Creating adsorption sites by doping heteroatoms into the graphitic structures of carbon electrodes is an effective strategy for improving lithium storage in lithium-ion batteries. In this work, we prepared carbon nanotubes with controllable morphology and controllable nitrogen-doping level by a one-step pyrolysis method through adjusting the amount of urea used during synthesis. Under the synergistic effects of high temperature and Ni-catalyst migration, the carbon nanosheets generated by pyrolysis become coiled into tube-like structures. Characterization using Raman and x-ray photoelectron spectroscopy revealed that the B and N atoms were successfully co-doped into the resultant carbon nanotubes. When the obtained materials were used as lithium-ion battery anodes, reversible specific capacities of 337.11 and 187.62 mA h g−1 were achieved at current densities of 100 and 2000 mA g−1, respectively. Moreover, a capacity of 140.53 mA h g−1 was retained after 2000 cycles at a current density of 2000 mA g−1. The mechanism of lithium storage in these carbon materials was elucidated using cyclic voltammetry tests. Regarding other functional applications, the synthesized composite carbon nanotube material could also be used in other energy-storage battery systems, such as in the sulfur-carrying structures of lithium-sulfur batteries and in the three-dimensional porous structures of sodium batteries.

  • We prepared carbon nanotubes with controllable morphology and nitrogen-doping levels through a one-step pyrolysis process by carefully adjusting the amount of urea used.

  • As lithium-ion battery anodes, the material exhibited reversible specific capacities of 337.11 and 187.62 mA h g−1 at current densities of 100 and 2000 mA g−1, respectively.

  • Capacity retention of 140.53 mA h g−1 was achieved after 2000 cycles at a current density of 2000 mA g−1.

With the ever-increasing global demand for energy and dramatic increases in air pollution, the need for storage of energy from renewable sources has aroused ever-increasing concern. Due to their long cycle life and high energy density, rechargeable lithium-ion batteries (LIBs) have become widely used as a highly effective energy-storage technology in consumer transportation, and they have dominated the market for portable electronic equipment for more than two decades. As one of the important components of LIBs, anode materials have a great influence on their performance; as such, developing new-generation anode materials to further improve the performance of LIBs is very important. Due to their abundance, low cost, large specific surface areas, superior chemical stability, and unique porosity, carbon-based materials have been extensively investigated as anode materials for fast-charging LIBs.

Graphite is one of the oldest carbon-based anode materials, and it has a >90% market share in commercial LIBs; it can provide a theoretical capacity of 372 mA h g−1, and it has a very flat and low potential (≈0.2 V vs Li/Li+).1,2 The three-dimensional crystal structure of graphite is composed of sp2 hybridization in the in-plane layer with interlayers connected by van der Waals forces.3 According to its source, graphite can be divided into three categories: natural, modified, and artificial. Natural graphite can be mined from metamorphic and igneous rocks. Modified graphite refers to natural graphite after it has been treated by some physical or chemical methods to change its surface structure and morphology to produce better cycling performance and higher specific capacity.4 Artificial graphite can be synthesized by graphitization of other carbon-based precursors (such as petroleum, asphalt, or coal) at temperatures above 2000 °C in an oxygen-free environment.1 After being assembled into batteries, co-embedding of solvents into natural graphite often causes stability problems, and the graphite electrode can easily fall off during charging and discharging processes over large numbers of cycles, which seriously affects cycling stability. Artificial graphite has better storage properties than natural graphite in LIBs, but it is relatively expensive to produce. In short, the dramatic degradation of the capacity of graphite anodes in rapid charge–discharge processes and the occurrence of lithium plating can cause significant safety risks. Furthermore, high-rate charging may cause high polarization, resulting in the loss of the electrochemical platform and reducing the potential; in such cases, the deposition of lithium metal may result in the formation of dendrites, leading to short circuits and the potential for explosions.

To address the problems outlined above and satisfy the increasing need for high-performance LIBs, scientists have paid increasing attention to the use of other allotropes of carbon for LIB anodes, such as graphene and carbon nanotubes (CNTs); however, complex processes and equipment are often required in the preparation processes for these materials. Pure graphene can undergo reconstruction during charge cycling, which can reduce the number of inserted lithium ions. Consequently, the capacity and lifespan of the battery are decreased.5 CNTs are also very attractive due to their typical tube-like structure with reserved hollow spaces, tunable diameters, outstanding mechanical strength, and excellent conductivity.6 These competitive properties make them good candidates for anode materials.7 Single-walled CNTs have been shown to exhibit reversible specific capacities from 300 to 600 mA h g−1,8–14 which is much higher than the theoretical maximum capacity of graphite, 372 mA h g−1; however, the practical capacity of LNBs using pure carbon-nanotube anodes is still very low. Lee et al. synthesized CNTs on Cu foil using plasma-enhanced chemical vapor deposition, with a low capacity measured at 64.94 mA h g−1.15,16

Doping heteroatoms into the graphitic structures of CNTs can significantly alter their properties, and when such materials are used as the anodes of LIBs, this has been shown to further improve their performance. Recently, continuous efforts have been made to fabricate doped carbon nanotubes with heteroatoms such as B, N, and S; these materials have exhibited good electrochemical performance in supercapacitors, LIBs, and field-effect transistors.17,18 In such a hybrid system, changes in the volume of CNTs can be used to induce an effective confining buffer for mechanical stress in charging and discharging reactions. In previous studies, it has been shown that doping of heteroatoms can endow carbon materials with excellent properties.10,12 The radii of heteroatoms tend to be greater than that of carbon atoms; as such, their introduction can increase the spacing between graphite layers, disturb the original ordered structures of carbon materials, and facilitate the insertion of lithium ions.18 Furthermore, the inserted heteroatoms have different physical and chemical properties from carbon atoms, and this can enhance the adsorption capacity of carbon materials for lithium ions and improve their electron-conduction capacity.19 Overall, the doping of heteroatoms into the carbon structure can generate new active sites, reduce the reaction and diffusion barriers, and improve the lithium storage capacity.

