Appl Biochem Biotechnol (2010) 160:2066–2074

DOI 10.1007/s12010-009-8707-8

Structure and Applications of a Rhamnolipid Surfactant

Produced in Soybean Oil Waste

Marcia Nitschke & Siddhartha G. V. A. O. Costa &

Jonas Contiero

Received: 9 March 2009 / Accepted: 5 July 2009 /

Published online: 4 August 2009
# Humana Press 2009

Abstract Soybean oil soapstock was utilized as an alternative carbon source for the
production of rhamnolipids by Pseudomonas aeruginosa LBI strain. The chemical
composition and properties of the rhamnolipid mixture obtained were determined to define
its potential applications. The chemical characterization of the rhamnolipid has revealed the
presence of ten different homologues. The monorhamnolipid RhaC10C10 and the
dirhamnolipid Rha2C10C10 were the main components of the mixture that showed
predominance of 44% and 29%, respectively, after 144-h of cultivation. The biosurfactant
was able to form stable emulsions with several hydrocarbons and showed excellent
emulsification for soybean oil and chicken fat (100%). The rhamnolipid removed 67% of
crude oil present in sand samples and presented antimicrobial activity against Bacillus
cereus and Mucor miehei at 64 μg/mL and inhibition of Neurospora crassa, Staphylococ-
cus aureus, and Micrococcus luteus at 256 μg/mL. The results demonstrated that the
rhamnolipid produced in soybean oil soapstock can be useful in environmental and food
industry applications.

Keywords Biosurfactants . Rhamnolipids . Soybean oil waste . Soapstock .

Pseudomonas aeruginosa

Introduction

Biosurfactants are defined as a class of surface-active molecules synthesized by

microorganisms. Some advantages of biosurfactants over synthetic surfactants such as

M. Nitschke (*)
Department of Physical-Chemistry, Institute of Chemistry, University of São Paulo—USP,
Av. Trabalhador São Carlense, 400, 13560-970 São Carlos, SP, Brazil
e-mail: nitschke@iqsc.usp.br

S. G. V. A. O. Costa : J. Contiero
Department of Biochemistry and Microbiology, Institute of Biological Sciences, São Paulo State
University—UNESP/Rio Claro, Av. 24-A, 1515—Bela Vista,
13506-900 P. O. Box 199, Rio Claro, SP, Brazil
Appl Biochem Biotechnol (2010) 160:2066–2074 2067

mild production condition, lower toxicity, higher biodegradability, and environmental

compatibility have prompted their applications in environmental protection as well as in
food, cosmetic, and pharmaceutical industries [1–3]. However, the production of
biosurfactants can be 50 times more expensive than the production of chemically derived
surfactants [4]; thus, the economics is the main factor against their widespread use.
It is well known that the disposal of wastes is a growing problem, and new
alternatives for their use should be studied as the treatment and disposal costs for
these wastes are a vast financial burden to various industries [1]. The use of alternative
low-cost substrates is an important strategy to facilitate industrial development of
biosurfactant production. To this end, good components seem to be agroindustrial by-
products or wastes, once these residues generally contain high levels of carbohydrates or
lipids to support growth and surfactant synthesis [5]. Olive oil mill effluent [6], oil
refinery wastes [7], waste fry oil [8], distillery and whey wastes [9], potato process
effluent [10], and cassava wastewater [11] are some examples on the application industrial
wastes as feedstock for biosurfactant production.
The world production of vegetable oils was about 134 million tons in 2008 and
soybean oil production represented about 37 million tons. Brazil produced 6 million
tons of soybean oil in 2008 and is considered one of the largest soybean and soybean
oil producer (USDA annual report). Great quantities of waste are generated by the oil
and fat industries: residual oils, tallow, soapstock, and frying oils. The production of
soapstock, one of the wastes of the oil neutralization process, corresponds to 2–3% of
the total oil production [12].
The rhamnolipids from Pseudomonas aeruginosa, which are among the most effective
biosurfactants, are essentially constituted by a mixture of the homologues species
Rha2C10C10 and RhaC10C10 [13]. The surfactant properties of rhamnolipids depend on its
composition and distribution of homologues that vary according to the bacterial strain,
culture conditions, and medium composition [14]. P. aeruginosa LBI strain was isolated
from a soil contaminated with hydrocarbons [15]. In previous work, we reported the
production of rhamnolipid surfactant by this strain using different edible oil refinery wastes
as carbon sources and soybean soapstock was the most promising substrate [16]. However,
to improve commercial exploitation of the process, it is important to know the properties
and potential applications of the product. The present work investigates the structure and
applications of the rhamnolipid surfactant obtained when soybean oil soapstock is used as
an alternative substrate.

