Linseed and soybean oil-based polyurethanes prepared via the non-isocyanate route and catalytic carbon dioxide conversion

Abstract

Soy- and linseed oil-based polyurethanes were prepared without using isocyanate monomers by curing carbonated soybean (CSBO) and linseed (CLSO) oils with different diamines. The conversion of the epoxidized seed oils with carbon dioxide to form cyclic carbonates was catalyzed with tetra-butylammonium bromide (TBAB) and silica-supported 4-pyrrolidinopyridinium iodide (SiO₂–(I)), which was readily recovered without requiring the conventional purification by solvent extraction of TBAB. Carbonate formation was monitored as a function of catalyst type and pressure, indicating slower conversion rates for the supported catalyst. Quantitative epoxy conversion was achieved and the amount of immobilized carbon dioxide increased from 15.2 wt% for CSBO to 19.6% for CLSO. The seed oil carbonates with variable carbonate content (20.2 to 26.8 wt%) were cured with 1,2-ethane diamine (EDA), 1,4-butane diamine (BDA) and isophorone diamine (IPDA) in order to examine the thermal and mechanical properties of the resulting non-isocyanate polyurethanes (NIPU). In contrast to the flexible conventional CSBO/EDA-NIPU, it was possible to increase NIPU glass transition temperatures from 17 °C to 60 °C and to improve stiffness, as expressed by Young's modulus, by three orders of magnitude. Higher crosslink densities accounted for reduced water swelling and toluene uptake.

---

Introduction

Both the increasing shortage of fossil resources and global warming due to carbon dioxide emissions stimulate the development of polymeric materials which do not require crude oil as feedstock and exhibit attractive carbon footprints. An important approach towards carbon dioxide neutral materials is to exploit carbon dioxide conversion reactions of intermediates from renewable resources such as plant seed oils for producing monomers. Another challenge in green polymer chemistry is to eliminate toxic intermediates in polymer synthesis. Several groups have addressed these challenges in polyurethane chemistry aiming at the development of non-isocyanate polyurethanes (NIPU) based upon renewable feedstocks and carbon dioxide conversion.^1–3 The synthetic strategy is outlined in Fig. 1.

Fig. 1 Strategy for NIPU formation.

