MICROSCOPY RESEARCH AND TECHNIQUE 77:170–175 (2014)
Porous TiO2 Nanowire Microsphere Constructed by Spray Drying and its Electrochemical Lithium Storage Properties CHUNJU LV,* TIAN HU, KANGYING SHU, DA CHEN, AND GUANGLEI TIAN College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, People’s Republic of China
KEY WORDS
microsphere; spray drying; agglomeration; lithium-ion battery
ABSTRACT Porous TiO2 nanowire microspheres with greatly decreasing agglomeration were successfully prepared by spray drying of hydrothermal reaction suspension, followed by calcination at 350 C. The as-obtained nanowire microspheres with TiO2-B structure reach an initial discharge capacity 210 mAh g21 with an irreversible capacity 25 mAh g21 at a current density of 20 mA g21. For the 450 C-calcined one with anatase TiO2 crystal structure, the initial discharge capacity is 245 mAh g21 but with a much higher irreversible capacity of 80 mAh g21. The hierarchical porous structure in the 350 C-calcined TiO2 nanowire microspheres collapsed at 450 C, annihilating the main benefit of nanostructuring. Microsc. Res. Tech. 77:170–175, 2014. V 2013 Wiley Periodicals, Inc. C
INTRODUCTION The carbon negative electrode widely used in the present commercial rechargeable Li-ion batteries suffers from a number of problems and cannot meet the performance requirements of some important applications satisfactorily, especially in the safety and rate performance (Fong et al., 12; Roth and Doughty, 22). As a result, there is now great interest in alternative anodes including transition metal oxides such as Fe2O3, Mn3O4, MoO3, WO3, and TiO2 (Chen et al., 10; Gao et al., 13; Hassan et al., 14; Li and Fu, 16; Murphy et al., 19 ; Zachau-Christiansen et al., 31). Especially, titanium oxide is regarded as a promising active lithium intercalation material which has been proven to have superior properties in terms of higher theoretical capacity of 335 mAh g21, a lower self-discharge rate, chemical stability, environmental benignancy, and low cost (Bruce et al., 8; Huang et al., 15; Ohzuku et al., 21; Zachau-Christiansen et al., 31). Satisfactory results have been obtained by using nanostructured TiO2, which can be beneficial to both power rate and cycle life (Armstrong et al., 1,2,b,c; Bavykin et al., 5; Bruce et al., 8; Deng et al., 11; Sudant et al., 24; Wagemaker et al., 26). Indeed, the basic requirement for many high-rate insertion battery materials is that they are prepared in a nanostructured form. In this way, the surface area can be maximized and the solid-state transport length minimized. As for TiO2, the dimensional confinement imposed by the TiO2 nanowire morphology makes it an ideal host for Li1 intercalation/ deintercalation and the controlled introduction of electrons into TiO2 nanowire. So far, different ways especially the hydrothermal reaction method have been applied successfully to synthesize TiO2 nanowires because the hydrothermal method is convenient, low cost, and easily adaptable to mass production. However, when we try to obtain the active TiO2 nanowire material in a solid form by means of the conventional centrifugal separation and subsequent oven drying of the product from the hydroC V
2013 WILEY PERIODICALS, INC.
thermal reaction, there will be a huge tendency toward agglomeration. Preventing agglomeration of nanoparticles during the drying progress of liquidphase preparation is one of the most important problems. Different drying methods to treat the precursors from liquid-phase preparation such as spray drying, microwave drying, and supercritical fluid drying were applied to prevent the agglomeration of nanoparticles (Bian et al., 6; Wang et al., 27; Ye et al., 30). Spraydrying technique exhibits advantages like low price, high efficiency. In the spray-dryers, a suspension is fed into the drying chamber, then the slip is atomized by pumping it at high pressure through a multi-nozzle array, after that the upward spiraling droplets encounter hot air and fed through a diffuser into the chamber (Vicent et al., 25). This makes it necessary to prepare nanoparticles granules with a good flowability and a size without inducing additional hard agglomerates. Nakahara et al. (20) reported excellent cycling performance of particulate Li4Ti5O12 prepared by spray drying process with LiOH and anatase type TiO2 as precursors, and it was found that granular morphology of the active Li4Ti5O12 was essential to the high rate capacity of the active materials. In the present investigation, we report the synthesis of porous TiO2 nanowire microspheres by spray drying of TiO2 nanowires suspension, which was obtained from a hydrothermal chemical reaction, followed by calcination at 350 C. The characteristics of the as-prepared powders such as morphology, particle size, and microstructure were investigated. The relationship *Correspondence to: Chunju Lv, College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, People’s Republic of China. E-mail: [email protected] Received 22 September 2013; accepted in revised form 19 November 2013 REVIEW EDITOR: Dr. Chuanbin Mao Contract grant sponsor: Zhejiang Analysis Test Project of China; Contract grant number: 2009F70010; Contract grant sponsor: National Natural Science Foundation of China; Contract grant numbers: 21003111, 21210102012. DOI 10.1002/jemt.22324 Published online 3 December 2013 in Wiley Online Library (wileyonlinelibrary.com).
TIO2 NANOWIRES BY SPRAY DRYING
171
Fig. 1. (a) SEM image and (b) EDS analysis of the oven-dried TiO2 nanowire precursor; (c) TGA curves and (d) XRD patterns of TiO2 nanowire precursor by different drying processes The bottom right inset in (a) is an enlarged SEM image.
