Ionic liquid-based random copolymers: a new type of polymer electrolyte with low glass transition temperature

Abstract

A new type of polymer electrolyte has been prepared from the side-chains of ionic liquids (IL) and an analogue of ethylene oxide (EO) directly grafted on a polyethylene oxide backbone. By tuning the two types of monomer composition during polymerization, a series of copolymers with different monomer ratios between IL : EO = 8 : 1 to 1 : 4 have been synthesized. All copolymerized polyionic liquids showed higher ionic conductivity than an IL homopolymer. By choosing a composition with IL : EO = 1 : 1, the corresponding polymer gave the highest conductivity, 1.2 × 10⁻⁴ S cm⁻¹ at room temperature, compared to 9.3 × 10⁻⁶ S cm⁻¹ from a homopolymer. We attribute this to one order of magnitude improvement of conductivity from a low glass transition temperature (T_g) below −40 °C.

---

1. Introduction

Compared to ionic liquids polyionic liquids (PILs) exhibit improved mechanical stability, processability and durability.^1–3 The corresponding polymer electrolytes are suitable for use in electrochemical devices due to their chemical and physical stability and negligible vapor pressure, preventing electrolyte leakage.^2,4–10 However, a critical drawback of PILs is that after polymerization the conductivity usually suffers a huge decrease in comparison to the monomers.^11–14

It is well known that a lower glass transition temperature (T_g) leads to higher conductivity.^1–3 One way to lower the T_g is to introduce anionic side-chains into a polymer such as bis(trifluoromethanesulfonyl)imide (TFSI). Baker's group has successfully grafted TFSI anions onto a polyethylene oxide (PEO) backbone using anionic ring-opening polymerization. The PILs obtained have a low glass transition temperature, about −14 °C. As polymer electrolytes they have moderate conductivity, around 10⁻⁵ S cm⁻¹ at 30 °C and close to 10⁻³ S cm⁻¹ at 90 °C.¹¹

Another way to reduce the T_g of polymers is to introduce a low T_g component to the polymer backbone, forming a copolymer.^15–17 Elabd and coworkers have studied the effect of introducing a low T_g component on the ionic conductivity of PILs.¹⁶ The copolymer was made by free-radical polymerization of a nonionic monomer, n-hexyl methacrylate (HMA) and an ionic monomer 1-(2-methacryloyl)ethyl-3-butylimidazolium tetrafluoroborate (MEBIMBF₄). A series of copolymers with different levels of HMA were obtained by tuning the HMA concentration during the free radical polymerization. The results indicated that the T_g of the copolymer decreased with increasing HMA composition, and the conductivity increased correspondingly.

In the present study, a second component chosen for copolymer synthesis was an ethylene oxide (EO) analogue. PEO is synthesized by the polymerization of ethylene oxide and has T_g ranging from −50 to −57 °C. However, ethylene oxide is a gas at room temperature, making it hard to process under normal polymerization conditions. In view of this, a liquid analogue of ethylene oxide, 2-((2-(2-(2-methoxyethoxy)ethoxy)ethoxy)methyl)oxirane (ME₃MO), was synthesized and used as monomer.¹⁸ The random copolymer polyEPCH-co-polyME₃MO was synthesized by cationic ring-opening polymerization by an activated monomer mechanism using trifluoroborate etherate, and 1,4-butanediol as initiator.¹⁹ After quaternarization and ion exchange reaction, the PIL copolymer electrolyte showed a conductivity more than one order of magnitude greater than that of a PIL homopolymer.

2. Experimental

2.1 Materials

Unless otherwise specified, all chemicals and solvents were of ACS reagent grade and were used as received without further purification. Epichlorohydrin (EPCH, 99%, Aldrich) was purified over CaH₂, distilled under vacuum and stored in a glove box until use. 1-Butylimidazole was distilled under vacuum and stored in a refrigerator. Toluene was dried by refluxing with sodium p-benzylphenol, distilled under nitrogen and stored in a glove box. Boron trifluoride diethyl etherate (BF₃OEt₂) (≥46% based on BF₃, Aldrich) and 1,4-butanediol (99%, Aldrich) were distilled under nitrogen and also stored in a glove box. Lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, >98%, TCI) was kept in a glove box and used without further purification.

