Bioresource Technology 97 (2006) 250256

Degradation of biodiesel under dierent storage conditions

D.Y.C. Leung *, B.C.P. Koo, Y. Guo
Department of Mechanical Engineering, University of Hong Kong, Pokfulam Road, Hong Kong, China Received 23 August 2004; received in revised form 3 February 2005; accepted 18 February 2005 Available online 7 April 2005

Abstract This paper aimed to investigate the biodiesel degradation characteristics under dierent storage conditions. The qualities of twelve biodiesel samples, which were divided into 3 groups and stored at dierent temperatures and environments, were monitored at regular interval over a period of 52 weeks. Experimental results demonstrated that the biodiesel under test degraded less than 10% within 52 weeks for those samples stored at 4 and 20 C while nearly 40% degradation was found for those samples stored at a higher temperature, i.e. 40 C. The results suggested that high temperature, together with air exposure, greatly increase the biodiesel degradation rate. The temperature or air exposure alone, however, had little eect on biodiesel degradation. Water content in biodiesel will enhance biodiesel degradation due to hydrolysis but its eect is much less than the above two factors. 2005 Elsevier Ltd. All rights reserved.
Keywords: Methyl ester; Purity; Acid value; Degradation rate; Biodiesel storage

1. Introduction Alternative fuels for diesel engines have attracted more and more attentions in the auto fuel market due to the depletion of fossil fuels in the world feedstock and the worsening of air pollution problems caused by the notorious emissions from motor vehicles. As early as 1900s, vegetable oils had already been used directly for operating diesel engines. However, carbon deposits and thickening of lubricating oil will occur in the engine when using vegetable oil as fuel mainly due to its high viscosities, low volatilities and polyunsaturated character, as well as its gum formation char-

Corresponding author. Tel.: +852 2859 7911; fax: +852 2858 5415. E-mail address: ycleung@hku.hk (D.Y.C. Leung).

acteristic due to oxidation and polymerization. This would damage the normal operation of an engine and reduce its life span even if they are blended with petroldiesel. Biodiesel, a vegetable oil derivative that has been developed for more than six decades, can substitute the direct use of the long chain vegetable oil. It became more widely used during the energy crises in 1970s. Three major processes: pyrolysis, micro-emulsication and transesterication, have been adopted to produce biodiesel for diesel engines. Among these three processes, transesterication (or alcoholysis) is the most commonly used process in producing biodiesel. This process transfers the ester of one alcohol chain to another by reuxing it with an excess of the second alcohol in the presence of a catalyst. In producing biodiesel, the vegetable oil or animal fat (triglycerides) reacts with alcohol (usually methanol or ethanol) to form esters (biodiesel) together with glycerol as a byproduct. Sodium hydroxide is usually used to catalyze the following main reaction:

0960-8524/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2005.02.006

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256

251

where R1, R2, R3 are saturated or unsaturated hydrocarbon chains and R 0 is an alkyl group. As shown above, biodiesel is a type of mono alkyl esters of long-chain fatty acids derived from renewable lipid feedstock, such as vegetable oils or animal fats. Previous studies demonstrated that biodiesel can drastically reduce emissions of total hydrocarbon, carbon monoxide, particulate matters, smoke density as well as producing low SOx, and aromatic compounds emissions since the feedstock of biodiesel is mainly vegetable oils which do not contain any sulphur or aromatic compounds (Knothe et al., 1997; Wang et al., 2000; Monyem and Van Gerpen, 2001; Al-Widyan et al., 2002). In addition, biodiesel has higher ash point, better biodegradability and lubricity, and lower toxicity than petroldiesel. Biodiesel in the market is mainly produced from agricultural products such as rapeseed, soy bean and palm oils. Life cycle analysis indicates that biodiesel is CO2 neutral as the biomass absorbs CO2 during their growth, and releases back to atmosphere after burning. Biodiesel can also be produced from used frying oil, grease trap waste oil, and waste tallow (Leung, 2001; Guo and Leung, 2003). By doing so, waste materials can be recycled and the cost of biodiesel can be reduced to a competitive level. Though it has so many advantages, some drawbacks are also found when using biodiesel in diesel engines, such as the slight reduction in power and torque, and the potential increase in NOx emission (Leung, 2003). Furthermore, biodiesel may dilute or contaminate the lubricating oil of engines. Even worse is that if the quality of biodiesel is not good and contains high degree of unsaturated fatty acids, it would tend to polymerize with the lubricating oil, forming sludge and increasing engine wear. Biodiesel also tends to crystallize at low temperatures. Under very cold weather, it may solidify and separate from diesel at fuel lines and lters if biodiesel is blended with diesel, causing problems in fuel pump. Biodiesel also has good bio-degradability, it is expected to degrade over 98% biologically within three weeks while diesel will only degrade 50% biologically within the same period (Williamson and Badr, 1998).

