A dichotomy in the enantioselective oxidation of aryl benzyl sulfides: A combined experimental and computational work

Article abstract of DOI:10.1016/j.tet.2018.03.005

Pentafluorobenzyl pentafluorophenyl sulfide is oxidised with moderate e.e. value and a low yield by the usually highly successful oxidation protocol based upon tert-butyl hydroperoxide (TBHP) in the presence of a titanium/hydrobenzoin complex. This disapp

Full text of DOI:10.1016/j.tet.2018.03.005

Accepted Manuscript
A dichotomy in the enantioselective oxidation of aryl benzyl sulfides: A combined
experimental and computational work
Maria Annunziata M. Capozzi, Andrea Bottoni, Matteo Calvaresi, Cosimo
Cardellicchio
PII:
S0040-4020(18)30248-5
10.1016/j.tet.2018.03.005
TET 29345
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 1 February 2018
Revised Date: 28 February 2018
Accepted Date: 4 March 2018
Please cite this article as: Capozzi MAM, Bottoni A, Calvaresi M, Cardellicchio C, A dichotomy in the
enantioselective oxidation of aryl benzyl sulfides: A combined experimental and computational work,
Tetrahedron (2018), doi: 10.1016/j.tet.2018.03.005.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
F
F
F
F
F
F
F
F
CHP, cat*
F
F
+
S
S
n-hexane, rt
F
F
-
F
F
O
F
F
F
F
F
F
76% yield
>98% ee
cat*= 5% Ti(O-i-Pr) /(S, S)-hydrobenzoin
4

ACCEPTED MANUSCRIPT
A dichotomy in the enantioselective oxidation of aryl benzyl sulfides: a
combined experimental and computational work
Maria Annunziata M. Capozzi,[b] Andrea Bottoni,*[c] Matteo Calvaresi,[c] Cosimo Cardellicchio*[a]
[a] Dr. C. Cardellicchio, Corresponding Author
CNR ICCOM - Dipartimento di Chimica
Università di Bari
via Orabona 4. 70125 Bari, Italy.
Fax: (+)39 – 0805539596
E-mail: cardellicchio@ba.iccom.cnr.it
[b] Dipartimento di Chimica. Università di Bari
via Orabona 4. 70125 Bari (Italy)
[c] Dipartimento di Chimica “G. Ciamician”,
Alma Mater Studiorum – Università di Bologna,
via F. Selmi 2, 40126 Bologna, Italy.
Abstract. Pentafluorobenzyl pentafluorophenyl sulfide is oxidised with moderate e.e. value and a
low yield by the usually highly successful oxidation protocol based upon tert-butyl hydroperoxide
(TBHP) in the presence of a titanium/hydrobenzoin complex. This disappointing result resisted until
the present work, in which the switch of the oxidation agent (from TBHP to cumene
hydroperoxide), suggested by our previous computations, yielded the enantiopure sulfoxide. This
valuable chiral compound was obtained in good yields (76%) without resorting to a
chromatographic separation. DFT computations uncovered that this favourable reactivity was
originated by a stabilizing π−π−stacking between the phenyl group of the oxidant and the
pentafluorophenyl moiety of the substrate.
Keywords: titanium; sulfoxide; DFT computations; hydrobenzoin; oxidation mechanism
Introduction
The synthesis of enantiopure sulfinyl compounds is a classic topic in asymmetric synthesis,1 since a
whole class of chiral sulfoxides were obtained for the first time at the beginning of the '60's by
using the Andersen-Mislow synthesis.2 In the following years, the easy availability of these
1

ACCEPTED MANUSCRIPT
compounds favoured their employment as chirality inducer in the asymmetric synthesis of many
bioactive compounds,3 or as ligands in asymmetric synthesis.1b, 4
Many other procedures were proposed in this last half century to overcome the limitations of the
original Andersen-Mislow synthesis.1 These procedures are usually divided into two main groups,
the first one being connected with the formation of a carbon-sulphur bond, whereas the second deals
with the formation of a carbon-oxygen bond.
The organometallic procedures, in which chiral sulfinyl compounds are transformed into the target
sulfoxides with the aid of an organometallic reagent, belong to the first group.1-2 The
enantioselective arylation of sulfenate anions, that stimulated a recent interest,5 can be collected
also in this group. On the other hand, the enantioselective oxidation reactions of prochiral sulfides,1
either with the aid of metal catalysed procedures6-7 or in the absence of metal catalysis,8 are clear
examples of the second group. Moreover, also combined approach were reported, such as the
displacement reaction of carbanionic leaving groups on enantiopure sulfinyl intermediates, obtained
via an enantioselective oxidation.9
The asymmetric catalysis methods, that are preferably selected by the pharmaceutical industry
because of the tolerance towards the functional groups of the target chiral bioactive sulfoxides,10
have been prevailing over the years (see, for instance, the case of the blockbuster anti-ulcer drug
(S)-omeprazole11). However, up to now, it has not been found a general method for the
enantioselective oxidation of any sulfide, the best result being the oxidation of a specific class of
similar sulfides.1, 6-7
During the last years, we contributed to this research with a series of papers12-19 on the
enantioselective oxidation of sulfides with tert-butyl hydroperoxide (TBHP) in the presence of a 1:2
complex between titanium i-propoxide and (S, S)- or (R, R)-hydrobenzoin (Chart 1, Ligand A), a
cheap and commercially available ligand.20
2

