Oxidation of olefins with H2O2 catalyzed by gallium(III) nitrate and aluminum(III) nitrate in solution

Article abstract of DOI:10.1016/j.molcata.2016.03.004

Soluble gallium and aluminum nitrates (simple salts of non-transition metals) are good catalysts for the epoxidation of olefins (cyclooctene, dec-1-ene) including terpenes (carvone, limonene) with hydrogen peroxide in ethyl acetate or tetrahydrofurane (THF). Typically, the gallium salt is more efficient in comparison with the aluminum derivative. Products are formed in yields up to 93%, turnover numbers (TONs) attained 40. Addition of trifluoroacetic acid or pyrazine-2-carboxylic acid (PCA) accelerates the reaction and improves the yield. In striking contrast, added 2,2′-bipyridine or phenanthroline dramatically inhibit the oxidation.

Full text of DOI:10.1016/j.molcata.2016.03.004

G Model
MOLCAA-9803; No. of Pages5
ARTICLE IN PRESS
Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Oxidation of olefins with H2O2 catalyzed by gallium(III) nitrate and
aluminum(III) nitrate in solution
Dalmo Mandellia,∗∗, Yuriy N. Kozlovb,c, Cezar A.R. da Silvaa, Wagner A. Carvalhoa,
Paolo P. Pescarmonad, Daniele de A. Cellaa, Polyana T. de Paivaa, Georgiy B. Shul’pinb,c,∗
a Center of Natural and Human Sciences, Federal University of ABC (UFABC), Santa Adélia Street, 166, Bangu, Santo André, SP 09210-170, Brazil
b Semenov Institute of Chemical Physics, Russian Academy of Sciences, Ulitsa Kosygina, dom 4, Moscow 119991, Russia
c Plekhanov Russian University of Economics, Stremyannyi pereulok, dom 36, Moscow 117997 Russia
d Chemical Engineering Department, University of Groningen, Nijenborgh 4, 9747 AG, Groningen, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Soluble gallium and aluminum nitrates (simple salts of non-transition metals) are good catalysts for the
epoxidation of olefins (cyclooctene, dec-1-ene) including terpenes (carvone, limonene) with hydrogen
peroxide in ethyl acetate or tetrahydrofurane (THF). Typically, the gallium salt is more efficient in com-
parison with the aluminum derivative. Products are formed in yields up to 93%, turnover numbers (TONs)
attained 40. Addition of trifluoroacetic acid or pyrazine-2-carboxylic acid (PCA) accelerates the reaction
and improves the yield. In striking contrast, added 2,2ꢀ-bipyridine or phenanthroline dramatically inhibit
Received 22 October 2015
Received in revised form 29 February 2016
Accepted 1 March 2016
Available online xxx
This article is dedicated to Prof. Dr. Ulf
Schuchardt, our good colleague and friend,
on the occasion of his 70th birthday.
the oxidation.
© 2016 Published by Elsevier B.V.
Keywords:
Carvone
Epoxidation
Limonene
Olefins
Terpenes
1. Introduction
lium and some other non-transitionmetals leads to the formation of
hydroxyl radicals which induce the oxidation of alkanes [13,15–20].
Valorization of natural products including biomass is a very
attractive task of contemporary chemistry [1]. Many natural
compounds, for example, terpenes, contain double bonds and
their epoxidation leads to the preparation of valuable product.
Usually epoxides are prepared by metal-catalyzed oxygenation
of compounds containing double bonds [2]. In recent decades,
heterogeneous [3–11] and soluble [12–14] derivatives of non-
transition elements aluminum and gallium have been reported
by Schuchardt, Sheldon, Mandelli, Rinaldi and Pescarmona to cat-
alyze olefin epoxidations with hydrogen peroxide. The interaction
of hydrogen peroxide with soluble derivatives of aluminum, gal-
Goldsmith and co-workers described epoxidation of olefins with
peroxyacetic acid catalyzed by gallium complexes with N-donor
ligands [21–23]. It is noteworthy that H2O2 was not effective in
that reaction.
