Norbornene-Functionalized Plant Oils for Biobased Thermoset Films and Binders of Silicon-Graphite Composite Electrodes

Abstract

We herein report the functionalization of plant oil with norbornene (NB) and subsequent polymerization to prepare biobased thermoset films and biobased binders for silicon/mesocarbon microbead (MCMB) composite electrodes for use in lithium-ion batteries. A series of NB-functionalized plant oils were prepared as biobased thermoset films via ring-opening metathesis polymerization (ROMP) in the presence of a second-generation Grubbs catalyst with tunable thermomechanical properties. Increasing the catalyst loading and cross-linking agent increased cross-link density, storage modulus (E′), and glass transition temperature (T_g), while the numbers of unreacted or oligomeric components in the films were reduced. High number of NB rings per triglyceride in the plant oil encouraged monomer incorporation to form a polymer network, therefore accounting for the high T_g and E′ values. Furthermore, the NB-functionalized plant oil and 2,5-norbornadiene (NBD) were copolymerized as bioderived binders for silicone/MCMB composite electrodes of lithium-ion batteries via ROMP during electrode preparation. Cell performance investigation showed that the silicone/MCMB composite electrode bearing the NBD-cross-linked NB-functionalized plant oil binder exhibited a higher C-rate and cycle-life performance than that using a conventional poly(vinylidene fluoride) (PVDF) binder. Finally, the electrode based on the bioderived binder exhibited a high specific charge capacity of 620 mA h g^–1 at 0.5 C.

Introduction

Concerns regarding the depletion of fossil resources in addition to ongoing environmental issues are driving demand for the replacement of unsustainable petroleum-based materials. As such, the utilization of renewable resources for the production of polymeric materials is becoming increasingly important from both the economic and environmental viewpoints. A wide range of biorenewable polymers has therefore been developed from renewable resources, such as plant oils, polysaccharides, lignin, sugars, cellulose, and rosin.^1,2 Among these, plant oils are considered to be the most promising starting material for the synthesis of polymers because of their inherent biodegradability and low toxicity. Indeed, the use of plant oils as renewable feedstocks for biobased thermosetting plastics has attracted both academic and commercial attention.

Thermoset polymers (thermosets) are commonly used as matrices in fiber-reinforced composites because of their high thermal and chemical resistances, as well as their good physical and mechanical properties. Biobased thermosets are commonly produced from plant oils using techniques including co-polymerization, free radical homopolymerization, and cationic polymerization. Olefin metathesis has also been employed to develop strong and tough thermosets, primarily through acyclic diene metathesis polymerization and ring-opening metathesis polymerization (ROMP). More specifically, three plant oil-based ROMP thermoset polymers have been produced by the copolymerization of norbornene (NB)-functionalized castor oil with cyclo-octene (CO),³ Dilulin (a commercial NB-functionalized linseed oil) with dicyclopentadiene (DCPD),⁴ and Dilulin with bicyclic NB-based cross-linkers (CLs) (bicyclo[2.2.1]hepta-2,5-diene)⁵ using a second-generation Grubbs catalyst. However, phase separation took place because of the differences in reactivity between the plant-oil-based monomers and the petroleum-based co-monomers, thereby compromising the resulting mechanical properties. Phase separation has also been observed when thermosetting of castor oil and palm oil was conducted using petroleum-based NB as a comonomer.^3,6 To address this issue, it is necessary to reduce the vegetable oil triglycerides to fatty alcohols, followed by functionalization with NB groups, such as 5-norbornene-2,3-dicarboxylate anhydride or 5-norbornene-2-carbonyl chloride.^7,8 These NB-functionalized biorenewable monomers readily undergo ROMP without requiring a comonomer, thereby enhancing the storage modulus (E′) and glass transition temperature (T_g) of the resultant thermosets.⁹

