Salophen and salen oxo vanadium complexes as catalysts of sulfides oxidation with H2O2: Mechanistic insights

Article abstract of DOI:10.1016/j.cattod.2012.03.032

The application of V(V) catalysts in oxidation of sulfides with peroxides offers an efficient procedure, that is compatible with different functional groups, and leads to good yields and selectivities. However, the understanding of the factors affecting the reactivity of different catalysts is still far to be complete. An experimental and theoretical study on a series of V(V) complexes containing variously substituted salen and salophen ligands is reported with the aim to correlate the activity of the catalysts with the electronic character of the vanadium center. The results obtained indicate that steric factors play a major role in determining the outcome of the reaction, often overcoming the electronic effects. Theoretical results suggest the intervention in the catalytic cycle of an hydroperoxo vanadium species.

Full text of DOI:10.1016/j.cattod.2012.03.032

Catalysis Today 192 (2012) 44–55
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Salophen and salen oxo vanadium complexes as catalysts of sulfides oxidation
with H2O2: Mechanistic insights
A. Colettia, P. Gallonia, A. Sartorelb, V. Contea,∗, B. Florisa
a Dipartimento di Scienze e Tecnologie Chimiche, Università di Roma Tor Vergata, via della Ricerca Scientifica snc, 00133 Roma, Italy
b Dipartimento di Scienze Chimiche, Università di Padova, via Marzolo 1, 35131 Padova, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The application of V(V) catalysts in oxidation of sulfides with peroxides offers an efficient procedure, that
is compatible with different functional groups, and leads to good yields and selectivities. However, the
understanding of the factors affecting the reactivity of different catalysts is still far to be complete. An
experimental and theoretical study on a series of V(V) complexes containing variously substituted salen
and salophen ligands is reported with the aim to correlate the activity of the catalysts with the electronic
character of the vanadium center. The results obtained indicate that steric factors play a major role in
determining the outcome of the reaction, often overcoming the electronic effects. Theoretical results
suggest the intervention in the catalytic cycle of an hydroperoxo vanadium species.
Received 8 November 2011
Received in revised form 5 March 2012
Accepted 8 March 2012
Available online 26 April 2012
Keywords:
Catalytc oxidation
Vanadium complexes
Substituent effect
51V NMR
© 2012 Elsevier B.V. All rights reserved.
DFT calculation
1. Introduction
Peroxovanadium complexes, formed upon interaction of the
metal ion with H2O2, depending on the nature of the ligands coor-
Oxidation reactions, and in particular their catalytic versions,
often represent key processes to transform bulk chemicals into
valuable materials. Millions of tons of such compounds are yearly
produced worldwide, and find applications in all areas of chemical
industries [1].
dinated to the metal and on the experimental conditions, can act
either as electrophilic oxygen transfer reagents or as radical oxi-
dants [11]. Typical electrophilic processes are the oxidation of
sulfides and tertiary amines and the epoxidation of allylic alco-
hol or simple alkenes [11–13]. The oxidation of alcohols and
the hydroxylation of aliphatic and aromatic hydrocarbons are
examples of homolytic reactivity [5,12,14]. Several review arti-
cles have appeared recently on vanadium-catalyzed oxidations
[5,15,16].
Additionally, the oxidation of sulfur containing compounds
remains an hot research topic both from a synthetic point of
view [11,17–19] i.e. the preparation of chiral sulfoxides and of
sulfones, as well as considering modern processes for desulfur-
ization of fuels [20]. The removal of sulfur from diesel oil and
gasoline is becoming more and more important, not only to pro-
tect car exhaust catalysts, but also because of the more strict
environment-protecting regulations. Up-to-date, hydrodesulfur-
ization is the method of choice to remove thiols, sulfides, and
disulfides, but with this process it is not possible to efficiently
remove aromatic sulfur-containing compounds, such as dibenzoth-
iophene and 4,6-dimethyldibenzothiophene. To achieve ultra-deep
desulfurization of fuels, oxidation reaction appears to be a promis-
ing procedure [21].
On the basis of economic and environmental reasons [2,3],
oxidation processes with H2O2 are sustainable processes: first of
all H2O2 is, besides dioxygen, the most appealing oxidant and
the one with the highest “atom efficiency” (47% of active oxy-
gen); in addition, aqueous hydrogen peroxide is readily accessible,
safe to use and upon reduction only water is formed [4–7]. Cat-
alytic systems based on different metals have been employed
in conjunction with hydrogen peroxide for oxidation reactions
[5]. Vanadium catalyzed oxidations with hydrogen peroxide have
attracted interest of several research groups because of the inter-
esting properties of this metal in terms of selectivity, reactivity
and stereoselectivity. In this perspective, these reactions are also
interesting because of the relationship with the reactivity of
vanadium-dependent haloperoxidases (VHPO) enzymes, species
able to activate hydrogen peroxide for the oxidation of halides,
thus producing halogenated compounds, and for stereoselective
sulfoxidations of appropriate substrates [8–10].
This background prompted us to explore the catalytic activity
of selected vanadium complexes in oxidation reactions starting
∗
Corresponding author. Tel.: +39 0672594014; fax: +39 0672594328.
E-mail address: valeria.conte@uniroma2.it (V. Conte).
from a model sulfide, Ph
S CH3, selectively to the corresponding
0920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.03.032

