Chinese Chemical Letters  2018, Vol. 29 Issue (4): 624-628   PDF    
Boron and nitrogen dual-doped carbon as a novel cathode for high performance hybrid ion capacitors
Jiangmin Jiang, Ping Nie, Shan Fang, Yadi Zhang, Yufeng An, Ruirui Fu, Hui Dou, Xiaogang Zhang    
Jiangsu Key Laboratory of Materials and Technology for Energy Conversion, College of Material Science and Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
Abstract: Hybrid ion capacitors have been considered as a very attractive energy source with high energy density and power density since it combines both merits of lithium ion batteries and supercapacitors. However, their commercial application has been limited by the mismatch of charge-storage capacity and electrode kinetics between the capacitor-type cathode and battery-type anode. Herein, B and N dual-doped 3D superstructure carbon cathode is prepared through a facile template method. It delivers a high specific capacity, excellent rate capability and good cycling stability due to the B, N dual-doping, which has a profound effect in control the porosity, functional groups, and electronic conductivity for the carbon cathode. The hybrid ion capacitors using B, N dual-doping carbon cathode and prelithiated graphite anode show a high energy density of 115.5 Wh/kg at 250 W/kg and remain about 53.6 Wh/kg even at a high power density of 10 kW/kg. Additionally, the novel hybrid device achieves 76.3% capacity retention after 2000 cycles tested at 1250 W/kg power density. Significantly, the simultaneous manipulation of heteroatoms in carbon materials provides new opportunities to boost the energy and power density for hybrid ion capacitors.
Key words: Hybrid ion capacitors     Boron-doping     Nitrogen-doping     Dual-doped carbon     Template method     Energy density    

With the rapid development of portable electronics and hybrid electric vehicles, there has been an urgent need for advanced energy devices with both high energy and power density as well as a long cycle life. Among various energy storage systems (ESS), lithium ion batteries (LIBs) and supercapacitors (SCs) are considered to be very attractive energy devices and dominated the markets in our daily life [1-4]. LIBs can deliver high energy density by the faradic redox mechanism, however, the power densities are limited and cycling life is poor [5-7]. Although SCs can offer high power densities and long cycling life, a relative unsatisfactory energy density due to the physical adsorption/desorption mechanism has restricted their wide application in the automotive industry [8-10]. Therefore, designing a novel ESS which can simultaneously combined the advantages of both LIBs and SCs has become an important research direction in the field of energy storage.

Hybrid ion capacitors (HIC), a kind of asymmetric supercapacitor with a battery-supercapacitor hybrid energy storage mechanism, bridges the gap between LIBs and SCs and has attracted tremendous research interest in recent years [11-14]. Generally, the lithium ion capacitors (LICs) composed of a high capacity battery-type anode and a high rate capability capacitortype cathode in organic electrolytes containing Li salts. In the typical LICs, high surface area commercial activated carbon, Li+ intercalation compounds such as graphite [15, 16], Li4Ti5O12 [17, 18], TiO2 [19], Nb2O5 [20], Li3VO4 [21], hard carbon [22], soft carbon [23] and graphene [24, 25], are usually used as the cathode and anode materials, respectively. Owing to the Li+ redox reaction is much slower than the anion adsorption/desorption process, thus the power performance was determined by the anode materials. Numerous researches have paid attention to improve the rate capability of various anode materials [26-28]. Actually, the energy characteristic of LICs depends on the cathode, although the commercial activated carbon (AC) having a high surface area up to 3000 m2/g, unfortunately, the specific capacitance of AC is limited by its inherent defect. Thus, designing a novel carbon material with excellent comprehensive performance used for the LICs is very meaningful.

Recently, modifying the physical and chemical properties of carbon-based materials to improve energy storage performance by doped the heteroatoms has attracted attention. It is found that the nitrogen (N) atoms doping can not only enhance the capacity of active materials through pseudocapacitance contribution but also promote electronic conductivity through doping effect [29]. Furthermore, the wettability and contact of carbon surface with electrolyte could be improved by the N atoms doping [30, 31]. Besides that, the boron (B) atom is an electron-deficient alternative to N atoms, which has a similar function and effect [32, 33]. Nevertheless, to the best of our knowledge, still little attention has been paid to the B and N dual-doped carbon materials as cathode for LICs.

In this work, we used nano-CaCO3 as template and sucrose as carbon source to synthesis honeycomb-like structure carbon cathode. The N and B atoms could dope into the 3D structure carbon easily by in situ activate process with ammonium borate (NH4HB4O7·3H2O) as an additive reagent. It was demonstrated that the B, N dual-doping has a significant effect in tuning the porosity, functional groups, electrical conductivity of the carbon cathode, and a higher specific capacity as well as excellent cycling performance has been achieved. Besides, the LICs have been assembled by using the B, N dual-doped carbon as the cathode and pre-lithiation graphite as the anode. Benefit from the synergistic effect of the B, N dual doping, the novel LICs shows 115.5 Wh/kg at 250 W/kg and remains at 53.6 Wh/kg even at a high power density of 10 kW/kg with a stable cycle life, which delivers a superior electrochemical performance than by using conventional carbon cathode material.

The morphologies of commercial nano-CaCO3, CC and BNC were characterized by SEM as shown in Fig. 1 and Fig. S1 (Supporting information), respectively. The particle size of the nano-CaCO3 was ca. 40–200 nm and a majority of them with about 40 nm (Figs. S1a and b in Supporting information). It should be noted here that the nano-CaCO3 is very cheap template, which can be easily dispersed in carbon precursor and removed with diluted HCl instead of the corrosive HF acid [34]. In addition, the CaCO3 template could be used as activator during the carbonization process, which provides a very simple and easy preparation for mass production [35]. Both of the CC and BNC cathodes show a honeycomb-like structure. The nano-CaCO3 in the precursor matrix works as templates for the mesopores and macropores structure after the acid pickling process (Figs. 1a and b, Figs. S1c and d in Supporting information). In fact, there are a lots of micropores derived from the selfactivation process of nano-CaCO3, which can be demonstrated by TEM measurement. As shown in Fig. 1c, the BNC delivered a 3D cross-linked superstructure and the thin carbon nanosheets were uniform distribution. Moreover, the high-resolution TEM (HRTEM) (Fig. 1d) revealed that the BNC nanosheets consist of welldeveloped amorphous structure with partial graphitic layer (insert the red arrow). The hierarchical porous carbon materials are beneficial to ion rapid transmission and the B, N dual-doped can significantly increase the active sites, improving the electrode kinetics of the carbon cathode.

Download:
Fig. 1. SEM (a, b) and TEM images (c, d) of BNC cathode.

To further study the microstructure of the CC and BNC cathode, the XRD patterns were performed in Fig. 2a. Two broad peaks at around 24° and 43° can be observed from all the carbon cathodes, corresponding to the crystallographic planes of (002) and (100) in the disorder carbon structure, which were consistent with the HRTEM results. It needs to be emphasized that the diffraction peaks of BNC slightly shifted to larger angles, exhibiting relatively high levels of graphitic character [36]. In addition, Raman spectra of the CC, BNC-0.5, BNC and BNC-2 are shown in Fig. 2b and Fig. S2 (Supporting information), the D band at 1350 cm-1 is assigned to defective or disordered graphitic structures, while the G band at 1580 cm-1 corresponding to the orderly graphitic layers structures. The ID/IG ratio of BNC-0.5, BNC and BNC-2 are smaller than that of the CC, suggesting that the degree of graphitization has been enhanced, which can be also proved by the XRD results.

Download:
Fig. 2. (a) XRD patterns and (b) Raman spectra of the CC and BNC. High-resolution B 1s (c) and N 1s (d) XPS spectra of the BNC.

The XPS analysis was performed to confirm the incorporation of heteroatoms in the 3D honeycomb-like carbon material. The survey XPS spectrum demonstrated the presence of C 1s, O 1s, B 1s and N 1s without any other impurities (Fig. S3 in Supporting information), indicating the successful B and N atoms doping by this simple method. The XPS revealed that the atomic percentage of B and N in BNC was 4.43% and 6.99%, respectively. In addition, the doping amount could be changed by adjusting the concentration of NH4HB4O7·3H2O (Table S1 in Supporting information). The high-resolution of B 1s spectrum could be divided into two components at 190.8 and 191.9 eV (Fig. 2c), corresponding to B-N/ BC3 and B-C2O/B-CO2 bonds, respectively [37]. Additionally, the high-resolution N 1s spectrum can be deconvoluted into four different signals with binding energies of 398.5, 400.2, 401 and 402.5 eV (Fig. 2d), corresponding to pyridinic-N (N1), pyrrolic/ pyridone-N (N2), quaternary-N (N3) and pyridine-N-oxide (N4), respectively [38].

The N2 adsorption-desorption isotherms and pore size distributions of the CC and BNC cathode are shown in Figs. 3a, a1, and b, b1. The CC exhibited combination of type-I and type-IV isotherm according to IUPAC classification with an extremely steep adsorption at low relative pressure and a sharp capillary condensation step at very high relative pressures (P/P0 > 0.9), indicating lots of micropores and large pores (Fig. 3a) [39]. Interestingly, the porous feature of BNC was changed dramatically, which exhibited combination of type-Ⅱ and type-Ⅳ isotherm characteristic with an H1-type hysteresis loop, indicating mesoporous-rich structure (Fig. 3b) [40]. Brunauer-Emmett-Teller specific surface areas (SBET) of the CC and BNC were 1045.2 and 967.7 m2/g, respectively, indicating that the B, N dual-doping would not reduce the SBET of CC cathode obviously. It should be noted here that the NH4HB4O7·3H2O is a stable solid material at room temperature, however, it would be decomposition released ammonia during the high temperature carbonation process, which could be realize heteroatom doping and pore-making process at the same time. In order to obtain the superior carbon cathode with higher graphitization and appropriate doping level, we used the BNC as the optimal cathode material in this work. Additionally, mesoporous-rich of the BNC cathode is favorable for ion accumulation and rapid diffusion.

Download:
Fig. 3. The N2 adsorption/desorption isotherms and pore size distributions of CC (a and a1) and BNC (b and b1). (c) The rate capability of CC and BNC. (d) Cycling performance of the BNC.

The electrochemical performance of CC and BNC cathode were evaluated using coin cells (2–4.2 V). As shown in Fig. 3c, the initial discharge specific capacity of BNC was 75.2 mAh/g, which is larger than 62.5 mAh/g for CC cathode at a current density of 0.1 A/g. The higher initial discharge capacity of BNC could be attributed to the dual-doped of N, B atoms and internal defects structure, which has be corroborated in the XPS and Raman patterns. Importantly, the BNC cathode can still deliver a high capacity of 51.4 mAh/g even at a high current density of 5 A/g, demonstrating its superior rate performance over CC cathode (only 44.5 mAh/g). Furthermore, Fig. 3d presented good cycle stability with high capacity retention of as high as 99.5% after 1500 cycles at 1 A/g. Doping with heteroatoms has been well demonstrated as a robust and versatile route to further boost their electronic conductivity and capacitance [41-43]. Here, the B, N dual-doped could markedly tune the interaction sites and improve the wettability towards carbon matrix (Fig. S4 in Supporting information). Owing to the B, N dualdoped 3D superstructure and excellent electrochemical performance, BNC is more suitable than CC to use as an ideal cathode material for LICs.

We have systematically investigated the electrochemical performance of LICs by using graphite as the anode, CC and BNC as the cathode, respectively. Before assembling LICs, the graphite anode was prelithiated by using a fast and efficient internal short approach the same to previous work [16, 29]. Fig. 4a shows the charge storage mechanism of the novel LICs. During the charge process, Li+ intercalation occurs within the prelithiated graphite anode, meanwhile, PF6- adsorption arises on the surface of BNC cathode, while the discharge is the opposite process. As shown in Fig. 4b, the EIS spectra of both LICs were measured after charging to the voltage of 4 V (vs. Li/Li+). It can be found that the BNC-LICs by using B, N dual-doping cathode has the obviously reduced total impedance compared with the CC-LICs, indicating the heteroatom doping could improve the electronic conductivity of carbon materials. In addition, Fig. 4c and Fig. S5 (Supporting information) show the galvanostatic charge-discharge curves of BNC-LICs and CC-LICs, respectively. Both of the curves display symmetric quasitriangular shapes, indicating combination a hybrid energy storage mechanism. Owing the BNC cathode shows greater specific capacity, thus the BNC-LICs delivered high energy densities compared with the CC-LICs under the same current density, which can be demonstrated by the discharge time of both LICs.

Download:
Fig. 4. (a) The charge storage mechanism of the LICs (BNC//Graphite). (b) EIS spectra of CC-LICs and BNC-LICs at 4 V. (c) Galvanostatic charge/discharge curves of BNC-LICs. (d) Ragone plot of the CC-LICs and BNC-LICs compared with other LICs reported in the literature. (g) The capacity retention of CC-LICs and BNC-LICs during 2000 cycles.

Ragone plot of the BNC-LICs and CC-LICs devices were compared with previously reported LICs (Fig. 4d). It clearly shows that the BNC-LICs has more superior electrochemical performance. From the standard calculation formulas, the energy density of this novel BCN-LICs was about 115.5 Wh/kg at a power density of 250 W/kg. Even at a high power density of 10 kW/kg, the BNC-LICs can still deliver a high energy density of 53.6 Wh/kg. For comparison, the CC-LICs obtained lower energy densities with 114.2 Wh/kg and 48.3 Wh/kg at the power densities of 250 W/kg and 10 kW/kg, respectively. It demonstrated that the LICs by using B, N dual-doping cathode showing a high energy density and the difference of two-types LICs was more obviously with increasing of current density. In addition, energy density and power capability of the present BNC-LICs were superior to previously reported LICs, such as TiNb2O7//GO [44], Li4Ti5O12//AC [45], TiO2//GO [46] and V2O5//AC [47] devices. Furthermore, the cycling stability of the CC-LICs and BNC-LICs were investigated at the power density of 1250 W/kg. As shown in Fig. 4e, the CC-LICs and BNC-LICs delivered capacity retentions of 71.6% and 59.3% after 2000 cycles, respectively. It should be noted here that the B, N dual-doping could increase the electronic conductively and improve the interfacial properties of carbon cathode but without reduce its stability during the cycling, indicating the heteroatom doping is an effective approach to modify the carbon materials.

In summary, we have successfully designed B, N dual-doped honeycomb-like BNC via a facile and controllable strategy. It has been revealed that B and N dual-doping results in improved graphitization, charge-storage active sites and electronic conductivity. As a consequence, the BNC electrode delivered a high specific capacity, excellent rate capability, good cycling stability and simultaneously reduced the mismatch of electrode kinetics between the cathode and anode for LICs. Additionally, the B and N dual-doped carbon cathode used for LICs configuration shows a high energy density of 115.5 Wh/kg, which can remains about 53.6 Wh/kg even at a high power density of 10 kW/kg. Significantly, simultaneous manipulation of heteroatoms in carbon materials provides new opportunities to boost the energy and power density for novel LICs.

Acknowledgments

We gratefully acknowledge financial support from the National Program on Key Basic Research Project of China (No. 2014CB239701), the National Natural Science Foundation of China (Nos. 51372116, 51672128, 21773118), Prospective Joint Research Project of Cooperative Innovation Fund of Jiangsu Province (No. BY2015003-7), and Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2018.01.029.

References
[1]
J.B. Goodenough, Energy Environ. Sci. 7 (2014) 14-18. DOI:10.1039/C3EE42613K
[2]
N.S. Choi, Z. Chen, S.A. Freunberger, et al., Angew. Chem. Int. Ed. 51 (2012) 9994-10024. DOI:10.1002/anie.201201429
[3]
G. Jeong, Y.U. Kim, H. Kim, Y.J. Kim, H.J. Sohn, Energy Environ. Sci. 4 (2011) 1986-2002. DOI:10.1039/c0ee00831a
[4]
P. Nie, X. Liu, R. Fu, et al., ACS Energy Lett. 2 (2017) 1279-1287. DOI:10.1021/acsenergylett.7b00286
[5]
[6]
J. Jiang, P. Nie, B. Ding, et al., ACS Appl. Mater. Interfaces 8 (2016) 30926-30932. DOI:10.1021/acsami.6b10038
[7]
P. Nie, J. Yuan, J. Wang, et al., ACS Appl. Mater. Interfaces 9 (2017) 20306-20312. DOI:10.1021/acsami.7b05178
[8]
H. Wang, Z. Xu, A. Kohandehghan, et al., ACS Nano 7 (2013) 5131-5141. DOI:10.1021/nn400731g
[9]
B. Duan, X. Gao, X. Yao, et al., Nano Energy 27 (2016) 482-491. DOI:10.1016/j.nanoen.2016.07.034
[10]
J. Qin, Z.S. Wu, F. Zhou, et al., Chin. Chem. Lett. 29 (2018) 582-586. DOI:10.1016/j.cclet.2017.08.007
[11]
V. Aravindan, J. Gnanaraj, Y.S. Lee, S. Madhavi, Chem. Rev. 114 (2014) 11619-11635. DOI:10.1021/cr5000915
[12]
K. Naoi, S. Ishimoto, J.I. Miyamoto, W. Naoi, Energy Environ. Sci. 5 (2012) 9363-9373. DOI:10.1039/c2ee21675b
[13]
D.P. Dubal, O. Ayyad, V. Ruiz, P. Gómez-Romero, Chem. Soc. Rev. 44 (2015) 1777-1790. DOI:10.1039/C4CS00266K
[14]
W. Zuo, R. Li, C. Zhou, et al., Adv. Sci. 4 (2017) 1600539. DOI:10.1002/advs.v4.7
[15]
S.R. Sivakkumar, A.G. Pandolfo, Electrochim. Acta 65 (2012) 280-287. DOI:10.1016/j.electacta.2012.01.076
[16]
J. Jiang, P. Nie, S. Dong, Y. Wu, X. Zhang, Acta Phys.-Chim. Sin. 33 (2017) 780-786.
[17]
L. Ye, Q. Liang, Y. Lei, et al., J. Power Sources 282 (2015) 174-178. DOI:10.1016/j.jpowsour.2015.02.028
[18]
R. Gokhale, V. Aravindan, P. Yadav, et al., Carbon 80 (2014) 462-471. DOI:10.1016/j.carbon.2014.08.086
[19]
Z. Le, F. Liu, P. Nie, et al., ACS Nano 11 (2017) 2952-2960. DOI:10.1021/acsnano.6b08332
[20]
H. Li, Y. Zhu, S. Dong, et al., Chem. Mater. 28 (2016) 5753-5760. DOI:10.1021/acs.chemmater.6b01988
[21]
L. Shen, H. Lv, S. Chen, et al., Adv. Mater. 29 (2017) 1700142. DOI:10.1002/adma.v29.27
[22]
W.J. Cao, J.P. Zheng, J. Power Sources 213 (2012) 180-185. DOI:10.1016/j.jpowsour.2012.04.033
[23]
M. Schroeder, M. Winter, S. Passerini, A. Balducci, J. Power Sources 238 (2013) 388-394. DOI:10.1016/j.jpowsour.2013.04.045
[24]
Y. Ma, H. Chang, M. Zhang, Y. Chen, Adv. Mater. 27 (2015) 5296-5308. DOI:10.1002/adma.201501622
[25]
W. Ahn, D.U. Lee, G. Li, et al., ACS Appl. Mater. Interfaces 8 (2016) 25297-25305. DOI:10.1021/acsami.6b08298
[26]
F. Zhang, Y. Tang, H. Liu, et al., ACS Appl. Mater. Interfaces 8 (2016) 4691-4699. DOI:10.1021/acsami.5b11705
[27]
R. Thangavel, B. Moorthy, D.K. Kim, Y.S. Lee, Adv. Energy Mater. 7 (2017) 1602654. DOI:10.1002/aenm.201602654
[28]
X. Han, P. Han, J. Yao, et al., Electrochim. Acta 196 (2016) 603-610. DOI:10.1016/j.electacta.2016.02.185
[29]
J. Jiang, P. Nie, B. Ding, et al., J. Mater. Chem. A 5 (2017) 23283-23291. DOI:10.1039/C7TA05972H
[30]
K. Huo, W. An, J. Fu, et al., J. Power Sources 324 (2016) 233-238. DOI:10.1016/j.jpowsour.2016.05.084
[31]
C. Xuan, Z. Peng, J. Wang, et al., Chin. Chem. Lett. 28 (2017) 2227-2230. DOI:10.1016/j.cclet.2017.09.009
[32]
J. Han, L.L. Zhang, S. Lee, et al., ACS Nano 7 (2013) 19-26. DOI:10.1021/nn3034309
[33]
J. Zhu, C. He, Y. Li, S. Kang, P.K. Shen, J. Mater. Chem. A 1 (2013) 14700-14705. DOI:10.1039/c3ta13318d
[34]
B. Xu, L. Peng, G. Wang, G. Cao, F. Wu, Carbon 48 (2010) 2377-2380. DOI:10.1016/j.carbon.2010.03.003
[35]
G. Yang, H. Han, T. Li, C. Du, Carbon 50 (2012) 3753-3765. DOI:10.1016/j.carbon.2012.03.050
[36]
J.P. Paraknowitsch, J. Zhang, D. Su, A. Thomas, M. Antonietti, Adv. Mater. 22 (2010) 87-92. DOI:10.1002/adma.v22:1
[37]
Q. Xia, H. Yang, M. Wang, et al., Adv. Energy Mater. 7 (2017) 1701336. DOI:10.1002/aenm.201701336
[38]
Z. Wang, Y. Xiong, S. Guan, Electrochim. Acta 188 (2016) 757-766. DOI:10.1016/j.electacta.2015.12.048
[39]
B. Xu, L. Shi, X. Guo, et al., Electrochim. Acta 56 (2011) 6464-6468. DOI:10.1016/j.electacta.2011.04.130
[40]
X. Yu, C. Zhan, R. Lv, et al., Nano Energy 15 (2015) 43-53. DOI:10.1016/j.nanoen.2015.03.001
[41]
C. Wang, F. Wang, Z. Liu, et al., Nano Energy 41 (2017) 674-680. DOI:10.1016/j.nanoen.2017.10.025
[42]
J. Jiang, Y. Zhang, P. Nie, et al., Adv. Sustain. Syst.(2017). DOI:10.1002/adsu.201700110
[43]
P. Nie, G. Xu, J. Jiang, et al., Small Methods(2017). DOI:10.1002/smtd.201700272
[44]
H. Li, L. Shen, J. Wang, et al., J. Mater. Chem. A 3 (2015) 16785-16790. DOI:10.1039/C5TA02929E
[45]
A. Banerjee, K.K. Upadhyay, D. Puthusseri, et al., Nanoscale 6 (2014) 4387-4394. DOI:10.1039/c4nr00025k
[46]
H. Kim, M.Y. Cho, M.H. Kim, et al., Adv. Energy Mater. 3 (2013) 1500-1506. DOI:10.1002/aenm.v3.11
[47]
Z. Chen, V. Augustyn, J. Wen, et al., Adv. Mater. 23 (2011) 791-795. DOI:10.1002/adma.201003658