- Email: [email protected]
Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities
Accepted Manuscript Title: Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities Author: Yoshiaki Matsuo Saeko Hirata Marc Dubois PII: DOI: Reference:
S0022-1139(16)30015-X http://dx.doi.org/doi:10.1016/j.jfluchem.2016.01.015 FLUOR 8724
To appear in:
FLUOR
Received date: Revised date: Accepted date:
3-10-2015 24-1-2016 25-1-2016
Please cite this article as: Yoshiaki Matsuo, Saeko Hirata, Marc Dubois, Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities, Journal of Fluorine Chemistry http://dx.doi.org/10.1016/j.jfluchem.2016.01.015 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.
Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities
Yoshiaki Matsuo,1,* Saeko Hirata1 and Marc Dubois2
1
Department of Materials Science and Chemistry, School of Engineering, University of
Hyogo, 2167 Shosha, Himeji, Hyogo, 671-2280, Japan 2
Institut de Chimie de Clermont-Ferrand (ICCF), Université Blaise Pascal, UMR
CNRS 6296, 24 Avenue Blaise Pascal, 63177, Aubiére, France
*E-mail [email protected]
Graphical Abstract
Stage 1 type oxygenated CxF with an interlayer spacing of 0.55 nm was obtained by the electrochemical oxidation of graphite in 47% HF aqueous solution at a current density of 200 mAcm-2 or hgiher.
Highlights
Electrochemical oxidation of graphite was performed at high current densities in 47% HF aqueous solution.
Covalent C-F bonging formed above 2.4 V vs Pb/PbF2.
Stage 1 type oxygenated fluorine-graphite intercalation compound with an interlayer spacing of 0.55 nm was obtained.
This material showed a high capacity of 550 mAh/g as a cathode of lithium primary battery.
Abstract Electrochemical oxidation of graphite was performed at high current densities in 47% HF aqueous solution. Cyclic voltammetry indicated that covalent C-F bonding formed above 2.4 V vs Pb/PbF2. The sample obtained at higher current densities than 200 mAcm-2 was stage 1 type material with an interlayer spacing of 0.55nm and it contained a considerable amount of oxygen, together with the covalently bonded fluorine. The discharge profile of this sample as a cathode of lithium primary battery was similar to that of C2.5F prepared under F2 gas atmosphere and the capacity reached 550 mAhg-1. This strongly indicated that not only fluorine but also oxygen in this sample was utilized.
Keywords: Hydrofluoric acid; Graphite; Intercalation; Electrochemical oxidation; Lithium primary battery
1. Introduction Graphite intercalation compounds of fluorine (hereafter abbreviated as CxF; x shows the carbon/fluorine ratio) are attracting much attention because of their unique carbon-fluorine bonding, c-axis structure and electrochemical properties, especially for the cathode materials of lithium primary battery with high voltage and energy density [1-4]. In CxF, fluorine atoms are ionically bonded to carbon at lower fluorine contents, however, they are covalently bonded when the fluorine contents increase. Interestingly, when the XPS measurement is performed for these materials, the peak position of the binding energy of F1s electron in CxF with larger fluorine contents locate between those observed for CxF with lower fluorine contents and graphite fluorides. This C-F bond also provides shifted a 19F NMR peak [5]. Therefore, it is usually called as “semi-ionic” or “semi-covalent” C-F bonding, though the bond length of it has been estimated as 0.14 nm [6] which is the same as that of covalent one. The slightly lower C-F bond order in CxF than those in poly(carbon monofluoride) ((CF)n) and poly(dicarbon monofluoride) ((C2F)n) is explained by the hyperconjugation involving the C-C bonds on the carbon sheets and C-F bond [6] and this seems responsible for the change in the F1s binding energy. Reflecting this change in the nature of C-F bonding depending on the fluorine contents various c-axis repeat distances have been observed.
Concerning the cathode of lithium primary battery, covalent type (CF)n (or it is also denoted as CFx.) has been commercialized, together with manganese oxide, however, it suffers from the relatively low discharge voltage and potential drop at the beginning of discharge. Recently it has been reported that the low discharge voltage can be somewhat improved by increasing the surface area of the sample by ball milling but it is still 2.5 V [7]. The discharge voltage of CxF is higher than graphite fluorides exceeding 3 V and no potential drop is observed probably because of its higher conductivity and the nature of C-F bonding. However, the capacity of it still lower than that of graphite fluoride, therefore, preparation of CxF with higher fluorine content has been extensively studied. In order to obtain CxF with high fluorine content, solid-gas reaction using fluorine gas in the presence of catalysts such as solid metal fluorides, gaseous halogen fluorides, etc has been usually employed [8-20]. Among them by using KAgF4 under high F2 gas pressure in anhydrous HF solution, CxF with the minimum x value (highest fluorine content) of 1.2 has been prepared [18]. The discharge capacity of CxF samples can reached more than 600 mAhg-1 with the discharge voltage higher than 3.0 V [14, 16, 18, 20]. Fluorine-intercalated graphite can be also prepared by the electrochemical oxidation of it in aqueous hydrofluoric acid solutions [21]. However, it was difficult to obtain CxF with high fluorine content was not obtained because of the
narrow electrochemical window of these solutions. Before reaching the potential to obtain highly oxidized graphite, oxygen evolution occur as the result of oxidation of water, which prevented CxF with high fluorine contents. Therefore, in our previous paper, we have electrochemically oxidized graphite in 47% aqueous solution of HF at high current densities in order to reach the potential needed to oxidize graphite at higher levels by increasing the overpotential for oxygen evolution reaction [22]. Highly fluorinated stage 1 “bi-intercalation type” [23], C2.8F with an interlayer spacing of 1.15 nm was obtained when graphite was oxidized at 100 mAcm-2. Moreover, a new covalent type stage 2 CxF was also successfully synthesized. In this study, graphite was oxidized at higher current densities than 100 mAcm-2 in order to obtain more highly oxidized CxF samples. The resulting materials were characterized in detail and the electrochemical performance of them was investigated.
2. Experimental Graphite sheet electrode (PERMA-FOIL; PF type, Toyo Tanso Co. Ltd., d(002)=0.3356 nm) was used as a working electrode. As a reference electrode, the Pb/PbF2 couple was prepared by immersing Pb wire in 47% HF aqueous solution overnight. Pt plate was used as a counter electrode. Cyclic voltammetry was performed at a potential scan rate
of 0.5 mVs-1 from OCV to various upper limit potentials of 2.2, 2.4 and 3.0 V vs Pb/PbF2 (-0.344 V vs NHE). Constant current oxidation was performed at current densities of 5 – 800 mAcm-2. The resulting materials are analyzed by X-ray diffraction (Rint-2100, CuKα) and X-ray photoelectron spectroscopy (Ulvac Phi-5000 MgKα) measurements. NMR measurements were performed using a BRUCKER AVANCE spectrometer, with working frequencies for
13
C, 1H, and
19
F of 73.4, 300.1 and
282.2 MHz, respectively. A magic angle spinning (MAS) probe (Bruker) operating with a 4 mm rotor was used. For MAS spectra, a simple sequence was performed with a single π/2 pulse length of 3.5, 4 and 3.5 μs for 1H, 19F and 13C, respectively. 1H and 13C chemical shifts were externally referenced to tetramethylsilane (TMS).
