Patent Application
20170053988
The current invention is directed toward a method for producing a semiconducting graphene oxide material in which the oxidation is confined to the graphene layer.
Graphene, a material comprising a lattice of carbon atoms positioned in a ‘honeycomb-type’ arrangement and tightly joined by in-plane sp2 bonds, has garnered much attention from research communities because its unique electrical and mechanical properties make it potentially ideally suited for various engineering applications, such as nanoelectronics and integrated circuits. Compared to other conductive and semiconductive materials, graphene has a superior carrier mobility and low resistivity, making it a promising candidate for integrated circuits. (Castro Neto A H, et al., Rev Mod Phys. 2009, 81, 109-62; Geim A and Novoselov K., Nature Materials. 2007, 6, 183-91; Novoselov K S, et al., Science. 2004, 306, 666-9; Novoselov K S, et al., Nature. 2005, 438, 197-200; Novoselov K S; McCann E, et al. Nat Phys. 2006, 2, 177-80; and Zhang Y, et al., Nature. 2005, 438, 201-4, the disclosures of which are incorporated herein by reference.)
However, graphene is inherently a semimetallic material—as opposed to a semiconductor material—and this limits its usability. (See, e.g., Oostinga J B, et al., Nature Materials. 2007, 7, 151-7; Ni Z H, et al., ACS nano. 2008, 2, 2301-5; Pereira V M, et al., Physical Review B. 2009, 80, 045401; Han M Y, et al., Phys Rev Lett. 2007, 98, 206805-8; Nakada K, et al., Phys Rev B. 1996, 54, 17954-61; and Ponomarenko L A, et al., Science. 2008, 320, 356-8, the disclosures of which are incorporated herein by reference.) As a result, researchers have employed a number of methods to introduce a finite band gap within graphene, and thereby convert it into a semiconductor. One approach to introduce an energy gap opening in graphene is to break its lattice symmetry using foreign atoms such as hydrogen, gold, nitrogen, oxygen, and organic molecular dopants. (See, e.g., Bostwick A, et al., Physical Review Letters. 2009, 103, 056404; Balog R, et al., Nat Mater. 2010, 9, 315-9; Sessi P, et al., Nano letters. 2009, 9, 4343-7; Geirz I, et al., Nano Lett. 2008, 8, 4603-7; Wehling T, et al., Nano letters. 2008, 8, 173-7; Luo Z, et al., Appl Phys Lett. 2009, 94, 111909-11; Leconte N, et al., ACS nano. 2010, 4, 4033-8; Nourbakhsh A, et al., Nanotechnology. 2010, 21, 435203-11; Kim D C, et al., Nanotechnology. 2009, 20, 375703, Alzina F, et al., Physical Review B. 2010, 82, 075422 Gokus T, et al., ACS nano. 2009, 3, 3963-8; Childres I, et al., New Journal of Physics. 2011, 13, 025008; Dong X, et al., Small. 2009, 5, 1422-6; and Lu Y H, et al., The Journal of Physical Chemistry B. 2008, 113, 2-5, the disclosures of which are incorporated herein by reference.)
For example, researchers have used wet oxidation methods to insert foreign atoms into the graphene structure. These impurities alter the sp2 carbon hybridization in graphene to a sp3 carbon hybridization, and eliminate the π-bonds that facilitate charge transport across the graphene plane. Consequently, with diminished charge transport the desired band gap is obtained. (See, e.g., Elias D C, et al., Science. 2009, 323, 610-3; Sofo J O, et al., Physical Review B. 2007, 75, 153401; and Boukhvalov D W, et al., Physical Review B. 2008, 77, 035427, the disclosures of which are incorporated herein by reference.) However, the wet oxidation process is less than ideal: it typically uses harsh chemicals, such as strong acids and oxidizing agents; it takes a significantly long time to complete; and it does not allow for the creation of site specific oxidation, which substantially limits the usability of this modified graphene. (See, e.g., Hummers W S and Offeman R E, J Am Chem Soc. 1958, 80, 1339; Park S and Ruoff R, Nature nanotechnology. 2009, 4, 217-24; Li D, et al., Nat Nanotechnol. 2008, 3, 101-5; Sun X, et al., Nano Res. 2008, 1, 203-12; Becerril H A, et al., Nano Lett. 2008, 2, 463-70, the disclosures of which are incorporated herein by reference.)