ACCEPTED MANUSCRIPT
O
O
HO
OH
HO
OH
(S, S)-Hydrobenzoin
(Ligand A)
(Ligand B)
Chart 1. Chiral ligands employed in the asymmetric oxidation of sulfides.
When applied to aryl benzyl sulfides,12, 15-19 this oxidation protocol was demonstrated to be very
convenient. First of all, it provides the desired sulfoxides in good yields with high enantiomeric
purity. The procedure is easy to perform (just a “mix and wait” at room temperature), uses cheap
and easily available reactants and yields the target sulfoxide without the formation of large amounts
of the corresponding sulfones (a drawback of many oxidation protocols1, 6). Moreover, when the e.e.
values are beyond the threshold of 81-85%, the enantiopure intermediate is easily obtained by
crystallization. Due to these favourable features, we have been able to build up a chemical library of
more than 40 enantiopure aryl benzyl sulfoxides,19 and to investigate their crystal structures15, 17-19
and the circular dichroism patterns.21
Moreover, we contributed to the comprehension of this oxidation process with mechanistic
investigations16, 19 and with DFT computations.15, 17 Our computations highlighted the importance
of the weak non-covalent interactions involving aryl groups in the highly enantioselective oxygen
transfer step.15, 17 For the sake of completeness, we must mention that other research groups, either
before22-23 or after24-25 our work, employed in some instances this oxidation system, often proposing
different experimental procedures.
In this very successful framework of reactivity and high enantioselectivity, the oxidation of
pentafluorobenzyl pentafluorophenyl sulfide emerges as a stumbling block.17 While the oxidation of
not fluorinated aryl benzyl sulfides are almost always successfully with our procedure,15-16, 19 and
the oxidation of aryl benzyl sulfide with only one of the phenyl groups fully fluorinated provides
even larger yields and ee values in comparison with the not fluorinated compounds,16-19 the same
oxidation protocol produces a pentafluorobenzyl pentafluorophenyl sulfoxide having a 61% e.e.
3

ACCEPTED MANUSCRIPT
value, but a yield of only 19%.17 Our theoretical model based on DFT computations provided also a
rationale for this disappointing result,17 which, however, pointed out the need for a different
oxidation procedure for this sulfide.
Results and discussions.
The purpose of our investigation was to find a new oxidation protocol to improve the quite
unsatisfactory results (Table 1, entry 1) previously obtained in the oxidation of pentafluorobenzyl
pentafluorophenyl sulfide 1a to pentafluorobenzyl pentafluorophenyl sulfoxide 1b.17
The oxidation process was preliminarily investigated with different procedures, such as the
hydroperoxide oxidation in the presence of a titanium/tartrate complex,11, 26 but poor ee values were
obtained. Poor enantioselectivity was obtained also by employing hydrogen peroxide in the
presence of a chiral vanadium catalyst.27 At this stage, we reasoned that it should be more fruitful to
return to our protocol, and to change it according to hints coming from our theoretical model, that
showed to be robust and reliable.
Table 1. Enantioselective oxidation of aryl benzyl sulfide with the titanium/(S, S)-
hydrobenzoin complex.
ox, cat*
O
S
S
Ar
Ar
Ar
n-hexane
Ar
cat* = 5% Ti(O-i-Pr)4/(S, S)-hydrobenzoin
Entry
Ar
Sulfide
1a
Oxidant
TBHP
TBHP
TBHP
CHP
Liganda
Product Yield (%)b
ee (%)c
61d
1
2
3
4
5
6
C6F5
C6F5
C6F5
C6H5
C6H5
C6F5
(R)-1b
(R)-1b
(R)-1b
(R)-2b
(R)-2b
(R)-1b
19d
26e
29
A
A
B
A
A
A
52 e
1a
20
1a
46
8
2a
TBHP
CHP
87f
76
>98f
>98
2a
1a
(a) A = (S, S)-1,2-bis-diphenyl-1,2-di-hydroxyethane (hydrobenzoin). B = (S, S)-1,2-bis-di-(2'-methoxyphenyl)-1,2-
dihydroxyethane (see Chart 1). (b) yields refer to pure isolated products (c) Determined by chiral HPLC (see
Experimental Section). (d) See ref. 17. (e) Inverse addition (see text). (f) See ref. 19.
4

