- Email: [email protected]
Effect of radical polymer cathode thickness on the electrochemical performance of organic radical battery
Available online at www.sciencedirect.com
Solid State Ionics 178 (2007) 1546 – 1551 www.elsevier.com/locate/ssi
Effect of radical polymer cathode thickness on the electrochemical performance of organic radical battery Jae-Kwang Kim a , Gouri Cheruvally a , Jae-Won Choi a , Jou-Hyeon Ahn a,⁎, Seo Hwan Lee b , Doo Seong Choi b , Choong Eui Song b,⁎ a
Department of Chemical and Biological Engineering, and ITRC for Energy Storage and Conversion, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, South Korea b Institute of Basic Science and Department of Chemistry, Sungkyunkwan University, Suwon, Gyeonggi-do 440-746, South Korea Received 14 March 2007; received in revised form 10 August 2007; accepted 25 September 2007
Abstract The influence of cathode thickness on the electrochemical performance of an organic radical battery (ORB) with the radical polymer poly(2,2,6,6-tetramethylpiperidinyloxy-4-yl methacrylate) (PTMA) as the cathode active material is presented. The ORB consists of lithium metal anode and PTMA cathode with an active material content of 40 wt.%. An increase in cathode thickness results in a decrease in specific capacity and discharge voltage of the cell and an increase in electrode/electrolyte interfacial resistance. The best performance is achieved with a thin cathode of 17 μm that shows nearly 100% utilization of the active material (∼111 mAh/g) at current densities up to 1.0 mA/cm2. The cell exhibits excellent highrate capability and cycle characteristics with a stable impedance behavior and an intact cathode structure on cycling. The results demonstrate that high performance can be achieved from the non-conductive PTMA cathode with higher active material content by using a thin and properly prepared cathode consisting of a uniform, nanometer range coating of the polymer layer on the conductive carbon particles. © 2007 Elsevier B.V. All rights reserved. Keywords: Organic radical battery; Cathode materials; Lithium secondary batteries; PTMA; Electrochemical property
1. Introduction Organic radical batteries (ORBs) are emerging as promising, environmentally benign, high-rate capable power sources. An ORB is composed of at least one of its electrodes based on an organic radical as the active material. The first report on the use of an organic radical as an electrode-active or charge-storage material in lithium battery was by Nakahara et al. in 2002 [1]. They employed the radical polymer PTMA, poly(2,2,6,6tetramethylpiperidinyloxy-4-yl methacrylate), as the cathode active material in lithium battery. Since then, a few more studies were reported that demonstrate the vast potential of ORBs, especially high-rate capability [2–7]. PTMA is a derivative of ⁎ Corresponding authors. Ahn is to be contacted at Fax: +82 55 753 1808. Song, Fax: +82 31 290 7075. E-mail addresses: [email protected] (J.-H. Ahn), [email protected] (C.E. Song). 0167-2738/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ssi.2007.09.009
polymethacrylate with a 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) radical in the repeating unit. Compounds containing TEMPO radical are known to show redox behavior in aprotic solvents in the range of 3.69–4.04 V vs. Li/Li+, independent of their structure [8,9]. TEMPO-based compounds have attracted interest as subjects of electron spin resonance [10] and molecular motion studies [11]. They find applications as polymeric stabilizer [12], oxidants of alcohols [13] and charge-storage materials [1,2]. Polyacetylene and polynorbonene derivatives carrying TEMPO also were synthesized recently and evaluated as materials for ORBs [14]. The nitroxide radical in TEMPO is characterized by good chemical stability [15] contributed by its resonance structures. Two redox couples can be exhibited by nitroxide radical: the anodic oxidation resulting in oxoammonium cation (p-type doping) and the cathodic reduction resulting in aminoxyl anion (n-type doping) [15]. The anodic oxidation of nitroxide radical in PTMA has been utilized in employing it as the cathode active
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
material of rechargeable ORBs [1–7]. A significantly small molecular weight of 30 per radical moiety, nearly 100% localized density of the unpaired electron in the radical and the extremely rapid electron transfer rate are important advantages of PTMA as a battery material. The fast reaction kinetics of PTMA during the charge–discharge operations, with an estimated electron transfer rate constant of N 10− 2 cm/s, leads to high power-rate capability for the battery [2,3]. Since PTMA is an insulator, the cathode composition usually consists of only a small proportion of the active material; the major portion being the conductive agent like carbon. Thus, many of the earlier studies have employed only 10 wt.% of PTMA with high amount (about 80 wt.%) of carbon [1,2,4]. For practical applications of ORB, it is essential to increase the active material content of cathode and attain higher specific energy from the battery. A few recent studies employ 30–50 wt. % of PTMA in the cathode [5–7]. The thickness of PTMA cathode needs to be optimized for deriving the maximum performance from the non-conductive active material. In this article we present the results of our evaluation of the influence of cathode thickness on the performance of a room temperature Li/PTMA cell with a relatively high PTMA content of 40 wt.%.