Some studies have been conducted focusing on how biodiesel stimulated the degradation of petroldiesel in aquatic environment or in sand column (Miller and Mudge, 1997; Zhang et al., 1998; Mudge and Pereira, 1999). However, there are very few studies concentrated on biodiesel degradation under dierent storage temperatures and storage environments such as in a sealed or ambient environment, and in an environment with or without the presence of water moisture. In-proper storage may degrade the quality of the biodiesel, thus aecting the engine performance and its life. International standards of biodiesel such as ASTM 6751 and DIN EN14214 specify the acid value and the purity of the biodiesel in order to ensure the quality of the biodiesel used. The purpose of this study is to investigate the biodiesel degradability characteristics under dierent storage situations. The study would provide insights on how biodiesel degrades with time under specic conditions. This information will be useful to both biodiesel producers and users for designing their biodiesel storage system and for maintaining the quality of biodiesel in the fuel or storage tank.

2. Materials used and testing methodology 2.1. Materials Neat edible rapeseed oil was used to produce the biodiesel for testing using methanol, sodium hydroxide and ethanoic acid. All chemicals used for the transesterication process were of analytical grade. Two types of test were conducted to determine the rate of degradation: the purity test and the acid value test. Methyl heptadecanoate was used as an internal standard for gas chromatographic (GC) analysis. For acid value test, standard sodium hydroxide solution (0.1 M) was used as titration solution while phenolphthalein was used as an indicator. Mixture of analytical grade diethyl ether and ethanol was used to act as a solvent to dissolve biodiesel titration in the acid value determination.

252

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256 Table 1 Fatty acid methyl ester prole of the sample at beginning of the experiment Fatty acid prole C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:0 C20:1 C22:0 C22:1 C24:0 Total Rapeseed methyl ester, concentration % by mass 0.1% 0.1 4.8 0.2 0.4 61.6 20.6 9.2 0.6 1.4 0.4 0.3 0.1 99.7

2.2. Analyses 2.2.1. Gas chromatographic analysis (purity test) Twelve dierent kinds of fatty acid methyl ester compositions were examined for the test samples as follows: C14:0, C16:0, C16:1, C18:0, C18:1, C18:2, C18:3, C20:0, C20:1, C22:0, C22:1 and C24:0. The composition of different fatty acid methyl esters in biodiesel was determined using a HP 6890 Series II Gas Chromatograph tted with a 3365/IIGC-ChemStation and a ame ionization detector. The capillary column was a HP-INNOWax (cross-linked PEG) column, which has a length of 30 m, a lm thickness of 0.5 lm and an ID of 0.32 mm. The split ratio was 45:1. The oven temperature was programmed to keep at 150 C for 1 min, then, heated at 15 C/min up to 225 C, followed by a rate of 5 C/ min to 260 C, where it was held for 7 min. Helium was used as the carrier gas with a ow rate of 2.1 ml/min and also as an auxiliary gas for the FID. One micro-litre of each diluted sample with analytical grade dichloromethane from BDH (England) was injected manually. The internal standard quantitative calculation was chosen with methyl heptadecanoate. Approximately 0.010.02 g of methyl heptadecanoate, which acted as internal standard, was put into a vial. Biodiesel sample (about ten times of the weight of the methyl heptadecanoate) was extracted from the testing asks and added to the same vial containing the standard. The mixture was diluted by dichloromethane to about 1.5 times the original volume and shaken vigorously to ensure uniformity. One micro-litre of the diluted sample was injected into the GC for determining the fatty acid methyl ester prole. As mentioned above, twelve dierent kinds of fatty acid methyl ester compositions were evaluated. Samples were taken at predetermined interval and each sample was analyzed by the GC for at least three times and the average reading was recorded. Table 1 shows the fatty acid methyl ester prole of the biodiesel determined by the GC before the start of the experiment that is used as a control sample. As can be observed, the biodiesel mainly consists of methyl 9Z-octadecenoate (methyl oleate) (i.e. C18:1), methyl 9Z, 12Z-octadecadienoate (methyl linoleate) (i.e. C18:2) and methyl 9Z, 12Z, 15Z-octadecatrienoate (methyl linolenate) (i.e. C18:3). According to Knothe et al. (1997), the feedstock with these compositions is Canola (rapeseed) oil. This vegetable oil consists of many unsaturated carbon molecules as indicated by its fatty acid prole. 2.2.2. Acid value analysis Acid value is a measure of the free fatty acid content of oil or fat, and is dened as the amount of potassium hydroxide (in mg) that is necessary to neutralize the free fatty acids contained in 1 g of oil sample. The acid value

