0 10 20 30 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 Drain Current (A) Gate Voltage (V) d 23 Figure 6. (a) Transport characteristics of 15 nm wide HIBL GNR array showing ON current of 4 µA and (b) Measured conductance of 15 nm, 10 nm, and 6 nm wide HIBL GNR at a gate voltage of -15 V. The linear equation conductance fit (G= σ(W-W 0 )/L) yielded a W 0 of 5.6 nm. Current degradation for EBL-patterned GNRs was rarely reported; however, our current degradation for 15 nm HIBL GNR compares favorably with limited data we can estimate from Ref. 24 , which reported a degradation of mobility (and thus current) by a factor of 15 when the GNR width goes from 1000 nm to 15 nm. We believe the main factor for the significant current degradation of 6 nm GNR is that 6 nm is close to the so-called inactive GNR edge width (W 0 ). The conductance of GNR can be described with a linear fit G= σ(W-W 0 )/L; where σ is the GNR sheet conductivity, W 0 is the GNR inactive edge width, W is the GNR width and L is the GNR length. The inactive GNR edge width is usually believed to be related to contributions from localized edge states scattering due to GNR edge roughness caused by etching. 25 Figure 6b plots the conductance of HIBL GNRs with widths of 15 nm, 10 nm, and 6 nm vs. GNR width, and a fit using the above equation yields W 0 of 5.6 nm, which is comparable to W 0 estimated for EBL- patterned GNRs. Furthermore, Raman characterization (Figure 3c) has revealed a similar I G /I D 0 2 4 6 8 10 12 14 16 0 5 10 15 20 25 Measurement Fitted Curve Vg = -15 V G( S) GNR width (nm) -30 -20 -10 0 10 20 30 2.0 2.5 3.0 3.5 4.0 15 nm GNR Vd = 0.1 V Drain Current ( A) Gate Voltage (V) W 0 ~ 5.6 nm a b 24 ratio for HIBL GNR in comparison to EBL patterned ones which confirms similar levels of defects created by HIBL and EBL processes. 8 In order to evaluate the effect of helium ions scattering in SiO 2, we performed a Monte Carlo probabilistic analysis based on the high energy stopping of light ions using commercial software. This simulation sheds light on the effect of defect states in SiO 2 on the gate control and Fermi level tunability in GNR array devices (Figure 7). Principally, in order to improve the device performance, patterning should be made on graphene before transfer. Alternatively, the whole device can be transferred from the substrate by using an oxide etcher and a polymer scaffold. A possible transfer-free method is based on free standing suspended GNR channel where the SiO 2 underneath the GNR is etched. 25 This method might have a low yield but would be interesting for scientific study purposes. b 56 nm SiO 2 500 nm Si 500 nm c a 10 μm SiO 2 GNR P++ Si V GS > 0 V GS < 0 25 Figure 7. a) An optical image of a HIBL patterned device after etching GNR showing a color contrast under the patterned area due to helium ion bombardment. b) Monte Carlo simulation of helium ions, with 30 KeV energy, scattering into the substrate used in this work. Each red line trace represents the scattering path of a single helium ion. c) Energy band diagram showing the effect of defect states on Fermi level modulation in the GNR array channel (i.e. blue lines) created by energetic helium ions in the SiO 2 . Green and red dashed lines represent the Fermi levels in both the P ++ Si back gate and GNR arrays as the gate voltage (V GS ) is swept to positive and negative values, respectively. In order to study the carrier transport in the patterned GNR arrays, we performed low temperature output characteristic (i.e. I d -V d ) measurements of a device to study the nature of thermionic activation of carriers through the graphene/GNR junction. For this device, in order to have higher operating currents, GNRs with a width of 5 nm and length of 200 nm were patterned. Figure 8a shows the temperature dependence of the I d -V d curves with clear reduction of the conductance as the temperature is reduced. The nonlinearity of the differential conductance near the zero bias point (i.e. V d = 0 V) is clearly shown for lower temperatures in Figure 8b. This effect indicates a potential barrier created for carriers transporting through graphene/GNR junction. Moreover, estimation of the activation energy (E A ) was carried out using the minimum conductance value for each temperature point. Figure 8c shows that the minimum conductance (G min ) points fit with the thermally activated carriers’ equation: G min = G 1 e -E A K B T Where G 1 is a constant, K B is Boltzmann’s constant, T is the absolute temperature, and E A is the activation energy. The curve fitting yielded an E A of 44 meV, and based on Figure 5b and 5c, we note that the Dirac point corresponds to a small negative gate voltage, and device at zero gate voltage shows n-type conduction. The E A derived based on output characteristics of the device at 26 zero gate voltage (Figure 8) should correspond to the energy difference between metal Fermi level and GNR conduction band, which should be slightly smaller than the energy difference between metal Fermi level and GNR valence band (as the gate voltage corresponding to the Dirac point is slightly negative). The GNR bandgap (E G ) is therefore estimated to be > 88 meV. Theoritical tight-binding models suggest that a GNR with 5 nm width will have a bandgap ~ 0 to 0.3 eV depending on the edge structure and number of dimer rows. 26 The value extracted from Figure 8 (i.e. E G ≥ 0.088 eV) falls within the expected theoretical values. 27 Figure 8.Temperature-dependent electrical measurements of 5nm wide and 200 nm long GNR array: a) Output characteristics of the device under different temperatures. b) Minimum differential conductance (G min ) variation versus drain voltage at different temperatures. c) Minimum conductance vs. inverse temperature and the corresponding curve fitting. The gate voltage was 0V. -0.4 -0.2 0.0 0.2 0.4 1E-8 1E-7 1E-6 300K 250K 200K 170 K 130 K 100K 77K G (S) Drain Voltage (V) 2 4 6 8 10 12 14 1E-9 1E-8 1E-7 1E-6 Measurement Fitted Curve G min (S) 1000/T(K -1 ) -0.4 -0.2 0.0 0.2 0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 300K 250K 200K 170 K 130 K 100 K 77K Drain Current ( A) Drain Voltage (V) a b c 28 One of the very promising applications of GNR devices is chemical sensing. 27 The reduced charge density caused by their small dimensions and the bandgap opening may provide higher sensitivity and a larger range of modulation of the electronic structure of the GNR compared with gapless graphene. In the light of this motivation, we have performed gas sensing experiments using a 5 nm half-pitch HIBL GNR array with 200 nm channel length. The target gas we used in the experiments was nitrogen dioxide (NO 2 ). High sensitivity to NO 2 is very important for health and environmental safety as concentrations above 0.2 ppm can cause respiratory irritation. 28 Figure 9a shows the changes of the I d -V d curves as a function of NO 2 concentrations. Nitrogen was used as a dilution gas to obtain NO 2 with different concentrations. It can be seen from Figure 9a that concentrations down to 20 ppb can be clearly detected with the GNR array device. The GNR array FETs exhibit ambipolar transport behavior, as can be discerned from Figure 5. Since NO 2 act as an electron acceptor, consequently, the GNR devices exhibit increased conductance with increasing the concentrations of NO 2 , As control experiments, after finishing the NO 2 sensing experiment and recovering the device back to the original conductance; the device was subjected to pure nitrogen environment to study the drift of the device conductance over time (Figure 10). This step is aimed to rule out the time drift of the conductance of the device as a source of false signal. Figure 9b is showing conductance change ∆G/G 0 of NO 2 sensing experiment. The sensor response is defined as: ∆G G 0 = G S − G 0 G 0 Where G 0 and G S are the conductance of the device before gas exposure and after 1000 seconds of exposure to certain concentration of gas. The apparent difference in ∆G/G 0 between the sensing and control experiments indicates the sensitivity of the GNR array device to NO 2 29 (Figures 9b and 10). Moreover, the measurement fits with the Langmuir isotherm for molecules absorbed on a surface. This indicates that charge transfer and monolayer molecular adsorption via site-binding govern the sensing mechanisms of the HIBL GNR array devices. Our device showed an unprecedented sensitivity for GNR FET based gas sensors. Novoselov et al. 9 has demonstrated single molecule sensitivity with a hall measurement system while a relatively small (4%) change of graphene device conductance was observed when exposed to 1 ppm NO 2 . Also, other graphene based sensors showed even less NO 2 sensitivity. 11 Comparatively, we observe a 53± 8% change in conductance for 0.4 ppm concentration of NO 2 gas. The high sensitivity of HIBL GNR array devices can be attributed to: (i) edge states of GNR which are more active sites for binding with NO 2 gas molecules than pristine sp2 bonded graphene surface, 29 (ii) the transition from a semi-metal to a semiconductor after HIBL resulting in a reduced charge density and better current modulation by NO 2 molecules. As previously mentioned, the charge traps in SiO 2 may screen the gate electric field and negatively impact the transconductance; however, chemical sensing involves modulation of carrier density in GNR by electron extraction using NO 2 and is not affected by charge traps inside SiO 2 . This demonstrates the potential for using HIBL GNR arrays for high sensitive sensing applications. 30 Figure 9. GNR array based NO 2 gas sensor a) Output characteristics of the GNR array sensor as a function of NO 2 gas concentrations. b) Conductance change (∆G/G 0 ) of the same GNR array sensor as a function of NO 2 gas concentration. Error bars are extracted from the control experiment of the same device with Nitrogen flow only c) Langmuir isotherm curve fitting for the same device showing agreement between measured conductance values (red squares) and fitted curve (black line). 0.00 0.01 0.02 0.03 0.04 0.05 2 4 6 8 10 12 14 Measurement Langmuir Isotherm fitting G 0 G 1/NO 2 Concentration (ppb -1 ) 0.0 0.1 0.2 0.3 0.4 0.5 0 100 200 300 400 500 400 ppb 200 ppb 100 ppb 40 ppb 20 ppb Original device Drain Current (nA) Drain Voltage (V) b c a 0 100 200 300 400 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 G/G 0 NO 2 Concentration (ppb) 31 Figure 10. Output characteristics of the 5 nm half-pitch GNR array device while subject to a control experiment to examine the device drift over time. In the experiment, nitrogen (a) and argon (c) was flowed with a constant flow rate and measurements were repeated every 1000 seconds. b, d) Conductance changes (∆G/G 0 ) of devices in (a) and (c) as a function of time when subjected to nitrogen (b) and argon (d). 2.3 Conclusion: In summary, we developed an efficient method for pattering CVD grown graphene into narrow and tunable widths, highly aligned, densely packed and high aspect ratio GNR arrays using HIBL. Moreover, Raman spectra of HIBL GNR showed the relative high quality of the patterning and the resultant material. Additionally, we have revealed the existence of activation energy of 44 meV in carrier transport through the graphene/GNR junction. Finally, we showed the exceptional NO 2 sensing performance of the GNR array device. The GNR array based sensor 0.0 0.2 0.4 0 100 200 300 400 500 5000 second 4000 second 3000 second 2000 second 1000 second Original device Drain Current (nA) Gate Voltage (V) 0.0 0.1 0.2 0.3 0.4 0.5 0 100 200 300 400 500 5000 second 4000 second 3000 second 2000 second 1000 second Original device Drain Current (nA) Drain Voltage (V) 0 1000 2000 3000 4000 5000 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 G/G 0 Time (second) 0 1000 2000 3000 4000 5000 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 G/G 0 Time (second) a b c d 32 exceeded the performance of other graphene based sensors and is sensitive to low NO 2 concentrations down to 20 ppb. 2.4 Methods: 2.4.1. Graphene growth and transfer: graphene was grown using a low pressure CVD growth on a 25 µm thick copper foil (99.98% in purity). The copper foil was annealed at 1000 ᴼC in hydrogen for 20 minutes. For graphene growth, temperature was maintained at 1000 ᴼC for 30 minutes and a mixture of hydrogen, argon and methane with flow rates of 4, 46 and 7 SCCM respectively was flowed at a pressure of 500 mTorr. Transfer was done by spin coating (2000 rpm, 1 min) two layers of methyl methacrylate (MMA) as a polymer scaffold on graphene and subsequently baking at 180 ᴼC for 1 min to solidify the film and improve the adhesion. Afterwards, copper was etched using a water diluted ferric chloride solution and the MMA/graphene film was rinsed with water and cleaned in a water diluted hydrochloric solution. The film was then transferred to the target substrate and MMA was removed using acetone and hydrogen annealing (350 Cº and 2 hours) subsequently. 2.4.2. Helium Ion Beam Lithography: The GNRs were patterned by direct helium ion beam milling using a helium ion microscope (Orion plus, Carl Zeiss SMT GmbH). The patterning was done using 5 µm aperture size and a beam spot 4, which exhibited a 0.9 pA beam current. The patterns were written as single pixel lines with step size of about 0.3 nm. The dose to pattern 5 nm half-pitch GNRs was 10 nC/cm and lager doses were used to patterns with larger features. 33 2.4.3. NO 2 Gas sensing: Gas sensing was carried out by exposing the GNR array FET device to Nitrogen diluted NO 2 gas in a closed chamber. Concentrations of NO2 were adjusted by changing the flow rates of both gases while keeping the total flow rate constant. For each curve, the device was exposed to the desired concentration for 1000 seconds and then the I d -V d curve was measured. Control device was subjected to the same measurement conditions (i.e. same time between measurements and total flow rate) but with Nitrogen gas only. 2.4.4. Helium Ion Scattering Simulation: Simulation was performed using a commercial program (SRIM). Simulation parameters were as follows. Helium Ions, 1000 data points, 30 KeV energy and SiO 2 /Si of 56 nm and 500 nm thicknesses, respectively) 34 2.5 References: 1. Sidorkin, V.; van Veldhoven, E.; van der Drift, E.; Alkemade, P.; Salemink, H.; Maas, D., Sub-10- nm nanolithography with a scanning helium beam. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 2009, 27 (4), L18-L20. 2. Winston, D.; Cord, B. M.; Ming, B.; Bell, D. C.; DiNatale, W. F.; Stern, L. A.; Vladar, A. E.; Postek, M. T.; Mondol, M. K.; Yang, J. K. W.; Berggren, K. K. In Scanning-helium-ion-beam lithography with hydrogen silsesquioxane resist, AVS: 2009; pp 2702-2706. 3. Ward, B. W.; Notte, J. A.; Economou, N. P. In Helium ion microscope: A new tool for nanoscale microscopy and metrology, AVS: 2006; pp 2871-2874. 4. Li, W.-D.; Wu, W.; Williams, R. S., Combined helium ion beam and nanoimprint lithography attains 4 nm half-pitch dense patterns. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures 2012, 30 (6), 06F304-4. 5. Lemme, M. C.; Bell, D. C.; Williams, J. R.; Stern, L. A.; Baugher, B. W. H.; Jarillo-Herrero, P.; Marcus, C. M., Etching of Graphene Devices with a Helium Ion Beam. ACS Nano 2009, 3 (9), 2674-2676. 6. Bell, D. C.; Lemme, M. C.; Stern, L. A.; Williams, J. R.; Marcus, C. M., Precision cutting and patterning of graphene with helium ions. Nanotechnology 2009, 20 (45), 455301. 7. Nakaharai, S.; Iijima, T.; Ogawa, S.; Suzuki, S.; Li, S.-L.; Tsukagoshi, K.; Sato, S.; Yokoyama, N., Conduction Tuning of Graphene Based on Defect-Induced Localization. ACS Nano 2013, 7 (7), 5694-5700. 8. Ryu, S.; Maultzsch, J.; Han, M. Y.; Kim, P.; Brus, L. E., Raman Spectroscopy of Lithographically Patterned Graphene Nanoribbons. ACS Nano 2011, 5 (5), 4123-4130. 9. Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P.; Katsnelson, M. I.; Novoselov, K. S., Detection of individual gas molecules adsorbed on graphene. Nat Mater 2007, 6 (9), 652-655. 10. Fowler, J. D.; Allen, M. J.; Tung, V. C.; Yang, Y.; Kaner, R. B.; Weiller, B. H., Practical Chemical Sensors from Chemically Derived Graphene. ACS Nano 2009, 3 (2), 301-306. 11. Pearce, R.; Iakimov, T.; Andersson, M.; Hultman, L.; Spetz, A. L.; Yakimova, R., Epitaxially grown graphene based gas sensors for ultra sensitive NO2 detection. Sensors and Actuators B: Chemical 2011, 155 (2), 451-455. 12. Reina, A.; Jia, X.; Ho, J.; Nezich, D.; Son, H.; Bulovic, V.; Dresselhaus, M. S.; Kong, J., Large Area, Few-Layer Graphene Films on Arbitrary Substrates by Chemical Vapor Deposition. Nano Letters 2008, 9 (1), 30-35. 13. Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S., Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils. Science 2009, 324 (5932), 1312-1314. 14. Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H., Large-scale pattern growth of graphene films for stretchable transparent electrodes. Nature 2009, 457 (7230), 706-710. 15. Ferrari, A. C.; Robertson, J., Interpretation of Raman spectra of disordered and amorphous carbon. Physical Review B 2000, 61 (20), 14095-14107. 16. Gillen, R.; Mohr, M.; Thomsen, C.; Maultzsch, J., Vibrational properties of graphene nanoribbons by first-principles calculations. Physical Review B 2009, 80 (15), 155418. 17. Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K., Raman Spectrum of Graphene and Graphene Layers. Physical Review Letters 2006, 97 (18), 187401. 18. Banhart, F.; Kotakoski, J.; Krasheninnikov, A. V., Structural Defects in Graphene. ACS Nano 2010, 5 (1), 26-41. 35 19. Duplock, E. J.; Scheffler, M.; Lindan, P. J. D., Hallmark of Perfect Graphene. Physical Review Letters 2004, 92 (22), 225502. 20. Yinxiao, Y.; Murali, R., Impact of Size Effect on Graphene Nanoribbon Transport. Electron Device Letters, IEEE 31 (3), 237-239. 21. Raza, H.; Kan, E. C., Armchair graphene nanoribbons: Electronic structure and electric-field modulation. Physical Review B 2008, 77 (24), 245434. 22. Hwang, Y.; Engel-Herbert, R.; Rudawski, N. G.; Stemmer, S., Effect of postdeposition anneals on the Fermi level response of HfO2/In0.53Ga0.47As gate stacks. Journal of Applied Physics 108 (3), -. 23. Lin, H. C.; Brammertz, G.; Martens, K.; de Valicourt, G.; Negre, L.; Wang, W.-E.; Tsai, W.; Meuris, M.; Heyns, M., The Fermi-level efficiency method and its applications on high interface trap density oxide-semiconductor interfaces. Applied Physics Letters 2009, 94 (15), -. 24. Han, M. Y.; Ã–zyilmaz, B.; Zhang, Y.; Kim, P., Energy Band-Gap Engineering of Graphene Nanoribbons. Physical Review Letters 2007, 98 (20), 206805. 25. Bao, W.; Liu, G.; Zhao, Z.; Zhang, H.; Yan, D.; Deshpande, A.; LeRoy, B.; Lau, C., Lithography-free fabrication of high quality substrate-supported and freestanding graphene devices. Nano Research 3 (2), 98-102. 26. Son, Y.-W.; Cohen, M. L.; Louie, S. G., Energy Gaps in Graphene Nanoribbons. Physical Review Letters 2006, 97 (21), 216803. 27. Huang, B.; Li, Z.; Liu, Z.; Zhou, G.; Hao, S.; Wu, J.; Gu, B.-L.; Duan, W., Adsorption of Gas Molecules on Graphene Nanoribbons and Its Implication for Nanoscale Molecule Sensor. The Journal of Physical Chemistry C 2008, 112 (35), 13442-13446. 28. ACGIH, Threshold Limit Values (TLVs) and Biological Exposure Indices (BEIs) - Nitrogen Dioxide. 2012. 29. Radovic, L. R.; Bockrath, B., On the Chemical Nature of Graphene Edges:â€‰ Origin of Stability and Potential for Magnetism in Carbon Materials. Journal of the American Chemical Society 2005, 127 (16), 5917-5927. 36 3. Deposition, Characterization, and Thin-Film-Based Chemical Sensing of Ultra-Long Chemically Synthesized Graphene Nanoribbons 3.1 Introduction: Theoretical predictions of quantum confinement in a graphene lattice have driven lots of research to verify that such confinement can be experimentally realized to create an energy bandgap in graphene. 1, 2 The creation of a bandgap in graphene is essential for the utilization of graphene in digital integrated circuits so that switching off the channel can be realized. Recently, lots of efforts were put into generating GNRs with various widths and lengths using lithographical 3-5 , chemical 6-8 and various other techniques 9-17 . However, patterning graphene using top-down approaches create GNRs with rough edges which can degrade the carrier transport. 18 Accordingly, the inability to create GNRs with specific edge structures results in significant sample-to-sample variations and disagreement with theoretical predictions on electronic properties. 1 In order to overcome such problems, various bottom-up chemical methods were developed to control the width and edge structure of such GNRs with high uniformity. 19-27 These methods allow very controllable tuning of the electronic edge structure of GNRs with the possibility of large scale synthesis which is desirable for future electronic and sensing applications of GNRs. Moreover, the latest advancement in chemical synthesis allowed the creation of ultra-long (> 200 nm) GNRs which are even processable in liquid-phase, thus facilitating the utilization of such GNRs in various applications. 