- Email: [email protected]
Development of asymmetric supercapacitors with titanium carbide-reduced graphene oxide couples as electrodes
Accepted Manuscript Development of asymmetric supercapacitors with titanium carbide-reduced graphene oxide couples as electrodes Adriana M. Navarro-Suárez, Katherine L. Van Aken, Tyler Mathis, Taron Makaryan, Jun Yan, Javier Carretero-González, Teófilo Rojo, Yury Gogotsi PII:
S0013-4686(17)32247-8
DOI:
10.1016/j.electacta.2017.10.125
Reference:
EA 30511
To appear in:
Electrochimica Acta
Received Date: 27 June 2017 Revised Date:
11 October 2017
Accepted Date: 18 October 2017
Please cite this article as: A.M. Navarro-Suárez, K.L. Van Aken, T. Mathis, T. Makaryan, J. Yan, J. Carretero-González, Teó. Rojo, Y. Gogotsi, Development of asymmetric supercapacitors with titanium carbide-reduced graphene oxide couples as electrodes, Electrochimica Acta (2017), doi: 10.1016/ j.electacta.2017.10.125. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Development of Asymmetric Supercapacitors with Titanium CarbideReduced Graphene Oxide Couples as Electrodes Adriana M. Navarro-Suáreza,b,†, Katherine L. Van Akena, Tyler Mathisa, Taron Makaryana, Jun Yana,c, Javier Carretero-Gonzálezd, Teófilo Rojob,e, and Yury Gogotsi*,a a
RI PT
Department of Materials Science & Engineering and A.J. Drexel Nanomaterials Institute, Drexel University, Philadelphia, PA 19104, USA.
b
CIC Energigune, Albert Einstein 48, 01510 Miñano, Alava, Spain.
c
SC
Key Laboratory of Superlight Materials and Surface Technology, Ministry of Education, College of Material Science and Chemical Engineering, Harbin Engineering University, Harbin 150001, China. d
e
M AN U
Institute of Polymer Science and Technology, ICTP-CSIC, Juan de la Cierva 3, 28006, Madrid, Spain. Inorganic Chemistry Department, P.O. Box 644, University of the Basque Country, 48080 Bilbao, Spain.
†
Current address: Department of Physics, Chalmers Univ. of Technol., 41296 Gothenburg, Sweden
Abstract
TE D
*Email: [email protected]
AC C
EP
Two-dimensional (2D) nanomaterials have attracted significant interest for supercapacitor applications due to their high surface to volume ratio. Layered 2D materials have the ability to intercalate ions and thus can provide intercalation pseudocapacitance. Properties such as achieving fast ion diffusion kinetics and maximizing the exposure of the electrolyte to the surface of the active material are critical for optimizing the performance of active materials for electrochemical capacitors (i.e. Supercapacitors). In this study, two 2D materials, titanium carbide (Ti3C2Tx) and reduced graphene oxide (rGO), were used as electrode materials for asymmetric supercapacitors, with the resulting devices achieving high capacitance values and excellent capacitance retention in both aqueous and organic electrolytes. This work demonstrates that Ti3C2Tx is a promising electrode material for flexible and high-performance energy storage devices.
Keywords: Supercapacitors; MXene; graphene; volumetric capacitance; 2D materials.
1
ACCEPTED MANUSCRIPT
1. Introduction
M AN U
SC
RI PT
Developing electrochemical energy storage devices with high energy densities, rapid charge/discharge capabilities, and long cycle lifetimes has been a subject of extensive research in the past decade for portable electronics and hybrid electric vehicles. [1–4] For instance, the ability to design miniaturized supercapacitors for applications such as on-chip energy storage, wearable electronics, etc., demands light and flexible materials with high volumetric capacitances.[5] However, the incorporation of supercapacitors into these applications requires improvement of the device’s energy density.[6] In the case of supercapacitors, which operate via physical adsorption of ions, this issue can be addressed by increasing the material’s active surface area or enlarging the operational voltage window.[7] For the latter, asymmetric devices have been proposed in order to advantageously utilize the operating voltage ranges of two different materials such as manganese dioxide (MnO2), graphene, or multi-walled carbon nanotubes (MWCNTs).[8] As for the former issue, incorporating pillar-like nanostructures for opening up densely packed two-dimensional (2D) nanomaterials has been explored as a possible route.[9]
EP
TE D
2D materials are inherently flexible, have high surface reactivity, high electrical conductivity, and large surface areas, which are attractive characteristics for materials for energy storage devices.[10] The exfoliation of graphite into single graphene layers and the subsequent research into graphene’s excellent physical and electronic properties in 2004 resulted in increased interest in 2D materials.[11] A recent addition to the family of 2D materials are MXenes, a large group (20+ members to date) of 2D transition metal carbides, nitrides, and carbonitrides that were discovered in 2011.[12] MXenes have a general formula of , where is an early transition metal (e.g. Ti, Nb), can be carbon or nitrogen, being equal to 1, 2 or 3, and represents surface functional terminations such as –OH, -O, or -F.
AC C
MXenes have shown great potential as electrode materials in supercapacitors.[13–21] Pristine, delaminated Ti3C2Tx is the most studied MXene and has achieved capacitance values up to 238 F.g-1 / 900 F.cm-3 in acidic electrolytes. [14,21] Mixed with reduced graphene oxide (rGO) in a symmetric cell configuration, Ti3C2Tx has demonstrated an outstanding capacitance of 154 F.g-1 at 2 A.g-1.[22] One disadvantage of Ti3C2Tx is that its potential window in aqueous electrolytes is narrow (less than 1 V). This translates to lower energy density when compared to carbon materials.[23] Most studies so far focused on Ti3C2Tx electrodes testing in 3-electrode cells,[17,20,24,25] which does not allow evaluation of the operating potential of devices with MXene electrodes or calculating energy and power density of the devices.
2
ACCEPTED MANUSCRIPT
SC
RI PT
In order to further increase the energy density of Ti3C2Tx-based supercapacitors, we have developed a full cell using Ti3C2Tx and rGO as the negative and positive electrodes, respectively, and tested the devices in both aqueous and organic electrolytes. By using an asymmetric device, the total electrochemical window of the supercapacitor is enlarged, resulting in increased energy and power densities. We have also studied the effect of disrupting the stacking of Ti3C2Tx layers by hybridizing MXene with MWCNTs, which increases the interlayer distance, resulting in better ion accessibility and transport for organic electrolytes. Through optimizing the physical properties of the active materials to match the electrolyte system, we achieved high capacitance for the Ti3C2Tx/MWCNT-rGO asymmetric device in an organic electrolyte. This strategy of matching and optimizing active materials to the electrolyte being used here can be applied to the development of other energy storage devices.
2.2 Materials Synthesis
M AN U
2. Experimental
2.2.1 Synthesis of 2D Titanium Carbide MXene
AC C
EP
TE D
The complete synthesis method for titanium carbide MXene is described elsewhere.[14] Briefly, to synthesize Ti3C2Tx, 20 mL of 9 M hydrochloric acid (HCl, Fisher Scientific) were added to 7.5 molar equivalents (2 g) of lithium fluoride (LiF, Alfa Aesar). The mixture was stirred until the salt was dissolved. Then, 2 g of the ternary titanium aluminum carbide powder (Ti3AlC2, < 38 µm particle size) were slowly added to the solution. The reaction mixture was held at 35 °C for 24 h while stirring a t 200 rpm. After 24 h, the mixture was washed with deionized water and separated by centrifugation at 3500 rpm, after which the supernatant was collected. The washing process was repeated until the pH of the supernatant became neutral (pH 6-7). In order to increase the surface area, the Ti3C2Tx solution was delaminated by bath sonication for 1 h under argon flow. Afterwards, the sonicated solution was centrifuged for another hour at 3500 rpm. Finally, the supernatant containing the Ti3C2Tx was collected. In order to prepare a flexible, free-standing Ti3C2Tx film, the solution was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum. The Ti3C2Tx/multi-walled carbon nanotube (MWCNT) composite was prepared as described elsewhere.[26] First, an aqueous 1 mg.mL-1 Ti3C2Tx suspension was prepared as well as an aqueous solution of 0.1 mg.mL-1 MWCNT in 8 mg.mL-1 aqueous solution of sodium dodecyl sulfate (SDS, Sigma Aldrich). Sandwich-like MXene/MWCNT films were prepared using an alternating filtration method. Specifically, 1 mL of the Ti3C2Tx dispersion was filtered through a polypropylene membrane to yield a thin Ti3C2Tx layer. Then, 1 mL of the MWCNT-SDS dispersion was filtered on top of the Ti3C2Tx layer. This alternating filtration was repeated several times to yield composite films composed of 5 and 4 3
ACCEPTED MANUSCRIPT
alternating Ti3C2Tx and MWCNTs layers, respectively. Afterwards, the film was washed with 200 milliliters of deionized water. The composite film was dried under vacuum at 120 °C. 2.2.2 Synthesis of Reduced Graphene Oxide
TE D
2.2.3 Hydrazine Reduction Method
M AN U
SC
RI PT
The graphite oxide was prepared following the Marcano-Tour method.[27] A 9:1 mixture of concentrated H2SO4/H3PO4 (360:40 mL) was added to a mixture of graphite powder (3.0 g, 1 wt. equiv.) and KMnO4 (18.0 g, 6 wt. equiv.). The reaction mixture was then heated to 50 °C and stirred overnight. The mixture was then c ooled to room temperature and poured onto ice (~400 mL) with 30 % H2O2 (3 mL). The solution was centrifuged (10000 rpm for 30 min), and the supernatant was decanted away. The remaining solid material was then washed in succession with 200 mL of water, 200 mL of 30 % HCl, and then water until a neutral pH was achieved. For each wash, the mixture was centrifuged (10000 rpm for 30 min) and the supernatant was decanted away. The graphite oxide was then sonicated in water for 1 hour and the exfoliated graphene oxide (GO) was collected. Thereafter, two different reduction methods were used. The first one was a chemical method that yields reduced graphene oxide (rGO) with carboxylic acid groups.[28] These groups can be beneficial for pseudocapacitance in aqueous electrolytes.[29] In the second method, the graphene oxide was reduced thermally, eliminating the oxygen functional groups.[30] Oxygen containing functionalities are known for being detrimental to capacitance in organic-based supercapacitors.[31]
AC C
EP
GO was reduced following the Li-Wallace’s method. [32] Briefly, 30 mL of GO in water (7 mg.mL-1) was heated to 85 °C. Then 2 mL of hydrazine hydra te (98 %, TCI America) and 1 mL of ammonium hydroxide (Fischer Scientific) were added to the GO. The reaction was kept under reflux during one hour, after which the solution was left to cool and centrifuged for 30 minutes; the reduced graphene oxide (rGOH) was then re-dispersed in ethanol (Decon Labs, Inc). In this case, hydrazine monohydrate acts as the reducing agent, and ammonium hydroxide is used to promote the colloidal stability of the GO sheets through electrostatic repulsion.[33] In order to prepare a free-standing film, the rGOH was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum. 2.2.4 Thermal Reduction Method The GO solution was freeze dried to produce a GO aerogel. Then, 200 mg of the freeze dried-GO was heated under argon flow at a rate of 10 °C.min -1 up to 900 °C and held there for 2 hours. After cooling, the reduced graphene oxide (rGOT) was re-dispersed in ethanol. Then, the rGOT was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum.
4
ACCEPTED MANUSCRIPT
2.3 Characterization Methods The characterization of the Ti3C2Tx, Ti3C2Tx-MWCNT, and rGO film morphologies was carried out using scanning electron microscopy (SEM, Zeiss Supra 50VP, Germany).
RI PT
The vibration frequencies of the materials were studied by Raman Spectroscopy. The Raman spectra were recorded on an inverted Renishaw inVia spectrometer equipped with a 60x objective, with a 632 nm laser as an excitation source (<1 mW laser power). The back-scattered light was reflected to a room-temperature operating detector after a 1200 lines/mm diffraction grating.
M AN U
2.3.1 Electrochemical Testing
SC
The interlayer distance of the Ti3C2Tx, Ti3C2Tx-MWCNT, and rGO films was calculated using X-Ray Diffraction (XRD). XRD was carried out on a Rigaku Smart Lab (Japan) diffractometer using Cu-Kα radiation (40 kV and 44 mA), a step scan of 0.02 °, 2θ range of 5 – 50 °, and a step time of 0.5 s.
