Paper Menu >>
Journal Menu >>
Advances in Chemical Engi neering and Science , 20 1 1, 1, 65-71
doi:10.4236/aces.2011.12011 Published Online April 2011 (http://www.scirp.org/journal/aces)
Copyright © 2011 SciRes. ACES
Case Study of Biodiesel-Diesel Blends as a Fuel
in Marine Environment
Tianxi Zhang1, Yapeng Chao1, Nan Liu1, Joe Thompson2, Manuel Garcia1,
Brian B. He2, Jon Van Gerpen2, Shulin Chen1*
1Department of Biol o gic al Sys t ems E ngineering, Washi n gton State Universit y, Pull m an , US A
2Department of Biolo gi cal a n d Agri cul t u ral En gi neeri n g, University of Idaho, Moscow, USA
Received February 16, 2011; revised March 24, 2011; accepted March 29, 2011
Formation of excessive sludge and fuel filter clogging were experienced in using biodiesel blends under ma-
rine environment. In this study, a field test was conducted in a marine ferry boat fueled by canola-based bio-
diesel blends. The sludge materials collected in the fuel purifier were characterized using inductively coupled
plasma (ICP), pyrolysis-GC/MS (Py-GC/MS), thermogravimetric analysis (TGA), and Karl-Fischer titration.
It was found that the sludge materials consisted of four distinctive fractions: organic materials derived from
diesel and canola biodiesel (major fraction), ash (11% - 14% w/w), water (~17% w/w), and bacteria. The
active bacteria were present in the sludge samples. It was suggested that bacterial contamination was one of
the major factors in contribution to the sludge formation.
Keywords: Biodiesel, Sludge, Filter Clogging, Microorganisms, Marine Environment
Formation of insoluble material in biodiesel blends has
drawn great attention as it causes clogging of filters .
One reason that blends-containing biodiesel is prone to
formation of insoluble material is the relatively high
temperature of crystallization (or solidification) com-
pared to petro-diesel. When biodiesel is used at low
temperatures, crystals tend to form and clog fuel filters.
When temperatures are even lower, biodiesel exhibits
jelly-like behavior and cannot be pumped. According to
ASTM standard D6751, three parameters are used to
evaluate biodiesel’s cold flow properties: cloud point
(CP), pour point (PP), and cold filter plugging point
(CFPP). Biodiesel has relatively high values for CP and
PP that limit its application as neat fuel (B100) under
low temperature conditions . Extensive research has
been conducted on cold flow properties of biodiesel from
various feedstocks. Dunn et al. summarized the cold
flow properties of methyl and ethyl esters of biodiesel
derived from several feedstocks and found that CP, PP,
and CFPP of these biodiesels varied quite widely .
Fuel additives are typically used to improve the cold
flow properties of petro-diesel under extremely low
temperature conditions . These additives usually con-
tain copolymers of ethylene and vinyl acetate or other
olefin-ester copolymers. In general, they act by distorting
the wax crystal shape and size to inhibit crystal growth,
thereby reducing PP temperatures. These additives seem
to work similarly in biodiesel blends: one study demon-
strated significant reduction of PP and CP for soy bio-
diesel and its blends with #2 diesel, with ethyl ester addi-
tives proving more effective than methyl ester .
Another reason for development of insoluble material
in biodiesel is that minor fuel components, such as sterol
glucosides and monoglycerides, can form precipitates
under certain conditions. These precipitates agglomerate
over time into flocs and sediments. The sterol glucoside
content in biodiesel may vary from supplier to supplier
based on both the feedstock origins and the processing
technologies used. The insoluble materials formed at low
temperature have recently been characterized from sev-
eral oil feedstocks, such as soybean, palm, yellow grease,
and cottonseed [4,5]. The authors reported isolation of
insoluble steryl glucosides from soybean-based oil, mo-
noglycerides from palm oil, and steryl glucosides and
monoglycerides from cottonseed oil.
Application of biodiesel blends in marine environment
is prone to even greater sludge formation than standard
land applications . It has been generally believed that
formation of insoluble material in marine conditions may
occur by a new mechanism, outside of low temperature
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
crystallization and precipitation of minor components.
This speculation was supported by the results from a
pilot test conducted by the Washington State Ferries
(WSF) in the Port of Seattle during a four month period
in 2004. B20 (20 percent biodiesel) fuel blend was used
in the test that resulted in the formation of excessive
sludge material, which led to the clogging of filters .
