Scribd
Upload
30 views12 pages

Fabris

Uploaded by

Gabriel Henrique Batista
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
30 views12 pages

Fabris

Uploaded by

Gabriel Henrique Batista
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 12
‘Computational Materials Since 140 (2017) 344 Contents lists available at ScienceDirect Computational Materials Science ELSEVIER, journal homepage: www. Isevier.com/locate/commatsci Porous graphene and graphenylene nanotubes: Electronic structure and strain effects (on Guilherme S.L. Fabris *', Chad E. Junkermeier °, Ricardo Paupitz** So Paulo State University (UNESP), Institute of Geasiences and Eat Sines, Ro Care, 1506-90 SP, Bai esearch Corporation of he Univesity of Hawa. Howaul, M9658 Unted tates ARTICLE INFO ABSTRACT ‘Arce toy Received 9 une 2017 Received in revise frm 16 Angus 2017 cepted Steer 207 “The unusual and unique mechanical and electronic properties of nanostructured carbon materials make them useful inthe construction of nanodevices. We investigate a new class of structures, called porous nanotubes, which are constructed from two recently synthesized two-dimensional matenals, namely the porous graphene (PG) andthe two-dimensional catbon allotrope known a graphenylene, aso known, 28 Biphenylene Carbon (BPC). We investigate this cass of quas-one-dimensional materials using the den- sity functional tight-binding (DFTB) method to optimize geometries and to calculate electronic structure features ofthese systems. Fr each type of porous nanetube, calculations were performed on tubes with several diameters and chirlitie Our rerlts show that the PG nanotubes have 2 wide band-gap, 3.36, and the graphenylene nanotubes have a semiconductor behavior with a band gap around atbon nanotubes Porous graphene Emnnenine Sinaaive 017. They also show tat a the diameter of D6 nanotube increases the band-gap decreases, whe Sininefss forte raphenlene nana the band gp increases In both ase the cbvewed gap vanation with Poros niotubes Increasing diameter is towards the value found for the respective two-dimensional membrane. Calculations on axially strained porous nanotubes show a decrease on the band gap of ~ 10% for some chiralities ofthe PG nanotube and an increase forthe graphenylene nanotubes gap that ean become a5, high as 100% These results are in contrast with the expected behavior fr carbon nanotubes, which show a linear dependence between gap opening and applied strain under simi 1. Introduction Graphene (1,2] isa two-dimensional array of hexagonal units of sp? bonded carbon atoms (Fiz. 1) which presents unusual and interesting electronic and mechanical properties [2]. Because of these properties it may be used in diverse applications, for example ‘electronics (3:4), water separation [9], nanomechanical resonators [6-8], chemical sensors (9), or in producing exotic materials [10— 14]. However, in its pristine form, graphene is a gapless semicon- ‘ductor. This characteristic imposes serious limitations to its use in practical electronic applications. Many approaches have been proposed to create a gap in graphene-like materials. One of the ‘most common strategies uses chemisorption methods, such as oxi- dation (15-17), hydrogenation |18-20), luorination 21-24), or some other adsorbate [9,25]. Another approach isto obtain inti sically hydrogenated structures, such as PG (Fig. 1b), whose syn- * Cowesponding autho. mal adres: poupteorcaespe (Paap) * Present aes: Grupo de Moselagem Siulacdo Noel ~ DM io Paulo State Univesity = UNESP, aa Peta 473, aur ~ So Pal, SP Bra hup:ibdiorg)10.1016) commasc.2017.05009 (0827-0256) 2017 Hkevier BY. Al ph reserved conditions '© 2017 Elsevier BY. All rights reserved thesis was recently achieved [26]. Other authors discussed the metallic behavior found for structures obtained by the combina- tion of biphenylenes in several configurations |27| witch, despite some structural similarities, are different from the structures dis- ‘cussed in the present work, which can present semiconducting or insulating behavior. One of these combinations is the so called gra- phenylene, which has a small band gap and delocalized frontier “orbitals 28-30]. A possible route to the synthesis of graphenylene ‘was proposed elsewhere [28] and its experimental realization, although obtained through a different route, was reported recently [31,32}, Further, Schlutter et al. reported recently that they were able to synthesize octafunctionalized biphenylene GNRS [33] ‘Another possibility is the combination of these geometries with functionalization [34 that can lead to interesting effects regarding gas adsorption [35]. The new generation of electronics may also be facilitated by producing any of the above structures using boron nitride (BN) oF some combination of carbon, boron, and nitrogen ‘atoms [36,37] In this context, we consider the possibilty ofstruc~ tures based on the PG and graphenylene primitive cells and on CNT architecture, the so-called porous nanotubes (PNT). The grapheny- lene version of the PNTs was proposed recently by Koch et al. (38 GS Faris Computational Maes Sence 140 (2017) 386-355, was Fig 1. Definitions of wit vectors a a yan riitive cells forthe two dimensional structures dicted thhout the text. (8) Graphene, 2) Pocus Graphene (Cee Phenylene Caton, and, taking into account the fact that density functional theory (DFT) calculations tends to underestimate gap openings, our results are in good agreement with their findings. In the present ‘work, we also study gap variations caused by axial strain on the proposed structures. In order to investigate structural and elec- tronic properties of this kind of material we investigate sixty-two representative one-dimensional structures. These are based on rine armehair-ke, fourteen 2igzag-like, and eight chical-like PG- based PNTS with’ or corresponding graphenylene-based NTs. Using DFTB we computed the electronic structure and studied the structural properties of each one of these structures. 2. Methods Geometry optimizations and electronic structure calculations of planar structures and PNTS were performed using a self- Consistent-charge (SCC) version of the density functional tight- binding (DFTB) method [39-41] in the DFTB* code |42]. This ‘methodology can be implemented in a low computational cost code and mixes advantages of tight binding methods and of DFT precision in the description of molecular and condensed matter Systems 45], This methedology is based ona second-order expan- sion of the Kohn-Sham energy defined in DFT (40,44). Band gap values predicted by DFTB may be underestimated, caused by the formulation being based only on valence electrons yet its useful in obtaining relative energy values orto explain trends ina series oflarge atomic structures. The particular parameterization used in the present work, pc [5,6] Slater-Koster files, was developed for solids and surfaces and has been used in modeling systems similar to those discussed here [13]. Convergence crteta forthe geometry optimization used force ciferences of 10-* with a SCC tolerance of 10-* For sampling the points in the Brillouin zone, we used a 10 1.1 Monkhorst-Pack grid (7) in which the 10 folding sin the direction related to the axis of the PNT. Also, as vacuum condi tion, it was considered a 50Adistance of empty space in directions perpendicular tothe nanotube axis. The molecular dynamics (MD) Simulations discussed in the present work were caried out using the Verlet algorithm, while temperatures were controlled through a Berendsen thermostat. 