AC C
EP
TE D
To quantify the amount of charge stored by the prepared materials, all tested films were cut to have the same dimensions and weight before being assembled in Swagelok-type cells. The materials were tested in a 3-electrode configuration to evaluate their capacitance and electrochemical stability window. Then, symmetric and asymmetric devices were assembled for further electrochemical analysis of the device performance. The cells were electrochemically studied by using a 1 M sulfuric acid (H2SO4, Sigma Aldrich) solution with a 3.5 M aqueous silver/silver chloride (Ag/AgCl) electrode as the reference. In the case of the organic electrolyte, 1 M tetraethyl ammonium tetrafluoroborate (Et4N+BF4-, Sigma Aldrich) in acetonitrile (ACN, Fischer) was used as the electrolyte and Ag wire as the reference or pseudo-reference electrode. For the 3electrode cell, overcapacitve activated carbon (YP-50, Kuraray) was used as the counter electrode. Upon assembly, all the cells were precycled between 50 and 200 times to ensure the electrolyte had permeated throughout the porous structure and the cell had stabilized. Cyclic voltammetry (CV) measurements at different scan rates ranging from 2 to 100 mV.s-1, as well as impedance measurements were performed at ambient conditions in a multichannel potentiostat/galvanostat (Biologic VMP3, France). The cycling performance of the asymmetric devices was tested at 20 mV.s-1 over 1000 cycles from 0 to 1.1 V and from 0 to 2 V in 1 M H2SO4 and 1 M Et4NBF4/ACN, respectively. Before and after the cycling test, electrochemical impedance spectroscopy (EIS) was measured within the ac frequency region from 10 mHz to 200 kHz with ac voltage amplitude of 10 mV at open circuit voltage. Capacitance values per electrode are typically reported, which is valid for symmetric cells and 3-electrode characterization. However, for asymmetric devices, these values are not accurate enough as each electrode has a different mass.[34] In this article, we will refer to capacitance when the electrode capacitance was calculated and to cell capacitance when referring to the capacitance of the whole device using both electrode mass. Cell capacitance is calculated using Eq. 1:
5
ACCEPTED MANUSCRIPT
Ccell =
∫ idt
Eq. 1
( m+ + m− ) ∆V
RI PT
where Ccell is the cell capacitance, i is the discharge current, ∆V is the voltage window, and m+ and m- correspond to the mass of the positive and negative electrode, respectively.[35] Ragone plots are calculated from the galvanostatic charge‒discharge curves. The gravimetric energy (Egrav) and power (Pgrav) corresponding to the mass of the active material per electrode are estimated by using the following equations respectively:
∫ iVdt ( m+ + m− )
Eq. 3
M AN U
Pgrav = Egrav / t
Eq. 2
SC
Egrav =
The volumetric capacitance, energy and power were calculated by considering the density of each electrode.
3. Results and Discussion 3.1 Materials Characterization
TE D
3.1.1 2D Titanium Carbide MXene and 2D Titanium Carbide MXene-MWCNT Composite
AC C
EP
During the synthesis of Ti3C2Tx from Ti3AlC2, the HCl and LiF dissociate and H+ and F- ions weaken the Ti-Al bonds. This results in an opening of the interlayer spacing and allows for further insertion that leads to the formation of AlF3 and H2,, which are removed during washing. The 2D Ti3C2Tx layers possess two exposed Ti atoms per unit formula that have surface functionalities such as hydroxyl and fluorine groups, which are present in the reaction media.[36] Recently, Dall’Agnese et al. proved that the addition of MWCNT improves the performance of Ti3C2Tx in organic electrolytes.[17] After delamination of the MXene, the MWCNT are added according to the procedure explained elsewhere.[18] In this case, the mass ratio of Ti3C2Tx to MWCNT in the film was 92:8. The insertion of MWCNT between the layers of d-Ti3C2Tx decreases the density of the MXene electrode film from 4.07 to 2.44 g.cm-3. In order to confirm the synthesis of Ti3C2Tx, to study its structure and the influence of the MWCNT on the Ti3C2Tx morphology, the films were characterized by SEM, Raman spectroscopy, and X-Ray diffraction. Figure 1(a) shows the cross-sectional scanning electron microscopy (SEM) image of a binder-free Ti3C2Tx film. This image shows that the entire film is composed of well-aligned stacked Ti3C2Tx MXene sheets. This layered morphology is similar to that of expanded graphite, as the nanolayers are clearly separated from each other. This shearing of the 2D particles allows the material to form 6
ACCEPTED MANUSCRIPT
flexible, free-standing films. Figure 1(b) shows a cross-sectional SEM image of a Ti3C2Tx/MWCNT composite. The anticipated increase in spacing between the Ti3C2Tx layers after the incorporation of the MWCNTs is evident in the image.
M AN U
SC
RI PT
Figure 1(c) shows the Raman spectrum of the Ti3C2Tx film. All the vibration bands are consistent with values reported in literature.[37] The four broad Raman peaks centered around 200, 369, 628 and 716 cm−1 are attributed to the vibrations from non-stoichiometric titanium carbide.[38,39] The Raman spectrum of the Ti3C2Tx/MWCNT film, shown in Figure 1(d), combines the features of both Ti3C2Tx and MWCNTs. The peaks observed at low frequencies correspond to the peaks attributed to Ti3C2Tx and are in agreement with the values of the MXene by itself. The (002) peak in the XRD pattern of d-Ti3C2 shifted from 7.1 ° to 6.4 ° for the sandwich-like Ti 3C2Tx/MWCNT film. This shift translates into an increase in the interlayer distance from 12.5 Å in the Ti3C2Tx to 13.8 Å in the Ti3C2Tx /MWCNT, which could be possibly attributed to the intercalation of surfactant molecules into Ti3C2Tx layers.[40] Ti3C2Tx does not have observable peaks at higher diffraction angles, yet the Ti3C2Tx/MWCNT film does. The peak at 1141.4 cm-1 corresponds to C-H vibration of the surfactant used during synthesis.[41] The peaks at 1295 and 1538.4 cm-1 correspond to the D and G bands of the MWCNTs, respectively.[42,43]
3.1.2 Chemically and Thermally Reduced Graphene Oxide
EP
TE D
The presence of surface functionalities, water, or impurities in the active material are the main causes of ageing in organic-based supercapacitors.[31] The hydrazine reduction of GO does not remove epoxide groups from the edges of the aromatic domains,[30] making the rGOH a poor candidate for organic-based supercapacitors. Thermal annealing of the GO above 700 °C eliminates hydroxyl and carboxyl gr oups from the material.[30] Herein, the GO is thermally reduced at 900 °C (labeled here as rGOT) in order to use it as an electrode material for organic-based supercapacitors.
AC C
To confirm the 2D structure of the rGOH and the rGOT films, the samples were also analyzed by SEM, Raman spectroscopy, and XRD. A cross section image of the rGOH and the rGOT films by SEM is shown in Figure 2(a) and 2(b). As with Ti3C2Tx, the reduced graphene oxide films have a layered structure. The Raman spectrum (Figure 2(c)) of both rGO films exhibit the typical fingerprint of carbon materials. The D band is around 1323 cm-1 while the G band is at 1598 cm-1, with an ID/IG ratio of 1.4 and 1.3. for rGOH and rGOT, respectively, indicating that the two reduction methods achieve similar rGO materials in terms of ordering of the 2D material. Figure 2(d) shows the X-Ray Diffraction of the rGOH and rGOT films. The materials show water-filled space between the layers ranging from 2.2 to 4.4 Å. The (002) peak of the rGOT is at 25.4 °, 1.7 ° to the right of the (002) peak of the rGOH. This shift towards higher angles indicates that rGOT is slightly less oxidized than rGOH. 7
ACCEPTED MANUSCRIPT
3.2 Electrochemical devices 3.2.1 Aqueous-Based Asymmetric Device
M AN U
SC
RI PT
According to the 3-electrode electrochemical studies of the Ti3C2Tx and the rGOH films, shown in Figures S1 and S2, the d-Ti3C2Tx film has high volumetric and gravimetric capacitances (230 F.g-1 and 934 F.cm-3 at 2 mV.s--1) as well as excellent capacitance retention (167 F.g-1 at 100 mV.s--1). It is important to mention that a peak at ~0.3 V corresponds to redox reaction of surface titanium atoms.[44] On the other hand, the rGOH film has a high gravimetric capacitance (251 F.g-1 at 2 mV.s--1) but its volumetric capacitance (248 F.cm-3 at 2 mV.s--1) and capacitance retention (129 F.g-1 at 100 mV.s--1) need to be improved. To compare the results from the asymmetric and symmetric devices, the capacitance values henceforth reported correspond to the cell capacitance. As these materials have complementary potential windows (from -0.2 to 0.35 V vs Ag/AgCl for Ti3C2Tx and from 0.0 to 0.9 V vs Ag/AgCl for rGOH), an asymmetric supercapacitor using these materials was constructed and tested. In order to assemble the asymmetric supercapacitor, the Ti3C2Tx film was used as the negative electrode and the rGOH film as the positive electrode. The amount of charge, Q, stored in each of the positive and negative electrodes in an asymmetrical supercapacitor must be the same and is governed by Eq. 4. Q = CSP+ * m+ * V+ = CSP- * m- * V-
Eq. 4
AC C
EP
TE D
Due to the asymmetry in the electrochemical behavior for the cations and anions, unequal specific capacitances for the positive (CSP+) and negative (CSP-) electrodes occur. Therefore, in order to maintain stable voltage conditions where V- = V+, the weight of the electrodes (m- and m+) should be unequal to compensate for the different specific capacitance values[45]. The mass balance was calculated using Eq. 5, taking into account the maximum capacitance of each material in a three-electrode setup with the same potential window.
m- = CSP+ * m+ / CSP-
Eq. 5
Then it follows that the ratio of rGOH to Ti3C2Tx is:
mrGOH / mTi3C2Tx = CTi3C2Tx / CrGOH = 230 / 251 = 0.9
Eq. 6
8
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
In an asymmetric device, the electrodes have dissimilar charge storage capacities; therefore, in this section the capacitance reported will be the one corresponding to the full cell. Figure 3 (a) shows the evolution of the cell capacitance with voltage for the asymmetric device at different scan rates, with mrGOH / mTi3C2Tx = 0.9. The capacitive behavior (rectangular shape of the voltammogram) is maintained at different scan rates indicating good electrical conductivity and ion diffusion inside the pores of the electrode materials. In the asymmetric device, the electrochemical window is expanded to 1.1 V as the electrodes have complementary working potential windows. Based on the evidence of peaks and faradaic reactions that started to become clear, the voltage window was not expanded further. The increase in the device’s operating voltage window will positively affect the energy and power delivered by it. Figure 3 (b) shows the contributions from each electrode when the cell is cycled at 10 mV.s-1. The potential windows corresponding to the rGOH and Ti3C2Tx were 0.71 and 0.39 V, respectively. To equilibrate the charges, other mass balance proportions were tried (not shown) but the Ti3C2Tx potential window was always limited to less than 0.4 V. The capacitance values achieved by the Ti3C2Tx electrode, and the rGOH electrode at 10 mV.s-1 are 195 and 114 F.g-1, respectively.
TE D
The cell specific capacitances and Ragone plots (calculated using the mass and dimensions of the active materials only) for the asymmetric and symmetric devices are shown in Figure 4. The asymmetric device exhibits intermediate capacitance retention between the symmetric cells of Ti3C2Tx and rGOH (Figure 4 (a and c)). The maximum cell capacitance achieved by the asymmetric device is 48 F.g-1 (78 F.cm-3). The use of Ti3C2Tx as a counterpart to rGOH in an asymmetric cell improves the capacitance retention (44 % at 10 A.g-1) as compared to that of the symmetric rGOH (33 % at 10 A.g-1) and expands the voltage window. However, the capacitance value is lower than the one achieved by the symmetric Ti3C2Tx device.
AC C
EP
The advantage of using rGOH as a counterpart to Ti3C2Tx, as explained before, is the expansion of the whole device’s operating potential window by compensating for the MXene electrode’s inherently moderate voltage window. In the Ragone plot (Figure 4(b and d)) the small voltage window causes the symmetric Ti3C2Tx device to present the lowest gravimetric energy values of the three devices. The symmetric rGOH device exhibits a larger energy density than the symmetric Ti3C2Tx device, but drops quickly as the power increases. As expected, the asymmetric device achieves the highest energy density owing to the enlarged voltage window, and the most stable energy density among the three devices. The asymmetric Ti3C2Tx/rGOH achieves up to 8 Wh.kg-1. The influence of the materials’ density can be observed in Figure 4 (c and d). As explained before, Ti3C2Tx is a denser material than rGOH. So, when comparing their volumetric capacitance retention, Ti3C2Tx exhibits the highest values, rGOH the lowest and the asymmetric device falls in between. This behavior is translated to the Ragone plot, where as in the gravimetric plot, the asymmetric device achieves the highest energy values. Nevertheless, in contrast with the gravimetric plot, Ti3C2Tx achieves larger energy values
9
ACCEPTED MANUSCRIPT
than rGOH, as its film density is four times higher. Table 1 summarizes the main results of the electrochemical testing of Ti3C2Tx and rGOH in aqueous (H2SO4) electrolyte in symmetric and asymmetric configurations.
