Kinetics and mechanistic study of the Bamberger rearrangement of N-phenylhydroxylamine to 4-aminophenol in acetonitrile-trifluoroacetic acid: A substrate acid complex as para selectivity driver

Article abstract of DOI:10.1016/j.apcata.2016.02.020

The paper deals with the kinetics of Bamberger rearrangement of N-phenylhydroxylamine to 4-aminophenol in acetonitrile as a solvent catalyzed by TFA. The influence of the water is also investigated being the rearranging moiety in the mechanism. It is evident that the presence of water depresses the rate of the rearrangement suggesting the rate-determining step does not involve water as the key reagent. In acetonitrile, at lower temperature than that of reaction, we evidence the formation of a PHA-TFA complex, and its equilibrium has been measured between 288 K and 298 K. We observe also the addition of water destroys this complex but the ternary equilibrium PHA, water and TFA cannot easily measured because of the complex solvent effect on the Uv-vis and NMR signals. Starting from this evidence, we have proposed a kinetic model, which takes into account all the equilibria, and the results of the fitting gives reliable thermodynamic and kinetic parameters of the entire process. Finally, we have suggested that the PHA-TFA complex is the key intermediate, which determines the regioselectivity of the rearrangement. In fact, preliminary quantum chemistry calculation have shown that the TFA interaction with the -NHOH group causes the hindering of the ortho position, thus favoring the attack of water to the para one.

Full text of DOI:10.1016/j.apcata.2016.02.020

Applied Catalysis A: General 516 (2016) 58–69
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Kinetics and mechanistic study of the Bamberger rearrangement of
N-phenylhydroxylamine to 4-aminophenol in
acetonitrile-trifluoroacetic acid: A substrate acid complex as para
selectivity driver
∗
Nicola de Fonzo, Giuseppe Quartarone, Lucio Ronchin , Claudio Tortato, Andrea Vavasori
Department of Molecular Sciences and Nanosystems, University Ca’ Foscari of Venice, Campus Via Torino 155, 30172 Venezia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The paper deals with the kinetics of Bamberger rearrangement of N-phenylhydroxylamine to 4-
aminophenol in acetonitrile as a solvent catalyzed by TFA. The influence of the water is also investigated
being the rearranging moiety in the mechanism. It is evident that the presence of water depresses the rate
of the rearrangement suggesting the rate-determining step does not involve water as the key reagent. In
acetonitrile, at lower temperature than that of reaction, we evidence the formation of a PHA-TFA complex,
and its equilibrium has been measured between 288 K and 298 K. We observe also the addition of water
destroys this complex but the ternary equilibrium PHA, water and TFA cannot easily measured because of
the complex solvent effect on the Uv–vis and NMR signals. Starting from this evidence, we have proposed
a kinetic model, which takes into account all the equilibria, and the results of the fitting gives reliable
thermodynamic and kinetic parameters of the entire process. Finally, we have suggested that the PHA-
TFA complex is the key intermediate, which determines the regioselectivity of the rearrangement. In fact,
preliminary quantum chemistry calculation have shown that the TFA interaction with the NHOH group
causes the hindering of the ortho position, thus favoring the attack of water to the para one.
Received 12 November 2015
Received in revised form 15 February 2016
Accepted 16 February 2016
Available online 21 February 2016
Keywords:
Bamberger rearrangement
Trifluoroacetic acid
4
-aminophenol
Kinetics and mechanism
©
2016 Elsevier B.V. All rights reserved.
1
. Introduction
Continuous Stirred Tank Reactor in which the biphasic reaction
medium is used to accomplish simultaneously the Pt catalyzed
hydrogenation of nitrobenzene and the acid catalyzed Bamberger
rearrangement of the intermediate N-phenylhydroxylamine (PHA)
[4,5,8]. The presence of the aqueous solution of H SO is the key for
Aminophenols are important raw materials for several products
in the field of dyes, polymer and pharmaceutics [1–3]. For instance,
acetaminophen (N-acetyl-4-aminophenol, also known as paraceta-
mol), benzoxazoles and substitutes compounds are drugs a widely
employed as analgesic, antipyretic and intermediates whose pro-
duction is in continuous growth [4–8]. More recently, a similar
reaction path has been employed in the synthesis of heterocycles
of industrial importance starting from nitrobenzene and glycerol
via modified Skraup reaction [9].
Industrial synthesis of paracetamol is based on 4-aminophenol,
which is obtained by different routes, however, the selective hydro-
genation of nitrobenzene is likely the most convenient from both
economic and environmental point of view [1,2,4–8]. The major
concern of this process is, however, the presence of H SO , which
2
4
obtaining high selectivity to the 4-aminophenol. The reason of such
a behavior is due to the easy protonation of the phenyl hydroxy-
lamine (pKa = 1.96 [10]), which is extracted from the organic phase.
−1
In the acid solution (H SO4 concentration 0.6–1 mol L ), the N-
2
phenylhydroxylamonium salt is readily formed and it undergoes
fast Bamberger rearrangement to the corresponding 4-hydroxy-
anilinium cation [11].
From environmental point of view, the major drawback of the
process is the neutralization of the acidic phase, with the con-
sequent by-production of sulfate salts, which are undesired low
values products and/or wastes. In addition, diluted sulphuric acid
causes huge corrosion concern, with the consequent increased
costs of the hydrogenation plant [1,2,4,5,8].
2
4
is origin of corrosion, safety, environmental and separation prob-
lems. The reaction (see Scheme 1) is industrially carried out in
Therefore, removal of H SO from the reaction is likely the most
2
4
important target for the improvement of the process. In this sense,
several researchers have recently proposed the use of biphasic liq-
uid system or gas phase reactions (water-nitrobenzene) employing
∗ Corresponding author.
E-mail address: ronchin@unive.it (L. Ronchin).
http://dx.doi.org/10.1016/j.apcata.2016.02.020
0
926-860X/© 2016 Elsevier B.V. All rights reserved.