In the present work we have discovered, for the first time, that
gallium nitrate in ethyl acetate (as well as in tetrahydrofuran, THF)
efficiently catalyzes epoxidation of certain olefins with H2O2. Alu-
minum nitrate [12,13,15] was used for comparison. A remarkable
accelerating effect of pyrazine-2-carboxylic acid (PCA) was found
in the present study. Finally, we were able to epoxidize dec-1-ene
by the Ga/PCA system. Terminal olefins can be epoxidized with
difficulty by usual reagents.
∗
2. Results and discussion
Corresponding author at: Semenov Institute of Chemical Physics, Russian
Academy of Sciences, Ulitsa Kosygina, dom 4, Moscow 119991, Russia.
∗∗
Corresponding author at: Center of Natural and Human Sciences, Federal Uni-
2.1. Oxidation of cyclooctene and dec-1-ene
versity of ABC (UFABC), Santa Adélia Street, 166, Bangu, Santo André, SP 09210-170,
Brazil.
In the first part of this work we carried out a detailed study
of cyclooctene oxidation with H2O2 catalyzed by nitrates of gal-
E-mail addresses: dalmo.mandelli@ufabc.edu.br (D. Mandelli),
Shulpin@chph.ras.ru, gbsh@mail.ru (G.B. Shul’pin).
http://dx.doi.org/10.1016/j.molcata.2016.03.004
1381-1169/© 2016 Published by Elsevier B.V.
Please cite this article in press as: D. Mandelli, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.03.004

G Model
MOLCAA-9803; No. of Pages5
ARTICLE IN PRESS
2
D. Mandelli et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
A
0.5
1
0.4
0.3
0.2
0.1
0
2
0
0.5
1.0
1.5
[cyclooctene]0 (M)
1
B
0.5
0.4
0.3
0.2
2
0.1
0
0.5
104[catalyst]0 (mol)
1.5
0
1.0
Fig. 2. Epoxidation of cyclooctene with H2O2 (4.0 M) containing 3.2 M H2O in the
presence of catalyst: Ga(NO3)3 (curves 1), Al(NO3)3 (curves 2). Graph A: dependence
of initial epoxidation rate W0 on initial concentration of cycloctene. Graph B: depen-
dence of initial epoxidation rate W0 on initial concentration of catalyst. Conditions:
solvent ethyl acetate, 80 ◦C.
decomposition of H2O2. Approaching a plateau is in accord with
an assumption that catalytically active oxidizing species reacts not
only with cyclooctene but also competitively with solvent.
Fig. 1. Graph A. Accumulation of epoxide in the oxidation of cyclooctene (1.0 M)
with H2O2 (4.0 M) containing 3.2 M H2O in the presence of catalyst (0.02 M):
Ga(NO3)3 (curve 1), Al(NO3)3 (curve 2) and without catalyst (curve 3). Conditions:
solvent ethyl acetate, 80 ◦C. Graph B. Yield of epoxide (subtracting the data for the
experiments in the absence of Ga nitrate) for the ratios H2O2/cyclooctene 1, 2, 3, 4
and 5 in the reaction catalyzed by Ga(NO3)3.
