ABSTRACT Graphene film has been demonstrated as promising active materials for electric double layer capacitors (EDLCs), mainly due to its excellent mechanical flexibility and freestanding

morphology. In this work, the distribution and variation pattern of electrolyte ions in graphene-film based EDLC electrodes are investigated with a 11B magic-angle spinning nuclear magnetic

resonance (MAS-NMR) spectroscopy. For neutral graphene films soaked with different amounts of electrolytes (1 M TEABF4/ACN), weakly and strongly adsorbed anions are identified based on the

resonances at different 11B chemical shifts. Unlike other porous carbonaceous materials, the strongly adsorbed anions are found as the major electrolyte anions components in graphene films.

will be useful for the design and fabrication of high performance graphene-film based EDLCs. SIMILAR CONTENT BEING VIEWED BY OTHERS UNRAVELING THE ENERGY STORAGE MECHANISM IN GRAPHENE-BASED

NONAQUEOUS ELECTROCHEMICAL CAPACITORS BY GAP-ENHANCED RAMAN SPECTROSCOPY Article Open access 04 July 2024 ELECTROLYTE GATING IN GRAPHENE-BASED SUPERCAPACITORS AND ITS USE FOR PROBING

NANOCONFINED CHARGING DYNAMICS Article 22 June 2020 CATION DESOLVATION-INDUCED CAPACITANCE ENHANCEMENT IN REDUCED GRAPHENE OXIDE (RGO) Article Open access 02 March 2024 INTRODUCTION

Electrochemical double layer capacitors (EDLCs), also known as supercapacitors or ultracapacitors, have attracted considerable attention due to their high power density, long cycle life and

low maintenance cost1,2,3. While extensive efforts have been devoted to the design and construction of high performance active materials, it is equally important to explore the underlying

charge storage mechanism of EDLCs4,5,6,7. A number of theoretical and experimental methods have been developed to investigate the structure and dynamics of electrode-electrolyte interface in

EDLCs8,9,10,11,12. Typically, molecular dynamic (MD) simulation is capable of describing the movement and adsorption of ions, as well as the formation of micro-structures of EDLCs at the

active materials and current collector interfaces, based on the calculation at molecular/atomic scales13,14,15; _in situ_ electrochemical quartz crystal microbalance (EQCM) is highly

effective at determining the mass variation of electrode within charging-discharging processes, by measuring the change in frequency of a quartz crystal resonator11,16; and _in situ_

infrared spectroscopy has been employed to experimentally monitor the migration of electrolyte ions within the pores of porous electrodes12. In recent years, nuclear magnetic resonance (NMR)

has emerged as an element-selective, highly localized and quantitative technique to obtain the atomic-scale information on the local environments with direct observation8,17. As for EDLCs

employing carbonaceous active materials, the species close to the solid surfaces experience a diamagnetic shielding (arising from the “ring current” in carbon electrodes18,19,20), leading to

the shift of NMR signals to low frequencies. This chemical shift, _i.e._, the so-called nucleus-independent chemical shift (NICS), can be used as a fingerprint to distinguish the adsorbed

species (ions and molecules) from the free ones in the bulk electrolyte17,18. NMR studies have been applied to porous materials for the understanding of the adsorption of molecules in micro-

and nano-sized pores. Harris _et al_. reported the adsorption of water and phosphorus-containing compounds on activated carbons21,22,23,24, where distinct resonances corresponding to the

adsorbed and non-adsorbed molecules were observed. Similar observation was also presented for the adsorption of water25,26, hydrogen26,27, and methanol28 inside carbon nanotubes (CNTs).

Forse _et al_.29 and Borchardt _et al_.30 unveiled the effects of relative pore/ion size on the adsorption of tetraethylammonium (TEA+) and tetrafluoroborate (BF4−) in carbide-derived

carbons and other porous carbon materials of well-defined, variable pore sizes. With _ex situ_ magic-angle spinning (MAS) NMR, comprehensive information on the anions, cations, and remaining

solvent molecules inside or outside the porosity of active carbons were provided by Deschamps _et al_.31 Moreover, based on the real-time observation on the charging process of EDLCs

employing TEABF4 in acetonitrile (ACN) electrolyte, Wang _et al_.20,32 presented the migration of ions between the electrodes and the changes in the nature of ion binding to the surface.

Such an _in situ_ measurements were also conducted on carbons soaked in different organic33,34 and aqueous electrolytes35,36. The above NMR studies provided useful insights into the

molecular mechanisms of the charge storage of EDLCs, including the adsorption of counter-ions into the electrode pores, expulsion of co-ions form pores, and ion exchange between anions and

cations8,17,37,38,39. Graphene is a two-dimensional nanomaterial with huge specific surface area and a series of unique physical, chemical and mechanical properties40,41. Thin film made of

graphene stacks, also known as “graphene paper” or “graphene film”, has attracted great attention for EDLCs, mainly due to its excellent mechanical flexibility and freestanding

morphology42,43. MD simulation10,44,45,46 and DFT calculation44,45 were conducted to explore the charge storage mechanisms. Typically, the influences of temperature46, electrolyte chemical

structure45, and applied voltage46 on the electric double layer structures and capacitance of EDLCs were unveiled with MD simulation. In our recent works, the crucial roles of channel

width10, edge effect44, and charge density10 on EDLCs performance were investigated with experimental research and numerical simulations. Despite the motion of particles has been well

described at molecular/atomic scale with the above theoretical simulations, a direct observation on the ion distribution and ion population upon charging in graphene films is highly needed.