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
59
H
HO
HO
+
NH
NH
H
NH2
N
+
H
H O
2
+
+
-
H O
-H
2
C
H
OH
Scheme 1. Commonly accepted mechanism for Bamberger rearrangement.
bifunctional Pt supported on solid acid but quite unsatisfactorily
results have been obtained for practical purposes [12–15]. Start-
ing from these considerations, a single liquid phase process could
be a catwalk to a more sustainable process by using solvent and
catalyst easily reusable [16–19]. In fact, very recently we pro-
pose the selective hydrogenation of nitrobenzene in a single liquid
phase reactions achieving 45% of aminophenol yield in one-step
2. Experimental
2.1. Materials
Nitrobenzene, aniline, 4-aminophenol, 2-aminophenol, trifluo-
roacetic acid, trifluoroacetic anhydride, were all Aldrich products,
their purity were checked by the usual methods (melting point,
TLC, HPLC, GC and GC–MS) and employed without any purification,
acetonitrile HPLC gradient grade was supplied by BDH, methanol,
nitromethane, dimethylformamide and dimethyl sulfoxide are ACS
reagent supplied by Aldrich. Phenylhydroxylamine was prepared
by following the classic procedure proposed by Kamm, scaled down
for gram amount synthesis [28].
[
16]. Bases of these researches have been the recent results in
the Beckmann rearrangement of ketoximes in CH C N CF COOH
3
3
solvent catalytic system [17–19]. The analogy between Beckmann
and Bamberger rearrangement originates from the idea that both
oximes and PHA are nitrogen-containing compounds, whose inter-
actions with TFA play the key roles in these reactions. Above all, the
mechanism of the Bamberger rearrangement is well known from
long time and it can be depicted as in Scheme 1, via a nitrenium ion
intermediate followed by nucleophilic attack of a water molecule
2.2. Equipment
[
11,20,21].
This mechanism does not explain, however, many features of
Products were identified by gas chromatography (GC), gas
the reaction, such as the regioselectivity of the hydroxylation in 4-
position of the phenyl ring [11,20,21]. This evidence is confirmed
also by using solid acid catalyst for which almost complete selec-
tivity in 4-position has been observed [22]. Besides, the Bamberger
rearrangement of PHA carried out in aqueous solution of carbon
dioxide shows a complete selectivity to para position [23]. In fact,
reactions whose selectivity toward the 4-position is not respected
are those carried out in the presence of metal catalyst, in which
the presence of a metal-organic intermediate determine a different
reaction product [24].
chromatography coupled mass spectrometry (GC–MS) and high
performance liquid chromatography (HPLC). GC and GC–MS anal-
yses were carried out with an Agilent 7890A, equipped with FID or
MS detector Agilent 5975C. Separation was achieved by a HP 5 col-
umn (I.D. 320 ␮m 30 m long), using helium as the carrier under the
following analysis conditions: injector 523 K, detector 543 K, flow
−
1
−1
1
1
mL min , oven 333 K for 3 min 523 K 15 K min and 523 K for
5 min.
Due to the thermal instability of the products, routine anal-
ysis were carried out by HPLC (Perkin Elmer 250 pump, LC 235
diode array detector and a C 8, 5 ␮m, 4 mm i.d. 25 cm long column)
analysis were carried out with CH C N H O as mobile phase in iso-
To say the truth, nobody detects a free nitrenium ion in solu-
tion, thus suggesting that this ion is not an existent intermediate,
but more likely, a hypothetical stage in concerted mechanisms
3
2
−
1
cratic 70% of CH CN at 1 mL min . The response factors obtained
3
[
20,21,25]. Recently, DFT calculations suggest the absence of the
by standard solutions of the pure products give conversion, yield
and selectivity.
A Perkin Elmer lambda 3 Spectrophotometer were employed for
the spectrophotometer measurements of the equilibria and of the
kinetics.
A Bruker Avance 300 spectrometer operating at 300.13 MHz
allows the recording of the 1H Nuclear Magnetic Resonance (NMR)
spectra of reagents and products.
nitrenium intermediate in the Bamberger rearrangement [26]. In
this study, the stabilization of a dication intermediate occurs via the
formation of a complex formed of four hydrogen-bonded molecule
of water surrounding the anilinium intermediate, which favors the
attack of the water in para position, for steric reasons. Actually, in
the heterogeneous Pd catalyzed hydrogenation of nitrobenzenes,
nitrenium ion is a likely intermediate, since its coordination with
Pd may stabilize a surface nitrenium complex [16,27].
Starting from these bases, the study of the Bamberger rearrange-
ment of PHA in non-aqueous solvent, in the presence of TFA, is
interesting, not only for synthetic purposes, but also for mech-
anistic ones. In addition, new insight on the mechanism of the
rearrangement may help to develop a more selective and sustain-
able one pot hydrogenation of nitrobenzene to 4-aminophenol.
Here we present a study on the kinetics of Bamberger rear-
rangements of PHA in acetonitrile catalyzed by TFA as a reusable
homogeneous acid catalyst.
Conductivity measurements were carried by using a Solartron
1
260 gain phase analyzer at 2000 Hz with a standard platinized
−
1
electrodes cell with 1 cm of cell constant.
2
.3. Equilibria of N-phenylhydroxylamine TFA in acetonitrile
The standard procedure to measure the interaction of PHA with
TFA in acetonitrile has been carried out by using a conventional
spectrophotometer in a 3 mL cuvette of 1 cm of optical length. The
additions of a concentrated solution of TFA in acetonitrile by a
micrometric micro-syringe in the cuvette gives the desired acid
concentration.