Materials and Methods

Microorganism and Culture Conditions P. aeruginosa LBI bacterial suspension, obtained

from a nutrient agar slant incubated for 24 h at 30 °C, was adjusted to DO610 0.65
(approx. 108 cfu/mL), and 1 mL of this culture was inoculated on a 250-mL Erlenmeyer
flask containing 50 mL of mineral salts medium [17] and 2% (w/v) of soybean oil as a
carbon source. The inoculum was incubated for 24 h, 30 °C, 200 rpm on a rotary shaker
(New Brunswick, USA), and an aliquot of 1 mL was added to the production medium.
Biosurfactant production was performed on 125-mL flasks containing 25 mL of salts
medium added of 2% (w/v total lipid) of soybean oil soapstock and incubated as
described above for 144 h. Initial pH of broth was adjusted to 6.8–6.9. Experiments were
conducted in three independent replicates. Error bars (when shown) represent standard
deviation.
2068 Appl Biochem Biotechnol (2010) 160:2066–2074

Rhamnolipid Quantification Rhamnolipids were quantified from the cell-free broth as

rhamnose [18]. Rhamnolipid content was determined by multiplying rhamnose values by 3
[19].

Biomass Estimation Cell growth was estimated by the protein content of the culture using
Lowry method [20].

Nitrate Concentration Nitrate concentration was estimated using the spectrophotometric

UV method [21].

Surface Activity Measurement Culture samples were centrifuged at 8,000 g for 20 min to
cell removal, and the supernatant was submitted to surface activity measurements. Surface
tension was determined with a Krüss Tensiometer (Krüss, Germany) using the ring method.
The critical micelle dilution (CMD−2) was measured as the surface tension of 100 times
diluted broth in distilled water.

Rhamnolipid Extraction Cells were removed from the culture broth by centrifugation at
10,000×g for 20 min. The biosurfactant was isolated from cell-free broth by acid precipitation
after adjusting the broth pH to 2.0 using 6 N H2SO4 and keeping it at 4 °C overnight. The
precipitate thus obtained was pelleted at 8,000×g for 20 min, redissolved in distilled water,
adjusted to pH 6.1, and applied to an adsorption chromatography column filled with
polystyrene resin, Amberlite XAD2 (Supelco, USA) previously equilibrated with 0.1 M
potassium phosphate buffer (pH 6.1). The adsorption of the active compound was assayed by
measuring the surface tension at the column outlet. Adsorption chromatography was terminated
when the surface tension dropped below 35 mN/m [22]. The column was then washed with
distilled water until a surface tension around 72 mN/m was attained. Rhamnolipids were eluted
with methanol, and the solvent was evaporated to dryness under vacuum.

Structural Characterization of Rhamnolipid The analyses were performed with a triple

quadrupole mass spectrometer Quattro II (Micromass, UK) equipped with a Z-spray
interface using electrospray ionization in negative mode. The instrument was interfaced to a
HP 1100 HPLC (Agilent Technologies, USA) equipped with a 150×4 mm Zorbax C8
reverse phase column [23].

Emulsifying Activity Hydrocarbon source (6 mL) was added to 4 mL of surfactant solution

(0.1% w/v) and vortexed at high speed for 2 min. After 24 h, the emulsification index (E24)
was calculated by dividing the measured height of emulsion layer by the mixture’s total
height and multiplying by 100.

