ES Materials & Manufacturing, 2018, 1, 67-71
Published online: 08 Oct 2018
Received 04 Sep 2018, Accepted 07 Oct 2018
Xiao-Chong Zhao,1 Pan Yang,1 Li-Jun Yang,1 Yu Cheng,1,3 Hui-Yuan Chen,3
Hu Liu,4,5 Gang Wang,2,* Vignesh Murugadoss,4,6
Subramania Angaiah,6,* and Zhanhu Guo4,*
1Institute of Materials, China Academy of Engineering Physics, Mianyang 621908, Sichuan, China.
2Institute of Chemical Engineering, Qinghai University, Xining 810016, China.
3Qinghai Nationalities University, Xining 810007, China.
4Integrated Composites Laboratory (ICL), Department of Chemical and Biomolecular Engineering, University of Tennessee, Knoxville, TN 37996 USA.
5National Engineering Research Center for Advanced Polymer Processing Technology, Zhengzhou University, Zhengzhou 450002, China.
6Electrochemical Energy Research Lab, Centre for Nanoscience and Technology,
Pondicherry University, Puducherry – 605 014.
TiO2 nanoparticles doped with Cu2+ are synthesized via a facile one-step solvothermal method with a uniform distribution of 50-60nm. The XPS results show that the Cu2+ are doped in TiO2 crystal lattice uniformly. Due to the smooth replacement of the part of Ti4+ sites by Cu2+ in the samples, more Ti4+ vacancies are formed, which is a benefit to the Li+ diffusion and enhanced electrochemical properties. At the 5C rate, the initial discharge capacity of TiO2 doped with 6 wt% Cu2+ reached 83.4 mAhg-1. After 100 charge-discharge cycles, the discharge capacity is still 76.5 mAhg-1, showing good cycling stability.
Table of Content
Effect of doping percentage of Cu2+ ions on the performance TiO2 nanoparticles as anode materials for lithium-ion battery were investigated for the first time.
Keywords: Cu2+ doping; TiO2; Electrochemical Performance; Lithium-ion Battery
Lithium-ion batteries (LIBs) have been widely used in portable electronic devices due to their superior properties such as high energy density, long cycle life, no memory effect and environmental friendliness.1-4 With rising interest in green electrode materials for LIBs, increasing attention has been paid to titanium dioxide (TiO2) anode material in recent years because of its long cycle life, low cost, and minimum environmental impact.5-9 Furthermore, the lithium insertion/extraction voltage of a TiO2 anode vs. Li+ /Li reaches 1.5 V, which can effectively avoid the formation of solid electrolyte interfaces (SEI) layers and lithium plating on the anode, and improve the safety of the batteries as compared with its carbon-based counterparts.10,11 However, TiO2 is also suffering from inherent drawbacks as a potential anode material in LIBs, such as the low ionic diffusivity, poor electronic conductivity in electrodes, and high resistance at the interface of electrode/electrolyte at high charge/discharge rates.12,13
To address these issues, a series of strategies have been developed to improve the structural integrity and electrical conductivity of TiO2-based materials, such as optimizing particle size distribution or surface morphology,14-17 and fabricating TiO2/carbon hybrids.18-20 Minella M et al synthesize titanium dioxide/reduced graphene oxide (TiO2-rGO) composites at different loadings of the carbonaceous phase and focus on the simplicity and low cost of the electrode production.13 Yu W et al synthesize anatase TiO2 ultra-thin nanosheets by a facile hydrothermal method reveals an excellent cycling performance.11
In this study, we have presented a facile one-step solvothermal method, which is based on the idea of increasing diffusion channel of Li+, for the synthesis of TiO2 nanoparticles doped with Cu2+. The effects of the Cu2+ doping on the structure, morphology and surface area of the TiO2 samples are investigated in detail. Furthermore, the Li+ diffusion in the LIBs during charge/discharge process is characterized by electrochemical measurements. Due to the uniform replacement of the part of Ti4+ sites by Cu2+, more Ti4+ vacancies are formed, which is a benefit to the Li+ diffusion.
