Food Chemistry 124 (2011) 1238–1243

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Microwave assisted extraction of anthocyanins from grape skins

A. Liazid a, R.F. Guerrero b, E. Cantos b, M. Palma a,*, C.G. Barroso a
a
Department of Analytical Chemistry, Faculty Sciences, University of Cádiz, P.O. Box 40, 11510 Puerto Real, Cádiz, Spain
b
Rancho de la Merced. IFAPA, Junta de Andalucía, Spain

a r t i c l e i n f o a b s t r a c t

Article history: A new method for the analysis of anthocyanins in grapes based on a systematic study of the extractability
Received 4 March 2009 of eleven anthocyanins from grapes has been developed. Microwave assisted extraction was applied as a
Received in revised form 17 February 2010 prior stage to the chromatographic determination of anthocyanins in the extracts. The stability of antho-
Accepted 18 July 2010
cyanins under the extraction conditions was checked using a standardised extract from grape skins.
Temperatures from 50 °C up to 150 °C were evaluated. A fractional factorial experimental design was
developed to analyse the influence on the extraction process of six different extraction variables: solvent
Keywords:
(mixtures of methanol and water), stirring, extraction temperature, extraction time, microwave power
Anthocyanins
Grapes
and extraction volume. The extraction solvent was the most important variable for the recovery of most
Microwave assisted extraction anthocyanins from grapes. Finally, the influence of the extraction time was also studied. With this new
method, anthocyanins can be extracted from grapes in 5 min, using 100 °C as extraction temperature
and 40% methanol in water as the extraction solvent. Repeatability and reproducibility were also
checked, the resulting RSDs (n = 9) were lower than 7% for glucosides, the main components, and lower
than 9% for the acyl derivatives, the compounds found in the lowest concentrations.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction anins of the remaining must. However, such methods do not allow
for the determination of the concentration of anthocyanins in the
Anthocyanins are the compounds responsible for the colours in grape, but rather in the must, and these levels depend directly on
several different fruits (Mazza & Miniati, 1993). In grapes, they are the process used for processing the grapes, which may produce
found almost exclusively in the skins, with only a limited number great variation.
of varieties showing these compounds in the pomace. They are of The methods developed for analytical determinations in grapes
particular importance in grapes, given that following fermentation have focussed on maceration of the skins with different solvents or
of the must, these compounds are responsible for red colour in red solvent mixtures. The extraction temperature has been proved as
wines (Cheynier, Moutounet, & Sarni-Machado, 2000). For this rea- the main extraction variable affecting the recovery of anthocyanins
son, and also due to the variations which occur in them as the using those methods (Vatai et al., 2008).
grape is ripening, their development in the grape must be charac- Maceration without applying techniques to speed up the
terised in order to determine the most suitable date for the harvest. extraction involves long extraction times which are not compatible
Anthocyanins are not the only compounds to be analysed for deter- with the analytical needs in the periods leading up to the grape
mining the best harvest time; however for red grapes they have a harvests. For this reason, the introduction of extraction techniques
special significance. As it is a solid sample, the analysis process which speed up the process has been evaluated, such as the use of
usually begins with extraction of the anthocyanins using organic solvents at different pH values (Morais et al., 2002). The influence
solvents either with or without water, and at different pH values of the pH is mainly due to the acid/base equilibrium of these com-
(Revilla, Ryan, & Martin-Ortega, 1998). pounds (Ribereau-Gayon, Glories, Maujean, & Dubourdieu, 2004)
As the grape is ripening, the speed of determination is of vital and directly affects their structure and therefore their extractabil-
importance, given that there may be great variation in the presence ity. However, it has been proven that the effect of changes in the
of anthocyanins due to the influence of climatic conditions pH can be compensated by the use of appropriate mixtures of sol-
(Downey, Dokoozlian, & Krstic, 2006). Different methods have been vents (Revilla et al., 1998). Methods based on more advanced
developed for extracting anthocyanins from grapes. The first meth- extraction techniques have also been developed, such as pressur-
ods were focussed on determining the composition of the anthocy- ised liquid extraction (Ju & Howard, 2003), sub- and supercritical
fluid extraction (Hasbay-Adil, Cetin, Yener, & Bayındırlı, 2007;
Mantell, Rodríguez, & De la Ossa, 2003) or the sea sand disruption
* Corresponding author. Tel.: +34 956 016360; fax: +34 956 016460.
E-mail address: [email protected] (M. Palma).
method (Manhita, Teixeira, & da Costa, 2006). For methods using

