An electrospinning technique was used to fabricate TiO2 nanofibers for use as binder-free electrodes for lithium-ion batteries. The as-electrospun nanofibers were calcined at 400–1,000°C and characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). SEM and TEM images showed that the fibers have an average diameter of ~100 nm and are composed of nanocrystallites and grains, which grow in size as the calcination temperature increases. The electrochemical properties of the nanofibers were evaluated using galvanostatic cycling and electrochemical impedance spectroscopy. The TiO2 nanofibers calcined at 400°C showed higher electronic conductivity, higher discharge capacity, and better cycling performance than the nanofibers calcined at 600, 800, and 1,000°C. The TiO2 nanofibers calcined at 400°C delivered an initial reversible capacity of 325 mAh·g−1 approaching their theoretical value at 0.1 C rate and over 175 mAh·g−1 at 0.3 C rate with limited capacity fading and Coulombic efficiency between 96 and 100%.
1. Introduction
Rechargeable lithium-ion batteries are among the most advanced electrical energy storage systems available today. Lithium-ion batteries have many notable advantages, particularly their higher volumetric and gravimetric energy densities, when compared with competing battery technologies. These features allow for lighter and smaller battery packs, a long cycle life, and a high-power rate. Nanomaterials also have many advantages, such as a higher surface-to-volume ratio and a shorter path length. Consequently, they have been used in energy storage technology to improve many material properties, leading to materials with high surface reactivity and high carrier transport.
Titanium dioxide (TiO2) is one of the most promising candidates for use as an anode material in lithium-ion batteries due to its relatively high charge-discharge potential compared to graphite, high energy density, low cost, and environmental friendliness [1–3]. However, like many ceramic battery materials, such as LiFePO4 [4], TiO2 has an intrinsically low Li-ion diffusivity and a low electronic conductivity [5], diminishing its potential for applications in lithium-ion batteries. One approach to resolving this problem and increasing the rate performance of TiO2 is to fabricate TiO2 in a one-dimensional nanostructured form. Nanowires [6], nanotubes [7, 8], and nanoribbons [8] of TiO2 have been studied and reported to demonstrate good performance due to their higher electrode/electrolyte contact area and shorter transport path lengths for electrons and Li-ions.
In this work, the structural, morphological, and electrochemical properties of TiO2 nanofibers, another one-dimensional nanostructured form of TiO2, with different anatase to rutile ratios were studied. An electrospinning technique was chosen to prepare the nanofibers due to its low cost and simple setup characteristics. In practice, this method may facilitate the scale-up and mass production of the electrodes for Li-ion batteries without the need of various steps in the conventional process such as mixing (of active materials with conductive additives and binders), casting, vacuum drying, and slitting. This technique can be used to fabricate nanofibers that are exceptionally long and uniform in diameter and that have various compositions.
Much research effort has been focused on the potential applications of electrospun TiO2 nanofibers for Li-ion batteries [8–11]. However, we found that this earlier work reported electrochemical testing of nanofibers using conventional composite electrodes. To ensure that the nanofibers are well mixed with a carbon additive and a binder, substantial grinding and mixing steps are required. Our objective in this work is to report the results obtained by using TiO2 nanofibers as electrode materials for Li-ion batteries directly rather than making conventional composite electrodes. The electrodes contain almost 100 wt% active material. Typical electrodes contain 80 wt% active material and 20 wt% conductive and binder additives. Here, we hope to employ the unique properties of highly crystalline, long, and 3D porous nanofibers without a polymer binder and a conductive additive so that we can evaluate the potential of nanofibers for use as binder-free electrodes for Li-ion batteries. TiO2 can be produced in various polymorphs such as anatase, rutile, and brookite. However, TiO2 in its anatase form is considered the most electroactive Li-insertion host among these polymorphs [12]. Therefore, previous studies focused on the electrochemical properties of the anatase structure. In this work, the electrochemical properties of rutile phase also were studied. Producing nanofibered TiO2 from various ratios of anatase to rutile phases has not been previously reported.