3. Results and discussion 3.1, From membranes to tubes In order to define a systematic way for the construction of the NTs, we took advantage of the geometrical similarities of the unit cells when comparing PG and graphenylene membranes with gra- phene. Using the well known definition of chiral indices for CNTs [18,49] we define a chiral vector for each one of our template pla- rat PG and graphenylene membranes. The construction of PNTs Fig. 2 Geometrical detintons and chia indies for two dimensions strates ‘rphenylene sheets The vectors A and ae respectively lie of and bused {h definng thecal vector Gy, with T being the anslaon vec. Shaded areas indieate the unvaled primatve cl for specie examples of evaities.Hipie ‘elon deliited by date lines nds te distances) PE with pri cll ofa (5.1) PNT ndiaed by shaded area (b)raphenene sheet with primitive cll fof (52) PNT indeated bythe shaded aca follow a recipe similar to that used in the case ofthe usual carbon nanotubes. In Fig. 2, chiral indices n and m are defined for both, PG (Pig. 2a) and graphenylene (Fig. 2b) PNTs as well as thet chiral vec- {or G,, primitive lattice vectors, translational vector T and quasi 6 (OSL Fabris Computational Material Slence M40 (2017) 344-355 Sct information and bad gap values fr pine cll of PG-based PATS Colum N and Uni Cell indicate the otal umber of atoms andthe taal length fe 3 one P6t6.0) == P6t8.2) 000 —0a2 ada 006 O08 Strain 040 oaa (b) Band-gap (eV) oa — BPCtaa) s+ BPC(6,0) BPC(8,2) oa] Boo 002 0.04 0.06 ‘O08 strain ‘oid ig. 10, Al stan fect on gap values for PG and grapenylonebased PTs of diferent chiral, namely (44 representative of armchair character, (60) epresenaive of igzag an (8.2) of chiral character a) PC-based PNT. () Graphenene-ased PNT. the energy increases smoothly with axial strain as shown in Fis. 9. Fig, L0(a) shows our results for PG based nanotubes where one can see that, starting with e— 0, higher strain values correspond to smaller gap values with a monotonic decrease for (6,0) and (8,2) tubes. On the other hand, the (4,4) tube presents a minimum at ‘¢~011 and the gap opening increases again for higher deforma- tions, although the values still smaller than 3.2 eV up to the max- imum strain considered. Fig. 10(b) shows the influence of strain in the axial direction for graphenylene-based nanotubes, which pre- sent a reversed trend if compared to that found for PG based PNTS. Gap values for graphenylene-based PNTs increase monotonically ‘with the increasing of strain. In both cases, higher strain (es, L/L ~ 0.2) causes plastic deformations of the PNTS, which coi cide with abrupt changes in gap values (not shown). Earlier studies considered usual CNTs [53.54] and reported a linear dependence between the gap opening and the applied axial strain, In the present work, we found that DFTB predicts 4 deviation from linearity in both cases. These deviations from linearity could be due to the fact that PNTS present several bonding types (double bonds, single bonds, C-H_ bonds) that ‘weakens the structure and facilitate greater atomic rearrange- ‘ment, In this context, the effect of band gap vatiation observed ‘while strain is applied can be understood as a consequence (at least partially) of electronic interactions of atomic orbitals