19
F chemical
shifts were referenced with respect to CFCl3. Discharge characteristics of CxF samples were investigated by constant current discharge at a current density of 20 mAg-1 in 1M LiClO4-EC/DMC electrolyte solution. The active materials were mixed with acetylene black as a conducting additive and PVDF as a binder at a weight ratio of 8:1:1. This mixture was sandwiched by Ni mesh and the pressed at 200kgcm-2 for 5 min. As a counter electrode, Li metal was used. Stage 1 type CxF sample with x=2.5 (based on the weight change during reaction) was prepared by the reaction of graphite powder with the mixture of HF (1 atm) and F2
(1 atm) gases at room temperature. After 24 h of reaction, F2 gas was added to compensate the consumed amount of it and then the reaction was continued for 8 h. Finally, the reactor was evacuated at room temperature for 1 h and then at 100°C for 12 h. Graphite oxide (hereafter GO) was also synthesized by the method based on the Brodie’s one according to our previous study [24]. These were used as references.
3. Results and discussion Fig.1 shows the cyclic voltammogram of graphite sheet in 47% HF aqueous solution at the scan rate of 0.5 mV/s. The curve measured with the upper limit of 3.0 V vs Pb/PbF2 was almost the same as that reported in our previous study [22], considering that the difference of the potentials of reference electrodes, Pb/PbF2 and Cu/CuF2 couples. Three broad oxidation peaks at around 2.0, 2.38 and 2.6 V vs Pb/PbF2 were observed. On the other hand, only one reduction peak at 1.3 V vs Pb/PbF2 was observed, indicating that covalent C-F bonding formed. When the upper limit potential was 2.2 V vs Pb/PbF2, a reduction peak at 1.8 V vs Pb/PbF2 appeared, together with small one at 1.95 V vs Pb/PbF2, indicating that graphite electrode was reversibly oxidized and reduced. On the other hand, no reduction peak corresponding to that at 2.38 V vs Pb/PbF2 was not observed and this peak disappeared during the subsequent cycles. At
this potential, some irreversible reaction occurs. These results suggest that reversible intercalation/de-intercalation of HF2- occurs below 2.2 V and covalent type of CxF formed above 2.4 V vs Pb/PbF2. Fig.2 shows the variation of potential during electrochemical oxidation of graphite at various current densities. When the current density was 800 mAcm-2, the potential linearly increased and then reached a plateau after charged 0.1 Fmol-1 (C10+). In the other cases, two potential plateaus were observed. The second plateau started after graphite was oxidized to less than C5+ (0.2 Fmol-1), which was accompanied by the vigorous oxygen gas evolution. The potential values of the first and second plateaus are summarized in Fig.3. In case of the sample oxidized at 5 mAcm-2, as already reported, stage 2 type CxHF2- formed at the beginning of the electrochemical oxidation and as the increase in the charge passed through the cell, covalent type stage 2 CxF gradually formed [22]. This is well explained by the results obtained above. The first plateau of 2.17 V is lower than 2.2 V, therefore, ionic type material was formed. On the other hand, the second plateau of 2.71 V vs Pb/PbF2 is enough high to form covalent type stage 2 CxF, though considerable charge passed through the cell was consumed by the oxygen evolution. Considering that stage 1 ”bi-intercalation type” CxF with large interlayer spacings formed, when the current densities higher than 30 mAcm-2, after the formation
of covalent type stage 2 CxF at 2.4 V vs Pb/PbF2, further intercalation of fluorine occurred at 3.3 V vs Pb/PbF2. Fig.4 shows the X-day diffraction patterns of graphite after oxidized at the current densities of 200 and 800 mAcm-2, together with that of C2.5F. A diffraction peak at around 2θ=16.5° (d=0.54 nm) was observed for both samples and no peaks ascribed to the (003)-(006) lines of stage 1 “bi-intercalation type” CxF [22] were observed. The peak positions were slightly higher than that observed for C2.5F, however, the pattern was very similar. This suggests that further oxidation of stage 1 “bi-intercalation type” CxF proceeded and stage 1 type CxF with smaller interlayer spacings formed. No further change was observed even when the current density increased. Table 1 shows the elemental analysis data of the samples oxidized at 200 and 800 mAcm-2. The contents of fluorine were unexpectedly low and the F/C ratios were 0.18 and 0.14, corresponding to the compositions of C5.7F and C7.1F, which were lower than that observed for the samples prepared at lower current densities [22]. Instead, it is suggested that the sample contained a large amount of oxygen. It has been reported that graphite oxide in which oxygen atoms are covalently bonded to carbon atoms forms when graphite is electrochemically oxidized in concentrated oxoacids such as perchloric acid [25, 26]. Therefore, it would be reasonable to think that oxygen atoms are
introduced into CxF in the present system, though it has still not been clear if the oxygen atoms in GO are from oxoacids or water in the previously reported systems. The content of oxygen slightly increased for the sample prepared at 800 mAcm-2, while that of fluorine was lower. Fig.5 shows the XPS spectra of the stage 1 type CxF sample obtained at 200 and 800 mAcm-2 in C1s, F1s and O1s regions, together with those of C2.5F and GO. The peak height of the F1s peak of CxF sample was normalized by C1s peak at 285 eV. In C1s region of C2.5F, two peaks ascribed to the carbons bonded to carbon and fluorine atoms were observed at 284.5 and 287.6 eV, respectively [13]. On the other hand, in CxF samples, similar peaks were obtained at around 285 and 286 eV, however, both peaks seem to consist of two ones (284.2, 285.0, 286.8 and 287.6 eV). The peaks at the lowest and highest energies are rather similar to those of C2.5F. On the other hand, the other two peaks are very similar to those of GO, which are assigned to C-C and C-O units [27]. In O1s region, a peak at 532.5 eV with a considerable intensity was observed and this peak position is also similar to that of GO, which is assigned to C-O in hydroxyl, ether or epoxy groups. On the other hand, almost no peak was observed for C2.5F. The F1s peak was observed at 686.2 eV which was almost the same as that of C2.5F and the intensity was almost half of that of C2.5F, reflecting the lower fluorine content estimated from the elemental analysis. These also indicate that
considerable amounts of oxygen atoms are introduced in the sample, together with fluorine. Since the spectral feature of the samples prepared at different current densities of 200 and 800 mAcm-2 was almost identical, suggesting that the oxidation of graphite was almost completed. The nature of chemical bondings was further investigated by solid state NMR measurements for the sample obtained by oxidizing graphite at a current densities of 800 mAcm-2. 19F MAS spectrum was first recorded with various spinning rates in order to decrease both strong
