Novel synthesis of a soy-based polyol for a polyurethane rigid foam

Abstract

A novel soy-based polyol was synthesised using a novel method. First, an allylic oxidation reaction was performed in a conventional flask. This step was followed by an epoxidation–hydroxylation reaction in a microflow system referring to the reported method, which was established by our team and could reduce the oligomerisation side reaction to improve the polyol quality. During the allylic oxidation process, hydroxyl groups were introduced to the oil main chain, which was confirmed by NMR, FTIR, and the determination of the hydroxyl number. The obtained soy-polyol labeled Polyol-a possessed a hydroxyl number of 378 mg KOH per g and a viscosity of 8825 mPa s. A corresponding soy-based polyurethane rigid foam labeled PU-a was also prepared. Compared with the previously obtained soy-based polyol without allylic oxidation treatment (i.e. Polyol-m), Polyol-a had a higher hydroxyl number. In addition, PU-a had a fine, uniform, and closed-cell morphology and exhibited excellent compression strength, thermal insulation, dimensional stability, and thermal stability. These properties were attributed to the high hydroxyl number of the soy-polyol, which created a high degree of crosslinking.

---

Introduction

Polyurethane (PU) is one of the most versatile polymers with a wide range of applications.^1,2 PUs have repeating urethane linkages in the main chain and are commonly synthesized by reacting polyisocyanates with polyols.^3,4 Over the past decade, increasing environmental concerns and the soaring price of crude oil have led to increasing interest in the exploration of vegetable oils to synthesize bio-based polyols for the production of PUs.^3,5 Normally, the routes for preparing bio-based polyols from vegetable oils involve hydroformylation, transesterification, ozonolysis followed by hydrogenation, and epoxidation followed by hydroxylation; the epoxidation–hydroxylation method is most frequently used due to it being a simple reaction and easy to control, as evidenced by the fact that the relevant products have already been commercialized.^6,7

Polyurethane rigid foam (PURF) accounts for approximately 23% of all PU production.⁶ Currently, the polyols that are applied to PURFs are mainly based on the epoxidation–hydroxylation method. The polyols generally have lower hydroxyl numbers compared with those of petrochemical polyols for PURF use, with the main drawbacks being the limited the amount of bio-based polyol in the foam preparation and a weakening of the foam properties. Researchers have increased the hydroxyl number via a multifunctional ring opener⁸ and further transesterification/transamination reactions,^9–12 along with several other advancements that have been made.

Microreaction technology is expected to make a revolutionary change in chemical synthesis. Based on process intensification, an improved heat transfer efficiency and a faster reaction rate could be achieved in microreactors with a high surface-to-volume ratio.^13,14 In the early stages, we have synthesised a high quality soy-polyol labeled Polyol-m for PURFs directly from soybean oil using a continuous micro-flow system, and the corresponding soy-based PURF labeled PU-m was also prepared. The results show that the micro-flow system has advantages for the preparation of high quality polyols.¹³

In this study, we report a new approach for producing soy-polyols with higher hydroxyl numbers. Unlike the previous procedure,¹³ the soybean oil was pretreated by allylic oxidation, which made the methylenes convert to secondary alcohols and introduced hydroxyl groups to the oil main chain. Then, the hydroxyl-contained soybean oil was further processed by epoxidation and hydroxylation in a microflow system referring to the reported method,¹³ and a soy-polyol with a new structure and a higher hydroxyl number was obtained (Schemes 1 and 2). In addition, a soy-based PURF was also prepared by reacting isocyanates with polyols containing a mixture of the soy-polyol and petroleum-based polyols. The thermal, physical, and mechanical properties of the foam were characterized and compared with PU-m which was prepared from Polyol-m without allylic oxidation treatment.

Scheme 1 Reaction scheme for soy-polyol synthesis.

Scheme 2 Synthesis of a novel soy-based polyol.

Experimental

Materials

Selenium dioxide, tert-butyl hydroperoxide (TBHP, 70%), hydrogen peroxide (30%), formic acid (98%), sulphuric acid (98%), and ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA-2Na) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All chemicals were reagent grade or better. FULINMEN® soybean oil with an iodine value of 130 mg I₂ per 100 g was supplied by COFCO (Beijing, China). WANNATE® PM-200, a polymeric diphenyl methane diisocyanate (MDI) with 30.5 wt% NCO, was obtained from Wanhua Chemical Group Co., Ltd. (Yantai, China). H4110III, H6305SA, and H403 are petroleum-based polyols with hydroxyl numbers of 430, 460, and 740 mg KOH per g, respectively, and were obtained by Nanjing Hongbaoli Co., Ltd. (Nanjing, China). The blowing agent cyclopentane was obtained from MEILONG Cyclopentane Chemical Co., Ltd. (Foshan, China). The foam stabilizer AK-8803 was supplied from Jiangsu Maysta Chemical Co., Ltd. (Nanjing, China). The catalysts Polycat® 8 and Polycat® 41 were obtained from Air Products and Chemicals, Inc.