Additional sludge materials were also found in the cen-
trifugal fuel purifiers. These purifiers function to sepa-
rate water and solid contaminants from fuel by differ-
ences in densities and centrifugal forces. The formation
of excessive sludge material was not clear and sludge
properties were not characterized in that pilot test.
Identifying the cause of fuel clogging in such a marine
environment is not a trivial task, because biodiesel-diesel
blends are not simply stored in a static and isolated en-
vironment on marine vessels. Biodiesel blends are trans-
ferred into one of the storage tanks in the vessel from
tanker trucks, and the fuel is therefore subject to mixing
in the storage tank. The fuel is then transferred from the
storage tank into day tanks through centrifugal fuel pu-
rifiers, pumps, and filters. Finally, the fuel is pumped
under pressure in service piping through filters, hoses,
and engine fuel injection equipment, into the engine,
with some fuel returning to the fuel oil day tanks. The
environmental conditions under which biodiesel is used
are therefore dependent not only on weather, but also on
the processes used on each particular marine vessel.
In order to understand the causes of filter clogging and
sludge properties, another pilot test was conducted dur-
ing 2008-2009 as a part of study. Sludge formation was
experienced again in this test. This study reports findings
from examination and characterization of the sludge ma-
terial obtained from that pilot test, including results ob-
tained using pyrolysis-GC/MS (Py-GC/MS), thermogra-
vimetric analysis (TGA), inductively coupled plasma
(ICP), and Karl Fisher (K-F) titration, as well as presence
of microbes and probable polysaccharides in the sludge
2. Materials and Methods
2.1. Pilot Test in the WSF System
The Tillikum, a Washington State Ferry vessel, was fu-
eled with canola-based biodiesel/diesel blends B5, B10,
and B20 and operated for a period of one year (April
2008-April 2009). This was the same vessel that expe-
rienced excessive filter clogging problems during the
pilot tests conducted in 2004 with soybean-based B20
biodiesel-diesel blends. The fuel tanks of this vessel were
carefully cleaned before the first canola B5 blend was
loaded. The objective of cleaning is to remove any tank
sediment coating the tank walls as the tanks were origi-
nal steel in over 50 years old. High pressure hot-water
was used to get tanks cleaned. Initially, the fuel system
appeared to be running well without excessive sludge
buildup in the purifier. However, sludge accumulation
within the fuel purifier was found in May 2008, about six
weeks after starting the tests. Excessive sludge formation
resulted in filter clogging. This sludge appeared to be
very similar to the samples collected in the 2004 tests. In
spite of twice-weekly cleaning to remove the sludge,
sludge buildup in the purifier continued to be recurred.
In a control test, another vessel run regular 100% Ul-
tra-Low Sulfur Diesel (ULSD). About 15,700 gallons of
ULSD diesel were burned. No excessive sludge, however,
was found during the test period.
Neat canola-based biodiesel (B100) which met current
standard specification (ASTM D 6751-07) was obtained
from Imperium Renewable Inc. (Seattle, Washington,
USA), and ultra low sulfur diesel (ULSD) was purchased
from the local market. Biodiesel blend fuels were pre-
pared by blending the B100 and ULSD at certain volu-
metric ratios. Reported here are the properties of two
sludge samples collected in the purifier which is between
storage tank and day tank on May 27th, 2008 (Sample #1)
and May 31st, 2008 (Sample #2).
2.3. Microbial Cultivation
In order to identify microbial types in the sludge, micro-
bial culture experiments were performed. Four types of
solid media were used for the microbial growth. Plate
count agar (PCA, Type 1) was designed for detecting
bacteria, potato dextran agar (PDA, Type 2) was used for
cultivating fungi that might be present in the sludge, malt
extract agar (MEA, Type 3) was used mainly for culti-
vating potential yeasts in the sludge, and anaerobic agar
(AA, Type 4) was designed for observing microbes that
could grow under anaerobic conditions. Components of
the media were summarized in Table 1 as below.
In addition, a small amount of the sludge was inocu-
lated and incubated. The procedure of microbial culture
Table 1. Components of culture media.