For example, using aerosol-assisted catalytic chemical vapor deposition, Bulusheva et al. obtained multi-walled nitrogen-doped carbon nanotubes;20 they found that the morphology and nitrogen content of these carbon nanotubes could be tuned by adjusting the concentration of toluene and acetonitrile in the raw material, and they obtained a capacity of 270 mA h g−1 at a current density of 0.2 mA cm−1, which is much higher than for pure carbon nanotubes. Tian and Zhu used sodium citrate and boric acid as source materials, which were calcined for a long period and underwent a series of acid-washing processes, to obtain B-doped CNTs with an average diameter of 150 nm; these showed excellent electrochemical performance for lithium storage applications.21 

Although the above-described works were well conducted and produced impressive results, the raw materials are toxic and harmful, or they require the use of acids for further cleaning; furthermore, their atomic doping contents are limited. In the present study, we used a Ni-based catalyst, boric acid as a B source, and urea as an N and C source, and we successfully fabricated B and N co-doped CNTs. The quantity of urea was adjusted to achieve controllable morphology and nitrogen content. The Ni metal formed by the catalyst helped to improve the conductivity of the CNTs. The resultant CNT materials, when working as an anode materials, defined as Ni@BTN samples for LIBs, were found to exhibit outstanding capacity under large currents.22 

All chemicals were purchased and used without any further purification. Boric acid (H3BO3) was purchased from a Chinese medical supplier. Nickel nitrate hexahydrate and urea were purchased from Jiuwei Chemical Industry, Ltd. Polyethylene glycol (PEG-2000) was purchased from Beijing Yinokai Technology Co., Ltd.

A total of 65 mg nickel nitrate hexahydrate, 0.15 g boric acid, 0.5 g PEG-2000, and x grams (x = 1, 2, 2.75) of urea were dissolved in 100 ml deionized water (18.2 MΩ cm−1) and stirred for 1 h. The Ni@BTN precursor solution was dried at 80 °C for 14 h, and the mixed dried product was then ground with an agate mortar for 5 min and transferred to alumina crucible before finally being put into a tube furnace for one-step pyrolysis. To prevent oxidation during pyrolysis, in an atmosphere with an argon gas flow rate of 20 sccm, the temperature was raised from room temperature at 5 °C min−1 to 900 °C for 4 h. When the tube furnace had returned to room temperature by natural cooling, the obtained samples were labeled as Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 depending on the quantity of urea used.

The crystal structures and lattice parameters of the samples were analyzed by x-ray diffraction (XRD; Rigaku MiniFlex600) with a Cu Kα (0.154 nm) radiation source from 10° to 60° at a rate of 10° min−1. The morphology and surface elements of the composites were characterized by scanning electron microscopy (SEM; Thermo Scientific Phenom Pharos) and energy-dispersive x-ray spectroscopy (EDS) mapping. The measurement mode was secondary electron detection, and the acceleration voltage was 15 kV. To reveal more structural information, the samples were examined by Raman spectroscopy with an incident laser light wavelength of 532 nm, in the 750–2000 cm−1 spectral range, using an InVia Raman spectrometer (Renishaw) with an air-cooled charge-coupled-device detector and a Leica microscope (50× short-distance objective); the spectral resolution was set at 10 cm−1. The exposure time for each measurement was 4 s, with 15 accumulations. Additionally, x-ray photoelectron spectroscopy (XPS) analysis was conducted on a Thermo Scientific ESCALAB 250Xi spectrometer using a monochromatic Al Kα radiation source ( = 1486.6 eV). Thermogravimetric analysis (TGA) was conducted on an STA 6000 (PerkinElmer) at a heating rate of 10 °C/min in an Ar atmosphere.

The Ni@BTNx samples were thoroughly blended with conductive additives (Super-P) and a polymeric binder (polyvinylidene fluoride) at a mass ratio of 8:1:1 in the organic solvent N-methylpyrrolidone. This was stirred for 5 h to obtain a slurry, and this slurry was then coated onto 9-µm-thick Cu foil. The electrodes were obtained after drying in a vacuum oven at 80 °C for 12 h to remove the solvent. The electrodes were then cut to 10-mm-diameter disks for assembly with active-material loadings around 1.50 mg cm−2. For testing, half a CR2032 cell was assembled in an argon glove box (water and oxygen content less than 1 ppm). Polypropylene film (commercial Celgard 2500) was used as the separator, and 1 M LiPF6 in a mixed solvent (the volume ratio of ethylene carbonate:ethyl methyl carbonate:dimethyl carbonate was 1:1:1) was used as the electrolyte; the lithium metal was used as the counter electrode, and the disks served as the anode. Cyclic voltammetry (CV) was performed on a CHI660D (Shanghai Chen Hua Instrument Co., China) electrochemical workstation at different scanning rates in the voltage window 0.01–3.00 V. Electrochemical impedance spectra (EIS) were obtained in a frequency range from 0.1 Hz to 100 kHz. Galvanostatic charge-discharge (GCD) and cycling-stability tests were carried out in the voltage range 0.01–3.00 V under a constant current density using a CT-4008T battery-test system (Neware Technology, China).

