electronic transport along hybrid mos monolayers

Electronic Transport along Hybrid MoS2 MonolayersGanesh Sivaraman,†,∥ Fabio A. L. de Souza,‡,∥ Rodrigo G. Amorim,§,¶ Wanderla L. Scopel,‡ Maria Fyta,*,†

and Ralph H. Scheicher*,§

†Institute for Computational Physics, Universitat Stuttgart, Allmandring 3, 70569 Stuttgart, Germany‡Departamento de Física, Universidade Federal do Espírito Santo-UFES, Vitoria, Espírito Santo, Brazil§Department of Physics and Astronomy, Materials Theory, Uppsala University, Box 516, SE-751 20 Uppsala, Sweden¶Departamento de Física, ICEx, Universidade Federal Fluminense - UFF, Volta Redonda, Rio de Janeiro, Brazil

ABSTRACT: Molybdenum disulfide (MoS2) is a two-dimensional material inwhich a semiconducting and a metallic phase can coexist. In this work, weinvestigate the electronic and transport properties of a hybrid MoS2 monolayercomposed of a metallic strip embedded in the semiconducting MoS2 phase. Usingquantum mechanical calculations within the density functional theory schemetogether with the non-equilibrium Greens functions approach, we study in detail thestructural and electronic properties of this hybrid material and its metal−semiconductor interface. A single point-defect analysis is performed in order toassess the stability of the hybrid system. Focus is given to the electronic transportproperties of the hybrid MoS2 monolayer extracted from the electronic transmissionspectra. These are linked to the local current across the monolayer. A clearasymmetry of the current flowing across the hybrid monolayer was found and wasattributed to the atomistic characteristics of the material’s interfaces. The resultssuggest strong potential for the application of hybrid MoS2 in the next generationbiosensing devices.

■ INTRODUCTION

Two-dimensional (2D) materials have attracted high interest inrecent years. Starting with graphene and its numerous potentialapplications, the research of 2D nanomaterials1,2 was followedby intense investigations on 2D transition metal dichalcoge-nides (TMDs).3,4 TMDs (MX2, with M = V, Mo, W, etc., and X= S, Se, etc.) are quasi two-dimensional layered materials withstrong interlayer ionic−covalent bonding. 2D TMDs can befound in two phases, semiconducting (2H) and metallic (1T).Liquid-phase exfoliation is the typical method to produce themonolayer TMDs from their layered counterparts.5,6 It wasshown that the transition from the 2H to the 1T phase ofMoS2, MoSe2, WS2, and WSe2 during their chemical exfoliationdepends on the MX2 composition of these materials.7

The most famous member of the 2D TMD family is MoS2,which has been used as a dry lubricant for many decades in itsbulk form. In the earlier studies, focus was given on MoS2-basednanoparticles, such as MoS2 nested inorganic fullerenes,nanotubes,8 and MoS2 nanoclusters9 used as catalysts.10 Theinvestigations then turned to MoS2 surfaces11,12 and theirability to adsorb hydrogen.13 The research on MoS2 has showna 4-fold increase since the year 2010 when the direct band gapin the single-layered structure was discovered.4,14 Thecoexistence of a metallic and semiconducting phase has beenreported in MoS2 in the past in several studies.15−17 Thisrepresents a distinct polymorphism in terms of structural andelectronic properties, a marked deviation from graphene. Thecoexistence of the semiconducting (2H) and metallic (1T)

phase in MoS2 monolayers has been characterized.18−21 Inprinciple, gliding only one S plane of MoS2 to the center of thehexagonal rings of the semiconducting 2H phase will graduallytransform the structure to the metallic 1T phase.15 During thistransformation, while the size of the 1T part increases, threedifferent boundariesα, β, and γemerge. The α boundary isrelated to the Mo−Mo distance shrinking, the β boundaryinvolves the Mo + S gliding, and the γ boundary is based on theS gliding.15 The recent development of controlled techniquesto induce 2H to 1T phase transition15,22 opens up promisingroutes for an atomically precise fabrication of single-layeredchemically homogeneous electronic devices. Lately, anotherimportant achievement in the field was the formation ofnanopores in MoS2.

23 These nanopores are formed using anelectrochemical reaction method,24 and it has been demon-strated that DNA can be translocated through them within asalt solution.25 These nanopores are efficient in discriminatingamong DNA nucleotides26 and can lead to a novel sequencingtechnique.27,28

In this study, we focus on hybrid monolayers of MoS2. Theterm “hybrid” refers to the combination of the two differentphases (2H and 1T) composing the monolayer. It accounts forthe polymorphicity of MoS2 linking to materials with tunablefunctionalities. The combination of different phases in

Received: August 5, 2016Revised: September 19, 2016

Article

pubs.acs.org/JPCC

© XXXX American Chemical Society A DOI: 10.1021/acs.jpcc.6b07917J. Phys. Chem. C XXXX, XXX, XXX−XXX

pubs.acs.org/JPCC

http://dx.doi.org/10.1021/acs.jpcc.6b07917

constructing the hybrid structures offers additional pathways toelectrons and is expected to enhance the electronic transportproperties along the hybrid monolayers. This would be relevantto electronics applications and also for sequencing DNA using ahybrid MoS2 in which a nanopore can be opened. Within asimilar context, the capability of MoS2 monolayers to act ashighly sensitive and selective gas sensors has also been exploredexperimentally.29−32 Accordingly, the electronic current con-finement on the entrenched 1T MoS2 nanoroad could make ahybrid MoS2 device more sensitive to the electronic character-istics of possibly adsorbed biomolecules than the single-phaseMoS2 devices. This article is organized as follows: we firstpresent the methodology used and then proceed to the resultson the structural, electronic, and transport properties of hybridMoS2 monolayers. Our results also include a point-defectanalysis, which reveals the stability characteristics of the hybridmonolayers. In the end, we discuss the implications of ourresults on novel nanobiotechnological applications of hybridMoS2 2D materials.

