Technologies, Am Stollen 19 B, D-38640 Goslar (Germany). 3) Ilmenau University of Technology, Institute of Materials Engineering, Department of materials for ...
Si-modified TiO2 nanotubes obtained by electrochemical anodization - promising anode material for energy storage application 1
2
2
3
Ivanov, S. ; Wulfmeier, H. ; Albrecht, D.; Grieseler, R. ; 3
2
Schaaf, P. ; Fritze, H. ; Bund, A.
1
Abstract Current studies in the area of battery materials suggest TiO2 as a promising candidate for replacement of the widely used graphite anodes. The increased attention in this field is associated to the good electrochemical stability and high operating voltage of TiO2. Si is another anode material of special interest due to a -1 very high theoretical capacity of ca. 4200 mAhg . Its main drawback is the volume expansion resulting from the mechanical stress accompanying the alloying with Li which induces structural instability. In order to minimize this problem recent studies are focusing on the design and development of nanostructured hybrid materials involving Si. The present work is concentrated on electrochemical properties of TiO2 nanotubes obtained by electrochemical anodization. TiO2 nanoporous structures were used as a template for Si deposition. The goal of the research is to combine the good electrochemical stability of TiO2 template with the high capacity of the Si.
Introduction Titania (TiO2) has been intensely investigated as a potential anode material for lithium-ion batteries. It has been concluded that because of its good electrochemical stability and high operating voltage TiO2 has a promising future in Li-ion battery technology. By introducing TiO2 the risk of battery overcharging is reduced, avoiding the growth of metallic Li dendrites during the Li ion insertion into the anode and corresponding thermal runaway of the battery [1]. A cell assembled according to this concept, combining high voltage anode and cathode, belongs to the category of third generation Li-ion batteries. An additional advantage of TiO2 is that it is nontoxic, low-cost and is easily available in various modifications. It is known that the
basic weakness of TiO2 is its comparatively low capacity for lithium ion insertion – an essential parameter needed for a good battery performance. The properties of TiO2 electrodes may improve dramatically when the material is porous because of a large surface area and open structure. A number of different approaches has been applied in order to obtain nano-porous TiO2. One recently applied method is Ti anodization, performed in fluoride containing media [2]. By this technique the obtained TiO2 layer consists of self-organized nano-tubes/nanopores with adjustable length, diameter and large active surface. Moreover the Ti/TiO2 system combines a current collector (Ti) and the active material (TiO2) with no binders and additives necessary. One additional positive side effect is that TiO2 nano-pores obtained by anodization are amorphous. It is known that in many experimental cases amorphous TiO2 has a higher capacity than the crystalline one. If needed, it can be transformed to a crystalline (rutile or anatase) form by annealing [1]. In order to obtain a material having better performance many active materials, including TiO2 are applied in a form of (nano)composites together with other materials. One of the possibilities for TiO2 modification is Si. Si is a potentially promising anode material for Li-ion batteries, having very high capacity of 4200 -1 mAhg for the formation of the Li-Si alloy [3]. Its main drawback is the volume expansion resulting from the mechanical stress accompanying Li alloying. This + structural change during the electrochemical cycling in Li is usually resulting in rapid capacity fade and mechanical decomposition of the structure [3]. A number of different approaches, including nanostructuring and composite formation, has been undertaken in order to improve the cycling stability of Si based electrodes. Most of these methods resulted in definite improvement of the cyclability of the electrodes. Beside the active materials for Li-ion batteries another object of high interest is the development of advanced and reliable electrolyte solutions. It was shown that in many cases the typically used alkyl carbonate based electrolyte can be + electrochemically oxidized below 4V vs. Li/Li [4]. The electrochemical instability can further bring operational and safety issues during battery handling. The larger potential window of imidazolium, ammonium and pyrrolidinium types of ionic liquids in comparison with ethylene carbonate / dimethyl carbonate (EC/DMC) based electrolytes was recently demonstrated. It was shown that ionic liquids can improve the electrochemical and thermal performance by their non-flammable nature and bring a good electrochemical stability at high positive potentials [5]. In the current research we suggest an approach combining the good stability and high surface area of TiO2 with the high capacity of the amorphous silicon. By this approach we expect that Si deposited in/on the stable TiO2 nanostructured matrix
could improve the electrochemical stability and increase the material capacity. Related to the high voltage application, Li ion insertion/extraction from/out of the Si + modified and non-modified samples is studied in Li containing ionic liquid - butyl+ methyl-pyrrolidinium bis(trifluoro)sulfonylimide [BMP TFSI ].
