Metal-Free Halogen(I) Catalysts for the Oxidation of Aryl(heteroaryl)methanes to Ketones or Esters: Selectivity Control by Halogen Bonding

Article abstract of DOI:10.1002/chem.201801717

Metal-free halogen(I) catalysts were used for the selective oxidation of aryl(heteroaryl)methanes [C(sp³)?H] to ketones [C(sp²)=O] or esters [C(sp³)?O]. The synthesis of ketones was performed with a catalytic amount of NBS in DMSO solvent. Experimental studies and density functional theory (DFT) calculations supported the formation of halogen bonding (XB) between the heteroarene and N-bromosuccinimide, which enabled imine–enamine tautomerism of the substrates. No additional activator was required for this crucial step. Isotope-labeling and other supporting experiments suggested that a Kornblum-type oxidation with DMSO and aerobic oxygenation with molecular oxygen took place simultaneously. A background XB-assisted electron transfer between the heteroarenes and halogen(I) catalysts was responsible for the formation of heterobenzylic radicals and, thus, the aerobic oxygenation. For selective acyloxylation (ester formation), a catalytic amount of iodine was employed with tert-butyl hydroperoxide in aliphatic carboxylic acid solvent. Several control reactions, spectroscopic studies, and Time-Dependent Density Functional Theory (TD–DFT) calculations established the presence of acetyl hypoiodite as an active halogen(I) species in the acetoxylation process. With the help of a selectivity study, for the first time we report that the strength of the XB interaction and the frontier orbital mixing between the substrates and acyl hypoiodites determined the extent of the background electron-transfer process and, thus, the selectivity of the reaction.

Full text of DOI:10.1002/chem.201801717

A Journal of
Accepted Article
Title: Metal-free Halogen(I) Catalysts for Oxidation of (Aryl)(heteroaryl)-
methanes to Ketones or Esters : Selectivity Control by Halogen
Bond
Authors: SOMRAJ GUHA and Govindasamy Sekar
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201801717
Link to VoR: http://dx.doi.org/10.1002/chem.201801717
Supported by

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
Metal–free Halogen(I) Catalysts for Oxidation of (Aryl)(heteroaryl)-
methanes to Ketones or Esters : Selectivity Control by Halogen
Bond
Somraj Guha,[a] and Govindasamy Sekar*[a]
Abstract: Use of metal–free halogen(I) catalysts has been
described for the selective oxidation of (aryl)(heteroaryl)methanes
(C(sp3)–H) to ketones (C(sp2)=O) or esters (C(sp3)–O). The
synthesis of ketone is carried out with catalytic amount of NBS in
DMSO solvent. Experimental studies and Density Functional Theory
(DFT) calculation supports the formation of halogen bond (XB)
between the heteroarene and NBS which enables imine–enamine
tautomerism of the substrates. No additional activator is required for
this crucial step. The isotope labelling and other supporting
experiments suggest that a Kornblum type oxidation with DMSO and
an aerobic oxygenation with molecular oxygen take place
simultaneously. A background XB assisted electron transfer (ET)
between heteroarenes and halogen(I) catalysts is responsible for the
formation of heterobenzylic radical and thus the aerobic oxygenation.
For selective acyloxylation (ester formation), catalytic amount of
iodine is employed with TBHP in aliphatic carboxylic acid solvent.
Several control reactions, spectroscopic study and TD–DFT
calculations establish the presence of acetyl hypoiodite as an active
halogen(I) species in acetoxylation process. With the help of a
selectivity study, for the first time we report that the strength of the
XB interaction and the frontier orbital mixing between the substrates
and acyl hypoiodites determine the extent of background ET process
and thus the selectivity of the reaction.
salts)[7] during a reaction is still unexplored. A well–thought–out
research on this topic can make the halogen–based reagents a
more versatile metal–free alternative. This possibility prompted
us to investigate how the easily available electrophilic halogen(I)
reagents can act as halogen bond donor catalysts,[8] and how
halogen–bonding interaction can influence the reactivity of the
halogen(I) reagents and in situ–generated halogen(I)
intermediates.[9] As a part of this ongoing research, herein, we
report
a
transition metal–free selective oxidation of
(aryl)(heteroaryl)methanes (C(sp3)–H) to ketones (C(sp2)=O) or
esters (C(sp3)–O) and the vital role of halogen bond in
maintaining the selectivity of these processes (Scheme 1).
Halogen–based
reagents
are
employed
as
efficient
organocatalysts and play important roles in developing transition
metal–free organic reactions.[1] Recently, the applications of
halogen–bonding (XB), an attractive noncovalent interaction
between the electrophilic region ( –hole) of
a substituted
halogen atom and
a Lewis base, has been successfully
introduced in organic synthesis.[2] In spite of its unorthodox
nature,[2a] this halogen–bonding interaction is already reported
as
a
competent tool for activation of functional group,[3]
photoactivation of halides to generate radical intermediates,[4]
synthesis of recyclable catalysts,[5] and transformation of
gaseous compounds to easily–handled condensed–phase liquid
reagents.[6]
The application of halogen bond in organic synthesis is still in its
nascent stage. Importantly, the role of halogen–bonding
interaction in controlling the reactivity of halogen based reagents
(such as influence of ligands on the reactivity of transition metal
[5]
Scheme 1. General mechanistic approaches for the oxidation of
heterobenzylic C(sp3)–H bond and importance of the heterobenzylic C(sp2)=O
and C(sp3)–O bonds containing products.
[a]
Somraj Guha, Prof. G. Sekar
Department of Chemistry, Indian Institute of Technology Madras
Chennai 600036, India
Eꢀmail: gsekar@iitm.ac.in; url: http://chem.iitm.ac.in/faculty/sekar/
The selective oxidation of heterobenzylic C(sp3)–H bonds to
C(sp2)=O or C(sp3)–O bonds are useful and straightforward
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
methods to obtain heterobenzyl ketones or alcohols (or their Results and Discussion
protected form, esters) respectively. So far, most of the reported
procedures for direct heterobenzylic oxidation describe the
formation of only C(sp2)=O bond (ketone) and not the C(sp3)–O
bond (alcohols or esters) at the heterobenzylic position.[10] In
fact, to the best of our knowledge, there are no reports available
on the selective partial oxidation of heterobenzylic C(sp3)–H
bonds to alcohols and very few reports are available on the
selective acetoxylation at heterobenzylic position.[11] Moreover,
We started our investigations using
bromosuccinimide (NBS)
as a halogen(I) source and 2–benzylpyridine 6a as a model
substrate. Our preliminary investigation was to get the proper
insight into the nature of interaction between the 2–
benzylpyridine and NBS by quantum chemical calculations in
vacuum (Figure 1A) and in DMSO solvent (Figure 1B). For
comparison, the 2–benzylpyridine/acetic acid and 2–
benzylpyridine/triflic acid systems were studied as well. NBS
was observed to act as a halogen bond donor (in both vacuum
and DMSO). Acetic acid was observed to act as a hydrogen
bond donor and not as a Bronsted acid to activate the pyridine
ring (in both vacuum and DMSO). On the contrary, triflic acid is
more likely to act as a Bronsted acid to form a “close to ionic”
intermediate (in both vacuum and DMSO). The XB interaction
energy between NBS and 2–benzylpyridine (ꢀ36.7 kcal molꢀ1 in
vacuum and ꢀ24.3 kcal molꢀ1 in DMSO) is very similar to the HB
interaction energy between acetic acid and 2–benzylpyridine (–
34.4 kcal molꢀ1 in vacuum and ꢀ24.4 kcal molꢀ1 in DMSO). On
contrary, the interaction energy between triflic acid and 2–
benzylpyridine is extremely high (–49.2 kcal molꢀ1 in vacuum and
ꢀ35.4 kcal molꢀ1 in DMSO).
these
heterobenzylic
acetoxylations
require
substrate
preactivation by N–oxide formation[11cꢀe] or excess amount of
transition metals under high pressure of oxygen.[11a, 11b]
There are two major facts behind the difficulties in selective
oxidation of heterobenzylic C(sp3)–H bonds to C(sp3)–O bonds.
