- Email: [email protected]
Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells Jung Sub Kim a,b, Hun-Gi Jung a, Wonchang Choi a, Haw Young Lee a, Dongjin Byun b, Joong Kee Lee a,* a
Advanced Energy Materials Processing Laboratory, Center for Energy Convergence Research, Korea Institute of Science and Technology, Hwarangno 14-gil 5, Seongbuk-gu, Seoul 136-791, Republic of Korea b Department of Material Science & Engineering, Korea University, Seoul 136-713, Republic of Korea
article info
abstract
Article history:
This study investigates bundle-type silicon nanorods (BSNR) that are aimed at improving
Received 28 September 2013
the discharge capacity and life cycle characteristics of secondary cells, by controlling the
Received in revised form
shape and etching depth of silicon thick-films produced by electroless etching. The pre-
20 January 2014
pared BSNR structure is composed of a columnar bundle, having a diameter of 100 nm and
Accepted 1 February 2014
lengths of 1.5 and 3.5 mm. The etching depths of the nanorods have a significant effect on
Available online xxx
the electrochemical performance characteristics, including the capacity fading and coulombic efficiency. Using a BSNR electrode therefore allows for an anode with a high
Keywords:
capacity and efficiency in lithium ion cells, and can help overcome the issues associated
Silicon
with conventional silicon thick-films. Furthermore, as a result of its unique self-relaxant
Electroless etching
structure, electrode deterioration is improved through mitigation of the volume change.
Metal-assisted chemical etching
Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Anode
reserved.
Surface modification
Introduction The power sources for portable electronics such as mobile phones, notebook computers, and personal digital assistants (PDAs) not only require miniaturization, but also need to be highly energized through having a high energy density. Interest in such energy applications is ever increasing, and secondary cells are becoming more commonly used for hybrid electric vehicles (HEVs). Consequently, silicon is being increasingly used as an anode material for lithium secondary
cells as a replacement for more conventional carbon materials [1]. At present, the graphite materials that are commonly used have a theoretical capacity of 372 mAh g1, whereas that of silicon is about 4200 mAh g1 [2]. However, when silicon is fabricated into an anode, the resulting cell has a charging capacity of only about 3260 mAh g1, a discharge capacity of 1170 mAh g1, and a coulombic efficiency of 35% [3]. Furthermore, when the cell is continuously charged and discharged over five cycles, its discharge capacity rapidly decreases to about 300 mAh g1, or about 10% of its initial value. The reason
* Corresponding author. Tel.: þ82 2 958 5252, þ82 10 8717 0978 (mobile); fax: þ82 2 958 5229. E-mail addresses: [email protected], [email protected] (J.K. Lee). 0360-3199/$ e see front matter Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2014.02.007
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 1 e Schematic illustration of BSNR electrode (a) Processes for preparation of the BSNR electrode, and (b) features of BSNR prepared by metal-assisted chemical etching.
for this rapid decrease is that the insertion of lithium forms a LieSi alloy (Li22Si5), which in turn causes a four-fold volume expansion. This results in a breakdown of the silicon structure and blockage of the electron pathway, ultimately causing the formation of a dead volume and a reduction in the silicon anode’s capacity. Consequently, as the cell is continuously charged and discharged, its total capacity is rapidly reduced. This phenomenon occurs regardless of whether it is a bulk silicon film, or micrometer-size particles [4]. Various methods have been previously proposed to solve these problems. For example, silicon has been grown on the surface of an electric accumulator in the form of a wire, and then used as an electrode. This method proved advantageous in that electrons could move more easily than in a conventional thin film, and the inner stress of the silicon during the charging and discharging of a cell is less than that of a conventional thin film. Overall, these advantages resulted in an excellent cycle performance. However, despite these strong
points, such electrodes cannot be used in practical applications because their performance is based on a thin-film forming process [5]. A method of controlling shape has also been proposed, in which silicon nanoparticles were formed into hollow siliconnanospheres to overcome the inherent weaknesses of silicon nanoparticles in cycling. However, this method proved to be problematic in that much time is needed to form the hollow silicon nanospheres, and the process itself is quite complicated [6]. In yet another approach, a composite of silicon and silica was first formed, and then just the silica was removed by chemical etching to form bulk silicon particles with numerous pores. This was aimed at overcoming the weaknesses of conventional bulk silicon particles in cycling, and differs from conventional methods in that the pores produced help alleviate volume expansion. However, this method also presents problems in terms of the length of time needed to form the
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
pores in the bulk silicon particles, the complexity of the process, and the need for high-temperature heat treatment [7]. The present study differs greatly from these aforementioned methods in that a silicon thick-film is only partially etched for a short period of time, thus producing a bundletype silicon nanorod (BSNR) structure on the surface of a copper current collector. The purpose of this work is to investigate the effects of the BSNR etching depth on the electrochemical characteristics when it is used as an anode material for lithium secondary batteries. The high aspect ratio of BSNRs, combined with their high surface area, provides not only a favorable amount of free space, but also helps them to endure physical stress.
Experimental Deposition of silicon thick-film Copper foils were used as a current collector, after being first cleaned for 10 min in an ultrasonic bath containing a mixture of acetone and ethanol. Silicon thick-films were then deposited onto the Cu foils to a thickness of c.a. 4 mm in a plasma-enhanced chemical vapor deposition (PECVD) reactor chamber. Silane (SiH4) gas was fed into the reactor to facilitate the deposition of the silicon thick-film. Typical operating conditions were a working pressure of 70 mTorr, a plasma power of 200 W, substrate temperatures of 300 C, a SiH4 flow rate of 10 sccm, and an argon flow rate of 30 sccm. The gas flow rates were all accurately controlled by a mass flow controller (MFC), and the deposition times were kept at 2 h for all specimens.
3
and water), which was contained in a dry room with a moisture content of less of than 0.5%. The electrochemical properties of the resulting cells were evaluated by galvanostatic chargeedischarge cycling (MACCOR series-4000). All cells were cycled at a constant current of 410 mA g1 at room temperature, and chargeedischarge cycling tests were performed with cut-off voltages of 2.0 to 0.0 V.
Characterization The pristine and modified silicon thick-films were observed by scanning electron microscopy (SEM, NOVA NanoSEM200, FEI Corp.), both before and after cycling. Following the 30th cycle, the samples were rinsed in DMC solution and then dried in an Ar-filled glove box in preparation for SEM analysis. Raman spectroscopy (Nicolet Almega XR dispersive Raman, Thermo electron co.) was used to determine the silicon nanostructure, using a 632 nm (red) laser excitation. In order to negate the heating effects of the laser, it was necessary to use a low laser power density with a 50x microscope objective, and an exposure time of 4 s. The resulting spectra were recorded at a 4 cm1 resolution between 4000 and 90 cm1; the diameter of the laser spot reaching the sample being about 2 mm. The silicon nanostructure was also investigated by X-ray diffraction (XRD, Rigaku co.), using Cu Ka radiation (l ¼ 1.54059 A). Electrochemical impedance spectroscopy (EIS) experiments with the fully charged cells (up to 0.0 V) were performed using an impedance/gain-phase analyzer (Solartron SI 1260) equipped with an electrochemical interface (Solartron SI 1286). The AC amplitude was 5 mV over a frequency range of 1 mHze100 kHz.
