Improving the cold flow properties of high-proportional waste cooking oil biodiesel blends with mixed cold flow improvers

Abstract

This study was conducted to improve the cold flow properties of biodiesel obtained from waste cooking oil. The fuel properties of biodiesel blends with 0# diesel in 10 vol% (B10), 20 vol% (B20), 30 vol% (B30), 40 vol% (B40), and 50 vol% (B50) were determined. Mixed cold flow improver (CFI) was composed of ethyl acetoacetate (EAA), iso-decyl methacrylate (EHMA) and iso-octyl methacrylate (IOMA) in various proportions. The fuel properties of B50 with mixed CFIs were also determined. Binary mixed CFIs performed the best improvement on cold flow properties of B50. With the blending of 2.5 vol% EAA and 10 vol% IOMA, the CFPP and PP of B50 with CFI were decreased by 11 °C and 12 °C, respectively, with respect to neat biodiesel. In addition, density, water content, kinematic viscosity, flash point, acid value, oxidation stability, and calorific value of formulated B50 were also determined. All the fuel properties of formulated B50 satisfied the GB19147-2009 (III) and GB19147-2013 (IV) for −10# automobile diesel fuels.

---

1. Introduction

With enduring population and economic growth, the deteriorating environment from pollution and the growing energy deficit have gradually become the hot global topics.¹ As an eco-friendly and renewable energy, biodiesel has increasingly captured the attention of scientists in recent years.

Biodiesel is mainly produced from vegetable oil such as soybean, canola, cotton, and palm, wild plants, engineered microalgae oil, animal fats, and some of waste oil from the catering industry.^2–4 Biodiesel has attracted considerable attraction because of its renewability, environmental protection, biodegradability and substitutability of petro-diesel, and it is considered to be a good substitute for petro-diesel fuels.^5,6 However, it contains a large content of saturated fatty acid esters, which are prone to form wax crystals at low temperatures. Crystal formation prevents the free flow of fuel along pipes and filters, consequently affecting the operation of engines.^7,8 Therefore, the poor cold flow properties greatly limit the application of biodiesel in cold climates. There are many methods to improve the cold flow properties of biodiesel, including the use of branched esters,^9,10 winterization,^11,12 blending petro-diesel with biodiesel,^13–15 and adding chemical additives.^16–18 Most often, blending petro-diesel with biodiesel exhibits a certain improvement on the basic fuel properties of biodiesel particular to its low temperature performance. Another option is adding chemical additives into the fuels, which are known as cold flow improvers (CFIs). Therefore, adding additives into biodiesel-diesel blend fuels seems to be a valid approach to enhance the fuel properties of biodiesel.

Waste cooking oil biodiesel, derived from the waste cooking oil from the catering industry, has the advantages of both a wider range of sources and lower cost with respect to the edible vegetable oil.² As a large energy production and consumption country, the amount of waste cooking oil generated in restaurants and homes is increasing rapidly due to the large population in China. Thus, biodiesel obtained from waste cooking oil has caused wide concern from a number of researchers, local governments, private companies, and other organizations in China in recent years.^19–21

In our previous investigation, ethyl acetoacetate (EAA) has proven to be a potential bio-based diluent depressant for improving the cold flow properties of a waste cooking oil-based biodiesel.²² Both the pour point (PP) and cold filter plugging point (CFPP) decreased by 4 °C after adding EAA at 20 vol%. Despite some reported additives being available for biodiesel, adding additives to improve biodiesel-diesel blends properties at low temperature is still a challenge and requires further research. Iso-decyl methacrylate (EHMA) and iso-octyl methacrylate (IOMA) are long-chain esters, which possess excellent compatibility with pure biodiesel that is composed of different fatty acid esters. They also have lower freezing points in contrast to pure biodiesel. Thus, adding EHMA, EAA and IOMA can effectively improve the cold flow properties without affecting other properties of biodiesel blends. Furthermore, the addition of iso-decyl methacrylate (EHMA) and iso-octyl methacrylate (IOMA) and their blends with EAA were never investigated to be an additive for biodiesel-diesel blends, particularly the high-proportional biodiesel blends.