Type 1PCA with pancreatic digest of casein, yeast extract,
dextrose, and agar
Type 2PDA with potato starch, dextrose, and agar
Type 3MEA with maltose, dextrose, glycerol, peptone, and agar
AA with agar with casein peptone, sodium chloride,
dextrose, sodium thioglycollate, soy peptone, L-cystine,
agar, sodium sulfoxyl formaldehyde, and methylene blue
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
was described elsewhere . To be briefly, 0.10 g of wet
sludge was weighed under sterile conditions, suspended
in 1.0 ml of sterile deionized water, and vortexed for ten
minutes. The suspended samples were diluted by a factor
of 5 000 and then the resultant samples were spread on
the culture plates of each type of media as described
above, respectively. The culture plates were incubated at
30℃ for 72 hours. Triplicate tests for the microbial cul-
ture were conducted using the sludge.
2.4. Analytical Methods
2.4.1. Trace Elemental Contents
The sludge samples were analyzed for trace elements by
inductively coupled plasma (ICP). The instrument used
was a Perkin Elmer Optima 3200 RL. The samples were
digested in nitric acid at approximately 120˚C in digested
tubes and then dried before the ICP analysis. Sample size
was 0.25 g after drying and grinding.
2.4.2. Char act eri zation of Organic Materi al s by
Py-GC/MS and TGA
Organic matter in the sludge was characterized using
pyrolysis-GC/MS (Py-GC/MS) and thermogravimetric
Py-GC/MS was carried out using a CDS pyroprobe
5000 series with an Aglient GC-MS (5975B inert XL
MSD). Samples were loaded into a quartz tube and kept
the oven (210˚C) to ensure adequate removal of oxygen
prior to pyrolysis. Samples were pyrolyzed by being
heated to 500˚C, and the resulting pyrolysis vapors were
separated by a (5% phenyl)-methylpolysiloxane non-
polar column. The gas flow rate was 1 ml/min and he-
lium was the carrier gas. The gas was then sent into a
mass spectrometer (Aglient Technologies 5975B Inert
XL MSD). The mass spectrometer conditions were transfer
line 150˚C, ion source 230˚C, and electron energy 70 eV.
The mass spectra of predominant peaks were then com-
pared to a mass spectra library to determine the com-
pound in a given peak.
TGA analysis was conducted using a Mettler-Toledo
TGA/SDTA851. Approximately 5 - 10 mg of samples
were loaded into an alumina pan, and vaporized in the
temperature range of 25˚C - 600˚C at a rate of 10˚C /min.
The samples were run under nitrogen atmosphere at a
flow rate of 20 ml/min.
2.4.3. Determination of Water Content by
Karl-Fisher (K- F) Titration
The sludge was a heterogeneous system; including or-
ganic matter and inorganic material (see “Results and
Discussion” Section). Water content in the wet sludge
was determined using the K-F titration method with a
Titroline KF Titrator from Schott Instruments GmbH.
Before the titration, the wet sludge samples were dis-
persed well in organic solvents in order to determine
water content in solvents. Several solvents, such as me-
thanol, ethanol, hexane, toluene, and pyridine, were
tested for the dispersion; only pyridine dispersed the
sludge well in the solvents tested. Wet sludge was dis-
persed in pyridine at final concentration of 2.7% (w/v).
The water content in the wet sludge was calculated by
the content of sludge in solvents and K-F titration.
2.4.4. Ion Chromatography (IC) Analysis
An aqueous sample collected in the purifier of the Tilli-
kum appeared to be highly viscous. It was hypothesized
that bacteria in the sludge produced viscous materials,
such as polysaccharides, that were dissolved into water.
In order to test this hypothesis, IC analysis was used to
identify monosaccharides in the digest of the aqueous
Sample preparation for IC analysis has been previous-
ly described . Samples were hydrolyzed to obtain
monosaccharides using acid (1.0 mol/L of H2SO4) at 100℃
for 2 hours, and then diluted to a certain concentration.
Five sugars, arabinose, galatose, glucose, xylose, and
fructose, were used as standard samples. Prior to analysis,
all samples including standard sugar samples were fil-
tered through 0.25-µm-pore-size polycarbonate mem-
branes (Nuclepore Corporation, Pleasanton, California).
All sugar analyses were carried out using a Dionex
ICS-3000 reagent free dual ion chromatography (IC)
system (Sunnyvale, California), which comprised a DP
dual gradient pump module, an EG dual eluent generator
(with one KOH reservoir cartridge in use for this work)
and a DC detector/chromatography module with three
programmable high-pressure six-port injector valves.