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
45
Scheme 1. Vanadium(V) salen and salophen complexes.
sulfoxide. The structure of the catalysts used in this study is
2. Procedures
reported in Scheme 1, they are all oxo vanadium(V) species, lig-
ated with Schiff bases derived from the appropriate substituted
salicylaldehydes and o-phenylendiamine or 1,2-ethanediamine
respectively.
Salophen and salen derivatives are considered “privileged lig-
ands” because their easy synthesis by condensation between
aldehydes and amines, they also coordinates a wide variety of
metals [22]. In addition, stereogenic centers or other elements of
chirality can be also introduced. Moreover after appropriate mod-
ification heterogeneous catalysts [23] can be prepared.
The variation of the coordination sphere was designed in order
to highlight correlation between the electronic distribution on
the metal center with the reactivity in oxidation reactions, sim-
ilar approach has been recently used with vanadium containing
strongly related salen and salan complexes [24].
The catalytic activity of some of the species here studied
was already recently analyzed in our labs, in the epoxidation of
cyclooctene [25]. Quite surprisingly, electron withdrawing groups
in the periphery of the salophen ligands reduced the catalytic activ-
ity of the vanadium complex toward double bonds. However, this
aspect was attributed to faster decomposition of the peroxide. On
the other hand, with oxo vanadium salophen complex possessing
four t-butyl groups in the backbone, the observed reactivity toward
cyclooctene was almost zero. In this latter case the prevalence of
the steric effect over the electronic one is evident, i.e. the bulki-
ness of the ligand heavily hamper the approach of the substrate
to the peroxo complex. Accordingly, in that work [25] no clear cut
correlation between the observed reactivity and the coordination
environment of the metal center was envisaged.
Moreover, it is to be kept in mind that in vanadium catalyzed
oxidations with H2O2 two competitive reactions occur: the oxida-
tion of the substrate and the vanadium catalyzed decomposition
of hydrogen peroxide [26], for that reason a higher activity of a
vanadium complex may simply result in a faster decomposition of
H2O2.
In this paper we present an experimental and theoretical study,
concerning the effects of steric and electronic modification of the
ligands on the catalytic activity of salophen and salen oxo vana-
dium(V) complexes in the oxidation of PhSMe with H2O2. This easy
to oxidize model substrate has been chosen with the aim to obtain
fast and clean oxidation process, good prerequisite for analyzing
a class of catalysts which have already shown ambiguous kinetic
behavior as function of the substituents present in the salophen or
salen scaffolds [25].
2.1. Experimental
Caution: The uncontrolled heating of large amounts of peroxides
must be avoided. Care should be exercised in order to avoid possible
explosion.
2.1.1. Instruments
1H NMR spectra were recorded on a Bruker Avance 300 MHz
spectrometer, with CDCl3, CD3CN or CD3COCD3 as solvents. 51V
NMR spectra were recorded on a Bruker Avance 400 MHz spectrom-
eter, in CD3CN. HPLC analyses were carried out with a Shimadzu
LC-10 ADvp instrument with UV–vis SPDM10Avp detector and
a Kromasil 100 C18 (250 mm × 4.6 mm, 5 ␮m) column with VV
complexes or a Metachem Metaphor 50 ODS-2 (250 mm × 4.6 mm,
5 ␮m) for oxidation reactions. UV–visible spectra were recorded on
a SHIMADZU 2450 spectrometer equipped with the UV Probe 2.34
program.
2.1.2. Materials
HPLC-purity grade MeCN was used. Methanol and
dichloromethane commercial grade purity were purchased
and used as such. Spectroscopic-grade solvents were used for
UV–vis spectra. Commercially available aqueous solution of H2O2
were used after iodometric titration (10.4 0.2 M H2O2 in water).
2.1.3. Synthesis of ligands
The Schiff bases used for the synthesis of the catalysts were
prepared by the well known reaction between salicylaldehyde
and diamine. Slight experimental variations were introduced with
respect to literature methods [27] and the resulting procedure was
successfully applied to a number of differently substituted aldehy-
des.
General procedure: Two equivalents of the appropriate sali-
cylaldehyde were dissolved in the minimum volume of boiling
methanol (generally, 20 ml) and added dropwise with one equiva-
lent of diamine (either 1,2-diaminoethane or 1,2-benzenediamine)
in 5 ml methanol. The solution was refluxed until all the aldehyde
disappeared (TLC analysis) and then cooled to room temperature,
thus causing precipitation of the Schiff base, as a yellow solid.
The filtered solid was washed with a small amount of methanol,
then with diethyl ether and dried. The following Schiff bases were
prepared:

46
A. Coletti et al. / Catalysis Today 192 (2012) 44–55
in CH2Cl2 [ꢀmax, nm (ε, M−1 cm−1)] 252 (41,000), 288 (27,000)
and 386 (3600). UV–vis spectra are consistent with literature data
[35,36].
Salophen, [1,2-bis-(salicylideneamino)benzene]: yield 95.3%;
5,5ꢀ-Cl2salophen,
[1,2-bis-(5-Cl-salicylideneamino)benzene]:
yield >99%; 5,5ꢀ-(t-Bu)2salophen [1,2-bis-(5-t-Bu-salicylidene-
amino)benzene]: yield 75%; 3,3ꢀ-(OMe)2salophen, [1,2-bis-(3-
OMe-salicylideneamino)-benzene]: yield 77%; 5,5ꢀ-(OMe)2
salophen, [1,2-bis-(5-OMe-salicylideneamino)-benzene]: yield
79%; 3,3ꢀ,5,5ꢀ-Cl4salophen [1,2-bis-(3,5-Cl2salicylideneamino)-
benzene]: yield 95%; 3,3ꢀ,5,5ꢀ-(t-Bu)4salophen [1,2-bis-(3,5-di-t-
Bu-salicylideneamino)benzene]: yield 75%.
2.1.5. Synthesis of VV complexes
The synthesis of these complexes was accomplished by a proce-
dure slightly different form that reported in the literature [34]. For
all compounds 1H NMR spectra are in agreement with the structure
(in CD3CN, aromatic protons resonate in the range 7.1–7.8 ppm,
CH N protons in the range 8.2–8.9 ppm, and CH2 protons of
salen complexes are around 4.4 ppm). By comparison with authen-
tic samples, TLC and HPLC analyses excluded the presence of VIV
species.
Salen, [1,2-bis-(salicylideneamino)ethane]: yield 88%; 5,5ꢀ-
Cl2salen, [1,2-bis-(5-Cl-salicylideneamino)ethane]: yield 67%;
5,5ꢀ-(t-Bu)2salen,
yield 91%;
[1,2-bis-(5-t-Bu-salicylideneamino)ethane]:
[1,2-bis-(3-methoxy-salicy-
3,3ꢀ-(OMe)2salen,
lideneamino)ethane]: yield 92%; 5,5ꢀ-(OMe)2salen, [1,2-bis-
(5-OMe-salicylideneamino)ethane]: yield 92%; 3,3ꢀ,5,5ꢀ-Cl4salen,
[1,2-bis-(3,5-Cl2-salicylideneamino)ethane]: yield 70%; 3,3ꢀ,5,5ꢀ-
General procedure: 200 mg of VIVO complex were dissolved in
30 ml CH2Cl2 under stirring. Dioxygen was bubbled 5 min into
the solution kept at 0 ◦C and O2 atmosphere was ensured with a
latex reservoir. 1.2 Equivalents trifluoromethanesulfonic acid were
rapidly added, causing darkening of the solution and precipitation
of a solid. The reaction mixture was allowed to reach room tem-
perature and stirred until disappearance of VIV species (5 ÷ 20 h).
Fine powder of VV complex was isolated after centrifugation of the
reaction mixture (6000 r.p.m.) and decantation of the supernatant
solution.
(t-Bu)4salen,
[1,2-bis-(3,5-(t-Bu)2-salicylideneamino)ethane]:
yield 72%.
All the compounds gave 1H NMR and UV–vis spectra consistent
with the structure and with literature data [25,28–30].
2.1.4. Synthesis of VIV complexes
The synthesis of these complexes was accomplished by a proce-
dure slightly different form that reported in the literature [31,32].
A number of the vanadium derivatives used in the literature were
tested as precursors, i.e. VO(acetylacetonate)2 [31], vanadyl sulfate
di-hydrate [29,33], and V(acetylacetonate)3 [34]. The best results
were obtained with VIII(acac)3, in terms of reproducibility of the
reaction and solubility of complexes.
The following VVO complexes were prepared.
[salophenVVO] CF3SO3, yield 95% UV–vis in MeCN [ꢀmax, nm
(ε, M−1 cm−1)] 242 (40,000), 304 (26,000) and 393 (10,000); [5,5ꢀ-
Cl2salophenVVO] CF3SO3, yield 98%, UV–vis in MeCN [ꢀmax, nm (ε,
M
−1 cm−1)] 303 (21,300), 408 (11,200); [5,5ꢀ-(t-Bu)2salophenVVO]
CF3SO3, yield 35%, UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 245
(42,700), 320 (24,300), 407 (14,400); [3,3ꢀ-(OMe)2salophenVVO]
CF3SO3, yield 75%, UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 220
(64,000), 251 (37,000), 302 (45,000), 310 (42,000), 340 (24,000) and
437 (7200); [5,5ꢀ-(OMe)2salophenVVO] CF3SO3, yield 82%, UV–vis
in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 215 (68,000), 242 (41,000),
292 (46000), 300 (45000), 341 (29000) and 435 (4900); [3,3ꢀ,5,5ꢀ-
Cl4SalophenVVO] CF3SO3, yield 35%, [ꢀmax, nm (ε, M−1 cm−1)] 313
(19200), 410 (13,700); [3,3ꢀ,5,5ꢀ-(t-Bu)4salophenVVO] CF3SO3, yield
87%, [ꢀmax, nm (ε, M−1 cm−1)] 244 (30,700), 326 (35,300), 416 sh
(9300).
General procedure: The Schiff base was dissolved in 100 ml
of boiling methanol, or suspended when scarcely soluble. The
equimolar amount of V(acac)3 was completely dissolved in the
minimum volume of MeOH with the help of sonication and added
dropwise to the solution (or to the suspension), causing immedi-
ate color change from yellow to green. After an overnight stirring
in an open vessel at room temperature, the reaction was stopped
and the precipitated solid was collected, washed with diethyl ether,
and dried. No trace of Schiff base was present. Eventually unreacted
V(acac)3 was washed off with warm acetone.
The following VIVO complexes were prepared, their purity was
checked with TLC and HPLC analyses.
[salenVVO] CF3SO3, yield 89%, UV–vis in MeCN [ꢀmax, nm (ε,
M
−1 cm−1)] 230 (30,000), 296 (14,000) and 347 (7400); [5,5ꢀ-
SalophenVIVO, yield 73%, UV–vis in MeCN [ꢀmax
,
nm (ε,
−1 cm−1)] 242 (40,000), 314 (22,000) and 396 (18,000); 5,5ꢀ-
Cl2salophenVIVO, yield 78%, UV–vis in MeCN [ꢀmax
nm (ε,
−1 cm−1)] 305 (12,000), 409 (10,700); 5,5ꢀ-(t-Bu)2salophenVIVO,
Cl2salenVVO] CF3SO3, yield 67%, UV–vis in MeCN [ꢀmax, nm (ε,
−1 cm−1)] 230 (30,000), 248 (24,000), and 285 (17,000); [5,5ꢀ-
M
M
,
(t-Bu)2salenVVO] CF3SO3, yield 5%, UV–vis in MeCN [ꢀmax, nm
(ε, M−1 cm−1)] 230 (33,400), 250 (27,500) 290 (19,000) and
348 (7500); [3,3ꢀ-(OMe)2salenVVO] CF3SO3, yield 58%, UV–vis in
MeCN [ꢀmax, nm (ε, M−1 cm−1)] 281 (12,000), 308 (10,000) and
359 (4200); [5,5ꢀ-(OMe)2salenVVO] CF3SO3, yield 89%, UV–vis
in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 288 (12,000) and 391 sh
(3000); [3,3ꢀ,5,5ꢀ-Cl4salenVVO] CF3SO3, yield 66%, UV–vis in MeCN
[ꢀmax, nm (ε, M−1 cm−1)] 293 sh (1300) and 345 sh (710);
3,3ꢀ,5,5ꢀ-(t-Bu)4salenVVO, yield 80%. UV–vis in CH2Cl2 [ꢀmax, nm
(ε, M−1 cm−1)] 263 (8000), 307 (5600) and 367 sh (2900).
Cyclic voltammetry (CV) characterization reported elsewhere
[37] confirmed the electronic effect of the substituents on the vana-
dium centers in these salenVIVO and salophenVIVO complexes. The
higher redox potentials of the salophen complexes with respect to
the salen ones indicate that the aromatic ring confers to vanadium
a stronger Lewis acidity.
M
yield 82%, UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 246 (40,900),
318 (22,700), 409 (15,400); 3,3ꢀ-(OMe)2salophenVIVO, yield 79%,
UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 221 (42,000), 301
(24,000), 313 (22,000) and 335 (13,000); 5,5ꢀ-(OMe)2salophenVIVO,
yield 87%, UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)] 216 (70,000),
243 (41,000), 289 (30,000), 300 (32,000), 337 (26,000) and 434
(9000); 3,3ꢀ,5,5ꢀ-Cl4SalophenVIVO: yield 78%, UV–vis in MeCN
[ꢀmax, nm (ε, M−1 cm−1)] 315 (9800), 412 (10,000); 3,3ꢀ,5,5ꢀ-
(t-Bu)4salophenVIVO, yield 87%, UV–vis in MeCN [ꢀmax, nm (ε,
M
−1 cm−1)] 250 (43,300), 327 (25,900), 416 (16,100).
SalenVIVO, yield 89%, UV–vis in MeCN [ꢀmax, nm (ε, M−1 cm−1)]
242 (39,000), 277 (18,000) and 362 (7900); 5,5ꢀ-Cl2salenVIVO,
yield 67% UV–vis in MeCN [ꢀmax, nm] 248 (49,000), 280 sh,
370 (7400); 5,5ꢀ-(t-Bu)2salenVIVO, yield 67%, UV–vis in MeCN
[ꢀmax, nm (ε, M−1 cm−1)] 246 (56,000), 278 (27,000), 370 (9200);
3,3ꢀ-(OMe)2salenVIVO, yield 93%, UV–vis in MeCN [ꢀmax, nm (ε,
M
−1 cm−1)] 224 (16,000), 296 (14,000) and 381 (2800); 5,5ꢀ-
(OMe)2salenVIVO, yield 86%, UV–vis in MeCN [ꢀmax
nm (ε,
2.1.6. Oxidation of phenyl methyl sulfide with H2O2, catalyzed by
VV complexes
The catalytic activity of all the complexes was tested in a probe
reaction, i.e. oxidation of PhSMe with H2O2 in MeCN. The reac-
tions were performed with the same initial concentration of all the
,
M
−1 cm−1)] 251 (20,000), 286 sh (8800) and 392 (9000); 3,3ꢀ,5,5ꢀ-
Cl4salenVIVO, yield 70%, UV–vis in MeCN: ꢀmax = 370 nm (does not
dissolve completely); 3,3ꢀ,5,5ꢀ-(t-Bu)4salenVIVO, yield 72%, UV–vis

