WO2012174040A1

WO2012174040A1 - Direct graphene growth on mgo (111) by physical vapor deposition: interfacial chemistry and band gap formation

Info

Publication number: WO2012174040A1
Application number: PCT/US2012/042140
Authority: WO
Inventors: Jefffy KELBER
Original assignee: University Of North Texas
Priority date: 2011-06-17
Filing date: 2012-06-13
Publication date: 2012-12-20
Also published as: EP2720809A1; EP2720809A4; KR20140089311A

Abstract

Graphene can be grown directly on MgO(111) by industrially practical and scalable methods: free radical-assisted chemical vapor deposition (CVD), and physical vapor deposition (PVD). Single layer and double layer films can be produced by PVD, with a ~ 2 monolayer thick film as the apparent limiting thickness. C(1s) x-ray photoemission spectra (XPS) indicate that in both layers, carbon atoms are in two different oxidation states. A band gap of - 0.5 -1 eV has been observed for the two layer film. The XPS, LEED and band gap findings indicate that the graphene/MgO interface is commensurate, and that the MgO surface layer is reconstructed, resulting in carbon->MgO charge transfer. The ability to grow MgO(111) films on Si(100) or Si(111)— reported in the literature— points to a direct path to the development of graphene-based field effect transistors (FETs) and spin-FETs on MgO(111)/Si(100).

Description

TITLE OF THE INVENTION

DIRECT GRAPHENE GROWTH ON MGO (111) BY

PHYSICAL VAPOR DEPOSITION: INTERFACIAL

CHEMISTRY AND BAND GAP FORMATION

Priority Data and Incorporation by Reference

This application claims benefit of priority to U.S. Provisional Patent Application Serial No. 61/497,971 filed June 17, 201 1 which is incorporated by reference in its entirety. This case also claims benefit of priority to U.S. Patent Application Serial No. 13/426,823, filed March 22, 2012, which is incorporated herein by reference in is entirety.

BACKGROUND OF THE INVENTION

Field of the Invention

[0001] Graphene's electronic properties, including low effective electron/hole mass [1-3], high charge carrier mobilities[3-7], long spin diffusion length [8-10]and polarizeability in the vicinity of a ferromagnet[l 1] have exciting potential applications in charge-based [12-17] and spintronics[8,9,18] devices. To date, however, most device studies have employed either (a) graphene sheets physically transferred from highly oriented pyrolytic graphite (HOPG) [1,2] or sheets grown on transition metals [19-21]and transferred onto S1O2 or other substrates, or (b) graphene grown via the high temperature evaporation of Si from SiC[3, 7,22,23]. The first approach presents obvious difficulties for industrial-scale production and also produces interfacial inhomogeneities[24,25] and other interactions [26] that severely impact device performance [8,9,16]. Graphene growth on SiC also presents issues, due to the high temperatures involved ([3,23], the fact that the graphene growth process is apparently not self-limiting, and the difficulties involved in integration of a high temperature SiC process with current Si CMOS devices.

[0002] The industrial-scale production of graphene-based devices requires the direct growth of graphene by practical and scalable methods, such as chemical or physical vapor deposition (CVD or PVD). Additionally, the development of FETs would be greatly facilitated by formation of a band gap, which would yield a truly "off state at room temperature, and increase the on/off device ratio. Such a gap has not been observed (in the absence of an applied field) for transferred single layer sheets. While a 0.26 eV band gap has been reported on the basis of angle-resolved photoemission for single-layer

graphene/SiC(0001) [27], such a band gap has not been detected by charge transport measurements [3,17]. In any case, a band gap of this magnitude may be too small, at least for conventional FET applications [28,29].

[0003] We have recently reported [30] the formation of well-ordered, ~2 -layer graphene on MgO(l 11) by exposure to thermally dissociated C2H4, (free radical-assisted CVD) followed by annealing in UHV at 1000 K to induce ordering. Photoemission spectroscopy (PES) indicated the film to be continuous, with formation of a well ordered LEED pattern and a valence band spectrum in excellent agreement with graphene sheets grown on, e.g., Ni(l 1 1) substrates [31]. PES and inverse photoelectron spectroscopy (IPES) revealed the existence of a ~ 0.5 eV -1 eV band gap for a ~ 2.5 monolayer graphene film [30], while core -level x- ray photoelectron spectroscopy (XPS) revealed the presence of carbon in two oxidation states, suggesting that at least some of the carbon atoms in the layer closest to the MgO surface partially oxidized [30]. Subsequently, we reported [32] the formation of - 1.5 monolayer thick graphene on MgO(l 1 1) by PVD— magnetron sputter deposition from a graphite target— at room temperature, followed by annealing in UHV. LEED data of that film, and the one produced by free radical assisted CVD, yielded apparently six-fold LEED patterns actually consisting of two three-fold patterns [32] due to unequal A site/B site LEED intensities. This inequality indicates unequal charge populations on A sites vs. B sites in the graphene lattice on the MgO(l 11) substrate. This lifting of the graphene lattice A site/B site chemical equivalency removes the HOMO/LUMO degeneracy [33], thus forming a band gap in graphene. A schematic of the proposed [32] mechanism of band gap formation in graphene on MgO(l 1 1) is shown in Fig. 1. Much of the detail of this process is set forth in U.S. Patent Application Serial Number 12/980,763, the entirety of which is incorporated herein-by -reference.