■ METHODOLOGY

Density functional theory (DFT)33,34 based simulations asimplemented in the code SIESTA35 were carried out. We usedthe generalized gradient approximation of Perdew−Burke−Erzernhof (PBE-GGA)36 and the norm-conserving Troullier−Martins pseudopotentials.37 For expanding the Kohn−Shamstates, we considered a double-ζ with polarization basis set(DZP). An energy shift of 0.01 Ry with a real space samplinggrid (mesh cutoff) of 200 Ry and 10 × 1 × 12 k-points withinthe Monkhorst−Pack scheme were used. This mesh was foundsuitable for the smallest unit in Figure 1a. The geometryrelaxations were performed until the net forces of each atomiccomponent became smaller than 0.01 eV/Å. The electronictransport calculations were performed using DFT combinedwith the non-equilibrium Green’s functions (NEGF) formal-ism, as implemented in TranSIESTA.38

We first carried out benchmark calculations separately onboth the pristine metallic and the semiconducting phases ofMoS2. For the pristine 1T phase, the lattice parameter of 3.23 Åand the Mo−S bond length of 2.47 Å were found. Because thepristine free-standing 1T phase is not very stable, there are noexperimental data for comparison. We also find that 2H MoS2is a semiconductor with a direct band gap, a lattice constant,and a Mo−S bond length of 1.64 eV, 3.23 Å, and 2.45 Å,respectively. These values are in very good agreement withprevious theoretical results.39−41

In this work, we study a monolayer hybrid MoS2 structure,which combines both the 1T (metallic) and the 2H(semiconducting) phase of MoS2. More specifically, a 1Tphase is embedded in the 2H phase as evident from Figure 1e.To investigate the electronic properties of the hybrid MoS2system, we consider different supercell sizes. The different unitcells used to generate these supercells are shown in Figure 1a−d. These correspond to 36, 48, 60, and 66 atoms per unit cell,respectively. By going from the 36 up to the 66 atom unit cellsystem, the width of the embedded 1T nanoribbon is increasedfrom 4.7 to 20.5 Å. A k-grid of 60 × 1 × 90 is used for theMonkhorst−Pack sampling of the density of states.The setup for the transport calculations across the hybrid

MoS2 monolayer is shown in Figure 1e. The hybrid MoS2device is composed of the two electrodes (left and right) andthe scattering region (device). Both parts, the leads and thescattering region, are hybrid, meaning that these are composedof both the 1T and the 2H phases. With this setup, theelectronic transport occurs along the 1T ribbon. The structureshown in Figure 1d was relaxed further up to a force toleranceof 0.001 eV/Å. The resulting relaxed structure formed the leadfor the transport setup. The device region was generated byreplicating four units of the leads along the z-direction. Finally,the transmission was computed by coupling the device to semi-infinite leads. Because of the large size of the system, theBrillouin zone sampling was restricted to the Γ-point. For the

Figure 1. Hybrid MoS2 structures for different widths of the metallic 1T ribbons. Four different unit cells were modeled: (a) a 36 atom unit cell with1 unit of 1T embedded into 2H, (b) a 48 atom unit cell with 3 units of 1T embedded in 2H, (c) a 60 atom unit cell with 5 units of 1T embedded in2H, and (d) a 66 atom unit cell with 6 units of 1T embedded in 2H. (e) The transport setup is depicted. Both the leads and the scattering region ofthe device are shown. The leads are built from a single unit cell of 66 atoms in a structure based on that shown in panel d but with relaxation of theatoms included. The scattering region is composed of four repeated unit cells each containing 66 atoms, in the same structure as in the leads. Theelectronic transport occurs along the z-direction. The S atoms are shown in brown and the Mo atoms in cyan. The same color coding will be usedthroughout this work.

The Journal of Physical Chemistry C Article

DOI: 10.1021/acs.jpcc.6b07917J. Phys. Chem. C XXXX, XXX, XXX−XXX

B


transport calculations, finite structures rather than periodicsupercells were considered.