analysis was based on titration of the sample. A solvent was required which was prepared by mixing diethyl ether with ethanol. Phenolphthalein indicator solution (1% phenolphthalein in 95% ethanol) was used as indicator for titration. The biodiesel sample was dissolved in the solvent and standard sodium hydroxide solution was used for the titration to measure the acid value of the sample using the following formula: Acid value Vol:NaOH ml Conc:NaOH M M:W:KOH Sample weight g For analyzing the acid values of the sample, a solvent made from a mixture of ethanol and diethyl ether at a ratio of 1:2 (V/V) was prepared. Fifty millilitres of the solvent was extracted and neutralized by 0.1 M standard NaOH solution to a pink colour that persisted for at least 30 s by 1% of phenolphthalein as the indicator. Approximately 5 g of biodiesel sample was weighed. The neutralized solvent, together with a few drops of the phenolphthalein indicator were added to the sample in an Erlenmeyer ask. The mixture was shaken until the solvent, sample and indicator were completely mixed. The mixture was titrated with 0.1 M standard NaOH solution and during which it was shaken vigorously until the appearance of the rst permanent pink colour of same intensity as that of the neutralized solvent before the addition. The pink colour should persist for at least 30 s indicating completion of the reaction and the volume of the standard NaOH solution added was recorded. Similar to the GC determination, each sample was analyzed for at least three times and the average value was taken. Although it is well known that ester measurement by GC analysis is more accurate than results given by acid value analysis for the determination of biodiesel degradation, the latter analysis was still carried out because

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256

253

of its comparatively simpler in determination as it involves titration only. Both equipment and reagents can be easily obtainable. Complicated calibration is not required and the process is not time consuming. The only drawback is its potential larger experimental error than the GC analysis, as there is diculty in precisely determining the end point based on colour change. In this study, both GC analysis and acid value analysis results correlate well as will be discussed in the next section. 2.3. Biodiesel sample preparation The catalyst for the transestericationsodium methoxide solution was prepared freshly by mixing a predetermined amount of methanol (22% by weight of oil) with NaOH (1.0% by weight of oil) in a container. The container is sealed to prevent NaOH from absorbing moisture and reacting with carbon dioxide in air, and to minimize loss of methanol due to evaporation. Biodiesel was produced with the normal transesterication reaction (Kusy, 1982), which was then cleaned thoroughly by washing with de-ionized water. The biodiesel produced has a purity of 99.7% which was used for all the subsequent tests. About 5 l of biodiesel was produced for the test, which was divided into 12 samples with 400 ml each. These 12 samples were further divided into three groups and stored in individual ask at dierent temperatures:

4 C (stored inside a refrigerator), 20 C (stored under room environment: temperature varied between 15 and 25 C during the whole experiment) and 40 C (stored in a hot-water bath). Each group consists of 4 samples: the rst sample was sealed by a glass stopper to avoid direct contact with air (type A sample); the second sample was opened and exposed to ambient air (type B sample); the third sample was added and well mixed with 5% water but the ask was sealed with a glass stopper (type C sample) and the last sample was added with 5% water and exposed to ambient air (type D sample). To facilitate comparison, the 12 samples were named as: A4, B4, C4, D4, A20, B20, C20, D20, A40, B40, C40 and D40. The numbers behind the alphabet, i.e. 4, 20 and 40, represent the temperature at which the samples were stored in a refrigerator, an air-conditioned environment and a water bath, respectively. All the samples were covered by aluminium foil paper so the biodiesel would not be decomposed by light.