RI PT
In order to further increase the energy density delivered by the asymmetric device, the materials were also tested in an organic electrolyte. However, as explained before the materials were slightly modified to improve their performance in organic electrolytes. 3.2.2 Organic Electrolyte Asymmetric Device
M AN U
SC
The assembly of an asymmetric supercapacitor with organic electrolyte is not as straightforward as the aqueous-based one. The potential windows of the Ti3C2Tx/MWCNT and rGOT are not complementary and rGOT has a better electrochemical performance in 1 M Et4NBF4/ACN than the composite (Figure S3). Nevertheless, the asymmetric supercapacitor was assembled to check whether the performance of the Ti3C2Tx/MWCNT could be improved by using rGOT as the positive electrode. rGOT was chosen as the positive electrode due to its larger potential window. Ti3C2Tx/MWCNT was chosen as the negative electrode as its larger interlayer spacing could accommodate Et4N+ ions. The mass balance was calculated using Eq. 5, taking into account the maximum capacitance values showed by each material in the 3-electrode configuration. The result is shown in Eq. 7. mrGOT / md-Ti3C2Tx/MWCNT = 42 / 71 = 0.6
Eq. 7
EP
TE D
Figure 5 shows the cyclic voltammograms of the asymmetric Ti3C2Tx/MWCNT and rGOT device using mrGOT / mTi3C2Tx/MWCNT = 0.6. The cell can operate at up to 2.3 V but the cell capacitance drops drastically when the scan rate is increased. Therefore, the asymmetric cells in this study were tested at 2 V (Figure 5 (a)). The performance of each electrode was studied with a pseudo-reference electrode (Figure 5 (b)). The potential window of rGOT is equal to 0.8 V and shows a capacitance of 83 F.g-1, while the Ti3C2Tx/MWCNT exhibits a potential window of 1.2 V and achieves 35 F.g-1.
AC C
Figure 6 (a and c) shows the cell capacitance values of the asymmetric device and the symmetric cells of Ti3C2Tx/MWCNT and rGOT. For these devices, the cell capacitance of the asymmetric device was the highest due to the interlayer distance of the active materials being suitable to accommodate the Et4N+ and BF4- ions. The XRD pattern indicated an interlayer distance for the Ti3C2Tx/MWCNT composite film of 1.38 nm and assuming the size of the unsolvated Et4N+ cation is 0.67 nm and the size of the solvated ion is 1.3 nm,[46] we can hypothesize that the similarity in size allows the insertion of ions in between the layers of the material. The rGOT has an interlayer distance ranging between 0.23 and 0.44 nm, which matches well with the solvated BF4- ion size of 0.33 nm.[47]
10
ACCEPTED MANUSCRIPT
RI PT
When taking into account the density of the materials (Figure 6 (c and d)), the symmetric cells achieve similar capacitance values at 2 mV.s-1. After increasing the scan rate above 5 mV.s-1, the symmetric rGOT retains 71 % of its initial capacitance while the symmetric Ti3C2Tx/MWCNT retains 84 %. Figure 6 shows the gravimetric (b) and volumetric (d) Ragone plots for the asymmetric devices in aqueous and organic electrolytes. As expected, the increase in voltage window results in higher energy density for the organic cell. Nevertheless, the aqueous electrolyte supercapacitor exhibits better energy retention owing to the higher conductivity of the aqueous electrolyte. Table 2 summarizes the performance of Ti3C2Tx-MWCNT and rGOT as electrodes for supercapacitors in organicbased electrolytes.
SC
3.3 Long-term Cycling Tests
M AN U
To assess the long-term cycling behavior of the MXene-rGO based supercapacitors, the asymmetric devices were tested for 1000 cycles and the evolution of their capacitance is plotted in Figure 7 (a) and 7 (c). After 1000 cycles at 20 mV.s-1, the capacitance retention of the asymmetric Ti3C2Tx/rGOH is 76 % of the initial value. The decrease in capacitance might be caused by the lower conductivity of the rGOH, as it has been proven that Ti3C2Tx can be cycled up to 10000 cycles without capacitance losses.[21] The coulombic efficiency for the device is close to 100 % (Figure 7 (a)), confirming that the process does not involve parasitic reactions.
AC C
EP
TE D
The aqueous asymmetric device was analyzed by EIS before and after cycling as shown in Figure 7 (b). The Nyquist plot of the asymmetric device before cycling shows a nearvertical line in the low frequency region and the lack of a noticeable semicircle at high frequencies. This behavior indicates low charge transfer resistance between the electrolyte and the electrodes. Both Nyquist plots for the asymmetric device have a region with a 45° slope, called the Warburg impedance [48], indicating there is a diffusion limitation for charge transfer. This is reasonable considering ion transport is occurring in electrodes that are composed of numerous nanofludic channels that result from restacking of the 2D material nanosheets during electrode fabrication. The increase (right-shift) in Re (Z) following cycling for the mid-frequency region of the Nyquist plot, which results in a noticeable semi-circle, indicates an increase in faradaic charge transfer resistance, which may contribute to the decrease in capacitance seen during cycling. The degradation of the carboxylic groups in rGOH might be responsible for the incremental increase in the faradaic charge transfer resistance.[49] In summary, we have found a matching positive electrode for MXene for operation with aqueous protic electrolytes. Thus, this work opens up exciting possibilities for using d-Ti3C2Tx and rGOH as a couple in asymmetric aqueous supercapacitors.
11
ACCEPTED MANUSCRIPT
SC
RI PT
The asymmetric Ti3C2Tx/MWCNT//rGOT device was cycled 1000 times at 20 mV.s-1 and the results of the long-term cycling are shown in Figure 7 (c) and 7 (d). As it can be seen in Figure 7 (c), the cell maintains 97% of its initial capacitance after 1000 cycles showing good long-term cycling performance. The coulombic efficiency during the cycling is maintained around 90%. This indicates that the charge storage process is not caused entirely by capacitive processes but some diffusion-limited or irreversible processes might be involved as well. In fact, Dall’Agnese et al.[17] already showed that Ti3C2Tx/MWCNT has redox peaks in organic electrolyte. Herein, these peaks might be overshadowed by the high capacitance of the material. Figure 7 (d) shows the EIS spectrum before and after the cycling test. After cycling there is an increase in the equivalent series resistance of the device; however, this change is small and seem not to affect the performance of the supercapacitor.
M AN U
There is only one report to date on the performance of Ti3C2Tx/MWCNT composite electrodes in organic electrolyte supercapacitors, where Ti3C2Tx/MWCNT electrodes were tested electrochemically in ET4NBF4/ACN in a 3-electrode configuration.[17] Capacitance values (for a single electrode) were reported to be at around 40 F.g-1. 4. Conclusions
EP
TE D
Pure Ti3C2Tx and a composite film made of Ti3C2Tx and MWCNT were tested as counterparts to chemically and thermally reduced graphene oxide electrodes in asymmetric supercapacitor devices in aqueous and organic electrolytes. The aqueous asymmetric supercapacitors achieved cell capacitances of 48 F.g-1 and 78 F.cm-3. These values were lower than those achieved by a symmetric Ti3C2Tx cell. However, the increase in the potential window of the asymmetric device, when compared to the symmetric cell, resulted in increased energy density. The organic asymmetric supercapacitor exhibited larger gravimetric and volumetric cell capacitances (30 F.g-1 and 41 F.cm-3, respectively) when compared to the symmetric cells of Ti3C2Tx and rGOT. The increase in the working potential window for the asymmetric organic-based device resulted in further increases in energy density when compared to the asymmetric aqueous device.
AC C
Acknowledgements
This work was financially supported by the Graphene Flagship (Grant agreement no: 604391. Call: FP7-ICT-2013-FET-F). AMNS was supported by CIC energiGUNE, the Basque Government Scholarship for pre-doctoral formation (PRE_2015_2_0096) and the Egonlabur Traveling Grant (EP_2016_1_0030). KLVA was funded 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. References 12
ACCEPTED MANUSCRIPT
P. Simon, Y. Gogotsi, Materials for electrochemical capacitors., Nat. Mater. 7 (2008) 845–854. doi:10.1038/nmat2297.
[2]
M. Armand, J.-M. Tarascon, Building better batteries, Nature. 451 (2008) 652–657. doi:10.1038/451652a.
[3]
T. Brousse, P.L. Taberna, O. Crosnier, R. Dugas, P. Guillemet, Y. Scudeller, Y. Zhou, F. Favier, D. Bélanger, P. Simon, Long-term cycling behavior of asymmetric activated carbon/MnO2 aqueous electrochemical supercapacitor, J. Power Sources. 173 (2007) 633–641. doi:10.1016/j.jpowsour.2007.04.074.
[4]
A. Khaligh, Z. Li, Battery, ultracapacitor, fuel cell, and hybrid energy storage systems for electric, hybrid electric, fuel cell, and plug-in hybrid electric vehicles: State of the art, IEEE Trans. Veh. Technol. 59 (2010) 2806–2814. doi:10.1109/TVT.2010.2047877.
[5]
M. Beidaghi, Y. Gogotsi, Capacitive energy storage in micro-scale devices: recent advances in design and fabrication of micro-supercapacitors, Energy Environ. Sci. 7 (2014) 867. doi:10.1039/c3ee43526a.
[6]
E. Faggioli, P. Rena, V. Danel, X. Andrieu, R. Mallant, H. Kahlen, Supercapacitors for the energy management of electric vehicles, J. Power Sources. 84 (1999) 261– 269. doi:10.1016/S0378-7753(99)00326-2.
[7]
E. Frackowiak, Carbon materials for supercapacitor application, Phys. Chem. Chem. Phys. 9 (2007) 1774. doi:10.1039/b618139m.
[8]
Z. Zhang, F. Xiao, L. Qian, J. Xiao, S. Wang, Y. Liu, Facile synthesis of 3D MnO2graphene and carbon nanotube-graphene composite networks for highperformance, flexible, all-solid-state asymmetric supercapacitors, Adv. Energy Mater. 4 (2014) 1400064. doi:10.1002/aenm.201400064.
[9]
Z. Yu, L. Tetard, L. Zhai, J. Thomas, Supercapacitor electrode materials: nanostructures from 0 to 3 dimensions, Energy Environ. Sci. 8 (2015) 702–730. doi:10.1039/C4EE03229B.
[10]
B. Mendoza-Sanchez, Y. Gogotsi, Synthesis of Two-Dimensional Materials for Capacitive Energy Storage, Adv. Mater. 28 (2016) 6104–6135. doi:10.1002/adma.201506133.
[12]
[13]
SC
M AN U
TE D
EP
AC C
[11]
RI PT
[1]
W. Choi, I. Lahiri, R. Seelaboyina, Y.S. Kang, Synthesis of Graphene and Its Applications: A Review, Crit. Rev. Solid State Mater. Sci. 35 (2010) 52–71. doi:10.1080/10408430903505036.
M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi, M.W. Barsoum, Two-dimensional nanocrystals produced by exfoliation of Ti3AlC2, Adv. Mater. 23 (2011) 4248–4253. doi:10.1002/adma.201102306. O. Mashtalir, M.R. Lukatskaya, A.I. Kolesnikov, E. Raymundo-Piñero, M. Naguib, M.W. Barsoum, Y. Gogotsi, The effect of hydrazine intercalation on the structure and capacitance of 2D titanium carbide (MXene), Nanoscale. 8 (2016) 9128–9133.
13
ACCEPTED MANUSCRIPT
doi:10.1039/C6NR01462C. M. Boota, B. Anasori, C. Voigt, M.Q. Zhao, M.W. Barsoum, Y. Gogotsi, Pseudocapacitive Electrodes Produced by Oxidant-Free Polymerization of Pyrrole between the Layers of 2D Titanium Carbide (MXene), Adv. Mater. 28 (2016) 1517– 1522. doi:10.1002/adma.201504705.
[15]
Z. Lin, D. Barbara, P.L. Taberna, K.L. Van Aken, B. Anasori, Y. Gogotsi, P. Simon, Capacitance of Ti3C2Tx MXene in ionic liquid electrolyte, J. Power Sources. 326 (2016) 575–579. doi:10.1016/j.jpowsour.2016.04.035.
[16]
J. Come, Y. Xie, M. Naguib, S. Jesse, S. V. Kalinin, Y. Gogotsi, P.R.C. Kent, N. Balke, Nanoscale Elastic Changes in 2D Ti3C2Tx (MXene) Pseudocapacitive Electrodes, Adv. Energy Mater. 6 (2016) 1–9. doi:10.1002/aenm.201502290.
[17]
Y. Dall’Agnese, P. Rozier, P.L. Taberna, Y. Gogotsi, P. Simon, Capacitance of twodimensional titanium carbide (MXene) and MXene/carbon nanotube composites in organic electrolytes, J. Power Sources. 306 (2016) 510–515. doi:10.1016/j.jpowsour.2015.12.036.