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
47
reactant species, at 25 ◦C in a thermostated vessel. Product analysis
was obtained by HPLC, to avoid thermal decomposition of prod-
ucts, adding a known amount of toluene as internal standard and
using the response factors obtained by calibration straight lines.
Stock solutions of sulfide and toluene in MeCN were used. The lat-
ter was periodically checked for the presence of oxidation product
that were always absent. Solutions of H2O2 and the catalyst were
prepared immediately before use.
0.5 ml of PhSMe 0.151 M in MeCN and 0.15 ml of catalyst
2.2 × 10−3 M in the same solvent were added to 5 ml MeCN. 0.5 ml of
the resulting solution were taken and diluted to 5 ml in a volumet-
ric flask containing 2 ml of PhMe 1.2 × 10−3 M in MeCN, to measure
initial sulfide concentration. The reaction started after addition of
0.18 ml freshly prepared H2O2 0.145 M in MeCN (obtained diluting
to 5 ml 70 ␮l 10.4 0.2 M H2O2 in water). Samples were taken at
selected reaction times (2, 5, 15, and 40 min) and analyzed at the
HPLC.
broader signals were detected in 1H NMR spectra with salen com-
plexes: i.e. signal for CH N at 9.23 ppm shows a ꢁ␯ = 7 Hz for
½
[3,3ꢀ,5,5ꢀ-(t-Bu)4salophenVVO] CF3SO3 while for the correspond-
ing [3,3ꢀ,5,5ꢀ-(t-Bu)4salenVVO] CF3SO3 derivative the same CH
N
signal at 8.86 ppm shows a ꢁ␯ = 22 Hz.
½
The model reaction has been the oxidation with H2O2 of PhSMe
to the corresponding sulfoxide, in MeCN at room temperature, the
ratio H2O2: Cat was ca. 50:1. A two-fold excess of the substrate
with respect to the oxidant was used in order to ensure selectiv-
ity toward the sulfoxide and also to keep roughly constant the
concentration of the substrate, at least at the beginning of the
reaction.
The sulfide consumption and the sulfoxide formation were
determined with quantitative HPLC analysis of the reaction mix-
ture, see Section 2 for details. Sulfide consumption and sulfoxide
yield, at selected reaction time, were taken as parameters to
compare the reactivity of the various catalysts, the two values
(see Tables) are equal within the experimental error 5%. The
data concerning the results with salophen complexes are col-
lected in Table 1 whilst those related to salen species are in
Table 2.
The Tables also contain the ꢂ␴ [42] which is the arithmeti-
cal sum of the ꢃ values for the substituents introduced into the
salophen and salen scaffolds. These values are given to sketch out
the electronic effect dictated by their presence. For the complexes
used, it can be noted that, spanning from a ꢂ␴ of about +1, due to
the presence of four chlorine atoms, to a value of about −0.8, when
four t-butyl groups are present, a large variation of the electronic
character of the metal ion, i.e. its Lewis acidity, should be obtained.
Looking at the data collected in the two tables some general
comments can be done: the catalytic effect of the salophen and
salen oxo vanadium(V) complexes is quite good; the reaction times
are short, most of the oxidations are complete in 15 min at room
temperature. This last observation indicates that the vanadium cat-
alyzed decomposition of H2O2 is not competing with the oxygen
transfer reaction.
However (see Fig. 1), no simple correlation can be detected
between the ꢂ␴ and the sulfoxide yields in both salophen and salen
series; additionally, no simple comparison can be made in equally
substituted couples of salophen and salen derivatives; this behavior
being indeed expected on the basis of our previous studies [25].
Therefore, a deeper insight of the results is necessary in order to
highlight, for each catalyst, if electronic or steric effects are playing
the major role in determining the reactivity.
2.2. Calculations
Pseudopotential LACVP was used for the energy optimization
without symmetry constraints. Gaussian 03 program was used
for geometry optimizations with the continuum solvent model
[38]. Geometry optimizations were performed with Density Func-
tional Theory (DFT) calculations using the B3LYP functional with
the 6-31G* basis set [39,40]. Geometries were optimized at the
B3LYP/LANL2DZ level, including solvent effects with the IEF-PCM
method (MeCN).
All the in vacuum quantum-mechanical calculations were car-
ried out by using Spartan’08 [41].
3. Results and discussion
3.1. Experimental
3.1.1. Reactivity
The complexes prepared and used in this work contained Schiff
base ligands possessing various substituents in the periphery of the
aromatic scaffold, see Scheme 1.
All salophen and salen molecules have been prepared
following literature procedures [27] starting from the appro-
priately substituted salicylaldehydes and o-phenylendiamine or
1,2-ethanediamine respectively. The various ligands have been
subsequently used to synthesize all VIV and VV complexes in very
good yields (see Section 2), following slightly modified literature
procedures [27,31,32,34] and the species obtained have been iden-
tified by comparison of their properties with literature data.
For comparison purposes, the VIV species, obtained in the first
step of the synthesis of vanadates derivatives, were also tested as
catalysts in the oxidation reactions, and the collected data were
comparable to those observed (and discussed below) with VV com-
plexes. However, the possible incursion of radical reactions, an
expected event when VIV derivatives are used as catalysts in reac-
tions with hydrogen peroxide [11], prompted us to examine the
reactivity in oxidation reaction with hydrogen peroxide only with
the catalysts in their highest oxidation state. In such a way, mini-
mization of the vanadium catalyzed decomposition of H2O2 should
occur.
Interestingly, the catalytic activity of substituted salophen com-
plexes, measured as sulfoxide yield at a specific reaction time (left
graph), roughly follows the trend of the ꢂ␴ values with the excep-
tion of the 3,3ꢀ-(OMe)2 derivative, this deviation from the expected
result will be discussed below. The observed tendency apparently
indicates that the reactivity somewhat follows the Lewis acidity
character of vanadium, ꢂ␴ +0.9 for 3,3ꢀ,5,5ꢀ-Cl4-salophen.
On the other hand, with substituted salen derivatives (right
graph) no simple correlation can be established between the sul-
foxide yields and the ꢂ␴ values; noteworthy, also in this series
the 3,3ꢀ-(OMe)2 derivative shows the highest reactivity. In addition,
within this catalysts series high reactivity is observed with 5,5ꢀ sub-
stituted ligands independently from their inductive effect, whereas
for 3,3ꢀ,5,5ꢀ-tetra substituted species the reactivity appears to track
the electron density around vanadium.
The various substituents introduced in the scaffold of the com-
plexes were chosen with the scope to modify the electronic
character of the VV ion. Furthermore, a consistent compari-
son between the salophen and salen structure should help in
divide the electronic from the structural effects, being the for-
mer species forced into a simil-planar configuration, while the
latter can experience a fluctional behavior [22]. Accordingly,
At this point of the discussion, a better understanding of the
results obtained in the reactivity studies can be likely obtained
looking also at the classical reaction mechanism [3–6,11] proposed
for metal catalyzed sulfides oxidation with peroxides.
The accepted reaction mechanism for vanadium catalyzed oxi-
dation processes with hydrogen peroxide requires the formation
of a peroxide vanadium complex upon coordination of H2O2 to the

48
A. Coletti et al. / Catalysis Today 192 (2012) 44–55
Table 1
V-salophen (0.00056 mmol) catalyzed oxidation of phenyl methyl sulfide (0.059 mmol) with H2O2 (0.026 mmol) in CH3CN (5.33 ml) at room temperature.
O
S
S
MeCN, 25°C
H2O2
+
cat.
5.6 x 10-4 mmol
0.026 mmol
0.059 mmol
b
Catalyst
Substituent
3,3ꢀ,5,5ꢀ-Cl4
t, min
% Sulfide consumed
% Sulfoxide formeda
ꢂ␴
2
5
15
93
93
94
94
96
94
0.908
O
N
N
V
Cl
O
O
Cl
CF3SO3
Cl
Cl
2
5
15
87
91
94
92
94
92
5,5ꢀ-Cl2
0.454
O
N
N
O
V
Cl
O
Cl
CF3SO3
2
5
15
54
90
95
56
98
100
H
0.000
O
V
N
O
N
O
CF3SO3
2
5
15
12
14
54
1.2
4.5
53
5,5ꢀ-(t-Bu)2
3,3ꢀ-(OMe)2
5,5ꢀ-(OMe)2
−0.394
−0.536
−0.536
O
V
N
N
O
O
CF3SO3
2
5
15
54
89
97
48
88
96
O
V
N
O
N
O
CF3SO3
O
O
2
5
15
27
40
70
9
26
62
O
N
O
N
O
V
O
O
CF3SO3