ACCEPTED MANUSCRIPT
In a first trial, we used a modification of the work-up procedure in which, after having obtained the
in situ titanium/hydrobenzoin complex, the oxidant was added before the sulfide (inverse addition).
In our previous work,16 we observed that this altered sequence of addition of the reactants provides
better yields, but a lower enantioselectivity. This fact was confirmed also in this reaction, in which
we observed (entry 2), a better yield (26%), but a lower enantioselectivity (52% ee). However, these
values remain unsatisfactory. We observed that in a work on the synthesis of omeprazole-like
molecule (that are indeed aryl benzyl sulfides), modified hydrobenzoin ligands were employed.24
Since in our theoretical model, the CH···π interactions28 play a crucial role, an increase of the
electronic density of the phenyl groups of the hydrobenzoin, due for example to a methoxy group,
should be beneficial. Thus, we synthesized the (S, S)-1,2-bis-di-(2'-methoxyphenyl)-1,2-
dihydroxyethane (Chart 1, ligand B) with an asymmetric dihydroxylation,29 and we used it as a
ligand of the titanium in the asymmetric oxidation of 1a (entry 3) with TBHP. We observed a
modest increase of the yield (29%, Table 1), but a drop of the enantioselectivity (20% ee, entry 3).
Probably, the presence of the methoxy group in the ortho-position of the phenyl groups of the
hydrobenzoin is able to interfere with the coordination modes of the ligand.19 Another opportunity
that we decided to evaluate was the variation of the oxidizing species. tert-Butyl hydroperoxide is a
cheap and easily available oxidant, and has the special merit of leaving only tert-butanol, after the
oxygen transfer. Other hydroperoxides were employed in the oxidation of sulfides, such as the furyl
hydroperoxide,30 or chiral hydroperoxides.31 However, these oxidants are not commercially
available. Our attention was turned towards cumene hydroperoxide (CHP), a cheap and available
oxidant, that had been sometimes used in previous works on sulfide oxidation in the presence of
titanium complexes.26 Generally speaking, CHP was considered a less performing oxidation agent
in comparison with TBHP. For example, a lower enantioselectivity was obtained by us, by using our
protocol, when CHP was used as the oxidant in the enantioselective oxidation of Sulindac sulfide
methyl ester.14 On the other hand, a sporadic case of a better enantioselectivity was observed by
5

ACCEPTED MANUSCRIPT
Kagan et al. in the switch from TBHP to CHP, when some aryl methyl sulfide were asymmetrically
oxidized in the presence of a complex between titanium and diethyl tartrate.26
As a first test, we used CHP with a titanium/hydrobenzoin complex to oxidise benzyl phenyl sulfide
2a, that is the simplest form of any aryl benzyl sulfide. The results were disappointing (entry 4),
because we obtained a 46% yield of the sulfoxide 2b, having a poor ee value (8%). This result is
analogous to the oxidation experiment of methyl p-tolyl sulfide reported by other authors with the
same oxidation reagents, but with a different experimental procedure.23 This result is particularly
disappointing, if it is compared with the identical oxidation performed by us19 with TBHP as the
oxidant (entry 5). The enantiopure sulfoxide 2b (>98%) was obtained with a good yield (87%), as
occurred for many other aryl benzyl sulfoxides.
However, we reasoned that CHP differs from TBHP in the substitution of a methyl with a phenyl
group. Since we found that the aryl groups play a crucial role in the mechanism that we proposed,15,
17 we decided that it was worthy to perform the oxidation of 1a with our catalytic system, and CHP
as the oxidant (entry 6). This result, even if counter-intuitive, was the turning point of this research,
because we obtained (R)-pentafluorobenzyl pentafluorophenyl sulfoxide 1b17 in an enantiopure
form (>98% ee), but also in good yields (76%, entry 4), limiting this number only to the crystals
that precipitated from n-hexane, as in our previous work,12 without resorting to a chromatographic
separation (see Experimental Section).
At this stage, the complementarity between the employment of CHP and TBHP in the oxidation
reaction of aryl benzyl sulfide in the presence of a complex between titanium and hydrobenzoin
emerges. The TBHP-protocol yielded enantiopure sulfoxides with almost every not fluorinated, or
partially fluorinated, aryl benzyl sulfide. In the singular case of pentafluorobenzyl
pentafluorophenyl sulfide 1a, only the CHP-protocol, that usually gave inferior results in the
previous cases, yielded the enantiopure sulfoxide 1b in good yields.
At this stage, as in our previous work,17 we performed a DFT computational investigation which
6

