Water content determination in biodiesel according to EN ISO 12937
R Schlink1 B Faas2
R. Schlink1, B. Faas
Introduction
Direct coulometric titration
Oven method
Ka
rl
F
is
ch
er
The presence of water in biodiesel reduces the calorific value and enhances
corrosion. Moreover, water promotes the growth of microorganisms and increases
the probability that oxidation products are formed during long-term storage.
These oxidation products can cause disturbances in the injection system and in the
engine itself. In view of this, the EN 14214 standard specifies a maximum water
content of 500 ppm for biodiesel.
EN ISO 12937 prescribes coulometric Karl Fischer titration (KFT) for determining
the water content of engine fuels In most cases the sample can be directly
By means of direct coulometric titration the water content of biodiesel
sample A (free of admixtures and antioxidants) is calculated for different
commercially available KF reagents.
In order to exclude reactions of additives with the iodine, the water content of
biodiesel sample B is determined by the KF oven method. The optimum oven
temperature should ensure complete water extraction in a reasonable time while
excluding side reactions. It is determined by means of a so-called heating curve
that is recorded at a heating rate of 2 ˚C/min over the temperature range 50 to
250 ˚C. In the heating curve, the amount of released water per unit time (drift) is
plotted as a function of the oven temperature.
Sample size
Water content
Sample A
rag
m
2
m
um
va
lue
m
um
v
alu
e
m
um
va
lue
x
mi
n
m
um
v
alu
e
x
max
val
u
e
x
1
enc
e
x
ma
x
- x
mi
n
atab
ili
ty
r
the water content of engine fuels. In most cases the sample can be directly
injected into the KF solution. In order to improve the solubility of the samples,
xylene is added to the KF reagent. In this work, we checked if some commercially
available KF reagents that contain solubilizers can be used. Additionally, we
determined the water content by an automated KF pipetting system and com-
pared the results to those obtained by manual KFT.
Many biodiesel fuels contain additives or impurities that can undergo side
reactions during coulometric KF titration. In these cases the fuel should not be
i j t d di
tl i t th KF
l ti
I t d th
l '
t
t t h
t
I
ti
f th
dd d KF
t
ll
lt
i th
The biodiesel sample B re-
leases its water content
between 50 and 110 ˚C.
Side reactions start at
about 150 ˚C. They are
more pronounced when
Di
ap
h
Mini
m
Maxi
m
Mini
m
Maxi
m
Mean
RS
D
Di
ff
er
e
Rep
ea
[mg]
[ppm]
[%] [ppm]
Coulomat AG (without solubilizer)
1600.4
2800.7 184.9
189.6
187.7 0.8
4.7
25.7
Coulomat AG + Xylene
967.1
1969.5 192.8
199.5
196.6 1.1
6.7
26.3
Coulomat AG Oil (CHCl3 + Xylene)
1761.2
1858.2 179.4
183.0
181.3 0.7
3.6
25.2
Coulomat A (CHCl3)
1339.6
2918.1 188.9
194.3
191.7 1.0
5.4
25.9
Coulomat AG-H (long-chain alcohol) –
1738.0
1763.1 186.1
191.5
189.1 0.9
5.4
25.8
1mean of ten determinations; 2 – = cell without diaphragm, = cell with diaphragm; 1 000 000 ppm = 100% (187.7 ppm = 0.019%)
228.545
injected directly into the KF solution. Instead, the sample's water content has to
be driven off at approx. 120 ˚C using a KF oven and transferred to the KF
coulometer titration cell in a flow of carrier gas. This process can also be fully
automated with the 774 Oven Sample Processor.
Repeatability r according to EN ISO 12937
According to EN ISO 12937, the test results must meet the following requirements
regarding repeatability:
Irrespective of the added KF reagent, all results are in the same ppm range.
However, the presence of admixtures in the biodiesel requires the addition of a
solubilizer. The numerical values for the difference xmax-xmin are much smaller than
those for the repeatability r. This clearly shows that direct KFT provides a far better
repeatability r for the water content in biodiesel than is required by EN ISO 12937.
Automated KF pipetting system
High sample throughputs should be handled by automation An automated
Based on the knowledge obtained from the heating curve, the following water
determinations were performed at an oven temperature of 120 ˚C.
more pronounced when
air is used as carrier gas.
