This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

This content was downloaded by: nwilson

IP Address: 137.205.165.171

This content was downloaded on 17/06/2015 at 10:40

Please note that terms and conditions apply.

van der Waals epitaxy of monolayer hexagonal boron nitride on copper foil: growth,

crystallography and electronic band structure

View the table of contents for this issue, or go to the journal homepage for more

2015 2D Mater. 2 025003

(http://iopscience.iop.org/2053-1583/2/2/025003)

Home Search Collections Journals About Contact us My IOPscience

2DMater. 2 (2015) 025003 doi:10.1088/2053-1583/2/2/025003

PAPER

van der Waals epitaxy of monolayer hexagonal boron nitride oncopper foil: growth, crystallography and electronic band structure

Grace EWood1, Alexander JMarsden1, James JMudd1,MarcWalker1,MariaAsensio2, JoseAvila2, Kai Chen2,GavinRBell1 andNeil RWilson1

1 Department of Physics, University ofWarwick, Coventry CV4 7AL,UK2 Synchrotron Soleil, L’Orme desMerisiers, Saint-Aubin, France

E-mail: Neil.Wilson@warwick.ac.uk

Keywords: monolayer hexagonal boron nitride, chemical vapour deposition, epitaxy, structural feedback, angle-resolved photoemissionspectroscopy

AbstractWe investigate the growth of hexagonal boron nitride (h-BN) on copper foil by low pressure chemicalvapour deposition (LP-CVD). At lowpressure, h-BN growth proceeds through the nucleation andgrowth of triangular islands. Comparison between the orientation of the islands and the localcrystallographic orientation of the polycrystalline copper foil reveals an epitaxial relation between thecopper and h-BN, even onCu(100) andCu(110) regions whose symmetry is notmatched to the h-BN.However, the growth rate is faster and the islandsmore uniformly oriented onCu(111) grains. Angleresolved photoemission spectroscopymeasurements reveal awell-defined band structure for the h-BN, consistent with a band gap of 6 eV, that is decoupled from the copper surface beneath. Theseresults indicate that, despite aweak interaction between h-BN and copper, van derWaals epitaxydefines the long range ordering of h-BN even on polycrystalline copper foils and suggest that largearea, single crystal,monolayer h-BN could be readily and cheaply produced.

1. Introduction

Interest inmonolayer hexagonal boron nitride (h-BN)has been stimulated by its complementarity to gra-phene, with demonstrations of its use as a dielectricsubstrate to improve the mobility of graphene or as atunnel barrier to add novel functionality to graphenedevices [1]. It is iso-structural to graphene, with alattice parameter only 1.8% larger, but is a wide bandgap semiconductor rather than a semi-metal. As withgraphene, the proof of principle experiments haveutilized mechanically exfoliated h-BN flakes, but largescale application will require scalable synthesis of largearea, high qualitymaterial.

Chemical vapour deposition (CVD) is proving tobe a promising technique to fulfil this need. CVD ofmonolayer h-BN on single crystal substrates in ultrahigh vacuum (UHV) was reported in the 1990s [2],but due to the high cost of the substrate and UHVgrowth chamber it is not a practical source of materialfor applications. Following shortly after the CVDgrowth of graphene on copper foil [3], recent reportsof h-BN growth on low cost substrates have shown the

potential for CVD to be a low cost route to high qualitymonolayer or few layer h-BN [4–11]. CVD growth ofh-BN proceeds through nucleation and growth ofislands that merge to form a continuous film. As aresult, similarly to graphene growth [12], CVD grownh-BN can form large area but polycrystalline films[10]. The grain boundaries within these films have dis-tinct properties from the pristine layers which, thoughinteresting in their own right [13], are undesirable forapplications such as the use of h-BN as a dielectricsubstrate.

In layeredmaterials such as graphene/graphite andtransition metal dichalcogenides, the inter-layerattraction arises through dispersion forces (van derWaals bonding) rather than strong chemical bonds.Unlike conventional materials, they do not have out-of-plane dangling bonds at their surface: in conven-tional epitaxy, the dangling bonds on the substratematerial result in the formation of strong chemicalbonds that require a close lattice-match between sub-strate and growth material. By contrast, layered mate-rials tend to interact with a surface of even a non-layered material via van der Waals bonding. As a

OPEN ACCESS

RECEIVED

20 January 2015

REVISED

6March 2015

ACCEPTED FOR PUBLICATION

23March 2015

PUBLISHED

6May 2015

Content from this workmay be used under theterms of theCreativeCommonsAttribution 3.0licence.