[18]
M.Q. Zhao, C.E. Ren, Z. Ling, M.R. Lukatskaya, C. Zhang, K.L. Van Aken, M.W. Barsoum, Y. Gogotsi, Flexible MXene/carbon nanotube composite paper with high volumetric capacitance, Adv. Mater. 27 (2015) 339–345. doi:10.1002/adma.201404140.
[19]
Y. Tang, J. Zhu, C. Yang, F. Wang, Enhanced supercapacitive performance of manganese oxides doped two-dimensional titanium carbide nanocomposite in alkaline electrolyte, J. Alloys Compd. 685 (2016) 194–201. doi:10.1016/j.jallcom.2016.05.221.
[20]
M. Ghidiu, M.R. Lukatskaya, M.-Q. Zhao, Y. Gogotsi, M.W. Barsoum, Conductive two-dimensional titanium carbide “clay” with high volumetric capacitance, Nature. 516 (2014) 78–81. doi:10.1038/nature13970.
[21]
M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Dall’Agnese, P. Rozier, P.L. Taberna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Cation intercalation and high volumetric capacitance of two-dimensional titanium carbide, Sci. Mag. 341 (2013) 1502–1505. doi:10.1126/science.1241488.
[22]
C. Zhao, Q. Wang, H. Zhang, S. Passerini, X. Qian, Two-Dimensional Titanium Carbide/RGO Composite for High-Performance Supercapacitors, ACS Appl. Mater. Interfaces. 8 (2016) 15661–15667. doi:10.1021/acsami.6b04767.
[24]
SC
M AN U
TE D
EP
AC C
[23]
RI PT
[14]
Y. Gogotsi, P. Simon, True performance metrics in electrochemical energy storage., Sci. Mag. 334 (2011) 917–918. doi:10.1126/science.1213003.
M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Y. Dall’Agnese, P. Rozier, P.L. Taberna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Cation Intercalation and High Volumetric Capacitance of Two-Dimensional Titanium Carbide, Sci. Mag. 341 (2013) 1502–1505. doi:10.1126/science.1241488.
14
ACCEPTED MANUSCRIPT
Y. Tang, J. Zhu, C. Yang, F. Wang, Enhanced Capacitive Performance Based on Diverse Layered Structure of Two-Dimensional Ti3C2 MXene with Long Etching Time, J. Electrochem. Soc. 163 (2016) A1975–A1982. doi:10.1149/2.0921609jes.
[26]
M.Q. Zhao, C.E. Ren, Z. Ling, M.R. Lukatskaya, C. Zhang, K.L. Van Aken, M.W. Barsoum, Y. Gogotsi, Flexible MXene/carbon nanotube composite paper with high volumetric capacitance, Adv. Mater. 27 (2015) 339–345. doi:10.1002/adma.201404140.
[27]
D.C. Marcano, D. V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L.B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano. 4 (2010) 4806–4814. doi:10.1021/nn1006368.
[28]
X. Yang, C. Cheng, Y. Wang, L. Qiu, D. Li, Liquid-mediated dense integration of graphene materials for compact capacitive energy storage., Sci. Mag. 341 (2013) 534–7. doi:10.1126/science.1239089.
[29]
Y.J. Oh, J.J. Yoo, Y. Il Kim, J.K. Yoon, H.N. Yoon, J.H. Kim, S. Bin Park, Oxygen functional groups and electrochemical capacitive behavior of incompletely reduced graphene oxides as a thin-film electrode of supercapacitor, Electrochim. Acta. 116 (2014) 118–128. doi:10.1016/j.electacta.2013.11.040.
[30]
X. Gao, J. Jang, S. Nagase, Hydrazine and thermal reduction of graphene oxide: Reaction mechanisms, product structures, and reaction design, J. Phys. Chem. C. 114 (2010) 832–842. doi:10.1021/jp909284g.
[31]
P. Azaïs, L. Duclaux, P. Florian, D. Massiot, M.A. Lillo-Rodenas, A. Linares-Solano, J.P. Peres, C. Jehoulet, F. Béguin, Causes of supercapacitors ageing in organic electrolyte, J. Power Sources. 171 (2007) 1046–1053. doi:10.1016/j.jpowsour.2007.07.001.
[32]
D. Li, M.B. Müller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous dispersions of graphene nanosheets., Nat. Nanotechnol. 3 (2008) 101–105. doi:10.1038/nnano.2007.451.
[33]
M.J. Fernández-Merino, L. Guardia, J.I. Paredes, S. Villar-Rodil, P. SolísFernández, A. Martínez-Alonso, J.M.D. Tascón, Vitamin C is an ideal substitute for hydrazine in the reduction of graphene oxide suspensions, J. Phys. Chem. C. 114 (2010) 6426–6432. doi:10.1021/jp100603h.
[35]
[36]
SC
M AN U
TE D
EP
AC C
[34]
RI PT
[25]
A.M. Navarro-Suárez, N. Casado, J. Carretero-González, D. Mecerreyes, T. Rojo, Full-cell quinone/hydroquinone supercapacitors based on partially reduced graphite oxide and lignin/PEDOT electrodes, J. Mater. Chem. A. 5 (2017) 7137–7143. doi:10.1039/C7TA00527J.
Y. Wang, Y. Song, Y. Xia, Electrochemical capacitors: mechanism, materials, systems, characterization and applications, Chem. Soc. Rev. 45 (2016) 5925–5950. doi:10.1039/C5CS00580A. P. Srivastava, A. Mishra, H. Mizuseki, K.-R. Lee, A.K. Singh, Mechanistic Insight into the Chemical Exfoliation and Functionalization of Ti3C2 MXene, ACS Appl.
15
ACCEPTED MANUSCRIPT
Mater. Interfaces. 8 (2016) 24256–24264. doi:10.1021/acsami.6b08413. M. Naguib, O. Mashtalir, M.R. Lukatskaya, B. Dyatkin, C. Zhang, V. Presser, Y. Gogotsi, M.W. Barsoum, One-step synthesis of nanocrystalline transition metal oxides on thin sheets of disordered graphitic carbon by oxidation of MXenes, Chem. Commun. 50 (2014) 7420–7423. doi:10.1039/C4CC01646G.
[38]
K.J. Cai, Y. Zheng, P. Shen, S.Y. Chen, TiCx–Ti2C nanocrystals and epitaxial graphene-based lamellae by pulsed laser ablation of bulk TiC in vacuum, CrystEngComm. 16 (2014) 5466–5474. doi:10.1039/c4ce00358f.
[39]
T. Hu, J. Wang, H. Zhang, Z. Li, M. Hu, X. Wang, Vibrational properties of Ti3C2 and Ti3C2T2 (T = O, F, OH) monosheets by first-principles calculations: a comparative study, Phys. Chem. Chem. Phys. 17 (2015) 9997–10003. doi:10.1039/C4CP05666C.
[40]
J. Yan, C.E. Ren, K. Maleski, B. Christine, Y. Gogotsi, C.W.V. Gmbh, C. Kgaa, Flexible MXene / Graphene Films for Ultrafast Supercapacitors with Outstanding Volumetric Capacitance, Adv. Funct. Mater. (2017). doi:10.1002/adfm.201701264.
[41]
J.-C. Valmalette, Z. Tan, H. Abe, S. Ohara, Raman scattering of linear chains of strongly coupled Ag nanoparticles on SWCNTs., Sci. Rep. 4 (2014) 5238. doi:10.1038/srep05238.
[42]
S. Osswald, M. Havel, Y. Gogotsi, Monitoring oxidation of multiwalled carbon nanotubes by Raman spectroscopy, J. Raman Spectrosc. 38 (2007) 728–736. doi:10.1002/jrs.1686.
[43]
L.M. Malard, M.A. Pimenta, G. Dresselhaus, M.S. Dresselhaus, Raman spectroscopy in graphene, Phys. Rep. 473 (2009) 51–87. doi:10.1016/j.physrep.2009.02.003.
[44]
M.R. Lukatskaya, S.-M. Bak, X. Yu, X.-Q. Yang, M.W. Barsoum, Y. Gogotsi, Probing the Mechanism of High Capacitance in 2D Titanium Carbide Using In Situ X-Ray Absorption Spectroscopy, Adv. Energy Mater. 5 (2015) 1500589. doi:10.1002/aenm.201500589.
[45]
M.R. Lukatskaya, B. Dunn, Y. Gogotsi, Multidimensional materials and device architectures for future hybrid energy storage, Nat. Commun. 7 (2016) 12647. doi:10.1038/ncomms12647.
[47]
[48]
SC
M AN U
TE D
EP
AC C
[46]
RI PT
[37]
J. Chmiola, C. Largeot, P.L. Taberna, P. Simon, Y. Gogotsi, Desolvation of ions in subnanometer pores and its effect on capacitance and double-layer theory, Angew. Chemie - Int. Ed. 47 (2008) 3392–3395. doi:10.1002/anie.200704894.
J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, P.-L. Taberna, Anomalous Increase in Carbon Capacitance at Pore Sizes Less Than 1 Nanometer, Sci. Mag. 313 (2006) 1760–1763. doi:10.1126/science.1132195. P.L. Taberna, P. Simon, J.F. Fauvarque, Electrochemical Characteristics and Impedance Spectroscopy Studies of Carbon-Carbon Supercapacitors, J.
16
ACCEPTED MANUSCRIPT
Electrochem. Soc. 150 (2003) A292. doi:10.1149/1.1543948. L. Lai, H. Yang, L. Wang, K. Teh, J. Zhong, H. Chou, L. Chen, W. Chen, Z. Shen, R.S. Ruoff, J. Lin, Preparation of Supercapacitor Electrodes through Selection of Graphene Surface Functionalities, ACS Nano. 6 (2012) 5941–5951. doi:10.1021/nn3008096.
EP
TE D
M AN U
SC
RI PT
[49]
AC C
Figure 1. Cross-sectional SEM images of the Ti3C2Tx film (a), and of the Ti3C2Tx/MWCNT film (b); Raman (c) and XRD (d) of the Ti3C2Tx and he Ti3C2Tx/MWCNT films. Numbers in the XRD pattern indicate the miller indices related to the diffraction peaks.
17
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 2. Cross-sectional SEM image (a and b), Raman spectra (c) and XRD patterns (d) of the rGOH and rGOT films. Number in the XRD pattern indicates the miller index related to the diffraction peak.
-1
Figure 3. Cyclic voltammetry at different scan rates of the asymmetric device (a) and at 10 mV.s of the performance of each electrode in the asymmetric device (b) in 1 M H2SO4. Numbers and arrows indicate the voltage/potential windows.
18
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 4. Comparison of the capacitance retention (a and c) and Ragone plots (b and d) calculated gravimetrically (top) and volumetrically (bottom) of the asymmetric device with the symmetric devices in 1 M H2SO4.
19
ACCEPTED MANUSCRIPT
Table 1. Summary of the performance of d-Ti3C2Tx and rGOH in aqueous supercapacitors. CCell -1 (F.g )
CCell -3 (F.cm )
Retention at 100 mV/s (%)
Energy Density -1 (Wh.kg )
Symmetric Ti3C2Tx
0.55
54
220
50
2
Symmetric rGOH
0.9
42
41
33
5
Asymmetric Ti3C2Tx/rGOH
1.1
48
78
48
8
TE D
M AN U
SC
RI PT
Cell Potential (V)
Aqueous-Based Supercapacitors
AC C
EP
Figure 5. Cyclic voltammetry of the asymmetric device in 1 M Et4NBF4/ACN. Cell capacitance variation -1 with voltage at different scan rates (a) and behavior of each electrode with potential at 10 mV.s (b). Numbers and arrows indicate the voltage/potential windows.
20
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 6. Comparison of the capacitance retention of the asymmetric device with the symmetric ones in 1 M EtN4BF4 (a and c). Ragone plot comparing the aqueous- with the organic-based supercapacitor (b and d). Values calculated gravimetrically (top) and volumetrically (bottom).
21
ACCEPTED MANUSCRIPT
Table 2. Summary of the performance of Ti3C2Tx-CNT and rGOT in organic-based supercapacitors. CCell -1 (F.g )
CCell -3 (F.cm )
Retention at 100 mV/s (%)
Energy Density -1 (Wh.kg )
Symmetric Ti3C2Tx-MWCNT
1.8
7
16
37
3
Symmetric rGOT
2.5
19
16
47
16
Asymmetric Ti3C2Tx-MWCNT//rGOT
2.0
30
41
50
20
AC C
EP
TE D
M AN U
SC
Organic-Based Supercapacitors
RI PT
Cell Potential (V)
22
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
-1
AC C
Figure 7. Capacitance retention during 1000 cycles at 20 mV.s of the Ti3C2Tx/rGOH (a) and the Ti3C2Tx/MWCNT//rGOT (c). Electrochemical Impedance spectroscopy before and after cycling of the asymmetrical device in 1 M H2SO4 (b) and in 1 M Et4NBF4/ACN (d).