Cyclic carbonates are attractive green intermediates which are non-toxic, biodegradable and represent excellent solvents with high boiling and flash points.⁴ Two different reaction pathways are feasible for reactions of cyclic carbonates react with nucleophiles: alkylation and ring opening. Weak nucleophiles such as phenol,⁵thiophenol⁶ or aniline⁷ are alkylated and CO₂ is split off.⁸ In contrast, stronger nucleophiles such as primary alkyl amines react with cyclic carbonates to afford ring opening and formation of N-hydroxy urethanes without loss of carbon dioxide.⁹ Reactivity depends upon the carbonate ring size. In contrast to conventional urethanes, obtained in isocyanate reactions, hydroxyl alkyl side chains are formed with the hydroxyl group being located at the β-carbon atom.^10–12Hydrogen bonds formed between the hydroxy groups and the carbonyl group of the urethanes can account for higher water absorption, improved resistance to organic solvents, and increased Young's modulus as well as higher tensile strength.¹³ In contrast to conventional PU the NIPUs are claimed to exhibit improved thermal stability because thermally labile biurets and allophanates groups are absent.¹⁴ Seed oils represent attractive raw materials because they are renewable and commercially available in large quantities.¹⁵ Unsaturated vegetable oils like soybean or linseed oil can be easily epoxidized via peracid oxidation.^16–21 These epoxidized vegetable oils, usually epoxidized soybean oil (ESBO) or epoxidized linseed oil (ELO), are used as plasticizers for polyvinylchloride. Several groups have examined NIPU formation. Wilkes and Tamami have pioneered the preparation of seed-oil based NIPU by means of diamine cure of carbonated soybean oil (CSBO) which is readily available by reacting epoxidized soybean oil (ESBO) with carbon dioxide.^1–3 Javni et al. converted ESBO into carbonated soybean oil (CSBO) at temperatures between 110 °C and 140 °C, CO₂ pressures ranging from atmospheric to 5.65 MPa. They demonstrated that both reaction rate and yield of the epoxy conversion depended on the pressure of CO₂, the reaction temperature, and the amount of catalyst terta-butylammonium bromide (TBAB).¹⁴ They also showed that the preferred amine to carbonate molar ratio in NIPU formation was 1. When excess amine was added, aminolysis of the seed oil ester groups occurred, thus lowering cross linking density. Most conventional NIPU derived from CSBO afford rather flexible NIPU with low strength and low glass temperature. For NIPU prepared by curing CSBO with EDA, Javni et al. reported a maximum tensile strength of 5.8 MPa, whereas Wilkes et al., obtained only 0.8 MPa. Such differences in mechanical properties of NIPU reflect changes of crosslink densities, which depend significantly upon the carbonate content of the carbonated seed oils. Since Wilkes et al. used CSBO with a lower degree of epoxy to carbonate conversions, the resulting CSBO carbonate content was lower, thus accounting for lower crosslink densities. The carbonate content of CSBO is restricted by the olefinic unsaturation of the soybean oil. For many NIPU applications, especially in the range of thermoplastic elastomers and engineering applications, it is highly desirable to improve both glass transition temperature and mechanical strength. In our research we have examined the preparation and cure of carbonated linseed oils (CLSO) from epoxidized linseed oil (ELSO) because ELSO have markedly higher epoxy content (8.9 g O/100 g) with respect to that of epoxidized soybean oil (6.8 g O/100 g). Moreover, the role of catalysts in CLSO preparation was investigated aiming at easy-to-recover solid phase catalysts which do not require purification by solvent extraction. Both Tamami et al. and Javni et al. used the homogeneous TBAB catalyst, which was removed by diluting the CSBO with ethyl acetate and subsequent extraction with water.^1–3,22,23 Since easy catalyst recovery is an important aspect in industrial production, supported alkylammonium catalysts are of special interest. According to Motokura et al., various alkyl ammonium catalysts immobilized on Si–O surface catalyze the conversion of propylene oxide or styrene oxide in order to afford propylene carbonate or styrenecarbonate.²⁴ At atmospheric CO₂ pressure higher carbonate yield and faster epoxy conversion were observed with respect to those of homogenous catalysts. The Si–OH on the silica surface acts as a weak acid to activate the epoxide, then the activated epoxide undergoes nucleophile attack by the halide anion (X⁻).²⁵ In our research we examined various homogeneous and heterogeneous catalysts in CLSO synthesis. The resulting CLSO were cured with 1,2-ethane diamine (EDA), 1,4-butane diamine (BDA) and isophorone diamine (IPDA) in order to establish the correlations between carbonate content and type of curing agent with solvent uptake, thermal and mechanical properties of linseed and soybean oil based NIPU.

Experimental

Materials

Epoxidized soybean oil (ESO) with epoxy oxygen content (EOC) of 6.8 g O/100 g corresponding to 4.2 mol epoxy groups per triglyceride was obtained from Cognis and epoxidized linseed oil (ELSO) with EOC of 8.9 g O/100 g, corresponding to 5.2 mol epoxy groups per triglyceride was obtained from HOBUM Oleo chemicals. Carbon dioxide grade N45 was supplied by Air Liquide. (3-Iodopropyl)-trimethoxysilan, (1-butyl)-imidazol, TBAB, EDA, BDA, and IPDA were supplied by Aldrich. As substrate for the heterogenous catalyst Aerosile 200 with a surface of 200 m² g⁻¹ from Evonik was used.

Preparation of carbonated seed oils

Typically the carbonated seed oils used for NIPUs were prepared according to the following procedures. The seed oil (5 kg) was placed in a 10 L reactor, carbon dioxide (10 bar pressure) and the catalyst TBAB (3.5 g per 100 g ESBO/ELO) were added, stirred and heated (140 °C). The reactions were terminated when complete conversion was indicated by means of IR, ¹H-NMR, titration measurements. Kinetics of the epoxy group conversion were measured using a 500 ml vessel by injecting different catalysts, and varying the CO₂ pressure atmospheric to 10 and 30 bar pressure. The EOC was determined by NMR spectroscopy and titration as reported in the literature.²⁶ The number of carbonate groups per triglyceride was calculated from the initial and final EOC content, assuming no side reactions.

