Abstract
In this study, new types of gel polymer blend electrolytes (GPBEs) were prepared with the synthesized PVC-g-PMMA graft copolymer, PMMA, plasticizers (propylene carbonate (PC), dioctyl phthalate (DOP)), and different loadings of Mg(ClO4)2 via the solution casting method using tetrahydrofuran as solvent. Fourier transform infrared (FTIR) spectra of the electrolytes showed mutual molecular interactions between Mg(ClO4)2 and organic moieties. The scanning electron microscopy images of the GPBEs showed their wrinkled surface morphology due to their low elastic modulus and high flexibility. Energy-dispersive X-ray (EDX) spectroscopy and mapping technique revealed the regular distributions of all atomic elements such as Cl, Mg, O, and C in the doped GPBEs. With increasing the Mg salt concentration, Young’s modulus and tensile strength of the GPBEs strongly decreased. Interestingly, the elongation at break of the GPBEs was higher than that of neat (undoped) GPBE and achieved the highest value of 215% at the salt content of 20 wt.%. The AC conductivity and ionic conductivity, as well as dielectric permittivity of plasticized PMMA/PVC-g-PMMA/Mg(ClO4)2 GPBE,s increased with frequency and Mg(ClO4)2 doping content. Ionic conductivity of the doped GPBEs can be achieved from to (S.cm-1) using Mg(ClO4)2 contents in the range from 10 to 40 wt.%. The doped GPBEs are thermally stable up to 100°C with very low weight losses. The GPBE doped with 20 wt.% of Mg(ClO4)2 can be used as a new type of electrolyte for developing Mg batteries.
1. Introduction
Nowadays, lithium batteries are widely used for energy storage devices in smart phones, tablets/laptops, electric vehicles, etc. [1, 2]. However, fire and explosion accidents of lithium batteries have occasionally occurred worldwide, some of which caused serious threats to user’s health [3–5]. The reasons for these accidents are known as the use of liquid electrolytes, the poor thermal stability of lithium salts, and the formation of oxygen in charging/recharging at high temperatures [3]. Therefore, it is necessary to develop new types of ionic batteries with low toxicity, greater safety, and lower cost than lithium batteries [5–11]. Over the past decades, gel polymer electrolytes have been extensively studied due to their high compliance, processability, mouldability, good electrode-electrolyte contact, and high ambient temperature conductivity. Besides, the increasing attention has been paid to the development of magnesium ion cells/batteries with the high specific capacity, efficiency, and good cyclability [12, 13]. Gel polymer electrolytes can be produced when the polymer swells up in organic solvents or plasticizers, which can provide better contact with the electrode surface than the dry solid polymer electrolytes (free of solvents/plasticizers) [13].
Poly(vinyl chloride) (PVC) and their blends have been employed which used as gel polymer electrolytes [14–20]. Several studies have shown that the tensile properties, [15] thermal stability, [16] ionic conductivity, and [15] charge-discharge properties of PVC can be improved by the blending with poly(methyl methacrylate) (PMMA) or using graft copolymer-like PVC-graft-PMMA and plasticizers to form gel polymer blend (GPB) electrolytes [14, 20].
In our previous study, we have reported the mechanical and dielectrical properties of GPBEs doped with Mg(ClO4)2 salt [21]. It was found that plasticizers are very important to improve the ionic conductivity of the GPBEs. However, it is necessary to enhance the mechanical and dielectrical properties for the GPBEs to expand the applicability of the materials. In this study, a graft copolymer named PVC-g-PMMA (PVCg) was synthesized and mixed with PMMA and plasticizers to prepare PMMA/PVCg gel polymer blend electrolytes. The graft copolymer was expected as a compatibilizer for improving mechanical and electrical properties of the GPBEs. The effects of magnesium perchlorate on these properties and morphology of the electrolytes were also investigated.
