Materials Research Bulletin 41 (2006) 1023–1037 www.elsevier.com/locate/matresbu
Evaluation of a cross-linked polyurethane acrylate as polymer electrolyte for lithium batteries P. Santhosh a, A. Gopalan a,*, T. Vasudevan a, Kwang-Pill Lee b a
b
Department of Industrial Chemistry, Alagappa University, Karaikudi 630003, India Department of Chemistry Education, Kyungpook National University, Daegu 702-701, South Korea
Received 6 September 2005; received in revised form 9 December 2005; accepted 16 December 2005 Available online 9 January 2006
Abstract A cross-linked polyurethane acrylate (CL-PUA) was synthesized by end capping 2,6-toluene diisocyanate (TDI)/poly(ethylene glycol) (PEG) based prepolymer with hydroxybutyl methacrylate (HBMA). Differential scanning calorimetry (DSC) and Fourier transform infra-red (FT-IR) spectroscopy measurements reveal the possible presence of significant interactions between lithium ions and soft/hard segments of the CL-PUA, when CL-PUA was complexed with lithium perchlorate (LiClO4). CL-PUA follows the VTF relationship for the ion transport. Predominant formation of contact ion pairs of LiClO4 has been observed through AC conductivity and DSC measurements. The lithium stripping–plating process is a reversible and implies better electrochemical stability in the working voltage range. Also, CL-PUA electrolyte shows better compatibility with lithium metal as inferred from impedance measurements and has a good cationic transference number suitable to be used as a solid polymer electrolyte. The addition of HBMA into PU matrix improves tensile strength of the CL-PUA. Swelling measurements of CL-PUA with plasticizer showed better dimensional stability. Also, a cell was constructed using CL-PUA as electrolyte and the performance was assessed. # 2006 Elsevier Ltd. All rights reserved. Keywords: A. Polymers; B. Chemical synthesis; C. DSC; C. TGA; D. Ionic conductivity
1. Introduction There is great interest in the development of high-performance sources of energy for applications such as mobile telephones, laptops, electrical vehicles and engine/battery hybrid vehicles. Among the energy sources, great attention is devoted to the studies of lithium and lithium-ion cells. The application of polymer electrolytes in lithium batteries in the place of liquid ones avoids the problem of electrolyte leakage from the devices and also enables the design of devices with large surfaces and any shape [1]. It has been found that polymer electrolytes based on host polymer matrices having oxyethylene chains yield higher ionic conductivities only at elevated temperatures [2,3]. Semi-crystalline character and high glass transition temperature are the reason for their low conductivity at ambient temperature. On the other hand, gel polymer electrolytes (GPE) based on polyacrylonitrile [4,5], poly(vinylidene fluoride) [6,7] and poly(methyl methacrylate) [8,9] show higher ionic conductivity in the order of mS/cm at room temperature. * Corresponding author. Tel.: +91 4565 228836; fax: +91 4565 225202. E-mail address:
[email protected] (A. Gopalan). 0025-5408/$ – see front matter # 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.materresbull.2005.12.004
1024
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
In recent years, polyurethane (PU) based polymers have received attention as polymer electrolytes for rechargeable lithium ion batteries [10–14]. The interest in using PU as matrix for polymer electrolyte is related to the possibility of increasing mechanical strength for linear polyethers due to their phase-separated microstructures. Furthermore, the low glass transition temperature (Tg) and hence higher segmental motion of the polyether soft segments leads to higher mobility of the dissolved ions. The hard segment domains, which are in the glassy state and either distributed or interconnected throughout the rubbery phase of the soft segment, act as reinforcing filler and hence contribute to the dimensional stability of the polymer electrolyte. It is perceptible from the earlier reports [10–14] that PU type polymer electrolyte with –(CH2–CH2–O)– units in it shows better conductivity in comparison with the other analogues. However, the dimensional stability of the electrolyte must be still improved to have the practical viability. Cross-linking of some of the units as side chains is one of the various ways to improve the mechanical property of a polymer [15]. Systematic studies of UV-curable urethane acrylate were carried out by Copper and co-workers [16]. Kim et al. [15] have studied the electrochemical and electrical properties of CL-PUA–LiCF3SO3 system using AC impedance spectroscopy. Sun et al. [17] reported the mechanical properties of urethane cross-linked poly(ethylene oxide-co-propylene oxide) glycerol ether–plasticizer (tetraethylene glycol dimethyl ether, or methyl formamide)– lithium perchlorate (LiClO4) based polymer electrolytes. They found that the elastic modulus and tensile strength of the materials unexpectedly decrease with increasing the salt concentration. This may be due to the predominance of intra-molecular coordination of the Li+ ions by the polymer. Preparation of a cross-linked polyurethane acrylate (CL-PUA) end capped with hydroxyethyl acrylate was reported [18]. DSC, FT-IR and 7Li MAS NMR spectroscopy were employed to identify the interactions of lithium ions