NIPU preparation

Typically NIPU was as prepared with a lab mixer from Molteni according to the following procedures: CSBO/CLSO was placed in a 500 mL aluminium can and degassed in vacuum at 70 °C, followed by amine addition. The reactions mixture was mixed thoroughly during 5 min. The viscous solution was poured into a mold and heated at 70 °C for 10 h, and then for 3 h at 100 °C. The polyurethanes prepared with the less reactive isophoronediamine as hardener were additionally heated for 12 h by 70 °C. NIPUs based on CLSO cured with EDA and BDA were prepared in laboratory scale with mixing thoroughly during 2 min in a 100 mL flask.

Material evaluation

Mechanical characterization. The tensile modulus of the composites was measured using a Zwick Z005 (Ulm, Germany) (ISO/DP 527). The data were taken at room temperature with preconditioning the samples for 1 week at 23 °C. The statistical average of the measurements on at least five test specimens was taken to obtain the mean value for all tests conducted. Dynamic mechanical tests were carried out on Rheometrics Solid Analyzer RSA II with a heating rate of 5 °C min⁻¹. DMA specimens were rectangular shaped and 50 mm × 6 mm × 2 mm. The testing was carried out between −20 °C and 90 °C in the 3-point-tension mode in a nitrogen atmosphere at 0.1% amplitude and the frequency of 1 Hz. The glass transition temperature was determined from the maximum of the tan δ.

Spectroscopy . FTIR spectra were recorded on a Bruker FTIR Vector 2200 spectrometer with a Goldengate unit. The spectra were recorded from the solid plaques using attenuated total reflectance (ATR) technique. The number of scans per one recording was 30 and the resolution 2 cm⁻¹. The ¹H-NMR spectra were recorded on a Bruker ARX 300 Spectrometer. As Solvent were used CDCl3 and the measurements takes place at 300 MHz.

Viscosity. All viscosities were recorded on a Brookfield RV-DV-II, cone type CPE 40 cone plate viscosimeter, cone angel 0.88. The measurement takes place at a rotation speed between 0.5–5.0 U min⁻¹, at 22.5 ± 0.1 °C and a sample volume of 0.5 ml.

Molecular weight. Molecular weight distribution was obtained on the THF GPC system consisting of the 510 pump and 410 Differential Refractometer at 40 °C with a flow rate of 1 mL min⁻¹ and polystyrene as standard.

Results and discussion

Preparation of carbonated linseed oil (CLSO)

As illustrated in Fig. 2, CLSO was prepared by converting the epoxy groups of ELSO with carbon dioxide in the presence of alkyl ammonium bromide catalysts. The bromide of the catalyst attacks the epoxide ring to afford ring opening followed by nucleophilic attack of the alkoxide at carbon dioxide. Subsequent ring closure reaction of the resulting carboxylate yields the five-membered carbonate.² The conversion was performed at 140 °C using atmospheric pressure as well as 10 bar and 30 bar carbon dioxide pressure. In addition to the well-known TBAB catalyst, silica-supported 4-pyrrolidinopyridinium iodide (SiO₂–(I)) were used as heterogeneous catalyst. SiO₂–(I) was prepared as reported from Motakura et al. by a silane-coupling reaction of 3-iodopropyltrimethoxysilane with silica (Aerosil 200). It was followed by quaternization of 4-pyrrolidinopyridine with SiO₂-supported propyl iodide.²⁴Elemental analysis of SiO₂–(I) showed carbon loading of 6.37 mmol g⁻¹ which was compared to a iodide loading of 0.49 mmol g⁻¹. Both catalysts were used with 3% molar concentration with respect to epoxy content (Table 1).