However, plant-oil-based ROMP thermosets tend to exhibit a low T_g and poor thermal properties. Their flexible triglyceride chains also render their mechanical strength weak at room temperature. Thus, to increase T_g and thermal properties, plant oil monomers are usually copolymerized with a petroleum-based NB monomer or CL such as DCPD,⁴ 5-ethylidene-2-norbornene,¹⁰ or NB (bicyclo[2.2.1]hept-2-ene),⁶ thereby promoting ROMP in the presence of Ru catalysts to give thermosets with good thermomechanical properties. However, their rigid structure, relatively low in situ polymerization rate, high-amount catalyst requirement, and nonrenewable characteristics restrict the use of petroleum-based CLs in thermosetting polymeric materials. Thus, to establish biobased polymers as industrial materials, it is necessary to explore novel NB-containing monomers. For example, monomers with a high reactivity and a low melting point can act as co-monomers or CLs for the nonrenewable NB (DCPD, NB, or CO) when preparing plant-oil-based thermosets via ROMP. Moreover, a cross-linking agent is normally required for the preparation of a thermally stable thermoset. For example, the use of naturally available molecules could be of interest, with one such example including isosorbide (IS), which is derived from glucose and is typically obtained from the depolymerization of cellulose or starch.^11,12 IS is a V-shaped diol with a unique bicyclic ring structure consisting of two cis-fused tetrahydrofuran rings and two secondary hydroxyl groups in the 2- and 5-positions,¹³ which impart a degree of rigidity to the resulting polymers. Its unique molecular structure, high thermal stability, biodegradability, and nontoxicity renders it a promising monomer and monomer precursor for the replacement of petroleum-based polymers.^14,15

To date, several applications have been reported for thermosets, ranging from basic to advanced materials. For example, their application in energy storage devices is of particular interest, where the thermoset could be used as a binder in battery electrodes. In this context, it should be noted that the key properties of binders for energy storage applications include good solvent resistance, high electrochemical stability, and appropriate flexibility.¹⁶ Conventional binders include fluorine-containing polymers, such as poly(vinylidene difluoride) (PVDF). However, because of environmental concerns, fluoro-free binders and biobased or bioderived polymers are gaining considerable attention.¹⁷

Thus, we herein report the incorporation of highly reactive NB functional groups into the IS polymer backbone. The resulting NB-functionalized IS (NB-IS) is then employed as a CL in the ROMP reaction of NB-functionalized plant oils. Three thermoset polymer films are synthesized via ROMP from olive oil (OO), rapeseed oil (RO), and soybean oil (SO), which differ in the amounts of polyunsaturated fatty acids present in their compositions. These monomers are subjected to ROMP without the use of a comonomer. The structures of the obtained thermosets are determined by nuclear magnetic resonance (NMR) spectroscopy and Fourier-transform infrared spectroscopy (FTIR), and their properties are characterized by dynamic mechanical analysis (DMA). The T_g, E′, and tangent (δ) peaks are used as metrics to probe structural network changes. In addition, the polymer thermal stability is determined using thermogravimetric analysis (TGA), and the thermomechanical properties of the prepared thermoset polymer films produced using different loadings of IS CL and different plant oils are compared. Furthermore, a comparison is made with films produced using a petroleum-based NB CL (1,4-bis[dimethyl[2-(5-norbornen-2-yl)ethyl]silyl]benzene). Finally, application of the NB-functionalized plant oil as a bio-based binder for Si/mesocarbon microbead (MCMB) composite anodes in lithium-ion batteries is also reported.

Results and Discussion

Structural Characterization

A biobased CL (NB-IS) was synthesized through the coupling of 5-norbornene-2-carbonyl chloride to hydroxyl (−OH) groups on IS. Three different NB-functionalized plant oil monomers were synthesized through epoxidation of the olefin groups and ring-opening of the epoxide groups (Scheme 1). The olefin groups in the plant oil triglycerides (represented by OO) were converted into oxirane rings using the in situ performic acid method. This produced EOO, as confirmed by ¹H NMR spectroscopy (Figure 1). Resonances ascribed to the olefin (a), inter-olefin (b), and α-olefin (d) protons (at 5.3, 2.8, and 2.0 ppm, respectively) disappeared and were replaced by resonances ascribed to protons originating from the oxirane and α-oxirane rings at 2.90–3.10 and 1.50 ppm, respectively. The absence of resonance at 5.3 ppm indicates that conversion of the double bond (C=C) was complete, thereby confirming the formation of EOO. Similar results were obtained for the epoxidized rapeseed oil (ERO) and epoxidized soybean oil (ESO). From the integrated intensities of the resonances in the ¹H NMR spectra, the average number of oxirane rings per EOO, ERO, and ESO unit were determined to be 1.3, 1.8, and 1.7, respectively. The characterization of the biobased CL and the modified vegetable oils were determined using ¹H NMR and FTIR spectroscopy, as shown in Figures S1–S4.

Scheme 1. Synthesis of the NB-Functionalized Plant Oils (Represented by NB-OO).

Figure 1.

¹H NMR spectra of OO, EOO, and NB-OO (in CDCl₃ at 25 °C). (Photographs of oil were taken with a digital camera).