with the soft and hard segments of CL-PUA. A series of CL-PUA were synthesized using 4,40 -methylene bis(phenyl isocyanate), polyethylene glycol, hydroxyl methacrylate and different reactive vinyl and divinyl diluents [19]. The electrolytes were prepared by UV radiation induced cross-linking of the PUA–diluent mixture. Depending upon the composition, these electrolytes exhibited wide range of mechanical and electrical properties and showed good compatibility with Li electrodes. In the present study, we have prepared a cross-linked polyurethane acrylate. Modification was made by reacting the end groups (NCO) of the PU prepolymer with hydroxybutyl methacrylate (HBMA) and subsequently cross-linked by polymerizing the unreacted vinyl bond in the HBMA units. Typically, a cross-linked polyurethane acrylate was synthesized by end capping 2,6-toluene diisocyanate (TDI)/poly(ethylene glycol) (PEG) based prepolymer with HBMA followed by thermal cross-linking. Characterization studies of the LiClO4 doped CL-PUA were made with differential scanning calorimetry (DSC), Fourier transform infra-red (FT-IR) spectroscopy and AC impedance measurements to bring out the phase transitions, interaction of Li+ ions with segments of CL-PUA and ionic conductivity, respectively. Electrochemical characteristics inform that the CL-PUA can be ideally suited for use in rechargeable lithium batteries. 2. Experimental 2.1. Materials Poly(ethylene glycol) (molecular weight: 400; E. Merck) was dehydrated under reduced pressure at 80 8C for 24 h before use. 2,6-Toluene diisocyanate, 1,4-butane diol, hydroxybutyl methacrylate (Aldrich), 1,10 -azobis(cyclohexane carbonitrile) (Aldrich) and the commercial battery electrolyte, LP 20 (Merck) were used as received. LiClO4 (Aldrich) was dehydrated at 120 8C under reduced pressure for 72 h. Li metal and LiCoO2 were purchased from Aldrich and used. All the other reagents and chemicals were used without further purification. 2.2. Synthesis of PUA Polyurethane acrylate was synthesized by a two-step addition process, where the prepolymer was made by reaction of excess of TDI with PEG and then capping the end NCO groups by the reaction with HBMA. The prepolymer was made by allowing the mixture of TDI and PEG to react at 85 8C for 6 h with constant stirring under the dry nitrogen blanket. After the prepolymer formation was over (confirmed by estimation of NCO groups by di-butyl amine back titration method), dimethyl formamide (DMF) was added to dissolve the prepolymer. The temperature of the mass was
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1025
reduced to 60 8C and then the required amount of HBMA was slowly added to the reaction mixture. The reaction was then allowed to continue for 2 h and then it was terminated by the addition of 2 mL of methanol. A molar ratio of 2:1:2 for TDI, PEG and HBMA, respectively, was used for the polymerization of CL-PUA. 2.3. Preparation of the cross-linked PUA and polymer electrolytes Solid polymer electrolytes (SPE) were prepared by mixing various concentrations of LiClO4 in DMF and PUA in DMF. 1,10 -Azobis(cyclohexane carbonitrile) was also mixed with the PUA for thermal cross-linking through the acrylate end groups. After homogeneous mixing, the solution was cast into teflon plates. Solvent removal and simultaneous cross-linking were done at 80 8C under reduced pressure for 72 h. The films were then kept inside the glove box before further experiments. The undoped film was also made in the same way but without having LiClO4 for swelling study in liquid electrolyte (LP 20). 2.4. Differential scanning calorimeter DSC experiments were carried out using a DSC 2010 differential scanning calorimeter (TA Instruments, USA) over a temperature range 150 to 150 8C at a scan rate of 10 8C/min. The dry samples were sealed in Al crucibles inside the glove box. The sealed samples were taken out from the glove box only at the time of DSC experiments. The samples were first annealed at 150 8C for 10 min, cooled down to 150 8C and then scanned. All the thermograms are base line corrected and calibrated against Indium metal. Glass transition temperature (Tg) was reported as the midpoint of the transition process and melting temperature was the peak temperature. 2.5. Fourier transform infra-red spectroscopy FT-IR spectra were taken at ambient temperature using Perkin-Elmer Rx1 instrument with a wave number resolution of 4 cm1. Samples for FT-IR were made by casting the polymer–salt mixture directly on KBr pellets and then simultaneously dried at 120 8C for 48 h. One hundred and twenty-five scans were signal averaged to increase the s/n ratio. Data processing was performed by Grams 386 software (Galactic Industries Corporation). 2.6. AC impedance measurements Impedance measurements of the polymer electrolytes were performed using thin films made by casting from solution drying (as described earlier). Film thickness was maintained in the range of 250–350 mm and the area of contact was 1.0 cm2. For the measurement of ionic conductivity, the samples were sandwiched between two stainless steel (SS 304) electrodes. Cell assembly was carried out in dry argon atmosphere inside the glove box. Conductivity measurements were performed using EG&G PAR 6310 Potentiostat/Galvanostat controlled by the frequency response analysis (FRA) under an oscillation potential of 10 mV. 2.7. Cyclic voltammetry and linear sweep voltammetry Cyclic voltammetry (CV) and linear sweep voltammetry (LSV) were performed with a cell made by using stainless steel as working electrode and Li metal as counter as well as reference electrode. CV experiments were performed in the potential range of 0.5 to 4.0 V (versus Li). For LSV, the potential of the working electrode was varied from 2.0 to 6.0 V (versus Li). A sweep rate of 5 mV/s was used in both cases. Electrochemical measurements were performed using Bio Analytical System 100 B (BAS 100BW) Electrochemical Analyzer. 