■ RESULTS AND DISCUSSION

Structural Characteristics and Defect Analysis. Weinitially considered a hybrid structure with zigzag edges alongthe interface. In this case, the fully relaxed atomic structure ledto buckling, in agreement with previous results.42 This bucklingdid not lead to any significant induced curvature outward fromor inward towards the hybrid structure. Theoretical studieshave shown that neither the armchair43 or the zigzag21

interfaces in hybrid MoS2 show out-of-plane distortion.Nevertheless, our findings demonstrate that the armchairinterface does not show inflection, but it is the most stableagainst buckling. Accordingly, the armchair interface will be theone considered herein. Thus, we begin our analysis with thestructural details of the hybrid MoS2 monolayers. Representa-tive fully relaxed structures with varying 1T regions are shownin Figure 2. Inspection of the side view of these reveals a planarsystem in which distorted bonds due to the interface areevident. From a structural point of view, it is clearly visible fromthe top views in Figure 2 that the interfaces are armchair-likeand not symmetric.Hence, we chose a nonsymmetric interface, which did not

lead to buckling consistent with the experimentally stablestructures.19 To generate this interface, the β- and γ-phases ofthe interface were shifted by one unit cell in the bottominterface with respect to the top one (“bottom” and “top” referto the interfaces clearly seen in Figure 1).43 From Figure 2, it isverified that the 2H phase does not change, whereas the 1Tphase presents a structural distortion induced by the interface.Each of the 1T phases contains a Mo−Mo trimer unit. In thepristine 1T phase, all of the Mo−Mo distances are equal to thelattice constant. In our hybrid model, the observed distortion inthe 1T region (for all 1T sizes modeled here) can be quantifiedby the Mo−Mo distance shortening along two of the Mo−Modistances. These distances are in the range of 2.75−2.85 Å. Thethird Mo−Mo distance remains undistorted, and its values areclose to the lattice parameter (3.15 Å). Similar experimentalresults have been reported for a Mo−Mo distance of 2.9 Å.17,44

Such a partial distortion in the Mo−Mo trimer unit of 1Tremains as the number of 1T units increases. This can beconfirmed also for the smallest unit cell taken (36 atoms/unitcell in Figure 1a) in which the 1T region is very thin and is

equal to about 4.7 Å. Further inspection of this in Figure 2reveals the atomically sharp interface in conjunction with theexperiments.19 No atom loss or affluence was observed in ourhybrid monolayers, in accordance with previous theoreticalinvestigation.21

To study the energetic stability of the hybrid system model,we have calculated the formation energy of single sulfurvacancies. Aiming to identify the regions which are energeticallymore favorable to form vacancies, we move along a sulfur lineacross the armchair interface of the monolayer as shown inFigure 3 and remove one sulfur atom (labeled and pointed by

the arrows in the figure) leaving the monolayer with a singlevacancy. The stability is cast in the form of the formationenergy of the single point sulfur defect in the monolayer. Thecalculation of the formation energy, Eform, for a single sulfurvacancy in hybrid MoS2 is based on the comparison of the totalenergies of the monolayers with and without the vacancy on thechemical potential of a single sulfur atom as calculated throughour simulations. Accordingly, the formation energy is giventhrough

Figure 2. Relaxed structures of the hybrid MoS2 monolayer for the smallest and largest unit cells from Figure 1a, d. The upper panels show topviews, while the lower panels show a side view of the relaxed structures. The gray rectangular regions in the top panel indicate the three periodic unitcells in the structures.

Figure 3. Single vacancy formation energies (Eform) of sulfur siteslocated along a vertical line crossing the 66 atom unit cell system. Eachbar in the graph gives the formation energy of a single S defect for thecorresponding S atoms from the string of encircled atoms below thegraph, as denoted by the respective arrow.



C


μ= + −E E Eform defect s pristine (1)

where Edefect is the total energy of the system with a single sulfurvacancy, Epristine is the total energy of the pristine monolayer,and μs is the chemical potential of S atoms. The μs is obtainedfrom the stable S8 ring, which is very similar to the bulk value ofμs.

45 The sulfur vacancy formation is an endothermic process,meaning that the more positive the formation energy, the moreunlikely the formation of the vacancy. The relative magnitudesof the calculated defect formation energies are shown in Figure3. The formation energy has its smallest value when sulfur atomnumber 9 is removed from the monolayer. This S atom islocated at the interface region. The variation in single vacancyformation energy exhibits another local minimum correspond-ing to the removal of sulfur atom number 5 slightly furtheraway from the interface. This removal corresponds to a defectformation energy which is about 0.5 eV higher than that foratom number 9 at the interface mentioned earlier.We have also investigated the sequential removal of sulfur

atoms along the two semiconductor−metallic interfaces in thehybrid MoS2 shown in Figure 3. The results (not shown)confirm that most of these sulfur atoms have a defect formationenergy of about 1 eV similar to the lowest energy in Figure 3denoting that indeed the interface atoms should be easier toremove. We have only investigated the removal of sulfur atomssince the formation energy of a molybdenum vacancy isexpected to be much larger (by 4 eV) than that for sulfur.46

Electronic Properties. The precise fabrication of electronicdevices built from hybrid materials such as the 1T/2H MoS2monolayer considered here15,22 requires a deep understandingof how the electronic structure could be affected by changingthe size, as well as the shape, of the 2H and 1T domains in theMoS2 heterostructure. We have therefore investigated theinfluence of the width of the embedded 1T-MoS2 on theelectronic properties of the hybrid monolayer. Through theanalysis of the electronic density of states (DOS) (shown onlyfor the largest width, see Figure 4), we find that for all the

different widths of the 1T ribbon considered here, the hybridmonolayer shows a metallic behavior.The total electronic DOS of the largest hybrid MoS2

