Decatungstate photocatalyzed benzylation of alkenes with alkylaromatics

Article abstract of DOI:10.1002/adsc.201300598

The direct benzylation of electrophilic al-kenes with alkylbenzenes has been achieved by decatungstate photocatalysis in a high ionic strength medium (5:1 acetonitrile/water mixed solvents, 0.5 M lithium perchlorate). The reaction has been extended to ring-substituted (Cl, F, CN) toluenes and further alkylbenzenes. Intramolecular selectivity (p-cymene, 4-methylanisole) and deuteration effects support the atom transfer character of the process that, however, requires a sufficient stabilization of the intermediate polar exciplex to occur. The reactivity of the deca-tungstate anion is compared with that of aromatic ketones, the benchmark of photochemical hydrogen/ electron transfer processes.

Full text of DOI:10.1002/adsc.201300598

FULL PAPERS
DOI: 10.1002/adsc.201300598
Decatungstate Photocatalyzed Benzylation of Alkenes with
Alkylaromatics
Hisham Qrareya,a Davide Ravelli,a Maurizio Fagnoni,a,* and Angelo Albinia,*
a
PhotoGreen Lab, Department of Chemistry, University of Pavia, Viale Taramelli 12, 27100
Fax : (+39)-038-298-7323; phone: (+39)-038-298-7316; e-mail: fagnoni@unipv.it or angelo.albini@unipv.it
Received: July 4, 2013; Revised: July 30, 2013; Published online: October 9, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201300598.
Abstract: The direct benzylation of electrophilic al- requires a sufficient stabilization of the intermediate
kenes with alkylbenzenes has been achieved by deca- polar exciplex to occur. The reactivity of the deca-
tungstate photocatalysis in a high ionic strength tungstate anion is compared with that of aromatic
medium (5:1 acetonitrile/water mixed solvents, 0.5M ketones, the benchmark of photochemical hydrogen/
lithium perchlorate). The reaction has been extended electron transfer processes.
to ring-substituted (Cl, F, CN) toluenes and further
alkylbenzenes. Intramolecular selectivity (p-cymene,
4-methylanisole) and deuteration effects support the Keywords: benzylation; C H bond activation; pho-
atom transfer character of the process that, however, tocatalysis; polyoxometalates; radicals
À
Introduction
benzenes in a number of cases. Some success has
been obtained, limited however by the above dichoto-
À
The direct activation of C H bonds in hydrocarbons my that shows off in the formation of bibenzyls on
remains a challenge for synthetic chemists, in particu- the one hand and in the reaction with the co-formed
lar under mild conditions that allow control of the radical or radical anion from A on the other one.
subsequent reactions. The most important approaches These processes compete with the desired trapping,
involve either metal catalysis[1] or homolytic hydrogen which usually involves an electrophilic alkene
abstraction.[2] In particular, radical benzylation start- (Scheme 1b).
ing directly from alkylbenzenes is of large synthetic
Polyoxoanions, such as the acetonitrile soluble tet-
interest,[3] but requires that a dichotomy is confronted, rabutylammonium decatungstate (TBADT), have
À
on the one hand strong activators have to be used be- been increasingly used for the activation of C H
cause the benzylic Ar-C H bond is still quite strong, bonds.[8] We recently used the same photocatalyst for
though less so than an Alk-C H bond (compare, for the benzylation of electron-poor olefins starting from
À
À
example, the bond dissociation energies – BDE – of
À
the C H bond in methane/cyclohexane and in tolu-
ene, corresponding to 105/99.5 and 88.5 kcalmolÀ1, re-
spectively).[4] On the other hand, the benzyl radical is
relatively persisting, a fact that often leads to dimeri-
zation, rather than to the desired trapping.[5,6]
Photochemistry offers an appealing path because
various classes of molecules (A in Scheme 1) are
known to interact with alkylbenzenes either as elec-
tron acceptors or as hydrogen accepting species in the
excited state.[5,7] 1,2,4,5-Tetracyanobenzene (TCB)
and
2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ) are examples of the oxidants, as are (aromat-
ic) ketones of hydrogen abstracting species (quinones
may follow either path, however, Scheme 1a). Photo-
Scheme 1. a) Photochemical generation of benzyl radicals
chemical activation has indeed been applied to alkyl- from alkylaromatics; b) fate of the benzyl radicals formed.
Adv. Synth. Catal. 2013, 355, 2891 – 2899
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2891