In this study, waste cooking oil biodiesel was prepared and compared with the standards of EN 14214 and ASTM D6371. EAA, EHMA, IOMA, and these mixed CFIs were firstly introduced into the blend fuels (10%, 20%, 30%, 40%, and 50% by volume) of waste cooking oil biodiesel and 0# diesel. The effects of these CFIs on the low temperature performance of B50 (50 vol% biodiesel) were determined. Other fuel properties, such as density, water content, kinematic viscosity (υ), flash point (FP), acid value (AV), oxidation stability (OS), and calorific value (CV), were also determined. All fuel properties of formulated biodiesel fuels were compared with the GB19147-2009 (III) and GB19147-2013 (IV) standards for −10# automobile diesel fuels.

2. Experimental

2.1 Materials

Waste cooking oil was obtained from the Shanghai Zhongming Chemical Co. Ltd. (Shanghai). Hejiawan 0# diesel was obtained from a Hejiawan gas station (Shanghai). EHMA, EAA, and IOMA were purchased from Aladdin Reagent Co., Ltd. (Shanghai). Potassium hydroxide (KOH) and methanol (CH₃OH) were obtained from Shanghai Titan Scientific Co., Ltd. (Shanghai).

2.2 Transesterification of waste cooking oil

The transesterification reaction was performed in a 500 ml three-necked flask equipped with a water-cooled condenser and a magnetic stirrer. The molar ratio of methanol/oil was 8 : 1, and the mass percent of KOH/oil was 0.75 wt%. CH₃OH and KOH were first reacted for a 15 minutes, then the waste cooking oil was added to the flask and the temperature was maintained at 65 °C for 120 min. After reaction, the mixture was kept in a separating funnel for 12 h to remove the lower glycerol layer. Residual methanol was removed with a rotary evaporator. The resulting biodiesel crude was washed several times with warm deionized water and subsequently dried with anhydrous calcium chloride. The fuel properties were then compared with the ASTM D6371 and EN 14214 (Table 1) specifications.

Table 1 Standards of biodiesel fuel

Specifications	PP (°C)	CFPP (°C)	Density (20 °C, g cm^−3)	Water content (mg kg^−1)	υ (40 °C, mm^2 s^−1)	FP (°C)	AV (mg KOH/g)	IP (110 °C, h)	TG (mass%)
ASTM D6751-02	—	—	0.87–0.89	Max 500	1.9–6.0	Min 100	Max 0.50	Min 6.0	Max 0.24
EN 14214	—	—	0.86–0.90	Max 500	3.5–5.0	Min 120	Max 0.50	Min 3.0	Max 0.25

---

2.3 Mixed cold flow improvers

Methacrylate CFIs have a better shear resistance and thus are attracting increasing attention. The mixed CFIs always exhibit special effects on improving the cold flow properties of diesel/biodiesel blends. This is usually associated with the synergistic effect among different PPDs. EHMA, EAA, and IOMA were blended with different ratios (vol%) that are listed in Table 5. All the abovementioned mixed CFIs were used for the experiments.

2.4 Blend operation

Waste cooking oil biodiesel was mixed with 0# diesel before adding mixed CFIs. The biodiesel blends were prepared in the following proportions: 10% (B10), 20% (B20), 30% (B30), 40% (B40), and 50% (B50) (v/v). After adding mixed CFIs into the blends, the mixtures were stirred at 500 rpm for 30 min, and then subjected to ultrasound at 40 °C for 10 min to ensure the mixture was well-blended.^23,24 For a greater degree of accuracy, measurements were performed at least in triplicate, and average values were calculated.