between microstructure, electrochemical properties as a lithium intercalation host, and processing technologies of the active material was discussed carefully. RESULTS AND DISCUSSION The morphology, composition, and structures of the as-obtained precursors by hydrothermal reaction at 180 C, acid treatment, and subsequent different drying processes are displayed in Figure 1. As clearly seen from the scanning electron microscope (SEM) image (Fig. 1a) of the oven-dried precursor, the material shows large-scale formation of cotton-like texture formed by thousands of entangled nanowires as suggested in the enlarged high-resolution inset image. Lin and coworkers (2008) also observed such texture which was composed of entangled TiO2-B nanofibers by the reaction of mesoporous anatase TiO2 at 90 C, followed by addition of 0.1 M HCl solution into the reaction precipitate to reach pH 1 and subsequent deionized water washing. In our research, the aggregation of curved nanowires is severe when the oven-drying process was adopted to deal with the acid and deionized waterwashed precipitate. Microscopy Research and Technique
The purity and chemical composition of the ovendried precursor are confirmed by energy dispersive spectroscopy (EDS) analysis (Fig. 1b), which mainly shows strong Ti and O signals. Only a very small amount of sodium residual from the raw NaOH stuff is detected. The atom ratio of Ti/O is very close to 3/7. At the same time, the thermal gravimetric analysis (TGA) curve (Fig. 1c) reveals that the oven-dried precursor contains 16 wt % weight loss in the whole range from room temperature to 800 C which could be partly attributed to the dehydration of the precursor. The reason that no H element signal is traced in the EDS spectrum comes from its technical limitation in the field of analyzing light elements. To further confirm the crystal structure of the oven-dried precursor, X-ray diffraction (XRD) analysis was carried out. As seen from the related XRD pattern (Fig. 1d), the profile for the obtained-product by hydrothermal reaction and acid treatment is found to be similar to that of the material prepared previously by Liu and coworkers (2008), which has been proposed to be due to H2Ti3O7 phase (JCPDS 41–0192). In a typical hydrothermal synthesis process, the layered hydrogen titanate H2TixO2x11
172
C. LV ET AL.
Fig. 2. Variable magnification SEM image of 350 C-calcined TiO2 nanowires from oven-dried (a,c) and spray-dried precursors (b,d).
phase was obtained by ion exchange through acid washing sodium titanate that was produced by an alkaline hydrothermal treatment. In our research, the Ti/O atom ratio of the as-obtained hydrogen titanate by the EDS analysis corresponds to that of H2Ti3O7. But the 16 wt % weight loss of the as-obtained hydrogen titanate after heating up to 800 C by TGA analysis is much higher than that of the complete dehydration of H2Ti3O7 to TiO2 which is calculated theoretically to be about 7 wt %. Moreover, nearly no weight loss is found above 450 C. So the weight loss below 450 C should be also related to the evaporation of the absorbed water and interlayer water except for the dehydration of structural water in the as-prepared hydrogen titanate with nanowire structure. In the present work, the spay-dried precursor was also characterized by TGA and XRD analysis. Because a certain amount of absorbed water and/or interlayer water had evaporated when the precipitate was spray dried at the inlet and outlet temperature of 150 and 80 C, a weight loss of 14 wt % for the spay-dried precursor which is a little lower than that of the 60 C oven-dried product is observed, as shown in Figure 1c. And the XRD profile for the spray-dried sample has no big difference from that of the oven-dried sample, implying that the drying process we applied has no influence on the crystal structure of the hydrogen titanate. Based on the TGA profiles, the oven-dried and spray-dried hydrogen titanate precursors were both
calcined at 350 C at which not only the evaporation of the absorbed and/or interlayer water but also the dehydration process of the structural water have almost finished. Figure 2 shows the morphology images of the two different calcination products, obviously suggesting the effect of drying process on the clustering of nanowires. When using the conventional oven drying, the overall SEM image from the ovendried precursor as shown in Figure 2a clearly reveals that it retains the same cotton-like morphology as the original oven-dried precursor where individual nanowires are severely clustered together. This clustering of nanowires will prohibit direct access of Li-ion and electrons to the whole surface of the primary nanowires and thereby annihilate the main benefit of nanostructuring. But the as-obtained product by spray drying and subsequent calcination at 350 C demonstrates a porous TiO2 nanowire microsphere morphology with a size ranging from 2 lm to 10 lm (Figs. 2b and 2d). As clearly seen from the higher magnification SEM image (Figs. 2c and 2d), the large amounts of TiO2 nanowires entanglement decreases greatly for the spray-dried product, compared with the conventional oven-dried one. Thus, the surface area of the TiO2 nanowires can be further maximized which will make it an ideal host for Li1 intercalation/deintercalation and the controlled introduction of electrons into TiO2 nanowire. Hereafter, the spray drying method is adopted to prepare much more TiO2 nanowire precursor for further investigation. Microscopy Research and Technique
TIO2 NANOWIRES BY SPRAY DRYING
To investigate the effect of annealing temperature, the spray-dried nanowire precursor with H2Ti3O7 phase was calcined at 350 C and 450 C for 3 h in air respectively. In our work, the spray-dried hydrogen titanate transforms into TiO2-B (JCPDS 46–1237) after heating at 350 C as shown in Figure 3, which is
Fig. 3. XRD patterns of TiO2 nanowires from the spray-dried precursor and subsequent calcinations at 350 C and 450 C, respectively.
consistent with those of TiO2-B reported previously (Nakahara et al., 20; Shieh et al., 23). Further increasing the temperature to 450 C, the diffraction peaks ascribed to anatase TiO2 phase (JCPDS 21-1272) are observed. And the peak at a 2h value of about 10 due to TiO2-B crystal structure is still present in the spectra, indicating that more thermal stable anatase TiO2 containing minor residual TiO2-B phase can be obtained via the post-heat treatment of the protonated titanate nanowires at 450 C for 3 h. Figure 4 shows SEM images of the TiO2 sample from the spray-dried precursor and subsequent calcination at 450 C. As can be seen from the low magnification SEM images in Figures 4a and 4b, the microsphere morphology within a small size range constructed by spray drying treatment is maintained. The higher magnification SEM image as shown in Figure 4c indicates that the as-obtained anatase TiO2 from the spray-dried hydrogen titanate and subsequent calcination at 450 C still keeps 1D shape. However, the hierarchical porous structure observed in the nanowire microspheres from the 350 C-calcined TiO2 product collapses after calcination at 450 C. Figure 5 displays the galvanostatic discharging– charging curves of two electrodes made from different TiO2 nanowires at a current rate of 20 mA g21 over the potential range of 1.0–2.5 V (versus Li1/Li). The discharging–charging profile for the 350 C-calcined
Fig. 4. Variable magnification on SEM image of TiO2 nanowires from the spray-dried precursor and subsequent calcination at 450 C.
Microscopy Research and Technique
173
174
C. LV ET AL.
Fig. 5. Initial charge–discharge profiles of TiO2 nanowires from the spray-dried precursor and subsequent calcination at 350 C and 450 C at a current rate of 20 mA g21 between 1.0 and 2.5 V.
