Optical Microscopy Unveils Rapid, Reversible Electrochemical

Jan 22, 2019 - ... University of California , Berkeley , California 94720 , United States ... Velický, Donnelly, Hendren, McFarland, Scullion, DeBene...
2 downloads 0 Views 2MB Size
Subscriber access provided by EKU Libraries

Communication

Optical microscopy unveils rapid, reversible electrochemical oxidation and reduction of graphene Wan Li, Michal Wojcik, and Ke Xu Nano Lett., Just Accepted Manuscript • Publication Date (Web): 22 Jan 2019 Downloaded from http://pubs.acs.org on January 23, 2019

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 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 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.

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 18 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

Nano Letters

Optical microscopy unveils rapid, reversible electrochemical oxidation and reduction of graphene Wan Li,† Michal Wojcik,† and Ke Xu*,†,‡ †Department of Chemistry, University of California, Berkeley, California 94720, USA ‡Division of Molecular Biophysics and Integrated Bioimaging, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA ABSTRACT. We unveil the reaction dynamics of monolayer graphene in electrochemical oxidation and reduction processes through interference reflection optical microscopy. At 300nanometer spatial resolution and 200-millisecond temporal resolution, we reveal rapid electrochemical oxidation of graphene, as well as its efficient electrochemical reduction back to the unoxidized state. We identify 1.4 V (vs. Ag/AgCl) as the onset voltage for oxidation, and show that the process is driven by free radicals generated in the electrolysis of water and so fully suppressible by a radical-trapping molecule. Moreover, we find the oxidation process to be spatially heterogeneous at the nanoscale, defect- and history-dependent, and characterized by a self-limiting effect unique to the two-dimensional system. We further demonstrate that electrochemical reduction rapidly reverses the oxidized graphene back to the unoxidized state in a controlled manner, and find strong dependency of reduction speed on the reduction voltage and pH, from which we conclude a one-to-one relationship between protons and electrons in the

ACS Paragon Plus Environment

1

Nano Letters 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 18

reduction process. Besides elucidating the electrochemical reaction mechanisms of graphene, our results point to new pathways to the controlled generation and fine-tuning of graphene derivatives through electrochemistry.

KEYWORDS. Graphene, oxidation, reversible reduction, electrochemistry, optical microscopy

TEXT. Chemistry has over the past decade emerged as a key pathway for the production, modification, and application of graphene and related two-dimensional (2D) materials.1-7 Graphene oxide (GO) represents the most studied chemically modified form of graphene:8 it serves as a starting material for the generation of different graphene derivatives,2,3,9 whereas reduced GO provides a route to graphene mass production.1,10 Electrochemistry provides a facile means for the production and property-tuning of graphene and GO products, and moreover, greatly expands the utilization of graphene and GO for sensor, battery, and supercapacitor applications.11-15 However, it remains a challenge to elucidate the electrochemical reaction dynamics of monolayer graphene, given that the system is only a single layer of carbon atoms. The detected electrochemical current is often overwhelmed by reactions of the electrolyte (e.g., Figure 1b below). Indeed, most previous electrochemical studies involving monolayer graphene provide the redox kinetics of the electrolyte as opposed to that of graphene.16-21 Key questions remain unanswered as under what conditions redox reactions occur for graphene, to what degrees such reactions are achieved, whether the reactions are suppressible or reversible, and what governs the reaction mechanisms. A recent study utilizes photoluminescence as a proxy for the oxidative

ACS Paragon Plus Environment

2

Page 3 of 18 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

Nano Letters

state of GO flakes,22 but the results are complicated by the undetermined photoluminescence mechanisms. Our lack of a clear understanding of graphene electrochemistry directly hinders applications. For the emerging electrochemical exfoliation and delamination approaches for graphene preparation,14,23-27 different voltages have been applied to the starting graphite and CVD graphene samples, and different degrees of oxidation are found for the products. What is the onset voltage for graphene oxidization, how fast is the process, and is the undesired oxidation suppressible and/or reversible? For the electrochemical tuning of the properties of graphene and GO,7,12,14,28-31 how to determine the oxidative state of the sample, and is the end product spatially homogenous? For the use of graphene as working electrodes in sensor, battery, and supercapacitor settings,11-15,32-35 what are the performance limits to avoid oxidization of the graphene electrode? To address these pressing questions, we here reveal the electrochemical redox kinetics of CVD-grown monolayer graphene with high spatiotemporal resolution. We recently showed that interference reflection microscopy (IRM), a label-free optical microscopy technique originated in cell biology,36,37 can be repurposed to provide outstanding image contrast for the nanoscale structure and reactions of graphene.38,39 Here, by integrating IRM with electrochemistry, we reveal rapid, reversible oxidation and reduction reactions of graphene. We find the electrochemical oxidation of graphene to be driven by free radicals generated in the electrolysis of water, spatially heterogeneous, defect- and history-dependent, and characterized by a unique self-limiting effect. We show that electrochemical reduction effectively reverses the oxidized graphene back to the unoxidized state, during which process the reduction voltage and time both