6
0
N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
Table 1
Results of equilibrium calculations by Koltof and Chantoony procedure. Run conditions: T 298 K, PHA 1 × 10 mol L
−
3
−1
.
Water (mol L 1
−
)
TFA (mol L
−1
)
H+ (mol L−1
)
PHAH+ (mol L−1
)
pKPHA-H
pKTFA
1 × 10−
1
2.49 × 10
1.31 × 10
7.78 × 10
3.89 × 10
−7
−4
−8
−5
3.46 × 10
−5
5.16
3.36
5.16
3.36
10.06
6.32
10.06
6.32
1
6
1
6
.2
−
1
2
2
−4
1 × 10
1 × 10
2.31 × 10
−
−5
.2
1.11 × 10
1 × 10−
8.18 × 10
−5
2
.4. Reaction kinetics
role of the solvent will be not negligible, in any case [30]. For these
reasons, a preliminary treatment on the equilibria in water, ace-
tonitrile and TFA is of interest for discussing the kinetics of the
rearrangement.
The Bamberger rearrangement of the PHA were car-
ried out under nitrogen (charged at atmospheric pressure)
in jacketed sealed glass reactor (40 mL) at several tem-
a
peratures and autogenous pressure. These are typical run
conditions of the reaction: temperatures 313–353 K, PHA con-
3
.2. Equilibria of TFA and PHA, in acetonitrile
−4
−3
−1
centrations 6 × 10 − 1.2 × 10 mol L
,
TFA concentrations
−
2
−1
−1
−1
In order to model the reaction an approximate evaluation of
1
× 10 –1 × 10 mol L
,
water concentrations 0.3–6 mol L
,
the pKa of PHA, TFA and the amount of the species in solution
has been carried out by using the procedures of Koltoff and Chan-
tooni applied to water acetonitrile mixtures [29,31]. Even though,
the values of pKa are an estimate, they allow an evaluation of the
ionic species in solution and thus verifying, if it exists a correlation
between these values and the kinetic parameters. Table 1 reports
some results of these calculations (detail and complete calculations
are in Supplementary materials). It is evident that increasing TFA
and water concentration increase both the H+ and the protonated
PHA.
These calculations put in evidence that the concentrations of the
ionic species are related both to solvation ability of the solvent and
to the acid-base properties of TFA and PHA. Then the augmentation
of the amount of water causes the increasing of the ionic species
but reaction rates, as we will show extensively in Section 3.4, does
not follow this trend.
On the basis of this calculation, the concentration of the ionic
species are negligible, but the evident Uv–vis spectral variation of
Fig. 1(A), as TFA concentration increases, at temperatures where the
rearrangement does not occur, it cannot be ascribed to a simple sol-
vent effect but more likely to a formation of a complex PHA-TFA.
Such a compound is likely to be an ionic couple, in fact, acetoni-
trile is well known to be a solvent whose solvating power does not
allow anion and cation separation but the formation of ionic cou-
ples is favored [30]. The spectral variation observed in Fig. 5(B) is
due to addition of water on the adduct TFA-PHA previously formed.
It is evident that the increase of PHA bands (the shoulder at 240 nm
raises) is related to the increase of water concentration. Likely, the
interaction between TFA and PHA decreases by adding water, but
solvent acetonitrile, reaction volume 25 mL.
The course of reactions was checked by sampling the liquid
phase at established intervals, and by diluting the samples at a suit-
able concentration for the analysis. The changes in the UV spectra
of the solution are recorded in a thermostated spectrophotometric
quartz cell (2 mL path-length 1 cm) at various temperatures. Since
strong solvent effect is evident for all bands, the samples were
analyzed by HPLC after dilution in the mixture of the HPLC elu-
ent (acetonitrile-water 70%). The spectral behavior of the kinetics
between 200 and 340 nm shows several wavelengths with a notice-
able variation but the more reliable data has been obtained at 220
and 275 nm, where the characteristic bands of the 4-aminophenol
are observed (see Supplementary materials).
The use of a computational software (Wolfram Mathematica)
allows the implementation of the minimization procedure by using
the built-in functions of the program. The procedure, for gaining the
constancy of the “kinetic factor” (kmod) by reagents concentrations,
consists in a minimization of the variance of the kmod. Such con-
stant is proportional to the kobs and to a power low equation in TFA
and H O concentrations, where the exponents are the minimization
2
variables of the variance (complete data set is in Supplementary
materials).
2.5. NMR measurements
The measurements were carried out in a 300 MHz nuclear mag-
netic resonance spectrometer, at 298 K by using deuterated DMSO
or deuterated chloroform at a concentration close to that of reac-
tions. All the chemical shifts were referred to the internal standard
tetramethylsilane.
−1
the band of neat PHA is not restored at least to 3.7 mol L of water.
In fact, the increase of the concentration of water causes not only
the rises of PHA, but also a modification of the solvent itself, thus
giving a different spectrum with respect to the starting PHA. It is
noteworthy, the formation of a complex with TFA in the presence
of two competing molecules, namely the PHA and the water, is a
complex phenomenon. A complete treatment of the argument is
beyond the scope of this work, but it will be took into account, as
an equilibrium (in a first approximation) for the calculation of the
kinetics (see Section 3.5) [30].This is in agreement with the results
observed employing the same solvent-acid system in the Beckmann
rearrangement of the cyclohexanone oxime [17,18]. In that case, at
temperature where cyclohexanone oxime does not rearrange, quite
strong interaction between cyclohexanone oxime and TFA exists,
3
. Results and discussion
3.1. Some preliminary consideration on bamberger
rearrangement
Bamberger rearrangement is one of those reactions whose
mechanisms have been remained in a kind of limbo for long time
because of the obscure nature of its regioselectivity, for which only
very recently a plausible explanation has been given [25]. However,
the reaction in the presence of non-aqueous solvent has not been
studied yet, and a kinetic study, in this environment, may help to
adding some new insight on this uncertain topic. As a matter of
fact, the reaction has almost two steps since the interactions of
the substrate with TFA is certainly occurring because of the quite
high acidity and basicity of TFA and PHA, respectively (in aqueous
solvent TFA pKa = 0.23 and PHA pKa = 1.96) [29,10]. After this step,
the rearrangement may follow several pattern of reaction, but the
−
1
and this equilibrium has an enthalpy variation of −19 kJ mol
.
This suggests the formation of a complex, which could be ascribed
to the ionic couple complex (TFA-cyclohexanone oxime), rather
than a real protonation equilibrium [17,18]. In the Beckmann
rearrangement, after the formation of the complex, the reaction
proceeds via an ester intermediate [17–19]. In the Bamberger

