Determination of sodium, potassium, calcium and magnesium ...

Analytica Chimica Acta 718 (2012) 116– 120
Contents lists available at SciVerse ScienceDirect
Analytica Chimica Acta
jou rn al hom epa ge: www.elsevier.com/locate/aca
Determination of sodium, potassium, calcium and magnesium cations in
biodiesel by ion chromatography
Lilia Basilio de Caland, Eva Lúcia Cardoso Silveira, Matthieu Tubino ∗
University of Campinas, Institute of Chemistry, POB 6154, 13083-970 Campinas, SP, Brazil
a r t i c l e i n f o
Article history:
Received 3 October 2011
Received in revised form
15 December 2011
Accepted 28 December 2011
Available online 8 January 2012
Keywords:
Biodiesel
Inorganic cations
Ion chromatography
Analysis
Biofuel
1. Introduction
a b s t r a c t
Biodiesel is defined as a mixture of mono-alkyl esters of long
chain fatty acids, obtained from oils and fats from vegetable and
animal sources. Its use is strongly associated with the replacement
of fossil fuels in diesel engines due to economical and ecological
aspects [1].
Vegetable oils and animal fats are mainly composed of triglycerides
(TGs) consisting of long chain fatty acids chemically bonded
to glycerol. Biodiesel is obtained through the transesterification
reaction, where TGs react with a short-chain alcohol in the presence
of a catalyst [2].
Many catalysts have been employed in biodiesel synthesis.
Some of them are alkaline and homogeneous, such as sodium and
potassium hydroxides and methoxides in alcoholic solution [3].
Heterogeneous catalysis is also used in the transesterification reaction
[4]. Residues of the catalyst used can be present in the prepared
biodiesel as impurities due to incomplete purification of the product.
Calcium and magnesium can be introduced into the biodiesel
during the purification process, either by washing with hard water
or by the addition of drying agents such as MgSO 4 and CaO. These
inorganic cations can also be inherently present in the raw-material
and persist in the final product [5].
∗ Corresponding author. Tel.: +55 19 3521 3133; fax: +55 19 3521 3023.
E-mail address: tubino@iqm.unicamp.br (M. Tubino).
0003-2670/$ – see front matter ©
2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.aca.2011.12.062
This work reports an ion chromatographic (IC) method for the quantitative determination of inorganic
cations (Na + , K + , Mg 2+ and Ca 2+ ) in biodiesel samples that were synthesized from different vegetable
oils and fat. The proposed method uses water extraction, heating and ultrasound. The limits of detection
(LOD) for each ion, in milligrams of the analyte per kilogram of biodiesel (mg kg −1 ), were respectively:
0.11 (Na + ); 0.42 (K + ); 0.23 (Ca 2+ ); and 0.36 (Mg 2+ ). The accuracy of the method was studied through
recovery tests. For comparison, two samples were also analyzed using an Inductively Coupled Plasma
Optical Emission Spectrometry (ICP-OES) procedure. The paired Student t test and the Snedecor F test
showed that both methods offer equivalent results in terms of accuracy and precision. The operational
simplicity, accuracy and precision of the proposed method suggest that it can be a good alternative for
the determination of inorganic cations in biodiesel samples.
© 2012 Elsevier B.V. All rights reserved.
In Europe, the regulation of biodiesel is established by norm EN
14214. In the United States, the regulation is norm ASTM D-6751. In
Brazil, Resolution ANP number 7/2008 (19/3/2008), of the National
Agency of Petroleum, Natural Gas and Biofuels [6] supplies the specifications
of pure biodiesel (B100) and the methodologies for its
characterization. In agreement with these legislations, the maximum
limit allowed, for the total of sodium plus potassium cations,
is 5 mg kg −1 , the same value is applied to the total of magnesium
plus calcium.
According to the European norms (EN 14108 and 14109) [7,8],
the determination of sodium and potassium in biodiesel samples
consists of sample dissolution with an organic solvent (xylene,
cyclohexane or petroleum ether) and analysis through flame atomic
absorption spectrometry, FAAS. The determination of calcium and
magnesium is regulated by EN 14538 [9] that indicates analysis
through Inductively Coupled Plasma Optical Emission Spectrometry
(ICP-OES). The monitoring of the concentrations of Na + , K + , Mg 2+
and Ca 2+ in biodiesel is necessary due to the fact that salts of these
metals form solid deposits in engines [10,11].
Due to its sensitivity and to the possibility of simultaneous
multi-element determination, the ion chromatography method
requires only small amounts of sample. The stability of the separation
system, good selectivity and speed of analysis allow the
range of IC applications to expand continuously [12]. Also, one can
note the relatively low cost of this analytical technique since it is
not necessary to use expensive reagents or solvents or to consume
specialty gases, while the equipment can be installed in a small
space in the laboratory.