2.1 Synthesis of TiO2 nanoparticles doped with Cu2+.
All materials were used as received without further purification. Tetrabutyl titanate, isopropanol, hydrofluoric acid (HF, 40 wt%), CuCl2⸱2H2O were purchased from Sigma-Aldrich.
Firstly, 30 ml of analytically pure isopropanol was taken in a 50ml polytetrafluoroethylene container. Secondly, a required quantity of CuCl2⸱2H2O powder (The Cu2+: Ti4+ mass ratio was controlled as 0, 6 wt%, 9 wt%, 12 wt%, corresponding to the samples of S-0, S-6wt%, S-9wt%, S-12wt%) was added to it. Thirdly, 5 ml tetrabutyl titanate was dropped into the above mixture with continuous magnetic stirring for 10 min. Finally, 0.6 ml HF was added dropwise into the above mixture.
After a solvothermal reaction in an oven for 6 h at 180 °C, the resulting precipitate consisting of TiO2 nanoparticles doped with Cu2+ was washed with distilled water and ethyl alcohol for three times, separately. After sintering at 500 °C for 2 h in an argon atmosphere, the TiO2 nanoparticles doped with Cu2+ were obtained.
2.2 Cell assembly
N-methyl-2-pyrrolidone (NMP, Aldrich) based slurry of the as-prepared doped with Cu2+ doped TiO2 nanoparticles was mixed with acetylene black (Shawinigan Black AB50, Chevron Corp., USA) as electronic conduction enhancer and poly-(vinylidene fluoride) as binder (PVDF, Solvay Solef-6020), in the weight ratio of 80 : 10 : 10. The mixture was then deposited over a copper foil as current collector using a standard “doctor blade” technique. After the evaporation of the solvent, the film was hot pressed (10 minutes at 70 °C and 200 bar) in order to improve the adhesion, cut into disks and outgassed, then transferred into an Ar-filled dry glove-box (MBraun Labstar, H2O and O2 content <1 ppm) for cell assembly. The electrode loading density of active materials is about 1.1 mg/cm2. Three-electrode T-cells were assembled by contacting in sequence the working electrode (having the above described composition), a 1.0 M lithium perchlorate (LiClO4, Aldrich) in a 1:1 v/v% mixture of ethylene carbonate (EC, Fluka) and diethyl carbonate (DMC, Aldrich) electrolyte soaked on a polypropylene separator (2400, Celgard) and a lithium foil (high purity lithium foils, Chemetall Foote corporation) counter electrode.
2.3 Characterization and Electrochemical Measurement
The X-Ray Diffraction (XRD) was characterized by TD-3500, Dandong Tongda, China. The scanning electron microscope (SEM) was characterized by S5200, Hitachi, Japan. The surface area and pore size distribution were characterized by QUADRASORB2QDS-evo, Quantachrome Instruments, America. The X-ray photoelectron spectroscopy (XPS) was characterized by ESCALAB 250 XI, Thermo Scientific, America. The CT2001a cell test instrument (LAND Electronic Co.) was used to test the galvanostatic discharge/charge cycling performance of cells with a potential range of 1.0 to 3.0 V (vs. Li/Li+) at 25 °C. The cyclic voltammetry (CV) and electrochemical impedance spectrum (EIS) characterization were carried out using a Solartron 1470E electrochemistry workstation.
