Study on the performance mechanism of methacrylate pour point depressant in soybean biodiesel blends

Abstract

The cold flow properties of soybean biodiesel blends with 0# diesel and methacrylate pour point depressant (10-320) were investigated in this study. Results showed that all of the fuel properties of the prepared biodiesel satisfied the qualification for ASTM D6751. Along with the increasing amount of pour point depressant (PPD) addition, the cold filter plugging point (CFPP) of the system was reduced. The addition of a constant amount of pour point depressant resulted in minimal values of CFPP in the system, as the volume ratio of biodiesel varied from 0% to 100%. The addition of 1% of 10-320 PPD in B60 produced a CFPP of −10 °C, which was the most significant reduction in biodiesel blends. Other fuel properties of B60 were in accordance with ASTM D6751 and EN 14214. Differential scanning calorimeter, low-temperature X-ray diffraction, and polarizing optical microscope were used to investigate the performance mechanism. Results showed that the crystallization rate of B60 was the slowest; the crystals were small, spherical particles; the particle size <20 μm; and the energy of the solid–liquid transition was the lowest. The crystal content in B60 was the least, the ratio of the peak area of the orthorhombic and monoclinic crystals was 0.086. The orthorhombic crystals were relatively less than the monoclinic crystals. As monoclinic crystals could easily pass through the filter, the CFPP of B60 was the lowest.

---

1 Introduction

Biodiesel is a renewable, biodegradable, and non-toxic fuel.^1–5 Biodiesel production is an intense research area because of the rapidly depleting energy reserves and fluctuating petroleum prices.^6–14 This fuel has many advantages, such as renewability, excellent environmental performance, biodegradability, and good combustion characteristic.^4,5,15,16 However, biodiesel consists of fatty acid methyl ester (FAME), which is the carbon chain that contains 14–16 carbon atoms. The congealing points of these FAMEs are low, and these low points lead to the poor low-temperature flow performance of biodiesel. This poor performance seriously influences its utilization in winter. Crystallization of biodiesel components at a relatively higher temperature during cold season causes fuel starvation and operability problems, as solidified materials clog fuel lines and filters. Several approaches are used to improve the low-temperature properties, such as pour point depressant (PPD) addition,^17,18 blending with petroleum diesel,^19–22 branched-chain esters,²³ and winterization.^24–26 Among these methods, branched-chain esters and winterization are technically complex, and the effect is not obvious. Therefore, these two methods do not have much practical value in use. Adding PPD and blending with petroleum diesel are the two more practical methods. In Europe and the United States, biodiesel blending with petroleum diesel has been popularized to a certain degree. The use of biodiesel blended with petroleum diesel and PPD addition can significantly improve the low-temperature performance.²⁷ Many studies have investigated biodiesel blends, but biodiesel composition is different from that of other sources.^19,20 Moreover, theory of biodiesel mechanism with PPD remains unclear, and thus studies on biodiesel mechanism with PPD at low temperature are needed. According to previous works, poly methacrylate have proved to be an effective cold flow improver for biodiesel and diesel.^17,18,28 Additionally, there is paucity of technical data in previous papers explicitly presenting the effects of methacrylate PPD on the cold flow properties of biodiesel. Thus, methacrylate PPD was selected to examine the effect of cold flow properties on soybean oil derived biodiesel blends in this work. Using standard methods, cloud point (CP), pour point (PP), cold filter plugging point (CFPP), induction period (IP), flash point (FP), kinematic viscosity (ν), and acid value (AV) were determined. Table 2 shows the comparative results of the prepared biodiesel and the biodiesel fuel standard ASTM D6751. Differential scanning calorimeter (DSC), polarizing optical microscope, and low-temperature X-ray diffraction (XRD) were used to investigate the performance mechanism, which is important to further develop new biodiesel PPDs.