considered here (see Figure 2b) is depicted in Figure 4a.Results are shown only for this structure, as the relevantproperties for the smaller structures are qualitatively similar.The top and bottom panels of this graph display the projectedand partial DOS (PDOS) of the same system. From Figure 4a,it can be seen that in the vicinity of the Fermi level the largestcontribution comes from the 1T phase. In addition, acomparison of the Mo and S contributions reveals that inboth phases, the largest contribution originates from the Morather than the S atoms. Regarding the different orbitalcontributions to the electronic structure, it is known from theliterature that in 2H MoS2 both valence and conducting bandsare mainly dominated by the 4d states of Mo with a slightcontribution from the 3p states of S.47 For the single-phasemetallic 1T-MoS2, the 4d states of Mo and the 3p states of Scontribute to the electronic states in the vicinity of the Fermilevel.47 The projected DOS decomposition with respect to the3p and 4d orbitals of the S and Mo atoms in the upper graph ofFigure 4a corroborate with previous observations. Indeed,around the Fermi level (from −0.5 to +0.5 eV), the 4d orbitalsof Mo and the 3p orbitals of S from the 1T part of themonolayer dominate. The respective orbitals of the 2H phasebecome more important further away from the Fermi level andare connected to the band gap edges of this phase.An additional important feature is related to the interface. As

mentioned previously, by construction, the interfaces betweenthe 2H and 1T phases in our hybrid MoS2 model are notsymmetric. This asymmetry prompted us to investigate thepartial DOS and the contributions of the Mo and S atoms ofthe 1T and 2H phases close to both interfaces. The differentatoms/regions included in the analysis are shown in Figure 4b.The hybrid structure is divided into three regions, the 2Hphase, the top 1T, and the bottom 1T phase. The PDOS of theMo and S atoms in the different parts of the hybrid monolayer

Figure 4. Density of states of the largest relaxed structure (66 atoms/unit cell) from Figure 1d. (a) The upper graph shows the total DOS and theprojected DOS with the contributions from the 3p states in S and the 4d states in Mo in both 1T and 2H phases in the hybrid MoS2. In the lowergraph, the partial DOS based on the contributions of the S and Mo atoms of the 2H and 1T phases close to the two (top and down) interfaces aregiven. The legends correspond to the structure in b, which shows two unit cells of the largest structure in Figure 1d. The two dotted vertical lines inthe lower graph indicate the energy interval (from −0.64 eV to −0.22 eV) used for the analysis of the local currents in the Electronic Transportsection. In all cases, the Fermi level (EF) is aligned with zero on the energy axis.



D


on both sides of the two interfaces are given in the lower panelof Figure 4a. It becomes clear that atoms from the top anddown part of the 1T ribbon make different contributions to thetotal DOS. From this figure, we note that the down part of the1T region contributes more to the total DOS than the top partof 1T close to the Fermi level, a clear confirmation of theinterface asymmetry. The contributions of the Mo atoms in allparts of the system are higher than those from S, as expected. Itis also clear that more states are associated with both Mo and Satoms of the down 1T part than in the case of the top 1T part.In view of the transport properties, which we will discuss next,this feature would relate to additional electronic pathways forthe movement of electrons along the down interface of thematerial compared to the top interface.Electronic Transport. We next turn to the calculation of

the electronic transport along the hybrid MoS2 monolayer.Previous theoretical studies only considered the coexistence ofthe 1T and 2H phases within the same monolayer.19,21 Thestudies did not investigate a single 1T ribbon embedded in the2H phase which more closely resembles the situation inexperimental studies15 that aim for the realization of electronictransport devices. In fact, the periodic boundary conditionsused in the DFT calculations induce a potential repeatingacross the periodic images of the hybrid MoS2 structure. Withinthe transport simulations, no such periodicity is assumed.Accordingly, the DOS from the transport calculations reveals asmall gap unlike the results from the periodic DFT simulations.For the structure in Figure 1e, the calculated electronic

transmission curves are drawn in Figure 5a. No transmission of

the electronic current is observed for the interval of about ±0.1eV around the Fermi level, consistent with the gap in the totalDOS in Figure 5b. The transport calculations were performedalso by considering leads and scattering regions larger than theones shown in Figure 1e and have led to similar results. Thissuccessful convergence test shows that the results presentedhere do not change when the electrode is made larger, as in a

typical hybrid MoS2 electronic device. Overall, the transmissionprofile close to the Fermi level is well structured, having manypeaks denoting a large transmission at the respective energies.In the inset of Figure 5a, the transmission curves for the single-phase MoS2, both the 1T and the 2H, are given for comparison.A clear metallic behavior for the 1T phase is shown, while the2H phase shows zero transmission between roughly −1.0 and1.0 eV. The transmission function of the hybrid MoS2 shows astepwise behavior indicating a ballistic transport. Although thetransmission of single-phase MoS2 shows a smoother behavior,the system is also ballistic. In the simulations, increasing thenumber of k-points perpendicular to the transport directionleads to a decrease of the transmission steps until reaching thelimit of the smooth curve. The length of the scattering regionalong the transport direction for the systems studied here ismuch smaller (≈22 Å) than the electron mean free path inMoS2.

