Catalysis Communications 78 (2016) 7–10

Contents lists available at ScienceDirect

Catalysis Communications

journal homepage: www.elsevier.com/locate/catcom

Short communication

Hydrolysis of triacylglyceridesin the presence of tin(IV) catalysts

Eid C. da Silva, Paula R. Mendes, Yariadner C. Brito, Mario R. Meneghetti, Simoni M.P. Meneghetti ⁎
Instituto de Química e Biotecnologia, Universidade Federal de Alagoas, Av. Lourival de Melo Mota, s/n°, Maceió-AL–57072-970, Brazil

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

Article history: Tin(IV)-based compounds (mono-n-butyltin hydroxide oxide (a), di-n-butyl-oxo-stannane (b), di-n-butyl tin
Received 2 June 2015 dilaurate (c) and tin oxide (d)) were used as catalysts in the hydrolysis of triacylglycerides (TGs). The effects
Received in revised form 25 January 2016 of the nature and amount of the catalysts, the nature of the TGs, the temperature and reaction time were evalu-
Accepted 28 January 2016
ated. Compared with the usual industrial processes, these systems showed excellent activities at mild tempera-
Available online 30 January 2016
tures and pressures, and low amounts of catalyst (ca. 0.3% w/woil). Fatty acid (FA) conversions of ca. 97% were
Keywords:
obtained at 180 °C, 4 h at oil:water:catalyst a molar ratio of 1:24:0.01 and, using soybean oil.
Hydrolysis © 2016 Elsevier B.V. All rights reserved.
Triacylglycerides
Fatty acid
Tin catalysts

1. Introduction
Triacylglycerides (TGs) and their derivatives, present in oils and fats
from animal or vegetal origin, are renewable sources that can produce
several useful products. They are strong candidates to replace fuels
and chemicals obtained from fossil sources. The chemical structure of
these species allows a diversity of reactions, such as transesterifications,
esterifications, oxidative polymerization and hydrolysis [1,2].
In the oleochemical industry, the hydrolysis is very important
because, via this reaction, highly pure fatty acids (FAs) and glycerol,
which have important applications in the production of soaps, surfac-
tants, lubricants, cosmetics, foods, pharmaceutics, plasticizers, paints,
biofuels, etc., are commonly produced [3,4].
Conventionally, the hydrolysis of an ester (mono-, di- or
triacylglycerides abbreviated as MGs, DGs and TGs, respectively) may
occur in the presence of basic or acid Bronsted catalysts or even en-
zymes. Independent of the nature of the catalyst used, the hydrolysis
of TGs is based on three consecutive steps: TGs are converted to DGs
that are converted to MGs and finally to glycerol, resulting in three
FA molecules (individually formed in each step) — see Figure S1 in
Supporting Information [2,5–7].
Due to the immiscibility of vegetable oils (or fats) in water, it is often
necessary to operate the hydrolysis process at high temperatures or
pressures to increase the compatibility of such components [3,4,7–9].
Consequently, the industrial processes are based on high energy
consumption and high costs due the use of a considerable amount of
water vapor overheated and use of large reactors made of materials
resistant to corrosion, which are associated at the poor quality of the
⁎ Corresponding author.
E-mail address: simoni.plentz@gmail.com (S.M.P. Meneghetti).
obtained products (FAs and glycerol) and require additional purification
steps [9].
Therefore, the investigation of alternative chemical or enzymatic
catalytic systems that help overcome the problems discussed above is
very important. Indeed, many proposals have been presented and
discussed in the scientific and technical literature, including the use of
Lewis acid species [3,9–12]. In terms of the hydrolysis mechanism, it is
possible to adopt the same considerations to transesterification and es-
terification reactions catalyzed by Lewis acids [12,13]. These reactions
involve the formation of a Lewis acid–base complex through an interac-
tion between the acylglycerides (TG, DG, and MG) at the metal center
via the oxygen of the carbonyl group.
There are some examples of use of tin(IV)-based homogeneous or
heterogeneous catalytic systems in esterification, transesterification
and polycondensation reactions to produce polymeric and intermediate
materials [14–16]. Recently, we reported the potential use of this family
of catalysts for fatty acid alkyl ester production [17–20].
However, to the best of our knowledge, no information is available
about the performance of tin(IV) compounds as catalysts for
acylglyceride hydrolysis. In this context, mono-n-butyltin hydroxide
oxide (a), di-n-butyl-oxo-stannane (b), di-n-butyl tin dilaurate (c)
and tin oxide (d) were used, as examples of tin(IV) compounds, to eval-
uate their potential as catalysts for the production of FAs from TGs.
2. Experimental
2.1. Materials
All of the chemicals were used as received without further purifica-
tion. Mono-n-butyltin hydroxide oxide (a), di-n-butyl-oxo-stannane
(b), di-n-butyl tin dilaurate (c), and tin oxide (d) were purchased