Crude Oil Removal A biosurfactant solution (0.1% w/v) was utilized to wash sand samples
contaminated with crude oil. The sand was added of 10% (w/w) of low viscosity crude oil
and maintained at room temperature for 3 days. Then, 5 g of sand (50–80 mesh) samples
were washed with rhamnolipid solution for 18 h at 200 rpm 30 °C. Aqueous solution was
removed, and the sand was dried at 50 °C for 24 h. The sand sample was washed two times
with dichloromethane (Tedia, USA), and the solvent was evaporated at 50 °C. The
remaining oil was determined gravimetrically, and the percentage of oil removal was
calculated using the following equation: Crudeoilremoved ð%Þ ¼ ðOi
OrÞ=Oi  100%,
where Oi is the initial oil in the soil (grams) before washing, and Or is the oil remaining in
the soil (grams) after washing [24].
Appl Biochem Biotechnol (2010) 160:2066–2074 2069

Antimicrobial Activity Antimicrobial activity of biosurfactant against bacteria, yeast, and

fungi was determined by minimal inhibitory concentration (MIC) using the microbroth
dilution method [25]. The bacterial strains were grown on Muller Hinton agar at 37 °C and
the yeasts and fungi on Sabouraud Dextrose agar at 30 °C.

Results and Discussion

The biosurfactant was obtained after growth of P. aeruginosa LBI using soybean oil
soapstock as the carbon source. The kinetics of rhamnolipid production using the
alternative substrate is shown in Fig. 1. The nitrate was consumed during first 24 h of
cultivation generating a biomass of 1.5 g/L. The bulk of rhamnolipid synthesis occurred
after cell reached the stationary phase as a typical secondary metabolite. After 144 h, the
rhamnolipid concentration was 11.7 g/L, and the conversion yield of carbon substrate to
biosurfactant was around 75%. Similar results were obtained by Benincasa et al. [15] using
sunflower oil soapstock as the carbon source. Bednarski et al. [7] also showed that the
production of glycolipid surfactants from Candida antarctica and Candida apicola was
increased from 7.3 to 13.4 g/L when soapstock was supplemented to culture medium.
The chemical characterization of the rhamnolipid revealed the presence of ten different
homologues: RhaC10C8, RhaC10C10, RhaC10C12, RhaC12C12, Rha2C10C8, Rha2C10C10,
RhaC8C10:1, RhaC10C10:1, Rha2C10C10:1, and Rha2C10C 12:1. The product extracted from
different cultivation times was analyzed, and the predominance of the homologues
RhaC10C10 and Rha2C10C10 as the main components of the mixture was found (Table 1).
The concentration of both homologues varied during time; however, the monorhamnolipid

Fig. 1 Time course of rhamnolipid production by P. aeruginosa LBI using soybean oil soapstock. Error
bars represent standard deviation
2070 Appl Biochem Biotechnol (2010) 160:2066–2074

Table 1 Distribution of homologues (percent) present on rhamnolipid mixture produced in soybean oil
soapstock at different cultivation times by P. aeruginosa LBI.

Rhamnolipid homologue Cultivation time (h)

24 48 72 96 120 144

Concentration (%)
RhaC10C8 1.56 1.9 2.06 4.44 4.25 4.41
RhaC10C10 42.18 59.02 66.86 38.44 40.85 43.97
RhaC10C12 ND ND ND 4.88 3.61 3.8
RhaC12C12 9.37 9.54 3.76 2.66 ND ND
Rha2C10C8 1.56 1.56 1.21 4.44 5.53 5.01
Rha2C10C10 45.31 24.3 23.05 42.66 42.76 29.55
RhaC8C10:1 ND 0.52 0.36 ND ND ND
RhaC10C10:1 ND 0.69 0.72 ND ND ND
Rha2C10C10:1 ND ND 0.12 0.66 1.06 0.73
Rha2C10C 12:1 ND 1.56 1.82 1.77 1.91 1.76