48 Accordingly, one might expect a ballistic transport forthese systems. A similar behavior would be expected even if thedevice length was increased by several factors. Note that thegap in the hybrid case is much smaller than that in the 2H case.Similar to the small DOS discrepancy between the periodic andthe finite structures in Figure 4 and Figure 5b, this mightindicate that in the limit of a large width of the 1T phase, thegap in the transmission T(E) might close.To understand the transmission curves in Figure 5a and to

unveil the microscopic mechanism behind them, the localcurrents transmitted across the hybrid system are visualized forthe energy values −0.64 eV (Figure 6a) and −0.22 eV (Figure

6b). The local currents at zero bias essentially map thetransmittance projection between two sites and can be obtainedfrom the non-equilibrium Greens functions and the trans-mission using a Keldysh formalism.38 Details on the fullderivation of the local currents are described elsewhere.49,50

Comparison of the two energy values reveals a qualitativedifference in the electronic transmission across the hybridMoS2. The energy value of −0.64 eV corresponds to a highertransmission than at −0.22 eV (referring to the dotted lines inFigure 5a). The corresponding local currents for these twoenergies also show different characteristics. Interestingly,

Figure 5. (a) The transmission spectra for the hybrid monolayer ofFigure 1e. The vertical dotted lines at −0.64 eV and −0.22 eVcorrespond to the energies used in the analysis of the wave functionsand to the local currents shown in Figure 6. The inset plots thetransmission as a function of energy (in units of eV) for the single-phase 1T and 2H MoS2 monolayer. (b) The total DOS for the samematerial. All data shown in this figure were obtained from NEGFcalculations.

Figure 6. Local currents for the setup shown in Figure 1e are depicted.The results are shown for two energies corresponding to the vicinity oftransmission peaks (see Figure 5a) at −0.64 eV (left panel) and −0.22eV (right panel).



E


current is not flowing along the whole 1T region as wasexpected. The local currents are rather associated with specificparts in the 1T ribbon. Figure 6a shows a significant currentacross the top interface and the top half of the 1T region. Theamount of local currents is less on the lower part of the 1Tphase. This is confirmed by the zero bias scattering stateeigenchannel wave functions (EWFs) (not shown). These wavefunctions are not equally spread across the metallic regionalong the transport direction. At this energy, the different signsof the EWFs on the bottom 1T part cancel out, so that there isonly a small amount of local current in this region. At −0.64 eV,there is current flowing across a different part of the metallicphase in the hybrid system compared to the case at −0.22 eV.There is more current flowing across the upper half of themetallic region and less current in the lower part at −0.64 eV.At −0.22 eV (Figure 6b), the EWFs are spread out across thelower part of the 1T material and the bottom interfacesupporting the finding that local currents are found at thebottom interface. The results indicate that at both energies,both Mo and S atoms are involved in promoting the flow of theelectrons and that electron motion is indeed associated mainlywith the metallic region and the interface, while the 2H phasevirtually does not participate. We have chosen two energyvalues which link the electronic properties shown in Figure 4 tothe asymmetry in the structure and transport properties inFigure 6 and the EWFs. Other energies should lead to differentcontributions to these properties.One can observe that by construction the bottom and the

top interfaces are not symmetric to each other (see Figures 1,4b) as mentioned in the Structural Characteristics and DefectAnalysis section on the structural characteristics of the hybridMoS2. Accordingly, depending on the energy, only one of thetwo interfaces is strongly related to the transport. Across thatinterface, a layer of Mo and S atoms from the 2H region isinvolved in the transport. This is evident from the localcurrents. Nevertheless, we have observed that the energy (orgating voltage) controls which parts of the metallic regions areinvolved in the transmission. As a way to unveil the mechanismwhich leads to a transmission along part of the metallic region,we turn again to the partial DOS in the lower graph of Figure4a. Because of the asymmetry of the interface, the down part ofthe 1T ribbon has shown a larger contribution to the PDOSclose to the Fermi level than the top 1T part (green and orangelines in the lower graph in Figure 4a). This observation canexplain the non-homogeneous current flow through the 1Tribbon. In addition, the higher DOS at −0.64 eV compared to−0.22 eV justifies why the currents are localized in the upper1T part in the former case and the lower 1T part in the lattercase as visualized in Figure 6. Nevertheless, on the interface ateach energy value, a current accumulation due to interfacestates has been observed. This analysis shows the strongdependence of the electron transport along the hybrid MoS2monolayer on its PDOS and the available electronic states atdifferent parts of the material. Accordingly, these results suggestthat a careful design of the hybrid interface would selectivelytune the electronic transport along the hybrid monolayer andwould control which part of the embedded ribbon is involvedin the transmission. Such a design would link to the exactexperimental gating voltages leading to the desired transmissionpattern.