Table 1 GC-MS analysis of soybean biodiesel

Name of fatty acid methyl esters (FAME)	Corresponding acid	Mass percent (%)
Methyl hexadecanoate	16 : 0	11.28
Methyl octadecanoate	18 : 0	5.35
Methyl oleate	18 : 1	25.17
Methyl linoleate	18 : 2	47.56
Methyl linolenate	18 : 3	9.15
Methyl eicosenoate	20 : 0	0.78
Saturated FAME	—	17.41
Unsaturated FAME	—	81.88

---

Table 2 Standard of biodiesel and physico-chemical properties of pure soybean biodiesel

Properties	ASTM D6751	Soybean biodiesel
AV, mg KOH per g	0.50 max	0.38
Free glycerol, mass%	0.020 max	0.01
Total glycerol, mass%	0.240 max	0.19
CP, °C	Report	−1
PP, °C	—	−5
CFPP, °C	—	−3
FP, °C	130 min	135
IP, 110 °C, h	3 min	5.1
ν, 40 °C, mm^2 s^−1	1.9–6.0	4.2

---

2 Experimental

2.1 Reagents and instruments

Soybean biodiesel (prepared in the laboratory²⁹), Hejiawan 0# diesel, and 10-320 PPD (methacrylate, Rohmax Corporation, Germany) were used in this study.

Agilent 7890A-5975c gas chromatography-mass spectrometry (GC-MS; Agilent Corporation, USA), SYP1022-2 multifunctional low temperature tester (Shanghai Boli Instrument Co. Ltd), DSC (DSC27HP, Mettler Corporation, Switzerland), X'Pert PRD XRD (PANalytical Corporation, the Netherlands), and polarizing optical microscope (Linkam Corporation, UK) were also used.

2.2 Experimental procedure

2.2.1 GC-MS of biodiesel. Shimadzu QP 2010/Plus GC-MS was used to analyze the biodiesel composition. The GC operation conditions were as follows: HP-Inn wax quartz capillary column (60 m × 0.25 mm × 0.25 μm); the temperature of the capillary column was first increased by 10 °C min⁻¹ from 60 °C to 150 °C and subsequently increased by 5 °C min⁻¹ from 150 °C to 230 °C; interface temperature was 250 °C; injector temperature was 250 °C; diffluent ratio was 100 : 1; high-purity helium carrier gas flow rate was 1 mL min⁻¹; and injection volume was 0.2 μL.

2.2.2 General properties. The following biodiesel properties were measured according to standard methods: AV (mg KOH per g): AOCS Cd 3d-63; CP (°C): ASTM D5773; PP (°C): ASTM D5949; IP (h, 110 °C): EN 14112; ν (mm² s⁻¹): ASTM D445; FP (°C, duplicates): ASTM D93; free glycerol (FG, mass%); and total glycerol (TG, mass%): ASTM D6584. In addition, the CFPP of biodiesel was evaluated according to ASTM D6371.

2.2.3 Differential scanning calorimeter. The DSC operating conditions were as follows: samples of 8–10 mg were placed in standard crucibles. Transition temperatures and enthalpies were determined by computer during the heating cycle at a scanning rate of 5 °C min⁻¹ and temperature ranging from 30 °C to −60 °C.³⁰

2.2.4 Polarizing optical microscope. A drop of sample was placed on a slide. Observation was performed using a polarizing optical microscope at a cooling rate of 10 °C min⁻¹. When crystal deposition was found in the biodiesel, the microscope focus was adjusted until clear observation. Subsequently, cooling was stopped. The temperature of the sample was established at room temperature at a cooling rate of 0.8 °C min⁻¹.

