Fuel 202 (2017) 196–205

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Two-step enzymatic production of environmentally friendly

biolubricants using castor oil: Enzyme selection and product
characterization
J. Greco-Duarte a, E.D. Cavalcanti-Oliveira a, J.A.C. Da Silva b, R. Fernandez-Lafuente c,⇑, D.M.G. Freire a,⇑
a
Department of Biochemistry, Institute of Chemistry, Federal University of Rio de Janeiro, Cidade Universitária, Centro de Tecnologia, Bl. A, Sl. 549, Ilha do Fundão, 21949-900 Rio
de Janeiro, Brazil
b
Cenpes, Centro de Pesquisas e Desenvolvimento Leopoldo Américo Miguez de Mello, Petrobras, 21941-970 Rio de Janeiro, Brazil
c
Departamento de Biocatalisis, ICP-CSIC, C/ Marie Curie 2, Campus UAM-CSIC, Cantoblanco, 28049 Madrid, Spain

h i g h l i g h t s

Castor oil hydrolysis using castor-seed lipases gives a high yield of unmodified COFFA.

Molecules with good lubricant characteristics are produced by the esterification of COFFA.

Estolides from COFFA esterification by CRL are the main reaction product.

CRL prefers ricinoleic acid as a nucleophile compared to polyols.

Estolide biolubricant properties are outstanding except for high acidity.

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

Article history: We performed two-step enzymatic biolubricant production: castor oil hydrolysis and esterification of
Received 20 February 2017 castor oil free fatty acids (COFFA) rich in ricinoleic acid. The hydrolysis step utilized castor seeds as cat-
Received in revised form 5 April 2017 alyst yielding 93.13 ± 5.9% COFFA in only 1 h. In the esterification step, C. rugosa lipase (CRL) decreased
Accepted 8 April 2017
the acidity by over 80% after 96 h, utilizing as polyols neopenthylglicol (NPG), trimethylolpropane
(TMP) and pentaerytritol (PE) and showed about 70% acidity decrease without polyols, indicating the pro-
duction of COFFA polymers (estolides). Nuclear magnetic resonance (NMR) confirmed estolides as the
Keywords:
main product, reaching a polymerization degree of 6/7 without polyols, 4 in the presence of PE and rici-
Biolubricant
Estolide
noleic acid dimers in the presence of NPG and TMP. Analyses of biolubricant properties showed that the
Castor oil best product was the estolide with a polymerization degree of 6/7: viscosity index of 162, oxidation sta-
Ricinoleic acid bility of 51 min and pour point of 42 °C.
Candida rugosa lipase Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction Currently used mineral lubricants derived from petroleum

The 21st century is marked by environmental degradation and
grave changes in the natural functioning of the biosphere [1]. It
is essential to develop green technologies that allow economic
growth without environmental disturbance [2]. Environmentally
friendly processes that are comparably efficient to those already
in use are being studied in order to develop technologies that allow
gradual replacement of conventional chemical processes by biolog-
ical processes [3].
⇑ Corresponding author.
E-mail addresses: rfl@icp.csic.es (R. Fernandez-Lafuente), freire@iq.ufrj.br
(D.M.G. Freire).
sources are clear examples of useful but environmentally danger-
ous products [4]. The current legislative environmental standards
have led to the replacement of mineral oils with more ecologically
benign synthetic oils, usually based on esters of long-chain fatty
acids from vegetable oils and polyols [5].
Vegetable oils are the mainly raw material for this purpose due
to its renewable character and the quality of the final product. Biol-
ubricants may be produced from chemical modification of veg-
etable oils that spare the structural problems of vegetable oils
making them able for the application as lubricant [4] and this pro-
cess can be made by chemical [6] or enzymatic catalysis [7–10].
Also vegetable oils present good viscosity index, good pour point
and good evaporation loss, but their wider application is limited
by the thermal and oxidative instabilities [11].