■ SUMMARY

In this work, we have probed the structural, electronic, andtransport properties, as well as the stability of hybrid MoS2monolayers. These hybrid systems are composed of a 1Tmetallic ribbon embedded in the 2H MoS2 phase forming aninterface with armchair edges. We have varied the size of theembedded metallic ribbon in order to assess its influence on theproperties of the hybrid system and have obtained convergedproperties. We have observed the formation of a stable, planarinterface which includes atomic sites of different stabilities.Following a single sulfur vacancy analysis along theheterostructure, we have identified the unstable regions of thehybrid system, which are located at the armchair interface. Theelectronic properties of the hybrid monolayer showed a clearmetallic character, as the 1T part of the heterostructureintroduces states in the electronic band gap of the 2H region.Clear electronic transmission signals across the structure, whichbecome negligible very close to the Fermi energy, have beenobserved and are attributed to the finite structures in thecalculations. Extrapolation to a large periodic hybrid MoS2 isexpected to promote electron transmission also close to theFermi level. Visualization of the local currents at differentenergies clearly manifests the flow of electrons across themetallic part and the interface of the material. The exactfeatures vary with the exact energy value and the asymmetry ofthe interface. The local currents have verified the stronginfluence of interface states on the transmission spectra.Interpretation of our results in view of the current high

interest in 2D MoS2 and in general on TMD structures wouldlead to 2D materials with tunable properties. This would be aspecial feature of 2D TMDs as opposed to graphene.Specifically, the variable properties of hybrid MoS2 wouldallow for a selective design of electronic devices as controllingthe interface between the semiconducting and the embeddedmetallic phase would lead to a modulation of the transmittedcurrent. Tuning the interface would control not only theamount of current flowing along a device but also the part ofthe device that is conducting. A further thorough investigationof this polymorphicity, as proposed through our work, wouldlead to new materials and novel applications in nanoelectronicswell beyond those based on graphene. Specifically, our pointdefect investigation is highly relevant to the nanoporeformation using an electrochemical reaction process in MoS2nanopores and the subsequent detection of translocatingDNA.26 Within this concept, our analysis provides a deeperunderstanding on the earliest stages of the nanopore formationprocess and a qualitative insight into the direction of poregrowth. The results denote that in order to begin a pore-opening process in a hybrid MoS2 monolayer, first a sulfur atomat the interface needs to be missing. Using then anelectrochemical reaction process, the pore opening can beinitiated in the neighborhood of the single point defect. Weconclude that on the basis of our results, the pore growth istherefore likely to start at the interface and to move favorably inthe direction of the 1T phase/interface rather than toward the2H phase. Accordingly, our work provides important insightsregarding the pathways for selectively opening a nanopore inhybrid MoS2 leading to properties superior to those of single-phase MoS2. These results are expected to have a large impacton the fabrication of novel nanopores, biosensors, and tunableelectronic devices.



F


■ AUTHOR INFORMATION

Corresponding Authors*E-mail: [email protected]; phone: +49 711/685-63935.*E-mail: [email protected]; phone: +4618-4715873.

Author Contributions∥Equally contributing authors.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

G. S. and M. F. acknowledge financial support from thecollaborative network SFB 716 “Dynamic simulations ofsystems with large particle numbers” funded by the GermanFunding Agency (Deutsche Forschungsgemeinschaft-DFG).The initial benchmarks calculations were performed on thecomputational resource ForHLR Phase I funded by theMinistry of Science, Research, and the Arts Baden-Wurttem-berg (MWK) and German funding agency (DeutscheForschungsgemeinschaft-DFG). This research was supportedin part by the bwHPC initiative and the bwHPC-C5 project(http://www.bwhpc-c5.de) funded by the MWK and DFG andprovided through associated computer services of the JUSTUSHPC facility at the University of Ulm. R. G. A. acknowledgesfinancial support from Carl Tryggers Stiftelse and R. H. S.thanks the Swedish Research Council. W. L. S acknowledgesBrazilian agency FAPES. F. A. L. S. thanks the CAPESFoundation (Ministry of Education of Brazil, Brasilia-DF70040-020) for a scholarship (Process No. 1640/14-3).

■ REFERENCES(1) Lebegue, S.; Eriksson, O. Electronic Structure of Two-Dimensional Crystals from Ab Initio Theory. Phys. Rev. B: Condens.Matter Mater. Phys. 2009, 79, 115409.(2) Lebegue, S.; Bjorkman, T.; Klintenberg, M.; Nieminen, R. M.;Eriksson, O. Two-Dimensional Materials from Data Filtering and AbInitio Calculations. Phys. Rev. X 2013, 3, 031002.(3) Mas-Balleste, R.; Gomez-Navarro, C.; Gomez-Herrero, J.;Zamora, F. 2D Materials: To Graphene and Beyond. Nanoscale2011, 3, 20−30.(4) Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F. AtomicallyThin MoS2: A New Direct-Gap Semiconductor. Phys. Rev. Lett. 2010,105, 136805.(5) Coleman, J. N.; Lotya, M.; O’Neill, A.; Bergin, S. D.; King, P. J.;Khan, U.; Young, K.; Gaucher, A.; De, S.; Smith, R. J.; et al. Two-dimensional nanosheets produced by liquid exfoliation of layeredmaterials. Science 2011, 331, 568−571.(6) Wang, K.; Wang, J.; Fan, J.; Lotya, M.; O’Neill, A.; Fox, D.; Feng,Y.; Zhang, X.; Jiang, B.; Zhao, Q.; et al. Ultrafast Saturable Absorptionof Two-Dimensional MoS2 Nanosheets. ACS Nano 2013, 7, 9260−9267.(7) Ambrosi, A.; Sofer, Z.; Pumera, M. 2H → 1T Phase Transitionand Hydrogen Evolution Activity of MoS2, MoSe2, WS2 and WSe2Strongly Depends on the MX2 Composition. Chem. Commun. 2015,51, 8450−8453.(8) Feldman, Y.; Wasserman, E.; Srolovitz, D. J.; Tenne, R. High-Rate, Gas-Phase Growth of MoS2 Nested Inorganic Fullerenes andNanotubes. Science 1995, 267, 222−225.(9) Lauritsen, J.; Bollinger, M.; Lægsgaard, E.; Jacobsen, K.; Nørskov,J.; Clausen, B.; Topsøe, H.; Besenbacher, F. Atomic-Scale Insight intoStructure and Morphology Changes of MoS2 Nanoclusters inHydrotreating Catalysts. J. Catal. 2004, 221, 510−522.