Phase composition and purity of the samples are characterized by XRD as shown in Figure 1. The pattern is obviously the same with that standard spectrum of anatase TiO2, and all the peaks can be well assigned to the TiO2 ((JCPDS NO.21-1272, I41/amd, a=b=3.7845Å, c =9.5143Å). This indicates that the as-prepared material has good crystallinity and high purity. When the Cu2+ doping amount is less than ≤9wt%, the XRD patterns have no indications of other crystalline derivatives, such as CuO or Cu2O, indicating that the Cu2+ is doped into the TiO2 lattice completely. Meanwhile, the obvious (004) peak reveals the wide existence of highly active (001) plane. The (001) plane of anatase TiO2 exposure can improve the Li+ diffusion coefficient, which is conducive to the fast charge/discharge process. However, when the Cu2+ doping amount reaches 12wt%, there is another extra peak appearing, which is corresponding to the peak of CuO (JCPDS NO.02-1042).11
Figure 1. XRD pattern of samples (a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt%
Figure 2 presents the morphological results characterized by SEM. The as-prepared samples show a relatively uniform distribution of 50-60nm. The nanoparticles trend a kind of progressive changing process from circular to square. This is consistent with the appearance of (004) plane in the XRD pattern. The SEM results indicate the Cu2+ doping has no obvious effects on the anatase TiO2 morphology. Meanwhile, the specific surface areas of the samples, (a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt% are 38.2 m2 g-1, 35.2 m2 g-1, 33.8 m2 g-1, 31.9 m2 g-1, respectively, showing a typically decreasing trend with the increasing Cu2+ doping.12,13
Figure 2. SEM image of samples (a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt%
Figure 3 presents the XPS energy spectrum of samples with 6wt% Cu2+ doping, full spectrum (a), O peak (b), Ti peak (c) and Cu peak (d), as well as the element mapping spectra of O (e), Ti (f) and Cu (g). The two peaks at 458.7 and 464.5 eV in Figure 3 (c) correspond to the Ti2p3/2 and Ti2p1/2 state divided from the Ti2p spin-orbit interaction energy. However, these Ti2p3/2 and Ti2p1/2 peaks shifted from the respective 459.2 and 464.7 eV peaks of pure TiO2 reported elsewhere,21,22 which possibly due to the greater number of oxygen vacancies in Cu2+ doped TiO2. This increment in the oxygen vacancies can contribute to electrical conductivity and thereby the enhanced electrochemical performance.23 The O1s spectrum reveals that the oxygen element exists in the form of O-Ti. Meanwhile, the electron binding energy of Cu2p3/2 and Cu2p1/2 is 932.1eV and 952.0eV, separately. These are consistent with the binding energy of Cu2+.24 From the mapping results of Figure 3 (e), (f), and (g), the O, Ti and Cu show a very uniform distribution, indicating that the Cu2+ is doped into the lattice of TiO2 uniformly.14
Figure 3. (a) XPS spectra of S-6wt%, (b) O1s, (c) Ti2p, (d) Cu2p, and
elemental mapping of (e) O, (f) Ti & (g) Cu
Figure 4 presents the initial charge/discharge curves of samples measured at a rate of 0.5C. The initial discharge capacity for the samples of S-0, S-6wt%, S-9wt%, and S-12wt% reaches 151.2, 156.5, 173.8, 138.2 mAhg-1, separately. The higher initial discharge capacity of the S-9wt% may due to their relatively higher electronic conductivity as evidenced in Figure 6, which results in rapid charge transport and a lower electrochemical polarization. These are also consistent with the trend that proper doping of Cu2+ is beneficial to improve the specific capacity.18,19
Figure 4. Initial discharge-charge curves of Samples (a) S-0 (b) S-6wt%
(c) S-9wt% (d) S-12wt% at 0.5C rate
To examine the rate performance and cyclic stability, the charge capacity at the 5C rate was investigated for 100 cycles as shown in Figure 5. The samples of S-0, S-6wt%, S-9wt%, and S-12wt% show initial discharge specific capacity of 76.3, 83.4, 69.4, 79.1 mAh g-1. The initial capacity discharge decreased by 46.7% and 60.06% for S-6wt% and S-9wt% at 5C compared to 0.5C, respectively, indicating that the S-6wt% is more suitable for higher current. This probably results from the comparatively larger volume change as well as copper ion dissolution of S-9wt% during charge-discharge cycle at higher currents.25 After 100 cycles, the capacity gradually dropped to 67.5, 76.5, 69.2, 60.2 mAhg-1, respectively. The Cu2+ doped TiO2 exhibit superior cyclic stability, implies their excellent structural stability during the lithium de-intercalation/intercalation process. These results refer that proper doping of Cu2+ can improve the rate performance and cycle stability obviously. Because the replacement from Ti4+ to Cu2+ in TiO2 lattice during the doping, more Ti4+ vacancies are formed, increasing the Li+ diffusion coefficient as well as benefiting the fast Li+ charge/discharge process. However, when the Cu2+ doping amount reaches 12 wt%, CuO appears as shown in Figure 1, which can decrease the conductivity of as-prepared samples. Furthermore, during the cycling process, the volume effect of CuO reaches about 170%, which may cause decreasing electrochemical properties seriously.