2.2 Synthesis of 2-((2-(2-(2-methoxyethoxy)ethoxy)ethoxy)methyl)oxirane (ME₃MO)

ME₃MO was synthesized according to the literature method.²⁰ NaH (60 wt% in mineral oil, 9 g, 0.23 mol) was rinsed with toluene and added to 200 mL THF. Triethylene glycol monomethyl ether (32.8 g, 0.2 mol) was then added and stirred for 2 h. The solution was cooled to 0 °C in an ice bath, followed by the gradual addition of epichlorohydrin (18.4 g, 0.2 mol). The solution was stirred between 0 °C and room temperature overnight. After reaction, the solution was neutralized using 30% methanolic H₂SO₄. The solution was filtered, concentrated using a rotavap and distilled under vacuum. Yield, 31.2 g (71%).

¹H NMR (CDCl₃, 500 MHz), δ (ppm) 3.66 (dd, J = 11.7, 3.0 Hz, 1H), 3.61–3.49 (br, 10H), 3.44–3.40 (m, 2H), 3.29 (dd, J = 11.7, 5.9 Hz, 2H), 3.25 (s, 3H), 3.05–3.00 (m, 1H), 2.66 (dd, J = 5.0, 4.2 Hz, 1H), 2.48 (dd, J = 5.0, 2.7 Hz, 1H). ¹³C NMR (CDCl₃, 500 MHz) δ (ppm) 71.55, 71.50, 70.28, 70.16 (overlapping signals), 70.06, 58.50, 50.32, 43.66. MS (TOF MS ES+): calculated for C₁₀H₂₀O₅Na⁺ (m/z), 243.1208; found, 243.1213.

2.3 Synthesis of polyEPCH by cationic ring-opening polymerization

PolyEPCH was synthesized by cationic ring-opening polymerization using the literature method.¹⁹ In a dry Schlenk flask, 1,4-butanediol (0.44 g, 4.8 mmol) was added to 22.5 mL dry toluene. This solution was degassed by a freeze–pump–thaw procedure three times and then protected by a nitrogen gas flow. To the solution, BF₃OEt₂ (0.12 g, 0.81 mmol) was slowly added under nitrogen and stirred 60 min at room temperature. The solution was cooled to 0 °C in an ice bath, followed by the dropwise addition of EPCH (7.5 g, 81 mmol). The polymerization was terminated after 8 h by adding a small amount of distilled water, and the polymer solution washed with sodium bicarbonate solution (5% w/v), followed by further washing with distilled water to neutral pH. The solvent was removed by vacuum distillation and the product dried overnight at 60 °C under vacuum. Yield, 6 g (80%).

¹H NMR (CDCl₃, 500 MHz) δ 3.78–3.66 (4H), 3.66–3.59 (1H).

2.4 Synthesis of polyME₃MO

PolyME₃MO was synthesized in a similar manner to polyEPCH. In a dry Schlenk flask, 1,4-butanediol (0.44 g, 4.8 mmol) was added to 22.5 mL dry toluene. The solution was degassed three times by a freeze–pump–thaw procedure and then protected by a nitrogen gas flow. To this solution BF₃OEt₂ (0.12 g, 0.81 mmol) was slowly added under a nitrogen atmosphere and stirred 60 min at room temperature. The solution was cooled to 0 °C in an ice bath, followed by the dropwise addition of ME₃MO (17.8 g, 81 mmol). The polymerization was stopped after 18 h by adding a small amount of distilled water. The purification method was in this case different, however, since polyME₃MO is soluble in water. The polymer solution was first washed with sodium bicarbonate solution (5% wt/v), and the polymer extracted using CHCl₃ and precipitated from hexane. Yield, 13.4 g (75%).

¹H NMR (CDCl₃, 500 MHz) δ 3.74–3.22 (20H).

2.5 Synthesis of polyEPCH-co-polyME₃MO random copolymer

PolyEPCH-co-polyME₃MO random copolymer was synthesized in a similar manner to polyEPCH. A series of copolymers were obtained by changing the molar concentration of ME₃MO monomer, keeping the concentration of EPCH monomer constant. For example, the polyEPCH-co-polyME₃MO random copolymer with a 1 : 1 monomer molar ratio (polyEPCH-co-polyME₃MO-1/1) was prepared as follows.