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
49
Table 1 (Continued)
b
Catalyst
Substituent
t, min
% Sulfide consumed
% Sulfoxide formeda
ꢂ␴
2
5
15
18
24
20
0
1.5
6
3,3ꢀ,5,5ꢀ-(t-Bu)4
−0.788
O
V
N
O
N
O
CF3SO3
a
Sulfide consumption and sulfoxide formation corresponds within the experimental error.
ꢂ␴ is the arithmetical sum of all the ꢃ values [42].
b
transfer to the substrate. On the contrary, when the ligand has a bent
conformation, like in the case of salenVV(O)(OOH) (see Fig. 2c) the
approach of the second oxygen to the metal is hampered as shown
by the longer distance in the calculated structure, thus preventing
the stabilization of the transition state.
Summarizing, theoretical calculations suggest that salophen
ligands force the hydroperoxovanadium complex in a planar con-
formation that allows the stabilization of the transition state of
the oxidation step, resulting in a higher reactivity with respect to
salen complexes, where the more stable bent conformation can be
adopted.
Returning to the experimental data, the higher reactivity of the
salophen derivatives, together with the observed relationship of
the sulfoxide yields with the Lewis acid character of the vanadium
center, appears to be related to the planar arrangement of such
species which in turn is dictated by the structure of the ligands.
On the other hand, with salen derivatives, which can adopt several
different conformations because of the mobility of the ligand, the
steric hindrance of the substituents combines, in an unpredictable
way, with the electronic character of the catalyst.
Scheme 2. Proposed catalytic cycle for sulfides oxidation with H2O2. L stands for
salen or salophen derivatives.
metal center, such intermediate is then the bona fide oxidant of
the substrate. For the sake of simplicity, we anticipate here that
this active species likely is an hydroperoxo one. The coordination
of the oxidant to the Lewis acid metal center activates the oxygen
toward the nucleophilic attack of an electron rich substrate, such
as the phenyl methyl sulfoxide. In the transition state of the oxi-
dation step, coordination of the second peroxidic oxygen atom to
vanadium helps the oxygen transfer to the substrate, subsequent
elimination of the sulfoxide and a water molecule regenerates the
catalyst precursor (Scheme 2).
The 3,3ꢀ dimethoxy substituted salophen and salen oxo-
vanadium complexes deserve a specific discussion. In fact, the
catalytic activity of [3,3ꢀ-(OMe)2salophenVVO]CF3SO3 and [3,3ꢀ-
(OMe)2salenVVO]CF3SO3 is much higher than that of all the other
derivatives in the corresponding series (see Fig. 1). This behavior
may suggest an easier formation of the metalperoxide species. In
that respect, as indicated in Fig. 3, the presence of 3,3ꢀ-(OMe)2 sub-
stituents may stabilize the approach of a H2O2 molecule to the
metal center, through the intervention of an hydrogen bond.
With salophen and salen species formation of the peroxidic
intermediate can be strongly influenced by the structure of the
precursor. In order to consider this structural aspect, energy mini-
mization of the peroxidic species, formed from salophen and salen
VVO precursors has been carried out and the outcome for the
unsubstituted hydroperoxo oxovanadium complexes (the plausi-
ble active oxidant species) is shown in Fig. 2.
For example, salophen ligands force the precursor and the pos-
sible peroxo complexes in a planar conformation, i.e. the ligand
resides on the plane of a simil-tetrahedrical shape (see Fig. 2a). On
the other hand with salen ligands, together with the planar confor-
mation (Fig. 2b) also a bent one can be adopted, in which one of the
nitrogen of the ligand is trans to the oxo group (see Fig. 2c). It is
interesting to note that from theoretical calculation (see following
paragraphs), this latter conformation appears to be favorite. The gas
phase optimized salenVV(O)(OOH) and salophenVV(O)(OOH) struc-
tures show that the hydroperoxo group has a different approach to
the metal depending on the conformation of the ligand. In partic-
ular, when the ligand has a planar conformation, like in the case
of salophenVV(O)(OOH), the peroxo group almost adopt the side-
on configuration around the metal, thus facilitating the oxygen
3.1.2. 51V NMR studies
It is well known that the number and the nature of the perox-
ometal species strongly depends on the reaction conditions, such as
the pH value, the ligand and the solvent. Since it is not always pos-
sible to obtain peroxovanadium complexes in the solid state, and
often their solution structure is different, their structure is often
investigated by means of several techniques. For example, 51V NMR
and electrospray ionization mass spectrometry (ESI-MS) or theo-
retical calculations have been recently used in order to understand
their behavior in catalytic cycles [43].
The most common structures of monoperoxovanadium com-
plexes are the side-on cyclic peroxo species, in which both
the oxygen atoms are coordinated to vanadium atom, and the
hydroperoxo species, in which only one oxygen is covalently linked
to the metal center. In this work we have tried to obtain infor-
mation on the formation in solution of the peroxo derivatives
from salophen and/or salen V(V) species, in the presence of dif-
ferent excesses of hydrogen peroxide. 51V NMR spectra have been
acquired for some of the vanadium(V) precursors in acetonitrile
and, in all cases, single peaks appeared around −600 ppm. To note,

50
A. Coletti et al. / Catalysis Today 192 (2012) 44–55
Table 2
V-salen (0.00056 mmol) catalyzed oxidation of phenyl methyl sulfide (0.059 mmol) with H2O2 (0.026 mmol) in CH3CN (5.33 ml) at room temperature.
O
S
S
MeCN, 25°C
H2O2
+
cat.
5.6 x 10-4 mmol
0.026 mmol
0.059 mmol
b
Catalyst
Substituents
3,3ꢀ,5,5ꢀ-Cl4
t, min
% Sulfide consumed
% Sulfoxide formed a
ꢂ␴
2
5
15
12
18
37
9
12
35
O
N
N
0.908
V
Cl
Cl
O
O
Cl
CF3SO3
Cl
Cl
2
5
15
76
98
96
76
99
98
5,5ꢀ-Cl2
0.454
O
N
O
N
O
V
Cl
CF3SO3
2
5
15
8
10
18
1
4
15
O
N
O
N
O
H
0.000
V
CF3SO3
2
5
15
92
88
92
94
98
96
O
5,5ꢀ-(t-Bu)2
3,3ꢀ-(OMe)2
5,5ꢀ-(OMe)2
3,3ꢀ,5,5ꢀ-(t-Bu)4
−0.394
−0.536
−0.536
−0.788
N
O
N
O
V
CF3SO3
2
5
15
>99
>99
>99
>99
>99
>99
O
V
N
N
O
O
CF3SO3
O
O
2
5
15
34
62
–
34
64
93
O
N
N
V
O
O
O
O
CF3SO3
2
5
15
92
88
92
94
98
96
O
N
V
N
O
O
CF3SO3
a
Sulfide consumption and sulfoxide formation corresponds within the experimental error.
ꢂ␴ is the arithmetical sum of all the ꢃ values [42].
b

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
51
Fig. 1. PhSOMe yields vs. ꢂ␴ of the ligands. Left: V-salophen catalyzed oxidation of PhSMe with H2O2 in CH3CN at room temperature at 2 min (experimental conditions
reported in Table 1). Right: V-salen catalyzed oxidation of PhSMe with H2O2 in CH3CN at room temperature at 15 min (experimental conditions reported in Table 2).
Fig. 2. Gas phase optimized structures of LVV(O)(OOH) complexes (L = salen or salophen). (a) Planar conformation of salenVV(O)(OOH). (b) Planar conformation of
salophenVV(O)(OOH). (c) Bent conformation of salenVV(O) (OOH).
the chemical shifts for salen and salophen species are very similar,
despite of the presence of different substituents on the aromatic
scaffold.
catalyst precursor, over a wide range of chemical shift (as an
example, see Fig. 4 for [salenVVO]CF3SO3). Peroxovanadium species
usually give signals at high fields [44]; in this case, the absence of
any signal may be due to the low solubility of peroxometal species
at room temperature. In fact, the formation of a yellow precipitate
was observed upon addition of a large excess of H2O2 to a solution
of salenVO+ in acetonitrile.
Upon addition of increasing amounts of H2O2 to a solution of
[salophenVVO]CF3SO3 or [salenVVO]CF3SO3, the signal of the pre-
cursor decreases and, for both complexes, no signals appear having
reference to vanadium free ligand species; in general, no other 51
V
signals were present in NMR spectra different from that of the
Fig. 4. 51V NMR spectra of [SalenVVO]CF3SO3 in acetonitrile (1 mM). (a) Initial solu-
tion (b) after addition of 2 equivalents of H2O2 to the previous solution; (c) after
addition of 160 equivalents of H2O2 to the initial solution.
Fig. 3. Plausible formation of an hydrogen bond between an approaching H2O2
molecule and the methoxy oxygens in [3,3ꢀ-(OMe)2salophenVVO]CF3SO3 complex.