Metal-assisted chemical etching of silicon thick-film
Results and discussion Modification of the aspect ratio of the silicon thick-films was carried out by metal-assisted chemical etching for 5 min at a temperature of either 10 or 25 C. As shown in Fig. 1(a), the entire process consisted of five steps: (1) the preparation of a cleaned current collector, (2) deposition of silicon as a starting material under constant conditions via PECVD, (3) attachment of a chemically stable polymer (PET) to prevent corrosion of the current collector (Cu) in acid solution; and metal deposition (0.01 M AgNO3, 1 M HF) of an Ag catalyst onto the silicon surface for 5 s, followed by silicon etching for 5 min in a solution of 0.036 M HF and 2.5 M H2O2 at temperatures of either 10 or 25 C, (4) removal of Ag metal by immersing in a 5% HNO3 solution for 20 min, followed by removal of the capping tape (PET), and (5) cell assembly using the Si nanostructure produced as the anode material.
Cell fabrication The BSNRs prepared by the two different etching temperatures were used as working electrodes, with lithium foil used as a counter electrode to complete the cell. A liquid electrolyte was used, consisting of 1 M LiPF6 dissolved in a mixture of ethylene carbonate (EC), ethylmethyl carbonate (EMC), and dimethyl carbonate (DMC) (1:1:1 vol%). The assembly of the lithium ion cell was carried out in an Ar-filled glove box (<1 ppm of oxygen
Controlling the porosity in silicon-based electrodes to allow for volume expansion has thus far proven to be quite a challenge. However, Unagami et al. [8] have proposed a reaction scheme involving the localized coupling of redox reactions with the metal surface to explain the metal-assisted etching process: Cathode reaction (at metal): H2O2 þ 2Hþ / 2H2O þ 2hþ
(A.1)
2Hþ þ 2e/H2[
(A.2)
Anode reaction: Si þ 4hþ þ 4HF / SiF4 þ 4Hþ
(A.3)
SiF4 þ 2HF / H2SiF6
(A.4)
Overall reaction:
Si þ H2O2 þ 6HF / 2H2O þ H2SiF6 þ H2[
(A.5)
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 2 e Top-view images of (a) pristine Si thick film, (b) Ag coated Si thick film, (c) dendrite formation of Ag and (d) formation of BSNR structure after elimination of Ag.
The deposition of metal facilitates the etching in HF and H2O2 solutions, and hole-injection is enabled by the reaction of H2O2 on the metal particles. The holes are injected into the silicon valence band, which subsequently leads to diffusion from the metal particles. The reduction of oxidizing metal ions to metallic species, and the dissolution of silicon, both occur spontaneously on the silicon surface. If the etching time is short, then the silicon thick-film is unable to form a porous silicon nanostructure. On the other hand, if the etching time is extremely long, then the etched structure is broken down by the corrosion of the Cu foil in acidic HF solution. Thus, in order to hinder such corrosion, the back side of the Cu foil was protected by a polymer film. As shown in Fig. 1(b), the use of BSNR morphologies containing numerous voids as an anode electrode provides the following functions: (1) improving the electrode reaction kinetics, due to an increase in the surface area caused by the networks of bundles; (2) acting as a self-relaxation material during the insertion and extraction of lithium due to the multitude of voids present; (3) providing good contact with the current collector by means of the direct deposition of Si onto Cu without a binder; and (4) a short lithium diffusion distance and efficient electron transfer when compared to a twodimensional electrode. BSNRs possessing these projected functions were fabricated by means of metal-assisted chemical etching, and were then evaluated as an anode material for lithium ion batteries. Representative images, such as those shown in Fig. 2, were observed in order to investigate any change in the morphologies of the silicon electrodes during each stage of the process. Fig. 2(a) shows that the silicon thick-film has a
smooth surface, which in order to prepare the silicon nanostructure, was immersed into AgNO3/HF solution for 5 s to deposit Ag particles. The resultant particle size distribution is important for understanding the physical and chemical properties related to the effect of varying the bundle diameter. The Ag particles observed on the silicon thick-film had sizes in the range of 20e250 nm, as shown in Fig. 2(b). The Ag was also found to form dendrites due to continuous reaction, as shown in Figs. 1(a) and 2(c) [9,10]. The Ag-coated Si thickfilms were next etched in HF/H2O2 solution for 5 min, and then dipped into HNO3 solution for 20 min to eliminate the Ag particles. This resulted in very different surface morphologies, as shown in Fig. 2(d). Those BSNRs with a corresponding diameter of 100 nm, and a large fraction of voids, offer a reduced volume change in response to the large variation in physical stress induced during the insertion and extraction of Liþ ions. Metal-assisted chemical etching is a simple and low-cost technique that can potentially be applied to the mass production of various silicon nanostructures with easily controlled parameters; such as the doping type, doping level, and etching temperature [11e16]. Huang et al. [17] showed that the channel depth changes as the etching time increases, and many other parameters have been examined in previous studies without considering the temperature. Pristine Si thickfilms, with a thickness of 4 mm, were therefore etched at different temperatures of 10 and 25 C. The cross-sectional images reveal different etching depths of about 1.5 and 3.5 mm, which correspond to aspect ratios of 15 (BNSR@15) and 35 (BNSR@35) in Fig. 3(b) and (c), respectively. This shows that the etching depth is a key parameter for the electrochemical
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
5
Fig. 3 e Cross-sectional images of (a) pristine Si thick film, (b) BSNR@15 and (c) BSNR@35 with different aspect ratios.