The XRD spectra of Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 shown in Fig. 1(a) reveal that the (111) and (200) planes of Ni metal nanoparticles and the (002) plane of graphitic carbon crystals exist in all three samples. According to JCPDS Card No. 04-0850, the peaks at 52.20° and 44.55° in Fig. 1 correspond to the (111) and (200) planes of Ni metal nanoparticles, which are obtained from the one-step pyrolysis process. This process can be divided into the following three steps:23,
2Ni(NO3)26H2O =2NiO +4NO2+6H2O
(1)
3NiO +4C = NiC3+ CO
(2)
NiC3=3Ni + C
(3)
First, the nickel nitrate is decomposed into nickel oxide and nitrogen dioxide gas [Eq. (1)]. Then, NiO spontaneously disperses on the carbon surface and is reduced to NiC3 by carbon [Eq. (2)]. Finally, as the temperature continues to rise, NiC3 decomposes to form carbon and Ni metal [Eq. (3)]. Therefore, Ni metal diffraction peaks can be observed in the spectra from our three prepared samples. The diffraction peaks located at around 26.10° for these three samples correspond to the typical graphitic carbon (002) crystal plane, according to JCPDS No. 41-1487. The broad diffraction peak indicates that some amorphous carbon structures exist in the three samples. The interlayer distances of the one-step-synthesized Ni@BTNx samples calculated using the Bragg equation are shown in Table I. It can be clearly seen that samples have a stable interlayer distance of 0.34 nm, which is larger than the radius of Li+ (0.059 nm).24 This large interlayer spacing can accommodate Li+ and reduce the resistance of Li+ insertion/delithiation during charge/discharge.
FIG. 1.

(a) XRD spectra from the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples. Raman spectra of Ni@BTNx: (b) Ni@BTN1, (c) Ni@BTN2, and (d) Ni@BTN2.75. (e) Complete XPS spectra from Ni@BTNx. High-resolution XPS spectra for: (f) B 1s, (g) C 1s, and (h) N 1s of the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples.

FIG. 1.

(a) XRD spectra from the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples. Raman spectra of Ni@BTNx: (b) Ni@BTN1, (c) Ni@BTN2, and (d) Ni@BTN2.75. (e) Complete XPS spectra from Ni@BTNx. High-resolution XPS spectra for: (f) B 1s, (g) C 1s, and (h) N 1s of the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples.

Close modal
TABLE I.

Interlayer distances of one step synthesis Ni@BTNx samples.

Sample nameDiffraction peak (°C)d(002) (nm)
Ni@BTN1 26.12 0.341 
Ni@BTN2 25.96 0.343 
Ni@BTN2.75 26.11 0.341 
Sample nameDiffraction peak (°C)d(002) (nm)
Ni@BTN1 26.12 0.341 
Ni@BTN2 25.96 0.343 
Ni@BTN2.75 26.11 0.341 

The XRD results in Fig. 1(a) indicate the presence of amorphous carbon in the samples; hence, Raman spectroscopy was used to investigate the degree of structural disorder. For short-range-ordered materials, Raman spectroscopy is a simple and effective method that is often used to obtain structural information such as the presence of structural defects. Raman signals are generated by lattice vibrations and are very sensitive to the degree of structural disorder. Figures 1(b)1(d) show Raman spectra from the one-step pyrolysis samples. All of the samples exhibit two broad overlapping bands, also known as the D and G bands, centered at ∼1350 and 1590 cm−1; these can be fitted into four different peaks. Research has shown that the presence of a lower-wave-number shoulder at about 1180 cm−1 indicates a disordered graphite lattice (polyene-like structure, A1g symmetry).8 There is a peak near 1480 cm−1 (A band) originating from the amorphous portion of carbon.8 The peak located at 1344 cm−1 represents the A1g symmetry breathing-mode vibration of full six-membered carbon. The G band (E2g symmetry) is related to the sp2 hybrid C–C bond of the graphite layer. Wang et al. demonstrated that the ID/IG values change when sp2 carbon transitions to sp3 carbon.25 This means that the ID/IG value can indicate the degree of defects and disorder in all samples.4,26 The ID/IG values of obtained from the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples, were 0.92, 0.95, and 0.99, respectively. Notably, the ID/IG value increases with the increasing N content, indicating that doping with N results in more structural defects and increases the disorder of the samples.27 

XPS was carried out to further evaluate the elemental composition and chemical state of the resultant carbon nanotubes. Figure 1(e) shows the XPS survey spectra of the one-step synthesized Ni@BTNx samples. The five peaks at around 192, 285, 399, 533, and 857 eV respectively correspond to the elements B, C, N, O, and Ni. Although a small amount of oxygen-containing functional groups and Ni nanoparticles are present in the samples, the impact of O and Ni in the cyclic lithium storage process is relatively small, and there is no specific analysis for it. For the Ni@BTN2 sample, a detailed analysis was conducted on the content and chemical bonding of C 1s, N 1s, and B 1s. The high-resolution B 1s spectrum shown in Fig. 1(f) can be fitted by three peaks, namely, 190.65 eV (B–C), 192.61 eV (B–N), and 193.44 eV (B–O), and the B content is 17.99 at. %. Deconvolution of the high-resolution C 1s spectrum in Fig. 1(g) reveals peaks located at 283.75, 284.34, 284.79, 285.54, and 286.69 eV, corresponding to O–C=O, C=O, C–C/N, C=C, and C–B, respectively, each representing different chemical bonding of C. The individual N 1s spectra in Fig. 1(h) reveal the distributions of N atoms of four different valence states. The central peaks are located at 397.97, 398.92, 399.57, 400.16, and 401.48 eV, corresponding to B–N–C, pyridinic-N, pyrrolic-N, graphitic-N, and oxidic-N chemical bonds, respectively, of which pyridinic-N and pyrrolic-N are N atoms on the six- and five-membered rings of carbon. These two types of N atom can improve the conductivity of the sample, enhancing the Li+-ion transport kinetics. Quantitative analysis revealed that the N contents of the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples were 8.15, 10.29, and 11.58 at. %, respectively.