28 Nevertheless, some challenges such as GNR deposition (i.e. both isolated individual GNRs and thin-films), characterization and electronic device measurements are not fully explored. 37 In this section we report optimized conditions for deposition and visualization of individual bottom-up chemically synthesized GNRs and films on Si/SiO 2 substrates, and their applications as devices and chemical sensors. This is in contrast to the previous report, where deposition of GNRs was performed only on conductive substrates, 28 and could not be directly applied for device studies. Other works, reporting electronic investigation of individual GNRs, employed top-down fabricated GNRs, 6–8 which might have significantly been affected by nonuniform width and undefined edge structures. We have studied different annealing conditions and their effect on GNR thin-film devices using attenuated-total-reflectance Fourier-transform infrared spectroscopy (ATR-FTIR), electrical measurements and Raman spectroscopy. Additionally, we have demonstrated an application of the GNR thin-film devices by measuring the sensitivity of such GNRs to NO 2 gas and observed ppb level sensitivity, highlighting the potential for using chemically synthesized GNRs for cost-conscious and scalable sensing applications. 3.2 Results and Discussion: Figure 1a shows the structure of the GNRs used in this study which were synthesized using recently reported methods. 23, 28 These GNRs have cove-type edge structures with width dimensions of 1.1 and 0.7 nm (Figure 1a). GNR 1a and 1b bear dodecyl and tert-butyl (t-Bu) groups, respectively, at the edges, which contribute to their enhanced processability. Additionally, based on the weight-average molecular weight of corresponding polyphenylene precursors, the average length of such GNRs can be estimated to be ~ 500 and ~390 nm for GNR 1a and 1b, respectively. 23, 28 Dispersions of GNRs in solvents are usually prepared by sonicating GNR powder in the solvent, followed by centrifugation (See Methods). Figure 2 shows the ultraviolet-visible (UV-vis) absorption spectrum of GNR 1a and 1b dissolved in various 38 solvents. Moreover, the optical bandgap of the GNRs can be extrapolated from Figure 2 which yields a value of ~1.84 eV comparable to theoretical predictions and other reports. 28, 29 Figure 1. a) Chemical structure of GNRs 1a and 1b. b) AFM tapping mode height image revealing synthesized GNR length > 500 nm. Inset showing a height profile of the GNR (at the dashed line) revealing a height of 0.785 nm. c) Raman spectrum of an individual GNR revealing D, G, 2D and D+D’ bands. Inset shows an individual GNR between two Ti/Pd electrodes. d) Current vs. drain voltage (I-V d ) characteristic of an individual GNR device after metal angle deposition with a channel length of ~ 20 nm. Inset showing an SEM image of a 20 nm gap between Ti/Pd and angle-deposited Pd. -1.0 -0.5 0.0 0.5 1.0 -30 -15 0 15 30 Drain Current (nA) Drain Voltage (V) [ [ n R R R 0 18 0 1 Height (nm) x (nm) 0.785 nm 100 nm a b c d R R R 1a: R = C 12 H 25 1b: R = t-Bu 0.7nm 1.1 nm 250 nm 1500 2000 2500 3000 Intensity (a.u.) Raman Shift (cm-1) D G 2D D+D’ 250 nm Ti/Pd Ti/Pd Ti/Pd Ti/Pd Pd 39 Figure 2. ultraviolet-visible (UV-vis) absorption spectrum of GNR 1a dissolved in 1-cyclohexyl- 2-pyrrolidone (CHP), chlorobenzene (CB) and tetrahydrofuran (THF). Notably, GNR 1b could be dispersed in deionized water with sodium dodecyl sulfate (SDS), where GNR 1a showed no dispersibility. Absorption peak position of ~540 - 580 nm was observed with the slight peak position variance attributed to different levels of GNRs aggregation in different solvents and the interaction of SDS with GNR. Moreover, the optical bandgap of the GNRs can be extrapolated from the x-intercept of the UV-vis spectrum in Figure 1b (dashed green line) which yields a value of ~1.84 eV. Because the GNRs are long and have extended aromatic structures, strong interaction between GNRs cause their aggregation in the solvent, which can be visually observed as partial precipitation in the dispersions in solvents such as chlorobenzene (CB) and tetrahydrofuran (THF), depending on the concentration and duration after preparation. In spite of the successful synthesis of GNR 1a and 1b, deposition and visualization of such individual GNR on various substrates has not been previously achieved. In order to image and characterize individual GNRs on a substrate of choice, aggregation of GNRs in the dispersion should be minimized while adhesion to the substrate must be promoted by the proper surface functionalization. Additionally, it is preferable to have a solvent with a high boiling point so that long incubation times are 400 600 800 GNR in CHP GNR in CB GNR in Water + SDS GNR in THF Absorbance (a.u.) Wavelength (nm) 40 possible. Based on those requirements, we functionalize the surface of a Si/SiO 2 wafer with dodecyltriethoxysilane while using 1-cyclohexyl-2-pyrrolidone (CHP) as a solvent for GNR incubation on the functionalized wafer for 24 hours (see Methods). CHP has a boiling point of ~ 290 ᴼ C in atmospheric pressure, and we have observed that GNR dispersions in CHP are very stable with little visible precipitation for the time frame of ten days, in sharp contrast to CB and THF. Figure 1b shows a typical atomic force microscope (AFM) image of a GNR with length > 500 nm and height of ~0.78 nm on the functionalized SiO 2 surface. Considering the error corresponding to AFM tip radius (i.e. ~2 nm, see Methods), the observed width of ~ 5.0 to 6.9 nm is consistent with the expected value for a single GNR 1a including the alkyl chains (~3.8 nm). Additionally, most GNRs observed by AFM were 300 to 500 nm long (Figure 3). In order to locate individual GNRs for recording a Raman spectrum, the measurement was done by patterning two Ti/Pd electrodes on an individual GNR (deposited using CHP as a solvent) using electron beam lithography (EBL) and subsequently focusing the laser on the located GNR (Figure 1c). The spectrum shows clear D, G, 2D and D+D’ peaks, which is consistent with GNR thin-films (Figure 5d) and previously reported spectrum measured on a powder sample of GNR 1a. 21, 28 Electrical measurements on an individual GNR with a channel length of ~ 100 nm (Figure 1c inset) revealed little current conduction in the GNR. Subsequently, we performed metal angle deposition to shorten the channel length to ~ 20 nm (Figure 1d inset) and observed current conduction in the GNR (Figure 1d) which confirms the conductivity of GNR 1a. 41 Figure 3. AFM images of typical individual GNRs on dodecyltriethoxysilane functionalized SiO 2 surface showing lengths of 300 to 500 nm. Figure 4 shows I-V d curves of an individual GNR device under gate biases of -50 V, 10 V and, 50 V. Very little gate dependence was observed, which prevented the determination of the electrical bandgap and the mobility. We attribute the absence of gate dependence in these devices to the screening of gate electric field by the closely spaced source and drain electrodes, as SiO 2 thickness is 300 nm while the channel length (or source-drain spacing) is ~ 20 nm. Fabrication of devices with much thinner dielectric is currently ongoing, and the results will be published in a separate paper. 100 nm 100 nm a b 42 Figure 4. I-V d curves of an individual GNR device at different gate biases. The weak P-type gate modulation is attributed to the electric field screening effect by the closely positioned source- drain. Moreover, any twists in the GNR structure will potentially influence the electrical properties of these GNRs as it will locally alter the energy bands and/or create insulating states at the twist location. Therefore, the fabrication of short channel devices (i.e. ~ 20 nm) is important to minimize the probability of twists or other structural defects within the channel. We have fabricated GNR thin-film devices by drop-casting the dispersion of GNR 1a in THF or CB as demonstrated in Figure 5a. First, 1 nm Ti/ 50 nm Au electrodes are patterned on top of P ++ Si/300 nm SiO 2 substrate (Figure 5 a, b). Then GNR dispersion (see Methods) is dropped on top of the substrate while heating the substrate to 120 ᴼ C to evaporate the solvent. Here, we used THF or CB as the solvent because they can vaporize rapidly and leave a GNR film behind, whilst CHP has a high boiling point and is suitable for incubation to get individual GNRs. Afterwards, the substrate is annealed in H 2 /Ar gases (see Methods). Figure 5c shows a scanning electron microscope (SEM) image of a GNR film between two Ti/Au electrodes for device study. Moreover, deposition of the GNRs was confirmed by Raman analysis (Figure 5d). 19, 21, 28 43 Figure 5. a) Scheme of GNR film device fabrication. b) Optical image of the fabricated devices. c) SEM image of a typical GNR film device with GNR films between two (falsely colored) Ti/Au electrodes. d) Raman spectrum of GNR films showing D, G, 2D and D+D’ bands in agreement with single GNR Raman spectrum. The annealing step after drop-casting the GNR is crucial to enhance the conductivity of GNR film devices. The motivation behind annealing the GNR films was to cut off the insulating alkyl chains from the GNR edges to reduce ribbon-to-ribbon junction resistance without affecting the basal plane. 30 In order to characterize the annealing effect on the GNR film, ATR- FTIR measurements were carried out on GNR films at different annealing temperatures (Figure 6 a, b and Figure 7). The spectra were normalized to the peak from conjugated C-C at ~1600 cm -1 . The clear decrease of the alkyl C-H peaks (i.e. ~1380, 1470 and 2850-2925 cm -1 ) relative to the conjugated C-C peak after annealing at 400 and 500 ᴼ C (Figure 6a and b) suggests the partial removal of alkyl chains from the GNR edges at such conditions. Additionally, the disappearance 1 μm Ti/Au Ti/Au GNR film 1500 2000 2500 3000 Intensity (a.u.) Raman Shift (cm -1 ) D G 2D D+D’ 200 μm Ti/Au Source Ti/Au Drain GNR in Organic Solvent 500 ºC in Ar/H 2 Source /Drain Pattern Drop-casting device in GNR dispersion Heat to evaporate solvent Anneal a b c d 44 of the peak at ~ 720 cm -1 , which is also a characteristic peak for alkyl chains, further confirms the removal of alkyl chains from GNR edges (Figure 7). 28 On the other hand, the peak at ~863 cm -1 , which corresponds to C-H bonds at the cove position of GNR, 28 is maintained, which suggests minimal damage to the GNR basal plane after thermal treatment. Moreover, Raman spectroscopy of GNR films after different annealing conditions was carried out to examine the effect of annealing on GNR quality, which revealed no apparent extra defects induced by the thermal treatment (Figure 8). Figure 6 c-d, show the current vs. drain voltage (I-V d ) and current vs. gate voltage (I-V gs ) of a GNR film device after annealing at 500 ᴼ C, respectively. Figure 6c inset shows I-V d for a thin-film device before (black curve) and after (red curve) annealing at 500 ᴼ C, which confirms that annealing can increase the conduction significantly. Although the GNR is semiconducting, the limited p-type current modulation may be attributed to an electric field screening effect in thick (i.e. 30-100 nm) GNR films and relatively high ribbon-to-ribbon junction resistance. 30, 31 Although the GNR is semiconducting, the limited p-type current modulation in the GNR films may be attributed to an electric field screening effect in thick (i.e. 30-100 nm) GNR films and relatively high ribbon-to-ribbon junction resistance. When the GNR film is much thicker than the electric field screening length (i.e. Debye length), the electric field cannot effectively modulate the charge carriers in the part of the channel that is farthest from the dielectric/channel interface and accordingly the device cannot be completely switched off. Moreover, poor contact between GNRs and/or twists along the length of GNRs can alter the energy band locally which makes the current modulation along the channel less effective and results in poor ON/OFF ratio. 45 Figure 6. a, b) ATR-FTIR normalized spectra of pristine, 400 and 500 ᴼ C annealed GNR films showing relative intensity decrease in peaks associated with alkyl C-H bonds for annealed samples. c) current vs. drain voltage (I-V d ) of a typical GNR film device under different gate voltages (V gs ). d) current vs. gate voltage (I-V gs ) of the same device under different drain voltages (V d ). Inset shows the I-V d of the device before and after annealing at 500 ᴼ C. Figure 7. ATR-FTIR spectra of pristine, 400 ᴼ C and 500 ᴼ C annealed GNR films showing the disappearance of peak at ~720 cm -1 associated with alkyl chains. The peak at ~862 cm -1 2600 2800 3000 3200 Absorbance (a.u.) Wavenumber (cm -1 ) Pristine 400 degree annealing 500 degree annealing 0 10 20 0 20 40 60 80 100 Drain Current (nA) Drain Voltage (V) V gs = -100 V to 100 V (50 V steps) 0 5 10 15 20 0 20 40 60 80 100 before Annealing After 500 C Annealing V gs = 0V Drain Current (nA) Drain Voltage (V) 1300 1400 1500 1600 Absorbance (a.u.) Wavenumber (cm -1 ) Pristine 400 degree annealing 500 degree annealing -100 -50 0 50 100 0 20 40 60 80 100 V d =20V V d =15V V d =10V V d =5V Drain Current (nA) Gate Voltage (V) a d Conjugated C-C Alkyl C-H b c Alkyl C-H 46 represents the C-H bond at the cove position and is unaffected by annealing. We hypothesize that the additional peak at ~850 cm -1 might correspond to similar C-H bond at the cove position of GNRs after removal of the alkyl chains or covalent bond formation with neighboring GNRs. Figure 8. a) Raman spectra of a GNR film before annealing and after sequential annealing at 300 ᴼ C, 400 ᴼ C, and 500 ᴼ C. b) I G /I D ratio (blue squares) and D-band full width half maximum (FWHM) of GNR films (red circles) vs. annealing temperatures. The I G /I D ratio even showed a slight increase at annealing temperatures of 400 and 500 ᴼ C compared to pristine GNR and GNR after annealing at 300 ᴼ C, which indicates that the annealing samples at conditions we used does not affect the GNR basal plane, as otherwise I G /I D would decrease due to defect formation. While we do not have a clear understanding of the cause of D-band broadening, we hypothesize that broadening of the D-band peak upon annealing might be derived from the removal of the alkyl chains and/or covalent bond formation between neighboring GNRs at the high temperatures. In order to demonstrate the applicability of GNR film devices, we used a GNR film device as a NO 2 gas sensor. High sensitivity to NO 2 is very crucial for human health as relatively low concentrations (i.e. 0.2 part per million (ppm)) can cause respiratory irritation. 32 For sensor measurements, we used Ar as a dilution gas for different NO 2 concentrations and exposed the device to different concentrations while monitoring the relative conductance (G/G 0 ) as a function 0 100 200 300 400 500 1.15 1.20 1.25 1.30 1.35 1.40 I G /I D Annealing Temperature ( o C) 66 88 110 D-Band FWHM (cm -1 ) 1000 1500 2000 Intensity (a.u.) Raman Shift (cm -1 ) Before Annealing 300 Cº Annealing 400 Cº Annealing 500 Cº Annealing D G a b 47 of time; where G is the absolute conductance of the device at any given time and G 0 is the initial conductance of the device before sensing. After exposing the device to a certain NO 2 concentration (green arrow), G/G 0 increases by electron extraction from NO 2 molecules (Figure 10a). Then, the sensing chamber is flushed with Ar to remove NO 2 from the chamber (red arrow) and recover G/G 0 before introducing the new concentration in the chamber. The conductance change of a device ∆G/G 0 is defined as: ∆G/G 0 = (G-G 0 )/G 0 and is considered a metric for device sensitivity for a certain concentration. The dependence of ∆G/G 0 on the NO 2 concentration is plotted in Figure 10b and c. The fitted Langmuir Isotherm indicates charge transfer as the sensing mechanism for GNR film devices. We note that the data point for 200 ppb NO 2 exposure is an outlier compared to the rest of the data, and the reason may be related to uncontrolled temporal perturbation to the device. Our GNR film sensors exhibit high sensitivities comparing favorably to other graphene-based NO 2 sensors 33-35 . For example, Novoselov et al. 33 and Kim et al. 35 reported ∆G/G 0 of ~ 4% and 0.7% for pristine graphene and ozone treated graphene devices, respectively, exposed to NO 2 concentration of 1 ppm. On the other hand, we observed ∆G/G 0 of ~5.6% for NO 2 concentration of 50 ppb. We attribute this enhanced NO 2 sensitivity to the semiconducting nature of chemically synthesized GNRs which allows higher level of current modulation via adsorption of NO 2 molecules than semimetal pristine graphene. Additionally, compared to graphene, edges of GNRs are more chemically active than pristine Sp 2 surface of graphene which might play an important role in the molecular adsorption and sensing mechanisms. In the future, it may be beneficial to study the sensing performance of GNRs of different width and length to understand the mechanisms behind the high sensitivity of GNR NO 2 sensors. 48 Figure 9 shows the I-V d curves of a device before annealing, and after annealing at 300 ᴼ C, 400 ᴼ C, and 500 ᴼ C. It can be clearly seen that annealing at higher temperatures led to higher conduction; however, we found that devices annealed at 300 ᴼ C and 400 ᴼ C showed unstable conduction and therefore were not used for sensing experiments. Figure 9. I-V d curves of a device before annealing, and after annealing at 300 ᴼ C, 400 ᴼ C, and 500 ᴼ C showing significant enhancement in conduction after annealing at 500 ᴼ C. 49 Figure 10. NO 2 gas sensing experiment using a GNR film device. a) Time domain normalized conductance (G/G 0 ) of the GNR film device whilst introducing different concentrations of NO 2 . Green arrows correspond to the device being exposed to a certain concentration of NO 2 while red arrows correspond to the device being flushed with Ar only. b) Measured conductance change (∆G/G 0 ) of the device in (a) as a function of NO 2 concentration (red squares) and the corresponding Langmuir isotherm curve fitting (black line). Langmuir isotherm fitted equation is written in bottom right corner. c) Inverse conductance change (G 0 /∆G) vs. inverse NO 2 concentration showing a linear relation that further confirms Langmuir isotherm fitting. 3.3 Conclusion: In summary, a method was developed to deposit individual GNRs on SiO 2 using CHP as a solvent and dodecyltriethoxysilane for surface functionalization. AFM of individual deposited GNRs revealed GNRs with a length of >500 nm and a thickness of ~ 0.78 nm. Additionally, Raman spectroscopy of individual deposited GNRs showed characteristic D, G, 2D, and D+D’ peaks and electrical measurements on an individual GNR confirmed the conductivity of such 0 200 400 600 800 0.98 1.00 1.02 1.04 1.06 1.08 1.10 1.12 400 ppb NO 2 G/G 0 Time (Second) 0 200 400 600 800 0.98 1.00 1.02 1.04 1.06 1.08 1.10 1.12 200 ppb NO 2 G/G 0 Time (Second) 0 200 400 600 800 0.98 1.00 1.02 1.04 1.06 1.08 1.10 1.12 100 ppb NO 2 G/G 0 Time (Second) 0 250 500 7501000 0.98 1.00 1.02 1.04 1.06 1.08 1.10 1.12 50 ppb NO 2 G/G 0 Time (Second) 0.005 0.010 0.015 0.020 10 12 14 16 18 20 Measurement Langmuir Isotherm Fitting G 0 / G 1/NO 2 Concentration (1/ppb) 0 100 200 300 400 0.00 0.02 0.04 0.06 0.08 0.10 Measurement Langmuir Isotherm Fitting G/G 0 NO 2 Concentration (ppb) Gas On Gas Off a b c 50 GNRs. Moreover, GNR film devices were fabricated and ATR-FTIR confirmed the partial removal of alkyl chains from the GNR edges upon annealing the devices at 400 and 500 ᴼ C which was further supported by the significant conductance increase of GNR film devices. Furthermore, we have demonstrated the exceptional NO 2 sensing performance of the GNR film device. The GNR film based sensor is sensitive to low NO 2 concentrations down to 50 ppb. 3.4 Methods: 3.4.1. Individual GNR deposition: The GNR dispersion is prepared by sonicating GNR powder in 1-cyclohexyl-2-pyrrolidone (CHP) for 30 min to 1 h in a glass vial. Then, the dispersion is centrifuged at 1500 rpm for 5 min. The SiO 2 surface is functionalized by incubating the substrate in (1:10) (dodecyltriethoxysilane : isopropyl alcohol (IPA)) solution for 10 mins and rinsing with IPA . The functionalized substrate is then covered completely with the CHP GNR dispersion and is incubated for 24 h. Afterwards, the wafer is rinsed with IPA and dried with a nitrogen gun. 3.4.2. GNR dispersion preparation (films): The dispersion is prepared by sonication of GNRs for 30-40 min in a low boiling point organic solvent (i.e. THF or CB) and is dropped on the heated substrate immediately after sonication to prevent aggregation of GNR in the solvent. 3.4.3. GNR annealing conditions: The sample is loaded into a furnace with H 2 and Ar flow rates of 15 and 250 sccm respectively. The pressure inside the furnace was set to 6 Torr and the furnace was heated to 500 ᴼ C for 2 h. 51 3.4.4. ATR-FTIR measurement: ATR-FTIR spectroscopy data was obtained using a Bruker Vertex 80v FTIR instrument with a PIKE MIRacle™ ATR unit equipped with a ZnSe crystal. Each sample was placed flush against the surface of the ZnSe crystal to obtain each spectrum. Data was obtained and analyzed using OPUS software with a scan from 500 to 4000 wavenumbers at 2cm -1 resolution at 32 scans per spectrum. Both the optical bench and sample chamber were under vacuum (2- 5 hPa) for each measurement to remove OH and CO stretching modes from water vapor and carbon dioxide in the atmosphere. 