MATERIALS AND METHODS
Shake flask system
A 2-litre Erlenmeyer flask (Fig. 1) contained 900 ml deionized distilled water (DIW), 100 ml of inoculum (acclimation medium), 1 ml of each stock solution (Table 1), and 10 mg/L carbon form the test compound. A reservoir holding 10 ml of barium hydroxide solution was suspended in the flask to trap the CO². After inoculation, the test flasks were sparged with CO² free air (by passage of compressed air through a series of three 2-litre bottles each containing 1000 ml of 5N Na(OH) to ensure that the trapped CO² came only from the microorganisms metabolizing the test substrate. The flasks were sealed and incubated with shaking in a dark room.

Figure 1. Shaker Flask System

Table 1. Medium employed for assay of CO² evolution.

Inoculum
1000 ml DIW dissolved in one gram of organic matter rich soil, 2 ml of activated (aerated) sewage mixed liquor, 50 ml of raw domestic sewage water, 25 mg/L of Ditco vitamin-free casamino acids, 25 mg/L of yeast extract, and 1 ml of each stock solution I, II, III (Table 1). For test system I and II (Table 2), test compounds were added at concentrations equivalent to 4, 8, and 8 mg carbon/L by calculation on days 0, 7, 11, respectively. At day 14, refilter the inoculum through glass wool. The inoculum is ready for use. For test system III, the acclimation period is 2 to 3 days to simulate a real oil spill situation.

Table 2. Test system design

Measurement of CO² evolution
The quantity of CO² evolved is measured by titration of the entire Ba(OH)₂ sample (10 ml of Ba(OH)₂ plus 10 ml of rinse water) with 0.1 N HCI to the phenolphthalein end point. After sampling, the reservoir was refilled with fresh Ba(OH)₂. All the samples were analyzed at least five times in a 28 day period to allow for a smooth biodegradation plot. Three millilitres of 20% H₂SO₄ were added on the day prior to terminating the test. The percent theoretical CO² evolved from the test compound was calculated at any sampling time from the formula:

% CO2 evolution = TF - CF / C x 100%

where,   (1)
TF = ml of 0.1 N HCI required to neutralize the Ba(OH)2 from 
the flask with the test substance;

CF = ml of 0.1 N HCI required to neutralize the Ba(OH)2 from 
the control flask;

C = Theoretical volume of the HCI required to neutralize the CO2 
converted from all the 10 mg carbon.

GC analysis
The GC method involved extraction of the samples with hexane and injection of a portion of the extract into a gas chromatograph. Quantitation was accomplished by using internal and external standards. In the extraction, the sample was first acidified to a pH of 2.0 or lower by adding 5 ml hydrochloric acid (1:1). One ml of internal standard was then added to the sample and shaken to mix well. Finally, 30 ml of hexane was added and the mixture was vigorously shaken for one minute. The layers are allowed to separate. The solvent layer was passed through a funnel containing sodium sulfate. The extract was transferred to a vial, sealed, and kept in a refrigerator (4°C) prior to the GC analysis.

 The parameters of the GC analysis for B100 were: methyl ester of heptadecanoic acid C17:0 (methyl 17:0) was the internal standard with the concentration at 0.003 g/ml and hexane was the solvent. The instrument used was an HP 5890 series II GC equipped with a flame ionization detector (FID), J&W DB-23 (30m x 0.25 mm I.D, 0.25 m film thickness) column. The temperatures were: injection port 250°C, detector 300°C, oven initial 215°C for 3 minutes, and final 230°C by a rate of 3 °C per minute. Five µl of extract were injected at a 50:1 split ratio.

 The parameters of the GC analysis of 2-D were: 2-fluorobiphenyl was the internal standard and methylene chloride was the extraction solvent. The instrument used was an HP series II 5890 gas chromatograph equipped with a capillary fused silica DB-5MS column (25 m by 0.21 mm by 0.33 µm; J&W Scientific, Folsom, Calif.) and MS interface. The interface temperature was set at 280°C. The MS detector was set initially at 250°C. A linear gradient of the oven temperature from 100°C to 320°C at 20°C per minute was used.

 The percentage of primary degradation was determined by subtracting the substrate concentration at sampling time from the initial substrate concentration, and the difference was divided by the initial concentration.

