Molecular Investigation of Oxidized Graphene: Anatomy of Double

Subscriber access provided by UNIV OF LOUISIANA

C: Energy Conversion and Storage; Energy and Charge Transport

Molecular Investigation of Oxidized Graphene: Anatomy of Double Layer Structure and Ion Dynamics Yu Zhang, Boris Dyatkin, and Peter T. Cummings J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.9b01617 • Publication Date (Web): 29 Apr 2019 Downloaded from http://pubs.acs.org on April 29, 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 28 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

The Journal of Physical Chemistry

Molecular Investigation of Oxidized Graphene: Anatomy of Double Layer Structure and Ion Dynamics Yu Zhang,† Boris Dyatkin,‡,§, and Peter T. Cummings†* †

Department of Chemical and Biomolecular Engineering, Vanderbilt University, Nashville,

Tennessee 37225, United States ‡

A.J. Drexel Nanomaterials Institute and Department of Materials Science and Engineering,

Drexel University, Philadelphia, Pennsylvania 19104, United States *

Corresponding author. E-mail: [email protected]

ABSTRACT We investigated the influence of surface oxidization of planar graphene electrodes on charge storage and ion dynamics of supercapacitors. Our approach compared two distinct ionic liquid (IL) electrolytes: tetraethylammonium tetrafluoroborate solvated in acetonitrile (TEA-BF4/ACN) and 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide solvated in acetonitrile (EMImTFSI/ACN). Both experimental electrochemical tests and molecular dynamics (MD) simulations showed positive, electrolyte-specific influences of hydroxyl-free electrode interfaces on capacitance. In the EMIm-TFSI/ACN system, the hydroxylated surface, due to its strong interaction with anions, impeded surface charge storage. On the other hand, in the case of TEABF4/ACN, the distribution and orientation of ACN across the system exerted vital influence on the capacitance, especially on the positive hydroxyl-free electrode. Furthermore, MD simulations of ion mobility with respect to electrode surface in the lateral and perpendicular directions revealed significantly slower diffusion performance on oxidized surface. Our efforts enhanced the level of

§ Current address: U.S. Naval Research Laboratory, Washington, DC 20375, USA ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 28

fundamental understanding of the effects of hydroxyl groups on electrode-electrolyte interfaces and resulting supercapacitor performance.

ACS Paragon Plus Environment

2

Page 3 of 28 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


1. INTRODUCTION Electric double layer capacitors (EDLCs) are energy storage devices, also known as supercapacitors, which offer high energy and power densities, broad operating temperature windows, and long cycle lifetimes.1 However, compared to batteries and fuel cells, EDLCs exhibit limited energy densities.2,3 EDLCs’ energy storage mechanism relies on electrosorption of electrolyte ions onto an electrode surface and their subsequent formation of an electrical double layer (EDL). In turn, the EDL determines the capacitance and charge/discharge dynamics of EDLCs.4–18 Although early supercapacitor designs have relied on aqueous or organic electrolytes, their operating potentials and energy densities were very limited. Novel ionic liquids (ILs), which have wider electrochemical windows and showcase low vapor pressure and high thermal stability, offer a promising pathway to higher operating voltages and energy densities.1,19 However, the slow ion dynamics and low electrical conductivity of ILs in their neat, solvent-free states may depress power densities.20,21 In practice, ILs are often solvated in organic solvents to improve their rate handling performance. 22–25 In order to extract maximum benefits from novel high-performance electrolytes, researchers, in complementary fashion, tailor and optimize electrode materials that offer high specific surface areas (SSAs) and good electronic conductivity.26 Graphene-based materials are exceptional candidates for electrodes, owing to over 2,600 m2 g-1 accessible SSA, zero-bandgap electron transport band structure, and high mechanical stiffness.27,28 The capacitive performance and charge storage mechanism of pristine graphene sheets have been extensively studied by both experiments and computational methods.5,12,15,29–32 However, commercially available, real-life graphene materials may contain edge sites and surface functional groups. These non-idealized structures differ from pristine modeled counterparts at the electrolyte-electrode interface, and, subsequently,


3


Page 4 of 28

diverge in capacitive performance.26 Certain surface species, such as nitrogen dopants, alter the band structure, increase quantum capacitance and, subsequently, show higher overall charge storage densities.33,34 Furthermore, MD simulations predict that edges of graphene sheets offer higher capacitance than flat basal planes.35,36 Several recent computational efforts have also assessed effects of oxygen groups on the capacitive performance, and discovered functionalization-induced steric hindrance and decreased molecular orientation abilities.37–39 Despite these recent advances, effects of surface functional groups on ion dynamics in graphenebased supercapacitors remain, to date, only partially understood. Hence, comprehensive knowledge of the effects of surface non-ideality on both capacitance and ion dynamics will substantially aid efforts to design novel supercapacitors that offer higher energy and power densities. In this work, we combined MD simulations and electrochemical tests to provide a comprehensive analysis of the effects of surface hydroxylation on the performance of graphene-based EDLCs. Two sets of organic IL salts solvated in acetonitrile (ACN), 1.5 M tetraethylammonium tetrafluoroborate (TEA-BF4/ACN) solution and a 1 wt. %: 1 wt. % solution of 1-ethyl-3methylimidazolium bis(trifluorosulfonyl)imide (EMIm-TFSI/ACN), were simulated on pristine graphene (PG) and oxidized graphene (OG) respectively. Our complementary approach relies on MD simulations and experimental results to dissect the EDL structure, and explain the influence of graphene surface groups on capacitive behavior and ion dynamics.