3. Results and discussions 3.1. Eect of storage temperature Temperature is one of the important factors that affect the degradation of many substances like food and oil that contain organic components. It is expected that

(i)
100 90

(ii)
100 90

Purity (%)

Purity (%)
0 10 20 30 Week 40 50 60

80 70 60 50

80 70 60 50
0 10 20 30 40 50 60

Week

100 90

(iii)
100 90

(iv)

Purity (%)

80 70 60 50
0 10 20 30 40 50 60

Purity (%)

80 70 60 50

10

20

30
Week

40

50

60

Week

Fig. 1. Purities of biodiesel degradation at dierent temperatures (i) sealed, (ii) with air exposure, (iii) sealed with water presence, (iv) with air exposure and water presence. () 0 C; (j) 20 C; (m) 40 C.

254

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256

temperature will also aect the degradation rate of biodiesel. Fig. 1(i)(iv) show the degradation of biodiesel under dierent temperatures and storage conditions. The degradation is represented by a reduction in purity of the biodiesel as determined from the GC. As shown in the gures, there is an increase in biodiesel degradation with increasing storage temperature. Those samples stored at the lowest temperature, that is, 4 C, has the lowest degradation rate while those samples stored at the highest temperature, i.e. 40 C, has the highest degradation rate but the drop in purity is relatively small for those samples stored under sealed condition (57% degradation over 52 weeks) irrespective of whether water is presence or not (see Fig. 1(i) and (iii)). For those samples exposed to ambient air, the rate of degradation has no signicant dierence from those stored under sealed condition except those stored at 40 C where dramatic drop ($40% over 52 weeks) in purity is found (Fig. 1(ii) and (iv)). The results indicated that high temperature, together with air exposure, play a signicant role in biodiesel degradation, probably due to the higher oxidation rate of biodiesel at higher temperature. Fig. 2(i)(iv) indicates the acid value of the biodiesel under the four dierent storage conditions for the whole testing period. The results are similar to those of the purity test. Those samples stored at 4 C has the lowest acid value while those maintained at 40 C has the high-

est acid value, indicating a higher degradation rate. A bigger rise in acid value occurs after week 13 for those samples stored at 40 C. This implies that the quality of biodiesel would easily become worse if it is stored at high temperature, but storage under sealed environment can improve the situation. The result further indicates that when the biodiesel degrades, its acid value increases too. This is because the fatty acid methyl ester molecules are broken down during degradation and the fatty acid chains increase the acid value of the biodiesel. Thus, the more the degradation, the higher will be its acidity. 3.2. Eect of air exposure and water content in biodiesel Fig. 3(i)(iii) show the comparisons of biodiesel degradation under the four dierent storage conditions at dierent temperatures. When the samples were stored at 4 and 20 C (Fig. 3(i) and (ii)), both the degradation rate and the degradation level are closed to each other for the four storage conditions. Their purity levels decrease from 99.7% to between 92.5% and 94.5% (refer to Table 2) after 52 weeks of storage and their purities rank in descending order is A > C > B > D. Both the effects of ambient air exposure and water content on biodiesel degradation were not prominent under these two storage temperatures. There is, however, a prominent

(i)
1.0

0.8 0.6 0.4 0.2 0.0 0 10 20 30

Week

20 18 16 14 12 10 8 6 4 2 0 0 10

(ii)

Acid value (mg KOH/g)

40

50

60

Acid value (mg KOH/g)

20

30
Week

40

50

60

1.0

(iii)

Acid value (mg KOH/g)

0.8 0.6 0.4 0.2 0.0 0 10 20 30

Week

Acid value (mg KOH/g)

20 18 16 14 12 10 8 6 4 2 0 0 10

(iv)

40

50

60

20

30
Week

40

50

60

Fig. 2. Acid values of biodiesel degradation at dierent temperatures (i) sealed, (ii) with air exposure, (iii) sealed with water presence, (iv) with air exposure and water presence. () 0 C; (j) 20 C; (m) 40 C.