1547
glass transition temperature (Tg) of PTMA was determined by differential scanning calorimetry (DSC: TA 2040) at a heating rate of 5 °C/min. The thermal stability was analyzed by thermogravimetric analysis (TGA: SDT Q600 TA, USA) in nitrogen atmosphere at a heating rate of 10 °C/min from 20 °C to 400 °C. The Swagelok® type Li/PTMA coin cell was fabricated by stacking lithium metal (300 μm thickness, Cyprus Foote Mineral Co.) anode and PTMA-based cathode with Celgard 2200 separator film. 1 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1 by vol) was used as the liquid electrolyte. The lithium salt and organic solvents were supplied by Aldrich. The cell assembly was performed under argon atmosphere in a glove box (H2O b 10 ppm). The charge–discharge and cycling properties of the cell were evaluated between 3.0 and 4.0 V at different current densities using an automatic galvanostatic charge–discharge unit, WBCS-3000 battery cycler (WonATech. Co.) at room temperature. Cyclic voltammetry measurement of the cell was conducted at a scan rate of 5 mV/s between 2.5 and 4.4 V and AC impedance of the cell was measured using IM6 frequency analyzer over a frequency range of 100 mHz to 2 MHz. 3. Results and discussion
2. Experimental PTMA was synthesized by the radical polymerization method, similar to the one reported by Nakahara et al. [1]. The synthesis steps followed are shown in Scheme 1. 2,2,6,6tetramethylpiperidine methacrylate monomer is polymerized first using 2,2'-azobisisobutyronitrile radical initiator and then oxidized with H2O2 in the presence of NaWO4 catalyst to get PTMA. For preparing the cathode, 40 parts by weight (pbw) of PTMA was mixed with 50 pbw of conductive, Super-P carbon and 10 pbw of poly(vinylidene fluoride) binder in N-methyl pyrrolidone solvent. The ingredients were mixed uniformly by ball milling at room temperature for 30 min and the slurry was cast on an aluminum foil and dried at 70 °C for 12 h. The film so obtained was cut into circular discs of 1.01 cm diameter for use as cathode. Cathode films of thicknesses varying between 17 and 64 μm were prepared using different spacers. The mass of cathode discs varied between 2.0 to 2.6 mg for these thicknesses. The surface morphology examination was performed with scanning electron microscope (SEM) (JEOL JSM 5600). The
Scheme 1. Synthesis of PTMA.
The radical polymer PTMA has a glass transition temperature of 74 °C that is convenient enough for easy processing, and good thermal stability with an initial decomposition temperature (corresponding to 10 wt.% decomposition) of 270 °C and peak decomposition temperature of 300 °C. The radical centers in PTMA are reported to be stable up to the decomposition temperature of the polymer and the storage stability of PTMA under ambient conditions is also good with no change in radical concentration for more than a year [3]. The SEM image of PTMA is shown in Fig. 1 (a). It reveals that at room temperature, the amorphous, glassy polymer has a homogenous, flake-like morphology. Fig. 1 (b) shows the morphology of PTMA-based cathode fabricated by blending PTMA, carbon powder (of average particle size 40 nm) and PVdF binder. Well-defined, uniform particle morphology is observed with an average particle size of 80 nm. This shows that the carbon particles have been coated with a very thin layer (thickness of ∼ 40 nm) of the polymer. During the cathode preparation by the solution process of ball mill blending of ingredients, the carbon particles with high surface area adsorb the polymer solution and get impregnated with it. Subsequent removal of solvent by evaporation leaves the thin coating (nanometer thick) of the active material on the conductive carbon particles. The uniform morphology of the electrode so obtained is highly beneficial in realizing good electrochemical performance for the Li/PTMA cell. Fig. 2 shows the CVof Li/PTMA cell with PTMA cathode of 17 μm thickness. A sharp, single redox couple with a redox half-wave potential at 3.6 V vs. Li+/Li is obtained. The peak separation between the anodic peak (at 3.75 V) and cathodic peak (at 3.45 V) is only 300 mV. Such low redox peak separation has been reported in earlier studies employing PTMA cathodes as well [1,2] and has been attributed to the fast
1548
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 3. CV of Li/PTMA cell with cathode of 64 μm thickness, during the first, second and fifth cycles (room temperature, 2.5 to 4.4 V, 5 mV/s).
electrode reaction kinetics of the radical polymer. The anodic and cathodic currents as well as the area under the peaks remain nearly the same indicating high coulomb efficiency for the charge–discharge processes. On repeated cycling, there is no significant change in the anodic and cathodic peak potentials or currents, as can be seen from the curves shown in the figure for second and fifth cycles. The CV behavior of Li/PTMA cell with PTMA cathode of higher thickness of 64 μm is presented in Fig. 3. Compared to the redox peaks shown by the thinner (17 μm) cathode, the thicker one shows much broader peaks. Although the redox half wave potential is still at 3.6 V, the anodic and cathodic peaks get shifted further apart and the peak separation is 700 mV. The peak current is also slightly less than that of the thinner cathode. These observations demonstrate that
electron transfer reaction as well as lithium ion diffusion becomes progressively difficult in thicker electrodes. Nevertheless, good reversibility on repeated cycling is exhibited by the thicker cathode as well. The initial charge–discharge properties of the cells with PTMA cathodes of different thicknesses at 1C current density (∼ 0.1 mA/cm2) are compared in Fig. 4. It shows that the specific capacity is influenced highly by the cathode thickness. As expected, the thinnest cathode exhibits the maximum specific capacity; 111 mAh/g is attained, which corresponds to 100% utilization of PTMA in the cathode (the theoretical capacity of PTMA is estimated as 111 mAh/g assuming one electron per monomer unit in the redox process) [1]. With cathodes of higher thickness, approximately two, three and four times the minimum of 17 μm, lower specific capacities of 80, 67 and 55 mAh/g are obtained respectively. The earlier studies with 10 wt.% PTMA cathode have reported discharge capacities of 77 mAh/g at a current density of 0.1 mA/cm2 with 150–250 μm thick cathode [1], and 81 mAh/g at 0.5 C (cathode thickness not mentioned) [4]. One of the reasons for the lower performance achieved in the above studies that contained significant amount of
Fig. 2. CV of Li/PTMA cell with cathode of 17 μm thickness, during the first, second and fifth cycles (room temperature, 2.5 to 4.4 V, 5 mV/s).