2. Experimental2.1. Preparation of TiO2 Nanofibers
The electrospun TiO2 nanofibers were prepared from precursors containing titanium (diisopropoxide) bis(2, 4-pentanedionate) 75 wt.% in 2propanol (TIAA), polyvinylpyrrolidone (PVP, Mn = 1,300,000), acetic acid, and ethanol. The preparation procedure of these nanofibers is shown in Figure 1. First, 5 mL of TIAA was dissolved in 10 mL ethanol and 10 mL acetic acid. Then, 2.5 g of PVP was dissolved in 22.5 mL ethanol. Both solutions were mixed using a magnetic stirrer at room temperature to form a viscous solution for electrospinning. The solution is loaded into a plastic syringe pump with a needle. The electrospinning system was subjected to a voltage of 15 kV. The solution was fed at a rate of 1 mL/h, and the as-electrospun nanofibers were calcined in air at 400–1,000°C for 3 hrs.
Synthesis procedure of TiO2 nanofibers via electrospinning and calcination at 400–1,000°C in air for 3 hrs.
2.2. Characterization of TiO2 Nanofibers
The crystalline phases were identified using X-ray diffraction (PW3710 MPD Control Systems, the Netherlands) with Cu Kα radiation (λ=0.15406 nm). The morphology of the nanofibers was observed using scanning electron microscopy (SEM; LEO 1450 VP, Cambridge, UK) and transmission electron microscopy (TEM; FEI Tecnai G2 20, Oregon, USA). Microanalysis of carbon content was performed using a LECO CHNS-932 elemental analyzer (LECO Corp., USA).
2.3. Preparation of Electrochemical Cells
The electrochemical performance of the fibers was studied using a square piece of ~3 mg·cm−2 of the TiO2 nanofibers without adding a binder or conductive carbon. Li-metal foil was utilized as the anode and the reference electrodes. The electrolyte was created by dissolving 1 M LiPF6 in a 1 : 1 (v/v) mixture of ethylene carbonate and diethyl carbonate. A microporous polymer (Celgard 2400) was used as the separator. When assembling the electrode, the Li foil was placed on inside of the Swagelok parts, and then the polymer separator was placed on the top. The calcined nanofiber piece was placed on top of the separator, and the electrolyte then added. A Cu disk was finally placed directly on top of the soaked nanofibers. The assembly process was done in an Ar-gas filled glove box. The cells were galvanostatically charged and discharged at 30°C over a voltage range of 0.5–3.0 V versus Li+/Li using a battery cycler constructed from National Instruments (Austin, TX, USA) modular electronic components. The battery cycle was operated by a computer using LabVIEW software. Several electrochemical cells made out of the same batch of the fibers have also been tested using a commercial electrochemical tester (Gamry REF 3000, PA, USA) for the system’s accuracy and yielded indistinguishable results (not shown). Electrochemical impedance spectroscopy (EIS) was performed over a frequency range of 0.1 to 106 Hz with an applied amplitude of 5 mV (Gamry REF 3000, PA, USA).
3. Results and Discussion3.1. Structural Characterization of TiO2 Nanofibers
X-ray diffraction analysis (shown in Figure 2) shows that the nanofibers consist of TiO2 crystallites in the form of either anatase, rutile, or mixed anatase-rutile phase depending on the calcination temperature.
XRD patterns of nanofibers calcined at different temperatures.
The nanofibers calcined at 400°C showed pure anatase phase. The nanofibers calcined at 800 and 1,000°C were composed of highly pure rutile phase, whereas the nanofibers calcined at 600°C consist of mixed anatase-rutile phase. The ratio between the rutile and anatase phases extracted from the XRD spectrum was calculated using the Spurr equation [13]
(1)FR=11+1.26[IA(101)/IR(110)],
where FR is the percentage content of the rutile phase and IA(101) and IR(110) are the integral intensities of the anatase and rutile phases, respectively. The ratio of rutile to anatase for the nanofibers calcined at 600°C was calculated to be 45%.
Figures 3 and 4 show SEM images and corresponding calculated particle size distribution of the as-spun and calcined nanofibers. The average diameter of the as-spun TiO2/PVP composite nanofibers was approximately 160–200 nm. The diameter of the nanofibers calcined at 400–800°C was approximately 100–120 nm.
SEM images of TiO2 nanofibers. (a) As-spun nanofibers. (b)–(e) Fibers calcined in air for 3 h at 400, 600, 800, and 1,000°C, respectively.