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
61
A
4
3
2
1
0
B
4.0
.5
3.0
[TFA]
3
2.5
2.0
1.5
1.0
0.5
0.0
[H O]
2
2
20
240
260
280
300
320
340
220
240
260
280
300
320
340
λ
(nm)
λ, (nm)
Fig. 1. A Spectral variation of PHA increasing TFA concentration (0–0.12 mol L 1), B addition of water after TFA (3 10−3–3.7 mol L ).
−
−1
-1
[TFA] = 0 mol L
-3
-1
[
TFA] = 6.9 x10 mol L
-2
-1
[
TFA] = 0.04.2 x10 mol L
-2
-1
[
TFA] = 7.4 x 10 mol L
-
1
[TFA] = 0.32 mol L
1
2
11
10
9
8
7
6
5
4
Chemical Shift (ppm)
Fig. 2. 1H NMR spectra of PHA in the presence of TFA. Run conditions: 300 MHz DMSO-d6, T 298 K, PHA concentration of 5 × 10 mol L . Chemical shift: TFA 0 mol L
−2
−1
−1
,
−
1
(
t, ı 6.71 ppm J = 7 Hz; d, ı 6.83 ppm, J = 8 Hz, ı 7.17 ppm J = 7 Hz, ␦ = 8.23 ppm, ␦ = 8.28 ppm); TFA 0.0069 mol L , (t, ı 6.76 ppm J = 7 Hz; d, ı 6.84 ppm, J = 8 Hz, ı 7.18 ppm
J = 7 Hz, broad ␦ = 5.3 ppm); TFA 0.042 mol L , (t, ı 6.82 ppm J = 7 Hz; d, ı 6.89 ppm, J = 8 Hz, ı 7.20 ppm J = 7 Hz, broad ␦ = 7.9 ppm); TFA 0.074 mol L , (t, ı 6.90 ppm J = 7 Hz;
d, ı 6.97 ppm, J = 8 Hz, ı 7.24 ppm J = 7 Hz, broad ␦ = 9.7 ppm); TFA 0.32 mol L , (t, ı 7.02 ppm J = 7 Hz; d, ı 7.08 ppm, J = 8 Hz, ı 7.28 ppm J = 7 Hz, broad ␦ = 11 ppm).
−1
−1
−
1
rearrangement, such an ester intermediate has never observed
in the reaction media. Furthermore, the NMR measurements (see
Fig. 2) give poor information on the nature of the TFA-THA complex,
because of the strong quenching of the signals due to the presence
of TFA (Fig. 3).
As a matter of fact, the quenching of the signal of the proton of TFA
and those of the NHOH group, together with the directly related
downfield of the phenyl protons with the amount of TFA, are fur-
ther proofs of the strong interaction between the acid and the PHA,
suggesting the formation of a stable complex (see Fig. 2).
It appears, however, that such a specie has the signals of the
aromatic protons very close to those of the starting PHA. In fact,
the addition of TFA does not noticeably modify the spectrum of the
protons of the phenyl ring, except for a downfield shift, likely due
to an interaction of TFA with the NHOH group. This strong inter-
action is in agreement with the absence of the NHOH at 8.25 ppm
and 8.21 ppm assigned to the aminic proton and to the hydroxyl
one, respectively. The addition of the TFA cause their disappear-
ing because of the exchange with that of the TFA, whose signals at
In order to verify the reactivity of PHA with TFA, which does
not give any product of esterification or amidation we study the
reactivity of PHA with trifluoroacetic anhydride, which gives N-
hydroxy-trifluoroacetanilide almost immediately, in agreement
with literature data (see Supplementary materials) [32]. The
spectrum of N-hydroxy-trifluoroacetanilide does not change by
increasing the temperature to 343 K, after 1 h of reactions, in the
presence of a stoichiometric amount of TFA, (see Fig. 2). This sug-
gests N-hydroxy-trifluoroacetanilide is quite stable at 343 K and no
rearrangement to 4-aminophenol or to its derivatives occurs. This
−
3
−1
6
.9 × 10 mol L is a very broad band centered at about 5.3 ppm.

6
2
N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
PHA 298 K
PHA TA 298 K
PHA TA 343 K
3
3
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
.5
.0
2
00
225
250
275
300
325
350
λ
(nm)
−
4
−1
Fig. 3. Reactivity of PHA with trifluoroacetic anhydride (TA) at 298 K and at 343 K for 1 h. Run conditions: Uv–vis spectra at 5 × 10 mol L solvent acetonitrile, optical
length 1 cm.
evidence establishes that N-hydroxy-trifluoroacetanilide is not an
intermediate of the Bamberger rearrangement.
3.3. Reaction rate: comparison of spectrophotometric and HPLC
measurements
Starting from these evidences the equilibrium intermediate is
neither an amide nor an ester of the TFA but it could be ascribed
to anionic couple or a kind of cluster complex. Apart, the spec-
troscopic evidence of NMR and UV Vis bands displacement, the
only information on the nature of this compound come from con-
ductivity measurements. In fact, the conductivity of the solution
The complex behavior of the UV–vis, spectra needs an inde-
pendent measurement of the reaction rate in order to correctly
choice the wavelength for measuring the kinetics of reaction. Actu-
ally, spectra of reagent and product change as the ratio of water
and TFA is modified, because of the strong effect that has this
parameter on the TFA-PHA complex equilibrium. The practical con-
sequence of this solvent effect is the impossibility of having an
unambiguous measurement of the kinetics simply by following the
spectral change. For this reason a comparison of kinetics obtained
by analysing with HPLC samples withdrawn at various time are
useful to assess the reliability of the UV–vis measurements. Table 2
reports the pseudo first order kinetic constants determined by
−
1
−1
at0.3 mol L of TFA in acetonitrile is 4 ␮S cm , which suggests
the presence of the TFA practically undissociated. This in agree-
ment with the calculations reported in Table 1.The addition of
−
3
−1
PHA at a concentration of 10 mol L raises the conductivity
−
1
to 150 ␮S cm , a value compatible with a conductivity in non-
aqueous solvent of a highly associated ionic compound, but not
with that of a free ion [33].
−
1
Until now the reflections on the nature of the specie detected
by spectroscopic measurements, does not give an evaluation of
the equilibrium of formation of such a complex, which is proba-
bly the key intermediate of the kinetics of reaction. For this reason,
we measure the equilibrium constant (Keq) by spectrophotometric
measurements of the complex PHA-TFA at 298, 293 and 288 K in
acetonitrile as a solvent (Scheme 2, Eq. (1)). At such temperatures
the rearrangement practically does not occur, since its kinetics is
very slow, thus allowing the measurement of the equilibrium of
formation of the PHA-TFA specie. We consider in the equilibrium
calculation the concentration of the species in solution and not the
activity, since there are no charge separation and each activity coef-
ficient is proportional to the specific Sechenov factor, which can be
englobed in the Keq [34].
using both techniques. At 0.3 mol L of water, the scarce variation
of the bands in the UV–vis spectra does not allow a reliable evalu-
ation of the kinetics by UV–vis measurements. However, the quite
good correspondence of the kinetic constants, obtained by UV–vis
−
1
and HPLC analysis at 1.2 and 6 mol L of water, suggests that the
kinetic constants measured with the two methods are comparable
(see Table 2) and for our purposes the two methods are equivalent.
The evaluation of the observed first order kinetic constant (kobs
)
allows the comparison of the reactions by varying the operative
variables but this behavior does not allow any consideration on
the reaction mechanism. In fact, the reaction is multi-step, since
spectrophotometric measurements evidences the PHA-TFA inter-
mediate. The formation of such a complex, however, is not evident
on the time vs. product profile of the overall kinetics (see Fig. 5).
This behavior is typical kinetics where the pre-equilibria are in
stationary state influencing the kobs but not the overall profile,
which remain first order, apparently. In fact, kobs is substantially
constant doubling substrate concentration (Table 2). This kind of
time vs. concentration profile (first order kinetics), without any
evident intermediate, even though in agreement with previous lit-
erature data in mineral acids, gives poor information on the reaction
mechanism [11,20,21].
PHA + TFA ꢀ PHA-TFA
PHA − TFA]
Scheme 2
(1)
[
[
Keq =
PHA] [TFA]
The measurements at various temperature give a linear Van’t
Hoff plot (Fig. 4), thus allowing the extrapolation of the equilib-
rium constant at the temperatures of reaction. As expected, the
equilibrium is moderately exothermic with an enthalpy of reaction
−
1
of c.a. –12 kJ mol . This value is in agreement with that measured
3.4. Influence of the operative variables on the reaction kinetics:
temperature, PHA, TFA, and water concentration
−
1
for the system TFA with cyclohexanone oxime (c.a. −19 kJ mol
)
[
17]. Clearly, the concentration of the cluster decreases as the tem-
perature increase and the negative entropy of formation suggests
the formation of the cluster causes a reduction of the global number
of molecules in solution.
In order to take information on the reaction path, we investi-
gate the role of some operative variable on the reaction kinetics.
Of particular interest is the role of water, on the reaction kinetics