Briefly, the mobile phase, at a flow rate of 1.0 ml/min,
consisted of ultrapure water (0.015 μS/cm; eluent A) and
250 mM NaOH (eluent B), with the following gradient:
0.0 min: 87% A, 13% B; 20.0 min: 87% A, 13% B; 40.0
min: 15% A, 85% B; 41.0 min: 100% B; 49.0 min: 100%
B; 50.0 min: 87% A, 13% B; and 65.0 min: 87% A, 13%
B. Due to the matrix interference, quantification was
carried out with standard addition.
3. Results and Discussion
3.1. Sludge Formation in the Purifier of the
Excessive sludge formation occurred on the purifier
walls when running canola-based B5. The sludge materi-
al was smooth, greasy and slightly grainy. Figure 1(a)
shows the buildup of sludge on the purifier inner walls,
and Figure 1(b) a typical microscopic image of a sludge
sample. Microdomains, ranging from 30 to 150 μm, were
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
Figure 1. A representative sludge sample from the vessel
using B5. (a) Sludge buildup on the purifier walls; (b) Mi-
croscopic image of the sludge (scale bar of 50 µm).
visible in this sample. Active microbes were found in the
microdomains (see “Bacterial Role and Sludge Forma-
tion” Section below).
3.2. Sludge Characterization
3.2.1. Content s of Ash an d Met al s
The sludge samples from the vessel Tillikum contained
11% - 14% (w/w) ash by drying method, as shown in
Table 2. The low ash content (< 15% w/w) suggests that
the major component did not come from insoluble inor-
ganic material. But the remaining 86% - 89% of sludge
components consisted of water and organic matter. Some
of this organic matter likely originated from the diesel
Table 3 shows the trace elements in two sludge sam-
ples collected on different days. Unexpectedly high lev-
Table 2. Ash analysis of sludge samples.
Sample Vessel Location Ash (% w/w)
Sample #1 Tillikum Purifier 11
Sample #2 Tillikum Purifier 14
Table 3. Trace element contents in the sludge samples.
Elements (mg/liter) Sample #1 Sample #2
Arsenic < 38 < 38
Barium 26 24
Beryllium < 0.38 < 0.38
Calcium 30 000 21 000
Cadmium < 1.5 < 1.5
Cobalt < 1.5 < 1.5
Chromium 28 37
Copper 92 50
Iron 1 600 2 400
Potassium 89 70
Magnesium 1 500 1 600
Manganese 41 33
Molybdenum 160 80
Sodium 4 400 6 000
Nickel 7 6.9
Phosphorus 4 400 4 100
Lead 34 18
Sulfur 5 000 4 500
Vanadium 4.1 5.8
Zinc 230 140
els of calcium, iron, sodium, and sulfur were found.
Since the fuel was stored in steel tanks, the source of the
iron might be rusty tank walls, but the sources of the
other metals were unknown. The sulfur content of the
sludge was high for both samples, even though the vessel
was using ULSD, which has sulfur content below 15 mg/L.
It should be noted that analysis was performed on sludge
that built up in the purifier, but not on the fuel itself.
Sulfur from the fuel may have accumulated in the sludge,
or sulfur-containing materials may have been picked up
from the hull of the vessel, which may contain dec-
3.2.2. Organic Matter in the Sludge
Figure 2(a) shows the Py-GC/MS chromatograph of the
sludge sample. The highest peak was at 31.5 min (Figure
2(a)). Figure 2(b) suggests that the MS pattern of this
peak is consistent with 8-octadecenoic acid methyl ester
(ODAME, C19H36O2), based on the library of standard
chemicals. Thus, ODAME is one of the sludge compo-
nents which could come from biodiesel based on its
Thermogravimetric analysis (TGA) was used to study
the thermal decomposition and thermal properties of or-
ganic matter from sludge . Dantas et al. investigated
the thermal behavior of corn oil-based biodiesel by
reacting with methanol and ethanol . They reported
that corn oil was thermally stable up to 336˚C, methyl
biodiesel up to 145˚C, and ethyl biodiesel up to 169˚C in
TGA was also used to characterize the organic materi-
al in sludge in this study. Figure 3 shows a derivative
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
Figure 2. (a) Py-GC/MS chromatograph of the sludge sam-
ple; (b) MS pattern of 8-octadecenoic acid methyl ester
thermogravimetric (DTG) curve converted from the
TGA. DTG presents the rate of weight change of the
samples versus the temperature change. There are two
Figure 3. Derivative thermogravimetric (DTG) curve of the
distinctive peaks apparent in Figure 3, suggesting that
the sludge consists of two major fractions with different
properties. The fraction within the temperature range of
430 - 490˚C could be heavy components that might not
come from diesel and biodiesel that could not get such
high temperature . However, the quantity of this frac-
tion was small, about 6% of the sludge. A large fraction
of the sludge evaporated at temperatures below 250˚C,
suggesting light components or compounds with low
molecular weight, including water. It is interesting that
three sub-peaks at 125, 130, and 136˚C were present. It is
not clear what specific compounds account for these
3.2.3. Water Content
Unlike the B5 fuel blend, which has low water content
(< 300 mg/L) , B5 sludge contains high levels of
water, which may facilitate microbial growth (see” Bac-
terial Role in Sludge Formation” Section). Table 4
presents the water contents of wet sludge samples using
pyridine as the K-F solvent. The solvent pyridine is
miscible with water; leading to that water in the sludge
could dissolve completely in the pyridine. The water
content of 17.2% w/w) was found, suggesting that water
is one of fractions in the sludge sample.