[0004] A major issue posed by the mechanism shown in Fig. 1 is how such strong C/MgO interactions yield a C_3V LEED pattern. The graphene C-C bond distance (2.5 A) is substantially shorter than the 0-0 nearest neighbor distance for a bulk-terminated MgO(l 11) (lxl) surface (2.9 A), even if, as previously reported [34], the MgO(lxl) surface layer is actually OH-terminated. Thus, the graphene lattice is expected to be incommensurate with the MgO(l 11) surface unit cell, with A site and B site carbon atoms experiencing a distribution of chemical interactions, resulting in equivalent average A site/B site charge distributions and LEED intensities. Indeed, the formation of well-defined C3V LEED patterns for single and double layer graphene on MgO indicates that the graphene and MgO surfaces are commensurate, implying a substantial reconstruction of either the graphene surface layer or the MgO(l 11) surface layer.

SUMMARY OF THE INVENTION

[0005] We present experimental evidence that physical vapor deposition (PVD) of carbon atoms on MgO(l 11) leads to single and double layer graphene films, with the double layer film as a limiting thickness. That is, at most two graphene monolayers are formed - no further graphene is formed if the process is continued. In each case, the XPS C(ls) core level spectrum indicates that the presence of two carbon oxidation states, consistent with the LEED data showing different A site/B site LEED intensities and electron densities. The LEED data also indicates that the axes of the graphene lattice and MgO(l 11) substrate are aligned rather than rotated azimuthally with respect to each other. These results indicate that the graphene layers are commensurate with the oxide substrate, and that substantial carbon^ oxide charge transfer occurs in both the first and second graphene layers. These results have important implications for growth of graphene on oxide substrates at temperatures compatible with Si CMOS integration, the formation of graphene band gaps, and the development of both FET and spin-FET structures.

[0006] Deposition, LEED and XPS studies were carried out in a system described previously [35,36]. Briefly, the system consisted of two separately pumped chambers. The introduction/deposition chamber (base pressure 1 x 10^~6 Torr) contained capabilities for PVD from a graphite target mounted on a DC sputter magnetron source. Sputter deposition was carried out in an Ar plasma at ambient substrate temperature at deposition rates of ~ 1 A/min. Sample transport to the analytical chamber (base pressure 5 x 10^~10Torr) occurred by magnetically-coupled feed-through without exposure to ambient. The analytical chamber contained capabilities for Ar ion sputtering, sample cooling/annealing (150 K-1200 K) as well as a hemispherical analyzer, non-monochromatic MgKa x-ray source operated at 15 keV and 300 W, and reverse-view LEED optics. XPS spectra were acquired in the constant pass energy mode (22 eV). XPS spectra were analyzed according to standard methods [37- 39]. To account for sample charging, the O(ls) peak maximum was referenced to a binding energy of 530.5 eV obtained for ultrathin epitaxial MgO films on Mo (100). [40]. LEED intensity analysis was done as described previously [32]. The MgO sample was a commercially-obtained (11 1) single crystal 1 cm x 1 cm, 0.5 mm thick, mounted on a Ta foil sample holder on a Mo sample transfer platen. The analysis of LEED intensities was also as previously described [32]. [0007] PVD of carbon onto MgO(l 1 1) and annealing at 1100 K results in formation of single or double layer graphene, with the second layer as a limiting thickness. LEED data indicate C_3V symmetry and a graphene axis aligned with that of the substrate. Consistent with C₃v symmetry, XPS data indicate the presence of carbon in two different bonding

environments in both the first and second layer, with strong carbon-oxygen interactions. These data, and the presence of a band gap reported previously [30,32] indicate MgO(l 1 1) reconstruction to form a commensurate graphene/MgO(l 1 1) interface. The above, combined with the ability to grow a MgO(l 11) film on Si(100) at 973 K [49], indicate a direct route to the industrially practical formation of FET and spin FET structures on Si(100).

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate exemplary embodiments of the invention, and, together with the general description given above and the detailed description given below, serve to explain the features of the invention.

[0009] FIG. 1 is Schematic of proposed band gap formation in PVD or CVD graphene on MgO(l 11), resulting from strong MgO/graphene interactions.

[0010] FIG. 2 is C(ls) and O(ls) XPS spectra (a,b,d, e) and corresponding LEED images

(c,f) of an as-received MgO(l 1 1) single crystal before (a-c) and after (d-f) exposure to (¾ at

10^"7Torr, 600 K for 120 min. LEED images acquired at 85 eVbeam energy.