Researchers have also experimented with using plasma oxidation, a dry oxidation method, to create a graphene oxide semiconductor material. This method is advantageous in a number of respects: it does not use any harmful chemicals; it is a more rapid process; and it allows for site specific oxidation. For example, Nourbakhsh et al. have fabricated and characterized such a graphene oxide layer. (See Bandgap Opening in Oxygen Plasma-Treated Graphene, Nourbakhsh et al. Nanotechnology. 2010, 21, 435203-11, the disclosure of which is incorporated herein by reference.)
Previous studies also show that the p-doping level, electron-electron scattering rate, and the total density of states of an UV/ozone treated graphene are dictated by the defect density associated with surface concentration of oxygenated functional groups and oxygen molecule. At a very low defect density, the p-doping level and electron-electron scattering rate increase in proportion to the increase in defect density. At a higher defect density, a continuous decay and smoothing of the van Hove singularities becomes apparent, and a further increase in the defect density results in a significant drop in the conductance. This indicates a strong Anderson metal-insulator transition, with an overall change in the carrier concentration in the order of 1012 cm−2. These studies also show that an increase in defect density becomes increasingly difficult as the oxygen adsorption reaches a constant value after a certain UV/ozone exposure time. (See, e.g., Leconte N, et al., ACS nano. 2010, 4, 4033-8; Nourbakhsh A, et al., Nanotechnology. 2010, 21, 435203-11; Kim D C, et al., Nanotechnology. 2009, 20, 375703, Alzina F, et al., Physical Review B. 2010, disclosed above.)
Similar electronic transport behaviors are also observed in oxygen plasma treated graphene, where the p-doping level increases with the increase of oxygen plasma exposure, rendering the oxidized graphene unipolar. As the oxygen plasma exposure increases further, the level of disorder in the structural symmetry of graphene becomes more pronounced, which leads to a decrease in conductance and mobility, as well as a transition from semimetallic to semiconducting behavior. However, the prior art has yet to develop a process for the production of a semiconductor graphene oxide material suitable for practical applications. For example, although Nourbakhsh et al. discuss characterizing a graphene oxide layer created by a plasma oxidation process, the authors do not provide any guidance on how to avoid the creation of oxides on the substrate surface. For example, the authors incorrectly suggest that the band gap that can be created using this dry oxidation process can be as high as 3.6 eV (they reached this figure via calculation). It has now been discovered that such high band-gaps are impossible absent the destruction of the graphene oxide band gap. Accordingly a need exists for improved fabrication processes capable of forming graphene oxide materials in which the oxidation is confined within the graphene layer such that they can be used in practical applications.
The present invention is directed to a novel fabrication method that allows for a versatile but precise manipulation of graphene so as to develop graphene oxide material that possesses an appreciable, and determinable, band gap, and in which the oxidation is confined to the graphene layer. This novel fabrication method can thus develop graphene oxides that are optimized for practical use. The process includes subjecting a graphene sample to a dry oxidation process for a pre-determined period of time, wherein the length of oxidation exposure determines the magnitude of the band gap.
In one embodiment of the invention, masks are used in conjunction with the oxidation process, and the masking and oxidation steps are optionally iterated in order to achieve a desired configuration.
In yet another embodiment of the invention, a UV/Ozone oxidation process, which can allow for a greater control in achieving a desired band gap, is used as the oxidation process.