performance of the silicon anodes, and is dependent on the etching temperature. Raman spectroscopy with a 632 nm red laser was used to obtain structural information regarding the BSNR fabricated by metal-assisted chemical etching. It is known that in the case of crystalline silicon, a sharp peak appears at 520 cm1 due to the transverse optical vibrational modes, as shown in Fig. 4(a). As the BNSRs have an amorphous structure, the transverse optical peak becomes broader and shifts to lower wavenumbers of about 35 cm1. It is clear that no sharp peak can be observed at 520 cm1 for the deposited films, and instead a broad peak is observed at 485 cm1 that suggests the deposited film is predominantly amorphous [18,19]. The XRD pattern of the BSNR is shown in Fig. 4(b), in which no peaks corresponding to crystalline silicon can be observed, thus further confirming that the silicon obtained by metal-assisted chemical etching is amorphous. The intensities of all peaks
are clearly observed at scattering angles (2q) of about 43.8 , 50.9 , and 74.5 , which are ascribed to the (111), (200), and (220) planes of the Cu current collector, respectively. The XRD data therefore matches well with the Raman spectra. Fig. 5(a) shows the initial voltage profiles of a pristine silicon thick-film, BSNR@15, and BSNR@35 in the potential range of 0e2 V, at a current density of 410 mA g1. All cell tests involved chargeedischarge cycling to a discharge depth of nearly 100%, without limiting the insertion capacity. As shown in Fig. 5(a), the pristine silicon thick-film exhibits charge and discharge capacities of 3604 and 3391 mAh g1, with a coulombic efficiency of 94% in the initial cycle. The charge-discharge capacities of the BSNR@15 and BSNR@35 electrodes are 3091, 2861, 3406, and 3203 mAh g1, indicating coulombic efficiencies of 92.5 and 94%, respectively. This unprecedentedly high coulombic efficiency in the initial cycle signifies an almost complete Liþ extraction during the charge
Fig. 4 e (a) Raman data (as a single-crystal silicon wafer for reference) and (b) XRD of BSNR for observation of the silicon structure. Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 5 e (a) Initial voltage profiles, (b) discharge capacities, (c) coulombic efficiencies and accumulated irreversible capacity of pristine silicon thick film, BSNR@15 and BSNR@35 in the potential range of 0e2 V at a constant current density of 410 mA gL1.
and discharge processes. Amorphous silicon was previously investigated by Netz [20], who made a cell of amorphous silicon and evaluated its electrochemical properties. Christensen et al. [21] reported that the charging profiles have two distinct voltage regions at 100 to 0.7 V and 70 to 0.05 V due to the phase transition from crystalline to amorphous, and the amorphous change to Li15Si4 in the first cycle, respectively. These phase transitions induce a large volume-expansion of the active silicon material, thus resulting in deterioration of the electrical conduction network. Amorphous silicon without a phase transition therefore leads to a high reversible reaction compared to crystalline silicon, due to its homogeneous volume change. Consequently, the high coulombic efficiencies observed in this study are likely to have originated from the amorphous silicon structure [22]. Fig. 5(b) shows the discharge capacity as a function of the cycle number for pristine Si, BSNR@15, and BSNR@35 at a constant current density. The discharge capacity of the pristine Si thick-film can be seen to gradually increase until the fourth cycle, and then drops to 150 mAh g1, or 4% of its initial discharge capacity, after 30 cycles. After the fifth cycle, the discharge capacity of the 4 mmthick pristine silicon electrode rapidly decreases, as the thick film generates mechanical failure during the insertion and extraction of Liþ. Fig. 5(c) shows the relationship between coulombic efficiency and the accumulated irreversible capacity of pristine Si, BSNR@15, and BSNR@35. It was worth noting that the coulombic efficiency of the pristine Si thickfilm is reduced to 64.7%, while simultaneously increasing in irreversible capacity. A solid electrolyte interphase (SEI) for pure silicon generally forms across a wide range, from 0.8 to
0.4 V, and can reappear during subsequent cycles due to cracks and the exposure of new silicon surfaces. In this case, the irreversible reaction is maintained with further progression of the cycles, thereby consuming active lithium due to the continuous formation of an unstable SEI layer. The respective sharp decline and increase of the coulombic efficiency and accumulated irreversible capacity suggests that the structure is completely fragmented after the fifth cycle. Conversely, the discharge capacity of BSNR@35 is maintained at 2411 mAh g1, indicating cycle retention of 75% after 30 cycles. More interestingly, the coulombic efficiencies of BSNR@35 are still quite stable after 30 cycles when compared to the pristine silicon thick-film and BSNR@15. Fig. 6 depicts top-view and cross-sectional images of the pristine Si thick-film and BSNR@35 after 30 cycles. In Fig. 6(a) and (c), it can be seen that there are large cracks present on the surface of the pristine Si thick film that eventually cause a loss of electrical contact with the current collector, as shown in Fig. 6(c). Although these cracks still remain, electrical contact is maintained with the current corrector in the case of BSNR@35, as shown in Fig. 6(b) and (d). This indicates that the silicon morphology is not greatly changed by the volume changes during the insertion and extraction of Liþ, but the empty spaces serve to relieve the mechanical failure [23]. Therefore the unique amorphous silicon nanostructure of BSNR@35, with its high amount of empty space, plays an important role in minimizing the deterioration of the electrical contact between the amorphous silicon and current collector that is caused by the self-alleviation of silicon against physical stress.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
7
Fig. 6 e Top-view and cross-sectional images of (a, c) pristine Si thick film and (b, d) BSNR@35 after 30 cycles.
An impedance spectroscopy technique was employed to identify the resistance, and the resulting Nyquist plots were investigated. The imaginary part of the impedance is plotted as a function of the real part over a wide range of frequencies, as shown in Fig. 7. Impedance spectra for the Si electrodes were obtained with different etching depths after the initial cycle in a frequency range of 1 mHze100 kHz, and the equivalent circuits utilized in the impedance analyses are given in the inset of Fig. 7; where RE is the resistance of the bulk electrolyte, QSEI is the space charge capacitance of the SEI layer, and RSEI is the resistance for Liþ conduction in the SEI layer. In addition, RCT, QDL, and ZW respectively represent the charge
Fig. 7 e EIS plots of pristine silicon thick film, BSNR@15, and BSNR@35 after 30 cycles within a frequency range of between 1 mHz and 100 kHz (inset: the equivalent circuit used to fit the impedance data).
transfer resistance, the double layer capacitance at the electrode surface, and the Warburg impedance. It is generally known that the semicircles formed in the high-frequency region reflect the interfacial characteristics of the electrodes, whereas the straight line in the low-frequency region represents the Warburg diffusion of lithium ions in the electrochemical cells containing silicon electrodes [24]. As shown in Fig. 7, it is clear that the semicircle diameter of BSNR@35 corresponding to 232 U is much smaller than that of the pristine Si thick-film corresponding to 650 U, indicating that the active site area for charge transfer decreases with lithium insertion and extraction cycling. This decrease in the active site area may be due to the reduced boundaries among the active material, current collector and electrolyte. The improved charge transfer at the current collector/electrode/
Fig. 8 e Schematic of morphological changes in pristine Si, BSNR@15, and BSNR35 before and after repeated cycling.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
electrolyte interface is established by the metal-assisted chemical etching process. Fig. 8 shows a schematic of the morphological changes in the pristine Si thick-film, BSNR@15, and BSNR@35 before and after cycling. The pristine Si thick-films used as an anode for lithium ion batteries show capacity fading, and a low cycle retention, due to physical stress and the loss of electrical contact between the active material and the current collector. Although the electrochemical performance of BSNR@15 is slightly better than that of the pristine Si thick-film, the results obtained are still unsatisfactory due to volume changes. However, the cycle retention, coulombic efficiency, and accumulated irreversible capacity of BSNR@35 are all quite stable by comparison due to an ample void space. The high aspect ratio of the BSNR can prevent the propagation of cracks over repeated cycling, resulting in increased connection between electrical contacts. As a result, the unique silicon nanostructure with a 3.5-mm etching depth, as-prepared by Ag-assisted chemical etching, gives improved cycling stability due to a reduction in the charge transfer resistance caused by surface cracks.
