Protection from Below: Stabilizing Hydrogenated ... - ACS Publications

Publication Date (Web): November 9, 2017. Copyright This article not ... Self-assembled monolayers (SAMs) using aliphatic (ATs) and aromatic ... Abstr...
0 downloads 0 Views 1MB Size
Subscriber access provided by University of Florida | Smathers Libraries

Article

Protection from below: stabilizing hydrogenated graphene using graphene underlayers Keith E. Whitener, Jeremy T Robinson, and Paul E. Sheehan Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03596 • Publication Date (Web): 09 Nov 2017 Downloaded from http://pubs.acs.org on November 14, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Protection from below: stabilizing hydrogenated graphene using graphene underlayers Keith E. Whitener, Jr.,*1 Jeremy T. Robinson,2 and Paul E. Sheehan1 1.

Chemistry Division, U.S. Naval Research Laboratory, Washington, DC 20375, United States.

2.

Electronic Science and Technology Division, U.S. Naval Research Laboratory, Washington, DC

20375, United States. KEYWORDS. Graphene functionalization, hydrogenated graphene, device stability, Raman spectroscopy

ABSTRACT:

We show that dehydrogenation of hydrogenated graphene proceeds much more slowly for bilayer systems than for single layer systems. We observe that an underlayer of either pristine or hydrogenated graphene will protect an overlayer of hydrogenated graphene against a number of chemical oxidants, thermal dehydrogenation, and degradation in an ambient environment over extended periods of time. Chemical protection depends on the ease of oxidant intercalation, with good intercalants such as Br2 demonstrating

much

higher

reactivity

than

poor

intercalants

such

as

1,2-dichloro-4,5-

dicyanonbenzoquinone (DDQ). Additionally, the rate of dehydrogenation of hydrogenated graphene at 300°C in H2/Ar was reduced by a factor of roughly 10 in the presence of a protective underlayer of graphene or hydrogenated graphene. Finally, the slow dehydrogenation of hydrogenated graphene in air at room temperature, which is normally apparent after a week, could be completely eliminated in

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

samples with protective underlayers over the course of 39 days. Such protection will be critical for ensuring the long-term stability of devices made from functionalized graphene.

Introduction The atomic thinness of graphene provides chemical and physical effects useful for surface engineering. For example, graphene applied to a substrate can tune its interactions with the external world, controlling the transmission of van der Waals forces from the substrate1 or even vastly changing the substrate’s wettability.2 Graphene can also place chemical groups on materials whose surface chemistry would otherwise be difficult to modify, thereby increasing its chemical versatility. This same atomic thinness also makes graphene uniquely chemically sensitive to an underlying substrate. For example, the chemical reactivity of graphene strongly depends on the substrate—single layer graphene on SiO2 (Gr) is far more reactive to diazonium grafting than bilayer graphene (Gr/Gr).3,4 As another example, fluorination of single layer graphene proceeds much more extensively when the substrate is SiO2 than when the substrate is copper or gold.5 In these cases, the higher reactivity of graphene on insulators is attributed to the inhomogeneous charge distribution (“charge puddles”) in the substrate directly beneath the graphene, causing some sites to become more reactive than others. Conductors delocalize these charge inhomogeneities, rendering the entire sheet of graphene less reactive.4 Many chemical reactions with graphene are reversible; in particular, hydrogenation of Gr is thermally, chemically, and mechanically reversible.6-9 While this property can be beneficial—for example, enabling patterning of electrically conductive regions of graphene in an insulating and n-doping hydrogenated graphene matrix—the reversion of hydrogenated graphene to pristine graphene raises the question of how best to stabilize devices made from these materials for the long-term. Indeed, worries over the stability of hydrogenated graphene were highlighted by Geim and Grigorieva, who stated, “Graphane (fully hydrogenated graphene) gradually loses its hydrogen and is unlikely to be useful for making heterostructures,”10 despite its shown promise in spintronics,11 in transferring ferromagnetism,12 and in strain-induced band gap formation.13 Thus, finding a way to chemically stabilize hydrogenated ACS Paragon Plus Environment

2

Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

graphene represents a critical step in enabling practical use of this unique material in heterostructure devices. In this paper, we report hydrogenated graphene with an underlayer of either pristine graphene (HG/Gr) or hydrogenated graphene (HG/HG) do not dehydrogenate under the same conditions as single layer hydrogenated graphene on SiO2 (HG). In particular, while small molecules such as bromine vapor will efficiently dehydrogenate all single layer and bilayer species, larger molecules such as the organic oxidizer DDQ only dehydrogenate HG, leaving the bilayers functionalized. We hypothesize that the difference in reactivity stems from the ability of an oxidizer to intercalate between graphene layers, based on observations of reactivity with a number of other oxidizers with varying intercalation ability. Other groups have reported layer-number dependence of hydrogenation stability when multilayer HG is heated in a vacuum.14 We corroborate this result in an H2/Ar atmosphere at 300°C; indeed, under these conditions, bilayer species show a roughly 10-fold reduction in the rate of hydrogen loss compared with monolayer species. HG/Gr also shows a significant stabilization versus HG with respect to dehydrogenation over a period of several weeks in ambient conditions (room temperature in air). Thus, we have demonstrated that multiple layers of graphene are far more robust to loss of chemical functionalization than single layers under a wide variety of conditions. These results are an important step toward designing functional graphene devices that are stable not merely under tightly controlled laboratory conditions, but also in practical settings with less forgiving environments. Methods Raman spectroscopy was performed on a Renishaw spectrometer with an excitation wavelength of 514 nm. Sheet resistivity measurements were carried out using a 4-point collinear probe method with a Keithley SCS-4200. Unless otherwise stated, all chemicals were purchased commercially and used without further purification. Graphene was synthesized using low-pressure chemical vapor deposition (CVD) with H2 and CH4 source gases at 1030°C.34,35 After growth, graphene was transferred to silicon wafers with 100-nm thermal oxide using a PMMA protective layer and etching the copper substrate. ACS Paragon Plus Environment