In still another embodiment, a remote indirect plasma oxidation treatment, which allows for more expedient oxidation while providing a safer oxidation treatment as compared to a direct plasma oxidation treatment, is used as the oxidation process.
These and other features, aspects, and advantages of the present invention will be more fully understood when considered with respect to the following detailed description, appended claims, and accompanying drawings, wherein:
The current invention is directed to a precise, versatile, and novel dry oxidation method of fabricating a semiconducting graphene material that can be used in a number of practical applications, such as: nanoelectronics, high frequency low noise field effect transistors which can be used for amplifiers, full wave rectifiers, RF resonators and switches, and integrated circuits. In particular, the current invention recognizes that absent very rigorous process parameters, graphene oxide layers are prone to the development of substrate oxides that can negatively impact the electronic characteristics and usability of the materials for practical applications, and has further discovered that by confining the oxidation to the graphene layer it is possible to prevent these substrate oxides.
The following sections will elaborate on these basic fabrication steps, and will also provide descriptions of alternative embodiments that may be used in accordance with the above fabrication steps:
A variety of substrates may typically be used in the fabrication of electronic devices. As described above, the current invention is directed to graphene oxide materials that may be used in practical electronic applications. Accordingly, any substrate suitable for use as a structural foundation for a practical electronic device may be used with the current invention. For example, some commonly used substrate materials include: silicon, silicon dioxide, aluminum oxide, sapphire, germanium, gallium arsenide, an alloy of silicon and germanium, and indium phosphide, and it should be understood that any of these substrate materials may be used with the fabrication process of the current invention. In making a selection of an appropriate substrate material in accordance with the current invention, it will be understood that it is preferable that the influence of the processing applications on the substrate be minimized. In some cases this can be difficult, because the substrate is typically bonded to the electrical device prior to various processing applications, and so they too are subjected to those processing applications. In particular, as discussed above and in following sections, it is important to the creation of high quality graphene oxide materials that the dry oxidation treatment be confined to the graphene layer. Accordingly, in some embodiments the substrate material can be chosen such that the material is resistant to the production of surface oxides from exposure to oxygen ions.
As described above, once a substrate material has been chosen, a graphene layer must be deposited thereon. With regard to this step of the fabrication process, it will be understood that the deposition of graphene can be achieved in any manner suitable for the formation of a continuous graphene layer on the chosen substrate.
For example, in one embodiment of the invention, graphene samples can be grown by chemical vapor deposition techniques on nickel coated SiO2/Si substrates at 900° C. under a flow of 25 sccm methane and 1500 sccm hydrogen precursor gases. The sample can then be exposed to vacuum-pyrolysis treatment at an elevated temperature of 250° C. and a mild vacuum at 2.5 torr for 24 hours to remove the residual oxygen adsorbed during the growth process. Alternatively, in another embodiment of the invention, the graphene substrate configuration can be achieved by depositing single-layer graphene (SLG) flakes by micromechanical exfoliation on n-doped Si substrates covered with a 90 nm thermally grown SiO2 film.
It should be understood that the above processes are merely provided as examples and simply represent possible embodiments of the invention that illustrate how the desired substrate-graphene configuration can be achieved. The enumeration of these embodiments is not meant to imply that they represent the only ways a substrate-graphene configuration can be achieved in accordance with this invention—any method of obtaining this graphene substrate configuration can be used consistent with the invention.
As previously described with regard to the flow chart in
Importantly, any of the above mentioned dry oxidation techniques would allow for the creation of specific desired band gaps within the graphene material to be achieved by controlling the oxidation process. Generally speaking, a longer exposure to the oxidation process will result in a larger band gap. The longer exposure time is understood to allow the dry oxidation treatment to induce oxygen adsorbates onto the graphene layer. These oxygen adsorbates introduce defects in the graphene's inherent sp2 structure, and thereby disrupt the graphene's inherent if-bond network (the if-bond network is what facilitates the electron mobility and charge transport across the graphene plane).