2.2.5 Low-temperature XRD. The XRD operating conditions were as follows: the tube voltage and current were set to 40 kV and 40 mA, respectively, and graphite monochromator Cu Kα radiation (λ = 1.542 Å) was selected.³¹

3 Results and discussion

3.1 Biodiesel composition

The main fatty acid methyl esters (FAME) composition and their mass percent in prepared soybean biodiesel are shown in Table 1. The prepared soybean biodiesel is mainly composed of methyl hexadecanoate (11.28%), methyl octadecanoate (5.35%), methyl oleate (25.17%), methyl linoleate (47.56%), methyl linolenate (9.15%), and methyl eicosenoate (0.78%). The saturated and unsaturated fatty acid methyl ester contents are 17.41% and 81.88%, respectively. The linoleic acid methyl ester content is the highest at 47.56%, and methyl eicosenoate content at only 0.78% is the lowest.

3.2 Fuel properties of soybean biodiesel

As shown in Table 2, all the fuel properties of the prepared soybean biodiesel well meet the specifications from ASTM D6751.

3.3 Effect of PPD on soybean biodiesel blends

Overall, along with the increasing amount of PPD added, the CFPP of the system was reduced. The additional constant amount of PPD resulted in minimal values for the CFPP of the system, as the volume ratio of biodiesel varied from 0% to 100% (Table 3). This result shows that blending biodiesel is more sensitive to PPD than pure biodiesel or pure petroleum diesel. Adding 1% 10-320 PPD to B60 produced a CFPP of −10 °C, which is the most significant reduction in biodiesel blends.

Table 3 The impact of PPD on the CFPP of soybean biodiesel blends^a

Sample	PPD content (wt%)
	0	0.2	0.4	0.6	0.8	1
a Note: B0 = pure 0# diesel, B20 = 20 vol% biodiesel + 80 vol% 0# diesel, B40 = 40 vol% biodiesel + 60 vol% 0# diesel, B60 = 60 vol% biodiesel + 40 vol% 0# diesel, B80 = 80 vol% biodiesel + 20 vol% 0# diesel, B100 = pure biodiesel.
B0	−1	1	−1	−2	−1	−1
B20	−1	−1	0	0	0	−1
B40	−1	−4	−5	−5	−6	−6
B60	−2	−7	−8	−9	−9	−10
B80	−4	−4	−5	−7	−8	−9
B100	−3	−3	−4	−4	−4	−5

---

10-320 PPD has no effect on the CFPP of B0 and B20 at about −1 °C. This observation indicated that the sensitivity of 10-320 PPD on petroleum and blends with petroleum-based was poor. The CFPP of B40, B60, B80, and B100 decreased with the increasing PPD content. Adding 1% additive produced the lowest CFPP. This finding implies that adding PPD is helpful for CFPP reduction, but adding too much will increase the cost. The CFPP of biodiesel blends without PPD was stable between −1 °C and −4 °C because of the mixed effect of biodiesel and petroleum diesel. 10-320 PPD hardly affected the CFPP of B0 and had some effect on the CFPP of B100 at a minimum of −5 °C.

Table 4 shows the effects of 10-320 PPD on the other physico-chemical properties other than CFPP. CP and PP was decreased with increasing PPD mass percentage. Upon addition of 1% of PPD, the CP and PP of B60 were reduced by 8 and 9 °C, respectively. AV and ν were also slightly increased within the acceptable limits. The addition of 1% PPD in B60 improved the IP from 5.5 h to 6.1 h, thereby simultaneously meeting the standard of EN 14214 and ASTM D6751. The required FP in EN 14214 and ASTM D6751 are ≥101 °C and ≥130 °C, respectively. Thus, the FP of B60 with 1% PPD was 103 °C within the limits of EN 14214.

Table 4 Other fuel properties of B60 with the addition of PPD

Sample	PPD, wt%	CP, °C	PP, °C	AV, mg KOH per g	ν, 40 °C, mm^2 s^−1	IP, 110 °C, h	FP, °C
B60	0	0	−4	0.33	3.35	5.5	91
	0.2	−5	−8	0.35	3.63	5.6	93
	0.4	−5	−9	0.37	3.82	5.6	94
	0.6	−6	−11	0.38	3.99	5.8	97
	0.8	−7	−11	0.39	4.17	5.9	100
	1.0	−8	−13	0.41	4.33	6.1	102
EN 14214	—	—	—	0.50 max	3.5–5.0	6.0 min	101 min
ASTM D6751	—	—	—	0.50 max	1.9–6.0	3 min	130 min

---

To explore the reason why biodiesel blends have better cold flow properties, DSC, low-temperature XRD, and polarizing optical microscope were used to examine the performance mechanism.