2. METHODS 2.1 Experimental Measurements


4

Page 5 of 28 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


Commercially available graphene nanoplatelets (XG Sciences, baked, “GNP-B”) were used as experimental electrode materials and processed according to previously determined procedures.37 Their structure was comprised of multilayer (< 100 layers) stacked graphene flakes with a ~2.5 μm cross-sectional planar area.40 The GNP-B powders were annealed at 1800 °C under a high vacuum (10-6 torr) in a graphite furnace to obtain the pristine graphene (PG) experimental material. A subset of PG powders were treated with flowing air in a quartz tube furnace at 485 °C to obtain oxidized experimental materials (OG). Prior results determined the specific surface area (SSA) of PG and OG to be 383.2 m2 g-1 and 430.5 m2 g-1, respectively. The electrode powders were bound together with 5 wt. % PTFE polymer, rolled into 100 μm thick freestanding films, and excised into 12 mm diameter discs. Electrochemical measurements used symmetrical cells arranged in a sandwich configuration with a three-electrode configuration.41 A chlorinated silver wire (Ag/AgCl) was used as a quasireference electrode42. All cells were carried out in an Ar-filled glovebox with a VMP3 Potentiostat. The pouch cells were filled with either 1.5 M TEA-BF4 solvated in acetonitrile or a 1:1 w/w ratio of EMIm-TFSI in acetonitrile. Cyclic voltammogram (CV) sweeps changed the inter-electrode potential at 2 mV s-1 – 1000 mV s-1 rates in the 0.0 ↔ +2.5 V range and measured current to derive capacitance, and the reference electrode monitored the individual potential changes of each electrode. Electrochemical impedance spectroscopy (EIS) fluctuated voltage with a 10 mV amplitude centered at 0.0 V (vs. open circuit potential) with a dampening frequency in the 200 kHz – 10 mHz range. 2.2 Molecular Dynamics Simulations As illustrated in Figure 1, each simulation consisted of an electrolyte in a channel enclosed by two electrodes. The walls were set 5 nm apart to achieve a bulk-like behavior in the channel center.


5


Page 6 of 28

The force fields for TEA-BF4 and ACN were adopted from literature;16 Lopes’ previous work provided the force fields for EMIm-TFSI.43,44 Simulations of oxidized graphene differed from those of pristine graphene by incorporating hydroxyl groups on both sides of the former and obtaining 2.8 % surface coverage. The configuration of hydroxyl group on the surface was taken from the DFT work of Yan et al,45 and the non-bonded parameters were defined by the OPLS-AA force field.46 Both random distribution and patterned distribution of hydroxyl groups were tested in our simulations (configurations shown in Figure S1 in the Supplementary Information (SI)). The Lorentz-Berthelot combining rule was used to calculate Lennard-Jones parameters for unlike atoms. Net surface charges were uniformly distributed to surface atoms on each electrode, and opposite surface charges on cathode and anode guaranteed charge neutrality of the system. The simulations were conducted in canonical NVT ensemble using MD package GROMACS 5.1.2.47 During each simulation, the positions of all carbon atoms in the electrodes were fixed, while the hydroxyl groups were allowed to rotate and bend. All bonds in the simulation box were constrained using the LINCS algorithm.48 The slab-PME method was used to compute the electrostatic interaction.49 For each run, the system was initially heated to 800 K for 5 ns, followed by 5 ns annealing to 300K. The system was further equilibrated at 300 K for 10 ns, before the final 20 ns of production run. The temperature was controlled by the Nose-Hoover thermostat with relaxation time of 0.4 ps. Each simulation was repeated 3 times with different initial configurations to ensure that the results were statistically representative. The potential distribution across the simulation channel can be calculated by integrating the 1-D Poisson equation50: ∅(𝑧) = −

1 0 𝜎 * (𝑧 − 𝑧 + )𝜌- (𝑧 + )d𝑧 + − 𝑧 𝜀) ) 𝜀)


(1)

6

Page 7 of 28 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


where 𝑧 is the perpendicular distance from the electrode surface, 𝜀) is the vacuum permittivity, 𝜌is the total space charge density across the channel, and 𝜎 is the surface charge density. The differential capacitance was then calculated by fitting the surface charge density as a function of electrode potential with a fourth-order polynomial function, followed by differentiating the fitted charge density with respect to electrode potential. To study the structure of electrical double layers, the center of mass (COM) number densities of ions and solvent were calculated in the direction perpendicular to surface. Since ions and solvent form layered structure near the electrode surface, in an effort to more clearly understand the ion and solvent dynamics, we divided the simulation channel into different regions. Three regions were defined based on the number densities: interfacial, transitional, and bulk-like. Figure S2 in SI shows the scheme of dividing simulation box of TEA-BF4/ACN on charged PG surfaces. The average location representing each region is calculated by 07

𝑧123 =

∫0) 𝑧𝜌5 (z)d𝑧

(2)

07

∫0) 𝜌5 (z)d𝑧

where 𝜌5 is the center-of-mass number density. The diffusion coefficients in each region were computed using the Einstein relation, and only the lateral mean square displacement (MSD) was considered. These 2-D self-diffusion coefficients can be obtained using the following formula: F

〈A𝒓(𝑡) − 𝒓(0)E 〉 𝐷 = lim=→? 4𝒕

(3)

where 𝒓(𝑡) represents the lateral position at 𝑡. Note that since molecules may enter and leave each region (interfacial, transitional and bulk-like regions; see figure S2 in SI), we only selected molecules that stayed in each region during the production run.

3. RESULTS AND DISCUSSION


7


Page 8 of 28

3.1 Experimental Electrochemistry Results Figure 2 shows the comparable experimental electrochemistry results for PG and OG electrodes, as normalized by the accessible SSA of each active material, in solvated organic salt electrolytes. The low-rate sweep of EMIm-TFSI solvated in ACN (Figure 2(a)) shows greater charge storage capacity of the defunctionalized PG electrode compared to the hydroxyl-rich OG electrode. This behavior holds true over the entire potential sweep range; Figure 2(b) shows approximately 3 μF cm-2 (roughly 30 %) higher charge storage density values for PG vs. OG. As described by the EISderived Nyquist plot (Figure 2(c)), the dynamics of EMIm-TFSI in ACN were similar for both PG and OG, with only a slight difference in the ionic impedance near the “knee” frequency region. In that regime, EMIm-TFSI showcased less impedance during electrosorption onto PG electrodes than on OG electrodes. However, the differences in dynamics were negligible in the low-frequency regime (where capacitive contributions to impedance are more dominant). This suggests that the equilibrium amount of ions that electrosorb on the surface, rather than intrinsically different dynamics, primarily account for improved performance of the defunctionalized graphene over its hydroxylated counterpart. As shown in Figure 2(d), the TEA-BF4/ACN electrolyte exhibited similar behavior and showcased higher capacitance for PG electrodes compared to OG electrodes. The difference, as measured across the entire CV sweep range (Figure 2(e)), shows a more significant increase in charge storage density for defunctionalized electrodes compared to the aforementioned EMIm-TFSI/ACN system. Furthermore, both PG and OG positive electrodes of the TEA-BF4/ACN electrolyte showed better performance than the negative electrodes of each respective system. This property was also more apparent than in the EMIm-TFSI/ACN-based electrolyte system. Electrolyte dynamics (Figure 2(f)) showed a slightly higher charge transfer resistance (the semi-circular high-frequency region) for