Moreover, the specific type of dry oxidation treatment used also impacts the variability of the band gap—the use of a plasma oxidation treatment increases the band gap much more rapidly than does the use of a UV/Ozone oxidation treatment. This difference in rapidity is thought to be a function of the different concentrations of reactive oxygen species per unit time present in both treatments.
Although a number of different dry oxidation processes may be used with the current invention to produce graphene oxides, and to engineer the band gap characteristics of these materials, with regard to the current invention it is of critical importance that the oxidation treatment should be confined to the graphene layer, and that, in turn, the substrate should not be exposed to the oxidation process. Exposure of the substrate to oxidation, and or leaching of the oxide from the graphene layer to the substrate can lead to the formation of oxides on the substrate, which can negatively impact the graphene oxide material. Accordingly, by controlling the process parameters to ensure that the oxidation is confined within the graphene layer it is possible to improve surface quality and produce substantially defect-free graphene oxide semiconductor material suitable for practical use.
In view of this, the inventive oxidation process is tailored to avoid the formation of substrate oxides and ensure confinement of the oxide within the graphene layer.
With the above process restrictions in mind, the following embodiments of the invention are provided as examples and illustrate how dry oxidation treatments can be applied to a graphene sample in accordance with the invention:
(1) In one embodiment of the invention, a UV/Ozone treatment is applied at standard room temperature and pressure for a selected period of time such that the concentration of the oxygen within the graphene oxide layer does not exceed 21%.
(2) In another embodiment of the invention, a remote oxygen plasma machine (for example, the Tepla M4L) is used under 20 Watts of RF power at a constant oxygen flow rate of 20 sccm and chamber pressure of 500 mTorr for an appropriate amount of time. Note that in no case should more than 50 Watts of RF power be used when using the plasma oxidation treatment method. Again, the oxygen content within the graphene layer must be confined to 21% or less.
It should be understood that these are simply embodiments of the invention that illustrate how dry oxidation treatments may be applied, and are not meant to limit the scope of the invention—any suitable dry oxidation treatment method may be used in conjunction with the invention.
Although the above discussion has described basic embodiments of the invention, in some alternative embodiments the fabrication process is used in conjunction with masking techniques to allow for the creation of graphene/graphene oxide materials with multi-function surfaces and/or variable band gap regions. Masking is a technique used in circuit manufacture that allows for the creation of multiple regions with distinct electrical properties within a single layer. Essentially, masks are employed prior to the material's treatment, thereby protecting the covered region from the treatment's influence.
Some such embodiments, a summary of which is shown in the flow chart provided in
Thus, in one embodiment of this invention, masking can be employed prior to the dry oxidation treatment. Any region protected by the mask during an initial dry oxidation treatment will retain graphene's inherent semimetallic properties. Using this technique, a graphene oxide material with multiple regions can be developed. Accordingly, it is possible to incorporate masking techniques with the current fabrication technique allows for the creation of more intricate graphene samples.
As shown in
Moreover, the masking and oxidation processes can be iterated to achieve even further intricate semiconductor patterns. For example, a masking layer can be employed during an initial oxidation process, and then removed during a second oxidation cycle. The resulting material would have two regions with different band gaps: the region that was subject to the initial mask could have some appreciable band gap, whereas the region that was not exposed to the masking but still subjected to both oxidation treatments would have and even greater band gap. Moreover, different masking patterns can be used between the multiple oxidation steps to achieve even further intricate patterns. Thus, in yet another embodiment of this invention, multiple masking and oxidation cycles are used to obtain multiple regions of varying band gaps.
Many applications will be made possible by having the ability to oxidize graphene layer at a particular location or with a specific pattern. These applications include graphene based 2D LEDs, high frequency transistors and solar cells. These devices will certainly take advantage of graphene's ballistic electron mobility behavior as well as its intrinsic strong and light weight properties.