Firstly, most of the reported oxidation procedures follow a radical
pathway where the heterobenzylic radical 1 is formed as a
intermediate and molecular oxygen is the sole oxidant (Scheme
1(a)). This highly reactive benzylic radical 1 immediately reacts
with molecular oxygen to form peroxo intermediate 2 which
converts to ketone via elimination of water. Secondly, if some
alcohols are formed by the cleavage of O–O bonds in 2, they get
further oxidized to ketones due to the presence of radical
species in the reaction mixture. However, Maes et al. reported
an alternative pathway that excludes the formation of
heterobenzylic radicals for the selective oxidation of
heterobenzylic C(sp3)–H to ketones.[10g] In this case,
stoichiometric amount of acetic acid activates the pyridine ring
and enables an imine–enamine tautomerism in the first step.
The enamine can attack electrophilic Cu(II) species to form
organometallic intermediate 3 followed by 4 and finally the
ketone (Scheme 1(b)).[10d]
In 2008, Bolm et al. reported the halogen–bonding activation of
pyridine ring with highly fluorinated bromoalkanes and
iodoalkanes.[12] We envisioned that the electrophilic halogen(I)
reagents or intermediates can activate the pyridine ring and
enable imine–enamine tautomerism. Furthermore, the enamine
can attack the electrophilic halogen(I) center to form 5, a non–
organometallic analogue of 3. At this point, nucleophilic
substitution of halide at benzylic position of 5 with DMSO can
lead to the formation of ketone (Kornblum oxidation)[13] and the
same with carboxylic acids (acetic acid) can lead to the
formation of esters (acetates), a protected form of alcohols
(Scheme 1(c)). For the first time, the oxygenated solvents are
utilized as the source of oxygen for this kind of heterobenzylic
oxidation. The two independent reactions follow a common
mechanistic route and a simple change in solvent helps to obtain
two different oxidized products (ketones and esters) selectively.
Both of these oxidized products with C(sp2)=O and C(sp3)–O
bonds often serve as biologically and medicinally active
compounds (Scheme 1(d)).[14] Moreover, the number of
heteroaromatic ring containing oral drugs is increasing in
market.[15] Presence of pyridine and thiazoles is very common in
F.D.A–approved drugs.[16] Hence, the development of metal–free
protocol for the functionalization of heterobenzylic C–H bond is
urgent and important to avoid even trace amount of transition
metal salt contamination with these compounds.
Figure 1. Structure of (a) NBS/2–benzylpyridine (b) acetic acid/2–
benzylpyridine and (c) triflic acid/2–benzylpyridine optimized with DFT using
M06–2X–D3 functional and 6ꢀ311G(d,p) for C,H; 6ꢀ311+G(d,p) for N,O;
LANL08d basis set in conjunction with the LANL2DZ effective core potential
(ECP) for Br in A) vacuum; B) in DMSO (IEFPCM) .[17]
After obtaining information from computational studies, the 1H
NMR shifts of the heterobenzylic protons on interaction of
pyridine ring with acetic acid, NBS, and triflic acid were checked
(Figure 2). An almost equal downfield shift of the heterobenzylic
protons was observed on interaction of pyridine ring with acetic
acid ( = 4.14 ppm to = 4.17 ppm) and NBS ( = 4.14 ppm to
= 4.19 ppm). On the other hand, the shift of the heterobenzylic
protons was remarkably high ( = 4.14 ppm to = 4.34 ppm)
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
10g]
when triflic acid (a strong acid) was allowed to interact with the
pyridine ring. These computational and 1H NMR analyses
indicated that NBS can be an appropriate alternative to acetic
acid for activation of pyridine ring and can enable the imine–
enamine tautomerism.
imine–enamine tautomerism of 6a,[10d,
thus, NBS should
enable the imine–enamine tautomerism of 6a too.
The formation of halogen bond between NBS and 2–
1
benzylpyridine was further confirmed when the H and 13C NMR
spectra of 1:1 mixture of NBS and 2–benzylpyridine was
recorded in CDCl3 (Figure 3). The 1H NMR peak assigned to the
methylenic protons of the succinimide shifted upfield ( = 2.97
ppm to
carbonyl carbon of the succinimide moiety shifted downfield (
173.2 ppm to = 175.2 ppm, (Figure 3(b)). Both these
= 2.86 ppm, (Figure 3(a)). The 13C NMR peak of
=
observations are in accordance with the previous report on
halogen–bonded adducts of N–halosuccinimide.[18] Furthermore,
when the spectra of a 1:1:1 mixture of NBS, 2–benzylpyridine
and DMSO was recorded in CDCl3, more downfield shift was
observed for the carbonyl carbon of the succinimide moiety.
Figure 3. (a) 1H NMR shifts assigned to the methylenic protons of the
succinimide and (b) 13C NMR shifts of carbonyl carbon of the succinimide on
interaction of pyridine ring with acetic acid, NBS, and triflic acid in CDCl3.
After confirming that NBS can enable an imine–enamine
tautomerism by the halogen–bonding activation of pyridine, the
oxidation of 2–benzylpyridine (1 mmol) 6a was tried with
stoichiometric amount of NBS (1 equiv) in of DMSO (2 mL) at
100 oC. Halogen bonds are weak non–covalent interactions.
However, a number of reports are available where halogen–
bonding and hydrogen–bonding catalysis occurs at high
20]
temperature.[3d,
Gratifyingly, 92% yield of ketone 7a was
isolated after 8 h from the commencement of the reaction (entry
1, Table 1). Use of 1 equivalent of –iodosuccinimide (NIS)
resulted in almost the same yield (90%) after 4 h (entry 2). A
large drop in the yield was observed when NIS (30 mol %) was
used (entry 3). However, 90 % yield of the ketone was observed
with only 30 mol
%
of NBS after 72
h
(entry 4).
Figure 2. 1H NMR shifts of the heterobenzylic protons due to interaction of
pyridine ring with acetic acid, NBS, and triflic acid in CDCl3.
Bromodimethylsulfonium bromide (BDMS, 30 mol%) also
worked efficiently (entry 5) but catalytic iodine was not effective
for this reaction (entry 6).
Absence of DMSO in the solvent system completely shut down
the reaction (entries 7 and 8) and a 1:1 mixture of DMSO and
water provided only 40% yield of the ketone (entry 9). The
volume of DMSO could be decreased up to 1 mL (entry 10, 92%
yield) but further decrease of solvent volume to 0.5 mL caused a
large drop in yield (entry 11, 62% yield). The amount of catalyst
could be decreased up to 20 mol % without any noticeable
change in reaction rate and product yield (entry 12). To our
pleasure, the long reaction time of 72 h could be reduced to 24 h
with a slight increase in the reaction temperature to 120 oC
(entry 13). Decreasing the reaction temperature to 80 oC led to a
drastic reduction in the yield of the ketone (entry 14).
A complete mechanistic study was performed to probe the
reaction pathway. Initially, our priority was to identify the key
halogen(I) active species in the reaction mixture. The reaction
could be started with either NBS or bromodimethylsulfoxonium
ion [DMSO–Br]+. It has already been reported by Sudalai et al.