19
F-19F homonuclear dipolar coupling and chemical shift
anisotropy as shown in Fig.6a. Two peaks at -147 and -170 ppm were mainly observed. A spinning speed of 4 kHz was sufficient to significant decrease of the linewidth of the resonance at -170 ppm, while narrowing was efficient at 7.5 kHz for the former line. These spectra were very similar to that observed for “(C2.5F)n” obtained by the fluorination of graphite at room temperature as previously reported and these lines are assigned to the covalent C-F bonds with lower covalency and F-, respectively [5, 6, 10, 28-32]. Fig.6b and c show the 1H and
13
C MAS spectra. One main line is observed at
1-2 ppm in the 1H spectrum, while C-O-C (with chemical shift of 60 ppm) [27, 33-36] and C-OH (70 ppm) [27, 33-34, 38, 39] are clearly observed in the 13C MAS spectrum, together with the line of sp2 carbons at 128 ppm. The absence of the line at 175 ppm due
to COOH groups in the 13C MAS spectrum indicates that the peak at 1-2 ppm in the 1H spectrum is ascribed to the C-OH groups (1.3 ppm) [36] and the presences of HF2species (13.5 ppm) and COOH (11.0 ppm) are also excluded. The line at 82 ppm in the 13
C MAS spectrum can be assigned to C-F bonds with weakened covalency [5, 6, 30-32,
37, 39], because of the co-existence with oxygenated groups. 19
F→13C cross-polarization was also performed. As expected, the C-F bond line at 82
ppm is favored with this sequence as shown in Fig.6d. Moreover, the line for sp2 carbon atoms
appears
because
of
the
interaction
with
neighboring
C-F
bonds
(hyperconjugation). Surprisingly, C-O-C line (at 60 ppm) is also observed, meaning that oxygenated groups are close enough from
19
F nuclei to be cross-polarized. Such an
observation evidences a homogenous distribution of both fluorinated and oxygenated groups. The narrow line width of sp2 carbon seems to evidence the interaction of all these atoms with either fluorinated or oxygenated groups. Based on these results, the sample prepared at a high current density of 200 and 800 mAcm-2 is not a simple CxF but oxygenated CxF. Considering the formation process of CxF during electrochemical oxidation of graphite, the structure might be expected as shown in Fig.7. The first one is the “bi-intercalation” type in which covalently bonded fluorine and oxygen atoms are alternatively inserted between carbon
layers. In the second type structure, both oxygen and fluorine are bonded to every carbon layer. If the oxidation of covalent type stage 2 CxF occurs after the formation of it is completed, the first type material can form. As shown above, the covalent type stage 2 CxF forms at the plateau of 2.5 V vs Pb/PbF2 shown in Fig.2. The potential started to increase when the carbon was oxidized to C10+, which is not sufficiently oxidized to form C5.7F or C7.1F as estimated from the elemental analysis. This means that oxidation and further fluorination occurred before the formation of covalent type stage 2 CxF was completed. In addition, since it is expected that the interlayer spacings of fluorine and oxygen containing layers are not identical, the interlayer distance should be the sum of the two layers. In such a case, additional (00ℓ) diffraction peaks should appear. Therefore, the second type structure seems more likely. The homogenous distribution of both fluorinated and oxygenated groups as revealed by the NMR study also support this conclusion. Covalent type stage 2 CxF with lower fluorine content was first formed and the additional fluorination and oxidation occurred simultaneously at the plateau above 3.5 V vs Pb/PbF2. The reason why the potential increased before the formation of fully fluorinated covalent type stage 2 CxF is not clear at this moment. However, considering that formation of covalent C-F bonding can reduce the electrical conductivity, the increase of resistance may explain
this phenomenon. Fig.8 shows the discharge curves of CxF sample prepared by oxidizing graphite at 200 mAcm-2 and C2.5F. The discharge voltage gradually decreased from 3.5 to 2.2 V and then it slightly more rapidly decreased. The potential drop at the beginning of discharge was not observed. This is very similar to that of C2.5F, except for the slightly lower discharge voltage at the beginning of the discharge. The capacity reached 550 mAhg-1, which was slightly higher than that observed for C2.5F (539 mAhg-1). This means that not only fluorine but also oxygen was utilized as expected from the previous studies on the electrochemical performance of GO [40-43]. The lower discharge voltage of the present sample than that of C2.5F would be ascribed to the existence of oxygen, based on the low discharge voltage of GO of 2-2.6 V [41-44]. In conclusion, stage 1 type oxygenated CxF with an interlayer spacing of 0.55 nm was obtained by the oxidation of graphite in 47% HF aqueous solution at high current density of 200 mAcm-2 or larger. Fluorine atoms are covalently bonded to carbons when graphite was oxidized above 2.4 V vs Pb/FbF2. The potential profile during reaction and the X-ray diffraction pattern indicated that both oxygen and fluorine are bonded to every carbon layer in the obtained sample. Moreover, NMR measurements indicated that oxygen functionalities are homogeniously distributed in the resulting samples.
When it is used for the cathode of lithium primary battery, the discharge voltage gradually decreased from 3.4 V without the drop at the beginning of discharge and the capacity reached 550 mAhg-1.
Acknowledge
The authors are grateful to Prof. R. Hagiwara of Kyoto University for supplying a CxF sample prepared under HF/F2 gas atmosphere. A part of this work was financially supported by the Research and Development Initiative for Scientific Innovation of New Generation Battery (RISING) Project under the auspices of the New Energy and Industrial Technology Development Organization (NEDO), Japan.
4. References [1] N. Watanabe, T. Nakajima and H. Touhara, Graphite Fluorides (1988) Elsevier, Amsterdam. [2] T. Nakajima and N. Watanabe, Graphite Fluorides and Carbon-Fluorine Compounds (1991) CRC Press, Boca Raton. [3] T. Nakajima, Fluorine-Carbon and Fluoride-Carbon Materials-Chemistry, Physics and Applications (T. Nakajima, ed.) (1995) Chap. 1, Marcel Dekker, New York. [4] T. Nakajima, Advanced Inorganic Fluorides-Synthesis, Characterization and Applications (T. Nakajima, B. Zemva and A. Tressaud, eds.) (2000) Chap. 15, Elsevier, Amsterdam. [5] J. Giraudet, M. Dubois, K. Gérin, C. Delabarre, A. Hamwi, F. Masin, J. Phys. Chem. B 111 (2007) 14143-14151. [6] Y. Sato, K. Itoh, R. Hagiwara; T. Fukunaga; Y. Ito, Carbon,42 (2004) 3243-3249. [7] M. Anji Reddy, B. Breitung, and M. Fichtner, ACS Appl. Mater. Interfaces, 5 (2013) 11207−11211. [8] T. Mallouk and N. Bartlett, J. Chem. Soc. Chem. Commun. (1983)103-104. [9] R. Hagiwara, M. Lerner and N. Bartlett, J. Chem. Soc. Chem. Commun. (1989) 573-574.