To further compare the contents of different types of N chemical bond in the samples, peak splitting of the high-resolution XPS spectrum for each sample was performed, and the results are shown in Table II. It can be seen that the content of pyridinic-N and pyrrolic-N were the highest in Ni@BTN2. It is clear that the N content of the one-step synthesized samples can be controlled by adjusting the amount of urea during the preparation process. Overall, the XPS results demonstrate that the one-step synthesis method can effectively co-dope B and N into the graphitic stricture of the resultant carbon materials with a controllable N content.28 

TABLE II.

Percentage contents of different N chemical bonds in the samples.

Chemical bondNi@BTN1 (%)Ni@BTN2 (%)Ni@BTN2.75 (%)
Oxidic-N 21.30 15.03 24.69 
Graphitic-N 42.91 19.90 33.97 
Pyridinic-N 17.53 25.57 18.79 
Pyrrolic-N 17.01 18.90 15.02 
B–N–C 1.25 20.60 7.53 
Chemical bondNi@BTN1 (%)Ni@BTN2 (%)Ni@BTN2.75 (%)
Oxidic-N 21.30 15.03 24.69 
Graphitic-N 42.91 19.90 33.97 
Pyridinic-N 17.53 25.57 18.79 
Pyrrolic-N 17.01 18.90 15.02 
B–N–C 1.25 20.60 7.53 

The tubular structures of the samples can be clearly observed in the SEM images shown in Figs. 2(a)2(c), confirming the successful preparation of CNTs. The morphology of the CNTs changes with increasing amount of urea. When adding small amounts of urea (Ni@BTN1, Ni@BTN2), the walls of the CNTs are smooth; when the amount of urea is increased to 2.75 g, the walls of the CNTs become rough. We believe that that the rough surface can be attributed to an increase in amorphous carbon resulting from the direct pyrolysis of urea. When the amount of urea is small, the Ni catalyst can significantly catalyze the urea to create high quality graphite-like carbon. As the amount of urea is increased, the catalytic process becomes insufficient, and this results in the production of a large amount of amorphous carbon. The ultimate distribution of doped non-metallic atoms affects the properties of the materials. The results of EDS of the Ni@BTN2 samples shown in Fig. 2(d) demonstrate the uniform distribution of B and N elements in the CNTs, with Ni existing in the form of nanoparticles. Figure 2(e) shows the results of TGA of the Ni@BTN2 and the BTN2 source. Under the condition of no catalyst, the weight decreases uniformly to zero as the temperature rises; with the catalyst, the weight ultimately remains at 1.45%. During the thermogravimetric tests, the weight changes mainly occurred before 400 °C, and the losses at these temperatures are mainly due to evaporation of surface water and decomposition of carbon-based raw materials. Above 400 °C, the weight of carbon materials containing nickel-salt catalysts hardly decreases as the temperature is further increased, while the weight of materials without catalysts decreases above 400 °C until reaching 0%. This process indicates that the presence of catalysts is crucial in the overall pyrolysis process and has a certain carbon-fixation effect. Figure 2(f) shows a schematic of the process of Ni-catalyzed growth of carbon nanotubes. Below 400 °C, the N and C obtained from urea decomposition combine with the carbon produced from PEG-2000 decomposition to form CN nanosheets. Above 400 °C, boric acid is decomposed into active groups containing B, and these are doped into the nanosheets. Nickel salts are dehydrated, pyrolyzed, and combined with carbon and distributed on the CN nanosheets, from the precursor to the generation of NiO.29 In the subsequent continuous heating process, NiO reacts with carbon to form NiC3, and this NiC3 is in turn reduced to Ni metal and encapsulated in carbon nanotubes. Due to the presence of Ni nanoparticles on the nanosheets, the local stress is not uniform, causing the nanosheets to wrinkle and curl.30–32 From the TEM image shown in Fig. 2(g), it can also be seen that Ni is encapsulated in carbon nanotubes in the form of nanoparticles.

FIG. 2.

SEM images of Ni@BTNx: (a) Ni@BTN1, (b) Ni@BTN2, and (c) Ni@BTN2.75. (d) Corresponding EDS mapping of Ni@BTN2. (e) TGA results for Ni@BTN2 and BTN2. (f) Schematic of the fabrication process. (g) TEM image of Ni@BTN2.

FIG. 2.

SEM images of Ni@BTNx: (a) Ni@BTN1, (b) Ni@BTN2, and (c) Ni@BTN2.75. (d) Corresponding EDS mapping of Ni@BTN2. (e) TGA results for Ni@BTN2 and BTN2. (f) Schematic of the fabrication process. (g) TEM image of Ni@BTN2.

Close modal

The insertion/disinsertion performance of Li+ was evaluated by GCD of the half batteries. First, the batteries were activated three times at a current of 50 mA g−1, and they were then charged and discharged at a constant current of 100 mA g−1. The initial capacities of the lithium half batteries assembled from Ni@BTN1, Ni@BTN2 and Ni@BTN2.75 were 265.70, 292.54, and 256.68 mA h g−1, respectively. As shown in Fig. 3(a), the capacity gradually increases with cycling; this is mainly due to the continuous infiltration of electrolytes and the continuous activation of electrode materials during the electrochemical cycling process, promoting Li+ insertion/disinsertion.33 The upward trend is divided into two stages: rising over the first 60 cycles, and remaining stable afterwards. The capacity continued to rapidly increase over the first 50 cycles, and this is attributed to two main causes: first, the activation is incomplete at the initial low current, and activation continues during the cycling process; second, new defects are created during the cycling process.34 The capacity remains basically stable after 60 cycles. After 100 cycles, the final capacity of the Ni@BTN1 battery was 281.85 mA h g−1, a capacity increase of 6.07%; for the Ni@BTN2 sample, the final capacity was 337.11 mA h g−1, a capacity increase of 15.23%; the capacity of the Ni@BTN2.75 sample reached 303.08 mA h g−1, an increase of 18.07%. The Ni@BTN2 sample maintained a coulombic efficiency of over 98% during charge–discharge cycles. This high coulombic efficiency may be attributed to the electrode surface remaining intact during low-current cycling; there is no formation of dead lithium or lithium dendrites in the electrode surface. Further verification can also be achieved through scanning tests on the surface and cross-sectional examination of the cycled electrode.

