Ch6.4.

2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 127

6.4.2
Stability of Biodiesel
Heinrich Prankl

Introduction
In 1997, the European Commission mandated the European Committee for
Standardization (CEN) to develop standards for biodiesel used as a fuel for diesel
engines and as heating fuel as well as the necessary standards for the analytical meth-
ods (1). During the drafting process, the lack of knowledge regarding biodiesel stabili-
ty became apparent. This was considered to be an important issue for which detailed
research was required. Between 2001 and 2003, the European Commission funded
the project “Stability of Biodiesel” (BIOSTAB; http://www.biostab.info; more details
are available through Ref. 2) which was carried out to clarify relevant questions about
stability determination methods, storage, and stabilization of biodiesel fuel. Both
European standards for biodiesel (EN14213, FAME as heating fuel, and EN14214,
FAME as automotive diesel fuel) have been available since July 2003.

Objectives
The objective of the project was to establish criteria and the corresponding analyti-
cal methods for determining the stability of biodiesel. The detailed aims were as
follows: (i) to develop appropriate methods for the determination of stability under
realistic conditions; (ii) to understand the influence of storage conditions on the
quality of biodiesel; (iii) to define a minimum level of natural and/or synthetic
antioxidants; and (iv) to determine the effects of fuel stability on the use of
biodiesel as automotive diesel fuel and as heating fuel.

Project Partners
Nine partners from industry, science, and research were involved in the project.
Seven were members of one or more working groups during the biodiesel stan-
dardization process. A very experienced biodiesel research consortium consisting
of the Bundesanstalt für Landtechnik (BLT, Austria; project coordination), Institut
des Corps Gras (ITERG, Pessac, France), Stazione Sperimentale Oli e Grassi
(SSOG, Milan, Italy), Institute of Chemistry of the University of Graz (Austria),
Graz University of Technology (TUG, Austria), OMV AG (Vienna, Austria),
TEAGASC (Oak Park Research Centre, Carlow, Ireland), NOVAOL (Paris,
France), and OLC-Ölmühle Leer Connemann (OLC, Germany) was assembled.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 128

Project Workplan
The project was divided into four thematic work areas. For each area, a leader was
responsible for coordination among the partners.

Determination Methods (Leader: ITERG, France)

The objective was to evaluate and develop accurate methods for the determination
of oxidation, storage, and thermal stability. In the area of oxidation stability, the
Rancimat test (EN 14112) had already been chosen for the biodiesel standards. The
relation between the induction period provided by this test and other quality para-
meters required clarification. Due to a lack of knowledge, no test method had been
chosen for thermal stability and storage stability. One of the main goals was to
select and develop a method for each item considering criteria such as reflecting
real conditions, correlating with quality parameters of biodiesel, precision, and
cost.

Storage Tests (Leader: SSOG, Italy)

Previous research demonstrated that storage conditions (e.g., temperature, light,
atmosphere, presence of prooxidant metals) have a strong effect on storage behav-
ior. The nature of the feedstock might also have a considerable influence on the
final result. The main task was to carry out a systematic study of the changes in
biodiesel samples obtained from different feedstocks and prepared by different
production technologies, during a long-term storage experiment under real-world
conditions.

Antioxidants (Leader: Institute of Chemistry, University Graz, Austria)

Natural antioxidants, such as tocopherols and carotenoids, delay the oxidation of
vegetable oils. The antioxidant effect depends on the type and amount of antioxi-
dant present. Antioxidants are also present in biodiesel derived from vegetable oils,
and the amounts will depend on the vegetable oil used and on the process technology.
Consequently, it may be necessary to add synthetic and natural antioxidants to the
biodiesel to improve oxidative stability. The objective of this task was to evaluate
the stabilizing effects of available natural and synthetic antioxidants in commercial
biodiesel. Over 20 synthetic and natural antioxidants were evaluated, and optimum
usage levels were determined.

Utilization of Biodiesel
Biodiesel is used both as automotive diesel fuel and as heating fuel. Bench and
field tests were carried out on vehicles and injection systems, as well as on heating
systems to establish a connection between laboratory test methods and effects dur-
ing use.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 129

Biodiesel as Automotive Diesel Fuel (Leader: University of Technology, Graz,

Austria). The test program comprised bench tests with three different injection
systems and test fuels with low-, standard-, and high-stability, long-term tests with
two diesel engines, a fleet test using biodiesel with low stability, and a fleet test
with blends of fossil diesel fuel and biodiesel with low stability.