23
S0013-4686(17)32247-8
DOI:
10.1016/j.electacta.2017.10.125
Reference:
EA 30511
To appear in:
Electrochimica Acta
Received Date: 27 June 2017 Revised Date:
11 October 2017
Accepted Date: 18 October 2017
Please cite this article as: A.M. Navarro-Suárez, K.L. Van Aken, T. Mathis, T. Makaryan, J. Yan, J. Carretero-González, Teó. Rojo, Y. Gogotsi, Development of asymmetric supercapacitors with titanium carbide-reduced graphene oxide couples as electrodes, Electrochimica Acta (2017), doi: 10.1016/ j.electacta.2017.10.125. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Development of Asymmetric Supercapacitors with Titanium CarbideReduced Graphene Oxide Couples as Electrodes Adriana M. Navarro-Suáreza,b,†, Katherine L. Van Akena, Tyler Mathisa, Taron Makaryana, Jun Yana,c, Javier Carretero-Gonzálezd, Teófilo Rojob,e, and Yury Gogotsi*,a a
RI PT
Department of Materials Science & Engineering and A.J. Drexel Nanomaterials Institute, Drexel University, Philadelphia, PA 19104, USA.
b
CIC Energigune, Albert Einstein 48, 01510 Miñano, Alava, Spain.
c
SC
Key Laboratory of Superlight Materials and Surface Technology, Ministry of Education, College of Material Science and Chemical Engineering, Harbin Engineering University, Harbin 150001, China. d
e
M AN U
Institute of Polymer Science and Technology, ICTP-CSIC, Juan de la Cierva 3, 28006, Madrid, Spain. Inorganic Chemistry Department, P.O. Box 644, University of the Basque Country, 48080 Bilbao, Spain.
†
Current address: Department of Physics, Chalmers Univ. of Technol., 41296 Gothenburg, Sweden
Abstract
TE D
*Email: [email protected]
AC C
EP
Two-dimensional (2D) nanomaterials have attracted significant interest for supercapacitor applications due to their high surface to volume ratio. Layered 2D materials have the ability to intercalate ions and thus can provide intercalation pseudocapacitance. Properties such as achieving fast ion diffusion kinetics and maximizing the exposure of the electrolyte to the surface of the active material are critical for optimizing the performance of active materials for electrochemical capacitors (i.e. Supercapacitors). In this study, two 2D materials, titanium carbide (Ti3C2Tx) and reduced graphene oxide (rGO), were used as electrode materials for asymmetric supercapacitors, with the resulting devices achieving high capacitance values and excellent capacitance retention in both aqueous and organic electrolytes. This work demonstrates that Ti3C2Tx is a promising electrode material for flexible and high-performance energy storage devices.
Keywords: Supercapacitors; MXene; graphene; volumetric capacitance; 2D materials.
1
ACCEPTED MANUSCRIPT
1. Introduction
M AN U
SC
RI PT
Developing electrochemical energy storage devices with high energy densities, rapid charge/discharge capabilities, and long cycle lifetimes has been a subject of extensive research in the past decade for portable electronics and hybrid electric vehicles. [1–4] For instance, the ability to design miniaturized supercapacitors for applications such as on-chip energy storage, wearable electronics, etc., demands light and flexible materials with high volumetric capacitances.[5] However, the incorporation of supercapacitors into these applications requires improvement of the device’s energy density.[6] In the case of supercapacitors, which operate via physical adsorption of ions, this issue can be addressed by increasing the material’s active surface area or enlarging the operational voltage window.[7] For the latter, asymmetric devices have been proposed in order to advantageously utilize the operating voltage ranges of two different materials such as manganese dioxide (MnO2), graphene, or multi-walled carbon nanotubes (MWCNTs).[8] As for the former issue, incorporating pillar-like nanostructures for opening up densely packed two-dimensional (2D) nanomaterials has been explored as a possible route.[9]
EP
TE D
2D materials are inherently flexible, have high surface reactivity, high electrical conductivity, and large surface areas, which are attractive characteristics for materials for energy storage devices.[10] The exfoliation of graphite into single graphene layers and the subsequent research into graphene’s excellent physical and electronic properties in 2004 resulted in increased interest in 2D materials.[11] A recent addition to the family of 2D materials are MXenes, a large group (20+ members to date) of 2D transition metal carbides, nitrides, and carbonitrides that were discovered in 2011.[12] MXenes have a general formula of , where is an early transition metal (e.g. Ti, Nb), can be carbon or nitrogen, being equal to 1, 2 or 3, and represents surface functional terminations such as –OH, -O, or -F.
AC C
MXenes have shown great potential as electrode materials in supercapacitors.[13–21] Pristine, delaminated Ti3C2Tx is the most studied MXene and has achieved capacitance values up to 238 F.g-1 / 900 F.cm-3 in acidic electrolytes. [14,21] Mixed with reduced graphene oxide (rGO) in a symmetric cell configuration, Ti3C2Tx has demonstrated an outstanding capacitance of 154 F.g-1 at 2 A.g-1.[22] One disadvantage of Ti3C2Tx is that its potential window in aqueous electrolytes is narrow (less than 1 V). This translates to lower energy density when compared to carbon materials.[23] Most studies so far focused on Ti3C2Tx electrodes testing in 3-electrode cells,[17,20,24,25] which does not allow evaluation of the operating potential of devices with MXene electrodes or calculating energy and power density of the devices.
2
ACCEPTED MANUSCRIPT
SC
RI PT
In order to further increase the energy density of Ti3C2Tx-based supercapacitors, we have developed a full cell using Ti3C2Tx and rGO as the negative and positive electrodes, respectively, and tested the devices in both aqueous and organic electrolytes. By using an asymmetric device, the total electrochemical window of the supercapacitor is enlarged, resulting in increased energy and power densities. We have also studied the effect of disrupting the stacking of Ti3C2Tx layers by hybridizing MXene with MWCNTs, which increases the interlayer distance, resulting in better ion accessibility and transport for organic electrolytes. Through optimizing the physical properties of the active materials to match the electrolyte system, we achieved high capacitance for the Ti3C2Tx/MWCNT-rGO asymmetric device in an organic electrolyte. This strategy of matching and optimizing active materials to the electrolyte being used here can be applied to the development of other energy storage devices.
2.2 Materials Synthesis
M AN U
2. Experimental
2.2.1 Synthesis of 2D Titanium Carbide MXene
AC C
EP
TE D
The complete synthesis method for titanium carbide MXene is described elsewhere.[14] Briefly, to synthesize Ti3C2Tx, 20 mL of 9 M hydrochloric acid (HCl, Fisher Scientific) were added to 7.5 molar equivalents (2 g) of lithium fluoride (LiF, Alfa Aesar). The mixture was stirred until the salt was dissolved. Then, 2 g of the ternary titanium aluminum carbide powder (Ti3AlC2, < 38 µm particle size) were slowly added to the solution. The reaction mixture was held at 35 °C for 24 h while stirring a t 200 rpm. After 24 h, the mixture was washed with deionized water and separated by centrifugation at 3500 rpm, after which the supernatant was collected. The washing process was repeated until the pH of the supernatant became neutral (pH 6-7). In order to increase the surface area, the Ti3C2Tx solution was delaminated by bath sonication for 1 h under argon flow. Afterwards, the sonicated solution was centrifuged for another hour at 3500 rpm. Finally, the supernatant containing the Ti3C2Tx was collected. In order to prepare a flexible, free-standing Ti3C2Tx film, the solution was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum. The Ti3C2Tx/multi-walled carbon nanotube (MWCNT) composite was prepared as described elsewhere.[26] First, an aqueous 1 mg.mL-1 Ti3C2Tx suspension was prepared as well as an aqueous solution of 0.1 mg.mL-1 MWCNT in 8 mg.mL-1 aqueous solution of sodium dodecyl sulfate (SDS, Sigma Aldrich). Sandwich-like MXene/MWCNT films were prepared using an alternating filtration method. Specifically, 1 mL of the Ti3C2Tx dispersion was filtered through a polypropylene membrane to yield a thin Ti3C2Tx layer. Then, 1 mL of the MWCNT-SDS dispersion was filtered on top of the Ti3C2Tx layer. This alternating filtration was repeated several times to yield composite films composed of 5 and 4 3
ACCEPTED MANUSCRIPT
alternating Ti3C2Tx and MWCNTs layers, respectively. Afterwards, the film was washed with 200 milliliters of deionized water. The composite film was dried under vacuum at 120 °C. 2.2.2 Synthesis of Reduced Graphene Oxide
TE D
2.2.3 Hydrazine Reduction Method
M AN U
SC
RI PT
The graphite oxide was prepared following the Marcano-Tour method.[27] A 9:1 mixture of concentrated H2SO4/H3PO4 (360:40 mL) was added to a mixture of graphite powder (3.0 g, 1 wt. equiv.) and KMnO4 (18.0 g, 6 wt. equiv.). The reaction mixture was then heated to 50 °C and stirred overnight. The mixture was then c ooled to room temperature and poured onto ice (~400 mL) with 30 % H2O2 (3 mL). The solution was centrifuged (10000 rpm for 30 min), and the supernatant was decanted away. The remaining solid material was then washed in succession with 200 mL of water, 200 mL of 30 % HCl, and then water until a neutral pH was achieved. For each wash, the mixture was centrifuged (10000 rpm for 30 min) and the supernatant was decanted away. The graphite oxide was then sonicated in water for 1 hour and the exfoliated graphene oxide (GO) was collected. Thereafter, two different reduction methods were used. The first one was a chemical method that yields reduced graphene oxide (rGO) with carboxylic acid groups.[28] These groups can be beneficial for pseudocapacitance in aqueous electrolytes.[29] In the second method, the graphene oxide was reduced thermally, eliminating the oxygen functional groups.[30] Oxygen containing functionalities are known for being detrimental to capacitance in organic-based supercapacitors.[31]
AC C
EP
GO was reduced following the Li-Wallace’s method. [32] Briefly, 30 mL of GO in water (7 mg.mL-1) was heated to 85 °C. Then 2 mL of hydrazine hydra te (98 %, TCI America) and 1 mL of ammonium hydroxide (Fischer Scientific) were added to the GO. The reaction was kept under reflux during one hour, after which the solution was left to cool and centrifuged for 30 minutes; the reduced graphene oxide (rGOH) was then re-dispersed in ethanol (Decon Labs, Inc). In this case, hydrazine monohydrate acts as the reducing agent, and ammonium hydroxide is used to promote the colloidal stability of the GO sheets through electrostatic repulsion.[33] In order to prepare a free-standing film, the rGOH was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum. 2.2.4 Thermal Reduction Method The GO solution was freeze dried to produce a GO aerogel. Then, 200 mg of the freeze dried-GO was heated under argon flow at a rate of 10 °C.min -1 up to 900 °C and held there for 2 hours. After cooling, the reduced graphene oxide (rGOT) was re-dispersed in ethanol. Then, the rGOT was filtered using a polypropylene membrane (3501 Coated PP, Celgard LLC) and dried under vacuum.
4
ACCEPTED MANUSCRIPT
2.3 Characterization Methods The characterization of the Ti3C2Tx, Ti3C2Tx-MWCNT, and rGO film morphologies was carried out using scanning electron microscopy (SEM, Zeiss Supra 50VP, Germany).
RI PT
The vibration frequencies of the materials were studied by Raman Spectroscopy. The Raman spectra were recorded on an inverted Renishaw inVia spectrometer equipped with a 60x objective, with a 632 nm laser as an excitation source (<1 mW laser power). The back-scattered light was reflected to a room-temperature operating detector after a 1200 lines/mm diffraction grating.
M AN U
2.3.1 Electrochemical Testing
SC
The interlayer distance of the Ti3C2Tx, Ti3C2Tx-MWCNT, and rGO films was calculated using X-Ray Diffraction (XRD). XRD was carried out on a Rigaku Smart Lab (Japan) diffractometer using Cu-Kα radiation (40 kV and 44 mA), a step scan of 0.02 °, 2θ range of 5 – 50 °, and a step time of 0.5 s.