8

Page 9 of 28 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


OG and suggested that increased resistance in the system stemmed from hydroxylated surface groups. While the “knee” frequency regions for both electrodes were similar, the oxidized electrodes exhibited slightly higher impedance in the low-frequency, purely capacitive region. 3.2 MD-Derived Capacitance The MD-derived computational comparison of capacitance between PG and OG electrodes is shown in Figure 3. Note that all electrode potentials from MD simulations are relative to the potential of zero charge (PZC). In agreement with experimental electrochemistry results, OG always yields lower differential capacitance than PG for both electrolytes. We also observed higher capacitance in the positive potential range than in the negative potential range. In both electrolytes, the size of anion is smaller than cation, which results in a thinner electrical double layer at the anode compared to that at the cathode, and yields higher capacitance at the anode.51,52 In addition, the differential capacitance discrepancy between PG and OG is more pronounced at positive potential than at negative potentials. This is particularly apparent for the TEA-BF4/ACN system: the differential capacitance is quite similar at potentials below -0.5 V. This indicates that the anode is more significantly affected by the hydroxyl groups than the cathode. We also examined the effects of hydroxyl patterns on the capacitance on OG. As shown in Figure S3 in SI, regularly and randomly patterned OG have very similar differential capacitance, but regularly distributed OG surfaces show slightly lower capacitance than the randomly distributed OG surfaces in both electrolytes. This is likely caused by the lower ion accessibility to the patterned OG surface.39 The results for OG presented in this paper were the averaged results from both regularly patterned and randomly distributed OG. Comparisons of the experimental results (in Figure 2) with computational findings (in Figure 3) highlight certain differences between experimental and MD-derived takeaways. In particular, we


9


Page 10 of 28

noticed that the experimental capacitance at slow sweep rates is two to three times higher than the MD-calculated capacitance. Certain intrinsic differences between the model and the experimental configuration most likely account for these deviations. Our MD systems represent graphene nanoplatelet surfaces as perfect basal planes. We incorporate hydroxyl groups on these (initially) flawless surfaces to simulate oxidized nanoplatelets. The experimental counterparts include additional structural factors that might affect capacitance but are convoluted, difficult to identify and quantify, and, therefore, essentially impossible to be implemented in our simulation model to the same extent. First, nanoplatelets are multilayered graphene flakes, and feature edges and surface curvature, which might yield higher capacitance.36,53 Second, assembly of platelet flakes into bulk binder-bound electrode films may create ion confinement in the form of interparticle slits. Previous results have demonstrated significant capacitance increases in cases of matching electrolyte and micropore dimensions.11,54 Finally, our simulation model used the fixed charge method, where net charges were evenly distributed on the surface atoms. A constant potential method, where each atomic charge is permitted to fluctuate according to its local environment, might yield a more accurate representation of the potential distribution of the system.55 In this paper, instead of quantitatively predicting the capacitance of oxidized graphene, we focus more on discussing the qualitative change in capacitance due to surface oxidization. 3.2 EDL Structure Figure S4 in SI shows number densities of both electrolytes on PG and OG at PZC. The PZC of EMIm-TFSI/ACN electrolyte dropped from 0.05 V to -0.09 V after hydroxylation. Similarly, TEA-BF4/ACN electrolyte has a PZC of 0.19 V on PG, and decreased to 0.03 V on OG. The decrease of PZC denotes the drop of excess charges in the EDL, indicating higher affinity of anions to the OG surface. In addition, hydroxylated surfaces attract more anions, and bring electrolyte


10

Page 11 of 28 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


species closer to the surface (Figure S4(c) and S4(d) in SI), likely through the interaction between the hydroxyl groups and the polar F atoms in the anions. However, the cations behave differently in these two electrolytes. [EMIm+] has very similar distribution on PG and OG, whereas [TEA+] was observed to have a much higher peak at 0.5 nm on OG surface than on PG surface. The [TEA+] layer at such a distance is likely formed due to the attraction of first [BF4-] layer at around 0.4 nm (Figure S4(d) in SI), whose peak is significantly higher on OG than on PG. Since OG accommodates higher ion concentrations in EDL, ACN is less concentrated in the first layers at PZC. The comparison of number densities at charged surfaces is shown in Figure S5 in SI. Of note, that these number densities also contain contributions from PZC. To exclude these biases, Figure 4 shows the relative number densities of the two electrolytes on charged surfaces by subtracting their corresponding number densities at PZC. Figure 4(a) and 4(b) show the relative number densities of EMIm-TFSI/ACN. By comparison, the cation shows relatively similar distribution between PG and OG, whereas the anion appears to have observable difference. More specifically, on the negative electrodes, anions act as co-ions and deplete from the surface during charging. Thus the anion curves have negative valleys in Figure 4(a). While the depth of the anion valleys are similar, the anion is much closer to OG than PG. As shown in Figure S5(a), there is an anion peak at around 0.45 nm, which impedes efficient charge screening and decreases capacitance of OG electrode. On positive electrodes, the anion has slightly low peak height, but is also slightly closer to the electrode surface. This is likely due to the competition between strong interaction between anion and the surface hydroxyls and the steric hindrance introduced by these hydroxyls. While the former draws the anions closer to the surface, the latter repels some anions from the EDL.39 The overall result as shown by the capacitance curve is that the capacitance decreases.