FIG. 3.

(a) Cycling performance of Ni@BTNx electrodes at a current density of 100 A g−1, (b) Charge and discharge curves for Ni@BTN2 electrodes at 100 A g−1.

FIG. 3.

(a) Cycling performance of Ni@BTNx electrodes at a current density of 100 A g−1, (b) Charge and discharge curves for Ni@BTN2 electrodes at 100 A g−1.

Close modal

Figure 4 shows the rate performances of the Ni@BTNx electrodes. Among the samples, the Ni@BTN2 electrode showed the best rate performance, with a charging capacity of 357.63–316.62 mA h g−1 (retention 88.53%) at a current density of 50 mA g−1. When the current density was increased to 100 mA g−1, the specific capacity reached 326.74 mA h g−1, which is slightly higher than the initial capacity at 100 mA g−1. Overall, this sample showed excellent cycling reversibility and stability at high-current charging and discharging. The high performance of the Ni@BTN2 sample can be attributed to its greater number of defects, larger interlayer distance (which is beneficial for electrolyte penetration), and its larger current density being beneficial for the removal of Li+. Further examination of the GCD curves [Fig. 4(b)] of Ni@BTN2 at different current densities reveals that as the current density is increased, the shape of the curve is well preserved, confirming the outstanding rate performance of Ni@BTN2. Next, the practical-application potential of Ni@BTN2 was examined by long-cycle tests with high current.

FIG. 4.

(a) Rate performances of Ni@BTNx electrodes at different current densities. (b) Charge/discharge curves for Ni@BTN2 electrodes at different current densities. (c) Cyclic performance of Ni@BTNx electrodes at a current density of 2000 mA g−1. EIS results of Ni@BTNx electrodes (d) before cycling and (e) after 2000 cycles. (f) CV curves for the first three cycles with Ni@BTN2 electrode at 0.1 mV s−1.

FIG. 4.

(a) Rate performances of Ni@BTNx electrodes at different current densities. (b) Charge/discharge curves for Ni@BTN2 electrodes at different current densities. (c) Cyclic performance of Ni@BTNx electrodes at a current density of 2000 mA g−1. EIS results of Ni@BTNx electrodes (d) before cycling and (e) after 2000 cycles. (f) CV curves for the first three cycles with Ni@BTN2 electrode at 0.1 mV s−1.

Close modal

We further tested the cycling stability of Ni@BTNx at a high current density of 2000 mA g−1 [Fig. 4(c)]. The results show that the Ni@BTN1 nanotubes retained 89.39% of their storage capacity after 2000 cycles. The initial delithiation capacities of the lithium half batteries assembled from Ni@BCN1, Ni@BCN2, and Ni@BCN2.75 were 130.12, 160.07, and 150.25 mA h g−1, respectively. As shown in Fig. 4(c), in the first 200 cycles, the capacities of the Ni@BTN1 and Ni@BTN2 samples continued to rise, but that of the Ni@BCN2.75 did not; this is attributed to two causes. First, when the small current suddenly changes to a large current, the polarization of the battery increases, resulting in a sudden decline in capacity; however, in the subsequent cycle, as the battery continues to adapt, its capacity begins to rise. Second, new defects are formed in the process of the circulation of a large current, resulting in a continuous increase in capacity. It was also found that the increase in capacity became slow after 250 cycles. Nonetheless, Ni@BTN2.75 performed a little differently due to it stabilizing at about 80 cycles. This may be because the Ni@BTN2.75 had a greater number of defects, and these mean it can quickly adapt to the rapid insertion of Li+ at large current. After 2000 cycles, the capacities of the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples finally stabilized at 116.32, 140.47, and 69.96 mA h g−1, respectively; the corresponding capacity-retention rates were 89.39%, 87.76%, and 46.56%. The half battery assembled from the Ni@BTN2 carbon material maintained a coulomb efficiency of more than 98% during the charge–discharge cycles, and its characteristics were found to be the best.

The good cyclic stability of Ni@BTNx can be attributed to the following facts. The outer surface of the carbon shell can suffer from higher stress during the lithiation/delithiation processes.8 The Ni NPs in the heteroatom-doped carbon nanotubes give better control over the strain produced during the lithiation/delithiation processes, thus avoiding the pulverization issue. This demonstrates that Ni@BTNx materials with a controllable N content and morphology as prepared by one-step pyrolysis have good cycle life as anode electrode materials for Li+ batteries.