ACS Paragon Plus Environment

3

Nano Letters 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 18

offer means to fine-tune the product. Finally, through pH-dependent experiments, we establish a one-to-one relationship of electrons and protons in the reduction process.

a Electrolyte Plastic well

b

Ag/AgCl reference

-

Cytoseal

+

Beam splitter

d

10 s @ 1.6V

I / I0 1.0 0.9

4.6×10−9 mol e

0.8

Au leads

Band-pass

Camera

0s

Vgraphene g

Glass Graphene

c

0

3.7×10−10 mol e

Lamp

0.6

Field diaphragm

e

0.7

0.5

45 s @ 1.6 V

f

4s@0V

g

120 s @ 0 V

Figure 1. IRM reveals rapid, reversible electrochemical oxidation and reduction of monolayer graphene. (a) Schematic of the setup. Vgraphene: voltage applied to graphene referred to the Ag/AgCl electrode. (b) Recorded electrochemical current at a graphene electrode in an oxidation-reduction cycle. The applied Vgraphene voltage sequence is marked at the top of the graph. Orange and cyan shaded areas give the integrated charges for the oxidation and reduction processes, respectively. (c) IRM image of the sample before reaction, showing graphene as predominately monolayer. “0” marks an area with exposed glass surface (no graphene coverage). Red arrows point to small islands of bilayers. (d-e) IRM images after the application of a Vgraphene = 1.6 V oxidation voltage for 10 s (d) and 45 s (e), respectively. (f-g) IRM images of the same sample after next applying a Vgraphene = 0 V reduction voltage for 4 s (f) and 120 s (g), respectively. Scale bars: 5 µm. CVD-grown monolayer graphene40 was deposited as a ~2×10 mm strip onto a glass coverslip, contacted with a pair of gold electrodes at the two ends, and assembled with a plastic well (Figure 1a). The exposed graphene surface to the electrolyte was ~2×4 mm. Electrolyte was a

ACS Paragon Plus Environment

4

Page 5 of 18 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

Nano Letters

potassium phosphate buffer (pH = 3 unless otherwise stated). The gold electrodes enabled conductivity measurement of the graphene sample, and in combination with an Ag/AgCl (3M NaCl) counter/reference electrode, converted graphene into the working electrode in a twoelectrode electrochemical system for redox reactions.29 The system was mounted on an IRM microscope38,39 to enable in situ monitoring of graphene reaction dynamics at diffraction-limited (~300 nm) spatial resolution and video-rate (up to 10 ms) temporal resolution. To match to the typical reaction speeds in this study, a framerate of 5 frames per second (200 ms per frame) was used to record a frame size of 1024×1024 pixels. Figure 1c shows the IRM image of a typical as-prepared sample before reaction. As expected,38 IRM signal was uniform for the monolayer graphene with a well-defined value of I/I0 =0.73, where I and I0 are the detected light intensities at the sample and at a bare glass substrate, respectively. Nanoscale defects, including small islands of bilayers (red arrows), wrinkles, and tears, were clearly resolved, as demonstrated previously.38 Upon the application of an oxidation voltage of 1.6 V (vs. Ag/AgCl), noticeable local changes in IRM signal were observed within a few seconds (Figure 1d and Movie S1): regions surrounding the nanoscale bilayers (red arrows in Figure 1c) brightened up (increased I/I0 value), indicating local oxidation of graphene into GO.39 The oxidation then propagated rapidly in 2D to generate micrometer-sized flower-like GO patterns (Figure 1e and Movie S1). This seedingpropagation behavior is reminiscent of that in the chemical oxidation of graphene,39 but the electrochemical oxidation here is faster by two orders of magnitude. Concurrently recorded anodic current on the graphene electrode was ~10 µA (Figure 1b), so that the integrated charge in 45 s corresponded to 4.6×10−9 mol electrons, one order of magnitude higher than the count of carbon atoms in the graphene electrode (~5×10−10 mol based on area). This result indicates that