52
A. Coletti et al. / Catalysis Today 192 (2012) 44–55
Table 3
Table 4
51V NMR chemical shifts for different(V) complexes in CH3CN.
Energetically preferred conformations for salenVV complexes as derived by DFT
calculations.
Catalyst
␦
51V NMR (ppm)
salenVV complex
Bent
Planar
ꢁGa (kcal mol−1
)
[salenVVO]CF3SO3
−597
−596
−584
−604
−595
[5,5ꢀ-(Cl)2salenVVO]CF3SO3
[5,5ꢀ-(t-Bu)2salenVVO]CF3SO3
[salophenVVO]CF3SO3
[salenVV(OH)(OOH)]+
[salenVVO(OOH)]
[salenVV(OO)(OH)]
[salenVVO(OO)]−
×
2.3
6.7
1.0
×
×
×
[3,3ꢀ,5,5ꢀ-(t-Bu)4salophenVVO]CF3SO3
14.6
a
Free energy difference between most and less stable conformations for each
species.
The IR spectrum of the yellow precipitate showed no signals
in the range of 918–954 cm−1 were the O O stretching is usually
present for side-on monoperoxo vanadium complexes [12], thus
supporting the hypothesis that salen and salophen oxovanadium
complexes form an end-on hydroperoxidic species in the presence
of H2O2 (see Section 3 below).
In such scheme the original state of the catalyst was consid-
ered to be a LVVO+ species (L stands for salen or salophen ligand)
originated from dissociation of LVVOCF3SO3 in solution; addition
of H2O2 to LVVO+ to form LVV(OH)(OOH)+ species were found to
be endothermic processes, with ꢁG = +18.1 and +21.4 kcal mol−1
for L = salen and salophen, respectively. For this reason, and con-
sidering also the steric hindrance of the ligands around the metal,
only the formation of monoperoxo species was taken into account.
Moreover, the fact that addition of H2O2 are endoergonic processes,
suggest that only small amounts of vanadium peroxo complexes
form in solution; this result is consistent with 51V NMR spectra,
where no signals different from the one of the precursor have been
detected.
Subsequent calculations were aimed at establishing the reactiv-
ity of the LVV(OH)(OOH)+ species. A crucial observation is that such
intermediates behave as strong acids, by releasing a proton from
the hydroxo group to form LVV(O)(OOH) derivatives (deprotona-
tion step in Scheme 4). This resulted by calculation of the pKa of the
species, by comparison with a reference acid/base couple according
to Wang et al. [48]. This procedure avoids direct calculation on H3O+
species, that are subject to relevant solvent effects, difficult to eval-
uate employing a continuum approximation. In this particular case,
the pyridinium/pyridine couple (pKa = 5.25) was chosen as the ref-
erence, in order to maintain the same total charge of the acid/base
couple under investigation, LVV(OH)(OOH)+/LVV(O)(OOH).
Calculated pKa for the LVV(OH)(OOH)+/LVV(O)(OOH) couples
were found −6.9 and −4.6 for salen and salophen derivatives,
respectively, thus confirming high acid behavior of the hydroxo
group coordinated to the vanadium center.
When salenVVOBr was used as precursor, 51V NMR spectra in
the presence of increasing amount of hydrogen peroxide showed
disappearance of the signal at −595 ppm, already after the addition
of ten equivalents of H2O2. In this conditions no formation of the
yellow precipitate was observed. This suggests that the formation
of the peroxovanadium species is more favorable when the less
tightly coordinated Br− is the counterion in place of CF3SO3−. In
agreement with such observation, when salenVVOBr was tested as
catalyst, a faster decomposition of hydrogen peroxide is observed
[5,26].
3.1.3. Theoretical studies
The unclear results obtained with the NMR spectroscopy
prompted us to apply DFT theoretical calculations (see Section 2.1
for details) in order to shed more light on the structure of the reac-
tive peroxo complexes formed in solution, when salophen and salen
vanadium(V) precursors are used.
Therefore, calculation on relevant species were performed at
B3LYP/lanl2dz level, including solvent effects with the IEF-PCM
method (acetonitrile). These include vanadium hydroperoxides
and Á2 peroxides, in different protonation forms, and in the pres-
ence of a residual oxo or hydroxo ligand (Scheme 3). Frequency
calculations were performed in order to confirm that the equilib-
rium structures were actual minimum energy species (Table 3).
Optimized distances between the vanadium center and the
coordinated oxygen of the peroxide moiety fall in the range
The LVV(O)(OOH) species were confirmed to be the most
probable ones, since possible isomers were observed at higher
energy. Indeed, LVV(OH)(OO) species are higher in energy by
12.5 and 4.1 kcal mol−1 for salen and salophen, respectively,
while a H-LVV(O)(OO) derivative (where the proton is located
at the oxygen atom of the L ligand) is 2.2–8.7 kcal mol−1 higher
when L = salophen (the same structure with L = salen failed to
converge). Protonation of the peroxide species, with no involve-
ment of the oxo group, was recently demonstrated by XANES
˚
1.82–2.05 A, while oxygen–oxygen distances of the peroxide group
˚
are found between 1.46–1.51 A; both are in good agreement with
literature values [45,46]. In general, a decrease of the V O(O) dis-
tance is observed increasing the positive charge of the species
[45,46]. When the peroxide is protonated (vanadium hydroperox-
ides), the protonated oxygen is not located at binding distance from
the vanadium center, confirming a monodentate mode of coordina-
tion of the O O group to the metal; differently, when the peroxide
moiety is not protonated both oxygen atoms are coordinated to
vanadium, in a Á2 mode [47].
As illustrated in Fig. 2, only planar conformations were consid-
ered for salophen complexes, while bent and planar conformations
were explored for all the salenVV species. To note, as reported
in Table 4, some of the salenVV derivatives are more stable in
the bent conformation, and in particular the V(O)(OOH) peroxo
complex, which is the reasonable active species responsible for
oxygen transfer (vide infra).
spectroscopy for
a
[HheidaV(O)(OO)]− complex (heida N-(2-
hydroxyethyl)iminodiacetic acid) to be the key step to activate the
species for oxygen transfer catalysis [49].
An alternative pathway (not represented in Scheme 4), where
water elimination from the LVV(OH)(OOH)+ leads to the forma-
tion of a LVV(OO)+ species was found to be endothermic by
5.4 kcal mol−1 in the case of L = salen, confirming that vanadium
complexes without oxo or hydroxo moieties are not stable [12]:
The accepted mechanism, already proposed in the literature, for
the formation of peroxidic species resulting from the reaction of
vanadium species with hydrogen peroxide [43,45,46] has been used
as the base to analyze the energetic path leading to the formation
of the peroxo species possibly responsible for oxygen transfer, see
Scheme 4.
Therefore, calculations suggest that
a
vanadium oxo-
hydroperoxo species, LVV(O)(OOH), is the most plausible
intermediate formed upon H2O2 addition to vanadium Salen
and Salophen complexes, followed by one proton removal. A