In a dry Schlenk flask, 1,4-butanediol (0.44 g, 4.8 mmol) was added to 22.5 mL dry toluene. The solution was degassed three times by a freeze–pump–thaw procedure and then protected by a nitrogen gas flow. To this solution, BF₃OEt₂ (0.12 g, 0.81 mmol) was added slowly under a nitrogen atmosphere and stirred 60 min at room temperature. The solution was cooled to 0 °C in an ice bath, followed by the dropwise addition of pre-mixed EPCH (7.5 g, 81 mmol) and ME₃MO (17.8 g, 81 mmol). The copolymer was purified similarly to polyME₃MO. Yield, 18.5 g (73%). For copolymers with different monomer molar ratios, the yield varied from 60 to 80%.

¹H NMR (CDCl₃, 500 MHz) δ 3.66–3.44 (22H), 3.39–3.36 (3H).

2.6 Synthesis of polyGBIMTFSI-co-polyME₃MO

A series of polyGBIMTFSI-co-polyME₃MO were synthesized based on different polyEPCH-co-polyME₃MO samples. PolyGBIMTFSI-co-polyME₃MO-1/1 was synthesized as follows.

PolyEPCH-co-polyME₃MO-1/1 (12.5 g, 40 mmol based on its repeating unit) was first placed in a 100 mL Schlenk flask, followed by the addition of 1-butylimidazole (14.9 g, 120 mmol) and a magnetic stirrer bar. The mixture was stirred until homogeneous. 1-Butylimidazole was again used both as reactant and solvent. The solution was degassed three times using the freeze–pump–thaw procedure, then placed in an oil bath at 115 °C and refluxed overnight under nitrogen. After reaction, the polyGBIMCl-co-polyME₃MO polymer was precipitated from ethyl ether, collected and dried under vacuum for 8 h. The polymer was dissolved in 100 mL Milli-Q water, followed by an excess of LiTFSI (17.2 g, 60 mmol). The product, polyGBIMTFSI-co-polyME₃MO-1/1, was collected by 30 min centrifugation, washed three times with Milli-Q water and dried overnight at 100 °C in vacuo. Yield, 23.7 g (87%).

¹H NMR (d₆-DMSO, 500 MHz) δ 9.20–8.98 (1H), 7.88–7.50 (2H), 4.50–3.00 (27H), 1.82–1.72 (2H), 1.28–1.14 (2H), 0.92–0.80 (3H).

2.7 Characterization

The ¹H and ¹³C NMR spectra were recorded on a Varian UnityPlus-500 spectrometer in CDCl₃ or DMSO-d₆ at room temperature, using the residual solvent proton signals as chemical shift standards. gHSQC and gHMBC 2D NMR spectra were recorded using an Agilent DDR2 500 MHz NMR spectrometer equipped with 7600AS 96 autosamplers. Polymer molecular weight was measured by gel permeation chromatography (GPC) at 35 °C using two PL-gel 10μ mixed-B columns in series. DMF was used as the eluting solvent at a flow rate of 1 mL min⁻¹, with a Waters 2410 differential refractometer as detector. Monodispersed poly(ethylene oxide) was used as a standard for calibrating molecular weight. Thermogravimetric analysis (TGA) was performed using a PerkinElmer TGA 7 at a heating rate of 10 °C min⁻¹ over a temperature range of 30 to 850 °C in air. TGA samples were dried in vacuo overnight before use. After increasing the temperature to 120 °C, samples were held 30 min in the TGA 7 before continuing the heating. Differential scanning calorimetry (DSC) was performed using a TA DSC Q100 instrument at a heating/cooling rate of 10 °C min⁻¹ under nitrogen.

2.8 Ionic conductivity

Ionic conductivity measurements were carried out in a home-made cell by impedance spectroscopy measurements, using an AC impedance analyzer HP 4192A over a frequency range 5 to 13 MHz with an applied voltage of 10 mV. The sample cell contained two steel disks serving as symmetrical electrodes, separated by a Teflon collar. All samples were equilibrated at a predetermined temperature 15 min before measurement. The resistance of the samples was estimated from a Nyquist plot, and the conductivity calculated from the electrode distance and cell cross-sectional area, as shown in eqn (1):where σ is the ionic conductivity of the sample, L is the distance between the two electrodes, A is the cell cross-sectional area, and R is the bulk resistance of the sample estimated from the Nyquist plot.