Table 1 Data of the used catalyst

Catalyst	Ratio halide per epoxy (mol%)	Halide loading^a (mmol g^−1)	Surface^b (m^2)
TBAB = tetra-butylammonium bromide, SiO2–(I) = silica-supported 4-pyrrolidinopyridinium iodide.a Elementary analysis.b BET.
TBAB	3.0	—	—
SiO2–(I)	3.0	0.49	287

---

Fig. 2 Conversion of epoxidized seed oil to carbonated vegetable oil.

The reaction of epoxides with CO₂ was monitored by ¹H-NMR and IR. The conversion of ELSO to CLSO is shown in Fig. 3 and Fig. 4. They show exemplarily the ¹H-NMR and FTIR spectra with TBAB as catalyst, at 140 °C and 10 bar CO₂ pressure.

Fig. 3 ¹H-NMR spectra from the kinetics of the ELSO conversion at 140 °C, 10 bar CO₂ pressure and TBAB as catalyst.

Fig. 4 IR spectra from the kinetics of the ELSO conversion at 140 °C, 10 bar CO₂ pressure and TBAB as catalyst.

The signals of the epoxy groups at δ = 2.80–3.20 ppm disappeared upon conversion of ELSO with CO₂, while the new signals at δ = 4.45–5.10 ppm corresponding to the cyclic carbonate groups, appeared. The FTIR-spectra in Fig. 4 showed an increasing band at 1803 cm⁻¹ which was assigned to the cyclic carbonate carbonyl. Using the glycerol ester carbonyl band at 1741 cm⁻¹ as reference it is possible to monitor quantitative the cyclic carbonate formation.

In Fig. 5a and the kinetics of epoxy conversions at 140 °C are displayed as a function of the CO₂ pressure in the presence of TBAB (Fig. 5a) and the heterogeneous SiO₂–(I) catalyst (Fig. 5b). Activities of both catalysts increased with increasing CO₂ pressure. At 30 and 10 bar carbon dioxide pressure, both catalysts afforded complete epoxy conversion, whereas at atmospheric pressure, even after 90 h, the epoxy conversion was incomplete. In contrast to the reaction of styrene oxide with CO₂ the homogeneous TBAB shows a faster conversion compared to SiO₂–(I).²⁴ At 30 bar the SiO₂–(I) catalyst reached complete reaction after 45 h whereas TBAB shows complete conversion already after 20 h. Most likely, the bulkiness and steric hindrance of ELSO plays an important role. Moreover, it should be noted that the much smaller styrene oxide is capable to enter small pores of the heterogeneous catalysts which are not accessible for ELSO. Similar observations reported by Javni et al. for CSBO, the TBAB catalysts is optimal in CLSO synthesis performed at 140 °C and 10 bar. Therefore, TBAB was used for scale-up of CLSO intermediates for NIPU preparation. The heterogeneous catalyst is easy to recover but requires further improvement with respect to its activity. This could be achieved by increasing pore size, tailoring pore compartments, increasing spacer length between support and alkyl ammonium groups, and by varying the alkyl ammonium cation groups and their counter ions. The CLO used for NIPU synthesis were made at these conditions.

Fig. 5 Kinetics of the ELSO epoxy conversion at different carbon dioxide pressures in the presence of TBAB (a) and SiO₂–(I) catalyst (b).

For NIPU synthesis CLSO and CSBO with 100% epoxy conversion were used in order to examine the role of the carbonate content. After purification and catalyst removal at 50 °C CLSO shows a viscosity of 27 000 ± 1000 mPa s and CSBO of 2500 ± 300 mPa s (Table 2). Obviously the increased carbonate content accounted for a viscosity increase by one order of magnitude. The viscosity increase for linseed with respect to soy bean products is in accordance with observations and modeling reported in the literature for acrylic derivatives of seed oils.²⁷ As a result of the higher carbonate functionality, the amount of immobilized carbon dioxide increased significantly from 15.2 wt% for CSBO to 19.6 wt% for CLSO.