3.4 Differential scanning calorimeter

DSC can quantitatively analyze energetic change in the phase change process in biodiesel blends with PPD. Thus, samples of B0, B60, and B100 with 1% PPD were tested by DSC. Fig. 1 and Table 4 show the DSC curves and data analysis, respectively.

Fig. 1 DSC curves of B0, B100, and B60 with 1% of PPD.

Srivastava³² and other researchers^31,33–36 considered the vertical axis to be the heat flow rate and the horizontal axis to be the temperature in DSC. The starting temperature of the peak (onset) in the curves indicates the starting precipitation temperature of wax crystals in diesel; the slope of the peak indicates the rate of precipitation of wax crystals in diesel; and the solid–liquid phase change energy (ΔH) indicates the stability of dispersion. The starting temperature of the peak of B60 (onset) is −1.634 °C between B0 (−0.284 °C) and B100 (−1.989 °C) because it is a mixture of B0 and B100. Two crystals precipitate in low temperature. The peak temperature (peak) is consistent with the temperature of the starting peak (onset). This finding indicates that the time from the beginning to the peak of the three diesel types is not significantly different. The absolute ΔH value of B60 is the least at 0.579 J g⁻¹. This finding indicates that the solid–liquid phase change energy of biodiesel blends is small, and the dispersion is more stable. The area of crystallization peak in B60 is the least at 5.78, which is far less than B0 and B100, and therefore its crystal content is the least (Table 4). From the slope of the crystallization peak, the crystallization rate of B100 is fast, whereas that of B0 is slow. The crystallization rate of B60 is the slowest. In conclusion, in B60 with PPD, the crystallization rate of wax crystals is slow, the crystal content is decreased, and the solid–liquid phase change energy is lessened (Table 5).

Table 5 Analysis of DSC on B0, B100, B60 with 1% of PPD

Sample	Onset (°C)	Peak (°C)	ΔH (J g^−1)	Area
B0	−0.284	−1.634	−2.463	23.64
B100	−1.989	−3.426	−1.275	11.96
B60	−1.634	−3.383	−0.579	5.78

---

3.5 X-ray diffraction

XRD can be used to analyze the lattice parameter and the structure of wax crystals in biodiesel.^31,34,35 However, the CFPP of biodiesel with PPD is low (−10 °C), and thus using the traditional XRD for analysis is difficult. Moreover, the temperature control is not precise, and the growth of wax crystals at low temperatures can only be roughly analyzed, adverse to the in-depth study of the mechanism. Therefore, we cooperated with the Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences to analyze the mechanism of biodiesel blends with low-temperature XRD.

The diffusion peak at 10–30° is obvious in biodiesel and 0# diesel. Sharp orthorhombic diffraction peaks occur at 21.4° and 23.9°, and sharp monoclinic diffraction peaks occur at 42.9°, 44.0°, 50.0°, 51.3°, and 52.3°. These findings indicate that amorphous and crystalline waxes are precipitated at low temperatures (Fig. 2). The diffusion peak in diesel, biodiesel, and biodiesel blends is broad, narrow, and moderate, respectively. This finding indicates that the content of amorphous wax in diesel is greater than that in biodiesel. According to previous investigations,^33,37 the XRD data were analyzed, and the results are shown in Table 6. The crystallization peak area of the three diesel types is in the order of B0 (42 879) > B100 (35 692) > B60 (29 690), which indicates that the crystal content in B60 decreases in the same temperature. The ratio of the orthorhombic area to the monoclinic peak area in B0 is 0.241 and that in B100 is 0.080. This finding shows that PPD changed the ratio of the two crystal types in B100 by co-crystallization. The content of the orthorhombic crystal precipitated is less than that of the monoclinic crystal precipitated. Gathering and growing monoclinic crystal is difficult (Table 7).³³

Fig. 2 XRD spectrums of B0, B100, and B60 with 1% of PPD at −10 °C.