3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

Twisted bilayer graphene was prepared by performing the graphene transfer procedure twice. Hydrogenation of Gr and Gr/Gr was carried out using a Birch reduction method which has been previously described.8 Transfer of HG onto Gr was carried out using a delamination method developed by our group which has been previously described.12 Reaction of graphene with bromine vapor was carried out at room temperature. Approximately 5 µL of Br2 liquid was added to a nitrogen-purged glass 20-mL scintillation vial with a polypropylene screw top. The top was sealed and the bromine vapor was allowed to fill the vessel. The graphene samples were then added and the vessel was resealed. The samples were allowed to react for 30 minutes. The vessel was then opened and purged with a nitrogen gun for several seconds before the samples were removed. Reaction of graphene with 1,2-dichloro-4,5-dicyanonbenzoquinone (DDQ) was carried out at room temperature. The samples were added to 10 mL of a saturated solution of DDQ in DCM. Reaction times varied between 90 minutes and 3 days, as indicated in the main text. Upon removal from the DDQ bath, the samples were rinsed with ethanol and dried under nitrogen gas. Results and Discussion We have previously examined several chemical routes to dehydrogenate HG.9 We report here that dehydrogenation of HG depends on the number of layers, and we examine the differences in reactivity when HG rests directly on silicon oxide, when it rests on HG (i.e., HG/HG), and when it rests on Gr (i.e., HG/Gr). To produce HG/HG, we treat a twisted bilayer graphene film15 with the Birch reduction, which hydrogenates both layers.8,16,17 To produce HG/Gr, we transfer Birch hydrogenated graphene, without loss of functionality,12 onto a pristine monolayer of CVD graphene. We note that, although the Birch reduction is a wet chemical technique, it has been shown not to leave residual lithium or ammonia in the hydrogenated graphene system after terminating the reaction.8,18,19

ACS Paragon Plus Environment

4

Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Because the electronic nature of the substrate could potentially impact the chemistry of the graphene overlayer, both an electrically insulating underlayer, HG, and a conductive underlayer, Gr, were used. Notably, HG/HG and HG/Gr exhibit comparable stability, with HG/Gr being only slightly less stable than HG/HG, indicating that the presence or absence of hydrogen on the underlayer and the concomitant changes in conductivity have little effect on the overlayer’s dehydrogenation. We employ Raman spectroscopy, optical microscopy and electronic transport measurements to follow functionalization changes in our HG systems. Raman spectroscopy is a sensitive probe of graphene’s defect density. In particular, the spectrum undergoes unambiguous changes when the sp2 conjugation of the graphene sheet is disturbed, making it a powerful tool for tracking the sp3 defect sites introduced by hydrogen chemisorption.6 We note that lattice vacancies, edges, and topological defects can also give rise to the same spectral features as hydrogen chemisorption.20 In fact, our chemical treatment likely produces some such defects, as in most cases these spectral features do not completely disappear upon restorative treatment. However, the extent of restoration is significant, and given that vacancy and topological defects are relatively high-energy structures (>5eV), which are not accessible to ‘healing’ under 300 °C thermal treatment or oxidant exposure, we can confidently assign the restoration of the sp2-carbon Raman spectra to H desorption alone. Four features of graphene’s Raman spectrum correlate with functionalization changes during chemical or thermal treatment. First is the presence of a ‘defect’ (D) peak at ~1345 cm-1, whose intensity relative to the first order G peak at ~1585 cm-1 changes non-monotonically as defect density increases. Second is the dramatic increase in background signal and sloped baseline in heavily hydrogenated species due to photoluminescence arising from electronically isolated polycyclic aromatic “islands” in a functionalized graphene “sea.”21-23 To quantify the high background signal from the broad photoluminescence and control for fluctuating laser power, we report the ratio of the intensity at 2000 cm-1 to the G peak maximum intensity and refer to this ratio as B/G (“B” for baseline) throughout. This ratio decreases as defects are removed and the baseline approaches zero. Third is the broadening of all ACS Paragon Plus Environment