The following embodiments are only exemplary and illustrative in nature, and are not meant to limit the scope of the invention.
Graphene samples used in the following studies were grown by chemical vapor deposition technique on nickel coated SiO2/Si substrates at 900° C. under a flow of 25 sccm methane and 1500 sccm hydrogen precursor gases. (See, Reina A, et al., Nano Lett. 2009, 9, 30-5; Brien M O and Nichols B, Sensors Peterborough NH. 2010, TR, 5047; and Reina A, et al., Nano Research. 2009, 2, 509-16, the disclosures of which are incorporated herein by reference.) These as-grown samples were then exposed to vacuum-pyrolysis treatment at an elevated temperature of 250° C. and a mild vacuum at 2.5 torr for 24 hours to remove the residual oxygen adsorbed during the growth process. In the following discussion, these vacuum-pyrolysis treated graphene samples are referred as pristine samples. The presence of monolayer graphene on the pristine samples was confirmed by Raman spectroscopy (Renishaw M1000) obtained with excitation energy of 2.41 eV.
Two different oxidation processes were applied to the pristine graphene samples. The first set of graphene samples were oxidized by UV/ozone treatment (Bioforce Nanosciences) at standard room temperature and pressure for 5 minutes, 30 minutes and 120 minutes. Another set of graphene samples were oxidized by remote oxygen plasma (Tepla M4L) under 20 Watts of RF power at a constant oxygen flow rate of 20 SCCM and chamber pressure of 500 mTorr for 5 seconds, 10 second, 30 seconds and 60 seconds. For brevity, in the discussion that follows, oxygen plasma treated samples are referred as O2P samples followed by the exposure time in seconds, and UV/ozone treated samples are referred as UVO samples followed by the exposure time in minutes.
Atomic-resolution images of treated graphene samples were obtained using a scanning tunneling microscope (Digital Instrument Nanoscope IIIa ECSTM) equipped with Pt/Ir scanning tip (Veeco, Inc) at constant height mode. During the imaging process, a graphene sample was placed on a flat sample stage and clamped from the top with a metal electrode that creates a direct contact with the graphene film. Tunneling I-V and differential conductance characteristics were obtained using the tunneling spectroscopy capability of the same scanning tunneling microscope. All scanning tunneling microscopy/spectroscopy (STM/STS) imaging and measurements were conducted at room pressure and temperature at a scan rate of 20.3 Hz and a stabilization voltage and current of 100 mV and 650 pA respectively. Differential conductance characteristic, dl/dV, of each sample was obtained by averaging dl/dV curves from mat least five randomly selected locations on the sample. At higher degrees of oxidation, these curves were obtained from regions that still exhibit reasonable atomic periodicity and can be imaged without excessive noise using STM. Surface chemistry characterizations were assessed using x-ray photoelectron spectroscopy (Surface Science M-Probe XPS). The XPS spectral analysis was performed using a Gaussian-Lorentzian curve-fit with Shirley baseline correction.
The atomically resolved image of pristine graphene samples obtained by scanning tunneling microscope (STM) exhibits a highly-symmetric hexagonal lattice structures, which is a typical signature of a pristine graphene layer. In agreement with previously reported study, these hexagonal lattice structures show an atomic spacing of ˜0.23 nm (
As revealed by scanning tunneling spectroscopy (STS), the tunneling I-V characteristics of the oxidized graphene samples exhibit a deviation from that of the pristine graphene samples around the zero-bias region, where a sign of tunneling current flattening start to occur once the graphene is oxidized. As graphene undergoes longer duration of oxidation, this flat region becomes wider and more apparent. For example, the lightly oxidized UVO5m and O2P5s graphene show a vague flat region of 0.2 eV and 0.3 eV, while the heavily oxidized UVO120m and O2P60s graphene show a flat region as large as 1.8 eV and 2.4 eV. This large deviation of I-V characteristics of the oxidized graphene to the as-grown one suggests a strong correlation between the oxidation process and the electronic characteristics of oxidized graphene.