ACS Paragon Plus Environment

5

Nano Letters 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 18

the measured current was mostly due to the electrolysis of water and so cannot be used to characterize the reaction dynamics of graphene as uniquely enabled by IRM. Remarkably, as we next applied a 0 V (vs. Ag/AgCl) reductive voltage, the flower-shaped GO patterns visualized by IRM rapidly diminished in seconds (Figure 1f and Movie S1), and mostly disappeared within 2 min (Figure 1g), albeit trace amounts of residual GO were still found surrounding the nanoscale bilayers (compare Figures 1g and 1c). IRM signal after the oxidationreduction cycle showed near-uniform I/I0 values of ~0.73, indicating effective reduction of GO back to graphene. This result again contrasts with the chemical reduction of GO, where only limited reduction is achieved at ~30 min.39 The integrated charge for the reduction period corresponded to 3.7×10−10 mol electrons (Figure 1b), consistent with reduction of the partially oxidized graphene. However, this possibility would otherwise be difficult to consider without corresponding IRM data showing that oxidation has occurred in the first place, and the measured current provides no spatial insights.

ACS Paragon Plus Environment

6

Page 7 of 18 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

Nano Letters

Figure 2. Confirming reversible graphene oxidation-reduction through micro-Raman spectroscopy and electrical transport measurements. (a) IRM image of a sample after electrochemical oxidation at 1.7 V (vs. Ag/AgCl) for 45 s. “0” marks an area with exposed glass surface. (b) Micro-Raman mapping of the D band intensity for the same area. (c) Correlation between the local Raman D band intensity and the IRM signal for the monolayer regions in (a,b). Each data point corresponds to a pixel in the Raman map; corresponding IRM signal is averaged over the matching 1×1 µm2 area. Black line is a linear fit. (d) Raman spectra taken at three different spots on the sample, from the center of a flower-shaped reaction pattern (magenta) to far away (orange). (e) Raman spectra of a sample that has gone through 6 oxidization-reduction cycles, at spots close to (teal) and away from (blue) a flower-shaped reaction pattern, compared to an unreacted sample (black). (f) I-V dependence for a sample before (black) and after (magenta) electrochemical oxidation, and then after subsequent electrochemical reduction (blue). Corresponding resistances are calculated from the I-V slopes and labeled in the graph. We employed micro-Raman spectroscopy and electrical transport characterizations to confirm reversible oxidation and reduction of graphene. Figure 2a shows the IRM image of a graphene sample after electrochemical oxidation. Raman mapping of the same area for the D-band at 1350 cm-1 (Figure 2b) showed features consistent with IRM results, so that stronger D-band intensities were found in regions closer to the nanoscale bilayers, indicating higher local degrees of oxidation. Plotting the Raman D band intensity in each pixel against the corresponding local IRM signal (Figure 2c) showed a good linear relationship; for the limit of zero D peak intensity, this linear relationship extrapolated to an IRM signal of 0.73, the expected value of pristine graphene. These results help validate the IRM signal as a good indicator for the oxidation progress. We also took full spectra for representative spots (Figure 2d). This showed increased D peak intensity, reduced 2D peak intensity, as well as the appearance of D’ and D+G peaks for spots closer to the reaction centers, again consistent with the increased oxidation degrees visualized by IRM. In contrast, even for samples that had gone through multiple electrochemical

ACS Paragon Plus Environment

7

Nano Letters 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 18

oxidation-reduction cycles, only weak D peaks were observed (Figure 2e), in agreement with the IRM-visualized effective reduction of GO back to graphene. X-ray photoelectron spectroscopy (XPS) also yielded consistent results (Figure S1): the spatially heterogeneous oxidized product had 13%, 14%, and 8% of total carbon as hydroxyl, epoxide, and carbonyl/carboxyl groups, respectively. Upon subsequent electrochemical reduction, only trace amounts (2-3% of total carbon) of residual epoxide and carbonyl/carboxyl groups remained. The measured electrical transport properties also showed reversible changes through the oxidization-reduction cycles. As-assembled devices had typical resistances of 5-12 kΩ, consistent with the strip-like sample geometry. After electrochemical oxidization, the resistance increased drastically to >100 kΩ (Figure 2f), consistent with the local conversion of conductive graphene into nonconductive GO. Remarkably, a low resistance was recovered as the sample was next electrochemically reduced back to the graphene state (Figure 2f). Even for samples that have gone through multiple electrochemical oxidization-reduction cycles, the measured electrical properties after the final reduction cycle were still fairly comparable to the initial sample: a ~20% increase in resistance was observed, and electrochemical gating experiments showed comparable resistance-gate voltage slopes and a small shift of the Dirac point (Figure S2), indicating modest changes in carrier mobility and density after the oxidization-reduction cycles. Together, our Raman, XPS, and transport characterizations all supported our IRM observation of reversible oxidization-reduction of graphene, as opposed to alternative effects, e.g., the deposition of carbon contaminations at the surface.41,42