[10] T. Mallouk, B.L. Hawkins, M.P. Conard, K. Zilm, G.E. Maciel and N. Bartlett, Philos. Trans. R. Soc. Lond., Ser. A 314 (1985) 179-187. [11] J.P. Lemmon and M.M. Lerner, Carbon 31 (1993) 437-443. [12] Y. Sato, T. Kume, R. Hagiwara and Y. Ito, Carbon 41 (2003) 351-357. [13] Y. Sato, S. Shiraishi, Z. Mazej , R. Hagiwara and Y. Ito, Carbon 41 (2003) 1971-1977. [14] A. Hamwi, M. Daoud and J.C . Cousseins, Synth. Metals 30 (1989) 23-31. [15] A. Hamwi, J. Phys. Chem. Solids, 57 (1996) 677-688. [16] P. Hany, R. Yazami and A. Hamwi, J. Power Sources 68 (1997) 708-710. [17] R. Yazami, P. Hany, P. Masset and A. Hamwi , Mol. Cryst. Liq. Cryst. 310 (1998) 397-402. [18] T. Nakajima, M. Koh, V. Gupta, B. Žemva and K. Lutar, Electrochim. Acta 45 (2000) 1655-1661. [19] V. Gupta, T. Nakajima, Y. Ohzawa and B. Žemva , Mol. Cryst. Liq. Cryst. 386 (2002) 25-31. [20] T. Nakajima, V. Gupta, Y. Ohzawa, H. Groult, Z. Mazej, and B. Žemva, J. Power Sources, 137 (2004) 80-87. [21] H. Takenaka, M. Kawaguchi, M. Lerner, and N. Bartlett, J. Chem. Soc. Chem.
Commun., (1987) 1431-1432. [22] Y. Matsuo, M. Segawa, J. Mitani and Y. Sugie, J. Fluorine Chem., 87 (1998) 145-150 [23] T. Nakajima, M. Molinier, M. Motoyama, Carbon, 29 (1991) 429-437. [24] J. Maruyama, S. Akita, Y. Matsuo, Y. Muramatsu, Carbon, 66 (2014) 327-333. [25] J. O. Besenhard, H. P. Fritz, Z. anorg. allg. Chem., 416 (1975) 106-116. [26] T. Nakajima and Y. Matsuo, Carbon, 32 (1994) 469-475. [27] T. Szabó, O. Berkesi, P. Forgó, K. Josepovits, Y. Sanakis, D. Petridis, I. Dékány, Chem. Mater., 18 (2006) 2740-2749. [28] C. Delabarre, K. Guérin, M. Dubois, J. Giraudet,Z. Fawal, A. Hamwi, J. Fluorine Chem. 126 (2005) 1078-1087. [29] C. Delabarre, M. Dubois, K. Guérin, J. Giraudet,Z. Fawal, A. Hamwi, J. Phys. Chem. Solids, 67 (2006) 1157-1161. [30] A. M. Panich, Synth. Met., 100 (1999) 169-185. [31] H. Touhara, F. Okino, Carbon, 38 (2000) 241-246. [32] M. Dubois, J. Giraudet, K. Guérin, A. Hamwi, Z. Fawal, P. Pirotte, F. Masin, J. Phys. Chem. B, 110 (2006) 11800-11808. [33] C. Hontoria-Lucas, A.J. López-Peinado, J.de D. López-González, M.L.
Rojas-Cervantes, R.M. Martín-Aranda, Carbon, 33 (1995) 1585-1592. [34] H.-K. Jeong, Y. P. Lee, R. J. W. E. Lahaye, M.-H. Park, K. H. An, I. J. Kim, C-W Yang, C. Y. Park, R. S. Ruoff, Y. H. Lee, J. Am. Chem. Soc., 130 (2008) 1362-1366. [35] D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L. B. Alemany, W. Lu, J.S. M. Tour, ACS Nano, 4 (2010) 4806–4814. [36] D. R. Dreyer, S. Park, C. W. Bielawski , R. S. Ruoff, Chem. Soc. Rev., 39 (2010) 228–240. [37] W. Zhang, P. Bonnet, M. Dubois, C.P. Ewels, K. Guérin, E. Petit, J. -Y. Mevellec, L. Vidal, D. Ivanov, A. Hamwi, Chem. Mater., 24 (2012) 1744-1751. [38] W. Zhang, M. Dubois, K. Guérin, P. Bonnet, H. Kharbache, F. Masin, A. P. Kharitonov, A. Hamwi, Phys. Chem. Chem. Phys., 12 (2010) 1388-1398. [39] J. Giraudet, M. Dubois, A. Hamwi, W. E. E.Stone, P. Pirotte, F. Masin, J. Phys. Chem. B., 109 (2004) 175-181. [40] R. Yazami, Ph. Touzain, Y. Chabre, D. Berger, M. Coulon, Rev. Chim. Minérale, 22 (1985) 398-411. [41] R. Yazami, Ph. Touzain, Synth. Metals, 12 (1985) 499-503. [42] Ph. Touzain, R. Yazami, J. Power Sources, 14 (1985) 99-104. [43] M. Mermoux, R. Yazami, Ph. Touzain, J. Power Sources, 20 (1987) 105-110.
Fig.1 Cyclic voltammograms of graphite at a scan rate of 0.5 mVs-1 with various upper limit potentials of 2.2 (thick), 2.4 (broken) and 3.0 (thin) V. 0.2
Current / mA
0.15 0.1
0.05 0 0.5
1
1.5
2
2.5
3
-0.05
-0.1
Potential / V vs Pb/PbF2
Fig.2 Variation of the potential during constant current oxidation of 5, 50, 200 and 800 mAcm-2.
Fig.3 Relationship between the potential values of the first (triangle) or second (rectangle) plateaus and current densities.
Fig.4 X-ray diffraction patterns of graphite after oxidized at (A): 200 and (B): 800 mAcm-2, together with that of (C): C2.5F.
Fig.5 C1s, F1s and O1s XPS spectra of (A): C2.5F and (B): graphite after oxidized at 200 mAcm-2, together with that of (C): GO.
Normalized intensity
C1s (A) (B) (C)
(D)
282
284
286 288 Binding energy / eV
290
292
F1s Normalized intensity
(A) (B) (C)
680
685
690
695
Binding energy / eV
Normalized intensity
O1s
528
(D)
(B) (C) (A) 530
532 534 Binding energy / eV
536
538
Fig.6 NMR spectra of graphite after oxidized at 800 mAcm-2. The symbol of “*” indicates the spinning sideband. (A): 19F MAS, (B): 1H, (C): 13C and (D): 13C CP MAS. C-F
*
(A)
F-
*
14kHz
*
10kHz
* *
(B)
C-OH
* 7.5kHz
* 4kHz
100
0 -100 -200 -300 -400 Chemical shift from CFCl3 / ppm
100
* *
*
50 0 -50 Chemical shift from TMS / ppm C=C
(C) C=C
-100 (D)
C-F C-O-C C-OH C-F C-O-C
250
150
50
Chemical shift from TMS / ppm
-50
250
150
50
Chemical shift from TMS / ppm
-50
Fig.7 Schematic view along c-axis of stage 1 type oxygenated CxF containing covalently attached oxygen and fluorine.
(A)
fluorine
(B)
oxygen
Fig.7 Discharge characteristics of graphite electrochemically oxidized at 200 mAcm-2 (thick line) and C2.5F prepared under HF/F2 gas atmosphere (broken line).