ACCEPTED MANUSCRIPT
can provide a rationale to the above discussed experimental data.
17
Computational Mechanistic Study. In our previous work,15,
we demonstrated that the
enantioselectivity is governed by the relative energies of the two diastereomeric octahedral
complexes that form when the oxidant (TBHP) is added to the initial adduct between the substrate
and the [(S,S)-hydrobenzoin]2Ti, before the oxygen transfer. They differ in the orientation of the
substrate with respect to the metal and the oxidant and originate two distinct diastereomeric paths
leading to the two sulfoxides characterized by opposite configuration of the newly formed sulphur
chiral centre. The energy difference between the two complexes (simply denoted here as M and M’)
was rather large (5.1 kcal mol-1) in the presence of benzyl phenyl sulfide as substrate. The more
stable and, thus, more populated species, was that (M’) corresponding to the sulfide approaching
orientation leading to configuration (R), i.e. the configuration experimentally observed. The larger
stabilization of this adduct was due to the presence of stabilizing T-shaped structures32-34 involving
the aromatic rings of the substrate and the [hydrobenzoin]2Ti species. The two complexes became
almost degenerate (the difference was only 0.4 kcal mol-1) when pentafluorobenzyl
pentafluorophenyl sulfide was used as a substrate and also the corresponding oxidation transition
states were very close in energy.17 Thus, the two complexes were similarly populated and, since the
activation barriers were very similar, the probability to follow the two diastereomeric reaction
channels became comparable leading to the lower e.e. values experimentally observed.
Since the energy difference between the adducts, that anticipate the oxygen transfer transition states,
was proved to be the key-factor governing the high or moderate observed enantioselectivity, we
replaced in our original model-system the oxidant TBHP with CHP and we computed the structure
and energy of the resulting M and M’ complexes, considering as substrate either pentafluorobenzyl
pentafluorophenyl sulfide or benzyl phenyl sulfide. We found that the trend of the M/M’ energy
difference in the two cases provides at least a qualitative explanation for the observed
enantioselectivity.
7

ACCEPTED MANUSCRIPT
In the case of pentafluorobenzyl pentafluorophenyl sulfide the nature of non-covalent interactions
involving the aromatic rings of the adducts (π-stacking interactions) change significantly when we
compare M to M’. The two complexes are schematically represented in Figure 1.
Figure 1. A schematic representation of M’ (top-side) and M (bottom side) complexes involving pentafluorobenzyl
pentafluorophenyl sulfide as substrate. Two different perspectives are given for M to show all important non-covalent
interactions. Bond lengths are in ångstroms.
8

ACCEPTED MANUSCRIPT
In M’ the two phenyl rings belonging to the substrate (ring A) and the oxidant CHP (ring B) lie in
approximately parallel planes resulting in a typical “sandwich structure”.32-35 Typical distances
between the most effectively interacting carbon atoms of the two rings are in the range 3.51-3.75 Å,
as evidenced in Figure 1, in agreement with the values computed in previous works.15, 17 Also, the
substrate ring A’ interacts with ring D of one hydrobenzoin moiety. However, in this case the two
phenyl planes are displaced to give a structural arrangement that was demonstrated to provide more
effective π−π stabilization.32-35 Furthermore, one fluorine atom of the substrate phenyl ring A’
interacts with a hydrogen of the hydrobenzoin bridge (F…HC distance = 2.79 Å) and the same
fluorine atoms is involved in non-covalent interactions with the hydrobenzoin phenyl ring C.
In M, where the orientation of the substrate is reversed with respect to M’, the substrate is arranged
in a folded conformation and the two rings A and B are too far away to give a significant stabilizing
interaction (the shortest distances between fluorine atoms of ring A and the carbon atoms of ring B
are 5.74 and 5.78 Å). This is the most important structural difference between the two adducts: the
other interactions involving rings A’, C and D do not change significantly and it is reasonable to
believe that they provide in the average a similar stabilizing effect in M an M’. The overall effect of
all above interactions is to make complex M’ significantly more stable (and more populated) than
M, and the computed energy difference (2.8 kcal mol-1) significantly increases with respect to the
reaction involving TBHP (only 0.4 kcal mol-1). Since the two oxidation transition states, TS and
TS’, are almost degenerate (the energy difference is only 0.3 kcal mol-1), the probability to follow
one or the other diastereomeric reaction channel is higher for the more populated intermediate M’.
Thus, it is the equilibrium distribution of M and M’ which determines the e.e. experimentally
observed. A schematic energy diagram for the two diastereomeric reaction pathways (simply
denoted as pro-(R) and pro-(S) reaction paths) is given in Figure 2.
9