Microflow system

The micro-flow system had been employed in our previous research.¹³ It was mainly supplied by Ehrfeld Mikrotechnik BTS and consisted of sandwich microreactors, LH25 slit-plate mixers, plunger pumps, and self-designed oil–water separators, as shown in Scheme 1. The reaction temperature was regulated using a Petite Fleur dynamic temperature control device (Huber Kältemaschinenbau GmbH, Germany).

Soy-polyol synthesis

The synthesis includes two parts: the allylic oxidation of soybean oil in a conventional flask^15,16 and the epoxidation–hydroxylation in the microflow system, which is illustrated in Schemes 1 and 2.

To a solution of soybean oil (0.10 mol) in ethyl acetate (600 mL), SeO₂ (40 mmol) and tert-butyl hydroperoxide (1.20 mol) were added. The reaction mixture was stirred at 40 °C for 36 h and quenched by saturated aqueous Na₂CO₃ (300 mL) for 1 h, and Na₂SO₃ was subsequently added to reduce the excess hydroperoxide. The organic phase was separated and later washed with brine (200 mL × 2), and dried with MgSO₄. After evaporation of the solvents, allylic oxidized soybean oil was obtained.

The epoxidation–hydroxylation process followed the methods that have been reported,^13,17 including the epoxidation of hydroxyl-contained soybean oil and the subsequent hydroxylation. The molar ratio of formic acid to hydrogen peroxide was 1 : 1, whereas sulphuric acid and EDTA-2Na account for 5 and 3 wt% of the pretreated soybean oil mass, respectively. The flow rate of oil was 0.9 mL min⁻¹, while the other mixture including hydrogen peroxide, formic acid, sulphuric acid, and EDTA-2Na, flowed at a rate of 5.3 mL min⁻¹. The two flows were pumped through the mixer into the reactor, where they were held for a residence time of 6 min at 75 °C.

In the hydroxylation stage, the flow rate of allylic-epoxidized soybean oil was also set as 0.9 mL min⁻¹ because the oil volume remained largely unchanged after epoxidation. Meanwhile, the rate of flow of the sulphuric acid solution with an acid concentration of 10 wt% was set to 2.2 mL min⁻¹. The two flows were mixed and held in the reactor for a residence time of 12 min at 75 °C. The sulphuric acid solution was recyclable because of the negligible change in concentration. After post-processing including neutralization, washing, drying, and rotary evaporation, a soy-polyol labeled Polyol-a was obtained.

Foam formulation

The soy-based PURF labeled PU-a was prepared using a free-rise method as has been reported,^6,13 and follows the formulation in Table 1.

Table 1 Rigid foam formulation

Ingredients	Soy-polyol	H4110 III	H6305 SA	H403	AK-8803	Polycat® 8	Polycat® 41	KAc solution (30 wt%)	Water	Cyclopentane	Isocyanate index
Formulation (parts by weight)	50	21	24	5	2.0	2.6	1.0	0.1	1.2	13	110

---

Characterization

The hydroxyl numbers and acid numbers were measured according to ASTM E1899-08 and D4662-08, respectively. The epoxy content was analysed according to AOCS Cd 9-57 (2009). The viscosities were measured at 25 °C using an SNB-1 digital viscometer (Techcomp Jingke Scientific Instruments Co. Ltd., China) according to ASTM D4878-08. ¹H NMR was recorded on magnet system 400′54 ascend purchased from Bruker Biospin AG. ¹H NMR spectra chemical shifts (δ) are reported in parts per million (ppm) referenced to TMS (0 ppm). A Fourier transform infrared (FTIR) spectrometer (TENSOR 27, Bruker Corp., Germany) equipped with a MIRacle™ single reflection ATR accessory (PIKE Technologies, US) was used to obtain FTIR spectra. The foam morphology was studied using an S-3400N scanning electron microscope (Hitachi High Technologies Inc., Japan). The apparent foam density was measured according to ASTM D1622-08. The compression strength was tested using a CMT4204 wicket computer-controlled electronic universal testing machine (MTS Systems Co. Ltd., China) according to ASTM D1621-10. The measurement of thermal conductivity (k value) was conducted using an HFM 436 Lambda heat flow meter (Netzsch, Germany). The dimensional stability was measured according to ISO 2796:1986. The thermostability of the PURF was studied by thermogravimetrical analysis (TGA) using a TG 209 F1 Thermogravimetric Analyser (Netzsch, Germany) at a heating rate of 10 °C min⁻¹ from 25 to 800 °C under nitrogen atmosphere. The dynamic properties of the PURF was evaluated by a Dynamic Mechanical Analyser (DMA Q800, TA Instruments, USA); a disk (15 mm diameter × 10 mm thickness) was required and tested in compression mode. The storage modulus (G′) and loss modulus (G′′) were recorded at 1 Hz and 0.1% strain over a temperature range of 25 °C to 220 °C and at a temperature ramp rate of 3 °C min⁻¹.