In this work, solid-state 11B MAS-NMR spectroscopy was applied to graphene-film based EDLCs using TEABF4/ACN electrolytes. The adsorption behaviors of BF4− anions in neutral graphene films

with different electrolyte feedings were investigated. Graphene-film based EDLCs were then assembled and the changes in the population of anions with different adsorption states in electrode

materials were measured at different charging voltages. The as-obtained results provide useful insights into the charge storage mechanisms in graphene nano-channels, which could be helpful

in designing high-performance EDLCs. RESULTS Figure 1a shows the 11B MAS-NMR spectra of graphene films soaked with different amounts (50~200 μL) of 1 M TEABF4/ACN electrolytes. Each spectrum

is fitted with mixed Gaussian/Lorentzian lineshapes, where the as-obtained NMR spectra are shown in black, the fitted components are shown in dark blue and purple, and the total fitted

lineshapes are shown in red. At a relatively low electrolyte loading of 50 μL, a single resonance was detected at the 11B chemical shift of −10.02 ppm. With the increase of the electrolyte

loading, this resonance increased in intensity and shifted in peak position to higher frequencies (_i.e_., the left-hand side of axis), reaching to −5.17 ppm with the electrolyte loading

volume of 200 μL. Meanwhile, a second resonance at higher frequency emerged with the electrolyte loading volumes lager than 75 μL. The chemical shift of the second resonance (−2.75~−2.28 

ppm) is close to that of crystalline TEABF4 (−1.13 ppm, see Fig. S1 in Supplementary Information). The different chemical shifts of the resonances provide an indication of their origin in

graphene materials29. As for the current graphene films with layered structure (see Fig. 1b), different adsorption sites can be specifically mapped to the positions on the graphene

nano-channel surfaces with different adsorption states (see Fig. 1c). As is well known, the resonances corresponding to adsorbed species on carbon surface shift to low frequency as compared

to free species, due to the diamagnetic contribution of the “ring current” in the carbon17. Consequently, the deconvoluted peaks at low frequencies (_e.g_., −10.02 ppm for electrolyte

loading volume of 50 μL) arise from the 11B atoms of strongly bound BF4− anions on carbon surface, which is herein referred to as “strongly adsorbed state”. Density functional theory (DFT)

calculation indicates that the “ring current” effects on adsorbed ions increased rapidly with the decreasing ion-surface distance, which is strongly related to the size of graphene domains,

pore width, and curvature of carbon surface18,32. On the other hand, the resonances at higher frequencies (_e.g_., −2.75 ppm for electrolyte loading volume of 75 μL) are assigned to the

weakly bound BF4− anions. These anions situated further from the graphene surfaces, corresponding to the “weakly adsorbed state”. The anions at this position were weakly affected by the

graphene surfaces and therefore presented a chemical shift close to that of crystalline TEABF4. Due to the rapid dynamic exchange of electrolyte ions between the weakly and strongly adsorbed

sites on the NMR timescale30,31, the observed spectrum is an averaging chemical shift for the anions at different states. Figure 2a shows the integrated intensity of experimental NMR

spectra as a function of electrolyte loadings. A quasi-lineal relation between the total intensity and BF4− anion population in graphene film was obtained, providing a relationship to

quantify the absolute number of ions for a given resonance intensity in the NMR spectrum. Figure 2b shows the deconvoluted intensities of the weakly and strongly adsorbed resonances for

different electrolyte loadings. Each value was normalized by the total intensity of the graphene films soaked with 200 μL 1 M TEABF4/ACN. Both the strongly and weakly adsorbed resonances

intensities increased with the increasing electrolyte loadings. For each certain electrolyte loading, the strongly adsorbed resonances presented a higher intensity than that of the weakly

adsorbed resonances. This observation is different from previous NMR results on some other porous carbon materials, where the weakly adsorbed (_i.e_., ex-pore) resonances presented a higher

intensity than the strongly adsorbed (_i.e_., in-pore) counterparts29,31,32. It can be attributed to the unique ion adsorption behaviors in graphene nano-channels. As for the activated

carbon electrode materials (_e.g_., prepared by mixing 95 wt% carbon powder with 5 wt% PTFE20,32,33,34), the “in-pore” resonances arise from the adsorbed ions on carbon surface within

micropores, and the “ex-pore” resonances are assigned to the ions in reservoirs of electrolyte in large space between primary carbon particles. As a consequence, the number of “in-pore” ions

is less than that of the “ex-pore” ions, leading to a relatively higher intensity of the weakly adsorbed resonances. In contrast, the graphene films used in the current work are binder-free