Experimental The chemicals for TiO2 nanotube layers formation were purchased from Merck. EC, DMC, LiPF6 and LiTFSI were obtained from Alfa Aesar. BMP, TFSI was purchased from IoLiTech. All the reagents, solutions and samples were dried in inert Ar atmosphere. Prior to the synthesis the Ti substrates were washed with distilled water, afterwards in acetone and treated under ultrasonication in ethanol and acetone. The electrochemical synthesis of TiO2 ordered nano-tubes was carried out in a two electrode electrochemical cell. The layers were synthesized by anodic oxidation in glycol or glycerol based electrolytes containing 0.5 % NH4F at 20 and 40 V. After the completion of the anodization the nanoporous layers were washed o o with water, dried at 60 C for 5 h, dried at 120 C at vacuum for another 5 h and transferred into a glove-box (M-BROWN), (Argon atmospere with less than 1 ppm water and oxygen). TiO2 samples were electrochemically characterized using potentiostat/galvanostat Biologic VMP3 in a three and two electrode configuration. Li metal counter and reference electrodes were used for electrochemical tests. Part of the Ti/TiO2 samples was modified by physical vapor deposition (PVD) of Si. The surface morphology of the samples was investigated using an ultrahigh resolution scanning electron microscope (FE-SEM, Hitachi S-4800).
Results and discussions TiO2 nanotube layers were synthesized by anodic oxidation in glycol and glycerol based electrolytes containing fluoride ions. Current transients of the anodization process carried out at 15 V in both electrolytes are shown in fig.1. The curves display plateau shaped maximum followed by a current decrease. This can be associated to different stages of nanotube formation [2]. Generally, the currents recorded during the anodization in glycol media are about three times higher than that of glycerol electrolyte due to the difference in the viscosities.
Fig.1 Current transients of Ti anodization performed at 10 V in glycol (___) and glycerol (---) based electrolytes. Inset: SEM image of typical TiO2 nanotube layer
Fig.2 Voltammetric curves of amorphous (--) and crystalline (___) TiO2 nanotube layers measured in 1M LiTFSI, [BMP][TFSI]. v = 0,5 mVs-1 Inset: galvanostatic cycling of TiO2 layers obtained after different anodization times.
After the anodization the surface morphology of the samples was characterized by scanning electron microscopy (SEM). A typical surface structure of TiO2 nanotubular layer is shown in fig.1 (see inset). The tube dimensions are strongly dependent on the applied voltage and electrolyte viscosity [2]. Depending on the synthesis parameters we obtained nanotubes diameter in 40 – 150 nm range. Due to its high significance to energy storage application, the mechanism of lithium insertion into TiO2 has been extensively studied. The electrochemical exchange of Li is represented by the following equation TiO2 +xLi+ +xe−↔ LixTiO2
(1)
The crystalline structure of anatase is tetragonal and contains distorted TiO6 octahedra. The latter allows a sequence of octahedral and tetrahedral vacancies. This allows an insertion of 0.5 Li per formula unit, which corresponds to a −1 theoretical capacity of 168 mAhg [1]. A two-phase mechanism has been suggested to describe the electrochemical insertion of lithium into anatase, involving equilibrium between Li-poor (tetragonal) and Li-rich (orthorhombic) phases. After the synthesis and drying procedures part of TiO2 samples was converted to crystalline anatase and both types of layers were electrochemically -1 cycled in Li-ion containing [BMP][TFSI] at 0,5 mVs (Fig.2). The cyclic voltammetry shows well-shaped quasi-reversible electrochemical couples that can be associated with the Li insertion/de-insertion into the TiO2 nanostructure. In contrast to anatase, the amorphous TiO2 nanotube layer has much broader voltammetric peaks due to
the amorphous nature of its structure. As a next step the layers were -2 galvanostatically cycled at 100 µA.cm (not shown here). In order to minimize the effect of solid-electrolyte interface (SEI) formation the further performed electrochemical measurements were limited to cathodic potentials of 1 V. The result for multiple constant current cycling showed a good electrochemical stability, no plateau region (typical for amorphous TiO2) and flat-type galvanostatic curve for anatase sample. It was observed that the first discharge takes much longer time than the subsequent ones. The latter finding could be explained by the above st mentioned SEI layer formation, which takes place irreversibly during the 1 discharge process with an onset at about 1 V. Furthermore, by prolonging the time of the anodization process the areal capacity of the material can be increased (Fig. 2 inset). The latter observation suggests that the growing structure of TiO2 is active for Li-ion intercalation and the capacity can be controlled by the anodization period. TiO2 nanotube layers were further modified by physical vapour deposition of Si. Cyclic voltammetry and initial data for galvanostatic cycling are shown in the next figures (fig.3, fig.4). Besides typical voltammetric couple for Li-ion intercalation into TiO2 structure at about 1.7 V, a new characteristic for Li-Si alloying process current peaks (at about 0.5 V) appeared in the voltammogram. Galvanostatic cycling of the -1 Si modified samples showed stable capacity of approximately 1.5 -2.4 Ahg for 40 cycles (fig.4). Our preliminary results suggested that the amount and morphology of both components of the hybrid material need to be further optimized.