Hisham Qrareya et al.
FULL PAPERS
Table 1. Optimization of the reaction conditions.
Scheme 2. Deacatungstate (TBADT) photocatalyzed benzy-
lation of electron-poor olefins investigated in previous[9] and
current work.
Entry 1a
2a
Conditions Conversion[a] 3 Yield[a]
(M) (M)
[%]
[%]
1
2
3
4
5
6
7
0.1
0.2
0.1
0.5
0.5
0.5
0.2
0.5
0.5
0.5
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
A
A
A
A
B
C
C
C
C
C
C
C
48
63
30
80
35[b]
38[b]
40[b]
38[b]
41[b]
75[c]
61[c]
4[b]
benzylsilanes[9] and we wondered whether the same
kind of reaction could be carried out by starting from
easily available and cheap alkylaromatics in the place
of silanes. This would result in a 100% atom-economi-
cal process (Scheme 2). In previous work, the deca-
tungstate cluster has been adopted in the photochemi-
cal functionalization of [60]fullerene by toluene and
a few derivatives[10] and in the oxidation of alkylben-
zenes.[11] The decatungstate-mediated photocatalytic
activation of other benzyl derivatives, such as arylal-
kanols, has been likewise studied in detail.[12] Further-
more, the same polyoxometalate proved to be an effi-
cient catalyst for the thermal oxidation of benzylic al-
cohols in the presence of hydrogen peroxide.[13]
84
100
100
20
8
65
8[d]
9[e]
10[f]
[b]
–
31[b]
84[b]
91[c]
11[g] 0.5
12[h] 0.5
100
100
[a]
Referred to the consumption of the limiting reagent
(that is 2a, except for entry 3).
GC yield using internal standard (n-undecane or n-do-
[b]
decane).
[c]
[d]
[e]
[f]
Isolated yield after column chromatography.
No TBADT.
Thermal reaction.
24 h irradiation at 366 nm.
10 h irradiation with SolarBox (see the Supporting Infor-
mation for details).
Results and Discussion
[g]
Product Studies
[h]
Direct sunlight irradiation (24 h over 3 days).
Explorative experiments demonstrated that the hoped
for benzylation did occur also when using alkylben-
zenes, but required a thorough work for tuning the
conditions and obtaining satisfactory results.
The previous study on benzyltrimethylsilanes[9] had
shown an improvement in the alkylation when passing
Toluene (1a) and fumaronitrile (2a) were chosen as from acetonitrile (conditions A) to acetonitrile/water
the reagents for the exploratory work. Thus, irradia- 5:1 (conditions B) and a further one, though smaller,
tion of an equimolar (0.1M) MeCN solution of 1a in passing to a 0.5M LiClO4 acetonitrile/water 5:1 so-
and 2a for 24 h (15 W low pressure phosphor-coated lution (conditions C). We tested whether a similar im-
lamps, lIRR =310 nm) in the presence of 2 mol% provement occurred also with toluene and indeed
TBADT afforded 2-benzylsuccinonitrile (3) as the found this to be the case. In particular, a 75% isolated
only detected product in 35% yield at 48% olefin yield of benzylated 3 was obtained at complete olefin
conversion (entry 1, Table 1). Using an excess conversion under conditions C (entry 6).
(2 equiv.) of the radical precursor resulted in an in-
Moreover, exploring whether under such conditions
creased conversion, while the yield remained almost the excess of alkylaromatics could be decreased re-
unchanged (entry 2). On the contrary, the use of an vealed that a lower excess of toluene (2 equiv.) still
excess of 2a (2 equiv.) gave unsatisfactory conversion led to complete conversion, with some decrease of
(entry 3). This prompted us to further increase the the yield (from 75 to 61%; entry 7), while 24 h irradia-
excess of 1a; indeed, with 5 equiv. an 80% conversion tion was required for complete consumption (see time
was obtained and the yield of 3 was maintained at the monitoring in Figures S1 and S2 in the Supporting In-
same level (entry 4). Further increasing the amount of formation). Blank experiments demonstrated that
toluene, up to an acetonitrile/toluene 1/2 mixture as both light and TBADT were required for the reaction
the solvent, marginally increased alkylation, but to proceed (entries 8 and 9).
caused the formation of some percent bibenzyl (as
from GC analysis; data not reported).
Finally, we explored whether different light sources
could be used. Low pressure, phosphor-coated lamps
2892
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2891 – 2899