Biodiesel as Heating Fuel (Leader: OMV, Austria). The objective of this task
was to investigate the effects of the fuel stability when used in heating systems.
The effects of fuel stability during the application and the operation parameters of
the residential heating systems using blended fuels were studied. The test program
comprised bench tests (emissions, operation, long-term testing) with heating sys-
tems and field tests with eight heating systems.

Results
Determination Methods
Oxidative Stability. Seven biodiesel samples (methyl esters of rapeseed oil, sun-
flower oil, used frying oil, and tallow) were evaluated with the Rancimat test (EN
14112; Fig. 1). Determination of quality parameters was carried out on aliquot

Fig. 1. Rancimat for determining the oxidation stability.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 130

samples every 0.5 h. At the end of the Rancimat induction period (RIP), the sam-
ples did not meet fatty acid methyl ester (FAME), or oil and fat specifications such
as viscosity, acid value (AV), ester content, or peroxide value (PV). The main con-
clusion was that the induction period determined by conductivity correlates well
with the degradation of quality parameters by the Rancimat test (3).

Storage Stability. At the beginning of the project, two test methods were evaluat-
ed, namely, ASTM D4625 (storage at 43°C for 24 wk) and an IP48/IP306-like
method at 90°C with air flow above the surface of the sample. For each method,
seven quality parameters were defined. Because it was difficult to correlate ASTM
D 4625 and the results of an accelerated method initially proposed (accelerated
IP48/IP306-like method at 90°C), it was decided to use the Rancimat apparatus
with special modifications for storage stability evaluation. A stream of purified air
(10 L/h) was passed over the surface of 3 g of sample heated to 80°C for 24 h.
Then the PV, ester content, and polymer content were measured. The modified
Rancimat test is suitable for use in terms of repeatability, significance, and ease of
handling. The PV shows the best correlation with ASTM D 4625 (storage at 43°C
for 24 wk). Using this method, “poor stability” and “good stability” samples can be
distinguished (4).

Thermal Stability. Initially, it was decided to keep the aging conditions of ASTM D
6468 (150°C, 180 or 90 min) because they were considered to be reasonably close to
real-world conditions. Seven quality parameters to be evaluated before and after the
aging test were defined. However, the variation of quality parameters (AV, Rancimat,
ester content) after the aging test was too low to be measured correctly. Finally, it was
decided to use the Rancimat apparatus with a procedure especially modified for ther-
mal stability evaluation. The sample (8 g) was aged for 6 h at 200°C in open tubes
with air exposure. After aging and cooling, the polymer content was determined by
high-performance liquid chromatography (HPLC). The modified Rancimat test is
suitable for use in terms of repeatability and ease of handling.

Storage Tests
A systematic study of the chemical and physical changes in 11 different samples of
biodiesel was carried out between July 2001 and October 2002. The results allow
an evaluation of the effects of different feedstocks and technologies as well as the
use of selected antioxidants on the chemical properties of aged biodiesel samples.
It was not possible to observe strong changes in 15 monitored characteristics. All
samples met the specification limits even at the end of the storage period, with the
exception of RIP; PV changes differed, depending on samples. For samples initially
not too oxidized, the PV increase was slow. For samples initially oxidized, the PV
first increased and then decreased due to the formation of secondary oxidation
products.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 131

The most important changes were recorded in oxidative stability as shown by the
Rancimat test. This means that aging of biodiesel occurs independently of the moni-
tored parameters, making biodiesel less stable with time. The Rancimat induction
period decreases with time. The rate depends on the quality of the sample and on stor-
age conditions. The Rancimat takes a picture of the actual situation, but it is impossi-
ble to predict the RIP value after long-term storage. There are aging processes that
cannot be observed by analyzing the parameters reported in EN 14213 and EN
14214; thus, a method for storage stability prediction is necessary.
Proper additization increases the induction period greatly and ensures that the
sample meets the specification for oxidative stability for at least 6 mon. Super-
additization procedures leading to an induction period >20 h have no meaning and
might have a negative effect on other parameters (e.g., Conradson carbon residue).
The need for correct storage and logistic solutions to avoid the contact of biodiesel
with air during its complete life cycle has been pointed out.