Test system design
A total of three test systems were conducted (Table 2). System I was to compare the biodegradability of different B100 fuels and diesel. System II was to test the biodegradability of mixtures of REE/diesel at different ratios (v/v). In system III varying concentrations of REE and diesel from low to high were tested to obtain the biodegradation rates. All results from the CO² evolution were compared to those from the GC analysis. Dextrose was the reference compound and the three flasks which contained Hg/Cl₂ were for inhibition control of microbial activity.

 Determination of biodegradation rates
Determination of biodegradation rates from CO² evolution (mineralization rates) employed the Warburg method used by Novak and Kraus (1973), but the method was modified. Varying substrate concentrations of REE were added into the test flasks and the rates of CO² evolution were measured. Initial slopes of all CO² evolution curves were taken to represent the rates of CO² evolution for the initial substrate concentration. These rates were multiplied by the initial concentrations of the test substrate to convert CO² evolution rates to mineralization rates, qms (mg L(-1) day(-1)). The rate for each concentration obtained was previously described and was plotted against the initial substrate concentration in each flask to determine the maximum mineralization rate q_Mmax

PART I
Biodegradability of REE, RME, SEE, SEE, NR, NS, 2-D

The composition of test substrates
The test substrates included: Rape Ethyl Ester (REE), Rape Methyl Ester (RME), Neat Rapeseed Oil (NR), Soyate Ethyl Ester (SEE), Soyate Methyl Ester (SME), Neat Soybean Oil (NS), and Phillips 2-D diesel. The composition and specific properties of B100 fuels are listed in Table 3 and of Phillips 2-D diesel in Table 4. REE and RME were made by the Department of Agricultural Engineering (Peterson et al., 1991) from neat Rapeseed oil and SEE and SME from neat soybean oil using a transesterification process. A more complete description of the process and a complete set of fuel characterization data is given in Peterson et al. (1994).

Table 3. Composition and main specific properties of B100.

Table 4. Composition and major physical properties of Phillips 2-D reference

RESULTS

Biodegradability of different B100 fuels (System I)

The percent accumulated CO² evolution of six B100 fuels REE, RME, SEE, SME, NR, and NS and 2-D in 28 days is summarized in Table 5 (all triplicates are averaged and an arithmetic mean, standard deviation, and RSD% is calculated). The maximum percent CO² evolution from the flasks of REE, RME, SEE, SME were between 85.54-88.49% in 28 days, the same as that of dextrose. It indicates there is no difference in their biodegradability. The maximum percent CO² evolution from neat rapeseed oil and neat soybean oil were 78.45 and 75.95% separately, which are slightly lower than their modified products. Yet, the CO² evolution for 2-D was only 26.24 % (average). The increase of the percent accumulated CO² evolution with time is graphically shown in Fig. 2. The data fits the exponential model well (Equation 2).

             y = a(1 - e-bx) for x  c         (2)

where y = the accumulated percent CO² evolved at day x; x = time in days; a = asymptote of CO² evolution curve (percent total CO²); b = rate constant of CO² evolution, day (-1); c = lag time before CO² evolution begins, days.

Table 5. CO² evolution from different substrates (conc. = 10 mg/L)

Figure 2. CO² evolution from seven B100 fuels and 2-D in 28 days

The substrate (REE and diesel) disappearance versus time and percent degradation by the GC analysis are given in Table 6 and plotted in Fig. 3. As they illustrate, within one day, 63.82% of fatty acids in REE disappeared while diesel was 26.99%. All the fatty acids in REE were not detectable after day 2. It is obvious that the primary biodegradation rates of REE and the diesel are much faster than the rates of mineralization (by CO² evolution method).
 

Table 6. Substance disappearance and percent degradation vs. time by GC analysis

Figure 3. The disappearance of REE and diesel in 2 days by GC analysis (conc. = 10 mg/L)

Biodegradation of B100/diesel mixture (System II)
The percent CO² evolution from REE/diesel mixtures increased with the increase of REE concentration in the mixture proportionally (Table 7 and Fig. 4). Higher REE concentrations in the mixture produced higher percentage of CO² evolution, and therefore, better biodegradability. This relation can be described by a linear equation Y = 0.596 X +0.207 with an R2 of 0.970, where X = the concentration of REE in a B100 mixture and Y = the maximum of CO² evolution after 28 days). The CO² evolution data fit the exponential model (Equation 2) very well by SAS non-linear regression (Fig. 4). The constants a and b for different mixtures are given in Table 7-8.