The presence of the NB unit in the NB-OO monomer, resulting from ring-opening of the epoxide groups, was confirmed by ¹H NMR spectroscopic observations. Resonances attributed to the oxirane ring protons at 2.9–3.1 ppm reduced in intensity, while new resonances appeared at 5.9–6.3 ppm, corresponding to the NB protons. In addition, resonances ascribed to the proton of the carbon atom attached to the NB moiety (at 4.8 ppm) and to the proton of the carbon atom attached to the hydroxyl group (at 3.6 ppm) were observed, followed by the opening of the oxirane ring. The remaining resonances, which were attributed to the protons present in the glycerol unit (α and β), suggested that the triglyceride structure remained after the reaction. Successful synthesis of the monomer was demonstrated by ¹³C NMR spectroscopy (Figure S5), whereby the successful grafting of NB molecules onto the OO triglycerides was confirmed. Similar ¹³C NMR results were observed for NB-RO and NB-SO, as shown in Figures S6 and S7. From the integrated intensities of the ¹H NMR resonances, the average numbers of NB molecules per NB-OO, NB-RO, and NB-SO unit were determined to be 1.0, 1.3, and 1.5, respectively; this was related to the unsaturated fatty acid content of each plant oil. Successful functionalization of the OO was further confirmed by the FTIR spectra, as described in Figures S8–S10.

Preparation of a Renewable Thermoset with Tunable Thermomechanical Properties via ROMP

To study the tunable thermomechanical properties of NB-functionalized plant oils via ROMP, a number of process parameters were investigated, including the type of EPO, the CL, the catalyst loading, and the curing temperature.

ROMP of NB-Functionalized with Different Plant Oils

The obtained biobased thermoset films (thicknesses ∼ 1.0 ± 0.1 mm) were flexible and slightly transparent, with an amber-like color (Figure S11). The successful ROMP reaction was confirmed by FTIR, as shown in Figure S12. In addition, from Figure 2a, it was apparent that the percentage of the soluble material in NB-OO decreased from 60 to 9 wt % upon increasing the catalyst loading, and the increased ROMP reactivity resulted in the formation of a more cross-linked insoluble polymer network. Similar results were observed for NB-RO (53–4 wt %) and NB-SO (28–3 wt %) (Table S1). At 0.5 wt % G2, ∼9 wt % of the NB-OO film was soluble, while NB-RO and NB-SO afforded solubilities of only 4 and 3 wt %, respectively. This is likely because NB-OO bears only one NB ring per triglyceride, thereby restricting the number of ROMP steps because of significant steric hindrance. This result suggested that the presence of additional NB rings (i.e., a larger NB ring density) played an important role in increasing the cross-link density of the resulting films. To investigate the influence of the curing temperature (Figure 2b), curing was performed at 80 and 120 °C for 12 h with a G2 catalyst loading of 0.125 wt %. The film cured at 80 °C exhibited a significantly higher soluble percentage (13 wt %) than the film cured at 120 °C, whereby the lower solubility indicates a greater cross-linking density because of a higher catalytic activity. The characteristics of the soluble fraction are described in Figure S13. Since the use of the 0.5 wt % G2 catalyst resulted in the lowest quantity of unreacted NB rings in the soluble portion, this loading was identified as being optimal for the production of highly insoluble thermoset films.

Figure 2.

Soxhlet extraction of biobased thermoset films from NB-functionalized obtained via ROMP. (a) Effect of the catalyst loading (inset image shows the thermoset film color) and (b) effect of the curing temperature (0.125 wt % G2). (Photographs of thermoset films were taken with a digital camera).

Dynamic Mechanical Properties of the Biobased Thermoset Films

DMA was used to elucidate the thermomechanical properties of the obtained films, more specifically, in terms of the E′, T_g, and cross-linking density. Thus, Figure 3 shows the E′ and tan δ curves as a function of temperature for the thermoset films. As indicated, all NB-OO thermoset films were rubbery at room temperature. In addition, the E′ decreased slightly as the temperature increased from −70 to −35 °C and then decreased rapidly upon heating from −20 to 0 °C (Figure 3a). This corresponds to the primary relaxation (T_α) peak related to the energy dissipation, where a maximum is observed in the tan δ curve. Following the α-relaxation at high temperatures, a plateau was observed in the storage modulus curves, suggesting the presence of a cross-linked network in the films.

Figure 3.

DMA thermograms of the thermoset films: (a) NB-OO with various catalyst loadings and (b) various NB-functionalized plant oils (0.5 wt % G2).