5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

peaks as defect density increases.24-26 Fourth is the monotonic intensity decrease of the second order defect peak (2D) peak at ~2690 cm-1 with respect to the G peak as defect density increases. Tracking all four spectral changes provides the fullest understanding of the hydrogenation process; however, because the D/G peak ratio is non-monotonic over the range of our experimental conditions, we only consider it qualitatively. The other three features—the B/G ratio, the D peak full width at half maximum (FWHM), and the G/2D ratio—are all monotonic with respect to defect density under the experimental conditions and will be handled quantitatively.25 Note that tracking multiple spectral features is necessary to follow the chemistry at all stages of processing since pristine graphene does not have a measurable D peak, while the G/2D ratio in bilayers depends on the twist angle.15 The impact of having variable twist angles, which leads to a variable enhancement of the G peak, is partly reflected in the large standard deviations for the G/2D and B/G ratios. We point out, however, that comparable standard deviations can also be seen in some single layer graphene measurements, where twist angle is not a factor. This effect reflects the fact that dehydrogenation is inhomogeneous across large areas of the graphene. Measured values of Raman spectral features (G/2D ratio, D FWHM, and B/G ratio) for the chemical dehydrogenation experiments are summarized in Figure 1, and are fully tabulated in the supplementary information (Table S1). We compared the Raman spectra of HG and HG/HG before and after exposure to two different oxidizers: bromine vapor and DDQ dissolved in dichloromethane (DCM). These spectra are shown in Figure 2, along with optical microscopy images taken before and after Birch hydrogenation and after subsequent exposure to oxidant. As graphene becomes more hydrogenated, its optical absorption decreases (i.e., it becomes more transparent).27,28 As shown in Figure 2a, exposure of HG to DDQ induces several dramatic changes in the Raman spectra. The G/2D ratio decreases from 2.80 ± 0.34 before the reaction to 0.57 ± 0.15 afterwards, the D peak FWHM decreases by 2× (from 58.2 ± 5.5 cm-1 to 28.6 ± 1.7 cm-1), and the B/G ratio decreases from 0.45 ± 0.07 to 0.11 ± 0.02. All of these changes are associated with removal of sp3 chemisorbed hydrogen defects and restoration of the sp2 graphene lattice. In contrast, the Raman spectra (Figure 2b) of HG/HG show no recovery even after 3 days of DDQ ACS Paragon Plus Environment

6

Page 7 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

exposure, with the G/2D ratio, D peak FWHM, and B/G ratio changing only slightly over the course of the reaction. The optical micrographs (Figures 2c-e) of the sample show the pre-Birch reduction contrast of Gr and Gr/Gr, as well as the dramatic decrease of contrast accompanied with the Birch hydrogenation. Upon DDQ exposure, the single layer regions regain much of their original optical contrast while the bilayer region does not. This observation contrasts with the hydrogenated samples that were exposed to bromine vapor, whose Raman spectra (Figures 2f, g) as well as optical contrast (Figures 2h-j) restore for both single layer and bilayer regions. The electrical resistivity of an HG film was used as another measure of chemical functionalization since hydrogenation disrupts electron pathways, dramatically increasing the sheet resistance.6,8 Using a four-point collinear probe technique and a Keithley 4200-SCS parameter analyzer, we measured the sheet resistance of Gr and Gr/Gr before hydrogenation, after hydrogenation, and after exposure to Br2 vapor and DDQ. Resistivity results are summarized in Table 1. Before hydrogenation, the sheet resistances for pristine Gr/Gr and Gr were 167.9 ± 33.5 Ω/□ and 670.9 ± 136.8 Ω/□, respectively. These samples were exposed to Br2 before measurement to control for any doping effect. After hydrogenation, both of these samples became highly resistive, giving open-circuit (>100 GΩ/□) values. Subsequent exposure to Br2 vapor recovers the electrical conductivity with the sheet resistances decreasing to ~5× that of the starting material. In the case of DDQ exposure, the sheet resistances for Gr/Gr and Gr before hydrogenation were 485.3 ± 39.2 Ω/□ and 927.4 ± 116.9 Ω/□, respectively. After hydrogenation, again both samples were open-circuit (>100 GΩ/□). After exposure to DDQ, the sheet resistances for Gr/Gr and Gr increased to roughly 10× that of the starting material. Thus, both Br2 and DDQ can restore electrical conductivity to Gr/Gr as well as Gr, but the restoration proceeds more completely with Br2. It is intriguing that both Br2 vapor and DDQ mostly restore conductivity to HG/HG, whereas only Br2 restores the Raman spectrum. This indicates that after exposure to DDQ, HG/HG remains heavily hydrogenated but the electronic band structure of the material changes sufficiently to allow conduction. It has been shown that Birch hydrogenation of multilayer graphene proceeds via intercalation,16 such ACS Paragon Plus Environment