The EIS results reveal the ion-transport behavior of the synthesized carbon material. As shown in Figs. 4(d)4(e), the Nyquist curve of the electrode material before cycling was composed of two different regions: the first is a semicircle in the high-frequency region, and this represents the charge-transfer resistance (Rct) at the interface between the electrode and the electrolyte; the other is an oblique line in the low-frequency region, which represents the Warburg-type resistance (Zw), indicating the ion-diffusion resistance inside the electrode material.35 The insets of Figs. 4(d) and 4(e) show equivalent-circuit diagrams before and after cycling, respectively. Before cycling, the calculated Rct values of Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 electrodes are 317.62, 128.80, and 228.32 Ω, respectively. The lower Rct value of the Ni@BTN2 electrode when compared to the Ni@BTN1 and Ni@BTN2.75 electrodes is due to its larger interfacial spacing, which is beneficial for ion transfer and diffusion. After 2000 cycles [Fig. 4(e)], it is clear that smaller arcs appear for the Ni@BTNx electrodes; these are associated with both a solid electrolyte interphase (SEI) passivation film on the surface (Rf) and charge transfer at the interface (Rct). The Rct value of Ni@BTN2 after 2000 cycles is much smaller (53.95 Ω) than that of the fresh Ni@BTN2 electrode, revealing the small charge-transfer impedance after 2000 cycles. The SEI film resistance values (Rf) of the Ni@BTN1, Ni@BTN2, and Ni@BTN2.75 samples were 9.52, 11.21, and 8.73 Ω, respectively. The resistance of the electrolyte (Rs) of Ni@BTN2 after 2000 cycles was smallest, at 3.91 Ω. This result can also further explain why the Ni@BTN2 electrode exhibits the best performance.

Cyclic voltammetry was used to study the storage mechanism of Li+, and kinetic analysis was performed on electrodes at different scanning rates. Figure 4(f) shows a CV diagram for the Ni@BTN2 sample, as obtained at 0.1 mV s−1. It can be seen that the CV curves of the first and second cycles do not overlap; this is because an SEI film was formed during the first cycle, consuming some lithium sources. Nonetheless, the second and third cycles completely overlap, indicating excellent cycling stability.

Figure 5 shows the CV results for Ni@BTN2 at different scan speeds with a fixed potential window ranging from 0.01 to 3.00 V. The absence of redox peaks in these results indicates that the charge storage and conversion processes depend on the charge accumulation on the surface of the electrode material.36 Clearly, there are two discharge plateaus in the first cycles of the GCD profiles, which is in good agreement with the reversible reduction peaks at around 1.5 V in the CV curves. Additionally, the shape of the CV profile is well maintained even at high scan rates [Fig. 5(a)], demonstrating rapid electrochemical reaction kinetics. The relationship between the peak current (i) and the scan rate (ν) also confirms this:
i=avb,
(4)
lni=blnv+lna,
(5)
where a and b are variable constants: b < 0.5 indicates a diffusion control process and b = 1 indicates a capacitive process. Fitting Eqs. (4) and (5), an obtained value of b around 0.8 indicates that both of these two storage mechanisms exist. This b value shows that the charge storage mechanism in the electrode material is a more capacitive process with limited lithium-ion diffusion. The contribution percentages of capacitance and diffusion at a fixed scan rate can be calculated as follows:37 
i=k1v+k2v0.5,
(6)
iv0.5=k1v0.5+k2,
(7)
where k1 and k2 are constants, k1ν refers to the capacitive process, and k2ν0.5 represents the sum of the diffusion control process. Therefore, the key step is to calculate the values of k1 and k2. Equation (6) can be converted to Eq. (7), as shown in Fig. 5(c); at a scan rate of 0.8 mV s−1, Ni@BTN2 has a capacitive charge storage percentage of 75.72%. When the scan rate is increased from 0.2 to 1.0 mV s−1, the capacitive charge storage performance of Ni@BCN2 increases from 61.00% to 77.18% in Fig. 5(d). The CV results and their interpretation show that the storage process of Li+ depends on the capacitive mechanism.
FIG. 5.

CV curves of the two batteries in the third cycle with voltage scans of (a) 0.05 mV s−1 and (b) 0.08 mV s−1. Kinetic analysis of Ni@BTN2: (c) CV curves at different sweep rates; (d) fitted curves of the b-values; (e) capacitive current contribution (red region) to the lithium-ion storage at 0.8 mV s−1; (f) contribution ratio of capacitive and diffusion controlled at different scan rates.

FIG. 5.

CV curves of the two batteries in the third cycle with voltage scans of (a) 0.05 mV s−1 and (b) 0.08 mV s−1. Kinetic analysis of Ni@BTN2: (c) CV curves at different sweep rates; (d) fitted curves of the b-values; (e) capacitive current contribution (red region) to the lithium-ion storage at 0.8 mV s−1; (f) contribution ratio of capacitive and diffusion controlled at different scan rates.

Close modal

Figure 6(a) shows surface and cross-sectional views of the electrode as obtained by SEM. The electrode is 16 μm thick before cycling [Fig. 6(a)]. Figure 6(b) shows the evolution of the morphology of the Ni@BTN2 electrode after 100 cycles of charge–discharge process at a current density of 100 mA g−1, and Fig. 6(c) shows the morphology after 2000 cycles of charge–discharge process at the current density of 2000 mA g−1. After 100 cycles, a uniform SEI film is formed on the surface, and its thickness increased to 27.64 µm. After 2000 cycles, the thickness is only slightly increased, to 28.34 µm. These results indicate that the Ni@BTN2 electrode exhibits good structural stability.

FIG. 6.

SEM images of the Ni@BTN2 electrode (a) before cycling, (b) after 100 cycles, and (c) after 2000 cycles.

FIG. 6.

SEM images of the Ni@BTN2 electrode (a) before cycling, (b) after 100 cycles, and (c) after 2000 cycles.

Close modal

In this work, carbon nanotubes were successfully prepared by one-step pyrolysis using urea as a raw material and Ni as catalyst; a controllable N content and controllable morphology were achieved. It was found that the prepared carbon nanotubes exhibited dual doping characteristics (N and B). When used as the anode of an LIB, the reversible specific capacity was found to reach 337.11 mA h g−1. After 2000 cycles, the reversible specific capacity remained at 89.39%, indicating good cyclic stability. This strategy for synthesis of dual-doped carbon nanotubes is expected to further develop the field of anode materials for Li+ batteries.

No specific funding was used for this project.

The authors have no conflicts to disclose.