http://dx.doi.org/10.1016/j.fuel.2017.04.036
0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
Moreover the production by green enzymatic routes together
with their biodegradability constitutes a clear example of green
chemistry, to be considered biodegradable, biolubricants must
decompose within one year through natural degradation [12].
Vegetable oils are known as typically 99% biodegradable, usually
falling to 90–98% after mixing with additives, while mineral oils
are only 20% biodegradable [13]. This is more one between the fac-
tors described here that make vegetable oils a promising substrate
for biolubricant production.
One of the ways to produce biolubricants is through the esteri-
fication of the free fatty acids [9,10] that can be obtained by the
vegetable oil hydrolysis. Ricinoleic acid esters have been suggested
as potential biolubricants, due to their structure. Ricinoleic acid is
the main component of the castor oil plant Ricinus communis L.,
reaching 70–90% of the total FFA content [14]. Castor-oil plant
seeds, in addition to their high oil content [15,16], contain certain
enzymes [17] of which lipases were used in this study [18,19].
Currently, the industrial hydrolysis of this oil to produce free
fatty acids is performed by thermal or chemical hydrolysis (usually
using alkaline catalysts) [20], which can cause oxidation of the
final product (see below). Ricinoleic acid can also be produced by
saponification followed by acidification [21]. This procedure,
besides the mild temperatures compared to thermal hydrolysis,
gives a product with color and odor degradations and a large quan-
tity of an acid sludge and hard water in the downstream process
[22], making it desirable to hydrolyze the oil using lipases [23] to
obtain ricinoleic acid.
Ricinoleic acid, formally known as 12-hydroxy-9-cis-
octadecenoic acid [24], is a fatty acid with specific physical proper-
ties due to the hydroxyl group in carbon 12 and the double bond
between carbons 9 and 10. This rigid configuration offers several
advantages, making this compound interesting for different appli-
cations such as cosmetics (emulsifiers), paper (as anti-foam), food
(as an additive), etc. [25]. However, the oxidative stability of rici-
noleic acid is questioned due to the hydroxyl group in its structure
[26]. The preservation of this hydroxyl group makes it unsuitable
to use chemical catalysis to produce the esters, and reinforces
interest in enzymatic catalysis for the modification of this acid,
which can be produced under milder conditions. However, the very
rigid structure of the acid may make it difficult to find a lipase that
can recognize it as a substrate. On the other hand, considering that
lubricants may be used at high temperatures, this instability of
ricinoleic acid could be a problem for their applications. One likely
solution for this is esterification of the hydroxyl group [27].
Enzymatic catalysis, mainly using lipases (triacylglycerol ester
hydrolases, E.C. 3.1.1.3 - IUPAC), has been used to successfully pro-
duce a variety of fuels like biodiesel and biolubricants [7–10,23]. In
vivo, lipases generally act at a substrate/ aqueous interface,
catalyzing the hydrolysis of triacylglycerol ester linkages,
producing free fatty acids and glycerol [28,29]. To fulfill this
function, lipases have a peculiar catalytic mechanism, called
interfacial activation, which involves the movement of an
oligopeptide that blocks the active center and the lipase adsorption
on the hydrophobic surface of the oil drop [30,31]. In vitro, lipases
can be used to catalyze several synthetic reactions such as
esterification and transesterification [32,33], which increases their
industrial importance.
According to [34], the thermal stability of the esters of ricinoleic
acid is directly related to the presence or absence of the hydrogen
in the b-carbon for the alcohol function. The presence of the hydro-
xyl group could cause decomposition of the ester molecule, pro-
ducing an acid and an alkene with little energy input. Thus, the
selected alcohols for this study were neopenthylglicol (NPG),
trimethylolpropane for (TMP) and pentaerytrithol (PE); the
absence of the b-hydrogen confers steric protection of the ester
linkage and improved thermal and oxidative stability.
197
An esterification reaction is a thermodynamically controlled
process [35–37] Thus, the thermodynamic constant of the process
and therefore the water activity (product of the reaction), concen-
trations and molar ratios of the substrates, pH, temperature, etc.
will determine the yield of the process, whereas the catalyst only
determines the reaction rate [35–37]. However, using this hydroxy
fatty acid, it is possible that, depending on the conditions and
enzyme employed, some estolide (where the hydroxyl groups of
one ricinoleic molecule can be esterified with the carboxylic group
of another molecule) can be produced [38], and this could depend
on the enzyme employed, affecting the overall ester yield. In addi-
tion to being a complication in the reaction design, the presence of
this estolide may be considered as merely a problem, or on the
contrary, may be interesting as a biolubricant (e.g., hydroxyl group
oxidation is no longer a problem).
We analyzed the properties of the several biolubricants pro-
duced: oxidation stability, pour point, viscosity at 40 and 100 °C,
viscosity index (VI), total acidity index (TAI) and water index. As
catalysts, we utilized some of the most frequently used enzymes:
lipase B from Candida antarctica (NovozymÒ 435) [39,40], lipases
from Rhizomucor miehei [41,42] and commercial extracts from Can-
dida rugosa lipases [43]. This last has been shown by zymography
to produce some esterification using castor oil biodiesel [44].
Here we present a new strategy to produce biolubricant mole-
cules based on ricinoleic acid, using castor oil as the raw material
in a solvent-free reaction medium, to avoid the use of any solvent
that could be considered a contaminant. The purpose is to obtain
ricinoleic acid esters through hydroesterification [45,46,33] in a
two-step enzymatic process: enzymatic hydrolysis of vegetable
oils and subsequent esterification of the free fatty acids obtained
by the previous step with the alcohol selected. Direct use of castor
oil as a substrate would eliminate the chemical or thermal catalysis
stage, making the process more ecologically benign and avoiding
undesired modifications.
2. Materials and methods
2.1. Materials
Castor oil was purchased from Campestre (São Bernardo do
Campo, Brazil); Castor seeds and ricinoleic acid were kindly
donated by Bioóleo Bahia (Feira de Santana, Brazil) and
Miracema-Nuodex (Jardim São José, Brazil) respectively. The poly-
ols neopentylglycol (NPG), trimethylolpropane (TMP) and pen-
taerythritol (PE) were purchased from Sigma-Aldrich (Jurubatuba,
Brazil). The lyophilized commercial lipase from Candida rugosa,
LipomodTM 34 MDP, was obtained from Biocatalysts Inc. (USA).
The immobilized enzymes from Rhizomucor miehei, LipozymeÒ
RM-IM, and Candida antarctica, NovozymÒ 435, were obtained
from Novozymes (Araucária, Brazil).
2.2. Methods
2.2.1. Determination of enzymatic activity
2.2.1.1. Activity determination of lipases from beans from castor-oil
plant. The hydrolytic activity of the lipases present in the dormant
castor beans was quantified by titration of butyric acid released in
the hydrolysis of tributyrin (5% w/v) in Triton X-100 (25% w/v) in
0.1 M sodium acetate buffer pH 4.0 at 30 °C using the fat-free ace-
tone preparation of lipases from the castor beans. This preparation
was made as described in [47]. One unit (U) of hydrolytic activity
was defined as the amount of enzyme that releases 1 mmole of
butyric acid per minute under the assay conditions.
2.2.1.2. Hydrolytic activity determination of commercial enzymes.
The hydrolytic activity of the commercial lipases was measured
198 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
using 0.025 M p-nitrophenyl laurate in acetonitrile: DMSO [1:1(v/
v)] (pNP-laurate) as substrate dissolved in 0.025 M phosphate buf-
fer at pH 7.0 as described in [48]. One unit (U) of hydrolytic activity
was defined as the amount of enzyme that releases 1 mmole of p-
nitrophenol per minute under the assay conditions.
2.2.2. Castor oil hydrolysis and recovery of COFFA
2.2.2.1. Hydrolysis. Castor oil was hydrolyzed using 30 g of castor
seeds and 150 mL of 0.1 M sodium acetate buffer pH 4, maintaining
a ratio (w/v) of 1:5. Castor seeds were blended together with the
buffer serving as a biocatalyst for hydrolyzing the endogenous oil
present in the seeds. This suspension was transferred to a jacketed
reactor where it was maintained at 30 °C under mechanical stirring
at 200 rpm.
Castor oil was also hydrolyzed with extra castor oil added to the
mixture of 0.1 M sodium acetate buffer pH 4.0 and seeds, before
blended, in the same volume as the buffer (150 mL). The ratio of
seeds/buffer/castor oil was 1:5:5 (w/v/v). The reactions of both sys-
tems formed an emulsion system and were performed in biological
duplicates and analytical triplicates.
This biocatalyst proposed in this paper is a new, green and low