Table 1 Elemental analysis data of CxF obtained at 200 and 800 mAcm-2 Current density /
H/%
C/%
F/%
O* / %
F/C
O/F
200
0.47
71.26
19.19
9.10
0.18
0.56
800
0.50
71.88
16.46
10.61
0.14
0.79
mAcm-2
*: balance
S0022-1139(16)30015-X http://dx.doi.org/doi:10.1016/j.jfluchem.2016.01.015 FLUOR 8724
To appear in:
FLUOR
Received date: Revised date: Accepted date:
3-10-2015 24-1-2016 25-1-2016
Please cite this article as: Yoshiaki Matsuo, Saeko Hirata, Marc Dubois, Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities, Journal of Fluorine Chemistry http://dx.doi.org/10.1016/j.jfluchem.2016.01.015 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.
Electrochemical oxidation of graphite in aqueous hydrofluoric acid solution at high current densities
Yoshiaki Matsuo,1,* Saeko Hirata1 and Marc Dubois2
1
Department of Materials Science and Chemistry, School of Engineering, University of
Hyogo, 2167 Shosha, Himeji, Hyogo, 671-2280, Japan 2
Institut de Chimie de Clermont-Ferrand (ICCF), Université Blaise Pascal, UMR
CNRS 6296, 24 Avenue Blaise Pascal, 63177, Aubiére, France
*E-mail [email protected]
Graphical Abstract
Stage 1 type oxygenated CxF with an interlayer spacing of 0.55 nm was obtained by the electrochemical oxidation of graphite in 47% HF aqueous solution at a current density of 200 mAcm-2 or hgiher.
Highlights
Electrochemical oxidation of graphite was performed at high current densities in 47% HF aqueous solution.
Covalent C-F bonging formed above 2.4 V vs Pb/PbF2.
Stage 1 type oxygenated fluorine-graphite intercalation compound with an interlayer spacing of 0.55 nm was obtained.
This material showed a high capacity of 550 mAh/g as a cathode of lithium primary battery.
Abstract Electrochemical oxidation of graphite was performed at high current densities in 47% HF aqueous solution. Cyclic voltammetry indicated that covalent C-F bonding formed above 2.4 V vs Pb/PbF2. The sample obtained at higher current densities than 200 mAcm-2 was stage 1 type material with an interlayer spacing of 0.55nm and it contained a considerable amount of oxygen, together with the covalently bonded fluorine. The discharge profile of this sample as a cathode of lithium primary battery was similar to that of C2.5F prepared under F2 gas atmosphere and the capacity reached 550 mAhg-1. This strongly indicated that not only fluorine but also oxygen in this sample was utilized.
Keywords: Hydrofluoric acid; Graphite; Intercalation; Electrochemical oxidation; Lithium primary battery
1. Introduction Graphite intercalation compounds of fluorine (hereafter abbreviated as CxF; x shows the carbon/fluorine ratio) are attracting much attention because of their unique carbon-fluorine bonding, c-axis structure and electrochemical properties, especially for the cathode materials of lithium primary battery with high voltage and energy density [1-4]. In CxF, fluorine atoms are ionically bonded to carbon at lower fluorine contents, however, they are covalently bonded when the fluorine contents increase. Interestingly, when the XPS measurement is performed for these materials, the peak position of the binding energy of F1s electron in CxF with larger fluorine contents locate between those observed for CxF with lower fluorine contents and graphite fluorides. This C-F bond also provides shifted a 19F NMR peak [5]. Therefore, it is usually called as “semi-ionic” or “semi-covalent” C-F bonding, though the bond length of it has been estimated as 0.14 nm [6] which is the same as that of covalent one. The slightly lower C-F bond order in CxF than those in poly(carbon monofluoride) ((CF)n) and poly(dicarbon monofluoride) ((C2F)n) is explained by the hyperconjugation involving the C-C bonds on the carbon sheets and C-F bond [6] and this seems responsible for the change in the F1s binding energy. Reflecting this change in the nature of C-F bonding depending on the fluorine contents various c-axis repeat distances have been observed.
Concerning the cathode of lithium primary battery, covalent type (CF)n (or it is also denoted as CFx.) has been commercialized, together with manganese oxide, however, it suffers from the relatively low discharge voltage and potential drop at the beginning of discharge. Recently it has been reported that the low discharge voltage can be somewhat improved by increasing the surface area of the sample by ball milling but it is still 2.5 V [7]. The discharge voltage of CxF is higher than graphite fluorides exceeding 3 V and no potential drop is observed probably because of its higher conductivity and the nature of C-F bonding. However, the capacity of it still lower than that of graphite fluoride, therefore, preparation of CxF with higher fluorine content has been extensively studied. In order to obtain CxF with high fluorine content, solid-gas reaction using fluorine gas in the presence of catalysts such as solid metal fluorides, gaseous halogen fluorides, etc has been usually employed [8-20]. Among them by using KAgF4 under high F2 gas pressure in anhydrous HF solution, CxF with the minimum x value (highest fluorine content) of 1.2 has been prepared [18]. The discharge capacity of CxF samples can reached more than 600 mAhg-1 with the discharge voltage higher than 3.0 V [14, 16, 18, 20]. Fluorine-intercalated graphite can be also prepared by the electrochemical oxidation of it in aqueous hydrofluoric acid solutions [21]. However, it was difficult to obtain CxF with high fluorine content was not obtained because of the
narrow electrochemical window of these solutions. Before reaching the potential to obtain highly oxidized graphite, oxygen evolution occur as the result of oxidation of water, which prevented CxF with high fluorine contents. Therefore, in our previous paper, we have electrochemically oxidized graphite in 47% aqueous solution of HF at high current densities in order to reach the potential needed to oxidize graphite at higher levels by increasing the overpotential for oxygen evolution reaction [22]. Highly fluorinated stage 1 “bi-intercalation type” [23], C2.8F with an interlayer spacing of 1.15 nm was obtained when graphite was oxidized at 100 mAcm-2. Moreover, a new covalent type stage 2 CxF was also successfully synthesized. In this study, graphite was oxidized at higher current densities than 100 mAcm-2 in order to obtain more highly oxidized CxF samples. The resulting materials were characterized in detail and the electrochemical performance of them was investigated.
2. Experimental Graphite sheet electrode (PERMA-FOIL; PF type, Toyo Tanso Co. Ltd., d(002)=0.3356 nm) was used as a working electrode. As a reference electrode, the Pb/PbF2 couple was prepared by immersing Pb wire in 47% HF aqueous solution overnight. Pt plate was used as a counter electrode. Cyclic voltammetry was performed at a potential scan rate
of 0.5 mVs-1 from OCV to various upper limit potentials of 2.2, 2.4 and 3.0 V vs Pb/PbF2 (-0.344 V vs NHE). Constant current oxidation was performed at current densities of 5 – 800 mAcm-2. The resulting materials are analyzed by X-ray diffraction (Rint-2100, CuKα) and X-ray photoelectron spectroscopy (Ulvac Phi-5000 MgKα) measurements. NMR measurements were performed using a BRUCKER AVANCE spectrometer, with working frequencies for
13
C, 1H, and
19
F of 73.4, 300.1 and
282.2 MHz, respectively. A magic angle spinning (MAS) probe (Bruker) operating with a 4 mm rotor was used. For MAS spectra, a simple sequence was performed with a single π/2 pulse length of 3.5, 4 and 3.5 μs for 1H, 19F and 13C, respectively. 1H and 13C chemical shifts were externally referenced to tetramethylsilane (TMS).