7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

that both sides of the top graphene layer of HG/HG are hydrogenated. The results here are consistent with DDQ dehydrogenating only the top surface of the top layer such that it resembles semihydrogenated graphene (sometimes referred to as “graphone”).29 This structure should retain its characteristic heavily defected Raman spectrum while having a much lower band gap, ~0.5 eV, than fully hydrogenated graphene, ~3.5 eV,29,30 consistent with the observed restoration of conductivity, though further studies are warranted. Another possible explanation, the formation of a conductive DDQgraphene charge-transfer complex, was ruled unlikely in light of our observation that exposure of HG/HG to the powerful electron acceptor tetracyanoquinodimethane (TCNQ) for 90 minutes does not restore conductivity. Given that stacking of the graphene layers could profoundly impact their reactivity, we developed an approach with internal controls. HG was transferred onto Gr using a hydrogen-assisted transfer technique to yield three distinct areas (Figure 3a) that could be simultaneously reacted with Br2 and DDQ.12 Critically, this sample configuration allows us to monitor the Raman spectra of HG and Gr independently, as well as together as a bilayer stack, to determine whether stacking impacts chemistry. The Raman spectrum of the unreacted HG/Gr stack is well approximated by simply summing the individual HG and Gr spectra (schematic in Figure 3b), indicating that the HG and Gr layers in the bilayer stack are only weakly electronically and vibrationally coupled. Indeed, comparing the direct sum of the HG and Gr spectra (HG+Gr, Figures 3c, e) to the single spectrum for the HG/Gr stack (HG/Gr, Figures 3d, f) is useful for following interlayer interactions. For example, after a 90-minute exposure to DDQ, the summed HG+Gr spectra (Figure 3c) exhibited extensive dehydrogenation while the HG/Gr spectrum (Figure 3d) does not show any change. In contrast, after a 30-minute exposure to Br2 vapor, both the HG+Gr summed Raman spectra (Figure 3e) and the HG/Gr spectrum (Figure 3f) show comparable loss of hydrogen. These comparisons support the picture that both DDQ and Br2 dehydrogenate single layer HG on SiOx, that DDQ fails to dehydrogenate HG on graphene, and that Br2 vapor dehydrogenates the HG/Gr stack. ACS Paragon Plus Environment

8

Page 9 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Delamination experiments on stacked films corroborate that an underlayer of graphene or HG protects HG from dehydrogenation (Figure S1). Hydrogenated graphene readily delaminates from silicon oxide when exposed to water, whereas pristine graphene does not.12 This difference likely stems from two effects: first, hydrogenation on the underside of the graphene separates the graphene from the substrate, weakening the van der Waals adhesion force which diminishes quickly with distance. Second, hydrogenation makes graphene electrically insulating, which weakens the low frequency response of the dielectric function and thereby diminishes the Hamaker constant, a measure of the overall strength of the van der Waals interactions. This observation also demonstrates that hydrogen bound to the top of the graphene does not induce the dangling bond on graphene to directly bind the substrate below it. After hydrogenated samples (HG/HG, HG, or HG/Gr) are exposed to Br2 vapor, they no longer delaminate from their substrates. In contrast, after exposure to DDQ, HG does not delaminate, but HG/HG delaminates cleanly from its substrate, indicating that HG/HG still bears enough hydrogen to weaken the van der Waals adhesion forces. On the other hand, HG/Gr seems to delaminate with significant tearing, even in the single layer regions. This may be owing to the fact that, as we see in the Raman spectra, DDQ is less effective at removing hydrogen than Br2 vapor. To understand the ability of an underlayer to protect graphene’s functionalization, we dehydrogenated these same systems (HG, HG/HG, HG/Gr) with a range of oxidizers, viz., iodine monochloride vapor, perfluorobutyliodide in hexane (PFBI) with UV light exposure, and chlorine and bromine dissolved in DCM. The extent of dehydrogenation is shown via a color map in Table 2, with darker squares indicating more complete dehydrogenation, as judged by the change in G/2D ratio, D peak FWHM, and B/G ratio from before to after reaction. The numbers in the table are the averaged ratios of the spectral values before and after the dehydrogenation reaction (cf. Figure 1). Lower numbers indicate more complete dehydrogenation. Full details of the analysis are given in the Supporting Information. Compounds that spontaneously form graphite intercalation complexes31 (Br2 vapor, ICl vapor) dehydrogenate both single layer and bilayer samples extensively. Bulkier oxidants (iodine from PFBI, DDQ) and solvent-complexed oxidants (Br2 and Cl2 in DCM) do not dehydrogenate multilayer ACS Paragon Plus Environment

9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 21

graphene or, at best, do so slowly. So, an oxidant’s ability to intercalate between graphene layers is correlated with its ability to dehydrogenate them. Indeed, optical images of HG/HG after exposure to Br2/DCM and Cl2/DCM (Figure S6) show patchy recovery of optical contrast consistent with a slow rate of intercalation between layers. Another potential source of instability of HG is heat: HG rapidly dehydrogenates at elevated temperatures, so we examined whether additional graphene layers would protect HG against thermal dehydrogenation. These studies also lend insight into the stabilization mechanism, as thermal dehydrogenation in an inert atmosphere does not depend on an oxidizer’s ability to intercalate between graphene layers. To examine this notion, HG, HG/Gr, and HG/HG were heated to 300°C under 10% H2/Ar gas, and Raman spectra were captured at multiple time points. Figure 4 plots the evolution of the G/2D ratio, D peak FWHM, and B/G ratio. The dehydrogenation of HG is heterogeneous across the film, as seen in optical images (Figure S7) as well as in the large standard deviations for intermediate times shown in Figure 4. It is clear, though, that the G/2D ratio and D peak FWHM for HG (Figure 4a, b, upward triangles) recover much more readily than for either HG/HG (Figure 4a, b, downward triangles) or for HG/Gr (Figure 4a, b, squares), indicating that HG dehydrogenates more quickly than either of the multilayer samples. The photoluminescence, as indicated by the B/G ratio, recovers relatively evenly for all species (Figure 4c), indicating that the level of hydrogenation required for photoluminescence is significantly higher than that required for the other Raman features, and the pendant hydrogens in this regime are quite thermally labile. Thermal dehydrogenation proceeds much more rapidly for HG than for HG/HG. Indeed, HG has almost completely reverted to Gr after only 30 minutes, whereas for HG/HG, the prominent Raman signatures of hydrogen chemisorption remain even after 120 min. Under the same conditions, HG/Gr stacks show only slight recovery. To further examine details of thermally treated HG samples, we employed the sample geometry shown in Figure 3 with internal controls. Specifically, the thermal treatment of the HG/Gr bilayer region (Figure 4, squares) was compared to the summed Raman spectra of HG+Gr (Figure 4, circles). The ACS Paragon Plus Environment