ACS Paragon Plus Environment

8

Page 9 of 18

a

0 1.4

0V

0 1.5

0V

0 1.6

0V

0 1.7

0V

0 1.7

0V

0 1.7

0V

Ig (µA)

Vg 30 15 0 -15 0.90 0.85

i

ii

iii

D

0.80

i

40

0 0

50 100 150

Cycle 6: 0 s A CB+ D+++

0

50 100 150

45 s @ 1.7 V

ii

A CB+ D+++

0

50 100 150

d

120 s @ 0 V

iii

0

time(s)

A CB+ D+++

0.90

50 100 150

Spot A

0.85

Cy6 Cy5 Cy4

0

50 100 150

Spot B

Cy6 Cy5 Cy4

0

50 100 150

Spot C

Cy6 Cy5 Cy4

0.80

Oxidation degree (%) 0

60

20

0.75 0.70

80

40

I/I0

c

60

20

0.75 0.70

Oxidation degree (%)

80

A B C

0 0

50 100 150

20 40 60 80 100

0

50 100 150

time (s)

0

Oxidation degree (%)

b I/I0

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

Nano Letters

50 100 150

Figure 3. In situ recording of the electrochemical oxidation and reduction kinetics of graphene over multiple cycles. (a) Electrochemical current at a graphene electrode in 6 consecutive oxidization-reduction cycles of varying oxidative voltages. Top of graph: The applied voltage sequences. (b) Corresponding time-dependent IRM signal (I/I0; left axis) and converted oxidation degree (right axis) for four spots from the center to the margin of a flower-shaped reaction pattern [marked as A-D in (c)]. (c) Maps of IRM-derived local oxidation degree for 3 time points in the final (6th) oxidization-reduction cycle, corresponding to before oxidation (i), at the end of 45 s of oxidation at 1.7 V (ii), and at the end of 120 s of reduction at 0 V (iii) [dotted lines in (b)]. Scale bars: 5 µm. (d) Overlay of the IRM signals in the 4th to 6th oxidization-reduction cycles for Spots A-C. The reversible electrochemical oxidization-reduction behavior we achieved allowed us to repeatedly oxidize and reduce the same graphene piece in multiple cycles, during which process IRM uniquely enabled the in situ monitoring of reaction progress with high spatiotemporal resolution. Figure 3 present the measured graphene-electrode electrochemical current (Figure 3a) for a sample in six consecutive oxidization-reduction cycles, together with the concurrently recorded IRM signal (I/I0) (Figure 3b) for 4 representative spots from the central to peripheral

ACS Paragon Plus Environment

9

Nano Letters 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 18

regions of a flower-shaped reaction pattern (spots A-D in Figure 3c). Varying oxidation voltages of 1.4, 1.5, 1.6, 1.7, 1.7, and 1.7 V (vs. Ag/AgCl) were applied in each cycle for the same duration of 45 s, whereas for reduction, the same reduction voltage of 0 V vs. Ag/AgCl was applied for 120 s (top of Figure 3a). Following our previous work,39 we converted the IRM signal linearly to the local oxidation degree of graphene, where I/I0 values of 0.73 and 0.97 corresponded to non-oxidized (0% oxidation degree) and fully oxidized (100% oxidation degree) graphene, respectively (right y-axis of Figure 3b). Figure 3c presents maps of the IRM-derived local oxidation degree for 3 specific time points of the final oxidization-reduction cycle. See Figure S3 and Movie S2 for the reaction maps of the other cycles. For the 1.4 V oxidation voltage, a low level of electrochemical current (~0.5 µA) was detected (Figure 3a), and IRM indicated that only very limited reaction occurred in regions close to the centers of the flower-shaped reaction patterns (Figure 3b). We further showed (Figure S4) that no detectable oxidation occurred at 1.3 V, in which case the measured electrochemical current was also minimal (10 µA currents were mostly due to the electrolysis of water. The strong correlation between the

ACS Paragon Plus Environment

10

Page 11 of 18

electrochemical current and oxidation rate points to the likelihood that the oxidation of graphene was due to reactive oxygen species (ROS) generated during the electrolysis of water.