Conclusions In order to mitigate the volume change during repeated chargeedischarge processes, bundled Si nanorods with a large number of voids and a high aspect ratio were prepared by Agassisted chemical etching process. These were then employed as an anode material in lithium ion batteries. The BSNR structure of the material has been found to advantageous, owing to its large pore volume and surface area. Consequently, it prevents the volume expansion of silicon during charging and discharging. In addition, the reaction area of the amorphous silicon nanorod structure and lithium increases, thus improving the coulombic efficiency and cycle performance of the lithium secondary cell. The cycle performance of the silicon nanostructure with a high aspect ratio represents a significant improvement over that of a pristine silicon thickfilm. The discharge capacity of the silicon nanostructure (BSNR@35) is retained at 2411 mAh g1 after 30 cycles, due to self-alleviation of the volume change. The electric conduction network is also well maintained, as the reduced volume change improves the charge transfer resistance. Therefore, the use of a high etching depth, achieved by controlling the temperature, presents an effective means to enhance the electrochemical performance of Si anodes in lithium ion batteries.
Acknowledgments This work was supported by the research grants of NRF2012M1A2A2671792, as funded by the National Research Foundation under the Ministry of Science, ICT & Future, Korea. Support was also provided by the KIST Institutional Program.
references
[1] Tarascon JM, Armand M. Issues and challenges facing rechargeable lithium batteries. Nature 2001;414:359e67. [2] Boukamp BA, Lesh GC, Huggins RA. All-solid lithium electrodes with mixed-conductor matrix. J Electrochem Soc 1981;128:725e9. [3] Ryu JH, Kim JW, Sung YE, Oh SM. Failure modes of silicon powder negative electrode in lithium secondary batteries. Electrochem Solid State Lett 2004;7:A306e9. [4] Maranchi JP, Hepp AF, Evans AG, Nuhfer NT, Kumta PN. Interfacial properties of the a-Si/Cu:active-inactive thin-film anode system for lithium-ion batteries. J Electrochem Soc 2006;153:A1246e53. [5] Chan CK, Peng H, Liu G, Mcilwrath K, Zhang XF, Huggins RA, et al. High-performance lithium battery anodes using silicon nanowires. Nat Nanotechnol 2008;3:31e5. [6] Ma H, Cheng F, Chen J, Zhao J, Li C, Tao Z, et al. Nest-like silicon nanospheres for high-capacity lithium storage. Adv Mater 2007;19:4067e70. [7] Kim H, Han B, Choo J, Cho J. Three-dimensional porous silicon particles for use in high-performance lithium secondary batteries. Angew Chem 2008;47:10151e4. [8] Unagami T. Formation mechanism of porous silicon layer by anodization in HF solution. J Electrochem Soc 1980;127:476e83. [9] Qiu T, Chu PK. Self-selective electroless plating: an approach for fabrication of functional 1D nanomaterials. Mater Sci Eng R 2008;61:59e77. [10] Qiu T, Wu XL, Siu GG, Chu PK. Intergrowth mechanism of silicon nanowires and silver dendrites. J Electron Mater 2006;35:1879e84. [11] Huang R, Zhu J. Silicon nanowire array films as advanced anode materials for lithium-ion batteries. Mater Chem Phys 2010;121:519e22. [12] Nakamura T, Hosoya N, Tiwari BP, Adachi S. Properties of silver/porous-silicon nanocomposite powders prepared by metal assisted electroless chemical etching. J Appl Phys 2010;108:104315-1e104315-6. [13] Peng K, Xu Y, Wu Y, Yan Y, Lee S, Zhu J. Aligned singlecrystalline Si nanowire arrays for photovoltaic applications. Small 2005;1:1062e7. [14] Chourou ML, Fukami K, Sakka T, Virtanen S, Ogata YH. Metal-assisted etching of p-type silicon under anodic polarization in HF solution with and without H2O2. Electrochim Acta 2010;55:903e12. [15] Li X, Bohn PW. Metal-assisted chemical etching in HF/H2O2 produces porous silicon. Appl Phys Lett 2000;77:2572e4. [16] Zhang ML, Peng KQ, Fan X, Jie JS, Zhang RQ, Lee ST, et al. Preparation of large-area uniform silicon nanowires arrays through metal-assisted chemical etching. J Phys Chem C 2008;112:4444e50. [17] Huang Z, Wang R, Jia D, Maoying L, Humphrey MG, Zhang C. Low-cost, large-scale, and facile production of Si nanowires exhibiting enhanced third-order optical nonlinearity. ACS Appl Mater Interfaces 2012;4:1553e9. [18] Epur R, Ramanathan M, Beck FR, Manivannan A, Kumta PN. Electrodeposition of amorphous silicon anode for lithium ion batteries. Mater Sci Eng B 2012;177:1157e62. [19] Lee JI, Lee KT, Cho J, Kim J, Choi NS, Park S. Chemicalassisted thermal disproportionation of porous silicon monoxide into silicon-based multicomponent systems. Angew Chem 2012;51:2767e71. [20] Netz A, Huggins RA, Weppner W. The formation and properties of amorphous silicon as negative electrode reactant in lithium systems. J Power Sources 2003;119e121:95e100.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
[21] Obrovac MN, Christensen L. Structural changes in silicon anodes during lithium insertion/extraction. Electrochem Solid State Lett 2004;7:A93e6. [22] Yoshio M, Tsumura T, Dimov N. Electrochemical behaviors of silicon based anode material. J Power Sources 2005;146:10e4.