Any further distribution ofthis workmustmaintainattribution to theauthor(s) and the title ofthework, journal citationandDOI.

© 2015 IOPPublishing Ltd

result, epitaxial growth can occur for these two-dimensional systems despite lattice mismatch andeven without matching symmetry. This van der Waalsepitaxy [14] can result in high quality, defect freematerial where the crystallographic orientation of thegrown layer is still defined by that of the substrate. Forgraphene growth, it has been shown that under low-pressure conditions the orientation of the graphenegrains can be determined by the substrate crystal-lography even on copper foil [15–17]. Under UHVconditions on single crystal substrates, h-BN has beenshown to grow epitaxially on Cu(111), with the h-BNlattice orientationally aligned to the hexagonal Cu(111) surface lattice despite the significant (1.8%) lat-tice mismatch [18, 19]. However, to the best of ourknowledge, there have been no systematic studies onwhether this epitaxial alignment persists in the moretechnologically relevant CVD growth on low cost cop-per substrates where the surface is often not Cu(111).Understanding and controlling the epitaxial growth ofh-BN should allow large area, monolayer, single crys-tal h-BN to be grown without the presence of grainboundaries yielding higher qualitymaterial.

In this paper we demonstrate the low pressureCVD (LP-CVD) growth of h-BN from ammonia bor-ane feedstock. Scanning electron microscopy (SEM)shows that independently nucleated islands large dis-tances from one another are frequently aligned in thesame orientation. Determination of the copper crys-tallography by electron backscatter diffraction(EBSD), correlated to quantification of the islandorientations, suggests that this alignment is due tointeraction with the copper surface. It is noticeablethat the epitaxial alignment is more prominent on Cu(111) regions and that the typical island sizes there arelarger. AFM shows that the copper surface restructuresand facets under the monolayer h-BN, but angleresolved photoemission spectroscopy (ARPES) showsthe band structure of the h-BN is electronically decou-pled from the copper indicating only a weakinteraction.

2.Methods

LP-CVD of h-BN. Copper foils (99.5% purity, 0.025mm thick, Alfa Aesar 46365) were cleaned by electro-polishing (10 s at 5 V, ∼1.5 A) in an electrolytecontaining orthophosphoric acid and urea, followed

by rinsing in acetone and isopropanol. The foil wasloaded to the centre of a 1 inch diameter quartz worktube inside a split tube furnace, evacuated down to abase pressure of < −10 7 mbar using a turbo pumpbacked by a dry diaphragm pump, then heated to

°1000 C under a flow of 10 standard cubic centimeters(sccm) of hydrogen (99.9995%). The foil was annealedfor 10 min at °1000 C prior to growth. 15 mg ofammonia borane (97% purity, Sigma Aldrich 682098)was placed in a quartz boat in a small chamber,separated from the furnace by a valve, and indepen-dently heated using heating tape. Growth was initiatedby opening the valve to the ammonia borane, nohydrogen was flowed during growth. After growth,hydrogen flowwas resumed and the furnace cooled to< °200 C before removing the foil.

SEM, EBSD and AFM. SEM images were taken ona Zeiss Supra 55-VP FEGSEM operating at 10 kV withthe Inlens detector. The same instrument was used forEBSD measurements, using an EDAX EBSD system.AFM images were taken with an Asylum ResearchMFP3D-SA operated inAC (tapping)mode.

Photoemission. ARPES and XPS measurementswere taken at the Antares beamline at SynchrotronSoleil using a Scienta R4000 electron analyser with 2Ddetection [20]. A photon energy of 700 eV was usedfor XPSmeasurements, and of 100 eV for ARPESmea-surements. The binding energy scale was calibratedusing the copper Fermi edge.

3. Results and discussions

A schematic of the experimental set-up used here forLP-CVD of h-BN is shown in figure 1, similar to thatused by Ismach et al [4]. Ammonia borane was used asa stoichiometric source of boron and nitrogen; it wascontained in a small chamber, separated from thegrowth furnace by a valve, with independent tempera-ture control.