Table 2 Data of the carbonated seed oils

Carbonate	Viscosity ^50°C (mPa s)	M (g mol^−1)	Carbonate (wt%)	Fixed CO2 (wt%)
a From GPC. b Measured by Brookfield viscosimeter. c Calculated from 100% epoxy conversion detected by ^1H-NMR.
CSBO	2800 ± 250^b	1648^a	20.7^c	15.2^c
CLSO	27000 ± 1000^b	1998^a	26.7^c	19.6^c

---

Preparation of the non-isocyanate polyurethanes (NIPU)

CSBO and CLSO as well as blends of both components were cured with 1,2 ethane diamine (EDA), 1,4-butane diamine (BDA) and isophorone diamine (IPDA) to produce NIPU for evaluation of thermal and mechanical properties as a function of carbonate content. Although the high carbonate content of CLSO is attractive with respect to immobilization of carbon dioxide and enhanced crosslink densities, high viscosity causes processing problems. Moreover, the higher carbonate functionality of CLSO is accompanied by much shorter gel times, especially when curing with reactive diamines such as EDA and BDA. To reduce the viscosity and increase the gel time CSBO was used as reactive diluent and was blended together with CLSO. As is apparent from Fig. 6, the viscosity correlates with carbonate content, which can be varied by blending together CSBO and CLSO. At a CLSO content of 50 wt% the viscosity was 6800 ± 500 mPas, which was the limit for processing when curing with EDA and BDA. In accord with Javni et al. the preferred amine/carbonate molar ratio was found to be 1.0. Cure was performed for 10 h at 70 °C and 3 h at 100°C until complete conversion of the carbonates took place.

Fig. 6 Uncured resin viscosity at 50 °C as a function of the carbonate content. The carbonate content was varied by mixing together CSBO with CLSO.

Fig. 7 IR spectra of CSBO before (bottom) and after (top) cure (10 h 70 °C, 3 h 100 °C) with EDA.

In contrast of EDA and BDA, the cyclic amine IPDA contains two amine groups with very different reactivities. The amine bonded to the primary C-atom is much more reactive with respect to the amine group at the secondary C-atom. In order to achieve complete conversion, the cure cycle was prolonged by post cure at 70 °C. As illustrated in Fig. 7, the cure reaction was monitored by IR-spectroscopy monitoring the disappearance of signals at 1803 cm⁻¹, typical for cyclic carbonate, and the appearance of new signals at 1710, 1654, 1540, and 3337 cm⁻¹, typical for carbonyl absorptions in urethane groups, N–H deformation of the urethane group, and hydroxyl groups, respectively. All curing reactions gave complete conversion of the cyclic carbonate groups.

Fig. 8 Swelling-time diagrams of NIPUs made of CLSO/CSBO: (a) solvent water, NIPU cured with EDA; (b) solvent toluene, NIPU cured with EDA; (c) solvent water, NIPU cured with BDA; (d) solvent toluene, NIPU cured with BDA; (e) solvent water, NIPU cured with IPDA; (f) solvent toluene, NIPU cured with IPDA.

Mechanical properties

The mechanical NIPU properties as a function of the carbonate content and CSBO/CLSO mixing ratio, respectively, are listed in Table 3. For all NIPUs cured with EDA and BDA glass temperature (T_g), Young's Modulus and tensile strengths increase with increasing carbonate content, whereas elongation at break decreases. In contrast to EDA the more flexible BDA lowers glass temperature and stiffness, as expected for reduced crosslink density and less hydrogen bridging. In contrast, the IPDA cure affords much higher stiffness with Young's modulus improved by three orders of magnitude when increasing the carbonate content. Moreover, it was possible to increase glass temperature up to 60 °C, as measured for CLSO/IPDA without adding CSBO. However, this drastic property improvement was accompanied by severe embrittlement, as indicated by losses of elongation at break. In contrast to glass temperature, elongation at break and stiffness, the tensile strength of IPDA-based NIPU were much less affected by the carbonate content. This indicated that the cross linking density did not increase in the same way like the NIPUs cured with EDA and BDA.