11


Page 12 of 28

Figure 4(c) and 4(d) shows EDL structure of the TEA-BF4/ACN system, where both cation and anion behave differently on PG and OG. Figure 4(c) shows that the counter-ion ([TEA+]) have lower peak heights after hydroxylation on negatively charged surfaces. The decreased [TEA+] peak on OG surface is likely a result of steric hindrance created by the surface hydroxyls, which may be more significant for big ions like [TEA+] than for smaller ions within the scope of this study. The effect of steric hindrance could also be inferred from the more dispersed distribution of [TEA+] between 0.4-0.6 nm in Figure S5(c). Accordingly, the co-ion has deeper valley on OG, meaning that less co-ions are presented in the EDL. On the contrary, Figure 4(d) shows that the peak of counter-ion [BF4-] nearly doubled after hydroxylation and the peak is also slightly closer to the surface. Anion has a peak height around 10 # nm-3 at 0.38 nm on PG, which, considering the small size of [BF4-], is far from saturation. Thus, the anion-affinitive OG surface has the capability to attract more anions to its interface than the PG surface. This intensified counter-ion layer also strengthens the co-ion ([TEA+]) layer at 0.53 nm. As a result, both counter-ions and co-ions have higher concentrations in EDL on OG than on PG. The synchronous increase or decrease of cation and anion complicate things, making it hard to interpret the capacitance from EDL structure. To better decode the charge storage mechanism in the TEA-BF4/ACN systems, the charge screening factor56 was calculated: 1 0 𝑓L (𝑧) = − * ∆𝜌- (𝑢)d𝑢 𝜎 )

(4)

where 𝜎 is the surface charge density, while ∆𝜌- (𝑢) is the variation of charge density as the surface charge density changed from 0 to 𝜎. If charge overscreening happened at the interface, fs will exceed 1. As described by Kornyshev et al56, overscreening is pronounced at a small voltage and, at higher potentials, is gradually replaced by the crowding of counter-ions until complete lattice saturation (at very large voltages).


12

Page 13 of 28 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


The charge screening factors with different surface charge densities can be found in Figure 5. Let’s first look at positive electrodes (Figure 5(b) and 5(d)), the peak height first increases as charging begins, but then decreases as surface charge density becomes large, which matches aforementioned Kornyshev’s theory.56 Because the OG surface is more occupied with counter-ions, the crowding of counter-ions happens at a lower surface charge density for OG compared to PG, which leads to earlier drop of peak height on OG as the surface charge increases. As a result, the screening factor is higher for OG with low surface charge and is higher for PG with high surface charge. At negative electrodes (Figure 5(a) and 5(c)), both PG and OG exhibit decreased peak heights at 0.32 nm as charging processes, where OG shows more substantial drops. According to Figure S5 in SI, there are no significant counter-ion or co-ion peaks at 0.32 nm, but only ACN has high peaks at this distance. Considering the high dipole moment of ACN (3.92 D), the distribution and orientation of ACN molecules near the electrode surface may play an important role on charge screening, and hence capacitance. Besides the decreasing first peaks in Figure 5(a) and 5(c), there are also nonnegligible peaks at 0.4 nm, which are the results of the over-screening of counter-ions at such distance. For both PG and OG, the peak at this distance shows an initial upward trend and followed by a downward trend, in response to growing surface charge densities. In an effort to further decouple the structure of ACN layer on electrode surface, the orientation order parameter is introduced in order to study the alignment of ACN along the direction perpendicular to the surface: 𝑃F (𝑐𝑜𝑠𝜃 ) =

1 (3〈𝑐𝑜𝑠 F 𝜃〉 − 1) 2

(5)

where the angle 𝜃 is formed by the normal vector of the surface and the vector from methyl C to N in ACN. denotes ensemble average. The range of orientation order parameter is from -0.5 to 1. If 𝑃F (𝑐𝑜𝑠𝜃) is -0.5, ACN lies parallel to the surface; if 𝑃F (𝑐𝑜𝑠𝜃) is 0, ACN adopts random


13


Page 14 of 28

orientation distribution; if 𝑃F (𝑐𝑜𝑠𝜃) is 1, ACN lies perpendicular to the surface. Figure S6 in SI shows the order parameter as a function of distance to the surface. Each curve starts with a value around -0.5 near the surface, suggesting a parallel orientation of ACN with respect to the surface, and then oscillates between negative and positive values, and, finally, converges to zero. This behavior implies that the ACN molecules orient themselves according to their local environments, resulting in local potential drop. To quantify the potential drop induced by ACN, we decomposed the total spatial charge in equation (1) into 2 parts: IL charge density and solvent charge density, respectively. 𝜌- = 𝜌-UV + 𝜌-XYZ

(6)

Then equation (1) can be rewritten as: ∅(𝑧) = \−

1 0 𝜎 1 0 * (𝑧 − 𝑧 + )𝜌-UV (𝑧 + )d𝑧 + − 𝑧] + \− * (𝑧 − 𝑧 + )𝜌-XYZ (𝑧 + )d𝑧 + ] 𝜀) ) 𝜀) 𝜀) )

(7)

In equation (5), the total potential drop is the sum of partial potential drop from IL and ACN. Figure S7 in SI shows the potential contributions generated from ACN with different surface charges. The partial potential drops from IL and ACN with regard to different surface charge densities were summarized in Figure 6. Figure 6(a) shows the partial potential drop contributed by IL, where the potential drops of OG are closer to 0. The specific capacitance is directly related to the inverse of the potential drop: `

𝐶=a

(8)

where 𝜎 is the surface charge density and 𝜎 is the potential drop. Based on this, the capacitance contributed purely by IL is higher on OG than on PG. However, the presence of ACN in EDL reverses the relative total capacitance.