Table 6 The XRD study results of B0, B100, and B60 with 1% of PPD

Sample	2-Theta	d (A)	Height	Area
B0	21.516	4.1266	1125	5562
	23.925	3.7163	403	2778
	42.961	2.1035	820	9135
	44.020	2.0554	438	5839
	50.058	1.8207	1218	12 444
	51.377	1.7770	335	4026
	52.237	1.7497	266	3095
B100	21.496	4.1304	338	909
	23.961	3.7107	382	1722
	42.963	2.1034	734	8636
	44.019	2.0554	442	5392
	50.057	1.8207	1251	13 129
	51.378	1.7770	324	3381
	52.237	1.7497	186	2523
B60	21.461	4.1370	306	1146
	23.974	3.7089	207	1195
	42.961	2.1035	671	7320
	44.019	2.0554	405	4270
	50.058	1.8206	1070	10 005
	51.379	1.7769	288	3291
	52.334	1.7467	213	2463

---

Table 7 Data analysis of XRD on B0, B100, B60 with 1% of PPD^a

B0		B100		B60
S	A	S	A	S	A
a Note: S indicates the relative area (orthorhombic area/monoclinic area); A indicates the total area of crystallization peak.
0.241	42 879	0.080	35 692	0.086	29 690

---

3.6 Polarizing optical microscope

A polarizing optical microscope can be used to observe the changes in wax crystals in biodiesel blends with and without PPD.^33,34,36,38 Thus, B0, B100 and B60 treated with or without 1% PPD were examined at −10 °C and 100× magnification (Fig. 3).

Fig. 3 Polarizing optical microscopy images of crystal morphologies of B0, B100 and B60 with or without PPD at −10 °C.

As shown in Fig. 3, before adding the PPD into the B0, B100 and B60 blends, the wax crystals have assumed in rod-like or strip-like shaped (Fig. 3a–c). All those crystals were characterized by their large number, particle size and concentrated distribution, which made them easy to form the wax crystal in a three-dimensional network structure by cross-linking, thereby losing their fluidity.

After adding 1% PPD, wax crystal changed in morphology. Many regulated group-like wax crystals, which had a particle size of about 50–100 μm, were presented in B0 at −10 °C, and they were gathered compactly. A number of spherical-like wax crystals were observed in B100, with particle size of about 20–50 μm. Particularly in B60 with the addition of 1% PPD, the particle size and amount of spherical-like wax crystals is the smallest (<20 μm), which can be obviously observed in Fig. 3. Thus, blending biodiesel (B60) was more sensitive to methacrylate PPD than biodiesel (B100) or petroleum diesel (B0). PPD effectively changed the crystallization behavior of wax crystals via transforming the wax crystal shape and restraining the formation of larger crystal.

3.7 Proposed performance mechanism

To explain the mechanism of why blending biodiesel was more sensitive to methacrylate PPD than biodiesel or petroleum diesel, we constructed the schematic diagram shown in Fig. 4 as follows.^28,34,38,39 In the figure, BD and PD are the crystals in pure biodiesel and petroleum diesel at low temperature, respectively. PPD changed the process of crystal growth, and the small crystal size was maintained by co-crystallization in B60. The compositions of diesel and biodiesel are different. Diesel is composed of n-alkanes, isoparaffin, pectin, and asphalt. Conversely, biodiesel is composed of the FAME of different carbon chains and saturation compositions. Two kinds of crystals were precipitated together at low temperatures. One crystal scattered around another crystal and produced some exclusion between diesel and biodiesel that inhibited aggregation and growth. The crystal size in the system was small. The crystals could pass more easily through the filter, thus causing a lower CFPP. The ratio of the orthorhombic area to the monoclinic peak area in B60 is similar to that in B100 at 0.086. This finding implies that co-crystallization in the two systems is similar. Furthermore, the total area of crystallization and crystal content were reduced, and the crystals uniformly dispersed in the system remained small in size. In sum, the CFPP of B60 is the lowest.