10

Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

G/2D ratio and D peak FWHM trends for the bilayer species restore much more slowly than the trends for the summed spectra, indicating that both graphene and hydrogenated graphene underlayers can protect the graphene overlayer from thermal dehydrogenation. We can get a rough estimate of the magnitude of this protective effect by noting that the evolution of the D peak FWHM (Figure 4b) is well-approximated by an exponential function that asymptotically decays to some value Γ0. For bilayer samples, we assume that the D peak is approximated by the sum of two Lorentzian lineshapes with the same peak center and FWHM. It is easily shown that this procedure produces a function with the same FWHM as before. Thus both single and bilayer species will decay to the same Γ0, which we fix at 25 cm-1. An exponential fit gives decay constants of -0.061 cm-1/min and -0.040 cm-1/min for HG+Gr and HG, respectively, and decay constants of -0.0029 cm-1/min and -0.0038 cm-1/min for HG/Gr and HG/HG, respectively. These results indicate very roughly that thermal dehydrogenation in single layer samples proceeds about 10 times faster than in bilayer samples. While the enhanced graphene reactivity due to the SiO2 substrate likely plays a role,4,32 additional mechanisms may be at play in this remarkable difference in dehydrogenation rates between single layer and multilayer samples. Zhang et al. reported that Birch hydrogenation of multilayer graphene proceeds from edges and defects in an intercalation process;16 microscopic reversibility would assert that the reverse process of dehydrogenation most likely proceeds from edges and defects as well. Hydrogen desorption from single layer graphene is highly entropically favored, as the gaseous products simply escape to infinity. However, in the interlayer space in bilayer hydrogenated graphene, desorbed hydrogen cannot simply escape to infinity. Instead, it must diffuse to an edge or a defect in order to leave the system. Meanwhile, the migrating hydrogen can readsorb to either of the surrounding graphene sheets, significantly slowing desorption kinetics. Thus, multilayers effectively protect one another against dehydrogenation in a fashion which is not possible for single graphene layers. Finally, the observations5,21 that functionalized graphene in ambient conditions slowly desorbs chemical groups over time motivated us to examine the time-dependent chemical stability of HG with ACS Paragon Plus Environment

11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 21

different underlayers. By comparing the Raman spectra from a HG/Gr stack (Figure 5, circles) with the summed HG+Gr spectra (Figure 5, squares), we find that single layers show significant reduction of G/2D ratio and D peak FWHM after about a week (in air at RT) whereas HG/Gr stacks do not show any appreciable recovery of their G/2D ratios or D peak FWHM after 39 days. In addition, the summed HG+Gr Raman spectra (Figure 5c squares, Figure S8a) begin to show immediate declines of the high background and photoluminescence which are characteristic of extensive hydrogenation, while HG/Gr (Figure 5c circles, Figure S8b) maintains a significant degree of photoluminescence, which requires very high levels of hydrogenation, even after 39 days. Thus, the use of multiple layers is a robust strategy to stabilize hydrogenated graphene against dehydrogenation in air over long times. Conclusions In conclusion, we have shown that both HG/HG and HG/Gr stacks are more stable to dehydrogenation than HG alone when exposed to high temperatures, to ambient environment, or to chemical treatment by a range of oxidizers. Given that HG is highly insulating and Gr is conductive, the electronic and van der Waals interactions between layers are starkly different, and yet the protective effect of an HG overlayer persists. In addition, as recent work has shown,15,33 twisting the bilayer affects the system’s electronic structure and could feasibly have some effect on the chemistry of the material. The observations that hydrogenation is preserved on twisted bilayers as well as on HG or Gr are therefore an indication that the interlayer electronic and van der Waals interactions do not strongly dictate the protective effect of the second graphene layer. Instead, the chemical protection provided by the intervening layer depends on its ability to prevent the oxidant from intercalating between graphene layers. For ‘bulky’ oxidants, or when otherwise good intercalants are complexed in solution, dehydrogenation is less effective than for ‘good’ intercalants in the gas phase, such as Br2 and ICl. A similar mechanism likely holds for dehydrogenation in ambient conditions: the most probable oxidizer is O2, which does not readily intercalate and therefore is inhibited from reaction in bilayer graphene. A similar, but somewhat more involved mechanism is proposed for thermal dehydrogenation. The ACS Paragon Plus Environment