2.5 Analysis

2.5.1 GC-MS analysis. GC-MS is one type of a frequently used and effective means of evaluating biodiesel composition.^25–27 In this study, GC-MS analysis was performed using a Shimadzu QP 2010/Plus GC-MS. The GC operating conditions were as follows: RXi-5Sil MS (30 m × 0.25 mm × 0.25 μm); initial capillary column temperature, 50 °C to 160 °C by 10 °C min⁻¹; final temperature 160 °C to 240 °C by 2 °C min⁻¹; interface and injector temperatures, 270 °C; gas flow rate, 1 ml min⁻¹; and injection volume, 1 μl.

2.5.2 Cold flow properties. PP is defined as the least temperature at which fuels may become pourable.⁶ CFPP is defined as a temperature at which a fuel is no longer filterable within a specified time limit. Both CFPP and PP are reported as the most important indicators of the cold flow properties of biodiesel.^16,22,28 The CFPP and PP were determined using an SYP1022-2 multifunction hypothermia tester (Shanghai Boli Instrument Co., Ltd., China) in accordance with the ASTM D6371 and ASTM D5949 standards.

2.5.3 Density and water content. Density of fuels has a significant effect on the atomization quality and nozzle range of fuel. Moreover, water content has a significant effect on the combustion properties of diesel. In this study, density and water content were measured according to the EN ISO 3675 and EN ISO 12937 standards, respectively.

2.5.4 Kinematic viscosity (υ). Kinematic viscosity (υ, mm² s⁻¹, 40 °C) is an indicator used to determine the flow properties of biodiesel fuels. υ was determined in accordance with ASTM D445-04 using the SYD-265B viscometer (Shanghai Boli Instrument Co., Ltd., China).

2.5.5 Flash point (FP). Although FP is not directly related to performance of an engine, it is of importance in connection with legal requirements and safety precautions. It is an important indicator used for evaluating the risk of oil fires.²⁸ FP is determined according to ASTM D93 using a SYP1002Z-II Pensky-Martens Closed Cup Tester (Shenkai Instrument Co., Ltd., China).

2.5.6 Acid value (AV). The acid number, as specified, is used to access the level of free fatty acids or processing acids that could be present in biodiesel fuels. Biodiesel with a high value of AV increases the fueling system deposits and the likelihood for corrosion. AV was determined using the ZD-3A automatic potentiometric titrator (Shanghai Boli Instrument Co., Ltd., China) according to the specification of AOCS Cd 3d-63.

2.5.7 Oxidation stability (OS). OS is an important indicator for fuels to resist the impacts of oxygen in the fuel properties. Biodiesel contains a large amount of unsaturated bonds that cause deterioration due to oxidation.^29,30 OS was determined using the SYP 2006 oxidation stability tester (Shanghai Boli Instrument Co., Ltd., China) according to the specification of SH/T 0175.

3. Result and discussion

3.1 Compositions of biodiesel from waste cooking oil

Waste cooking oil biodiesel is obtained via trans-esterification, with high yields (>96 wt%, with respect to oil). A biodiesel total ion chromatogram is presented in Fig. 1. The main compositions of biodiesel and their mass percentages are provided in Table 2. Waste cooking oil biodiesel mainly contains methyl hexadecanoate (19.87 wt%), methyl linoleate (29.12 wt%), and methyl oleate (38.65 wt%). It also contains less amount of methyl tetradecanoate (0.74 wt%), methyl 9-hexadecenoate (1.03 wt%), and methyl octadecanoate (6.67 wt%). These results are consistent with previous reports.^4,31

Fig. 1 Total ion chromatogram of waste cooking oil biodiesel.

Table 2 GC-MS analysis of biodiesel

Time/min	Name of fatty acid methyl esters	Corresponding acid	Components	Total/%
23.67	Methyl tetradecanoate	C14:0		0.74
30.64	Methyl 9-hexadecenoate	C16:1		1.03
31.77	Methyl hexadecanoate	C16:0		19.87
38.98	Methyl linoleate	C18:2		29.12
39.35	Methyl oleate	C18:1		38.65
40.36	Methyl octadecanoate	C18:0		6.67
∑saturated	—	—	—	27.28
∑unsaturated	—	—	—	68.80

---

3.2 Fuel properties of biodiesel and its blends

Table 3 shows the fuel properties of biodiesel, diesel, and biodiesel-diesel blends without CFIs. The prepared biodiesel (B100) meets the specifications that are listed in Table 1.