The evolution of electronic characteristics of oxygenated graphene can be understood by investigating the tunneling differential conductance, which is proportional to the local density of states (LDOS), at various durations and types of oxidation process. The tunneling differential conductance presented herein was calculated numerically by taking the first derivative of the tunneling current with respect to the bias voltage (
In contrast to that of the pristine graphene samples, the tunneling differential conduction spectra of the oxidized graphene samples shows a sign of flattening around the zero-energy region, suggesting a considerable suppression in the LDOS around the zero-energy. The mildly oxidized UVO5m graphene show a narrow flat region of about 0.2 eV around the zero-energy region. The suppression in the LDOS becomes much more pronounced as the graphene samples undergo a prolonged oxidation time. In fact, the heavily oxidized UVO120m and O2P60s graphene show extended suppression in the LDOS up to 1.8 eV and 2.4 eV (
In agreement to the previous studies, the extent of the energy gap of oxidized graphene seems to depend heavily on the oxidation time, where longer exposure time to UV/ozone and oxygen plasma treatments results in larger energy gap opening. (See, Alzina, Gokus & Childres, cited above.) It is important to note that the increase of energy gap opening in oxygen plasma treated graphene is significantly faster than that in UV/ozone treated graphene. For instance, after only 60 seconds of oxygen plasma treatment, the O2P60s graphene has an energy gap of 2.44 eV. In contrast, 120 minutes of UV/ozone treatment gives the UVO120m graphene an energy gap of 1.93 eV. Such difference may be caused by different concentrations of reactive oxygen species per unit time present in both treatments. Note that in both treatments the energy gap opening does not increase linearly with the increase of oxidation time (
It can be expected that the opening of energy gap in the LDOS of oxidized graphene is induced by the introduction of disordered defects in the sp2 structure of graphene due to the presence of oxygen adsorbates. These defects produce a strong disruption to the π-bond network that facilitates the electron mobility and charge transport across the graphene plane. (Elias, cited above.) Since the magnitude of such disruption depends heavily on the spatial distribution of defects sites and the degree of induced localization, an increase of defect density will certainly reduce the electron mobility. (See, Bostwick & Luo, cited above.) In addition, any alteration to the π-bond network near the defect sites further distorts the electron-phonon couplings and electron-electron interactions. (See, Luo, cited above; and Manes J L, Physical Review B. 2007, 76, 045430, the disclosure of which are incorporated herein by reference.) Therefore, the energy gap in the LDOS observed in this study can be associated with the electron mobility gap introduced by disordered defects in the .pi.-bond network. (See, Childres, cited above.)
A more meaningful relation can be obtained by correlating the energy gap of the oxidized graphene to its oxygen-to-carbon atomic ratio (O/C ratio). Basically, the O/C ratio represents the surface concentration of oxygen adsorbates. In this study, the O/C ratio of graphene samples was obtained from the x-ray photoelectron spectroscopy (XPS) survey scans. Multiple peaks related to C 1s, O 1s, Si 2s and Ni 2p can be seen on the XPS spectra of all graphene samples (
As expected, the correlation between energy gap of an oxidized graphene and its O/C ratio is monotonic, where a higher O/C ratio yields a larger energy gap, regardless of the oxidation method used. This finding implies that the observed energy gap opening is indeed introduced by defects created by oxygen adsorbates, which create disruption in the .pi.-bond network. These defects also induce a localization effect, where each of the oxidized site acts as a strongly repulsive hard wall barrier, and the degree of such localization is dictated by the spatial distribution and the density of the oxidized sites. (See, Luo, cited above.) In addition, the correlation confirms the existence of an O/C ratio threshold around 15%, above which the energy gap opening increases exponentially (
In contrast to the previous studies, a slight increase in O/C ratio below this threshold does not increase the opening of energy gap dramatically. (See, Leconte & Kim, cited above.) At such regime, however, the experimental data seem to agree with the prediction done by electronic band structure calculation, where an O/C ratio of ˜12% yields an energy gap opening of ˜0.2 eV. (See, Nourbakhsh, cited above.) On the other hand, a slight increase in O/C ratio above this threshold results in drastic increase of energy gap opening, which is not quite in agreement with the prediction done by electronic band structure calculation. Experimental data show that an energy gap opening of ˜1.5 eV can be obtained by an O/C ratio of just ˜18%. Clearly, such prediction underestimates the energy gap opening because the electronic gap calculation is only valid for graphene that retains its structural integrity. At a high O/C ratio (greater than 21%), the defect density becomes extremely high such that it is unlikely that the band structure has survived. This implies that the observed opening of energy gap beyond this limit may not be a band gap, and therefore is not suitable for use in electronics applications.