Results and discussion

Properties of the soybean oil derivatives

By comparing the ¹H NMR spectra of the allylic oxidized product with soybean oil, the hydroxyl peak at δ 4.49 ppm of chemical shifts was detected (Fig. S1 and S2, ESI†). The FTIR spectra of soybean oil and its derivatives are shown in Fig. 1. The peaks at 2927, 2855, 1461, and 724 cm⁻¹ were all assigned to –CH₂ groups, while the bands at 1743 and 1240 cm⁻¹ were assigned to C=O and C–O stretching vibrations, respectively. For soybean oil, the peak appearing at 3009 cm⁻¹ was assigned to the C–H stretching vibrations of unsaturated double bonds; after allylic oxidation, this peak remained and a hydroxyl peak at 3400 cm⁻¹ appeared, as well as a peak at 1098 cm⁻¹, which was assigned to secondary hydroxyl groups. Combined with the NMR analysis, the change of the spectrum for allylic oxidized soybean oil indicated that hydroxyl groups were introduced to the oil main chain. For allylic-epoxidized soybean oil, a weak peak attributed to the epoxy group at 823 cm⁻¹ appeared, which disappeared after the hydroxylation, as noted for Polyol-a. In addition, the gradual increase in the intensity of the hydroxyl peak at 3400 cm⁻¹ was attributed to a slight ring-opening side reaction that occurred during the epoxidation of the allylic-epoxidized soybean oil and the complete opening reaction for Polyol-a.

Fig. 1 FTIR spectra of soybean oil and its derivatives.

As shown in Table 2, the oil possessed a hydroxyl number of 135 mg KOH per g after the allylic oxidation, indicating that hydroxyl groups were successfully introduced to the oil main chain. The minor increase in the hydroxyl number of the allylic-epoxidized soybean oil was due to the ring opening side reaction that accompanied the epoxidation process. The trend in the change of hydroxyl numbers was consistent with the infrared analysis. Compared with Polyol-m, Polyol-a possessed a higher hydroxyl number and a slightly increased viscosity. The alteration of viscosities was similar to that of the hydroxyl numbers, which resulted from the oligomers introduced by oligomerisation and is a consequence of the gradually enhanced intermolecular forces from increasing amounts of hydroxyl groups.

Table 2 Properties of the soybean oil derivatives

Sample	Hydroxyl number (mg KOH per g)	Acid number (mg KOH per g)	Epoxy content (%)	Viscosity (mPa s)
Allylic oxidized soybean oil	135	0.1	—	615
Allylic-epoxidized soybean oil	169	0.5	7.0	2050
Polyol-a	378	0.6	<0.1	8825
Polyol-m	300	1.5	0.68	7550

---

Performances of the soy-based foam

The foam morphology was studied by scanning electron microscopy (SEM) and the images of the foam cross-section are shown in Fig. 2. PU-a had polygon-shaped closed-cell structures; the cell structure was uniform, in spite of some negligible broken-walled cells. The polyol viscosity plays an important role in the formation of the foam cellular structure, and the moderate viscosity of the soy-polyol is beneficial to the process of cell nucleation.

Fig. 2 Scanning electron micrographs of PU-a.

The primary properties of foams are summarised in Table 3. As seen in Table 3, PU-a exhibited more excellent compression strength than PU-m; although the differences in thermal insulation and dimensional stability of the two foams were not obvious, the properties exhibited well compared with commercialized products. The good performances were attributed to the uniform cell structure and high hydroxyl number of Polyol-a, which created a high degree of crosslinking.