Table 3 The fuel properties of biodiesel-diesel blends

Sample	CFPP (°C)	PP (°C)	Density (20 °C, g cm^−3)	Water content (mg kg^−1)	υ (40 °C, mm^2 s^−1)	FP (°C)	AV (mg KOH/ml)	OS (total insolubles, mg/100 ml)	CV (kJ kg^−1)
B0	−4	−10	0.812	87.4	3.04	71	0.07	1.07	42 750
B10	−3	−8	0.819	124.9	3.09	76	105	1.09	40 586
B20	−3	−7	0.823	150.0	3.17	83	107	1.12	40 973
B30	−2	−7	0.829	173.4	3.32	89	0.22	1.19	41 208
B40	−2	−6	0.835	192.3	3.45	93	0.25	1.26	41 730
B50	0	−4	0.840	213.4	3.57	102	0.29	1.35	42 089
B100	2	1	0.882	277.2	4.29	142	0.37	1.98	40 223

---

By adding 0# diesel into biodiesel fuel, the CFPP and PP decreased with increasing the proportion of diesel. As the proportion of 0# diesel reached 50 vol% (B50), the CFPP dropped to 0 °C. The CFPP values of blends were within the limits of standards (Table 4) for 0# diesel, but did not satisfy the specifications for −10# automobile diesel fuels.

Table 4 Standards of automobile diesel fuels (0# and −10# diesel)

Specification	PP (°C)	CFPP^c (°C)	Density (20 °C, g cm^−3)	Water content (mg kg^−1)	υ (40 °C, mm^2 s^−1)	FP (°C)	AV (mg KOH/ml)	OS (total insolubles, mg/100 ml)
a GB19147-2009 (III) denotes the standard of Automobile diesel fuel (III) set by China.b GB19147-2013 (IV) denotes the standard of Automobile diesel fuel (IV) set by China, and it is used for taking the place of Automobile diesel fuel (III).c The maximum of CFPP specified in GB19147-2009 (III) and GB19147-2013 (IV) for 0# and −10# automobile diesel fuels were 4 °C and −5 °C, respectively.
GB19147-2009 (III)^a	−	Max 4/−5	790–840	Trace	1.8–7.0	Min 55	Max 7	Max 2.5
GB19147-2013 (IV)^b	−	Max 4/−5	790–840	Trace	1.8–7.0	Min 55	Max 7	Max 2.5

---

Density, water content, FP, AV, υ, and OS significantly increased with increasing the amount of biodiesel in the blends. All these properties are presented in Table 4. The CV of 0# diesel was higher than of biodiesel. The CV of the biodiesel-diesel blends was changed with the change in biodiesel fraction. However, the CV was changed and ranged from 40 223 kJ kg⁻¹ to 42 750 kJ kg⁻¹.

3.3 Effect of CFIs on high-proportional biodiesel blends

3.3.1 Effect of CFIs on cold flow properties. While much research has been conducted on blends of B2–B20 to overcome problems related to low temperature operability, little research has been conducted on blends containing 50% (B50) biodiesel.³² Thus, EHMA, EAA, and IOMA were combined to further enhance the low temperature performance of biodiesel blends B50. Table 5 shows the effect of mixed CFIs on the flow properties of B50. Moreover, all the results were compared with the standards listed in Table 4.