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
53
Scheme 3. Representation of relevant species considered in DFT calculations, with optimized interatomic distances indicated in black and red for salen and salophen species,
respectively. The salen or salophen ligand is represented as “L” for clarity reasons.
Scheme 4. Energies for the formation pathways of various vanadium based peroxo complexes with L = salen (black values) or salophen (red values). The energy of the
reactions involving H+ ions were not directly calculated, but determined indirectly from the calculation of the pKa of the competent species. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of the article.)
potential, second deprotonation of LVV(O)(OOH) to give an anionic
LVV(O)(OO)− species was also considered; in these cases calculated
pKa are 4.1 and 7.8 for L = salen and salophen, respectively (in this
case acetic acid/acetate has been used as the reference acid/base
couple, in order to preserve the charges of the species in the
equilibrium process). It is interesting to notice that the calculated
pKa compare reasonably with the values of 5.4–5.9 observed for
the vanadium-dependent haloperoxidases enzymes [50,51]. Nev-
ertheless, deprotonated peroxides such as LVV(O)(OO)− are known
to be poorly active in oxygen transfer [45,46,49], therefore, even
if they could actually be formed in solution according to the pKa
values discussed above, they are not the reasonable catalytically
active species in the present system.
A final argument concerns the possible oxygen transfer activity
of these LV(O)(OOH) peroxides, namely if they can be the actual
species responsible for sulfide oxidation.
As discussed above, the optimized geometries of LV(O)(OOH)
peroxides strongly suggest a monodentate binding of the perox-
ide to the vanadium center, and it occurs with the non protonated
peroxidic oxygen, while the protonated oxygen is not located at
˚
˚
binding distance from the vanadium (V O(H) = 2.962 A and 2.520 A
for L = salen and salophen, respectively). This differs from other
vanadium hydroperoxo species reported in the literature, where
the calculated V O(H) distance is much shorter [45,46] and can be
related to the higher steric hindrance of salen and salophen ligands.
As a result, two different scenario can be envisaged: (a) the
LV(O)(OOH) are the actual species for oxygen transfer; indeed
the O O ␴* orbitals, usually recognized as the one giving the
most important contribution in the electrophilic oxidative activ-
ity of metal peroxides [53–55], are found at reasonable energy
levels for both species (Fig. 5); (b) the LV(O)(OOH) species are in
equilibrium with other forms, where also the second protonated
oxygen is coordinated to vanadium, and these latter species are
responsible for oxygen transfer:
Therefore, the presence of the LVV(O)(OOH) peroxo complexes
in solution is the most reliable scenario. This is in agreement with
results found in the literature for similar vanadium complexes with
model salan ligands [52], where the oxo hydroperoxo vanadium
derivatives appears to be the most favorable active species in alkene
epoxidation processes. In such paper, the salan ligand is coordi-
nated to V atom in a bent conformation, in which two O and one
N occupy the equatorial positions, while the remaining O is in the
axial one; this conformation is consistent with the bent arrange-
ment of the ligand around the vanadium atom that we found for
salenVO peroxo complexes.
In the case of salenV(O)(OOH) peroxocomplex, the energies
for the bent and planar conformations of all the 5,5ꢀ-substituted
derivatives were calculated in the gas phase in order to under-
stand whether the presence of substituents could affect their
conformations. The comparison between the two conformations,
represented by their energy differences, is reported in Table 5.
Interestingly, in all cases the most favorable structure is the bent
one. For the unsubstituted salen complex, energies in acetoni-
trile as the model solvent were evaluated, obtaining comparable
results.
Table 5
Differences of energies for planar and bent conformations of differently substituted
salenV(O)(OOH) species.
salenV(O)(OOH) complex
ꢁE (bent-planar) (kcal mol−1
)
[5,5ꢀ-(t-Bu)2salenVVO(OOH)]
[5,5ꢀ-(MeO)2salenVVO(OOH)]
[salenVVO(OOH)]
−7.71
−8.79
−7.46 (−6.7 in CH3CN)
−7.84
[5,5ꢀ-(Cl)2salenVVO(OOH)]