Table 7. CO² evolution from REE/diesel mixtures

Figure 4. CO² evolution from REE/diesel mixtures after non-linear regression when conc. = 10 mg/L

Table 8. The dynamics constants of CO² evolution for different REE/diesel mixtures.

Again, the GC analysis shows a much faster degradation rate in the mixture. At day 1, 63.62% of the REE/diesel mixture was degraded and about 96% disappeared after day 2. An interesting result is that the diesel in the mixture was degraded twice as fast as diesel alone, 62.8% versus 26.99% at day 1 (Table 9 and Fig. 5). It suggests that in the presence of REE, microorganisms use the fatty acids as an energy source to promote the degradation of diesel. A comparison of the GC identification in these two cases is shown in Fig. 6. From the GC peak identification of diesel (top part), there is no significant change in the peak counts between time 0 and day 1 when diesel alone was the sole carbon source. However, the diesel peaks in the REE/diesel mixture (bottom part of Fig. 6, the peaks for fatty acids of REE are at the right and so do not show) decreased significantly within one day (only the internal standard peak, retention time = 7.072 min., remains the same which indicates the same extraction efficiency).

 The biodegradation pattern of the mixture can be seen by comparing the percent degradation of the blend and the diesel in the blend (Table 9) at day 1, 63.62.81% versus 62.80%. It is clear that the microorganisms attacked the fatty acids in REE and alkane chains in the diesel at the same time and at almost same rates instead of favoring the fatty acids only.

Table 9. The disappearance of diesel in different media

Figure 5. Degradation of 2-D in the 50%REE/50%2-D and 100% 2-D in 2 days with GC analysis.

Figure 6. GC analysis of the disappearance of the diesel in two different medial at time O and day 1. The top part is the case of diesel alone as the sole carbon source and shows almost no difference in the area counts between time O and day 1. The bottom part demonstrates the significant change of the area counts of the REE/diesel blend (50/50) between time 0 and day 1. The larger peaks on the right are REE's. In both cases, the concentration of the diesel is 10 mg/L or 10 ppm.

Biodegradation rates of B1 (System III)
The mineralization rates (qM) for REE determined by the CO² evolution versus REE concentrations is shown in Fig. 7. When the REE concentration is below 180 mg L(-1), the rates increase with an increase in REE concentrations linearly (R2 = 0.95). However, when REE concentration is greater than 200 mg L(-1), the qMs stay almost constant. The maximum mineralization rate for REE was approximately 25 mg L(-1) day(-1). The pattern of the CO² evolution rates of the diesel changing with the concentration was almost the same as that of REE (Fig. 8), but the maximum mineralization rate for the diesel was about 12.5 mg L (-1) day(-1), half that of REE.

Figure 7. The dynamic response of mineralization rates of REE determined by CO² evolution to the increase of REE concentration (0-1000 mg/L).

Figure 8. The dynamic response of mineralization rates of 2-D diesel determined by CO² evolution to the increase of diesel concentration (0-500 mg/L).

REE disappearance with time and volumetric substrate removal rate, qv for each REE concentration are summarized in Table 10. The maximum degradation rate, q(vmax), for REE is about 160 mg L(-1) day(-1), which is 6.4 times higher than the maximum mineralization rate, q(Mmax) (= 25 mg L(-1) day(-1)), obtained by CO² evolution. The q(v)s plotted against REE concentrations is illustrated in Fig. 9.

Figure 9. The dynamic response of biodegradation rates of REE determined by GC analysis to the increase of REE concentration (0-1000 mg/L)

Table 10. REE disappearance with time by GC analysis

By comparing Fig. 6 and Fig. 9, one can see that both follow the same pattern.

Calibration
A comparison of percent REE degradation in 14 days by CO² evolution and in 5 days by the GC method (because both have different rates) is summarized in Table 11 and graphically shown in Figure 10. Apparently, they have the same trend. Their relation could be described by linear equation with a slope=1, %degradation = %CO² evolution + 11.0% (Fig 11). From a comparison calibration curve (Fig. 12), one can see that both maximum CO² evolution rate and substrate depletion rate happen within 5 days.

Figure 10. A comparison of percent degradation from CO² evolution in 14 days and GC analysis in 5 days when REE concentration = 100 mg/L(-1).