3.3. Bacterial Role in Sludge Formation
3.3.1. Observation of Bacteria Presence in the Sludge
Active microbes in the sludge samples were first ob-
served under a microscope. A great number of active
bacteria, both round and rod-shaped, were found in the
microdomains (Figure 4). The microbes tend to live in
the microdomains, which could be water-rich. It ap-
peared that there were several bacterial species present in
the samples, but no yeasts or other fungi were observed.
Table 4. Water content in the wet sludge.
Dispersed Solvent H2O in wet sludge % (w/w)
Pyridine 17.2 ± 2.2
Figure 4. Image of microbes in the sludge sample (scale bar
of 10 µm).
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
It is not uncommon for microorganisms to grow in
hydrocarbon fuels. For example, Lutz et al. (2006) re-
ported that bacteria commonly present in natural envi-
ronments could aerobically biodegrade palm methyl or
ethyl biodiesel . Biodegradation of rapeseed oil me-
thyl ester (RME) by microorganisms has also been ob-
served . Biodiesel could be more favored carbons
source than diesel fuel to some microbes . These
reports further confirm our findings that active microbes
are present in the biodiesel fuel.
Culture of sludge microbes yielded large quantities of
bacteria on each of the four types of media tested. How-
ever, either fungal or yeast colonies were not found, even
on the appropriate media (PDA for fungal and MEA for
yeast). Therefore, it was suggested that bacteria were the
dominant microbes in the sludge samples studied. The
bacteria counts in the sludge from three types of culture,
including aerobic and/or anaerobic conditions, were
about 107 to 108 CFU/mg (colony-forming unit) sludge
by plate counting. Therefore, the bacteria grew well in
both anaerobic and aerobic conditions, without a signifi-
It should be noted that multiple strains of bacteria
were present in the sludge. In our previous study, isola-
tion of bacteria was conducted from the sludge samples.
Three bacterial strains as predominant growing bacteria
were isolated, and further identified as Klebsiella oxyto-
ca, Klebsiella nov. sp., and Staphylococcus epidermidis
using molecular biological method as well morphologi-
cal, biochemical, physiological properties .
Microbe origin is an interesting question. The fuel
samples of B100 collected were analyzed using ASTM
methods. They all met the ASTM standard specification,
indicating good quality of fuel used. The biodiesel fuel
blends and the sludge samples were tested using micro-
bial kits. The fuel blends were shown to be microbial
negative results while the sludge samples were positive,
which further confirmed active microbes present in the
sludge. Thus the results suggested that the biodiesel and
fuel blends were not sources of microbial contamination.
Marine environments (such as water, air) could be re-
lated with the contamination. For example, the high hu-
midity appears to promote microbial growth. Further
investigation is required to clarify this question about the
3.3.2. IC An a lysis of Visc o us Material in t he A queous
High viscosity in the aqueous solution was observed in
the sludge samples from the purifier. It is speculated that
microbes produced polysaccharides, which contributed
to the high viscosity in the solution. Significant amounts
of exopolysaccharides were produced by the isolated
Klebsiella oxytoca using the PCA medium (Type 1 in
Table 1) . In order to confirm this speculation in the
sludge, the IC measurement was also used in the sludge
samples to determine monosaccharide presence. Table 5
lists the monosaccharide presence of the hydrolyzed
aqueous solution from the sludge sample. Galactose and
glucose were found in this sample, while significant
amount of xylose and fructose were not detected. Mono-
saccharides are probably breakdown products of poly-
saccharides typically produced by the isolated bacteria.