3.4.5. AFM measurement: We used AFM model: Bruker Dimension Icon Scanning Probe Microscope (SPM) with a PeakForce tapping mode. The tip used was Scan Asyst-Air by Bruker with a nominal tip radius of 2 nm and a maximum radius of 12 nm. 52 3.5 References: 1. Son, Y .-W.; Cohen, M. L.; Louie, S. G., Phys. Rev. Lett. 2006, 97, 216803. 2. Han, M. Y .; Ã–zyilmaz, B.; Zhang, Y .; Kim, P ., Phys. Rev. Lett. 2007, 98, 206805. 3. Liang, X.; Wi, S., ACS Nano 2012, 6, 9700-9710. 4. Abbas, A. N.; Liu, G.; Liu, B.; Zhang, L.; Liu, H.; Ohlberg, D.; Wu, W.; Zhou, C., ACS Nano 2014, 8, 1538–1546. 5. Son, J. G.; Son, M.; Moon, K.-J.; Lee, B. H.; Myoung, J.-M.; Strano, M. S.; Ham, M.-H.; Ross, C. A., Adv. Mater. 2013, 25, 4723-4728. 6. Wang, X.; Dai, H., Nat. Chem. 2010, 2, 661-665. 7. Wang, X.; Ouyang, Y .; Li, X.; Wang, H.; Guo, J.; Dai, H., Phys. Rev. Lett. 2008, 100, 206803. 8. Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H., Science 2008, 319, 1229-1232. 9. Bai, J.; Duan, X.; Huang, Y ., Nano. Lett. 2009, 9, 2083-2087. 10. Abramova, V.; Slesarev, A. S.; Tour, J. M., ACS Nano 2013, 7, 6894-6898. 11. Kato, T.; Hatakeyama, R., Nat. Nanotechnol. 2012, 7, 651-656. 12. Sokolov, A. N.; Yap, F. L.; Liu, N.; Kim, K.; Ci, L.; Johnson, O. B.; Wang, H.; Vosgueritchian, M.; Koh, A. L.; Chen, J.; Park, J.; Bao, Z., Nat. Commun. 2013, 4, No. 2402. 13. Pan, Z.; Liu, N.; Fu, L.; Liu, Z., J. Am. Chem. Soc. 2011, 133, 17578-17581. 14. Solís-Fernández, P .; Yoshida, K.; Ogawa, Y .; Tsuji, M.; Ago, H., Adv. Mater. 2013, 25, 6562-6568. 15. Wei, D.; Xie, L.; Lee, K. K.; Hu, Z.; Tan, S.; Chen, W.; Sow, C. H.; Chen, K.; Liu, Y .; Wee, A. T. S., Nat. Commun. 2013, 4, 1374. 16. Xie, L.; Wang, H.; Jin, C.; Wang, X.; Jiao, L.; Suenaga, K.; Dai, H., J. Am. Chem. Soc. 2011, 133, 10394-10397. 17. Jiao, L.; Zhang, L.; Wang, X.; Diankov, G.; Dai, H., Nature 2009, 458, 877-880. 18. Pan, M.; Girao, E. C.; Jia, X.; Bhaviripudi, S.; Li, Q.; Kong, J.; Meunier, V.; Dresselhaus, M. S., Nano. Lett. 2012, 12, 1928-1933. 19. Cai, J.; Ruffieux, P .; Jaafar, R.; Bieri, M.; Braun, T.; Blankenburg, S.; Muoth, M.; Seitsonen, A. P .; Saleh, M.; Feng, X. ; Müllen, K.; Fasel, R., Nature 2010, 466, 470-473. 20. Ruffieux, P .; Cai, J.; Plumb, N. C.; Patthey, L.; Prezzi, D.; Ferretti, A.; Molinari, E.; Feng, X.; Müllen, K.; Pignedoli, C. A.; Fasel, R., ACS Nano 2012, 6, 6930-6935. 21. Schwab, M. G.; Narita, A.; Hernandez, Y .; Balandina, T.; Mali, K. S.; De Feyter, S.; Feng, X.; Müllen, K., J. Am. Chem. Soc. 2012, 134, 18169-18172. 22. Koch, M.; Ample, F.; Joachim, C.; Grill, L., Nat. Nanotechnol. 2012, 7, 713-717. 23. Tan, Y .-Z.; Yang, B.; Parvez, K.; Narita, A.; Osella, S.; Beljonne, D.; Feng, X.; Müllen, K., Nat Commun. 2013, 4, No. 2646. 24. Jensen, S. r. A.; Ulbricht, R.; Narita, A.; Feng, X.; Müllen, K.; Hertel, T.; Turchinovich, D.; Bonn, M., Nano Lett. 2013, 13, 5925-5930. 25. Kim, K. T.; Jung, J. W.; Jo, W. H., Carbon. 2013, 63, 202-209. 26. Kim, K. T.; Lee, J. W.; Jo, W. H., Macromol. Chem. Phys. 2013, 214, 2768-2773. 27. Vo, T. H.; Shekhirev, M.; Kunkel, D. A.; Morton, M. D.; Berglund, E.; Kong, L.; Wilson, P . M.; Dowben, P . A.; Enders, A.; Sinitskii, A., Nat Commun. 2014, 5, No. 3189. 28. Narita, A.; Feng, X.; Hernandez, Y .; Jensen, S. r. A.; Bonn, M.; Yang, H.; Verzhbitskiy, I. A.; Casiraghi, C.; Hansen, M. R.; Koch, A. H. R. ; Fytas, G.; Ivasenko, O.; Li, B.; Mali, K. S.; Balandina, T.; Mahesh, S.; De Feyter, S.; Müllen, K., Nat. Chem. 2014, 6, 126-132. 29. Osella, S.; Narita, A.; Schwab, M. G.; Hernandez, Y .; Feng, X.; Müllen, K.; Beljonne, D., ACS Nano 2012, 6, 5539-5548. 53 30. Do, J.-W.; Estrada, D.; Xie, X.; Chang, N. N.; Mallek, J.; Girolami, G. S.; Rogers, J. A.; Pop, E.; Lyding, J. W., Nano Lett. 2013, 13, 5844-5850. 31. Sangwan, V. K.; Ortiz, R. P .; Alaboson, J. M. P .; Emery, J. D.; Bedzyk, M. J.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C., ACS Nano 2012, 6, 7480-7488. 32. ACGIH, Threshold Limit Values (TLVs) and Biological Exposure Indices (BEIs) - Nitrogen Dioxide. 2012. 33. Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P .; Katsnelson, M. I.; Novoselov, K. S., Nat. Mater. 2007, 6, 652-655. 34. Pearce, R.; Iakimov, T.; Andersson, M.; Hultman, L.; Spetz, A. L.; Yakimova, R., Sens. Actuators, B 2011, 155, 451-455. 35. Chung, M. G.; Kim, D. H.; Lee, H. M.; Kim, T.; Choi, J. H.; Seo, D. k.; Yoo, J.-B.; Hong, S.-H.; Kang, T. J.; Kim, Y . H., Sens. Actuators, B 2012, 166-167, 172-176. 54 4. Vapor-Phase Transport Deposition, Characterization, and Applications of Large Nanographenes 4.1 Introduction: Advances in organic chemistry have allowed the bottom-up synthesis of graphene nanostructures such as graphene nanoribbons (GNRs) with various lengths and edge structures 1-6 as well as large polycyclic aromatic hydrocarbons (PAHs), namely nanographenes. 7-19 This progress has provided a high degree of design and tunability to apply such chemically synthesized structures to diverse applications. Specifically, large nanographenes, which we define here as structurally defined PAHs possessing more than 90 sp 2 carbon atoms in the aromatic core, are highly promising with their extended aromatic cores, which lower their bandgaps and predicted to enhance their intrinsic charge-carrier mobilities. 13 The processing and deposition of the nanographenes into films on substrates in a reliable and controllable manner without destroying their structure is of extreme importance to the development of various applications utilizing such large molecules, and yet, proven to be challenging. So far, liquid- phase processing is the most commonly used method to prepare films of nanographenes. 1,4,5,20 Liquid-phase process starts with dispersing nanographene molecules in a solvent, followed by drop-casting, spin-coating, zone-casting 21 or incubating the dispersion on top of a target substrate. However, processing of large nanographenes in liquid phase is difficult due to their low solubility in solvents and their tendency to stack and form aggregates in the solution. 1,3,20,22 To the best of our knowledge, there is hitherto no report on field-effect transistors (FETs) of large nanographenes, except for several studies on chemically synthesized GNRs, 4,5,20 which are regarded as large nanographenes in our definition. In addition, liquid-phase processing of such GNRs has so far yielded FETs with rather low current on/off ratios only up to ~20. 4,5,20 Besides 55 liquid-phase processing, another possible approach would be vacuum sublimation. However, it has been perceived that thermal cracking prevents the deposition of large nanographene through sublimation. 22 We have therefore developed another deposition method of nanographenes, using soft-landing mass spectrometry, but this method cannot supply enough material for device fabrication. 22 Consequently, a reliable method to prepare films usable in electronic, optical, and sensing applications is highly desired for the fundamental study of the large nanographene molecules, and also for the progress and development of various applications. In this section, we report our finding that films of large nanographene molecules can be prepared using vacuum sublimation without destroying their aromatic core structures. Specifically, we developed a vapor-phase transport (VPT) approach, in which the large nanographene molecules are sublimated on a substrate in a vacuum-sealed glass tube. These films comprise of stacked large nanographene molecules maintaining their aromatic cores, without fusion or fragmentation. FETs based on such films exhibited current on/off ratios in the range of 140 – 170, which represents a significant progress compared to previously reported liquid-phase-processed large nanographene FETs. 4,5 We chose two kinds of large nanographene molecules. One is C 96 , 22-24 bearing 96 sp 2 carbons in the aromatic core. This molecule has been considered to be too large for vacuum sublimation. 22 Secondly, we chose a patch-like nanographene molecule (denoted graphene nanopatch (GNP)), which is even larger than C 96 and features a uniform width of ~2.1 nm and estimated lengths of approximately 1–10 nm. 3 The GNPs employed here are shorter analogues of GNRs reported in Ref. 3. We chose the GNPs instead of the GNRs, since sublimation of larger GNRs is more challenging. We used atomic force microscopy (AFM), matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), and Raman spectroscopy to characterize the film morphology, 56 to compare the pristine material with the sublimated films, and to ensure that the deposited large nanographenes in the films maintained their basal plane structures. Moreover, we showed that GNP and C 96 films can be used in the fabrication of thin-film transistors (TFT) with improved current on/off ratios by sublimating the molecules of choice onto substrates with prefabricated electrodes. Besides the observation of uniform films, we have also observed crystal-like island structures upon sublimation of GNPs, which have shown even higher electrical conductance than the films. Our VPT approach to prepare pristine nanographene molecular films suggest great promise for the future implementation of large nanograhpenes in electronic, optoelectronic and sensing applications. 57 4.2 Results and Discussion: 58 Figure 1. Molecule structure, vapor-phase transport deposition, and characterization of the nanographene molecules . a) Molecule structure of graphene nanopatch (GNP) (dangling bonds at edges indicate longitudinal repeat of such structure). b) Molecular structure of C 96 molecules used in this study. Carbon and hydrogen atoms are represented in grey and white respectively. c- e) Scheme and image of GNP film preparation setup used. GNP molecules were loaded into a glass tube along with the target substrate and subsequently vacuum sealed using a mechanical pump and a flame. Then, the sealed glass tube was loaded into a furnace where it was heated to 515 ᴼC for 50 minutes. Finally, the glass tube was opened and various electrical and optical characterization were applied to the sample. e) A vacuum-sealed glass tube loaded with GNP/C 96 and a substrate with metal electrodes. f, g) Digital camera images of a P ++ Si/ 300 nm SiO 2 substrate with Ti/Au electrodes before and after sublimation of GNP. AFM images of (h) GNP and (i) C 96 deposited on Si/SiO 2 substrate showing a flat and uniform films with root-mean- square roughness ~1 nm. j) AFM amplitude image showing a GNP island with multiple step edges. Inset shows an AFM height profile revealing a height of ~ 123 nm. Fig 1 a and b illustrate the molecular structures of GNP and C 96 , used in this study, respectively. Since GNPs are made through polymerization of tailor-made monomers to form corresponding polyphenylene precursors, followed by cyclodehydrogenation reaction, the GNPs, as depicted in Fig 1a, possess a certain distribution of lengths. 3 By tuning the polymerization conditions, the lengths of the GNPs can be controlled. The film deposition process started with the loading of the desired amount of nanographene molecules and the target substrate in a glass tube (Fig 1c). Then, the glass tube was connected to a vacuum pump for ~ 120 sec and subsequently flame-sealed to keep the inside chamber in vacuum (Fig 1d). Afterwards, the sealed glass tube was loaded into a furnace and was kept at 515 ᴼC for 50 mins, where the sublimation of GNP/C 96 took place (Fig 1e). Subsequently, the furnace was slowly cooled down to room temperature in a period of ~ 2 hours. By experimentally varying temperatures between 450 and 650 ᴼC, 515 ᴼC was consequently chosen as the lowest temperature that allowed formation of optically visible films. Finally, the glass tube is opened and the films are characterized. Fig 1f and g show a P ++ Si/ 300 nm SiO 2 substrate with Ti/Au electrodes before and after GNP sublimation, revealing a clear optical contrast. In addition, AFM of sublimated GNP films (Fig 59 1h) and sublimated C 96 films (Fig 1i) showed very flat and uniform films with a root-mean- square (RMS) roughness of ~ 1 nm. The uniformity and smooth morphology of films is important to pursue various applications utilizing large nanographene molecules, because the rough morphology deteriorates the electronic and optical properties of the films. Interestingly, we found that under certain conditions, macro scale GNP crystal-like structures were formed after sublimation. These crystal-like structures were found to partially cover some areas of the substrate. An AFM image of a GNP crystal-like island is shown in Fig 1j revealing a thickness of ~123 nm with clearly visible step edges indicating the layered nature of the GNP crystals. This study demonstrates that VPT techniques can form crystalline structures of large nanographenes, which has not been reported before. No crystals were observed on C 96 films under the sublimation conditions we tried. An important question regarding sublimation of such large nanographenes is whether sublimation cracks and breaks the molecules into smaller fragments. Maintaining the molecular basal plane is essential to preserve the intrinsic electronic and optical properties of the molecules. To investigate this issue, we used MALDI-TOF MS to analyze the sublimated films. MALDI- TOF MS analysis of a sublimated film C 96 showed four major peaks at masses ~ 1182, 1183, 1184, and 1185 (Fig. 2a and 3), which correspond to the isotopic distribution of sublimated C 96 molecules (i.e. C 96 H 30 ) and are consistent with the simulated results, as shown in Fig. 2a inset. This result proved that the sublimated C 96 films indeed maintained their chemical identity after sublimation onto films. On the other hand, GNPs have dodecyl chains at the peripheral positions to enhance their dispersibility in organic solvents. Fig. 2b shows the MALDI-TOF MS spectrum of the GNP molecules in its pristine powder form before sublimation. Multiple peaks can be observed with an average interval of 1000 gmol -1 , and the underlying reason is that the GNP 60 molecules have a certain distribution in the number of repeating primary units due to the synthesis process. 3 Fig 2c shows the MALDI-TOF MS spectrum of GNP molecules after treating the GNP powder inside the glass tube at 515 ᴼC. The MALDI-TOF MS spectrum showed a change in the average peak-to-peak mass interval from 1000 gmol -1 in the case of pristine GNP powder (Fig 2b) to 700-800 gmol -1 (Fig 2c) in the thermally treated GNP powder. The peak-to- peak interval corresponds to the mass of the primary repeating unit used as a building block to synthesize GNPs. 3 It is known that the energy required to remove dodecyl side chains is much lower than the energy to break bonds in the graphene basal plane. 25,26 We therefore believe the reduction in the mass of such unit at high temperature indicates the partial removal of dodecyl side chains, which is depicted in the molecular structure of the primary unit in the inset of Fig 2c. Importantly, we believe the partial removal of dodecyl side chains is beneficial for charge transport as the dodecyl side chains might significantly degrade the conductivity of GNP films due to their insulating nature. 20 61 Figure 2. Investigations of the chemical composition and structure of the sublimated films. a) Normalized MALDI-TOF MS spectrum of C 96 and the reference DCTB matrix used in the spectroscopy. Inset shows the peaks in the C 96 spectrum in agreement with the simulated C 96 MS, which confirms the preservation of the molecular structure of C 96 after sublimation into films. b) MALDI-TOF MS spectrum of the synthesized GNP powder with the corresponding structure of the repeating unit shown in the inset. c) MALDI-TOF MS spectrum of GNP powder after sublimation showing a reduced average mass interval, indicating the partial removal of alkyl chains at the edges. A suggested structure for the repeating unit after sublimation is shown in the inset (dashed black lines indicate peak positions). Figure 3. MALDI-TOF MS characterization of C 96 films deposited via VPT a) MALDI-TOF MS spectra of C 96 /DCTB (black) and DCTB matrix (red) for a wider mass range. b) Zoom in 500 1000 1500 2000 C 96/DCTB DCTB Intensity (a.u.) Mass (m/z) 500 1000 1500 2000 C 96 /DCTB Intensity (a.u.) Mass (m/z) a b C 96 62 image in the dashed area of (a), which shows a peak from C 96 (red arrow) that doesn’t appear in the spectrum of only DCTB, confirming the sublimation of undamaged C 96 . All the other signals can be assigned to the clusters of DCTB matrix (blue arrows) and their fragments. The signals at m/z = 250, 500, 750 1000, 1250, and 1500 are obviously the peaks of DCTB clusters, because the molecular weight of DCTB is 250 Da. The other peaks, especially in the mass range of 500– 750 Da and 750–1000 Da, could be attributed to the recombination of fragment ions with DCTB matrix clusters. One of the most promising applications for large nanographenes is TFTs. Band-like transport, which enhances the TFT device performance, was observed in highly ordered graphene-based monolayers. 17 The high tunability of the electronic properties of these molecules makes them attractive for various TFT applications. However, for some recently studied chemically synthesized GNRs, the liquid-phase process of TFTs yielded nondepletable (i.e. low current on/off ratio) transistors which may be attributed to poor control over the film thickness, uniformity, and roughness due to the aggregation of the large nanographene molecules in solution. 4,5,20 When the film in the channel area of a TFT has a varying thickness and/or roughness, the carrier densities in different areas of the channel would be modulated by the gate electric field differently, and thicker films may not be fully depleted of carriers by the gate electric field, thus leading to poor current on/off ratios. We have also tried preparing liquid- phase-processed GNP TFTs, which showed nonuniform film thickness, significantly large roughness (~ 10-250 nm RMS roughness), and nondepletable drain current-gate voltage (I d -V g ) characteristics with current on/off ratios ~ 2-3, consistent with previous reports (Fig 4). 4,5,20 63 Figure 4. Electrical measurements of liquid-phase dispersed GNP TFT. a) I d -V g characteristics of a liquid-phase dispersed GNP TFT on P ++ Si/ 300 nm SiO 2 with Ti/Au electrodes at different drain biases showing a non-depletable FET with a current on/off ratio of ~ 2-3. b) I d -V d characteristics of the liquid-phase processed GNP FET at different gate voltages. c) Optical image of a liquid-phase dispersed GNP TFT showing visible aggregates. d) AFM image of a liquid-phase dispersed GNP revealing large particles and RMS roughness of ~ 10 – 250 nm. In contrast, we observed that using sublimated GNP films with smooth morphologies significantly improved the current on/off ratio of GNP TFTs. We sublimated GNP films over prefabricated Ti/Au electrodes on P ++ Si/ 300 nm SiO 2 substrate for TFT device measurements. The sublimation of GNP films was verified by Raman spectroscopy, and the spectrum revealed similar G band and D band peaks with the same G/D intensity ratios (i.e. ~2) for the sublimated film and the liquid-phase-processed films (Fig 5a), further supporting the preservation of the aromatic core structure of GNPs. The difference in the Raman spectrum signal-to-noise ratios a b c d 5 µm 200 nm -200 nm 200 µm 64 between liquid-phase-processed films and sublimated films is because liquid-phase-based deposition led to thicker films and therefore better signal-to-noise ratio than sublimation-based deposition. We have analyzed the position and the full-width at half-maxima (FWHM) of the Raman peaks for both GNPs and C 96 (Table 1). First, it can be observed that the FWHM of G band (and D band) for liquid-phase-processed and sublimated GNPs are quite different while they are similar for liquid-phase-processed and sublimated C 96 . This could be due to the dodecyl side chain removal under high temperature in sublimated GNPs, which might induce stronger aggregations and affect the Raman peaks when compared to liquid-phase-processed GNPs, which are processed at room temperature and thus experience no side chain removal. 