0308-8146/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2010.07.053
 A. Liazid et al. / Food Chemistry 124 (2011) 1238–1243 1239

supercritical fluids (pure CO2 or CO2 with an organic modifier) MA, USA). The malvidin-3-glucoside was obtained from Polyphe-
extraction pressure was the main variable affecting the recovery nols A.S. (Sandnes, Norway).
of anthocyanin, as it mainly determines the density of the extrac-
tion fluid (Ju et al., 2003; Mantell et al., 2003). For methods based 2.2. Samples
on pressurised liquid extraction, the main extraction variable was
the extraction temperature, as several anthocyanins were de- The grape variety used to develop the MAE method was Tintilla
graded at temperatures higher than 60 °C using that technique de Rota. The skins were separated manually from the rest of the
(Ribereau-Gayon et al., 2004). Other extraction techniques have fruit. All the material was ground with a conventional mixer until
been checked, however most of the resulting methods were fo- a homogenous sample had been obtained for the analysis. The
cussed on the recovery of anthocyanins and on the kinetics of ground sample obtained was kept in the freezer at 20 °C until
the process instead of on the evaluation as a routine analysis meth- its analysis, so that it would be unchanged throughout the
od (Luque-Rodríguez, Luque de Castro, & Pérez-Juan, 2007; Maier, procedure.
Göppert, Kammerer, Schieber, & Carle, 2008).
This study has considered the possibility of using microwave as-
sisted extraction (MAE) to develop a method for determination of 2.3. Extraction process
anthocyanins in grapes. MAE equipments allow for the simulta-
neous treatment of various samples, therefore it could become a The extraction of grape derived phenolic compounds by means
useful tool in monitoring the concentration of anthocyanins during of microwaves was carried out under different extraction condi-
the ripening of the grapes with a reduced time per sample. tions according to the experimental design shown in Table 1. Each
Working with related compounds, a method based on MAE has experiment in the experimental design was run in duplicate. The
been previously developed for determination of the phenols in variables studied were: solvent (50–80% MeOH in water), magnetic
grape seeds (Hong, Yaylayan, Raghaven, Pare, & Belanger, 2001). stirring, temperature (50–100 °C), time (5–20 min), microwave
It was found that the most significant extraction variable was the power (100–500 W), solvent volume (25–50 mL).
solvent used, which was more significant than the microwave The extraction using microwaves was carried out in an Ethos
power applied and even the extraction time. Such a result is usual 1600 microwave extractor (Milestone, Shelton, CT, USA).
for this type of method, where the microwave transmission/ The final extraction method was as follows: using 2 g of ground
absorption properties and the extraction properties of the solvents grape skins the extraction variables has to be fixed at 40% metha-
used must work together. nol in water as extraction solvent, 100 °C as extraction tempera-
When developing methods based on MAE, in addition to the ture, 500 W as system power and 25 mL as extraction volume.
influence of the solvent, the other variables to be evaluated, apart Moreover magnetic stirring has to be used during the extraction.
from time, are usually microwave power, extraction temperature, The extraction time was 5 min.
sample quantity and the solvent volume (Carro, Garcia, & Cela, Minitab v10.0 (State College, PA, USA) software was used for the
1997). development and evaluation of the results of the experimental de-
In this study, a systematic study was carried out based on a frac- sign. A fractional factorial design (26–2) was used, carrying out a to-
tional factorial experimental design for determining the influence tal of 16 extractions in duplicate instead of the 64 possible
of these variables, as well as their interactions, on the extraction combinations of the six extraction variables evaluated. This kind
of anthocyanins from grape skins. The method was later optimised of experimental design has produced good results in the extraction
by evaluating the influence of the most significant variables. methods development previously for different compounds (Liazid,
Barbero, Palma, Brigui, & Barroso, 2007; Palma & Taylor, 1999; Pal-
ma, Taylor, Zoecklein, & Douglas, 2000). Table 1 shows the extrac-
2. Methods and materials tion conditions assayed and the relative recoveries according to the
type of anthocyanin (glycoside or acyl derivatives). Graphic analy-
2.1. Products and reagents sis of the principal effects and the interactions between the vari-
ables was used for interpretation of the results.
The methanol used (Panreac, Barcelona, Spain) was HPLC grade. An already well established solid–liquid method of extraction
The water was supplied by a Milli-Q system, Millipore (Bedford, was used as a reference. Analyses were carried out according to

Table 1
Conditions and results of the extractions based on the fractional factorial experimental design.