Fig. 4 The performance mechanism of PPD in the soybean biodiesel blends.

4 Conclusions

(1) The effect of methylacrylate PPD 10-320 on the cold flow properties of biodiesel blends was investigated. The results show that adding 1% additive to B60 produced a CFPP of −10 °C.

(2) DSC analysis shows that in B60 with PPD, the crystallization rate of wax crystals was slow, the crystal content was decreased, and the solid–liquid phase change energy is lessened in the system at about 0.579 J g⁻¹. The system was stable, and the CFPP was the lowest.

(3) According to the low-temperature XRD analysis, the crystal content in B60 was less, and the ratio of the orthorhombic area to the monoclinic peak area was 0.086. The content of the orthorhombic crystal precipitated was relatively less than that of the monoclinic crystal precipitated. The monoclinic crystal could easily pass through the filter, and thus B60 had the lowest CFPP.

(4) The polarizing optical microscope showed a large number of small, spherical-like wax crystalline in B60, with a particle size <20 μm. These particles went through the filter easily, thus the CFPP was reduced. This outcome was caused by PPD changing the process of crystal growth by co-crystallization. The compositions of diesel and biodiesel were different. Thus, some exclusion between diesel and biodiesel was produced that inhibited aggregation and growth. The crystals remained small in size.

(5) The proposed performance mechanism is that PPD changes the process of crystal growth and maintains the small crystal size by co-crystallization. Moreover, the compositions of diesel and biodiesel are different. Two kinds of crystals were precipitated together at low temperatures. Thus, one crystal scattered around another crystal, and it produced some exclusion between diesel and biodiesel that inhibited aggregation and growth.

Acknowledgements

This project was supported by the Shanghai Leading Academic Discipline Project (Project Number J51503), National Natural Science Foundation of China (Project Number 20976105), Science and Technology Commission of Shanghai Municipality (Project Number 09QT1400600), ShuGuang Project (Project Number 11SG54), Innovation Program of Shanghai Municipal Education Commission (Project Number 11ZZ179) and Innovation Program of Shanghai Municipal Education Commission (Project Number 09YZ387), Shanghai Talent Development Funding (Project Number 201335).