2.8. DC polarization measurement DC polarization cell was constructed by sandwiching the electrolyte film between symmetric lithium metal electrodes. A small constant DC potential difference of about 10 mV was applied across the symmetric cell and the
1026
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
current was measured as a function of time until it reaches a constant value. By considering the potential drop occurring at surface layers on the electrode, the cation transport number t+ was determined by using the relation, tþ ¼
Iss ðDV Ii Ri Þ Ii ðDV Iss Rss Þ
where Ii and Iss are the initial and steady-state currents, Ri and Rss the initial and steady-state resistance of the passivated layers and DV is the applied potential; AC impedance spectra were recorded before and after the current relaxation measurement without interruption of the DC bias, to permit Ri and Rss to be evaluated. 2.9. Thermogravimetric analysis Thermogravimetric analysis (TGA) of the prepared polymer electrolyte samples was performed under nitrogen atmosphere using Perkin-Elmer TGA 7/DX Thermal Analyzer with a scan rate of 20 8C/min. 2.10. Mechanical strength Tensile strength was measured at room temperature by means of a Universal tensile machine (Instron model 5565, Lloyd) at a full-out velocity of 50 mm min1. The polymer electrolyte sample thickness was 100 mm. Measurements were performed five times for each sample, and the average value was calculated. 2.11. Charge–discharge cycling tests LiCoO2 and Li were employed as cathode and anode, respectively. The area of both electrodes was fixed as 2.0 cm2. Li/PE/LiCoO2 laminated cells were assembled by pressing Li, PE and LiCoO2, sealed by polyethylene film and laminated by an aluminium foil. A cell was assembled in the dry argon atmosphere inside the glove box. Cyclic voltammetry and AC impedance measurements were made for the laminated cell. The observed cyclic voltammetric patterns for the lithium insertion–deinsertion redox process and impedance phenomena ensured the interfacial contact between CL-PUA and Li electrode. The charge–discharge cycling tests of the laminated cell electrolyte were conducted under galvanostatic conditions in a dry argon atmosphere. Also, discharge curves were obtained at different current rates in order to obtain the rate capacity of the cell at room temperature. 3. Results and discussion 3.1. Differential scanning calorimeter DSC was used to examine the effect of LiClO4 on the morphology based thermal transitions of CL-PUA and glass transition temperature of soft segments (SS) in polyether units. DSC results (Table 1) of the pristine CL-PUA and LiClO4 doped CL-PUA are presented. The undoped CL-PUA has a crystalline endothermic peak at 46.5 8C and a low temperature endothermic transition at 65.4 8C, corresponding to the melting temperature (Tm) and Tg of SS, Table 1 DSC data for CL-PUA with different LiClO4 concentrations Film
Tg of SS (8C)
DTg/DC (8C/(mM g))
Tg of HS (8C)
Tm (8C)
0 a b c d e f
65.4 60.2 51.4 39.8 32.3 10.2 0.2
– 15.4 21.8 25.5 27.4 32.4 24.2
+27.5 +30.2 +38.6 +48.1 +30.1 +28.3 +27.6
46.5 – – – – – –
Film of CL-PUA doped with different concentrations of LiClO4: (0) undoped CL-PUA, (a) 0.1 mM, (b) 0.2 mM, (c) 0.3 mM, (d) 0.4 mM, (e) 0.5 mM and (f) 0.6 mM LiClO4 doped CL-PUA films.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1027
respectively. Tg of SS increases with the concentration of LiClO4 added to that. The increase in Tg of SS is due to the partial arresting of local motion of the polymer segments through the formation of transient cross-links between Li+ ions and groups in CL-PUA. Otherwise, polyether chains become stiffened due to the ion–dipole interaction between the Li+ ions and oxygen atoms in polyether units. The change in Tg with concentration of LiClO4 (DTg/DC) increases with LiClO4 concentration and reaches a maximum at 0.5 mM LiClO4 and thereafter decreases (Table 1 and Fig. 1B). The plasticizing effect created by the formation of charge neutral contact ion pairs at higher LiClO4 concentration causes this trend [20]. The neutral contact ion pairs do not have tendency to form cross-links and hence cannot contribute to the increase in Tg of SS. The second thermal event occurring around 27.5 8C is attributed to an endothermic process of the hard segment (HS). The addition of LiClO4 to CL-PUA also influenced the Tg of HS. It is evident from Table 1 that Tg of HS tends to increase with the increase in LiClO4 concentration up to 0.3 mM and then decreases. The phase interchange mixing of hard and soft segments induced by Li+ ions may be the reason for the observation. The thermal transition corresponding to crystalline region of CL-PUA was absent. This informs that doped CL-PUA is transformed into amorphous upon the addition of LiClO4.