Figure 4. Charge capacity as a function of cycle number at the 5C rate.
(a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt%
Figure 6 presents the electrochemical impedance spectroscopy (EIS) results for the samples of S-0, S-6wt%, S-9wt%, and S-12wt% after 100 cycles. The respective equivalent circuit is shown in the inset of Figure 6. All these EIS curves are composed of one semicircle and one straight line. The semicircle in the high-frequency range is consistent with the charge transfer process, responding to the charge transfer impedance (Rct). The smaller the diameter of the semicircle, representing lower Rct, the better the electrochemical properties of as-prepared samples. The line in low-frequency range responds to Li+ body diffusion process in the active material, responding to the Warburg impedance. It can be seen that the samples of S-6wt% and S-9wt% with proper Cu2+ doping, show a lower Rct compared to that of pure TiO2 samples. This is consistent with the initial charge/discharge curves as shown in Figure 4 and Figure 5. However, when more Cu2+ are doped into TiO2, the samples of S-12wt% show a basically equivalent Rct with that of pure TiO2 samples. That is because too much Cu2+ doping cause the appearance of CuO, which decreases the conductivity of active materials.24
Figure 6. The EIS spectra after 100 cycles; (a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt%. Inset: A Respective equivalent circuit for fitting.
To better elucidate the electrochemical mechanism of the Cu2+ doping anatase TiO2 nanoparticles, the cyclic voltammogram (CV) in the range between 1.0 and 3.0 V are analyzed and the result is displayed in Figure 7. In the cycle, a pair of reduction/oxidation peaks at 1.68/2.09 V belong to the Li+ insertion/extraction processes between TiO2 and Li0.5TiO2. When the Cu2+ doping amount is proper for the samples of S-6wt% and S-9wt%, the peak current densities become higher than that of S-0. Meanwhile, the redox peaks become narrower. Both the higher current density and narrower redox peak indicates that the conductivity becomes better for the samples of S-6wt% and S-9wt%. The positions of the redox peaks almost remain unchanged in the 5th cycles, revealing the enhanced cyclability. When the Cu2+ doping amount is 12wt%, the peak current densities decrease to that of S-0. The appearance of CuO in the as-prepared samples results in the decreasing reactive activity of as-prepared samples. Furthermore, the redox peaks move to keep away from each other. This infers that the electrode polarization happens with the appearance of CuO.20,24
Figure 7. Cyclic voltammetry of samples(a) S-0 (b) S-6wt% (c) S-9wt% (d) S-12wt% at a scanning rate of 0.2 mV s-1
A facile one-step solvothermal method is deployed to synthesize TiO2 nanoparticles doped with different content of Cu2+. The XPS results show that Cu2+ is doped in TiO2 crystal lattice uniformly. At the rate of 5C, the initial discharge capacity of samples doped with 6 wt% Cu2+ reached 83.4 mAhg-1. After 100 charge-discharge cycles, the discharge capacity is still 76.5 mAhg-1, showing good cycling stability. Because of the replacement from Ti4+ to Cu2+ in TiO2 lattice during the doping, more Ti4+ vacancy is formed, increasing the Li+ diffusion coefficient as well as benefiting the fast Li+ charge/discharge process. All these efforts result in enhanced electrochemical properties.
The authors express their gratitude for the support provided by the Qinghai International Cooperation Project (No. 2014-HZ-816), National Natural Science Foundation of China (NSFC No. 51402267 and 51508484) and the State Key Laboratory of New Ceramic and Fine Processing Tsinghua University (KF201715) for providing financial support.
Authors declare no conflict of interest