19
F chemical
shifts were referenced with respect to CFCl3. Discharge characteristics of CxF samples were investigated by constant current discharge at a current density of 20 mAg-1 in 1M LiClO4-EC/DMC electrolyte solution. The active materials were mixed with acetylene black as a conducting additive and PVDF as a binder at a weight ratio of 8:1:1. This mixture was sandwiched by Ni mesh and the pressed at 200kgcm-2 for 5 min. As a counter electrode, Li metal was used. Stage 1 type CxF sample with x=2.5 (based on the weight change during reaction) was prepared by the reaction of graphite powder with the mixture of HF (1 atm) and F2
(1 atm) gases at room temperature. After 24 h of reaction, F2 gas was added to compensate the consumed amount of it and then the reaction was continued for 8 h. Finally, the reactor was evacuated at room temperature for 1 h and then at 100°C for 12 h. Graphite oxide (hereafter GO) was also synthesized by the method based on the Brodie’s one according to our previous study [24]. These were used as references.
3. Results and discussion Fig.1 shows the cyclic voltammogram of graphite sheet in 47% HF aqueous solution at the scan rate of 0.5 mV/s. The curve measured with the upper limit of 3.0 V vs Pb/PbF2 was almost the same as that reported in our previous study [22], considering that the difference of the potentials of reference electrodes, Pb/PbF2 and Cu/CuF2 couples. Three broad oxidation peaks at around 2.0, 2.38 and 2.6 V vs Pb/PbF2 were observed. On the other hand, only one reduction peak at 1.3 V vs Pb/PbF2 was observed, indicating that covalent C-F bonding formed. When the upper limit potential was 2.2 V vs Pb/PbF2, a reduction peak at 1.8 V vs Pb/PbF2 appeared, together with small one at 1.95 V vs Pb/PbF2, indicating that graphite electrode was reversibly oxidized and reduced. On the other hand, no reduction peak corresponding to that at 2.38 V vs Pb/PbF2 was not observed and this peak disappeared during the subsequent cycles. At
this potential, some irreversible reaction occurs. These results suggest that reversible intercalation/de-intercalation of HF2- occurs below 2.2 V and covalent type of CxF formed above 2.4 V vs Pb/PbF2. Fig.2 shows the variation of potential during electrochemical oxidation of graphite at various current densities. When the current density was 800 mAcm-2, the potential linearly increased and then reached a plateau after charged 0.1 Fmol-1 (C10+). In the other cases, two potential plateaus were observed. The second plateau started after graphite was oxidized to less than C5+ (0.2 Fmol-1), which was accompanied by the vigorous oxygen gas evolution. The potential values of the first and second plateaus are summarized in Fig.3. In case of the sample oxidized at 5 mAcm-2, as already reported, stage 2 type CxHF2- formed at the beginning of the electrochemical oxidation and as the increase in the charge passed through the cell, covalent type stage 2 CxF gradually formed [22]. This is well explained by the results obtained above. The first plateau of 2.17 V is lower than 2.2 V, therefore, ionic type material was formed. On the other hand, the second plateau of 2.71 V vs Pb/PbF2 is enough high to form covalent type stage 2 CxF, though considerable charge passed through the cell was consumed by the oxygen evolution. Considering that stage 1 ”bi-intercalation type” CxF with large interlayer spacings formed, when the current densities higher than 30 mAcm-2, after the formation
of covalent type stage 2 CxF at 2.4 V vs Pb/PbF2, further intercalation of fluorine occurred at 3.3 V vs Pb/PbF2. Fig.4 shows the X-day diffraction patterns of graphite after oxidized at the current densities of 200 and 800 mAcm-2, together with that of C2.5F. A diffraction peak at around 2θ=16.5° (d=0.54 nm) was observed for both samples and no peaks ascribed to the (003)-(006) lines of stage 1 “bi-intercalation type” CxF [22] were observed. The peak positions were slightly higher than that observed for C2.5F, however, the pattern was very similar. This suggests that further oxidation of stage 1 “bi-intercalation type” CxF proceeded and stage 1 type CxF with smaller interlayer spacings formed. No further change was observed even when the current density increased. Table 1 shows the elemental analysis data of the samples oxidized at 200 and 800 mAcm-2. The contents of fluorine were unexpectedly low and the F/C ratios were 0.18 and 0.14, corresponding to the compositions of C5.7F and C7.1F, which were lower than that observed for the samples prepared at lower current densities [22]. Instead, it is suggested that the sample contained a large amount of oxygen. It has been reported that graphite oxide in which oxygen atoms are covalently bonded to carbon atoms forms when graphite is electrochemically oxidized in concentrated oxoacids such as perchloric acid [25, 26]. Therefore, it would be reasonable to think that oxygen atoms are
introduced into CxF in the present system, though it has still not been clear if the oxygen atoms in GO are from oxoacids or water in the previously reported systems. The content of oxygen slightly increased for the sample prepared at 800 mAcm-2, while that of fluorine was lower. Fig.5 shows the XPS spectra of the stage 1 type CxF sample obtained at 200 and 800 mAcm-2 in C1s, F1s and O1s regions, together with those of C2.5F and GO. The peak height of the F1s peak of CxF sample was normalized by C1s peak at 285 eV. In C1s region of C2.5F, two peaks ascribed to the carbons bonded to carbon and fluorine atoms were observed at 284.5 and 287.6 eV, respectively [13]. On the other hand, in CxF samples, similar peaks were obtained at around 285 and 286 eV, however, both peaks seem to consist of two ones (284.2, 285.0, 286.8 and 287.6 eV). The peaks at the lowest and highest energies are rather similar to those of C2.5F. On the other hand, the other two peaks are very similar to those of GO, which are assigned to C-C and C-O units [27]. In O1s region, a peak at 532.5 eV with a considerable intensity was observed and this peak position is also similar to that of GO, which is assigned to C-O in hydroxyl, ether or epoxy groups. On the other hand, almost no peak was observed for C2.5F. The F1s peak was observed at 686.2 eV which was almost the same as that of C2.5F and the intensity was almost half of that of C2.5F, reflecting the lower fluorine content estimated from the elemental analysis. These also indicate that
considerable amounts of oxygen atoms are introduced in the sample, together with fluorine. Since the spectral feature of the samples prepared at different current densities of 200 and 800 mAcm-2 was almost identical, suggesting that the oxidation of graphite was almost completed. The nature of chemical bondings was further investigated by solid state NMR measurements for the sample obtained by oxidizing graphite at a current densities of 800 mAcm-2. 19F MAS spectrum was first recorded with various spinning rates in order to decrease both strong
19
F-19F homonuclear dipolar coupling and chemical shift
anisotropy as shown in Fig.6a. Two peaks at -147 and -170 ppm were mainly observed. A spinning speed of 4 kHz was sufficient to significant decrease of the linewidth of the resonance at -170 ppm, while narrowing was efficient at 7.5 kHz for the former line. These spectra were very similar to that observed for “(C2.5F)n” obtained by the fluorination of graphite at room temperature as previously reported and these lines are assigned to the covalent C-F bonds with lower covalency and F-, respectively [5, 6, 10, 28-32]. Fig.6b and c show the 1H and
13
C MAS spectra. One main line is observed at
1-2 ppm in the 1H spectrum, while C-O-C (with chemical shift of 60 ppm) [27, 33-36] and C-OH (70 ppm) [27, 33-34, 38, 39] are clearly observed in the 13C MAS spectrum, together with the line of sp2 carbons at 128 ppm. The absence of the line at 175 ppm due
to COOH groups in the 13C MAS spectrum indicates that the peak at 1-2 ppm in the 1H spectrum is ascribed to the C-OH groups (1.3 ppm) [36] and the presences of HF2species (13.5 ppm) and COOH (11.0 ppm) are also excluded. The line at 82 ppm in the 13
C MAS spectrum can be assigned to C-F bonds with weakened covalency [5, 6, 30-32,
37, 39], because of the co-existence with oxygenated groups. 19
F→13C cross-polarization was also performed. As expected, the C-F bond line at 82
ppm is favored with this sequence as shown in Fig.6d. Moreover, the line for sp2 carbon atoms
appears
because
of
the
interaction
with
neighboring
C-F
bonds
(hyperconjugation). Surprisingly, C-O-C line (at 60 ppm) is also observed, meaning that oxygenated groups are close enough from
19
F nuclei to be cross-polarized. Such an
observation evidences a homogenous distribution of both fluorinated and oxygenated groups. The narrow line width of sp2 carbon seems to evidence the interaction of all these atoms with either fluorinated or oxygenated groups. Based on these results, the sample prepared at a high current density of 200 and 800 mAcm-2 is not a simple CxF but oxygenated CxF. Considering the formation process of CxF during electrochemical oxidation of graphite, the structure might be expected as shown in Fig.7. The first one is the “bi-intercalation” type in which covalently bonded fluorine and oxygen atoms are alternatively inserted between carbon
layers. In the second type structure, both oxygen and fluorine are bonded to every carbon layer. If the oxidation of covalent type stage 2 CxF occurs after the formation of it is completed, the first type material can form. As shown above, the covalent type stage 2 CxF forms at the plateau of 2.5 V vs Pb/PbF2 shown in Fig.2. The potential started to increase when the carbon was oxidized to C10+, which is not sufficiently oxidized to form C5.7F or C7.1F as estimated from the elemental analysis. This means that oxidation and further fluorination occurred before the formation of covalent type stage 2 CxF was completed. In addition, since it is expected that the interlayer spacings of fluorine and oxygen containing layers are not identical, the interlayer distance should be the sum of the two layers. In such a case, additional (00ℓ) diffraction peaks should appear. Therefore, the second type structure seems more likely. The homogenous distribution of both fluorinated and oxygenated groups as revealed by the NMR study also support this conclusion. Covalent type stage 2 CxF with lower fluorine content was first formed and the additional fluorination and oxidation occurred simultaneously at the plateau above 3.5 V vs Pb/PbF2. The reason why the potential increased before the formation of fully fluorinated covalent type stage 2 CxF is not clear at this moment. However, considering that formation of covalent C-F bonding can reduce the electrical conductivity, the increase of resistance may explain
this phenomenon. Fig.8 shows the discharge curves of CxF sample prepared by oxidizing graphite at 200 mAcm-2 and C2.5F. The discharge voltage gradually decreased from 3.5 to 2.2 V and then it slightly more rapidly decreased. The potential drop at the beginning of discharge was not observed. This is very similar to that of C2.5F, except for the slightly lower discharge voltage at the beginning of the discharge. The capacity reached 550 mAhg-1, which was slightly higher than that observed for C2.5F (539 mAhg-1). This means that not only fluorine but also oxygen was utilized as expected from the previous studies on the electrochemical performance of GO [40-43]. The lower discharge voltage of the present sample than that of C2.5F would be ascribed to the existence of oxygen, based on the low discharge voltage of GO of 2-2.6 V [41-44]. In conclusion, stage 1 type oxygenated CxF with an interlayer spacing of 0.55 nm was obtained by the oxidation of graphite in 47% HF aqueous solution at high current density of 200 mAcm-2 or larger. Fluorine atoms are covalently bonded to carbons when graphite was oxidized above 2.4 V vs Pb/FbF2. The potential profile during reaction and the X-ray diffraction pattern indicated that both oxygen and fluorine are bonded to every carbon layer in the obtained sample. Moreover, NMR measurements indicated that oxygen functionalities are homogeniously distributed in the resulting samples.
When it is used for the cathode of lithium primary battery, the discharge voltage gradually decreased from 3.4 V without the drop at the beginning of discharge and the capacity reached 550 mAhg-1.
Acknowledge
The authors are grateful to Prof. R. Hagiwara of Kyoto University for supplying a CxF sample prepared under HF/F2 gas atmosphere. A part of this work was financially supported by the Research and Development Initiative for Scientific Innovation of New Generation Battery (RISING) Project under the auspices of the New Energy and Industrial Technology Development Organization (NEDO), Japan.
4. References [1] N. Watanabe, T. Nakajima and H. Touhara, Graphite Fluorides (1988) Elsevier, Amsterdam. [2] T. Nakajima and N. Watanabe, Graphite Fluorides and Carbon-Fluorine Compounds (1991) CRC Press, Boca Raton. [3] T. Nakajima, Fluorine-Carbon and Fluoride-Carbon Materials-Chemistry, Physics and Applications (T. Nakajima, ed.) (1995) Chap. 1, Marcel Dekker, New York. [4] T. Nakajima, Advanced Inorganic Fluorides-Synthesis, Characterization and Applications (T. Nakajima, B. Zemva and A. Tressaud, eds.) (2000) Chap. 15, Elsevier, Amsterdam. [5] J. Giraudet, M. Dubois, K. Gérin, C. Delabarre, A. Hamwi, F. Masin, J. Phys. Chem. B 111 (2007) 14143-14151. [6] Y. Sato, K. Itoh, R. Hagiwara; T. Fukunaga; Y. Ito, Carbon,42 (2004) 3243-3249. [7] M. Anji Reddy, B. Breitung, and M. Fichtner, ACS Appl. Mater. Interfaces, 5 (2013) 11207−11211. [8] T. Mallouk and N. Bartlett, J. Chem. Soc. Chem. Commun. (1983)103-104. [9] R. Hagiwara, M. Lerner and N. Bartlett, J. Chem. Soc. Chem. Commun. (1989) 573-574.
[10] T. Mallouk, B.L. Hawkins, M.P. Conard, K. Zilm, G.E. Maciel and N. Bartlett, Philos. Trans. R. Soc. Lond., Ser. A 314 (1985) 179-187. [11] J.P. Lemmon and M.M. Lerner, Carbon 31 (1993) 437-443. [12] Y. Sato, T. Kume, R. Hagiwara and Y. Ito, Carbon 41 (2003) 351-357. [13] Y. Sato, S. Shiraishi, Z. Mazej , R. Hagiwara and Y. Ito, Carbon 41 (2003) 1971-1977. [14] A. Hamwi, M. Daoud and J.C . Cousseins, Synth. Metals 30 (1989) 23-31. [15] A. Hamwi, J. Phys. Chem. Solids, 57 (1996) 677-688. [16] P. Hany, R. Yazami and A. Hamwi, J. Power Sources 68 (1997) 708-710. [17] R. Yazami, P. Hany, P. Masset and A. Hamwi , Mol. Cryst. Liq. Cryst. 310 (1998) 397-402. [18] T. Nakajima, M. Koh, V. Gupta, B. Žemva and K. Lutar, Electrochim. Acta 45 (2000) 1655-1661. [19] V. Gupta, T. Nakajima, Y. Ohzawa and B. Žemva , Mol. Cryst. Liq. Cryst. 386 (2002) 25-31. [20] T. Nakajima, V. Gupta, Y. Ohzawa, H. Groult, Z. Mazej, and B. Žemva, J. Power Sources, 137 (2004) 80-87. [21] H. Takenaka, M. Kawaguchi, M. Lerner, and N. Bartlett, J. Chem. Soc. Chem.
Commun., (1987) 1431-1432. [22] Y. Matsuo, M. Segawa, J. Mitani and Y. Sugie, J. Fluorine Chem., 87 (1998) 145-150 [23] T. Nakajima, M. Molinier, M. Motoyama, Carbon, 29 (1991) 429-437. [24] J. Maruyama, S. Akita, Y. Matsuo, Y. Muramatsu, Carbon, 66 (2014) 327-333. [25] J. O. Besenhard, H. P. Fritz, Z. anorg. allg. Chem., 416 (1975) 106-116. [26] T. Nakajima and Y. Matsuo, Carbon, 32 (1994) 469-475. [27] T. Szabó, O. Berkesi, P. Forgó, K. Josepovits, Y. Sanakis, D. Petridis, I. Dékány, Chem. Mater., 18 (2006) 2740-2749. [28] C. Delabarre, K. Guérin, M. Dubois, J. Giraudet,Z. Fawal, A. Hamwi, J. Fluorine Chem. 126 (2005) 1078-1087. [29] C. Delabarre, M. Dubois, K. Guérin, J. Giraudet,Z. Fawal, A. Hamwi, J. Phys. Chem. Solids, 67 (2006) 1157-1161. [30] A. M. Panich, Synth. Met., 100 (1999) 169-185. [31] H. Touhara, F. Okino, Carbon, 38 (2000) 241-246. [32] M. Dubois, J. Giraudet, K. Guérin, A. Hamwi, Z. Fawal, P. Pirotte, F. Masin, J. Phys. Chem. B, 110 (2006) 11800-11808. [33] C. Hontoria-Lucas, A.J. López-Peinado, J.de D. López-González, M.L.
Rojas-Cervantes, R.M. Martín-Aranda, Carbon, 33 (1995) 1585-1592. [34] H.-K. Jeong, Y. P. Lee, R. J. W. E. Lahaye, M.-H. Park, K. H. An, I. J. Kim, C-W Yang, C. Y. Park, R. S. Ruoff, Y. H. Lee, J. Am. Chem. Soc., 130 (2008) 1362-1366. [35] D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L. B. Alemany, W. Lu, J.S. M. Tour, ACS Nano, 4 (2010) 4806–4814. [36] D. R. Dreyer, S. Park, C. W. Bielawski , R. S. Ruoff, Chem. Soc. Rev., 39 (2010) 228–240. [37] W. Zhang, P. Bonnet, M. Dubois, C.P. Ewels, K. Guérin, E. Petit, J. -Y. Mevellec, L. Vidal, D. Ivanov, A. Hamwi, Chem. Mater., 24 (2012) 1744-1751. [38] W. Zhang, M. Dubois, K. Guérin, P. Bonnet, H. Kharbache, F. Masin, A. P. Kharitonov, A. Hamwi, Phys. Chem. Chem. Phys., 12 (2010) 1388-1398. [39] J. Giraudet, M. Dubois, A. Hamwi, W. E. E.Stone, P. Pirotte, F. Masin, J. Phys. Chem. B., 109 (2004) 175-181. [40] R. Yazami, Ph. Touzain, Y. Chabre, D. Berger, M. Coulon, Rev. Chim. Minérale, 22 (1985) 398-411. [41] R. Yazami, Ph. Touzain, Synth. Metals, 12 (1985) 499-503. [42] Ph. Touzain, R. Yazami, J. Power Sources, 14 (1985) 99-104. [43] M. Mermoux, R. Yazami, Ph. Touzain, J. Power Sources, 20 (1987) 105-110.
Fig.1 Cyclic voltammograms of graphite at a scan rate of 0.5 mVs-1 with various upper limit potentials of 2.2 (thick), 2.4 (broken) and 3.0 (thin) V. 0.2
Current / mA
0.15 0.1
0.05 0 0.5
1
1.5
2
2.5
3
-0.05
-0.1
Potential / V vs Pb/PbF2
Fig.2 Variation of the potential during constant current oxidation of 5, 50, 200 and 800 mAcm-2.
Fig.3 Relationship between the potential values of the first (triangle) or second (rectangle) plateaus and current densities.
Fig.4 X-ray diffraction patterns of graphite after oxidized at (A): 200 and (B): 800 mAcm-2, together with that of (C): C2.5F.
Fig.5 C1s, F1s and O1s XPS spectra of (A): C2.5F and (B): graphite after oxidized at 200 mAcm-2, together with that of (C): GO.
Normalized intensity
C1s (A) (B) (C)
(D)
282
284
286 288 Binding energy / eV
290
292
F1s Normalized intensity
(A) (B) (C)
680
685
690
695
Binding energy / eV
Normalized intensity
O1s
528
(D)
(B) (C) (A) 530
532 534 Binding energy / eV
536
538
Fig.6 NMR spectra of graphite after oxidized at 800 mAcm-2. The symbol of “*” indicates the spinning sideband. (A): 19F MAS, (B): 1H, (C): 13C and (D): 13C CP MAS. C-F
*
(A)
F-
*
14kHz
*
10kHz
* *
(B)
C-OH
* 7.5kHz
* 4kHz
100
0 -100 -200 -300 -400 Chemical shift from CFCl3 / ppm
100
* *
*
50 0 -50 Chemical shift from TMS / ppm C=C
(C) C=C
-100 (D)
C-F C-O-C C-OH C-F C-O-C
250
150
50
Chemical shift from TMS / ppm
-50
250
150
50
Chemical shift from TMS / ppm
-50
Fig.7 Schematic view along c-axis of stage 1 type oxygenated CxF containing covalently attached oxygen and fluorine.
(A)
fluorine
(B)
oxygen
Fig.7 Discharge characteristics of graphite electrochemically oxidized at 200 mAcm-2 (thick line) and C2.5F prepared under HF/F2 gas atmosphere (broken line).
Table 1 Elemental analysis data of CxF obtained at 200 and 800 mAcm-2 Current density /
H/%
C/%
F/%
O* / %
F/C
O/F
200
0.47
71.26
19.19
9.10
0.18
0.56
800
0.50
71.88
16.46
10.61
0.14
0.79
mAcm-2
*: balance