Fig. 6. Cycling performance of TiO2 nanowires from the spraydried precursor and subsequent calcination at 350 C at a current rate of 20 mA g21 between 1.0 and 2.5 V.
TiO2 nanowires with TiO2-B structure is characterized by a monotonous potential decrease corresponding to a solid solution insertion mechanism. Such mechanism is related to TiO2-B structure but has nothing to do with its morphology such as powder, nanoribbon, nanofiber, nanosheet which was evidenced in other studies (Brousse et al., 7; Liu et al., 18; Wang et al., 28). The initial discharge specific capacity for the asprepared TiO2 nanowires calcined at 350 C is 210 mAh g21 with a small irreversible capacity (charge capacity less than discharge for the anode materials) of about 25 mAh g21. The potential profiles for the 450 C-calcined TiO2 nanowires containing the main anatase phase and residual TiO2-B can be divided into two characteristic domains, which are induced by the two coexisting phases. One domain consists of the plateaus from two phase intercalation, which is a typical shape for anatase TiO2 (Chen et al., 9; Wang et al., 29). The discharge and charge potential plateaus are presented at about 1.75 and 1.95 V (vs. Li1/Li) for lithium insertion and extraction, respectively, which is related to the fraction of the anatase phase. Another domain beyond the plateaus characterized by a slopping potential decrease is a further Li-insertion into layer nanowires or the residual TiO2-B. As shown in Figure 5, at the first cycle at 20 mA g21, the as-prepared TiO2 nanowires calcined at 450 C can reach a high discharge specific capacity of 245 mAh.21. However, a much higher irreversible capacity about 80 mAh g21 than that of the 350 C-calcined TiO2 electrode material emerges, which resulted from the collapse of the hierarchical porous structure after heating at higher temperature. The loosely-attached individual nanowires in the 350 C-calcined TiO2 with porous microsphere morphology facilitate their contact with electrolyte and hence maximize the electrode/electrolyte contact area, favoring the Li ion transport. Figure 6 shows the cycling performance of TiO2 naowires from the spry-dried precursor and subsequent calcination at 350 C at a current rate of 20 mA g21 between 1.0 and 2.5 V. The charge specific capacities tend to decrease with increasing cycling: approximately 42.9% (90 mAh g21) of its initial discharge
capacity is retained after 20 cycles. The average capacity fading was close to 2.8% at each cycle, showing unsatisfying cycling stability. In this aspect, further work is being carried out to improve the lithium storage performance of the porous TiO2 nanowire microspheres. EXPERIMENTAL The preparation of titanate nanowires was referred in previous literatures (Armstrong et al., 1,2,b). TiO2 powders (Degussa P25, ca. 80% anatase, 20% rutile) were used as received. All other chemicals were of analytical reagent grade and used without further purification. Deionized water was used throughout. In a typical synthesis process, 1.0 g commercial P25 powders were added into aqueous NaOH solution (60 mL, 10 mol L21). After stirring for 1 h, the resulting suspension was transferred into a Teflon-lined stainless steel autoclave (100 mL). The autoclave was maintained at 180 C for 12 h and then naturally cooled to room temperature. Subsequently, the obtained precipitates were washed with HCl (0.1 mol L21) solution and then with deionized water until the pH value approaching about 7.0. Subsequently, the precipitates were dispersed into 120 mL deionized water to form homogeneous slurry with the assistance of ultrasound. Afterwards, the homogeneous slurry was spray-dried at the inlet and outlet temperature of 150 and 80 C, respectively. Finally, the dried powders were heated at different temperature for 3 h in air in a covered alumina crucible with a heating rate of 2 C min21. For comparison, the acid and deionized water-washed precipitate was dried in oven at 60 C for 12 h and then developed at the same calcination conditions. XRD measurements were performed on a diffractometer (Bruker D2 PHASER) with Cu Ka radiation in the range from 10 to 90 . The particle size and morphology of the samples were studied with a SEM (JEM-5610) coupled with an EDS. The TGA was performed using a METTLER TOLEDO TGA/DSC 1LF1600 apparatus. The experiments were performed by heating from 30 to 800 C at 5 C min21 in air environment. Microscopy Research and Technique
TIO2 NANOWIRES BY SPRAY DRYING
Coin cells (CR 2015) were constructed in a dry room to investigate Li-ion intercalation behavior. The working electrodes were prepared from the as-synthesized active materials, carbon black as a conducting agent, and poly(vinylidene difluoride) (PVDF) binder dissolved in n-methyl pyrolidinone(NMP, 8 wt %) at a weight ratio of 80:10:10. The slurry was mixed overnight to obtain a homogeneous black paste which is then spread onto a copper foil current collector. The solvent is removed at 120 C in vacuum for 12 h. Copper foil pieces with 14 mm diameter were then cut off and used as the electrode in homemade cells which were assembled inside Ar filled glove box. The average amount of TiO2 nanowires in a round piece of electrode is about 5 mg. Each electrode was carefully weighted before use and several electrodes were tested to assume the reproducibility of the electrochemical behavior. The coin cells also used Li metal as a counter electrode, a Celgard 2500 as a separator, and a 1 M LiPF6 dissolved in ethylene carbonate, dimethyl carbonate, and ethylene methyl carbonate (1:1:1, vol/vol/ vol) as an electrolyte. The assembled cells were aged for 12 h before testing and then were galvanostatically charged and discharged over a voltage range of 2.5–1.0 V vs. Li1/Li at a current density of 20 mA g21 using a LAND cell test (Land-CT 2001A) system. The specific capacity was calculated based on the mass of active materials in the electrode. CONCLUSION In summary, TiO2 nanowires were prepared on a large scale by a hydrothermal reaction followed by acid washing, drying processing, and post-heat treatment at different temperature in air. The drying process has an obvious effect on the morphology of TiO2 nanowires. The TiO2 nanowires after conventional oven drying have a huge tendency toward agglomeration. But hierarchical porous TiO2 nanowire microspheres within a size of 2–10 lm are achieved by spray drying. Moreover, the spray drying method has the advantage of high efficiency which is beneficial for practical use. We also investigated the effect of annealing temperature on the spray-dried nanowires. At higher temperature, TiO2-B phase began to transform into more thermal stable anatase TiO2 phase. The nanowire microsphere morphology constructed by spray drying was retained, but the hierarchical porous structure was toppled down. The as-obtained TiO2 calcined at 350 C demonstrated better first coulombic efficiency as anode materials for lithium-ion batteries than that of the TiO2 material calcined at 450 C, implying a promising anode candidate for lithium-ion batteries. REFERENCES Armstrong AR, Armstrong G, Canales J, Bruce PG. 2004. TiO2-B nanowires. Angew Chem Int Ed 43:2286–2288. Armstrong AR, Armstrong G, Canales J, Bruce PG. 2005a. TiO2–B nanowires as negative electrodes for rechargeable lithium batteries. J Power Sources 501–506. Armstrong AR, Armstrong G, Canales J, Garcia R, Bruce PG. 2005b. Lithium-ion intercalation into TiO2-B nanowires. Adv Mater 17: 862–865. Armstrong G, Armstrong AR, Canales J, Bruce PG. 2005c. Nanotubes with the TiO2-B structure. Chem Commun 19:2454–2456.