3. Results and discussion

The synthesis of 2-((2-(2-(2-methoxyethoxy)ethoxy)ethoxy)methyl)oxirane (ME₃MO) is shown in Fig. 1(a). Triethylene glycol monomethyl ether was first deprotonated by sodium hydride and then reacted with epichlorohydrin (EPCH). The product ME₃MO was obtained in 71% yield and characterized by ¹H NMR, shown in Fig. 1(b). Peak 7 in the ¹H NMR spectrum was assigned to the terminal CH₃ group in the PEO oligomer chain, due to its distinct chemical shift and integral value of 3.03. The remaining peaks were assigned according to 2D NMR, gHSQC and gHMBC (Fig. 2(a) and (b)). From gHSQC, most of the peaks could be assigned, apart from peak 6 in Fig. 1(b). The peaks at 44 and 51 ppm in the ¹³C spectrum were assigned to the C of the CH₂O group in the oxirane ring, and that at 72 ppm to the C in the CH₂O group adjacent to the oxirane ring. Based on the correlation between the carbons and protons, the peaks from 1 to 5 in ¹H NMR were assigned as follows. Peak 6 was assigned according to gHMBC. Since the protons of peak 6 interacted with the carbon of peak 7, peak 6 was assigned to the protons attached to the C closest to peak 7 in the ¹³C spectrum. Finally, the broad peak with an integral of 10.66 in ¹H NMR was assigned to the remaining five CH₂ protons in the PEO oligomer chain.

Fig. 1 (a) Schematic illustration of the synthesis of ME₃MO monomer, and (b) the corresponding ¹H NMR spectrum.

Fig. 2 (a) gHSQC and (b) gHMBC of the ME₃MO monomer.

Using BF₃OEt₂ and 1,4-butanediol as initiator, cationic ring-opening polymerization was used to synthesize polyepichlorohydrin (polyEPCH), polyME₃MO and polyEPCH-co-polyME₃MO, as shown in Fig. 3(a). Six copolymer samples of different monomer composition, and polyEPCH and polyME₃MO homopolymer, were synthesized. The molecular weight data are summarized in Table 1. PolyEPCH homopolymer had a M_n of 2.2 × 10³ g mol⁻¹, and polyME₃MO homopolymer had M_n of 3.4 × 10³ g mol⁻¹. The M_n of copolymer increased with the increase in polyME₃MO composition.

Fig. 3 (a) Schematic illustration of the synthesis of polyGBIMTFSI-co-polyME₃MO, and (b) its corresponding HNMR spectrum.

Table 1 Molecular weight of polyEPCH, polyME₃MO and polyEPCH-co-polyME₃MO synthesized using BF₃OEt₂ and 1,4-butanediol as initiator

[M]1/[M]2/[diol]/[ BF3OEt2]^a	Mn × 10^−3^b g mol	Mw × 10^−3^b g mol	Mp × 10^−3^b g mol	PDI^b
a [M]1, [M]2, [diol] and [BF3OEt2] represent molar concentrations of EPCH, ME3MO, 1,4-butanediol and BF3OEt2, respectively.b Mn is the number average molecular weight, Mw the weight average molecular weight, Mp the peak molecular weight, and PDI is the polydispersity index. Data were obtained by GPC, using PEO as standard.
100/0/6/1	2.2	3.3	2.8	1.50
0/100/6/1	3.4	4.8	3.6	1.41
100/12.5/6/1	3.9	5.8	4.0	1.49
100/25/6/1	4.4	6.4	5.0	1.45
100/50/6/1	4.9	7.4	6.0	1.51
100/100/6/1	5.4	8.7	7.1	1.61
100/200/6/1	7.3	11.7	9.4	1.60
100/400/6/1	12.0	19.8	15.4	1.65

---

PolyGBIMTFSI-co-polyME₃MO was synthesized by quaternarizing polyEPCH-co-polyME₃MO with 1-butylimidazole, followed by an ion exchange reaction. Only the polyEPCH component in the copolymer was quaternarized, whereas polyME₃MO was left intact. A high yield was achieved for all the samples. The structure of the polyGBIMTFSI-co-polyME₃MO was characterized by ¹H NMR, as shown in Fig. 3(b). The peaks between 3.0 and 4.5 ppm represented protons from the polyGBIMTFSI backbone, all the protons from polyME₃MO and one CH₂ from the butyl group. The three peaks above 7.4 ppm were assigned to the three protons from the imidazolium ring, and the three peaks below 1.8 ppm were the protons from CH₂CH₂CH₃ of the butyl group on the imidazolium ring. The integral of CH₃ group was set as 3, and the integral of broad peaks between 3.0 and 4.5 ppm was 12.23. The integrals corresponded very well to the composition of the copolymer, polyGBIMTFSI-co-polyME₃MO-4/1.