Decatungstate Photocatalyzed Benzylation of Alkenes with Alkylaromatics
Table 2. TBADT photocatalyzed addition of (substituted)
toluene(s) to electron-poor olefins.
(lIRR =366 nm) were found to be less effective than
310 nm lamps, while the use of a SolarBox (xenon
lamp with filters simulating solar light irradiation con-
ditions) allowed us to increase the reaction yield up
to 84% (entries 10 and 11). A further increase (isolat-
ed yield 91%) was observed upon exposure to direct
sunlight for 24 h (entry 12; see the Supporting Infor-
mation, Figure S3 for details).
The scope of the reaction was then explored by
using different substrates under the most favorable re-
action conditions determined above (that is, those of
entry 6 in Table 1, with further elaboration when ap-
propriate). Thus, benzylation by toluene was extended
to further carboxyl derivatives, viz. maleic anhydride
(2b) and to maleic imide (2c) as well as to the N-
methyl and N-phenyl derivatives (2d and 2e). Yields
around 50% were consistently obtained (entries 1 and
2, Table 2).
The reaction was likewise satisfactory with esters,
viz. dimethyl fumarate (2f, 75%) and maleate (2g,
50% upon prolonged irradiation; entry 3). On the
other hand, the yield was limited to 29% with unsatu-
rated ketone 2h (entry 4). A few substituted toluenes
were tested in the benzylation of fumaronitrile, with
results ranging from good in the case of 4-fluoro- and
4-chlorotoluene (1b and 1c, yield around 80%) to sat-
isfactory in the case of 4-cyanotoluene (1d, 50% yield
upon SolarBox irradiation; entries 5–7).
Increasing the number of alkyl groups appeared to
favor addition to olefins, resulting in higher yields
with polyalkylbenzenes (see Table 3). Thus, alkylation
of fumaronitrile was achieved in ca. 80% isolated
yield both with isomeric xylenes (1e–g) and with me-
sitylene (1h), albeit less satisfactorily with poorly solu-
ble durene (1i, at any rate 60% yield).
Introducing alkyl groups different from methyl, as
in the case of ethylbenzene (1j), indane (1k) and
cumene (1l), led to the best results in the series (80–
90% yield).
In order to gain mechanistically useful evidence,
some competition experiments were carried out.
Thus, TBADT photocatalysis was applied to mole-
cules possessing two potentially reactive sites. It was
found that 4-methylanisole (21) reacted preferentially
at the benzylic site in the alkylation of 2a. Thus,
a product ratio 22/23=2.3 was observed in neat aceto-
nitrile and experiments in mixed aqueous medium led
to doubling the yield of 22 and a moderate increase
[a]
Isolated yields after column chromatography; conver-
sions and yields are based on olefin consumption. A
100% olefin consumption was obtained, except where
otherwise noted.
SolarBox irradiation (10–20 h: see the Supporting Infor-
mation).
[b]
[c]
40 h irradiation; traces of bibenzyl were also detected by
GC analysis.
of the yield of 23. The effect was confirmed in the 41 and 68%, respectively; ratio 25/26=1.6, 1.6 and
presence of lithium perchlorate, so that the ratio 22/ 1.8, Scheme 4).
23 increased to 3.4 and 3.9 (Scheme 3). p-Cymene
Finally, the benzylation of 2a was tested with a deu-
(24) showed a preference for reaction at the tertiary terated derivative. Irradiation in the presence of an
position rather than at the primary, a trend that in equimolecular mixture of toluene (1a) and toluene-d8
this case was maintained, while the yield of the two (1a-d8) evidenced a preference for the activation of
products increased in a parallel way in going from the former compound, as from GC-MS analysis (ratio
conditions A, to B and to C (total yield 25+26=26, 3/3-d7 =2.38 under conditions C, Scheme 5; see Fig-
ure S4 in the Supporting Information for details).
Adv. Synth. Catal. 2013, 355, 2891 – 2899
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2893