cost method to hydrolyze castor oil by crushing castor bean seeds
in the reaction medium.
2.2.2.2. Recovery of COFFA. Both reactions described above (Sec-
tion 2.2.2.1) were stopped by extracting the free fatty acids with
ethyl acetate - used in the ratio 1:2 (v/v) of reaction mixture. The
mixture then formed two layers: 1) the upper layer containing sol-
vent and COFFA solved, and 2) the lower layer formed an emulsion
containing crushed seeds, buffer ethyl acetate and glycerol. The
upper layer (1) was recovered and stored and the lower layer (2)
was centrifuged for 5 min at 10,000 rpm to break up the emulsion
recovering the solvent with COFFA and to separate the biocatalyst.
After this procedure, the biocatalyst was washed with more
300 mL of ethyl acetate to recover the still remaining COFFA. The
solvent stored was weighed, and to it, 10% (w/w) of anhydrous
sodium sulfate was added and maintained under constant mag-
netic stirring for 30 min to absorb the remaining water in the sol-
vent. Afterward, the anhydrous sodium sulfate was decanted and
the COFFA-rich solvent was filtered with filter paper and then fully
evaporated in a rotary evaporator to recover the COFFA.
The COFFA content was determined by titration in an automatic
titrator (Mettler Toledo). At fixed intervals, the increase in acidity
was analyzed by titrating triplicate samples of 100 mL COFFA dis-
solved in acetone and ethanol (45 mL) 1:1 (v/v) with 0.4 M NaOH
until the end point at pH 11.
2.2.3. COFFA enzymatic esterification
2.2.3.1. Evaluation of different enzymes in esterification reac-
tions. COFFA were used as substrates in the esterification reactions
catalyzed by the different lipases with the polyols NPG, TMP and PE
at 2.5:1, 3.75:1 and 5:1 molar ratios, respectively. These ratios
were selected to maintain the stoichiometry between the hydrox-
yls of the polyol and COFFA molecules for the three cases. The
esterification reactions were conducted in thermostated reactors
(maximum volume of 30 mL) at 40 °C, under constant magnetic
stirring (200 rpm). The reaction medium consisted of 12g of COFFA
and 0.48g (4% w/w COFFA) of Lipomod 34 MDP, Lipozyme RM-IM
or Novozym 435 as biocatalysts. After 24 and 96 h, the mixture
was centrifuged (10,000 rpm for 5 min) and the supernatant was
stored at 4 °C until analysis of the COFFA by the titration method,
using an automatic titrator (Mettler Toledo) with 0.04 M NaOH as
the titrating reagent until the end point of pH 11 was reached. The
reactions were performed in biological duplicates and analytical
triplicates. The percentage of acidity reduction (conversion to
ester) was calculated as in Eq. (1).
%Acidity reduction ¼ 100  ðCOFFAi  COFFAf Þ=COFFAi ð1Þ
where:
COFFAf: COFFA remaining in the system.
COFFAi: COFFA fed to the system.
2.2.3.2. Analysis of the enzymatic production of estolides. After the
most suitable enzyme was selected, new reactions were performed
under the same conditions mentioned above in Section 2.2.3.1,
however performing the reaction without added polyols in order
to investigate the production of estolides from ricinoleic acid.
The reactions were also performed in biological duplicates and
analytical triplicates.
2.2.4. Polyol solubilization
The temperature of the mixture between COFFA and NPG, TMP
or PE at 2.5:1, 3.75:1 and 5:1 molar ratios, respectively, was slowly
increased from 40 °C, under constant magnetic stirring, until
reaching the melting point of the polyols and/or 150 °C (at this
temperature the color of the COFFA changed, indicating thermal
degradation). After melting, the mixture was maintained under
constant magnetic stirring until the temperature decreased to
40 °C, which was the enzymatic reaction temperature. Using PE,
the solid was heated only until reaching 260 °C (its melting point)
and the liquefied polyol was added to the COFFA solution at 40 °C.
Because the polyol solidified at mild temperature and for the rea-
sons discussed below, PE could not be used to characterize the
reaction product.
2.2.5. Scale-up of biolubricant ester production
The reactions for the characterization of the biolubricant prop-
erties were scaled up for reactors with 210 g COFFA with the fol-
lowing reaction composition:
a) Home-made COFFA + Lipomod 34 MDP (4% w/w) + NPG
(molar ratio 2.5:1);
b) Home-made COFFA + Lipomod 34 MDP (4% w/w) + TMP
(molar ratio 3.75:1);
c) Home-made COFFA + Lipomod 34 MDP (4% w/w);
d) Commercial COFFA + Lipomod 34 MDP (4% w/w);
As both NPG and TMP are solids, the temperature of the mixture
was raised until the polyol crystals were completely solubilized,
and then reduced to the operating temperature of 40 °C before
the enzyme was added. All reactions were performed under con-
stant mechanical stirring at 200 rpm, for 24 h in reactions ‘‘a”
and ‘‘b”, and for 96 h in reactions ‘‘c” and ‘‘d”. At the end of all reac-
tions, the reaction mixture was centrifuged (10,000 rpm, 5 min)
and the supernatant was stored at 4 °C until use.
2.2.6. Biolubricant characterization
The biolubricants produced were characterized in terms of pour
point, kinematic viscosity, viscosity index (VI), oxidative stability
by the rotating pressure vessel method (RPVOT) and total acidity
index (TAI).
Pour point was analyzed according to the ASTM D97 method.
Kinematic viscosity at 40 and 100 °C was determined according
to ASTM D445 and the viscosity index was determined through
calculation from the ASTM D2270 method, which takes into
account the product viscosities at 40 and 100 °C. The oxidative sta-
bility was determined using the RPVOT according to ASTM D2272
as described in [9].
Also, the TAI was determined through potentiometric titration,
using KOH (0.1 mol l–1) and titrating until all acids were neutral-
ized. The end point is expressed in mg KOH g–1 sample.
 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
2.2.7. Nuclear magnetic Resonance analysis
The Nuclear Magnetic Resonance (NMR) analysis was per-
formed with the products from the reactions without solubiliza-
tion of the polyols (96 h), Section 2.2.3.2; and afterward for
reactions ‘‘a” and ‘‘b”, Section 2.2.5. Before the analysis, the reac-
tion medium was centrifuged at 10,000 rpm for 10 min in order
to obtain only the medium-soluble particles. Afterward, the reac-
tion samples from the bench-scale reactions were submitted to
1
H and 13C NMR for structural identification of the reaction prod-
ucts. An Agilent INOVA-300 (7.05T) spectrometer was employed.
Samples were prepared in deuterated chloroform (CDCl3) at
25 °C, at 5 (1H) and 20% 13C concentration, and compared with
pure chemical compounds present in the spectra library [44].
2.2.8. Gas chromatography
The free fatty acids present in the home-made and commercial
COFFA were methylated according to the method adapted by [49]
The fatty acid methyl esters were analyzed in a gas chromatograph
(CG2010; Shimadzu, Japan) equipped with an Omegawax 320 cap-
illary column (30 m  0.32 mm  0.25 lm; Sigma, São Paulo, Bra-
zil). The following conditions were used: injector temperature:
260 °C; carrier gas: helium at 4.0 mLmin1; sample injection vol-
ume of 1 lL with split ratio: 1:20. The column temperature gradi-
ent was 150 °C for 5 min, increased at 2 °C min1 to 210 °C, and
remained constant for 30 min. The flame ionization detector tem-
perature was 280 °C [50]. The results of the biological triplicates
were compared with each other, and the two samples were used
for the t-test with 95% confidence in order to confirm that they
were equivalent.
3. Results and discussion
3.1. Castor oil hydrolysis using endogenous seed enzymes
A strategy for the hydrolysis of castor oil was developed using
the dormant lipases present in the castor beans themselves
[18,19]. However, the COFFA was extracted after hydrolysis using
ethyl acetate. This solvent is described in the ‘‘Guide for industry”
[51] in Class 3, which includes solvents that pose no risk to human
health and have low toxicity. This strategy aimed to produce a biol-
ubricant obtained completely by enzymatic routes and fully envi-
ronmentally benign.
The product of the hydrolysis of only the oil contained in the
seeds (COFFA) by the endogenous lipases with activity of
225.7 ± 0.5 U/g (endogenous oil hydrolysis) reached 93.13 ± 5.9%
acidity in 45 min of reaction, finally reaching 94.62 ± 0.84% in 4 h
(Fig. 1). Despite the rapid hydrolysis of castor oil, the oil content
in the seeds varied from 45 to 50% of the total mass [52], generat-
ing only a small amount of COFFA. Thus, to increase the amount of
COFFA, additional castor oil was added in a ratio of 1:5:5 (w/v/v)
seeds: buffer: castor oil. Then, in comparison to the previous reac-
tion, hydrolysis with the added oil was also performed and its
kinetic hydrolysis (castor oil hydrolysis) is shown in Fig. 1.
The reaction with additional castor oil produced 87.83 ± 3.67%
acidity in 4 h, reaching 91.9 ± 0.65% acidity in 6 h of reaction. Com-
parison with the amount obtained from the endogenous oil hydrol-
ysis (94.62 ± 0.84%) shows a decrease in productivity from
approximately 94 g (endogenous oil hydrolysis) to 15.33 g (castor
oil hydrolysis) of free fatty acids h–1, which was expected due to
the addition of 6-fold more oil for the same amount of enzyme.