materials with a unique two-dimensional layered structure. Both the weakly and strongly adsorbed anions are mainly located inside the well-defined nano-channels between adjacent graphene

sheets. It is reasonable that the strongly adsorbed anions are the major component of anions in graphene films, which could play a leading role in charge storage process of graphene-film

based EDLCs. Figure 2c shows the changes of chemical shift for the weakly and strongly adsorbed resonances with the increasing electrolyte loadings. The chemical shifts of the weakly

adsorbed resonances were almost kept as a constant. The increasing chemical shifts of the strongly adsorbed resonances could be attributed to the exchange of strongly bound anions on the

graphene surface with the weakly bound anions outside the adsorption layer20. Since the NMR timescale is much longer than the timescale of ion exchange, the fast exchange of electrolyte

anions between the weakly and strongly adsorbed sites makes these subpopulations resolved in the NMR spectrum47. With the increasing electrolyte loading volumes, the number of anions

involved in exchange increased gradually (as shown in Fig. 2b), leading to a time-averaged ion-distribution in the pore volume of graphene films23,48. Consequently, the measured chemical

shifts of BF4− anions were the NICS averaged over the pore space and had a less negative value than that of anions adsorbed on graphene surfaces48. Figure 2d schematically shows the ion

adsorption process within the graphene channels, based on the analysis on the NMR spectra. For the relatively low electrolyte loadings of less than 75 μL, the absence of weakly adsorbed

resonances indicates that the BF4− anions are preferentially adsorbed on the graphene surface (as schematically shown in the “status 1”). This phenomenon could be attributed to the high

adsorption ability of graphene48. Similar observation was also reported in the NMR study on the adsorption of methanol on CNTs28. The weakly adsorbed anions were discovered at the

electrolyte loading volume of 75 μL (as schematically shown in the “status 2”). However, it is different from the observation in previous NMR studies that the weakly (_i.e_., the ex-pore)

adsorption does not occur until the strongly (_i.e_., the in-pore) adsorption has attained saturation17. In the current work, the strongly adsorbed resonances intensity did not reach a

maximum value (_i.e_., the strongly adsorption was not saturated), when the weakly adsorbed anions were observed. Both the weakly and strongly resonances intensities increased with the

increasing electrolyte loadings, indicating that the strongly and weakly adsorption were carried out simultaneously. For the relatively high electrolyte loadings of more than 75 μL, the

intensities of both the strongly and weakly adsorbed resonances increased as a function of electrolyte loadings, indicating that the number of BF4− anions in this two adsorption sites

increased gradually. This observation could be attributed to two possible adsorption processes of electrolyte ions. With the increase of electrolyte loadings, the electrolyte ions entered

into the inner space of the graphene channels, and the volume occupied by the electrolyte ions increased gradually. Meanwhile, the packing density of the ions within graphene nano-channels

increased with the increasing electrolyte loadings, and therefore more ions were stockpiled in graphene films (as schematically shown in the “status 3”). To further describe the ion packing

behavior during the adsorption process, a coefficients _η_ was introduced: where _V_ion and _V_pore represent the volume of electrolyte ions and the pore volume of graphene film,

respectively. According to N2 adsorption-desorption isotherms, the graphene film used in the current work has a pore volume (_V_pore) of 2.11 cm3 g−1, which is in agreement with the results

of other researches49,50. Assuming that there are equal numbers of anions and cations in graphene film, the total volume occupied by full-solvated electrolyte ions, _V_ion, can be calculated

as: where _d_cation represents the diameter of the full-solvated cations (TEA+), _d_anion represents the diameter of the full-solvated anions (BF4−), and _λ_ refers to the packing factor of

full-solvated electrolyte ions. For TEABF4/ACN electrolyte, _d_cation and _d_anion are 1.3 nm (TEA+) and 1.16 nm (BF4−), respectively51. Since the value of _d_cation nearly equals to that

of _d_anion, the minimum value of _V_ion can be estimated by considering a maximum packing density of solvated anions and cations (for equal spheres, the packing factor _λ_ has a range from

52.36 to 74.05%)52. The numbers of anions and cations per gram of graphene film at each loading volume can be calculated by the following equation: where _c_ is the electrolyte

concentration, _V_ represents the loading volume, _N__a_ refers to the Avogadro’s number and _m_ is the sample weight. For the electrolyte loading volumes of 50 and 75 μL, the coefficients

_η_ were calculated as 0.63 and 0.95, respectively. These values are less than 1, which means that the volumes occupied by full-solvated electrolyte ions are less than the pore volume of

graphene film. The coefficients _η_ were calculated as 1.3, 1.6, 1.9 and 2.5 at the electrolyte loading volumes of 100, 125, 150 and 200 μL, respectively. It suggests that the total volume

occupied by electrolyte ions is higher than the pore volume of graphene films, indicating the partial desolvation of ion solvation shells during the adsorption process. Figure 3a shows the