The cross-link density (v_e) values were then calculated at a temperature of 40 °C above the T_g in the rubbery plateau of the storage modulus curve, using the rubbery theory of elasticity of eq 1(18,19)

1

where v_e is the cross-link density, E′ (Pa) is the storage modulus at T_g + 40 °C in the rubbery plateau, R is the universal gas constant 8.31 J·mol^–1 K^–1, and T is the absolute temperature (K) at T_g + 40 °C. The calculated cross-link density for the thermoset films is given in Table 1. As indicated, the cross-link density of the NB-OO films increased as the catalyst loading increased [v_e = 38 mol m^–3 (0.0625 wt %, run 1) versusv_e = 657 mol m^–3, run 4], suggesting that an increased catalyst loading produced a more efficient ROMP reaction. In addition, T_g was determined from the temperature giving the maximum peak in the tan δ curve. Table 1 (runs 1–4) and Figure 3a show the T_g values of the NB-OO thermoset films, where it was apparent that T_g increased from −23 to −8 °C as the catalyst loading increased from 0.0625 to 0.5 wt %. This was attributed to the increased cross-linking and cross-link density, restricting the mobility of the polymer.

Table 1. DMA Results and Thermal Stability of the Biobased Thermoset Films Prepared Using Different Catalyst Loadings and Plant Oils.

run	polymer	cat. (G2) (wt %)	E′ at 25 °C (MPa)a	Tg (°C)b	ve (mol m–3)c	tan δ	T5d (°C)	Tmaxe (°C)
1	NB-OO	0.0625	0.26	–23	38	1.59	362	486
2	NB-OO	0.125	1.26	–19	172	1.14	355	487
3	NB-OO	0.25	3.11	–13	414	0.84	335	489
4	NB-OO	0.5	5.00	–8	657	0.73	314	490
5	NB-RO	0.5	27.9	21	880	0.62	341	486
6	NB-SO	0.5	40.0	24	895	0.68	344	484

^aStorage modulus at 25 °C.

^bGlass transition temperature at the maximum tan δ curves.

^cCross-link density calculated at 40 °C above T_g.

^dTemperature at 5% weight loss.

^eMaximum thermal degradation temperature.

Table 1 (runs 1–4) also shows the E′ values obtained at room temperature. More specifically, the E′ values increased linearly with the catalyst loading [E′ = 0.26 MPa (0.0625 wt %, run 1) versus E′ = 5.0 MPa (run 4, 0.5 wt %)], which was attributed to restriction of the molecular motion because of the increased cross-link density (runs 1–4), which significantly decreased the energy dissipated through the polymer. In addition, the E′ value at room temperature increased significantly as the catalyst loading was increased, while the tan δ peak shifted to a higher temperature and its value decreased. The tan δ values of all NB-OO thermoset films were in the range of 0.73–1.59.

All NB-functionalized plant oils showed a linear positive relationship between the catalyst loading and the values of T_g, E′, and v_e (Table S2 and Figure S14). However, the absolute values differed, giving a ranking from low to high of NB-OO < NB-RO < NB-SO (Table 1, runs 1, 5, and 6, and Figure 3b). At 0.5 wt % G2, the E′ at room temperature was 5 MPa for NB-OO, 28 MPa for NB-RO, and 40 MPa for NB-SO, while the T_g was −8 °C for NB-OO, 21 °C for NB-RO, and 24 °C for NB-SO, and the cross-link density values were 657 mol m^–3 for NB-OO, 880 mol m^–3 for NB-RO, and 895 mol m^–3 for NB-SO (Table 1, runs 1, 5, and 6). Similarly, the polyunsaturated fatty acid content was related to the average number of NB rings per triglyceride, with values of 1.0, 1.3, and 1.5 being obtained for NB-OO, NB-RO, and NB-SO, respectively. These results suggest that the additional reactive NB rings present in the monomer structure were effectively incorporated into the thermoset film networks, thereby resulting in higher cross-linking. The tan δ values are also given in Table 1. More specifically, all three NB-OO, NB-RO, and NB-SO thermoset films (at 0.5 wt % G2) exhibited tan δ values >0.3 at a temperature of ∼60 °C, thereby rendering them potential candidates for damping applications.²⁰