The calculated particle size distribution of the TiO2 nanofibers. (a) As-spun nanofibers. (b)–(d) Fibers calcined in air for 3 h at 400, 600, 800, and 1,000°C, respectively.
The reduction in size of the nanofibers after calcination was attributed to the loss of PVP from the nanofibers. It is believed that the PVP is decomposed to form amorphous carbon in air at 300°C [14]. A microanalysis of carbon content performed using a LECO CHNS-932 elemental analyzer shows that the nanofibers calcined at 400°C–1000°C in this study contained approximately 0.35, 0.09, 0.14, and 0.74 weight percent (wt%), respectively. After calcination at 800°C, the morphology of the nanofibers was changed, exhibiting connected particles rather than continuous fibers. Figure 3(e) shows an SEM image of the samples calcined at 1000°C. The samples appear to consist of linked particles with sizes ranging from 100 to 200 nm and have the same diameter as the fibers. These changes in morphology are related to a drastic change in crystal structure from anatase to rutile, as observed in other electrospun fibers such as vanadium pentoxide (V2O5) [15], sodium cobalt spinel (NaCo2O4) [16], and barium strontium titanate [17].
TEM images (Figure 5) show the morphology of the nanofibers at higher magnification. As shown, the nanofibers are composed of nanocrystallites ranging from 5–10 nm in size for the samples calcined at 400°C to 50–70 nm for the nanofibers calcined at 600°C and 50–200 nm for the fibers calcined at 800°C. The grain size of the fibers calcined at 1,000°C was approximately 100–300 nm. The primary crystallites of the nanofibers calcined at 400°C appear to be well connected.
TEM images of carbon-coated TiO2 nanofibers (a)–(d) calcined in air for 3 h at 400, 600, 800, and 1,000°C, respectively.
3.2. Electrochemical Properties of TiO2 Nanofibers
Electrochemical impedance spectroscopy (EIS) technique can be used to study electronic conductivity of the nanofibers. Figure 6 shows the characteristic shape of a Nyquist plot, with a semicircle in the high-frequency region connected by a straight line to the low-frequency region. The semicircle in the high-frequency region corresponds to charge-transfer resistance at the electrode/electrolyte interface. The straight line at low frequencies represents a solid-state diffusion-controlled process of Li-ions [18]. Here, we can see that the charge-transfer resistance of the nanofibers having anatase structure is smaller than that of the samples containing rutile phase. This lower charge-transfer resistance suggests better electron and lithium-ion transport kinetics and is responsible for the better electrochemical performance of the nanofibers with anatase structure.
EIS spectra of TiO2 nanofibers calcined at 400–1,000°C at frequencies ranging from 0.1 to 106 Hz.
The electrochemical performance of the nanofibers calcined at 400–1000°C, which featured polymorphs of pure anatase, mixed rutile and anatase, and pure rutile, was evaluated. Figure 7 shows charge-discharge profiles for the first few cycles of the nanofibers at 0.1 C (0.1 C means that it takes 10 hrs to fully charge this electrode to the cutoff voltage) assuming a theoretical capacity of 335 mAhg−1. As seen, the discharge capacity of the samples calcined at 800–1000°C is lower than those of the samples calcined at 400°C and 600°C. This result shows that the rutile phase in the nanofibrous form of TiO2 is also less electrochemically active than the anatase form. For the nanofibers calcined at 400°C, the low C-rate discharge capacity is very high and approaches its theoretical value. The effect of crystallite size is also evident. It can be seen that when the crystallite size decreases, the electrochemical performance is enhanced. The special properties of these nanostructures, that is, their high surface-to-volume ratio and short lithium diffusion path length with high electronic conductivity, are likely responsible for this observed result. The effects of particle size on enhancing electrochemical performance of battery electrodes have also been observed in other battery materials such as Li4Ti5O12 titanate anodes [19] and LiFePO4 olivine cathodes [20].
Charge-discharge curves of TiO2 nanofibers calcined at 400–1,000°C at a current rate of 0.1 C.
Voltage plateaus at 1.78 V and 1.90 V are observed in the discharge and charge curves, respectively. This finding is related to the phase transformation from-to LixTiO2 (anatase, I41/amd) to-from LiyTiO2 (orthorhombic distortion, Imma) during cycling, where LixTiO2 and LiyTiO2 represent the Li-poor and Li-rich phases of this compound, respectively [21]. The nanofibers calcined at 400°C exhibit the largest plateau region, followed by the fibers calcined at 600, 800, and 1000°C, respectively. Another interesting feature appears at the beginning of the charge and discharge processes of the samples containing different polymorphs. The samples containing the rutile phase show similar voltage profiles, with very similar values of x and y, regardless of the particle size. The nanofibers containing pure anatase phase (calcined at 400°C) show much larger x and y values. This result is interesting and may explain why the anatase form is more electrochemically active than the rutile form. The extended nonstoichiometric behavior in lithium-rich and lithium-poor phases has been correlated to the enhanced rate capability of olivine cathode material as well [22]. As has been established, a transformation related to a topotactic insertion mechanism is much faster than that related to the two-phase transformation mechanism. Hence, samples with large values of x and y should exhibit enhanced electrochemical performance.
Figure 8 shows the rate performance and cyclic stability of the nanofibers with pure anatase structure. Each data point is calculated from the average of three electrodes. The data points for the samples calcined at 600–1,000°C are the averages of three cycles obtained from three different electrodes as well. Clearly, the results obtained for the nanofibers calcined at 400°C containing pure anatase phase are much better than those obtained for the nanofibers calcined at 600–1000°C containing mixed anatase-rutile and pure rutile structures. The cyclic stability of the nanofibers calcined at 400°C was evaluated at a discharge rate of 0.3 C. As shown, the nanofibers exhibit a stable cycle life and almost unchanged capabilities after approximately 50 cycles. The Coulombic efficiency at all C rates is at least 96% and almost 100% at higher C-rates. These Coulombic efficiency numbers are much higher than those found for nanofibers reported previously [8–11]. Therefore, an electrode made from binder-free TiO2 nanofibers possesses acceptable capacity, stable cycling behavior, and acceptable rate performance, showing a promising potential for application as an anode material in lithium-ion batteries, though an optimization of synthesis conditions needs further investigation. Other as-spun nanofibers made from other electrode materials should have been evaluated in order to take the full benefits of the fabrication of nanofibers using the electrospinning method.
Capacity versus cycle number at different current rates of TiO2 nanofibers calcined at 400°C. Results obtained at 0.1–0.3 C for the samples calcined at 600–1,000°C are also shown (inset).
4. Conclusions
In this study, TiO2 nanofibers were prepared using an electrospinning technique. The electrochemical properties of the resulting long and 3D porous fibers were then evaluated directly (as opposed to making composite electrodes) for use as an anode active material in lithium-ion batteries without an additive or a binder. The nanofibers with anatase TiO2 structure exhibited a high lithium storage capacity, a stable cycle life, and good rate capability, while the nanofibers containing rutile phase show poor performance. The enhanced reversible capacity and cycling performance of the anatase TiO2 nanofibers are attributed to the large surface area of the nanofibers, small nanocrystallite size, large Li nonstoichiometric parameters (x, y) of TiO2, and the increased electronic conductivity. This nanofiber material shows promising potential for use as a binder-free anode for rechargeable lithium-ion batteries.
Conflict of Interests
The authors declare that they have no conflict of interests.
Acknowledgments
This project was financially supported by the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission through the Advanced Functional Materials Cluster of Khon Kaen University. Nonglak Meethong acknowledges support by the Nanotechnology Center (NANOTEC), NSTDA, Ministry of Science and Technology, Thailand, through its program of Center of Excellence Network.