9
[23] Shin HC, Corno JA, Gole JL, Liu M. Porous silicon negative electrodes for rechargeable lithium batteries. J Power Sources 2005;139:314e20. [24] Ruffo R, Hong SS, Chan CK, Huggins RA, Cui Y. Impedance analysis of silicon nanowire lithium ion battery anodes. J Phys Chem C 2009;113:11390e8.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells Jung Sub Kim a,b, Hun-Gi Jung a, Wonchang Choi a, Haw Young Lee a, Dongjin Byun b, Joong Kee Lee a,* a
Advanced Energy Materials Processing Laboratory, Center for Energy Convergence Research, Korea Institute of Science and Technology, Hwarangno 14-gil 5, Seongbuk-gu, Seoul 136-791, Republic of Korea b Department of Material Science & Engineering, Korea University, Seoul 136-713, Republic of Korea
article info
abstract
Article history:
This study investigates bundle-type silicon nanorods (BSNR) that are aimed at improving
Received 28 September 2013
the discharge capacity and life cycle characteristics of secondary cells, by controlling the
Received in revised form
shape and etching depth of silicon thick-films produced by electroless etching. The pre-
20 January 2014
pared BSNR structure is composed of a columnar bundle, having a diameter of 100 nm and
Accepted 1 February 2014
lengths of 1.5 and 3.5 mm. The etching depths of the nanorods have a significant effect on
Available online xxx
the electrochemical performance characteristics, including the capacity fading and coulombic efficiency. Using a BSNR electrode therefore allows for an anode with a high
Keywords:
capacity and efficiency in lithium ion cells, and can help overcome the issues associated
Silicon
with conventional silicon thick-films. Furthermore, as a result of its unique self-relaxant
Electroless etching
structure, electrode deterioration is improved through mitigation of the volume change.
Metal-assisted chemical etching
Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Anode
reserved.
Surface modification
Introduction The power sources for portable electronics such as mobile phones, notebook computers, and personal digital assistants (PDAs) not only require miniaturization, but also need to be highly energized through having a high energy density. Interest in such energy applications is ever increasing, and secondary cells are becoming more commonly used for hybrid electric vehicles (HEVs). Consequently, silicon is being increasingly used as an anode material for lithium secondary
cells as a replacement for more conventional carbon materials [1]. At present, the graphite materials that are commonly used have a theoretical capacity of 372 mAh g1, whereas that of silicon is about 4200 mAh g1 [2]. However, when silicon is fabricated into an anode, the resulting cell has a charging capacity of only about 3260 mAh g1, a discharge capacity of 1170 mAh g1, and a coulombic efficiency of 35% [3]. Furthermore, when the cell is continuously charged and discharged over five cycles, its discharge capacity rapidly decreases to about 300 mAh g1, or about 10% of its initial value. The reason
* Corresponding author. Tel.: þ82 2 958 5252, þ82 10 8717 0978 (mobile); fax: þ82 2 958 5229. E-mail addresses: [email protected], [email protected] (J.K. Lee). 0360-3199/$ e see front matter Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2014.02.007
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 1 e Schematic illustration of BSNR electrode (a) Processes for preparation of the BSNR electrode, and (b) features of BSNR prepared by metal-assisted chemical etching.
for this rapid decrease is that the insertion of lithium forms a LieSi alloy (Li22Si5), which in turn causes a four-fold volume expansion. This results in a breakdown of the silicon structure and blockage of the electron pathway, ultimately causing the formation of a dead volume and a reduction in the silicon anode’s capacity. Consequently, as the cell is continuously charged and discharged, its total capacity is rapidly reduced. This phenomenon occurs regardless of whether it is a bulk silicon film, or micrometer-size particles [4]. Various methods have been previously proposed to solve these problems. For example, silicon has been grown on the surface of an electric accumulator in the form of a wire, and then used as an electrode. This method proved advantageous in that electrons could move more easily than in a conventional thin film, and the inner stress of the silicon during the charging and discharging of a cell is less than that of a conventional thin film. Overall, these advantages resulted in an excellent cycle performance. However, despite these strong
points, such electrodes cannot be used in practical applications because their performance is based on a thin-film forming process [5]. A method of controlling shape has also been proposed, in which silicon nanoparticles were formed into hollow siliconnanospheres to overcome the inherent weaknesses of silicon nanoparticles in cycling. However, this method proved to be problematic in that much time is needed to form the hollow silicon nanospheres, and the process itself is quite complicated [6]. In yet another approach, a composite of silicon and silica was first formed, and then just the silica was removed by chemical etching to form bulk silicon particles with numerous pores. This was aimed at overcoming the weaknesses of conventional bulk silicon particles in cycling, and differs from conventional methods in that the pores produced help alleviate volume expansion. However, this method also presents problems in terms of the length of time needed to form the
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
pores in the bulk silicon particles, the complexity of the process, and the need for high-temperature heat treatment [7]. The present study differs greatly from these aforementioned methods in that a silicon thick-film is only partially etched for a short period of time, thus producing a bundletype silicon nanorod (BSNR) structure on the surface of a copper current collector. The purpose of this work is to investigate the effects of the BSNR etching depth on the electrochemical characteristics when it is used as an anode material for lithium secondary batteries. The high aspect ratio of BSNRs, combined with their high surface area, provides not only a favorable amount of free space, but also helps them to endure physical stress.
Experimental Deposition of silicon thick-film Copper foils were used as a current collector, after being first cleaned for 10 min in an ultrasonic bath containing a mixture of acetone and ethanol. Silicon thick-films were then deposited onto the Cu foils to a thickness of c.a. 4 mm in a plasma-enhanced chemical vapor deposition (PECVD) reactor chamber. Silane (SiH4) gas was fed into the reactor to facilitate the deposition of the silicon thick-film. Typical operating conditions were a working pressure of 70 mTorr, a plasma power of 200 W, substrate temperatures of 300 C, a SiH4 flow rate of 10 sccm, and an argon flow rate of 30 sccm. The gas flow rates were all accurately controlled by a mass flow controller (MFC), and the deposition times were kept at 2 h for all specimens.
3
and water), which was contained in a dry room with a moisture content of less of than 0.5%. The electrochemical properties of the resulting cells were evaluated by galvanostatic chargeedischarge cycling (MACCOR series-4000). All cells were cycled at a constant current of 410 mA g1 at room temperature, and chargeedischarge cycling tests were performed with cut-off voltages of 2.0 to 0.0 V.
Characterization The pristine and modified silicon thick-films were observed by scanning electron microscopy (SEM, NOVA NanoSEM200, FEI Corp.), both before and after cycling. Following the 30th cycle, the samples were rinsed in DMC solution and then dried in an Ar-filled glove box in preparation for SEM analysis. Raman spectroscopy (Nicolet Almega XR dispersive Raman, Thermo electron co.) was used to determine the silicon nanostructure, using a 632 nm (red) laser excitation. In order to negate the heating effects of the laser, it was necessary to use a low laser power density with a 50x microscope objective, and an exposure time of 4 s. The resulting spectra were recorded at a 4 cm1 resolution between 4000 and 90 cm1; the diameter of the laser spot reaching the sample being about 2 mm. The silicon nanostructure was also investigated by X-ray diffraction (XRD, Rigaku co.), using Cu Ka radiation (l ¼ 1.54059 A). Electrochemical impedance spectroscopy (EIS) experiments with the fully charged cells (up to 0.0 V) were performed using an impedance/gain-phase analyzer (Solartron SI 1260) equipped with an electrochemical interface (Solartron SI 1286). The AC amplitude was 5 mV over a frequency range of 1 mHze100 kHz.