that result in some degree of repulsion combined with orbital recom- binations that modify the band structure. In Fig. 11 it is possible to compare the projected density of states (PDOS) for represen- tative porous nanotubes investigated in the present work. in that figure, it is possible to identify that 2p states are the main con- tribution for the majority of states located close to the Fermi level (located at 0 eV in the figure). This characteristics indicates that, upon deformations, the interaction and repulsion effects SL aie Computational Materia Science 140 (2017) 24 (@) = Dos (b) 00s = 0s cs) | = 0s cis » = ro0s cts = 0s cis : —r008 every ev Energy (ev © = @ a 7 0s cis) . = 500s cs) 008 ci) = F008 ct 4 —bo08 ts : Ea ! B40 8 ' gx 3 Tc a = v ? + nero ev Energy Fig. 11 ect of axa non P00S fortwo representative porous nanatbes. nthe horizontal xis e ndcates the erm eve ney.) Graphene base nanotube ‘thea! indexes (82) wah no stain (e= 02) Ie povtat to ote ha the green curve (2 obi contibuton) lose to rea amd acca conedes wh te hovzontal axis in blk and canna be seen lay (0 Porous papene based nanotube lnelaxed geomety(¢~ O(c) Caphenylene based be wth ~ NOR Again ‘he green curve, that desroes 2 orlascontiution fs small hat mest clacden with the horzntal ais (2) Poros Graphene seed nanacbe with ~ 10% CFOt Interpretation ofthe references acolo nth gre legend the reader Is ferred tothe web version ofthis ate) between 2p states ate the responsible for the gap modifications as deseribed by Fis. 10. In order to evaluate the quality of results obtained in the DFTB, approximation level, we cattied some DFT calculations for repre- sentative porous nanotubes considering both types, PG and gra- phenylene based PNTS using different exchange-correlation potentials (PBE [55] and B3LYP [56,57)). Our calculations indicate ‘that DFTB results are in relative accordance with DFT results in the sense that it predicts approximately the relative variation on sgap values in both cases (PG-based and graphenylene-based). Also, DFT predicts correctly the trends in these band gap openings, that js: an increasing gap with strain in the case of graphenylene-based tubes and decreasing gaps with strain for PG-based nanotubes. la terms of absolute values, DFTB predictions are closer to B3LYP results for the pure carbon porous nanotubes than PBE. This can be interpreted asa favorable result for DFTB ifone considers calcu- lations reported elsewhere by Goddard and coworkers [58] where ‘they conclude that B3LYP provides the best description for the gap) ‘opening of carbon nanotubes. On the other hand, for porous agraphene-based nanotubes, relative variations of bandgap open- ings predicted by DFTB approximation still in accordance with [BSLYP and PBE, but absolute values are closer to PBE results. 4. Conclusions We investigate the geometrical configuration and electronic structure of two new classes of nanotube architectures. The nan- fotubes were built using two different repeating units, the frst based on porous graphene and the second based on graphenylene structure. Band gaps vary depending on tube diameters. Porous based tubes present higher band gaps than that found for PG nanosheets, while graphenylene-based tubes present gap values smaller than theit two dimensional counterpatt. Variation on the dlameter of these tubes revealed a variation tendency on gap val- ues of both nanotube types towards their respective planar sheets band gaps. Porous based nanotubes present a decrease in gap val- ues and graphenylene type present increase of band gap values for increasing diameters. The