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

1.
Wang
G
,
Yu
M
,
Feng
X
.
Carbon materials for ion-intercalation involved rechargeable battery technologies
.
Chem Soc Rev
2021
;
50
(
4
):
2388
2443
.
2.
Shin
M C
,
Kim
J H
,
Nam
S
,
Oh
Y J
,
Jin
H J
,
Park
C R
,
Zhang
Q
,
Yang
S J
.
Atomic-distributed coordination state of metal-phenolic compounds enabled low temperature graphitization for high-performance multioriented graphite anode
.
Small
2020
;
16
(
33
):
e2003104
.
3.
Yacoby
A
.
Tri and tri again
.
Nat Phys
2011
;
7
(
12
):
925
926
.
4.
Zhou
J
,
Ma
K
,
Lian
X
,
Shi
Q
,
Wang
J
,
Chen
Z
,
Guo
L
,
Liu
Y
,
Bachmatiuk
A
,
Sun
J
,
Yang
R
,
Choi
J H
,
Rummeli
M H
.
Eliminating graphite exfoliation with an artificial solid electrolyte interphase for stable lithium-ion batteries
.
Small
2022
;
18
(
15
):
e2107460
.
5.
Kim
D W
,
Jung
S M
,
Senthil
C
,
Kim
S S
,
Ju
B K
,
Jung
H Y
.
Understanding excess Li storage beyond LiC6 in reduced dimensional scale graphene
.
ACS Nano
2021
;
15
(
1
):
797
808
.
6.
Wang
G X
,
Ahn
J-h
,
Yao
J
,
Lindsay
M
,
Liu
H K
,
Dou
S X
.
Preparation and characterization of carbon nanotubes for energy storage
.
J Power Sources
2003
;
119-121
:
16
23
.
7.
Yu
M F
,
Lourie
O
,
Dyer
M J
,
Moloni
K
,
Kelly
T F
,
Ruoff
R S
.
Strength and breaking mechanism of multiwalled carbon nanotubes under tensile load
.
Science
2000
;
287
(
5453
):
637
640
.
8.
Yang
S
,
Huo
J
,
Song
H
,
Chen
X
.
A comparative study of electrochemical properties of two kinds of carbon nanotubes as anode materials for lithium ion batteries
.
Electrochim Acta
2008
;
53
(
5
):
2238
2244
.
9.
Mi
C H
,
Cao
G S
,
Zhao
X B
.
A non-GIC mechanism of lithium storage in chemical etched MWNTs
.
J Electroanal Chem
2004
;
562
(
2
):
217
221
.
10.
Meunier
V
,
Kephart
J
,
Roland
C
,
Bernholc
J
.
Ab initio investigations of lithium diffusion in carbon nanotube systems
.
Phys Rev Lett
2002
;
88
(
7
):
075506
.
11.
Prem Kumar
T
,
Ramesh
R
,
Lin
Y Y
,
Fey
G T-K
.
Tin-filled carbon nanotubes as insertion anode materials for lithium-ion batteries
.
Electrochem Commun
2004
;
6
(
6
):
520
525
.
12.
Kawasaki
S
,
Hara
T
,
Iwai
Y
,
Suzuki
Y
.
Metallic and semiconducting single-walled carbon nanotubes as the anode material of Li ion secondary battery
.
Mater Lett
2008
;
62
(
17-18
):
2917
2920
.
13.
Morris
R S
,
Dixon
B G
,
Gennett
T
,
Raffaelle
R
,
Heben
M J
.
High-energy, rechargeable Li-ion battery based on carbon nanotube technology
.
J Power Sources
2004
;
138
(
1-2
):
277
280
.
14.
Nishidate
K
,
Hasegawa
M
.
Energetics of lithium ion adsorption on defective carbon nanotubes
.
Phys Rev B
2005
;
71
(
24
):
245418
.
15.
Lee
S
,
Kwon
S
,
Kim
K
,
Kang
H
,
Ko
J M
,
Choi
W
.
Preparation of carbon nanowall and carbon nanotube for anode material of lithium-ion battery
.
Molecules
2021
;
26
(
22
):
6950
.
16.
Zheng
Z
,
Morgan
M
,
Maji
P
,
Xia
X
,
Zu
X
,
Zhou
W
.
CuO nanorod arrays by gas-phase cation exchange for efficient photoelectrochemical water splitting
.
RSC Adv
2023
;
13
(
6
):
3487
3493
.
17.
Cai
C
,
Han
S
,
Zhang
X
,
Yu
J
,
Xiang
X
,
Yang
J
,
Qiao
L
,
Zu
X
,
Chen
Y
,
Li
S
.
Ultrahigh oxygen evolution reaction activity in Au doped co-based nanosheets
.
RSC Adv
2022
;
12
(
10
):
6205
6213
.
18.
Zhu
Y
,
Huang
Y
,
Chen
C
,
Wang
M
,
Liu
P
.
Phosphorus-doped porous biomass carbon with ultra-stable performance in sodium storage and lithium storage
.
Electrochim Acta
2019
;
321
:
134698
.
19.
Chen
C
,
Huang
Y
,
Meng
Z
,
Lu
M
,
Xu
Z
,
Liu
P
,
Li
T
.
N/O/P-rich three-dimensional carbon network for fast sodium storage
.
Carbon
2020
;
170
:
225
235
.
20.
Bulusheva
L G
,
Okotrub
A V
,
Kurenya
A G
,
Zhang
H
,
Zhang
H
,
Chen
X
,
Song
H
.
Electrochemical properties of nitrogen-doped carbon nanotube anode in Li-ion batteries
.
Carbon
2011
;
49
(
12
):
4013
4023
.
21.
Xia
T
,
Zhu
Y
.