Antioxidants
On the basis of the findings of an extensive literature survey, 20 natural and syn-
thetic antioxidants were selected, all of which are commercially available at an
affordable price. The stabilizing effect of the selected antioxidants was evaluated
with biodiesel prepared from four different raw materials: RME (rapeseed oil
methyl ester), SME (sunflower oil methyl ester), UFOME (used frying oil methyl
ester), and TME (tallow methyl ester). Both distilled and undistilled biodiesel sam-
ples were evaluated, and Rancimat induction times were used to indicate oxidation
stabilities. Antioxidants with a good stabilizing effect were tested at different con-
centrations to determine the optimum antioxidant levels. The effect of natural
antioxidants on the oxidative stability of biodiesel was also evaluated. The relatively
high stability of RME, which cannot be attributed solely to the natural antioxidants
detected, is being investigated further.
Generally, the limit for the proposed oxidative stability parameter could be
achieved by adding antioxidants to all different types of biodiesel. Within the vari-
ety of antioxidants, synthetic products were more effective than natural antioxi-
dants. The efficiency and the required amount of the different antioxidants strongly
depend on the feedstock and technology used for biodiesel production (5). Under
the given conditions, no significant negative influence of antioxidants on fuel
behavior was observed. The influence of antioxidants on engine performance was
not investigated within the project. However, to minimize possible negative
effects, it is recommended that antioxidants be used at very low concentrations.
The present report does not include any recommendations for the use of specific
antioxidants. Long-term engine tests must be conducted to study the influence of
synthetic antioxidants on engine performance.
Tocopherols (α-, δ-, and γ-) delay the oxidation of SME, RME, waste cooking
oil methyl ester (WCOME), and TME in some cases by more than a factor of 10

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 132

compared with methyl esters without tocopherols. γ-Tocopherol was the most and
α-tocopherol the least effective of the three natural antioxidants. Their antioxidant
effect increased with concentration up to an optimum level. Above the optimum
level, the increase in antioxidant effect with concentration was relatively small.
The stabilizing effect of tocopherols also depended on the composition of the
methyl ester. The order of effectiveness was as follows: TME > WCOME > RME >
SME.
Oxidation of unsaturated fatty compounds begins with the build-up of perox-
ides. Irreversible oxidation, indicated by viscosity increase, starts only after the
peroxides reach a certain level. Tocopherols stabilize the unsaturated fatty com-
pound esters by reducing the rate of peroxide formation, thereby extending the
time required to reach the peroxide level at which viscosity starts to increase.
The carotenoids, astaxanthin and retinoic acid, had no detectable effect on the sta-
bility of SME. Similarly, β-carotene added to camelina oil methyl ester (CME) along
with some α-tocopherol, to give the same maximum absorbance at 448 nm as RME,
had no stabilizing effect on the methyl ester. However, a carotenoid, at much higher
level than β-carotene, was detected in RME, but it was not present in less stable
methyl esters such as CME and SME. The detected carotenoid did not extend the peri-
od of stability of RME, but it changed its oxidation pattern by reducing both the rate
of peroxide formation and of viscosity increase during oxidation.

Utilization of Biodiesel
Biodiesel as Automotive Diesel Fuel. As bench tests, long-term tests were car-
ried out with three different modern injection systems (heavy-duty common rail,
passenger car distribution pump, passenger car common rail) on the test bench.
Three fuel qualities, i.e., RME with low (induction period 1.8–3.5 h according to
EN 14112), standard (induction period 6 h), and high stability (induction period
14–18 h), were used.
Wear and sedimentation were normal for the runtime of each injection system.
All enumerated effects were more salient in those parts that were operated in test
runs with fuels with lower oxidative stability. Fatty deposits were detected only in
those parts of the system that were operated under very strong conditions (RME
low oxidation stability, no change of the fuel during the complete test run at the
injection system test bed). In all other systems, no critical sedimentations were
detected. After the test runs, the functionality was given for each system tested.
Two long-term, real-world engine tests were carried out on the test bench
fueled by biodiesel with low (induction period 3.5 h according to EN 14112) and
high stability (induction period 20 h). The direct injection diesel engines were
equipped with a modern common rail injection system. The duration of each test
was 500 h. The measured differences in emissions and power between diesel and
RME operation at the beginning of each test were considered normal for biodiesel
use. The power loss and difference in the injection amount after 250 h of runtime