20 On the other hand, C 96 molecules do not possess any side chains, which make the Raman spectra of the sublimated and liquid-phase-processed samples chemically identical. Fig 5b shows the I d -V g characteristics of the sublimated GNP TFT. Unlike liquid-phase-processed GNP TFT, the sublimated GNP films with a thickness of ~2 nm (measured using an ellipsometer) and ~3-5 nm (measured using an AFM) (Fig 6), showed a reasonably depletable P-type TFT with a current on/off ratio of 1.74 х 10 2 (Fig 5b inset). The p-type behavior is consistent with other chemically synthesized large nanographene TFTs. 4,5,20 We believe the VPT deposition played a key role for the observation of improved current on/off ratios, as VPT allowed us to produce a thin and uniform film of GNP, which ensures that the gate electric field can deplete all carriers in the channel. Because of the strong aggregation in dispersions, the films deposited from the dispersion cannot avoid the high roughness. In addition, the packing of molecules in the films is also an important factor that affects electron transport, as electrons have to move from one molecule to another before reaching the electrode. Fig 5c shows the drain current – drain voltage (I d -V d ) characteristics of a GNP TFT. The nonlinearity of I d -V d curves indicates a significant 65 Schottky barrier for holes transporting from Au electrodes to GNP films. The relatively low current density (i.e. I ON /W= 0.2 nA/µm) for GNP films might be due to both the large Schottky barrier formed at the contacts and hole transport between molecules. Nevertheless, the demonstration of an operational GNP TFT (i.e. depletable I d -V g ) is an encouraging step towards the utilization of such molecules and also other large nanographenes in future electronic applications. We note that it would not be meaningful to calculate the mobility of such TFTs due to the existence of a large Schottky barrier which would significantly underestimate the intrinsic mobility of such films. We carried out TFT measurements on sublimated C 96 films as well. Raman spectroscopy of both liquid-phase-processed and sublimated C 96 films showed qualitatively similar spectra for both with G/D ratio of ~1 (Fig 5d). Unlike GNPs, Raman spectra of C 96 show 2D and D+D’ peaks. Fig 5e shows the I d -V g characteristics of the sublimated C 96 TFT devices. The devices based on C 96 films revealed a current on/off ratio of 1.4 х 10 2 , and I ON /W = 0.027 nA/µm. A significant Shottky barrier evident by highly nonlinear I d -V d curve is also observed in C 96 TFT devices (Fig 5f). 66 Figure 5. Raman spectra and electrical measurements of sublimated GNP/C 96 film TFTs on P ++ Si/ 300 nm SiO 2 substrates with Ti/Au electrodes. a) Raman spectra of GNP films prepared by liquid-phase processing and sublimation. The spectra show G and D bands characteristic of GNP as well as qualitatively similar intensity ratios confirming the sublimation of GNP films. Laser wavelength used was 532 nm. b) I d -V g characteristics of a GNP film TFT at different drain biases. Inset shows a logarithmic scale I d -V g curve of GNP film TFT showing a current on/off ratio of ~174. c) I d -V d characteristics of the GNP film TFT at different gate voltages showing highly non-linear curves indicating a Schottky contacted TFT. d) Raman spectra of C 96 films prepared by liquid-phase processing and sublimation. Both spectra show similar G, D, 2D, and D+D’ bands characteristic of C 96 as well as qualitatively similar G- and D-bands intensity ratios. e) I d -V g characteristics of the C 96 TFT at different drain biases. Inset shows a logarithmic scale I d -V g curve of the C 96 film TFT showing a current on/off ratio of ~140. f) I d -V d characteristics of the C 96 film TFT at different gate voltages showing non-linear curves indicating a Schottky contacted TFT. 1500 2000 2500 3000 Intensity (a.u.) Raman Shift (cm -1 ) Solution dispersed Sublimated -60 -40 -20 0 20 40 60 80 0.00 0.05 0.10 0.15 0.20 V d = 60 V V d = 30 V V d = 50 V V d = 20 V V d = 40 V V d = 10 V Drain Current ( A) Gate Voltage (V) -60 -30 0 30 60 90 1E-9 1E-8 1E-7 V d = 60 V Drain Current (A) Gate Voltage (V) 0 20 40 60 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 V g = -60 V V g = -40 V V g = -20 V V g = 0 V V g = 20 V V g = 40 V V g = 60 V V g = 80 V Drain Current ( A) Drain Voltage (V) a b c d G D 0 20 40 60 0.00 0.02 0.04 0.06 0.08 V g = -80 V V g = -40 V V g = 0 V V g = 40 V V g = 80 V Drain Current ( A) Drain Voltage (V) -90 -60 -30 0 30 60 90 0.000 0.005 0.010 0.015 0.020 0.025 0.030 V d = 60 V V d = 30 V V d = 50 V V d = 20 V V d = 40 V V d = 10 V Drain Current ( A) Gate Voltage (V) e 1500 2000 2500 3000 Intensity (a.u.) Raman Shift (cm -1 ) Solution dispersed Sublimated -90 -60 -30 0 30 60 90 1E-10 1E-9 1E-8 V d = 60 V Drain Current (A) Gate Voltage (V) G D f 2D D+D’ 67 Table 1. Analysis of Raman spectral peaks of liquid-phase-processed and sublimated GNPs/C 96 . In the above table, the full-width at half-maximum (FWHM), G and D band positions of different samples are listed. It can be seen that for GNPs, the FWHM is different for liquid- phase-processed and sublimated samples, while the FWHM is the same for liquid-phase- processed and sublimated C 96 . This is due to the removal of dodecyl side chains in the case of sublimated GNPs. Figure 6. AFM characterization of GNP thin-film thickness. a) AFM height image of a Si/SiO 2 substrate after GNP deposition scratched gently with a micro probe to expose the substrate and analyze the height profile. The three dashed lines correspond to the height profile taken at different positions. b) An optical image of the substrate in (a) revealing an optical contrast in the probe scratched area. Inset shows the three height profiles taken at positions marked in (a) with heights of ~ 5.3 nm (black), 4.8 nm (blue), and 2.9 nm (red). The prospect of using nanographenes in various applications relies heavily on the processing of these molecules into various forms (e.g. films) and the applicability of such forms. An important prospect of using GNPs is in the field of optoelectronics and photonics due to their 2 µm 10 nm -10 nm a b 0.0 0.2 0.4 0.6 0.8 1.0 -3 -2 -1 0 1 2 3 4 5 Height (nm) X ( m) 68 adjustable lengths and edge structures which directly modify the optical properties of this material. Accordingly, the optical characterization of the sublimated GNP films is very important for the development of novel optical devices utilizing these films. In order to characterize the intrinsic optical properties of GNP, photoluminescence (PL) measurements of GNP were carried out in a dispersion in 1,2-dichlorobenzene (DCB), where the interaction between dispersed GNPs is minimized. Fig 8a shows a typical PL spectrum of GNP dispersion under a 300 nm excitation wavelength. As it can be seen, the fluorescence (i.e. PL) of GNPs when exciting the dispersion with a 300 nm source reveal two peaks at ~350 nm and 666 nm while the absorption spectrum of such dispersion shows a peak at 300 nm and another broad peak at 410 nm. To investigate the radiative nature of absorption, we measured the excitation response of PL peaks at 350 nm and 666 nm. Fig 8b shows the fluorescence intensity at 666 nm (black) and 350 nm (red) as a function of excitation wavelength revealing a peak around ~ 299 nm where the absorption of the material mostly leads to a radiative deactivation. On the other hand, by observing both PL excitation and absorption spectra, we conclude that GNP absorption for wavelengths >350 nm, including the peak at 410 nm, mostly comprise of nonradiative deactivation processes. In order to assess the applicability of sublimated GNP films in optoelectronic and photonic applications, the fluorescence of films of thickness between ~ 2 to 5 nm must not be quenched by nonradiative scattering mechanisms induced by the substrate and interaction between GNPs. This PL quenching has been observed in other nanomaterials such as carbon nanotubes (CNTs). To study such issues, transient photoluminescence (TPL) measurements were carried out to compare the fluorescence lifetimes of GNP dispersions and GNP films. Due to the limited available choices for laser wavelengths, we chose a laser excitation wavelength of 400 nm and a PL emission at 450 nm (Fig 7). Fig 8c shows TPL 69 measurement of GNP dispersion and a GNP film for emission at 450 nm. Table 2 summarizes the TPL fitting parameters, including carrier lifetimes, for both films and dispersions. For GNP dispersion, the longest carrier lifetime is 9.1 ns and it represents the majority of the radiative population (i.e. 58 %). On the other hand, the longest carrier lifetime for GNP films is 4.3 ns and it represents only ~20% of the total radiative processes. The difference in the main radiative component carrier lifetime between the dispersion and the films is due to different GNP interaction mechanisms. The variation in GNP/GNP interaction, GNP/solvent interaction and GNP/substrate interaction in the case of GNP films compared to dispersions lead to a slight alteration of the electronic properties of GNPs and their carrier lifetimes. Moreover, the increase in the percentage of short lived species from GNP dispersion to films can be explained by surface polar scattering mechanisms due to the substrate effect and other scattering mechanisms induced by GNP/GNP junctions. Additionally, due to the inability to sublimate GNP species with the largest molecular weight, the distribution of species in the GNP dispersion and GNP films is different, and this lead to three different lifetimes in GNP dispersion compared to only two in the GNP film. Table 2. Carrier lifetimes for GNP films and dispersions extracted from fitting parameters using fractional intensities of positive decay components. Carrier lifetimes GNP dispersion GNP film τ 1 9.1 ns (57.62 %) 4.3 ns (20.19%) τ 2 2 ns (33.57%) 1 ns (79.81%) τ 3 0.3 ns (8.81 %) 70 Figure 7. Steady-state PL spectrum of GNP dispersed in DCB showing a PL peak around 450 nm when exciting the sample with a 400 nm source. 400 450 500 550 600 650 700 750 800 0 500 1000 1500 2000 2500 3000 400 nm excitation PL Intensity (a.u.) Wavelength (nm) 71 Figure 8. Investigations of the optical properties of the sublimated GNP dispersion and films. a) Steady-state PL spectrum of GNP dispersed in DCB (black) showing a clear PL peak around 350 nm and 666 nm when exciting the sample with a 300 nm source and the corresponding UV/vis absorption spectrum of GNP dispersed in DCB (red) showing peaks at ~300 nm and 410 nm. The x-axis is broken at the position of the peak of twice the excitation wavelength (i.e. 600 nm) a b c 0 20 40 60 80 GNP Dispersion GNP Dispersion (FIT) GNP Film GNP Film (FIT) IRF 450 nm Emission PL Counts (a.u.) Time (ns) 300 400 500 700 800 0.0 0.2 0.4 0.6 0.8 1.0 PL (300 nm excitation) Absorption Normalized Intensity (a.u.) Wavelength (nm) 250 300 350 400 450 500 550 600 650 700 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 PL @ 666 nm PL @ 350 nm PL Intensity (a.u.) Excitation Wavelength (nm) 72 b) PL excitation spectrum with emission fixed at 666 nm (black) and 350 nm (red) for GNP dispersed in DCB as a function of source wavelength. Maximum PL response corresponds to excitation at ~299 nm which agrees with the absorption peak indicating radiative deactivation processes at such wavelength. The x-axis is broken at the position of the peak of half the PL wavelength (i.e. 333 nm) c) Normalized transient PL measurements of emission at 450 nm for both GNP dispersed in DCB (red) and sublimated GNP film on P ++ Si/ 300 nm SiO 2 (black) when exciting the samples with a 400 nm laser. Moreover, the fittings for both GNP dispersion and film TPL measurements are shown in blue and green, respectively. Additionally, the instrument response function (IRF) is plotted in pink with 22ps time resolution. The average intensity-weighted carrier lifetimes for GNP solution and sublimated films are 5.913 ns and 1.636 ns respectively. Figure 9. Characterization of GNP islands. a) Optical image of a rectangular GNP island and b) the corresponding Raman mapping of G band intensity using a 457 nm laser. c) A typical Raman spectrum of the GNP island (black) in (a) showing G and D bands consistent with GNP films and Raman spectrum of a location away from the GNP crystal on the same substrate (red). d) A SEM image of a GNP island bridging two Ti/Au electrodes on a P ++ Si/ 56 nm SiO 2 . e) I d -V g characteristics of a GNP crystal TFT in (d) at different drain biases. f, g) I d -V d characteristics of the GNP crystal TFT in (d) at different gate voltages. At higher values of V d , the curves appear to be slightly nonlinear indicating Schottky contacts. 5 μm 5 μm 1200 1400 1600 1800 Intensity (a.u.) Raman Shift (cm -1 ) Not on Crystal On Crystal 457 nm Laser 500 nm -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.10 -0.05 0.00 0.05 0.10 0.15 V g = -20 V V g = -10 V V g = 0 V V g = 10 V V g = 20 V Drain Current ( A) Drain Voltage (V) -20 -10 0 10 20 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 V d = 0.1 V V d = 0.2 V V d = 0.3 V V d = 0.4 V V d = 0.5 V Drain Current ( A) Gate Voltage (V) 0.0 0.5 1.0 1.5 2.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 V g = -20 V V g = -10 V V g = 0 V V g = 10 V V g = 20 V Drain Current ( A) Drain Voltage (V) a b c d e f g G D 73 When the GNP concentration inside the tube is sufficient, we observed that many GNP crystal-like islands grew on top of the target substrate. Fig 9a shows an optical image of the crystal-like structure in Fig 1j. These crystal-like islands grew with various sizes and heights. We found that for sublimated materials with density under 0.3 mg/cm 3 , only films were formed and no islands would grow. When the sublimation starts, GNPs tend to form films covering the SiO 2 surface. When the SiO 2 surface is covered, GNPs start to stack up in three-dimensional (3D) structures forming the crystal-like islands. This is why the concentration of the molecule inside the chamber must be sufficient for such a process. In addition, we believe that the crystal-like islands appear to exhibit a screw-dislocation pattern, which is known as a mechanism for evolution of two-dimensional (2D) growth to 3D growth (Fig 1j). Raman mapping of the GNP crystal-like islands was carried out using a 457 nm laser, and the GNP island in the resultant image can be easily distinguished from the surrounding species (Fig 9b). The Raman spectrum of the crystal-like islands shows the characteristic G band and D band of GNP molecules with a G/D ratio of ~2.2 (Fig 9c), suggesting that these islands may have similar quality with pristine GNPs. In order to measure the electronic device performance of GNP islands, we fabricated pre- patterned substrates with electrodes defined using electron beam lithography (EBL) with channel lengths of 100 nm to 500 nm, and then GNP sublimation was carried out. Fig 9d and 10 show scanning electron microscope (SEM) images of some of the observed GNP islands bridging the EBL electrodes or attached to the metal electrodes. We have carried out electrical device measurements for GNP islands bridging metal electrodes. Fig 9e shows the I d -V g characteristics of GNP island in Fig 9d. As it can be seen, the device is nondepletable, which might be due to the thickness of the GNP island (i.e. >100 nm) causing a gate electric field screening effect. Nevertheless, the conductance of the device is greatly enhanced (I ON /W ~ 1500 nA/µm at V d =0.5 74 V) and an effective device mobility, using the device transconductance (see Methods) in the linear region and considering all crystal-like islands bridging the electrodes, is estimated to be ~ 1 cm 2 /V·s. Fig 9f and g show the I d -V d characteristics of GNP island devices in two different voltage ranges. The extracted mobility is underestimated because of both the Schottky barrier observed in the slightly nonlinear I d -V d curve (Fig 9 f, g) and also the electric field screening effect which may significantly affect the device mobility estimation as well. The growth and the measurement of GNP crystals might be an important route to create thinner crystalline GNP films for high-mobility TFTs. 27 Moreover, XRD was carried out for GNP islands to investigate their structural properties. XRD revealed a peak corresponding to a lattice distance of ~ 3.1 Å, whereas no obvious feature was observed for GNP flat films without islands (Fig 11b). This distinction between the XRD of GNP films and GNP islands might be either because thin GNP films (i.e. < 5 nm) are not totally crystalline due to the mismatch with the substrate or that diffraction from such thin films cannot be resolved by the XRD equipment used in the experiment (see Methods). 75 Figure 10. a- d) SEM images of GNP crystals formed on Ti/Au electrode. Figure 11. a) UV-Vis absorption spectrum of GNP dispersed in THF. The optical bandgap of the GNP can be extrapolated from the x-intercept yielding a value of ~1.75 eV. b) X-ray diffraction (XRD) spectrum of a sample with GNP crystals (red) and a flat GNP film (black) indicating an ordered assembly of GNP in the case of crystals. The XRD peak in the crystal sample corresponds to a crystal constant of ~3.1 A. 1 µm 1 µm 1 µm 1 µm a b c d 25 26 27 28 29 30 crystal film Intensity (a.u.) 2  (degree) ~3.1 A 400 600 800 Absorption (a.u.) Wavelength (nm) ~1.75 eV a b 76 4.3 Conclusion: In summary, we have developed a method to deposit GNP and C 96 films using VPT in a vacuum-sealed glass tube. Remarkably, the pristine aromatic core structure of such large nanographene molecules is preserved. AFM characterization of these films revealed smooth and flat deposited films with RMS roughness of ~ 1 nm which could not be achieved using liquid- phase processing techniques. Additionally, we have characterized the sublimated molecules using MALDI-TOF MS and confirmed that the basal plane structure of large nanographene, C 96 , was preserved after sublimation. This result overrides previous hypotheses and expectations that such large molecules will be subject to cracking and disintegration if thermally sublimated. Moreover, electrical device measurements of GNP and C 96 based TFTs showed the creation of a depletable and operational TFT as well as the possibility of achieving relatively high mobility in FETs based on GNP crystalline samples. The results presented in this work may influence the understanding of sublimation and crystallization of such large nanographenes and could lay the foundation for future use of nanographenes in electronic and optoelectronic applications. 4.4 Methods: 4.4.1. Synthesis of large nanographenes: Both C 96 and GNPs were synthesized following reported procedures. 3,23 For the preparation of GNPs, polyphenylene precursors with a weight-average molecular weight of 5200 g/mol, a number-average molecular weight of 3500 g/mol, and a polydispersity index of 1.5, based on size-exclusion chromatography analysis against a polystyrene standard, was used. 77 4.4.2. Sealed glass tube sublimation: The GNP/C 96 powder was loaded along with the target substrate in a ¼ inch glass tube. Then, a mechanical pump was connected to the open end of the glass tube, which was pumped for ~120 seconds. Afterwards, the glass tube was flamed at the side away from the powder and target substrate until that end was sealed. The tube was then loaded into a vacuum furnace and heated up to 515 ᴼC in 15 minutes, maintained at that temperature for ~ 50 minutes, and then cooled down with the furnace cover closed until it reached room temperature (~ 2 hours). 4.4.3. MALDI-TOF MS: Mass spectra of powder samples were recorded using a Bruker Reflex II utilizing a 337 nm nitrogen laser, calibrated against poly(ethylene glycol) (3,000 gmol –1 ), through solid-state sample preparation with tetracyanoquinodimethane (TCNQ) as matrix. 28 Mass spectrometry experiments on films were performed on a SYNAPT G2 Si instrument (Waters Corp., Manchester, UK) with matrix-assisted laser desorption/ionization (MALDI) source. For the analysis of the sublimated films, the sample wafer was fixed on the MALDI sample plate by a double side tape. DCTB ({(2E)-2-methyl-3-[4-(2-methyl-2-propanyl)phenyl]-2-propen-1-ylidene}malononitrile) matrix was subsequently sublimed on the surface of each wafer for the measurements. The mass range of 150~2000 Da was recorded with a mass resolution around 10000. 4.4.4. XRD: We used Rigaku Ultima IV powder/thin film diffractometer for the purpose of analyzing films and crystals. The sweep rate used was 0.04 deg/minute and the X-ray generator has a Cu target with a maximum power output of 3 kW. 4.4.5. TFT fabrication and measurement: Fabrication was done using regular photolithography techniques on a P ++ Si/ 300 nm SiO 2 substrate with Ti/Au electrodes. 78 Devices had a variable channel length of 3 – 8 µm and a channel width of 1mm. Device measurement was done on using an Agilent 4156B Semiconductor Analyzer. 4.4.6. UV/vis Absorption: The absorption spectrum was taken with a Perkin Elmer Lambda 950 UV/Vis/NIR spectrometer with an integrating sphere. 4.4.7. Raman Spectroscopy: Raman spectrum was taken using a Renishaw (inVia) Raman microscope with an objective lens of 100x. Laser wavelengths used were 532 nm and 457 nm. 4.4.8. Mobility calculation: The mobility of GNP crystal devices were calculated using μ = g m L/(C ox V ds W), where C ox is the capacitance of the gate oxide per unit area, L and W are the channel length and width, respectively, V ds is the drain bias, and g m is the peak transconductance calculated from the I d -V g slope. 4.4.9. TPL: Transient PL measurements were carried out using an excitation wavelength of 400 nm obtained from the second harmonic generation of the fundamental 800 nm laser (250 kHz Ti:sapphire amplifier, Coherent RegA 9050). The emission was collected at 450 nm using a R3809U-50 Hamamatsu PMT with a Becker and Hickl SPC 630 detection module (22 ps time resolution). 