http://dx.doi.org/10.1016/j.catcom.2016.01.032
1566-7367/© 2016 Elsevier B.V. All rights reserved.
8 E.C. da Silva et al. / Catalysis Communications 78 (2016) 7–10

from Sigma-Aldrich (MO, USA). Soybean oil (commercial grade) was Table 1
supplied by Bunge Alimentos (Brazil), coconut oil was purchased from Nomenclature and formulas of the catalysts.

Sococo (Brazil), and castor oil was acquired at a local market. All of Nomenclature General formulaa
the oils were used as received and were analyzed (viscosity and acidity) a Mono-n-butyl tin hydroxide
using standard methods described by the Association of Official Analyt- oxide
ical Chemists (AOAC, reapproved 1997) [21]. The fatty acid composi-
b Di-n-butyl-oxo-stannane
tions of the oils were determined via gas chromatography (GC) using
the official AOCS methods Ce 1-62 and Ce 2-26 (A.O.C.S., 1998) [21].
c Di-n-butyl tin dilaurate

2.2. Hydrolysis reactions

Hydrolysis reactions were performed in a 100-mL batch stainless

steel reactor coupled to a manometer, temperature probe and a me-
chanic stirrer operated at 1000 rpm. All of the reactions were carried
out using an oil:water:catalyst molar ratio of 1:x:y (x = 3, 6, 12, or 36
and y = 0.01, 0.02 or 0.03) at different temperatures (140, 160 or d Tin oxide SnO2
180 °C) for 0.25 to 8 h. The endogenous pressure, observed during the a
Squematic representation of the chemical species because these species are more
experiments, was ca. of 3 bar at 140 °C, 4 bar at 160 °C and 5 bar at likely to displaying intramolecular coordination or intermolecular association lead-
180 °C. After the appropriate reaction time, the obtained product was ing to dimers, oligomers, or polymers [13].