The thermal stability of polyGBIMTFSI and its random copolymers was studied using TGA, as shown in Fig. 4. PolyGBIMTFSI and polyGBIMTFSI-co-polyME₃MO with a higher molar ratio of polyGBIMTFSI (increased from 8/1 to 2/1) showed similar excellent thermal stability, with thermal degradation temperatures above 375 °C. On the other hand, polyGBIMTFSI-co-polyME₃MO, with a higher molar ratio of polyME₃MO (increased from 1/1 to 1/4) showed relatively poor thermal stability, and began to lose weight at about 200 °C. As shown in a previous study, polyEPCH also started to lose weight around 200 °C.¹¹ The improvement in the thermal stability of polyGBIMTFSI arose from quaternarization, replacing the weak bonding between the pendant CH₂–Cl and CH₂–O(CH₂CH₂O)₃CH₃.

Fig. 4 TGA traces of polyGBIMTFSI, polyME₃MO and their random copolymers of varying molar composition.

The glass transition temperatures (T_g) of polyGBIMTFSI, polyME₃MO and their random copolymers were also studied by DSC. As shown in Table 2, polyGBIMTFSI had the highest T_g (−25 °C), and polyME₃MO the lowest (−52 °C). All the random copolymers showed T_g between these values. With its higher molar composition of polyME₃MO, the copolymer had a lower T_g.

Table 2 T_g of polyGBIMTFSI, polyME₃MO and their random copolymers of varying molar composition

Sample	Tg (°C)
PolyGBIMTFSI	−25
PolyGBIMTFSI-co-polyME3MO (8/1)	−29
PolyGBIMTFSI-co-polyME3MO (41)	−33
PolyGBIMTFSI-co-polyME3MO (2/1)	−36
PolyGBIMTFSI-co-polyME3MO (1/1)	−41
PolyGBIMTFSI-co-polyME3MO (1/2)	−43
PolyGBIMTFSI-co-polyME3MO (1/4)	−45
PolyME3MO	−52

---

The ionic conductivities of polyGBIMTFSI and polyGBIMTFSI-co-polyME₃MO were measured over a temperature range from 25 to 90 °C, as shown in Fig. 5. The ionic conductivity data agreed well with the T_g data. PolyGBIMTFSI, with the highest T_g, showed the lowest ionic conductivity. With increasing polyME₃MO the conductivity increased up to polyGBIMTFSI-co-polyME₃MO-1/1. However, at higher compositions of polyME₃MO the conductivity began to decrease, and polyGBIMTFSI-co-polyME₃MO-1/4 had lower conductivity than its 1/1 analogue. This phenomenon is readily understood, and similar results have been reported in the literature.¹⁶ Introduction of polyME₃MO lowered the T_g, which in turn had a positive effect on the ionic conductivity of the copolymer. On the other hand, increasing the polyME₃MO composition meant a decrease in the polyGBIMTFSI composition, resulting in lower concentration of charge carriers and a negative effect on the ionic conductivity of the copolymer. Initially the conductivity of the copolymers increased with increasing polyME₃MO composition, since the reduction in T_g had a more significant effect on ionic conductivity than the reduction in charge carrier concentration. However, in the case of polyGBIMTFSI-co-polyME₃MO-1/2 and polyGBIMTFSI-co-polyME₃MO-1/4, as the concentration of charge carriers became very low the negative effect overwhelmed the positive effect.

Fig. 5 Ionic conductivity of polyGBIMTFSI and its random copolymers of varying molar composition.

PolyGBIMTFSI-co-polyME₃MO-1/1 showed the highest conductivity (around 1.2 × 10⁻⁴ S cm⁻¹ at 25 °C), which was more than one order of magnitude higher than that of the polyGBIMTFSI homopolymer (9.3 × 10⁻⁶ S cm⁻¹ at 25 °C). On the other hand, the conductivity of a common ionic liquid, 1-methyl-3-propylimidazolium iodide (MPII), was 1.1 × 10⁻³ S cm⁻¹ at 25 °C.^10,11 The conductivity of polyGBIMTFSI-co-polyME₃MO-1/1 was less than one order of magnitude lower than that of MPII. Introducing a low T_g component to form a PIL copolymer proved to be very effective in improving the ionic conductivity of polyionic liquids.

4. Conclusions

2-((2-(2-(2-Methoxyethoxy)ethoxy)ethoxy)methyl)oxirane (ME₃MO), a liquid analogue of ethylene oxide, was synthesized and copolymerized with epichlorohydrin. The resulting copolymer (polyEPCH-co-polyME₃MO) was quaternarized with 1-butylimdazole, followed by an ion exchange reaction, to form a PIL copolymer (polyGBIMTFSI-co-polyME₃MO). By tuning the monomer composition, a series of PIL copolymers from polyGBIMTFSI-co-polyME₃MO-8/1 to polyGBIMTFSI-co-polyME₃MO-1/4 were synthesized. All the monomers and polymers were characterized by 1D and 2D NMR. The polymerized ionic liquids in the study showed a degradation temperature above 350 °C. All PIL copolymer samples showed higher ionic conductivity than the PIL homopolymer up to 1.2 × 10⁻⁴ S cm⁻¹ at room temperature, which was more than one order of magnitude higher than the PIL homopolymer (9.3 × 10⁻⁶ S cm⁻¹ at 25 °C). It was found that the introduction of a component such as ethylene oxide, with low T_g or a PEO backbone, to form a PIL copolymer proved very effective in improving the ionic conductivity of PIL. Ongoing research will include potential applications such as lithium polymer batteries, solid supercapacitors or solid-state dye-sensitized solar cells.

Acknowledgements

This study was supported financially by the National Science Foundation (DMR–0934568) and the MSU Center for Alternative Energy Storage Research and Technology (CAESRT). W.Y. is also indebted to Dr James McCusker, Dr Keith Promislow and Dr Lawrence Drzal for valuable discussions.

References

1. D. Mecerreyes, Prog. Polym. Sci., 2011, 36, 1629–1648.
2. J. Y. Yuan and M. Antonietti, Polymer, 2011, 52, 1469–1482.
3. O. Green, S. Grubjesic, S. W. Lee and M. A. Firestone, Polym. Rev., 2009, 49, 339–360.
4. H. Ohno and K. Ito, Chem. Lett., 1998, 1, 751–752.
5. M. Yoshizawa and H. Ohno, Chem. Lett., 1999, 1, 889–890.
6. H. Ohno, Electrochim. Acta, 2001, 46, 1407–1411.
7. M. Yoshizawa and H. Ohno, Electrochim. Acta, 2001, 46, 1723–1728.
8. S. Washiro, M. Yoshizawa, H. Nakajima and H. Ohno, Polymer, 2004, 45, 1577–1582.
9. W. Yuan, H. Zhao, H. Y. Hu, S. Wang and G. L. Baker, ACS Appl. Mater. Interfaces, 2013, 5, 4155–4161.
10. H. Y. Hu, W. Yuan, H. Zhao and G. L. Baker, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 121–127.
11. H. Y. Hu, W. Yuan, L. S. Lu, H. Zhao, Z. Jia and G. L. Baker, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 2104–2110.
12. M. Dobbelin, I. Azcune, M. Bedu, A. R. de Luzuriaga, A. Genua, V. Jovanovski, G. Cabanero and I. Odriozola, Chem. Mater., 2012, 24, 1583–1590.
13. H. Ohno, M. Yoshizawa and W. Ogihara, Electrochim. Acta, 2004, 50, 255–261.
14. Y. Zhang, Y. W. Zhu, G. Lin, R. S. Ruoff, N. P. Hu, D. W. Schaefer and J. E. Mark, Polymer, 2013, 54, 3605–3611.
15. Y. S. Ye, J. H. Choi, K. I. Winey and Y. A. Elabd, Macromolecules, 2012, 45, 7027–7035.
16. H. Chen, J. H. Choi, D. Salas-de La Cruz, K. I. Winey and Y. A. Elabd, Macromolecules, 2009, 42, 4809–4816.
17. W. Yuan, H. Zhao and G. L. Baker, Org. Electron., 2014, 15, 3362–3369.
18. Z. Jia, W. Yuan, H. Zhao, H. Y. Hu and G. L. Baker, RSC Adv., 2014, 4, 41087–41098.
19. A. U. Francis, S. Venkatachalam, M. Kanakavel, P. V. Ravindran and K. N. Ninan, Eur. Polym. J., 2003, 39, 831–841.
20. S. J. Jungk, J. A. Moore and R. D. Gandour, J. Org. Chem., 1983, 48, 1116–1120.

---

Footnotes

† These authors contributed equally to this study.

‡ Professor Gregory L. Baker passed away on 18 October 2012.

---