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
63
H
O
H
H
O
N
H
OH
O
H
N
O
H
F
F
+
OH
H
F
F
F
F
H
H
PHA-TFA
Scheme 2. Equilibrium of PHA, TFA and the complex PHA-TFA.
Fig. 4. Vant’Hoff plot of the equilibrium PHA-TFA: ꢀH = −12.6 kJ mol−1, ꢀS = −11.3. J K−1 mol
◦
◦
−1
.
Table 2
Comparison of the kobs obtained with UV–vis and HPLC at 343 K TFA 0.1 mol L in CH3CN as a solvent, reaction volume 25 mL.
−
1
5
5
5
Water
PHA
HPLC10 xkobs
UV–vis (220 nm)10 xkobs
UV–vis (275 nm)10 xkobs
−
−1
−1
−1
−1
)
(
mol L 1
.3
.2
)
(mol L
)
(s
)
(s
/
)
(s
/
0
1
6
6
6
1 × 10−3
1 × 10−3
1 × 10−3
61.0
48.1
18.4
/
49.0
22.9
22.8
23.0
49.0
23.0
23.1
22.9
−
4
3
1.2 × 10
6 × 10−
/
A
B
Fig. 5. Comparison of two fittings of first order equation on the kinetic profile 4-hydroxyaniline formation measured by both spectrophotometric UV–vis 220 nm (A) and
−
3
−1
−1
−1
−1
HPLC (B) analysis. Run conditions: temperature (343 K), PHA (10 mol L ), TFA (10 mol L ), water (6 mol L ), solvent acetonitrile, reaction volume 25 mL.
being an active participating reagent on the mechanism [26]. On
the basis of well-established literature data, water enters in the
reaction pathway as an intermolecular reagent, in fact, the migra-
tion of the hydroxyl group is not specific but it comes from the

6
4
N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
A
B
-
1
0
.0005
TFA 0.1 mol L
0
0
0
0
0
0
.0005
3
3
3
13 K
33 K
43 K
-1
TFA 0.06 mol L
TFA 0.01 mol L
-
1
.0004
.0003
.0002
.0001
.0000
0.0004
0.0003
0.0002
0.0001
k
k
1
0
1
2
3
4
5
6
0
1
2
3
4
5
6
-1
-1
water (mol L )
Water (mol L )
−
3
−1
−1
−1
Fig. 6. Influence of water on the kobs at various TFA concentrations and temperatures. Run conditions: PHA (10 mol L ), TFA (10 mol L ), solvent acetonitrile reaction
volume 25 mL., point 1 extrapolated from Arrhenius plot.
0
0
0
0
0
0
0
0
0
0
0
.00050
.00045
.00040
.00035
.00030
.00025
.00020
.00015
.00010
.00005
.00000
3
33 K
343 K
3
53 K
k
0.00
0.02
0.04
0.06
0.08
0.10
-1
TFA (mol L )
−
3
−1
Fig. 7. Influence of the TFA concentration on the kobs at various temperature. Run conditions: PHA (10 mol L ), water (150 mmol), solvent acetonitrile, reaction volume
5 mL.
2
solvent [11,20,21]. For this reason, water concentration is a fun-
damental parameter also for this study, even though, we use
acetonitrile as a solvent. Fig. 6 shows the influence of water amount
on the kobs, there is a monotone decreasing of the kobs as the amount
of water raises. This behavior suggests that the limiting step is not
the attack of the water as nucleophile, even though it has a clear
influence on the overall kinetics.
also on the energy of the various steps of the process, thus suggest-
ing the presence of temperature dependent equilibria. In any case,
these behavior are in agreement with those measured in sulfuric
acid by other authors, then it is likely that despite of, different sol-
vent and reaction environment a similar reaction path may occurs
[11,20,21].
The amount of acid, as expected, influences directly the kobs
,
which increases monotonically as the TFA concentration rises. This
behavior is expected since the rate of rearrangement to an electron-
deficient nitrogen is generally directly related to the amount of acid
3.5. The kinetics of reaction
As already pointed out in previous sections, the equation rate
is first order, but the influence of the operative variables on the
kobs suggests the presence of equilibria affecting the rate constant.
In fact, the effective kinetic constant depend from the temperature
only. Starting from these considerations, we have to remember that
in Section 3.2, we observed the formation of the PHA-TFA com-
plex and we measured its equilibrium in acetonitrile as a solvent.
In addition, the presence of water influences this equilibrium, but
due to the shift of the bands, in the UV–vis region, the equilibrium
of the system acetonitrile-TFA-PHA-water is not possible to mea-
sure reliably. Then, it is necessary to determine these equilibria by
an indirect procedure. It appears from what reported in Section 3.4
that the kobs depends from both TFA and water concentration. For
this reason, we plot kobs vs TFA/water mole fractions (Fig. 9A). Start-
ing from this evidence, we define a kmod, obtained by combination
(
e.g. in the Beckmann rearrangement [35–37]). In this case, how-
ever, it is more probably related to the formation of an intermediate
whose equilibrium is influenced by the simultaneous presence of
TFA and PHA. On the contrary, as pointed out in Section 3.2, the
PHA protonation depends directly to PHA, TFA and water concen-
trations but the amount of the latter is inversely proportional to the
kobs. This suggests that the catalysis is not due to proton exchange
but likely via the formation of the PHA-TFA complex Fig 7.
The Arrhenius Plots (Fig. 8) at various TFA and water con-
centration show temperature factors (apparent activation energy)
comprised between 79 kJ mol and 104 kJ mol .
−1
−1
This behavior suggests that both TFA and water concentration
has a clear influence not only on the observed kinetic constant but