14

Page 15 of 28 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


As shown in Figure 6(b), the potential drop contributed by ACN has the opposite sign to the potential drop by IL, which reduces the absolute value of the total potential drop and thus enlarges capacitance according to equation (8). Put in another way, the electrode could store the amount of charge at a lower potential, yielding higher capacitance. Previous findings have also shown that adding appropriate solvents to the ionic liquids electrolyte may help to increase the capacitance.57– 59

The work presented herein confirmed the integral role of the distribution and orientation of polar

solvent species near electrode surfaces. Furthermore, the total capacitance has both contributions from IL and ACN. While the PG has lower partial capacitance from IL, its total capacitance is higher, especially at positive potentials. From Figure 6S, we know that the orientation distribution of ACN does not differ obviously between PG and OG. The reason that PG has much higher potential contribution from ACN at positive electrode is attributed to difference of ACN concentrations. As shown in Figure 4(d) and Figure S5(d), the positive OG surface accommodates more ions than PG surface, squeezing out ACN molecules, and the ACN peak is almost halved. Similarly, the negative electrodes adsorb bulky cations ([TEA+]), leading to decreased ACN concentrations. As a result, the potential drop yielded by ACN on these surfaces is less prominent. Our approach comprehensively dissected the structure of double layers to decouple the relative influence of key factors on capacitance of pristine and oxidized graphene. While the anion played the major role in the drop of capacitance of EMIm-TFSI/ACN after hydroxylation, the capacitive behavior of TEA-BF4/ACN involves the participation of both IL and solvents. ACN exhibits an oscillatory orientation distribution across the EDL and significantly influences capacitance, especially on positive electrodes. 3.3 Ion and Solvent Dynamics


15


Page 16 of 28

Since the ions and solvent both form layered structure near electrode surfaces, heterogeneity of ion and solvent dynamics can be observed in the direction perpendicular to the electrode surface. Ion density is conventionally considered a major factor that influences the ion dynamics. Subsequently, in order to find a way to boost charging dynamics, previous studies have correlated ion density in slit pores with diffusion coefficients.60,61 As shown in Figure 7(a), the self-diffusion coefficients drop when ion/solvent approaches surface. This trend is more significant on OG, as the diffusion coefficients in the interfacial region decrease by more than an order of magnitude compared with those in bulk-like region. As a comparison, the diffusion coefficients on PG exhibit only slight decreases during the transition from bulk-like region to interfacial region. In addition, the diffusion coefficients of cation and anion are very similar, suggesting strong correlations between cation and anions. Figure 7(b) shows the average number densities in each region. In the interfacial region, OG brings ions closer to the surface, resulting in higher ion concentrations near the oxidized surfaces. These conditions prohibit fast ion dynamics. In the bulk-like region, as the number densities recover to bulk number densities, the diffusion coefficients are similar for both PG and OG. However, in the transitional region, even though the number densities are similar to bulk-like regions, the diffusion coefficients are still much lower than those in the bulk-like region. This suggests that a number of intricate factors that convolute the diffusive behavior of ion/solvent in the interface. It cannot be solely explained by the ion density, and both ion-ion and ion-wall interactions play an important role. Similar trends are observed when the surfaces are charged. As shown in Figure 7(c), ions that approach surfaces exhibit lower diffusion coefficients, and this trend is more obvious on OG. The lateral diffusion coefficients presented above only illustrate how fast ions move in the directions parallel to the surface. However, the mobility of ions in the direction perpendicular to


16

Page 17 of 28 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


the surface is also important. To that end, the residence time autocorrelation functions examine the time scale during which ions/solvents remain within the interfacial region. The function was calculated using the following formula: 𝐶c (𝑡) =

〈𝑅(0)𝑅(𝑡)〉 〈𝑅(0)𝑅(0)〉

(9)

where 𝑅(𝑡) is a binary function that equals to 1 if a molecule within the interfacial region at time 0 is also found in the interfacial region at time t, and 0 otherwise. Figure 8 shows 𝐶c (𝑡) at PZC, positive electrode, and negative electrode, respectively. At PZC, [TEA+] has higher residence time, as shown in Figure 8(a). Therefore, even though [TEA+] has higher lateral diffusion coefficients than [BF4-], it is less mobile in the direction perpendicular to the surface. The curves of OG are higher than those of PG, which indicates longer residence time and lower mobility for ions and solvent on OG. On charged surfaces, counter-ions always exhibit the longest residence time, indicating that they are strongly adsorbed by the electrodes, and reluctant to swap with ions in the transitional region. Moreover, the hydroxyl groups increase the residence time of counter-ions in the interfacial region. The dynamics results for EMIm-TFSI/ACN systems are similar, and can be found in Figure S8 in SI.

4. CONCLUSIONS We evaluated the charge storage capabilities of graphene electrodes with pristine and hydroxylated surfaces and the influence of different interfaces on two distinct IL/solvent electrolyte systems. Experimental results and MD simulations both showed that oxygen moieties decreased differential capacitance for both systems. An in-depth analysis of MD simulations allowed us to fully analyze the EDL structure. In the EMIm-TFSI/ACN system, interactions of the [TFSI-] anion play a more significant role and decrease capacitance on hydroxylated electrodes. In TEA-BF4/ACN systems,


17


Page 18 of 28

due to layered distribution across the electrode-to-electrode channel, both ions and the ACN solvent molecules affect total capacitance. Especially on positive graphene electrodes, a high ACN concentration layer, coupled with a potential drop yield, compensates for the overall potential and increases capacitance. The simulation and experimental data sets also provided insights into dynamics of ions and solvents on differently functionalized planar electrodes. The lateral diffusion coefficient drops as ion or solvent approaches surface, and oxidization amplifies this effect by an order of magnitude. Residence time autocorrelation functions show the exchanging rate of ions/solvent between interfacial region and transitional region. Counter-ions near OG surface exhibit longer residence times than on the PG surface, indicating reduced mobility in the direction perpendicular to the surface, i.e., less ion exchange between the interface and the transitional region and, eventually, the bulk-like region. The combined MD simulation and electrochemical measurements provided a comprehensive picture of the effects of surface hydroxylation of graphene on the capacitance and ion/solvent dynamics.

5. SUPPORTING INFORMATION Regularly and randomly distributed oxidized graphene and their differential capacitance, number density and angle distribution of electrolyte, partial potential distribution of acetonitrile, and residence time autocorrelation function for EMIm-TFSI/ACN.