analyzed by titration, and the reaction conversion was calculated

based on the increase in the acid index for the products in relation to
acid index for the initial content of TGs, according to the AOCS Cd3d63
standard method [21]. For that, 0.5 g of sample was diluted in 10 mL
3.1. Influence of the catalyst nature, temperature and reaction time
of diethyl ether:ethanol and alcohol (2:1 v:v) solution and 0.2 mL of
phenolphthalein was added. This solution was titrate, under stirring,
The catalysts a, b, c and d were used in the hydrolysis of soybean oil
with the potassium hydroxide solution (0.01 mol.L−1) until pink
at 140 and 160 °C and at several reaction times, employing the molar
color, which must persist for at least 10 s. This determination is rune
ratio of oil:water:catalyst of 1:24:0.01. Fig. 1 presents the results for
in triplicate.
both temperatures in comparison with the reaction in the absence of
For the samples characterized by HPLC, a Shimadzu chromatograph
catalyst.
(CTO-20A) equipped with a UV detector (λ = 205 nm) and a reverse
The catalyst d showed the same behavior observed in the reactions
phase column (Shim-pack VP-ODS C18, 250 mm × 4.6 mm, 5 mm) at
run in the absence of catalyst, indicating that this compound did not
40 °C. The injection volume was 10 μL, and the flow rate was 1 mLmin−1-
act as a catalyst in this type of reaction, regardless of the temperature
using methanol and a 2-propanol:n-hexane solution (5:4 v:v) with a
or reaction time considered.
composition gradient in the mobile phase: 100% methanol at 0 min,
When the catalysts a, b and c were used, significant conversions
50% methanol and 50% of the 2-propanol:n-hexane solution at 10 min,
were attained, especially at 160 °C, demonstrating the positive influence
and isocratic elution for 10 min. The chromatograms were analyzed
of the temperature on the reaction conversion of TGs to FAs. When an-
using LabSolutions software.
alyzing the evolution in relation to the reaction time, the conversions
In order to investigate the Sn concentration presents in the reaction
were enhanced with as the reaction time increased.
medium in the case of solid catalysts, reactions using a and b were in-
The best conversion results were obtained with catalyst a at 160 °C,
vestigated using oil:water:catalyst molar ratio of 1:24:0.01 at 180 °C
which could be justified by the better access to the reaction site, accord-
for 4 h. After the appropriate reaction time, the oil and water phase
ing to the mechanism mentioned in the introduction. The results
were separated, and filtered through a 0.45 μm Millipore filter. After
observed in the presence of b and c were comparable but lower than
this procedure, the Sn content (mg/kg) was investigated in the two
that obtained for a due the presence of two butyl substituents (in b)
phases, by radial view Inductively Coupled Plasma Optical Emission
and two butyl and two carboxylate substituents (in c).
Spectrometry (ICP OES) using Spectro Arcos equipment.
Additionally, since the catalysts a and b are solids, the Sn content
Catalysts a and b were characterized using: (i) TM3000 TableTop
(mg/kg) was investigated on the filtrated oil and water phase, after re-
SEM with a magnification of 1000× and SwiftED 3000 X-Ray software
action (oil:water:catalyst molar ratio of 1:24:0.01 at 180 °C for 4 h).
(Oxford Instruments, USA); (ii) FTIR spectra for evaluated pyridine
Considering the amount of catalyst employed, if the whole catalyst is
adsorption at 100 °C, 150 °C and 200 °C on a Shimadzu infrared spectro-
in the aqueous phase the concentration of Sn expected is 2749 mg/kg.
photometer (model Prestige-21) and (iii) isotherms were investigated
On the contrary, if the totality of catalyst stays in the oily phase, the
(specific surface areas and pores diameter) at 77 K with a Nova 2200e
estimated Sn concentration will be 1361 mg/kg. In the aqueous phase,
instrument (Quantachrome Instruments).
it was detected b0.010 mg/kg of Sn, for both catalysts. In the oily
phase, 40 mg/kg and 69 mg/kg of Sn were detected for a and b, respec-
3. Results and discussion tively. From these results, it is possible to assume that the catalysts a and
b, are really solids with low solubility, since just a small amount of
In this work, catalysts based on tin(IV) compounds, presented in catalyst can be found, either in the aqueous or in the oily phases.
Table 1, were employed as catalysts in the hydrolysis of TGs. The catalyt- Furthermore, a reaction at oil:water molar ratio of 1:24 at 180 °C
ic performance of these chemical species and the influence of several (1 to 4 h) was studied, using the amount of Sn detected for a
reaction parameters in the reaction yields were investigated. (40 mg/kg), after the procedure described above (see Figure S3 in
Three different natural sources of TG were used (soybean, coconut Supporting Information). The results, if compared those obtained at
and castor oils), and these three oils exhibited different characteristics oil:water:catalyst molar ratio of 1:24:0.01, show proportional conver-
in terms of their FA compositions (Table S1 in Supporting Information). sions with the catalyst amount. It is important to mention that it is not
Complementarily, the viscosity and acidity were also determined possible dismiss that this Sn amount, detected by ICP in the oily phase,
(Table S2 in Supporting Information). The acidity values were suitable can be associated to solid catalyst, since the minor particles of a and b
for use in hydrolysis. can permeate through the filter.
 E.C. da Silva et al. / Catalysis Communications 78 (2016) 7–10 9

Fig. 1. Conversion (%) of TGs by hydrolysis in the presence of a, b, c, d and without catalyst (oil:water:catalyst molar ratio = 1:24:0.01; temperature = 140, 160°or 180 °C).

Textural analysis and FTIR spectra with adsorbed for pyridine
technique were used to characterize solid catalysts a and b. In both
cases, it was not possible to obtain suitable results, since textural char-
acterization shows that no N2 adsorption occurred and no absorption
bands due to pyridine absorption were detected in FTIR spectra, sug-
gesting that these materials present small surface area and low porosity.
The images of SEM also confirm this trend (see Figure S4 in Supporting
Information).
3.2. Influence of the water amount
The hydrolysis of TGs due to the nature of the reactants involved (a
large-chain organic molecule and water) occurred in the interface
formed by the oil–water emulsion [22]. For this reason, it was important
to investigate the influence of the proportion of these reactants. To this
end, four different amounts of water were evaluated (oil:water:catalyst:
1:24:0.01, 1:3:0.01, 1:6:0.01, 1:12:0.01 or 1:36:0.01) at 160 and 180 °C