influence of axial strain on the band gap ‘opening vas also investigated. Our results indicate that PG-based NTs can have their band gap decreased by ~ 10% in some cases. For graphenylene-based PNTS we found that strain application ‘an induce a substantial increase on the band gap opening, chat ‘an reach up to 100% in some cases. Having these adjustable phys- ical properties make these systems good candidates for the con- struction of nano-sensors or nano-devices. as (OSL Fabris Computational Material Slence M40 (2017) 344-355 ‘Acknowledgments [Nanotube models where produced using Greenwood: A library for creating molecular models and processing molecular dynamics simulations [59]. Funding This work was supported in part by the Brazilian Agencies CNPq, CAPES, FAPESP. Ricardo Paupitz thanks the financial support of Fapesp (Grant 2014/15521-9) and CNPa (Grant 308298/2014-4). Appendix A. Supplementary material ‘Supplementary data associated with this article can he found, in the online version, at http:|jdx doi org/10.1016)j commatsei.2017. 09.009. References UI KS. Noein AK Gi SMa, Jang hg SD 1 Scence 306 (5636 (2004) 666-658 BR) AM CastoNero, F Guinea, MLR. Peres, KS. Novesdov, AK. Geim, The eczonte properies of graphene Re. Mod Ps 81 (1) (2008) 109-162, {3} Z Wang. CP Puls. E_Staley. Zhang A Tod. Xu, CA Mowsare Hollander} Robinson ¥. Li, Technolgy eady use of single layer sapien 35 a transparent etre for hyd photovatae devs. Phys. LoW= Sioensional ‘Syst Nanostact 44 (2) (2011) 821-5289 olor Taioio)ipnyse2011 1000. tepinawssecedeecomiscencs aeleps158654771 1003651. 1 KE, Whtener Reversible graphene funcionatzation for. electronic applications a review. in; Armerican Chem! Society. 2074. pp. 41-34, int oedtorg/10.1081o% 2014-1163 appeal U5) Lach RA Uno Xu, ¥. Ca baju, W. Zhang, RL Sac LV. Visio, PE. Fulvio, P Ganesh, e al. Aqueous proton wanser cross singe ayer szaphene Nat Comm. 6 (2013. (61 Le Robinson. MX Zalattdinoy. CE. Junkermeier 3 Culberson TL Reinecke, Stine, PE Sheehan, BH, Houston, ES. Show, Structural atlormatons emily modified graphene, Sod State Commun. 152 (21) (2012) 1980" 1998, pd. erg] 10.1016) ssez01z 04051, ‘ate ‘wiv stencediectcomistenceaticepi/300381098}2002578> 17 NOC Zaldiney Robinson CE Junker Culberson Ti Reinecke. Re'stine, PE sheehan, BI. Houston, ES. Shot, Engineering sapere ‘mechanical systems, Nano Let 12 (8) (2012) 4212-8248. ide) Toutoayfa0V8055, ip ipubeace nde VO.103019018059 “hts. acs. 102113018050 [s) {1 ebinson, MA. Zleluino, CD. Cress LC Culbeson, AL Friedman, A al Graphene stained by ef ACS Nano 0 (0) (0) ml Mi: ‘aoa hpi org/10.102 esa. 7000823 Io] SCoHermnde, CJC Henne CE Jantermeter SD Tao Ef. ezaes 8 Stine, T-fobinson. EH, Lock. DR. Bors, RD. Pate J, Caldwell TL Reinecke PE Sinehan SG Welton chemical gradient om graphene o dive droplet maton, ‘ACS Nao 7 (6) (2013) 4748"4755, hep foesorw 10 102 An04257, pao: 23650405, 1101 Vols D. Galo, A. Joi, Geometic and electronic sractre of caon ‘anotube nor "sper carbon nnobes Nanoteenlogy 17 (3) (2006) 617-004 In1] VR, colds R08. os Santos 0, alva, opolglaly cased macomelecles Imade_of single walled. carbon nangtubee"super-flleenes,. Nanos ‘Nanoechao. 0 (7) (2010) 978-4383. na] F- Ouyang. Peng Zi. 2 Ui, Banden opening in graphene ano tices: {hemising hall ACS Nano 5 (3) (2041) 4023-4030, [13] RPauple CE junkermeler, AT van ul, brani Fullerenes generated fom porous statues, PCCP 16 (40) (2008) 25815-25822, 013) Hotoeojacpossea, ee ta ss) WOs000344989500038. 