Table 5 Effect of mixed CFIs on the cold flow properties of B50

Sample	Mixed CFIs (vol%)			PP/°C	CFPP/°C	Sample	Mixed CFIs (vol%)			PP/°C	CFPP/°C
	EHMA	EAA	IOMA				EHMA	EAA	IOMA
EHMA–EAA	2.5	2.5	—	−8	−5	EAA–IOMA	—	2.5	2.5	−8	−6
	5	2.5	—	−9	−5		—	5	2.5	−9	−8
	10	2.5	—	−9	−7		—	10	2.5	−10	−8
	2.5	5	—	−9	−6		—	2.5	5	−9	−6
	5	5	—	−8	−6		—	5	5	−9	−7
	10	5	—	−9	−7		—	10	5	−10	−7
	2.5	10	—	−9	−7		—	2.5	10	−11	−9
	5	10	—	−9	−6		—	5	10	−10	−8
	10	01	—	−10	−7		—	10	10	−10	−7
EHMA–IOMA	2.5	—	2.5	−8	−5	EHMA–EAA–IOMA	2.5	2.5	2.5	−9	−5
	5	—	2.5	−8	−6		10	5	2.5	−5	−5
	10	—	2.5	−9	−6		5	10	2.5	−10	−7
	2.5	—	5	−8	−6		5	2.5	5	−9	−6
	5	—	5	−8	−6		2.5	5	5	−9	−5
	10	—	5	−9	−7		10	10	5	−10	−7
	2.5	—	10	−9	−7		10	2.5	10	−9	−7
	5	—	10	−9	−7		5	5	10	−10	−7
	10	—	01	−10	−7		2.5	10	10	−12	−7

---

As it can be seen from Table 5, all the formulated blends had low CFPP values, which were within the limits of GB19147-2009 (III) and GB19147-2013 (IV) for −10# automobile diesel fuels (−5 °C). Mixed CFIs (EAA–IOMA) performed better improving on the PP and CFPP compared to that of EHMA–EAA and EHMA–IOMA. After adding 2.5 : 10 (EAA–IOMA), the CFPP reached the lowest value of −9 °C, which has a 9 °C and 11 °C depression with respect to B50 and B100, and PP were decreased by 7 °C and 12 °C with respect to B50 and B100, respectively. This can be attributed to the very low freezing point of EAA (−45 °C) and IOMA (−50 °C), and their synergistic effect, thereby retarding the aggregation of crystals and inhabiting the formation of larger crystals at a low temperature. Thus, the CFPP of B50 with mixed CFIs of EAA and IOMA is low.

In comparison, EHMA–EAA–IOMA has the biggest PP depression of 8 °C and 13 °C with respect to B50 and B100. However, the ternary mixed CFIs (EHMA–EAA–IOMA) did not always produce better effects with respect to the binary mixed CFIs in both values of PP and CFPP. The binary mixed CFIs have turned out to be the most effective CFIs in this study. One of the best PP and CFPP depression was observed at 2.5 : 10 (EAA–IOMA).

3.3.2 Effect of CFI on other properties. The addition of mixed CFI to biodiesel blends also had an impact on various fuel properties other than PP and CFPP. The effects of mixed CFIs (EAA–IOMA) on the density, water content, FP, AV, OS, CV of B50 are listed in Table 6.

Table 6 Effect of mixed CFIs (EAA–IOMA) on other properties of B50

Sample	Mixed CFIs (vol%)			Density (20 °C, g cm^−3)	Water content (mg kg^−1)	υ (40 °C, mm^2 s^−1)	FP (°C)	AV (mg KOH/ml)	OS (total insolubles, mg/100 ml)	CV (kJ kg^−1)
	EHMA	EAA	IOMA
EAA : IOMA	—	2.5	2.5	0.843	246.8	3.31	92	0.28	1.39	41 731
	—	5	2.5	0.854	244.2	3.14	89	0.27	1.41	41 013
	—	10	2.5	0.859	234.9	2.90	78	0.25	2.34	40 423
	—	2.5	5	0.855	246.1	3.17	87	0.26	1.52	41 545
	—	5	5	0.860	242.5	3.09	81	0.24	1.59	40 937
	—	10	5	0.864	235.1	2.83	76	0.22	2.21	41 121
	—	2.5	10	0.853	243.6	3.03	84	0.23	1.98	41 338
	—	5	10	0.861	241.0	2.94	81	0.20	1.79	41 033
	—	10	10	0.869	231.7	2.83	73	0.18	2.37	40 347

---

Both density and water content slightly changed after adding mixed CFIs in B50. The density of treated B50 changed in the 843–869 g cm⁻³ range whereas their water content increased within the limits of GB19147-2009 (III) and GB19147-2013 (IV).