High-temperature hydrolysis of oils containing hydroxy fatty
acids, such as castor oil, is not recommended. These conditions
generate undesirable changes in the free fatty acids from thermal
decomposition of triacylglycerol, which leads to changes in color
or odor and also reduces the content of free fatty acids [53].
Besides, the commercial production of free fatty acids is carried
199
100
Acidity (g/100g sample)
80
60
40
20
0
0 4 8 12 16 20 24
Time (h)
Endogenous oil hydrolysis Castor oil hydrolysis
Fig. 1. Comparison between hydrolysis using only crushed seeds or with castor oil
added in the process. For both hydrolysis procedures, castor seeds (with a lipase
activity of 226 U/g) were crushed in a mixer together with 0.1 M buffer pH 4.0 in a
seed:buffer ratio of 1:5 (w/v) in the absence or presence of additional castor oil
[1:5:5(w/v/v)seed/buffer/castor oil]. After trituration of the seeds, both mixtures
were incubated at 30 °C under magnetic stirring. The oil was extracted and its
acidity was determined by titration with 0.04 M NaOH.
out through the noncatalytic hydrolysis of triglycerides using
superheated steam in large reactors made of expensive
corrosion-resistant materials, making the process costly and
energy-intensive [54].
Enzymatic hydrolysis of castor oil has been described previ-
ously. Khaskheli et al. [55], studying optimization of the hydrolysis
of castor oil, employed 0.01% of Rhizopus oryzae lipase in an oil-
water emulsion system in a 1:4 oil:water ratio, in which the reac-
tion was conducted for 12 h, at 37 °C in order to obtain 90% free
fatty acids. Gomes et al. [56], with the main objective of using
the hydrolyzed castor oil in producing aromas, developed an enzy-
matic hydrolysis procedure using commercial lipases. They
obtained the best result with Lipozyme TL-IM, achieving approxi-
mately 95% free fatty acids at 25 h, at pH 8 and 27 °C. Usually,
the enzymatic processes reported in the literature require longer
times to obtain a high level of COFFA, and use commercial
enzymes, which increases the final cost and requires a solvent to
recover the COFFA from the medium.
In the present study, however, it was possible to perform a
hydrolysis with a high reaction rate, using lipases with only the
cost of seed crushing, very mild reaction conditions, and using a
low-toxicity solvent to recover the COFFA, so that this process con-
forms to the aims of green chemistry.
Thus, hydrolysis of the castor oil using endogenous enzymes
from the seeds as catalysts was chosen as a first step in the new
route of biolubricant production using castor oil.
The FFA compositions of the COFFA produced by our methodol-
ogy and a commercial COFFA supplied by Miracema are very sim-
ilar (Table 1). However, commercial samples have more than a 4-
fold higher percentage of undetermined compounds. This index
of non-identified fatty acids of the commercial COFFA may be
related to products of thermal degradation of the castor oil and
COFFA. This difference in the quality of the samples from the two
processes is perceptible not only from the lack of odor but also
from the appearance of the two samples at similar concentrations,
as seen in Fig. 2. Commercial COFFA (B) is reddish, while home-
made COFFA (A) is yellowish, closer to the color of castor oil (C)
This suggests that the commercial hydrolysis process partially
degrades the acids of the castor oil.
3.2. Production of biolubricants by COFFA esterification
3.2.1. Evaluation of different enzymes in esterification reactions
The COFFA obtained in step 3.1 was used for the esterification
reactions to obtain esters from NPG, TMP and PE, catalyzed by
200 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
Table 1
Comparison between the fatty acid composition of home-made and commercial
COFFA.
Free fatty acids (FFA) Home-made (g fatty acid/ Commercial(g fatty acid/
100 g of sample) 100 g of sample)
C16:0 1.02 ± 0.01 1.01 ± 0.05
C18:0 0.70 ± 0.01 0.39 ± 0.03
C18:1n9 2.73 ± 0.02 2.82 ± 0.15
C18:2n6 4.96 ± 0.06 3.48 ± 1.18
C18:3n3 0.47 ± 0.03 0.43 ± 0.02
C22:0 1.43 ± 0.31 2.17 ± 0.89
C18:1-OH 87.57 ± 0.15 85.61 ± 2.75
Non-identified 0.95 ± 0.25 4.308 ± 2.7
the commercial lipases Lipomod 34 MDP, Lipozyme RM-IM and
Novozym 435. The results of the acidity reduction (%) in these reac-
tions are shown in Table 2.
All enzymes evaluated catalyzed some degree of conversion to
esters using NPG and TMP. However, using PE, the reaction pro-
gressed only partly when CRL Lipomod 34 MDP was used (Table 2).
The use of NPG reached different conversions by the use of the dif-
ferent lipases. Lipozyme RM-IM showed 24.8 ± 0.3% conversion at
24 h and reached 61.0 ± 0.6% in 96 h of reaction; Novozymes 435
achieved in 24 h of reaction 23.0 ± 0.2% reaching 55.1 ± 0.2% at
96 h; and, over all the enzymes, Lipomod 34 MDP reached
80.6 ± 1.4% at 24 h and 90.2 ± 0.9% in 96 h of reaction. The use of
TMP show results quite inferiors for all lipases: Lipozyme RM-IM
which the conversion to esters at 24 h was almost negligible
(3.5 ± 1.5%), reaching 27.8 ± 0.3% in 96 h and Novozym 435 that
performed slightly better with the use of TMP, reaching 13 ± 0.5%
in 24 h; however, this conversion was statistically the same at
96 h – 16.4 ± 0.4% –, Lipomod 34 MDP, however, presents
80.6 ± 1.4% conversion in only 24 h and 90.7 ± 1.5% conversion at
96 h. The use of PE, however, did not show any result when it used
Lipomod RM-IM nor Novozyme 435 but for Lipomod 34 MDP the
acidity reduction was approximately 74.9 ± 0.3% in 96 h.
The commercial enzyme from Candida rugosa has been reported
as a specific catalyst for the modification of castor bean derivatives,
to produce either ricinoleic acid oligomers known as estolides [36],
or biolubricants from castor biodiesel [9]. According to Da Silva
et al. [9], better results were obtained for the reaction between cas-
tor biodiesel and TMP with the use of the C. rugosa lipase Lipomod
34 MDP than with the lipases Lipozyme RM-IM and Novozym 435.
The enzymatic reaction produced a higher conversion (approxi-
mately 98%) compared to chemical catalysis (60%).
The better performance of this enzyme for castor oil free fatty
acids may be related to the active site architecture of CRL. Some-
how, this singular architecture favors the recognition of ricinoleic
Fig. 2. Color comparison between home-made COFFA (A), commercial COFFA (B) and castor oil (C).
acid. This fatty acid has a hydroxyl group in position 12, which
forms an ‘‘L” shape that probably fits better in the active site of
CRL [43].
3.2.2. Analysis of the enzymatic production of estolides
Some reports have described the formation of oligomeric polye-
sters of hydroxylated fatty acids known as estolides [57]. Bódalo
et al. [38] described the use of CRL to produce estolides from rici-
noleic fatty acid. This acid is the majority component of COFFA,
comprising about 88% of its fatty acid composition (see Table 1).
This acid has a hydroxyl group at carbon 12 that enables the ester-
ification reaction between this hydroxyl and the carboxyl of differ-
ent molecules, forming oligomeric polyesters of ricinoleic acid [58],
as seen in Fig. 3 showing the estolide formation catalyzed by CRL
[58].
In view of the possibility that the reduction in acidity of the
reaction samples could be related not only to the production of
polyol derivatives but also to the formation of estolides, a
reaction was performed using only COFFA and CRL (Lipomod 34
MDP) as the biocatalyst under similar reaction conditions to those
used in the presence of polyols (see Section 2.2.3.2). Fig. 4 shows
the high percentage of acidity reduction in this reaction, using
Lipomod 34 MDP as biocatalysts. The result was similar to that
obtained in the presence of polyols, and suggested that the
estolide formation may be the main reaction using the lipases
Lipomod 34 MDP, since the other two enzymes produced no
reduction in the acidity in this reaction medium without polyols
(data not shown).
Using NPG or TMP, an acidity reduction of about 90% after 96 h
of reaction was observed, and using PE this reduction reached 80%
in 96 h. Without polyols, the initial acidity decreased to approxi-
mately 40% in 24 h and 70% in 96 h of reaction. This result not only
showed that the lipase Lipomod 34 MDP is capable of catalyzing
the esterification between the ricinoleic acid, forming estolides,
but also suggested that estolides may be also a main reaction pro-
duct even in the presence of polyols.
3.2.3. NMR analyses of the reaction products
NMR analyses were performed for the final reaction products
(96 h) obtained from the reactions in Section 2.2.3.2. Table 3 sum-
marizes the composition of the different final reaction media.
The data in Table 3 show that the reactions performed without
polyol and with PE produced only estolide molecules. It is possible
that the PE was not properly dissolved during the reaction. This
polyol did not appear in the NMR analyses, as occurred with the
other polyols (Table 3). As the NMR analyses were performed with
only the supernatant from the reaction-medium centrifugation
(section 2.2.7), this result confirms that the PE concentration in
 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205 201