54
A. Coletti et al. / Catalysis Today 192 (2012) 44–55
[11] V. Conte, O. Bortolini, in: Z. Rappoport (Ed.), The Chemistry of Peroxides –
Transition Metal Peroxides Synthesis and Role in Oxidation Reactions, Wiley
Interscience, 2006, pp. 1053–1128.
[12] V. Conte, B. Floris, Dalton Transactions 40 (2011) 1419–1436, and references
therein.
[13] M.L. Ramos, L.L.G. Justino, H.D. Burrows, Dalton Transactions 40 (2011)
4374–4383, and references therein.
[14] See as an example M.V. Kirillova, M.L. Kuznetsov, V.B. Romakh, L.S. Shul’pina,
J.J.R. Fráusto da Silva, A.J.L. Pombeiro, G.B. Shul’pin., Journal of Catalysis 267
(2009) 140–167.
[15] D.C. Crans, J.J. Smee, E. Gaidamauskas, L. Yang, Chemical Reviews 104 (2004)
849–902.
[16] J.A.L. da Silva, J.J.R. Frausto da Silva, A.J.L. Pombeiro, Coordination Chemistry
Reviews 255 (2011) 2232–2248.
Fig. 5. Representation of the O O ␴* orbital for SalenV(O)(OOH) (LUMO + 6) and
SalophenV(O)(OOH) (LUMO + 10) species.
[17] C. Bolm, Coordination Chemistry Reviews 237 (2003) 245–256.
[18] E. Wojaczyska, J. Wojaczyski, Chemical Reviews 110 (2010) 4303–4356.
[19] J. Gätjens, B. Meier, Y. Adach, H. Sakurai, D. Rehder, European Journal of Inor-
ganic Chemistry (2006) 3575–3585.
Equilibria of this type have already been considered for other
V(V) species, where the Á2 coordination of the hydroperoxide was
necessary to observe oxygen transfer activity and the free energy
of the reaction was influenced by the presence of Lewis bases as
coligands [56]. Equilibria of this type are currently under evaluation
for salen and salophen derivatives. However, being the distance
of the protonated oxygen shorter in the salophen complexes, they
may likely establish more easily such equilibrium. On this basis
the higher activity of salophen complexes with respect to salen
ones could be explained, further support to this is given by their
observed higher Lewis acidity as measured with CV experiments
[37].
[20] C. Song, Catalysis Today 86 (2003) 211–263.
[21] P.S. Kulkarni, C.A.M. Afonso, Greem Chemistry 12 (2010) 1139–1149.
[22] P.G. Cozzi, Chemical Society Reviews 33 (2004) 410–421.
[23] A. Zulauf, M. Mellah, X. Hong, E. Schulz, Dalton Transactions 39 (2010)
6911–6935.
[24] P. Adão, M.R. Maurya, U. Kumar, F. Avecilla, R.T. Henriques, M.L. Kusnetsov, J.
Costa Pessoa, I. Correia, Pure and Applied Chemistry 81 (2009) 1279–1296.
[25] V. Conte, F. Fabbianesi, B. Floris, P. Galloni, D. Sordi, I.W.C.E. Arends, M. Bonchio,
D. Rehder, D. Bogdal, Pure and Applied Chemistry 81 (2009) 1265–1277.
[26] M. Bonchio, V. Conte, F. Di Furia, G. Modena, S. Moro, J.O. Edwards, Inorganic
Chemistry 33 (1994) 1631–1637.
[27] C.P. Horwitz, P.J. Winslow, J.T. Warden, C.A. Lisek, Inorganic Chemistry 32
(1993) 82–88.
[28] N.F. Choudhary, N.G. Connelly, P.B. Hitchcock, G.J. Leigh, Journal of the Chemical
Society, Dalton Transactions (1999) 4437–4446.
[29] M.P. Webereski Jr., C.C. McLauchlan, C.G. Hamaker, Polyhedron 25 (2006)
119–123.
Full mechanistic study, including evaluation of transition states,
is in progress and will be reported in a separate communication.
[30] A.L. Singer, D.A. Atwood, Inorganica Chimica Acta 277 (1998) 157–162.
[31] C.J. Chang, J.A. Labinger, H.B. Gray, Inorganic Chemistry 36 (1997) 5927–5930.
[32] J.A. Bonadies, C.J. Carrano, Journal of the American Chemical Society 108 (1986)
4088–4095.
[33] M. Salavati-Niasari, A. Badiei, K. Saberyan, Chemical Engineering Journal 173
(2011) 651–658.
[34] E. Tsuchida, K. Yamamoto, K. Oyaizu, N. Iwasaki, F.C. Anson, Inorganic Chemistry
33 (1994) 1056–1063.
[35] T. Ben, I. Zid, A. Keheder, Ghorbel, Reaction Kinetics Mechanisms 100 (2010)
131–143.
4. Conclusions
The experimental and theoretical results obtained, even though
do not allow to draw a decisive picture of the mechanism, let to
emphasize that steric factors play a major role in determining the
outcome of the reaction, often overcoming the electronic effects.
Theoretical results suggest the intervention in the catalytic cycle of
an hydroperoxo vanadium species.
[36] J. Zhao, W. Wang, Y. Zhang, Journal of Inorganic and Organometallic Polymers
18 (2008) 441–447.
[37] A. Coletti, C.J. Whiteoak, V. Conte, A.W. Kleij, ChemCatChem
http://dx.doi.org/10.1002/cctc.201100398.
4 (2012),
Acknowledgements
[38] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Rob, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Peters-
son, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.
Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P.
Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Strat-
mann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari,
J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Ste-
fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. John-
son, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Gaussian, Inc.,
Wallingford, CT, 2003.
Research carried out in the framework of COST D40 Action
“Innovative Catalysis – New Processes and Selectivities”. Experi-
mental work of S. Ummarino and G. Di Carmine is acknowledged.
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.cattod.2012.03.032.
[39] A.D. Becke, Journal of Chemical Physics 98 (1993) 5648–5652.
[40] C. Lee, W. Yang, R.G. Parr, Physical Review B-Condensed Matter Material Physics
37 (1988) 785–789.
References
[41] Spartan’08 Wavefunction, Inc. Irvine, CA.
[42] L.P. Hammett, Physical Organic Chemistry, 2nd ed., McGraw-Hill, New York,
1970, p. 367.
Use of arithmetical ꢂ␴ to measure electronic effects, when two or more
substituents are present, is a standard procedure for evaluation of free energy rela-
tionships.
[1] T. Punniyamurthy, S. Velusamy, J. Ikbal, Chemical Reviews 105 (2005)
2329–2363.
[2] P.T. Anastas, J.T. Warner, Green Chemistry Theory and Practice, Oxford Univer-
sity Press, 1998.
[3] R.A. Sheldon, I. Arends, U. Hanefeld, Green Chemistry and Catalysis, Wiley-VCH,
Weinheim, Germany, 2007.
[4] G. Strukul, Catalytic Oxidation with Hydrogen Peroxide as Oxidant, Kluwer
Academic Publisher, Netherlands, 1992.
[5] V. Conte, B. Floris, Inorganica Chimica Acta 363 (2010) 1935–1946, and refer-
ences therein.
[6] V. Conte, A. Coletti, B. Floris, G. Licini, C. Zonta, Coordination Chemistry Reviews
255 (2011) 2165–2177, and references therein.
[7] G. Licini, V. Conte, A. Coletti, M. Mba, C. Zonta, Coordination Chemistry Reviews
255 (2011) 2345–2357, and references therein.
[8] A. Butler, Coordination Chemistry Reviews 187 (1999) 17–35, and refs. cited
therein.
[43] O. Bortolini, V. Conte, Journal of Inorganic Biochemistry 99 (2005) 1549–1557.
ˇ
[44] M. Casny´, D. Rehder, Dalton Transactions (2004) 839–846, and refs. cited
therein.
[45] G. Zampella, P. Fantucci, V.L. Pecoraro, L. De Gioia, Journal of the American
Chemical Society 127 (2005) 953–960.
[46] C.J. Schneider, G. Zampella, C. Greco, V.L. Pecoraro, L. De Gioia, European Journal
of Inorganic Chemistry (2007) 515–523.
[47] In Ref. [43], trigonal bipyramid vanadium complexes are studied where the
˚
protonated oxygen is found at binding distances of 2.03–2.08 A from the vana-
dium centre. The different results obtained for salophen and salen systems,
where that oxygen is located at larger distances, may be ascribed to the their
octahedral structure, which hampers the binding of a further coordinating
atom.
[9] V.M. Dembitsky, Tetrahedron 59 (2003) 4701–4720.
[10] A.G.J. Ligtenbarg, R. Hage, B.L. Feringa, Coordination Chemistry Reviews 237
(2003) 89–101.
[48] F. Ding, J.M. Smith, H. Wang, Journal of Organic Chemistry 74 (2009) 2679–2691.

A. Coletti et al. / Catalysis Today 192 (2012) 44–55
55
[49] C.J. Schneider, J.E. Penner-Hahn, V.L. Pecoraro, Journal of the American Chemical
Society 130 (2008) 2712–2713.
[53] I.V. Yudanov, P. Gisdakis, C. Di Valentin, N. Rösch, European Journal of Inorganic
Chemistry (1999) 2135–2145.
[50] R.R. Everett, J.R. Kanofsky, A. Butler, Journal of Biological Chemistry 265 (1990)
4908–4914.
[51] J.W.P.M. Van Schijndel, P. Barnett, J. Roelse, E.G.M. Vollenbroek, R. Wever, Euro-
pean Journal of Biochemistry 225 (1994) 151–157.
[52] P. Adão, J. Costa Pessoa, R.T. Henriques, M.L. Kuznetsov, F. Avecilla,
M.R. Maurya, U. Kumar, I. Correia, Inorganic Chemistry 48 (2009)
3542–3561.
[54] E. Kühn, A.M. Santos, R.W. Rocksy, E. Herdweck, W. Scherer, P. Gisdakis, I.V.
Yudanov, C. Di Valentin, N. Rösch, Chemistry-A European Journal 5 (1999)
3603–3615.
[55] F.C. Di Valentin, P. Gisdakis, I.V. Yudanov, N. Rösch, Journal of Organic Chemistry
65 (2000) 2996–3004.
[56] S. Lovat, M. Mba, H.C. Abbenhuis, D. Vogt, C. Zonta, G. Licini, Inorganic Chemistry
48 (2009) 4724–4728.