0.90

0V

0 1.7

a

0V

0 1.7

b

0V

0 1.7

c

+++ +

0V

0 1.7

d

+++ +

0V 80

e

+++ +

+++ +

+++ +

60

0.85 45 s @ 1.7 V

0.80

45 s @ 1.7 V

45 s @ 1.7 V

45 s @ 1.7 V

45 s @ 1.7 V

40 20

0.75 0.70

Oxidation degree (%)

Vg 0 1.7

I/I0

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

Nano Letters

0 0

50

100

150

0

50

100

150

0

50

100

time (s)

150

0

50

100

150

0

50

100

150

Figure 4. The electrochemical oxidation of graphene is effectively suppressed by the freeradical trap TEMPO. (a-e) Time-dependent IRM signal (I/I0; left axis) and converted oxidation degree (right axis) in 5 consecutive oxidization-reduction cycles of identical voltage sequences (top of graphs), for four spots from the center to the margin of a flower-shaped reaction pattern. Electrolyte in the 1st, 3rd, and 5th cycles (a,c,e) was a potassium phosphate buffer (pH = 3), whereas 10 mg/mL of TEMPO was added into the same buffer for the 2nd and 4th oxidization-reduction cycles (b, d). Insets: Maps (25×25 µm2 in size) of IRM-derived local oxidation degree at the end of each oxidation session (45 s at 1.7 V). Previous work has identified hydroxyl radicals (HO·) as the major active ROS during water electrolysis.43,44 To examine whether radical attack of the graphene basal plane underlies the electrochemical oxidation of graphene, we added into the electrolyte 2,2,6,6-tetramethyl-1piperidinyloxy (TEMPO), a stable radical that traps reactive radicals including hydroxyl radicals45 and has been shown to lessen graphene oxidation during the electrochemical exfoliation of graphite.27 Remarkably, we found that the addition of 10 mg/mL TEMPO consistently and near completely suppressed the electrochemical oxidation of graphene (Figure 4bd) for samples that otherwise undergo reversible oxidation-reduction reactions in identical electrochemical redox cycles (Figure 4ace). These results indicate that free radicals generated in the electrolysis of water indeed drive graphene oxidation, and also establish TEMPO as an ACS Paragon Plus Environment

11

Nano Letters 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 18

effective suppressor for the process. We also examined the possibility of suppressing the electrochemical oxidation with Co2+;46 notable suppression was found, although under our conditions, the reversible formation and dissolution of cobalt oxide deposits complicated the results (Figure S5). Further examination of the oxidation-reduction cycles indicated that for all regions, the initial oxidation reaction rate, as indicated by the slopes of the reaction progress curves at the onsets of the oxidation cycles, became progressively higher in consecutive cycles, even when the same oxidation voltage was applied (Figure 3d). This sequence-dependent result suggests that although graphene was efficiently reduced (as evidenced above by IRM, Raman spectroscopy, and conductivity measurements), residual defects left from earlier oxidization-reduction cycles may seed and accelerate the oxidation reaction for later cycles. The oxidation dynamics is thus determined by both the oxidative voltage and the local density of defects. Interestingly, we further found that independent of position in the sample or the reaction sequence, the detected oxidation progress in all processes “saturated” at an IRM signal of ~0.880.90, corresponding to an oxidation degree of ~65-70% (Figures 3, 4, and S3). Even for regions close to reaction centers, fast oxidation terminated abruptly within a few seconds to plateau at this value (magenta curves in Figure 3b). Such plateauing of oxidation progress is not observed in chemical oxidation,39 and may be rationalized by a novel self-limiting process unique to the 2D system: electrochemical reactions are driven by local currents at the electrode surface, but the oxidized product of graphene, GO, has significantly lower conductivity (see also Figure 2e). Local conductor-insulator transitions due to the oxidation reaction per se may thus self-limit local electrochemical oxidation progresses to a particular value. Consistent with this interpretation, substantial drops in electrochemical current was observed over time as growing