11B MAS-NMR spectra of positive and negative electrodes at different charging voltages. For the positive electrodes, the peak intensity of NMR lines first decreased and then increased with

an increasing applied voltage from 0 to 2.5 V. Two resonances corresponding to weakly and strongly adsorbed BF4− anions were unveiled by deconvolution of these as-obtained NMR spectra. The

strongly adsorbed resonances played a predominant role on the charging process. For the negative electrodes, similar observations were also obtained upon charging. Figure 3b presents the

fitted intensity of strongly and weakly resonances for positive electrodes, based on the deconvoluted NMR spectra. The total intensity of experimental NMR spectrum at 0 V was used as a

reference. An abnormal decrease in the intensity of strongly adsorbed resonances was observed between 0 and 1.0 V, which means a number of BF4− anions were departed from the graphene channel

during the charging process. It suggests that the charge storage in graphene electrode materials was not driven by the adsorption of anions as the traditional theory supposed17. In this

case, the ejection of TEA+ cations was the primary mechanism for the charging process. At the charging voltage higher than 1.0 V, an abrupt increase in the intensity of strongly adsorbed

resonances was observed. It indicates that a large number of BF4− anions were adsorbed into the graphene nano-channels upon charging. Therefore, the adsorption of anions played a more

significant role in charge storage between 1.0 and 2.5 V. Similar trends were observed for the negative electrodes (see Fig. 3c). When the charging voltages was lower than 1.0 V, a rapid

decrease in intensity of the strongly adsorbed resonances showed up. It suggests that a large number of strongly adsorbed BF4− anions were expulsed from the graphene nano-channels and that

this played a more significant role in the charge storage process. As the charging voltage increase from 1.0 to 2.5 V, the abnormal increase in intensity of the strongly adsorbed resonances

indicates that a number of BF4− anions enter into the graphene channels during charging. However, for the negative electrodes, surplus cations were supposed to counteract the negative charge

that was developed on the electrode surface. As a consequence, the mechanisms of charge storage was dominated by the adsorption of TEA+ cations in the voltage range operated in the current

work. DISCUSSION The results presented in this study highlight the complexity of charge storage mechanisms within the graphene nano-channels. To better understand the charge storage process

that had taken place in graphene films upon charging, the electronic charge and ionic charge stored on the positive electrodes were calculated respectively. Figure 4a shows the electronic

charge stored on the positive electrodes as a function of the increasing charging voltage. It was calculated by integrating the current intensity of current-time curves (Supplementary

Information, Fig. S2) without the contribution of self-discharge. Figure 4b shows the BF4− ionic charge stored on positive electrodes at different charging voltages, where the BF4− ionic

charge was calculated based on the relationship between anion population and NMR spectra intensity shown in Fig. 2a. As the charging voltage increased from 0 to 2.5 V, the BF4− ionic charge

(_i.e_., the population of BF4− anions) within graphene nano-channels decreased firstly and then increased sharply with the charging voltages higher than 1.0 V. In order to understand the

migration of TEA+ cations in electrodes, the TEA+ ionic charge stored on positive electrodes was calculated based on the principle of charge conservation. Figure 4c shows the TEA+ ionic

charge (_i.e_., the population of TEA+ cations) within the graphene channel decreased greatly at the charging voltage lower than 1.0 V. As discussed in the previous section, the expulsion of

cations was the primary mechanisms of charge storage in graphene electrodes. Excess cations have been expelled from the graphene nano-channels upon charging to balance the electronic charge

that accumulated in the graphene surface. Meanwhile, the population of anions decreased abnormally, possibly because amounts of BF4− anions were taken out of the channel by the cations

during the charging process. Similar observation was also reported by Luo _et al_.36, where a nonlinear behavior of both Na+ and BF4− on negative charging above 0.6 V was observed. This

abnormal phenomenon was caused by the competing effect between the ion-ion correlations and the ion-surface electrostatic interactions. In this case, the decreasing TEA+ cations

concentration favors the dragging of BF4− anions out of the graphene channels because of the strongly ion-ion correlations between electrolyte ions. As the charging voltage increased from

1.0 to 2.5 V, a gradually increase in the TEA+ ionic charge was observed, which means that a number of TEA+ cations were brought into graphene films by anions upon charging. Figure 4d

schematically shows the possible charge storage process in graphene nano-channels. At low charging voltages ranging from 0 V to 1.0 V, the charge storage process was driven by the expulsion

of TEA+ cations. Meanwhile, a number of anions were taken out of graphene films by cations because of the ion-ion correlations between BF4− anions and TEA+ cations (as schematically shown in

the “stage 1”). An opposite phenomenon showed up at the charging voltages of 1.0 V~2.5 V. In such a voltage range, the charge storage process is dominated by the adsorption of BF4− anions.