2. Experimental
2.1. Materials
Polyvinyl chloride (PVC, SG-660, k index of 65-67) is a commercial product of Plastic & Chemical Corp., Ltd. (TPC Vina, Vietnam). Methyl methacrylate (MMA, 99%), poly(methyl methacrylate) (PMMA, weight average molecular weight of 120000), and propylene carbonate (PC, 99.7%) were purchased from Aldrich (USA). Tetrahydrofuran (THF, 99.7%), chloroform (CH3Cl, 99.5%), and Mg(ClO4)2 were reagent grade products of Xilong Co., Ltd., China. Ethyl alcohol (ethanol, 99.7%), methyl alcohol (methanol, 99.7%), and dioctyl phthalate (DOP, 95%) were provided by Duc Giang Chemical and Detergent powder Joint Stock Company, Vietnam. Chemicals were of analytical grade and used as received.
2.2. Preparation of Electrolyte Films
Graft copolymer of PVC-g-PMMA (labeled as PVCg) with the PMMA graft content of 20.3 wt% had been prepared as stated in the previous study [22]. As similarly, the weight ratio of PVC : PMMA was kept as 1 : 1 in this study. Therefore, the compositions of GPBEs were prepared as shown in Table 1, where the weight of PMMA in PVC-g-PMMA was calculated of 0.115 g and that of PVC was 0.45 g, or PMMA : PVC weight ratio of 1 : 1. In an example of preparation of an electrolyte film, PMMA, PVCg, plasticizers, and Mg(ClO4)2 were charged in a flask containing THF. The mixture was stirred at 40°C for 4 hours to obtain a homogenous solution which was then poured into a (cm) release paper. The THF solvent was allowed to evaporate naturally in a fume hood for 24 h, then complete drying in vacuum oven at 40°C for 24 h. The collected film was stored in a desiccator for at least 24 h before characterizations. The compositions and sample labels are described in Table 1.
2.3. Characterization
FTIR spectra of all samples were performed on a Fourier transform infrared spectrometer (Nicolet/Nexus 670, USA) with 32 scans, 4 cm-1 resolution, and in wave number ranging from 400 to 4000 cm-1 at room temperature.
The tensile properties of the electrolyte samples were conducted on a universal testing machine (Zwick V.2.5, Germany) with a crosshead speed of 50 mm/min, in accordance with ASTM D882 for thin plastic films.
Complex dielectric properties (real part— [] and imaginary part— []) of electrolyte films were measured by using an Agilent E4980A instrument at - impedance parameters using the 16451B test fixture, in the frequency range of 25 Hz-1 MHz (according to the limit of the Agilent E4980A instrument), with a peak amplitude of 1.0 V and applied voltage to the ground electrode of 0 V (Earth ground), at room temperature (24-28°C), as per ASTM D150 standard. Electrolyte films were sandwiched between stainless steel (SS) electrodes as the SS/GPBEs/SS configuration or Mg electrodes as the SS/Mg/GPBEs/Mg/SS configuration. Mg electrodes with a diameter of 5 mm were cut from the Mg sheet.
AC conductivity () and dielectric constant () of electrolyte films were determined from measured - values (capacitive reactance and resistance of complex impedance) as the equations at every frequency (): [23–25]. where is the thickness of the electrolyte film, is the (contact) area of the working electrode (with 0.005 m in diameter), is the frequency, is the absolute value of (measured capacitive reactance), is the measured resistance component, and is the vacuum permittivity ( F/m).
Morphology, element composition, and elemental mapping of the samples were analyzed by scanning electron microscopy (SEM) combined with energy dispersive X-ray (EDX) spectroscopy (JSM-6510LV SEM instrument, Jeol, Japan).
Thermal Gravimetric Analysis (TGA) was carried out in a NETZSCH TG 209F1 (Germany) instrument to study the thermal degradation properties of the electrolyte under nitrogen gas with a flow rate of 40 ml/min, from room temperature to 650°C with a heating rate of 10°C/min, and specimen weight of about 6-7 mg. All above measurements were carried out in laboratory rooms with relative humidity of 40-50%, and the sample was measured right after removed from the vacuum environment.