Notes and references

1. M. Fan, J. Huang, J. Yang and P. Zhang, Appl. Energy, 2013, 108, 333–339.
2. J. Dias, J. Araújo, J. Costa, M. Alvim-Ferraz and M. Almeida, Energy, 2013, 53, 58–66.
3. H. V. Lee, Y. Taufiq-Yap, M. Hussein and R. Yunus, Energy, 2013, 49, 12–18.
4. S. Hama and A. Kondo, Bioresour. Technol., 2013, 135, 386–395.
5. A. Silitonga, H. Ong, T. Mahlia, H. Masjuki and W. Chong, Fuel, 2013, 108, 855–858.
6. M. K. Lam, K. T. Lee and A. R. Mohamed, Appl. Catal., B, 2009, 93, 134–139.
7. Y. Ali and M. Hanna, Bioresour. Technol., 1994, 50, 153–163.
8. A. Demirbaş, Energy Sources, 2002, 24, 835–841.
9. A. K. Agarwal and L. Das, J. Eng. Gas Turbines Power, 2001, 123, 440–447.
10. F. Guo, Z.-L. Xiu and Z.-X. Liang, Appl. Energy, 2012, 98, 47–52.
11. S. K. Hoekman and C. Robbins, Fuel Process. Technol., 2012, 96, 237–249.
12. M. M. Roy, W. Wang and J. Bujold, Appl. Energy, 2013, 106, 198–208.
13. F. Qiu, Y. Li, D. Yang, X. Li and P. Sun, Appl. Energy, 2011, 88, 2050–2055.
14. B. Salamatinia, A. Z. Abdullah and S. Bhatia, Fuel Process. Technol., 2012, 97, 1–8.
15. B. Wang, S. Li, S. Tian, R. Feng and Y. Meng, Fuel, 2013, 104, 698–703.
16. G. Knothe, Prog. Energy Combust. Sci., 2010, 36, 364–373.
17. J. Wang, L. Cao and S. Han, Fuel, 2014, 117, 876–881.
18. L. Cao, J. Wang, K. Liu and S. Han, Appl. Energy, 2014, 114, 18–21.
19. A. Sarin, R. Arora, N. Singh, R. Sarin, R. Malhotra and K. Kundu, Energy, 2009, 34, 2016–2021.
20. C.-W. Chiu, L. G. Schumacher and G. J. Suppes, Biomass Bioenergy, 2004, 27, 485–491.
21. D. Qi and C. Lee, J. Taiwan Inst. Chem. Eng., 2014, 45, 504–507.
22. R. Candeia, M. Silva, J. Carvalho Filho, M. Brasilino, T. Bicudo, I. Santos and A. Souza, Fuel, 2009, 88, 738–743.
23. P. V. Bhale, N. V. Deshpande and S. B. Thombre, Renewable Energy, 2009, 34, 794–800.
24. Á. Pérez, A. Casas, C. M. Fernández, M. J. Ramos and L. Rodríguez, Bioresour. Technol., 2010, 101, 7375–7381.
25. T. H. Doğan and H. Temur, Fuel, 2013, 108, 793–796.
26. M. G. Gómez, R. Howard-Hildige, J. Leahy and B. Rice, Fuel, 2002, 81, 33–39.
27. H. Tang, S. O. Salley and K. S. Ng, Fuel, 2008, 87, 3006–3017.
28. S. Han, P. Wang, Y. Wang, Y. Song and T. Ren, Process Saf. Environ. Prot., 2010, 88, 41–46.
29. M. Wu, G. Wu, Y. Han and P. Zhang, China Oils Fats, 2002, 28, 65–67.
30. M. Garcia-Perez, T. T. Adams, J. W. Goodrum, K. Das and D. P. Geller, Bioresour. Technol., 2010, 101, 6219–6224.
31. J. Zhang, C. Wu, W. Li, Y. Wang and Z. Han, Fuel, 2003, 82, 1419–1426.
32. S. Srivastava, R. Tandon, P. Verma, D. Pandey and S. Goyal, Fuel, 1995, 74, 928–931.
33. Q.-Z. Jiang, Z.-Z. Song, J.-J. Ge and F.-L. Zhao, Zhongguo Shiyou Daxue Xuebao, 2006, 30, 118–122.
34. Y. Chevalier, B. Fixari, S. Brunel, E. Marie and P. de Guio, Polym. Int., 2004, 53, 475–483.
35. W. Chen, Z. Zhao and C. Yin, Fuel, 2010, 89, 1127–1132.
36. H. Hamada, H. Kato, N. Ito, Y. Takase, H. Nanbu, S. Mishima, H. Sakaki and K. Sato, Eur. J. Lipid Sci. Technol., 2010, 112, 1323–1330.
37. H. Zhang and B.-X. Shen, Acta Pet. Sin., Pet. Process. Sect., 2006, 22, 74.
38. N. U. Soriano, V. P. Migo, K. Sato and M. Matsumura, Eur. J. Lipid Sci. Technol., 2005, 107, 689–696.
39. T. Liu, L. Fang, X. Liu and X. Zhang, Fuel, 2015, 143, 448–454.

---