At the same time, a number of TEA+ cations were taken into the graphene nano-channels by BF4− anions upon charging because of the strongly ion-ion correlation electrolyte ions (as

schematically shown in the “stage 2”). The charge storage in negative electrodes has the similar process with the increasing charging voltage (Supplementary Information, Fig. S3). Two

distinct charging stages were observed on the graphene-film based EDLCs charged at voltage values varying from 0 to 2.5 V. At low charging voltages (0~1.0 V), the expulsion of co-ions (TEA+

cations for positive electrodes, BF4− anions for negative electrodes) played an important role in the charge storage process. In this stage, the co-ions were expelled from graphene

nano-channels with the increasing of charging voltages. At a higher charging voltage (1.0~2.5 V), the charge storage process was dominated by the adsorption of counter-ions (BF4− anions for

positive electrodes, TEA+ cations for negative electrodes). Large numbers of counter-ions were adsorbed in to the graphene channels to balance the electronic charge that accumulated in the

graphene surface. During the charging process, the abnormal decrease or increase in the number of counter-ions or co-ions within graphene channels arised from the strongly ion-ion

correlation between cations and anions. CONCLUSIONS In summary, 11B MAS-NMR measurements were performed on graphene-film based EDLCs. In neutral graphene films, the anions were

preferentially adsorbed on graphene surfaces at relatively low electrolyte loading volumes (<50 μL). The ion population of weakly and strongly adsorbed anions increased as a function of

electrolyte loadings, and partial desolvation of ion solvation shells was found at higher loading volumes (>100 μL). Unlike previous NMR studies on other porous carbons and CNTs, the

number of strongly adsorbed anions was much more than that of the weakly adsorbed anions at each certain electrolyte loading, mainly due to the unique two-dimensional layered structure of

graphene films. The NMR experiments carried out at different voltages showed that the charging process can be divided into two distinct charge storage stages for different voltage ranges

(_i.e._, ejection of co-ions at relatively low voltages and adsorption of counter-ions at higher voltages). The abnormal decrease or increase in the number of counter- or co-ions (_e.g._,

the abnormal decrease in the number of anions in the positive electrodes between 0 and 1.0 V) could be attributed to the ion-ion correlation between cations and anions. The results obtained

in the current NMR measurements provide useful insights for advancing the optimization of graphene-film based EDLCs. METHODS PREPARATION OF GRAPHENE FILM Graphite oxide (GO) powder (250 mg)

synthesized by modified Hummer’s method was dispersed in deionized water (1 L) and ultrasonicated (FB15150, 300 W, Fisher Scientific) for 1.5 h. The resulting GO dispersion was then refluxed

in a 95 °C oil bath for 2 h with 4 mL ammonia solution (ca. 25~28 wt% in water, Sinopharm Chemical Reagent Co. Ltd) and 206 μL hydrazine hydrate (85 wt% in water, Sinopharm Chemical Reagent

Co. Ltd). Graphene film was fabricated by vacuum filtration of the resulting dispersion through a membrane filter of 0.22 μm in pore size, and dried in desiccator for a week before

transferred to glovebox. PREPARATION OF NMR SAMPLES In an argon glovebox, a series amounts (50, 75, 100, 125, 150, 200 μL) of 1 M tetraethylammonium tetrafluoroborate (TEABF4, 99.0%,

Sigma-aldrich Co.) in acetonitrile (ACN, 99.8%, Sigma-aldrich Co.) electrolyte were added into as-prepared graphene films (60 mg) and soaked for 12 h in an airtight container. Then these

samples were taken out of the container and dried in argon environment for 10 min. Graphene-film based EDLCs were assembled in a two electrode system with 1 M TEABF4/ACN electrolyte. In

order to eliminate the effect of irreversible changes that could take place at the first few charging cycles, the EDLCs were cycled by galvanostatic charge-discharge cycle for 30 mins with

charge/discharge rate of 1 A g−1. After that, the EDLCs were held at the desired constant voltages (_i.e_., 0, 0.5, 1.0, 1.5, 2.0, 2.5 V) for 30 min and then disassembled quickly within 1 

min. Then the graphene films electrode materials were separated from the current collector and dried under argon environment for 10 mins. The electrochemical measurements were performed on

an electrochemical workstation (PGSTAT302N, Metrohm Autolab B.V) at room temperature. NMR EXPERIMENTS Slid-state NMR experiments were performed on a Bruker AVANCE III 400 MHz spectrometer

operating at 128.42 MHz for 11B, and equipped with a double-resonance magic-angle spinning (MAS) probe, supporting MAS rotors of 3.2 mm outer diameter. 11B Slid-state NMR spectra were

recorded at MAS frequency of 10 kHz, employing the direct excitation method with rf-nutaion frequency of 140 kHz for 11B. 11B chemical shift was referenced to NaH4B at −42.06 ppm. A recycle

delay of 5 s was used to collect a total of 64 scans for each sample. NMR spectra with a spectral width of 608.37 ppm (78125.00 Hz) were recorded in all cases. To ensure the experimental

parameters were proper in the current work, the spin-lattice (T1) relaxation time for 11B were measured with an inversion-recovery pulse sequence in separate experiments (see Fig. S4 in