Table 3 Mechanical properties of synthesized NIPUs

Amine	CLO/CSBO (wt%)	Carbonate content^a (wt%)	T g ^b (°C)	E-modulus (MPa)	σ yield (MPa)	ε break (%)
EDA = 1,2-ethane diamine; BDA = 1,4-butane diamine; IPDA = isophorone diamine.a Carbonate content of the carbonated seed oil mixture.b Measured by DMA with tan δ method.
EDA	0/100	20.8	20	4 ± 1	6 ± 1	240 ± 30
	20/80	22.0	24	6 ± 1	6 ± 1	220 ± 20
	33/67	22.8	29	25 ± 5	9 ± 1	180 ± 20
	43/57	23.3	33	60 ± 10	11 ± 2	140 ± 20
	50/50	23.8	33	80 ± 10	14 ± 2	140 ± 30
	100/0	26.7	55	180 ± 30	18 ± 2	57 ± 20
BDA	0/100	20.8	17	2 ± 1	2 ± 0	310 ± 10
	20/80	22.0	23	7 ± 1	5 ± 1	250 ± 20
	33/67	22.8	23	15 ± 1	8 ± 1	240 ± 20
	43/57	23.3	25	26 ± 3	11 ± 0	240 ± 10
	50/50	23.8	28	42 ± 3	12 ± 2	190 ± 30
	100/0	26.7	45	300 ± 50	17 ± 2	84 ± 30
IPDA	0/100	20.8	40	50 ± 10	5 ± 0	200 ± 30
	20/80	22.0	40	80 ± 10	4 ± 1	60 ± 20
	33/67	22.8	40	120 ± 10	5 ± 1	30 ± 10
	43/57	23.3	45	280 ± 20	4 ± 1	2 ± 2
	50/50	23.8	50	480 ± 50	4 ± 2	1 ± 0
	60/40	24.4	52	720 ± 70	7 ± 3	2 ± 1
	67/33	24.8	52	900 ± 100	6 ± 4	1 ± 1
	80/20	25.5	52	1170 ± 80	7 ± 5	1 ± 1
	100/0	26.7	60	1460 ± 120	10 ± 3	1 ± 0

---

Solvent swelling

Swelling as a function of carbonate content of the monomers, hydroxyl functionality, and crosslink density, respectively, was examined by immersing NIPU samples in water and toluene. The question of water uptake is critical with respect to NIPU applications because NIPU have high hydroxyl group content, increasing with increasing carbonate content. The results are illustrated in Fig. 8. The water uptake is rather low and falls in range of that typical for conventional polyurethanes and polyamide materials. It increases with increasing carbonate content, which accounts for polarity increase and higher hydroxyl group content. Highest water absorption was observed for the EDA-based NIPU which has highest hydroxyl group content among the prepared NIPU families. When the more hydrophobic IPDA was incorporated into NIPU, water absorption was reduced markedly, as expected for rendering NIPU more hydrophobic in the presence of the cycloaliphatic segments. In contrast to water absorption, toluene uptake is much faster and more extensive with swelling equilibria reached within a few days. Especially IPDA/NIPU exhibited massive toluene swelling and even weight-loss. Most likely, toluene soluble side products, e.g., resulting from chain scission by aminolysis contribute to swelling and weight loss. Javni et al. assumed that a different compounds like mono-, diglycerides, free glycerol as fatty acid amides could be soluble in toluene. Moreover, macro cycles appear to be feasible due to the two-stage cure reaction of IPDA with two amine groups of very different reactivity.

Conclusions

Catalytic conversion of epoxide groups with carbon dioxide was used to prepare the carbonated linseed oil CLSO. In addition to the homogeneous TBAB catalyst, the silica-supported alkylpyridinium iodide SiO₂–(I) was employed successfully to catalyze this carbon dioxide conversion reaction and to enable facile catalyst recovery by filtration, thus eliminating solvent extraction. Although quantitative conversion was achieved with both homogeneous and heterogeneous alkylammonium catalysts more research is needed to improve the catalyst activity of supported catalysts. In comparison to soy-based CSBO, the carbonate content of carbonated seed oils was increased substantially, as reflected by increasing the content of immobilized carbon dioxide from 15.2 wt% for CSBO to 19.6 wt% for CLSO. Mixing together CSBO and CLSO enabled variation of carbonate content in order to explore the influence of curing agents and the carbonate content on the thermal and mechanical properties of NIPU. With increasing crosslink density stiffness, and glass transition temperature increase significantly. Using the cycloaliphatic curing agent IPDA for curing CLSO, the glass transition temperature increased from 20 to 60 °C, while the stiffness, as reflected by the Young's modulus, increased three orders of magnitude at the expense of elongation at break. While most conventional CSBO-based systems are rather soft and flexible, this approach indicates that NIPU can be developed to meet the demands of engineering applications with higher dimensional stability and stiffness. In conclusion, CLSO represents an interesting formulation component for NIPU based upon renewable resources as resource effective class of materials and approach to immobilize carbon dioxide.