As shown in the SEM images in figure 2, character-istically triangular h-BN islands nucleate indepen-dently and grow, with the average island sizeincreasing with growth time. Under the conditionsused here, the islands start to coalesce after around20 min and a continuous h-BN sheet is formed after60 min. At this stage, small white triangles and whitelines can be observed in the SEM images, indicative ofmulti-layer regions and wrinkles respectively. The

Figure 1. Schematic of the LP-CVDgrowth setup.

2

2DMater. 2 (2015) 025003 GEWood et al

typical island size here when coalescence starts isμ∼10 m or μ∼50 m2. The growth rate appears to

decrease over time, consistent with a copper surfacecatalysed reaction.

The triangular islands are strongly indicative of h-BN growth, consistent with zigzag nitrogen termi-nated edges as the most energetically favourable con-figuration under these conditions [21]. However, itshould be noted that under other growth conditions,e.g. higher partial pressures of the active B-N species,the island shape can deviate from triangular, withreports for example of hexagonal islands in atmo-spheric pressure CVD of h-BN on copper foils [6, 7].Here the presence of h-BN is confirmed by XPS mea-surements (see supporting information). Even thoughwe have not performed a systematic optimization ofthe growth conditions, the island size here is similar toor larger than that reported by Tay et al [6]. In agree-ment with their results, we found that electropolishingthe copper foil prior to use was essential for repro-ducible growth of such larger islands.

Previous reports of h-BN growth from ammoniaborane precursor have found significant deposition ofBN particulates, which could be mostly removed byfilters between the ammonia borane and growth fur-nace [9]. However, no particulates are apparent inthese SEM images and this is further verified by AFM(see later). The lower ammonia borane temperatureused here, and the separation of the ammonia boraneto a small side chamber, result in a clean growth.

A striking feature of the triangular islands infigure 2 is that despite nucleating independently, i.e.the nucleation of one island is not determined byanother, they often have the same orientation. Priorreports have also noted that the triangles appear to bealigned, but have not investigated this effect fur-ther [9, 17].

In the SEM image shown in figure 3(a), triangularh-BN islands can be seen on three copper grains. Thedarker, more circular regions are graphene: unfortu-nately, we occasionally found that h-BN growth wascontaminated with graphene, despite no hydrocarbon

Figure 2. SEMof LP-CVDgrowthwith growth times of (a) 10 min, (b) 15 min, (c) 20 min, (d) 30 min, (e) 40 min and (f) 60 min.

3

2DMater. 2 (2015) 025003 GEWood et al

being deliberately introduced to the system. Strudwicket al have shown that copper foils can contain smallamounts of carbon that during annealing in hydrogencan precipitate as graphene, with the amounts varyingfrom foil to foil [22]. This could explain the growth ofgraphene here without an external carbon source andis consistent with the graphene forming before h-BNgrowth as h-BN can be seen to grow from the grapheneislands but not the reverse. We note that most h-BNislands nucleate and grow separately from the gra-phene islands, and hence the presence of small frac-tions of graphene in some of these samples does notcompromise the conclusions drawn below on theorientations of the h-BN islands. We have found thatgraphene and h-BN can be easily distinguished in theSEM by their different contrast, particularly in multi-layer regions. In conventional secondary electrondetection images, wrinkles and multi-layer islands aredark for graphene but light for h-BN [23].

The dashed boxes on figure 3(a) mark sections ofthe image from the three different copper grains. Ineach, the triangular h-BN islands appear to be orien-tated in a consistent direction. Closer inspection sug-gests that the orientation of h-BN islands in themiddlecopper grain, marked by the dashed blue rectangle, isin a slightly different direction to the other two.