Table 3 Foam primary properties

Foam	Density (kg m^−3)	Compression strength (kPa)	k value (mW (m K)^−1)	Dimensional stability (%)			Tg (°C)
				Length	Width	Thickness
PU-a	37.5	205	20.0	0.4	0.4	0.3	161.1
PU-m	36.2	187	20.1	0.4	0.3	−0.6	155.3

---

The thermal stability of the soy-based foams was evaluated by TGA. The decomposition of the foams mainly occurred in two steps: first, the degradation of urethane bonds started at approximately 180 °C. Next, the polyols contributed to degradation at higher temperatures.^2,18 As shown in Fig. 3, PU-a exhibited less weight losses than PU-m below 320 °C, which was attributed to the higher hydroxyl number of Polyol-a; in other temperature ranges, PU-a exhibited worse thermal stability which may be caused by the more apparent thermolability of Polyol-a.

Fig. 3 TGA and derivative TGA curves of the soy-based foams.

Fig. 4 shows the thermo-mechanical properties of the soy-based foams. A tan δ curve, denoted as a visible peak, showed the transition of a foam from its glassy state to a rubbery state. Combined with the information in Table 3, PU-a exhibited a higher glass-transition temperature (T_g), which was also attributed to a higher cross-linking density.

Fig. 4 tan δ vs. temperature of the soy-based foams.

Conclusions

A soy-based polyol was successfully synthesised using a novel method. First, allylic oxidation in a conventional flask was followed by epoxidation–hydroxylation in the microflow system. The corresponding soy-based PU rigid foam was also prepared by reacting isocyanates with polyols, containing a mixture of the soy-polyol and petroleum-based polyols. The foam, namely PU-a, exhibited polygon-shaped and uniform closed-cell structures, and it contained a high cross-linking content, which was due to characteristics of the soy-polyol such as high hydroxyl number and moderate viscosity. Finally, the foam exhibited excellent compression strength, thermal insulation, dimensional stability, and thermal stability.

Acknowledgements

The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (No. 21522604, U1463201 and 21402240), the Natural Science Foundation of Jiangsu Province (No. BK20150031 and BY2014005-03), and a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

References

1. M. A. Mosiewicki, P. Rojek, Sł. Michałowski, M. I. Aranguren and A. Prociak, J. Appl. Polym. Sci., 2015, 132, 41602.
2. A. A. Septevani, D. A. C. Evans, C. Chaleat, D. J. Martin and P. K. Annamalai, Ind. Crops Prod., 2015, 66, 16–26.
3. R. Chen, C. Zhang and M. R. Kessler, J. Appl. Polym. Sci., 2015, 132, 41213.
4. M. Heinen, A. E. Gerbase and C. L. Petzhold, Polym. Degrad. Stab., 2014, 108, 76–86.
5. C. Zhang and M. R. Kessler, ACS Sustainable Chem. Eng., 2015, 3, 743–749.
6. D. Ji, Z. Fang, W. He, Z. Luo, X. Jiang, T. Wang and K. Guo, Ind. Crops Prod., 2015, 74, 76–82.
7. M. Desroches, M. Escouvois, R. Auvergne, S. Caillol and B. Boutevin, Polym. Rev., 2012, 52, 38–79.
8. L. T. Yang, C. S. Zhao, C. L. Dai, L. Y. Fu and S. Q. Lin, J. Polym. Environ., 2012, 20, 230–236.
9. X. Kong, G. Liu and J. M. Curtis, Eur. Polym. J., 2012, 48, 2097–2106.
10. V. B. Veronese, R. K. Menger, M. M. C. Forte and C. L. Petzhold, J. Appl. Polym. Sci., 2011, 120, 530–537.
11. A. Palanisamy, B. S. Rao and S. Mehazabeen, J. Polym. Environ., 2011, 19, 698–705.
12. D. Ji, Z. Fang, Z. Wan, H. Chen, W. He, X. Li and K. Guo, Adv. Mater. Res., 2013, 724–725, 1681–1684.
13. D. Ji, Z. Fang, W. He, K. Zhang, Z. Luo, T. Wang and K. Guo, ACS Sustainable Chem. Eng., 2015, 3, 1197–1204.
14. N. Zhu, Z. Zhang, W. Feng, Y. Zeng, Z. Li, Z. Fang, K. Zhang, Z. Li and K. Guo, RSC Adv., 2015, 5, 31554–31557.
15. M. A. Warpehoski, B. Chabaud and K. B. Sharpless, J. Org. Chem., 1982, 47, 2897–2900.
16. M.-E. J. Garcia, U. Fröhlich and U. Koert, Eur. J. Org. Chem., 2007, 2007, 1991–1999.
17. W. He, Z. Fang, D. Ji, K. Chen, Z. Wan, X. Li, H. Gan, S. Tang, K. Zhang and K. Guo, Org. Process Res. Dev., 2013, 17, 1137–1141.
18. X. Luo, A. Mohanty and M. Misra, Ind. Crops Prod., 2013, 47, 13–19.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20855j

---