Microscopy Research and Technique
175
Bavykin DV, Friedrich JM, Walsh FC. 2006. Protonated titanates and TiO2 nanostructured materials: synthesis, properties, and applications. Adv Mater 18:2807–2824. Bian H, Yang Y, Wang Y, Tian W. 2012. Preparation of nanostructured alumina–titania composite powders by spray drying, heat treatment and plasma treatment. Powder Technol 219:257–263. Brousse T, Marchand R, Taberna P, Simon P. 2006. TiO2 (B)/activated carbon non-aqueous hybrid system for energy storage. J Power Sources 158:571–577. Bruce PG, Scrosati B, Tarascon JM. 2008. Nanomaterials for rechargeable lithium batteries. Angew Chem Int Ed 47:2930–2946. Chen JS, Tan YL, Li CM, Cheah YL, Luan D, Madhavi S, Boey F, Archer LA, Lou XW. 2010a. Constructing hierarchical spheres from large ultrathin anatase TiO2 nanosheets with nearly 100% exposed (001) facets for fast reversible lithium storage. J Am Chem Soc 132: 6124–6130. Chen JS, Zhu T, Yang XH, Yang HG, Lou XW. 2010b. Top-down fabrication of a-Fe2O3 single-crystal nanodiscs and microparticles with tunable porosity for largely improved lithium storage properties. J Am Chem Soc 132:13162–13164. Deng D, Kim MG, Yang JY, Cho J. 2009. Green energy storage materials: Nanostructured TiO2 and Sn-based anodes for lithium-ion batteries. Energy Environ Sci 2:818–837. Fong R, Sacken U, Dahn JR. 1990. Studies of lithium intercalation into carbons using nonaqueous electrochemical cells. J Electrochem Soc 137:2009–2013. Gao J, Lowe MA, Hector Abruna D. 2011. Spongelike nanosized Mn3O4 as a high-capacity anode material for rechargeable lithium batteries. Chem Mater 23:3223–3227. Hassan MF, Guo ZP, Chen Z, Liu HK. 2010. Carbon-coated MoO3 nanobelts as anode materials for lithium-ion batteries. J Power Sources 195:2372–2376. Huang S, Kavan L, Exnar I, Greatzel M. 1995. Rocking chair lithium battery based on nanocrystalline TiO2 (Anatase). J Electrochem Soc 142:L142–L144. Li WJ, Fu ZW. 2010. Nanostructured WO3 thin film as a new anode material for lithium-ion batteries. Appl Surf Sci 256:2447–2452. Li Q, Zhang J, Liu B. 2008. Synthesis of high-density nanocavities inside TiO2–B nanoribbons and their enhanced electrochemical lithium storage properties. Inorg Chem 47:9870–9873. Liu S, Jia H, Han L, Wang J, Gao P, Xu D, Yang J, Che S. 2012. Nanosheet-constructed porous TiO2–B for advanced lithium ion batteries. Adv Mater 24:3201–3204. Murphy DW, Cava RJ, Zahurak SM, Santoro A. 1983. Ternary LixTiO2 phases from insertion reactions. Solid State Ionics 9:413–417. Nakahara K, Nakajima R, Matsushima T, Majima H. 2003. Preparation of particulate Li4Ti5O12 having excellent characteristics as an electrode active material for power storage cells. J Power Sources 117:131–136. Ohzuku T, Kodoma T, Hirai T. 1985. Electrochemistry of anatase titanium dioxide in lithium nonaqueous cells. J Power Sources 14:153– 166. Roth EP, Doughty DH. 2004. Thermal abuse performance of highpower 18650 Li-ion cells. J Power Sources 128:308–318. Shieh DL, Ho CH, Lin JL. 2008. Study of preparation of mesoporous TiO2-B nanofibers from mesoporous anatase TiO2 and interaction between CH3I and TiO2-B. Microporous Mesoporous Mater 109: 362–369. Sudant G, Baudrin E, Larcher D, Tarascon JM. 2005. Electrochemical lithium reactivity with nanotextured anatase-type TiO2. J Mater Chem 15:1263–1269. Vicent M, Sanchez E, Santacruz I, Moreno R. 2011. Dispersion of TiO2 nanopowders to obtain homogeneous nanostructured granules by spray-drying. J Eur Ceram Soc 31:1413–1419. Wagemaker M, Borghols WJH, Mulder FM. 2007. Large impact of particle size on insertion reactions. A case for anatase LixTiO2. J Am Chem Soc 129:4323–4327. Wang B, Zhang W, Zhang W, Mujumdar AS, Huang L. 2005. Progress in drying technology for nanomaterials. Drying Technol 23:7-32. Wang Q, Wen ZH, Li JH. 2006. Solvent-Controlled Synthesis and electrochemical lithium storage of one-dimensional TiO2 nanostructures. Inorg Chem 45:6944–6949. Wang J, Bai Y, Wu M, Yin J, Zhang WF. 2009. Preparation and electrochemical properties of TiO2 hollow spheres as an anode material for lithium-ion batteries. J Power Sources 191: 614–618. Ye X, Lin Y, Wang C, Engelhard MH, Wang Y, Wai CM. 2004. Supercritical fluid synthesis and characterization of catalytic metal nanoparticles on carbon nanotubes. J Mater Chem 14:908–913. Zachau-Christiansen B, West K, Jacobsen T, Atlung S. 1988. Lithium insertion in different TiO2 modifications. Solid State Ionics 28: 1176–1182.