So the monosaccharide composition is associated to the
polysaccharides which are unclear. Relationship between
bacteria and polysaccharides needs further investigation.
Extracellular polymeric substances (EPSs), such as
polysaccharides produced by bacteria, represent a class
of polymeric materials with a wide variety of potentially
useful applications [15,16]. Polysaccharides of high mo-
lecular weight, such as xanthan gum, a widely used food
additive, make aqueous solutions viscous . It is
therefore likely that polysaccharides produced by bacte-
ria resulted in the high viscosity of the aqueous solution.
3.3.3. Corrective Action in the Pilot Test of the
As discussed above, active bacteria were present in the
sludge samples from the purifier, and microbial conta-
mination occurred in the pilot tests of the Tillikum. To
prevent bacterial growth, a commercial biocide (Biobor
JF) was added to the fuel at dose of one gallon of the
biocide per 10 000 gallons of fuel. The excess sludge
problem disappeared after this biocide application. The
Tillikum was then shifted to canola-based B10, and later
to B20, for an additional four months. The Tillikum ran
biodiesel blends under the same conditions that would
have been used for regular diesel, without any problems.
These results provide a strong indication that bacteria
played an important role in sludge formation in the Tilli-
kum purifier. Biocide application is therefore highly
recommended for marine conditions to prevent excess
sludge formation and filter clogging. The presence of the
biocide does not have a negative influence on engine
operation. Biocide products are typically pesticides,
which inhibit the growth of microbes over long periods
of time in very low concentrations. Further investigation
requires understanding the interaction between biocide
and specific microbial species. This information would
be useful to screen better biocides in terms of higher
Table 5. Monosaccharide composition analysis in the
aqueous solution by ion chromatography.
Sample Arabinose Galactose Glucose Xylose Fructose
Sludge– + + – –
Note: 1: The samples were hydrolyzed with 1.0 mol/L H2SO4 and then
diluted 1 000-fold before sampling. 2: “–” indicates no sugar was detected;
“+” indicates sugar was detected.
T. X. ZHANG ET AL.
Copyright © 2011 SciRes. ACES
performance and lower cost.
Excess sludge was formed in the purifier of the Tillikum
when the vessel was fueled by canola-based B5. The
sludge sample contained metal (11% - 14% ash), water
(~17%), major fraction of organic materials, and bacteria.
Active bacteria present in the sludge grew in culture me-
dia under both anaerobic and aerobic conditions. It is
suggested that the bacteria played a key role in sludge
formation, as shown by the absence of sludge problems
after addition of biocide to the fuel for the remainder of
the study period.
This work was supported by the Department of Energy
under Award Number (DE-FG36-06GO86032) through a
contract (200700001) between the Puget Sound Clean
Air Agency (PSCAA) and Washington State University
(WSU). The authors are thankful to Imperium Rene-
wables Inc (IRI). for providing biodiesel fuel, Washing-
ton State Ferries (WSF) for the pilot tests and samples
collecting, the Analytical Science Laboratory at the Uni-
versity of Idaho (UI) for ICP analysis, and the Franceschi
Microscopy and Imaging Center at WSU for microscope
images. The authors also gratefully acknowledge the
individuals for their great assistance in the project: Mr.
Tom Hudson from the PSCAA, Mr. Paul Brodeur and
Mr. Scott Calhoun from WSF, Mr. John Herkes of the UI,
Mr. Todd Ellis from IRI., Mr. Jake Millan, Mr. David W.
Larsen, Mr. Paul S. Smith, and Ms. Lisa Renehan from
the Glosten Associates, Mr. Shi-Shen Liao and Ms. Joan
Million from WSU.
 U. S. Department of Energy, “Biodiesel Handling and
Use Guideline,” 2nd Edition, 2006.
 R. O. Dunn, “Cold Weather Properties and Performance
of Biodiesel,” In: G. Knothe, J. Krahl and J. Van Gerpen
Eds., The Biodiesel Handbook, AOCS Press, Urbana,
2005, pp. 83-121. doi:10.1201/9781439822357.ch6.3
 D. S. Shrestha, J. Van Gerpen and J. Thompson, “Effec-
tiveness of Cold Flow Additives on Various Biodiesel,
Diesel and their Blends,” Transactions of ASABE, Vol. 51,
No. 14, 2008, pp. 1365-1370.