Sample
size
Water
content
regarding repeatability:
The difference between two results, obtained by the same person
under identical test conditions, may exceed the following value r for
the repeatability only once in 20 cases:
where x is the mean value of all test results given as a mass fraction in
percent rounded off to 0 001%
r 0.018
x
74
,
High sample throughputs should be handled by automation. An automated
pipetting system transfers biodiesel sample B (stabilized with antioxidants) to the
coulometric cell.
Sample A
D
iap
hrag
m
2
Mi
nimu
m val
u
e
Maxi
mu
m value
Mi
nimu
m val
u
e x
min
Maxi
mu
m value
x
ma
x
Mean v
alu
e x
1
RSD
Di
ff
er
en
ce
x
max
- x
min
R
ep
eat
a
b
ili
ty
r
[mg]
[ppm]
[%] [ppm]
Coulomat AG Oven
– 1694.7
2060.7 177.6
188.4
183.9 1.9
10.8
25.4
1mean of ten determinations; 2 – = cell without diaphragm, = cell with diaphragm; 1 000 000 ppm = 100% (183.9 ppm = 0.018%)
Sample B
Manual KFT
Automated KF pipetting
system
Mean value water content x1 [ppm]
228.5 223.5
Relative standard deviation [%]
0.41 0.40
100
2
5
x
28.4
[%]
100 98
Maximum value x
max [ppm]
229.3
224.3
Instrumentation
percent rounded off to 0.001%.
Summary
Both manual direct injection KFT (228.5 ppm) and the automated KF pipetting
system (223.5 ppm) yielded comparable water contents. Moreover the results
comply with EN ISO 12937.
The results of the oven method comply with the requirements of EN ISO 12937.
The small variation between the results obtained by direct coulometric injection
(181.3…196.6 ppm) and those by the oven method (183.9 ppm) indicates that
the investigated biodiesel sample A contained no additives that reacted with the
iodine, as these would have interfered with the direct injections.
mean of ten determinations; cell without diaphragm, cell with diaphragm; 1 000 000 ppm 100% (183.9 ppm 0.018%)
Maximum value x
max [ppm]
229.3
224.3
Minimum value x
min [ppm]
226.8 221.8
Difference x
max – xmin [ppm]
2.5
2.5
Repeatability r
28.3
28.0
1mean of six determinations
Direct coulometric titration
Automated KF pipetting
Oven method
(manual KFT)
system (direct injection)
756 Coulometer
831 KF Coulometer
831 KF Coulometer
728 Magnetic Stirrer
801 Magnetic Stirrer
728 Magnetic Stirrer
815 USB Robotic Sample Processor 774 Oven Sample Processor
Summary
The higher the water content in biodiesel, the lower is its stability. Therefore EN 14214 limits the water content of fatty acid methyl esters to 500 ppm. Two biodiesel
samples were analyzed, one with (sample B) and one without (sample A) antioxidants and admixtures. Using different solubilizers, the water content of biodiesel
sample A was compared with that obtained by the Karl Fischer oven method. The biodiesel contamination with water was determined for sample B by direct coulometric
Karl Fischer Titration (KFT) and the automated KF pipetting system.
Both the direct injection methods, utilizing different solvents as solubilizers, and the oven method yielded comparable water contents for the investigated biodiesel
sample A (181.3…196.6 ppm). The good agreement between the results of direct injection and the oven method points to the fact that the investigated biodiesel
sample contained no interfering additives, as these would have reacted in the direct titrations. However, in case that some additives undergo side reactions, the KF oven
815 USB Robotic Sample Processor 774 Oven Sample Processor
800 Dosino
Dosing Unit 10 mL
1 Metrohm AG, CH-9101 Herisau/Switzerland, [email protected], phone +41 71 353 86 64
2 Deutsche Metrohm GmbH & Co. KG, D-70794 Filderstadt/Germany
Download a copy of this poster from http://www.metrohm.com/com/Applications (search for 8.000.6077EN).
p
g
,
,
g
,
method offers a largely matrix-independent solution as only the water of the sample is transferred to the titration cell of the KF Coulometer. Accordingly, the water
content for sample B, determined by the KF Pipetting system (223.5 ppm) agreed well with that obtained by manual KFT (228.5 ppm).
All three tested methods, i.e. the direct coulometric titration, the automated KF pipetting system and the oven method are ideally suited for the determination of the
water content in biodiesel and provide a far better repeatability r for the water content in biodiesel than is required by EN ISO 12937.