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 133

were higher than expected. Engine inspection revealed normal abrasion for a 500-h
runtime and no significant differences due to diesel operation.
For the fleet test with pure biodiesel, four passenger cars were operated with a
low-stability biodiesel from July 2001 to November 2002. The test fuel was
preaged by a special treatment with temperature and air. The oxidation stability
determined by the induction period (Rancimat, 110°C) could be reduced from 7 h
to <2 h. Two cars were equipped with a distributor injection pump; the other cars
had a unit injector system. The cars were used in typical operation, mainly on the
highway. Distance and fuel consumption were recorded in a log book. Some tem-
peratures (engine oil, fuel filter, fuel tank, ambient) were recorded automatically
during the entire test period. The total driving distance ranged from 21,000 to
60,000 km/car.
Initial and final tests were conducted at the roller test bed, comprising perfor-
mance and exhaust emission analyses (CO, HC + NOx, particles). The differences
in emissions and performance could not be assigned to the biodiesel operation (in
combination with the results of the injection system check after the test run).
Before and after the field test, all limited emissions were below the EURO 3 level
(EURO 3 limit vehicle group 2, 1305 kg < reference weight < 1760 kg).
After the tests, the injection systems were inspected by the manufacturer. All sys-
tems were functioning normally. Swelling of elastomers was found in a distributor
injection pump, which can lead to leakage particularly when using petrodiesel fuel.
RME deposits were found on several parts of a distributor injection pump. Traces of
oxidation and notable corrosion were found on some parts of the unit injectors. One
fuel filter blocked and some fuel filters had to be changed during the winter. It should
be noted that the results of the field test were obtained from four individual cars. Thus,
a general conclusion about performance with low-stability biodiesel cannot be drawn.
This would require an extensive fleet of cars and total coverage of all field influences.
For the vehicle fleet test with a diesel fuel/FAME blend, a 19-mon test run with
four diesel vehicles (light- and heavy-duty passenger cars) was carried out. The per-
formance of a diesel fuel (EN 590) blended with 5% UFOME was evaluated.
Parameters controlled regularly during the test were fuel quality, cleanness of the fuel
storage and supply system, and its operability, vehicle exhaust emissions, engine
lubricant performance, drivability of the vehicles in warm and cold conditions, clean-
ness and wear of the vehicles’ fuel systems and fuel injection equipment.
After ~66,000 km driven in daily traffic, there were no significant complaints
about the operability of the vehicles registered. No excessive wear or deposit
build-up occurred during this period compared with the operation with diesel fuel,
which was known from previous fleet tests. The engine lubricant gave a perfor-
mance similar to that experienced with pure diesel fuel (regular drain interval
15,000 km), and there was no need for premature change of the lubricant. A mod-
erate increase in exhaust emissions was measured during the test interval.
The main components of the fuel storage and delivery system did not suffer
from deterioration. The quality of the fuel in terms of most specification parame-

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 134

ters was generally constant. Microbial contamination did not occur, probably due to
proper cleaning and especially drying of the logistic system at the start of the test. The
oxidation stability in the bottom layers of the storage vessels increased dramatically
and significantly exceeded the limit of EN 590 (25 g/m3; EN ISO 12205).

Biodiesel as Heating Fuel. In general, the bench tests in the 1-h heating cycle test
gave two different results depending on the operating mode of the heating system, i.e.,
stationary or nonstationary conditions. Under stationary conditions, the emissions
were as low as expected and all units met the general standards for heating units.
However, during the starting procedure for the heating unit (semi-warm start, nonsta-
tionary condition), concentrations of the hydrocarbons and the carbon monoxide in the
flue gas were higher than some seconds after the start. These irregularities depended
in large part on the technology of the heating units. In addition, those heating systems
that have high emissions during the starting process have worse CO and CxHy emis-
sions with an increased percentage of FAME in the blend.
In the long-term test of three different heating units with different fuels, one fact
can be seen clearly. The blends made from FAME products having critical stability
(aged artificially, stored for 1 y, or distilled and stored for 1.5 y) caused problems in
the oil feed rate counter. In the field test, no significant problems have been reported
to date. But it must be noted that the conditions for the test were as perfect as possible.
Issues to be resolved include the following: (i) microbiological attack of the
blends by bacteria or microbes because of the fast biological decomposition of
FAME; (ii) the storage stability of the blends, which is influenced by the materials of
the tank, plastics, softening agents; (iii) higher foaming behavior during filling, which
leads to tank runover and problems with correct measurement of volume at the filling
procedure; (iv) decomposition of the Euromarker (European tax marker) or other
additives; (v) swelling effect of the used plastic material, i.e., sealing, pipes, parts in
the flow counter, pumps, and nozzles.