The presence of oxygen adsorbates in the oxidized graphene samples was further investigated using high-resolution XPS scans. Curve-fitting and deconvolution of the high-resolution XPS spectra of C 1s was performed using a Gaussian-Lorentzian peak shape with Shirley baseline correction. Deconvolution of the C 1s XPS spectra of both oxygen plasma and UV/ozone treated samples shows four distinct peaks associated with sp2 C—C (284.7±0.1 eV, FWHM 0.9 eV), C—O (285.2±0.1 eV, FWHM 1.45 eV), C═O (286.7±0.1 eV, FWHM 1.45 eV), and O—C═O (288.6±0.1 eV, FWHM 4 eV). (See, Yang D, et al., Carbon. 2009, 47, 145-52, the disclosure of which is incorporated herein by reference.) The C 1s XPS spectra of the pristine graphene samples show a very strong C—0 peak which may be caused by a significant presence of hydroxyl or epoxide groups at the edge. These spectra also show a relatively weak C═O peak and the absence of a peak associated with O—C═O group (
As mentioned earlier, the energy gap opening of graphene samples can be correlated to their oxygen adsorbates concentration. The pristine graphene samples are expected to have a very low surface concentration of oxygenated functional groups. On the other hand, higher surface concentration of these groups is expected to be found in graphene samples that have larger energy gap. Although the correlation is not exactly linear, the surface concentration of these groups does increase as the increase of energy gap opening of the graphene samples (
An oxygen uptake by the graphene layer during the oxidation process can also be observed in the O 1s and Ni2p XPS spectra of both oxygen plasma and UV/ozone treated samples. The O 1s spectra show three distinct components associated with O—C═O (531.5±0.3 eV, FWHM 1.4 eV), C═O (532.6±0.3 eV, FWHM 1.3 eV), and C—O (533.5±0.3 eV, FWHM 1.4 eV). (See, Yang, cited above.) The area percentage of each O 1s spectral component agrees with the corresponding component in the C 1s spectra. A major O 1s peak, which can be seen obviously in UVO120m and O2P60s samples, may be associated with absorbed hydroxide species or water (535.2±0.7 eV, FWHM 1.9 eV). (See, Biesinger M, et al., Surface and Interface Analysis. 2009, 41, 324-32, the disclosure of which is incorporated herein by reference.) An additional weak O 1s peak (536.2±0.8 eV, FWHM 2.1 eV) may be associated with physisorption of oxygen. (See, Biesinger, cited above.) The main Ni metal 2p3/2 spectral component can be found at 532.5±0.5 eV with FWHM of 1 eV, while the second and third ones can be found at binding energy shifts of +3.65 eV and +6.05 eV respectively, with FWHM of 2.6 eV and 2.8 eV respectively.