ND not detected

(RhaC10C10) was the major component of the final product extracted after 144 h of
cultivation. The accepted pathway for rhamnolipid synthesis postulates that the mono-
rhamnolipids are the precursors of the dirhamnolipids once the inactivation of rhl genes
required for the synthesis of RhaC10C10 abolish rhamnolipid production [26]. Then, the
changes in rhamnolipid distribution may account for the stages of their biosynthesis.
The level of rhamnolipid production by P. aeruginosa UG2 was affected by culture
conditions but the distribution of homologues did not vary between cultures [27]. The
rhamnolipid mixture obtained by these authors using corn oil as substrate had 60% of the
dirhamnolipid (Rha2C10C10) form. In our case, the monorhamnolipid showed an average
predominance of 48.5% in contrast with 34.5% for the dirhamnolipid form. These results
confirm the previous observations that the monorhamnolipid form predominates in culture
when P. aeruginosa LBI was grown in soybean oil soapstock [16]. Nevertheless, Benincasa
et al. [28] showed that after 84 h of cultivation in sunflower oil soapstock, this strain
produced a rhamnolipid mixture where the dirhamnolipids isoforms predominate. These
differences can be attributed to the carbon source composition, culture conditions as well as
to the loss of some components during the purification steps. A recent report suggested that
the amount of unsaturated fatty acids in the carbon source reflects in the unsaturation of
rhamnolipid carbon chains. The authors reported that the amount of unsaturated fatty acids
in substrate was 50%, and the unsaturated carbon chains in rhamnolipid mixtures were
30.9% [12], whereas in our work the unsaturated fatty acids concentration in soybean oil
soapstock was 77% but the proportion of unsaturated rhamnolipid molecules was very low
(Table 1). Figure 2 shows the high-performance liquid chromatography of the rhamnolipid
extracted after 144 h of cultivation. The main components of the mixture showed retention
times of 10.05 and 10.49 min corresponding to RhaC10C10 and Rha2C10C10, respectively.
The mass spectrum shows the presence of the pseudomolecular ions of m/z 503 and 649,
which correspond to the [M–H]− forms of RhaC10C10 and Rha2C10C10. As can be seen, the
relative intensity also demonstrates the predominance of the monorhamnolipid homologue
(Fig. 3). The identity of the pseudomolecular ions present in rhamnolipid mixtures were
confirmed by their fragmentation profiles.
Appl Biochem Biotechnol (2010) 160:2066–2074 2071

Fig. 2 High-performance liquid chromatography of rhamnolipid isolated after 144 h of cultivation. a Peak
separation profile, b RhaC10C10 peak, c Rha2C10C10 peak

The properties of the product obtained after 144 h of cultivation using soybean oil
soapstock as carbon source were investigated. The surface tension of 0.1% (w/v)
rhamnolipid solution was 26.9 mN/m, an interfacial tension against hexadecane 1.25 mN/m
and a CMC of 51.5 mg/L [16]. The properties of rhamnolipids are dependent on bacterial
strain, medium composition, and culture conditions that determine the composition and
distribution of homologue molecules present on final product [14]. Values of CMC ranging
from 10 to 234 mg/L and surface tension from 25 to 31 mN/m have been reported to different
rhamnolipid mixtures [1]. However, as in this case, the mono- and dirhamnolipid were also
the major molecules present in biosurfactant mixtures differing primarily on their
predominance and concentration. To illustrate, the commercial rhamnolipid JBR 599 (Jeneil

Fig. 3 Mass spectrometry (ESI-MS) showing the predominance of m/z 503 (RhaC10C10) and m/z 649
(Rha2C10C10) ions
2072 Appl Biochem Biotechnol (2010) 160:2066–2074

Biosurfactant Co., USA) has a purity of 99.9% and is composed of 51% of monorhamno-
lipids (37.7% RhaC10C10) and 44.9% of dirhamnolipids (33.2% Rha2C10C10). This product
shows a CMC from 100 to 125 mg/L, a surface tension of 29.0 mN/m, and an interfacial
tension of 0.3 mN/m against paraffin oil.
The potential applications of the biosurfactant obtained in soybean soapstock were also
investigated. The ability to form and stabilize emulsions is one of the most important
features to be considered for the practical application of a surfactant. The emulsifying
activity of rhamnolipid against different hydrophobic substrates is shown in Fig. 4. The
biosurfactant was able to form emulsion with all hydrocarbons tested, which were stable up
to 30 days. The rhamnolipid product was able to emulsify efficiently the aromatic
hydrocarbons toluene and benzene as well as hexane, heptane, and kerosene showing
potential to environmental applications such as bioremediation of pollutants and oil clean
up. The excellent levels of emulsification observed for soybean oil (100%), chicken fat
(100%), and mineral oil (71%) suggest that it can be useful for food industry. The results
observed also suggest that the rhamnolipid mixture obtained from soybean oil soapstock
forms emulsions more efficiently with long-chain fatty acids and triglycerides that
predominate in oil and fats.
The capacity of aqueous rhamnolipid solution in removing crude oil from contaminated
sand was also investigated. The washing with rhamnolipid was able to remove 67% of
crude oil present in sand samples showing the potential of the product in oils spill accidents.
Urum and Pekdemir [29] showed that a 0.1% of commercial rhamnolipid solution removed
about 75% of crude oil from soil samples.
Antimicrobial activity of the rhamnolipid revealed activity against Bacillus cereus
(64 μg/mL) and Mucor miehei (64 μg/mL) and some inhibition of Neurospora crassa,
Staphylococcus aureus, and Micrococcus luteus at 256 μg/mL (Table 2). A rhamnolipid
mixture obtained from P. aeruginosa AT10 growing on soybean waste fatty acids as carbon