6. AKNOWLEDGEMENTS This research was supported by the Fluid Interface Reactions, Structures and Transport (FIRST) Center, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences. This research used resources of the National Energy


18

Page 19 of 28 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


Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy. Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for U.S. DOE under Contract No. DE-AC05-00OR22725. This research used resources of the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

REFERENCES (1) Simon, P.; Gogotsi, Y. Materials for Electrochemical Capacitors. Nat Mater 2008, 7, 845– 854. (2) Simon, P.; Gogotsi, Y.; Dunn, B. Where Do Batteries End and Supercapacitors Begin? Science 2014, 343, 1210–1211. (3) Miller, J.; Simon, P. Electrochemical Capacitors for Energy Management. Science 2008, 321, 651–652. (4) Kornyshev, A. A. Double-Layer in Ionic Liquids: Paradigm Change? J. Phys. Chem. B 2007, 111, 5545–5557. (5) Feng, G.; Huang, J.; Sumpter, B. G.; Meunier, V.; Qiao, R. A “Counter-Charge Layer in Generalized Solvents” Framework for Electrical Double Layers in Neat and Hybrid Ionic Liquid Electrolytes. Phys. Chem. Chem. Phys. 2011, 13, 14723–14734. (6) Feng, G.; Qiao, R.; Huang, J.; Sumpter, B. G.; Meunier, V. Atomistic Insight on the Charging Energetics in Subnanometer Pore Supercapacitors. J. Phys. Chem. C 2010, 114, 18012–18016. (7) Li, S.; Zhang, P.; Fulvio Pasquale, F.; Hillesheim Patrick, C.; Feng, G.; Dai, S.; Cummings Peter, T. Enhanced Performance of Dicationic Ionic Liquid Electrolytes by Organic Solvents. J. Phys. Condens. Matter 2014, 26, 284105. (8) Feng, G.; Jiang, X.; Qiao, R.; Kornyshev, A. A. Water in Ionic Liquids at Electrified Interfaces: The Anatomy of Electrosorption. ACS Nano 2014, 8, 11685–11694. (9) Li, S.; Feng, G.; Fulvio, P. F.; Hillesheim, P. C.; Chen, L.; Dai, S.; Cummings, P. T. Molecular Dynamics Simulation of Binary Mixture of Ionic Liquids near Carbon Electrode: The Capacitive Performance and Temperature Effects. Abstr. Pap. Am. Chem. Soc. 2013, 245, 1. (10) Dyatkin, B.; Osti, N. C.; Gallegos, A.; Zhang, Y.; Mamontov, E.; Cummings, P. T.; Wu, J.; Gogotsi, Y. Electrolyte Cation Length Influences Electrosorption and Dynamics in Porous Carbon Supercapacitors. Electrochim. Acta 2018, 283, 882–893. (11) Feng, G.; Cummings, P. T. Supercapacitor Capacitance Exhibits Oscillatory Behavior as a Function of Nanopore Size. J. Phys. Chem. Lett. 2011, 2, 2859–2864. (12) Feng, G.; Li, S.; Presser, V.; Cummings, P. T. Molecular Insights into Carbon Supercapacitors Based on Room-Temperature Ionic Liquids. J. Phys. Chem. Lett. 2013, 4, 3367–3376.


19


(13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30)

Page 20 of 28

Feng, G.; Li, S.; Atchison, J. S.; Presser, V.; Cummings, P. T. Molecular Insights into Carbon Nanotube Supercapacitors: Capacitance Independent of Voltage and Temperature. J. Phys. Chem. C 2013, 117, 9178–9186. Feng, G.; Jiang, D. E.; Cummings, P. T. Curvature Effect on the Capacitance of Electric Double Layers at Ionic Liquid/Onion-like Carbon Interfaces. J. Chem. Theory Comput. 2012, 8, 1058–1063. Li, S.; Feng, G.; Cummings, P. T. Interfaces of Dicationic Ionic Liquids and Graphene: A Molecular Dynamics Simulation Study. J. Phys. Condens. Matter 2014, 26, 284106. Feng, G.; Huang, J.; Sumpter, B. G.; Meunier, V.; Qiao, R. Structure and Dynamics of Electrical Double Layers in Organic Electrolytes. Phys. Chem. Chem. Phys. 2010, 12, 5468–5479. Feng, G.; Qiao, R.; Huang, J.; Dai, S.; Sumpter, B. G.; Meunier, V. The Importance of Ion Size and Electrode Curvature on Electrical Double Layers in Ionic Liquids. Phys. Chem. Chem. Phys. 2011, 13, 1152–1161. Li, S.; Van Aken, K. L.; McDonough, J. K.; Feng, G.; Gogotsi, Y.; Cummings, P. T. The Electrical Double Layer of Dicationic Ionic Liquids at Onion-like Carbon Surface. J. Phys. Chem. C 2014, 118, 3901–3909. Zhu, Q.; Song, Y.; Zhu, X.; Wang, X. Ionic Liquid-Based Electrolytes for Capacitor Applications. J. Electroanal. Chem. 2007, 601, 229–236. Stoppa, A.; Zech, O.; Kunz, W.; Buchner, R. The Conductivity of Imidazolium-Based Ionic Liquids from (-35 to 195) °c. A. Variation of Cations Alkyl Chain. J. Chem. Eng. Data 2010, 55, 1768–1773. Brandt, A.; Pohlmann, S.; Varzi, A.; Balducci, A.; Passerini, S. Ionic Liquids in Supercapacitors. MRS Bull. 2013, 38, 554–559. Lin, R.; Huang, P.; Ségalini, J.; Largeot, C.; Taberna, P. L.; Chmiola, J.; Gogotsi, Y.; Simon, P. Solvent Effect on the Ion Adsorption from Ionic Liquid Electrolyte into SubNanometer Carbon Pores. Electrochim. Acta 2009, 54, 7025–7032. Sheng, K.; Sun, Y.; Li, C.; Yuan, W.; Shi, G. Ultrahigh-Rate Supercapacitors Based on Eletrochemically Reduced Graphene Oxide for Ac Line-Filtering. Sci. Rep. 2012, 2, 247. Béguin, F.; Presser, V.; Balducci, A.; Frackowiak, E. Carbons and Electrolytes for Advanced Supercapacitors. Adv. Mater. 2014, 26, 2219–2251. Zhong, C.; Deng, Y.; Hu, W.; Qiao, J.; Zhang, L.; Zhang, J. A Review of Electrolyte Materials and Compositions for Electrochemical Supercapacitors. Chem. Soc. Rev. 2015, 44, 7484–7539. Zhan, C.; Lian, C.; Zhang, Y.; Thompson, M. W.; Xie, Y.; Wu, J.; Kent, P. R. C.; Cummings, P. T.; Jiang, D. E.; Wesolowski, D. J. Computational Insights into Materials and Interfaces for Capacitive Energy Storage. Adv. Sci. 2017, 4. Meyer, J. C.; Geim, A. K.; Katsnelson, M. I.; Novoselov, K. S.; Booth, T. J.; Roth, S. The Structure of Suspended Graphene Sheets. Nature 2007, 446, 60–63. Allen, M. J.; Tung, V. C.; Kaner, R. B. Honeycomb Carbon: A Review of Graphene. Chem. Rev. 2010, 110, 132–145. Shi, M.; Kou, S.; Yan, X. Engineering the Electrochemical Capacitive Properties of Graphene Sheets in Ionic-Liquid Electrolytes by Correct Selection of Anions. ChemSusChem 2014, 7, 3053–3062. Kislenko, S. a; Samoylov, I. S.; Amirov, R. H. Molecular Dynamics Simulation of the Electrochemical Interface between a Graphite Surface and the Ionic Liquid BMIM PF6.