Fig. 1. (A) DSC thermograms for CL-PUA doped with various LiClO4 concentrations: (a) 0.1 mM, (b) 0.2 mM, (c) 0.3 mM, (d) 0.4 mM, (e) 0.5 mM and (f) 0.6 mM. (B) Variation of soft and hard segments Tg of CL-PUA with LiClO4 concentration.
1028
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
Fig. 2. IR absorbance in the –NH stretch region for: the undoped (a) and 0.2 mM LiClO4 doped CL-PUA (b) (inset shows the deconvolution of the – NH stretching region for undoped CL-PUA).
3.2. FT-IR spectroscopy FT-IR spectra inform that on adding LiClO4 to CL-PUA, Li+ ions interact with groups in CL-PUA. The results of FT-IR spectra were analyzed in the three frequency regions, including: (1) free and H-bonded –NH stretching mode, (2) urethane carbonyl symmetric stretching vibration and (3) C–O–C stretching vibrations of PEG, C(O)–O–C stretching vibrations of the hard segment and C(O)–O–C stretching vibrations of acrylate group. Fig. 2 represents the absorbance values in the –NH stretching frequency region. The changes in the relative absorbance values are analyzed for pristine CL-PUA and CL-PUA doped with different concentrations of LiClO4. The FT-IR spectra of –NH stretching frequency were deconvoluted and are presented in Table 2. Four regions were identified and assigned [21]. The peak in the region includes: (i) 3532 cm1 for free –NH and hydrogenbonded –NH stretching vibration, (ii) 3431 cm1 for carbonyl group, (iii) 3358 cm1 for H-bonded –NH stretching formed with the carbonyl and (iv) 3312 cm1 for the H-bonded –NH stretching formed with ether oxygen, respectively. Table 2 indicates that the frequency of the peak positions is lowered on adding LiClO4 to CL-PUA. The shifts in the peak positions are due to the interactions of Li+ ions with coordination sites in CL-PUA. A clear shift in the frequency of free –NH position to the lower frequency from 3532 to 3475 cm1 was noticed on adding LiClO4. This informs that Li+ ions are coordinated to the nitrogen atoms of the free –NH groups. The bands corresponding to the H-bonded –NH groups also showed shifting to lower frequency on the addition of LiClO4. Table 2 also presents the respective peak areas for the deconvoluted regions of –NH bands. Peak areas were normalized against the total area of the –NH band.
Table 2 Deconvolution data of FT-IR spectra of CL-PUA in the –NH stretching region Sample
0 a b c d
LiClO4 (mM)
– 0.1 0.2 0.3 0.4
Peak positions (cm1)
Percent area a
1
2
3
4
1
2
3
4
3532 3520 3509 3489 3475
3431 3430 3429 3430 3430
3358 3350 3341 3333 3325
3312 3300 3296 3281 3275
16.1 13.8 11.9 10.8 9.6
38.4 36.5 32.9 30.1 28.9
26.2 25.4 23.8 22.2 20.9
19.2 24.1 30.9 36.9 39.9
Peak 1: free –NH bonding. Peak 2: overtone of C O. Peak 3: hard–hard segment H-bonding. Peak 4: hard–soft segment H-bonding. a Peak areas are based on total –NH stretching band area.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1029
Fig. 3. IR absorbance in the C O stretch region for: the undoped (a) and 0.2 mM LiClO4 doped CL-PUA (b) (inset shows the deconvolution of the C O stretching region for undoped CL-PUA).