ACCEPTED MANUSCRIPT
Figure 2. A schematic energy diagram showing the diastereomeric reaction pathways that originate from the two
complexes M and M’.
A different trend of non-covalent stabilizing interactions in the comparison between M and M’ was
observed for the non-fluorinated substrate (benzyl phenyl sulfide). In both M and M’ the two
phenyl rings A(substrate) and B(CHP) are involved in π-stacking and T-shaped interactions: typical
values of carbon(A)-carbon(B) distances and hydrogen(A)-carbon(B) distances are reported in
Figure 3. Similarly, interactions between phenyl ring A’ and ring C can be recognized in both
adducts. Even if, in general, T-shaped interactions are more stabilizing than π-stacking
interactions,32-34 the presence of similar A-B and A’-C interactions in both M and M’ reduces the
energy difference between the two adducts, which becomes 1.9 kcal mol-1 (it was 5.1 kcal mol-1
when TBHP was used as oxidant). The two corresponding transition states (see Figure S2 in the SI
file) are again almost degenerate (energy difference is 0.7 kcal mol-1). Thus, the difference in the
population of the two complexes decreases for the non-fluorinated substrate, with a consequent
decrease of the enantiomeric excess, even if, in this particular case, it remains only on qualitative
10

ACCEPTED MANUSCRIPT
basis.
Figure 3. A schematic representation of M’ (top-side) and M (bottom side) complexes involving benzyl phenyl sulfide
as substrate. Bond lengths are in ångstroms.
11

ACCEPTED MANUSCRIPT
In summary, these computations show that the different nature of the oxidant CHP (compared to
TBHP) is responsible for the increased e.e. observed with pentafluorobenzyl pentafluorophenyl
sulfide. The CHP phenyl ring activates specific stabilizing π-stacking interactions with the
pentafluorophenyl group only in M’ where the two aromatic rings are arranged in a suitable relative
position. This is not possible in M where the pentafluorobenzyl group is more distant from the CHP
phenyl ring because of the methylene unit positioned between sulphur and the aromatic ring. In this
case, the substrate adopts a more convenient folded conformation. Furthermore, with benzyl phenyl
sulfide the presence of the CHP phenyl ring activates similar interactions in M and M’ and the
difference between the two intermediates becomes less pronounced.
Conclusions
The stumbling block, represented by the enantioselective oxidation of pentafluorobenzyl
pentafluorophenyl sulfide in the presence of a titanium hydrobenzoin complex, was overcome. The
counter-intuitive choice of a different, usually less performing oxidant (CHP) resolves the synthetic
problems. The previously reported protocol, based on TBHP as oxidant, and the new one, in which
TBHP was replaced by CHP, are complementary and should be used according to the sulfide that
should be oxidised.
A theoretical model based on DFT computations can explain the trend of the the improvement of
e.e. when TBHP is replaced by CHP as oxidant. As demonstrated in our previous work17 the
enantioselectivity is dictated by the energy difference between the two diastereomeric octahedral
complexes (M and M’) that form when the oxidant (CHP in the present case, TBHP in the previous
work) is added to the initial adduct between the substrate and the [(S, S)-hydrobenzoin]2Ti, before
the oxygen transfer.
The M/M’ energy difference is in turn determined by non-covalent interactions, whose importance
depends on the structural features of substrate and oxidant. In the present case, the phenyl ring of
CHP is responsible for important stabilizing π-stacking interactions, which can be activated only
12