Supplementary Information for more details). All NMR spectra were fitted with mixed Gaussian/Lorentzian function lineshapes using the DMFIT software53. In order to estimate the errors caused

by deconvolution, fits were repeated up to four times for each series of data34. The standard errors were calculated by the following equation: where _n_ (=4) refers to the number of data

points for each case, _x__i_ is the value of these data point, and represents the average values of _x__i_. ADDITIONAL INFORMATION HOW TO CITE THIS ARTICLE: Li, K. _et al_. Solid-state NMR

Study of Ion Adsorption and Charge Storage in Graphene Film Supercapacitor Electrodes. _Sci. Rep._ 6, 39689; doi: 10.1038/srep39689 (2016). PUBLISHER'S NOTE: Springer Nature remains

neutral with regard to jurisdictional claims in published maps and institutional affiliations. REFERENCES * Miller, J. R. & Simon, P. Materials science. Electrochemical capacitors for

energy management. Science 321, 651–652 (2008). Article  CAS  PubMed  Google Scholar  * Zhang, L. L. & Zhao, X. S. Carbon-based materials as supercapacitor electrodes. Chem. Soc. Rev.

38, 2520–2531 (2009). Article  CAS  PubMed  Google Scholar  * Simon, P. & Gogotsi, Y. Materials for electrochemical capacitors. Nature Mater. 7, 845–854 (2008). Article  CAS  ADS  Google

Scholar  * Bo, Z. et al. Emerging energy and environmental applications of vertically-oriented graphenes. Chem. Soc. Rev. 44, 2108–2121 (2015). Article  CAS  PubMed  Google Scholar  *

Chmiola, J. et al. Anomalous increase in carbon capacitance at pore sizes less than 1 nanometer. Science 313, 1760–1763 (2006). Article  CAS  ADS  PubMed  Google Scholar  * Kondrat, S., Wu,

P., Qiao, R. & Kornyshev, A. A. Accelerating charging dynamics in subnanometre pores. Nature Mater. 13, 387–393 (2014). Article  CAS  ADS  Google Scholar  * Huang, J. S., Sumpter, B. G.

& Meunier, V. A universal model for nanoporous carbon supercapacitors applicable to diverse pore regimes, carbon materials, and electrolytes. Chem. Eur. J 14, 6614–6626 (2008). Article 

CAS  PubMed  Google Scholar  * Forse, A. C., Merlet, C., Griffin, J. M. & Grey, C. P. New Perspectives on the Charging Mechanisms of Supercapacitors. J. Am. Chem. Soc. 138, 5731–5744

(2016). Article  CAS  PubMed  PubMed Central  Google Scholar  * Ilott, A. J., Trease, N. M., Grey, C. P. & Jerschow, A. Multinuclear _in situ_ magnetic resonance imaging of

electrochemical double-layer capacitors. Nat. Commun. 5, 4536 (2014). Article  CAS  ADS  PubMed  Google Scholar  * Bo, Z. et al. Molecular Insights into Aqueous NaCl Electrolytes Confined

within Vertically-oriented Graphenes. Sci. Rep. 5, 14652 (2015). Article  CAS  ADS  PubMed  PubMed Central  Google Scholar  * Levi, M. D., Salitra, G., Levy, N., Aurbach, D. & Maier, J.

Application of a quartz-crystal microbalance to measure ionic fluxes in microporous carbons for energy storage. Nature Mater. 8, 872–875 (2009). Article  CAS  ADS  Google Scholar  * Richey,

F. W., Dyatkin, B., Gogotsi, Y. & Elabd, Y. A. Ion dynamics in porous carbon electrodes in supercapacitors using _in situ_ infrared spectroelectrochemistry. J. Am. Chem. Soc. 135,

12818–12826 (2013). Article  CAS  PubMed  Google Scholar  * Wang, S., Li, S., Cao, Z. & Yan, T. Molecular dynamic simulations of ionic liquids at graphite surface. J. Phys. Chem. C 114,

990–995 (2009). Article  CAS  Google Scholar  * 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][PF(6)]. Phys. Chem. Chem. Phys. 11, 5584–5590 (2009). Article  CAS  PubMed  Google Scholar  * Merlet, C. et al. On the molecular origin of supercapacitance in

nanoporous carbon electrodes. Nature Mater. 11, 306–310 (2012). Article  CAS  ADS  Google Scholar  * Levi, M. D. et al. Electrochemical quartz crystal microbalance (EQCM) studies of ions and

solvents insertion into highly porous activated carbons. J. Am. Chem. Soc. 132, 13220–13222 (2010). Article  CAS  PubMed  Google Scholar  * Griffin, J. M., Forse, A. C. & Grey, C. P.

Solid-state NMR studies of supercapacitors. Solid State Nucl. Magn. Reson. 74–75, 16–35 (2016). Article  CAS  PubMed  Google Scholar  * Forse, A. C., Griffin, J. M., Presser, V., Gogotsi, Y.