(10) McBride, K. L.; Head, J. D. DFT Investigation of MoS2Nanoclusters Used as Desulfurization Catalysts. Int. J. QuantumChem. 2009, 109, 3570−3582.(11) Cristol, S.; Paul, J. F.; Payen, E.; Bougeard, D.; Clemendot, S.;Hutschka, F. Theoretical Study of the MoS2 (100) Surface: AChemical Potential Analysis of Sulfur and Hydrogen Coverage. 2.Effect of the Total Pressure on Surface Stability. J. Phys. Chem. B 2002,106, 5659−5667.(12) Huang, M.; Cho, K. Density Functional Theory Study of COHydrogenation on a MoS2 Surface. J. Phys. Chem. C 2009, 113, 5238−5243.(13) Alexiev, V.; Prins, R.; Weber, T. DFT Study of MoS2 andHydrogen Adsorbed on the (101¯0) Face of MoS2. Phys. Chem. Chem.Phys. 2001, 3, 5326−5336.(14) Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.;Strano, M. S. Electronics and Optoelectronics of Two-DimensionalTransition Metal Dichalcogenides. Nat. Nanotechnol. 2012, 7, 699−712.(15) Lin, Y.-C.; Dumcenco, D. O.; Huang, Y.-S.; Suenaga, K. AtomicMechanism of the Semiconducting-to-Metallic Phase Transition inSingle-Layered MoS2. Nat. Nanotechnol. 2014, 9, 391−396.(16) Song, I.; Park, C.; Choi, H. C. Synthesis and Properties ofMolybdenum Disulphide: from Bulk to Atomic Layers. RSC Adv.2015, 5, 7495−7514.(17) Voiry, D.; Mohite, A.; Chhowalla, M. Phase Engineering ofTransition Metal Dichalcogenides. Chem. Soc. Rev. 2015, 44, 2702−2712.(18) Eda, G.; Yamaguchi, H.; Voiry, D.; Fujita, T.; Chen, M.;Chhowalla, M. Photoluminescence from Chemically Exfoliated MoS2.Nano Lett. 2011, 11, 5111−5116.(19) Eda, G.; Fujita, T.; Yamaguchi, H.; Voiry, D.; Chen, M.;Chhowalla, M. Coherent Atomic and Electronic Heterostructures ofSingle-Layer MoS2. ACS Nano 2012, 6, 7311−7317.(20) Qiao, W.; Yan, S.; Song, X.; Zhang, X.; He, X.; Zhong, W.; Du,Y. Luminescent monolayer MoS2 Quantum Dots Produced by Multi-Exfoliation Based on Lithium Intercalation. Appl. Surf. Sci. 2015, 359,130−136.(21) Guo, X.; Yang, G.; Zhang, J.; Xu, X. Structural, Mechanical andElectronic Properties of In-Plane 1T/2H Phase Interface of MoS2Heterostructures. AIP Adv. 2015, 5, 097174.(22) Kappera, R.; Voiry, D.; Yalcin, S. E.; Branch, B.; Gupta, G.;Mohite, A. D.; Chhowalla, M. Phase-Engineered Low-ResistanceContacts for Ultrathin MoS2 Transistors. Nat. Mater. 2014, 13, 1128−1134.(23) Liu, K.; Feng, J.; Kis, A.; Radenovic, A. Atomically ThinMolybdenum Disulfide Nanopores with High Sensitivity for DNATranslocation. ACS Nano 2014, 8, 2504−2511.(24) Feng, J.; Liu, K.; Graf, M.; Lihter, M.; Bulushev, R. D.;Dumcenco, D.; Alexander, D. T. L.; Krasnozhon, D.; Vuletic, T.; Kis,A.; et al. Electrochemical Reaction in Single Layer MoS2: NanoporesOpened Atom by Atom. Nano Lett. 2015, 15, 3431−3438.(25) Farimani, A. B.; Min, K.; Aluru, N. R. DNA Base DetectionUsing a Single-Layer MoS2. ACS Nano 2014, 8, 7914−7922.(26) Feng, J.; Liu, K.; Bulushev, R. D.; Khlybov, S.; Dumcenco, D.;Kis, A.; Radenovic, A. Identification of Single Nucleotides in MoS2Nanopores. Nat. Nanotechnol. 2015, 10, 1070−1076.(27) Building a Better Nanopore. Nat. Nanotechnol. 2016, 11, 105;editorial.10.1038/nnano.2016.15(28) Branton, D.; Deamer, D. W.; Marziali, A.; Bayley, H.; Benner, S.A.; Butler, T.; Di Ventra, M.; Garaj, S.; Hibbs, A.; Huang, X.; et al. ThePotential and Challenges of Nanopore Sequencing. Nat. Biotechnol.2008, 26, 1146−1153.(29) Cho, B.; Hahm, M. G.; Choi, M.; Yoon, J.; Kim, A. R.; Lee, Y.-J.;Park, S.-G.; Kwon, J.-D.; Kim, C. S.; Song, M.; et al. Charge-Transfer-Based Gas Sensing Using Atomic-layer MoS2. Sci. Rep. 2015, 5, 8052.(30) Liu, G.; Rumyantsev, S. L.; Jiang, C.; Shur, M. S.; Balandin, A. A.Selective Gas Sensing With h-BN Capped MoS2 HeterostructureThin-Film Transistors. IEEE Electron Device Lett. 2015, 36, 1202−1204.