12

Page 13 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

presence of multiple graphene layers likely impedes desorption or deintercalation of hydrogen atoms from the interlayer interstices. Hydrogen atoms may also, at elevated temperatures, desorb and re-adsorb between the various layers in multilayer species. The stabilization of hydrogenated graphene afforded by stacking multiple layers will certainly prove useful for researchers wishing to build hydrogenated graphene devices which are robust and durable to chemical and thermal changes. FIGURES AND TABLES

Figure 1. Summary of analysis of Raman spectra for the different graphene species and chemical oxidation conditions examined in this paper. The spectral values of G/2D ratio, D peak FWHM, and B/G ratio are normalized to 1 before the oxidation reaction takes place (gray horizontal line). The triplet of gray error bars indicate 1 standard deviation for G/2D ratio, D peak FWHM, and B/G ratio, respectively, before oxidation. The triplet of red bars indicates the spectral values of G/2D ratio, D peak FWHM, and B/G ratio, respectively, after the oxidation reaction. The red error bars also indicate 1 standard deviation. ACS Paragon Plus Environment

13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 21

Figure 2. Stabilization of hydrogenated bilayer graphene against chemical oxidation. a) Raman spectra of HG exposed to DDQ in DCM for 0 min, 90 min, and 3 days. b) Raman spectra of HG/HG exposed to DDQ in DCM for 0 min, 90 min, and 3 days. c) Optical image of Gr/Gr and Gr before Birch reduction. d) Optical image of HG/HG and HG after Birch reduction. e) Optical image of HG/HG and HG after exposure to DDQ. All scale bars in c-e are 100 µm. f) Raman spectra of HG exposed to Br2 vapor for 0 min and 30 min. g) Raman spectra of HG/HG exposed to Br2 vapor for 0 min and 30 min. h) Optical image of Gr/Gr and Gr before Birch reduction. i) Optical image of HG/HG and HG after Birch reduction. j) Optical image of Gr/Gr and Gr after exposure to Br2 vapor. All scale bars in h-j are 100 µm.

ACS Paragon Plus Environment

14

Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3. Spectral analysis of hydrogenated graphene on graphene. a) Experimental setup for monitoring HG, Gr, and the HG/Gr stack on the same SiOx wafer. b) Schematic for Raman spectrum analysis. See text for full details. c) Additive Raman spectra of HG+Gr upon exposure to DDQ for 0 min and 90 min. d) Raman spectra of HG/Gr stack upon exposure to DDQ for 0 min and 90 min. e) Additive Raman spectra of HG+Gr upon exposure to Br2 vapor for 0 min and 30 min. f) Raman spectra of HG/Gr stack upon exposure to Br2 vapor for 0 min and 30 min.

ACS Paragon Plus Environment

15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

Figure 4. Evolution of the Raman spectra of hydrogenated graphenes with respect to time for thermal dehydrogenation at 300°C under H2/Ar. a) G/2D peak ratios of Raman spectra versus heating time for HG, HG/HG, HG/Gr, and HG+Gr. b) D peak FWHM of Raman spectra versus heating time for HG, HG/HG, and HG/Gr. c) B/G ratios of Raman spectra versus heating time for HG, HG/HG, and HG/Gr.

ACS Paragon Plus Environment

16

Page 17 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5. Stability of hydrogenated graphene in air over time. a) G/2D peak ratios of Raman spectra versus time for HG/Gr and HG+Gr. b) D peak FWHM of Raman spectra versus time for HG/Gr and HG+Gr. c) B/G ratios of Raman spectra versus time for HG/Gr and HG+Gr. Error bars indicate one standard deviation.

ACS Paragon Plus Environment

17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 21

Table 1. Sheet resistance measurements of Gr/Gr and Gr before and after hydrogenation, and after exposure to Br2 or DDQ. All values are in Ω/□. Sample

Oxidant

Before hydrogenation

After Hydrogenation

After Oxidation

Br2 vapor

670.9 ± 136.8

>100 GΩ/□

2280. ± 523

DDQ

927.4 ± 116.9

>100 GΩ/□

8768 ± 1351

Br2 vapor

167.9 ± 3.5

>100 GΩ/□

713.6 ± 46.5

DDQ

485.3 ± 39.2

>100 GΩ/□

4684 ± 1945

Gr

Gr/Gr

Table 2. Oxidants and their ability to dehydrogenate HG, HG/Gr, and HG/HG. Darker squares correlate with spectral features indicative of more fully dehydrogenated species. Experimental procedures, details of the analysis, spectra, and images can be found in the SI. Oxidant\Sample HG

HG/Gr

HG/HG

DDQ

0.316

0.988

0.946

Cl2/DCM

0.258

0.606

0.810

PFBI + UV

0.339

0.520

0.863

Br2/DCM

0.225

0.467

0.604

ICl

0.230

0.418

0.283

Br2 vapor

0.202

0.270

0.240

More hydrogen

Less hydrogen

Supporting Information. The following files are available free of charge. Additional experimental procedures, detailed Raman analysis, optical images and Raman spectra (PDF). AUTHOR INFORMATION Corresponding Author *Keith Whitener: [email protected] Author Contributions ACS Paragon Plus Environment