& Grey, C. P. Ring Current Effects: Factors Affecting the NMR Chemical Shift of Molecules Adsorbed on Porous Carbons. J. Phys. Chem. C 118, 7508–7514 (2014). Article  CAS  Google

Scholar  * Lazzeretti, P. Ring currents. Prog. Nucl. Magn. Reson. Spectrosc. 36, 1–88 (2000). Article  CAS  Google Scholar  * Wang, H. et al. Real-time NMR studies of electrochemical

double-layer capacitors. J. Am. Chem. Soc. 133, 19270–19273 (2011). Article  CAS  PubMed  Google Scholar  * Harris, R. K., Thompson, T. V., Norman, P. R., Pottage, C. & Trethewey, A. N.

High-resolution 2H solid-state NMR of 2H2O adsorbed onto activated carbon. J. Chem. Soc., Faraday Trans. 91, 1795–1799 (1995). Article  CAS  Google Scholar  * Harris, R. K., Thompson, T. V.,

Norman, P. R. & Pottage, C. Adsorption competition onto activated carbon, studied by magic-angle spinning NMR. J. Chem. Soc., Faraday Trans. 92, 2615–2618 (1996). Article  CAS  Google

Scholar  * Harris, R. K., Thompson, T. V., Norman, P. R. & Pottage, C. Phosphorus-31 NMR studies of adsorption onto activated carbon. Carbon 37, 1425–1430 (1999). Article  CAS  Google

Scholar  * Dickinson, L. M., Harris, R. K., Shaw, J. A., Chinn, M. & Norman, P. R. Oxygen-17 and deuterium NMR investigation into the adsorption of water on activated carbon. Magn.

Reson. Chem. 38, 918–924 (2000). Article  CAS  Google Scholar  * Sekhaneh, W., Kotecha, M., Dettlaff-Weglikowska, U. & Veeman, W. S. High resolution NMR of water absorbed in single-wall

carbon nanotubes. Chem. Phys. Lett. 428, 143–147 (2006). Article  CAS  ADS  Google Scholar  * Chen, Q. et al. Identification of endohedral water in single-walled carbon nanotubes by H-1 NMR.

Nano Lett. 8, 1902–1905 (2008). Article  CAS  ADS  PubMed  Google Scholar  * Shen, K. & Pietrass, T. H-1 and H-2 NMR of hydrogen adsorption on carbon nanotubes. J. Phys. Chem. B 108,

9937–9942 (2004). Article  CAS  Google Scholar  * Liu, X. et al. NMR Study of Preferential Endohedral Adsorption of Methanol in Multiwalled Carbon Nanotubes. J. Phys. Chem. C 116, 7803–7809

(2012). Article  CAS  Google Scholar  * Forse, A. C. et al. Nuclear magnetic resonance study of ion adsorption on microporous carbide-derived carbon. Phys. Chem. Chem. Phys. 15, 7722–7730

(2013). Article  CAS  PubMed  Google Scholar  * Borchardt, L., Oschatz, M., Paasch, S., Kaskel, S. & Brunner, E. Interaction of electrolyte molecules with carbon materials of

well-defined porosity: characterization by solid-state NMR spectroscopy. Phys. Chem. Chem. Phys. 15, 15177–15184 (2013). Article  CAS  PubMed  Google Scholar  * Deschamps, M. et al.

Exploring electrolyte organization in supercapacitor electrodes with solid-state NMR. Nature Mater. 12, 351–358 (2013). Article  CAS  ADS  Google Scholar  * Wang, H. et al. _In situ_ NMR

spectroscopy of supercapacitors: insight into the charge storage mechanism. J. Am. Chem. Soc. 135, 18968–18980 (2013). Article  CAS  PubMed  PubMed Central  Google Scholar  * Griffin, J. M.

et al. Ion counting in supercapacitor electrodes using NMR spectroscopy. Faraday Discuss. 176, 49–68 (2014). Article  CAS  ADS  PubMed  Google Scholar  * Griffin, J. M. et al. _In situ_ NMR

and electrochemical quartz crystal microbalance techniques reveal the structure of the electrical double layer in supercapacitors. Nature Mater. 14, 812–819 (2015). Article  CAS  ADS  Google

Scholar  * Luo, Z. X. et al. Dehydration of Ions in Voltage-Gated Carbon Nanopores Observed by _in Situ_ NMR. J. Phys. Chem. Lett. 6, 5022–5026 (2015). Article  CAS  PubMed  Google Scholar

  * Luo, Z. X., Xing, Y. Z., Ling, Y. C., Kleinhammes, A. & Wu, Y. Electroneutrality breakdown and specific ion effects in nanoconfined aqueous electrolytes observed by NMR. Nat. Commun.

6, 6358 (2015). Article  CAS  ADS  PubMed  Google Scholar  * Lee, S. I. et al. B-11 NMR study of the BF4- anion in activated carbons at various stages of charge of EDLCs in organic

electrolyte. Carbon 44, 2578–2586 (2006). Article  CAS  Google Scholar  * Azais, P. et al. Causes of supercapacitors ageing in organic electrolyte. J. Power Sources 171, 1046–1053 (2007).