Finally, the 1H NMR shifts of the heterobenzylic protons on
interaction of pyridine ring with acetic acid, NBS, and triflic acid
were checked in DMSO–d6 to verify the presence of XB
interaction in a polar solvent such as DMSO. Both the HB
interaction between acetic acid with 6a and the XB interaction
between NBS and 6a was appeared as weak interaction in
DMSO (Figure S12, Supporting Information). The binding
1
constant of the XB and HB complexes were calculated with H
NMR titration followed by fitting the shifts (1:1 model) in
Bindfit.[17, 19] . The binding constant of XB complex between NBS
and 2–benzylpyridine
constant of HB complexation between acetic acid and 2–
benzylpyridine
(HB) = 0.03 ± 0.001 Mꢀ1. However, this
(XB) = 1.17 ± 0.04 Mꢀ1 and the binding
observation established that NBS can interact with 6a more
effectively than with acetic acid in DMSO solvent. As the
previous reports already established that acetic acid can enable
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
that bromodimethylsulfoxonium ion intermediate is formed due
to the interaction of DMSO with NBS.[21] To prove the formation
Finally, the possible reversible process associated with the
interaction between NBS and 6a (Figure 4(b), Equation 1) and
the interaction between NBS and DMSO (Figure 4(b), Equation
Table 1: Optimization of reaction conditions for ketone formation[a]
2) were thermodynamically quantified. The
(–Log ) for
these two processes were calculated by DFT.[17] The self–
consistent reaction field (scrf) model was used to account for
solvent effects of DMSO. The results showed that the
of
reaction (ii) is significantly higher than the of reaction (i)
(Figure 4(b)). The formation of halogen–bonded complex
between NBS and 6a is more thermodymanically favorable than
the formation of bromodimethylsulfoxonium ion by dissociation
of NBS in DMSO. Hence, the initial reaction started with the
interaction of NBS with 2–benzylpyridine.[22]
entry
catalyst
solvent
temp (oC)
time (h)
yield (%)
1
NBS[b]
NIS[b]
NIS
DMSO
100
100
100
100
100
100
100
100
100
100
100
100
120
80
8
92
90
30[c]
90
82
43[c]
ꢀ
2
DMSO
4
3
DMSO
72
72
72
48
48
48
48
72
72
72
24
48
4
NBS
DMSO
5
BDMS
Iodine
NBS
DMSO
6
DMSO
7
toluene
CH3CN
DMSO/water[d]
DMSO[e]
DMSO[f]
DMSO[e]
DMSO[e]
DMSO[e]
8
NBS
ꢀ
9
NBS
40
92
62
94
96
10
Figure 4. (a) 13C NMR spectra of NBS (spectra A) and 1:1 mixture of NBS and
10
11
12
13
14
NBS
2–benzylpyridine (spectra B) in [D6]DMSO (b) Calculated
for the
NBS
formation of halogen–bonded adduct between NBS and 6 [Equation (i)] and
the same for the formation of bromodimethylsulfoxonium ion [Equation (ii)]
(M06–2X–D3/6ꢀ311G(d,p) for C,H; 6ꢀ311+G(d,p) for N,O; LANL08d basis set
in conjunction with the LANL2DZ effective core potential (ECP) for Br
/IEFPCM(DMSO)).[17]
NBS[g]
NBS[g]
NBS[g]
[a] Reaction conditions: 6a (1 mmol), catalysts (30 mol %), solvent (2 mL);
isolated yield, [b] catalyst (1 equiv), [c] 30–35% alcohol phenyl(pyridinꢀ2ꢀ
yl)methanol was isolated [d] 1:1 ratio of DMSO and water, [e] DMSO (1 mL), [f]
DMSO (0.5 mL), [g] NBS (20 mol %)
Several controlled reactions were carried out to predict the
complete reaction pathway. First, the diphenylmethane 8 was
taken as a starting material to check the importance of the
imine/enamine tautomerism in this reaction. No benzophenone 9
was observed after 24 h of the reaction (Scheme 2(a)). This
result suggested that the heterobenzylic C–H bonds can be
selectively oxidized with this method and that imine–enamine
tautomerism is a mandatory step in the pathway.
of this intermediate, Sudalai and co–workers have reported that,
two characteristic 13C peaks for the carbonyl groups of NBS
were observed when equimolar amount of NBS and DMSO were
mixed and the 13C NMR of the same was recorded in CDCl3.
According to them, one peak was observed at = 177.8 ppm
(corresponding to free succinimide anion carbonyl peaks) and
The reaction was carried out in DMS18O (75 atom % 18O) solvent
to find the source of the oxygen in this oxidation process. It was
observed that the 18O–labeled ketone 18O–7a was formed along
with 16O–labeled ketone 7a with the ratio of 0.72:1 (Scheme
2(b)). However, only 2% of the product was 18O–labeled ketone
18O–7a when the reaction was carried out in normal DMS16O (98
atom % 16O, Scheme 2(c)). The reaction was also carried out
with 2 equiv of H218O in the presence of dry DMS16O (98 atom %
16O). The 18O–labeled ketone 18O–7a was formed along with
16O–labeled ketone 7a with the ratio of 0.02:1 (Scheme 2(d)).
These 18O–labeling experiments suggested that the reaction
followed two competitive pathways. In one pathway, the
heterobenzylic radical was formed in the reaction mixture and
quenched by aerial oxygen. In another pathway, the
heterobenzylbromide was formed which underwent nucleophilic
substitution with DMSO (Kornblum oxidation), as per our
expectation.
another was found at
DMSO coordinated NBS and showed sufficient downfield shift
with respect to free NBS in CDCl3, = 173.2 ppm).[21]
= 179.4 ppm (corresponding to the
However, to our surprise, when we recorded the 13C NMR
spectrum of equimolar amount of NBS and DMSO in presence
of equimolar amount of 2–benzylpyridine 6a in CDCl3, only one
peak at = 178.2 ppm was observed (Figure. 3(b), Spectra A).
This observation indicated that bromodimethylsulfoxonium ion is
not formed in the presence of 6a. The halogen–bonded complex
between NBS and 6a might be more thermodynamically stable
than the complex between NBS and DMSO.
To further prove this hypothesis, the 13C NMR spectrum of NBS
was recorded in [D6]DMSO. The characteristic peak of
succinimide anions were observed at = 177.5 ppm (Figure 4(a),
spectrum A). Gratifyingly, when 13C NMR spectrum of equimolar
mixture of 6a and NBS was recorded in [D6]DMSO, the peak at
= 177.5 ppm disappeared (Figure 4(a), spectrum B). It clearly
proved that NBS was not decomposing in DMSO in the
presence of 6a.
Several radical scavengers were screened at optimized
condition for the oxidation of 6a to 7a to detect the background
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
radical pathway.[17] In presence of TEMPO (1 equiv), the
reaction was unaffected (Scheme S5, entry 1, supporting
information). However, the yield of the product was decreased to
75% when BHT (1 equiv) was used (Scheme S5, entry 2). The
amount of BHT was increased to 2 equiv and the yield of 7a was
decreased up to 60% (Scheme S5, entry 3). It was observed
that the yield of 7a did not decrease further even after increasing
the amount of BHT up to 4 equiv (Scheme S5, entry 4, 5).
2–benzylpyridine enables the imine–enamine tautomerism. This
tautomerism can be represented with the structures 10 and 11.
The enamine attacks NBS to form the heterobenzylbromide 12.
The heterobenzylbromide 12 undergoes Kornblum oxidation to
form the ketone 7a via intermediate 13. The byproduct HBr is
immediately oxidized by DMSO to form bromodimethylsulfonium
bromide (BDMS) intermediate 14. This BDMS can act as an
alternative halogen(I) source for NBS and can convert the 2–
benzylpyridine to heterobenzylbromide 12 (entry 5, Table 1).