Image analysis was used to robustly and indepen-dently quantify the orientations of the triangular h-BNislands in the SEM images, using the directionalityplugin for the ImageJ analysis software [24]. Thismea-sures the orientations of edges within an image, pro-ducing histograms of the total amount (i.e. length) of

edges at a given angle. For a single equilateral trianglethis would result in three equal height peaks in the his-togram separated by 60°. The histograms infigures 3(b)–(d) show the edge orientations fromsections of the three different copper grains shown inthe corresponding SEM images. In each histogramthree peaks can be observed, separated by 60° as expec-ted, indicating the preferred orientation of the trian-gular islands in each image. However, from thehistograms it is clear that there is a roughly 10° differ-ence in the orientation of the islands in (c) comparedto those in (b) and (d). The observation that the pre-ferred orientation changes from one grain to the nextand back again is not consistent with the orientationbeing determined by the gas flow direction, as sug-gested by Han et al [9], but rather indicates that it isdue to interactionwith the substrate.

The importance of the interaction between h-BNand copper surface can also be seen in AFM topo-graphy images, as shown in figure 4. The copper sur-face has restructured under the h-BN monolayer,forming sharply angled corrugations. This interfacialrestructuring is similar to that observed for grapheneon copper [15]. As in the case of graphene on copper[15], for partial coverage of h-BN this structural feed-back facetting effect is only seen on areas of the samplecovered by h-BN. The faceting extends across largeareas, as shown in figures 4(c) and (f). It appears thatwhere the average surface plane is close to a crystal-lographic low-index face such as (100), the interactionwith h-BN stabilizes facetting to that (lower energy)crystallographic face such that the surface orientation

Figure 3. (a) SEM image of copper foil with triangular h-BN islands. Panels (b)–(d) showdirectionality analysis of the sections ofimage as indicated, revealing the preferred orientation of the triangular islands in each.

4

2DMater. 2 (2015) 025003 GEWood et al

is predominantly (100). A detailed analysis of the ener-getic balance for such structural feedback is beyondthe scope of the present paper.

EBSD was used to investigate the crystallographyof the copper foil and compared to SEM images todetermine the correlation between substrate

Figure 4.AFM topography images of h-BNon copper foil, the h-BNhere fully covers the graphene surface. Panels (d)–(f) are thecorresponding three-dimensional views of the surface topographies shown in (a)–(c).

Figure 5.EBSD and SEMof h-BNon copper.(a) SEM image and (b) EBSDof the same area.Marked on are the orientations of thegrains. Panels (c) and (d) are histograms of the edge orientations from selected areas of the SEM images, as indicated. The trianglesshown in (e) and (f) are indicative of the orientations of the islands, as shown by the dashed triangles in (c) and (d) respectively; thearrows show crystallographic directions extracted from the EBSDdata in (b) for the respective copper grains.

5

2DMater. 2 (2015) 025003 GEWood et al

crystallography and h-BN island orientation. EBSDshows that after growth the copper foil is polycrystal-line but with macroscopic copper grains with lateraldimensions of typically millimeters or larger. The sur-face orientation is mostly aligned near Cu(100), butwith Cu(111) microstripe inclusions (see supportinginformation), as previously reported for these copperfoils after graphene growth [15] and [25]. An exampleis shown in figure 5 where EBSD and SEM of the sameregion reveals correlation between the orientation ofthe triangular islands and the copper surface orienta-tion. EBSD reveals a Cu(111)microstripe between twovicinal Cu(100) regions. The h-BN islands are clearlylarger on theCu(111) region and are aligned relative toone another. The directionality analysis of this region,figure 5(d), quantitatively shows this. Comparisonbetween the typical triangle orientation and the sur-face crystallography shows that the islands are alignedwith one edge parallel to the Cu [1̄10] direction(extracted from the EBSD data), as shown infigure 5(f). Analysis of the island orientations on theCu(100) region is more complicated in this image, butcareful inspection shows that islands frequently havean edge parallel to the Cu [011̄]direction, as shown infigures 5(c) and (e).

The correlation between the crystallography of thecopper substrate and h-BN island orientation suggeststhat the copper surface is playing an important role.The growth of h-BN on copper is thought to involvesurface mediated processes, with evidence that there isalso some B diffusion through the bulk of the copper[11]. The islands nucleate and grow during the flow ofammonia-borane feedstock, whilst the copper foil is at

°1000 C. At this temperature, it is known that the cop-per surface is mobile [26], but despite this previousstudies of graphene growth on copper under similarconditions have found that the crystallography of the

copper surface still plays an important role in definingthe orientation of the graphene [15, 16].