It is important to note that the energy gap opening reported herein is not induced by the production of nickel oxide or hydroxide during the oxidation process. High-resolution O 1s XPS spectra show the non existence of peaks associated with Ni—O for all samples, even after 60 seconds of oxygen plasma and 120 minutes of UV/ozone treatments (
It will be understood from the findings mentioned above that the oxygen uptake is indeed caused by the oxidation of graphene, as long as the samples are not over oxidized. For the heavily oxidized samples, i.e. UVO120m, O2P30s, and O2P60s samples, a small fraction of oxygen uptake is caused by the production of Ni(OH)2. Such an uptake may also induce an energy gap in the LDOS, rendering the data invalid for energy gap larger than ˜1.9 eV. If the energy gap data from these heavily oxidized samples are omitted, one can find an almost linear relation between the energy gap opening and the overall surface concentration of the oxygenated functional groups.
The effect of both oxygen plasma and UV/ozone treatments to the electronic structure of graphene can be literally seen from the evolution of the atomically resolved images obtained by scanning tunneling microscope (STM). As graphene is oxidized, defects on the hexagonal lattice structure due to the presence of oxygen adsorbates in the form of oxygenated functional groups start to occur. For samples with low concentration of oxygen adsorbates, i.e. O2P5s and UVO5m samples, the degree of disorder is quite low such that the hexagonal lattice structures still can be recognized from the raw STM images without using any further image processing technique (
At even higher concentration of oxygen adsorbates, i.e. UVO120m and O2P60s samples, the defect density becomes very high, such that the hexagonal patterns cannot be recognized anymore from the raw STM images (
Applicant has disclosed a novel fabrication method that allows for a versatile but precise manipulation of graphene so as to develop within it an appreciable band gap while at the same time ensuring that oxidation is confined within the graphene layer. Such a graphene material can be very well-suited for a host of practical electronic applications.
The experimental findings presented herein show the effect of dry oxidation processes, in particular oxygen plasma and UV/ozone treatments, to introduce an energy gap opening in graphene. The opening of the gap itself can be correlated to the surface concentration of oxygen adsorbates, where the energy gap increases strongly as the increase of oxygen adsorbates concentration. In fact, an increase of oxygen-to-carbon atomic ratio from ˜9% to ˜21% is enough to increase the energy gap opening from 0 eV to ˜2.4 eV. Note that a significantly observable energy gap opening occurs when the oxygen adsorbates concentration is higher than the oxygen-to-carbon atomic ratio threshold of ˜15%. At high oxygen adsorbates concentration (˜21%), the defect density becomes extremely high such that it is unlikely that the band structure has survived. This implies that the observed opening of energy gap may be associated with a mobility gap, not a band gap. The presence of oxygenated functional groups, e.g. C—O, C═O and O—C═O groups, induces distortion to the .pi.-bond network of the graphene such that the hexagonal patterns become much less apparent and the rectangular unit cells become much more pronounced. In general, the oxygen plasma treatment gives a much faster rate of oxidation than the UV/ozone treatment. On the other hand, the slower oxidation rate of UV/ozone treatment may provide a better control over the degree energy gap opening.
Those skilled in the art will appreciate that the foregoing examples and descriptions of various preferred embodiments of the present invention are merely illustrative of the invention as a whole, and that variations in the fabrication methodologies of the present invention, may be made within the spirit and scope of the invention. Accordingly, the present invention is not limited to the specific embodiments described herein but, rather, is defined by the scope of the appended claims. Unless otherwise specified, all of the references disclosed herein are incorporated by reference into the specification.
This filing is a continuation application of U.S. patent application Ser. No. 14/095,454, filed Dec. 3, 2013, which is a continuation application of U.S. patent application Ser. No. 13/316,771, filed Dec. 12, 2011, now U.S. Pat. No. 8,609,458, which claims the benefit of and priority to U.S. Provisional Patent Application No. 61/422,092, filed Dec. 10, 2010, all of which are incorporated by reference herein in their entireties and for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 61422092 | Dec 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14095454 | Dec 2013 | US |
| Child | 15254296 | US | |
| Parent | 13316771 | Dec 2011 | US |
| Child | 14095454 | US |