The technique of ion chromatography is nowadays widely used
in analysis of biological [12–14] and environmental [15] samples,
oil products [16] and water [17–20] for the simultaneous determination
of ions, particularly at the level of trace concentrations.
There are a few cases where this technique was explored for the
analysis of biodiesel for the determination of metallic ions [21,22]
but, as far we know, no full detailed papers using this technique are
reported in the literature.
In this study an ion chromatographic (IC) method for simultaneous
determination of Na + , K + , Mg 2+ and Ca 2+ in samples of biodiesel
is developed. The ions are extracted from the matrix using water,
heating and ultrasound. The detection is done using conductimetry.
The proposed extraction procedure of the ions from the sample
is very attractive because the use of organic solvents is avoided.
2. Experimental
2.1. Materials and methods
2.1.1. Reagents and solutions
All reagents were of analytical grade. The aqueous solutions
were prepared with deionized water, ultrapure Type I
(18.2 M cm), obtained from a Milli-Q system (Millipore, Bedford,
MA, USA).
Standard stock 1000 mg L −1 solutions of sodium and potassium
ions were prepared by weighing adequate masses of the respective
anhydrous salts, NaCl and KCl, into distilled, deionized water.
In the case of magnesium and calcium, stock solutions containing
1000 mg L −1 were provided by Sigma–Aldrich (St. Louis, MO,
USA). Mixed cation standard solutions of 1.00, 2.00, 3.00, 4.00, 5.00
and 0.20, 0.40, 0.60, 0.80 and 1.00 mg L −1 were used to calibrate
the ion chromatograph. All standards and samples were stored in
polyethylene bottles.
Recovery assays to test the extraction effectiveness from
biodiesel were done using sodium cyclohexanebutyrate, potassium
cyclohexanebutyrate, calcium cyclohexanebutyrate and magnesium
cyclohexanebutyrate. Known amounts of the aforementioned
salts were added to biodiesel samples, then submitted to the extraction
procedure described in Section 2.3.
2.1.2. Samples
Vegetable oils and bovine fat were purchased in local shops.
Seven samples of different types of biodiesel were used in this
study. With the exception of the biodiesel from rough soy oil, which
was fabricated and furnished by an industry, all were prepared in
our laboratory [23]. Biodiesel prepared from sunflower oil was used
to perform experimental tests in order to optimize the extraction
process with ultrasound. The optimized process was then applied
to seven other samples of biodiesel. Six of them were prepared
from vegetable oils using the transesterification with methanol,
i.e., for soy, rough soy, corn, sunflower and canola oils, and also
through the transesterification with ethanol for canola oil. A seventh
sample was prepared from animal fat (bovine) provided by a
slaughterhouse, using the procedure with methanol.
2.2. Extraction of the cations with aqueous solution
Each sample was treated using ultrasound assisted extraction,
in order to transfer the ions from biodiesel to an acidic aqueous
solution. Aliquots (10 ± 0.001 g) were directly weighed into centrifuge
tubes. 20 mL of deionized water and 50 �L of a 1.0 mol L −1
HNO 3 aqueous solution (in order to maintain the pH of the samples
at a suitable level for IC analyses) were added. The mixture was
shaken for 1 min in a vortex apparatus (Motion II). Sequentially,
the tube was partially immersed in a thermostatic bath set at 85 ◦ C
for 30 min. The tube was then placed in an ultrasonic bath (Unique
L.B. Caland et al. / Analytica Chimica Acta 718 (2012) 116– 120 117
Ultraclean 14000) operating at a frequency of 50 kHz and sonication
for 15 min. In sequence it was centrifuged at 1000 rpm for 5 min.
The aqueous phase was separated and collected in a 25 mL volumetric
flask and deionized water was added up to the mark. Before
being injected into the chromatograph, aliquots from this solution
were filtered through a 0.7 �m glass microfiber filter.
2.3. Instruments
Analyses were conducted using an ion chromatograph equipped
with a conductivity detector (Metrohm model 761), a dialysis system
and a 100 �L injection loop. A cation exchange column with
carboxylic groups on silica gel, model Metrosep C4 100, with 7 �m
particles, was used. Its dimensions were: 100 mm length and 4 mm
internal diameter.
In this work an optimization of the eluent solutions aiming
was carried out to obtain better chromatographic separation of the
studied cations. Two eluent solutions were employed (oxalic acid;
nitric/dipicolinic acid) in various concentrations. It was observed,
based on the investigated chromatographic parameters, that the
eluent dipicolinic/nitric acid at 1.7 mmol L −1 (dipicolinic acid) and
2.5 mmol L −1 (nitric acid) enabled a shorter analysis time and quite
satisfactory separation of the cations. Therefore, this solution was
employed, at a flow rate of 0.9 mL min −1 . The eluent was always
previously degassed with ultrasonication for 15 min.