Fig. 4. Initial charge and discharge capacities of Li/PTMA cells with PTMAbased cathodes of varying thicknesses (room temperature, 1 C rate, 3.0 to 4.0 V).
Fig. 1. SEM images of: (a) radical polymer, PTMA (b) PTMA-based cathode.
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 5. Discharge–charge cycles of Li//PTMA cells with cathodes of 17 μm and 64 μm thicknesses, at 1 C current density (room temperature, 3.0 to 4.0 V).
carbon (∼80 wt.%) could be the higher thickness of the cathode employed. Nishide et al. employed PTMA cathode prepared by mixing the polymer with 20–50 wt.% graphite fibers (exact formulation not disclosed) and reported a high discharge capacity of 110 mAh/g at 1 C [3]. The high performance was attributed to the cathode structure consisting of a thin layer of polymer (50– 100 nm thick) coating on the conductive graphite fibers (150 nm). These results clearly demonstrate the importance of reducing the thickness of cathode to the minimum as possible in employing the insulating polymer as active material. This becomes more relevant in the case of a cathode with a higher content of the insulating polymer, as in the present study (40 wt.%). We observe that high PTMA utilization is realizable with enhanced PTMA content from a thinner cathode having a nanometer layer coating of the polymer on the conductive carbon. The cathode thickness does not seem to affect the coulomb efficiency of the cells: thus, the respective charge and discharge capacities are the same for all cells. However, the difference between the charge and discharge voltages (ΔV) increases with an increase in cathode thickness. The electrochemical processes take place at well-defined plateau voltages in the case of thinner cathodes, whereas, almost continuously increasing/decreasing curves are observed for thicker ones. ΔV increases from 0.03 V to 0.40 V between cathodes of 17 and 64 μm respectively (an average voltage was taken for cathode of 64 μm thickness, without a definite plateau voltage). This observation is an indication of the polarization taking place in the cell with thicker cathodes. The results highlight the need for a proper design and fabrication of the radical polymer cathode with the lowest possible thickness, incorporating adequate amounts of conductive agent. A similar thickness effect had been reported for other battery electrodes with active materials of insulating nature, like sulfur [16] and lithium iron phosphate [17] and is attributed to a larger electrode resistance and a slower Li+ ion conduction though the thicker electrode films. A comparison of the time of discharging followed by charging of the cells with PTMA cathodes of 17 and 64 μm thicknesses is given in Fig. 5. The process was carried out at a current density of 1 C (∼0.1 mA/cm2), allowing a time gap of 10 s between
1549
charging and discharging. The superior performance of the thinner cathode is obvious; it performs almost ideally, with a discharge and charge time of one h each. The thicker one lags behind with a 50% performance level. The cell with thinner cathode maintains the performance level at higher current densities also. Thus, at a current density of 10 C (∼1 mA/cm2), which is much higher than that of conventional batteries, the ORB can be charged in about 6 min. Hence, it follows that the rapid electron transfer reaction of the organic radical that is responsible for the fast electrochemical process has not been hindered by confining it in the cathode layer. This could be made possible by employing a thin electrode consisting of a nanometer thick layer of active material coating efficiently on the conductive carbon. The cycle performances of Li/PTMA cells with radical cathodes of different thicknesses were evaluated for over 100 cycles and the results are shown in Fig. 6. The cells showed a stable performance: the thinner cathodes of 17 and 36 μm result in ∼90% retention of their initial capacities after 100 cycles and the thicker ones of 50 and 64 μm show ∼70% retention. The excellent cycle performance of the cells results from a combination of different favorable factors such as the chemical stability of the nitroxide radical, the simple redox reaction involving a singleelectron transfer, and the absence of any structural change for the active material during charge–discharge processes [3]. The amorphous nature of PTMA polymer that might result in a partial swelling in the presence of the organic solvents in the electrolyte might also be helpful in providing good counterion mobility during the redox process. A stable electrode/electrolyte interfacial property is essential for attaining good cycle performance from a battery. The study of impedance behavior of the cell during cycling would provide an understanding of the changes in the electrode/electrolyte interfacial property. Fig. 7 compares the impedance spectra obtained for cells with PTMA cathodes of 17 and 64 μm thicknesses before subjecting to cycling and also for the cathode of 17 μm thickness after cycling. Typical semi-circle shaped impedance spectra are observed showing the real-axis intercepts at high frequency region corresponding to the bulk electrolyte resistance (Re) and at low frequency region corresponding to
Fig. 6. Cycle performance of Li/PTMA cells with PTMA-based cathodes of varying thicknesses (room temperature, 1 C rate, 3.0 to 4.0 V).
1550
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 7. Impedance spectra of Li/PTMA cells: with cathode of 17 μm thickness, before cycling, and after 30, 100 charge–discharge cycles; and with cathode of 64 μm thickness before cycling (room temperature, 10 C-rate, 3.0 to 4.0 V).
the electrode/electrolyte interface resistance (Rf). Re was the same for the cells with different cathode thicknesses. During cycling of the cell also Re did not change appreciably (range of 2–4 Ω). This shows that there is no significant reduction in ionic conductivity of the electrolyte or mobility of ions with repeated cycling of the cell. However, Rf varied with cathode thickness: 473 Ω for thicker one vs. 346 Ω for thinner one. On cycling of the cell with 17 μm thick cathode, Rf decreased considerably to 306 and 232 Ω respectively after 30 and 100 cycles. The gradual decrease in Rf is a result of the stable electrode/electrolyte interfaces formed which permit the easy passage of ions and indicates good compatibility of the electrolyte with both lithium and PTMA electrodes. The interface at the lithium metal surface is especially important for stable cycling, since any unwelcome reactions between the electrolyte and lithium could result in the formation of lithium dendrites affecting the performance of the cell on cycling. For evaluating the rate-capability of ORB, the performance of the cell with PTMA cathode of 17 μm thickness was evaluated at different current densities from 1 C to 50 C (0.1 to 5 mA/cm2). The initial discharge capacities and cycle properties
Fig. 8. Initial discharge capacities of Li/PTMA cells at different C-rates (cathode thickness 17 μm, room temperature, 3.0 to 4.0 V).