Table 11. A comparison of degradation

Figure 11. A comparison calibration (REE concentration = 100 mg/L(-1).

Figure 12. Calibration curve (REE concentration = 100 mg/L(-1)).

DISCUSSION
The experiment shows that B100 degrades much faster degraded than diesel. There are several reasons explaining this. First, the rate of a catalyzed reaction is regulated by the amount of catalyzing enzymes that are present in the cell. In other words, for a biochemical process to occur rapidly, appropriate enzymes must be available. B100 is a natural product consisting of pure fatty acids, the enzymes responsible for their breakdown also naturally exist. All the fatty acids are even hydrocarbon chains in ester form with two oxygen atoms attached which makes them very biologically active. The enzymes such as Acetyl-CoA dehydrogenase can recognize and attack them immediately. In the process of degradation, fatty acids are oxidized at the carbon (hence oxidation) and degraded to acetic acid and a fatty acid with two fewer carbons (Zubay, 1993). On the other hand, diesel consists of a large amount of alkane and alkene (hydrocarbon chains from C: 10-C:20) without oxygen attached, therefore, they are not biologically active. The natural enzymes may not recognize them. However, bacteria have very strong adaptability. They can either produce new enzymes or mutate the original genes of enzymes to adapt new products, especially during times of starvation. According to Pavel Pitter (Pitter and Chudoba, 1990), initial-oxidation of the terminal CH3 group to the carboxylic group is the main metabolic pathway during the biochemical oxidation of alkanes (monoterminal oxidation). This is followed by-oxidation of the aliphatic chain. Finally, the composition of diesel is much more chemically complicated then B100. Except for alkanes and alkenes, it also contains aliphatic cyclic hydrocarbons, polycyclic aromatic hydrocarbons (PAHs), and alkylbenzenes, as well as their derivatives such as toluene, xylenes, PCBs (phenyl and biphenyls), and so on. Alkylbenzenes are toxic to microorganisms. Benzene itself is very stable, and therefore, needs more energy for microorganisms to open the ring (Cole, 1993).

Another theoretical question to address is why mineralization rates determined by CO² evolution method is much slower and lower than the degradation determined by the GC analysis. There are at least three reasons that account for this. First of all, microbial breakdown of fatty acids to CO² and H₂O is a complex process and consists of a series of reactions. It begins in the cytosol and completes in the mitochondrion and involves totally different enzymes located in different places within the cell. At the first stage of metabolism, substrates are transported across the plasma membrane of a cell by specific enzymes. At the second stage, they are transported again into the mitochondrion where degradation of fatty acids occurs by oxidation with two carbon atoms less in each degradation cycle. Apparently, it takes some time for cells to digest 16-22 carbon chains. The GC method detects only the results of the first stage (substrate disappearance) while the CO² method examines the latter one (digestion). Secondly, a portion of the carbons from the substrates will be assembled into cell structure instead of being converted to CO² . Higher substrate concentrations, more significant cell growth, more carbon incorporated into cells, and therefore, less CO² detected. Thirdly, it is possible that some fatty acids are broken down to intermediates accumulated in the media so that there are no fatty acids detected by GC but degradation of these intermediates continues by the microorganisms. Hejalar and Chudoba (1986) found that organic substances excreted by activated sludge microorganisms into the cultivation medium can be both end products and intermediates. Whereas intermediates are readily re-assimilated during the growth of a mixed culture. Why are intermediates accumulated? According to Dawes (Hejzlar and Chudoba, 1986), some bacteria, e.g. Pseudomonas aeruginosa, can metabolize glucose in two ways. At low concentrations glucose is completely oxidized inside the cells whereas at high concentrations it blocks itself from this method and it is instead oxidized via the extracellular pathway (by periplasmatic enzymes) to gluconic acid which released from the cells.

It was noticed from the results of System I that neat rapeseed oil and soybean oil have slightly lower percent degradation. Their higher viscosity may limit their solubility, therefore, limit their biodegradability.
 

PART II
Biodegradability with HySEE
HySEE, one of the prospective B100 fuels, is made from waste hydrogenated soybean oil recovered from a potato processing plant. This report tested the biodegradability of HySEE and HySEE/diesel blends in the aquatic environment and compared them with that of REE and diesel by the CO² evolution method (EPA, 1982).