ACS Paragon Plus Environment

12

Page 13 of 18 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

Nano Letters

areas of the graphene surface became oxidized and thus less conductive for electrochemical currents (last three panels of Figure 3a). Together, we have shown that the electrochemical oxidation of graphene is driven by free radicals generated during the electrolysis of water, and characterized by rich spatiotemporal dynamics controlled by the applied voltage, intrinsic and reaction-induced local defects, as well as reaction-induced changes in electrical conductivity. We next examined the reduction kinetics. IRM indicated rapid reduction (abrupt jump in I/I0) as the applied voltage was switched from oxidative to reductive (0 V vs. Ag/AgCl), so that substantial reduction was apparent for all cycles within ~1 s (Figure 3bd). The reaction slowed down as the sample gradually approached the limit of fully reduced graphene, with spots closer to the oxidation center taking a longer time to reach the baseline. At the end of each reduction cycle of 120 s, the sample largely recovered its initial IRM signal (I/I0 ~ 0.73) (Figure 3b-d), indicating effective reduction back to graphene. We further showed that electrochemical reduction can still be effectively achieved after waiting for an extended period of time as opposed to immediately done after oxidation (Figure S6). Closer examination of the reduction kinetics showed independence of sample history. Overlay of the IRM signals for three spots of the sample in three consecutive oxidization-reduction cycles (Figure 3d) showed coinciding reduction kinetics, in contrast to the accelerating oxidation kinetics over time. This process-independent behavior allowed us to directly probe the effects of applied voltage on reduction kinetics by repeatedly oxidizing a sample to the above-discussed limit of ~65% oxidation degree, and then reducing the same sample with varying reduction voltages (Figure 5a). Whereas the application of 0 V and −0.1 V reduction voltages (vs. Ag/AgCl) led to rapid, effective reduction (oxidation degree ~0%) within a 120 s timeframe, significantly decreased reduction speeds were found for the more positive (less reductive)

ACS Paragon Plus Environment

13

Nano Letters

voltages (Figure 5a). Plotting the oxidation degree at the end point (120 s) of each reduction cycle showed a sigmoidal dependence on the reduction voltage (Figure 5b), thus demonstrating that this parameter may be harnessed, together with the reaction time, to control reaction progress and the final state of the product.

60 0.5 V

40

0.4 V

0.80

0.3 V

20

0.2 V 0.1 V

0.75

0.0 V

0

−0.1 V

0

50 time (s)

100

120 s @ Vred

pH = 7

0.80

I/I0

Vred

0.85

b

40

pH = 3

Oxidation degree (%)

0.84

a

Oxidation degree (%)

0.90

I/I0

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 18

30 20

250 mV

0.76

10 0

0.72 -0.4

-0.2

0.0

0.2

Vred (V)

0.4

0.6

Figure 5. Dependence of reduction kinetics on voltage and pH. (a) IRM signal (left axis) and converted oxidation degree (right axis) for a spot in a sample that was repeated oxidized to the same oxidation limit, and then reduced with varying voltages Vred in the range of −0.1 to 0.5 V in 7 different electrochemical redox cycles. (b) The end-point IRM signal (left axis) and oxidation degree (right axis) as a function of reduction voltage for the same sample [colored data points matching the colored curves in (a); pH = 3], vs. results from another sample at pH = 7 (black data points). Each data point is averaged from 8 different spots at the central areas of the flower-shaped reaction patterns, with the standard deviations drawn as error bars. Black curves are guides to the eyes. We next found that the reduction kinetics depended strongly on pH, a result that offers further insights into the reduction mechanism. As we switched from the pH = 3 buffer used in the above experiments to a pH = 7 buffer, the reduction speed decreased notably for the same reduction voltages. Effective reduction (oxidation degree ~0%) was only achieved at −0.3 V (vs. Ag/AgCl), whereas more positive reduction voltages led to considerable residual GO after the 120 s reduction period (Figure 5b). The overall trend of voltage dependence resembles that

ACS Paragon Plus Environment

14

Page 15 of 18 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

Nano Letters

observed at pH =3, but the curve was uniformly shifted to the negative direction by 250 mV. Dividing this shift by the difference in pH (4 units) gave ~62 mV/pH, a value comparable to the 59.2 mV constant in the Nernst equation. The electrochemical reduction of GO has been proposed through the mechanism:8,28 GO + a H+ +b e− → rGO + c H2O Our ~62 mV/pH result above indicates a:b ~1, i.e., a 1:1 ratio between protons and electrons during the electrochemical reduction of GO. In summary, through IRM we have resolved the reaction dynamics of graphene in electrochemical oxidation and reduction processes with high spatiotemporal resolution. This allowed us to reveal fast, reversible redox processes, as well as rich spatiotemporal heterogeneity. We identified 1.4 V (vs. Ag/AgCl) as the onset voltage for the electrochemical oxidation of graphene, and elucidated that the process is driven by free radicals generated in the electrolysis of water, meanwhile establishing TEMPO as an effective suppressor for the oxidation. We also found the oxidation process to be spatially heterogeneous, defect- and history-dependent, and characterized by a self-limiting effect unique to the two-dimensional system. We further showed that electrochemical reduction could effectively reverse the oxidized graphene back to the unoxidized state, and revealed a strong dependence of reduction speed on the reduction voltage and pH, from which we concluded a one-to-one relationship between protons and electrons in the process. Besides shedding new light on the electrochemical reaction mechanisms of graphene, our demonstrated ability to control reactions through electrochemistry points to new pathways to the controlled generation and fine-tuning of graphene derivatives.