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
65
B
-
-
-
-
-
-
-
60
65
70
75
80
85
90
A-60
-
1
-1
TFA 0.1 mol L
water 1.2 mol L
-
1
-
1
-64
-68
-72
TFA 0.06 mol L
TFA 0.012 mol L
water 3 mol L
water 6 mol L
-1
-1
k
-76
k
-80
-84
-88
-92
-95
-100
0
.00284 0.00288 0.00292 0.00296 0.00300
0.0029
0.0030
0.0031
0.0032
-
1
-1
1
/T (K )
1/T (K )
−
1
−1
−1
−3
−2
−1
−2
−1
;
Fig. 8. Arrhenius Plot at various TFA concentrations (A) and water amount (B): (A) 92 kJ mol in TFA 0.1 mol L , 97 kJ mol in TFA 6 × 10 mol L and in TFA 10 mol L
−
1
−1
−1
−1
−1
−1
−1
(
B) 79 kJ mol (water 1.2 mol L ), 81 kJ mol (water 3 molL ), 92 kJ mol (water 6 mol L ). Run conditions: PHA (10 mol L ), solvent acetonitrile, reaction volume
5 mL.
2
B
0.007
A
0
.0005
.0004
.0003
353 K
43 K
313 K
33 K
0.006
0.005
0.004
0.003
0.002
0.001
0.000
3
53 K
3
0
0
343 K
313 K
33 K
3
3
k
0.0002
k
0
0
.0001
.0000
0
.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
0
.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
TFA (mole fractions)
TFA (mole fraction)
Fig. 9. Comparison of kobs (a) and kmod (B) at various temperatures in function of the TFA mole fraction.
of the kobs with the initial water and TFA concentrations. Eq. (2)
shows the dependence of the reaction rate from kobs, as a pseudo
first order kinetics and it defines the kmod as a power low equa-
tion in water and TFA concentrations (details of the calculations
are in Supplementary materials). It appears in Fig. 9 B that kmod
On the light of these calculations, it appears a direct relationship
between the rate of reaction with TFA and PHA and an inverse one
with water concentration. This behavior is a further proof that TFA
dissociation and PHA protonation are not the determining stages,
since they depends directly by water and TFA concentrations and in
acetonitrile these processes are negligible. Then, we could suppose
that the PHA-TFA complex is the key intermediate in the reaction
kinetics, and we suppose Eq. (3)is the rate of reaction, where k is
the actual kinetic constant Fig. 11.
does not change with the mole fraction TFA/H O, thus suggesting
2
its independence from the species present in solution.
TFA]0.69
H2O]0.5
[H2O]0.5
[TFA]0.69
r = kobs [PHA] = kmod [PHA] [
[
thuskmod = kobs
(2)
r = k [PHA − TFA]
(3)
Then, we obtain Eq. (4) by rearranging Eqs. (1) and (2), since kmod
has a behavior typical of a “true” kinetic constant, we reasonably
In Fig. 10 B, it is evident, that kmod depends only by the tem-
perature, and the Arrhenius plot is linear (Fig. 9). The power low
0
.69
−0.5
supposek = kmod, then we calculate the equilibrium concentration
expression [PHA] [TFA]
[H O]
2
represents the equilibria in
of the PHA-TFA complex by Eq. (5).
solution of TFA, water and PHA, from which derives the PHA-TFA
complex, then, we suppose this is key intermediate, from which
depends the kinetics of reaction.
TFA]0.69
[
r = kobs [PHA] = k [PHA − TFA] = kmod [PHA]
(4)
(5)
H2O]0.5
[
This behavior of the kmod suggests that these constant is strictly
related to the actual kinetic of reaction. If we consider the activation
parameter obtained by fitting the Eyring-Evans-Polanyi equation
TFA]0.69
[
PHA − TFA] = [PHA] [
H2O]0.5
[
ꢀ
−1
we obtain the enthalpy of the activated state of ꢀH = 77.6 kJ mol
−
1
−1
and the positive entropy of activation (177 J mol
K ), which
After these considerations, we try to calculate the equilibria in
could be related to a dissociative mechanism for the activated state
30]. In agreement with a displacement of a molecule of water from
PHA to the solvent.
solution taking into account for the presence of water. Despite of
the complexity of the three component system (water, TFA, ace-
tonitrile), we suppose the equilibrium of the species in solution
[

6
6
N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
ꢀ
−1
ꢀ
−1
−1.
Fig. 10. Arrhenius plot of the modified kinetic constant kmod. Eyring-Evans-Polanyi equation parameters: ꢀH = 77.6 kJ mol , ꢀS = 177 J K mol
Fig. 11. Experimental vs calculated r0 values.
H
O
H
O
O
H
H
F
O
+
F
α
O
O
H
H
OH
F
H
F
F
F
TFA-W
molecules of water
α
Scheme 3. Equilibrium of TFA, H2O and the complex TFA-W.
could be described by the equilibrium of formation the complex
PHA-TFA (Scheme 2 and Eq. (1)) and the equilibrium of water with
TFA (Scheme 3and Eq. (6)).
describes the competing equilibrium of water with TFA on forming
a TFA-W cluster. Besides, in Eq. (6), we assume the equilibrium of
the TFA-W complex depends on a variable number of molecules of
TFA and water and the exponent␣ of Eq. (6) takes into account for
the stoichiometric ratio of the complex Fig. 12.
By reducing the simultaneous Eqs. (1), (6)–(8), where Eqs. (7)
and (8) are the mass balance of TFA and PHA, respectively, we
obtain an expression of the PHA-TFA concentration in function of
the [PHA] , [TFA] , [H O], concentrations at time 0, and of Keq, K , ␣.
˛H O + TFA ꢀ TFA-W Scheme 3
2
0
0
2
S
As described in Section 3.1, Eq. (5) takes into accounts for the equi-
librium between TFA, PHA and the complex PHA-TFA, while Eq. (6)
[
[
PHA − TFA]
Keq =
(1)
PHA] [TFA]