Fig. 4 Emulsifying activity of the rhamnolipid obtained in soybean oil soapstock

Appl Biochem Biotechnol (2010) 160:2066–2074 2073

Table 2 Antimicrobial activity of the rhamnolipid produced in soybean oil soapstock.

Microorganisms MIC (μg/mL)

Escherichia coli ATCC 11229 >500

Bacillus cereus ATCC 10876 64
Micrococcus luteus ATCC 4698 256
Pseudomonas aeruginosa ATCC 15442 >500
Rhodococcus equi ATCC 6939 >500
Salmonella cholerasius ATCC 10708 >500
Serratia marcescens ATCC 1953 >500
Staphylococcus aureus ATCC 25923 256
Candida glabrata ATCC 15126 >500
Kluyveromyces marxianus ATCC 16045 >500
Aspergillus niger ATCC 16404 >500
Mucor miehei ATCC 26282 64
Neurospora crassa ATCC 9277 256
Rhizopus microsporus ATCC 22959 >500

source also found a MIC of 64 μg/mL against B. cereus; however, MIC for S. aureus was
128 μg/mL and for M. luteus 32 μg/mL [30]. Benincasa et al. [28] reported that the
rhamnolipid mixture containing 28.9% of Rha2C10C10, 23% Rha2C10C12:1, 23.4%
RhaC10C10, 11.3% RhaC10C12, 7.9% RhaC10C12:1, and 5.5% RhaC10C12 showed a MIC
of 4 μg/mL for B. cereus, 8 μg/mL for S. aureus, 128 μg/mL for M. luteus, and excellent
levels of inhibition against fungi. The variation observed in results can be due to the
differences in homologues composition of the rhamnolipid mixtures evaluated. When
comparing the antimicrobial activity of two rhamnolipid mixtures obtained from different P.
aeruginosa strains, Haba et al. [31] observed that P. aeruginosa AT10 rhamnolipids were
most effective against fungi, and P. aeruginosa RL47T2 was more efficient against bacteria.
The effects observed were attributed to the differences in homologue composition of both
rhamnolipid mixtures. Owing to their intrinsic properties, surface-active compounds
interfere with cell surfaces and disrupt microbial membranes. There is still little information
available about antimicrobial action of rhamnolipid surfactants and the contribution of each
homologous species on the final activity of the product.
We have demonstrated that the use of soybean oil soapstock as an alternative low-cost
substrate for rhamnolipid production generates a product with potential industrial
applications and, moreover, a process that can contribute to decrease the disposal of
wastes in the environment.

Conclusions

The rhamnolipid produced by P. aeruginosa LBI using soybean oil soapstock as carbon
source was characterized in terms of chemical structure. The product obtained showed the
predominance of RhaC10C10 and Rha2C10C10 homologues at a final concentration of 44%
and 29%, respectively, after 144 h of cultivation. The results demonstrated that the
rhamnolipid can be useful in environmental and food industry applications.
2074 Appl Biochem Biotechnol (2010) 160:2066–2074

Acknowledgments This work was supported by the Fundação de Amparo a Pesquisa do Estado de São
Paulo (FAPESP). We thank Cargill S.A. for supplying the soybean oil soapstock and François Lépine and
Sylvain Milot from Institut Armand-Frappier (Canada) for their help on structural characterization of
rhamnolipids.