AC C
EP
TE D
To quantify the amount of charge stored by the prepared materials, all tested films were cut to have the same dimensions and weight before being assembled in Swagelok-type cells. The materials were tested in a 3-electrode configuration to evaluate their capacitance and electrochemical stability window. Then, symmetric and asymmetric devices were assembled for further electrochemical analysis of the device performance. The cells were electrochemically studied by using a 1 M sulfuric acid (H2SO4, Sigma Aldrich) solution with a 3.5 M aqueous silver/silver chloride (Ag/AgCl) electrode as the reference. In the case of the organic electrolyte, 1 M tetraethyl ammonium tetrafluoroborate (Et4N+BF4-, Sigma Aldrich) in acetonitrile (ACN, Fischer) was used as the electrolyte and Ag wire as the reference or pseudo-reference electrode. For the 3electrode cell, overcapacitve activated carbon (YP-50, Kuraray) was used as the counter electrode. Upon assembly, all the cells were precycled between 50 and 200 times to ensure the electrolyte had permeated throughout the porous structure and the cell had stabilized. Cyclic voltammetry (CV) measurements at different scan rates ranging from 2 to 100 mV.s-1, as well as impedance measurements were performed at ambient conditions in a multichannel potentiostat/galvanostat (Biologic VMP3, France). The cycling performance of the asymmetric devices was tested at 20 mV.s-1 over 1000 cycles from 0 to 1.1 V and from 0 to 2 V in 1 M H2SO4 and 1 M Et4NBF4/ACN, respectively. Before and after the cycling test, electrochemical impedance spectroscopy (EIS) was measured within the ac frequency region from 10 mHz to 200 kHz with ac voltage amplitude of 10 mV at open circuit voltage. Capacitance values per electrode are typically reported, which is valid for symmetric cells and 3-electrode characterization. However, for asymmetric devices, these values are not accurate enough as each electrode has a different mass.[34] In this article, we will refer to capacitance when the electrode capacitance was calculated and to cell capacitance when referring to the capacitance of the whole device using both electrode mass. Cell capacitance is calculated using Eq. 1:
5
ACCEPTED MANUSCRIPT
Ccell =
∫ idt
Eq. 1
( m+ + m− ) ∆V
RI PT
where Ccell is the cell capacitance, i is the discharge current, ∆V is the voltage window, and m+ and m- correspond to the mass of the positive and negative electrode, respectively.[35] Ragone plots are calculated from the galvanostatic charge‒discharge curves. The gravimetric energy (Egrav) and power (Pgrav) corresponding to the mass of the active material per electrode are estimated by using the following equations respectively:
∫ iVdt ( m+ + m− )
Eq. 3
M AN U
Pgrav = Egrav / t
Eq. 2
SC
Egrav =
The volumetric capacitance, energy and power were calculated by considering the density of each electrode.
3. Results and Discussion 3.1 Materials Characterization
TE D
3.1.1 2D Titanium Carbide MXene and 2D Titanium Carbide MXene-MWCNT Composite
AC C
EP
During the synthesis of Ti3C2Tx from Ti3AlC2, the HCl and LiF dissociate and H+ and F- ions weaken the Ti-Al bonds. This results in an opening of the interlayer spacing and allows for further insertion that leads to the formation of AlF3 and H2,, which are removed during washing. The 2D Ti3C2Tx layers possess two exposed Ti atoms per unit formula that have surface functionalities such as hydroxyl and fluorine groups, which are present in the reaction media.[36] Recently, Dall’Agnese et al. proved that the addition of MWCNT improves the performance of Ti3C2Tx in organic electrolytes.[17] After delamination of the MXene, the MWCNT are added according to the procedure explained elsewhere.[18] In this case, the mass ratio of Ti3C2Tx to MWCNT in the film was 92:8. The insertion of MWCNT between the layers of d-Ti3C2Tx decreases the density of the MXene electrode film from 4.07 to 2.44 g.cm-3. In order to confirm the synthesis of Ti3C2Tx, to study its structure and the influence of the MWCNT on the Ti3C2Tx morphology, the films were characterized by SEM, Raman spectroscopy, and X-Ray diffraction. Figure 1(a) shows the cross-sectional scanning electron microscopy (SEM) image of a binder-free Ti3C2Tx film. This image shows that the entire film is composed of well-aligned stacked Ti3C2Tx MXene sheets. This layered morphology is similar to that of expanded graphite, as the nanolayers are clearly separated from each other. This shearing of the 2D particles allows the material to form 6
ACCEPTED MANUSCRIPT
flexible, free-standing films. Figure 1(b) shows a cross-sectional SEM image of a Ti3C2Tx/MWCNT composite. The anticipated increase in spacing between the Ti3C2Tx layers after the incorporation of the MWCNTs is evident in the image.
M AN U
SC
RI PT
Figure 1(c) shows the Raman spectrum of the Ti3C2Tx film. All the vibration bands are consistent with values reported in literature.[37] The four broad Raman peaks centered around 200, 369, 628 and 716 cm−1 are attributed to the vibrations from non-stoichiometric titanium carbide.[38,39] The Raman spectrum of the Ti3C2Tx/MWCNT film, shown in Figure 1(d), combines the features of both Ti3C2Tx and MWCNTs. The peaks observed at low frequencies correspond to the peaks attributed to Ti3C2Tx and are in agreement with the values of the MXene by itself. The (002) peak in the XRD pattern of d-Ti3C2 shifted from 7.1 ° to 6.4 ° for the sandwich-like Ti 3C2Tx/MWCNT film. This shift translates into an increase in the interlayer distance from 12.5 Å in the Ti3C2Tx to 13.8 Å in the Ti3C2Tx /MWCNT, which could be possibly attributed to the intercalation of surfactant molecules into Ti3C2Tx layers.[40] Ti3C2Tx does not have observable peaks at higher diffraction angles, yet the Ti3C2Tx/MWCNT film does. The peak at 1141.4 cm-1 corresponds to C-H vibration of the surfactant used during synthesis.[41] The peaks at 1295 and 1538.4 cm-1 correspond to the D and G bands of the MWCNTs, respectively.[42,43]
3.1.2 Chemically and Thermally Reduced Graphene Oxide
EP
TE D
The presence of surface functionalities, water, or impurities in the active material are the main causes of ageing in organic-based supercapacitors.[31] The hydrazine reduction of GO does not remove epoxide groups from the edges of the aromatic domains,[30] making the rGOH a poor candidate for organic-based supercapacitors. Thermal annealing of the GO above 700 °C eliminates hydroxyl and carboxyl gr oups from the material.[30] Herein, the GO is thermally reduced at 900 °C (labeled here as rGOT) in order to use it as an electrode material for organic-based supercapacitors.
AC C
To confirm the 2D structure of the rGOH and the rGOT films, the samples were also analyzed by SEM, Raman spectroscopy, and XRD. A cross section image of the rGOH and the rGOT films by SEM is shown in Figure 2(a) and 2(b). As with Ti3C2Tx, the reduced graphene oxide films have a layered structure. The Raman spectrum (Figure 2(c)) of both rGO films exhibit the typical fingerprint of carbon materials. The D band is around 1323 cm-1 while the G band is at 1598 cm-1, with an ID/IG ratio of 1.4 and 1.3. for rGOH and rGOT, respectively, indicating that the two reduction methods achieve similar rGO materials in terms of ordering of the 2D material. Figure 2(d) shows the X-Ray Diffraction of the rGOH and rGOT films. The materials show water-filled space between the layers ranging from 2.2 to 4.4 Å. The (002) peak of the rGOT is at 25.4 °, 1.7 ° to the right of the (002) peak of the rGOH. This shift towards higher angles indicates that rGOT is slightly less oxidized than rGOH. 7
ACCEPTED MANUSCRIPT
3.2 Electrochemical devices 3.2.1 Aqueous-Based Asymmetric Device
M AN U
SC
RI PT
According to the 3-electrode electrochemical studies of the Ti3C2Tx and the rGOH films, shown in Figures S1 and S2, the d-Ti3C2Tx film has high volumetric and gravimetric capacitances (230 F.g-1 and 934 F.cm-3 at 2 mV.s--1) as well as excellent capacitance retention (167 F.g-1 at 100 mV.s--1). It is important to mention that a peak at ~0.3 V corresponds to redox reaction of surface titanium atoms.[44] On the other hand, the rGOH film has a high gravimetric capacitance (251 F.g-1 at 2 mV.s--1) but its volumetric capacitance (248 F.cm-3 at 2 mV.s--1) and capacitance retention (129 F.g-1 at 100 mV.s--1) need to be improved. To compare the results from the asymmetric and symmetric devices, the capacitance values henceforth reported correspond to the cell capacitance. As these materials have complementary potential windows (from -0.2 to 0.35 V vs Ag/AgCl for Ti3C2Tx and from 0.0 to 0.9 V vs Ag/AgCl for rGOH), an asymmetric supercapacitor using these materials was constructed and tested. In order to assemble the asymmetric supercapacitor, the Ti3C2Tx film was used as the negative electrode and the rGOH film as the positive electrode. The amount of charge, Q, stored in each of the positive and negative electrodes in an asymmetrical supercapacitor must be the same and is governed by Eq. 4. Q = CSP+ * m+ * V+ = CSP- * m- * V-
Eq. 4
AC C
EP
TE D
Due to the asymmetry in the electrochemical behavior for the cations and anions, unequal specific capacitances for the positive (CSP+) and negative (CSP-) electrodes occur. Therefore, in order to maintain stable voltage conditions where V- = V+, the weight of the electrodes (m- and m+) should be unequal to compensate for the different specific capacitance values[45]. The mass balance was calculated using Eq. 5, taking into account the maximum capacitance of each material in a three-electrode setup with the same potential window.
m- = CSP+ * m+ / CSP-
Eq. 5
Then it follows that the ratio of rGOH to Ti3C2Tx is:
mrGOH / mTi3C2Tx = CTi3C2Tx / CrGOH = 230 / 251 = 0.9
Eq. 6
8
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
In an asymmetric device, the electrodes have dissimilar charge storage capacities; therefore, in this section the capacitance reported will be the one corresponding to the full cell. Figure 3 (a) shows the evolution of the cell capacitance with voltage for the asymmetric device at different scan rates, with mrGOH / mTi3C2Tx = 0.9. The capacitive behavior (rectangular shape of the voltammogram) is maintained at different scan rates indicating good electrical conductivity and ion diffusion inside the pores of the electrode materials. In the asymmetric device, the electrochemical window is expanded to 1.1 V as the electrodes have complementary working potential windows. Based on the evidence of peaks and faradaic reactions that started to become clear, the voltage window was not expanded further. The increase in the device’s operating voltage window will positively affect the energy and power delivered by it. Figure 3 (b) shows the contributions from each electrode when the cell is cycled at 10 mV.s-1. The potential windows corresponding to the rGOH and Ti3C2Tx were 0.71 and 0.39 V, respectively. To equilibrate the charges, other mass balance proportions were tried (not shown) but the Ti3C2Tx potential window was always limited to less than 0.4 V. The capacitance values achieved by the Ti3C2Tx electrode, and the rGOH electrode at 10 mV.s-1 are 195 and 114 F.g-1, respectively.
TE D
The cell specific capacitances and Ragone plots (calculated using the mass and dimensions of the active materials only) for the asymmetric and symmetric devices are shown in Figure 4. The asymmetric device exhibits intermediate capacitance retention between the symmetric cells of Ti3C2Tx and rGOH (Figure 4 (a and c)). The maximum cell capacitance achieved by the asymmetric device is 48 F.g-1 (78 F.cm-3). The use of Ti3C2Tx as a counterpart to rGOH in an asymmetric cell improves the capacitance retention (44 % at 10 A.g-1) as compared to that of the symmetric rGOH (33 % at 10 A.g-1) and expands the voltage window. However, the capacitance value is lower than the one achieved by the symmetric Ti3C2Tx device.
AC C
EP
The advantage of using rGOH as a counterpart to Ti3C2Tx, as explained before, is the expansion of the whole device’s operating potential window by compensating for the MXene electrode’s inherently moderate voltage window. In the Ragone plot (Figure 4(b and d)) the small voltage window causes the symmetric Ti3C2Tx device to present the lowest gravimetric energy values of the three devices. The symmetric rGOH device exhibits a larger energy density than the symmetric Ti3C2Tx device, but drops quickly as the power increases. As expected, the asymmetric device achieves the highest energy density owing to the enlarged voltage window, and the most stable energy density among the three devices. The asymmetric Ti3C2Tx/rGOH achieves up to 8 Wh.kg-1. The influence of the materials’ density can be observed in Figure 4 (c and d). As explained before, Ti3C2Tx is a denser material than rGOH. So, when comparing their volumetric capacitance retention, Ti3C2Tx exhibits the highest values, rGOH the lowest and the asymmetric device falls in between. This behavior is translated to the Ragone plot, where as in the gravimetric plot, the asymmetric device achieves the highest energy values. Nevertheless, in contrast with the gravimetric plot, Ti3C2Tx achieves larger energy values
9
ACCEPTED MANUSCRIPT
than rGOH, as its film density is four times higher. Table 1 summarizes the main results of the electrochemical testing of Ti3C2Tx and rGOH in aqueous (H2SO4) electrolyte in symmetric and asymmetric configurations.