In order to understand the relation between the h-BN island orientations and the surface crystal-lography, it is necessary to consider the copper surfacecrystallography. Figure 6 shows the atomic arrange-ments of the Cu(111), Cu(110) and Cu(100) surfaces,with h-BN islands overlaid. The bulk structure of cop-per is face centred cubic (fcc), with a room tempera-ture lattice parameter of =a 0.361Cu nm.

The Cu(111) surface has hexagonal symmetry,with surface lattice parameter of = =a a 2 0.255Cu

nm, as shown in figure 6(a). The in plane room tem-perature lattice parameter of h-BN is 0.250 nm, 1.8%smaller than the copper surface lattice. Density func-tional calculations have predicted that the N atomsreside on top of the surface metal atoms whereas the Batoms should lie in the hollow sites [27]. As figure 6 (a)shows, there are two possible orientations that satisfythis but which are distinct due to the second layer ofcopper atoms: on the fcc hollow site there is no Cuatom in the second layer, whilst by contrast in the hcpsite there is a Cu atom in the second layer immediatelybelow the hollow site. The h-BN island on the left-hand side of figure 6(a) shows the N on the hcp sitewhilst the island on the right-hand side has N on thefcc site. DFT calculations indicate that the fcc site ispreferable [27], although experimentally for h-BN Ni(111) it has been shown that the B atom can also stablysit on the hcp position [28]. The islands should havean edge parallel to the Cu [1̄10] direction as shown.The experimental results in figure 5 are thus consistentwith this picture and show that both the fcc and hcpsites can be stable in this case. Note that as Auwärteret al explained for h-BN on Ni(111) [28], although theh-BN lattice has the same orientation in the fcc andhcp case the two islands cannot be seamlessly joineddue to their different atomic registration and so linear

Figure 6.Atomic structure ofmonolayer h-BNon different orientations of copper surface, showing the top two atomic layers of eachcopper surface orientation. Panel (a) is for h-BNonCu(111), (b) h-BNonCu(110), and (c) h-BNonCu(100).

6

2DMater. 2 (2015) 025003 GEWood et al

defects would be expected to mark the boundariesbetween two such domains. We note that in generalcomparison of the substrate surface lattice with theoverlayer crystal structure is not the only factor thatcould define the orientation: for example, interactionsbetween the edges of a growing island and the sub-strate surface could strongly influence the geometry,particular for the small island sizes where the orienta-tion isfirst pinned.

On Cu(110), figure 6(b), the surface lattice is rec-tangular and so the match to the hexagonal h-BN isless clear. However, the rectangular Cu(110) surfacehas a periodicity of = =a a 2 0.255Cu nm in onedirection and hence can share a common periodicitywith the h-BN. This would result in two (equivalent)favoured orientations of the h-BN islands aligned intheCu [1̄00]/ Cu [100]directions as shown.

The square lattice of the Cu(100) surface, with lat-tice parameter again = =a a 2 0.255Cu nm, gives alattice match in four different directions as shown infigure 6(c). This common periodicity could result in 4,equivalent, preferred orientations of the triangular h-BN islands with edges parallel to the Cu[011̄] or Cu[011] direction as shown (i.e. rotated 90° relative toone another).

Figure 7 shows a large area of a substrate (morethan 1mmby around 0.5mm)which EBSD showed to

be a single copper grain with orientation vicinal Cu(100). The image was taken with a high pixel densityand magnified views of the two areas highlighted bydashed squares are shown in figures 7(b) and (c).Visual inspection of these images suggests that fourorientations are present. Directionality analysis of theentire SEM image in figure 7(a) is shown in figure 7(d)and clearly reveals a prevalence of edges separated by30° from one another, consistent with the four islandorientations rotated 90° from one another as shown infigure 6(c). Comparison with the EBSD data from thesame area indicated that the island orientations wereconsistent with the expected copper crystallographicdirections.