Porous TiO2 Nanowire Microsphere Constructed by Spray Drying and its Electrochemical Lithium Storage Properties CHUNJU LV,* TIAN HU, KANGYING SHU, DA CHEN, AND GUANGLEI TIAN College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, People’s Republic of China
KEY WORDS
microsphere; spray drying; agglomeration; lithium-ion battery
ABSTRACT Porous TiO2 nanowire microspheres with greatly decreasing agglomeration were successfully prepared by spray drying of hydrothermal reaction suspension, followed by calcination at 350 C. The as-obtained nanowire microspheres with TiO2-B structure reach an initial discharge capacity 210 mAh g21 with an irreversible capacity 25 mAh g21 at a current density of 20 mA g21. For the 450 C-calcined one with anatase TiO2 crystal structure, the initial discharge capacity is 245 mAh g21 but with a much higher irreversible capacity of 80 mAh g21. The hierarchical porous structure in the 350 C-calcined TiO2 nanowire microspheres collapsed at 450 C, annihilating the main benefit of nanostructuring. Microsc. Res. Tech. 77:170–175, 2014. V 2013 Wiley Periodicals, Inc. C
INTRODUCTION The carbon negative electrode widely used in the present commercial rechargeable Li-ion batteries suffers from a number of problems and cannot meet the performance requirements of some important applications satisfactorily, especially in the safety and rate performance (Fong et al., 12; Roth and Doughty, 22). As a result, there is now great interest in alternative anodes including transition metal oxides such as Fe2O3, Mn3O4, MoO3, WO3, and TiO2 (Chen et al., 10; Gao et al., 13; Hassan et al., 14; Li and Fu, 16; Murphy et al., 19 ; Zachau-Christiansen et al., 31). Especially, titanium oxide is regarded as a promising active lithium intercalation material which has been proven to have superior properties in terms of higher theoretical capacity of 335 mAh g21, a lower self-discharge rate, chemical stability, environmental benignancy, and low cost (Bruce et al., 8; Huang et al., 15; Ohzuku et al., 21; Zachau-Christiansen et al., 31). Satisfactory results have been obtained by using nanostructured TiO2, which can be beneficial to both power rate and cycle life (Armstrong et al., 1,2,b,c; Bavykin et al., 5; Bruce et al., 8; Deng et al., 11; Sudant et al., 24; Wagemaker et al., 26). Indeed, the basic requirement for many high-rate insertion battery materials is that they are prepared in a nanostructured form. In this way, the surface area can be maximized and the solid-state transport length minimized. As for TiO2, the dimensional confinement imposed by the TiO2 nanowire morphology makes it an ideal host for Li1 intercalation/ deintercalation and the controlled introduction of electrons into TiO2 nanowire. So far, different ways especially the hydrothermal reaction method have been applied successfully to synthesize TiO2 nanowires because the hydrothermal method is convenient, low cost, and easily adaptable to mass production. However, when we try to obtain the active TiO2 nanowire material in a solid form by means of the conventional centrifugal separation and subsequent oven drying of the product from the hydroC V
2013 WILEY PERIODICALS, INC.
thermal reaction, there will be a huge tendency toward agglomeration. Preventing agglomeration of nanoparticles during the drying progress of liquidphase preparation is one of the most important problems. Different drying methods to treat the precursors from liquid-phase preparation such as spray drying, microwave drying, and supercritical fluid drying were applied to prevent the agglomeration of nanoparticles (Bian et al., 6; Wang et al., 27; Ye et al., 30). Spraydrying technique exhibits advantages like low price, high efficiency. In the spray-dryers, a suspension is fed into the drying chamber, then the slip is atomized by pumping it at high pressure through a multi-nozzle array, after that the upward spiraling droplets encounter hot air and fed through a diffuser into the chamber (Vicent et al., 25). This makes it necessary to prepare nanoparticles granules with a good flowability and a size without inducing additional hard agglomerates. Nakahara et al. (20) reported excellent cycling performance of particulate Li4Ti5O12 prepared by spray drying process with LiOH and anatase type TiO2 as precursors, and it was found that granular morphology of the active Li4Ti5O12 was essential to the high rate capacity of the active materials. In the present investigation, we report the synthesis of porous TiO2 nanowire microspheres by spray drying of TiO2 nanowires suspension, which was obtained from a hydrothermal chemical reaction, followed by calcination at 350 C. The characteristics of the as-prepared powders such as morphology, particle size, and microstructure were investigated. The relationship *Correspondence to: Chunju Lv, College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, People’s Republic of China. E-mail: [email protected] Received 22 September 2013; accepted in revised form 19 November 2013 REVIEW EDITOR: Dr. Chuanbin Mao Contract grant sponsor: Zhejiang Analysis Test Project of China; Contract grant number: 2009F70010; Contract grant sponsor: National Natural Science Foundation of China; Contract grant numbers: 21003111, 21210102012. DOI 10.1002/jemt.22324 Published online 3 December 2013 in Wiley Online Library (wileyonlinelibrary.com).
TIO2 NANOWIRES BY SPRAY DRYING
171
Fig. 1. (a) SEM image and (b) EDS analysis of the oven-dried TiO2 nanowire precursor; (c) TGA curves and (d) XRD patterns of TiO2 nanowire precursor by different drying processes The bottom right inset in (a) is an enlarged SEM image.