Fig. 9. Cycle performance of Li/PTMA cells at different C-rates (cathode thickness 17 μm, room temperature, 3.0 to 4.0 V).
are compared in Fig. 8 and 9 respectively. There is no decrease in discharge capacity up to 10 C-rate; a decrease is observed only at very high C-rates. With increasing current density, the discharge voltage also shows a decreasing trend. Thus, ΔV is minimal (0.03 V) at 1 C-rate, but increases to 0.84 V at 50 Crate. This is an expected behavior of the electrochemical cell where practical limitations inherent of electrodes and electrolytes do not meet the demands of very fast electrode reactions at higher C-rates. The performance of the cell on cycling is nearly constant for current densities between 1 C and 20 C. At 50 C, after a few initial cycles with lower performance, a stable capacity of ∼ 75% of that delivered at 1 C is achieved. This remarkably good rate-capability of ORB is a direct consequence of the fast electrode reaction kinetics in the thin polymer layer. The structural integrity of the cathode is also an important factor that contributes to high rate-capability and cycle property. The SEM image of the PTMA cathode that underwent 100 charge– discharge cycles at 10 C-rate is shown in Fig. 10. The cathode structure remained nearly the same as its initial state (shown in Fig. 1 (b)). Since the redox reaction of PTMA does not involve any structural change, and the liquid electrolyte solvents employed here do not dissolve away the active material from the original position, the cathode structure and composition remain intact upon repeated cycling.
Fig. 10. SEM image of PTMA cathode after 100 cycles of charge–discharge at 10 C-rate.
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
1551
4. Conclusion
References
The effect of cathode thickness on the electrochemical performance of ORB with the insulating polymer radical, PTMA as the cathode active material, has been investigated. PTMA cathode containing 40 wt.% of the active material is prepared by a solution (in an organic solvent) casting method which results in a uniform, nanometer thick coating of PTMA over the conductive carbon. It is observed that cathode thickness has a profound influence on the specific capacity of the cell, whereas, the cycle property remained nearly unchanged. Li/PTMA cell with the thinnest electrode shows sharp redox peaks in CVand is capable of delivering the full discharge capacities at reasonably high C-rates and performs extraordinarily well over repeated cycling. The cell with thicker electrode exhibits higher electrode/ electrolyte interfacial resistance and polarization of the cell. By employing sufficiently thin PTMA cathode consisting of a thin, nanometer range covering of PTMA layer over the carbon particles, room temperature ORB with good performance could be realized.
[1] K. Nakahara, S. Iwasa, M. Satoh, Y. Morioka, J. Iriyama, M. Suguro, E. Hasegawa, Chem. Phys. Lett. 359 (2002) 351. [2] H. Nishide, S. Iwasa, Y.-J. Pu, T. Suga, K. Nakahara, M. Satoh, Electrochim. Acta 50 (2004) 827. [3] H. Nishide, T. Suga, Electrochem. Soc. Interface Winter (2005) 32. [4] H. Li, Y. Zou, Y. Xia, Electrochim. Acta. 52 (2007) 2153. [5] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 163 (2007) 1110. [6] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 165 (2007) 870. [7] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 165 (2007) 398. [8] F. MacCorquodale, J.A. Crayston, J.C. Walton, D.J. Worsfold, Tetrahedron Lett. 31 (1990) 771. [9] K. Nakahara, S. Iwasa, J. Iriyama, Y. Morioka, M. Suguro, M. Satoh, E.J. Cairns, Electrochim. Acta 52 (2006) 921. [10] O.H. Griffith, J.F.W. Keana, S. Rottschaefer, T.A. Warlick, J. Am. Chem. Soc. 89 (1967) 5072. [11] Z. Veksli, W.G. Miller, Macromolecules 10 (1977) 686. [12] M. Dagonneau, V.B. Ivanov, E.G. Rozantsev, V.D. Sholle, E.S. Kagan, J. Macromol. Sci., Rev. Macromol. Chem. Phys. C22 (1982) 169. [13] C.D. Anderson, K.J. Shea, S.D. Rychnovsky, Org. Lett. 7 (2005) 4879. [14] T. Katsumata, M. Satoh, J. Wada, M. Shiotsuki, F. Sanda, T. Masuda, Macromol. Rapid Comm. 27 (2006) 1206. [15] L.B. Volodarsky, V.A. Reznikov, Synthetic Chemistry of Stable Notroxides, CRC Press, Florida, 1993. [16] S.E. Cheon, K.S. Ko, J.H. Cho, S.W. Kim, E.Y. Chin, H.T. Kim, J. Electrochem. Soc. 150 (2003) A800. [17] D.Y.W. Yu, K. Donoue, T. Inoue, M. Fujimoto, S. Fujitani, J. Electrochem. Soc. 153 (2006) A835.
Acknowledgements This research was supported by the Ministry of Information and Communication (MIC), Korea, under the Information Technology Research Center (ITRC) support program supervised by the Institute of Information Technology Assessment (IITA), the Division of Advanced Batteries in NGE Program (Project No. 10016439), and the Korea Research Foundation (KRF-2005-005-J11901). Gouri Cheruvally is thankful to the KOFST for the award of Brain Pool Fellowships.