Metal-assisted chemical etching of silicon thick-film
Results and discussion Modification of the aspect ratio of the silicon thick-films was carried out by metal-assisted chemical etching for 5 min at a temperature of either 10 or 25 C. As shown in Fig. 1(a), the entire process consisted of five steps: (1) the preparation of a cleaned current collector, (2) deposition of silicon as a starting material under constant conditions via PECVD, (3) attachment of a chemically stable polymer (PET) to prevent corrosion of the current collector (Cu) in acid solution; and metal deposition (0.01 M AgNO3, 1 M HF) of an Ag catalyst onto the silicon surface for 5 s, followed by silicon etching for 5 min in a solution of 0.036 M HF and 2.5 M H2O2 at temperatures of either 10 or 25 C, (4) removal of Ag metal by immersing in a 5% HNO3 solution for 20 min, followed by removal of the capping tape (PET), and (5) cell assembly using the Si nanostructure produced as the anode material.
Cell fabrication The BSNRs prepared by the two different etching temperatures were used as working electrodes, with lithium foil used as a counter electrode to complete the cell. A liquid electrolyte was used, consisting of 1 M LiPF6 dissolved in a mixture of ethylene carbonate (EC), ethylmethyl carbonate (EMC), and dimethyl carbonate (DMC) (1:1:1 vol%). The assembly of the lithium ion cell was carried out in an Ar-filled glove box (<1 ppm of oxygen
Controlling the porosity in silicon-based electrodes to allow for volume expansion has thus far proven to be quite a challenge. However, Unagami et al. [8] have proposed a reaction scheme involving the localized coupling of redox reactions with the metal surface to explain the metal-assisted etching process: Cathode reaction (at metal): H2O2 þ 2Hþ / 2H2O þ 2hþ
(A.1)
2Hþ þ 2e/H2[
(A.2)
Anode reaction: Si þ 4hþ þ 4HF / SiF4 þ 4Hþ
(A.3)
SiF4 þ 2HF / H2SiF6
(A.4)
Overall reaction:
Si þ H2O2 þ 6HF / 2H2O þ H2SiF6 þ H2[
(A.5)
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 2 e Top-view images of (a) pristine Si thick film, (b) Ag coated Si thick film, (c) dendrite formation of Ag and (d) formation of BSNR structure after elimination of Ag.
The deposition of metal facilitates the etching in HF and H2O2 solutions, and hole-injection is enabled by the reaction of H2O2 on the metal particles. The holes are injected into the silicon valence band, which subsequently leads to diffusion from the metal particles. The reduction of oxidizing metal ions to metallic species, and the dissolution of silicon, both occur spontaneously on the silicon surface. If the etching time is short, then the silicon thick-film is unable to form a porous silicon nanostructure. On the other hand, if the etching time is extremely long, then the etched structure is broken down by the corrosion of the Cu foil in acidic HF solution. Thus, in order to hinder such corrosion, the back side of the Cu foil was protected by a polymer film. As shown in Fig. 1(b), the use of BSNR morphologies containing numerous voids as an anode electrode provides the following functions: (1) improving the electrode reaction kinetics, due to an increase in the surface area caused by the networks of bundles; (2) acting as a self-relaxation material during the insertion and extraction of lithium due to the multitude of voids present; (3) providing good contact with the current collector by means of the direct deposition of Si onto Cu without a binder; and (4) a short lithium diffusion distance and efficient electron transfer when compared to a twodimensional electrode. BSNRs possessing these projected functions were fabricated by means of metal-assisted chemical etching, and were then evaluated as an anode material for lithium ion batteries. Representative images, such as those shown in Fig. 2, were observed in order to investigate any change in the morphologies of the silicon electrodes during each stage of the process. Fig. 2(a) shows that the silicon thick-film has a
smooth surface, which in order to prepare the silicon nanostructure, was immersed into AgNO3/HF solution for 5 s to deposit Ag particles. The resultant particle size distribution is important for understanding the physical and chemical properties related to the effect of varying the bundle diameter. The Ag particles observed on the silicon thick-film had sizes in the range of 20e250 nm, as shown in Fig. 2(b). The Ag was also found to form dendrites due to continuous reaction, as shown in Figs. 1(a) and 2(c) [9,10]. The Ag-coated Si thickfilms were next etched in HF/H2O2 solution for 5 min, and then dipped into HNO3 solution for 20 min to eliminate the Ag particles. This resulted in very different surface morphologies, as shown in Fig. 2(d). Those BSNRs with a corresponding diameter of 100 nm, and a large fraction of voids, offer a reduced volume change in response to the large variation in physical stress induced during the insertion and extraction of Liþ ions. Metal-assisted chemical etching is a simple and low-cost technique that can potentially be applied to the mass production of various silicon nanostructures with easily controlled parameters; such as the doping type, doping level, and etching temperature [11e16]. Huang et al. [17] showed that the channel depth changes as the etching time increases, and many other parameters have been examined in previous studies without considering the temperature. Pristine Si thickfilms, with a thickness of 4 mm, were therefore etched at different temperatures of 10 and 25 C. The cross-sectional images reveal different etching depths of about 1.5 and 3.5 mm, which correspond to aspect ratios of 15 (BNSR@15) and 35 (BNSR@35) in Fig. 3(b) and (c), respectively. This shows that the etching depth is a key parameter for the electrochemical
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
5
Fig. 3 e Cross-sectional images of (a) pristine Si thick film, (b) BSNR@15 and (c) BSNR@35 with different aspect ratios.