Alternatively, this unstable intermediate can be transformed to
[Me2S→Br2] charge–transfer complex 14a which further reacts
with succinimide to regenerate NBS.[24] The halogen–bonding
interaction between NBS and 6a can also enable a background
electron transfer (ET) from 6a to NBS and the radical ion pair
10a is generated.[23b] The radical cation of 6a can be converted
to heterobenzylic radical 15 by eliminating a proton. The ketone
7a is formed by the quenching of oxygen with 15.
Scheme 2. Controlled experiments to probe the mechanistic pathway
Scheme 3. Plausible mechanistic pathways for ketone formation
We also tried to find the possible reason behind the formation of
heterobenzylic radical as a background reaction. The possibility
of thermal decomposition of NBS to generate bromine radical
(Br●) followed by the abstraction of hetero benzylic C–H bond to
form heterobenzylic radical can be ruled out. This was confirmed
when no ketone was formed from diphenylmethane 8 under
optimized condition in spite of the negligible difference in
homolytic bond dissociation energy (BDE) of benzylic C–H of 8
(326 kJ moleꢀ1) and 6a (320 kJ moleꢀ1)(Scheme 2(a)).[10d]
Furthermore, when 8 was subjected to oxidation under the
standard conditions in the presence of 2–MePyridine (which
resembled 2–benzylpyridine and can act as a XB acceptor),
The scope of this metal–free protocol was tested for the benzylic
oxidation of a number of heteroarene substrates (Scheme 4).
Good to excellent yields of ketones were observed for all the 2–
benzylpyridines that contain electron–donating as well as
electron–withdrawing arenes (7b–7i). When the arene ring is
substituted by thiophene ring, yield of the ketone was reduced to
61% (7j). Substrate that contained 4–substituted pyridine also
provided the corresponding product in good yield (7k, 85%).
However, substrate that contained 3–substituted pyridine
couldn’t result in any oxidized product (7l). This observation
again established the vital role of XB assisted imine–enamine
tautomerism of the substrates in this oxidation protocol.
ketone
9 was not formed (Scheme 2e). Hence, it was
hypothesized that a halogen bond assisted electron transfer
process might be the reason behind the formation of the
heterobenzylic radical. Roshokha and coworkers reported that
halogen bonds can result in a significant lowering of the
activation barrier for the electron transfer (ET) from the XB
acceptor to XB donor.[23] The electron transfer from 6a to NBS
could generate the radical cation of 6a which successively
transformed to the heterobenzylic radical.
Several 2–benzylbenzothiazoles were also found to provide
good yield of ketones (7m–7q). Substrate with simple benzene
ring at heterobenzylic position provided 57% yield of ketone
(7m).
A variety of 2–benzylbenzothiazoles with electron–
withdrawing arenes, electron–donating arenes and naphthalene
ring present at the heterobenzylic positions were tested and
good yields of ketones were obtained (7n–7q). Finally, the
reaction was tried for a 1 g scale and 87% yield of the ketone
was obtained which showed the scalability of the process (entry
7a).
Based on the supporting experiments and the controlled
reactions, the mechanistic pathway has been proposed for this
reaction (Scheme 3). The halogen bond between the NBS and
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
After the development of NBS catalyzed oxidation of
heterobenzylic C(sp3)–H bonds to C(sp2)–O bond, we tried to
develop the selective oxidation of heterobenzylic C(sp3)–H
bonds to C(sp3)–O bond. The 2–benzylpyridine 6a was taken as
the model substrate and acetic acid was taken as the
nucleophilic solvent. However, the mechanistic study for the
tert–butylhydroperoxide (TBHP, 5–6 (M) in decane, 2 equiv)
(Scheme 5(d)). A quick optimization of reaction conditions[17]
revealed that only iodine (30 mol %) is required along with TBHP
(2 equiv, 5–6 (M) in decane) and AcOH (2 mL) at 100 oC to
complete the reaction in 1.5 h with 85% yield of acetoxylated
product (Scheme 6).
Scheme 5. Preliminary screening of conditions to increase the selectivity of the
acetoxylation of 6a
Scheme 4. Substrate scope for the NBS catalyzed hetero–benzylic oxidation.
ketone formation revealed the possibility of a background ET
reaction which could lead to the formation of heterobenzyl
radical 15 followed by the formation of the ketone 7a (Scheme 3).
As per the assumption, a major problem of selectivity was
encountered when NBS (1 equiv) was heated with 6a (1 equiv)
at 100 oC in the presence of acetic acid (3 mL). The reaction
was completed within 15 h, but, 35% yield of the acetoxylated
product 16a was isolated along with 40% yield of the ketone 7a
(Scheme 5(a)).
To get better selectivity, several electrophilic iodine sources
were checked as catalyst. Only 20% yield of the acetoxylated
product was obtained with 20% yield of the ketone when
molecular iodine (1 equiv) was employed (Scheme 5(b)). Better
selectivity for the formation of acetoxylated product 16a over the
ketone 7a was observed when more electrophilic iodine sources
such as NIS was employed as a halogen(I) source (Scheme
5(c)). However, the yield of acetoxylated product 7a couldn’t be
raised.
Next, we thought to employ the in situ–generated hypoiodites as
active species. The in situ–generated inorganic hypoiodites and
iodates are well known metal–free alternative for oxidative
coupling reactions.[25] These electrophilic hypervalent iodines are
unstable but can be generated in situ by the reaction of iodides
or iodine with peroxides.[26] Gratifyingly, 85% yield of the
acetoxylated product 16a was obtained with only 10% of 7a
within 1.5 h of the commencement of the reaction when 6a was
treated with iodine (precatalyst, 30 mol % o) in the presence of
Scheme 6. Optimized reaction conditions for acetoxylation of 6a
The high selectivity of this process prompted us to go through a
detailed mechanistic investigation and to find the key factor that
controlled the selectivity of acetoxylation over the ketone
formation. Few controlled experiments were designed to find the
key intermediate of this oxidation protocol. When the reaction
was carried out without the iodine precatalyst, neither
acetoxylation nor the ketone formation was observed after 3 h
from the commencement of the reaction (Figure 5A, Equation (i)).
The presence of the radical trapper TEMPO enhanced the rate
of the reaction and the yield of the product was not reduced
(Figure 5A, Equation (ii)). This indicated the possibility of a non–
radical pathway and the formation of non–radical intermediates
in the reaction.
Iodine can be converted to hypoiodous acid (HOI) or hypoiodites
26cꢀf]
(IOꢀ) in the presence of peroxides.[26a,
According to the
previous reports, hypoiodous acid can be converted to transient
acetyl hypoiodite (CH3COOI) in the presence of acetic acid.[26b,
27]
Acetyl hypoiodite is a well–known electrophilic iodine(I)
source which cannot be isolated from the reaction mixture but
can be generated in situ and can be used for the
stereoselective dihydroxylation of olefins,[28] iodination of
aromatic compounds,[29] cleavage of vicinal diols,[30] and
oxidation of alcohols.[9a, 31] We envisaged that acetyl hypoiodite
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
might be the active halogen(I) intermediate in this acetoxylation
reaction.
acetyl hypoiodite than molecular iodine (Figure 5B). Hence, the
interaction energy between 2–benzylpyridine and acetyl
hypoiodite (ꢀ33.0 kcal molꢀ1) was higher than that of 2–
benzylpyridine and molecular iodine (ꢀ25.6 kcal molꢀ1). The
stronger XB interaction enabled the imine–enamine tautomerism
more efficiently. Thus, the acetoxylation reaction was more
efficient in iodine/TBHP/acetic acid system where acetyl
hypoiodite is the key XB donor species than iodine/acetic acid
system where iodine is the key XB donor species (Scheme 5d
and 5b respectively).