3. Results and Discussion
3.1. FTIR Spectra
Figure 1 displays the FTIR spectra of the raw materials and synthesized samples. Referred to previous studies [22], it is very easy to assigned all absorption peaks for specific groups of PMMA (Figure 1(a)), PVCg (Figure 1(b)), and GPBE (Figure 1(c)). For more details, Table 2 lists some main band assignments for PMMA, PVC, and GBPE samples without and with magnesium perchlorate salt (20, 40 wt.%). Comparing with the spectra of neat PMMA and PVCg, the spectrum of the GPBE.0 sample (Figure 1(d)) appears some new peaks attributed to the presence of plasticizers, the peaks at 1805 cm-1 and 1732 cm-1 can be assigned for C=O stretching vibrations () of the PC plasticizer and PMMA, and the doublet at 1600 and 1580 cm-1 is assigned for aromatic ring of the DOP plasticizer (Table 2 column 5), whereas comparing with spectrum of the GPBE.0 sample, the FTIR spectra of GPBEs doped with different weight amounts of Mg(ClO4)2 (Figures 1(e) and 1(f)) show some differences. The first is the strong absorption band of (OH) in the wavenumber region from 3700 to 3100cm-1 due to the moisture absorption of Mg(ClO4)2 from the air during FTIR recording. The second difference is the disappearance of three peaks (651, 637, and 616 cm-1), instead of only one peak at 627 cm-1 which appeared in the region of 660-600 cm-1. In addition, the band attributed to C=O of PC appearing at 1808 cm-1 in spectrum of GPBE0 is shifted to 1780 and 1770 cm-1 in spectrum of GPBE.20 and GPBE.40, respectively. This shift increases with increasing Mg(ClO4)2 content, indicating the strong interactions between Mg(ClO4)2 and organic moieties (polymers and plasticizers) in the electrolytes.
3.2. Surface Morphology of Electrolyte Films
Figure 2 represents the SEM images of GPBE.0, GPBE.10, and GPBE.20 films. It can be seen that there are wrinkles on the surfaces of two kinds of samples (undoped and doped with magnesium salt), which are caused by the high flexibility of the films containing high plasticizer content. It was reported that PMMA/PVC blend films have craters or islands due to the immiscibility of ethylene/propylene carbonate plasticizer with PVC, which caused the separation of the PVC-rich phase and plasticizer-rich phase [26]. In order to improve the miscibility of plasticizer with the PMMA/PVC blend, ethylene carbonate was replaced by DOP which is known as one of the best plasticizer for PVC; moreover, PVC-g-PMMA graft copolymer was also employed in this study. As a result, the SEM images of new GPBEs clearly reveal that no craters or islands were observed on the surface of GPBEs. In addition, it is also interesting to see that the number of wrinkles on the surface of the Mg(ClO4)2-doped electrolyte film decreases with increasing Mg salt content.
(a)
(b)
(c)
(d)
To clarify the phase compositions at the wrinkles and the surface of the GPBE samples, elemental mapping of the GPBE.20 sample at its surface was conducted. In comparison with the original SEM image (Figure 2(d)) that was selected for EDS analysis, Figure 3(a) shows that the main chemical elements of GPBE.20 including Mg, C, O, and Cl are all appeared and regularly distributed in its layered EDS mapping image, even at the wrinkles. Other EDS mapping images of single elements such as Cl, Mg, and O also confirm their regular distributions in the GPBE.20 sample. In other words, the wrinkles are not the interphase, and there is not any phase separation in the electrolyte sample. It means that using PVC-g-PMMA and DOP can improve the compatibility for the PMMA/PVC blend (50 : 50 wt./wt.).
(a)
(b)
(c)
(d)
Figure 4 displays a selected EDX spectrum of the GPBE.20 electrolyte film and the composition table result. The chemical elements of the electrolyte were detected with relevant peaks labeled with C, O, Mg, and Cl. It is noted that the unlabeled peak (between Mg and Cl peaks) is of Pt coat and not taken into account. The table result (inside Figure 4) verifies that magnesium atom appears on the surface of the electrolyte films (1.75wt.%) with lower concentration than those in the bulk (mathematically evaluated as 2.31 wt.%).