The values corresponding to –NH groups H-bonded to the ether oxygen are more for the LiClO4 added CL-PUA in comparison to pristine CL-PUA. In the presence of Li+ ions, the extend of –NH groups, H-bonded to the carbonyl oxygen, becomes higher. With the addition of LiClO4, the Li+ ions preferentially coordinate to the easily accessible ether groups and hence reduce the extent of –NH groups that are H-bonded to the ether oxygen. The carbonyl stretch region for the pristine CL-PUA and CL-PUA doped with 0.2 mM LiClO4 is presented in Fig. 3. The deconvoluted FT-IR spectra (Fig. 3) of doped samples are used to calculate the percent areas (Table 3) for the carbonyl stretching vibrations. On deconvolution, three modes of vibrations are identified: the free or non-Hbonded C O (1791–1769 cm1) and two stretching vibrations associated with disordered (1785 cm1) and ordered (1771 cm1) H-bonded C O [22]. The shifting in the frequencies of the theses bands to lower on addition of LiClO4 indicates the coordination of electron deficient Li+ ions with the carbonyl oxygen atoms. The trend of changes in the absorbance in the ether stretch region for the pristine CL-PUA and CL-PUA doped with 0.2 mM LiClO4 and the deconvoluted results for the four vibrational modes are presented in Fig. 4 and Table 4. (1) The peak in the region of 1089 cm1 is assigned to the C(O)–O–C stretch of the acrylate groups [23], (2) the second peak in the region of 1080–1045 cm1 is assigned to the C–O–C stretch of PEG, (3) the peak in the region of 1025–1005 cm1 is due to the C(O)–O–C stretch of the urethane groups and (4) the fourth one in the region of 992 cm1 is assigned to the H-bonded C–O–C stretch of PEG [22]. The acrylate C(O)–O–C stretching band does not show any change in its position upon the addition of LiClO4. The frequencies of C(O)–O–C stretch of the urethane groups and the C–O–C stretch show a marginal shift to lower frequency with the addition of lithium ions. A comparison of the percent area reveals that addition of LiClO4 released a proportion of H-bonded C–O–C groups.
Table 3 Deconvolution data of FT-IR spectra of CL-PUA in the C O stretching region Sample
0 a b c d
LiClO4 (mM)
– 0.1 0.2 0.3 0.4
Peak positions (cm1)
Percent areaa
1
2
3
1
2
3
1791 1786 1780 1776 1769
1785 1777 1771 1765 1754
1771 1760 1742 1731 1725
41.4 46.4 49.2 52.5 55.3
20.2 18.5 17.2 16.3 14.1
38.3 35.1 33.4 31.2 30.5
Peak 1: free carbonyl. Peak 2: disordered H-bonded carbonyl. Peak 3: ordered H-bonded carbonyl. a Peak areas are based on total C O stretching band area.
1030
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
Fig. 4. IR absorbance in the ether stretch region for: the undoped (a) and 0.2 mM LiClO4 doped CL-PUA (b).
3.3. Ionic conductivity The ionic conductivity obtained from AC impedance measurements as a function of LiClO4 concentration at 35 8C is shown in Fig. 5A. An increase in conductivity with LiClO4 concentration up to 0.5 mM was noticed. This observation is in accordance with the increase in the number of charge carriers and decrease in the free volume. The number of charge carriers increases with the concentration of LiClO4. A decrease in average free volume is expected as we have inferred with the increase in Tg (as a result of the interaction of Li+ ions with the ether oxygen) on addition of LiClO4. At low concentrations of LiClO4, the increase in the number of charge carriers dominates, whereas at high LiClO4 concentration, the reduction of free volume takes precedence [24]. When the LiClO4 concentration reaches 0.5 mM or more, the decrease in free volume becomes more pronounced than the increase in number of charge carriers. At this stage, the lower fraction of free volume may no longer compensate the increase in the number of charge carriers. As a result, a decrease in conductivity with increase in LiClO4 concentration was noticed at higher LiClO4 concentrations. Additionally, at higher LiClO4 concentrations, considerable amount of LiClO4 remains as ion pairs, which do not contribute to the conductivity of the electrolyte. In Fig. 5B, the ionic conductivity data for CL-PUA with various concentrations of LiClO4 are presented. A VTF relationship for ion transport could be seen for CL-PUA. It can be seen from Fig. 5 that the conductivity of sample e ([LiClO4] = 0.5 mM) is the highest among others at the temperature range studied. Also, sample e shows higher conductivity than sample f ([LiClO4] = 0.6 mM). The higher conductivity of sample e compared to sample f may be due to higher number of charge carriers at relatively higher LiClO4 concentration. The lower conductivity for sample f
Table 4 Deconvolution data of FT-IR spectra of CL-PUA in the ether stretching region Sample
0 a b c d
LiClO4 (mM)
– 0.1 0.2 0.3 0.4
Peak positions (cm1)
Percent areaa
1
2
3
4
1
2
3
4
1089 1085 1083 1085 1084
1080 1075 1065 1058 1045
1025 1020 1015 1011 1005
992 990 990 992 991
31.2 28.5 25.6 25.4 21.5
28.1 35.1 42.0 45.6 52.2
25.2 23.7 22.9 20.8 20.1
15.5 12.4 9.4 8.2 6.2
Peak 1: C(O)–O–C stretch of the acrylate. Peak 2: C–O–C stretch of PEG. Peak 3: C(O)–O–C stretch of the urethane. Peak 4: H-bonded C–O–C stretch of PEG. a Peak areas are based on total ether stretching band area.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1031
Fig. 5. (A) Variation of conductivity of CL-PUA electrolyte with LiClO4 concentration at 35 8C. (B) Temperature dependence of ionic conductivity of CL-PUA doped with different amounts of LiClO4: (a) 0.1 mM, (b) 0.2 mM, (c) 0.3 mM, (d) 0.4 mM, (e) 0.5 mM and (f) 0.6 mM.