Experiment Extraction variables Relative recovery*

Solvent (%MeOH in water) Stirring Temp (°C) Time (min) Power (w) Volume (mL) Glucosides Acyl derivatives
1 80 No 100 5 100 50 32 22
2 80 Yes 50 20 100 25 70 85
3 80 Yes 100 20 500 50 86 64
4 50 Yes 100 20 100 50 83 49
5 50 No 50 20 100 50 100 100
6 80 No 50 20 500 50 60 68
7 80 Yes 100 5 500 25 77 79
8 50 No 100 20 500 25 93 66
9 80 Yes 50 5 100 50 74 74
10 50 Yes 100 5 100 25 79 85
11 50 No 100 5 500 50 75 80
12 80 No 100 20 100 25 84 82
13 50 Yes 50 5 100 25 72 68
14 50 Yes 50 20 500 25 76 67
15 80 No 50 5 500 25 65 69
16 50 No 50 5 500 50 78 89
*
All the recoveries are shown relative to the amount found using the most effective conditions: experiment No. 5 (100%).
1240 A. Liazid et al. / Food Chemistry 124 (2011) 1238–1243

the protocol of Cantos et al. (Cantos, Espín, & Tomás-Barberán, compounds detected, comparing them with the extracts which
2002) with some modifications. Briefly, grapes were peeled with showed the highest concentrations of each of the determined
the help of a sharp knife, and the skins were homogenised in an anthocyanins.
Ultraturrax T-25 (Janke and Kunkel, Ika-Labortechnick) at As can be seen in Fig. 1, the temperature which produced the
24,000 rpm. for 1 min after the addition of 2 mL of a solution best recovery of anthocyanins from grape skins was 100 °C, with
MeOH/formic acid (95:5) per gram of grape skin. After this, the ex- recoveries found to be lower at temperatures above 100 °C. This
tracts were subjected to extraction for 5 h in dark, cold conditions. indicates that degradation must occur in compounds of this type
The extract was centrifuged at 5000g for 5 min in a Centromix cen- when those temperatures are used. Therefore, 100 °C is the maxi-
trifuge (Selecta, Barcelona), filtered through a 0.22 lm membrane mum temperature to be used in developing the microwave as-
filter Millex-HV13 (Millipore Corporation, USA) and analysed by sisted extraction method.
HPLC. It can also be observed that there is no clear difference between
glucosides (compounds 1–5) and the acyl derivatives (compounds
6–11). Rather, they all showed similar characteristics: a sudden de-
2.4. Liquid chromatography system
crease in recovery, of approximately 50–60% in all cases, on
increasing the temperature from 100 to 125 °C.
The samples (20 lL) were analysed using a Waters (Milford,
MA, USA) HPLC system with a pump model 1525 and a diode array
detector Waters 996. Separations were achieved on a Teknokroma 3.2. Development of the method
Mediterranean sea column (Tecknokroma, Barcelona, Spain) (RP-
18, 25  0.46 cm; 5 lm particle size) and a guard column of the After setting the maximum extraction temperature the next
same material, at 30 °C. The mobile phase consisted of water with step was to study the influence on the recovery of the factors
5% formic acid (solvent A) and HPLC grade methanol (solvent B) at which can influence the process.
a flow rate of 1 mL/min. The elution program involved gradient To do this, a fractional factorial experimental design was used,
elution from 20% B to reach 30% at 10 min, 35% B at 30 min, 40% with a total of 16 extractions instead of full factorial, which would
B at 45 min, 50% B at 60 min, 60%B at 70 min, and 95% B at require 64 (26) extractions.
75 min. Quantification was done at 520 nm. All compounds were The factors to be considered were the extraction variables, i.e.
quantified using the calibration curve obtained using the malvi- the percentage of methanol in water (50–80%), the presence or
din-3-glucoside (LOD = 0.074 ppm, LOQ = 0.240 ppm). The HPLC not of stirring during the extraction process, the temperature (con-
analyses were run in triplicate. sidering 100 °C as the maximum value), the duration of the extrac-
HPLC–MS analyses were done for the identification of chro- tion (5–20 min), the power supplied (100–500 W) and the volume
matographic peaks. The results have been previously discussed of solvent used (25–50 mL).
(Guerrero et al., 2009). Table 1 shows the values of each of these variables in the exper-
iments carried out, and the relative recovery of each of the com-
pounds analysed. All the concentrations are shown relative to the
3. Results and discussion amount found using the most effective conditions (100%). This
means that recovery was not calculated relative to the total
3.1. Determination of stability amount of anthocyanins present in the samples, but relative to
the highest concentration found in the extracts. Due to the differ-
A study was carried out to determine the stability of the antho- ence in polarity between the glucoside type anthocyanins and the
cyanins at different working temperatures in the microwave sys- acyl derivatives of these anthocyanins, the mean recoveries of each
tem. To determine stabilities of anthocyanin type compounds, of these types of compounds are shown separately.
their recoveries were determined in the samples of grape skins of For a better interpretation of the results, graphic analysis of the
the Tintilla de Rota variety in a 20 min extraction in the microwave same was carried out, which was also undertaken independently
system. To do this, approximately 2 g of sample were subjected to for the glucosides and for the acyl derivatives. The results are
an extraction with 25 mL of methanol:water (50/50) at different shown in Fig. 2.
temperatures between 50 and 150 °C, using 500 W system power. In the case of the glucoside type compounds (Fig. 2a), it can be
A comparison was made of the chromatographic areas of the 11 seen that the variables which have the most intense effect on