[0011] FIG. 3 is PVD of from a graphite target onto the carbon substrate (Fig. 2d-f). (a)

C(ls) spectrum after PVD at room temperature; (b) corresponding O(ls) spectrum; (c) LEED image; (d)C(ls) spectrum after annealing in UHV at 1 100 K for 5 hours; (e) corresponding

O(ls) spectrum; (f) corresponding LEED image; (g) C(ls) spectrum after further annealing

(an additional 6 hours— 11 hours total annealing) at 1 100 K in UHV; (h) corresponding 0(1 s) spectrum; (i) corresponding LEED image. All LEED images acquired at 85 eV beam energy. Labels on LEED spots in (f) and (i) refer to A site and B site LEED intensities of the graphene lattice (See Table I). LEED of single layer graphene (i) also displays an inner ring of diffraction spots consistent with the in plane lattice constant of bulk MgO(l 11 1)

[0012] FIG. 4 is a commensurate graphene/MgO(l 11) Interface. (A)Ball-and-stick and (B) space-filling models for a commensurate graphene/MgO(l 1 1) surface with the O surface layer so that the 0-0 distance is that of the graphene lattice constant, 2.5 A. (A) is viewed slightly off-axis so as to highlight carbon atom interactions with Mg or O sites. Atomic diameters for the space-filling model are given in the legend.

DETAILED DESCRIPTION OF THE INVENTION

[0013] XPS C(ls) and O(ls) spectra, and corresponding LEED images are displayed in Figure 2 for an as-received MgO(l 11) single crystal before (Fig. 2 a-c) and after (Fig. 2 d-f) annealing in O2 (10^~7Torr) for 120 minutes. The data show that prior to annealing in O2, the sample is covered with a multilayer carbon film on a hydroxylatedMgO surface without long- range order. At this stage, the XPS-derived average thickness of the carbon overlayer is 5.0 A, or 3.6 monolayers (ML), based on a carbon covalent diameter of 1.4 A in graphene. The C(ls) spectrum (Fig. 2a) is well fit with two carbon bonding environments, one at 285 eV, generally indicative of alkyl carbon [41] (but see below), and a second environment at -286.5 eV, consistent with carbon in an oxidized environment, such as CHO or COH [39,41]. There is no corresponding LEED pattern (Fig. 2c), indicating that at this point the surface carbon layer and MgO substrate surface lack long-range order. Unsurprisingly, the MgO surface region is heavily hydroxylated, as indicated by a O(ls) feature at > 532 eV binding energy, attributable to surface OH species [39], and a smaller feature 534 eV suggesting bound H₂0 (Fig. lb). After annealing, the total carbon coverage is reduced to - 1.1 ML. The C(ls) spectrum indicates the presence of two bonding environments of approximately equal intensity. The higher binding energy feature, at - 286 eV is still consistent with the presence of oxidized carbon [39,41], while the second feature near 284.5 eV is indicative of graphitic carbon [41]. The corresponding O(ls) XPS spectrum (Fig. 2d) and LEED pattern (Fig. 2e) are similar to those reported [34] for a clean, OH-terminated MgO(l 1 l)(lxl) surface. These data therefore indicate that annealing in (¾ at 600 K results in a partially oxidized and graphitic carbon layer on top of an ordered, hydroxylatedMgO surface substrate. The resemblance of the LEED image (Fig. 2f) to previous results [32,34] indicates that annealing to 600 K in the presence of (¾ induces long-range ordering of the MgO surface, but not of the remaining carbon overlayer.

[0014] Exposure of the annealed surface to sputter deposition from a graphite target results in the XPS and LEED images displayed in Fig. 3 a-c. The C(ls) spectrum (Fig. 3a) is still well-fit by two environments, at binding energies near 285 eV and 286 eV, with a total average carbon thickness of ~ 3.8 ML. The O(ls) spectrum (Fig. 3b) indicates an apparent increase in surface hydroxylation, not surprising upon exposure to an Ar plasma environment during the brief sputter deposition step. The LEED image (Fig. 3c) remains unchanged, indicating that the C overlayer is still disordered but that the MgO substrate remains ordered.

[0015] Subsequent annealing of the disordered sample in UHV to 1100 K for 5 hours yields the surface shown in Fig. 3d-f. Average carbon overlayer thickness is ~ about 2 ML. The two carbon bonding environments indicated in the C(ls) spectrum (Fig. 3d) have equal intensities at binding energies of 286 and 285.3 eV. The O(ls) spectrum (Fig. 3e) shows little change except for some possible de-hydroxylation. Importantly, the LEED image (Fig.3f) now displays the 6 spot pattern expected for graphene[30,32]. This pattern, similar to previous reports [32] exhibits unequal A site/B site LEED intensities (Fig. 3f, and Table I).