Test substrates
The test substrates included: HySEE, HySE /diesel blends (v/v) 80/20 and 20/80, Rapeseed Ethyl Ester (REE), 2-D, and dextrose as a reference. The composition and specific properties of HySEE, REE, and 2-D are listed in Table 12.

Table 12. Fuel Properties and fatty acid composition.

Quality control
Duplicate experiments were conducted and in each experiment triplicate flasks were set up for each substrate.

RESULTS AND CONCLUSION

The percent accumulated CO² evolution from all the substrates in 28 days is summarized in Table 13 and graphically shown in Figure 13 (the six samples for each substrate in two experiments were averaged and an arithmetic mean, standard deviation, and RSD% was calculated). The maximum percentage of CO² evolution from HySEE was 89.5%, almost the same as REE and dextrose, 91.9% and 89.81%, respectively. Whereas the CO² evolution from diesel was only 28.7%. Therefore, by EPA definition, HySEE is a "readily biodegradable" compound in the aquatic environment.

 The percent CO² evolution from HySFF/diesel mixtures increased with the increase of HySEE concentration in the blend, 77.62% vs. 42.5% for HySEE80/diesel20 and HySEE20/diesel80, respectively.

Table 13. Percentage of CO. evolution from the substrates in 28 days.

Fig. 13. 28 day accumulated carbon dioxide evolution.

CONCLUSIONS
1. All the B100 fuels are readily biodegradable compounds according to EPA standard (EPA, 1982) and have a relatively high biodegradation rate in the aquatic environment. The maximum mineralization rate for REE determined by CO² evolution is 25 mg L(-1) day(-1), while the diesel is 12.5 mg L(-1) day(-1), only half of REE s. The maximum volumetric substrate removal rate for REE obtained from GC analysis is about 160 mg L(-1) day(-1).

2. B100 can promote and speed up the biodegradation of diesel. The more B100 present in a B100/diesel mixture, the faster the degradation rate. The biodegradation pattern in a B100/diesel mixture is that microorganism metabolize both B100 and diesel at the same time and at almost the same rates.

3. The mineralization rate for a compound determined from CO² evolution is much slower than its primary degradation rate from GC. However, the results from CO² evolution method have good correlation with that from GC analysis. It is therefore possible to use the CO² evolution method to estimate the biodegradation rate for an organic compound and it is economical and environmentally safe.

REFERENCES
 1. Cole, G., Mattney. 1993. Assessment and remediation of petroleum contaminated sites. Lewis Publishers, pp38-57
 2. EPA 560/6-82-003, PB82-233008. 1982. Test guidelines: chemical fate - aerobic aquatic biodegradation.
 3. Heizlar, J. And Chudoba, J. 1986. Microbial polymers in the aquatic environment production by activated sludge microorganisms under different conditions. Water Resources. Vol. 20, No. 10, pp1209-12 6.
 4. Lyman, W.J., W.F. Reehl, and D.H. Rosenblatt. 1990. Handbook of chemical property estimation methods -- Environmental behavior of organic compounds. American Chemical Society, Washington, DC.
 5. Novak, J.T. and D.L. Kraus. 1973. Degradation of long chain fatty acids by activated sludge. Water Research Pergamon Press, 7:843-851.
 6. Reece, D.L. 1995. On-road testing of rapeseed B100. M.S. Thesis. Agricultural Engineering Department, University of Idaho, Moscow, ID 83844-2060.
 7. Peterson, C.L., M. Feldman, R. Korus, and D.L. Auld. 1991. Batchtype transesterification process for winter rape oil. ASAE paper, St. Joseph, MI 49085-9659.
 8 Peterson, C.L., D.L. Reece, and S.M. Beck. 1994. Processing, characterization and performance of eight lipids. ASAE Paper No. 946531. ASAE, St. Joseph, MI 49085-9659.
 9. Peterson C.L., D.L. Reece, B.L. Hammond, J.C. Thompson, and S.M. Beck. 1995. Commercialization of Idaho B100 (HySEE) form ethanol and waste vegetable oil. ASAE Paper No. 956738. ASAE, St. Joseph, MI 49085-9659.
 10. Pitter, Pavel and Jan Chudoba. 1990. Biodegradability of organic substances in the aquatic environment. CRC Press pp167.
 11. Zubay, Geoffrey. 1993. Biochemistry (third edition). Columbia University, Wm. C. Brown Publishers, pp445.