Solid State Ionics 178 (2007) 1546 – 1551 www.elsevier.com/locate/ssi
Effect of radical polymer cathode thickness on the electrochemical performance of organic radical battery Jae-Kwang Kim a , Gouri Cheruvally a , Jae-Won Choi a , Jou-Hyeon Ahn a,⁎, Seo Hwan Lee b , Doo Seong Choi b , Choong Eui Song b,⁎ a
Department of Chemical and Biological Engineering, and ITRC for Energy Storage and Conversion, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, South Korea b Institute of Basic Science and Department of Chemistry, Sungkyunkwan University, Suwon, Gyeonggi-do 440-746, South Korea Received 14 March 2007; received in revised form 10 August 2007; accepted 25 September 2007
Abstract The influence of cathode thickness on the electrochemical performance of an organic radical battery (ORB) with the radical polymer poly(2,2,6,6-tetramethylpiperidinyloxy-4-yl methacrylate) (PTMA) as the cathode active material is presented. The ORB consists of lithium metal anode and PTMA cathode with an active material content of 40 wt.%. An increase in cathode thickness results in a decrease in specific capacity and discharge voltage of the cell and an increase in electrode/electrolyte interfacial resistance. The best performance is achieved with a thin cathode of 17 μm that shows nearly 100% utilization of the active material (∼111 mAh/g) at current densities up to 1.0 mA/cm2. The cell exhibits excellent highrate capability and cycle characteristics with a stable impedance behavior and an intact cathode structure on cycling. The results demonstrate that high performance can be achieved from the non-conductive PTMA cathode with higher active material content by using a thin and properly prepared cathode consisting of a uniform, nanometer range coating of the polymer layer on the conductive carbon particles. © 2007 Elsevier B.V. All rights reserved. Keywords: Organic radical battery; Cathode materials; Lithium secondary batteries; PTMA; Electrochemical property
1. Introduction Organic radical batteries (ORBs) are emerging as promising, environmentally benign, high-rate capable power sources. An ORB is composed of at least one of its electrodes based on an organic radical as the active material. The first report on the use of an organic radical as an electrode-active or charge-storage material in lithium battery was by Nakahara et al. in 2002 [1]. They employed the radical polymer PTMA, poly(2,2,6,6tetramethylpiperidinyloxy-4-yl methacrylate), as the cathode active material in lithium battery. Since then, a few more studies were reported that demonstrate the vast potential of ORBs, especially high-rate capability [2–7]. PTMA is a derivative of ⁎ Corresponding authors. Ahn is to be contacted at Fax: +82 55 753 1808. Song, Fax: +82 31 290 7075. E-mail addresses: [email protected] (J.-H. Ahn), [email protected] (C.E. Song). 0167-2738/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ssi.2007.09.009
polymethacrylate with a 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) radical in the repeating unit. Compounds containing TEMPO radical are known to show redox behavior in aprotic solvents in the range of 3.69–4.04 V vs. Li/Li+, independent of their structure [8,9]. TEMPO-based compounds have attracted interest as subjects of electron spin resonance [10] and molecular motion studies [11]. They find applications as polymeric stabilizer [12], oxidants of alcohols [13] and charge-storage materials [1,2]. Polyacetylene and polynorbonene derivatives carrying TEMPO also were synthesized recently and evaluated as materials for ORBs [14]. The nitroxide radical in TEMPO is characterized by good chemical stability [15] contributed by its resonance structures. Two redox couples can be exhibited by nitroxide radical: the anodic oxidation resulting in oxoammonium cation (p-type doping) and the cathodic reduction resulting in aminoxyl anion (n-type doping) [15]. The anodic oxidation of nitroxide radical in PTMA has been utilized in employing it as the cathode active
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
material of rechargeable ORBs [1–7]. A significantly small molecular weight of 30 per radical moiety, nearly 100% localized density of the unpaired electron in the radical and the extremely rapid electron transfer rate are important advantages of PTMA as a battery material. The fast reaction kinetics of PTMA during the charge–discharge operations, with an estimated electron transfer rate constant of N 10− 2 cm/s, leads to high power-rate capability for the battery [2,3]. Since PTMA is an insulator, the cathode composition usually consists of only a small proportion of the active material; the major portion being the conductive agent like carbon. Thus, many of the earlier studies have employed only 10 wt.% of PTMA with high amount (about 80 wt.%) of carbon [1,2,4]. For practical applications of ORB, it is essential to increase the active material content of cathode and attain higher specific energy from the battery. A few recent studies employ 30–50 wt. % of PTMA in the cathode [5–7]. The thickness of PTMA cathode needs to be optimized for deriving the maximum performance from the non-conductive active material. In this article we present the results of our evaluation of the influence of cathode thickness on the performance of a room temperature Li/PTMA cell with a relatively high PTMA content of 40 wt.%.