in comparison to sample e may be attributed to the larger extends of ion pair formation with the increase in LiClO4 concentration. This observation is consistent with the results from DSC (Table 1) studies. 3.4. Cyclic voltammetry studies Cyclic voltammetry was performed to study the kinetics of Li deposition–stripping process of the CL-PUA electrolyte. Fig. 6 shows the repetitive cyclic voltammograms for CL-PUA at a sweep rate of 5 mV/s. The
Fig. 6. Cyclic voltammogram of CL-PUA.
1032
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
Fig. 7. Current relaxation plot for transference number measurement of CL-PUA (inset: complex impedance plot; initial resistance and steady state resistance measured before and after the current transient).
voltammetric curves indicate that the process at the lithium interface is basically reversible in the CL-PUA medium and the cyclability is good with the high recovery of lithium. On a cathodic sweep, a peak at approximately 0.32 V versus Li was observed which corresponds to the deposition of lithium on the SS electrode. At the same time, stripping of Li from the SS electrode could be noticed at 0.12 V. The peak current values for the lithium stripping process decreases upon cycling. This may be due to the formation of a passive layer on the surface of SS electrode. 3.5. Transport number Cation transport number, t+, in the sample electrolytes was determined by the application of 10 mV DC potential across the test cell (Li/CL-PUA/Li). The current decays immediately and asymptotically approaches steady state (Fig. 7). The decrease in current may be due to the growth of passivating layers at the electrode [25]. A stable current response was noticed after 2 h. The impedance response of cell (Fig. 7, inset) was monitored at an initial time and at a steady state current condition. The t+ value of CL-PUA was determined to be 0.54. The reported t+ values for SPEs range from 0.06 to 0.2 [26]. For a gel polymer systems, t+ values of order 0.4–0.5 have been found for poly(bis methoxy ethoxy ethoxy) phosphazene by Abraham and Alamgir [27] and Matsuda et al. reported that value of t+ = 0.56 for PEO–PMMA systems [28]. However, t+ of 0.54 in this case is lower than that of PAN based (0.6–0.8) [29] or PMMA based gel electrolyte (0.5–0.7) [30]. In the present case, the t+ of about 0.54 for CL-PUA systems is an acceptable one for a SPE. 3.6. Impedance spectra The impedance response of the cell using CL-PUA as polymer electrolyte at various times is shown in Fig. 8. Fig. 8 reveals that the curves are similar in shape with arcs in high frequency region and straight lines in low frequency region. The diameter of the semicircle increases as a function of time (Fig. 8, inset). There may be two possible explanations for this occurrence. Lithium may have reaction with deleterious moisture or with CL-PUA. The cell was fabricated in a glove box excluding the moisture. Hence, the variation in the resistance may be attributed to the growth of a passivating film on the Li electrode surface due to the reaction of Li metal with CL-PUA. Further, it is to be noted that the increase in interfacial resistance, Ri, of the CL-PUA electrolyte was gradual and occurred up to a storage time of 20 days. Afterwards, no change in Ri beyond 20 days was noticed. The steady increase in Ri with storage time may be due to the possible stabilizing role of the acrylate component in CL-PUA. Acrylate can impart good adhesiveness to solid electrolytes while making them stable to atmosphere moisture [31,32]. Also, the presence of acrylate groups helps to reduce the effects of moisture on the electrolyte and to retard the growth of passivation film between the electrolyte and the lithium electrode.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1033
Fig. 8. Nyquist plot for the cell having CL-PUA as polymer electrolyte measured at 2, 5, 10, 15, 17 and 20 days.