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256

255

(i)
100 90

80 70 60 50 0 10 20 30 40 50 60

Week
100

(ii)

eect of ambient air exposure and water moisture on biodiesel degradation for the samples stored at 40 C. The samples with ambient air exposure (B) and the samples with ambient air exposure and water presence (D) have the lowest purities (about 60% after 52 weeks) while the purities of samples A and C (stored under sealed environment) were about 93%. The rate of degradation increases sharply for samples B and D starting from week 13. The purity of A40 and C40 reduce at an average rate of 0.15% per week over the entire testing period while the purity of B40 and D40 reduce at a rate of about 0.5% per week for the rst 13 weeks and then at about 0.9% per week for the remaining 39 weeks.

Purity (%) Purity (%)

90 80
1.0

(i)

60 50 0 10 20 30 40 50 60

Acid value (mg KOH/g)

70

0.8 0.6 0.4 0.2 0.0 0 10 20 30 40 50 60

Week
100 90 (iii)

Purity (%)

80 70 60 50 0 10 20 30 40 50 60
1.0

Week
(ii)

Acid value (mg KOH/g)

0.8 0.6 0.4 0.2 0.0

Week
Fig. 3. Purities of biodiesel degradation at dierent storage temperatures (i) 4 C, (ii) 20 C, (iii) 40 C. () sealed; (j) with air exposure; (m) presence of water; () with air exposure and water presence.

10

20

30

40

50

60

Week
Table 2 Purities and acid values of twelve samples at week 0 and week 52 Sample Purity (%) 0.1% Week 0 A4 B4 C4 D4 A20 B20 C20 D20 A40 B40 C40 D40 99.7 Week 52 94.5 93.8 93.9 93.6 93.6 92.7 93.3 92.5 92.9 60.2 92.5 58.3 Acid value (mg KOH/g) 0.01 mg KOH/g Week 0 0.15 Week 52 0.22 0.22 0.22 0.20 0.28 0.28 0.27 0.28 0.75 17.69 0.93 19.30
20

(iii)

Acid value (mg KOH/g)

15 10 5 0

10

20

30

40

50

60

Week
Fig. 4. Acid values of biodiesel degradation at dierent storage temperature. (i) 4 C, (ii) 20 C, (iii) 40 C. () sealed; (j) with air exposure; (m) presence of water; () with air exposure and water presence.

256

D.Y.C. Leung et al. / Bioresource Technology 97 (2006) 250256

Clearly, air exposure and high temperature are responsible for the high biodiesel degradation rate obtained. On the other hand, the hydrolysis eect of water on biodiesel is much smaller than the eect of air exposure since the purity of sample D is only approximately 2% lower than that of sample B at the end of the test. It can be concluded that the eect of hydrolysis on biodiesel is small as compared with the air exposure eect on biodiesel. The results suggest that biodiesel may not degrade quickly under humid conditions but air exposure and high temperature are the main factors aecting the degradation rate. Fig. 4(i)(iii) compare the acid value of the samples stored under the four dierent conditions at constant temperatures of 4, 20 and 40 C respectively. As can be seen from these gures and also Table 2, the acid values of the samples increase from 0.15 to 0.22 mg KOH/g at 4 C and to 0.28 mg KOH/g at 20 C. Once again, the similar acid value result obtained at 4 and 20 C is consistent with the purity test results, indicating that these samples have similar degradation rate and purities. However, there is an order of magnitude increase in the acid values for those samples maintaining at 40 C. As shown in Table 2, at the end of the test (i.e. week 52), sample D40 has the highest acid value (19.3 mg KOH/g), followed by B40 (17.7 mg KOH/g). Obviously both B40 and D40 have much higher acid values than their counterparts A40 (0.75 mg KOH/g) and C40 (0.93 mg KOH/g), both of which were stored under a sealed environment. Again, this agrees with the purity test result shown in Fig. 3. From the above observations, the acid value has a direct correlation with the purity of the sample. The more the sample degrades, the higher the acid value will be. Therefore, the acid value test can provide alterative supporting evidence on biodiesel degradation. 4. Conclusions The characteristics of the degradation of biodiesel under dierent storage conditions are studied experimentally. Results indicated that the degradation characteristics of those biodiesel stored at temperature below 20 C were similar. The samples purities under these temperatures decreased from 99.7% to 92.5% after 52 weeks. When biodiesel was stored at 40 C, the purities decreased to about 60% after 52 weeks when it was exposed to ambient air irrespective of whether water is present in the samples or not and the degradation rate increased signicantly starting from week 13. For those samples keeping at 40 C under sealed condition, the purities were dropped to 92% after 52 weeks, which is similar to those stored at lower temperatures. The acid values of those samples stored at temperature below