It is interesting to note that the epitaxial orienta-tions determined here are different from those pre-viously observed for LP-CVD of graphene on copperfoil [15], where the graphene lattice was found to ori-ent ±8° from the Cu[010] direction on Cu(100). Thelattice parameter of graphene at room temperature is0.246 nm, 3.6% less than aCu and 1.8% less than the h-BN lattice parameter. Accounting for thermal expan-sion, the lattice parameter of copper will be relativelylarger at the growth temperature (as evidenced by thewrinkles in graphene and h-BN after growth). Itappears that the difference in lattice parameterbetween graphene and h-BN is sufficient to result in

Figure 7. (a) Lowmagnification, high pixel density SEM image of h-BN islands on a singlemillimetre sized vicinal Cu(100) region ofcopper foil. Panels (b) and (c) showmagnified sections of the imagewith clearly resolved triangular h-BN islands. The directionalityhistogram in (d) was obtained from the entire image and clearly shows peaks in edge orientation separated by 30°.

7

2DMater. 2 (2015) 025003 GEWood et al

differences in their epitaxial alignment (i.e. misalignedfrom the high symmetry axis for graphene and alignedfor h-BN), although the general phenomena of mis-match epitaxy and structural feedback/interfacialrestructuring appear to be consistent for both gra-phene and h-BNon copper.

The clear epitaxial relation between h-BN and thecopper surface indicates an interaction between thetwo. In order to gain insight into the strength of thisinteraction, and to determine the electronic quality ofthe h-BN, ARPES was used to measure the band-structure of the h-BN monolayer on copper foil. Itshould be noted that these measurements were takenafter exposure to air, and hence some oxygen inter-calation may have taken place [11]. The samples wereannealed in UHV prior to measurement to return thesamples to close to pristine condition, as measured byXPS. Figure 8 shows the dispersion of the h-BN bandsmeasured on the copper foil. The orientation of thecopper surface in this region is not known, but as thesample is predominantly vicinal Cu(100) we expect itto be Cu(100). The high intensity between 2 and 4 eVis due to the copper d states. In figure 8(b) the data hasbeen modified by momentum distribution curve nor-malization (each line at constant energy is divided bythe average of that line), making the h-BN π band dis-persion more apparent. The h-BN valence band max-imum is in the copper d states, but can be seen to be∼3 eV below the Fermi level consistent with a 6 eVband gap and similar to that found for h-BN on singlecrystal Cu(111) by Roth et al [19]. The dispersion canbe approximately described by a nearest neighbourtight binding model, which is shown overlaid on the

experimental data in figure 8(c). The tight bindingmodel used is of the form

ε ε

ε ε

=+

−

−

+

+ +

( )( )

E k k

tk a

k a k a

,2

( )

2

4 cos2

cos3

2cos

2

1

4

x y

B N

B N

y

x y

2

2

2⎛⎝⎜⎜

⎛⎝⎜

⎞⎠⎟

⎞⎠⎟

with parameters: on site energies ε ε= − =B N 2.05 eV,hopping parameter =t 2.25 eV and lattice spacing

=a 0.250 nm. The two dashed lines overlaid corre-spond to dispersion relations in the ΓM direction(darker green and shorter dashes) and the ΓK direc-tion (lighter green and longer dashes). These representthe two limiting cases in the dispersion relation.

The dispersion of the h-BN is similar to that foundby Roth et al on single crystal Cu(111), despite the dif-ferent surface orientation of the copper here. The h-BN is electronically decoupled from the surface, main-taining its insulating character. This demonstrates thatthere is only a weak interaction between the coppersurface and physisorbed h-BNmonolayer.

4. Conclusions

As with graphene, the h-BN radically alters the surfacetopography and hence surface crystallography of the

Figure 8.ARPES data from a full coverage h-BN sample. Left; raw image, centre;momentumdistribution curve (MDC) normalizedimage, right;MDCnormalized imagewith the dispersions of a tight-bindingmodel for h-BNoverlaid in green; the tight bindingdispersion in the ΓM direction is the darker green linewith shorter dashes, and in the ΓK direction is the lighter green linewithlonger dashes.