3.7. Electrochemical stability The electrochemical stability window of CL-PUA electrolyte is determined by linear sweep voltammetry [33]. The onset of the current in the anodic high voltage is resulted from a decomposition process associated with the electrode [34] and this onset voltage is taken as the upper limit of the electrolyte stability. Fig. 9 shows the current–voltage curve of CL-PUA electrolyte. The current onset occurs at 5.5 V versus Li, suggesting the high anodic stability of the electrolyte (Fig. 9). The current increased steeply as the applied voltage increased above this cell voltage (5.5 V). The low residual current level up to 5.5 V with the absence of any peak on the low voltage range confirms the purity of the electrolyte. In particular, it is relevant to note that there is no peak at 4.25 V versus Li, indicating the absence of moisture or water in the electrolyte [35].
3.8. Mechanical property The stress–strain curves of CL-PUA and simple PU are presented in Fig. 10. Tensile strength (stress) as high as 3 103 kPa at an elongation-at-break value (strain) of 58% was observed in the case of PU electrolyte, whereas strain value of 80% was observed in the case of CL-PUA electrolyte. The high elastic property of CL-PUA over simple PU is attributed to the effect of cross-linked network structure in CL-PUA.
Fig. 9. Electrochemical stability of the CL-PUA performed by linear sweep voltammetry.
1034
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
Fig. 10. Tensile strength behavior of simple PU and CL-PUA films (modulus 500 kPa).
3.9. Thermal analysis TGA curves for the undoped and doped CL-PUA are presented in Fig. 11. Pure CL-PUA decomposes in a single step beginning at 452 8C (Fig. 11). The introduction of LiClO4 promotes the decomposition of the backbone structure of CL-PUA at lower temperature. A decreasing trend in the decomposition temperature for the samples with increasing LiClO4 concentration can be seen (inset of Fig. 11). This behavior may be explained on the basis of weakening of the C O bond, caused by the decrease in electron density due to the interaction of Li+ ions with the oxygen atoms [19]. Alternatively, perchlorate oxidation of PEG phase can cause this thermal event to start from 400 8C. The observed trend in thermal transitions at higher LiClO4 concentration may arise from the predominant perchlorate oxidation.
Fig. 11. Thermograms of CL-PUA containing different amounts of LiClO4: (a) 0.1 mM, (b) 0.2 mM, (c) 0.3 mM, (d) 0.4 mM, (e) 0.5 mM and (f) 0.6 mM.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
1035
Fig. 12. Temperature dependence of ionic conductivity of gelled CL-PUA doped with 0.5 mM LiClO4.
Fig. 13. (A) The charge–discharge performance of Li/CL-PUA/LiCoO2 and Li/(gelled CL-PUA)/LiCoO2. (B) Rate capacity profile of Li/CL-PUA/ LiCoO2 cell at various rates.
1036
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037
3.10. Gel electrolytes As the conductivity of as synthesized CL-PUA was comparatively lower (108 S/cm) in its solid state, gel polymer electrolyte was prepared by loading the plasticizer (LP 20). Swelling of the electrolyte was continued until the swollen percentage reached the required value. The dimensional stability of the aforementioned GPE was found to be very good for practical use. The temperature dependence of the conductivity for the gelled CL-PUA is presented in Fig. 12. The conductivity of gelled CL-PUA follows Arrhenius relationship. A comparison of the conductivities of CLPUA and gelled CL-PUA with an addition of similar concentration of LiClO4 was made. Higher conductivity was noticed for gelled CL-PUA over CL-PUA. 3.11. Charge–discharge cycling tests In order to investigate the utility of the CL-PUA as polymer electrolyte, a cell consisting of Li metal as anode, LiCoO2 as cathode and CL-PUA as electrolyte was constructed. The charge–discharge performance of Li/CL-PUA/ LiCoO2 and Li/(gelled CL-PUA)/LiCoO2 cells at the first cycle is given in Fig. 13A. The cell at C/10 rate achieves a capacity of 101 mAh/g. The lower capacity at C/10 rate was due to the low ionic conductivity of the CL-PUA. A comparison among the discharge curves of both