ACCEPTED MANUSCRIPT
when the substrate pentafluorophenyl group can face the CHP phenyl ring: this occurs in complex
M’ leading to the observed configuration R of the sulfoxide product. When the position of the
substrate is reversed (from up to down) as in complex M (leading to configuration (S)), these
stabilizing π-stacking interactions disappear. The resulting energy difference computed with CHP is
2.8 kcal mol-1, which is much larger than the value of 0.4 kcal mol-1 computed in the presence of
TBHP as oxidant.
Experimental section
Sulfides 1a and 2a, and sulfoxides 1b and 2b had been previously reported17,19 and fully
characterised. Enantioselective oxidation of sulfides were performed according to our previous
publications.13-19 The innovative CHP-oxidation is reported beneath.
Enantioselective oxidation of pentafluorobenzyl pentafluorophenyl sulfide (1a) with cumene
hydroperoxide in the presence of a titanium/(S, S)-hydrobenzoin catalyst. A solution of Ti(O-i-
Pr)4 99.999% (28 mg, 0.1 mmol) in 5 mL of n-hexane was added to a solution of finally grounded
(S, S)-hydrobenzoin (43 mg, 0.2 mmol) in 10 mL of n-hexane under a nitrogen atmosphere. The
mixture was stirred for 1 hour at room temperature. A solution of pentafluorobenzyl
pentafluorophenyl sulfide17 1a (0.76 g, 2 mmol) in 10 mL of n-hexane was then added and the
mixture was stirred for 30 minutes. After this time, 0.41 mL of a commercial solution of cumene
hydroperoxide 80% was added and the stirring was continued for 21 hours. During this time, a
white solid precipitated. 0.6 g of this solid, constituted by enantiopure sulfoxide 1b, was filtered
(76% yield). Further batch of 1b could be obtained starting from the remaining solution, provided
that the residual cumyl alcohol was preliminarily distilled with a low-pressure kugelrohr apparatus.
25
(R)-Pentafluorobenzyl pentafluororophenyl sulfoxide17 1b (n-hexane/ethanol 9:1). [α]D = + 36.9
(c= 0.85, CHCl3). The enantiopurity was checked with HPLC (Chiralcel OD-H column; n-hexane/i-
propanol 7:3; 0.5 mL/min flow rate. t(R) = 23.1 min.; t(S) = 28.0 min; α = 1.29). The absolute
configuration had been previously attributed by CD spectra,17, 21c and it follows our empirical rule,
13

ACCEPTED MANUSCRIPT
according to which (R)-sulfoxides are obtained by our oxidation protocol, when (S, S)-hydrobenzoin
was employed as a ligand of the titanium.
Computational methods. All DFT computations reported in the present paper, were carried out
with the Gaussian 09 series of programs.36 Since the investigated model-system involves aryl
groups in the substrate and ligands, it is reasonable to believe that a reliable estimate of non-
covalent interactions involving π systems (π-π stacking interactions, T-shaped interactions, H-
contacts) is essential in the computation of the potential surface. Recently various new hybrid
functional have been proposed for a reliable description of non-covalent interactions. In the interests
of consistency with previous work we used the hybrid functional MPWB1K,37 which was
demonstrated to be capable of treating non-covalent interactions and medium-range correlation
effects.38-39 The model-system was described with the DZVP basis,40 which is a Local Spin Density
(LSD)-optimised basis set of double-zeta quality that includes polarisation functions and is suitable
to describe weak hydrogen and π interactions such as those occurring in the system investigated
here. The transition vector of the various transition states was analysed by means of frequency
computations.
Acknowledgements
We thank Dr. Francesco Contillo for some preliminary experiments. Thanks are due to the
“Phoebus” Project “Tecnologie plastiche per la realizzazione di celle solari e sorgenti organiche per
illuminazione ad elevata efficienza, uniformità e brillanza” in the framework of the Reti di
Laboratori Pubblici di Ricerca. This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors. The authors declare no conflict of
interest.
Supplementary data. Representation of the calculated transition states; Cartesian coordinates of
the calculated intermediates.
14