1547
glass transition temperature (Tg) of PTMA was determined by differential scanning calorimetry (DSC: TA 2040) at a heating rate of 5 °C/min. The thermal stability was analyzed by thermogravimetric analysis (TGA: SDT Q600 TA, USA) in nitrogen atmosphere at a heating rate of 10 °C/min from 20 °C to 400 °C. The Swagelok® type Li/PTMA coin cell was fabricated by stacking lithium metal (300 μm thickness, Cyprus Foote Mineral Co.) anode and PTMA-based cathode with Celgard 2200 separator film. 1 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1 by vol) was used as the liquid electrolyte. The lithium salt and organic solvents were supplied by Aldrich. The cell assembly was performed under argon atmosphere in a glove box (H2O b 10 ppm). The charge–discharge and cycling properties of the cell were evaluated between 3.0 and 4.0 V at different current densities using an automatic galvanostatic charge–discharge unit, WBCS-3000 battery cycler (WonATech. Co.) at room temperature. Cyclic voltammetry measurement of the cell was conducted at a scan rate of 5 mV/s between 2.5 and 4.4 V and AC impedance of the cell was measured using IM6 frequency analyzer over a frequency range of 100 mHz to 2 MHz. 3. Results and discussion
2. Experimental PTMA was synthesized by the radical polymerization method, similar to the one reported by Nakahara et al. [1]. The synthesis steps followed are shown in Scheme 1. 2,2,6,6tetramethylpiperidine methacrylate monomer is polymerized first using 2,2'-azobisisobutyronitrile radical initiator and then oxidized with H2O2 in the presence of NaWO4 catalyst to get PTMA. For preparing the cathode, 40 parts by weight (pbw) of PTMA was mixed with 50 pbw of conductive, Super-P carbon and 10 pbw of poly(vinylidene fluoride) binder in N-methyl pyrrolidone solvent. The ingredients were mixed uniformly by ball milling at room temperature for 30 min and the slurry was cast on an aluminum foil and dried at 70 °C for 12 h. The film so obtained was cut into circular discs of 1.01 cm diameter for use as cathode. Cathode films of thicknesses varying between 17 and 64 μm were prepared using different spacers. The mass of cathode discs varied between 2.0 to 2.6 mg for these thicknesses. The surface morphology examination was performed with scanning electron microscope (SEM) (JEOL JSM 5600). The
Scheme 1. Synthesis of PTMA.
The radical polymer PTMA has a glass transition temperature of 74 °C that is convenient enough for easy processing, and good thermal stability with an initial decomposition temperature (corresponding to 10 wt.% decomposition) of 270 °C and peak decomposition temperature of 300 °C. The radical centers in PTMA are reported to be stable up to the decomposition temperature of the polymer and the storage stability of PTMA under ambient conditions is also good with no change in radical concentration for more than a year [3]. The SEM image of PTMA is shown in Fig. 1 (a). It reveals that at room temperature, the amorphous, glassy polymer has a homogenous, flake-like morphology. Fig. 1 (b) shows the morphology of PTMA-based cathode fabricated by blending PTMA, carbon powder (of average particle size 40 nm) and PVdF binder. Well-defined, uniform particle morphology is observed with an average particle size of 80 nm. This shows that the carbon particles have been coated with a very thin layer (thickness of ∼ 40 nm) of the polymer. During the cathode preparation by the solution process of ball mill blending of ingredients, the carbon particles with high surface area adsorb the polymer solution and get impregnated with it. Subsequent removal of solvent by evaporation leaves the thin coating (nanometer thick) of the active material on the conductive carbon particles. The uniform morphology of the electrode so obtained is highly beneficial in realizing good electrochemical performance for the Li/PTMA cell. Fig. 2 shows the CVof Li/PTMA cell with PTMA cathode of 17 μm thickness. A sharp, single redox couple with a redox half-wave potential at 3.6 V vs. Li+/Li is obtained. The peak separation between the anodic peak (at 3.75 V) and cathodic peak (at 3.45 V) is only 300 mV. Such low redox peak separation has been reported in earlier studies employing PTMA cathodes as well [1,2] and has been attributed to the fast
1548
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 3. CV of Li/PTMA cell with cathode of 64 μm thickness, during the first, second and fifth cycles (room temperature, 2.5 to 4.4 V, 5 mV/s).
electrode reaction kinetics of the radical polymer. The anodic and cathodic currents as well as the area under the peaks remain nearly the same indicating high coulomb efficiency for the charge–discharge processes. On repeated cycling, there is no significant change in the anodic and cathodic peak potentials or currents, as can be seen from the curves shown in the figure for second and fifth cycles. The CV behavior of Li/PTMA cell with PTMA cathode of higher thickness of 64 μm is presented in Fig. 3. Compared to the redox peaks shown by the thinner (17 μm) cathode, the thicker one shows much broader peaks. Although the redox half wave potential is still at 3.6 V, the anodic and cathodic peaks get shifted further apart and the peak separation is 700 mV. The peak current is also slightly less than that of the thinner cathode. These observations demonstrate that
electron transfer reaction as well as lithium ion diffusion becomes progressively difficult in thicker electrodes. Nevertheless, good reversibility on repeated cycling is exhibited by the thicker cathode as well. The initial charge–discharge properties of the cells with PTMA cathodes of different thicknesses at 1C current density (∼ 0.1 mA/cm2) are compared in Fig. 4. It shows that the specific capacity is influenced highly by the cathode thickness. As expected, the thinnest cathode exhibits the maximum specific capacity; 111 mAh/g is attained, which corresponds to 100% utilization of PTMA in the cathode (the theoretical capacity of PTMA is estimated as 111 mAh/g assuming one electron per monomer unit in the redox process) [1]. With cathodes of higher thickness, approximately two, three and four times the minimum of 17 μm, lower specific capacities of 80, 67 and 55 mAh/g are obtained respectively. The earlier studies with 10 wt.% PTMA cathode have reported discharge capacities of 77 mAh/g at a current density of 0.1 mA/cm2 with 150–250 μm thick cathode [1], and 81 mAh/g at 0.5 C (cathode thickness not mentioned) [4]. One of the reasons for the lower performance achieved in the above studies that contained significant amount of
Fig. 2. CV of Li/PTMA cell with cathode of 17 μm thickness, during the first, second and fifth cycles (room temperature, 2.5 to 4.4 V, 5 mV/s).
Fig. 4. Initial charge and discharge capacities of Li/PTMA cells with PTMAbased cathodes of varying thicknesses (room temperature, 1 C rate, 3.0 to 4.0 V).