The υ of B100 and B50 were up to 4.29 mm² s⁻¹ and 3.57 mm² s⁻¹, respectively. As show in Table 6, υ became significantly low after adding the mixed CFIs and the υ values still satisfied the standards.

The required FP in GB19147-2009 (III) and GB19147-2013 (IV) are ≥55 °C. Although the addition of mixed CFIs decreased the FP of the biodiesel blends, the FP of B50 with CFIs are still within the limits.

The total insolubles were used to present the OS of the diesel fuels. Adding mixed CFIs decreased the amount of total insoluble, thereby improving the OS of biodiesel blends. The maximum value specified in GB19147-2009 (III) and GB19147-2013 (IV) was 2.5 mg ml⁻¹ (Table 3). Thus, all those values that are listed in Table 6 are not beyond the limits.

The CV and AV were also marginally changed with the changes in the proportion and amount of mixed CFI within acceptable ranges.

4. Conclusions

This study explored the blending effect and impact of a mixed cold flow improving additive (EHMA, EAA and IOMA) on cold flow properties of biodiesel blends as well as the impact of mixed CFIs on other important fuel properties. On the basis of the obtained results, the following conclusion can be reached:

(1) Cold flow properties of waste cooking oil biodiesel can be enhanced by blending with petro-diesel. The PP and CFPP linearly reduced with increasing the concentration of petro-diesel in blends. The CFPP of blends were within the limits of GB19147-2009 (III) and GB19147-2013 (IV) for 0# diesel, but did not satisfy the specifications for −10# diesel fuels.

(2) EHMA, EAA and IOMA were mixed in various proportions. Binary mixed CFIs (EAA–IOMA) performed better improvement on cold flow properties than that of EHMA–EAA, EHMA–IOMA, and EHMA–EAA–IOMA. The best PP and CFPP depression was observed in 2.5 : 10 (EAA–IOMA). The CFPP and PP were decreased by 11 °C and 12 °C, respectively, with respect to the neat biodiesel.

(3) The addition of CFIs in B50 decreased the density, υ, AV, and FP, and the OS was increased. The formulated B50 satisfied the GB19147-2009 (III) and GB19147-2013 (IV) for −10# automobile diesel fuels.

Acknowledgements

This project was supported by the Shanghai Leading Academic Discipline Project (Project Number J51503), the National Natural Science Foundation of China (Project Number 20976105), the Shanghai Association for Science and Technology Achievements Transformation Alliance Program (Project Number LM201559), the Shanghai Municipal Education Commission boosting project (Project Number 15cxy39), the Science and Technology Commission of Shanghai Municipality Project (Project Number 14520503200), the Shanghai Municipal Education Commission (Plateau Discipline Construction Program), and the Shanghai Talent Development Funding (Project Number 201335).