3.3. Tensile Properties
Table 3 shows that Young’s modulus of GPBE films strongly decreases from 2.75 to 0.65 MPa with increasing Mg(ClO4)2 contents from 0 to 40 wt.%, and their tensile strength also decreases from 1.72 to 0.84 MPa. In fact, these free-standing films can be commonly mounted in the tensile test machine. The elongation at break of Mg(ClO4)2-doped GPBEs is higher than that of the undoped GPBEs and reaches the highest value of 215% with 20 wt.%-doped content. In comparison with the previous study, tensile strength and modulus of present GPBEs (using PVC-g-PMMA) are several folders higher than plasticized PMMA/PVC/Mg(ClO4)2 electrolytes with equal salt contents. It demonstrates that the PVC-g-PMMA graft copolymer has played an important role as a compatibilizer to increase the adhesion between polymers and plasticizers.
3.4. Dielectric Properties
Figure 5 displays logarithm plots of AC conductivity and dielectric permittivity as a function of frequency (from 25 Hz to 1 MHz) for the GPBEs with different doping contents of Mg(ClO4)2. Figure 5(a) indicates that AC conductivity of all doped GPBEs increases with frequency and with magnesium salt concentration. The AC conductivities of doped GPBEs are several orders higher than undoped GPBE (GPBE.0). For example, the AC conductivity at 1 kHz of neat GPBE.0 is about S/cm. However, the values of GPBE.10, GPBE.20, GPBE.30, and GPBE.40 samples are of , , , and S/cm, respectively. Figure 5(b) shows the strongly increase of dielectric permittivity of Mg(ClO4)2-doped electrolytes with the salt concentration. It should be noted that the weights of polymers and plasticizers (Table 3) are kept the same for all GPBEs. The enhancements in AC conductivity and dielectric permittivity are only related to the increase of the ion carrier concentration, which are generated by the electrolysis of the magnesium salt in the GPBEs.
(a)
(b)
Figures 6 and 7 display the Nyquist complex impedance plots of GPBEs with different doping contents of Mg(ClO4)2 using SS/GPBE/SS and SS/Mg/GPBE/Mg/SS electrode configurations, in the range from 25 Hz to 1 MHz at room temperature. The impedance plots exhibit two regions, a part of semicircle and a tail, which could be modelled to the equivalent circuit drawn inside Figure 6(a), (where is a bulk resistance, is a double-layer capacitor, is a charge transfer resistance, and is an interfacial impedance of GPBEs) [23]. Although the first region is not a full semicircle, when completed by fitting, the resistances (, ) can be evaluated via the intercepts of the semicircle with the real part axis at low- and high-frequency regions, respectively. It is easy to recognize that the diameters of the fitted semicircles (~2 resistances) are reduced by increasing Mg(ClO4)2 doping contents.
(a)
(b)
For more details, Table 4 shows the evaluated values of and of GPBE films from Figures 6 and 7. For the SS/GPBE/SS configuration, obtained results indicate that and of electrolyte samples decrease with increasing salt content from 10 to 40 wt.%. Based on equation (1), the ionic conductivity of GPBEs doped with Mg(ClO4)2 (GPBE.10, 20, 30, 40) increases from to (S.cm-1). With the SS/Mg/GPBE/Mg/SS configuration, obtained results also show a similar decrease of and of electrolyte samples with increasing the salt content. However, the lower value of for the SS/Mg/GPBE configuration at the same content of Mg(ClO4)2 confirmed that equilibrium was established between Mg metal and Mg2+ ions in the GPPE [27]. Table 4 also shows that the intercept () values using the SS/Mg/GPBE configuration are higher than using for the GPBE samples with the SS/GPBE configuration.