Fig. 2. Conversion (%) of TGs by hydrolysis in the presence of a using soybean, coconut and castor oils (oil:water:catalyst molar ratio = 1:x:0.01 (x = 3, 6, 12); temperature = 160 °C).
10 E.C. da Silva et al. / Catalysis Communications 78 (2016) 7–10

using a (see Figure S2 in Supporting Information). Through the analysis Table 2

of results obtained, it was possible to assume that the reactions involv- Conversion (determined by HPLC and titration) of TGs to DGs, MGs and FAs (%) by hy-
drolysis (oil:water:catalyst b molar ratio = 1:6:0.01; temperature = 160 °C), using
ing less amount of water were distinguished from the others in terms of soybean oil.
conversion, clearly indicating that the water content had a significant
influence on the reaction course. It is important to mention that, at Amount (%, HPLC)
Time (h) AG (titration)
160 °C, when employing more than 12 moles of water, a tendency for TG DG MG + AG
the conversion values to stabilize was detected. 0 93.9 5.3 0.8 –
0.5 72.4 19.0 8.6 1.7
3.3. Influence of the TG nature 1 65.0 25.0 10.0 4.4
2 46.0 27.9 26.0 12.7
3 36.0 31.6 32.5 24.4
As already mentioned, three oils exhibiting different FA composi- 4 25.4 25.3 49.3 34.8
tions (see Table S1) were submitted to hydrolysis at 160 °C using a as 5 18.8 24.5 56.7 44.3
the catalyst and employing oil:water:catalyst molar ratios of 1:3:0.01, 6 15.9 23.6 60.5 51.4
1:6:0.01, and 1:12:0.01 (Fig. 2). 7 13.6 22.8 63.7 57.0
8 10.6 21.1 68.3 64.9
The results showed a tendency of the conversion to FAs:
coconut N soybean N castor oils. It was possible to suggest that this be-
havior could be related to mass transfer effects on the reaction medium
that influenced the emulsion formation. The viscosity values increased
4. Conclusions
in the following order: coconut oil b soybean oil ≪ castor oil (see
Table S2), which supported the observed tendency. The viscosity values
The tin(IV)-based catalysts investigated in this study were very
displayed by the coconut and soybean oils, despite their different com-
promising in vegetable oil hydrolysis reactions, principally the a com-
positions, could be related to the large content of carboxylate moieties
plex. The augmentation of the temperature, reaction time and amount
based on saturated short chains for coconut oil (82.0% of saturated
of catalyst had a positive influence on the conversion of TGs to FAs. It
chains, especially C8 and C14 ones) and the high degree of unsaturation
is important to highlight that the catalysts studied here showed excel-
for soybean oil. Castor oil was also considered relatively unsaturated but
lent results at mild temperatures compared with the severe conditions
was based mainly on ricinoleic acid (88.6%), which contains an OH
normally employed in industrial processes. Moreover, the relative
group at C12 (that leads to the formation of intermolecular hydrogen
catalyst content for high conversion of TGs was quite low, at least for
bonds).
catalyst a (ca. 80% of conversion using 0.5 wt% based on the weight of
A lower viscosity resulted is better reaction dynamics, leading to en-
the oil).
hanced conversions. From this perspective, the lower yields observed
using castor oil could be explained by its high viscosity, which
prevented good fluidity of the reaction mixture. Moreover, another Appendix A. Supplementary data
factor that must be considered is based on the interaction of water
molecules with the hydroxyl groups of ricinoleic acid, as well as inter- Supplementary data to this article can be found online at http://dx.
molecular hydrogen bonds. doi.org/10.1016/j.catcom.2016.01.032.