Acknowledgements

The authors gratefully acknowledge financial support by Volkswagen. We would like to thank as well our co-workers Roman Erath and Mathias Menzel for their enthusiastic support.

References

1. G. L. Wilkes, S. Sohn, and B. Tamami, US20040230009A1, 2004.
2. B. Tamami, S. Sohn and G. L. Wilkes, J. Appl. Polym. Sci., 2004, 92, 883.
3. G. L. Wilkes, S. Sohn, and B. Tamami, US10780909, 2006.
4. J. H. Clements, Ind. Eng. Chem. Res., 2003, 42, 663–674.
5. K. Kem, US4261922, 1981.
6. P. Strege, and G. Doorakian, US4310708, 1982.
7. S. King, US5104982, 1992.
8. P. Tundo, A. Perosa and F. Zecchini, Methods and Reagents for green Chemistry, Wiley, New Jersey, 2007.
9. J. Malkemus, US2627524, 1953.
10. G. Rokicki and M. Lewandowski, Angew. Makromol. Chem., 1987, 148, 53–66.
11. T. Bürgel and M. Fedtke, Polym. Bull., 1991, 27, 171–177.
12. H. Tomita, F. Sanda and T. Endo, J. Polym. Sci., Part A: Polym. Chem., 2001, 39, 3678–3685.
13. O. L. Figovsky and L. D. Shapovalov, Macromol. Symp., 2002, 187, 325.
14. I. Javni, D. P. Hong and Z. S. Petrovic', J. Appl. Polym. Sci., 2008, 108, 3867–3875.
15. H. Baumann, M. Bühler, H. Fochem, F. Hirsinger, H. Zoebelein and J. Falbe, Angew. Chem., 1988, 100, 41–62.
16. K. D. Carlson, R. Kleiman and M. O. Bagby, J. Am. Oil Chem. Soc., 1994, 71, 175.
17. J. Falbe, Methodicum Chimicum, Thieme Verlag, Stuttgart, 1975.
18. D. Swern, G. N. Billen, T. W. Findley and J. T. Scanlan, J. Am. Chem. Soc., 1940, 62, 2305.
19. T. W. Findley, D. Swern and J. T. Scalan, J. Am. Chem. Soc., 1945, 67, 412.
20. A. Campanella, M. A. Baltanás, M. C. Capel-Sánchez, J. M. Campos-Martín and J. L. G. Fierro, Green Chem., 2004, 6, 330.
21. G. L. Crocco, W. F. Shum, J. G. Zajacek, and H. S. J. Kesling, US5166372, 1992.
22. K. M. Doll and S. Z. Erhan, Green Chem., 2005, 7, 849–854.
23. Z. Li, Y. Zhao, S. Yan, X. Wang, M. Kang, J. Wang and H. Xiang, Catal. Lett., 2008, 123, 246–251.
24. K. Motokura, S. Itagaki, Y. Iwasawa, A. Miyaji and T. Baba, Green Chem., 2009, 11, 1876–1880.
25. T. Takahashi, T. Watahiki, S. Kitazume, H. Yasuda and T. Sakakura, Chem. Commun., 2006, 1664–1666.
26. R. A. Holser, J. Oleo Sci., 2007, 56, 629–632.
27. J. L. Scala and R. P. Wool, J. Appl. Polym. Sci., 2005, 95, 774.

---