 H. Tang, C. Rhet and C. De Guzman, “Formation of In-
solubles in Palm Oil-, Yellow Grease-, and Soybean
Oil-Based Biodiesel Blends after Cold Soaking at 4˚C,”
Journal of the American Oil Chemists’ Society, Vol. 85,
No. 12, 2008, pp. 1131-1182.
 H. Tang, S. O. Salley and K. Y. S. Ng, “Fuel Properties
and Precipitate Formation at Low Temperature in Soy-,
Cottonseed-, and Poultry Fat-Based Biodiesel Blends,”
Fuel, Vol. 87, No. 13-14, 2008, pp. 3006-3017.
 S. Chen, “Report of Findings from 2004 WSF Biodiesel
Pilot Test,” Pullman, 2007.
 Y. Chao, N. Liu, T. Zhang and S. Chen, “Isolation and
Characterization of Bacteria from Engine Sludge Gener-
ated from Biodiesel-Diesel Blends,” Fuel, Vol. 89, No.
11, 2010, pp. 3358-3364. doi:10.1016/j.fuel.2010.05.041
 S. Apirattananusorn, S. Tongta, S. W. Cui and Q. Wang,
“Chemical, Molecular, and Structural Characterization of
Alkali Extractable Nonstarch Polysaccharides from Job’s
Tears,” Journal of Agricultural and Food Chemistry, Vol.
56, No. 18, 2008, pp. 8549-8557. doi:10.1021/jf801231y
 J. Dweck, and C. M. S. Sampaio, “Analysis of the Ther-
mal Decomposition of Commercial Vegetable Oils in Air
by Simultaneous TG/DTA,” Journal of Thermal Analysis
and Calorimetry, Vol. 75, No. 2, 2004, pp. 385-391.
 M. B. Dantas, M. M. Conceio, Jr., V. J. Fernandes, N. A.
Santos, R. Rosenhaim, A. L. B. Marques, I. M. G. Santos
and A. G. Souza, “Thermal and Kinetic Study of Corn
Biodiesel Obtained by the Methanol and Ethanol Routes,”
Journal of Thermal Analysis and Calorimetry, Vol. 87,
No. 3, 2007, pp. 835-839.
 B. He, J. Thompson, D. Routt and J. Van Gerpen,
“Moisture Absorption in Biodiesel and Its Petro-Diesel
Blends,” Applied Engineering in Agriculture, Vol. 23, No.
1, 2007, pp. 71-76.
 J. Lutz, M. Chavarra, M. L. Arias and J. F. Mata-Segreda,
“Microbial Degradation of Palm (Elaeis guineensis) Bio-
diesel,” Revista de Biologia Tropical, Vol. 54, No. 1,
2006, pp. 59-63.
 T. Schleicher, R. Werkmerster, W. Russ and R. Mey-
er-Pittoff, “Microbiological Stability of Biodiesel-Diesel
Mixtures,” Biore source Technology, Vol. 100, No. 2, 2009,
pp. 724-730. doi:10.1016/j.biortech.2008.07.029
 M. Owsianiak, L. Chrzanowski, A. Szulc, J. Stainewski,
A. Olszanowski, A. K. Olejnik-Schidt and H. J. Heipieper,
“Biodegradation of Disel/Biodiesel Blends by a Consor-
tium of Hydrocarbon Degraders: Effect of the Type of
Blend and the Addition of Biosurfactants,” Bioresource
Technology, Vol. 100, No. 3, 2009, pp. 1497-1500.
 R. De Philippis, R., C. Sili, R. Paperi and M. Vincenzini,
“Exopolysaccharide-Producing Cyanobacteria and Their
Possible Exploitation: A Review,” Journal of Applied
Physiology, Vol. 13, 2001, pp. 293-299.
 S. Kumar, K. Mody and B. Jha, “Bacterial Exopolysac-
charides—A Perception,” Journal of Basic Microbiology,
Vol. 47, 2007, pp. 103-117. doi:10.1002/jobm.200610203
 S. Rosalam and R. England, “Review of Xanthan Gum
Production from Unmodified Starches by Xanthomonas
comprestris sp.,” Enzyme and Microbial Technology, Vol.
39, No. 2, 2006, pp. 197-207.