Table 2
Acidity reduction (%) or conversion to esters from the ester synthesis catalyzed by lipases Novozym 435, Lipozyme RM-IM and Lipomod 34 MDP at 4% (w/w of COFFA). The
reactions, described in the subsection 2.2.3.1, were conducted at 40 °C in a stirred-bed reactor using COFFA and NPG, TMP or PE polyols as the substrate in COFFA:polyol molar
ratios of 2.5:1, 3.75:1 and 5:1, respectively.

NPG TMP PE
24 h RM-IM 24.8 ± 0.3 3.5 ± 1.5 0
Novozymes 435 23.0 ± 0.2 13 ± 0.5 0
Lipomod 34 MDP 80.6 ± 1.4 65.2 ± 0.3 48.9 ± 0.8
96 h RM-IM 61.0 ± 0.6 27.8 ± 0.3 0
Novozymes 435 55.1 ± 0.2 16.4 ± 0.4 0
Lipomod 34 MDP 90.2 ± 0.9 90.7 ± 1.5 74.9 ± 0.3

Fig. 3. Estolide formation from the polymerization of ricinoleic acid.

100 result shows the inability of these enzymes to produce estolides

from ricinoleic acids. The other two polyols dissolved, although
Acidity reducon (%)

80 not immediately, in the reaction medium, allowing some ester

formation.
60 The presence of polyols using Lipomod 34 MDP reduced the
degree of polymerization of the estolide. The reaction performed
40 in absence of polyols produced hexameter or heptamer molecules.
However, in the presence of PE, the maximum degree of estolide
20 polymerization was tetramers, showing that the presence of a solid
powder of this polyol hindered the formation of larger estolide
0 molecules, even without being soluble, perhaps by somehow
24 h 96 h affecting the aggregates of the enzyme (Table 3).
NPG TMP PE Without polyols Although the production of esters from NPG and TMP was
observed, the desired peracylated product (dineopentyl rici-
Fig. 4. Acidity reduction (%) from the ester synthesis and control reactions
noleate) was obtained only with NPG. Using TMP, mono- and
catalyzed by the commercial lipase Lipomod 34 MDP at 4% (w/w of COFFA). The
reactions were conducted at 40 °C and aliquots for titration analysis are taken in 24 dimethylpropyl ricinoleates were observed, but no trimethylpropyl
and 96 h. The reaction were conducted in a stirred-bed reactor using COFFA and ricinoleate was detected (Table 3). However, all free fatty acids had
NPG, TMP or PE polyols as substrates in COFFA:polyol molar ratios of 2.5:1; 3.75:1 been consumed to produce mainly dimers of ricinoleic acid (com-
and 5:1, respectively. The control reaction that was performed under the same prising 70.7 and 65.9% of the total weight, respectively).
conditions, in the absence of polyols.
A likely explanation for these results is that the delay in the sol-
ubilization of the polyols may permit the formation of dimers of
the reaction medium was negligible. On the other hand, the utiliza- ricinoleic acid before the polyol solubilization, and this prevents
tion of PE as a substrate in the reactions catalyzed by Lipozyme the synthesis of polyol-peracylated compounds. Therefore, solubi-
RM-IM and Novozym 435 did not produce any ester (Table 2). This lization of the polyols was optimized before starting the reaction.
202 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205

Table 3
NMR analysis of the supernatant obtained by centrifugation of the reactions between home-made COFFA with and without polyols NPG, TMP and PE (molar ratios 2.5:1, 3.75:1
and 5:1, respectively) the reactions are described in the Section 2.2.3.2.