The rate dependence of the oxygenation is first order with
respect to the initial hydrogen peroxide at [H2O2]0 < 3 M for both
catalysts. Dependences of initial rates on concentrations of the cat-
alysts are shown in Fig. 2B. In these cases gallium derivative is also
more active. This observation is in accordance with conclusions
made on the basis of the DFT calculations [16]. Indeed, a theoretical
study demonstrated that the acidity of the gallium and their com-
plexes in aqueous solutions is the highest when compared with the
other metals of Group III of the periodic table. It was also argued
that the activity of catalysts [M(H2O)n]3+ in the alkane oxidation
with H2O2 depends on the stability of the complex and activation
of hydrogen peroxide. According to the calculations, this activity
in another pathway generating hydroxyl radicals follows the series
Al ≈ La <Y ≈ In < Sc < Ga [16]. The dependence of initial epoxidation
rate on the amount of catalyst (N, mol) correlates with the cata-
lyst solubility under the studied conditions. Thus, at N < m0 where
m0 = 0.5 × 10−4 mol (that is at catalyst concentration < 10−2 M) cat-
alyst is completely dissolved in the reaction mixture and the
solution is homogeneous. However, at N > 0.5 × 10−4 mol a portion
of introduced catalyst (N − m0) is not dissolved and some turbidity
of the reaction mixture can be noticed. Assuming that reaction rate
is proportional to catalyst concentration in homogeneous solution
and that solid catalyst is inactive in the oxidation, that is
lium or aluminum which gives rise to the formation of the epoxide
and diol (Scheme 1). It should be noted that diol was not detected
by GC for the reaction time < 8 h. Diol in a small concentration
was found only at time 24 h. For the cases of both catalysts the
products were formed predominantly during the first hours of the
reaction. Total yield of the epoxide was not changed significantly in
aliquots removed at 6, 8 and 24 h reaction. In the case of Ga(NO3)3
these values were 71, 77 and 72%, respectively. Some reduction of
yield at 24 h reaction time can be due to the formation of over-
oxidation products. We measured also the efficiency of hydrogen
peroxide usage. This efficiency was calculated taking into account
the amount (mol) of remaining hydrogen peroxide in the reac-
tion solution after 24 h, the amount of H2O2 initially added to the
medium and amount of formed oxidized product. In the case of the
reaction catalyzed Ga3+ and Al3+, efficiency for hydrogen peroxide
was 76 and 75%, respectively.
Accumulation of epoxide in the oxidation of cyclooctene is
shown in Fig. 1. It can be seen in Fig. 1, Graph B that after 1.5 h
some acceleration of the process takes place. It can be assumed
that after this time (1.5 h) the system begins to generate hydroxyl
radicals. This was totally confirmed by the epoxidation with car-
vone and it is one of the most interesting features of the work (see
Supplementary data, Appendix A).
N
W0 = ␣ atN ≤ m0and
V
m0
Fig. 2A demonstrates dependence of the initial reaction rate
on initial concentration of cyclooctene. It can be clearly seen that
the gallium derivative is noticeably more active than the cor-
responding aluminum salt. The mode of dependence of W0 on
initial cyclooctene concentration (approaching the limited value
when [cyclooctene]0 growths) indicates that the olefin is epoxi-
dized by an intermediate species which is produced in Ga-catalyzed
W0 = ␣
atN > m0
V
where ␣ are effective pseudo-first order rate constants and V is the
reaction solution volume, we will see that given above explanation
is in full agreement with experimental results presented in Fig. 2B.
We studied effects of various additives on the yields of products.
Figs. S1 and S1a show that addition of acetic acid leads to some
Please cite this article in press as: D. Mandelli, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.03.004

G Model
MOLCAA-9803; No. of Pages5
ARTICLE IN PRESS
D. Mandelli et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
3
A
10:1
4:1
100
1:1
50
0:1
0
0
1
2
4
3
B
10:1
100
1:1
4:1
50
Fig. 5. Accumulation of epoxides in the oxidation of carvone (1.0 M) with H2O2
(4.0 M) containing 3.2 M H2O in the presence of catalyst (0.02 M): Ga(NO3)3 and co-
catalyst PCA in the molar ratio PCA:catalyst 4:1. Conditions: solvent ethyl acetate,
80 ◦C.
0:1
0
0
1
2
Time (h)
4
3
2.2. Oxidation of carvone
Fig. 3. Accumulation of epoxide in the oxidation of cyclooctene (1.0 M) with H2O2
(4.0 M) containing 3.2 M H2O in the presence of catalyst (0.02 M): Ga(NO3)3 (Graph
A), Al(NO3)3 (Graph B) and co-catalyst PCA in the molar ratio PCA:catalyst 0:1, 1:1,
4:1 and 10:1. Conditions: solvent ethyl acetate, 80 ◦C.
We have found the similarity in catalytic behavior of Al3+ and
Ga3+ in epoxidation of (R)−carvone. This study was carried out sim-
ilarly to that performed by Rinaldi, Schuchardt and co-workes [12].
In our case THF was replaced by ethyl acetate because efficiency of
the latter was higher for the epoxidation of cyclooctene. Carvone
conversion. after 3 h at 80 ◦C in the case of both metals was ca.