electrolytes (CL-PUA and gelled CL-PUA) reveals that the capacity of CL-PUA electrolyte is slightly lower than that of gelled CL-PUA. The reduced capacity is due to the lower diffusion rate of lithium ions in the CL-PUA as compared with its gel counterpart. The rated capacity of Li/LiCoO2 cell using CL-PUA is shown in Fig. 13B. The Li/LiCoO2 cell at C/10 rate reached close to 93% normal capacity. The reduced capacity at high rate is due to the lowering of diffusion coefficient of lithium ions in the lattice of LiCoO2. The diffusion rate of lithium ions in the CL-PUA is lower in comparison with that in gelled CL-PUA. At the C/5 and C/1 rates, the cell delivered about 83 and 67% of the full capacity at an average load voltage of 3.5 V. 4. Conclusions A cross-linked polyurethane acrylate was prepared by end capping 2,6-toluene diisocyanate/poly(ethylene glycol) based prepolymer with hydroxybutyl methacrylate. Significant interactions between lithium ions and soft/hard segments of the CL-PUA were observed, when complexed with LiClO4. The addition of HBMA into PU matrix improves tensile strength of CL-PUA. CL-PUA showed high electrochemical stability window and better compatibility with lithium metal. Electrochemical characteristics inform that the CL-PUA can be used as polymer electrolyte in rechargeable lithium batteries. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
M. Armand, J.M. Chabagno, M. Duclot, Fast Ion Transport in Solids, Amsterdam, North-Holland, 1979, p. 131. H.S. Lee, X.Q. Yang, J. McBreen, Z.S. Xu, T.A. Skotheim, Y. Okamoto, J. Electrochem. Soc. 141 (1994) 886. J. Barthel, R. Buestrich, E. Carl, H.J. Gores, J. Electrochem. Soc. 143 (1996) 3565. I. Nicotera, L. Coppola, C. Oliviero, A. Russo, G.A. Ranieri, Solid State Ionics 167 (2004) 213. H.S. Min, J.M. Ko, D.W. Kim, J. Power Sources 119–121 (2003) 469. S.W. Choi, S.M. Jo, W.S. Lee, Y.R. Kim, Adv. Mater. 15 (2003) 2027. J.R. Kim, S.W. Choi, S.M. Jo, W.S. Lee, B.C. Kim, Electrochim. Acta 50 (2004) 69. J. Vondrak, J. Reiter, J. Velicka, B. Klapste, M. Sedlarikova, J. Dvorak, J. Power Sources 146 (2005) 436. A.K. Hjelm, T. Eriksson, G. Lindbergh, Electrochim. Acta 48 (2002) 171. N. Yoshimoto, H. Nomura, T. Shirai, M. Ishikawa, M. Morita, Electrochim. Acta 50 (2004) 275. H.L. Wang, H.M. Kao, M. Digar, T.C. Wen, Macromolecules 34 (2001) 529. H.L. Wang, H.M. Kao, T.C. Wen, Macromolecules 33 (2000) 6910. W.C. Chen, H.H. Chen, T.C. Wen, M. Digar, A. Gopalan, J. Appl. Polym. Sci. 91 (2004) 1154. A. Pattanayak, S.C. Jana, Polymer 46 (2005) 5183. C.S. Kim, B.H. Kim, K. Kim, J. Power Sources 84 (1999) 12. S. Velankar, J. Pazos, S.L. Copper, J. Appl. Polym. Sci. 62 (1996) 1361. J. Sun, D.R. MacFarlane, M. Forsyth, Solid State Ionics 85 (1996) 137. M. Digar, S.L. Hung, H.L. Wang, T.C. Wen, A. Gopalan, Polymer 43 (2002) 681. M. Digar, S.L. Hung, H.L. Wang, T.C. Wen, A. Gopalan, Polymer 43 (2002) 1615.
P. Santhosh et al. / Materials Research Bulletin 41 (2006) 1023–1037 [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]
S. Schantz, L.M. Torell, J.R. Stevens, J. Chem. Phys. 94 (1991) 6862. M. Solomon, M. Xu, E.M. Eyring, S. Petrucci, J. Phys. Chem. 98 (1994) 8234. J.D. van Heumen, J.R. Stevens, Macromolecules 28 (1995) 4268. C.J. Pouchert (Ed.), The Aldrich Library of Infra-Red Spectra1997. Z.L. Peng, B. Wang, S.Q. Li, S.J. Wang, H. Liu, H.Q. Xie, Phys. Lett. A 194 (1994) 228. P.G. Bruce, C.A. Vincent, J. Electroanal. Chem. 225 (1987) 1. J.B. Goodenough, Solid State Ionics 69 (1994) 184. K.M. Abraham, A. Alamgir, Chem. Mater. 3 (1991) 339. Y. Matsuda, M. Morita, H. Tsutsumi, Polym. Adv. Technol. 4 (1993) 209. F. Croce, F. Gerance, G. Dautzenberg, S. Passerini, G.B. Appetecchi, B. Scrosati, Electrochim. Acta 39 (1994) 2187. G.B. Appetecchi, F. Croce, B. Scrosati, Electrochim. Acta 40 (1995) 991. T. Mani, R. Mani, J.R. Stevens, Solid State Ionics 60 (1993) 113. R. Mani, T. Mani, J.R. Stevens, J. Polym. Sci., Part A: Polym. Chem. 30 (1992) 2025. G.B. Appetecchi, F. Croce, B. Scrosati, J. Power Sources 66 (1997) 77. S. Slane, N. Salomon, J. Power Sources 55 (1995) 7. G.B. Appetecchi, F. Croce, A. D Paolis, B. Scrosati, J. Electroanal. Chem. 463 (1999) 248.
1037