20 C, increased moderately while the acid values of those samples maintained at 40 C surged to very high values, particularly for those samples stored with air exposure. In conclusion, it can be observed that temperature and air exposure are two important factors aecting the degradation of biodiesel. If biodiesel is stored in a high temperature environment and allowed to expose to ambient air, the degradation rate will be greatly increased. The temperature or air exposure alone, however, had little eect on biodiesel degradation. Water content in biodiesel will enhance biodiesel degradation due to hydrolysis but its eect is much less than other factors. Attention should therefore be paid to the above factors, particularly for long term storage of biodiesel.

Acknowledgement This study is funded by the William Mong Research Grant of the University of Hong Kong.

References
Al-Widyan, M., Tashtoush, G., Abu-Qudais, M., 2002. Utilization of ethyl ester of waste vegetable oils as fuel in diesel engines. Fuel Process. Technol. 76, 91103. Guo, Y., Leung, Y.C., 2003. Analysis on the biodiesel production using grease trap oils from restaurants. Macro Rev. 16, 421426. Knothe, G., Dunn, R.O., Bagby, M.O., 1997. Biodiesel: the use of vegetable oils and their derivatives as alternative diesel fuels. In: Saha, B.C., Woodward, J. (Eds.), Fuels and Chemicals from Biomass. American Chemical Society, Washington, DC, pp. 172 208. Kusy, P.F., 1982. Transesterication of vegetable oils for fuels. In: Proc. Int. Conf. on Plant and Vegetable Oils as Fuels, Fargo, ND. Leung, D.Y.C., 2001. Development of a clean biodiesel fuel in Hong Kong recycled oil. J. Water. Air Soil Pollut. 130, 277282. Leung, D.Y.C., 2003. Feasibility study of using biodiesel as motor fuel in Hong Kong. Final Report submitted to the Environment Protection Department of the HKSAR Government by the University of Hong Kong. 30pp. Miller, N.J., Mudge, S.M., 1997. The eect of biodiesel on the rate of removal and weathering characteristics of crude oil within articial sand columns. Spill Sci. Technol. Bull. 4, 1733. Monyem, A., Van Gerpen, J.H., 2001. The eect of biodiesel oxidation on engine performance and emissions. Biomass Bioenergy 20, 317 325. Mudge, S.M., Pereira, G., 1999. Stimulating the biodegradation of crude oil with biodiesel preliminary results. Spill Sci. Technol. Bull. 5, 353355. Wang, W.G., Lyons, D.W., Clark, N.N., Gautam, M., Norton, P.M., 2000. Emissions from nine heavy trucks fuelled by diesel and biodiesel blend without engine modication. Environ. Sci. Technol. 34, 933939. Williamson, A.M., Badr, O., 1998. Assessing the viability of using rape methyl ester (RME) as an alternative to mineral diesel fuel for powering road vehicles in the UK. Appl. Energy 59, 187214. Zhang, X., Peterson, C., Reece, D., Haws, R., Mller, G., 1998. Biodegradability of biodiesel in the aquatic environment. Trans. ASAE 41, 14231430.