18

Page 19 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

KEW performed experiments. KEW and JTR synthesized materials. KEW, JTR, and PES designed experiments and wrote the manuscript. All authors have given approval to the final version of the manuscript. ACKNOWLEDGMENT This work was supported by the Defense Threat Reduction Agency (DTRA) under funding document HDTRA1722938 and the NRL Nanoscience Institute. REFERENCES (1) Tsoi, S.; Dev, P.; Friedman, A. L.; Stine, R.; Robinson, J. T.; Reinecke, T. L.; Sheehan, P. E. van der Waals Screening by Single-Layer Graphene and Molybdenum Disulfide. ACS Nano 2014, 8, 12410-12417. (2) Rafiee, J.; Mi, X.; Gullapalli, H.; Thomas, A. V.; Yavari, F.; Shi, Y. F.; Ajayan, P. M.; Koratkar, N. A. Wetting transparency of graphene. Nat. Mater. 2012, 11, 217-222. (3) Koehler, F. M.; Jacobsen, A.; Ensslin, K.; Stampfer, C.; Stark, W. J. Selective Chemical Modification of Graphene Surfaces: Distinction Between Single- and Bilayer Graphene. Small 2010, 6, 1125-1130. (4) Wang, Q. H.; Jin, Z.; Kim, K. K.; Hilmer, A. J.; Paulus, G. L. C.; Shih, C.-J.; Ham, M.H.; Sanchez-Yamagishi, J. D.; Watanabe, K.; Taniguchi, T.; Kong, J.; Jarillo-Herrero, P.; Strano, M. S. Understanding and controlling the substrate effect on graphene electron-transfer chemistry via reactivity imprint lithography. Nat. Chem. 2012, 4, 724-732. (5) Stine, R.; Lee, W.-K.; Whitener, K. E.; Robinson, J. T.; Sheehan, P. E. Chemical Stability of Graphene Fluoride Produced by Exposure to XeF2. Nano Lett. 2013, 13, 4311-4316. (6) Elias, D. C.; Nair, R. R.; Mohiuddin, T. M. G.; Morozov, S. V.; Blake, P.; Halsall, M. P.; Ferrari, A. C.; Boukhvalov, D. W.; Katsnelson, M. I.; Geim, A. K.; Novoselov, K. S. Control of Graphene's Properties by Reversible Hydrogenation: Evidence for Graphane. Science 2009, 323, 610613. (7) Felts, J. R.; Oyer, A. J.; Hernández, S. C.; Whitener Jr, K. E.; Robinson, J. T.; Walton, S. G.; Sheehan, P. E. Direct mechanochemical cleavage of functional groups from graphene. Nat. Commun. 2015, 6. (8) Whitener Jr, K. E.; Lee, W. K.; Campbell, P. M.; Robinson, J. T.; Sheehan, P. E. Chemical hydrogenation of single-layer graphene enables completely reversible removal of electrical conductivity. Carbon 2014, 72, 348-353. (9) Whitener, K. E.; Lee, W. K.; Stine, R.; Tamanaha, C. R.; Kidwell, D. A.; Robinson, J. T.; Sheehan, P. E. Activation of radical addition to graphene by chemical hydrogenation. RSC Adv. 2016, 6, 93356-93362. (10) Geim, A. K.; Grigorieva, I. V. Van der Waals heterostructures. Nature 2013, 499, 419425. (11) Friedman, A. L.; van ’t Erve, O. M. J.; Robinson, J. T.; Whitener, K. E.; Jonker, B. T. Hydrogenated Graphene as a Homoepitaxial Tunnel Barrier for Spin and Charge Transport in Graphene. ACS Nano 2015, 9, 6747-6755. (12) Whitener, K. E.; Lee, W. K.; Bassim, N. D.; Stroud, R. M.; Robinson, J. T.; Sheehan, P. E. Transfer of Chemically Modified Graphene with Retention of Functionality for Surface Engineering. Nano Lett. 2016, 16, 1455-1461. ACS Paragon Plus Environment