79 4.5. References: 1 Narita, A.; Feng, X.; Hernandez, Y.; Jensen, S. r. A.; Bonn, M.; Yang, H.; Verzhbitskiy, I. A.; Casiraghi, C.; Hansen, M. R.; Koch, A. H. R.; Fytas, G.; Ivasenko, O.; Li, B.; Mali, K. S.; Balandina, T.; Mahesh, S.; De Feyter, S.; Müllen, K. Nat. Chem. 2014, 6, 126-132. 2 Vo, T. H.; Shekhirev, M.; Kunkel, D. A.; Morton, M. D.; Berglund, E.; Kong, L.; Wilson, P. M.; Dowben, P. A.; Enders, A.; Sinitskii, A. Nat. Commun. 2014, 5, 3189. 3 El Gemayel, M.; Narita, A.; Dossel, L. F.; Sundaram, R. S.; Kiersnowski, A.; Pisula, W.; Hansen, M. R.; Ferrari, A. C.; Orgiu, E.; Feng, X.; Mullen, K.; Samori, P. Nanoscale 2014, 6, 6301-6314. 4 Kim, K.T.; Lee, J.W.; Jo, W.H. Macromol. Chem, Physic. 2013, 214, 2768-2773. 5 Kim, K.T.; Jung, J.W.; Jo, W.H. Carbon 2013, 63, 202-209. 6 Narita, A.; Verzhbitskiy, I. A.; Frederickx, W.; Mali, K. S.; Jensen, S. A.; Hansen, M. R.; Bonn, M.; De Feyter, S.; Casiraghi, C.; Feng, X.; Müllen, K. ACS Nano 2014, 8, 11622-11630. 7 Yamamoto, Y.; Zhang, G.; Jin, W.; Fukushima, T.; Ishii, N.; Saeki, A.; Seki, S.; Tagawa, S.; Minari, T.; Tsukagoshi, K.; Aida, T. P. Natl. Acad. Sci. 2009, 106, 21051–21056. 8 Xiao, S.; Tang, J.; Beetz, T.; Guo, X.; Tremblay, N.; Siegrist, T.; Zhu, Y.; Steigerwald, M.; Nuckolls, C. J. Am. Chem. Soc. 2006, 128, 10700-10701. 9 Lee, J.-H.; Choi, S.-M., Pate; B.D., Chisholm; M.H.; Han, Y.-S. J. Mater. Chem. 2006, 16, 2785- 2791. 10 Xiao, S.; Myers, M.; Miao, Q.; Sanaur, S.; Pang, K.; Steigerwald, M. L.; Nuckolls, C., Molecular Wires from Contorted Aromatic Compounds. Angew. Chem. 2005, 117, 7556-7560. 11 Pisula, W.; Feng, X.; Müllen, K. Chem. Mater. 2010, 23, 554-567. 12 Sergeyev, S.; Pisula, W.; Geerts, Y.H. Chem. Soc. Rev. 2007, 36, 1902-1929. 13 Wu, J.; Pisula, W.; Müllen, K. Chem. Rev. 2007, 107, 718-747. 14 Shklyarevskiy, I. O.; Jonkheijm, P.; Stutzmann, N.; Wasserberg, D.; Wondergem, H. J.; Christianen, P. C. M.; Schenning, A. P. H. J.; de Leeuw, D. M.; Tomović, Ž.; Wu, J.; Müllen, K.; Maan, J. C. J. Am. Chem. Soc. 2005, 127, 16233-16237. 15 Craats, A. M. v. d.; Warman, J. M.; Fechtenkötter, A.; Brand, J. D.; Harbison, M. A.; Müllen, K. Adv. Mater. 1999, 11, 1469-1472. 16 Tsao, H.N.; Räder, H.J.; Pisula, W.; Rouhanipour, A.; Müllen, K. Phys. status solidi (a) 2008, 205, 421-429. 17 Käfer, D.; Bashir, A.; Dou, X.; Witte, G.; Müllen, K.; Wöll, C. Adv. Mater. 2010, 22, 384-388. 18 Harris, K. D.; Xiao, S.; Lee, C. Y.; Strano, M. S.; Nuckolls, C.; Blanchet, G. B. J. Phys. Chem. C 2007, 111, 17947-17951. 19 Chen, L.; Hernandez, Y.; Feng, X.; Müllen, K. Angew. Chem. Int. Edit. 2012, 51, 7640-7654. 20 Abbas, A. N.; Liu, G.; Narita, A.; Orosco, M.; Feng, X.; Müllen, K.; Zhou, C. J. Am. Chem. Soc. 2014, 136, 7555-7558. 21 Pisula, W.; Menon, A.; Stepputat, M.; Lieberwirth, I.; Kolb, U.; Tracz, A.; Sirringhaus, H.; Pakula, T.; Müllen, K. Adv. Mater. 2005, 17, 684-689. 22 Rader, H. J.; Rouhanipour, A.; Talarico, A. M.; Palermo, V.; Samori, P.; Müllen, K. Nat Mater 2006, 5, 276-280. 23 Iyer, V.S.; Wehmeier, M.; Brand, J.D.; Keegstra, M.A.; Müllen, K. Angew. Chem. Int. Edit. 1997, 36, 1604-1607. 24 Tomović, Ž.; Watson, M.D.; Müllen, K. Angew. Chem. Int. Edit. 2004, 43, 755-758. 25 CRC Handbook of Chemistry and Physics, edited by David R. Lide (Taylor and Francis group, Boca Raton, FL, 2006), Vol. 87th ed. 26 Müller, M.; Kübel, C.; Müllen, K. Chem. Eur. J. 1998, 4, 2099-2109. 80 27 Yuan, Y.; Giri, G.; Ayzner, A. L.; Zoombelt, A. P.; Mannsfeld, S. C. B.; Chen, J.; Nordlund, D.; Toney, M. F.; Huang, J.; Bao, Z. Nat. Commun. 2014, 5, 3005. 28 Przybilla, L.; Brand, J.-D.; Yoshimura, K.; Räder, H.J.; Müllen, K. Anal. Chem. 2000, 72, 4591-4597. 81 5. Black Phosphorus Gas Sensors 5.1 Introduction: Recently, the rediscovery of black phosphorus (BP) 1-3 as a new single-element two- dimensional (2D) layered material has sparked the interest of scientists in various fields. Electronic and optical properties showed great promise for using BP in numerous applications. The field-effect transistor (FET) of few-layer BP exhibited high charge mobility, anisotropic transport behavior, high operating frequencies, and relatively high current on/off ratios, making BP a potential candidate for future electronics. 1-6 The recently reported device optimization techniques of BP FETs have yielded transistors with even better performance (e.g. higher mobility and lower contact resistance). 7-10 In Addition, optical applications including photovoltaics (PV), photodetectors, and imaging devices were created using BP FETs with different device structures. 9, 11-14 Moreover, passivation and stability of black phosphorus has also been studied. 15, 16 On the other hand, other applications such as chemical sensing of BP remain only theoretically explored. 17 Chemical sensing using various nanomaterials is one of the most promising applications, due to the inherent large surface-to-volume ratios. A variety of nanomaterials including carbon nanotubes 18, 19 , nanowires, 20, 21 and graphene 22, 23 were extensively studied for chemical and gas sensing applications. In the 2D family, both exfoliated and chemical-vapor-deposited (CVD) MoS 2 with various thicknesses were used for chemical sensing. The sensitivity of these sensors varied significantly depending on flake thickness, metal contacts, method of synthesis, and other factors. 24-28 For example, chemical sensitivity of MoS 2 FETs to nitrogen dioxide (NO 2 ) varied 82 from a few hundred parts per million (ppm) in exfoliated samples to a few parts per billion (ppb) in monolayer CVD samples. 26, 28 NO 2 is a common gas produced as a byproduct in industrial plants and vehicles. This gas is hazardous to humans and can cause a number of health problems. According to the U.S. Department of Environmental Protection Agency (EPA), exposure to NO 2 concentrations larger than 53 ppb can cause possible health problems. 29 Consequently, detection of this gas with sensitivities better than the aforementioned limit is of extreme importance. BP, being a 2D material, is predicted to be sensitive to various chemicals with comparable or better sensitivities than MoS 2 and graphene because the adsorption energies, of molecules such as NO 2 and NO are larger with BP than with graphene and MoS 2 . 17 Based on our knowledge; there is yet no systematic experimental verification of gas and/or chemical sensing of BP FETs. In this section, we investigated the chemical sensing performance of multilayer BP FET to NO 2 gas. We studied the stability of our BP sensors by Raman spectroscopy of flakes before and after sensing, which revealed no difference in the spectra, indicating the multilayer BP was stable for the timeframe, and repeated sensing we used. In our experiment, we exposed the BP FET to varying concentrations of NO 2 and monitored the relative conductance change in the device. The BP FET showed a systematic increase in conductance with varying concentrations, indicative of hole doping charge transfer caused by NO 2 molecules. The multilayer BP sensor exhibited a clear conductance change to NO 2 concentrations as low as 5 ppb comparing favorably with the performance of almost all other 2D sensors including monolayer MoS 2 . 26, 28 Moreover, the devices showed a good recovery to the original conductance after flushing the device with argon, 83 suggesting a reversible adsorption and desorption of NO 2 . The relative conductance change fitted fairly well with Langmuir Isotherm for molecular adsorption on a surface. This implies that NO 2 molecular adsorption via site binding and charge transfer are the sensing mechanisms for our BP devices. Additionally, we studied the adsorption and desorption rates of NO 2 molecules on the BP surface and derived the rate constants for various NO 2 concentrations. We also examined the drain current vs. drain voltage (I d -V d ) and drain current vs. gate voltage (I d -V g ) of the BP FET under varying concentrations of NO 2 , which showed a systematic increase in conductance and good consistency with Langmuir Isotherm. The results presented in this study may stimulate further study on the interaction between 2D materials and gas molecules, and may lead to various sensing applications. 84 5.2 Results and Discussion: Figure 1. Schematic, images, electronic properties and stability of multilayer BP FET used for chemical sensing. a) Scheme of a multilayer BP FET. b) An optical image of the multilayer BP flake between two Ti/Au electrodes used in this study. The BP flake is bordered by a dashed black line to guide the eye. c) I d -V d curves of the multilayer BP FET in (b) showing linear curves, which is a characteristic of an ohmic contacted device. Inset shows an I d -V g curve -40 -20 0 20 40 -10 -5 0 5 10 Vg = -60 V Vg = -30 V Vg = 0 V Vg = 30 V Vg = 60 V Drain Current ( A) Drain Voltage (mV) -60-40-20 0 20 40 60 4 5 6 7 8 9 10 Vd = 50 mV Drain Current ( A) Gate Voltage (V) 10 µm Ti/Au Source 300 nm SiO 2 P ++ Si Ti/Au Drain a b c d 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 Height (nm) X ( m) 3 µm 55 nm 0 500 1000 1500 2000 -0.4 -0.2 0.0 0.2 0.4 In Air G/G 0 Time (Second) 300 400 500 600 Intensity (a.u.) Raman Shift (cm -1 ) Pristine After NO 2 e f 85 measured at 50 mV V d with a nondepletable current. d) An AFM height profile of the BP flake revealing a height of 55 nm. Inset showing an AFM amplitude image of the multilayer BP FET in (b). e) a bias stress test of a BP multilayer flake in air showing a maximum fluctuation in relative conductance of less than 4%. f) Raman spectra of multilayer BP FET before (red) and after (black) exposure to 800 ppb NO 2 for 30 mins showing qualitatively identical spectra (A 1 g at ~362 cm -1 , A 2 g at ~466 cm -1 and B 2g at ~440 cm -1 ). Fig 1a shows a schematic of the BP device used in this study. First, chemically synthesized BP flakes (see Methods) were exfoliated, using a scotch tape, on a P ++ Si/300 nm SiO 2 substrate and subsequently patterned with contact metals (0.5 nm Ti/ 50 nm Au) as source and drain electrodes. In the back gate configuration, the P ++ Si acts as a back gate and the 300 nm SiO 2 is the dielectric. Fig 1b reveals an optical image showing the multilayer BP FET used for NO 2 sensing. The I d -V d curves of the device at different back gate voltages and an inset of an I d -V g curve are shown in Fig 1c. The linearity of I d -V d curves suggests Ohmic contacts between Au and multilayer BP. The nondepletable performance of the multilayer BP flake is due to the electric field screening effect in thick BP flakes. 1-3 Fig. 1d shows an atomic force microscope (AFM) image of the multilayer BP sensor and a height profile revealing a thickness of ~55 nm. The use of thick BP flakes is of extreme importance to the stability of the sensor and it was recently applied to other applications of BP such as imaging. 11 Generally, a stable performance over the sensing experiment is required for reliable sensing. Specifically, BP FETs, using relatively thin BP, have displayed degradation in performance under ambient conditions due to the oxidation of phosphorus. Recently, several solutions were developed to encapsulate and passivate BP FETs to maintain good performance under ambient conditions. 15, 30, 31 These methods are not applicable for sensing applications since direct exposure of the device active material to the chemical is required. Accordingly, we decided to use thick BP flakes for our sensing experiments to enhance the stability of the device and reduce degradation under 86 exposure to NO 2 . Although thinner BP flakes may theoretically offer better sensing performance because of the larger surface-to-volume ratio and larger bandgaps (i.e. reduced charge density), thinner flakes maybe more affected by oxidation than thicker flakes. Based on our experiments, we observed that continuous electrical measurements of thin BP flakes in air made the BP FET fail (Fig 2). On the other hand, a bias stress test of a thick BP flake in air showed relatively stable performance with a conductance variation less than 4% (Fig 1e). To isolate the effect of NO 2 on BP from other species such as oxygen and water vapor, we carried out the sensing experiments in an argon environment (i.e. NO 2 diluted in argon). Nonetheless, the relatively stable conductance value in air is promising for more practical future sensing applications. To further investigate the effect of NO 2 exposure on BP flakes we compared the Raman spectrum of BP flakes before and after exposure to NO 2 . Due to the anisotropic nature of BP, care was taken to keep the laser polarization, for a specific flake, in the same direction in all Raman measurements. Fig 1f shows a Raman spectroscopy taken on a multilayer BP flake before and after exposure to 800 ppb NO 2 for ~30 min. It can be seen that all peaks associated with BP (A 1 g at ~362 cm -1 , A 2 g at ~466 cm -1 , and B 2g at ~440 cm -1 ) remain in the same positions and show similar relative peak intensity ratios before and after exposure. This suggests minimal chemical degradation of multilayer BP flakes used in our experiments during NO 2 exposure. 87 Figure 2. Stability of thin BP flake FET. a) I d -V g curve of a ~10 nm flake BP FET under a 50 mV V d before failing. (b) Optical image of the device in (a) after repeated measurements in air showing a breaking point in the BP channel pointed by a red arrow. The NO 2 sensing experiment starts by loading the device in a gas chamber while flushing the system with argon for 10 mins. Afterwards, the NO 2 gas is diluted with argon to produce various concentrations and the BP device is subsequently exposed to the desired concentration. After the exposure to a certain NO 2 concentration, the system is flushed with argon for 300 seconds to partially recover the device and to observe a conductance change opposite to the NO 2 exposure period. Finally, when the device is exposed to all desired concentrations, the system is flushed with argon until the device recovers to the original conductance value. Fig. 3 shows the results from the above-described sensing experiment. In Fig. 3a, the relative conductance change ∆G/G 0 is plotted vs. time (where ∆G = G-G 0 , G is the instantaneous conductance of the device, and G 0 is the conductance of the device before exposure to NO 2 ). The inset of Fig 3a illustrates the point in time where the device is exposed to 5 ppb NO 2 concentration (i.e. ON) and when the device is flushed with argon (i.e. OFF). It can be clearly observed that the BP sensor responds to NO 2 concentrations down to 5 ppb evident by a conductance change of 2.9 %. The relatively high -60 -40 -20 0 20 40 60 0.0 0.5 1.0 1.5 2.0 2.5 Drain Current ( A) Gate Voltage (V) a b 88 sensitivity to NO 2 for multilayer BP is very interesting and may be further improved by reducing the layer number and increasing the surface-to-volume ratio. Additionally, it can be seen that the conductance change is monotonic and systematically increases as the concentration increases from 5 ppb to 40 ppb. After exposure to all NO 2 concentrations, the device recovers while being flushed with argon in a period of ~ 35 mins and can be used again for another round of sensing (Fig 3a). In Fig. 3b, the measured ∆G/G 0 is plotted vs. NO 2 gas concentration. ∆G/G 0 is extracted by taking the difference of the ON and OFF values (Fig. 3a inset) for each concentration. The data points in Fig. 3b fit with Langmuir Isotherm for molecules adsorbed on a surface with equation: ∆𝐺 𝐺 0 = 0.257 1+ 21.8 𝐶 (𝑝𝑝𝑏 ) (where C is the concentration in ppb). The fitting further confirms that charge transfer is the sensing mechanism for NO 2 sensing in the multilayer BP devices. Moreover, repeated sensing experiments revealed very similar sensing performances of the same device suggesting a stable performance over the timeframe of our experiment (Fig 4). Figure 3. NO 2 gas sensing performance of multilayer BP FET. a) Relative conductance change (∆G/G 0 ) vs. time in seconds for a multilayer BP sensor showing a sensitivity to NO 2 concentrations (5-40 ppb). Inset shows a zoomed in image of a 5 ppb NO 2 exposure response with identification of points in time where the NO 2 gas is switched on and off. b) ∆G/G 0 plotted vs. NO 2 concentration applied to the BP FET showing an agreement between the measured 0 1000 2000 3000 4000 5000 0.00 0.05 0.10 0.15 0.20 0.25 0.30 G/G 0 Time (Second) 5 ppb 10 ppb 40 ppb 20 ppb 0 250 500 750 0.00 0.04 G/G0 Time (Second) ON OFF a b 0 10 20 30 40 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 Measured Langmuir Isotherm G/G 0 NO 2 Concentration (ppb) 89 values (red squares) and the fitted Langmuir isotherm. The equation in the bottom right is the fitted Langmuir isotherm. Figure 4. Repeatability of BP sensor. (a) Relative conductance change (∆G/G 0 ) vs. time in seconds for a multilayer BP sensor for a first time sensing and b) For a second time showing similar response to various concentrations of NO 2 . We define the sensor response time as the time required to change the conductance after introducing either NO 2 (conductance increase) or argon (conductance decrease) in a specific range by 90%. Accordingly, we calculated the response time of our sensor to be in the range of ~280-350 seconds for different concentrations. This value is an indication of the rate the molecules are adsorbed on BP surface and it is comparable to other reports using other 2D materials, such as MoS 2 as gas sensors. 24, 26, 28 To further analyze the adsorption and desorption of NO 2 molecules on BP, we extracted the absorption/desorption rate constants (τ) of the multilayer BP device. The first order rate equation is of the form: 𝐺 = 𝐺 𝑓 + (𝐺 0 − 𝐺 𝑓 )𝑒 −𝑡 /𝜏 , Where G is the instantaneous conductance, G f is the final conductance after the end of an adsorption/desorption period, G 0 is the initial conductance before an adsorption/desorption period, and t is the time. 32 Fig. 5a, shows the measured conductance decrease associated with NO 2 desorption and the fitted curve when the BP FET is flushed with argon after being in a 20 0 2000 4000 6000 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 1st Time sensing G/G 0 Time (Second) 1 ppb 5 ppb 20 ppb 40 ppb 100 ppb 200 ppb 400 ppb 0 2000 4000 6000 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 2nd time sensing G/G 0 Time (Second) 1 ppb 5 ppb 20 ppb 40 ppb 100 ppb 200 ppb 400 ppb a b 90 ppb NO 2 environment. The measured and fitted curves show a nearly perfect agreement with a fitting error of ~ 1.5 %. This agreement suggests that there is only one time constant and therefore only one mechanism associated with NO 2 molecular adsorption/desorption in our experiment. Fig. 5b plots the adsorption and desorption rate constants (τ) vs. different concentrations of NO 2 . The extracted τ values varied from ~130 sec for a concentration of 5 ppb to ~840 sec for a concentration of 40 ppb. It can be observed that τ, which is a metric for how fast the NO 2 adsorption/desorption process is, decreases as the gas concentration increases and saturates at high concentrations. Figure 5. NO 2 gas sensing adsorption and desorption rate study. a) Measured time-dependent conductance change of a multilayer BP FET (black) flushed with argon after being exposed to 20 ppb NO 2 , showing a continuous decrease in conductance and the corresponding fitted curve (red) using the equation in the bottom of (a). b) Rate constant (τ) extracted from curve fitting vs. NO 2 concentration showing a saturation behavior for higher concentrations of NO 2 . The surprisingly high sensitivity of the thick (i.e. 55 nm) multilayer BP flake compared to other multilayer 2D materials such as MoS 2 is worth noting. 25, 26 For instance, He et al. observed a ~1% change in ∆G/G 0 when exposing an 18 nm MoS 2 flake to 1200 ppb NO 2 , while we observe a 2.9% change in ∆G/G 0 when exposing a 55 nm BP flake to 5 ppb NO 2 (240 times 0 100 200 300 146 148 150 152 154 156 158 160 Conductance ( S) Time (sec) 20 ppb Desporption (measured) Fitting a b 0 10 20 30 40 100 200 300 400 500 600 700 800 900 Desorption Adsorption (sec) NO 2 Concentration (ppb) 91 lower in concentration). 25 This staggering difference in response of multilayer samples is potentially material dependent. As we mentioned earlier, the high adsorption energies of NO 2 to BP is one reason causing the high sensitivity of our BP gas sensors. 17 Additionally, it was theoretically predicted and experimentally observed that BP layers have less out-of-plain conductance than other 2D materials such as graphene and MoS 2 . 