20

Page 21 of 28 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


(31) (32) (33) (34) (35) (36) (37)

(38) (39) (40) (41) (42) (43) (44) (45) (46) (47)

Phys. Chem. Chem. Phys. 2009, 11, 5584–5590. Merlet, C.; Rotenberg, B.; Madden, P. a; Taberna, P.-L.; Simon, P.; Gogotsi, Y.; Salanne, M. On the Molecular Origin of Supercapacitance in Nanoporous Carbon Electrodes. Nat. Mater. 2012, 11, 306–310. Vatamanu, J.; Cao, L.; Borodin, O.; Bedrov, D.; Smith, G. D. On the Influence of Surface Topography on the Electric Double Layer Structure and Differential Capacitance of Graphite/Ionic Liquid Interfaces. J. Phys. Chem. Lett. 2011, 2, 2267–2272. Jeong, H. M.; Lee, J. W.; Shin, W. H.; Choi, Y. J.; Shin, H. J.; Kang, J. K.; Choi, J. W. Nitrogen-Doped Graphene for High-Performance Ultracapacitors and the Importance of Nitrogen-Doped Sites at Basal Planes. Nano Lett. 2011, 11, 2472–2477. Zhan, C.; Zhang, Y.; Cummings, P. T.; Jiang, D. Enhancing Graphene Capacitance by Nitrogen: Effects of Doping Configuration and Concentration. Phys. Chem. Chem. Phys. 2016, 18, 4668–4674. Yang, H.; Yang, J.; Bo, Z.; Zhang, S.; Yan, J.; Cen, K. Edge Effects in VerticallyOriented Graphene Based Electric Double-Layer Capacitors. J. Power Sources 2016, 324, 309–316. Zhan, C.; Zhang, Y.; Cummings, P. T.; Jiang, D. en. Computational Insight into the Capacitive Performance of Graphene Edge Planes. Carbon N. Y. 2017, 116, 278–285. Dyatkin, B.; Zhang, Y.; Mamontov, E.; Kolesnikov, A. I.; Cheng, Y.; Meyer, H. M.; Cummings, P. T.; Gogotsi, Y. Influence of Surface Oxidation on Ion Dynamics and Capacitance in Porous and Nonporous Carbon Electrodes. J. Phys. Chem. C 2016, 120, 8730–8741. Park, S. W.; DeYoung, A. D.; Dhumal, N. R.; Shim, Y.; Kim, H. J.; Jung, Y. J. Computer Simulation Study of Graphene Oxide Supercapacitors: Charge Screening Mechanism. J. Phys. Chem. Lett. 2016, 7, 1180–1186. Deyoung, A. D.; Park, S. W.; Dhumal, N. R.; Shim, Y.; Jung, Y.; Kim, H. J. Graphene Oxide Supercapacitors: A Computer Simulation Study. J. Phys. Chem. C 2014, 118, 18472–18480. Hendricks, T. R.; Lu, J.; Drzal, L. T.; Lee, I. Intact Pattern Transfer of Conductive Exfoliated Graphite Nanoplatelet Composite Films to Polyelectrolyte Multilayer Platforms. Adv. Mater. 2008, 20, 2008–2012. Stoller, M. D.; Ruoff, R. S. Best Practice Methods for Determining an Electrode Material’s Performance for Ultracapacitors. Energy Environ. Sci. 2010, 3, 1294–1301. Van Aken, K. L.; Beidaghi, M.; Gogotsi, Y. Formulation of Ionic-Liquid Electrolyte to Expand the Voltage Window of Supercapacitors. Angew. Chemie - Int. Ed. 2015, 54, 4806–4809. Canongia Lopes, J. N.; Deschamps, J.; Pádua, A. A. H. Modeling Ionic Liquids Using a Systematic All-Atom Force Field. J. Phys. Chem. B 2004, 108, 2038–2047. Lopes, J. N. C.; Pádua, A. A. H. Molecular Force Field for Ionic Liquids Composed of Triflate or Bistriflylimide Anions. J. Phys. Chem. B 2004, 108, 16893–16898. Yan, J. A.; Chou, M. Y. Oxidation Functional Groups on Graphene: Structural and Electronic Properties. Phys. Rev. B - Condens. Matter Mater. Phys. 2010, 82, 21–24. Rizzo, R. C.; Jorgensen, W. L. OPLS All-Atom Model for Amines: Resolution of the Amine Hydration Problem. J. Am. Chem. Soc. 1999, 121, 4827–4836. Abraham, M. J.; Murtola, T.; Schulz, R.; Páall, S.; Smith, J. C.; Hess, B.; Lindah, E. Gromacs: High Performance Molecular Simulations through Multi-Level Parallelism from


21


(48) (49) (50) (51) (52) (53) (54) (55) (56) (57) (58) (59) (60) (61)

Page 22 of 28

Laptops to Supercomputers. SoftwareX 2015, 1–2, 19–25. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. LINCS: A Linear Constraint Solver for Molecular Simulations. J. Comput. Chem. 1997, 18, 1463–1472. Yeh, I.-C.; Berkowitz, M. L. Ewald Summation for Systems with Slab Geometry. J. Chem. Phys. 1999, 111, 3155. Spohr, E. Molecular Simulation of the Electrochemical Double Layer. Electrochim. Acta 1999, 44, 1697–1705. Fedorov, M. V.; Kornyshev, A. a. Ionic Liquid near a Charged Wall: Structure and Capacitance of Electrical Double Layer. J. Phys. Chem. B 2008, 112, 11868–11872. Lockett, V.; Sedev, R.; Ralston, J.; Horne, M.; Rodopoulos, T. Differential Capacitance of the Electrical Double Layer in Imidazolium-Based Ionic Liquids: Influence of Potential, Cation Size, and Temperature. J. Phys. Chem. C 2008, 112, 7486–7495. Sacci, R. L.; Matsumoto, R.; Thompson, M. W.; Sanders, N. C.; Cummings, P. T. Scalable Screening of Soft Matter: A Case Study of Mixtures of Ionic Liquids and Organic Solvents. J. Phys. Chem. B 2019, acs.jpcb.8b11527. Dahl, I.; Kjølseth, C.; Fjeld, H.; Prytz, Ø.; Inge, P.; Estournès, C.; Haugsrud, R.; Norby, T. Anomalous Increase in Carbon Capacitance at Pore Sizes Less Than 1 Nanometer. Science (80-. ). 2006, 313, 1760–1763. Merlet, C.; Péan, C.; Rotenberg, B.; Madden, P. A.; Simon, P.; Salanne, M. Simulating Supercapacitors: Can We Model Electrodes as Constant Charge Surfaces? J. Phys. Chem. Lett. 2013, 4, 264–268. Bazant, M. Z.; Storey, B. D.; Kornyshev, A. A. Double Layer in Ionic Liquids: Overscreening versus Crowding. Phys. Rev. Lett. 2011, 106, 6–9. Liu, W.; Yan, X.; Lang, J.; Xue, Q. Effects of Concentration and Temperature of EMIMBF 4/Acetonitrile Electrolyte on the Supercapacitive Behavior of Graphene Nanosheets. J. Mater. Chem. 2012, 22, 8853–8861. Uralcan, B.; Aksay, I. A.; Debenedetti, P. G.; Limmer, D. T. Concentration Fluctuations and Capacitive Response in Dense Ionic Solutions. J. Phys. Chem. Lett. 2016, 7, 2333– 2338. Liu, K.; Wu, J. Boosting the Performance of Ionic-Liquid-Based Supercapacitors with Polar Additives. J. Phys. Chem. C 2016, 120, 24041–24047. He, Y.; Qiao, R.; Vatamanu, J.; Borodin, O.; Bedrov, D.; Huang, J.; Sumpter, B. G. Importance of Ion Packing on the Dynamics of Ionic Liquids during Micropore Charging. J. Phys. Chem. Lett. 2016, 7, 36–42. Kondrat, S.; Wu, P.; Qiao, R.; Kornyshev, A. a. Accelerating Charging Dynamics in Subnanometre Pores. Nat. Mater. 2014, 13, 387–393.


22

Page 23 of 28 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


Figure 1. (a) Snapshot of the simulation system. (b) Molecular structures of ions and solvent.

Figure 2 (a) Cyclic voltammogram sweep of pristine (PG) and oxidized (OG) graphene nanoplatelet electrodes in EMIm-TFSI/ACN at 2 mV s-1. (b) Rate handling comparison for individual positive and negative electrodes for PG and OG in EMIm-TFSI/ACN in the 2 mV s-1 – 1000 mV s-1 sweep range. (c) Nyquist impedance plot that compares electrosorption dynamics of


23


Page 24 of 28

EMIm-TFSI/ACN in PG and OG. (d) 2 mV s-1 CV sweep of PG and OG in TEA-BF4/ACN electrolyte. (e) Range of CV sweeps for PG and OG in TEA-BF4/ACN. (f) Nyquist impedance plot for PG and OG in TEA-BF4/ACN.

Figure 3. Differential capacitance of (a) EMIm-TFSI/ACN and (b) TEA-BF4/ACN


24

Page 25 of 28 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


Figure 4. Number densities relative to PZC: (a) EMIm-TFSI/ACN on negative electrodes (-5 µC/cm2); (b) EMIm-TFSI/ACN on positive electrodes (5 µC/cm2); (c) TEA-BF4/ACN on negative electrodes (-5µC/cm2); (d) TEA-BF4/ACN on positive electrodes (5µC/cm2).

Figure 5. Screening factor as a function of distance to (a) negative PG electrode surface, (b) positive PG surface, (c) negative OG surface, and (d) positive OG surface. The legend denotes the absolute surface charge density of each curve.

Figure 6. Partial potential drop contributed by (a) IL and (b) ACN.


25


Page 26 of 28

Figure 7. (a) Diffusion coefficients and (b) average number densities of TEA-BF4/ACN at PZC. (c) Diffusion coefficients in different regions with surface charge densities of ±5 µC/cm2 (left side is positively charged and right side is negatively charged).


26

Page 27 of 28 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


Figure 8. Residence time autocorrelation function of TEA-BF4/ACN at (a) PZC, (b) positive electrode (5 µC/cm2), and (c) negative electrode (-5 µC/cm2).


27


Page 28 of 28

TOC Graphic


28

Molecular Investigation of Oxidized Graphene: Anatomy of Double

Recommend Documents