RI PT
In order to further increase the energy density delivered by the asymmetric device, the materials were also tested in an organic electrolyte. However, as explained before the materials were slightly modified to improve their performance in organic electrolytes. 3.2.2 Organic Electrolyte Asymmetric Device
M AN U
SC
The assembly of an asymmetric supercapacitor with organic electrolyte is not as straightforward as the aqueous-based one. The potential windows of the Ti3C2Tx/MWCNT and rGOT are not complementary and rGOT has a better electrochemical performance in 1 M Et4NBF4/ACN than the composite (Figure S3). Nevertheless, the asymmetric supercapacitor was assembled to check whether the performance of the Ti3C2Tx/MWCNT could be improved by using rGOT as the positive electrode. rGOT was chosen as the positive electrode due to its larger potential window. Ti3C2Tx/MWCNT was chosen as the negative electrode as its larger interlayer spacing could accommodate Et4N+ ions. The mass balance was calculated using Eq. 5, taking into account the maximum capacitance values showed by each material in the 3-electrode configuration. The result is shown in Eq. 7. mrGOT / md-Ti3C2Tx/MWCNT = 42 / 71 = 0.6
Eq. 7
EP
TE D
Figure 5 shows the cyclic voltammograms of the asymmetric Ti3C2Tx/MWCNT and rGOT device using mrGOT / mTi3C2Tx/MWCNT = 0.6. The cell can operate at up to 2.3 V but the cell capacitance drops drastically when the scan rate is increased. Therefore, the asymmetric cells in this study were tested at 2 V (Figure 5 (a)). The performance of each electrode was studied with a pseudo-reference electrode (Figure 5 (b)). The potential window of rGOT is equal to 0.8 V and shows a capacitance of 83 F.g-1, while the Ti3C2Tx/MWCNT exhibits a potential window of 1.2 V and achieves 35 F.g-1.
AC C
Figure 6 (a and c) shows the cell capacitance values of the asymmetric device and the symmetric cells of Ti3C2Tx/MWCNT and rGOT. For these devices, the cell capacitance of the asymmetric device was the highest due to the interlayer distance of the active materials being suitable to accommodate the Et4N+ and BF4- ions. The XRD pattern indicated an interlayer distance for the Ti3C2Tx/MWCNT composite film of 1.38 nm and assuming the size of the unsolvated Et4N+ cation is 0.67 nm and the size of the solvated ion is 1.3 nm,[46] we can hypothesize that the similarity in size allows the insertion of ions in between the layers of the material. The rGOT has an interlayer distance ranging between 0.23 and 0.44 nm, which matches well with the solvated BF4- ion size of 0.33 nm.[47]
10
ACCEPTED MANUSCRIPT
RI PT
When taking into account the density of the materials (Figure 6 (c and d)), the symmetric cells achieve similar capacitance values at 2 mV.s-1. After increasing the scan rate above 5 mV.s-1, the symmetric rGOT retains 71 % of its initial capacitance while the symmetric Ti3C2Tx/MWCNT retains 84 %. Figure 6 shows the gravimetric (b) and volumetric (d) Ragone plots for the asymmetric devices in aqueous and organic electrolytes. As expected, the increase in voltage window results in higher energy density for the organic cell. Nevertheless, the aqueous electrolyte supercapacitor exhibits better energy retention owing to the higher conductivity of the aqueous electrolyte. Table 2 summarizes the performance of Ti3C2Tx-MWCNT and rGOT as electrodes for supercapacitors in organicbased electrolytes.
SC
3.3 Long-term Cycling Tests
M AN U
To assess the long-term cycling behavior of the MXene-rGO based supercapacitors, the asymmetric devices were tested for 1000 cycles and the evolution of their capacitance is plotted in Figure 7 (a) and 7 (c). After 1000 cycles at 20 mV.s-1, the capacitance retention of the asymmetric Ti3C2Tx/rGOH is 76 % of the initial value. The decrease in capacitance might be caused by the lower conductivity of the rGOH, as it has been proven that Ti3C2Tx can be cycled up to 10000 cycles without capacitance losses.[21] The coulombic efficiency for the device is close to 100 % (Figure 7 (a)), confirming that the process does not involve parasitic reactions.
AC C
EP
TE D
The aqueous asymmetric device was analyzed by EIS before and after cycling as shown in Figure 7 (b). The Nyquist plot of the asymmetric device before cycling shows a nearvertical line in the low frequency region and the lack of a noticeable semicircle at high frequencies. This behavior indicates low charge transfer resistance between the electrolyte and the electrodes. Both Nyquist plots for the asymmetric device have a region with a 45° slope, called the Warburg impedance [48], indicating there is a diffusion limitation for charge transfer. This is reasonable considering ion transport is occurring in electrodes that are composed of numerous nanofludic channels that result from restacking of the 2D material nanosheets during electrode fabrication. The increase (right-shift) in Re (Z) following cycling for the mid-frequency region of the Nyquist plot, which results in a noticeable semi-circle, indicates an increase in faradaic charge transfer resistance, which may contribute to the decrease in capacitance seen during cycling. The degradation of the carboxylic groups in rGOH might be responsible for the incremental increase in the faradaic charge transfer resistance.[49] In summary, we have found a matching positive electrode for MXene for operation with aqueous protic electrolytes. Thus, this work opens up exciting possibilities for using d-Ti3C2Tx and rGOH as a couple in asymmetric aqueous supercapacitors.
11
ACCEPTED MANUSCRIPT
SC
RI PT
The asymmetric Ti3C2Tx/MWCNT//rGOT device was cycled 1000 times at 20 mV.s-1 and the results of the long-term cycling are shown in Figure 7 (c) and 7 (d). As it can be seen in Figure 7 (c), the cell maintains 97% of its initial capacitance after 1000 cycles showing good long-term cycling performance. The coulombic efficiency during the cycling is maintained around 90%. This indicates that the charge storage process is not caused entirely by capacitive processes but some diffusion-limited or irreversible processes might be involved as well. In fact, Dall’Agnese et al.[17] already showed that Ti3C2Tx/MWCNT has redox peaks in organic electrolyte. Herein, these peaks might be overshadowed by the high capacitance of the material. Figure 7 (d) shows the EIS spectrum before and after the cycling test. After cycling there is an increase in the equivalent series resistance of the device; however, this change is small and seem not to affect the performance of the supercapacitor.
M AN U
There is only one report to date on the performance of Ti3C2Tx/MWCNT composite electrodes in organic electrolyte supercapacitors, where Ti3C2Tx/MWCNT electrodes were tested electrochemically in ET4NBF4/ACN in a 3-electrode configuration.[17] Capacitance values (for a single electrode) were reported to be at around 40 F.g-1. 4. Conclusions
EP
TE D
Pure Ti3C2Tx and a composite film made of Ti3C2Tx and MWCNT were tested as counterparts to chemically and thermally reduced graphene oxide electrodes in asymmetric supercapacitor devices in aqueous and organic electrolytes. The aqueous asymmetric supercapacitors achieved cell capacitances of 48 F.g-1 and 78 F.cm-3. These values were lower than those achieved by a symmetric Ti3C2Tx cell. However, the increase in the potential window of the asymmetric device, when compared to the symmetric cell, resulted in increased energy density. The organic asymmetric supercapacitor exhibited larger gravimetric and volumetric cell capacitances (30 F.g-1 and 41 F.cm-3, respectively) when compared to the symmetric cells of Ti3C2Tx and rGOT. The increase in the working potential window for the asymmetric organic-based device resulted in further increases in energy density when compared to the asymmetric aqueous device.
AC C
Acknowledgements
This work was financially supported by the Graphene Flagship (Grant agreement no: 604391. Call: FP7-ICT-2013-FET-F). AMNS was supported by CIC energiGUNE, the Basque Government Scholarship for pre-doctoral formation (PRE_2015_2_0096) and the Egonlabur Traveling Grant (EP_2016_1_0030). KLVA was funded 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. References 12
ACCEPTED MANUSCRIPT
P. Simon, Y. Gogotsi, Materials for electrochemical capacitors., Nat. Mater. 7 (2008) 845–854. doi:10.1038/nmat2297.
[2]
M. Armand, J.-M. Tarascon, Building better batteries, Nature. 451 (2008) 652–657. doi:10.1038/451652a.
[3]
T. Brousse, P.L. Taberna, O. Crosnier, R. Dugas, P. Guillemet, Y. Scudeller, Y. Zhou, F. Favier, D. Bélanger, P. Simon, Long-term cycling behavior of asymmetric activated carbon/MnO2 aqueous electrochemical supercapacitor, J. Power Sources. 173 (2007) 633–641. doi:10.1016/j.jpowsour.2007.04.074.
[4]
A. Khaligh, Z. Li, Battery, ultracapacitor, fuel cell, and hybrid energy storage systems for electric, hybrid electric, fuel cell, and plug-in hybrid electric vehicles: State of the art, IEEE Trans. Veh. Technol. 59 (2010) 2806–2814. doi:10.1109/TVT.2010.2047877.
[5]
M. Beidaghi, Y. Gogotsi, Capacitive energy storage in micro-scale devices: recent advances in design and fabrication of micro-supercapacitors, Energy Environ. Sci. 7 (2014) 867. doi:10.1039/c3ee43526a.
[6]
E. Faggioli, P. Rena, V. Danel, X. Andrieu, R. Mallant, H. Kahlen, Supercapacitors for the energy management of electric vehicles, J. Power Sources. 84 (1999) 261– 269. doi:10.1016/S0378-7753(99)00326-2.
[7]
E. Frackowiak, Carbon materials for supercapacitor application, Phys. Chem. Chem. Phys. 9 (2007) 1774. doi:10.1039/b618139m.
[8]
Z. Zhang, F. Xiao, L. Qian, J. Xiao, S. Wang, Y. Liu, Facile synthesis of 3D MnO2graphene and carbon nanotube-graphene composite networks for highperformance, flexible, all-solid-state asymmetric supercapacitors, Adv. Energy Mater. 4 (2014) 1400064. doi:10.1002/aenm.201400064.
[9]
Z. Yu, L. Tetard, L. Zhai, J. Thomas, Supercapacitor electrode materials: nanostructures from 0 to 3 dimensions, Energy Environ. Sci. 8 (2015) 702–730. doi:10.1039/C4EE03229B.
[10]
B. Mendoza-Sanchez, Y. Gogotsi, Synthesis of Two-Dimensional Materials for Capacitive Energy Storage, Adv. Mater. 28 (2016) 6104–6135. doi:10.1002/adma.201506133.
[12]
[13]
SC
M AN U
TE D
EP
AC C
[11]
RI PT
[1]
W. Choi, I. Lahiri, R. Seelaboyina, Y.S. Kang, Synthesis of Graphene and Its Applications: A Review, Crit. Rev. Solid State Mater. Sci. 35 (2010) 52–71. doi:10.1080/10408430903505036.
M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi, M.W. Barsoum, Two-dimensional nanocrystals produced by exfoliation of Ti3AlC2, Adv. Mater. 23 (2011) 4248–4253. doi:10.1002/adma.201102306. O. Mashtalir, M.R. Lukatskaya, A.I. Kolesnikov, E. Raymundo-Piñero, M. Naguib, M.W. Barsoum, Y. Gogotsi, The effect of hydrazine intercalation on the structure and capacitance of 2D titanium carbide (MXene), Nanoscale. 8 (2016) 9128–9133.
13
ACCEPTED MANUSCRIPT
doi:10.1039/C6NR01462C. M. Boota, B. Anasori, C. Voigt, M.Q. Zhao, M.W. Barsoum, Y. Gogotsi, Pseudocapacitive Electrodes Produced by Oxidant-Free Polymerization of Pyrrole between the Layers of 2D Titanium Carbide (MXene), Adv. Mater. 28 (2016) 1517– 1522. doi:10.1002/adma.201504705.
[15]
Z. Lin, D. Barbara, P.L. Taberna, K.L. Van Aken, B. Anasori, Y. Gogotsi, P. Simon, Capacitance of Ti3C2Tx MXene in ionic liquid electrolyte, J. Power Sources. 326 (2016) 575–579. doi:10.1016/j.jpowsour.2016.04.035.
[16]
J. Come, Y. Xie, M. Naguib, S. Jesse, S. V. Kalinin, Y. Gogotsi, P.R.C. Kent, N. Balke, Nanoscale Elastic Changes in 2D Ti3C2Tx (MXene) Pseudocapacitive Electrodes, Adv. Energy Mater. 6 (2016) 1–9. doi:10.1002/aenm.201502290.
[17]
Y. Dall’Agnese, P. Rozier, P.L. Taberna, Y. Gogotsi, P. Simon, Capacitance of twodimensional titanium carbide (MXene) and MXene/carbon nanotube composites in organic electrolytes, J. Power Sources. 306 (2016) 510–515. doi:10.1016/j.jpowsour.2015.12.036.