between microstructure, electrochemical properties as a lithium intercalation host, and processing technologies of the active material was discussed carefully. RESULTS AND DISCUSSION The morphology, composition, and structures of the as-obtained precursors by hydrothermal reaction at 180 C, acid treatment, and subsequent different drying processes are displayed in Figure 1. As clearly seen from the scanning electron microscope (SEM) image (Fig. 1a) of the oven-dried precursor, the material shows large-scale formation of cotton-like texture formed by thousands of entangled nanowires as suggested in the enlarged high-resolution inset image. Lin and coworkers (2008) also observed such texture which was composed of entangled TiO2-B nanofibers by the reaction of mesoporous anatase TiO2 at 90 C, followed by addition of 0.1 M HCl solution into the reaction precipitate to reach pH 1 and subsequent deionized water washing. In our research, the aggregation of curved nanowires is severe when the oven-drying process was adopted to deal with the acid and deionized waterwashed precipitate. Microscopy Research and Technique
The purity and chemical composition of the ovendried precursor are confirmed by energy dispersive spectroscopy (EDS) analysis (Fig. 1b), which mainly shows strong Ti and O signals. Only a very small amount of sodium residual from the raw NaOH stuff is detected. The atom ratio of Ti/O is very close to 3/7. At the same time, the thermal gravimetric analysis (TGA) curve (Fig. 1c) reveals that the oven-dried precursor contains 16 wt % weight loss in the whole range from room temperature to 800 C which could be partly attributed to the dehydration of the precursor. The reason that no H element signal is traced in the EDS spectrum comes from its technical limitation in the field of analyzing light elements. To further confirm the crystal structure of the oven-dried precursor, X-ray diffraction (XRD) analysis was carried out. As seen from the related XRD pattern (Fig. 1d), the profile for the obtained-product by hydrothermal reaction and acid treatment is found to be similar to that of the material prepared previously by Liu and coworkers (2008), which has been proposed to be due to H2Ti3O7 phase (JCPDS 41–0192). In a typical hydrothermal synthesis process, the layered hydrogen titanate H2TixO2x11
172
C. LV ET AL.
Fig. 2. Variable magnification SEM image of 350 C-calcined TiO2 nanowires from oven-dried (a,c) and spray-dried precursors (b,d).
phase was obtained by ion exchange through acid washing sodium titanate that was produced by an alkaline hydrothermal treatment. In our research, the Ti/O atom ratio of the as-obtained hydrogen titanate by the EDS analysis corresponds to that of H2Ti3O7. But the 16 wt % weight loss of the as-obtained hydrogen titanate after heating up to 800 C by TGA analysis is much higher than that of the complete dehydration of H2Ti3O7 to TiO2 which is calculated theoretically to be about 7 wt %. Moreover, nearly no weight loss is found above 450 C. So the weight loss below 450 C should be also related to the evaporation of the absorbed water and interlayer water except for the dehydration of structural water in the as-prepared hydrogen titanate with nanowire structure. In the present work, the spay-dried precursor was also characterized by TGA and XRD analysis. Because a certain amount of absorbed water and/or interlayer water had evaporated when the precipitate was spray dried at the inlet and outlet temperature of 150 and 80 C, a weight loss of 14 wt % for the spay-dried precursor which is a little lower than that of the 60 C oven-dried product is observed, as shown in Figure 1c. And the XRD profile for the spray-dried sample has no big difference from that of the oven-dried sample, implying that the drying process we applied has no influence on the crystal structure of the hydrogen titanate. Based on the TGA profiles, the oven-dried and spray-dried hydrogen titanate precursors were both
calcined at 350 C at which not only the evaporation of the absorbed and/or interlayer water but also the dehydration process of the structural water have almost finished. Figure 2 shows the morphology images of the two different calcination products, obviously suggesting the effect of drying process on the clustering of nanowires. When using the conventional oven drying, the overall SEM image from the ovendried precursor as shown in Figure 2a clearly reveals that it retains the same cotton-like morphology as the original oven-dried precursor where individual nanowires are severely clustered together. This clustering of nanowires will prohibit direct access of Li-ion and electrons to the whole surface of the primary nanowires and thereby annihilate the main benefit of nanostructuring. But the as-obtained product by spray drying and subsequent calcination at 350 C demonstrates a porous TiO2 nanowire microsphere morphology with a size ranging from 2 lm to 10 lm (Figs. 2b and 2d). As clearly seen from the higher magnification SEM image (Figs. 2c and 2d), the large amounts of TiO2 nanowires entanglement decreases greatly for the spray-dried product, compared with the conventional oven-dried one. Thus, the surface area of the TiO2 nanowires can be further maximized which will make it an ideal host for Li1 intercalation/deintercalation and the controlled introduction of electrons into TiO2 nanowire. Hereafter, the spray drying method is adopted to prepare much more TiO2 nanowire precursor for further investigation. Microscopy Research and Technique
TIO2 NANOWIRES BY SPRAY DRYING
To investigate the effect of annealing temperature, the spray-dried nanowire precursor with H2Ti3O7 phase was calcined at 350 C and 450 C for 3 h in air respectively. In our work, the spray-dried hydrogen titanate transforms into TiO2-B (JCPDS 46–1237) after heating at 350 C as shown in Figure 3, which is
Fig. 3. XRD patterns of TiO2 nanowires from the spray-dried precursor and subsequent calcinations at 350 C and 450 C, respectively.
consistent with those of TiO2-B reported previously (Nakahara et al., 20; Shieh et al., 23). Further increasing the temperature to 450 C, the diffraction peaks ascribed to anatase TiO2 phase (JCPDS 21-1272) are observed. And the peak at a 2h value of about 10 due to TiO2-B crystal structure is still present in the spectra, indicating that more thermal stable anatase TiO2 containing minor residual TiO2-B phase can be obtained via the post-heat treatment of the protonated titanate nanowires at 450 C for 3 h. Figure 4 shows SEM images of the TiO2 sample from the spray-dried precursor and subsequent calcination at 450 C. As can be seen from the low magnification SEM images in Figures 4a and 4b, the microsphere morphology within a small size range constructed by spray drying treatment is maintained. The higher magnification SEM image as shown in Figure 4c indicates that the as-obtained anatase TiO2 from the spray-dried hydrogen titanate and subsequent calcination at 450 C still keeps 1D shape. However, the hierarchical porous structure observed in the nanowire microspheres from the 350 C-calcined TiO2 product collapses after calcination at 450 C. Figure 5 displays the galvanostatic discharging– charging curves of two electrodes made from different TiO2 nanowires at a current rate of 20 mA g21 over the potential range of 1.0–2.5 V (versus Li1/Li). The discharging–charging profile for the 350 C-calcined
Fig. 4. Variable magnification on SEM image of TiO2 nanowires from the spray-dried precursor and subsequent calcination at 450 C.
Microscopy Research and Technique
173
174
C. LV ET AL.
Fig. 5. Initial charge–discharge profiles of TiO2 nanowires from the spray-dried precursor and subsequent calcination at 350 C and 450 C at a current rate of 20 mA g21 between 1.0 and 2.5 V.
Fig. 6. Cycling performance of TiO2 nanowires from the spraydried precursor and subsequent calcination at 350 C at a current rate of 20 mA g21 between 1.0 and 2.5 V.