The imine–enamine tautomerism of 2–benzylpyridine 6a could
also be enabled by the formation of hydrogen–bonded complex
between 6a and acetic acid. However, the calculation of change
in free energy associated with two independent processes
demonstrated that the formation of halogen–bonded adduct
between 6a and acetyl hypoiodite is more feasible (
kcal molꢀ1) than the formation of hydrogen–bonded adduct
between 6a and acetic acid (
= ꢀ0.8 kcal molꢀ1, Figure 6A).
= ꢀ7.8
Thus, acetylhypoiodite was the key intermediate which enabled
the imine–enamine tautomerism of 6a.
Finally, the halogen–bonding interaction between 6a and acetyl
hypoiodite was demonstrated using UV–vis spectroscopy and
molecular modeling with time dependent DFT (TD–DFT)
approach. A mixture of 1:1:1 iodine, TBHP and acetic acid was
stirred for 0.5 h at 100 oC and the UV–vis absorption of the same
was recorded in acetic acid at 10ꢀ4 (M) concentration.[17] An
absorption band appeared with the maxima at 474.9 nm for
acetyl hypoiodite. However, when 6a (1 equiv) was added to the
same solution and the UV–vis spectra of the sample was
recorded, the band at 474.9 nm completely disappeared and two
new bands appeared with the maxima at 358.7 nm (blue–shifted
Figure 5. A) Controlled experiment to probe the mechanistic route and the key
intermediate for the acetoxylation of 6a; B) Electrostatic potential map of (i)
iodine and (ii) acetyl hypoiodite at 0.001 electron bohrꢀ3 isodensity surface
highlighting the ꢀhole region; C) Optimized structure and calculated
interaction energy for (i) 6a.Iodine complex and (ii) 6a.acetyl hypoiodite
complex. (M06–2X–D3/6ꢀ311G(d,p) for C,H; 6ꢀ311+G(d,p) for N,O; LANL08d
basis set in conjunction with the LANL2DZ effective core potential (ECP) for I/
IEFPCM(acetic acid))
The detection of hypoiodites in the reaction mixture is difficult
due to its transient nature. However, Wei and coworkers were
band, λBS) and 291.3 nm (charge–transfer band, λCT) (Figure
–
able to detect the iodite (IO2 ) and hypoiodite ions (IOꢀ) with
17, 32]
6B).[9b,
The charge–transfer (CT) band merged with the
mass spectroscopy.[26e] Pleasingly, we were also able to detect
broad absorption band of 6a with the maxima at 270 nm. To
overcome this problem, the UV–vis spectra of the same sample
were recorded with a 10ꢀ4 (M) solution of 6a in acetic acid as
blank. The absorption band of 6a was automatically cancelled
out and the charge–transfer band with the maxima at 291.3 nm
clearly emerged (Figure 6C).[17]
The DFT calculation showed that the HOMO–LUMO gap of
acetyl hypoiodite increased upon halogen–bonding complex
formation with 6a (from 6.59 eV to 7.10 eV, Figure 6D). This
observation satisfactorily corroborated the observed blue shift in
UV–vis experiment (from 474.9 nm to 358.7 nm). The TD–DFT
calculation predicted the presence of one absorption maxima at
360.1 nm with low oscillator strength ( = 0.0081) and another
absorption maxima with higher oscillator strength ( = 0.0505)
than the previous one for the XB complex of CH3COOI with 6a.
These results corroborated with the experimental values, i.e.
358.7 nm with low absorbance and 291.3 nm with higher
absorbance.
acetyl hypoiodite (
/
186.10 [M+]) with GC–MS in an 1:1:1
mixture of iodine, TBHP and acetic acid.[17] Moreover, when the
reaction was carried out with iodine (1 equiv) in the presence of
silver acetate (AgOAc, 1 equiv) and in the absence of any
oxidizing agent, the reaction was completed within 1.5 h and
65% of the acetoxylated product 16a was obtained along with
30% of the ketone (Figure 5A, Equation (iii)). However, no
product was formed when only AgOAc (1 equiv) was employed
in the absence of iodine source (Equation (iv)). These controlled
reactions confirmed the active hand of AcOI in the
accomplishment of the reaction as AcOI is formed irreversibly in
a 1:1 mixture of iodine and silver acetate in acetic acid.[28] Finally,
the reaction was carried with catalytic HI in absence and
presence of TBHP (Equation (v) and Equation (vi)). No products
were formed in the first case while 65% of acetoxylated products
16a along with 25% of ketones 7a were formed in the second
case. These reactions indicated that if HI is generating in the
reaction mixture it could not catalyze the reaction. The oxidation
of HI to iodine (I) active species (acetyl hypoiodite) was
mandatory for the completion of the reaction.
Furthermore, this calculation revealed that 43.7% of the 292.4
nm absorption (CT band) was contributed by the excitation of
electrons from HOMOꢀ1 to LUMO+1 and 52.1% was contributed
by the excitation of electrons from HOMOꢀ1 to LUMO+2 (Figure
6E). These orbitals were largely located on the nitrogen of
pyridine and the iodine atom of acetyl hypoiodite. Hence, the
Further quantum chemical studies of electronic distribution on
iodine and acetyl hypoiodite employing DFT showed that the
size of the σ–hole (the electro–positive region through which the
XB interaction takes place with Lewis bases) was higher in
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
TD–DFT simulation of the band at 291.3 nm clearly showed that
this band appeared due to excitation of the electrons
participating in the halogen–bonding interaction between the
pyridine moieties of 6a and the iodine of acetyl hypoiodite.
Hence, we screened the straight and branched chain aliphatic
acids having the
= 4.76). As a result, 60%, 53%, 45% and 43% yield of
acyloxylated product was obtained in propanoic acid (
values near the
value of acetic acid
(
=
4.86), butanoic acid (
= 4.83), pentanoic acid (p = 4.83)
a
and 2–methylpropanoic acid (
= 4.88) solvents respectively
(entries 3–6, Figure 7A). Although, the reactions proceeded in all
these solvents, the selectivity was lost and substantial amount of
the ketone 7a was isolated in the case of pentanoic acid and 2–
methylpropanoic acid (entries 5 & 6, Figure 7A).
The preliminary experiments indicated that,
a weak XB
interaction could not enable the imine–enamine tautomerism of
6a and thus the selectivity and efficiency of the reaction was less
(Scheme 5 and Figure 5). Hence, initially it was intuited that the
XB interactions between 6a and acyl hypoiodites were reduced
with the increase in the number of carbon in the aliphatic chain
of acyl hypoiodites. The increasing +I effect of the aliphatic
moiety of acyl hypoiodites could reduce the electrophilicity and
the size of –hole of iodine atom.