Fig. 1. SEM images of: (a) radical polymer, PTMA (b) PTMA-based cathode.
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 5. Discharge–charge cycles of Li//PTMA cells with cathodes of 17 μm and 64 μm thicknesses, at 1 C current density (room temperature, 3.0 to 4.0 V).
carbon (∼80 wt.%) could be the higher thickness of the cathode employed. Nishide et al. employed PTMA cathode prepared by mixing the polymer with 20–50 wt.% graphite fibers (exact formulation not disclosed) and reported a high discharge capacity of 110 mAh/g at 1 C [3]. The high performance was attributed to the cathode structure consisting of a thin layer of polymer (50– 100 nm thick) coating on the conductive graphite fibers (150 nm). These results clearly demonstrate the importance of reducing the thickness of cathode to the minimum as possible in employing the insulating polymer as active material. This becomes more relevant in the case of a cathode with a higher content of the insulating polymer, as in the present study (40 wt.%). We observe that high PTMA utilization is realizable with enhanced PTMA content from a thinner cathode having a nanometer layer coating of the polymer on the conductive carbon. The cathode thickness does not seem to affect the coulomb efficiency of the cells: thus, the respective charge and discharge capacities are the same for all cells. However, the difference between the charge and discharge voltages (ΔV) increases with an increase in cathode thickness. The electrochemical processes take place at well-defined plateau voltages in the case of thinner cathodes, whereas, almost continuously increasing/decreasing curves are observed for thicker ones. ΔV increases from 0.03 V to 0.40 V between cathodes of 17 and 64 μm respectively (an average voltage was taken for cathode of 64 μm thickness, without a definite plateau voltage). This observation is an indication of the polarization taking place in the cell with thicker cathodes. The results highlight the need for a proper design and fabrication of the radical polymer cathode with the lowest possible thickness, incorporating adequate amounts of conductive agent. A similar thickness effect had been reported for other battery electrodes with active materials of insulating nature, like sulfur [16] and lithium iron phosphate [17] and is attributed to a larger electrode resistance and a slower Li+ ion conduction though the thicker electrode films. A comparison of the time of discharging followed by charging of the cells with PTMA cathodes of 17 and 64 μm thicknesses is given in Fig. 5. The process was carried out at a current density of 1 C (∼0.1 mA/cm2), allowing a time gap of 10 s between
1549
charging and discharging. The superior performance of the thinner cathode is obvious; it performs almost ideally, with a discharge and charge time of one h each. The thicker one lags behind with a 50% performance level. The cell with thinner cathode maintains the performance level at higher current densities also. Thus, at a current density of 10 C (∼1 mA/cm2), which is much higher than that of conventional batteries, the ORB can be charged in about 6 min. Hence, it follows that the rapid electron transfer reaction of the organic radical that is responsible for the fast electrochemical process has not been hindered by confining it in the cathode layer. This could be made possible by employing a thin electrode consisting of a nanometer thick layer of active material coating efficiently on the conductive carbon. The cycle performances of Li/PTMA cells with radical cathodes of different thicknesses were evaluated for over 100 cycles and the results are shown in Fig. 6. The cells showed a stable performance: the thinner cathodes of 17 and 36 μm result in ∼90% retention of their initial capacities after 100 cycles and the thicker ones of 50 and 64 μm show ∼70% retention. The excellent cycle performance of the cells results from a combination of different favorable factors such as the chemical stability of the nitroxide radical, the simple redox reaction involving a singleelectron transfer, and the absence of any structural change for the active material during charge–discharge processes [3]. The amorphous nature of PTMA polymer that might result in a partial swelling in the presence of the organic solvents in the electrolyte might also be helpful in providing good counterion mobility during the redox process. A stable electrode/electrolyte interfacial property is essential for attaining good cycle performance from a battery. The study of impedance behavior of the cell during cycling would provide an understanding of the changes in the electrode/electrolyte interfacial property. Fig. 7 compares the impedance spectra obtained for cells with PTMA cathodes of 17 and 64 μm thicknesses before subjecting to cycling and also for the cathode of 17 μm thickness after cycling. Typical semi-circle shaped impedance spectra are observed showing the real-axis intercepts at high frequency region corresponding to the bulk electrolyte resistance (Re) and at low frequency region corresponding to
Fig. 6. Cycle performance of Li/PTMA cells with PTMA-based cathodes of varying thicknesses (room temperature, 1 C rate, 3.0 to 4.0 V).
1550
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
Fig. 7. Impedance spectra of Li/PTMA cells: with cathode of 17 μm thickness, before cycling, and after 30, 100 charge–discharge cycles; and with cathode of 64 μm thickness before cycling (room temperature, 10 C-rate, 3.0 to 4.0 V).
the electrode/electrolyte interface resistance (Rf). Re was the same for the cells with different cathode thicknesses. During cycling of the cell also Re did not change appreciably (range of 2–4 Ω). This shows that there is no significant reduction in ionic conductivity of the electrolyte or mobility of ions with repeated cycling of the cell. However, Rf varied with cathode thickness: 473 Ω for thicker one vs. 346 Ω for thinner one. On cycling of the cell with 17 μm thick cathode, Rf decreased considerably to 306 and 232 Ω respectively after 30 and 100 cycles. The gradual decrease in Rf is a result of the stable electrode/electrolyte interfaces formed which permit the easy passage of ions and indicates good compatibility of the electrolyte with both lithium and PTMA electrodes. The interface at the lithium metal surface is especially important for stable cycling, since any unwelcome reactions between the electrolyte and lithium could result in the formation of lithium dendrites affecting the performance of the cell on cycling. For evaluating the rate-capability of ORB, the performance of the cell with PTMA cathode of 17 μm thickness was evaluated at different current densities from 1 C to 50 C (0.1 to 5 mA/cm2). The initial discharge capacities and cycle properties
Fig. 8. Initial discharge capacities of Li/PTMA cells at different C-rates (cathode thickness 17 μm, room temperature, 3.0 to 4.0 V).