Effect of CL on the Thermomechanical Properties of the Thermoset Films

In an attempt to increase the value of T_g, two types of CLs, petroleum-based NB with a silyl benzene ring linker (CL1) and a biobased CL from the NB-IS (CL2), were examined. The cross-linked networks of NB-plant oil both in the presence and absence of a CL are shown in Figure 4 and summarized in Table 2. No significant change in either E′ or the cross-link density was observed with a CL1 loading of 5 wt %, while upon increasing this loading to 10%, the E′ value of NB-OO increased from 5.0 (pristine) to 7.03 MPa, while that of NB-RO increased from 27.9 to 294 MPa and that of NB-SO increased from 40.0 to 278 MPa. The cross-link density (υ_e) also increased from 657 to 878 mol m^–3 for NB-OO, from 880 to 1112 mol m^–3 for NB-RO, and from 895 to 1720 mol m^–3 for NB-SO. This suggested that the addition of CL1 increased the degree of cross-linking. Furthermore, the T_g values increased significantly from −8 to −2 °C for NB-OO, from 21 to 34 °C for NB-RO, and from 24 to 40 °C for NB-SO. A CL1 loading >10 wt % produced brittle films that broke during preparation. This was attributed to the presence of an ethyl silyl benzene ring in CL1, which caused the films to become more rigid.

Figure 4.

Cross-linked networks formed using (a) NB-plant oil without CL, (b) NB-plant oil with 4-bis[dimethyl[2-(5-norbornen-2-yl)ethyl]silyl]benzene (CL1), and (c) NB-plant oil with NB-IS (CL2). (d) Photographic image of the biobased thermoset film and the brittle film. (Photographs of thermoset films were taken with a digital camera).

Table 2. DMA Results and Thermal Stability of the Biobased Thermoset Films Obtained Using Different Plant Oils with Different CLs.

runa	polymer	E′ at 25 °C (MPa)b	Tg (°C)c	ve (mol m–3)d	tan δ	T5e (°C)	Tmaxf (°C)
1	NB-OO	5.00	–8	657	0.73	362	490
2	NB-OO-5CL1	4.98	–8	658	0.72	359	484
3	NB-OO-10CL1	7.03	–2	878	0.67	363	524
4	NB-OO-5CL2	9.1	2	1055	0.63	350	482
5	NB-OO-10CL2	12.9	7	1167	0.55	346	482
6	NB-RO	27.9	21	880	0.62	341	486
7	NB-RO-5CL1	44.4	26	884	0.66	341	486
8	NB-RO-10CL1	294	34	1112	0.58	347	484
9	NB-RO-5CL2	116	32	1010	0.50	325	480
10	NB-SO	40.0	24	895	0.68	344	484
11	NB-SO-5CL1	41.4	25	1000	0.66	330	505
12	NB-SO-10CL1	278	40	1720	0.59	347	507
13	NB-SO-5CL2	149	35	1408	0.59	314	486

^aReaction conducted in the presence of 0.5 wt % G2.

^bStorage modulus at 25 °C.

^cGlass transition temperature at the maximum tan δ curve.

^dCross-link density calculated at 40 °C above T_g.

^eTemperature at 5% weight loss.

^fMaximum thermal degradation temperature.

Upon the application of biobased CLs, the E′, T_g, and v_e values increased at a CL2 loading of 5 wt %. More specifically, at room temperature, the E′ of NB-OO increased from 5.0 (pristine) to 9.1 MPa (5 wt %), that of NB-RO increased from 27.9 to 116 MPa, and that of NB-SO increased from 40 to 149 MPa. In addition, the T_g values of all thermoset films also increased by ∼10 °C. This was attributed to an increase in the cross-link density through linkage of the oxygen-containing heterocycles. Furthermore, upon increasing the CL2 loading in the NB-OO film from 5 to 10 wt %, the value of E′ increased to 12.9 MPa, while the cross-link density increased to 1167 mol m^–3. Raising the CL2 loading beyond 10 wt % produced brittle films that broke during preparation. Indeed, no specimens of NB-RO or NB-SO films could be prepared at 10 wt % CL2 because of the extent of cross-linking and the resulting increased brittleness.

The thermoset films prepared using CL2 were found to exhibit significantly higher T_g values. More specifically, at a CL loading of 5 wt %, the T_g was 25 °C for NB-SO-CL1 and 35 °C for NB-SO-CL2. This was attributed to the shorter and more rigid IS present in CL2, which restricted the chain motion, and so implied that the biobased NB-IS is a potential substitute for petroleum-based NB in ROMP-based polymer networks.