33-35 This low conductance in the out-of-plane direction in BP compared to the in-plane conductance may explain the high sensitivity we observe. Since only the top most BP layer and edges are exposed to NO 2 during sensing, and since the metal contact to the multilayer BP flake is mostly to the top layer, the transport and doping of the top layers may dominantly control the conductance of the device because of the low conductance in the out-of-plane direction. To further explore the conductance of the out-of-plane direction in BP, we fabricated a vertical structure comprised of a monolayer graphene bottom contact to multilayer BP and a 0.5 nm Ti/ 50 nm Au top contact (Fig. 6). It can be observed that the ON current value in the vertical BP transistor structure (Fig. 6) is approximately two orders of magnitude lower than the traditional lateral transport FET structure (i.e. 10 µA for V d = 0.05V in lateral FET compared to 20 - 100 nA for V d = 0.1 V in vertical FET). Moreover, comparing to a vertical MoS 2 with flake thicknesses in the same range and similar device structure, BP vertical FET structure revealed a vertical current density of 0.5 - 2.4 A/cm 2 at V d = 0.2 V (Fig. 6), while MoS 2 at the same V d exhibited a current density of ~800 A/cm 2 . 36 This notably large difference further supports our explanation and hypothesis about the observed high sensitivity in multilayer BP sensors compared to its other 2D counterparts. 92 Figure 6. Vertical transport in multilayer BP flakes. a, c) Optical microscope image of a vertical FET comprised of a bottom monolayer CVD graphene electrode/BP/top (Ti/Au) electrode with a P ++ Si/ 300 nm SiO 2 back gated structure. Dashed line is a guide to the eye of the monolayer CVD graphene border. b, d) I d -V G curves for the devices in (a) and (c) respectively under V d =0.2 V. We further characterize the electronic properties of the multilayer BP FET via observing the change in I d -V d and I d -V g curves after exposing the device to different concentrations. First, the device was flushed with argon to clear the system from any contaminants or residual gas species. Then the device was exposed to a specific NO 2 concentration for 500 sec. Subsequently, -10 -5 0 5 10 15 0 50 100 150 200 250 300 V d =0.2 V Drain Current (nA) Gate Voltage (V) 5 µm Au Electrode graphene BP 3 х 3 µm Top Au Electrode 5 µm graphene BP 3 х 3 µm Top Au Electrode Au Electrode -10 -5 0 5 10 15 0 10 20 30 40 50 V d = 0.2 V Drain Current (nA) Gate Voltage (V) a b c d 93 measurements of I d -V d and I d -V g curves were recorded while the device is still exposed to NO 2 . Afterwards, the device was flushed with argon for 300 sec before exposing the device to a new concentration of NO 2 . Fig. 7a shows the I d -V g curves of the BP device under different concentrations. An upshift in the curves with increasing NO 2 concentrations associated with extra hole doping was observed. Fig. 7b plots the on current (I ON ) (defined as current at V g = -60 V) extracted from Fig 7a vs. the concentration of NO 2 the BP device was exposed to. The measured data points reveal a saturation behavior at higher concentrations and follow the Langmuir Isotherm with equation: 𝐼 𝑂𝑁 = 1.19×10 −5 1+ 0.836 𝐶 (𝑝𝑝𝑏 ) (A), which further supports that charge transfer is the main mechanism for BP FET sensors. The inset of Fig. 7b shows that 1/I ON vs. 1/C(ppb) have a linear relationship, which is another representation of the Langmuir Isotherm in Fig. 7b. I d -V d curves at different concentrations of NO 2 are plotted in Fig. 7c. As can be seen, the conductance increases monotonically with increasing NO 2 concentrations. Moreover, the I d -V d curves maintain their linearity with various NO 2 concentrations indicating minimal effect of Schottky barrier modulation induced by NO 2 exposure under the conditions we used in the experiment. Fig. 7d shows the conductance (G) of our BP sensor extracted from Fig. 7c vs. the concentration of NO 2 . The fitted Langmuir Isotherm equation is: 𝐺 = 1.6×10 −4 1+ 1.59 𝐶 (𝑝𝑝𝑏 ) (S), which agrees with the measured data points. similar to Fig. 7b, the inset of Fig. 7d shows a linear dependence of 1/G vs. 1/C(ppb) which agrees with the Langmuir Isotherm fitting. 94 Figure 7. Electronic properties of multilayer BP FET under different NO 2 gas concentrations. a) I d -V g curves of multilayer BP FET under different concentrations of NO 2 showing a clear upshift in the curves as the concentrations increase. b) Measured I ON vs. NO 2 concentration (red) and the corresponding fitted Langmuir isotherm (black) showing a good agreement. Inset shows 1/I ON vs.1/NO 2 concentration exhibiting a linear relation further illustrating Langmuir isotherm fitting. c) I d -V d curves of the same device under different NO 2 concentrations showing a monotonic increase in conductance with increasing NO 2 concentrations. d) Measured G vs. NO 2 concentration (red) and the corresponding fitted Langmuir isotherm (black) showing good agreement. Inset shows 1/G vs.1/NO 2 concentration exhibiting a linear relation. -60 -40 -20 0 20 40 60 4 5 6 7 8 9 10 11 12 Before NO2 5 ppb 10 ppb 20 ppb 40 ppb V d = 0.05 V Drain Current ( A) Gate Voltage (V) 0 10 20 30 40 50 0 2 4 6 8 Before NO2 5 ppb 10 ppb 20 ppb 40 ppb V g = 0 V Drain Current ( A) Drain Voltage (mV) a b 5 10 15 20 25 30 35 40 45 10.0 10.2 10.4 10.6 10.8 11.0 11.2 11.4 11.6 11.8 12.0 I ON ( A) NO 2 Concentration (ppb) 0 5 10 15 20 25 30 35 40 45 120 125 130 135 140 145 150 155 160 G( S) NO 2 Concentration (ppb) c d 0.05 0.10 0.15 0.20 84000 90000 96000 1/I ON (A -1 ) 1/NO 2 Concentration (ppb -1 ) Fitting Measured 0.05 0.10 0.15 0.20 6300 7200 8100 Fitting Measured 1/G (S -1 ) 1/NO 2 Concentration (ppb -1 ) 95 5.3 Conclusion: In summary, we experimentally demonstrated NO 2 gas sensing performance of multilayer BP FETs. The BP sensors were sensitive to NO 2 concentration down to 5 ppb making them comparable in sensitivity to the best 2D material based sensors. Raman spectroscopy comparison revealed no apparent change in the spectra before and after exposure to NO 2 , which shows that thick BP flakes can maintain their relative stability after sensing. Moreover, The BP device sensing performance fitted well with the Langmuir Isotherm for molecules adsorbed on a surface, which confirms charge transfer as the dominant mechanism for sensing. The systematic increase in conductance with increasing NO 2 concentrations suggests NO 2 molecules withdraw electrons and dope BP flakes with holes. These results lay the ground work for BP to be applied to various sensing applications including chemical, gas, and bio-sensors. 5.4 Methods: 5.4.1 BP synthesis: We synthesized BP samples from red phosphorus (Chempur, 99.999+ %) and tin/tin (IV) iodide (Sn/SnI 4 = 10/5 mg per 250 mg batch) in evacuated (p < 10 -3 mbar) silica ampoules according to literature procedures. 37 Subsequently, the temperature of the starting materials was raised to 650 °C in a period of 8 hours and that temperature was held for 5 hours. Then, the oven chamber was cooled down to 550 °C in a period of 7.5 hours and was kept at that temperature for 6 hours. Eventually, the mixture was cooled to room temperature. 5.4.2. BP device fabrication: BP flakes were exfoliated using a commercial tape on a P ++ Si/ 300 nm SiO 2 substrate with alignment marks patterned. After the flakes were located using optical microscopy, electron beam lithography (EBL) defined 96 electrodes were patterned on the target BP flake. Subsequently, electron beam evaporation of 0.5 nm Ti as an adhesion layer and then 50 nm thick Au layer for contacts were carried out. Then, the sample was soaked in acetone for ~30 mins to do metal lift-off. Afterwards, the devices were manually bonded using indium wire bonds after mounting the substrate on a chip carrier. Finally, devices were loaded inside the gas sensing chamber and measurements were recorded. 5.4.3. NO 2 Gas sensing: Gas sensing was carried out by exposing the BP FET device to NO 2 gas diluted in argon in a closed chamber. Concentrations of NO 2 were adjusted by changing the flow rates of both gases while keeping the total flow rate constant. For each curve, the device was exposed to the desired concentration for 500 sec and then flushed with argon for 300 sec. Similar procedure was followed to measure I d -V d and I d -V g for BP devices. 97 5.5. References: 1. Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D.; Chen, X. H.; Zhang, Y., Black phosphorus field-effect transistors. Nat. Nanotechnol. 2014, 9, 372-377. 2. Xia, F.; Wang, H.; Jia, Y., Rediscovering black phosphorus as an anisotropic layered material for optoelectronics and electronics. Nat. Commun. 2014, 5. 3. Liu, H.; Neal, A. T.; Zhu, Z.; Luo, Z.; Xu, X.; Tomànek, D.; Ye, P. D., Phosphorene: An Unexplored 2D Semiconductor with a High Hole Mobility. ACS Nano 2014, 8, 4033-4041. 4. Koenig, S. P.; Doganov, R. A.; Schmidt, H.; Castro Neto, A. H.; Özyilmaz, B., Electric field effect in ultrathin black phosphorus. Appl. Phys. Lett. 2014, 104, 103106. 5. Wang, H.; Wang, X.; Xia, F.; Wang, L.; Jiang, H.; Xia, Q.; Chin, M. L.; Dubey, M.; Han, S.-j., Black Phosphorus Radio-Frequency Transistors. Nano Lett. 2014, 14, 6424-6429. 6. Zhang, X.; Xie, H.; Liu, Z.; Tan, C.; Luo, Z.; Li, H.; Lin, J.; Sun, L.; Chen, W.; Xu, Z. et al. Black Phosphorus Quantum Dots. Angew. Chem., Int. Ed., DOI: 10.1002/anie.201409400. 7. Kamalakar, M. V.; Madhushankar, B. N.; Dankert, A.; Dash, S. P., Low Schottky Barrier Black Phosphorus Field-Effect Devices with Ferromagnetic Tunnel Contacts. Small, DOI:10.1002/smll.201402900. 8. Li, L.; Ye, G. J.; Tran, V.; Fei, R.; Chen, G.; Wang, H.; Wang, J.; Watanabe, K.; Taniguchi, T.; Yang, L. et al. Quantum Oscillations in Black Phosphorus Two-dimensional Electron Gas. arXiv 2014, arXiv:1411.6572. 9. Buscema, M.; Groenendijk, D. J.; Steele, G. A.; van der Zant, H. S. J.; Castellanos-Gomez, A., Photovoltaic effect in few-layer black phosphorus PN junctions defined by local electrostatic gating. Nat. Commun. 2014, 5. 10. Du, Y.; Liu, H.; Deng, Y.; Ye, P. D., Device Perspective for Black Phosphorus Field-Effect Transistors: Contact Resistance, Ambipolar Behavior, and Scaling. ACS Nano 2014, 8 (10), 10035-10042. 11. Engel, M.; Steiner, M.; Avouris, P., Black Phosphorus Photodetector for Multispectral, High- Resolution Imaging. Nano Lett. 2014, 14, 6414-6417. 12. Buscema, M.; Groenendijk, D. J.; Blanter, S. I.; Steele, G. A.; van der Zant, H. S. J.; Castellanos- Gomez, A., Fast and Broadband Photoresponse of Few-Layer Black Phosphorus Field-Effect Transistors. Nano Lett. 2014, 14, 3347-3352. 13. Youngblood, N.; Chen, C.; Koester, S. J.; Li, M., Waveguide-integrated black phosphorus photodetector with high responsivity and low dark current. Nat. Photonics 2015, doi:10.1038/nphoton.2015.23. 14. Yuan, H.; Liu, X.; Afshinmanesh, F.; Li, W.; Xu, G.; Sun, J.; Lian, B.; Ye, G.; Hikita, Y.; Shen, Z. et al. Broadband Linear-Dichroic Photodetector in a Black Phosphorus Vertical p-n Junction. arXiv 2014, arXiv:1409.4729. 15. 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., Effective Passivation of Exfoliated Black Phosphorus Transistors against Ambient Degradation. Nano Lett. 2014, 14, 6964-6970. 16. Ziletti, A.; Carvalho, A.; Campbell, D. K.; Coker, D. F.; Castro Neto, A. H., Oxygen Defects in Phosphorene. Phys. Rev. Lett. 2015, 114, 046801. 17. Kou, L.; Frauenheim, T.; Chen, C., Phosphorene as a Superior Gas Sensor: Selective Adsorption and Distinct I-V Response. J. Phys. Chem. Lett. 2014, 5, 2675-2681. 18. Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng, S.; Cho, K.; Dai, H., Nanotube Molecular Wires as Chemical Sensors. Science 2000, 287, 622-625. 98 19. Qi, P.; Vermesh, O.; Grecu, M.; Javey, A.; Wang, Q.; Dai, H.; Peng, S.; Cho, K. J., Toward Large Arrays of Multiplex Functionalized Carbon Nanotube Sensors for Highly Sensitive and Selective Molecular Detection. Nano Lett. 2003, 3, 347-351. 20. Zhang, D.; Liu, Z.; Li, C.; Tang, T.; Liu, X.; Han, S.; Lei, B.; Zhou, C., Detection of NO2 down to ppb Levels Using Individual and Multiple In2O3 Nanowire Devices. Nano Lett. 2004, 4, 1919-1924. 21. Zou, X.; Wang, J.; Liu, X.; Wang, C.; Jiang, Y.; Wang, Y.; Xiao, X.; Ho, J. C.; Li, J.; Jiang, C. et al. Rational Design of Sub-Parts per Million Specific Gas Sensors Array Based on Metal Nanoparticles Decorated Nanowire Enhancement-Mode Transistors. Nano Lett. 2013, 13, 3287-3292. 22. Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P.; Katsnelson, M. I.; Novoselov, K. S., Detection of individual gas molecules adsorbed on graphene. Nat. Mater. 2007, 6, 652-655. 23. Abbas, A. N.; Liu, G.; Liu, B.; Zhang, L.; Liu, H.; Ohlberg, D.; Wu, W.; Zhou, C., Patterning, Characterization, and Chemical Sensing Applications of Graphene Nanoribbon Arrays Down to 5 nm Using Helium Ion Beam Lithography. ACS Nano 2014, 8, 1538-1546. 24. Li, H.; Yin, Z.; He, Q.; Li, H.; Huang, X.; Lu, G.; Fam, D. W. H.; Tok, A. I. Y.; Zhang, Q.; Zhang, H., Fabrication of Single- and Multilayer MoS2 Film-Based Field-Effect Transistors for Sensing NO at Room Temperature. Small 2012, 8, 63-67. 25. He, Q.; Zeng, Z.; Yin, Z.; Li, H.; Wu, S.; Huang, X.; Zhang, H., Fabrication of Flexible MoS2 Thin- Film Transistor Arrays for Practical Gas-Sensing Applications. Small 2012, 8, 2994-2999. 26. Late, D. J.; Huang, Y.-K.; Liu, B.; Acharya, J.; Shirodkar, S. N.; Luo, J.; Yan, A.; Charles, D.; Waghmare, U. V.; Dravid, V. P. et al. Sensing Behavior of Atomically Thin-Layered MoS2 Transistors. ACS Nano 2013, 7, 4879-4891. 27. Perkins, F. K.; Friedman, A. L.; Cobas, E.; Campbell, P. M.; Jernigan, G. G.; Jonker, B. T., Chemical Vapor Sensing with Monolayer MoS2. Nano Lett. 2013, 13, 668-673. 28. Liu, B.; Chen, L.; Liu, G.; Abbas, A. N.; Fathi, M.; Zhou, C., High-Performance Chemical Sensing Using Schottky-Contacted Chemical Vapor Deposition Grown Monolayer MoS2 Transistors. ACS Nano 2014, 8, 5304-5314. 29. U.S. Department of Environmental Protection Agency. http://www.epa.gov/Air/Nitrogenoxides/. 30. Xi, L.; Rahbarihagh, Y.; Hwang, J. C. M.; Han, L.; Yuchen, D.; Ye, P. D., Temporal and Thermal Stability of Al2O3-Passivated Phosphorene MOSFETs. IEEE Electron Device Lett. 2014, 35, 1314-1316. 31. Doganov, R. A.; O'Farrell, E. C. T.; Koenig, S. P.; Yeo, Y.; Ziletti, A.; Carvalho, A.; Campbell, D. K.; Coker, D. F.; Watanabe, K.; Taniguchi, T. et al. Accessing the transport properties of pristine few-layer black phosphorus by van der Waals passivation in inert atmosphere. arXiv 2014, arXiv:1412.1274. 32. Zhang, Y.; Kolmakov, A.; Chretien, S.; Metiu, H.; Moskovits, M., Control of Catalytic Reactions at the Surface of a Metal Oxide Nanowire by Manipulating Electron Density Inside It. Nano Lett. 2004, 4, 403-407. 33. Low, T.; Rodin, A. S.; Carvalho, A.; Jiang, Y.; Wang, H.; Xia, F.; Castro Neto, A. H., Tunable optical properties of multilayer black phosphorus thin films. Phys. Rev. B 2014, 90, 075434. 34. Takahashi, T.; Gunasekara, N.; Ohsawa, H.; Ishii, H.; Kinoshita, T.; Suzuki, S.; Sagawa; T; Kato, H.; Miyahara, T.; Shirotani, I., Angle-resolved photoemission study of black phosphorus: Interlayer energy dispersion. Phys. Rev. B 1986, 33, 4324-4326. 35. Han, C. Q.; Yao, M. Y.; Bai, X. X.; Miao, L.; Zhu, F.; Guan, D. D.; Wang, S.; Gao, C. L.; Liu, C.; Qian, D. et al. Electronic structure of black phosphorus studied by angle-resolved photoemission spectroscopy. Phys. Rev. B 2014, 90, 085101. 36. Yu, W. J.; Li, Z.; Zhou, H.; Chen, Y.; Wang, Y.; Huang, Y.; Duan, X., Vertically stacked multi- heterostructures of layered materials for logic transistors and complementary inverters. Nat. Mater. 2012, 12, 246-252. 99 37. Köpf, M.; Eckstein, N.; Pfister, D.; Grotz, C.; Krüger, I.; Greiwe, M.; Hansen, T.; Kohlmann, H.; Nilges T., Access and in situ Growth of Phosphorene-Precursor Black Phosphorus, J. Cryst. Growth 2014, 405, 6-8. 100 6. Room Temperature Black Arsenic-Phosphorus Photodetectors at 5 µm 6.1 Introduction The recent rediscovery of semiconducting two-dimensional (2D) materials including various transition-metal dichalcogenides (TMDCs) 1, 2 , and black phosphorus (BP) 3-5 has led the way to various unique applications, including electronics and optoelectronics. Due to the direct bandgap of monolayer TMDCs and black phosphorus, and the tunability of their bandgaps with varying thicknesses, they were a natural choice for utilization in photonic and optoelectronic devices. Since most explored TMDCs have direct bandgaps only for monolayer samples 6 , most of the recent studies focused on monolayer TMDCs for applications of photodetectors and emitters except for very limited studies. 7 Additionally, the bandgap of most studied TMDCs have values larger than 1eV, which limits their optical applications to ultraviolet (UV), visible, and near infrared (NIR) parts of the spectrum. 6 Consequently, numerous studies on TMDC photodetectors based on photoconduction (i.e. photocurrent with an applied bias voltage) 7-10 or photovoltaic effect (i.e. photocurrent with a zero applied bias voltage) 1, 11, 12 were demonstrated for UV, visible and NIR spectral bands. Although impressive performances, in terms of photoresponsivity and detectivity, of TMDC photodetectors were reported, 7, 8 problems including slow response times, 8, 9 due to charge-trap states, 13, 14 significantly reduced photoresponsivity at large incident powers and longer wavelengths, 7, 8 and inapplicability of TMDC photodetectors in wavelengths longer than ~1240 nm were difficult to overcome. BP, which has a direct bandgap of 2.2 eV for monolayers and ~0.3 eV for bulk, has opened a route for 2D optoelectronic applications to extend to the short-wave infrared (SWIR) (0.9-2.5 µm), and part of the mid-wave infrared (MWIR) (3-5 μm). 15, 16 Various studies demonstrating photodetection in UV, visible, 17-20 NIR, 21 SWIR 22, 23 and MWIR (up to 3.39 μm) 24 were 101 presented in the past few years, emphasizing the potential for using BP in wideband optoelectronic applications. Unlike TMDCs, atomic defects and vacancies in BP do not affect the optical and electronic properties significantly. 25 As a result, faster photoresponse times are possible in BP. MWIR and LWIR (8-14 μm) are important bands for applications in imaging objects that emit thermal radiation. For example, blackbody radiation from objects with temperatures ranging from 300K to1000K will have a peak exitance at ~9.7 μm to 2.9 μm, respectively. Additionally, missile guidance systems and imaging systems that are tolerant to smoke, fog or dust are some of the other important applications of MWIR and LWIR bands. 16 The most commonly used material for photodetector applications in MWIR and LWIR is Hg 1-x Cd x Te (MCT). MCT is a popular choice due to its tunable composition, which controls the bandgap, and high absorption in MWIR and LWIR. However, low-temperature operation and the use of special equipment, such as molecular beam epitaxy (MBE), for growth are the main disadvantages of MCT photodetectors. Other traditionally used MWIR photodetectors include InSb photodetectors and InAsSb pyramidal arrays. 16 Recently, a room temperature BP photodetector at 3.39 μm was demonstrated, which opened new frontiers for 2D optoelectronic applications. 24 Nevertheless, an extension to longer wavelength in MWIR and LWIR spectral bands requires different materials due to the bandgap of BP (~0.3 eV). Based on our knowledge, other nanomaterials such as graphene and carbon nanotubes were used in NIR and SWIR photodetector application, but not MWIR and LWIR. 26-30 Recently, we have presented a method to alloy arsenic (As) and phosphorus (P) in a 2D crystal structure that is similar to BP (i.e. orthorhombic structure with a puckered honeycomb lattice). The resultant black arsenic-phosphorus (b-As x P 1-x ) alloy has a tunable bandgap down to 102 ~0.15 eV, corresponding with an As concentration of 83%. 31 With this new 2D layered material, optical absorption up to ~8260 nm is possible, which pushes the potential optoelectronic applications of such a material to cover MWIR band and reach LWIR band. In this study, we utilized the small bandgap of b-As 0.83 P 0.17 to make a room temperature MWIR photodetectors at a wavelength of 5 μm. This wavelength is beyond the detection limit of BP and is the longest wavelength detected by any 2D layered material. The b-As 0.83 P 0.17 photodetector showed a photoresponsivity of 4.5 mA/W at an incident beam power of 100 mW and an applied bias voltage of -1V. We have studied the photocurrent dependence on the applied gate voltage (V gs ) and bias voltage (V ds ). Accordingly, we confirmed that the photocurrent originates from photoexcited carriers over the small bandgap as well as the photogating effect. Moreover, due to the in-plane anisotropy in the crystal structure of b-As 0.83 P 0.17 , 31 a polarization dependent photocurrent was observed, and the photocurrent ratio of the two principle directions was ~ 1.66. This unique property adds an additional degree of freedom in designing novel optoelectronic devices using b-As x P 1-x . Furthermore, to examine the photodetector response time, we examined the photocurrecnt as a function of a 5μm laser chopping speed, and observed a small (~15.9%) drop in the photocurrent at a 10 kHz chopping speed compared to the unchopped signal. This result emphasizes the potential for using b-As x P 1-x photodetectors in high speed applications, such as space communication. Finally, we showed that the utilization of b-As 0.83 P 0.17 photodetectors can include the detection of shorter wavelengths in visible (532 nm) and NIR (785 nm) spectral ranges, and observed two different photocurrent mechanisms, photoconduction and photovoltaic effects. The results presented in this study, presents the first step for using b- As x P 1-x in high speed and room temperature optoelectronic applications in MWIR and potentially LWIR spectrua. 