Substrates Compound Molar (%) Weight (%)

COFFA + NPG (molar ratio 2,5:1) 6,7 1,3
NEOPENTHYLGLYCOL (NPG)

13,6 17,1

DINEOPENTHYL RICINOLEATE
15 10,9
MONONEOPENTHYL RICINOLEATE

64,7 70,7

RICINOLEIC ACID DIMER

COFFA + TMP(molar ratio 3,75:1) 9,8 2,5
TRIMETHYLOLPROPANE (TMP)

15 19,7

DIMETHYLOLPROPYL RICINOLEATE

15,1 11,9

MONOMETHYLPROPYL RICINOLEATE
60,1 65,9
RICINOLEIC ACID DIMER

COFFA + PE(molar ratio 5:1) 100 100

RICINOLEIC ACID OLIGOMERS (n = 2)

COFFA 100 100

RICINOLEIC ACID OLIGOMERS (n = 4 or 5)

3.2.4. Effect of polyol solubilization

100
Before the enzyme addition, the reaction medium was heated
Acidity reducon (%)

until complete solubilization of the different polyols as described 80

in Section 2.2.4.
Using PE, although the temperature of the reaction medium was 60
increased to 150 °C it remained fully insoluble. To solve this prob-
lem, PE was incubated at 260 °C (the melting point) and was incor- 40
porated as a liquid in the COFFA sample preheated to 40 °C.
However, the PE solidified when the temperature decreased. 20
Therefore we discarded the use of this polyol in this medium. PE
has been used in chemical catalysis in similar reactions, but at 0
higher temperatures, for example in the transesterification with 16 h 24 h 48 h 96 h
castor oil at 200 °C [59] and with palm oil at 158 °C [60]; however, TMP NPG
the oxidative stability of the products obtained was not reported.
Both NPG and TMP were solubilized by heating the reaction Fig. 5. Acidity reduction (%) from the ester synthesis, with fully solubilized polyols,
catalyzed by the commercial lipase Lipomod 34 MDP at 4% (w/w of COFFA). The
medium to between 56 and 58 °C and they remained in solution
biolubricant synthesis reactions were conducted at 40 °C for 96 h in a stirred-bed
even when the temperature was reduced again to 40 °C. Thus, it reactor using COFFA and NPG or TMP polyols as substrates in COFFA:polyol molar
was possible to have these two polyols soluble before starting ratios of 2.5:1 and 3.75:1 respectively.
the reaction.
Solubilizing the polyols allowed us to reduce the time required
to obtain an approximately 90% reduction of acidity from 96 h to final reaction mixtures are shown in Table 4. Using pre-
24 h, suggesting that the lack of polyol solubility hindered the solubilized alcohols, the polyol esters increased while the forma-
esterification of the COFFA (Fig. 5). tion of ricinoleic acid dimers decreased. However, the dimer of rici-
NMR analyses of these new reaction products (‘‘a” and ‘‘b” of noleic acid remained the main component (53.3% using NPG and
Section 2.2.5) were performed, and the main components of the 65.4% using TMP of the total weight), while the peracylated polyols
 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205 203

Table 4
NMR analysis of the supernatant obtained by centrifugation of the reactions ‘‘a” and ‘‘b” between home-made COFFA and the fully solubilized polyols NPG and TMP (molar ratios
2.5:1 and 3.75:1 respectively). The reactions are described in the Section 2.2.5.

Substrates Compounds Molar (%) Weight (%)

COFFA + NPG (molar ratio 2,5:1) 17,4 3,9
NEOPENTHYLGLYCOL (NPG)

15,1 21,8
DINEOPENTHYL RICINOLEATE

25,1 20,9
MONONEOPENTHYL RICINOLEATE

42,4 53,3
RICINOLEIC ACID DIMER

COFFA + TMP (molar ratio 3,75:1) 4,4 1

TRIMETHYLOLPROPANE

15,3 18,1

DIMETHYLOLPROPYL RICINOLEATE

7,4 5,5

MONOMETHYLPROPYL RICINOLEATE

6 10
TRIMETHYLOLPROPYL RICINOLEATE

66,4 65,4
RICINOLEIC ACID DIMER

were a minority product (21.8% dineopenthyl ricinoleate and only
10% trimethylolpropyl ricinoleate of the total weight)
This suggests that Lipomod 34 MDP lipase has a preferential
affinity for ricinoleic acid as a nucleophile reactant compared to
the polyols used, even when these polyols are present in molar
excess.
3.2.5. Product characterization as lubricants
Table 5 shows the main features of the products obtained using
COFFA produced by our methodology (home-made COFFA) with
previously solubilized polyols and with the products obtained
without polyols using a commercial COFFA or our home-made
COFFA
The lubricating properties of all samples were satisfactory. Both
the pour point and oxidative stability were similar and higher than
those previously reported for biolubricants [9,10,25,61]. Surpris-
ingly, when only long-chain estolides were present, the viscosity
index (162 and 154 for home-made and commercial COFFA,
respectively) was higher than the results for the other products.
Viscosity is the most important index of these products, since it
relates to the formation of a protective film on the metal, which
protects against friction [62]. A high viscosity index such as this
means that the viscosity of the fluid does not change with
increased temperature [9,62], which is desirable for any lubricant
application. However, the IAT was higher using only estolides
(40.4 and 58.4 mg KOH g–1 of home-made and commercial COFFA,
respectively) than the NPG and TMP products (7.3 and 4.3 mg KOH
g–1 of sample, respectively). This can be explained by the acidic
characteristic of the estolide molecule from the carboxylic tip of
ricinoleic acid. The acidity of the final product might be decreased
using different strategies presently under study in our laboratory,
e.g., esterification of estolides with a simple alcohol.
On the other hand, the use of home-made or commercial COFFA
has some clear differences. The home-made product had a lower
pour point (–42 °C versus –36 °C) and a 1.4-fold increase in the
oxidative stability (51 min versus 37 min). These differences must
be attributed to the different hydrolysis processes: the enzymatic
process preserved the ricinoleic acid structure, while the thermal

Table 5
Biolubricant properties of the products from the ester reactions between home-made COFFA and NPG and TMP (molar ratios 2.5:1 and 3.75:1, respectively) and without polyols
using commercial and home-made COFFA. All reactions were catalyzed by Lipomod 34 MDP at 4% (w/w of COFFA) and are described in the Section 2.2.5.

Assay Reaction substrates

COFFA + NPG COFFA + TMP Home-made COFFA Commercial COFFA
2
Viscosity 40 °C (mm /s) 172.0 278.1 379.9 408.18
Viscosity 100 °C (mm2/s) 17.60 24.49 41.43 41.54
Viscosity index 111 112 162 154
Oxidative estability (min) 50 54 51 37
Pour point (°C) 42 39 -42 -36
IAT potentiometric (mg KOH/g of sample) 7.3 4.3 40.4 58.4
204 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
process produced byproducts that affected the biolubricant fea-
tures. The product obtained from the esterification of ricinoleic
acid obtained using enzymatically produced COFFA is better than
any other biolubricant molecules described in the literature, and
is produced in an eco-friendly way.
4. Conclusion
The new two-step enzymatic process described here permits
one to obtain, in a green and very efficient way, new products with
properties that surpass those of other biolubricants reported in the
literature. The use of endogenous lipases from castor-oil seed is a
low-cost and efficient method to obtain intact COFFA, and the fur-
ther estolide formation catalyzed by the enzymes from C. rugosa
allows the formation of hexamers/heptamer estolides of ricinoleic
acid. Ricinoleic acid is a better nucleophile for this enzyme than the
polyols employed, whose reactions always generate estolides as
the main product. These estolides have promising lubricant charac-
teristics (better than the products obtained using the polyols) for
industrial use. They also improve the properties of the product
derived from commercial COFFA obtained by thermal treatment,
confirming that the new route is not only more sustainable, but
also more efficient. However, some properties such as excess acid-
ity need to be improved, and are presently a focus of our research
group.
Acknowledgements
The authors are grateful to CNPq (Brazil, Conselho Nacional de
Desenvolvimento Científico e Tecnológico), PVE Project from CNPq,
CAPES (Brazil, Coordenação de Aperfeiçoamento de Pessoal de
Nível Superior), FAPERJ (Brazil, Fundação Carlos Chagas Filho de
Apoio à Pesquisa do Estado do Rio de Janeiro), MINECO (Spain, pro-
ject number CTQ2013-41507-R) and Petrobras, ANP, MCTI and
FINEP for scholarships and financial support.
References
[1] Tollefson J. How much longer can Antarctica’s hostile ocean delay global
warming? Nature 2016;539(7629):346–8.
[2] Anastas P, Eghbali N. Green chemistry: principles and practice. Chem Soc Rev
2010;39(1):301–12.
[3] Ferreira-Leitão VS, Cammarota MC, Aguieiras ECG, de Sá LRV, Fernandez-
Lafuente R, Freire DMG. The protagonism of biocatalysis in green chemistry
and its environmental benefits. Catalysts 2017;7:9. http://dx.doi.org/
10.3390/catal7010009.
[4] Panchal TM, Patel A, Chauhan DD, Thomas M, Patel JV. A methodological
review on bio-lubricants from vegetable oil based resources. Renew Sustain
Energy Rev 2017;70:65–70.
[5] Kania D, Yunus R, Omar R, Rashid SA, Jan BM. Review of biolubricants in
drilling fluids: Recent research, performance, and applications. J Petrol Sci Eng
2015;135:177–84.
[6] McNutt J, He SQ. Development of biolubricants from vegetable oils via
chemical modification. J Ind Eng Chem 2016;36:1–12.
[7] Nagendrama P, Kaul S. Development of ecofriendly/biodegradable lubricants:
an overview. Renew Sustain Energy Rev 2012;16:764–74.
[8] Dormo N, Belafi-Bako K, Bartha L, Ehrenstein U, Gubicza L. Manufacture of an
environmental-safe biolubricant from fusel oil by enzymatic esterification in
solvent-free system. Biochem Eng J 2004;21:229–34.
[9] Da Silva JAC, Soares VF, Fernandez-Lafuente R, Habert AC, Freire DMG.
Enzymatic production and characterization of potential biolubricants from
castor bean biodiesel. J Mol Catal B Enzym 2015;122:323–9.
[10] Åkerman CO, Hagströma AEV, Mollaahmada MA, Karlssonb S, Hatti-Kaul R.
Biolubricant synthesis using immobilised lipase: process optimisation of
trimethylolpropane oleate production. Process Biochem 2011;46:2225–31.
[11] Bart JCJ, Gucciardi E, Cavallaro S. Renewable feedstocks for lubricant
production. In: Biolubricants Science and technology. Woodhead Publishing
Limited; 2013.
[12] Whitby RD. Understanding the global lubricants business-regional markets,
economic issues and profitability. In: Course notes. Oxford, England: The
Oxford Princeton Programme; 2005.
[13] Makkonen I. Environmentally compatible oils, Field Note No. General, Pointe-
Claire, 39. Que: Forest Engineering Research Institute of Canada (FERIC); 1994.
[14] Goswami D, Sen R, Basu JK, De S. Maximization of bioconversion of castor oil
into ricinoleic acid by response surface methodology. Bioresour Technol
2009;100(18):4067–73.
[15] Scarpa A, Guerci A. Various uses of the castor oil plant (Ricinus communis L.) a
review. J Ethnopharmacol 1982;5:117–37.
[16] Bale AT, Adebayo RT, Ogundele DT, Bodunde VT. Fatty acid composition and
physicochemical properties of castor (Ricinus Communis L.) seed obtained
from malete, moro local government area, kwara state. Nigeria. Chem Mater
Res 2013;3(12).
[17] Kermode AR, Bewley JD. The role of maturation drying in the transition from
seed development to germination: II. Post-germinative enzyme production
and soluble protein synthetic pattern changes within the endosperm of
Ricinus communis L. seeds. J Exp Bot 1985;36(12):1916–27.
[18] Avelar MHM, Cassimiro DMJ, Santos KC, Domingues RCC, De Castro HF,
Mendes AA. Hydrolysis of vegetable oils catalyzed by lipase extract powder
from dormant castor bean seeds. Ind Crops Prod 2013;44:452–8.
[19] Fuchs C, Vine N, Hills MJ. Purification and characterization of the acid
lipase from the endosperm of castor oil seeds. J Plant Physiol 1996;149(1–2):
23–9.
[20] Maher KD, Bressler DC. Review Pyrolysis of triglyceride materials for the
production of renewable fuels and chemicals. Bioresour Technol 2007;98
(12):2351–68.
[21] Satyarthi JK, Srinivas D, Ratnasamy P. Hydrolysis of vegetable oils and fats to
fatty acids over solid acid catalysts. Appl Catal A 2011;391(1–2):427–35.
[22] Noor IM, Hasan M, Ramchandran KB. Effect of operating variables on
thehydrolysis rate of palm by lipase. Process Biochem 2003;39:13–20.
[23] Hermansyah H, Wijanarko A, Dianursanti D, Gozan M, Praswasti PDKW,
Arbianti R, et al. Kinetic model for triglyceride hydrolysis using lipase: review.
Makara J Technol 2007;11(1):30–5.
[24] Goodrum JW, Geller DP. Influence of fatty acid methyl esters from
hydroxylated vegetable oils on diesel fuel lubricity. Bioresour Technol
2005;96(7):851–5.
[25] Beopoulos A, Verbeke J, Bordes F, Guicherd M, Bressy M, Marty A, et al.
Metabolic engineering for ricinoleic acid production in the oleaginous yeast
Yarrowia lipolytica. Appl Microbiol Biotechnol 2014;98:251–62.
[26] Azevedo DMP, Lima EF. O Agronegócio da Mamona no Brasil. 21ª
ed. Embrapa: Brasília; 2001.
[27] Yao L, Hammond EG, Wang T, Bhuyan S, Sundararajan S. Synthesis and
physical properties of potential biolubricants based on ricinoleic acid. J Am Oil
Chemist’s Soc 2010;87(8):937–46.
[28] Angajala G, Pavan P, Subashini R. Lipases: an overview of its current challenges
and prospectives in the revolution of biocatalysis. Biocatal Agric Biotechnol
2016;7:257–70.
[29] Sharma R, Chisti Y, Banerjee UC. Production, purification, characterization and
applications of lipases. Biotechnol Adv 2001;19:627–62.
[30] Derewenda ZS. Structure and function of lipases. Adv Protein Chem
1994;45:1–52.
[31] Verger R. ‘‘Interfacial activation” of lipases: facts and artifacts. Trends
Biotechnol 1997;15:32–8.
[32] Aarthy M, Saravanan P, Gowthaman MK, Rose C, Kamini NR. Enzymatic
transesterification for production of biodiesel using yeast lipases: an overview.
Chem Eng Res Des 2014;92:1591–601.
[33] Aguieiras ECG, Cavalcanti-Oliveira ED, Freire DMG. Current status and new
developments of biodiesel production using fungal lipases. Fuel
2015;159:52–67.
[34] Mouloungui Z, Pelet S. Study of theacyl transfer reaction: structure and
properties of glycerol carbonate esters. Eur J Lipid Sci Technol
2001;103:216–22.
[35] Kasche V. Mechanism and yields in enzyme catalysed equilibrium and
kinetically controlled synthesis of b-lactam antibiotics, peptides and other
condensation products. Enzyme Microb Technol 1986;8(1):4–16.
[36] Kasche V, Haufler U, Riechmann L. Equilibrium and kinetically controlled
synthesis with enzymes: semisynthesis of penicillins and peptides. Methods
Enzymol 1987;136:280–92.
[37] Švedas VK, Margolin AL, Berezin IV. Enzymatic synthesis of b-lactam
antibiotics: a thermodynamic background. Enzyme Microb Technol 1980;2
(2):138–44.
[38] Bódalo-Santoyo A, Bastida-Rodríguez J, Máximo-Martín MF, Montiel-Morte
MC, Murcia-Almagro MD. Enzymatic biosynthesis of ricinoleic acid estolídeos.
Biochem Eng J 2005;26:155–8.
[39] Haigha KF, Vladisavljevića GT, Reynoldsb JC, Nagya Z, Sahac B. Kinetics of the
pre-treatment of used cooking oil using Novozyme 435 for biodiesel
production. Chem Eng Res Des 2014;92(4):713–9.
[40] Anderson EM, Larsson KM, Kirk O. One biocatalyst - many applications: the use
of Candida antarctica B-lipase in organic synthesis. Biocatal Biotransform
1998;16(3):181–204.
[41] Andualema B, Gessesse B. Microbial lipases and their industrial applications:
review. Biotechnology 2012;11(3):100–18.
[42] Rodrigues RC, Fernandez-Lafuente R. Lipase from Rhizomucor miehei as an
industrial biocatalyst in chemical process. J Mol Catal B Enzym 2010;64(1–2):
1–22.
[43] De María PD, Sánchez-Montero JM, Sinisterra JV, Alcántara AR. Understanding
Candida rugosa lipases: an overview. Biotechnol Adv 2006;24(2):180–96.
[44] Duarte JG, Leone-Ignacio K, Da Silva JAC, Fernandez-Lafuente R, Freire DMG.
Rapid determination of the synthetic activity of lipases/esterases via
transesterification and esterification zymography. Fuel 2016;177:123–9.
 J. Greco-Duarte et al. / Fuel 202 (2017) 196–205
[45] Pourzolfaghar H, Abnisa F, Daud WMAW, Aroua MK. A review of the enzymatic
hydroesterification process for biodiesel production. Renew Sustain Energy
Rev 2016;61:245–57.
[46] De Sousa JS, Cavalcanti-Oliveira EA, Aranda DAG, Freire DMG. Application of
lipase from the physic nut (Jatropha curcas L.) to a new hybrid
(enzyme/chemical) hydroesterification process for biodiesel production. J
Mol Catal B Enzym 2010;65:133–7.
[47] Cavalcanti EDC, Maciel FM, Villeneuve P, Lago RCA, Machado OLT, Freire DMG.
Acetona powder from dormant seeds of Ricinus communis L. Appl Biochem
Biotechnol 2007;136–140:57–66.
[48] Gutarra MLE, Godoy MG, Maugeri F, Rodrigues MI, Freire DMG, Castilho LR.
Production of an acidic and thermostable lipase of the mesophilic fungus
Penicillium simplicissimum by solid-state fermentation. Bioresour Technol
2009;100:5249–54.
[49] Lepage G, Roy CC. Direct transesterification of all classes of lipids in a one-step
reaction. J Lipids Res 1986;27:114–20.
[50] Sousa JS, Torres AG, Freire DMG. Nutritional enrichment of vegetable oils with
long-chain n-3 fatty acids through enzymatic interesterification with a new
vegetable lipase. Grasas y Aceites Int J Fats Oils 2015;66:2.
[51] Food and drug administration. Guide for industry Q3C – Tables and list
2012. In: http://www.fda.gov/downloads/drugs/guidancecomplianceregulatory-
information/guidances/ucm073395.pdf [access 02.01.17].
[52] Tambascia MB, Teixeira JPF. Mamona: determinação quantitativa do teor de
óleo. Bragantia 1986;45(1):23–7.
[53] Holliday RL, King JW, List GR. Hydrolysis of vegetable oils in sub- and
supercritical water. Ind Eng Chem Res 1997;36(3):932–5. http://dx.doi.org/
10.1021/ie960668f.
205
[54] Ngaosuwan K, Lotero E, Suwannakarn K, Goodwin Jr JG, Praserthdam P.
Hydrolysis of triglycerides using solid acid catalysts. Ind Eng Chem Res
2009;48(10):4757–67. http://dx.doi.org/10.1021/ie8013988.
[55] Khaskheli AA, Talpur FN, Ashraf MA, Cebeci A, Jawaid S, Afridi HI. Monitoring
the Rhizopus oryzae lipase catalyzed hydrolysis of castor oil by ATR-FTIR
spectroscopy. J Mol Catal B Enzym 2015;113:56–61.
[56] Gomes N, Braga A, Teixeira JA, Belo I. Impact of lipase-mediated hydrolysis of
castor oil on c-decalactone production by yarrowia lipolytica. J Am Oil Chem
Soc 2013;90(8):1131–7.
[57] Pelaez M, Orellana C, Marqués A, Busquets M, Guerrero A, Manresa A. Natural
estolides produced by Pseudomonas sp. 42A2 grown on oleic acid: Production
and characterization. JAOCS 2003;80(9):859–66.
[58] Bódalo A, Bastida J, Máximo MF, Montiel MC, Murcia MD, Ortega S. Influence of
the operating conditions on lipase-catalysed synthesis of ricinoleic acid
estolídeos in solvente-free systems. Biochem Eng J 2009;44:214–9.
[59] Valero MF, Gonzalez A. Polyurethane adhesive system from castor oil modified
by a transesterification reaction. J Elastomers Plast 2012;44(5):433–42.
[60] Aziz NAM, Yunus R, Rashid U, Syam AM. Application of response surface
methodology (RSM) for optimizing the palm-based pentaerythritol ester
synthesis. Ind Crops Prod 2014;62:305–12.
[61] Cavalcante IG, Rocha NRC, Maier ME, Lima APD, Neto DMA, Brito DHA, et al.
Synthesis and characterization of new esters of oleic acid and glycerol
analogues as potential lubricants. Ind Crops Prod 2014;62:453–9.
[62] Bart JCJ, Gucciardi E, Cavallaro S. Principals of lubrication. In: Biolubricants
Science and Technology. Woodhead publishing Series in Energy; 2013. 10-22.