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
67
Fig. 12. Van’t Hoff plot of the equilibrium TFA-W complex: ꢀH = −13.5 kJ mol−1, ꢀS = − 23 J K−1 mol
◦
◦
−1
.
Fig. 13. Steric hindrance towards a nucleophilic attack to the para positions of the PHA-TFA complex.
[
TFA − W]
Table 3 shows some results of the fitting; it appears that the
value of ␣ is lower than 1 suggesting that the TFA-W complex is
likely a cluster composed of more than one molecules of TFA and
water. In any case, the interaction of TFA with water explains the
depressive effect of water on the reaction kinetics, thus compet-
ing in the formation of the PHA-TFA complex. Ks is the equilibrium
constant of TFA, water and the TFA-W complex, whose physical
meaning can be related also with the modification of the solvent
as water concentration increases. Also in this case, it seems that,
despite of the indirect procedure for obtaining the equilibrium con-
stants, the model holds, since the Van’t Hoff plot is consistent and
the value of the enthalpy of formation (moderately exothermic-
Ks =
(6)
˛
[
TFA] [H2O]
TFA]0 = [TFA] + [PHA − TFA] + [TFA − W]
PHA]0 = [PHA] + [PHA − TFA]
[
[
(7)
(8)
Then, simultaneous equations 1, 6–8 have as unknown
parameters␣ and Ks. By the minimization of the square residuals
between the calculated [PHA-TFA] (by solving the simultaneous
Eqs. (1), (6)–(8)) and the values obtained from Eq. (5) (derived
from the measured rate of reaction) we obtain the values of the
two parameters (␣ and Ks). From a mathematical point of view, the
solutions of the simultaneous equation is a bi-parametric approach
for the calculation of the equilibria that takes into accounts the
complexity of the system. It is noteworthy that at first attempt, we
have tried a model, with one molecule of water and one of TFA in the
equilibrium with the TFA-W complex (namely ␣ = 1). In this case,
the correspondence, between the calculated initial rate of reaction
and those measured, was quite poor (see Fig. 13 black triangles, see
also Supplementary materials). From a practical point of view, the
necessity of a further parameter to describe these complex equi-
libria is reasonable, since the formation of a TFA-W complex is
dependent on the solvent. In fact, water modify the characteristic
of the solvent, for instance, favors charge separation. This proce-
dure, however, is useful since it introduces a single parameter (␣)
directly related to the equilibrium of the solvent but having an evi-
dent physical meaning. Fig. 13 (open circles) shows the goodness
of the model, being evident a good correspondence between cal-
culated vs. the experimental initial reaction rates, additionally the
points are well behaved around the middle of the line.
−
1
1
3.5 kJ mol ) of the cluster is in agreement with the constant of
formation of these kind of complex [38]. Entropy of formation is
−
1
−1
negative for 23 J K mol suggesting the involvement of lower
number of molecules passing from the free state to the associated
one. These results are physically consistent, and then it is a fur-
ther proof of the goodness of the model. On these bases, we could
attempt an explanation of the kinetics and of the regioselectivity
of the reaction Fig. 12.
Until now, we have shown the role of the PHA-TFA complex as
a stationary state specie, whose equilibrium influences the over-
all kinetics. This explanation on the kinetic mechanism, however,
does not straightly elucidate the regioselectivity towards the para
position. The complex PHA-TFA should be responsible also of the
regioselectivity towards the para position if we consider the ortho
position is sterically hindered. In fact, the −NHOH group interacts
with the COOH one of the TFA resulting in a hindered struc-
ture in ortho position. This is evidenced in Fig. 13, which shows a