3.4. Influence of the catalyst amount References

The influence of the catalyst amount was also evaluated using cata-
lyst a and soybean oil at 160 °C and several reaction times. The
oil:water:catalyst molar ratios employed were 1:6:0.01, 1:6:0.02 and
1:6:0.03 (see Figure S5 in Supporting Information). As expected, the
conversion was enhanced with augmentation of the amount of catalyst
present in the reaction medium, ever since there are more catalytic sites
available.
3.5. Evaluation of the amounts of TGs, FAs, DGs and MGs during the
hydrolysis reaction
In more detail, for soybean oil hydrolysis using catalyst a at 160 °C
and an oil:water:catalyst molar ratio of 1:6:0.01, the contents of TG,
FA, DG and MG present in the reaction medium were determined be-
tween 0 and 8 h and are depicted in Figure S6 (Supporting Information)
and Table 2.
As already mentioned, the hydrolysis of TGs is based and three
consecutive steps, those TGs are converted to DGs, then to MGs and
then finally to glycerol, resulting in three FA molecules (individually
formed in each step) [2,5–7]. This trend was also confirmed, during
the reaction progress in the presence of catalyst b, a Lewis acid species
(Table 2).
In addition, the results obtained by titration (Table 2) were comple-
mentary to those obtained by HPLC, since by HPLC was not possible to
distinguish the signals related to the FAs and MGs, due to the co-
elution. Regardless, when high conversion values were attained, the
results were very similar (at 8 h, 68.3% by HPLC and 64.9% by titration),
showing the high conversion of MG into FA at a long reaction time.
[1] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[2] P.A.Z. Suarez, S.M.P. Meneghetti, M.R. Meneghetti, C.R. Wolf, Quim. Nova 30 (2007)
667–676.
[3] H. Luo, K. Xue, W. Fan, C. Li, G. Nan, Z. Li, Ind. Eng. Chem. Res. 53 (2014)
11653–11658.
[4] J.K. Satyarthi, Appl. Catal., A 391 (2011) 427–435.
[5] S. Pinto, F.M. Lanças, J. Braz. Chem. Soc. 17 (2006) 85–89.
[6] W.C. Wang, N. Thapaliya, A. Campos, L.F. Stikeleather, W.L. Roberts, Fuel 95 (2012)
622–629.
[7] P.H.L. Moquin, F. Temelli, Fluids 45 (2008) 94–101.
[8] T.A. Patil, T.S. Raghunathan, H.S. Shankar, Ind. Eng. Chem. Res. 27 (5) (1988)
735–739.
[9] K. Ngaosuwan, E. Lotero, K. Suwannakarn, J.G. Goodwin Jr., P. Praserthdam, Ind. Eng.
Chem. Res. 48 (10) (2009) 4757–4767.
[10] K. Ramani, R. Boopathy, A.B. Mandal, G. Sekaran, Catal. Commun. 14 (2011) 82–88.
[11] C.J. Yow, K.Y. Liew, J. Am. Oil Chem. Soc. 76 (1999) 529–533.
[12] M.B. Alves, F.C. Medeiros, P.A.Z. Suarez, Ind. Eng. Chem. Res. 49 (2010) 7176–7718.
[13] M.R. Meneghetti, S.M.P. Meneghetti, Catal. Sci. Technol. 5 (2015) 765–771.
[14] J.P. LatereDwan'Isa, A.K. Mohanty, M. Misra, L.T. Drzal, J. Mater. Sci. 39 (2004)
2081–2087.
[15] M.B. Siddaramaiah, J. Mater. Sci. 39 (2004) 4615–4623.
[16] H. Lee, S.J. Kima, B.S. Ahna, W.K. Lee, H.S. Kima, Catal. Today 87 (2003) 139–144.
[17] J. Dupont, P.A.Z. Suarez, M.R. Meneghetti, S.M.P. Meneghetti, Energy Environ. Sci. 2
(2009) 1258–1265.
[18] D.A.C. Ferreira, M.R. Meneghetti, S.M.P. Meneghetti, C.R. Wolf, Appl. Catal., A 317
(2007) 58–61.
[19] D.R. Mendonça, J.P.V. da Silva, R.M. de Almeida, C.R. Wolf, M.R. Meneghetti, S.M.P.
Meneghetti, Appl. Catal., A 365 (2009) 105–109.
[20] Y.C. Brito, D.A.C. Ferreira, D.M.A. Fragoso, P.R. Mendes, C.M.J. de Oliveira, M.R.
Meneghetti, P.A.Z. Suarez, S.M.P. Meneghetti, Appl. Catal., A 443-444 (2012)
202–206.
[21] A.O.C.S, in: D. Firestone (Ed.), Official Methods and Recommended Practices of the
American Oil Chemists' Society, fifth ed.American Oil Chemists' Society, Champaign,
IL, 1998.
[22] L.L. Schramm, Emulsions: fundamentals and applications in the petroleum industry,
in: Petroleum Recovery Institute (Ed.), Advances in Chemistry Series 231, American
Chemical Society, Washington DC, 1992.