Article  CAS  ADS  Google Scholar  * Forse, A. C. et al. NMR Study of Ion Dynamics and Charge Storage in Ionic Liquid Supercapacitors. J. Am. Chem. Soc. 137, 7231–7242 (2015). Article  CAS 

PubMed  PubMed Central  Google Scholar  * Geim, A. K. Graphene: status and prospects. Science 324, 1530–1534 (2009). Article  CAS  ADS  PubMed  Google Scholar  * Zhang, L. L., Zhou, R. &

Zhao, X. S. Graphene-based materials as supercapacitor electrodes. J. Mater. Chem. 20, 5983–5992 (2010). Article  CAS  Google Scholar  * Gwon, H. et al. Flexible energy storage devices

based on graphene paper. Energ. Environ. Sci. 4, 1277–1283 (2011). Article  CAS  Google Scholar  * Wang, G. K. et al. Flexible Pillared Graphene-Paper Electrodes for High-Performance

Electrochemical Supercapacitors. Small 8, 452–459 (2012). Article  CAS  PubMed  Google Scholar  * Yang, H. et al. Edge effects in vertically-oriented graphene based electric double-layer

capacitors. J. Power Sources 324, 309–316 (2016). Article  CAS  ADS  Google Scholar  * Feng, G. et al. The importance of ion size and electrode curvature on electrical double layers in ionic

liquids. Phys. Chem. Chem. Phys. 13, 1152–1161 (2011). Article  CAS  PubMed  Google Scholar  * Vatamanu, J., Borodin, O. & Smith, G. D. Molecular Insights into the Potential and

Temperature Dependences of the Differential Capacitance of a Room-Temperature Ionic Liquid at Graphite Electrodes. J. Amer. Chem. Soc. 132, 14825–14833 (2010). Article  CAS  Google Scholar 

* Anderson, R. J. et al. NMR methods for characterizing the pore structures and hydrogen storage properties of microporous carbons. J. Am. Chem. Soc. 132, 8618–8626 (2010). Article  CAS 

PubMed  Google Scholar  * Xing, Y. Z., Luo, Z. X., Kleinhammes, A. & Wu, Y. Probing carbon micropore size distribution by nucleus independent chemical shift. Carbon 77, 1132–1139 (2014).

Article  CAS  Google Scholar  * Lian, P. C. et al. Large reversible capacity of high quality graphene sheets as an anode material for lithium-ion batteries. Electrochim. Acta 55, 3909–3914

(2010). Article  CAS  Google Scholar  * Gao, Y. J., Ma, D., Wang, C. L., Guan, J. & Bao, X. H. Reduced graphene oxide as a catalyst for hydrogenation of nitrobenzene at room temperature.

Chem. Commun. 47, 2432–2434 (2011). Article  CAS  Google Scholar  * Chmiola, J., Largeot, C., Taberna, P. L., Simon, P. & Gogotsi, Y. Desolvation of ions in subnanometer pores and its

effect on capacitance and double-layer theory. Angew. Chem. Int. Ed. Engl. 47, 3392–3395 (2008). Article  CAS  PubMed  Google Scholar  * Conway, J. H. & Sloane, N. J. A. Sphere packings,

lattices and groups. Vol. 290 (Springer Science & Business Media, 2013). * Massiot, D. et al. Modelling one- and two-dimensional solid-state NMR spectra. Magn. Reson. Chem. 40, 70–76

(2002). Article  CAS  Google Scholar  Download references ACKNOWLEDGEMENTS Financial support for this work was provided by the National Natural Science Foundation of China (No. 51306159) and

the Foundation of National Excellent Doctoral Dissertation of China (No. 201238). AUTHOR INFORMATION AUTHORS AND AFFILIATIONS * State Key Laboratory of Clean Energy Utilization, Institute

for Thermal Power Engineering, College of Energy Engineering, Zhejiang University, Hangzhou, 310027, Zhejiang Province, China Kecheng Li, Zheng Bo, Jianhua Yan & Kefa Cen Authors *

Kecheng Li View author publications You can also search for this author inPubMed Google Scholar * Zheng Bo View author publications You can also search for this author inPubMed Google

Scholar * Jianhua Yan View author publications You can also search for this author inPubMed Google Scholar * Kefa Cen View author publications You can also search for this author inPubMed 

Google Scholar CONTRIBUTIONS Z.B. and K.C. designed this research; K.L. carried out experiments; K.L., Z.B., and J.Y. contributed to the analysis and discussion of the data; K.L and Z.B.

drafted the manuscript; and all authors commented on the final manuscript. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing financial interests. ELECTRONIC

SUPPLEMENTARY MATERIAL SUPPLEMENTARY INFORMATION RIGHTS AND PERMISSIONS This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party

material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons

license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ Reprints

and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Li, K., Bo, Z., Yan, J. _et al._ Solid-state NMR Study of Ion Adsorption and Charge Storage in Graphene Film Supercapacitor Electrodes.

_Sci Rep_ 6, 39689 (2016). https://doi.org/10.1038/srep39689 Download citation * Received: 28 July 2016 * Accepted: 25 November 2016 * Published: 21 December 2016 * DOI:

https://doi.org/10.1038/srep39689 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently

available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative