Oxidative functionalization of benzylic C-H bonds by DDQ

Article abstract of DOI:10.1039/c2nj40021a

C-H activation of the methyl group of toluene and related ArCH₃ derivatives by 2,3-dichloro-4,5-dicyano-1,4-benzoquinone (DDQ) gives insertion products, ArCH₂O[C₆Cl₂(CN)₂]OH via a rate-determining hydride abstraction by DDQ. The resulting benzylic ether can undergo reactions with phosphines to give benzylic phosphonium salts (Wittig reagents) and with phosphites to give phosphonate esters (Horner-Wadsworth- Emmons reagents). The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012.

Full text of DOI:10.1039/c2nj40021a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 1141–1144
www.rsc.org/njc
LETTER
Oxidative functionalization of benzylic C–H bonds by DDQw
Victor S. Batista,* Robert H. Crabtree,* Steven J. Konezny, Oana R. Luca and
Jeremy M. Praetorius
Received (in Montpellier, France) 10th January 2012, Accepted 17th February 2012
DOI: 10.1039/c2nj40021a
C–H activation of the methyl group of toluene and related
ArCH3 derivatives by 2,3-dichloro-4,5-dicyano-1,4-benzoqui-
none (DDQ) gives insertion products, ArCH2O[C6Cl2-
(CN)2]OH via a rate-determining hydride abstraction by
DDQ. The resulting benzylic ether can undergo reactions with
phosphines to give benzylic phosphonium salts (Wittig reagents)
and with phosphites to give phosphonate esters (Horner–
Wadsworth–Emmons reagents).
yield (eqn (1)). Crystallization of this material yielded light purple
crystals suitable for X-ray crystallography,w which showed that
the product results from the oxidative insertion of DDQ into a
C(sp3)-H bond of toluene (Fig. 1). Activation of secondary
benzylic and vinylic positions by DDQ has previously been
reported and has been a topic of much interest in C–C bond
forming reactions.3,4
High potential quinones, such as 2,3-dichloro-4,5-dicyano-1,4-
benzoquinone (DDQ), are powerful oxidants capable of
performing a wide variety of organic transformations. DDQ is a
highly oxidizing quinone that is indefinitely stable in dry conditions.
It is widely employed for dehydrogenation of organic molecules to
form aromatic and a,b-unsaturated carbonyls; and the oxidation of
activated methylene and hydroxy groups to carbonyl compounds.
Our group has recently investigated DDQ as an electrocatalyst
for the dehydrogenation of secondary amines and N-containing
heterocycles,1 with the goal of using the heterocycles as liquid
phase carriers for storing n(H+ + eꢀ) in a virtual hydrogen
storage scheme that avoids the need for free H2 at any stage.2 We
demonstrated that the active quinone could be electrochemically
regenerated after the dehydrogenation of N-phenylbenzylamine,
allowing near quantitative yield of the corresponding imine with
only catalytic amounts of DDQ. Although the regeneration of
DDQ occurs at high potential (B0.9 V), this provides precedent
for the use of organic oxidants as organocatalysts in electro-
chemical applications.
ð1Þ
The C–C bond lengths of the hydroquinone fragment of 1a
indicate that DDQ has undergone a complete two-electron
reduction from the quinone to the arene form upon insertion
into the C–H bond of toluene.
The oxidation of benzylic C–H bonds by DDQ to give
quinol ethers, such as 1a, has been reported previously, but
only with substrates that were considerably more activated,5–7
e.g. 3-(p-methoxyphenyl)propene.8 Furthermore, reactions
with these activated substrates give significant amounts of
2 : 1 (substrate:DDQ) products, ArCH2O[C6Cl2(CN)2]-
OCH2Ar, whereas toluene only gives 1a.9 For example, reaction
of DDQ with diphenylmethane gave (Ph2CH)O[C6Cl2-
(CN)2]O(CHPh2) in 82% yield, along with a stoichiometric
amount of HO(C6Cl2(CN)2)OH, as the only products
(eqn (2)).10
While screening the stoichiometric reaction of DDQ with other
potential energy carriers, a highly coloured material precipitated
from reactions run in refluxing toluene. A control reaction in the
absence of the heterocycle but still in toluene as solvent gave the
same dark colour and highly coloured precipitate. After sixteen
hours at reflux, the dark precipitate was collected by filtration and
washed with diethyl ether to give 1 as a pale purple solid in 78%
ð2Þ
Yale Chemistry Department, 225 Prospect St., New Haven, CT
06511, USA. E-mail: victor.batista@yale.edu,
robert.crabtree@yale.edu; Fax: +1 203 432 6144;
Tel: +1 203 432 6672, +1 203 432 3915
w Electronic supplementary information (ESI) available: Electronic
structure calculations, experimental procedures and characterization.
CCDC reference number 864075. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c2nj40021a
Fig. 1 ORTEP diagram of 1a (50% probability elipsoids): one of the
two independent molecules in the asymmetric unit shown.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1141–1144 1141

View Article Online
cation and [DDQ-H]ꢀ in a single step; (3) Hydrogen atom
transfer from toluene to DDQ generating the tolyl radical and
[DDQ-H]ꢁ in a single step.
ð3Þ
From competition experiments, the linear free energy
relationship and the deuterium kinetic isotope effect (DKIE)
were both determined in order to help differentiate the three
plausible mechanisms. The linear free energy relationship was
determined by competition experiments: toluene and the other
arylmethane were both dissolved in deuterated tetrachloro-
ethane (d2-TCE) along with 0.1 equiv. of DDQ and heated to
105 1C for 16 h after which time the product ratio was
determined from 1H NMR data. The results of the competition
experiments obey a good linear correlation (R2 = 0.96) with the
substituent s+ values11 and show a r+ value of ꢀ3.60. The
large, negative r+ value indicates a significant positive charge
buildup on the benzylic carbon in the rate-determining
transition state.
Table 1 Reaction of arylmethanes with DDQ
Entry
Substrate
Conditionsa
Time (h)
% Yieldb
1
2
3
4
5
6
7
p-H (1a)
p-Ph (1b)
p-Cl (1c)
p-Cl (1c)
p-tBu (1d)
p-F (1e)
A
B
A
B
A
B
A
16
16
16
22
18
24
18
78
67
65
60
81
40
68
m-F (1f)
a
A: neat 2 mmol DDQ in 2 mL of ArCH3, 110 1C, Ar; B: 1.5 equiv. of
DDQ, 0.4 M ArCH3 in PhCl, 140 1C in a sealed tube. Isolated yields.
b
To differentiate between a two-step mechanism involving
electron transfer followed by transfer of the hydrogen atom
from a single-step mechanism, the deuterium kinetic isotope
effect of the reaction was investigated. A similar competition
experiment was carried out comparing protiated and deuterated
toluene, yielding a primary DKIE of 5.2 (eqn (4)).
The scope of the reaction was explored with a series of
substituted toluenes featuring electron-withdrawing or weakly
electron-donating groups to give the analogous products
(Table 1). For substrates available on large scale, the reaction
could be performed in neat reactant giving the analogous
C–H insertion products after precipitation with diethyl ether.
In the case of substrates conveniently available in only
modest quantities, such as 4-fluorotoluene, the same reactions
occurred at 140 1C in PhCl. All reactions gave the expected
products and were obtained in moderate to good yields in all
cases but took somewhat longer with substrates having more
electron withdrawing substituents.
ð4Þ
This large, primary DKIE is strongly indicative of C–H
bond breaking occurring during the rate-determining transi-
tion state as was previously observed by Floreancig and
coworkers.12 The primary DKIE combined with the large
negative r+ value is consistent with a one-step mechanism
and more specifically rate-determining hydride abstraction
from toluene by DDQ. There are, however, various radical
mechanisms that show better correlation with s+ rather than
s values,13,14 a fact that has been ascribed to the presence of a
highly polar transition state.15 The magnitude of the r+ values
observed in these instances is believed to be a function of the
electronegativity difference between the substrate and the
reactive species; in our case toluene and DDQ, respectively.
Thus, to differentiate between a hydride mechanism and a
radical mechanism with a highly polar transition state, the
reaction coordinate was investigated with DFT calculations of
solution-phase free energies at the BH&H/6-311++G(d,p)
level. The overall reaction is found to be exergonic, releasing
28.7 kcal/mol. Initially, as previously calculated for the dehydro-
genation of N-phenylbenzylamine by DDQ,1 toluene and DDQ
form a charge transfer complex; in this instance leading to a
stabilization of 0.7 kcal/mol. From this ground state, reaction
coordinates corresponding to both hydrogen radical and
hydride abstraction were calculated to determine both transition
state energies and the ground state energies of the required
intermediates.
Mechanistic investigations
An experimental-computational combined study gave insight
into the mechanism of benzylic C–H activation by DDQ. For
the C–H activation, the three most plausible scenarios
(Scheme 1) are: (1) Electron transfer from the arylmethane
to DDQ to generate the radical cation of toluene and the
radical anion of DDQ, followed by hydrogen atom transfer to
generate the tolyl cation and [DDQ-H]ꢀ in a two-step process;
(2) Hydride transfer from toluene to DDQ generating the tolyl
Reaction to form the radical pair proceeds through a
transition state that requires 36.7 kcal/mol. Conversely,
Scheme 1 Possible mechanistic pathways for the C–H abstraction
from toluene by DDQ.
c
1142 New J. Chem., 2012, 36, 1141–1144
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Scheme 2 The previously reported reaction of 1,1-dimethyltetralin
with DDQ to form 1,2-dimethylnaphthalene and the cationic inter-
mediates proposed to account for the observed rearrangement.
Nucleophilic displacement from ArCH2(DDQ)H
Fig. 2 Linear free energy relationship of the relative rates of reaction
determined from competition experiments of toluene and substituted
arylmethane with 0.1 equivalents of DDQ in d2-TCE. The relationship
is plotted as the log of product ratios from the competition reactions
versus reported s+ constants for the arylmethane substituent employed
in the competing substrate.
Given the extremely electron-withdrawing nature of the hydro-
quinone ring in the products we postulated that these groups
would serve as satisfactory leaving groups in substitution
reactions. Unfortunately, this same reasoning also makes the
proton of the –OH of this molecule fairly acidic and thus only
non-basic nucleophiles could be employed. To suit this restriction
we investigated the reaction of 1a with triphenylphosphine. The
product of this reaction, benzyl(triphenyl)phosphonium salt, is a
well-known reagent in the Wittig reaction for the synthesis of
alkenes. Thus, alkenes could be accessed from aldehydes and
arylmethanes in three steps.17
hydride abstraction to give an ion pair requires an input of
only 24.5 kcal/mol and is clearly favoured over the radical
mechanism by a DDGz of 412 kcal/mol (Fig. 2). Formation of
the products from these two sets of intermediate pairs is likely
a fast reaction of low barrier, limited only by reorganization of
the two molecules to align the reactive carbon and oxygen
atoms by a mutual rotation. We therefore conclude from our
experimental and computational results that C–H activation
of toluene and its derivatives most likely occurs through a
hydride mechanism as depicted in Scheme 1 and Fig. 3.
Our mechanistic results are further corroborated by a
previous report detailing the reaction of 1,1-disubstituted
tetralins with high potential quinones to form the corres-
ponding 1,2-substituted naphthalenes. For example, reaction
of 1,1-dimethyltetralin with two equivalents of DDQ in refluxing
toluene was reported to give 1,2-dimethylnapthalene in 84%
yield (Scheme 2).16 The skeletal rearrangement of 1,1-dimethyl-
tetralin in this reaction implicates methyl migration in conjugated
carbocation intermediates. It was also observed that similar
molecules lacking benzylic C–H bonds, such as octahydro-
octamethylanthracene, did not undergo dehydrogenation under
analogous conditions.
Triphenylphosphine and 1a react in refluxing mesitylene to
give the expected benzyl(triphenyl)phosphonium salt 2 in 93%
yield (eqn (5)). The product is isolated as a precipitate from the
reaction mixture, initially as a black solid, but following
extraction from CH2Cl2 and recrystallization from acetone/
diethyl ether is obtained as a light brown solid.
ð5Þ
Using the same idea, Horner–Wadsworth–Emmons reagents,
such as diethyl benzylphosphonate can also be prepared. This can
be done in an analogous manner to preparation from benzyl
bromide. 1a reacts with triethylphosphite to give the phosphonate
in 93% yield after purification (eqn (6)). These reagents undergo
similar olefination reactions with aldehydes, generally with higher
E/Z ratios than their phosphonium counterparts.
ð6Þ
Conclusions
We have isolated benzylic ethers resulting from the oxidative
C–H activation of arylmethanes by DDQ. These products
were isolated for a number of relatively unactivated substrates.
A combination of experimental and computational studies
suggests that this C–H activation occurs through hydride
abstraction from the benzylic methyl group by DDQ. Further-
more, we have demonstrated that the benzylic ether products
can be reacted with phosphines to generate phosphonium
Fig. 3 Free energy diagram for the hydrogen abstraction from the
methyl group of toluene by DDQ, with the radical pathway shown in
red and the hydride pathway shown in blue.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1141–1144 1143

View Article Online
reagents suitable for the Wittig reaction. We have now devel-
oped a synthetic pathway to generate b-substituted styrenes
from arylmethane and aldehyde starting materials.
3 Y. Zhang and C.-J. Li, J. Am. Chem. Soc., 2006, 128, 4242–4243.
4 L. Liu and P. E. Floreancig, Org. Lett., 2009, 11, 3152–3155.
5 R. Foster and I. Horman, J. Chem. Soc. B, 1966, 1049–1053.
6 M. Lemaire, J. Doussot and A. Guy, Chem. Lett., 1988,
1581–1584.
7 D. J. Wallace, A. D. Gibb, I. F. Cottrell, D. J. Kennedy, K. M.
J. Brands and U. H. Dolling, Synthesis, 2001, 1784–1789.
8 E. F. Kiefer and F. E. Lutz, J. Org. Chem., 1972, 37,
1519–1522.
9 M. Mella, M. Freccero and A. Albini, J. Chem. Soc., Chem. Commun.,
1995, 41–42.
10 H. D. Becker, J. Org. Chem., 1969, 34, 1203–1210.
11 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165–195.
12 H. H. Jung and P. E. Floreancig, Tetrahedron, 2009, 65,
10830–10836.
13 R. L. Huang and K. H. Lee, J. Chem. Soc. C, 1966, 935–937.
14 S. S. Kim, S. Y. Choi and C. H. Kang, J. Am. Chem. Soc., 1985,
107, 4234–4237.
Acknowledgements
This material is based upon work supported as part of the Center
for Electrocatalysis, Transport Phenomena, and Materials
(CETM) for Innovative Energy Storage, an Energy Frontier
Research Center funded by the U.S. Department of Energy, Office
of Science, Office of Basic Energy Sciences under Award Number
DE-SC00001055. The authors would also like to thank Nathan
D. Schley for X-ray crystallography and useful discussions.
Thanks to Gary Yeager (General Electric), David R. Edwards
(UNC Chapel Hill) and Kenneth B. Wiberg for help.
15 A. A. Zavitsas and J. A. Pinto, J. Am. Chem. Soc., 1972, 94,
7390–7396.
16 E. A. Braude, L. M. Jackman, R. P. Linstead and G. Lowe,
J. Chem. Soc., 1960, 3123–3132.
Notes and references
17 N. J. Lawrence and F. Muhammad, Tetrahedron, 1998, 54,
15361–15370. Using this previously reported Wittig-like process,
(E)-stilbene was synthesized in 41% isolated yield.
1 O. R. Luca, T. Wang, S. J. Konezny, V. S. Batista and
R. H. Crabtree, New J. Chem., 2011, 35, 998–999.
2 R. H. Crabtree, Energy Environ. Sci., 2008, 1, 134–138.
c
1144 New J. Chem., 2012, 36, 1141–1144
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012