[0016] Further annealing of the double layer graphene film on MgO(l 1 1) (six additional hours, for a total of 11 hours) at 1100 K in UHV results in a decrease in the intensity of C(ls) spectrum (Fig. 3g), and the average total thickness of graphene layer is now 1 ML. The C(ls) XPS spectrum (Fig. 3g) is still well fit by two separate components near 286 eV and 285 eV binding energies, although the two components are no longer of equal intensity. The O(ls) spectrum (Fig. 3h) shows little change, but the LEED pattern (3i) is now more complex. The C(ls) spectrum (Fig. 3g) , and LEED image (Fig. 3i) now resemble those previously reported [32] upon annealing of an adventitious carbon layer in UHV. Further, the LEED image corresponding to the graphene lattice still displays unequal A site/B site intensities (Table I), in agreement with previous results [32] . In addition to the six LEED spots in Fig. 3i (labeled A and B), there is an inner ring of six spots (marked by arrows, corresponding to smaller reciprocal lattice and therefore larger direct lattice unit cell.

Assuming that the larger hexagonal pattern (A, B, Fig. 3i) corresponds to a graphene direct lattice spacing of 2.5 A, the inner ring corresponds to a direct lattice spacing of ~ 4A -4.5A, or close to 4.2 A of the in plane direct lattice constant for bulk MgO(l 11)[42].

[0017] The results shown in Fig. 3 were repeatable by additional PVD of carbon onto single layer graphene, followed by annealing at 1100 K. The total average carbon thickness, however, was never more than 2 ML, indicating that 2 ML is a limiting thickness for graphene on MgO(l 11) under these conditions.

Table I: LEED intensities for PVD carbon on MgO(l 1 1)

lumbers in parenthesis are the standard deviation

2A and B site positions in LEED spectra as in Fig. 3

[0018] The LEED spectrum for the single layer film (Fig. 3 i) also shows an inner ring of six spots. Assuming that the outer ring of six spots (2 of which are labeled "A", and "B" in Fig. 3i) corresponds to the graphene lattice, a rough estimation of the corresponding direct lattice for the inner ring of spots is ~ 4.0-4.5 A, consistent with the in-plane MgO(l 11) lattice constant of 4.2 A for bulk MgO[42].

The Graphene/MgOmi) Interface

[0019] XPS C(ls) spectra for both the single and double layer graphene samples (Fig. 3 d,g) indicate two C(ls) environments with binding energies of - 286 and -285 eV .

Similarly, the corresponding LEED spectra (Fig. 3f,i) indicate that A sites and B sites in the graphene lattice have strong/weak intensities (Table I) and therefore corresponding electron densities. Neither of these is consistent with a graphene lattice (2.5 A lattice constant) in contact with the O -terminated MgO(l 1 1) surface (2.9 A between nearest neighbor oxygen sites. The incommensurate nature of the graphene and MgO(l 1 1) surface oxygen lattices would ensure that both A and B sites in the graphene lattice would experience an ensemble of chemical environments, resulting in a true Οβν LEED pattern (equal A site/B site lattice intensities), as observed for graphene/Ru(0001)[30,32], where there is a similar graphene/substrate lattice mismatch. Thus, the LEED for both double and single layer graphene on MgO(l 11) (Figs. 3f, i and Table I) indicate a reconstruction at the MgO and graphene interface to form a commensurate grapheneoverlayer on the MgO substrate.

[0020] An analysis of the XPS core level spectra for double and single layer

graphene/MgO(l 1 1) (Fig.3) confirms the presence of two carbon bonding environments, and provides further information on bonding interactions and charge transfer at the

graphene/MgO interface. The charge-corrected C(ls) spectra for single and double layer graphene (Fig. 3) can be analyzed in terms of corresponding spectra for partially oxidized single walled carbon nanotubes (SWCNTs) [41]. The chemical shift (ΔΕ) of a core level binding energy, relative to that of an elemental standard, is generally proportional to the change in atomic charge population (Aq) in the ground state (before emission of a photoelectron) introduced by bonding to electronegative or electropositive nearest neighbors [37,41,43]: (1) where V_rei corresponds to the relaxation energy of the system in the final state due to the presence of a localized core hole [37,43].

[0021] In general, V_rei is the same, or nearly so, for similar systems, allowing semiquantitative or quantitative correlation of binding energy shifts with ground state atomic charge densities for structurally similar materials. For an unmodified carbon site on a SWCNT, the binding energy was reported as 284.5 eV, appropriate for sp²- hybridized carbon. A C(ls) binding energy near 286 eV was attributed to CHO groups [41].