The results of thermal stability investigation for the NB-plant oil thermoset films are summarized in Tables 1 and 2, where the degradation temperature for 5% weight loss (T₅) and the maximum degradation temperature (T_max) were used as benchmarks to evaluate the thermal stability. The thermograms showed two stages of degradation behavior (Figures S15–S19). In the first stage (250–350 °C), thermal degradation was attributed to the evaporation and decomposition of soluble components (unreacted monomer and oil fragments),²¹ while the second stage at 350–500 °C revealed degradation of the polymer network, including cleavage of the carbon–carbon double bonds and ester groups. The weight loss percentage was found to decrease as the catalyst loading increased, which was attributed to the presence of a large cross-linking network in the film, and a reduction in the quantities of soluble components or oligomers. This result was consistent with the Soxhlet extraction results (Figure 2a). However, all NB-plant oil thermoset films presented similar T_max values in the range of 483–490 °C (Table S2). At the same catalyst loading (0.5 wt % G2), the T₅ value of the NB-OO film was higher (T₅ = 362 °C) than those of the NB-RO and NB-SO films (341 and 344 °C, respectively). This was expected due to the fact that the NB-RO and NB-SO main network structures contain higher amounts of oxirane and hydroxyl (−OH) groups, which undergo debonding and oxidization at lower temperatures. The polymer network of the NB-OO film also contains greater amounts of soluble unreacted triglycerides and oligomers, which can act as plasticizers, thereby increasing thermal stability.²²

The polymer linkage also played an important role in increasing the cross-link density and reducing the thermal degradation properties. More specifically, the thermoset with oxygen-containing heterocyclic linkages (CL2) presented a higher cross-link density than that with the ethyl silyl benzene ring linkage (CL1); however, the thermal degradation of the CL1-based thermoset was slightly higher. This result implies that the presence of aromatic linkages in a thermoset polymer may improve its heat resistance. Therefore, the incorporation NB-IS into the NB-plant oil macromolecular chains would be expected to reduce the high-temperature performance of the thermoset, as the thermal stability of the oxygen-containing heterocyclic linkage is lower than that of the silyl benzene ring linkage.¹⁵ However, CL2 could be considered a promising natural CL to enhance the thermomechanical properties of thermoset films.

Electrochemical Performance

The application of the NB-functionalized plant oil polymer as a binder for Si electrodes in lithium-ion batteries is one option to use the biobased thermoset in an energy storage device. Thus, Figure 5a,b shows the initial five cyclic voltammetry (CV) cycles of the Si/MCMB composite electrodes containing the NBD-cross-linked NB-OO and PVDF binders, respectively, whereby the electrodes present a cathodic peak at 1.25 V versus Li/Li⁺, which was attributed to the reduction of fluoroethylene carbonate (FEC). Furthermore, strong cathodic and anodic peaks were observed at 0.01–0.25 V, which correspond to the lithiation and delithiation of Si, thereby indicating a reversible process. In addition, Figure 5c,d shows the charge and discharge curves at different C-rates for the Li||Si/MCMB electrodes containing the NBD-cross-linked NB-OO and PVDF binders, respectively. As indicated, the discharge capacities of the electrodes containing the NBD-cross-linked NB-OO binder at 0.1, 0.2, 0.5, and 1 C were 605, 541, 413, and 279 mA h g^–1, respectively. However, when PVDF was employed as the binder, the discharge capacities at 0.1, 0.2, 0.5, and 1 C were 569, 424, 292, and 196 mA h g^–1, respectively. Furthermore, the C-rate performance shows that the Si/C composite electrode containing the NBD-cross-linked NB-OO binder exhibits a superior performance, which may be due to its low resistance caused by the present of hydroxyl, ester, and ether groups.

Figure 5.

CV curves of the Si electrodes (for the initial five cycles) using the (a) NBD-cross-linked NB-OO binder, and the (b) PVDF binder. Charge and discharge curves of Li | 1.0 M LiPF₆-EC/DEC/FEC (45:45:10, w/w) | Si electrode using the (c) NBD-cross-linked NB-OO binder, and the (d) PVDF binder at charge–discharge rates of 0.1, 0.2, 0.5, and 1.0 C.

To investigate the resistance of the electrodes, electrochemical impedance spectroscopy (EIS) was performed. As shown in Figure 6a, the Nyquist plots for the Li||Si/MCMB electrodes containing the NBD-cross-linked NB-OO and PVDF binders exhibit quasi-semicircles. The diameter of the quasi-semicircle along the x-axis may include both the resistance of the solid–electrolyte interface (SEI) film of the electrode (R_SEI) and the charge-transfer resistance (R_ct). Furthermore, to investigate the resistance of the electrolyte (R_s), R_SEI, R_ct, and the Warburg impedance (W) in detail, the Nyquist plots were simulated in Zview Version 3.2c using the equivalent circuit, as shown in Figure 6b. The results (Table S3) show that the R_s values of both electrodes were approximately 14 Ω. The R_SEI and R_ct values of the Li||Si/MCMB electrode containing the NBD-cross-linked NB-OO binder were 2.7 and 11.3 Ω, respectively, whereas those of the Li||Si/MCMB electrode with the PVDF binder were 4.2 and 13.2 Ω, respectively. Therefore, the overall resistance of the cell with the NBD-cross-linked NB-OO binder is less than that of the cell with the PVDF binder. As a result, the electrode with the NBD-cross-linked NB-OO binder exhibits a superior C-rate performance.

Figure 6.

(a) Nyquist plots of the Li||Si/MCMB electrodes containing the NBD-cross-linked NB-OO and PVDF binders and (b) corresponding equivalent circuit model.

Moreover, Figure 7 shows the cycle-life performances of the cells, whereby the initial charge capacities of both cells were ∼620 mA h g^–1 at 0.5 C. The charge capacity of the electrode containing the PVDF binder was lower than that containing the NBD-cross-linked NB-OO binder over 100 cycles, and the coulombic efficiency of the cell based on the NBD-cross-linked NB-OO binder was slightly higher than that where PVDF was employed. However, a decrease in charge capacity was observed after 100 cycles, which might be because of the catalyst present in the composite electrode; therefore, the effect of the catalyst or post-treatment catalyst removal should be investigated in future studies.

Figure 7.

(a) Charge capacities at various C-rates and (b) cycle-life performances and coulombic efficiencies of cells with the Li||Si/MCMB electrodes containing the NBD-cross-linked NB-OO and PVDF binders.

These results constitute a preliminary study into the use of NBD-functionalized plant oil as a biobased lithium-ion battery binder, whereby the obtained electrochemical results are promising in terms of comparison with a commercial PVDF binder. Although these results indicate the potential for NB-functionalized plant oil to be employed in energy storage systems, the binder stability requires further improvement prior to an application.

Conclusions

We herein reported our investigation into a simple method for the preparation of thermosetting films from NB-functionalized plant oils (OO, RO, and SO) using ROMP. This was achieved in the presence of a ruthenium-carbene (second-generation Grubbs) catalyst, and it was found that the T_g and E′ values, in addition to the thermal stability, were tunable. More specifically, the E′ and T_g values increased significantly as the catalyst loading increased, while the number of unreacted or oligomeric components in the films decreased. The optimal catalyst loading was found to be 0.5 wt %. In addition, the quantity of unsaturated fatty acids present in the plant oil structure was determined to be related to the average number of NB rings per triglyceride unit, and the number of NB rings in the monomer was related to the cross-link density of the thermosets and therefore to the T_g and E′ values. All three biobased thermoset films exhibited good thermal stability up to 340 °C. Moreover, the prepared NB-IS was demonstrated to be a natural CL that can enhance the E′ and T_g values of thermoset films, while slightly reducing their initial thermal stability. Furthermore, the NB-functionalized plant oil was used as a bioderived binder via copolymerization with another diene monomer and applied in a lithium-ion battery composite electrode, which showed promising performance when compared with an electrode containing a commercial PVDF binder. In particular, the electrode with the NBD-cross-linked NB-OO binder exhibited a superior C-rate performance. These results therefore indicate that the biobased NB-IS is a potential alternative to petroleum-based NB, and offers a promising route to more environmentally friendly polymers for application as biobased binders in energy storage systems.

Experimental Section

The materials and functional plant oil synthesis procedure are described in the Supporting Information section.^6,15,23 The functionalized plant oils were characterized by ¹H and ¹³C NMR spectra (Bruker AVANCE III HD 600 MHz), FTIR (Perkin Elmer Spectrum 100) at wavenumbers between 4000 and 600 cm^–1 and a resolution of 4 cm^–1 with 32 scans. TGA was performed on a Mettler-Toledo TGA/DSC 3+ thermogravimetric analyzer. A sample of each specimen (∼10 mg) was heated from 30 to 700 °C at a heating rate of 20 °C min^–1 under a nitrogen flow of 30 mL min^–1. DMA was carried out on a Mettler-Toledo DMA1 analyzer in the three-point bending mode at 1 Hz. Samples were cut into rectangular shapes of 10 mm × 5 mm × 1 mm. Samples were cooled to −70 °C and then held isothermally for 3 min. The samples were then heated to 150 °C at a rate of 3 °C min^–1.

Soxhlet extraction of the thermoset films was carried out at 60 °C for 24 h using methylene chloride as solvent. After extraction, the insoluble portion was dried in an oven at 60 °C for 12 h. The remaining solution was concentrated under reduced pressure and dried in vacuo. The soluble percentage of the thermoset films was calculated using eq 2(24)

2

where m₀ is the initial weight and m₁ is the dry weight after extraction.

Preparation of the Biobased Thermoset Films via ROMP

The second-generation Grubbs catalyst G2, RuCl₂(PCy₃)(H₂IMes)(CHPh) (0.0625–0.5 wt % based on the entire monomer and CL), was dissolved in dichloromethane (DCM) (0.4 mL) and added to mixtures of the three monomers (NB-OO, NB-RO, or NB-SO, 2.0 g) at different CL loadings (0–15 wt %) in DCM (1.0 mL per 1000 mg of total monomer and CL). The resulting solutions were then precooled to 0 °C in an ice bath and stirred vigorously in a 20 mL glass vial for 30 s to obtain a homogeneous solution. The solution was cast in a mold, placed in an oven, cured at 40 °C for 1 h, then post-cured at 120 °C for 12 h. The resulting thermoset films were rubbery and slightly transparent but retained the color of the monomers. To investigate the influence of the catalyst loading on the efficiency of the ROMP reaction, Soxhlet extraction with DCM was carried out for 24 h to remove any soluble materials (i.e., oligomeric NB-plant oil or unreacted triglyceride oil) from the films.

Preparation of the Si/MCMB Composite Electrodes

The main components of the silicon/MCMB composite electrode were graphite (75 wt %, MCMB, China Steel Chemical Corp., Taiwan), Si (5 wt %), Super P (10 wt %), NBD-cross-linked NB-OO binder (7.5 wt %), NB-OO, NBD (2.5 wt %). The second-generation Grubbs catalyst and oxalic acid were added at 0.125 and 1 wt %, respectively, relative to the total content of the main components of the electrode. All components were mixed in THF (0.5 mL) to form a slurry, which was cast onto a copper foil, and then dried in a vacuum oven at 120 °C for 12 h. The dried Si/MCMB composite electrode was cut into a circular shape with a diameter of 13 mm.

Electrochemical Measurements of the Si/MCMB Composite Electrodes

The electrochemical performance of the Si/MCMB composite electrode containing the NBD-cross-linked NB-OO (or PVDF) as a binder was examined using a coin-cell setup (CR 2032). The coin cells were assembled in an Ar-filled glovebox. For the CV measurements, the scan range was 3.0–0.01 V, and a scan rate of 0.1 mV s^–1 was employed at 30 °C. The Si/MCMB composite electrode was used as the working electrode and Li metal was used as the counter and reference electrodes. A polypropylene (PP) porous membrane was employed as a separator, and a 1.0 M LiPF₆-EC/DEC/FEC (ethylene carbonate/diethylene carbonate/fluoroethylene carbonate) solution (45:45:10, v/v) was used as an electrolyte. To determine the cell performance, the C-rate performance was measured at charge and discharge rates of 0.1, 0.2, 0.5, and 1 C at 30 °C, and the cycle-life performance was obtained at a charge–discharge rate of 0.5 C at 30 °C. For the EIS measurements, the cells were discharged and charged for two cycles, and then, the cell voltage was maintained at 1.2 V with a perturbation amplitude of 5 mV in the frequency range of 0.1–100 kHz.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c02645.

• Materials, functional plant oil synthesis procedure, structural characterization of NB-IS, structural characterization of NB-functionalized plant oils, photo of biobased thermoset films, FTIR spectra of biobased thermoset films, Soxhlet extraction analysis of NB-RO and NB-SO thermoset films, ¹H NMR spectra of soluble materials in thermoset films, dynamic mechanical and thermal properties of biobased thermoset films, TGA of biobased thermoset films, and equivalent circuit and EIS results (PDF)

Author Contributions

All authors contributed to the preparation of the manuscript. The following is a list of individual contributions. D.L.: investigation and validation; C.S. and K.N.: co-supervision and resources; J-T.L.: testing of electrochemical properties; S.K.: supervision, conceptualization, methodology, visualization, original draft preparation, writing, reviewing, and editing; S.K.: funding acquisition and co-supervision.

Thammasat University and Thailand Research Fund

The authors declare no competing financial interest.