ASSOCIATED CONTENT

ACS Paragon Plus Environment

15

Nano Letters 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 18

Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Materials and methods; Figures S1−S6 (PDF) Video S1 (AVI) Video S2 (AVI)

AUTHOR INFORMATION Corresponding Author * [email protected] Notes The authors declare no competing financial interest. ACKNOWLEDGMENT We thank Dr. Michael Ross (P. Yang Group) for help with Raman experiments. K.X. acknowledges support from the Sloan Research Fellowship and the Bakar Fellows Award. M.W. acknowledges support from the NSF Graduate Research Fellowship under DGE 1106400. Work at the Molecular Foundry was supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

REFERENCES (1) Park, S.; Ruoff, R. S. Nat. Nanotechnol. 2009, 4, 217-224. (2) Georgakilas, V.; Otyepka, M.; Bourlinos, A. B.; Chandra, V.; Kim, N.; Kemp, K. C.; Hobza, P.; Zboril, R.; Kim, K. S. Chem. Rev. 2012, 112, 6156-6214.

ACS Paragon Plus Environment

16

Page 17 of 18 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

Nano Letters

(3) Kuila, T.; Bose, S.; Mishra, A. K.; Khanra, P.; Kim, N. H.; Lee, J. H. Prog. Mater. Sci. 2012, 57, 1061-1105. (4) Yan, L.; Zheng, Y. B.; Zhao, F.; Li, S. J.; Gao, X. F.; Xu, B. Q.; Weiss, P. S.; Zhao, Y. L. Chem. Soc. Rev. 2012, 41, 97-114. (5) Chua, C. K.; Pumera, M. Chem. Soc. Rev. 2013, 42, 3222-3233. (6) Bottari, G.; Herranz, M. A.; Wibmer, L.; Volland, M.; Rodriguez-Perez, L.; Guldi, D. M.; Hirsch, A.; Martin, N.; D'Souza, F.; Torres, T. Chem. Soc. Rev. 2017, 46, 4464-4500. (7) Kaplan, A.; Yuan, Z.; Benck, J. D.; Rajan, A. G.; Chu, X. S.; Wang, Q. H.; Strano, M. S. Chem. Soc. Rev. 2017, 46, 4530-4571. (8) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. Chem. Soc. Rev. 2010, 39, 228240. (9) Dreyer, D. R.; Todd, A. D.; Bielawski, C. W. Chem. Soc. Rev. 2014, 43, 5288-5301. (10) Pei, S. F.; Cheng, H. M. Carbon 2012, 50, 3210-3228. (11) Chen, D.; Tang, L. H.; Li, J. H. Chem. Soc. Rev. 2010, 39, 3157-3180. (12) Pumera, M. Chem. Soc. Rev. 2010, 39, 4146-4157. (13) Ambrosi, A.; Chua, C. K.; Bonanni, A.; Pumera, M. Chem. Rev. 2014, 114, 7150-7188. (14) Ambrosi, A.; Chua, C. K.; Latiff, N. M.; Loo, A. H.; Wong, C. H. A.; Eng, A. Y. S.; Bonanni, A.; Pumera, M. Chem. Soc. Rev. 2016, 45, 2458-2493. (15) Fu, W. Y.; Jiang, L.; van Geest, E. P.; Lima, L. M. C.; Schneider, G. F. Adv. Mater. 2017, 29, 1603610. (16) Li, W.; Tan, C.; Lowe, M. A.; Abruna, H. D.; Ralph, D. C. ACS Nano 2011, 5, 22642270. (17) Valota, A. T.; Kinloch, I. A.; Novoselov, K. S.; Casiraghi, C.; Eckmann, A.; Hill, E. W.; Dryfe, R. A. W. ACS Nano 2011, 5, 8809-8815. (18) Tan, C.; Rodriguez-Lopez, J.; Parks, J. J.; Ritzert, N. L.; Ralph, D. C.; Abruna, H. D. ACS Nano 2012, 6, 3070-3079. (19) Guell, A. G.; Ebejer, N.; Snowden, M. E.; Macpherson, J. V.; Unwin, P. R. J. Am. Chem. Soc. 2012, 134, 7258-7261. (20) Yuan, W. J.; Zhou, Y.; Li, Y. R.; Li, C.; Peng, H. L.; Zhang, J.; Liu, Z. F.; Dai, L. M.; Shi, G. Q. Sci. Rep. 2013, 3, 2248. (21) Guell, A. G.; Cuharuc, A. S.; Kim, Y. R.; Zhang, G. H.; Tan, S. Y.; Ebejer, N.; Unwin, P. R. ACS Nano 2015, 9, 3558-3571. (22) Chen, T.; Zhang, Y. W.; Xu, W. L. ACS Nano 2016, 10, 8434-8442. (23) Su, C. Y.; Lu, A. Y.; Xu, Y. P.; Chen, F. R.; Khlobystov, A. N.; Li, L. J. ACS Nano 2011, 5, 2332-2339. (24) Wang, Y.; Zheng, Y.; Xu, X. F.; Dubuisson, E.; Bao, Q. L.; Lu, J.; Loh, K. P. ACS Nano 2011, 5, 9927-9933. (25) Low, C. T. J.; Walsh, F. C.; Chakrabarti, M. H.; Hashim, M. A.; Hussain, M. A. Carbon 2013, 54, 1-21. (26) Abdelkader, A. M.; Cooper, A. J.; Dryfe, R. A. W.; Kinloch, I. A. Nanoscale 2015, 7, 6944-6956. (27) Yang, S.; Bruller, S.; Wu, Z. S.; Liu, Z. Y.; Parvez, K.; Dong, R. H.; Richard, F.; Samori, P.; Feng, X. L.; Mullen, K. J. Am. Chem. Soc. 2015, 137, 13927-13932. (28) Zhou, M.; Wang, Y. L.; Zhai, Y. M.; Zhai, J. F.; Ren, W.; Wang, F. A.; Dong, S. J. Chem.-Eur. J. 2009, 15, 6116-6120.