TiO2 nanowires with TiO2-B structure is characterized by a monotonous potential decrease corresponding to a solid solution insertion mechanism. Such mechanism is related to TiO2-B structure but has nothing to do with its morphology such as powder, nanoribbon, nanofiber, nanosheet which was evidenced in other studies (Brousse et al., 7; Liu et al., 18; Wang et al., 28). The initial discharge specific capacity for the asprepared TiO2 nanowires calcined at 350 C is 210 mAh g21 with a small irreversible capacity (charge capacity less than discharge for the anode materials) of about 25 mAh g21. The potential profiles for the 450 C-calcined TiO2 nanowires containing the main anatase phase and residual TiO2-B can be divided into two characteristic domains, which are induced by the two coexisting phases. One domain consists of the plateaus from two phase intercalation, which is a typical shape for anatase TiO2 (Chen et al., 9; Wang et al., 29). The discharge and charge potential plateaus are presented at about 1.75 and 1.95 V (vs. Li1/Li) for lithium insertion and extraction, respectively, which is related to the fraction of the anatase phase. Another domain beyond the plateaus characterized by a slopping potential decrease is a further Li-insertion into layer nanowires or the residual TiO2-B. As shown in Figure 5, at the first cycle at 20 mA g21, the as-prepared TiO2 nanowires calcined at 450 C can reach a high discharge specific capacity of 245 mAh.21. However, a much higher irreversible capacity about 80 mAh g21 than that of the 350 C-calcined TiO2 electrode material emerges, which resulted from the collapse of the hierarchical porous structure after heating at higher temperature. The loosely-attached individual nanowires in the 350 C-calcined TiO2 with porous microsphere morphology facilitate their contact with electrolyte and hence maximize the electrode/electrolyte contact area, favoring the Li ion transport. Figure 6 shows the cycling performance of TiO2 naowires from the spry-dried precursor and subsequent calcination at 350 C at a current rate of 20 mA g21 between 1.0 and 2.5 V. The charge specific capacities tend to decrease with increasing cycling: approximately 42.9% (90 mAh g21) of its initial discharge
capacity is retained after 20 cycles. The average capacity fading was close to 2.8% at each cycle, showing unsatisfying cycling stability. In this aspect, further work is being carried out to improve the lithium storage performance of the porous TiO2 nanowire microspheres. EXPERIMENTAL The preparation of titanate nanowires was referred in previous literatures (Armstrong et al., 1,2,b). TiO2 powders (Degussa P25, ca. 80% anatase, 20% rutile) were used as received. All other chemicals were of analytical reagent grade and used without further purification. Deionized water was used throughout. In a typical synthesis process, 1.0 g commercial P25 powders were added into aqueous NaOH solution (60 mL, 10 mol L21). After stirring for 1 h, the resulting suspension was transferred into a Teflon-lined stainless steel autoclave (100 mL). The autoclave was maintained at 180 C for 12 h and then naturally cooled to room temperature. Subsequently, the obtained precipitates were washed with HCl (0.1 mol L21) solution and then with deionized water until the pH value approaching about 7.0. Subsequently, the precipitates were dispersed into 120 mL deionized water to form homogeneous slurry with the assistance of ultrasound. Afterwards, the homogeneous slurry was spray-dried at the inlet and outlet temperature of 150 and 80 C, respectively. Finally, the dried powders were heated at different temperature for 3 h in air in a covered alumina crucible with a heating rate of 2 C min21. For comparison, the acid and deionized water-washed precipitate was dried in oven at 60 C for 12 h and then developed at the same calcination conditions. XRD measurements were performed on a diffractometer (Bruker D2 PHASER) with Cu Ka radiation in the range from 10 to 90 . The particle size and morphology of the samples were studied with a SEM (JEM-5610) coupled with an EDS. The TGA was performed using a METTLER TOLEDO TGA/DSC 1LF1600 apparatus. The experiments were performed by heating from 30 to 800 C at 5 C min21 in air environment. Microscopy Research and Technique
TIO2 NANOWIRES BY SPRAY DRYING
Coin cells (CR 2015) were constructed in a dry room to investigate Li-ion intercalation behavior. The working electrodes were prepared from the as-synthesized active materials, carbon black as a conducting agent, and poly(vinylidene difluoride) (PVDF) binder dissolved in n-methyl pyrolidinone(NMP, 8 wt %) at a weight ratio of 80:10:10. The slurry was mixed overnight to obtain a homogeneous black paste which is then spread onto a copper foil current collector. The solvent is removed at 120 C in vacuum for 12 h. Copper foil pieces with 14 mm diameter were then cut off and used as the electrode in homemade cells which were assembled inside Ar filled glove box. The average amount of TiO2 nanowires in a round piece of electrode is about 5 mg. Each electrode was carefully weighted before use and several electrodes were tested to assume the reproducibility of the electrochemical behavior. The coin cells also used Li metal as a counter electrode, a Celgard 2500 as a separator, and a 1 M LiPF6 dissolved in ethylene carbonate, dimethyl carbonate, and ethylene methyl carbonate (1:1:1, vol/vol/ vol) as an electrolyte. The assembled cells were aged for 12 h before testing and then were galvanostatically charged and discharged over a voltage range of 2.5–1.0 V vs. Li1/Li at a current density of 20 mA g21 using a LAND cell test (Land-CT 2001A) system. The specific capacity was calculated based on the mass of active materials in the electrode. CONCLUSION In summary, TiO2 nanowires were prepared on a large scale by a hydrothermal reaction followed by acid washing, drying processing, and post-heat treatment at different temperature in air. The drying process has an obvious effect on the morphology of TiO2 nanowires. The TiO2 nanowires after conventional oven drying have a huge tendency toward agglomeration. But hierarchical porous TiO2 nanowire microspheres within a size of 2–10 lm are achieved by spray drying. Moreover, the spray drying method has the advantage of high efficiency which is beneficial for practical use. We also investigated the effect of annealing temperature on the spray-dried nanowires. At higher temperature, TiO2-B phase began to transform into more thermal stable anatase TiO2 phase. The nanowire microsphere morphology constructed by spray drying was retained, but the hierarchical porous structure was toppled down. The as-obtained TiO2 calcined at 350 C demonstrated better first coulombic efficiency as anode materials for lithium-ion batteries than that of the TiO2 material calcined at 450 C, implying a promising anode candidate for lithium-ion batteries. REFERENCES Armstrong AR, Armstrong G, Canales J, Bruce PG. 2004. TiO2-B nanowires. Angew Chem Int Ed 43:2286–2288. Armstrong AR, Armstrong G, Canales J, Bruce PG. 2005a. TiO2–B nanowires as negative electrodes for rechargeable lithium batteries. J Power Sources 501–506. Armstrong AR, Armstrong G, Canales J, Garcia R, Bruce PG. 2005b. Lithium-ion intercalation into TiO2-B nanowires. Adv Mater 17: 862–865. Armstrong G, Armstrong AR, Canales J, Bruce PG. 2005c. Nanotubes with the TiO2-B structure. Chem Commun 19:2454–2456.