However, to our surprise, a counter–intuitive result was obtained
when we tried to compare the strength of XB interaction
between the acyl hypoiodites and 6a with DFT (Figure 7A and
7B). Gratifyingly, a thorough literature survey revealed that
noncovalent interactions such as halogen bond often shows
these kinds of uncommon trends.[34] These counter–intuitive
trends signify the charge–transfer character and substantial
mixing of frontier orbitals (HOMO of donor and LUMO of
acceptor) in a noncovalent adduct due to the presence of
energetically low–lying LUMO on the acceptor moiety (here, XB
donor).[34c] In addition, according to Rosokha and coworkers,
the mixing of frontier orbitals causes the lowering of the energy
Figure 6. A) Calculation of free energy change for (i) HB adduct formation
between 6a and acetic acid (ii) XB adduct formation between 6a and acetyl
hypoiodite. B) UV–vis spectra of 6a, I2+TBHP+acetic acid (1:1:1) and 6a+
barrier for electron transfer (ET) from XB acceptor to XB
35]
donor.[23b,
Hence, the energy of the HOMO of 6a and the
I2+TBHP+acetic acid (1:1:1:1) in acetic acid (path length
= 1 cm) at
energy of the LUMOs of acyl hypoiodites were calculated with
DFT to compare qualitatively the possibilities of mixing of
LUMOs of the acyl hypoiodites with the HOMO of 6a (Figure
7C) .[17]
concentrations of 10ꢀ4 (M) for each species. C) UV–vis spectra of 6a+
I2+TBHP+acetic acid (1:1:1:1) in acetic acid (path length
= 1 cm) at
concentrations of 10ꢀ4 (M) for each species keeping 10ꢀ4 (M) solution of 6a in
acetic acid as blank; D) Energy difference between the frontier orbitals of
CH3COOI (left) and of CH3COOI.6a. E) TD–DFT simulation of UV–vis
spectroscopic data of XB adduct of 6a and acetyl hypoiodite.
Pleasingly, this analysis clearly showed that other acyl
hypoiodites generated from propanoic acid, butanoic acid,
pentanoic acid and 2–methylpropanoic acid had energetically
low–lying LUMOs in comparison to the LUMO of acetyl
hypoiodites (Figure 7C). Thus, the possibility of the ET process
by orbital–mixing was increased in these solvent and was
highest for pentanoic acid and 2–methylpropanoic acid (entry 5
& 6, Figure 7A). This selectivity study indicated that, a particular
interaction energy between acyl hypoiodites and 6a was
required to maintain the selectivity of acyloxylation process. The
weaker XB interaction could not enable the imine–enamine
tautomerism and the stronger XB interaction increased the
possibility of background ET process. In both the cases, the
efficiency and the selectivity of the process were reduced.
Furthermore, this selectivity study pointed out that role of
halogen–bonding should be considered seriously during the
designing and mechanistic analysis of halogen–based
organocatalysis and the quantum chemical calculations can be a
helpful means in this aspect.
Next, we tried to find out the key factor that controlled the
selectivity of the acetoxylation over the ketone formation. For
this, the reaction was carried out in several aliphatic organic
acids with different
product 17a nor the ketone 7a was observed when a stronger
acid such as α–chloro acetic acid ( = 2.85) was used as the
values (Figure 7A). Neither acyloxylated
solvent (entry 2, Figure 7A). It is worth mentioning that iodine
remains in almost free molecular state (uncomplexed) in a
strong and polar acidic medium whereas it forms a complex (XB
conmplex) in weak acidic medium.[33] These XB complexes are
the pre–reactive species for the formation of acyl hypoiodites.
The peroxide increases the electrophilicity of iodine by oxidizing
it to iodine(I) and facilitates the formation of acyl hypoiodites. As
the formation of pre–reactive complex is an elusive process in
strong acid, it can be concluded that the required acyl hypoiodite
intermediate couldn’t be generated in –chloro acetic acid and
thus, no oxidative process was observed in this solvent.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
A plausible mechanism for the iodine catalyzed acyloxylation
reaction is designed based on the results of supporting
experiments and quantum chemical calculations (Figure 7D).
The acyl hypoiodite 18 is generated from the reaction of acid,
TBHP and iodine. This acyl hypoiodite 18 undergoes feasible
formation of XB adduct 19. Although, the substrate 6a can form
the HB adduct 20 with solvent acid, the process is less
thermodynamically favorable (Figure 6A).The intermediate 19
can be converted to the heterobenzyl iodide 22 by the means of
product 23. The by–product HI can further be oxidized to 18 in
the presence of TBHP. The ketone is generated by XB assisted
ET from 6a to 18 through the intermediacy of radical ion pair 19a
and heterobenzylic radical 15.
The scope of the substrates was investigated to probe the
efficiency of the process. The 2–benzylpyridines bearing
electron–rich arenes provided acetoxylated product in good to
excellent yields (16b–16f, Scheme 6). The methoxy group was
well tolerated (16b). Moreover, the acetoxylation was not
observed at other benzylic position when the starting materials
that contain more than one benzylic position were subjected to
the reaction conditions (16c–16f).
imine–enamine tautomerism
intermediate 21. The
nucleophilic substitution of iodide with acid RCOOH at
heterobenzylic position leads to the formation of acyloxylated
A
B)
C)
D)
Figure 7. A) Selectivity study using different organic aliphatic acids with different
. B) Optimized structures of the acyl hypoiodites; (M06–2X–D3/6ꢀ311G(d,p)
for C,H; 6ꢀ311+G(d,p) for N,O; LANL08d basis set in conjunction with the LANL2DZ effective core potential (ECP) for I) in vacuum;
interaction energy; = free energy change of complexation . C) Frontier orbitals of the 6a and acyl hypoiodites showing comparative possibilities of orbital
mixing. The energies of the orbitals were calculated with DFT.[17] D) Plausible reaction mechanism.
= BSSE corrected
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
OMe
Me
Et
Conclusions
N
N
N
N
The selective oxidation of (aryl)(heteroaryl)methanes (C(sp3)–
H) to ketones (C(sp2)=O) or esters (C(sp3)–O) with metal–free
halogen(I) catalysts is reported. The oxidation of
(aryl)(heteroaryl)methanes to ketone has been designed with
catalytic amount of NBS in DMSO. No external activator is
required for imine–enamine tautomerism, as, NBS itself acts
as the XB donor activator. Unlike the previous reports,
Isotope–labeling experiments reveal that the solvent with
nucleophilic oxygen atom can act as the source of oxygen.
OAc
OAc
OAc
OAc
1.5 h, 85%
1h, 75% (1gm scale)
16a
1 h, 72%
16b
1 h, 90%
16d
1.5 h, 67%
16c
Cl
F
N
N
N
N
Me
OAc Me
3 h, 65%
OAc
OAc
OAc
1.5 h, 88%
0.75 h, 65%
16g
0.5 h, 65%
OMe
16e
16f
16h
Me
Me
N
S
N
S
N
OAc
OAc
1 h, 56%
16i
N
OAc
OAc
15 h, 0%
16j[a]
However,
a
background XB–assisted ET process is
0.5 h, 55%
16k
0.25 h, 51%
16l
responsible for the generation of heterobenzylic radicals in the
reaction mixture which can be quenched with molecular
oxygen to generate the ketone too.
Ph
S
S
N
S
N
Cl
The acetoxylation reaction takes place in acetic acid in the
presence of catalytic amount of iodine and stoichiometric
amount of TBHP. The formation of acetyl hypoiodite is
established with the help of several control experiments, GC–
MS and UV–vis spectroscopy in combination with TD–DFT.
The vital role of halogen bond in controlling the background ET
process and maintaining selectivity of the reaction has been
established for the first time. This result encourages the new
possibilities of developing metal–free routes for oxidative
functionalization of heterobenzylic C–H bond with other
nucleophiles by avoiding the formation of ketone to access
new medicinally and biologically important molecules.
N
OAc
OAc
OAc
10 min, 62%
1.25 h, 30%
25 min, 56%
OMe
16m
16n
16o
Cl
S
S
Me
Me
O
N
O
N
O
O
20 min, 40%
16p[b]
3 h, 25%
16q[c]
Scheme 7. Substrate scope for iodine catalysed acyloxylation reaction. In all
the cases very trace amount of corresponding ketones have been formed.
[a] Neither ketone nor acetoxylated product were isolated. [b] 30%
corresponding ketone 7o was isolated. [c] 25% corresponding ketone 7n
was isolated.
Acknowledgements ((optional))
.