8

2DMater. 2 (2015) 025003 GEWood et al

copper, suggesting structural feedback between theoverlayer and substrate. This occurs despite theinteraction between the h-BN and copper being onlyweak, with ARPES showing that the h-BN is electro-nically decoupled from the copper surface, as expectedin van der Waals epitaxy of a layered material such ash-BN. It is particularly surprising that this weakinteraction is sufficient to induce epitaxial ordering,aligned to the high symmetry axes of the substrate,even when the symmetry of surface and overlayer arenot matched. However, the degree of orientation isstronger on Cu(111), suggesting a stronger interactionwhen the symmetry is matched. The growth rate alsoappears to be faster on Cu(111) than Cu(100). For thecold-rolled copper foils studied here, which arepredominantly Cu(100) after annealing, the fourequivalent orientations of h-BN will lead to a pre-valence of high angle (30°) grain boundaries. Control-ling this epitaxy will enable the growth of high quality,large area, single crystal, monolayer h-BN films by lowpressure CVD.

Acknowledgments

EPSRC is thanked for support through studentships toAJM and JJM and through grant number EP/K005200/1.We acknowledge support via theCALIPSOTransNa-tional Access Program from the European Commu-nityʼs Seventh Framework Programme (FP7/2007-2013)under grant agreement number 312284.

References

[1] YankowitzM,Xue J and LeRoyB J 2014 J. Phys.: Condens.Matter 26 303201

[2] NagashimaA, TejimaN,GamouY,Kawai T andOshimaC1995Phys. Rev.B 51 4606–13

[3] Li X et al 2009 Science 324 1312–4[4] IsmachA et al 2012ACSNano 6 6378–85[5] Gao Y, RenW,MaT, Liu Z, Zhang Y, LiuWB,Ma LP,

MaX andChengHM2013ACSNano 7 5199–206[6] Tay RY,GriepMH,MallickG, Tsang SH, SinghR S,

Tumlin T, Teo EHT andKarna S P 2014Nano Lett. 14839–46

[7] Tay RY,WangX, Tsang SH, LohGC, SinghR S, LiH,MallickG andTong TeoEH2014 J.Mater. Chem.C 2 1650

[8] KimG, JangAR, JeongHY, Lee Z, KangD J and ShinH S 2013Nano Lett. 13 1834–9

[9] Han J, Lee J Y, KwonHandYeo J S 2014Nanotechnology 25145604

[10] GibbAL, AlemN,Chen JH, EricksonK J, Ciston J, GautamA,LinckMandZettl A 2013 J. Am. Chem. Soc. 135 6758–61

[11] Kidambi PR, BlumeR, Kling J,Wagner J B, Baehtz C,WeatherupR S, Schloegl R, Bayer BC andHofmann S 2014Chem.Mater. 26 6380–92

[12] Huang PY et al 2011Nature 469 389–92[13] Liu Y, ZouX andYakobsonB I 2012ACSNano 6 7053–8[14] KomaA 1992Thin Solid Films 216 72–76[15] WilsonNR et al 2013NanoRes. 6 99–112[16] Murdock AT, Koos A, BrittonTB,Houben L, Batten T,

Zhang T,WilkinsonA J,Dunin-Borkowski RE, LekkaCE andGrobert N 2013ACSNano 7 1351–9

[17] Brown L et al 2014Nano Lett. 14 5706–11[18] Joshi S et al 2012Nano Lett. 12 5821–8[19] Roth S,Matsui F, Greber T andOsterwalder J 2013Nano Lett.

13 2668–75[20] Avila J, Razado-Colambo I, Lorcy S, Lagarde B,Giorgetta J L,

Polack F andAsensioMC2013 J. Phys.: Conf. Ser. 425 192023[21] Liu Y, Bhowmick S andYakobsonB I 2011Nano Lett. 11

3113–6[22] Strudwick A J,WeberNE, SchwabMG,KettnerM,Weitz RT,

Wünsch J R,MüllenK and SachdevH2015ACSNano 931–42

[23] Sutter P and Sutter E 2014APLMater. 2 092502[24] Schneider C a, RasbandWS and Eliceiri KW2012Nat.

Methods 9 671–5[25] Chen S, CaiW, Piner RD, Suk JW,WuY, RenY, Kang J and

Ruoff R S 2011Nano Lett. 11 3519–25[26] Wang Z J et al 2015ACSNano 9 1506–19[27] Laskowski R, Blaha P and SchwarzK 2008Phys. Rev.B 78

045409[28] AuwärterW,MuntwilerM,Osterwalder J andGreber T 2003

Surf. Sci. 545 735–40

9

2DMater. 2 (2015) 025003 GEWood et al