Fig. 1. Recovery of the 11 anthocyanins quantified in the extracts at different temperatures. Identification of anthocyanins: 1, delphinidin3-glucoside; 2, cyanidin 3-
glucoside; 3, petunidin 3-glucoside; 4, peonidin3-glucoside; 5, malvidin3-glucoside; 6, peonidin 3-acetylglucoside; 7, malvidin 3-acetylglucoside; 8, malvidin 3-p-
coumaroylglucoside (cis); 9, malvidin 3-caffeoylglucoside; 10, petunidin 3-p-coumaroylglucoside; 11, malvidin 3-p-coumaroylglucoside.
 A. Liazid et al. / Food Chemistry 124 (2011) 1238–1243 1241

recovery are the composition of the extraction solvent and the It was observed that the effect of a decrease in recovery when
extraction time. increasing the working temperature only occurs in the cases where
In the case of the acyl derivatives (Fig. 2b), the composition of the volume of extraction was 50 mL, whilst when using a volume
the solvent is still equally important with regard to recovery. How- of 25 mL this effect does not occur, rather the opposite occurs
ever, the other significant variable is the temperature of the extrac- and the recovery is greater. The stability of anthocyanins was
tion, not the extraction time. Given that the influence is negative, checked using 25 mL of solvent, when using 50 mL additional
i.e. the higher the temperature, the lower the recovery, it was microwaves should be applied to reach the working temperature,
decided to study the interaction effects between the temperature i.e. the sample received a higher amount of energy, so using
and the other variables in an attempt to explain this fact. 50 mL degradation could be produced when using the highest tem-
The most relevant interaction for temperature was that found perature checked (100 °C). Degradation in MAE mainly depends on
with the extraction volume. the extraction temperature, but also it depends on the energy ap-

Fig. 2. (a) Principal effects on recovery for glucosides. Stirring conditions: 1 = No, +1 = Yes. (b) Principal effects on recovery for acyl derivatives. Stirring conditions: 1 = No,
+1 = Yes.
1242 A. Liazid et al. / Food Chemistry 124 (2011) 1238–1243

Fig. 3. Recovery of the 11 anthocyanins quantified in the extracts using different percentages of methanol in the extraction solvent.