60% (Fig. S6a), but with a mass balance of approximately 65% (Fig.
S6b), which indicates that some other volatile products are formed.
Internal epoxide was mainly formed in the beginning of the reac-
tion (Fig. S6c). Both systems, Ga and Al, are very fast in generating
nucleophilic species OOH which attacks the double internal bond
situated closer to the carbonyl group and therefore bearing lower
electron density. Thus, within only 0.5 h of the reaction almost all
internal epoxide was formed for both catalysts (Fig. S6c). The metals
have similar activities and form the same final amount of internal
epoxide (13%, after 3 h of the reaction). However, the formation
of the species. OOH is faster in the case of Ga3+: after only 15 min
there was a 12% yield of the internal epoxide with this metal (with
a total yield of 13% i.e. 92% of total yield) whereas yield of the inter-
nal epoxide in the case of Al was 9% (and total 13%) or 69% of total
yield. The use of carvone as molecule model for understanding the
mechanism was useful in one experiment made in the presence of
PCA. The results are shown in Fig. 5. PCA favored the formation of
external epoxide, probably improving the generation of hydroxyl
radicals, the main species responsible for the attack at the external
double bond. On the other hand, the protons from carboxylic acid
(pKa = 2.9) could react the OOH− anions, species responsible for the
attack at the internal bond depressing the formation of the inter-
nal epoxide. This also confirms the mechanism proposed by Rinaldi
et al [12].
Fig. 4. Accumulation of epoxide in the oxidation of dec-1-ene (1.0 M) with H2O2
(4.0 M) containing 3.2 M H2O in the absence of catalyst (violet curve) in the presence
of catalyst (0.02 M): Ga(NO3)3 (green curve) and co-catalyst + PCA in the molar ratio
PCA:catalyst 4:1 (blue curve). Conditions: solvent ethyl acetate, 80 ◦C.
increase of the epoxide yield. The epoxidation is not efficient in
the absence of the gallium salt although acetic acid is present in
the solution (Fig. S1a). Trifluoroacetic acid is also co-catalyst as Fig.
S2 demonstrates. The best result has been obtained for Ga(NO3)3
and ten-fold excess of CF3COOH: yield of the epoxide was 70% after
45 min. A strong accelerating effect has been found for pyrazine-2-
carboxylic acid, PCA [24], in the cases of both gallium and aluminum
salts (Fig. 3A and B, respectively). The epoxide yield attained 93%
in the case of gallium salt after 4 h. Fig. S3 shows that the order of
the reaction in PCA is >1 in the case of gallium salt. In contrast to
PCA, diamines such as 2,2ꢀ-dipyridyl and phenanthroline exhibited
inhibiting effects (Figs. S4,S5).
2.3. Oxidation of limonene
Like carvone, terpene limonene also contains two double bonds:
internal and external ones (Scheme 1). Table 1 summarizes the
results on oxidation of limonene with both catalysts in two sol-
vents. It can be seen that internal, external epoxides and diepoxides
are produced in approximately equal yields. Yields of isomeric diols
are 2–3 times higher than yield of each this product. Total yield of
epoxides is lower than in the case of carvone containing electron-
withdrawing carbonyl group.
Fig. 4 demonstrates the strong effect of added PCA on initial reac-
tion rate in the dec-1-ene epoxidation. Aluminum nitrate exhibited
much less activity under these conditions.
Please cite this article in press as: D. Mandelli, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.03.004

G Model
MOLCAA-9803; No. of Pages5
ARTICLE IN PRESS
4
D. Mandelli et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
OH
+
O
OH
Cyclooctene
O
Epoxide
O
Diol
+
O
O
O
External epoxide
R-Carvone
Internal epoxide
OH
O
O
OH
+
+
+
O
O
Limonene
Internal epoxide External epoxide
Diepoxide
Diol
Scheme 1. Oxidation of cyclooctene, carvone and limonene.
Table 1
Oxidation of limonene with H2O2 catalyzed by gallium and aluminum nitrates.