For comparison purposes samples of sunflower oil and bovine
fat biodiesels were also analyzed using ICP-OES (PerkinElmer,
Norwalk, EUA). The following spectral lines were used: sodium
588.995 nm; potassium 766.490 nm; magnesium 280.271 nm and
calcium 317.933 nm.
3. Results and discussion
3.1. The analytical method
The determination of Na + , K + , Mg 2+ and Ca 2+ in biodiesel is
of importance as their presence results in solid deposits inside
engines. However, as they usually are present in low concentrations
in a matrix as complex as biodiesel, the analytical procedures
are not trivial.
The results obtained in this work demonstrate that it is possible
to simultaneously analyze these alkaline and alkaline-earth cations
using ion chromatography. The ions were initially extracted using
water as extracting solvent, in the presence of acid, with the aid of
heating and ultrasound. This extraction employs the higher affinity
of the inorganic cations for a highly polar solvent such as water, in
comparison to a non-polar organic solvent (biodiesel).
The literature describes different optimum ultrasonic exposition
times, varying from 5 to 180 min, in order to achieve complete
extraction for different samples [24]. To verify the efficiency of
the ultrasound extraction, known amounts of each analyte were
separately added to sunflower biodiesel samples and exposed to
ultrasound during different times, from 0 to 30 min.
The results are shown in Fig. 1 where it can be observed
that larger samples amounts (10 g) and longer ultrasound assisted
extractions were beneficial to the recovery yields. The application
of ultrasound for 10 min (Na + , K + and Ca 2+ ) and 15 min (Mg 2+ )
ensures good recoveries. Aiming to analyze simultaneously these
four metals, 15 min of sonication was employed. A decrease in the
recovery of Mg 2+ was observed beyond ca. 15 min of sonication.
According to Souza et al. [25] this decreasing could indicate the
formation of non-soluble species, reducing the analyte concentration
in the solution. However, it is difficult to imagine what kind of
insoluble species could be formed in the presence of HNO 3 and in
such low concentrations.

118 L.B. Caland et al. / Analytica Chimica Acta 718 (2012) 116– 120
Fig. 1. Recovery (%) from a sunflower biodiesel containing (2.00 ± 0.01) mg L −1 of each cation after different ultrasound treatments with two biodiesel amounts 5 g (•) and
10 g (�).
If we consider the data exposed in Fig. 1 and calculate the overall
mean (among the extraction values from 0 to 30 min) of the extraction
recovery for 10 g of biodiesel, the value for sodium, whose salts
are very soluble, is equivalent to that of magnesium, indicating that
other phenomena are probably affecting the process. Sodium and
magnesium are neighbors in the third row of the periodic table and
potassium and calcium are neighbors in the fourth row. Curiously,
the extraction recoveries can be placed in “pairs” as functions of
their mean atomic masses, i.e.,: Na + , 87.7% and 23.0; Mg 2+ , 86.4%
and 24.3; K + , 96.8% and 39.1; Ca 2+ , 100.3% and 40.1. This correlation
indicates that there is some influence of the cation mass in the
extraction process. Therefore, phenomena other than formation of
insoluble species are probably occurring as, for example, adsorption
on the walls of the flask, among other possibilities. However,
a detailed study is beyond the aim of the present work.
3.2. Concentrations of the Na + , K + , Mg 2+ and Ca 2+ ions in the
aqueous extract
Fig. 2 shows an ion chromatogram obtained in the analysis of
a biodiesel from soy oil where the signals of sodium, potassium,
magnesium and calcium ions are clearly shown. Table 1 reports
the retention times, the analytical equations, and the detection and
quantitation limits.
Table 2 also shows the figures of merit for the determination of
Na + , K + , Ca 2+ and Mg 2+ in biodiesel by the proposed method, after
ultrasound assisted aqueous extraction.
The linearities of the analytical curves are quite satisfactory as
the coefficient of determination, R 2 ≥ 0.994. The limits of detection
(LOD) and limits of quantitation (LOQ) obtained are adequate for
the analysis of these cations in biodiesel as they are lower than
the highest concentrations allowed by the norms, i.e., 5 mg kg −1
for sodium plus potassium and also for magnesium plus calcium,
(Brazil, ANP resolution n. 07/2008; and by the fuel legislations of
Fig. 2. Chromatogram obtained in the analysis of biodiesel from soy oil containing
sodium, potassium, magnesium and calcium ions.