Comparison of observed C(ls) binding energy shifts with ab initio calculations of Mulliken atomic charge populations yielded a value for k (equ. l) of 4.38 eV/electron [41]. Assuming similar V_rei terms for a SWCNT as for graphene, a C(ls) binding energynear 286eV (Fig. 3d,g) is consistent with direct C-0 interaction, while a binding energy near 285 eV for the remaining carbon atoms suggests a charge transfer of ~ 0.1 eV from carbon atoms in this bonding environment, relative to carbon atoms in HOPG. The O(ls) spectra for single layer (Fig. 3f) and double layer (Fig. 3e) graphene/MgO(l 11) both display a small shoulder near 532 eV, and a main lattice oxygen peak referenced to 530.5 eV. This is similar to the spectrum for the MgO(l 1 l)(lxl) surface prior to PVD (Fig. 3b). Since O(ls) binding energies for C-0 bonding environments are quite close to those for H-0 bonding

environments -e.g., esters vs. alchohols, at 532.2 eV[44], the O(ls) spectra in Fig.3f and Fig. 3e are consistent with C-0 bond formation . Thus, a comparison of C(ls) and O(ls) XPS data for graphene/MgO(l 11) (Fig. 3) to that of modified SWCNTs indicates significant graphene-^ MgO charge transfer, a conclusion entirely consistent with the LEED data, and with inverse photoemission data [30]. Both single and two layer graphene grown directly on MgO(l 11) are p-type.

[0022] The XPS and LEED data also suggest that the MgO(l 11) surface reconstructs to yield a ~ 10% shortening of 0-0 nearest neighbor bond distance relative to that for bulk- terminated MgO(l 11). This is because the stretching of the graphene lattice to become commensurate with the bulk-terminated MgO(l 1 1) lattice would require of - 16% not only in the first, but also in the second graphene layer, and which is also inconsistent with PES and IPES data [30]. Further, the LEED data for single layer graphene /MgO(l 1 1) (Fig. 3i) indicates that the graphene lattice is aligned with that of the MgO substrate, rather than rotated azimuthally, and that this is true as well for the second layer (Fig. 3f). On the basis of these data, a commensurate graphene/MgO(l 11) interface is displayed in Fig. 4.

[0023] Inspection of Fig. 4 shows that carbon atoms in the first layer are either over O or over Mg sites, in accord with the LEED and XPS data.

[0024] Strong overlayer- MgO(l 11) charge transfer is generally observed for metal overlay ers on MgO(l 1 1) surfaces [45]. The MgO(l 1 1) surface is polar, and therefore unstable unless terminated by bonding to another species, such as OH groups [34] or carbon atoms. In contrast, metal interactions with the non-polar MgO(100) surface typically yield negligible charge transfer, and the strong charge transfer associated with bonding to the (11 1) surface may also be accompanied by significant surface reconstruction [45].

Implications for Device Fabrication and Integration with Si CMOS

[0025] The formation of a band gap in graphene is critical for the development of graphene FETs, in order to obtain high On/Off current ratios. A band gap of - 0.5 eV- 1 eV is satisfactory for FET applications [28,29]. Although transport/mobility measurements for graphene/MgO(l 1 1) have not yet been published, the existing data indicate that

graphene/MgO(l 1 1) has exciting potential for FET applications. Additionally, non-local spin valve operation at room temperature has been demonstrated using a graphene sheet transferred from HOPG to Si0₂[8]_. While a band gap is not essential for a spin-FET or similar devices, it is not a hindrance [46], and graphene grown on MgO(l 1 1) is therefore of strong potential interest for both FET and spin-FET applications. Previous studies of graphene sheets physically transferred to high dielectric substrates [4,5,15] indicate that the proximity of a high dielectric medium to the graphene layer should significantly enhance electron and hole mobility, important for both FET and spin-FET devices.

[0026] The ability to grow graphene at consistent thicknesses is also of obvious importance for device fabrication. This is potentially troublesome for graphene/SiC(0001), where multilayer growth has been observed [22,23]. While films varying between 1-2 ML thick have been grown with great care on 2 inch SiC(0001) wafers [17,47], permitting the fabrication of GHz FETs, this issue is obviously a concern for industrial-scale processing. The data reported here, however, suggest that growth of graphene on MgO(l 1 1) by PVD has a limiting thickness of 2 ML, of considerable importance of actual device

fabrication/production. A similar conclusion was recently reached for graphenenanoflakes grown by CVD on highly porous MgO substrates [48] for capacitor applications. The reasons for this apparent limiting thickness are not obvious, nor is it yet certain that this applies to, e.g., free-radical based growth processes [30]. Such a limiting thickness, however, would greatly assist in the production of large-scale wafers with consistent graphene structures, and therefore the fundamental surface chemistry regarding graphene growth on MgO(l 11) and other oxides is of direct importance to practical device fabrication.

[0027] In addition to the specific issue of graphene growth on a dielectric substrate, the integration of graphene devices with Si CMOS requires that epitaxial dielectric film formation on Si(100) should be feasible, and that graphene growth be achieved at temperatures compatible with oxide/Si interfacial stability. MgO(l 11) films have been grown on Si(100) at 973 K [49], entirely compatible with graphene/MgO(l 1 1) growth processes of - 1000 K - 1 100 K. Since MgO[10], alumina [47] and Hf0₂ oxides have all been deposited on graphene substrates at mildly elevated temperatures, the fabrication of such gate oxides is entirely compatible with graphene/MgO/Si. Therefore, there now exists a direct path to the practical and scalable fabrication of graphene/MgO-based FET and spin- FET structures on Si(100).