114]. Peim, & Paupizz PAS. Auteto, DS. Calvae Inowzanicsrapeaylene: 2 ‘ote two-dimensional materiel wit unable bandgap Ph, Chem. € 118 (41) (204s) za670~23674 pando or 1010219502119. perme Paupit R Auteto, PAS. Cava, D. Co to IS>OS.00034333 360007. [15] S'Stankvieh, DA’ Di, RD, Piner A Kellan, A Aleiahamnes, Yi ia ST Nguyen, RS. fol, Synths of grapone-based nanoseets a ‘emi teduttion of exoiated praphte oxide Cavbon 45 (7) (2007) 1338— 1565, ep ix ovary (0 0to)earoon 200702034. sae svi scieneaet com stent piS00086223070817. ne 5c, 5. Han, Me Wang, Ki” Wang. Kane, A chemical out ta graphene Tor desicesppcatons Nap Ltt (11) (2007) 3394-1398, 17 ot, Gap: cling al ce, NAL Rate. > (1) (2008) 118) 0, Sofa. AS. Chaudhari, GD. Barber. Graphane: a. twovdimensional hydrocarbon. Pays. Re. B75 (2007) 4S3a0h, coo 101105) Pysev8 9515890. hip:fiapsorgiy 10 105/Physkev8 75153401 1n9] 5 fy Mv Han] Match, TH eine, Kim ML tegen, LE Bre Reveribe isl plane hydrogenation of graphene Nano Ut. 8 (12) (2008), 1597-4602 (20). Has air. MohludinS Morozo,P. Blke Mt, Halsall A. Frran Boukhvalo, M, Kateneleon, A.ceim, K: Nevaeelow, Control ofgraphene's Propet: by reversible hydrogenation: evidence for graphane, See 523, (S914) 2005) 610-613, 121) 17. Robinson. Js. Burgess. CE Junkermeier S. Best, TL. Reinecke. EK ovin, MK Zalalutdnoy, JW. fawn, Culberson, PE, Sheehan. ES. ‘Snow roperies af fuonaod graphene ms Nano et 10(8)(2010) 3001 5300s, apd doorg/toto2H/aniO1e37p. cit fpubsacs-r cep 1. ‘tztjaltonasrp> tp: pbs aes erldaabs10.1021/n01437p>, [22] RPaupite, PAS. Autret. $B. Leos. 5 Scnvasan, Tan Duin DS Clo, Graphene to oropaphene and Muorographane: a theoreti stdy (a) CE Juke, SE aad Rea, Hey Moriated phn, 124) D. Solenov.c Junkermeer. TL Reinecke. KA. Veizhanin, Tunable adsorbate- ‘sorbate interactions on graphene. Prys Rev. Let 1112013) 115502.) ‘Seaovongt0 1 To3/thyskevete TT it1330%, eh inkap orca 0 TosPonyevtete tnt 1ss02> [25] CE junkermesr 0 Soleo, 1 Reinecke, Adsorption af NH on graphene in the presence of defects and adsorbtes J Phys. Chem. € 117 (0) (2018) 2783 2798, itp iéxdo.org( 101021 jps0aaion, cht Hpubsacsoryop0 dapat n> hep) us seo do sb 1024p 3038 [26] Mie, Mees, an ea Manoa, Plea 0. Cong, Coming, aster, Es Reger X Fg, Millen, fasel, Porous graphene: [471 Hy. Montost ack. Special pins fr billouin-zone iterations. Phys. few. B13 (1979) 5188-5192, it; dover 11 103]Phystev8 135188 “iip/inkapeotda10.1103/ysRev, 133 185> [as] JC. Chae. se. 5. Roce, Hectonic and’ uanspot proper of anotubes. er. Modern Piys. 79 (2) (2007) 677-732, My. oe. 10. 103)RevModPiys78677 [49] M. DresehausG Dressel. Sat, Physics of carbon nanotubes, Carbon 33 Gy (i9ns) a89-801, ftp or] 10, 101s oe e223 99 0017-8 [so] W-Humpirey.& Dale Schulte VD = visual molecular dynamics, el Crap (1986) 35-38. [51] HR Gutirez N Pere-Lpez,AL Has, A. Berkdemit, . Wane. Rv. F.tper Uras WH ces Termes Ne Terones. Extraordinary room-temperature Dotolumanescence in triangular WSP meonoayers Nana Late 13 (8) (2013) Sa7-3454 hp edorg 1010201026559, pa: 23194085. {521 Si sion, te ox So Ste ac, ate, Oxon Unerty Fes, [53] CAC Lin YH Huang. Wu, Leung Sain eects onthe band gp and work function” of zigzag single-walled carbon nanotubes 2nd. graphene: a Cermparatve study. Comput Phys. Commun 185 (5) (2014) 1422-1428. ‘iupedosong/t0 10 ehepe 20740200, cute ‘wor scienceetcom/scienceatlepi/S01 046551400081 > [54] LW bing XH. Yan JX Cae, DL Wang. Tang. 08 Yang Curvature an srain ffecs on electronic properties of single wall carbon nanotubes. ys ope ater 15 (2200) US, i sexton 5) 14 a= [55] JP. Perdew. K Bute: M,Eezehot Generalized gradient approximation made Simple. Ps Rev. Let. 77 (1996) 3865-868, peda! 101 108) Dayskeuia77sees, —”

You might also like