References

1. Nitschke, M., Costa, S. G. V. A. O., & Contiero, J. (2005). Biotechnology Progress, 21, 1593–1600.
2. Banat, I. M., Makkar, R. S., & Cameotra, S. S. (2000). Applied Microbiology and Biotechnology, 53,
495–508.
3. Singh, P., & Cameotra, S. S. (2004). Trends in Biotechnology, 22, 142–146.
4. Deleu, M., & Paquot, M. (2004). C.R Chimie, 7, 641–646.
5. Nitschke, M., & Costa, S. G. V. A. O. (2007). Trends in Food Science & Technology, 18, 252–259.
6. Mercade, M. E., & Manresa, M. A. (1994). Journal of the American Oil Chemists' Society, 71, 61–64.
7. Bednarski, W., Adamczak, M., Tomasik, J., & Plaszczyk, M. (2004). Bioresource Technology, 95, 15–18.
8. Shah, V., Jurjevic, M., & Badia, D. (2007). Biotechnology Progress, 23, 512–515.
9. Dubey, K., & Juwakar, A. (2001). W. J. Microbial Biotechnology, 17, 61–69.
10. Fox, S. L., & Bala, G. A. (2000). Bioresource Technology, 75, 235–240.
11. Nitschke, M., & Pastore, G. M. (2003). Applied Biochemistry and Biotechnology, 105–108, 295–301.
12. Benincasa, M., & Accorsini, F. R. (2008). Bioresource Technology, 99, 3843–3849.
13. Lang, S., & Wagner, F. (1987). In N. Kosaric & W. L. Cairns (Eds.), Biosurfactants and Biotechnology:
Structure and properties of biosurfactants (pp. 21–45). New York: Marcel Dekker.
14. Guerra-Santos, L., Kapelli, O., & Fiechter, A. (1984). Applied and Environmental Microbiology, 48,
302–305.
15. Benincasa, M., Contiero, J., Manresa, A., & Moraes, I. O. (2002). Journal of Food Engineering, 54,
283–288.
16. Nitschke, M., Costa, S. G. V. A. O., Haddad, R., Gonçalves, L. A. G., Eberlin, M. N., & Contiero, J.
(2005). Biotechnology Progress, 21, 1562–1566.
17. Robert, M., Mercade, M. E., Bosch, M. P., Parra, J. L., Espuny, M. J., Manresa, A., et al. (1989).
Biotechnology Letters, 11, 871–874.
18. Chandrasekaran, E. V., & Bemiller, J. N. (1980). In L. Wrhiste & M. L. Wolfrom (Eds.), Methods in
carbohydrate chemistry: constituent analysis of glycosaminoglycans (pp. 89–96). New York: Academic.
19. Itoh, S., Honda, H., Tomota, F., & Suzuki, T. (1971). Journal of Antibiotics, 24, 855–859.
20. Lowry, O. H., Rosebrought, N. J., Farr, A., & Randall, R. J. (1951). Journal of Biological Chemistry, 66,
265–274.
21. Eaton AD, Clesceri LS and Greenberg AE (1995). In: Standard methods for the examination of water
and wastewater: nitrogen (nitrate) ultraviolet spectrophotometric screening method. American Public
Health Association, Baltimore: United Book Press, Inc., pp. 4–85–4–86.
22. Reiling, H. E., Thanei-Wyss, U., Guerra-Santos, L. H., Hirt, R., Kappeli, O., & Fiechter, A. (1986).
Applied and Environmental Microbiology, 51, 985–989.
23. Deziel, E. F., Lepine, F., Dennie, D., Boismenu, D., Mamer, A. O., & Villemur, R. (1999). Biochimica et
Biophysica Acta, 1440, 244–252.
24. Urum, K., Pekdemir, T., & Çopur, M. (2004). Journal of College Interface Science, 276, 456–464.
25. Woods, G. L., & Washington, J. A. (1995). In P. R. Murray (Ed.), Manual of clinical microbiology:
antibacterial susceptibility tests: dilution and disk diffusion methods (pp. 1327–1341). Washington:
ASM.
26. Ochsner, U. A., Hembach, T., & Fiechter, A. (1995). Advances in Biochemical Engineering/
Biotechnology, 53, 89–118.
27. Mata-Sandoval, J. C., Karns, J., & Torrents, A. (2001). Microbiological Research, 155, 249–256.
28. Benincasa, M., Abalos, A., Oliveira, I., & Manresa, A. (2004). Anton Leeuw Int J G, 85, 1–8.
29. Urum, K., & Pekdemir, T. (2004). Chemosphere, 57, 1139–1150.
30. Abalos, A., Pinazo, A., Infante, M. R., Casals, M., García, F., & Manresa, A. (2001). Langmuir, 17,
1367–1371.
31. Haba, E., Abalos, A., Jauregui, O., Espuny, M. J., & Manresa, A. (2003). Journal of Surfactants and
Detergents, 6, 155–161.