Entry
Solvent
Catalyst
Internal epoxide
External epoxide
Diepoxide
Diol
19
13
19
9
1
2
3
4
Ethyl acetate
Ethyl acetate
THF
Ga(NO3)3
Al(NO3)3
Ga(NO3)3
Al(NO3)3
7
5
4
3
5
4
5
4
7
7
6
5
THF
Yield (%) is given based on starting limonene. Conditions. Limonene, 1.0 M; H2O2, 4.0 M; catalyst, 0.02 M. 80 ◦C, 3 h.
3. Conclusions
concentrations, and CH3NO2 as internal standard. The products
formed from natural compounds were identified by using a gas
chromatography coupled to a mass spectrometer (GC–MS) Shi-
madzu 2010-plus. The turbidity of the system was monitored using
a turbidimeter PoliControl AP2000.
In olefin epoxidation with hydrogen peroxide gallium nitrate is
more efficient in comparison with aluminum derivative. Additives
of trifluoroacetic acid or pyrazine-2-carboxylic acid (PCA) accel-
erate the reaction. Noteworthy, that in striking contrast, added
2,2ꢀ-bipyridine or phenanthroline dramatically inhibit the oxida-
tion.
Acknowledgements
The authors thank the “Science without Borders Program, Brazil-
Russia”, CAPES (grant A017-2013) and the Russian Foundation for
Basic Research (grant 16-03-00254) for support. D.M. is grateful to
CNPq (grants 303828/2010-2, 311585/2013-2, 472130/2012-9 and
490043/2013-5) for support. G.B.S. expresses his gratitude to CAPES
and the Center of Natural and Human Sciences, Federal University
of ABC (UFABC), Santo André, São Paulo, for making it possible for
him to stay at the UFABC, as an invited Professor and to perform a
part of the present work. We are also thankful to Ms. Andressa V.
Müller for her help with the experiments.
4. Experimental
The experiments on olefin oxidations were carried out in ace-
tonitrile in thermostated Pyrex cylindrical vessels (total volume
10 mL) with vigorous stirring. In a typical experiment, initially,
a portion of the aqueous solution of H2O2 (aqueous 70% hydro-
gen peroxide, Solvay, solution was used as received) was added
to the solution of a substrate, catalyst 1 (gallium nitrate) in ace-
tonitrile. Catalyst 1 was used in the form of stock solutions in
H2O2. (CAUTION. The combination of air or molecular oxygen and
H2O2 with organic compounds at elevated temperatures may be
explosive!). After certain periods, the reaction mixture was
analyzed by GC using Shimadzu-HP 2010 gas chromato-
graph equipped with a polyethylene glycol column (Innowax,
25 m × 0.2 mm × 0.4 ␮m) coupled to the flame ionization detector
and an automatic injector measuring concentrations of prod-
ucts. Authentic samples of all oxygenated products were used to
attribute the peaks in chromatograms (comparison of retention
times was carried out for different regimes of GC-analysis). The
quantification of products was accomplished by constructing cal-
ibration curves for the products, by using standards of known
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2016.03.
004.
References
[1] (a) C.O. Tuck, E. Pérez, I.T. Horváth, R.A. Sheldon, M. Poliakoff, Science 337
(2012) 695–699;
Please cite this article in press as: D. Mandelli, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.03.004

G Model
MOLCAA-9803; No. of Pages5
ARTICLE IN PRESS
D. Mandelli et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
5
(b) G. Centi, R.A. van Santen (Eds.), Catalysis for Renewables, WILEY-VCH
Verlag GmbH & Co., Weinheim, 2007.
[2] (a) E.T. Denisov, I.B. Afanas’ev, Oxidation and Antioxidants in Organic
Chemistry and Biology, Taylor& Francis Group, Boca Raton, 2005;
(b) S.T. Oyama (Ed.), Mechanisms in Homogeneous and Heterogeneous
Epoxidation, Elsevier, 2008;
(c) B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2473.
[3] R. Rinaldi, U. Schuchardt, J. Catal. 227 (2004) 109–116.
[4] P.P. Pescarmona, P.A. Jacobs, Catal. Today 137 (2008) 52–60.
[5] G. Staica, M. Santiago, P.A. Jacobs, J. Pérez-Ramírez, P.P. Pescarmona, Appl.
Catal. A: Gen. 371 (2009) 43–53.