plied to the sample. Additional experiments were done using 50, 15 min and it was therefore decided to use 5 min as the extraction
75 and 100 °C with 50 mL of solvent. It was found that a 12% lower time.
recovery was found using 100 °C than using 75 °C, whilst no signif-
icant differences were found for 50 and 75 °C. Therefore, 25 mL 3.4. Analytical characteristics of the method
was selected as the extraction volume.
In view of the significant differences in the influences of the dif- The final extraction conditions were as follows: the presence of
ferent variables, it was decided to optimise the most significant stirring, 100 °C as extraction temperature, 500 W as system power,
variable, this being the composition of the solvent, leaving the 25 mL as extraction volume, 40% methanol in water as extraction
other variables fixed at the most favourable conditions based on solvent and 5 min as extraction time. Using these extraction condi-
the principal effects shown in Fig. 2. These conditions are: the pres- tions, the repeatability and the reproducibility of the method
ence of stirring, 100 °C as extraction temperature, 500 W as system developed were determined by carrying out a total of 15 extrac-
power and 25 mL as extraction volume. As regards the extraction tions, nine on the first day and a further three on each of the fol-
time, it is clear in both cases that this variable favours the increase lowing two days. The results as regards the relative standard
of the extraction of acyl derivatives and in particular, glucosides. In deviation of the areas of the chromatographic peaks for intraday
any case, this variable must be optimised at a later point in order to analyses were found between 0.9% (Peonidin 3-glucoside) and
guarantee quantitative recoveries of the compounds. 6.8% (Cyanidin 3-glucoside). For interday analyses the relative
standard deviation were found ranging from 3.0% (Delphinidin 3-
3.3. Optimisation of the composition of the extraction solvents and the glucoside) and 8.5% (Malvidin 3-coumaroylglucoside).
extraction time It can be observed that the best variation found for the gluco-
sides is below 7%, whilst for the acyl derivatives it is below 9%.
Assays were carried out using mixtures of methanol in water The greatest variation found for the acyl derivatives could be
between 50% and 20%. The results of these extractions are shown attributed generally to the smaller chromatographic area of these
in Fig. 3, where it can be seen that the conditions which produce compounds.
the best recovery for all the compounds, are those which use 40%
methanol in water. It must be noted here that, low differences 3.5. Comparison of the method with the classical method
were found for glucosides, whereas for the acyl derivatives there
are appreciable differences between the recoveries found with Table 2 shows the results of the analysis of samples of the Tin-
the different solvents. On the basis of these results, the percentage tilla de Rota grape variety carried out using the solid–liquid macer-
of methanol in water was fixed at 40%. ation classical method and the developed MAE method. For these
As regards the determination of the time necessary for extrac- analyses the same sample preparation protocol prior to the extrac-
tion, the recovery of the eleven compounds was studied using tion process was used (Section 2.3).
times between 2.5 and 30 min. The results are shown in Fig. 4. It It can be observed that the results are of the same order of mag-
can be observed that there is a decrease in recovery when using nitude, with statistically significant differences being found for a
extraction times of 30 min, compared to the values obtained using total of six compounds, three of these having lower recoveries in
15 min. There were no significant differences found between 5 and the case of the MAE method and three of them having higher

Fig. 4. Recovery of the 11 anthocyanins quantified in the extracts at different extraction time.
 A. Liazid et al. / Food Chemistry 124 (2011) 1238–1243 1243

Table 2
Quantities of anthocyanins found using the developed method (MAE) and the reference method expressed as mg of malvidin/
kg FW.

Solid–liquid maceration MAE (mg/kg) Relative recovery

(mg/kg)
Delphinidin 3-glucoside 63.5 ± 0.3 67.1 ± 0.4 105.7
Cyanidin 3-glucoside 35.9 ± 0.3 27.7 ± 0.4 77.4*
Petunidin 3-glucoside 73.6 ± 0.3 90.2 ± 2.4 122.6*
Peonidin 3-glucoside 371.0 ± 5.3 325.1 ± 1.72 87.6
Malvidin 3-glucoside 680.4 ± 14.3 953.0 ± 24.0 140.1*
Peonidin 3-acetylglucoside 43.6 ± 0.4 36.2 ± 0.6 83.0*
Malvidin 3-acetylglucoside 86.9 ± 0.9 116.0 ± 4.6 133.5*
Malvidin 3-coumaroylglucoside (cis) Nq 19.8 ± 0.6
Malvidin 3-caffeoylglucoside Nq 51.3 ± 3.5
Petunidin 3-p-coumaroylglucoside Nq 17.5 ± 0.5
Malvidin 3-p-coumaroylglucoside 191.0 ± 1.4 154.0 ± 7.4 80.6*
Total anthocyanins 1545.9 1857.9 117.6
*
Significant differences (p < 0.05).

recoveries. It must also be noted that with the new method it is References
possibly to quantify in the samples the compounds malvidin 3-
coumaroylglucoside (cis), malvidin 3-caffeoylglucoside, petunidin Cantos, E., Espín, J. C., & Tomás-Barberán, F. A. (2002). Journal of Agricultural and Food
Chemistry, 50, 5691.
3-p-coumaroylglucoside, which do not reach the quantification Carro, N., Garcia, C. M., & Cela, R. (1997). Analyst, 122, 325.
limit using the classical method. Cheynier, V., Moutounet, M., & Sarni-Machado, P. (2000). In C. Flanzy (Ed.), Enologia
fundamentos científicos y tecnológicos (pp. 115–124). Madrid: Mundi-Prensa.
Downey, M. O., Dokoozlian, N. K., & Krstic, M. P. (2006). American Journal of Enology
4. Conclusions and Viticulture, 57, 257.
Guerrero, R. F., Liazid, A., Palma, M., Puertas, B., González, R., Gil, A., et al. (2009).
The variable having the most significant influence on the Food Chemistry, 112, 949.
Hasbay-Adil, I., Cetin, H. I., Yener, M. E., & Bayındırlı, A. (2007). Journal of
extraction of anthocyanins from grapes by means of microwave as- Supercritical Fluids, 43, 55.
sisted extraction was the extraction solvent, the other variables Hong, N., Yaylayan, V. A., Raghaven, G. S., Pare, J. R., & Belanger, J. M. (2001). Natural
being less significant. With the developed method, similar recover- Products Letters, 15, 197.
Ju, Z. Y., & Howard, L. R. (2003). Journal of Agricultural and Food Chemistry, 51, 5207.
ies were obtained to those found with the maceration extraction Liazid, A., Barbero, G. F., Palma, M., Brigui, J., & Barroso, C. G. (2007).
method, with a notable reduction in the extraction time from 5 h Chromatographia, 66, 571.
to 5 min. Using the new method three additional acyl derivatives Luque-Rodríguez, J. M., Luque de Castro, M. D., & Pérez-Juan, P. (2007). Bioresource
Technology, 98, 2705.
can be determined in the samples analysed which are not recov- Maier, T., Göppert, A., Kammerer, D. R., Schieber, A., & Carle, R. (2008). European
ered in significant levels with the classical method to allow them Food Research and Technology, 227, 267.
to be quantified. Manhita, A. C., Teixeira, D. M., & da Costa, C. T. (2006). Journal of Chromatography. A,
1129, 14.
The extraction system using microwaves also allows treatment
Mantell, C., Rodríguez, M., & De la Ossa, E. M. (2003). Separation Science and
of up to 10 samples to be carried out simultaneously, thereby sig- Technology, 38, 3689.
nificantly reducing the analysis time per sample. The reproducibil- Mazza, G., & Miniati, E. (1993). Anthocyanins in fruits, vegetables and grains. Florida:
CRC press Inc..
ity and repeatability values for the method were adequate for
Morais, H., Ramos, C., Forgacs, E., Cserhati, T., Matos, N., Almeida, V., et al. (2002).
the initial target values, they were RSD < 8% and RSD < 9%, Journal of Chromatographia, 57, S173.
respectively. Palma, M., & Taylor, L. T. (1999). Analytica Chemica Acta, 391, 321.
Palma, M., Taylor, L. T., Zoecklein, B. W., & Douglas, L. S. (2000). Journal of
Agricultural and Food Chemistry, 48, 775.
Acknowledgments Revilla, E., Ryan, J.-M., & Martin-Ortega, G. (1998). Journal of Agricultural and Food
Chemistry, 46, 4592.
This work was carried out with the funding received for the pro- Ribereau-Gayon, P., Glories, Y., Maujean, A., & Dubourdieu, D. (2004). Handbook of
enology, vol. 2.. Chichester, UK: John Wiley and Sons Ltd..
ject FQM-01282/2005 funded by the Consejería de Innovación, Vatai, T., Skerget, M., Kneza, Z., Kareth, S., Wehowski, M., & Weidner, E. (2008).
Ciencia y Empresa de la Junta de Andalucía and the project Journal of Supercritical Fluids, 45, 32.
RTA2008-00014 funded by INIA.