Fig.3 CV curves of crystalline (___) and amorphous (---) TiO2 nanotubes modified by PVD of Si and equal amount of sputtered Si on bare Ti substrate (….) in 1M LiTFSI, [BMP,TFSI]. v = 0,5 mV s-1
Conclusions
Fig.4 Galvanostatic cycling of Si modified TiO2 layers in 1M LiTFSI, [BMP,TFSI]. PVD is performed at 50 W for 450 s (▪) and at 200 W for 100 s (□) at room temperature.
Ti/TiO2 nanotube structures were synthesised by electrochemical anodization of Ti foil. The nanostructured TiO2 layers were tested as a current collector – anode + material system for Li-ion intercallation. Li insertion-deinsertion in Ti/TiO2 nanotube structures was measured in LiTFSI, [BMP][TFSI] as a non-flammable, non-volatile ionic liquid electrolyte. TiO2 nanotube structure displayed a good cyclability in Li containing [BMP] [TFSI]. Ti/TiO2 - Si nanoporous composite was suggested as a potential anode material with high performance for Li-ion batteries application. The Ti/TiO2 samples were modified by Si physical vapor deposition. The electrochemical behavior of Ti/TiO2 - Si was studied by cyclic voltammetry and chronopotentiometry + in 1M Li TFSI , [BMP] [TFSI]. The galvanostatic cycling of Si modified and nonmodified samples showed good electrochemical stability. The conditions for the formation of the hybrid TiO2 - Si anode and correspondingly its electrochemical performance for energy storage application need further to be optimized.
Literature [1]
Ortiz, G.; Hanzu, I.; Knauth, P.; Lavela, P.; Tirado, J.; Djenizian, T.; Electrochim. Acta, 2009, 54, p. 4262
[2]
Macak, J.M.; Schmuki, P.; Electrochim. Acta, 2006, 52, p. 1258
[3]
Kasavajjula, U.; Wang, C.; Appleby, A.; J. Power Sources, 2007, 163, p. 1003.
[4]
Baranchugov, V.; Markevich, E.; Pollak, E.; Salitra, G.; Aurbach, D., Electrochem. Commun.; 2007, 9, p. 796.
[5]
Markevich, E.; Baranchugov, V.; Aurbach, D., Electrochem. Commun., 2006, 8, p. 1331.
Acknowledgements Financial support by the DFG funded program WeNDeLIB (project 6: “Linking of model and commercial active materials for lithium ion batteries by in-situ determination of thermodynamic and kinetic data”) is gratefully acknowledged. 1) Ilmenau University of Technology, Department of electrochemistry and Electroplating, Gustav-Kirchhoff-Straße 6, D-98693 Ilmenau (Germany) 2) Clausthal University of Technology, Institute of Energy Research and Physical Technologies, Am Stollen 19 B, D-38640 Goslar (Germany) 3) Ilmenau University of Technology, Institute of Materials Engineering, Department of materials for electrical engineering, Gustav-Kirchhoff-Straße 5, D-98693 Ilmenau (Germany)