performance of the silicon anodes, and is dependent on the etching temperature. Raman spectroscopy with a 632 nm red laser was used to obtain structural information regarding the BSNR fabricated by metal-assisted chemical etching. It is known that in the case of crystalline silicon, a sharp peak appears at 520 cm1 due to the transverse optical vibrational modes, as shown in Fig. 4(a). As the BNSRs have an amorphous structure, the transverse optical peak becomes broader and shifts to lower wavenumbers of about 35 cm1. It is clear that no sharp peak can be observed at 520 cm1 for the deposited films, and instead a broad peak is observed at 485 cm1 that suggests the deposited film is predominantly amorphous [18,19]. The XRD pattern of the BSNR is shown in Fig. 4(b), in which no peaks corresponding to crystalline silicon can be observed, thus further confirming that the silicon obtained by metal-assisted chemical etching is amorphous. The intensities of all peaks
are clearly observed at scattering angles (2q) of about 43.8 , 50.9 , and 74.5 , which are ascribed to the (111), (200), and (220) planes of the Cu current collector, respectively. The XRD data therefore matches well with the Raman spectra. Fig. 5(a) shows the initial voltage profiles of a pristine silicon thick-film, BSNR@15, and BSNR@35 in the potential range of 0e2 V, at a current density of 410 mA g1. All cell tests involved chargeedischarge cycling to a discharge depth of nearly 100%, without limiting the insertion capacity. As shown in Fig. 5(a), the pristine silicon thick-film exhibits charge and discharge capacities of 3604 and 3391 mAh g1, with a coulombic efficiency of 94% in the initial cycle. The charge-discharge capacities of the BSNR@15 and BSNR@35 electrodes are 3091, 2861, 3406, and 3203 mAh g1, indicating coulombic efficiencies of 92.5 and 94%, respectively. This unprecedentedly high coulombic efficiency in the initial cycle signifies an almost complete Liþ extraction during the charge
Fig. 4 e (a) Raman data (as a single-crystal silicon wafer for reference) and (b) XRD of BSNR for observation of the silicon structure. Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 5 e (a) Initial voltage profiles, (b) discharge capacities, (c) coulombic efficiencies and accumulated irreversible capacity of pristine silicon thick film, BSNR@15 and BSNR@35 in the potential range of 0e2 V at a constant current density of 410 mA gL1.
and discharge processes. Amorphous silicon was previously investigated by Netz [20], who made a cell of amorphous silicon and evaluated its electrochemical properties. Christensen et al. [21] reported that the charging profiles have two distinct voltage regions at 100 to 0.7 V and 70 to 0.05 V due to the phase transition from crystalline to amorphous, and the amorphous change to Li15Si4 in the first cycle, respectively. These phase transitions induce a large volume-expansion of the active silicon material, thus resulting in deterioration of the electrical conduction network. Amorphous silicon without a phase transition therefore leads to a high reversible reaction compared to crystalline silicon, due to its homogeneous volume change. Consequently, the high coulombic efficiencies observed in this study are likely to have originated from the amorphous silicon structure [22]. Fig. 5(b) shows the discharge capacity as a function of the cycle number for pristine Si, BSNR@15, and BSNR@35 at a constant current density. The discharge capacity of the pristine Si thick-film can be seen to gradually increase until the fourth cycle, and then drops to 150 mAh g1, or 4% of its initial discharge capacity, after 30 cycles. After the fifth cycle, the discharge capacity of the 4 mmthick pristine silicon electrode rapidly decreases, as the thick film generates mechanical failure during the insertion and extraction of Liþ. Fig. 5(c) shows the relationship between coulombic efficiency and the accumulated irreversible capacity of pristine Si, BSNR@15, and BSNR@35. It was worth noting that the coulombic efficiency of the pristine Si thickfilm is reduced to 64.7%, while simultaneously increasing in irreversible capacity. A solid electrolyte interphase (SEI) for pure silicon generally forms across a wide range, from 0.8 to
0.4 V, and can reappear during subsequent cycles due to cracks and the exposure of new silicon surfaces. In this case, the irreversible reaction is maintained with further progression of the cycles, thereby consuming active lithium due to the continuous formation of an unstable SEI layer. The respective sharp decline and increase of the coulombic efficiency and accumulated irreversible capacity suggests that the structure is completely fragmented after the fifth cycle. Conversely, the discharge capacity of BSNR@35 is maintained at 2411 mAh g1, indicating cycle retention of 75% after 30 cycles. More interestingly, the coulombic efficiencies of BSNR@35 are still quite stable after 30 cycles when compared to the pristine silicon thick-film and BSNR@15. Fig. 6 depicts top-view and cross-sectional images of the pristine Si thick-film and BSNR@35 after 30 cycles. In Fig. 6(a) and (c), it can be seen that there are large cracks present on the surface of the pristine Si thick film that eventually cause a loss of electrical contact with the current collector, as shown in Fig. 6(c). Although these cracks still remain, electrical contact is maintained with the current corrector in the case of BSNR@35, as shown in Fig. 6(b) and (d). This indicates that the silicon morphology is not greatly changed by the volume changes during the insertion and extraction of Liþ, but the empty spaces serve to relieve the mechanical failure [23]. Therefore the unique amorphous silicon nanostructure of BSNR@35, with its high amount of empty space, plays an important role in minimizing the deterioration of the electrical contact between the amorphous silicon and current collector that is caused by the self-alleviation of silicon against physical stress.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
7
Fig. 6 e Top-view and cross-sectional images of (a, c) pristine Si thick film and (b, d) BSNR@35 after 30 cycles.
An impedance spectroscopy technique was employed to identify the resistance, and the resulting Nyquist plots were investigated. The imaginary part of the impedance is plotted as a function of the real part over a wide range of frequencies, as shown in Fig. 7. Impedance spectra for the Si electrodes were obtained with different etching depths after the initial cycle in a frequency range of 1 mHze100 kHz, and the equivalent circuits utilized in the impedance analyses are given in the inset of Fig. 7; where RE is the resistance of the bulk electrolyte, QSEI is the space charge capacitance of the SEI layer, and RSEI is the resistance for Liþ conduction in the SEI layer. In addition, RCT, QDL, and ZW respectively represent the charge
Fig. 7 e EIS plots of pristine silicon thick film, BSNR@15, and BSNR@35 after 30 cycles within a frequency range of between 1 mHz and 100 kHz (inset: the equivalent circuit used to fit the impedance data).
transfer resistance, the double layer capacitance at the electrode surface, and the Warburg impedance. It is generally known that the semicircles formed in the high-frequency region reflect the interfacial characteristics of the electrodes, whereas the straight line in the low-frequency region represents the Warburg diffusion of lithium ions in the electrochemical cells containing silicon electrodes [24]. As shown in Fig. 7, it is clear that the semicircle diameter of BSNR@35 corresponding to 232 U is much smaller than that of the pristine Si thick-film corresponding to 650 U, indicating that the active site area for charge transfer decreases with lithium insertion and extraction cycling. This decrease in the active site area may be due to the reduced boundaries among the active material, current collector and electrolyte. The improved charge transfer at the current collector/electrode/
Fig. 8 e Schematic of morphological changes in pristine Si, BSNR@15, and BSNR35 before and after repeated cycling.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
electrolyte interface is established by the metal-assisted chemical etching process. Fig. 8 shows a schematic of the morphological changes in the pristine Si thick-film, BSNR@15, and BSNR@35 before and after cycling. The pristine Si thick-films used as an anode for lithium ion batteries show capacity fading, and a low cycle retention, due to physical stress and the loss of electrical contact between the active material and the current collector. Although the electrochemical performance of BSNR@15 is slightly better than that of the pristine Si thick-film, the results obtained are still unsatisfactory due to volume changes. However, the cycle retention, coulombic efficiency, and accumulated irreversible capacity of BSNR@35 are all quite stable by comparison due to an ample void space. The high aspect ratio of the BSNR can prevent the propagation of cracks over repeated cycling, resulting in increased connection between electrical contacts. As a result, the unique silicon nanostructure with a 3.5-mm etching depth, as-prepared by Ag-assisted chemical etching, gives improved cycling stability due to a reduction in the charge transfer resistance caused by surface cracks.