3.5. Thermal Stability
Figure 8 illustrates the TGA diagrams of the undoped (or neat) GBPE and the GPBEs that were doped with magnesium perchlorate salt at different loadings. Table 5 displays the thermo analysis results from the above TGA diagrams. Figure 8 and Table 5 clearly present some differences between the undoped GPBE sample (GPBE.0) and doped GBPE samples (GPBE.10, 20, 30, 40); in other words, magnesium perchlorate affects the thermal stability of the neat GPBE. In the temperature region below 200°C, the TGA line of the GPBE.0 sample keeps almost constant, which indicates it is thermo-stable in this temperature range. Meanwhile, the TGA curves of doped GPBEs show the first weight loss in this temperature region, due to the removal of adsorbed water [28]. Table 5 shows that the weight loss at 150°C of the doped GPBEs increases with increasing magnesium salt contents. This may result from the present of Mg(ClO4)2 as a weak Lewis acid property which can accelerate the degradation of PVC. However, with temperature lower than 100°C, the GPBE.10 and GBPE.20 samples are thermally stable with low weight losses, e.g., 1.65 wt.% and 2.72 wt.%, respectively.
In the temperature region from 200 to 650°C, both two types of undoped and doped electrolyte samples undergo thermal degradation of organic moieties and decomposition of magnesium salt. In this temperature range, the GPBE.0 sample undergoes degradation with two distinct stages, the first stage starts from 220°C to 330°C with the onset temperature of 277.4°C, and this can be attributed to the dehydrochlorination of PVC and unsaturated groups of PMMA [16, 29, 30]. The second stage starts from 330 to 650°C with char residual of about 8 wt.%. In 200-280°C range, the TGA curves of doped electrolyte samples also show their second stages of thermal degradation, in which their curves shift to lower temperature compared to that of the GPBE.0 sample. This means that magnesium perchlorate salt strongly reduces the thermal stability of the plasticized PVCg/PMMA blend. Table 5 shows that the onset temperatures of doped electrolyte samples strongly reduce when increasing the salt contents. This phenomenon can be due to the decomposition of perchlorate groups that has generated the oxide atoms which accelerate the oxidation and dehydrochlorination of PVC as well as the degradation of other organic substances in the GPBEs.
It should be noted that stabilizers had not been mixed with PVC while preparing plasticized PVC-g-PMMA/PMMA blend electrolytes in order to investigate the sole effect of magnesium perchlorate on its thermal stability and other electrical properties. This suggests that stabilizers for PVC should be used in order to limit the thermal stability of the Mg(ClO4)2-doped GPBEs.
4. Conclusion
Mg(ClO4)2-doped GPBEs have been prepared by using the solution casting method. The obtained results showed that PVC-g-PMMA played an important role as the compatibilizer for enhancing mechanical and dielectrical properties of GPBEs. Some observable changes in FTIR spectra of GPBEs indicate the molecular interactions between the salt and plasticizers/polymers in the GPBEs. SEM images and EDX mapping technique demonstrated that Mg(ClO4)2 regularly distributed into the plasticizers/polymers system. Young’s modulus and tensile strength of GPBEs strongly reduced with the loading Mg(ClO4)2 content. The elongation at break of the doped GPBEs was higher than the undoped GPBE and reached the highest value of 215% at the doping content of 20 wt.%. The AC and ionic conductivities of GPBEs increased with the loading Mg(ClO4)2 content. Magnesium perchlorate reduced the thermal stability of GPBE films; however, the doped GPBEs are thermally stable with low weight losses at temperature lower than 100°C. Stabilizers and anhydrous magnesium perchlorate and water-free media should be used when preparing GBPE films for the next studies. In overall consideration of mechanical, dielectrical properties and thermal stability of the GPBEs, the plasticized PMMA/PVC-g-PMMA with 20 wt.% Mg(ClO4)2 can be chosen as a best candidate of electrolyte for magnesium batteries with working temperature lower than 100°C.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
All authors declare that they have no conflicts of interest.
Acknowledgments
This work is completed under the financial support from the Vietnam Academy of Science and Technology (VAST) under grant number VAST03.05/20-21.