This observation highlighted the excellent chemo selectivity of
this process where only the heterobenzylic position could be
acetoxylated. The yield of the acetoxylated product was low in
case of the substrate bearing 2–methyl arene (16f). This
observation supported that the nucleophilic substitution at
heterobenzylic position by acetic acid could be a crucial step in
the mechanistic pathway.
The 2–benzylpyridines that contain electron withdrawing
arenes provided moderate yields of the acetoxylated product
and the reactions were very fast (16g and 16h). The reaction
was efficient for 4–benzylpyridine (16i) but did not work for 3–
benzylpyridine (16j). This observation pointed out the
importance of imine–enamine tautomerism during this
acetoxylation process. Several benzothiazoles were also found
to provide good yields of the acetoxylated products (16k–16o).
The yield was low in the case of the benzothiazole bearing 2–
substituted arene (16m). This observation further emphasized
the possibility of nucleophilic substitution at heterobenzylic
position by acetic acid. When the solvent was changed to
propanoic acid, the yield of the acyloxylated product was
reduced (16p, 16q). Finally, the reaction was tried at 1 g scale
and 75% yield of the acetoxylated product was obtained
proving the scalability of the process (16a).
We thank DST, New Delhi (SB/S1/OCꢀ72/2013), and IIT
Madras (CHY/17ꢀ18/847/RFIR/GSEK) for financial support.
Keywords: halogen bond • NBS catalyst• iodine catalyst •
selective oxidation • (Aryl)(heteroaryl)ꢀmethanes
[1]
a) H. Zhang, K. Muñiz,
A. Studer,
2017, , 4122ꢀ4125; b) X. Wang,
2017, , 1712ꢀ1724; c) S.ꢀK. Wang, X.
You, D.ꢀY. Zhao, N.ꢀJ. Mou, Q.ꢀL. Luo,
11757ꢀ11760; d) S. Wang, X. Zhang, C. Cao, C. Chen, C. Xi,
2017, , 4515ꢀ4519; e) S. Maksymenko, K. N. Parida, G. K.
Pathe, A. A. More, Y. B. Lipisa, A. M. Szpilman, 2017,
6312–6315; f) T. Dohi, D. Koseki, K. Sumida, K. Okada, S. Mizuno, A.
2017,
,
Kato, K. Morimoto, Y. Kita,
2017,
, 3503ꢀ3508;
g) G. L. Tolnai, U. J. Nilsson, B. Olofsson,
2016, , 11226ꢀ11230;
2016, 128,11392 –11396 h)
2016, , 15812ꢀ
A. H. Sandtorv, D. R. Stuart,
15815;
2016, 128 ,16044 –16047 i) I. Saikia, A. J.
2016, , 6837ꢀ7042; j) F. Malmedy,
2016, , 16072ꢀ16077; k) Y.ꢀF. Liang, X. Li,
Borah, P. Phukan,
T. Wirth,
X. Wang, M. Zou, C. Tang, Y. Liang, S. Song, N. Jiao,
2016,
, 12271ꢀ12277; l) M. S. Yusubov, V. V. Zhdankin,
2015, , 49ꢀ67; m) C. Huo, M. Wu, F.
2015, , 4708ꢀ4711;
2012, , 14638ꢀ14642; o)
2003, , 1407ꢀ
Chen, X. Jia, Y. Yuan, H. Xie,
n) M. Lamani, K. R. Prabhu,
N. E. Leadbeater, M. Marco,
1409;
2003,
, 1445.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
[2]
[3]
a) Y. Zhao, Y. Cotelle, N. Sakai, S. Matile,
2016,
LesterꢀZeiner, J. Danao, J. Able, M. Cueva, S. Talreja, T. Kornecook,
H. Chen, A. Porter, R. Hungate, J. Treanor, J. R. Allen,
, 4270ꢀ4277; b) G. Cavallo, P. Metrangolo, R. Milani, T. Pilati, A.
Priimagi, G. Resnati, G. Terraneo,
2601; c) D. Bulfield, S. M. Huber,
2016,
2016,
, 2478ꢀ
2014, , 6632ꢀ6641; b) R. Shankar, S. S. More, M. V. Madhubabu,
, 14434ꢀ
N. Vembu, U. K. Syam Kumar,
Easmon, G. Heinisch, G. Pürstinger, T. Langer, J. K. Österreicher, H.
H. Grunicke, J. Hofmann, 1997, , 4420ꢀ4425; d) E.
Frank, J. Gearien, M. Megahy, C. Pokorny, 1971,
551ꢀ553; e) A. Pelser, D. G. Müller, J. du Plessis, J. L. du Preez, C.
2012,
, 1013ꢀ1020; c) J.
14450; d) T. M. Beale, M. G. Chudzinski, M. G. Sarwar, M. S. Taylor,
2013, , 1667ꢀ1680.
a) D. von der Heiden, S. Bozkus, M. Klussmann, M. Breugst,
2017, , 4037ꢀ4043; b) M. Saito, Y. Kobayashi, S. Tsuzuki, Y.
,
Takemoto,
2017,
2017,
, 7653ꢀ7657;
Goosen,
2002, , 239ꢀ244; f) N. A. Salvi, S.
, 7761 –7765 c) J.ꢀP. Gliese, S. H. Jungbauer, S. M.
R. Udupa, A. Banerji,
Barouh, H. Dall, D. Patel, G. Hite,
1998,
, 201ꢀ203; g) V.
Huber,
2017, , 12052ꢀ12055; d) A. Matsuzawa, S.
1971, , 834ꢀ836.
Takeuchi, K. Sugita,
2016,
, 2863ꢀ2866; e) M.
[15] T. J. Ritchie, S. J. F. Macdonald, R. J. Young, S. D. Pickett,
Breugst, E. Detmar, D. v. d. Heiden,
2016, , 3203ꢀ3212;
2011, , 164ꢀ171.
f) Y. Takeda, D. Hisakuni, C.ꢀH. Lin, S. Minakata,
318ꢀ321; g) S. H. Jungbauer, S. M. Huber,
2015,
,
[16] E. Vitaku, D. T. Smith, J. T. Njardarson,
10257ꢀ10274.
2014,
,
2015,
, 12110ꢀ12120; h) L. Zong, X. Ban, C. W. Kee, C.ꢀH. Tan,
[17] See supporting information for details.
2014,
12043–12047 i) N. Tsuji, Y. Kobayashi, Y. Takemoto,
2014, , 13691ꢀ13694; j) W. He, Y.ꢀC. Ge, C.ꢀH. Tan,
2014, , 3244ꢀ3247.
, 11849ꢀ11853;
2014,
,
[18] I. Castellote, M. Moron, C. Burgos, J. AlvarezꢀBuilla, A. Martin, P.
GomezꢀSal, J. J. Vaquero,
2007, 1281ꢀ1283.
2016,
[19] a) D. Brynn Hibbert, P. Thordarson,
,
12792ꢀ12805; b) J. P. Wojciechowski, A. D. Martin, M. Bhadbhade, J.
[4]
[5]
[6]
Y. Wang, J. Wang, G.ꢀX. Li, G. He, G. Chen,
1445.
2017, 1442ꢀ
E. A. Webb, P. Thordarson,
2016, , 4513ꢀ4517; c)
P. Thordarson,
.
2011,
, 1305ꢀ1323; d)
S. Dordonne, B. Crousse, D. BonnetꢀDelpon, J. Legros,
2011, , 5855ꢀ5857.