HeB. L.DongB.LiH. L.Preparation and electrochemical properties of Ag-modified TiO2 nanotube anode material for lithium-ion battery2007934254302-s2.0-3384767623710.1016/j.elecom.2006.10.008SubramanianV.KarkiA.GnanasekarK. I.EddyF. P.RambabuB.Nanocrystalline TiO2 (anatase) for Li-ion batteries200615911861922-s2.0-3374805302310.1016/j.jpowsour.2006.04.027ReddyM. A.KishoreM. S.PralongV.CaignaertV.VaradarajuU. V.RaveauB.Room temperature synthesis and Li insertion into nanocrystalline rutile TiO2200688129913032-s2.0-3374669307510.1016/j.elecom.2006.05.021BavykinD. V.FriedrichJ. M.WalshF. C.Protonated titanates and TiO2 nanostructured materials: synthesis, properties, and applications20061821280728242-s2.0-3375104412010.1002/adma.200502696ShimH. W.ChoI. S.HongK. S.ChoW. I.KimD. W.Li electroactivity of iron (II) tungstate nanorods201021462-s2.0-7865013639610.1088/0957-4484/21/46/465602465602FangF.FutterJ.MarkwitzA.KennedyJ.UV and humidity sensing properties of ZnO nanorods prepared by the arc discharge method200920242-s2.0-6764918589610.1088/0957-4484/20/24/245502245502ParkS. J.KimY. J.LeeH.Synthesis of carbon-coated TiO2 nanotubes for high-power lithium-ion batteries201119611513351372-s2.0-7995298530910.1016/j.jpowsour.2011.01.105HanH.SongT.BaeJ. Y.NazarL. F.KimH.PaikU.Nitridated TiO2 hollow nanofibers as an anode material for high power lithium ion batteries2011445324536YangZ.DuG.MengQ.GuoZ.YuX.ChenZ.GuoT.ZengR.Synthesis of uniform TiO2@carbon composite nanofibers as anode for lithium ion batteries with enhanced electrochemical performance20122258485854ZhangX.KumarP. S.AravindanV.LiuH. H.SundaramurthyJ.MhaisalkarS. G.DuongH. M.RamakrishnaS.MadhaviS.Electrospun TiO2-graphene composite nanofibers as a highly durable insertion anode for lithium ion batteries2012116281478014788ZhuP.WuY.ReddyM. V.NairA. S.ChowdariB. V. R.RamakrishnaS.Long term cycling studies of electrospun TiO2 nanostructures and their composites with MWCNTs for rechargeable Li-ion batterie20122531537MurphyD. W.CavaR. J.ZahurakS. M.SantoroA.Ternary LixTiO2 phases from insertion reactions19839-1014134172-s2.0-0020897996SpurrR. A.Quantitative analysis of anatase-rutile mixtures with an X-ray diffractometer195729576076210.1021/ac60125a0062-s2.0-0038535661NuansingW.NinmuangS.JarernboonW.MaensiriS.SeraphinS.Structural characterization and morphology of electrospun TiO2 nanofibers2006131147155ViswanathamurthiP.BhattaraiN.KimH. Y.LeeD. R.Vanadium pentoxide nanofibers by electrospinning20034965775812-s2.0-003777469110.1016/S1359-6462(03)00333-6MaensiriS.NuansingW.Thermoelectric oxide NaCo2O4 nanofibers fabricated by electrospinning20069911041082-s2.0-3374565566210.1016/j.matchemphys.2005.10.004MaensiriS.NuansingW.KlinkaewnarongJ.LaokulP.KhemprasitJ.Nanofibers of barium strontium titanate (BST) by sol-gel processing and electrospinning200629725785832-s2.0-3364607088210.1016/j.jcis.2005.11.005LeviM. D.AurbachD.Diffusion coefficients of lithium ions during intercalation into graphite derived from the simultaneous measurements and modeling of electrochemical impedance and potentiostatic intermittent titration characteristics of thin graphite electrodes1997101234641464710.1021/jp97019112-s2.0-0006785152WagemakerM.BorgholsW. J. H.MulderF. M.Large impact of particle size on insertion reactions. A case for anatase LixTiO2200712914432343272-s2.0-3424713726210.1021/ja067733pMeethongN.HuangH. Y. S.CarterW. C.ChiangY. M.Size-dependent lithium miscibility gap in nanoscale Li1−xFePO42007105A134A1382-s2.0-3394732657310.1149/1.2710960AldonL.KubiakP.PicardA.JumasJ. C.Olivier-FourcadeJ.Size particle effects on lithium insertion into Sn-doped TiO2 anatase2006186140114062-s2.0-3364553646910.1021/cm051445vMeethongN.HuangH. Y. S.SpeakmanS. A.CarterW. C.ChiangY. M.Strain accommodation during phase transformations in olivine-based cathodes as a materials selection criterion for high-power rechargeable batteries2007177111511232-s2.0-3424857272010.1002/adfm.200600938