Fig. 9. Cycle performance of Li/PTMA cells at different C-rates (cathode thickness 17 μm, room temperature, 3.0 to 4.0 V).
are compared in Fig. 8 and 9 respectively. There is no decrease in discharge capacity up to 10 C-rate; a decrease is observed only at very high C-rates. With increasing current density, the discharge voltage also shows a decreasing trend. Thus, ΔV is minimal (0.03 V) at 1 C-rate, but increases to 0.84 V at 50 Crate. This is an expected behavior of the electrochemical cell where practical limitations inherent of electrodes and electrolytes do not meet the demands of very fast electrode reactions at higher C-rates. The performance of the cell on cycling is nearly constant for current densities between 1 C and 20 C. At 50 C, after a few initial cycles with lower performance, a stable capacity of ∼ 75% of that delivered at 1 C is achieved. This remarkably good rate-capability of ORB is a direct consequence of the fast electrode reaction kinetics in the thin polymer layer. The structural integrity of the cathode is also an important factor that contributes to high rate-capability and cycle property. The SEM image of the PTMA cathode that underwent 100 charge– discharge cycles at 10 C-rate is shown in Fig. 10. The cathode structure remained nearly the same as its initial state (shown in Fig. 1 (b)). Since the redox reaction of PTMA does not involve any structural change, and the liquid electrolyte solvents employed here do not dissolve away the active material from the original position, the cathode structure and composition remain intact upon repeated cycling.
Fig. 10. SEM image of PTMA cathode after 100 cycles of charge–discharge at 10 C-rate.
J.-K. Kim et al. / Solid State Ionics 178 (2007) 1546–1551
1551
4. Conclusion
References
The effect of cathode thickness on the electrochemical performance of ORB with the insulating polymer radical, PTMA as the cathode active material, has been investigated. PTMA cathode containing 40 wt.% of the active material is prepared by a solution (in an organic solvent) casting method which results in a uniform, nanometer thick coating of PTMA over the conductive carbon. It is observed that cathode thickness has a profound influence on the specific capacity of the cell, whereas, the cycle property remained nearly unchanged. Li/PTMA cell with the thinnest electrode shows sharp redox peaks in CVand is capable of delivering the full discharge capacities at reasonably high C-rates and performs extraordinarily well over repeated cycling. The cell with thicker electrode exhibits higher electrode/ electrolyte interfacial resistance and polarization of the cell. By employing sufficiently thin PTMA cathode consisting of a thin, nanometer range covering of PTMA layer over the carbon particles, room temperature ORB with good performance could be realized.
[1] K. Nakahara, S. Iwasa, M. Satoh, Y. Morioka, J. Iriyama, M. Suguro, E. Hasegawa, Chem. Phys. Lett. 359 (2002) 351. [2] H. Nishide, S. Iwasa, Y.-J. Pu, T. Suga, K. Nakahara, M. Satoh, Electrochim. Acta 50 (2004) 827. [3] H. Nishide, T. Suga, Electrochem. Soc. Interface Winter (2005) 32. [4] H. Li, Y. Zou, Y. Xia, Electrochim. Acta. 52 (2007) 2153. [5] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 163 (2007) 1110. [6] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 165 (2007) 870. [7] K. Nakahara, J. Iriyama, S. Iwasa, M. Suguro, M. Satoh, E.J. Cairns, J. Power Sources 165 (2007) 398. [8] F. MacCorquodale, J.A. Crayston, J.C. Walton, D.J. Worsfold, Tetrahedron Lett. 31 (1990) 771. [9] K. Nakahara, S. Iwasa, J. Iriyama, Y. Morioka, M. Suguro, M. Satoh, E.J. Cairns, Electrochim. Acta 52 (2006) 921. [10] O.H. Griffith, J.F.W. Keana, S. Rottschaefer, T.A. Warlick, J. Am. Chem. Soc. 89 (1967) 5072. [11] Z. Veksli, W.G. Miller, Macromolecules 10 (1977) 686. [12] M. Dagonneau, V.B. Ivanov, E.G. Rozantsev, V.D. Sholle, E.S. Kagan, J. Macromol. Sci., Rev. Macromol. Chem. Phys. C22 (1982) 169. [13] C.D. Anderson, K.J. Shea, S.D. Rychnovsky, Org. Lett. 7 (2005) 4879. [14] T. Katsumata, M. Satoh, J. Wada, M. Shiotsuki, F. Sanda, T. Masuda, Macromol. Rapid Comm. 27 (2006) 1206. [15] L.B. Volodarsky, V.A. Reznikov, Synthetic Chemistry of Stable Notroxides, CRC Press, Florida, 1993. [16] S.E. Cheon, K.S. Ko, J.H. Cho, S.W. Kim, E.Y. Chin, H.T. Kim, J. Electrochem. Soc. 150 (2003) A800. [17] D.Y.W. Yu, K. Donoue, T. Inoue, M. Fujimoto, S. Fujitani, J. Electrochem. Soc. 153 (2006) A835.
Acknowledgements This research was supported by the Ministry of Information and Communication (MIC), Korea, under the Information Technology Research Center (ITRC) support program supervised by the Institute of Information Technology Assessment (IITA), the Division of Advanced Batteries in NGE Program (Project No. 10016439), and the Korea Research Foundation (KRF-2005-005-J11901). Gouri Cheruvally is thankful to the KOFST for the award of Brain Pool Fellowships.