Table 1
Analytical curves (A = a + bC) a ; coefficient of determination, R 2 ≥ 0.994); limits of
detection (LOD in mg kg −1 ) and limits of quantitation (LOQ in mg kg −1 ) for the
analysis of Na + , K + , Ca 2+ and Mg 2+ in biodiesels through ion chromatography.
Na + K + Ca 2+ Mg 2+
Retention time
(min)
2.99 4.00 5.34 7.00
a −1.10 ± 1.82 −1.58 ± 4.71 1.11 ± 3.46 7.6 ± 12.4
b 53.8 ± 1.8 37.3 ± 1.4 48.7 ± 1.0 114.4 ± 3.7
sb 1.82 4.71 3.46 12.4
LODb 0.11 0.42 0.23 0.36
LOQc 0.33 1.25 0.70 1.07
a A = Area, in (�S cm−1 ) min; a = linear coefficient; b = slope; C = concentration in
mg kg−1 .
b LOD = 3.3 s/b, where s is the standard deviation of the linear coefficient and b is
the slope of the analytical curve.
c LOD = 10 s/b [27].
USA, EU and Australia). Similar limits of determination and of quantitation
were obtained by Piovezan et al. [26] using liquid–liquid
extraction and capillary electrophoresis for the quantitative analysis
of such cations.
In order to verify the efficiency of the extraction process used,
tests were performed by adding known quantities of these four
cations to a biodiesel prepared from sunflower oil. Extraction was
done as described. The results are shown in Table 2.
The sample of biodiesel from sunflower oil presented recoveries
between minimums of 90.6% (Mg 2+ ) and maximums of 108.2% (K + ),
with 15 min of sonication, indicating that the extraction procedure
used was sufficiently efficient to remove the cations present in the
mixture of the long chain alkyl esters.
L.B. Caland et al. / Analytica Chimica Acta 718 (2012) 116– 120 119
Currently, in analytical procedures, recovery percentages in the
range from 70 to 120%, with precision of about 20%, are accepted.
However, depending on matrix complexity, this range can be
extended from 50% to 120% with precision of 15% [27]. Considering
the extraction efficiency obtained in this work, the proposed
method presents accuracy sufficient to be applied for the determination
of the four metallic cations analyzed in biodiesel samples.
With respect to the precision obtained it also can be considered
sufficient.
3.3. Statistical comparison of the proposed method with ICP-OES
The accuracy of the method was determined through recovery
experiments performed with a biodiesel prepared from sunflower
oil. The aqueous extract was analyzed with the method proposed
herein. The IC results were compared with those obtained with
the ICP-OES procedure. In these experiments known quantities of
an organic salt of each cation studied were directly added to the
biodiesel.
With the objective to demonstrate the equivalence, in terms of
accuracy and precision of the proposed method with ICP-OES, analyses
were performed using the two procedures for the same matrix.
The statistical comparison was done using the paired Student-t test
and the Snedecor F-test. The results can be seen in Table 3.
The results shown in Table 3 indicate that for the sunflower and
for the bovine fat biodiesels analyzed, the equivalence between the
two methods in terms of precision and accuracy is at the 95% confidence
level. In the case of the soy biodiesel, equivalence of the
precision at the 95% confidence level was observed for the four
analytes. As for accuracy with the soy biodiesel, potassium agreed
Table 2
Results obtained with the recovery experiments of added analyte using ion chromatography with sunflower biodiesel as matrix. Extraction was performed with heating at
85 ◦ C and 15 min of sonication.
Cation Initial concentration a (mg kg −1 ) Added (mg kg −1 ) Found a (mg kg −1 ) Recovery (%)
Na + 3.56 ± 0.05 1.148
5.200
K +

120 L.B. Caland et al. / Analytica Chimica Acta 718 (2012) 116– 120
Table 4
Analysis of various biodiesels using the proposed method and the analytical equations reported in Table 1.
Na + (mg kg −1 ) RSD (%) K + (mg kg −1 ) RSD (%) Mg 2+ (mg kg −1 ) RSD (%) Ca 2+ (mg kg −1 ) RSD (%)
Sunflower 3.56 ± 0.05 1.7 0.91 ± 0.13 14.7 0.13 ± 0.07 5.3 2.40 ± 0.08 3.3
Corn 0.99 ± 0.06 6.1 0.36 ± 0.06 16.6 0.06 ± 0.02 33.3 0.43 ± 0.08 18.6
Soy 1.86 ± 0.25 13.4 0.35 ± 0.02 5.7 0.22 ± 0.05 22.7 1.13 ± 0.01 0.9
Soy (crude) 0.92 ± 0.09 10.0 0.39 ± 0.03 7.8