Implications for Graphene Growth on other Oxides

[0028] The data for graphene growth on MgO have significant implications for graphene growth by PVD or CVD on other oxides. A commensurate graphene/MgO(l 1 1) surface layer, and therefore the ability of the MgO(l 11) surface to reconstruct (Fig. 4), is responsible for band gap formation. This reconstruction may also be essential to the growth of macroscopic graphene layers, as opposed to nanoflakes. Because of the unstable surface energy associated with a polar, O^2" (or Mg⁺²)- terminated surface layer, MgO(l 11) is prone to reconstruction in the presence of overlayers[34,45]. This implies that graphene growth should not occur on non-polar MgO surfaces, such as MgO(100), which are less prone to relaxation [45], and therefore induce strain during graphene growth due to graphene/substrate lattice mismatch. Indeed, CVD on such "non- 1 11" surfaces results only in graphenenanoflake formation [48,50]. Corroboration for this hypothesis comes from graphene growth on Α1₂θ₃(0001), which has an 0-0 nearest neighbor distance of- 3 A, but which is non-polar and therefore less prone to the type of reconstruction necessary to insure a commensurate graphene/sapphire interface [51]. Attempted CVD of graphene directly on Αΐ₂θ₃(0001) at moderate temperatures (< -1300 K) produces only nanoflake formation[52]. Large scale, high mobility graphene sheets have been grown on Α1₂θ₃(Ό001), but at much higher temperatures (> 1700 K), suggesting that such extreme growth conditions are necessary to overcome the problem of graphene/alumina lattice mismatch. Importantly, in that effort a weak graphene/alumina interaction was deduced from the Rama spectra[53], consistent with an incommensurate interface, and suggesting the absence of a graphene band gap for graphene/Al₂O₃(0001).

[0029] The above data suggest that hexagonal oxide (11 1) surfaces, with a tendency to reconstruct, may be favorable candidates for direct graphene growth. This category includes other AO type oxides of the rocksalt structure (e.g., NiO (1 1 1), AI2O_{3 (}m₎,HfO (1 11) but perhaps also other oxides, including perhaps the titanates (MT1O4 wherein M is divalent). In selecting oxide surfaces as candidates for graphene growth, the ability to grow the oxide(l 11) film on Si(100) is an obvious limitation for CMOS integration. The large number of potential candidates, however, is exciting in the potential for forming multifunctional, non-volatile spin and charge devices.

REFERENCES

[0030] [1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jian, M.I. Katsnelson, I.V. Grigorieva, S.V. Dubonos, A.A. Firsov, Nature 438 (2005) 197.

[0031] [2] A.H.C. Neto, F. Guineau, N.M.R. Peres, K.S. Noveoselov, A.K. Geim, Reviews of Modern Physics 81 (2009) 109. [0032] [3] C. Berger, Z. Song, T. Li, Z. Li, A.Y. Ogbazghi, R. Feng, Z. Dai, A.N.

Marchenkov, Congrad, Edward H. First, Phillip N., W.A. de Heer, Journal of Physical Chemistry B 108 (2004) 19912.

[0033] [4] R.S. Shishir and D.K. Ferry, Journal of Physics Condensed Matter 21 (2009) 232204.

[0034] [5] F. Chen, J. Xia, D.K. Ferry, N. Tao, Nano Lett. 9 (2009) 2571.

[0035] [6] C.R. Dean, A.F. Young, I. Meric, C. Lee, L. Wang, S. Sorgenfrei, K. Watanabe,

T. Taniguchi, P. Kim, K.L. Shepard, J. Hone, Nature, Nanotech. 5 (2010) 722.

[0036] [7] C. Berger, Z. Spong, X. Li, X. Wu, N. Brown, C. Naud, D. Mayou, T. Li, J.

Hass, A. Marchenkov N., E.H. Conrad, P.N. First, W.A. de Heer, Science 312 (2006) 1 191.

[0037] [8] N. Tombros, C. Jozsa, M. Popinciuc, H.T. Jonkman, B.J.v. Wees, Nature (2007)

571.

[0038] [9] S. Cho, Y. Chen, M.S. Fuhrer, Applied Physics Letters 91 (2007) 123105.

[0039] [10] W. Han, K. Pi, H. Wang, M. McCreary, Y. Li, W. Bao, P. Wei, J. Shi, C.N.

Laun, R.K. Kawakami, Proceedings of SPIE 7398 (2009) 739819-1.

[0040] [11] H. Haugen, D. Huertas-Hernando, A. Brataas, Physcal Review B 77 (2008)

1 15406.

[0041] [12] X. Liang, Z. Fu, S.Y. Chou, Nanoletters 7 (2007) 3840.