[18]
M.Q. Zhao, C.E. Ren, Z. Ling, M.R. Lukatskaya, C. Zhang, K.L. Van Aken, M.W. Barsoum, Y. Gogotsi, Flexible MXene/carbon nanotube composite paper with high volumetric capacitance, Adv. Mater. 27 (2015) 339–345. doi:10.1002/adma.201404140.
[19]
Y. Tang, J. Zhu, C. Yang, F. Wang, Enhanced supercapacitive performance of manganese oxides doped two-dimensional titanium carbide nanocomposite in alkaline electrolyte, J. Alloys Compd. 685 (2016) 194–201. doi:10.1016/j.jallcom.2016.05.221.
[20]
M. Ghidiu, M.R. Lukatskaya, M.-Q. Zhao, Y. Gogotsi, M.W. Barsoum, Conductive two-dimensional titanium carbide “clay” with high volumetric capacitance, Nature. 516 (2014) 78–81. doi:10.1038/nature13970.
[21]
M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Dall’Agnese, P. Rozier, P.L. Taberna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Cation intercalation and high volumetric capacitance of two-dimensional titanium carbide, Sci. Mag. 341 (2013) 1502–1505. doi:10.1126/science.1241488.
[22]
C. Zhao, Q. Wang, H. Zhang, S. Passerini, X. Qian, Two-Dimensional Titanium Carbide/RGO Composite for High-Performance Supercapacitors, ACS Appl. Mater. Interfaces. 8 (2016) 15661–15667. doi:10.1021/acsami.6b04767.
[24]
SC
M AN U
TE D
EP
AC C
[23]
RI PT
[14]
Y. Gogotsi, P. Simon, True performance metrics in electrochemical energy storage., Sci. Mag. 334 (2011) 917–918. doi:10.1126/science.1213003.
M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Y. Dall’Agnese, P. Rozier, P.L. Taberna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Cation Intercalation and High Volumetric Capacitance of Two-Dimensional Titanium Carbide, Sci. Mag. 341 (2013) 1502–1505. doi:10.1126/science.1241488.
14
ACCEPTED MANUSCRIPT
Y. Tang, J. Zhu, C. Yang, F. Wang, Enhanced Capacitive Performance Based on Diverse Layered Structure of Two-Dimensional Ti3C2 MXene with Long Etching Time, J. Electrochem. Soc. 163 (2016) A1975–A1982. doi:10.1149/2.0921609jes.
[26]
M.Q. Zhao, C.E. Ren, Z. Ling, M.R. Lukatskaya, C. Zhang, K.L. Van Aken, M.W. Barsoum, Y. Gogotsi, Flexible MXene/carbon nanotube composite paper with high volumetric capacitance, Adv. Mater. 27 (2015) 339–345. doi:10.1002/adma.201404140.
[27]
D.C. Marcano, D. V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L.B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano. 4 (2010) 4806–4814. doi:10.1021/nn1006368.
[28]
X. Yang, C. Cheng, Y. Wang, L. Qiu, D. Li, Liquid-mediated dense integration of graphene materials for compact capacitive energy storage., Sci. Mag. 341 (2013) 534–7. doi:10.1126/science.1239089.
[29]
Y.J. Oh, J.J. Yoo, Y. Il Kim, J.K. Yoon, H.N. Yoon, J.H. Kim, S. Bin Park, Oxygen functional groups and electrochemical capacitive behavior of incompletely reduced graphene oxides as a thin-film electrode of supercapacitor, Electrochim. Acta. 116 (2014) 118–128. doi:10.1016/j.electacta.2013.11.040.
[30]
X. Gao, J. Jang, S. Nagase, Hydrazine and thermal reduction of graphene oxide: Reaction mechanisms, product structures, and reaction design, J. Phys. Chem. C. 114 (2010) 832–842. doi:10.1021/jp909284g.
[31]
P. Azaïs, L. Duclaux, P. Florian, D. Massiot, M.A. Lillo-Rodenas, A. Linares-Solano, J.P. Peres, C. Jehoulet, F. Béguin, Causes of supercapacitors ageing in organic electrolyte, J. Power Sources. 171 (2007) 1046–1053. doi:10.1016/j.jpowsour.2007.07.001.
[32]
D. Li, M.B. Müller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous dispersions of graphene nanosheets., Nat. Nanotechnol. 3 (2008) 101–105. doi:10.1038/nnano.2007.451.
[33]
M.J. Fernández-Merino, L. Guardia, J.I. Paredes, S. Villar-Rodil, P. SolísFernández, A. Martínez-Alonso, J.M.D. Tascón, Vitamin C is an ideal substitute for hydrazine in the reduction of graphene oxide suspensions, J. Phys. Chem. C. 114 (2010) 6426–6432. doi:10.1021/jp100603h.
[35]
[36]
SC
M AN U
TE D
EP
AC C
[34]
RI PT
[25]
A.M. Navarro-Suárez, N. Casado, J. Carretero-González, D. Mecerreyes, T. Rojo, Full-cell quinone/hydroquinone supercapacitors based on partially reduced graphite oxide and lignin/PEDOT electrodes, J. Mater. Chem. A. 5 (2017) 7137–7143. doi:10.1039/C7TA00527J.
Y. Wang, Y. Song, Y. Xia, Electrochemical capacitors: mechanism, materials, systems, characterization and applications, Chem. Soc. Rev. 45 (2016) 5925–5950. doi:10.1039/C5CS00580A. P. Srivastava, A. Mishra, H. Mizuseki, K.-R. Lee, A.K. Singh, Mechanistic Insight into the Chemical Exfoliation and Functionalization of Ti3C2 MXene, ACS Appl.
15
ACCEPTED MANUSCRIPT
Mater. Interfaces. 8 (2016) 24256–24264. doi:10.1021/acsami.6b08413. M. Naguib, O. Mashtalir, M.R. Lukatskaya, B. Dyatkin, C. Zhang, V. Presser, Y. Gogotsi, M.W. Barsoum, One-step synthesis of nanocrystalline transition metal oxides on thin sheets of disordered graphitic carbon by oxidation of MXenes, Chem. Commun. 50 (2014) 7420–7423. doi:10.1039/C4CC01646G.
[38]
K.J. Cai, Y. Zheng, P. Shen, S.Y. Chen, TiCx–Ti2C nanocrystals and epitaxial graphene-based lamellae by pulsed laser ablation of bulk TiC in vacuum, CrystEngComm. 16 (2014) 5466–5474. doi:10.1039/c4ce00358f.
[39]
T. Hu, J. Wang, H. Zhang, Z. Li, M. Hu, X. Wang, Vibrational properties of Ti3C2 and Ti3C2T2 (T = O, F, OH) monosheets by first-principles calculations: a comparative study, Phys. Chem. Chem. Phys. 17 (2015) 9997–10003. doi:10.1039/C4CP05666C.
[40]
J. Yan, C.E. Ren, K. Maleski, B. Christine, Y. Gogotsi, C.W.V. Gmbh, C. Kgaa, Flexible MXene / Graphene Films for Ultrafast Supercapacitors with Outstanding Volumetric Capacitance, Adv. Funct. Mater. (2017). doi:10.1002/adfm.201701264.
[41]
J.-C. Valmalette, Z. Tan, H. Abe, S. Ohara, Raman scattering of linear chains of strongly coupled Ag nanoparticles on SWCNTs., Sci. Rep. 4 (2014) 5238. doi:10.1038/srep05238.
[42]
S. Osswald, M. Havel, Y. Gogotsi, Monitoring oxidation of multiwalled carbon nanotubes by Raman spectroscopy, J. Raman Spectrosc. 38 (2007) 728–736. doi:10.1002/jrs.1686.
[43]
L.M. Malard, M.A. Pimenta, G. Dresselhaus, M.S. Dresselhaus, Raman spectroscopy in graphene, Phys. Rep. 473 (2009) 51–87. doi:10.1016/j.physrep.2009.02.003.
[44]
M.R. Lukatskaya, S.-M. Bak, X. Yu, X.-Q. Yang, M.W. Barsoum, Y. Gogotsi, Probing the Mechanism of High Capacitance in 2D Titanium Carbide Using In Situ X-Ray Absorption Spectroscopy, Adv. Energy Mater. 5 (2015) 1500589. doi:10.1002/aenm.201500589.
[45]
M.R. Lukatskaya, B. Dunn, Y. Gogotsi, Multidimensional materials and device architectures for future hybrid energy storage, Nat. Commun. 7 (2016) 12647. doi:10.1038/ncomms12647.
[47]
[48]
SC
M AN U
TE D
EP
AC C
[46]
RI PT
[37]
J. Chmiola, C. Largeot, P.L. Taberna, P. Simon, Y. Gogotsi, Desolvation of ions in subnanometer pores and its effect on capacitance and double-layer theory, Angew. Chemie - Int. Ed. 47 (2008) 3392–3395. doi:10.1002/anie.200704894.
J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, P.-L. Taberna, Anomalous Increase in Carbon Capacitance at Pore Sizes Less Than 1 Nanometer, Sci. Mag. 313 (2006) 1760–1763. doi:10.1126/science.1132195. P.L. Taberna, P. Simon, J.F. Fauvarque, Electrochemical Characteristics and Impedance Spectroscopy Studies of Carbon-Carbon Supercapacitors, J.
16
ACCEPTED MANUSCRIPT
Electrochem. Soc. 150 (2003) A292. doi:10.1149/1.1543948. L. Lai, H. Yang, L. Wang, K. Teh, J. Zhong, H. Chou, L. Chen, W. Chen, Z. Shen, R.S. Ruoff, J. Lin, Preparation of Supercapacitor Electrodes through Selection of Graphene Surface Functionalities, ACS Nano. 6 (2012) 5941–5951. doi:10.1021/nn3008096.
EP
TE D
M AN U
SC
RI PT
[49]
AC C
Figure 1. Cross-sectional SEM images of the Ti3C2Tx film (a), and of the Ti3C2Tx/MWCNT film (b); Raman (c) and XRD (d) of the Ti3C2Tx and he Ti3C2Tx/MWCNT films. Numbers in the XRD pattern indicate the miller indices related to the diffraction peaks.
17
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 2. Cross-sectional SEM image (a and b), Raman spectra (c) and XRD patterns (d) of the rGOH and rGOT films. Number in the XRD pattern indicates the miller index related to the diffraction peak.
-1
Figure 3. Cyclic voltammetry at different scan rates of the asymmetric device (a) and at 10 mV.s of the performance of each electrode in the asymmetric device (b) in 1 M H2SO4. Numbers and arrows indicate the voltage/potential windows.
18
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 4. Comparison of the capacitance retention (a and c) and Ragone plots (b and d) calculated gravimetrically (top) and volumetrically (bottom) of the asymmetric device with the symmetric devices in 1 M H2SO4.
19
ACCEPTED MANUSCRIPT
Table 1. Summary of the performance of d-Ti3C2Tx and rGOH in aqueous supercapacitors. CCell -1 (F.g )
CCell -3 (F.cm )
Retention at 100 mV/s (%)
Energy Density -1 (Wh.kg )
Symmetric Ti3C2Tx
0.55
54
220
50
2
Symmetric rGOH
0.9
42
41
33
5
Asymmetric Ti3C2Tx/rGOH
1.1
48
78
48
8
TE D
M AN U
SC
RI PT
Cell Potential (V)
Aqueous-Based Supercapacitors
AC C
EP
Figure 5. Cyclic voltammetry of the asymmetric device in 1 M Et4NBF4/ACN. Cell capacitance variation -1 with voltage at different scan rates (a) and behavior of each electrode with potential at 10 mV.s (b). Numbers and arrows indicate the voltage/potential windows.
20
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 6. Comparison of the capacitance retention of the asymmetric device with the symmetric ones in 1 M EtN4BF4 (a and c). Ragone plot comparing the aqueous- with the organic-based supercapacitor (b and d). Values calculated gravimetrically (top) and volumetrically (bottom).
21
ACCEPTED MANUSCRIPT
Table 2. Summary of the performance of Ti3C2Tx-CNT and rGOT in organic-based supercapacitors. CCell -1 (F.g )
CCell -3 (F.cm )
Retention at 100 mV/s (%)
Energy Density -1 (Wh.kg )
Symmetric Ti3C2Tx-MWCNT
1.8
7
16
37
3
Symmetric rGOT
2.5
19
16
47
16
Asymmetric Ti3C2Tx-MWCNT//rGOT
2.0
30
41
50
20
AC C
EP
TE D
M AN U
SC
Organic-Based Supercapacitors
RI PT
Cell Potential (V)
22
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
-1
AC C
Figure 7. Capacitance retention during 1000 cycles at 20 mV.s of the Ti3C2Tx/rGOH (a) and the Ti3C2Tx/MWCNT//rGOT (c). Electrochemical Impedance spectroscopy before and after cycling of the asymmetrical device in 1 M H2SO4 (b) and in 1 M Et4NBF4/ACN (d).
23