ACCEPTED MANUSCRIPT
References
1. a) Han J, Soloshonok VA, Klika KD, Drabowicz J, Wzorek A, Chem. Soc. Rev. 2018; 47:
1307. b ) Otocka S, Kwiatkowska M, Madalińska L, Kiełbasiński P, Chem Rev. 2017; 117:
4147. c) O' Mahony GE, Ford A, Maguire AR, J. Sulfur Chem. 2013; 34: 301. d)
Wojaczyńska E, Wojaczyński J, Chem. Rev. 2010; 110: 4303.
2. Andersen KK, Gaffield W, Papanikolaou NE, Foley JW, Perkins RI, J. Am. Chem. Soc.
1964; 86: 5637. Mislow K, Green MM, Laur P, Melillo JT, Simmons T, Ternay AL Jr., J.
Am. Chem. Soc. 1965; 87: 1958.
3. Posner, G.H. Asymmetric synthesis using α-sulfinyl carbanions and β-unsaturated sulfoxides
in The chemistry of sulphones and sulphoxides (Eds Patai, S.; Rappoport, Z.; Stirling,
C.J.M.); John Wiley and Sons: New York, 1988; pp. 823-849. Carreño, MC Chem. Rev.
1995; 95: 1717. Solladié, G. Optically active β-ketosulfoxides in asymmetric synthesis in
Advanced Asymmetric Synthesis (Ed. Stephenson, G.R.); Chapman and Hall: London, 1996;
pp 60-92.
4. Trost BM, Rao, M, Angew. Chem. Int. Ed. Engl. 2015; 54: 5026.
5. a) Maitro G, Vogel S, Sadaoui M, Prestat G, Madec D, Poli G, Org. Lett. 2007; 9: 5493. b)
Gelat F, Jayashankaran, J, Lohier JF, Gaumont AC, Perrio S, Org. Lett. 2011; 13: 3170. c)
Zong L, Ban X, Kee CW, Tan CH, Angew. Chem. Int. Ed. Engl. 2014; 53: 11849. d) Jia T,
Zhang M, McCollom SP, Bellomo A, Montel S, Mao J, Dreher SD, Welch CJ, Regalado EL,
Williamson RT, Manor BC, Tomson NC, Walsh PJ, J. Amer. Chem. Soc. 2017; 139: 8337. e)
Zong L, Tan CH, Acc. Chem. Res. 2017; 50: 842. f) Wang L, Chen M, Zhang P, Li W, Zang
J, J. Amer. Chem. Soc. 2018; 140: xxxx. DOI: 10.1021/jacs.8b00178.
6. Bryliakov KP, Talsi EP, Curr. Org. Chem. 2012; 10: 1215.
15

ACCEPTED MANUSCRIPT
7. For recent report on less frequently used metal complexes, see Ye X, Moeljadi AMP, Chin
KF, Hirao H, Zong L, Tan CH, Angew. Chem. Int. Ed. Engl. 2016; 55: 7101. Zong L, Wong
C, Moeljadi AMP, Ye X, Ganguly R, Li Y, Hirao H, Tan CT, Nature: Commun. 2016; 7:
article number 13455.
8. Stingl KA, Tsogoeva SB, Tetrahedron: Asymm. 2010; 21: 1055.
9. Capozzi MAM, Cardellicchio C, Naso F., Eur. J. Org. Chem. 2004: 1855.
10. Legros J, Dehli JR, Bolm C, Adv. Synth. Catal. 2005; 347: 19.
11. Cotton H, Elebring T, Larsson M, Li L, Sorensen H, von Unge S, Tetrahedron: Asymmetry
2000; 11: 3819. Seenivasaperumal M, Federsel HJ, Szabó KJ, Adv. Synth. Catal. 2009; 351:
903.
12. Capozzi MAM, Cardellicchio C, Naso F, Rosito V, J. Org. Chem. 2002; 67: 7289.
13. Cardellicchio C, Hassan Omar O, Naso F, Capozzi MAM, Capitelli F, Bertolasi V,
Tetrahedron: Asymm. 2006; 17: 223.
14. Naso F, Cardellicchio C, Affortunato F, Capozzi MAM, Tetrahedron: Asymm. 2006; 17:
3226.
15. Naso F, Capozzi MAM, Bottoni A, Calvaresi M, Bertolasi V, Capitelli F, Cardellicchio C,
Chem. Eur. J. 2009; 15: 13417.
16. Capozzi MAM, Centrone C, Fracchiolla G, Naso F, Cardellicchio C, Eur. J. Org. Chem.
2011: 4327.
17. Capozzi MAM, Capitelli F, Bottoni A, Calvaresi M, Cardellicchio C, ChemCatChem 2013;
5: 210.
18. Capozzi MAM, Capitelli F, Cardellicchio C, Cryst. Growth Des. 2014; 14: 5442.
19. Capozzi MAM, Terraneo G, Cavallo G, Cardellicchio C, Tetrahedron 2015; 71: 4810.
20. Okano K, Tetrahedron 2011; 67, 2483.
16