Microscopy Research and Technique
175
Bavykin DV, Friedrich JM, Walsh FC. 2006. Protonated titanates and TiO2 nanostructured materials: synthesis, properties, and applications. Adv Mater 18:2807–2824. Bian H, Yang Y, Wang Y, Tian W. 2012. Preparation of nanostructured alumina–titania composite powders by spray drying, heat treatment and plasma treatment. Powder Technol 219:257–263. Brousse T, Marchand R, Taberna P, Simon P. 2006. TiO2 (B)/activated carbon non-aqueous hybrid system for energy storage. J Power Sources 158:571–577. Bruce PG, Scrosati B, Tarascon JM. 2008. Nanomaterials for rechargeable lithium batteries. Angew Chem Int Ed 47:2930–2946. Chen JS, Tan YL, Li CM, Cheah YL, Luan D, Madhavi S, Boey F, Archer LA, Lou XW. 2010a. Constructing hierarchical spheres from large ultrathin anatase TiO2 nanosheets with nearly 100% exposed (001) facets for fast reversible lithium storage. J Am Chem Soc 132: 6124–6130. Chen JS, Zhu T, Yang XH, Yang HG, Lou XW. 2010b. Top-down fabrication of a-Fe2O3 single-crystal nanodiscs and microparticles with tunable porosity for largely improved lithium storage properties. J Am Chem Soc 132:13162–13164. Deng D, Kim MG, Yang JY, Cho J. 2009. Green energy storage materials: Nanostructured TiO2 and Sn-based anodes for lithium-ion batteries. Energy Environ Sci 2:818–837. Fong R, Sacken U, Dahn JR. 1990. Studies of lithium intercalation into carbons using nonaqueous electrochemical cells. J Electrochem Soc 137:2009–2013. Gao J, Lowe MA, Hector Abruna D. 2011. Spongelike nanosized Mn3O4 as a high-capacity anode material for rechargeable lithium batteries. Chem Mater 23:3223–3227. Hassan MF, Guo ZP, Chen Z, Liu HK. 2010. Carbon-coated MoO3 nanobelts as anode materials for lithium-ion batteries. J Power Sources 195:2372–2376. Huang S, Kavan L, Exnar I, Greatzel M. 1995. Rocking chair lithium battery based on nanocrystalline TiO2 (Anatase). J Electrochem Soc 142:L142–L144. Li WJ, Fu ZW. 2010. Nanostructured WO3 thin film as a new anode material for lithium-ion batteries. Appl Surf Sci 256:2447–2452. Li Q, Zhang J, Liu B. 2008. Synthesis of high-density nanocavities inside TiO2–B nanoribbons and their enhanced electrochemical lithium storage properties. Inorg Chem 47:9870–9873. Liu S, Jia H, Han L, Wang J, Gao P, Xu D, Yang J, Che S. 2012. Nanosheet-constructed porous TiO2–B for advanced lithium ion batteries. Adv Mater 24:3201–3204. Murphy DW, Cava RJ, Zahurak SM, Santoro A. 1983. Ternary LixTiO2 phases from insertion reactions. Solid State Ionics 9:413–417. Nakahara K, Nakajima R, Matsushima T, Majima H. 2003. Preparation of particulate Li4Ti5O12 having excellent characteristics as an electrode active material for power storage cells. J Power Sources 117:131–136. Ohzuku T, Kodoma T, Hirai T. 1985. Electrochemistry of anatase titanium dioxide in lithium nonaqueous cells. J Power Sources 14:153– 166. Roth EP, Doughty DH. 2004. Thermal abuse performance of highpower 18650 Li-ion cells. J Power Sources 128:308–318. Shieh DL, Ho CH, Lin JL. 2008. Study of preparation of mesoporous TiO2-B nanofibers from mesoporous anatase TiO2 and interaction between CH3I and TiO2-B. Microporous Mesoporous Mater 109: 362–369. Sudant G, Baudrin E, Larcher D, Tarascon JM. 2005. Electrochemical lithium reactivity with nanotextured anatase-type TiO2. J Mater Chem 15:1263–1269. Vicent M, Sanchez E, Santacruz I, Moreno R. 2011. Dispersion of TiO2 nanopowders to obtain homogeneous nanostructured granules by spray-drying. J Eur Ceram Soc 31:1413–1419. Wagemaker M, Borghols WJH, Mulder FM. 2007. Large impact of particle size on insertion reactions. A case for anatase LixTiO2. J Am Chem Soc 129:4323–4327. Wang B, Zhang W, Zhang W, Mujumdar AS, Huang L. 2005. Progress in drying technology for nanomaterials. Drying Technol 23:7-32. Wang Q, Wen ZH, Li JH. 2006. Solvent-Controlled Synthesis and electrochemical lithium storage of one-dimensional TiO2 nanostructures. Inorg Chem 45:6944–6949. Wang J, Bai Y, Wu M, Yin J, Zhang WF. 2009. Preparation and electrochemical properties of TiO2 hollow spheres as an anode material for lithium-ion batteries. J Power Sources 191: 614–618. Ye X, Lin Y, Wang C, Engelhard MH, Wang Y, Wai CM. 2004. Supercritical fluid synthesis and characterization of catalytic metal nanoparticles on carbon nanotubes. J Mater Chem 14:908–913. Zachau-Christiansen B, West K, Jacobsen T, Atlung S. 1988. Lithium insertion in different TiO2 modifications. Solid State Ionics 28: 1176–1182.