[20] a) P. Zhangjin, F. Zhenwei, L. Beili, C. Jiajia,
, 1761ꢀ1767; b) V. V. D., R. Edward, W. Eléna, M. Joseph,
2017, , 3085ꢀ3089; 2017, , 3131ꢀ
2018,
F. Sladojevich, E. McNeill, J. Börgel, S.ꢀL. Zheng, T. Ritter,
2015,
, 3712ꢀ3716;
2015,
,
,
3783–3787
a) J. R. Khusnutdinova, D. Milstein,
3135
[21] P. K. Prasad, R. N. Reddi, A. Sudalai,
[7]
2015,
2016, , 500ꢀ503.
12236ꢀ12273;
M. Lautens,
2015,
, 12406 –12445b) K. Fagnou,
2002, , 26ꢀ47;
[22] The reaction mixture was injected into GCꢀMS during the course of
reaction and no peak for bromodimethyl sulfoxonium ion was
observed. This observation indicated that the bromodimethyl
sulfonium ion did not generated in reaction mixture at 120 oC too.
However, thermodynamically less stable bromodimethyl sulfonium ion
can form in the reactiom mixture via a kinetically controlled pathway
and can react with 6a. In this case, bromodimethyl sulfonium ion can
form XB interaction with pyridine ring of 6a and activate it and an
alternative mechanism for the reaction can be drawn as below :
2002,
, 26–49. c) P. W. N. M. van Leeuwen, P. C. J. Kamer, J. N.
H. Reek, P. Dierkes,
2000,
, 2741ꢀ2770; d) D. J.
Berrisford, C. Bolm, K. B. Sharpless,
1995,
,
1059ꢀ1070;
Austin,
. 1995,
, 1159. e) A. Padwa, D. J.
1994, , 1797ꢀ1815;
1994,
, 1881 –1899.
[8]
[9]
I. Kazi, S. Guha, G. Sekar,
2017, , 1244ꢀ1247.
a) S. Guha, I. Kazi, A. Nandy, G. Sekar,
, 5497ꢀ5518; b) S. Guha, I. Kazi, P. Mukherjee, G. Sekar,
2017, , 10942ꢀ10945.
2017,
2017,
[10] a) W. Jin, P. Zheng, W.ꢀT. Wong, G.ꢀL. Law,
, 1588ꢀ1593; b) D. P. Hruszkewycz, K. C. Miles, O. R. Thiel, S. S.
Stahl, 2017, , 1282ꢀ1287; c) H. Wang, Z. Wang, H.
Huang, J. Tan, K. Xu, 2016, , 5680ꢀ5683; d) H. Sterckx, J.
De Houwer, C. Mensch, I. Caretti, K. A. Tehrani, W. A. Herrebout, S.
Van Doorslaer, B. U. W. Maes,
Ren, L. Wang, Y. Lv, G. Li, S. Gao,
2016, , 346ꢀ357; e) L.
2015, , 2078ꢀ2081; f)
However, bromodimethyl sulfonium ions are transient at reaction
temperature as the formation of bromodimethyl sulfonium ions follows
a kinetically controlled pathway (K.C.P). Thus, very few molecules of
6a can undergo this alternative route to generate intermediate 12 and
15 at reaction temperature.
J. Liu, X. Zhang, H. Yi, C. Liu, R. Liu, H. Zhang, K. Zhuo, A. Lei,
2015, , 1261ꢀ1265; . 2015,
,1277–1281. g) J. De Houwer, K. Abbaspour Tehrani, B. U. W.
Maes,
2012,
, 2745ꢀ2748;
2012,
, 2799–2802
[23] a) S. Rosokha,
Rosokha, M. K. Vinakos,
2017,
, 315ꢀ332; b) S. V.
2014, , 1809ꢀ
2008, , 641ꢀ
[11] a) B. Akhlaghinia, H. Ebrahimabadi, E. K. Goharshadi, S. Samiee, S.
Rezazadeh,
2012,
2010, , 7259ꢀ7261; c) V. J.
1961,
1958,
, 67ꢀ72; b) H. Jiang, H.
1813; c) S. V. Rosokha, J. K. Kochi,
653.
Chen, A. Wang, X. Liu,
Traynelis, A. I. Gallagher, R. F. Martello,
4365ꢀ4368; d) V. J. Traynelis, R. F. Martello,
, 6590ꢀ6593; e) V. Boekelheide, W. J. Linn,
1954, , 1286ꢀ1291.
,
[24] H. F. Askew, P. N. Gates, A. S. Muir,
265ꢀ274.
1991,
,
[25] a) M. Uyanik, K. Ishihara,
French, S. Bissmire, T. Wirth,
2012, , 177ꢀ185; b) A. N.
2004, , 354ꢀ362.
[12] A. Bruckmann, M. A. Pena, C. Bolm,
2008,
, 900ꢀ902.
1959,
[26] a) Y. Su, X. Zhou, C. He, W. Zhang, X. Ling, X. Xiao,
[13] a) N. Kornblum, W. J. Jones, G. J. Anderson,
2016, , 4981ꢀ4987; b) T. Froehr, C. P. Sindlinger, U. Kloeckner, P.
1959,
, 4113ꢀ4114; b) H. Nace, J. Monagle,
Finkbeiner, B. J. Nachtsheim,
2011,
, 3754ꢀ3757; c) M.
, 1792ꢀ1793.
Uyanik, D. Suzuki, T. Yasui, K. Ishihara,
2011,
[14] a) E. Hu, N. Chen, M. P. Bourbeau, P. E. Harrington, K. Biswas, R. K.
Kunz, K. L. Andrews, S. Chmait, X. Zhao, C. Davis, J. Ma, J. Shi, D.
, 5331ꢀ5334;
2011, 1 , 5443–5446 d) Y. Yan, Z.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
Wang,
2011,
2011, , 1082ꢀ1086; f) M. Uyanik, H. Okamoto, T.
Yasui, K. Ishihara, 2010, , 1376ꢀ1379.
, 9513ꢀ9515; e) C. Zhu, Y. Wei,
[27] E. T. Urbansky, B. T. Cooper, D. W. Margerum,
1997,
, 1338ꢀ1344.
[28] R. B. Woodward, F. V. Brutcher,
211.
1958,
1970, 1689ꢀ1691.
, 209ꢀ
[29] Y. Ogata, I. Urasaki,
[30] R. C. Cambie, D. Chambers, P. S. Rutledge, P. D. Woodgate,
1978, 1483ꢀ1485.
[31] T. R. Beebe, B. A. Barnes, K. A. Bender, A. D. Halbert, R. D. Miller, M.
L. Ramsay, M. W. Ridenour,
[32] C. Reid, R. S. Mulliken,
1975, , 1992ꢀ1994.
1954, , 3869ꢀ3874.
1953, , 552ꢀ555.
[33] R. E. Buckles, J. F. Mills,
[34] a) W. Zierkiewicz, M. Michalczyk,
2017,
, 125;
b) S. M. Huber, E. JimenezꢀIzal, J. M. Ugalde, I. Infante,
2012, , 7708ꢀ7710; c) F. Bessac, G. Frenking,
2003, , 7990ꢀ7994.
[35] S. V. Rosokha, C. L. Stern, J. T. Ritzert,
8774ꢀ8788.
2013,
,
This article is protected by copyright. All rights reserved.

10.1002/chem.201801717
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 2:
FULL PAPER
Page No. – Page No.
Metal–free Halogen(I) Catalysts for
Oxidation of (Aryl)(heteroaryl)-
methanes to Ketones or Esters :
Selectivity Control by Halogen Bond
Interaction matters: The role of halogen–bonding interaction in the selective
oxidation of (Aryl)(heteroaryl)ꢀmethanes to Ketones or Esters with Bromine(I) and
Iodine(I) catalysts, respectively, has been thoroughly demonstrated.
This article is protected by copyright. All rights reserved.