6
8
N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
Table 3
Table of the thermodynamic and of the kinetic constants. Run Conditions: PHA (1 × 10 mol L−1), TFA (1 × 10−1 mol L ), solvent acetonitrile, reaction volume 25 mL.
−
3
−1
TFA (mol L 1
−
)
H2O (mol L
−1
)
T (K)
105 x kobs (s−1
)
103 x kmod
Keq
Ks
␣
105 x PHA-TFA (mol L−1
)
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
1
2
2
1
1
1
2
2
2
2
2
2
1
1
1
1
1
1
1
6
1
1
1
1
6
6
6
1
1
1
1
1
1
1
1
1
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
× 10
6
6
6
1.2
3
6
1.2
3
6
1.2
3
353
353
353
343
343
343
343
343
343
343
343
343
313
313
313
333
333
333
50.1
34.9
9.72
52.1
32.9
23.2
37.1
23.8
16.5
10.8
7.18
4.99
8.01
4.43
2.84
20.6
13.5
10.4
6.28
6.32
6.25
2.73
2.74
2.70
2.75
2.82
2.74
2.89
2.86
2.86
0.12
0.24
0.24
1.09
1.07
1.12
20
20
20
22
22
22
22
22
22
22
22
22
33
33
33
25
25
25
5.85
0.679
0.679
0.679
0.623
0.623
0.623
0.601
0.601
0.601
0.592
0.592
0.592
0.699
0.699
0.699
0.618
0.618
0.618
8.23
5.43
1.15
18.6
11.9
8.24
13.1
8.36
5.80
3.80
2.43
1.69
44.4
21.4
12.9
18.6
11.9
8.24
5.85
5.85
7.69
7.69
7.69
6.95
6.95
6.95
4.08
4.08
4.08
6
0.1
1.2
3
1.2
3
10.5
10.5
10.5
8.8
8.86
8.86
6
by-dimensional representation (it is more clear with respect to the
tridimensional one) of the PHA-TFA complex obtained by a pre-
liminary energy minimization of the structure by SCF calculation
of the couple PHA-TFA without charge separation (details in Sup-
plementary materials). The real structure of the PHA-TFA specie, it
is more likely to be a complex cluster of molecules, whose struc-
ture could be object of a theoretical study, but this is beyond the
scope of this work. In this way, the attack of water, which is a quite
weak nucleophile, toward the para position is more probable than
the ortho one. Formation of cluster as intermediate for explaining
the regioselectivity of the reaction has been already proposed in a
computational study, where the formation of a cluster of molecule
of water stabilizes a dication intermediate deriving from the pro-
tonated PHA in aqueous solution [25]. In acetonitrile, the charge
separation does not significantly occur then the formation of the
PHA-TFA complex plays the major role on activating the molecule
for the rearrangement. Clearly, the presence of water, which com-
petes with PHA on forming cluster with the acid, depresses the
formation of the PHA-TFA complex, thus reducing the reaction rate
of the rearrangement.
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.apcata.2016.02.
020.
References
[
[
[
1] S. Mitchell, R. Waring, Aminophenols, vol. 2, 4th edition, John Wiley & Sons,
1999.
2] S.K. Tanielyan, J.J. Nair, N. Marin, G. Alvez, R.J. McNair, D. Wang, R.L.
Augustine, Org. Process Res. Dev. 11 (2008) 681.
3] K. Hunger, Industrial Dyes, 25, Wiley-VCH Verlag, Weinheim, 2003.
[4] C.V. Rode, M.J. Vaidya, R. Jaganathan, R.V. Chaudhari, Chem. Eng. Sci. 56
2001) 1299–1304.
(
[
[
5] http://www.dsir.gov.in/reports/techreps/tsr131.pdf.
6] J.R. Fritch, O.S. Fruchney, T. Horlenko, D.A. Aguilar, C.B. Hilton, P.S. Snyder, W.J.
Seelinger, U.S. Pat. 5,155,273 to Hoechst-Celanese Co. (1992).
[
[
7] Y. Gao, F. Wang, S. Liao, D. Yu, React. Kinet. Catal. Lett. 64 (1998) 351–357.
8] K. Min, J. Choi, Y. Chung, W. Ahn, R. Ryoo, P.K. Lim, Appl. Catal. A: Gen. 337
(
2008) 97–104.
[9] H. Saggadi, D. Luart, N. Thiebault, I. Polaert, L. Estel, C. Len, Catal. Commun. 44
2014) 15–18.
10] I.M.C. Brighente, L.R. Vottero, A.J. Terezani, R.A. Yunes, J. Phys. Org. Chem. 4
1991) 107–112.
[11] T. Sone, Y. Tokuda, T. Sakai, S. Shinkai, O. Manabe, J. Chem. Soc. Perkin Trans. 2
1981) 298–301.
12] S. Wang, Y. Ma, Y. Wang, W. Xue, X. Zhao, J. Chem. Technol. Biotechnol. 83
2008) 1466–1471.
(
[
(
4
. Conclusion
(
[
In this work, we found that the Bamberger rearrangement in
(
acetonitrile catalyzed by TFA does not depend from the protonation
of the PHA. In fact, it is strictly related to the formation of a PHA-TFA
complex, which is responsible of the kinetics as well of the regios-
electivity towards the para position on forming the aminophenol.
The equilibrium of the PHA-TFA intermediate is measured spec-
trophotometrically and its value is employed in the determination
of the kinetic parameters. The role of water on depressing the reac-
tion rate is also connected with its equilibrium with TFA on forming
a TFA-W adduct, which in turn causes a decreasing of the concentra-
tion of the PHA-TFA complex. The proposed kinetic model takes into
accounts both equilibria, giving a good fitting of the experimental
data and it allows the calculation of the equilibrium constant of
TFA with water. In addition, the model permits the calculation of
the enthalpy and entropy of all the steps of the reaction, suggesting
a dissociative transition state for the kinetics.
[13] A. Deshpande, F. Figueras, M.L. Kantam, K.J. Ratnam, R.S. Reddy, N.S. Sekhar, J.
Catal. 275 (2010) 250–256.
[
[
14] S. Wang, B. He, Y. Wang, X. Zhao, Catal. Commun. 24 (2012) 109–113.
15] T. Komatsu, T. Hirose, Appl. Catal. A: Gen. 276 (2004) 95–102.
[16] G. Quartarone, L. Ronchin, A. Tosetto, A. Vavasori, Appl. Catal. A: Gen. 475
(2014) 169–178.
[
[
[
17] L. Ronchin, A. Vavasori, M. Bortoluzzi, Catal. Comun. 10 (2008) 251–256.
18] L. Ronchin, A. Vavasori, J. Mol. Catal. A: Chem. 313 (2009) 22–30.
19] G. Quartarone, E. Rancan, L. Ronchin, A. Vavasori, Appl. Catal. A: Gen. 472
(2014) 167–177.
[
[
[
[
20] G. Kohnstam, W.A. Petch, D.L.H. Williams, J. Chem. Soc. Perkin II (1984)
423–427.
21] T. Sone, K. Hamamoto, Y. Siji, S. Shinkai, O. Manabe, J. Chem. Soc. Perkin II 159
(1984) 1596–1598.
22] K.J. Ratnam, R.S. Reddya, N.S. Sekhara, M.L. Kantama, A. Deshpande, F.
Figueras, Appl. Catal. A: Gen. 348 (2008) 26–29.
23] S. Liu, Y. Hao, J. Jiang, Ind. Eng. Chem. Res. 53 (2014) 8372–8375.
[24] A. Vavasori, M. Capponi, L. Ronchin, Chem. List. 106 (2013) 1191.
[
[
[
25] J.C. Fishbein, R.A. McClelland, Can. J. Chem. 74 (1996) 1321–1328.
26] S. Yamabe, G. Zeng, W. Guan, S. Sakaki, Beil. J. Org. Chem. 9 (2013) 1073–1082.
27] E.A. Gelder, S.D. Jackson, C.M. Lok, Chem. Commun. (2005) 522–524.
Acknowledgements
[28] O. Kamm, Org. Synt. Coll. 4 (1925) 57.
[
29] S. Espinosa, E. Bosch, M. Rose ı´ s, J. Chromatogr. A 964 (2002) 55–66.
[30] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 2nd ed., VCH,
Weinheim, 1988, pp. 79.
Financial support by Ca’ Foscari University of Venice is gratefully
acknowledged (ADIR fund 2014).
[31] M.K. Chantooni, I.M. Kolthoff, Anal. Chem. 51 (1979) 133–140.

N.d. Fonzo et al. / Applied Catalysis A: General 516 (2016) 58–69
69
[
[
[
[
32] P.S. Branco, S. Prabhakar, A.M. Lobo, D.J. Williams, Tetrahedron 30 (1992)
335–6360.
33] J. Barthel, H.J. Gores, G. Schmeer, R. Wachter, Non-Aqueous Electrolyte Solut.
Chem. Modem Technol. 111 (1983) 33–144.
34] A. Pietropolli Charmet, G. Quartarone, L. Ronchin, C. Tortato, A. Vavasori, J.
Phys. Org. Chem. 27 (2014) 680–689.
35] N.C. Marziano, L. Ronchin, C. Tortato, O. Tonon, R. Bertani, Int. J. Chem. Kin. 36
[36] N.C. Marziano, L. Ronchin, C. Tortato, A. Vavasori, C. Badetti, J. Mol. Catal. A:
Chem. 277 (2007) 221–232.
[37] N.C. Marziano, L. Ronchin, C. Tortato, A. Vavasori, M. Bortoluzzi, J. Mol. Catal.
A: Chem. 290 (2008) 79–87.
6
[38] M. Ilczyszyn, H. Ratajczak, J. Chem. Soc. Faraday Trans. 91 (1995) 1611–1618.
(
2004) 419–426.