G

mailto:[email protected]

mailto:[email protected]

http://www.bwhpc-c5.de


(31) Perkins, F. K.; Friedman, A. L.; Cobas, E.; Campbell, P.;Jernigan, G.; Jonker, B. T. Chemical Vapor Sensing with MonolayerMoS2. Nano Lett. 2013, 13, 668−673.(32) 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. ACSNano 2013, 7, 4879−4891.(33) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Phys.Rev. 1964, 136, B864−B871.(34) Kohn, W.; Sham, L. J. Self-Consistent Equations IncludingExchange and Correlation Effects. Phys. Rev. 1965, 140, A1133−A1138.(35) Soler, J. M.; Artacho, E.; Gale, J. D.; García, A.; Junquera, J.;Ordejon, P.; Sanchez-Portal, D. The SIESTA Method for Ab InitioOrder-N Materials Simulation. J. Phys.: Condens. Matter 2002, 14,2745.(36) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized GradientApproximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868.(37) Troullier, N.; Martins, J. L. Efficient pseudopotentials for plane-wave calculations. Phys. Rev. B: Condens. Matter Mater. Phys. 1991, 43,1993−2006.(38) Brandbyge, M.; Mozos, J.-L.; Ordejon, P.; Taylor, J.; Stokbro, K.Density-functional method for nonequilibrium electron transport.Phys. Rev. B: Condens. Matter Mater. Phys. 2002, 65, 165401.(39) Ding, Y.; Wang, Y.; Ni, J.; Shi, L.; Shi, S.; Tang, W. FirstPrinciples Study of Structural, Vibrational and Electronic Properties ofGraphene-Like MX2 (M= Mo, Nb, W, Ta; X= S, Se, Te) monolayers.Phys. B 2011, 406, 2254−2260.(40) Johari, P.; Shenoy, V. B. Tuning the Electronic Properties ofSemiconducting Transition Metal Dichalcogenides by ApplyingMechanical Strains. ACS Nano 2012, 6, 5449−5456.(41) Kan, M.; Wang, J.; Li, X.; Zhang, S.; Li, Y.; Kawazoe, Y.; Sun, Q.;Jena, P. Structures and Phase Transition of a MoS2 Monolayer. J. Phys.Chem. C 2014, 118, 1515−1522.(42) Duerloo, K.-A. N.; Li, Y.; Reed, E. J. Structural PhaseTransitions in Two-Dimensional Mo- and W-Dichalcogenide Mono-layers. Nat. Commun. 2014, 5, No. 10.1038/ncomms5214,DOI: 10.1038/ncomms5214.(43) Hu, Z.; Zhang, S.; Zhang, Y.-N.; Wang, D.; Zeng, H.; Liu, L.-M.Modulating the Phase Transition Between Metallic and Semi-conducting Single-Layer MoS2 and WS2 Through Size Effects. Phys.Chem. Chem. Phys. 2015, 17, 1099−1105.(44) Heising, J.; Kanatzidis, M. G. Structure of Restacked MoS2 andWS2 Elucidated by Electron Crystallography. J. Am. Chem. Soc. 1999,121, 638−643.(45) Santosh, K.; Longo, R. C.; Addou, R.; Wallace, R. M.; Cho, K.Impact of Intrinsic Atomic Defects on the Electronic Structure ofMoS2 Monolayers. Nanotechnology 2014, 25, 375703.(46) Hong, J.; Hu, Z.; Probert, M.; Li, K.; Lv, D.; Yang, X.; Gu, L.;Mao, N.; Feng, Q.; Xie, L. Exploring Atomic Defects in MolybdenumDisulphide Monolayers. Nat. Commun. 2015, 6, 6293.(47) Tang, Q.; Jiang, D. Stabilization and Band-Gap Tuning of the1T-MoS2 Monolayer by Covalent Functionalization. Chem. Mater.2015, 27, 3743−3748.(48) Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, i. V.; Kis,A. Single-layer MoS2 transistors. Nat. Nanotechnol. 2011, 6, 147−150.(49) Paulsson, M.; Brandbyge, M. Transmission Eigenchannels fromNonequilibrium Green’s Functions. Phys. Rev. B: Condens. MatterMater. Phys. 2007, 76, 115117.(50) Okabayashi, N.; Paulsson, M.; Ueba, H.; Konda, Y.; Komeda, T.Inelastic Tunneling Spectroscopy of Alkanethiol Molecules: High-Resolution Spectroscopy and Theoretical Simulations. Phys. Rev. Lett.2010, 104, 077801.



H

http://dx.doi.org/10.1038/ncomms5214


electronic transport along hybrid mos monolayers

Documents