References

1. N. N. A. N. Yusuf, S. K. Kamarudin and Z. Yaakub, Energy Convers. Manage., 2011, 52, 2741–2751.
2. D. S. Patle, S. Sharma, Z. Ahmad and G. P. Rangaiah, Energy Convers. Manage., 2014, 85, 361–372.
3. M. Kouzu and J.-s. Hidaka, Fuel, 2012, 93, 1–12.
4. A. Molaei Dehkordi and M. Ghasemi, Fuel Process. Technol., 2012, 97, 45–51.
5. F. Ma and M. A. Hanna, Bioresour. Technol., 1999, 70, 1–15.
6. I. M. Monirul, H. H. Masjuki, M. A. Kalam, N. W. M. Zulkifli, H. K. Rashedul, M. M. Rashed, H. K. Imdadul and M. H. Mosarof, RSC Adv., 2015, 5, 86631–86655.
7. H. Imahara, E. Minami and S. Saka, Fuel, 2006, 85, 1666–1670.
8. G. Knothe and R. O. Dunn, J. Am. Oil Chem. Soc., 2009, 86, 843–856.
9. G. Knothe, Energy Environ. Sci., 2009, 2, 759.
10. G. Knothe, Energy Fuels, 2008, 22, 1358–1364.
11. A. Perez, A. Casas, C. M. Fernandez, M. J. Ramos and L. Rodriguez, Bioresour. Technol., 2010, 101, 7375–7381.
12. M. E. G. Gómez, R. Howard-Hildige, J. J. Leahy and B. Rice, Fuel, 2002, 81, 33–39.
13. Y.-H. Chen, B.-Y. Huang, T.-H. Chiang and T.-C. Tang, Fuel, 2012, 94, 270–273.
14. M. Serrano, R. Oliveros, M. Sánchez, A. Moraschini, M. Martínez and J. Aracil, Energy, 2014, 65, 109–115.
15. P. Benjumea, J. Agudelo and A. Agudelo, Fuel, 2008, 87, 2069–2075.
16. P. Ma, Y. Xue, W. Zhao, G. Lan, Z. Hang, F. Liu and S. Han, RSC Adv., 2015, 5, 90144–90149.
17. P. C. Smith, Y. Ngothai, Q. Dzuy Nguyen and B. K. O'Neill, Renewable Energy, 2010, 35, 1145–1151.
18. P. V. Bhale, N. V. Deshpande and S. B. Thombre, Renewable Energy, 2009, 34, 794–800.
19. S. Sirisomboonchai, M. Abuduwayiti, G. Guan, C. Samart, S. Abliz, X. Hao, K. Kusakabe and A. Abudula, Energy Convers. Manage., 2015, 95, 242–247.
20. S. Kagawa, K. Takezono, S. Suh and Y. Kudoh, Energy Policy, 2013, 55, 362–368.
21. Y. Chen, B. Xiao, J. Chang, Y. Fu, P. Lv and X. Wang, Energy Convers. Manage., 2009, 50, 668–673.
22. L. Cao, J. Wang, K. Liu and S. Han, Appl. Energy, 2014, 114, 18–21.
23. R. Feng, J. Yao and L. Guan, Ultrasonics handbook, Publishing Company of Nanjing University, Nanjing, 1999.
24. J. Ji, J. Wang, Y. Li, Y. Yu and Z. Xu, Ultrasonics, 2006, 44(1), e411–414.
25. G. Knothe, J. Am. Oil Chem. Soc., 2006, 83, 823–833.
26. M. Tariq, S. Ali, F. Ahmad, M. Ahmad, M. Zafar, N. Khalid and M. A. Khan, Fuel Process. Technol., 2011, 92, 336–341.
27. S. Tianniam, T. Bamba and E. Fukusaki, J. Biosci. Bioeng., 2010, 109, 89–93.
28. J. Wang, W. Zhao, Y. Ai, H. Chen, L. Cao and S. Han, RSC Adv., 2015, 5, 45575–45581.
29. R. O. Dunn, Fuel Process. Technol., 2005, 86, 1071–1085.
30. E. Sendzikiene, V. Makareviciene and P. Janulis, Pol. J. Environ. Stud., 2005, 14, 335–339.
31. A. Talebian-Kiakalaieh, N. A. S. Amin, A. Zarei and I. Noshadi, Appl. Energy, 2013, 102, 283–292.
32. C.-W. Chiu, L. G. Schumacher and G. J. Suppes, Biomass Bioenergy, 2004, 27, 485–491.

---

Footnote

† These authors contributed equally to this work.

---