Conclusions In order to mitigate the volume change during repeated chargeedischarge processes, bundled Si nanorods with a large number of voids and a high aspect ratio were prepared by Agassisted chemical etching process. These were then employed as an anode material in lithium ion batteries. The BSNR structure of the material has been found to advantageous, owing to its large pore volume and surface area. Consequently, it prevents the volume expansion of silicon during charging and discharging. In addition, the reaction area of the amorphous silicon nanorod structure and lithium increases, thus improving the coulombic efficiency and cycle performance of the lithium secondary cell. The cycle performance of the silicon nanostructure with a high aspect ratio represents a significant improvement over that of a pristine silicon thickfilm. The discharge capacity of the silicon nanostructure (BSNR@35) is retained at 2411 mAh g1 after 30 cycles, due to self-alleviation of the volume change. The electric conduction network is also well maintained, as the reduced volume change improves the charge transfer resistance. Therefore, the use of a high etching depth, achieved by controlling the temperature, presents an effective means to enhance the electrochemical performance of Si anodes in lithium ion batteries.
Acknowledgments This work was supported by the research grants of NRF2012M1A2A2671792, as funded by the National Research Foundation under the Ministry of Science, ICT & Future, Korea. Support was also provided by the KIST Institutional Program.
references
[1] Tarascon JM, Armand M. Issues and challenges facing rechargeable lithium batteries. Nature 2001;414:359e67. [2] Boukamp BA, Lesh GC, Huggins RA. All-solid lithium electrodes with mixed-conductor matrix. J Electrochem Soc 1981;128:725e9. [3] Ryu JH, Kim JW, Sung YE, Oh SM. Failure modes of silicon powder negative electrode in lithium secondary batteries. Electrochem Solid State Lett 2004;7:A306e9. [4] Maranchi JP, Hepp AF, Evans AG, Nuhfer NT, Kumta PN. Interfacial properties of the a-Si/Cu:active-inactive thin-film anode system for lithium-ion batteries. J Electrochem Soc 2006;153:A1246e53. [5] Chan CK, Peng H, Liu G, Mcilwrath K, Zhang XF, Huggins RA, et al. High-performance lithium battery anodes using silicon nanowires. Nat Nanotechnol 2008;3:31e5. [6] Ma H, Cheng F, Chen J, Zhao J, Li C, Tao Z, et al. Nest-like silicon nanospheres for high-capacity lithium storage. Adv Mater 2007;19:4067e70. [7] Kim H, Han B, Choo J, Cho J. Three-dimensional porous silicon particles for use in high-performance lithium secondary batteries. Angew Chem 2008;47:10151e4. [8] Unagami T. Formation mechanism of porous silicon layer by anodization in HF solution. J Electrochem Soc 1980;127:476e83. [9] Qiu T, Chu PK. Self-selective electroless plating: an approach for fabrication of functional 1D nanomaterials. Mater Sci Eng R 2008;61:59e77. [10] Qiu T, Wu XL, Siu GG, Chu PK. Intergrowth mechanism of silicon nanowires and silver dendrites. J Electron Mater 2006;35:1879e84. [11] Huang R, Zhu J. Silicon nanowire array films as advanced anode materials for lithium-ion batteries. Mater Chem Phys 2010;121:519e22. [12] Nakamura T, Hosoya N, Tiwari BP, Adachi S. Properties of silver/porous-silicon nanocomposite powders prepared by metal assisted electroless chemical etching. J Appl Phys 2010;108:104315-1e104315-6. [13] Peng K, Xu Y, Wu Y, Yan Y, Lee S, Zhu J. Aligned singlecrystalline Si nanowire arrays for photovoltaic applications. Small 2005;1:1062e7. [14] Chourou ML, Fukami K, Sakka T, Virtanen S, Ogata YH. Metal-assisted etching of p-type silicon under anodic polarization in HF solution with and without H2O2. Electrochim Acta 2010;55:903e12. [15] Li X, Bohn PW. Metal-assisted chemical etching in HF/H2O2 produces porous silicon. Appl Phys Lett 2000;77:2572e4. [16] Zhang ML, Peng KQ, Fan X, Jie JS, Zhang RQ, Lee ST, et al. Preparation of large-area uniform silicon nanowires arrays through metal-assisted chemical etching. J Phys Chem C 2008;112:4444e50. [17] Huang Z, Wang R, Jia D, Maoying L, Humphrey MG, Zhang C. Low-cost, large-scale, and facile production of Si nanowires exhibiting enhanced third-order optical nonlinearity. ACS Appl Mater Interfaces 2012;4:1553e9. [18] Epur R, Ramanathan M, Beck FR, Manivannan A, Kumta PN. Electrodeposition of amorphous silicon anode for lithium ion batteries. Mater Sci Eng B 2012;177:1157e62. [19] Lee JI, Lee KT, Cho J, Kim J, Choi NS, Park S. Chemicalassisted thermal disproportionation of porous silicon monoxide into silicon-based multicomponent systems. Angew Chem 2012;51:2767e71. [20] Netz A, Huggins RA, Weppner W. The formation and properties of amorphous silicon as negative electrode reactant in lithium systems. J Power Sources 2003;119e121:95e100.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 4 ) 1 e9
[21] Obrovac MN, Christensen L. Structural changes in silicon anodes during lithium insertion/extraction. Electrochem Solid State Lett 2004;7:A93e6. [22] Yoshio M, Tsumura T, Dimov N. Electrochemical behaviors of silicon based anode material. J Power Sources 2005;146:10e4.
9
[23] Shin HC, Corno JA, Gole JL, Liu M. Porous silicon negative electrodes for rechargeable lithium batteries. J Power Sources 2005;139:314e20. [24] Ruffo R, Hong SS, Chan CK, Huggins RA, Cui Y. Impedance analysis of silicon nanowire lithium ion battery anodes. J Phys Chem C 2009;113:11390e8.
Please cite this article in press as: Kim JS, et al., Bundle-type silicon nanorod anodes produced by electroless etching using silver ions and their electrochemical characteristics in lithium ion cells, International Journal of Hydrogen Energy (2014), http:// dx.doi.org/10.1016/j.ijhydene.2014.02.007