103 6.2 Results and Discussion: Figure 1a illustrates an optical extinction spectrum of b-As 0.83 P 0.17 , where T and T 0 are the optical transmission in b-As 0.83 P 0.17 and the background substrate, respectively. The absorption edge of the b-As 0.83 P 0.17 sample appears to be corresponding to a photon energy of ~0.16 eV (7750 nm). An optical image of the b-As 0.83 P 0.17 device used for photodetection is shown in figure 1b, where the b-As 0.83 P 0.17 flake thickness is ~22 nm. A thickness > 15 nm usually corresponds to the bulk properties of the material, where no change in the electronic and optical properties is observed by increasing the thickness of the flake. 31 The choice of a thick b- As 0.83 P 0.17 flake was made, because it was previously observed that thicker flakes are more stable under continuous illumination and suffer from negligible degradation. 22 The device was fabricated on a 285 nm SiO 2 / P ++ Si substrate to be used for back gating. The source/drain electrodes were 1/50 nm Cr/Au, where the Cr is used as an adhesion layer. It is noteworthy to mention that b-As 0.83 P 0.17 has the highest As content we can currently synthesize, and accordingly the smallest bandgap material we possess. Figure 1c shows the drain current vs. gate voltage characteristics (I d - V gs ) of the b-As 0.83 P 0.17 device in dark and illuminated with a 5μm laser of 100 mW power (i.e. corresponding to irradiance of 1.11×10 -4 W/μm 2 ). A clear photocurrent is observed when V gs < 0 and the device is hole-doped. As can be seen, the current on/off ratio for this b-As 0.83 P 0.17 device is low, which can be explained by the thick flake (22 nm) not being fully depleted. In figure 1d, the drain current vs. drain voltage (I d - V ds ) of the b- As 0.83 P 0.17 device is shown under dark and illuminated with a 5μm laser having a power of 100 mW power. A clear change in conductance is apparent, and a photocurrent value of 23.35 µA is observed at V ds = 1V and V gs = -10V. The linear I d - V ds curves indicate ohmic contacts between Au and As 0.83 P 0.17 . 104 Figure 1. Optical and electronic characteristics of b-As 0.83 P 0.17 . a) FTIR absorption spectrum of b-As 0.83 P 0.17 , showing an absorption edge corresponding to ~0.16 eV. The dashed blue line is a guide to the absorption edge of b-As 0.83 P 0.17 . b) Optical image and an AFM height profile (inset) of the BAsP FET photodetector, revealing a thickness of ~22 nm. The blue dashed line is the location where the height profile was taken. c) I d - V gs of the b-As 0.83 P 0.17 device in (b), in dark (black) and illuminated (red) with a 5μm laser (100 mW power and V ds =50 mV). d) I d - V ds of the b-As 0.83 P 0.17 device in (b), in dark (black) and illuminated (red) with a 5μm laser (100 mW power and V gs =- 10 V). In order to study the photocurrent mechanisms of the b-As 0.83 P 0.17 device, further investigation of the photocurrent dependence on V ds and V gs was undertaken. The photocurrent is equal to the difference between device current under illumination and the device dark current. Figure 2a is a photocurrent map of the b-As 0.83 P 0.17 device showing the change in the photocurrent with V ds and V gs . From the first observation, it is apparent that the photocurrent is 1000 2000 3000 4000 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Energy (eV) Extinction (1-T/T 0 ) Wavenumber (cm -1 ) B-As 0.83 P 0.17 0.1 0.2 0.3 0.4 0.5 5 µm 0.0 0.5 1.0 1.5 2.0 -10 0 10 22 nm Height (nm) X  m) -40 -20 0 20 40 1 2 3 4 5 Drain Current ( A) Gate Voltage (V) V ds =50mV = 5 m (100mW) Dark Illuminated 0.0 0.5 1.0 0 25 50 75 V gs = -10 V (illuminated) V gs = -10 V (dark) = 5 m (100 W) Drain Current ( A) Drain Voltage (V) a) c) b) d) 105 larger for V gs < 0, which is a consequence of the device being p-type (figure 1c) due to the Au metal contact aligning with the valence band of b-As 0.83 P 0.17 . Moreover, the absolute value of the photocurrent decreases as V gs changes from 0 to -50 V, and has a peak value at V gs = -10 V. This is due to the increase of the energy barrier for holes on the drain side at larger negative gate voltage, which increases the recombination probability and accordingly reduces the photocurrent. 24 Another important observation is that the photocurrent changes sign when 0<V gs <10 V and then changes sign again when 40<V gs <50 V. Guo et al. explained this as being a result of photogating, which is a result of trapping of photoexcited electrons in available trap states above the fermi energy. These negatively charged traps will yield a positive shift in the charge neutrality point (i.e. the point of minimum conductance in the I d -V gs curve, where a switch between hole and electron conduction occurs) as shown in figure 2c. The crossing between the dark and illuminated curves (figure 2c) is what causes the sign shift in the photocurrent for 0<V gs . The source of these mid-gap traps maybe from interfacial charge traps, and is likely not from intrinsic defects and dislocations. 25 In figure 2d, I d -V ds family curves under dark and illumination (5 μm wavelength and 100 mW) conditions with different applied V gs values are presented. Similar to figure 2b, it can be seen that the photocurrent decreases as the gate voltage is swept from -10 V to -60 V, confirming previous conclusions. In order to assess the photoresponsivity and photogenerated carrier loss mechanism, we plotted the photocurrent as a function of laser power (figure 2e). It can be seen that the photocurrent fits linearly with the laser power (I p (photocurrent)= 7.4 × 10 -8 × P(mW)), which means that the charge traps are mostly filled at these power levels, 24 and the carrier loss mechanism is through monomolecular recombination. 1 The photoresponsivity for all power levels is represented by the slope in figure 2e. The photoresponsivity reaches a peak value or 4.5 mA/W at V ds = -1 V. This value can be 106 higher as the photoresponsivity values for 2D materials are usually significantly higher for lower power levels due to the photogating effect through charge traps. 8, 24 Due to the in-plane anisotropic nature of b-As 0.83 P 0.17 , the charge masses and absorption coefficients are functions of in-plane orientation. 4, 31 As a results, the photocurrent is anisotropic with the incident polarization orientation angle, when the laser polarization is parallel to the b-As 0.83 P 0.17 plane. In figure 2f, the photocurrent is presented in a polar plot as a function of a 5μm laser incident polarization angle. The two principle directions are shown as 0 and 90 deg, and the photocurrent ratio of the two principle directions was ~ 1.66. The anisotropic nature of As 0.83 P 0.17 may be used in future polarization-selective applications. Figure 2. Detailed analysis of photocurrent in As 0.83 P 0.17 devices. a) Photocurrent map of the b- As 0.83 P 0.17 device vs. V gs and V ds . b) Photocurrent vs. V gs at V ds = -0.2, -0.5, -1 V. c) I d vs. V gs showing photogating effect and a shifted charge neutrality point under 5 μm illumination. d) I d - V ds family curves under dark and 5 μm illumination (100 mW). e) Measured photocurrent vs. laser power (black squares) and a linearly fitted curve (red). f) Polar plot of photocurrent vs. incident laser polarization angle showing anisotropic photocurrent. One of the important metrics of photodetectors is the speed, which is dictated by the response time of the photodetector. High operating speeds are important for applications of optical 0.0 0.5 1.0 0 50 100 Vgs= -50 V (iIIuminated) Vgs= -50 V (Dark) Vgs= -40 V (iIIuminated) Vgs= -40 V (Dark) Vgs= -30 V (iIIuminated) Vgs= -30 V (Dark) Vgs= -20 V (iIIuminated) Vgs= -20 V (Dark) Vgs= -10 V (iIIuminated) Vgs= -10 V (Dark) (100 mW) power Drain Current ( A) Drain Voltage (V) -40 -20 0 20 40 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Photocurrent ( A) Gate Voltage(V) Drain Voltage (V) -25.00 -12.50 0 12.50 25.00 0 10 20 30 40 50 1.6 1.8 2.0 Drain Current ( A) Gate Voltage (V) Vds=50mV = 5 m (100mW) Dark Illuminated -50-40-30-20-10 0 10 20 30 40 50 -25 -20 -15 -10 -5 0 5 10 15 20 25 Vds = -1 V Vds = -0.5 V Vds=-0.2 V Photocurrent ( A) Gate Voltage (V) a) c) b) d) 0 20 40 60 80 100 -1 0 1 2 3 4 5 6 7 8 9 10 Photocurrent ( A) Laser Power (mW) Vgs=-20V, Vds=0.5V I p =7.4 x 10 -8 x P (fit) 3 4 5 6 7 8 0 30 60 90 120 150 180 210 240 270 300 330 3 4 5 6 7 8 Photocurrent ( A) e) f) 107 communications, and high-speed imaging. For TMDC photodetectors, response times in the range of seconds down to a few milliseconds were reported, due to the slow trapping and detrapping processes. 8-10 For BP integrated with a waveguide, 3-dB cutoff frequencies up to 3 GHz were reported for 1.55 μm photodetection, 23 but a 3-dB cutoff frequency of only 2.2 kHz was reported for MWIR photodetection at 3.39 μm. 24 In order to measure the response time of our photodetectors, we used the scheme shown in figure 3a. The 5 μm laser is modulated with an optical chopper at a certain speed, and the modulated photocurrent from the b-As 0.83 P 0.17 device is measured using a lock-in amplifier. The resulting data is shown in figure 3b, which shows a small (15.9%) drop of the photocurrent from the unchopped signal compared to 10 kHz chopped signal. Since the maximum chopping speed we can experimentally reach using our setup is 10 kHz, it was not possible to determine the 3-dB cutoff frequency of our photodetector because it is more than 10 kHz. Nevertheless, this result is encouraging to further study the speed limit of b-As 0.83 P 0.17 MWIR photodetectors, which may be suitable for high speed applications, such as space communication. Figure 3. B-As 0.83 P 0.17 5μm Photodetector speed dependent response study. a) Schematic of the measurement setup used to measure the response time. b) Photocurrent vs. chopper speed showing a 15.9% photocurrent drop at 10 KHz relative to the unchopped photocurrent. 10 100 1000 10000 3 4 5 6 7 Vgs = -20V, Vds = 0.5V  = 5 m (100mW) Photocurrent ( A) Chopper Speed(Hz) Mirror Chopper controller B-As 0.83 P 0.17 device IR laser Lock-in amplifier a) b) 108 We wanted to assess the potential for using b-As 0.83 P 0.17 photodetector with shorter wavelengths. Figure 4a and 4b show a time domain b-As 0.83 P 0.17 device response to 785 nm (NIR) and 532 nm (visible green) lasers, respectively. The measurement reveals a repeatable response to both laser wavelengths under on and off conditions. Next, we use the small laser spot size of 532 nm laser (<1μm), to study the photocurrent mechanisms in b-As 0.83 P 0.17 device. Figure 4c shows I d -V ds curves of the b-As 0.83 P 0.17 device when it is illuminated with a 532 nm laser. The 532 nm laser positions are shown on the right of figure 4c, where the green laser spot on the device indicates the position of the laser during the measurement. For Pos1, the laser is positioned outside the active area of the device so that any second order effects, such as thermal effects, can be detected. It is clear that Pos1 and dark current curves are similar and no noticeable difference can be deduced. Next we move the laser spot closer to both metal electrodes, to observe the electric field effect close to the contacts. The electric field profile at the Au electrode/ b-As 0.83 P 0.17 interface is due to the work function difference between the two. Accordingly, the electric field direction will be opposite for the two electrodes. For Pos2 and Pos5, where the laser position is close to the left electrode, the I d -V ds curves move to the bottom due to the photovoltaic effect. A short circuit current and an open circuit voltage of 41.1 nA and 2 mV can be observed for Pos2 I d -V ds curve. On the other hand, Pos3 and Pos4 show I d -V ds curves that are shifted up, also due to the photovoltaic effect and the electric field forcing photoexcited charges in a direction opposite to the case of Pos2 and Pos5. For example, a short circuit current and an open circuit voltage of – 57.6 nA and – 3.07 mV can be observed for Pos4 I d -V ds curve. For Pos6, the laser spot has been optically adjusted to cover the whole active area of the device. As a result, the photovoltaic effect, from the left and right electrodes, cancels out, which yields a zero short-circuit current and a zero open-circuit voltage. Accordingly, we can 109 observe a photocurrent when |V ds |>0 V, which is the photoconduction mechanism. For the 5 μm photodetection, the beam size is approximately 30 × 30 μm, which is larger than the photodetector active area. Consequently, no photovoltaic effect is observed for the 5 μm b- As 0.83 P 0.17 photodetector. Figure 4. NIR and visible photodetection with b-As 0.83 P 0.17 devices. a) I d vs time for a b- As 0.83 P 0.17 device while a 785 nm laser is switched on and off. b) I d vs time for a b-As 0.83 P 0.17 1 2 3 4 5 6 7 8 9 10 0.33 0.34 0.35 0.36 0.37 0.38 0.39 0.40  = 785nm P = 10% Vds = 50 mV Drain Current ( A) Time (Second) 1 2 3 4 5 6 7 8 9 10 0.35 0.36 0.37 0.38 0.39 0.40 0.41 0.42  = 532 nm P = 0.1 % Vds = 50 mV Drain Current ( A) Time (Second) ON OFF ON OFF -3 -2 -1 0 1 2 3 -0.10 -0.05 0.00 0.05 0.10 Dark Pos1 Pos2 Pos3 Pos4 Pos5 Pos6  = 532 nm P= 1% Drain Current ( A) Drain Voltage (mV) Pos1 Pos2 Pos3 Pos4 Pos5 Pos6 a) b) c) 110 device while a 532 nm laser is switched on and off. c) I d -V ds of a b-As 0.83 P 0.17 device when a 532 nm laser is placed on positions shown on the right. 6.3 Conclusion: In summary, we have demonstrated the use of b-As 0.83 P 0.17 as a photodetector for 5 µm wavelength. We analyzed of the b-As 0.83 P 0.17 device parameters under illumination, and we found that the photocurrent mechanism is due to both, the photogating effect and photoconduction. The photoresponsivity for the power levels (20 mW-100 mW) was found to be 4.5 mA/W. Additionally, the photocurrent depends on the incident polarization, due to the in- plane anisotropy of the b-As 0.83 P 0.17 crystal. Moreover, we tried to measure the cutoff frequency of the b-As 0.83 P 0.17 photodetector, and found that the 3-dB cutoff frequency is higher than the maximum chopper frequency in our measurement setup (10 kHz). Finally, we used b-As 0.83 P 0.17 devices as photodetectors in NIR (785 nm) and visible (532 nm) spectral ranges. The results of this study is a demonstration that shows the promising potential of using b-As 0.83 P 0.17 as MWIR photodetectors at room temperature, and the high operation speeds of such photodetectors may be used in applications of high speed communication and imaging. 6.4 References: 1. Pospischil, A.; Furchi, M. M.; Mueller, T., Solar-energy conversion and light emission in an atomic monolayer p-n diode. Nat Nano 2014, 9 (4), 257-261. 2. RadisavljevicB; RadenovicA; BrivioJ; GiacomettiV; KisA, Single-layer MoS2 transistors. Nat Nano 2011, 6 (3), 147-150. 3. Liu, H.; Neal, A. T.; Zhu, Z.; Luo, Z.; Xu, X.; TomÃ¡nek, D.; Ye, P. D., Phosphorene: An Unexplored 2D Semiconductor with a High Hole Mobility. ACS Nano 2014, 8 (4), 4033-4041. 4. Xia, F.; Wang, H.; Jia, Y., Rediscovering black phosphorus as an anisotropic layered material for optoelectronics and electronics. Nat Commun 2014, 5. 5. Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D.; Chen, X. H.; Zhang, Y., Black phosphorus field-effect transistors. Nat Nano 2014, 9 (5), 372-377. 6. Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S., Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nat Nano 2012, 7 (11), 699- 712. 111 7. Choi, W.; Cho, M. Y.; Konar, A.; Lee, J. H.; Cha, G.-B.; Hong, S. C.; Kim, S.; Kim, J.; Jena, D.; Joo, J.; Kim, S., High-Detectivity Multilayer MoS2 Phototransistors with Spectral Response from Ultraviolet to Infrared. Advanced Materials 2012, 24 (43), 5832-5836. 8. Lopez-Sanchez, O.; Lembke, D.; Kayci, M.; Radenovic, A.; Kis, A., Ultrasensitive photodetectors based on monolayer MoS2. Nat Nano 2013, 8 (7), 497-501. 9. Zhang, W.; Chiu, M.-H.; Chen, C.-H.; Chen, W.; Li, L.-J.; Wee, A. T. S., Role of Metal Contacts in High-Performance Phototransistors Based on WSe2 Monolayers. ACS Nano 2014, 8 (8), 8653-8661. 10. Ma, C.; Shi, Y.; Hu, W.; Chiu, M.-H.; Liu, Z.; Bera, A.; Li, F.; Wang, H.; Li, L.-J.; Wu, T., Heterostructured WS2/CH3NH3PbI3 Photoconductors with Suppressed Dark Current and Enhanced Photodetectivity. Advanced Materials 2016, 28 (19), 3683-3689. 11. Baugher, B. W. H.; Churchill, H. O. H.; Yang, Y.; Jarillo-Herrero, P., Optoelectronic devices based on electrically tunable p-n diodes in a monolayer dichalcogenide. Nat Nano 2014, 9 (4), 262-267. 12. Ross, J. S.; Klement, P.; Jones, A. M.; Ghimire, N. J.; Yan, J.; Mandrus, D. G.; Taniguchi, T.; Watanabe, K.; Kitamura, K.; Yao, W.; Cobden, D. H.; Xu, X., Electrically tunable excitonic light- emitting diodes based on monolayer WSe2 p-n junctions. Nat Nano 2014, 9 (4), 268-272. 13. Amani, M.; Taheri, P.; Addou, R.; Ahn, G. H.; Kiriya, D.; Lien, D.-H.; Ager, J. W.; Wallace, R. M.; Javey, A., Recombination Kinetics and Effects of Superacid Treatment in Sulfur- and Selenium-Based Transition Metal Dichalcogenides. Nano Letters 2016, 16 (4), 2786-2791. 14. Amani, M.; Lien, D.-H.; Kiriya, D.; Xiao, J.; Azcatl, A.; Noh, J.; Madhvapathy, S. R.; Addou, R.; Kc, S.; Dubey, M.; Cho, K.; Wallace, R. M.; Lee, S.-C.; He, J.-H.; Ager, J. W.; Zhang, X.; Yablonovitch, E.; Javey, A., Near-unity photoluminescence quantum yield in MoS2. Science 2015, 350 (6264), 1065- 1068. 15. Wang, X.; Jones, A. M.; Seyler, K. L.; Tran, V.; Jia, Y.; Zhao, H.; Wang, H.; Yang, L.; Xu, X.; Xia, F., Highly anisotropic and robust excitons in monolayer black phosphorus. Nat Nano 2015, 10 (6), 517- 521. 16. N. K. Dhar, R. D. a. A. K. S., Advances in Infrared Detector Array Technology. In Optoelectronics – Advanced Materials and Devices, Ballato, S. L. P. a. J. M., Ed. InTech: 2013. 17. Wu, J.; Koon, G. K. W.; Xiang, D.; Han, C.; Toh, C. T.; Kulkarni, E. S.; Verzhbitskiy, I.; Carvalho, A.; Rodin, A. S.; Koenig, S. P.; Eda, G.; Chen, W.; Neto, A. H. C.; Ã–zyilmaz, B., Colossal Ultraviolet Photoresponsivity of Few-Layer Black Phosphorus. ACS Nano 2015, 9 (8), 8070-8077. 18. Huang, M.; Wang, M.; Chen, C.; Ma, Z.; Li, X.; Han, J.; Wu, Y., Broadband Black-Phosphorus Photodetectors with High Responsivity. Advanced Materials 2016, 28 (18), 3481-3485. 19. Buscema, M.; Groenendijk, D. J.; Steele, G. A.; van der Zant, H. S. J.; Castellanos-Gomez, A., Photovoltaic effect in few-layer black phosphorus PN junctions defined by local electrostatic gating. Nat Commun 2014, 5. 20. Miao, J.; Zhang, S.; Cai, L.; Wang, C. C., Black Phosphorus Schottky Diodes: Channel Length Scaling and Application as Photodetectors. Advanced Electronic Materials 2016, 2 (4), n/a-n/a. 21. Buscema, M.; Groenendijk, D. J.; Blanter, S. I.; Steele, G. A.; van der Zant, H. S. J.; Castellanos- Gomez, A., Fast and Broadband Photoresponse of Few-Layer Black Phosphorus Field-Effect Transistors. Nano Letters 2014, 14 (6), 3347-3352. 22. Engel, M.; Steiner, M.; Avouris, P., Black Phosphorus Photodetector for Multispectral, High- Resolution Imaging. Nano Letters 2014, 14 (11), 6414-6417. 23. Youngblood, N.; Chen, C.; Koester, S. J.; Li, M., Waveguide-integrated black phosphorus photodetector with high responsivity and low dark current. Nat Photon 2015, doi:10.1038/nphoton.2015.23. 24. Guo, Q.; Pospischil, A.; Bhuiyan, M.; Jiang, H.; Tian, H.; Farmer, D.; Deng, B.; Li, C.; Han, S.-J.; Wang, H.; Xia, Q.; Ma, T.-P.; Mueller, T.; Xia, F., Black Phosphorus Mid-Infrared Photodetectors with High Gain. Nano Letters 2016, 16 (7), 4648-4655. 112 25. Liu, Y.; Xu, F.; Zhang, Z.; Penev, E. S.; Yakobson, B. I., Two-Dimensional Mono-Elemental Semiconductor with Electronically Inactive Defects: The Case of Phosphorus. Nano Letters 2014, 14 (12), 6782-6786. 26. Xia, F.; Mueller, T.; Lin, Y.-m.; Valdes-Garcia, A.; Avouris, P., Ultrafast graphene photodetector. Nat Nano 2009, 4 (12), 839-843. 27. Itkis, M. E.; Borondics, F.; Yu, A.; Haddon, R. C., Bolometric Infrared Photoresponse of Suspended Single-Walled Carbon Nanotube Films. Science 2006, 312 (5772), 413-416. 28. Goykhman, I.; Sassi, U.; Desiatov, B.; Mazurski, N.; Milana, S.; de Fazio, D.; Eiden, A.; Khurgin, J.; Shappir, J.; Levy, U.; Ferrari, A. C., On-Chip Integrated, Siliconâ€“Graphene Plasmonic Schottky Photodetector with High Responsivity and Avalanche Photogain. Nano Letters 2016, 16 (5), 3005- 3013. 29. Mueller, T.; Xia, F.; Avouris, P., Graphene photodetectors for high-speed optical communications. Nat Photon 2010, 4 (5), 297-301. 30. Huang, Z.; Gao, M.; Yan, Z.; Pan, T.; Liao, F.; Lin, Y., Flexible infrared detectors based on p-n junctions of multi-walled carbon nanotubes. Nanoscale 2016, 8 (18), 9592-9599. 31. Liu, B.; Köpf, M.; Abbas, A. N.; Wang, X.; Guo, Q.; Jia, Y.; Xia, F.; Weihrich, R.; Bachhuber, F.; Pielnhofer, F.; Wang, H.; Dhall, R.; Cronin, S. B.; Ge, M.; Fang, X.; Nilges, T.; Zhou, C., Black Arsenic– Phosphorus: Layered Anisotropic Infrared Semiconductors with Highly Tunable Compositions and Properties. Advanced Materials 2015, 27 (30), 4423-4429.