19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 21

(13) Das, D.; Bhattacharyya, S.; Munoz, E.; Singh, A. K. Strain-induced chiral symmetry breaking leads to large Dirac cone splitting in graphene/graphane heterostructure. Phys. Rev. B 2016, 94. (14) Luo, Z. Q.; Yu, T.; Kim, K. J.; Ni, Z. H.; You, Y. M.; Lim, S.; Shen, Z. X.; Wang, S. Z.; Lin, J. Y. Thickness-Dependent Reversible Hydrogenation of Graphene Layers. ACS Nano 2009, 3, 1781-1788. (15) Robinson, J. T.; Schmucker, S. W.; Diaconescu, C. B.; Long, J. P.; Culbertson, J. C.; Ohta, T.; Friedman, A. L.; Beechem, T. E. Electronic Hybridization of Large-Area Stacked Graphene Films. ACS Nano 2012, 7, 637-644. (16) Zhang, X.; Huang, Y.; Chen, S. S.; Kim, N. Y.; Kim, W.; Schilter, D.; Biswal, M.; Li, B.; Lee, Z.; Ryu, S.; Bielawski, C. W.; Bacsa, W. S.; Ruoff, R. S. Birch-Type Hydrogenation of Few-Layer Graphenes: Products and Mechanistic Implications. J. Am. Chem. Soc. 2016, 138, 14980-14986. (17) Hirsch, A.; Englert, J. M.; Hauke, F. Wet Chemical Functionalization of Graphene. Acc. Chem. Res. 2013, 46, 87-96. (18) Yang, Z. Q.; Sun, Y. Q.; Alemany, L. B.; Narayanan, T. N.; Billups, W. E. Birch Reduction of Graphite. Edge and Interior Functionalization by Hydrogen. J. Am. Chem. Soc. 2012, 134, 18689-18694. (19) Pekker, S.; Salvetat, J. P.; Jakab, E.; Bonard, J. M.; Forro, L. Hydrogenation of carbon nanotubes and graphite in liquid ammonia. J. Phys. Chem. B 2001, 105, 7938-7943. (20) Eckmann, A.; Felten, A.; Mishchenko, A.; Britnell, L.; Krupke, R.; Novoselov, K. S.; Casiraghi, C. Probing the Nature of Defects in Graphene by Raman Spectroscopy. Nano Lett. 2012, 12, 3925-3930. (21) Schafer, R. A.; Englert, J. M.; Wehrfritz, P.; Bauer, W.; Hauke, F.; Seyller, T.; Hirsch, A. On the Way to GraphanePronounced Fluorescence of Polyhydrogenated Graphene. Angew. Chem.-Int. Edit. 2013, 52, 754-757. (22) Wang, B.; Sparks, J. R.; Gutierrez, H. R.; Okino, F.; Hao, Q. Z.; Tang, Y. J.; Crespi, V. H.; Sofo, J. O.; Zhu, J. Photoluminescence from nanocrystalline graphite monofluoride. Appl. Phys. Lett. 2010, 97. (23) Shang, J. Z.; Ma, L.; Li, J. W.; Ai, W.; Yu, T.; Gurzadyan, G. G. The Origin of Fluorescence from Graphene Oxide. Sci. Rep. 2012, 2. (24) Jorio, A.; Dresselhaus, M.; Saito, R.; Dresselhaus, G. F.: Raman Spectroscopy in Graphene Related Systems; Wiley-VCH: Weinheim, Germany, 2011. (25) Martins Ferreira, E. H.; Moutinho, M. V. O.; Stavale, F.; Lucchese, M. M.; Capaz, R. B.; Achete, C. A.; Jorio, A. Evolution of the Raman spectra from single-, few-, and many-layer graphene with increasing disorder. Phys. Rev. B 2010, 82, 125429. (26) Lucchese, M. M.; Stavale, F.; Ferreira, E. H. M.; Vilani, C.; Moutinho, M. V. O.; Capaz, R. B.; Achete, C. A.; Jorio, A. Quantifying ion-induced defects and Raman relaxation length in graphene. Carbon 2010, 48, 1592-1597. (27) Putz, S.; Gmitra, M.; Fabian, J. Optical conductivity of hydrogenated graphene from first principles. Phys. Rev. B 2014, 89. (28) Son, J.; Lee, S.; Kim, S. J.; Park, B. C.; Lee, H. K.; Kim, S.; Kim, J. H.; Hong, B. H.; Hong, J. Hydrogenated monolayer graphene with reversible and tunable wide band gap and its fieldeffect transistor. Nat. Commun. 2016, 7. (29) Zhou, J.; Wang, Q.; Sun, Q.; Chen, X. S.; Kawazoe, Y.; Jena, P. Ferromagnetism in Semihydrogenated Graphene Sheet. Nano Lett. 2009, 9, 3867-3870. (30) Feng, L.; Zhang, W. X. The structure and magnetism of graphone. Aip Adv 2012, 2. (31) Enoki, T.; Suzuki, M.; Endo, M.: Graphite Intercalation Compounds and Applications; Oxford University Press: New York, 2003. (32) Paulus, G. L. C.; Wang, Q. H.; Strano, M. S. Covalent Electron Transfer Chemistry of Graphene with Diazonium Salts. Acc. Chem. Res. 2013, 46, 160-170. (33) Dong, J. C.; Zhang, K. L.; Ding, F. Hydrogenation of bilayer graphene: A small twist makes a big difference. Nano Res. 2015, 8, 3887-3897. ACS Paragon Plus Environment

20

Page 21 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(34) Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S. Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils. Science 2009, 324, 1312-1314. (35) Li, X. S.; Magnuson, C. W.; Venugopal, A.; Tromp, R. M.; Hannon, J. B.; Vogel, E. M.; Colombo, L.; Ruoff, R. S. Large-Area Graphene Single Crystals Grown by Low-Pressure Chemical Vapor Deposition of Methane on Copper. J. Am. Chem. Soc. 2011, 133, 2816-2819. TOC Graphic:

ACS Paragon Plus Environment

21