ACCEPTED MANUSCRIPT
21. a) Pescitelli G, Di Pietro S, Cardellicchio C, Capozzi MAM, Di Bari L, J. Org. Chem. 2010;
75: 1143. b) Padula D, Di Pietro S, Capozzi MAM, Cardellicchio C, Pescitelli G, Chirality
2014; 26: 462. c) Berardozzi R, Guido CA, Capozzi MAM, Cardellicchio C, Di Bari L,
Pescitelli G, Eur. J. Org. Chem. 2015; 5554. d) Górecki M, Capozzi MAM, Albano G,
Cardellicchio C, Di Bari L, Pescitelli G, Chirality, 2018; 30: 29.
22. Yamamoto K, Ando H, Shuetake T, Chikamatsu H, J. Chem. Soc. Chem. Commun, 1989:
754.
23. Donnoli MI, Superchi S, Rosini C, J. Org. Chem. 1998; 63, 9392.
24. Jiang B, Zhao XL, Dong JJ, Wang WJ, Eur. J. Org. Chem. 2009: 987.
25. Seto M, Miyamoto N, Aikawa K, Aramaki Y, Kanzai N, Iizawa Y, Baba M, Shiraishi M,
Biorg. Med. Chem. 2005; 13: 363.
26. Zhao SH, Samuel O, Kagan HB, Tetrahedron 1987; 43: 5135.
27. Bolm C, Bienewald F, Angew. Chem. Int. Ed. Engl. 1995; 34: 2640.
28. Nishio M, Umezawa Y, Fantini J, Weiss MS, Chakrabarti P, Phys. Chem. Chem. Phys. 2014;
16: 12648. Nishio M, Phys. Chem. Chem. Phys. 2011; 13: 13873. Cardellicchio C, Capozzi
MAM, Alvarez-Larena A, Piniella JF, Capitelli F, CrystEngComm, 2012; 14: 3972. Desiraju
GR, Vittal JJ, Ramanan A, Crystal Engineering. A Textbook. World Scientific: Singapore,
2011.
29. Wallace TW, Wardell I, Li KD, Leeming P, Redhouse AD, Challand SR, J. Chem. Soc.
Perkin Trans 1; 1995: 2293.
30. Scettri A, Bonadies F, Lattanzi A, Tetrahedron: Asymm. 1996; 7: 629.
31. Adam W, Korb MN, Roschmann KJ, Saha-Möller CR, J. Org. Chem. 1998; 63: 3423.
32. Janowsky T, Pulay P, Chem. Phys. Lett. 2007; 447: 27.
33. Pitoňák M, Neogrády P, Řezáč J, Jurečka P, Urban M, Hobza P, J. Chem. Theory Comput.
2008; 4: 1829.
17

ACCEPTED MANUSCRIPT
34. Sinnokrot MO, Sherill CD, J. Phys. Chem A 2006; 110: 10656.
35. Meyer EA, Castellano RK, Diederich F, Angew. Chem. Int. Ed. 2003; 42: 1210.
36. Frisch M, Trucks G, Schlegel H, Scuseria G, Robb M, Cheeseman J, Scalmani G, Barone V,
Mennucci B, Petersson G, Nakatsuji H, Caricato M, Li X, Hratchian H, Izmaylov A, Bloino
J, Zheng G, Sonnenberg J, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M,
Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Montgomery J Jr, Peralta J, Ogliaro F,
Bearpark M, Heyd J, Brothers E, Kudin K, Staroverov V, Kobayashi R, Normand J,
Raghavachari K, Rendell A, Burant J, Iyengar S, Tomasi J, Cossi M, Rega N, Millam JM,
Klene M, Knox J, Cross J, Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann R,
Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski J, Martin R, Morokuma K,
Zakrzewski V, Voth G, Salvador P, Dannenberg J, Dapprich S, Daniels A, Farkas Ö,
Foresman J, Ortiz J, Cioslowski J, Fox D, Gaussian 09, revision D.1, Gaussian Inc.,
Wallingford, CT.
37. Zhao Y, Truhlar G, J. Phys. Chem. A 2004; 108: 6908.
38. Zhao Y, Truhlar G, J. Phys. Chem. A 2005; 109: 5656.
39. Zhao Y, Tishchenko O, Truhlar G, J. Phys. Chem. B 2004; 109: 19046.
40. Godbout N, Salahub DR, Andzelm J, Wimmer E, Can. J. Chem. 1992, 70, 560.
18