Abstract (if available)

Abstract In the past decades, nanomaterials have attracted strong attention due to their versatile applications. Specifically, nanomaterials gained special focus in the fields of electronics and sending due to their flexibly adjustable electronic properties enabled by engineering their size, structure, chemical identity, and other material characteristics. In recent years, materials such as carbon nanotube, nanowires, graphene, transition metal dichalcogenides (TMDCs), and black phosphorus (BP) were utilized in a wide range of electronic applications such as digital, radio frequency (RF), bio-sensing, chemical sensing, and optical devices. In this letter, I will present our recent efforts to use graphene nanoribbons (GNRs), BP, and black arsenic-phosphorus (BAsP) in various applications.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

One-dimensional nanomaterials for electronic and sensing applications

PDF

Synthesis, characterization, and device application of two-dimensional materials beyond graphene

PDF

Nanomaterials for energy storage devices and electronic/optoelectronic devices

PDF

One-dimensional nanostructures for chemical sensing, transparent electronics, and energy conversion and storage devices

PDF

One-dimensional nanomaterials: synthesis and applications

PDF

GaAs nanowire optoelectronic and carbon nanotube electronic device applications

PDF

Semiconducting metal oxide nanostructures for scalable sensing applications

PDF

Ultrasonic microelectromechanical system for microfluidics, cancer therapeutics and sensing applications

PDF

Applications of one-dimensional structured nanomaterials as biosensors and transparent electronics

PDF

Zero-dimensional and one-dimensional nanostructured materials for application in photovoltaic cells

PDF

Chemical vapor deposition of graphene and two-dimensional materials: synthesis, characterization, and applications

PDF

Nanomaterials for macroelectronics and energy storage device

PDF

Substrate-induced effects in thermally cycled graphene & electron and thermoelectric transports across 2D material heterostructures

PDF

Surface functionalization of nanomaterials and the development of nanobiosensors

PDF

Nanostructured silicon for lithium-ion battery anode

PDF

In-situ characterization of nanoscale opto-electronic devices through optical spectroscopy and electron microscopy

PDF

In2O3 COVID-19 biosensors and two-dimensional materials: synthesis and applications

PDF

Nanomaterials for bio-imaging and therapeutics

PDF

Fabrication and characterization of yoroidal resonators for optical process improvement

PDF

Piezoelectric MEMS for acoustic sensing, contactless RF switching, and resonant mass sensing

Asset Metadata

Creator Abbas, Ahmad (author)

Core Title Novel nanomaterials for electronics, optoelectronics and sensing applications

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Electrical Engineering

Publication Date 03/23/2017

Defense Date 08/31/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemical sensing,nanomaterial,nanotechnology,OAI-PMH Harvest,photodetector,sensor

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Zhou, Chongwu (committee chair), Goo, Edward (committee member), Wang, Han (committee member), Wu, Wei (committee member)

Creator Email aabbas@usc.edu,ahmad.nabil.abbas@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-304162

Unique identifier UC11279494

Identifier etd-AbbasAhmad-4799.pdf (filename),usctheses-c40-304162 (legacy record id)

Legacy Identifier etd-AbbasAhmad-4799.pdf

Dmrecord 304162

Document Type Dissertation

Format application/pdf (imt)

Rights Abbas, Ahmad

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA