Effects of electrolytes on the capacitive behavior of nitrogen/phosphorus co-doped nonporous carbon nanofibers: an insight into the role of phosphorus groups

Abstract

The capacitive properties of nitrogen/phosphorus co-doped nonporous carbon nanofibers and nitrogen doped nonporous carbon nanofibers are comprehensively and comparatively investigated in different aqueous electrolytes in order to identify the role of phosphorus groups in improving the capacitive performance of carbon. The introduction of phosphorus groups is favourable for the adsorption of electrolyte ions onto the carbon surface, especially protons, and thus greatly enhances the electric double layer capacitance.

---

The emerging need for high-power energy storage/release systems has promoted the development of supercapacitors, which can provide a higher power density as well as a much longer cycle life as compared to batteries.^1,2 Carbon materials, such as activated carbons, carbon nanofibers, carbon nanotubes, and graphene, have been the most widely studied electrode materials for supercapacitors because of their structural versatility.^3,4 Although tremendous efforts have been made to develop high-performance carbon electrodes, the energy density for carbon-based electrodes is still orders of magnitude lower than for batteries. Therefore, extensive and intensive research has been devoted to increasing the specific capacitance of carbon materials by introducing metal oxides,⁵ conducting polymers⁶ and/or functional groups (such as nitrogen and phosphorus groups)^7,8 into carbon materials. Transitional metal oxides and conducting polymers can provide a much larger pseudocapacitance than functional groups. Nevertheless, these materials experience a fast capacitance degradation during the charge–discharge processes, limiting their practical applications. Heteroatom functionalization of carbon materials will not deteriorate the cyclability, making it a practical way to develop high-capacitance carbon electrodes for commercial use.

Nitrogen and phosphorus co-doped carbons have recently attracted much attention due to their outstanding electrochemical performances.^9–13 For instance, Nasini et al. reported the preparation of a nitrogen/phosphorus co-doped mesoporous carbon which presented a small surface area of 479 m² g⁻¹ but a very high capacitance (271 F g⁻¹ in 1 M H₂SO₄ and 236 F g⁻¹ in 6 M KOH).¹³ Nitrogen groups have been proven to be electrochemically active groups which can offer a high pseudocapacitance through reversible redox reactions.^14,15 However, the role of phosphorus groups is not known exactly. Therefore, it is important to identify the true role of phosphorus groups in carbon-based electrodes for a future optimal design and fabrication of high-performance nitrogen/phosphorus co-doped carbon materials. Often, the co-existence of micropores and heteroatom groups makes it complex to study the true role of the heteroatom groups. Since we recently have synthesized nitrogen/phosphorus co-doped nonporous carbon nanofibers by electrospinning a precursor solution, containing polyacrylonitrile (PAN) and phosphoric acid, and subsequent thermal treatments,¹² the impact of the porosity on the capacitance can be avoided, making it possible to identify the role of phosphorus groups. Hence, the present work investigates the capacitive behavior of such nitrogen/phosphorus co-doped nonporous carbon nanofibers (N/P-NPCNFs) in different electrolytes (e.g. 1 M H₂SO₄, 0.5 M Li₂SO₄, 1 M Na₂SO₄ and 0.5 M K₂SO₄). Also, the capacitive behavior of pure nitrogen-doped nonporous carbon nanofibers (N-NPCNFs), derived from PAN, were studied in these electrolytes. A new insight into the role of phosphorus groups in enhancing the electric double layer (EDL) capacitance of carbon materials was provided.

Fig. 1 shows typical SEM images of the precursor nanofibers and the carbon nanofibers. All nanofibers demonstrate long, continuous fibrous morphologies. Obviously, the introduction of phosphoric acid greatly changed the physical properties of the precursor solution, leading to different surface morphologies and diameters for the precursor nanofibers (Fig. 1a and b). Pure PAN nanofibers possess a very smooth surface while the H₃PO₄–PAN composite nanofibers present a rough surface as well as a larger average diameter. After carbonization, the surface morphologies of N/P-NPCNFs and N-NPCNFs are both smooth (Fig. 1c and d), and the average diameter of N/P-NPCNFs is still much larger than that of N-NPCNFs.

Fig. 1 SEM images of (a) pure PAN nanofibers, (b) H₃PO₄–PAN composite nanofibers, (c) nitrogen-doped carbon nanofibers, and (d) nitrogen/phosphorus co-doped carbon nanofibers.

As proposed elsewhere,¹² the N1s spectra (Fig. 2a) can be deconvoluted into four different components: the peaks at 398.3, 400.4, 401.1 and 403.3 eV are ascribed to pyridinic nitrogen and the P=N bond, pyridonic/pyrrolic nitrogen (pyrrole-like nitrogen), quaternary nitrogen and the P–N bond, and pyridine-N-oxide, respectively.^16,17 It is clear that the content of the pyrrole-like nitrogen in the N/P-NPCNFs is much greater than that in the N-NPCNFs according to the XPS spectra. The O1s spectra can be divided into five regions (Fig. 2b), which represent C=O including quinones and non-bridging oxygen in the phosphate groups (P=O) (531.0 eV), oxygen single bonded to carbon in C–O and C–O–P groups (532.6 eV), oxygen single bonds in hydroxyl groups (533.6 eV), and carboxylic groups (–COOH) and/or water (535.0 eV).^18,19 The structure of the P2p peak (Fig. 2c) points to the presence of three major phosphorus groups differentiated by their binding energies: C–O–P groups (134.2 eV), C–PO₃ or C₂–PO₂ groups (133.1 eV), and C₃–P groups (132.2 eV).¹⁸ More details about the XPS analyses can be found in ref. 12.

Fig. 2 XPS spectra of N-NPCNFs and N/P-NPCNFs.

Fig. 3 shows the cyclic voltammograms of N-NPCNFs in different electrolytes at a scan rate of 10 mV s⁻¹. The cyclic voltammogram of the N-NPCNF electrode is similar to a triangular shape in 1 M H₂SO₄ (Fig. 3a), indicating a poor EDL behavior. However, the N-NPCNF electrode still exhibits a relatively high capacitance due to the pseudocapacitive interactions between the H⁺ ions and the heteroatom (nitrogen and oxygen) groups. Previous research showed that nitrogen-doped carbons store energy by a simple electrostatic interaction between electrolyte ions and charge at the electrode surface in neutral electrolytes.²⁰ Thus, a very poor capacitive behavior for the N-NPCNF electrode in 0.5 M Li₂SO₄ (Fig. 3b), 1 M Na₂SO₄ (Fig. 3c) and 0.5 M K₂SO₄ (Fig. 3d) was observed, showing tremendously distorted rectangular shapes and an extremely small capacitance, thus confirming the nonporous characteristics of N-NPCNFs.

Fig. 3 Cyclic voltammograms of N-NPCNFs obtained in different electrolytes at 10 mV s⁻¹.

Fig. 4 shows the cyclic voltammograms of N/P-NPCNFs in different electrolytes at a scan rate of 10 mV s⁻¹. It is worth noting that although the specific surface area of the N/P-NPCNFs is much smaller than that of the N-NPCNFs (see ref. 12), the N/P-NPCNF electrodes present greatly improved capacitive properties in all electrolytes due to the introduction of the phosphorus groups. The cyclic voltammogram of the N/P-NPCNF electrode is a nearly-rectangular shape in 1 M H₂SO₄ (Fig. 4a), indicating that the phosphorus groups mainly enhance the EDL capacitance. Obviously, no difference in the redox humps for N-NPCNFs and N/P-NPCNFs is observed in the H₂SO₄ electrolyte, implying that the phosphorus groups are not electrochemically active. In the neutral electrolytes, the N/P-NPCNF electrode shows the best capacitive behavior in 0.5 M K₂SO₄ with the cyclic voltammogram presenting a slightly distorted rectangular shape (Fig. 4d). Though the cyclic voltammogram of the N/P-NPCNF electrode in 0.5 M Li₂SO₄ is far from a rectangular shape (Fig. 4b), it is much better than that of the N-NPCNF electrode in 0.5 M Li₂SO₄. As it is expressed by the cyclic voltammogram in Fig. 4c, the capacitive behavior of the N/P-NPCNF electrode in 1 M Na₂SO₄ falls in between those in 0.5 M Li₂SO₄ and 0.5 M K₂SO₄. These results suggest that the phosphorus-functionalized carbon surface is very attractive to the electrolyte ions due to the presence of the oxygen-rich phosphorus groups.

Fig. 4 Cyclic voltammograms of N/P-NPCNFs obtained in different electrolytes at 10 mV s⁻¹.

It is considered that the galvanostatic charge–discharge measurement is a more accurate technique to estimate the capacitance, especially the pseudocapacitance. Therefore, specific capacitances of the N-NPCNFs and the N/P-NPCNFs in different electrolytes as a function of current density are plotted in Fig. 5a and b, respectively. The specific capacitance was calculated from the galvanostatic charge–discharge profiles based on the following equation:

C = iΔt/mΔV (1)

where C (F g⁻¹) is the specific capacitance, i (A) refers to the discharge current, ΔV (V) represents the potential window within the discharge time Δt (s), and m (g) corresponds to the amount of active material on the electrode.²¹ In order to roughly estimate the EDL capacitance originating from the phosphorus groups, we speculate that the total capacitance is the sum of the capacitances that come from the nitrogen, oxygen and phosphorus groups. Therefore, the capacitance provided by the phosphorus groups can be roughly obtained by the following equation:

C_p = C_N/P − C_N (2)

where C_p (F g⁻¹) is the capacitance induced by the phosphorus groups, and C_N/P (F g⁻¹) and C_N (F g⁻¹) represent the specific capacitances of the N/P-NPCNF electrode and the N-NPCNF electrode, respectively. The capacitance values calculated according to eqn (1) and (2) are listed in Table 1. Obviously, the capacitance of the N-NPCNF electrode in neutral electrolytes is negligible, while the capacitance of the N/P-NPCNF electrode increased greatly in neutral electrolytes as well as in the acidic electrolyte. These results agree well with the results of the CV measurements. The capacitance derived from the phosphorus groups in different electrolytes decreases in the order of H₂SO₄ > K₂SO₄ ≈ Li₂SO₄ > Na₂SO₄. In general, the difference in capacitance as well as in rate capability may be related to (I) the crystal radius of the ions, (II) the radius of the ionic hydration sphere in an aqueous solution, (III) the conductivity of the electrolyte, (IV) the mobility of the ions, and (V) the solvation/desolvation energy of the ions in aqueous electrolytes. It is well known that H⁺ ions and alkali metal ions are strongly solvated in an aqueous solution with a decrease of the ion–solvent complex diameter in the order of Li⁺ > Na⁺ > K⁺ > H⁺, and the crystal radius of the ions decreases in the sequence of K⁺ > Na⁺ > Li⁺ > H⁺.^22–24 It seems that the EDL capacitance provided by the phosphorus groups is more related to the crystal radius of the electrolyte ions. This suggests that the adsorbed electrolyte ions by the phosphorus groups might partially desolvate, shortening the distance between the carbon surface and the adsorbed ions, which may explain the tremendously high EDL capacitance per unit area (∼8.5 F m⁻² in acidic electrolyte), which only originated from the phosphorus groups. It should be noted that the larger capacitance in 0.5 M K₂SO₄ compared to that in 1 M Na₂SO₄ could be attributed to the high content of pyrrole-like nitrogen groups in the N/P-NPCNFs (accounting for 65.4% of the total nitrogen groups¹²), which are proven to strongly bind to the K⁺ ions.¹⁵ This was further verified by the similar capacitance of the N/P-NPCNF electrode in 1 M Na₂SO₄ and 0.5 M K₂SO₄ at a high current density of 5 A g⁻¹, because the advantage of the strong adsorption of K⁺ to the pyrrole-like nitrogen configurations diminished under conditions of fast ion transport. Thus, the nitrogen groups contribute a lot to the total capacitance of the N/P-NPCNFs in the K₂SO₄ electrolyte at low current densities.

Fig. 5 Specific capacitance as a function of current density for (a) N-NPCNFs and (b) N/P-NPCNFs in different electrolytes; Nyquist plots of (c) N-NPCNFs and (d) N/P-NPCNFs in different electrolytes.

Table 1 Specific capacitance (F g⁻¹) of the samples in each electrolyte at 0.5 A g⁻¹ and the roughly estimated capacitance (F g⁻¹) from the phosphorus groups

	1 M H2SO4	0.5 M Li2SO4	1 M Na2SO4	0.5 M K2SO4
N-NPCNFs	121	2.4	1.5	1.5
N/P-NPCNFs	223	78	53	82
CP	102	75.6	51.5	80.5

---

In order to clearly evaluate the effects of the electrolyte ions on the capacitive properties, an EIS measurement was performed and the Nyquist plots are shown in Fig. 5c and d. The linear part in the Nyquist plots in the high-frequency region is related to the ion diffusion process; the semicircle in the medium-frequency region is a measure of the interfacial charge transfer resistance (R_ct); and the intercept value of the curve with the real axis in the high-frequency region represents the equivalent series resistance (R_ERS).^15,25 For the N-NPCNF electrode (Fig. 5c), the linear part of the Nyquist plot obtained in 1 M H₂SO₄ exhibits the largest slope value, indicating a fast formation rate of EDL,¹⁵ maybe due to the alkaline properties of the nitrogen groups (facilitating the adsorption of the H⁺ ions). Furthermore, the semicircle confirms the presence of redox reactions between the heteroatom (mainly nitrogen) groups and the H⁺ ions, and the small diameter of the semicircle suggests a small R_ct. Obviously, the R_ERS varies according to the electrolyte, decreasing in the sequence of Li₂SO₄ > K₂SO₄ > Na₂SO₄ > H₂SO₄. The R_ERS of the N/P-NPCNF electrode in different electrolytes shows the same trend (Fig. 5d). These results indicate that the conductivity of the electrolytes, in the present case, decreases in the order of 1 M H₂SO₄ > 1 M Na₂SO₄ > 0.5 M K₂SO₄ > 0.5 M Li₂SO₄. However, the R_ERS of the N/P-NPCNF electrode is always smaller than that of the N-NPCNF electrode in every electrolyte, suggesting a decrease of the contact resistance between the electrode and the electrolyte due to the greatly enhanced surface wettability of the carbon nanofibers induced by the phosphorus groups. Furthermore, the nearly-vertical line in the Nyquist plot of the N/P-NPCNF electrode obtained in 1 M H₂SO₄ shows a much better capacitive behavior as compared to that of the N-NPCNF electrode, confirming that the introduction of the phosphorus groups really facilitate the adsorption of the H⁺ ions. It is further confirmed by the smaller R_ct (0.27 Ω) as compared to that (0.48 Ω) of the N-NPCNF electrode in 1 M H₂SO₄.

Interestingly, relatively large semicircles were observed in the Nyquist plots of the N/P-NPCNF electrodes in neutral electrolytes, whereas no semicircles were observed in the Nyquist plots of the N-NPCNF electrodes in neutral electrolytes. This could be attributed to the enhanced adsorption of the electrolyte ions onto the surface of the nonporous carbon nanofibers, leading to a greatly increased charge transfer and thus enhanced redox reactions. Apparently, in neutral electrolytes, the N/P-NPCNF electrode in 0.5 M K₂SO₄ presents the smallest R_ct, confirming a strong binding between the K⁺ ions and the pyrrole-like nitrogen groups and thus giving rise to a relatively fast charge transfer. Furthermore, the slope value of the line in the Nyquist plot for the N/P-NPCNF electrode in 0.5 M Li₂SO₄ is the smallest, suggesting a very slow ion diffusion process. Therefore, it can be concluded that the smallest electrolyte conductivity and the low diffusion coefficient of the Li⁺ ions lead to the worst rate performance of the N/P-NPCNF electrode in 0.5 M Li₂SO₄ (12.8% capacitance retention in the current range of 0.5–5 A g⁻¹). In addition, the rate capability is also related to the binding energy between the electrolyte ions and the surface groups. In this case, the phosphorus groups have stronger interactions with the H⁺ ions due to the hydrogen bonding between the H⁺ ions and oxygen. The excellent rate capability (84.3% capacitance retention) of the N/P-NPCNF electrode in 1 M H₂SO₄ may be explained by the following reasons: high electrolyte conductivity, high ionic mobility, and the hydrogen bonding between the H⁺ ions and the oxygen-rich phosphorus groups (which lead to a great number of the H⁺ ions to locate at the electrode/electrolyte interface and thus shorten the ion diffusion distance from the electrolyte to the carbon surface).

Conclusions

In summary, cyclic voltammetry, galvanostatic charge–discharge and electrochemical impedance spectroscopy measurements have been employed to comparatively and comprehensively study the electrochemical performances of N/P-NPCNFs and N-NPCNFs in different electrolytes (1 M H₂SO₄, 0.5 M Li₂SO₄, 1 M Na₂SO₄ and 0.5 M K₂SO₄). The electrochemical measurements show that the N-NPCNFs demonstrate no capacitive properties in neutral electrolytes, and that the phosphorus groups play a crucial role in improving the EDL behavior of the nonporous carbon nanofibers by greatly enhancing the surface wettability of carbon and the ability to adsorb electrolyte ions. Furthermore, the specific capacitance and rate capability of the N/P-NPCNFs in different electrolytes are related to the crystal size of the ions, electrolyte conductivity and ionic mobility. In addition, the superior capacitive behavior of the N/P-NPCNFs in an acidic electrolyte may also be correlated with the hydrogen bonding between the H⁺ ions and the oxygen-rich phosphorus groups.

Acknowledgements

The authors acknowledge the financial supports from National Natural Science Foundation of China (no. 51072013, 51272021, and 51142004) and Natural Science Foundation of Jiangsu Province (BK20131147).

Notes and references

1. A. Izadi-Najafabadi, T. Yamada, D. N. Futaba, M. Yudasaka, H. Takagi, H. Hatori, S. Iijima and K. Hata, ACS Nano, 2011, 5, 811.
2. M. G. Hahm, A. L. M. Reddy, D. P. Cole, M. Rivera, J. A. Vento, J. Nam, H. Y. Jung, Y. L. Kim, N. T. Narayanan, D. P. Hashim, C. Galande, Y. J. Jung, M. Bundy, S. Karna, P. M. Ajayan and R. Vajtai, Nano Lett., 2012, 12, 5616.
3. H. Jiang, P. S. Lee and C. Li, Energy Environ. Sci., 2013, 6, 41.
4. S. Bose, T. Kuila, A. K. Mishra, R. Rajasekar, N. H. Kim and J. H. Lee, J. Mater. Chem., 2012, 22, 767.
5. A. Jena, N. Munichandraiah and S. A. Shivashankar, J. Power Sources, 2013, 237, 156.
6. Z. Niu, P. Luan, Q. Shao, H. Dong, J. Li, J. Chen, D. Zhao, L. Cai, W. Zhou, X. Chen and S. Xie, Energy Environ. Sci., 2012, 5, 8726.
7. Y. S. Yun, J. Shim, Y. Tak and H.-J. Jin, RSC Adv., 2012, 2, 4353.
8. D. Hulicova-Jurcakova, A. M. Puziy, O. I. Poddubnaya, F. Suárez-García, J. M. D. Tascón and G. Q. Lu, J. Am. Chem. Soc., 2009, 131, 5026.
9. C. Wang, L. Sun, Y. Zhou, P. Wan, X. Zhang and J. Qiu, Carbon, 2013, 59, 537.
10. C. Wang, Y. Zhou, L. Sun, Q. Zhao, X. Zhang, P. Wan and J. Qiu, J. Phys. Chem. C, 2013, 117, 14912.
11. C. Wang, Y. Zhou, L. Sun, P. Wan, X. Zhang and J. Qiu, J. Power Sources, 2013, 239, 81.
12. X. Yan, Y. Liu, X. Fan, X. Jia, Y. Yu and X. Yang, J. Power Sources, 2014, 248, 745.
13. U. B. Nasini, V. G. Bairi, S. K. Ramasahayam, S. E. Bourdo, T. Viswanathan and A. U. Shaikh, J. Power Sources, 2014, 250, 257.
14. M. Zhong, E. K. Kim, J. P. McGann, S.-E. Chun, J. F. Whitacre, M. Jaroniec, K. Matyjaszewski and T. Kowalewski, J. Am. Chem. Soc., 2012, 134, 14846.
15. F. M. Hassan, V. Chabot, J. Li, B. K. Kim, L. Ricardez-Sandoval and A. Yu, J. Mater. Chem. A, 2013, 1, 2904.
16. W. B. Perry, T. F. Schaaf and W. L. Jolly, J. Am. Chem. Soc., 1975, 97, 4899.
17. J. R. Pels, F. Kapteijn, J. A. Moulijn, Q. Zhu and K. M. Thomas, Carbon, 1995, 33, 1641.
18. A. M. Puziy, O. I. Poddubnaya, R. P. Socha, J. Gurgul and M. Wisniewski, Carbon, 2008, 46, 2113.
19. R. Arrigo, M. Hävecher, S. Wrabetz, R. Blume, M. Lerch, J. McGregor, E. P. J. Parrott, J. A. Zeitler, L. F. Gladden, A. Knop-Gericke, R. Schlögl and D. S. Su, J. Am. Chem. Soc., 2010, 132, 9616.
20. D. Hulicova, J. Yamashita, Y. Soneda, H. Hatori and M. Kodama, Chem. Mater., 2005, 17, 1241.
21. L. Zhao, L.-Z. Fan, M.-Q. Zhou, H. Guan, S. Qiao, M. Antonietti and M.-M. Titirici, Adv. Mater., 2010, 22, 5202.
22. X. Zhang, X. Wang, L. Jiang, H. Wu, C. Wu and J. Su, J. Power Sources, 2012, 216, 290.
23. W. Gu, M. Sevilla, A. Magasinski, A. B. Fuertes and G. Yushin, Energy Environ. Sci., 2013, 6, 2465.
24. K. Fic, G. Lota, M. Meller and E. Frackowiak, Energy Environ. Sci., 2012, 5, 5842.
25. L.-Q. Mai, A. Minhas-Khan, X. Tian, K. M. Hercule, Y.-L. Zhao, X. Lin and X. Xu, Nat. Commun., 2013, 4, 2924.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed experimental procedures. See DOI: 10.1039/c4ra02299h

---