Summary
The method for determining oxidative stability (EN 14112-Rancimat test) correlates
well with the development of quality parameters. Biodiesel shows high resistance
against temperature. A method for detecting thermal stability using the Rancimat
equipment by evaluation of the polymer content was proposed. A method for the
determination of the storage stability using the Rancimat equipment was also pro-
posed. Samples with poor and good stability can be clearly separated.
Biodiesel can be stored under normal storage conditions without a dramatic
change in quality parameters for a period of 1 y. Most changes affect oxidative stability
and PV, depending on the quality and the storage conditions. The right additization
ensures that the samples continue to meet the specifications after a storage period of 1
y. Nevertheless, proper storage and fuel transport/logistics are absolutely necessary.
Contact with air, water, and sunlight must be avoided.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 135

The limit for oxidative stability can be reached by appropriate addition of antioxi-
dants with all different types of biodiesel. Synthetic antioxidants are more effective
than natural ones. The efficiency and the amounts of the different antioxidants
required depend strongly on the feedstock and biodiesel production technology. No
significant negative influence of antioxidants on fuel performance has been observed
to date. The influence of additives on engine performance was not investigated within
the project. To minimize possible negative effects, however, it is recommended that
antioxidants be used at very low concentrations. α-, δ- and γ-Tocopherol significantly
delay the oxidation of unsaturated fatty compounds, with γ-tocopherol as the most
effective.
The results of bench tests with injection systems showed that functionality was
present with all injection systems after long-term tests with different fuel stabilities.
Wear and sedimentation were normal for the runtime. Deposits similar to fat were
found only in parts being operated under very strong conditions. No effects related to
fuel stability were observed in long-term engine tests.
The injection systems of four vehicles from a field test fueled with low-stability
biodiesel were acceptable after the test. However, traces of oxidation could be found
on some parts of the unit injector. Fuel deposits were found on several parts of a unit
injector. Although no problems in relation to the low stability of the fuel occurred, it is
recommended to ensure that the minimum stability required by the standards is met to
avoid problems during application under very sharp conditions.
The field test using a low-stability biodiesel in a 5% blend with fossil diesel did
not have negative effects in terms of wear, deposits, engine lubricant, fuel storage, and
delivery, microbiological contamination, or fuel quality. Critical points that remain
include oxidative stability (EN ISO 12205 as prescribed by EN 590) and the absolute
requirement to clean and keep clean the storage infrastructure (vessels, pipelines) of
biodiesel.
When using biodiesel as a heating fuel, no differences existed in short-term bench
tests with five heating units under stationary conditions. In contrast, the HC- and CO-
emissions during the starting procedure were higher than with pure fossil heating fuel.
The results depend significantly on the technology of the heating units. In long-term
tests of three units with different fuels, problems occurred in the oil feed rate counter
when using fuels with very low stability.
To date, no significant problems have been reported from a field test with eight
heating units during two heating seasons. Fossil heating fuel containing 5% FAME
(RME, UFOME, and antioxidants) was used. Open questions are focused on microbi-
ological attack, the storage stability of blends, as well as foaming and material com-
patibility of blends.

Acknowledgments
The paper is based on the results of the project “Stability of Biodiesel” carried out in the 5th
framework program of the European Commission. Many thanks to the project partners for
providing the results and for the excellent teamwork.

Copyright © 2005 AOCS Press

Ch6.4.2(Biodies)(127-136)(Co#1) 6/6/05 3:44 PM Page 136

References
1. M/245 Mandate to CEN for the Elaboration and Adoption of Standards Concerning
Minimum Requirement Specification Including Test Methods for Fatty Acid Methyl
Ester (FAME) as Fuel for Diesel Engines and for Heating, European Commission, 29
January 1997.
2. Stability of Biodiesel Used as a Fuel for Diesel Engines and Heating Systems.
Presentation of the BIOSTAB Project Results, Graz, July 3, 2003, published by BLT
Wieselburg, Graz, Austria, 2003, (http://www.biostab.info).
3. Lacoste, F., and L. Lagardere, Quality Parameters Evolution During Biodiesel
Oxidation Using Rancimat Test, Eur. J. Lipid Sci. Technol. 105: 149–155 (2003).
4. Bondioli, P., A. Gasparoli, L. Della Bella, and S. Tagliabue, Evaluation of Biodiesel
Storage Stability Using Reference Methods, Eur. J. Lipid Sci. Technol. 104: 777–784
(2002).
5. Mittelbach, M., and S. Schober, The Influence of Antioxidants on the Oxidation
Stability of Biodiesel, J. Am. Oil Chem. Soc. 80: 817–823 (2003).

Copyright © 2005 AOCS Press