[0042] [13] A. Das, S. Pisana, B. Charkraborty, S. Piscanec, S.K. Saha, U.V. Waghmare,

K.S. Novoselov, H.R. Krishnamurthy, A.K. Geim, A.C. Ferrari, A.K. Sood, Nature 3 (2008)

210.

[0043] [14] M. Jin, H. Jeong, J.W. Yu, D.J. Bae, B.R. Kang, Y.H. Lee, J. Phys. D. : Appl. Phys. 42 (2009) 135109.

[0044] [15] X. Hong, A. Posadas, K. Zou, C.H. Ahn, J. Zhu, Physical Review Letters 102 (2009) 136808. [0045] [16] V.E. Dorgan, M. Bae, E. Pop, Appl. Phys. Lett. 97 (2010) 082112.

[0046] [17] Y.-. Lin, C. Dimitrakopoulos, K.A. Jenkins, D.B. Farmer, FL-. Chiu, A. Grill,

P. Avouris, Science 327 (2010) 662.

[0047] [18] V.M. Karpan, P.A. Khomyakov, A.A. Starikov, G. Giovannetti, M.

Zwierzycki, M. Talannana, G. Brocks, J. van den Brink, P.J. Kelly, Phys. Rev. B. 78 (2008) 195419.

[0048] [19] X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S.K. Banerjee, L. Colobmo, R.S. Ruoff, Science 324 (2009) 1312.

[0049] [20] A. Reina, S. Thiele, X. Jia, S. Bhaviripudi, M.S. Dresselhaus, J.A. Schaefer, J. Kong, Nano. Res. 2 (2009) 509.

[0050] [21] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M.S. Dresselhaus, J. Kong, Nanoletters 9 (2009) 30.

[0051] [22] W.A. de Heer, C. Berger, X. Wu, P.N. First, E.H. Conrad, X. Li, T. Li, M. Sprinkle, J. Hass, M.L. Sadowski, M. Potemski, G. Martinez, Solid State Communications 143 (2007) 92.

[0052] [23] I. Forbeaux, J.-. Themlin, J.-. Debever, Physical Review B 58 (1998) 16396.

[0053] [24] A. Deshpande, W. Bao, F. Miao, C.N. Lau, B.J. LeRoy, Phys. Rev. B. 79 (2009) 20541 1.

[0054] [25] Y. Zhang, V.W. Brar, C. Girit, Zettl, Alex Crommie, Michael F., Nature Physics 5 (2009) 722.

[0055] [26] J. Chen, C. Jang, S. Xiao, M. Ishigami, M.S. Fuhrer, Nature Nanotechnology 3 (2008) 206.

[0056] [27] S.Y. Zhou, G.-. Gweon, A.V. Federov, P.N. First, W.A. de Heer, D.-. Lee, F. Guinea, A.H. Castro Neto, A. Lanzara, Nature Materials 6 (2007) 770-776. [0057] [28] M. Cheli, P. Michetti, G. lannaccone, IEEE Transactions on Electron Devices 57 (2010) 1936.

[0058] [29] P. Michetti, M. Cheli, G. lannaccone, Applied Physics Letters 96 (2010) 133508.

[0059] [30] L. Kong, C. Bjelkevig, S. Gaddam, M. Zhou, Y.H. Lee, G.H. Han, H.K. Jeong, N. Wu, Z. Zhang, J. Xiao, P.A. Dowben, J.A. Kelber, J. Phys. Chem. C. 114 (2010) 21618.

[0060] [31] Y.S. Dedkov, M. Fonin, U. Rudiger, C. Laubschat, Phys. Rev. Lett. 100 (2008) 107602.

[0061] [32] S. Gaddam, C. Bjelkevig, S. Ge, K. Fukutani, P.A. Dowben, J.A. Kelber, J. Phys. Cond. Matt. 23 (201 1) (in press).

[0062] [33] P.A. Cox, The Electronic Structure and Chemistry of Solids, Oxford

University Press, 1987.

[0063] [34] V.K. Lazarov, R. Plass, H.-. Poon, D.K. Saldin, M. Weinert, S.A. Chambers, M. Gajdardziska-Josifovska, Phys. Rev. B 71 (2005) 1 15434.

[0064] [35] J.A. Wilks, N.P. Magtoto, J.A. Kelber, V. Arunachalam, Applied Surface Science, 253 (14) (2007) 6176.

[0065] [36] J.A. Wilks and J.A. Kelber, Appl. Surf. Sci. 255 (23) (2009) 9543.

[0066] [37] D. Briggs and M.P. Seah, 1 (1990) 657.

[0067] [38] K. Shepherd and J. Kelber, Applied Surface Science 151 (1999) 287.

[0068] [39] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray

Photoelectron Spectroscopy, Physical Electronics, 1995.

[0069] [40] J.S. Corneille, J. He, W. Goodman, Surf. Sci. 306 (1994) 269.

[0070] [41] A. Jung, R. Graupner, L. Ley, A. Hirsch, Phys. Stat. Sol. B. 243 (2006) 3217.