Hierarchical B-doped carbon nanotube with enhanced electrochemical lithium storage
.
Microporous Mesoporous Mater
2019
;
284
:
276
282
.
22.
Siddiqui
S E
,
Rahman
M A
,
Kim
J H
,
Sharif
S B
,
Paul
S
.
A review on recent advancements of Ni-NiO nanocomposite as an anode for high-performance lithium-ion battery
.
Nanomaterials
2022
;
12
(
17
):
2930
.
23.
Zhang
J
,
Ai
Y
,
Wu
J
,
Zhang
D
,
Wang
Y
,
Feng
Z
,
Sun
H
,
Liang
Q
,
Sun
T
,
Yang
Y
.
Nickel-catalyzed synthesis of 3D edge-curled graphene for high-performance lithium-ion batteries
.
Adv Funct Mater
2019
;
30
(
9
):
1904645
.
24.
Nayak
P K
,
Yang
L
,
Brehm
W
,
Adelhelm
P
.
From lithium-ion to sodium-ion batteries: Advantages, challenges, and surprises
.
Angew Chem Int Ed Engl
2018
;
57
(
1
):
102
120
.
25.
Wang
Y
,
Alsmeyer
D C
,
McCreery
R L
.
Raman spectroscopy of carbon materials: Structural basis of observed spectra
.
Chem Mater
2002
;
2
(
5
):
557
563
.
26.
Zhao
Y
,
Hu
Z
,
Fan
C
,
Liu
Z
,
Zhang
R
,
Han
S
,
Liu
J
,
Liu
J
.
Constructing high-performance N-doped carbon nanotubes anode by tuning interlayer spacing and the compatibility mechanism with ether electrolyte for sodium-ion batteries
.
Chem Eng J
2022
;
446
:
137427
.
27.
Obraztsov
I
,
Bakandritsos
A
,
Šedajová
V
,
Langer
R
,
Jakubec
P
,
Zoppellaro
G
,
Pykal
M
,
Presser
V
,
Otyepka
M
,
Zbořil
R
.
Graphene acid for lithium-ion batteries—Carboxylation boosts storage capacity in graphene
.
Adv Energy Mater
2021
;
12
(
5
):
2103010
.
28.
Li
B
,
Guo
J
,
Yuan
Z
,
Xiang
X
,
Jiang
X
,
Nie
J
,
Zhong
B
,
Yuan
X
,
Zu
X
.
Microstructure evolution and performance degradation of sol-gel silica films induced by helium ion irradiation
.
Opt Mater
2024
;
147
:
114689
.
29.
Tabassum
H
,
Zou
R
,
Mahmood
A
,
Liang
Z
,
Wang
Q
,
Zhang
H
,
Gao
S
,
Qu
C
,
Guo
W
,
Guo
S
.
A universal strategy for hollow metal oxide nanoparticles encapsulated into B/N Co-doped graphitic nanotubes as high-performance lithium-ion battery anodes
.
Adv Mater
2018
;
30
(
8
):
1705441
.
30.
Raccichini
R
,
Varzi
A
,
Passerini
S
,
Scrosati
B
.
The role of graphene for electrochemical energy storage
.
Nat Mater
2014
;
14
(
3
):
271
279
.
31.
Qiu
X
,
Yan
X
,
Pang
H
,
Wang
J
,
Sun
D
,
Wei
S
,
Xu
L
,
Tang
Y
.
Isolated Fe single atomic sites anchored on highly steady hollow graphene nanospheres as an efficient electrocatalyst for the oxygen reduction reaction
.
Adv Sci
2018
;
6
(
2
):
1801103
.
32.
Fu
N
,
Liu
Y
,
Liu
R
,
Wang
X
,
Yang
Z
.
Metal cation-assisted synthesis of amorphous B, N Co-doped carbon nanotubes for superior sodium storage
.
Small
2020
;
16
(
20
):
e2001607
.
33.
Tang
Y
,
Wang
X
,
Chen
J
,
Wang
X
,
Wang
D
,
Mao
Z
.
PVP-assisted synthesis of g–C3N4–derived N-doped graphene with tunable interplanar spacing as high-performance lithium/sodium ions battery anodes
.
Carbon
2021
;
174
:
98
109
.
34.
Chen
Y X
,
Liu
H C
,
Xie
W Q
,
Shen
Z
,
Xia
J L
,
Nie
Z Y
,
Xie
J P
.
Diatom frustules decorated with Co nanoparticles for the advanced anode of Li-ion batteries
.
Small
2023
;
19
(
30
):
e2300707
.
35.
Hu
S
,
Ge
Z
,
Tan
J
,
Lai
H
,
Feng
T
,
Zhang
S
,
Xu
Z
,
Zhou
H
,
Cao
X
,
Zhu
G
,
Wu
M
.
Dual-functionally modified N/S doped hierarchical porous carbon and glycerol-engineered polyacrylonitrile carbon nanofibers combine for high-performance lithium-ion capacitors
.
J Power Sources
2023
;
558
:
232624
.
36.
Liu
T
,
Qu
Y
,
Liu
J
,
Zhang
L
,
Cheng
B
,
Yu
J
.
Core-shell structured C@SiO2 hollow spheres decorated with nickel nanoparticles as anode materials for lithium-ion batteries
.
Small
2021
;
17
(
49
):
e2103673
.
37.
Xu
H
,
Zhao
L
,
Liu
X
,
Huang
Q
,
Wang
Y
,
Hou
C
,
Hou
Y
,
Wang
J
,
Dang
F
,
Zhang
J
.
Metal-organic-framework derived core-shell N-doped carbon nanocages embedded with cobalt nanoparticles as high-performance anode materials for lithium-ion batteries
.
Adv Funct Mater
2020
;
30
(
50
):
2006188
.