[14] C. Bour, V. Gandon, Coord. Chem. Rev. 279 (2014) 43–57.
[15] D. Mandelli, K.C. Chiacchio, Y.N. Kozlov, G.B. Shul’pin, Tetrahedron Lett. 49
(2008) 6693–6697.
[16] A.S. Novikov, M.L. Kuznetsov, A.J.L. Pombeiro, N.A. Bokach, G.B. Shul’pin, ACS
Catal. 3 (2013) 1195–1208.
[17] M.L. Kuznetsov, F.A. Teixeira, N.A. Bokach, A.J.L. Pombeiro, G.B. Shul’pin, J.
Catal. 313 (2014) 135–148.
[18] B.G.M. Rocha, M.L. Kuznetsov, Y.N. Kozlov, A.J.L. Pombeiro, G.B. Shul’pin, Catal.
Sci. Technol. 5 (2015) 2174–2187.
[19] M.L. Kuznetsov, B.G.M. Rocha, A.J.L. Pombeiro, G.B. Shul’pin, ACS Catal. 5
(2015) 3823–3835.
[6] C. Aprile, E. Gobechiya, J.A. Martens, P.P. Pescarmona, Chem. Commun. 46
(2010) 7712–7714.
[7] E.E. Macias, C.A. Deshmane, J.B. Jasinski, M.A. Carreon, P. Ratnasamy, Catal.
Commun. 12 (2011) 644–650.
[20] A.V. Müller, C.A.R. da Silva, W.A. Carvalho, P.P. Pescarmona, G.B. Shul’pin, D.
Mandelli, Epoxidation of cyclooctene catalyzed by Ga(NO3)3, in: Materials of
4th International IUPAC Conference on Green Chemistry, Foz do Iguac¸ u,
Brasil, 2012.
[8] C.A. Deshmane, J.B. Jasinski, P. Ratnasamy, M.A. Carreon, Catal. Commun. 15
(2011) 46–51.
[9] W. Lueangchaichaweng, L. Li, Q.-Y. Wang, B.-L. Su, C. Aprile, P.P. Pescarmona,
Catal. Today 203 (2013) 66–75.
[10] X. Collard, L. Li, W. Lueangchaichaweng, A. Bertrand, C. Aprile, P.P.
Pescarmona, Catal. Today 235 (2014) 184–192.
[11] (a) S. Rahman, C. Santra, R. Kumar, J. Bahadur, A. Sultana, R. Schweins, D. Sen,
S. Maity, S. Mazumdar, B. Chowdhury, Appl. Catal. A: Gen. 482 (2014) 61–68;
(b) A.J. Bonon, Y.N. Kozlov, J.O. Bahú, R. Maciel Filho, D. Mandelli, G.B.
Shul’pin, J. Catal. 319 (2014) 71–86.
[21] W. Jiang, J.D. Gordon, C.R. Goldsmith, Inorg. Chem. 51 (2012) 2725–2727.
[22] W. Jiang, J.D. Gordon, C.R. Goldsmith, Inorg. Chem. 52 (2013) 5814–5823.
[23] (a) M.L. McKee, C.R. Goldsmith, Inorg. Chem. 53 (2014) 318–326;
(b) C.A. Koellner, N.A. Piro, W.S. Kassel, C.R. Goldsmith, C.R. Graves, Inorg.
Chem. 54 (2015) 7139–7141;
(c) F. Bronston, S. Ting, Q. Zhang, C.R. Goldsmith, Polyhedron (2016), http://dx.
doi.org/10.1016/j.poly.2015.12.040.
[24] A.M. Kirillov, G.B. Shul’pin, Coord. Chem. Rev. 257 (2013) 732–754.
[12] R. Rinaldi, H.F.N. de Oliveira, H. Schumann, U. Schuchardt, J. Mol. Catal. A:
Chem. 307 (2009) 1–8.
[13] M.L. Kuznetsov, Y.N. Kozlov, D. Mandelli, A.J.L. Pombeiro, G.B. Shul’pin, Inorg.
Chem. 50 (2011) 3996–4005.
Please cite this article in press as: D. Mandelli, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.03.004