ACS Paragon Plus Environment

17

Nano Letters 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 18

(29) Ramesh, P.; Itkis, M. E.; Bekyarova, E.; Wang, F. H.; Niyogi, S.; Chi, X. L.; Berger, C.; de Heer, W.; Haddon, R. C. J. Am. Chem. Soc. 2010, 132, 14429-14436. (30) Liu, F.; Xue, D. F. Chem.-Eur. J. 2013, 19, 10716-10722. (31) Toh, S. Y.; Loh, K. S.; Kamarudin, S. K.; Daud, W. R. W. Chem. Eng. J. 2014, 251, 422434. (32) Liu, C. G.; Yu, Z. N.; Neff, D.; Zhamu, A.; Jang, B. Z. Nano Lett. 2010, 10, 4863-4868. (33) Shao, Y. Y.; Wang, J.; Wu, H.; Liu, J.; Aksay, I. A.; Lin, Y. H. Electroanalysis 2010, 22, 1027-1036. (34) Li, N.; Chen, Z. P.; Ren, W. C.; Li, F.; Cheng, H. M. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 17360-17365. (35) Kim, H.; Park, K. Y.; Hong, J.; Kang, K. Sci. Rep. 2014, 4, 5278. (36) Curtis, A. S. G. J. Cell Biol. 1964, 20, 199-215. (37) Limozin, L.; Sengupta, K. ChemPhysChem 2009, 10, 2752-2768. (38) Li, W.; Moon, S.; Wojcik, M.; Xu, K. Nano Lett. 2016, 16, 5027-5031. (39) Wojcik, M.; Li, Y. Q.; Li, W.; Xu, K. J. Am. Chem. Soc. 2017, 139, 5836-5841. (40) Li, X. S.; Cai, W. W.; An, J. H.; Kim, S.; Nah, J.; Yang, D. X.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S. Science 2009, 324, 1312-1314. (41) Zou, Y. Q.; Walton, A. S.; Kinloch, I. A.; Dryfe, R. A. W. Langmuir 2016, 32, 1144811455. (42) Hurst, J. M.; Li, L.; Liu, H. T. Carbon 2018, 134, 464-469. (43) Marselli, B.; Garcia-Gomez, J.; Michaud, P. A.; Rodrigo, M. A.; Comninellis, C. J. Electrochem. Soc. 2003, 150, D79-D83. (44) Jeong, J.; Kim, J. Y.; Yoon, J. Environ. Sci. Technol. 2006, 40, 6117-6122. (45) Samuni, A.; Goldstein, S.; Russo, A.; Mitchell, J. B.; Krishna, M. C.; Neta, P. J. Am. Chem. Soc. 2002, 124, 8719-8724. (46) Ejigu, A.; Fujisawa, K.; Spencer, B. F.; Wang, B.; Terrones, M.; Kinloch, I. A.; Dryfe, R. A. W. Adv. Funct. Mater. 2018, 28, 1804357.

TOC Graphic Oxidation degree (%) 0 20 40 60 80 100

5 µm

Electrochemical oxidation

Electrochemical reduction

5 µm

ACS Paragon Plus Environment

5 µm

18