[0071] [42] T. Susaki, S. Kumada, T. Katase, K. Matsuzaki, M. Miyakawa, H. Hosono,

Appl. Phys. Exp. 2 (2009) 091403. [0072] [43] E.A. Hoffmann, T. K5rtvelyesi, E. Wilusz, L.S. Korugic-Karasz, F.E. Karasz, Z.A. Fekete, Journal of Molecular Structure: THEOCHEM 725 (1-3) (2005) 5.

[0073] [44] G.P. Lopez, D.G. Castner, B.D. Ratner, Surf, and Int. Anal. 17 (1991) 267.

[0074] [45] J. Goniakowski and C. Noguera, Physical Review B 66 (2002) 085417.

[0075] [46] H.C. Koo, J.H. Kwon, J. Eom, J. Chang, S.H. Han, M. Johnson, Science 325 (2009) 1515.

[0076] [47] J.S. Moon, D. Curtis, M. Hu, D. Wong, C. McGuire, P.M. Campbell, G. Jernigan, J.L. Tedesco, B. VanMil, R. Myers-Ward, C. Eddy Jr., D.K. Gaskill, IEEE Elect. Dev. Lett. 30 (2009) 650.

[0077] [48] G. Ning, Z. Fan, G. Wang, J. Gao, W. Qian, F. Wei, Chem. Commun. 47 (2011) 5976.

[0078] [49] X.Y. Chen, K.H. Wong, C.L. Mak, X.B. Yin, M. Wang, J.M. Liu, Z.G. Liu, J. Appl. Phys. 91 (2002) 5728.

[0079] [50] M.H. Rummeli, A. Bachmatiuk, A. Scott, F. Bornert, J.H. Warner, V.

Hoffman, J. Lin, G. Cuniberti, B. Buchner, ACSnano 4 (2010) 4206.

[0080] [51 ] J.A. Kelber, Surface Science Reports 31 (2007) 271.

[0081] [52] Y. Miyasaka, A. Nakamura, J. Temmyo, Jap. J. Appl. Phys. 50 (2011)

04DH12.

[0082] [53] M.A. Fanton, J.A. Robinson, B.E. Weiland, M. LaBella, K. Trumbell, R. Kasarda, C. Howsare, M. Hollander, D.W. Snyder, Abstract for the Graphene 2011

Conference (Bilbao, Spain) (2011)

Claims

WHAT IS CLAIMED IS;

1. A method for growing a one or two monolayer film of graphene on a substrate, comprising:

Depositing a first layer of graphene on said substrate by physical vapor deposition to form a carbon layer on the substrate,

annealing said carbon layer to form an ordered graphene layer one or two monolayers thick on said substrate,

wherein further deposition and annealing of carbon on said graphene does not result in the formation of additional graphene layers.

2. The method of Claim 1, wherein said annealing is conducted at 1000 - 1100 °K.

3. The method of Claim 1, wherein said depositing and annealing steps comprise depositing a first layer of carbon, followed by annealing at no more than 600 °K to provide an a partially oxidized, graphitic carbon layer and subsequently sputter depositing a carbon layer on said partially oxidized graphitic carbon layer followed by annealing at about 1000 - 1 100 °K to form a film of at most 2 monolayers of graphene on said substrate.

4. The method of Claim 1, wherein said graphene film exhibits an film exhibits a band gap of at least about 0.5 eV.

5. The method of Claim 1, wherein said substrate is a rock salt oxide or a titanate.

6. The method of Claim 5, wherein said substrate is a rock salt oxide.

7. The method of Claim 6, wherein said rock salt oxide is selected from the group consisting of MgO (1 11), O (11 1), A1₂0₃ (1 11) and HfO (11 1).

8. The method of Claim 7, wherein said rock salt oxide is MgO (1 11).

9. The method of Claim 3, wherein said graphene formation is self-limiting, such that no further graphene is formed on said film, despite further deposition and annealing of carbon thereover.

10. The method of Claim 1, wherein said substrate is formed on a layer of a device comprising Si.

11. The method of Claim 10, wherein said Si layer is Si (100).

12. A composition of matter comprising a substrate on which is formed a film of one or two monolayer graphene exhibiting a band gap of at least 0.5 eV.

13. The composition of Claim 12, wherein said substrate is comprised of a rock salt oxide or a titanate.

14. The composition of Claim 13, wherein said titanate is of the formula MT1O₄, wherein M is divalent.

15. The composition of Claim 12, wherein 13, wherein said substrate is a rock salt oxide selected from the group consisting of MgO(l 11), NiO(l 1 1), A1₂0₃ (1 11) and HfO (11 1).

16. The composition of matter of Claim 12, wherein said substrate is formed on a layer of Si.

17. The composition of matter of Claim 16, wherein said Si is Si (100).

18. The composition of matter of Claim 17, wherein said composition of matter is a field effect transistor.