Hisham Qrareya et al.
FULL PAPERS
and monoreduced TBADT ([W10O32]5À or protonated
[HW10O32]4À) absorbed.
Table 3. TBADT photocatalyzed addition of alkylaromatics
to fumaronitrile.
In the presence of toluene, the transients observed
were unchanged with respect to neat acetonitrile, that
is the absorption of the tungstate excited state de-
cayed (t ca. 60 ns) and was substituted by that of the
reduced catalyst, which was persistent in the time-
window explored up to a few ms (see Figure 1a). This
species is known to arise by H abstraction from the
solvent, a reversible reaction on a longer time scale.
Likewise, adding toluene caused only a minimal
change on the signal in acetonitrile/water (see Fig-
ure 1b). On the other hand, the short-lived transient
was clearly quenched when the experiment was re-
peated in a solution containing lithium perchlorate
(see Figure 1c, initial part) and a bimolecular rate
constant of 5.2ꢁ108 LmolÀ1 sÀ1 was determined (see
Figure 2).
Under these conditions, the persistent species
formed was more intense, more than twice as much as
in the absence of toluene (see Figure 1c). Moreover,
the effect was similar when NaClO4 was substituted
for LiClO4 (data not shown). An analogous experi-
ment with toluene-d8 caused a less effective quench-
ing (2.1ꢁ108 LmolÀ1 sÀ1, Figure 2). For the sake of
comparison we carried out similar flash experiments
with benzyltrimethylsilane. In this case, quenching of
the first-formed transient was apparent already in
neat acetonitrile (see Figure 1d, initial part, bimolecu-
lar rate constant of 7.1ꢁ108 LmolÀ1 sÀ1) and doubled
(1.6ꢁ109 LmolÀ1 sÀ1) in water-containing medium, but
the persistent signal did not increase, indeed de-
creased particularly in the latter case (see Figure 1d,
e). In the presence of lithium perchlorate quenching
remained at the same value observed under condi-
tions B and the persistent signal was unaffected (see
Figure 1f).
[a]
Isolated yields after column chromatography; conver-
sions and yields are based on olefin consumption. A
100% olefin consumption was obtained, except where
otherwise noted.
20 h SolarBox irradiation.
Direct sunlight irradiation (24 h over 3 days).
0.2M alkylaromatic was used.
0.15M 1i was used due to solubility limitations.
Traces of 2,2’,3,3’-tetrahydro-1H,1’H-1,1’-biindene (19’, as
a mixture of diastereoisomers) were also obtained (see
Benzylation of Electron-Poor Olefins, Mechanism
[b]
[c]
Benzylation of olefins starting directly from alkylben-
zenes has only a few precedents in the literature. As
an example, Naito and co-workers reported the radi-
cal benzylation of electron-poor olefins by means of
the Et3B/O2 system at 1808C.[15] Quite common is the
use of peroxides[16,17] for the generation of the radical
intermediate, generally in the presence of metal cata-
lysts.[17] Other approaches involve either the use of
high temperatures[18] or catalytic systems, such as
[d]
[e]
[f]
the Supporting Information for details).
Flash Photolysis Experiments
Mechanistic evidence was sought also in laser flash those based on transition metals[19] or enzymes.[20]
photolysis experiments. Flashing an acetonitrile solu- Photochemical contributions are the benzylation of
tion of TBADT generated, as previously reported, tetracyanoethylene upon irradiation in the presence
a transient absorption at 780 nm.[12,14] Extensive work of alkylaromatics,[21] the photocatalyzed (e.g., by 1,4-
has demonstrated that at this wavelength both the re- dicyanonaphthalene) alkylation of olefins[5,7] and the
active excited state (indicated as wO, populated in the decatungstate-mediated benzylation of [60]fullerene
picosecond domain from the spectroscopic state)[12,14] with alkylaromatics.[10]
2894
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2891 – 2899

Decatungstate Photocatalyzed Benzylation of Alkenes with Alkylaromatics
Scheme 3. Photocatalyzed reaction between 4-methylanisole (21) and fumaronitrile (2a).
Scheme 4. Photocatalyzed reaction between p-cymene (24) and fumaronitrile (2a).
refers only to conditions C, acetonitrile-water mixture
in the presence of lithium perchlorate. The origin of
this requirement must thus be understood for assess-
ing the significance of the method. A key issue in-
volves the generation of radicals. As discussed in pre-
vious literature, activation by excited TBADT in dif-
ferent cases is best envisaged either as an electron
transfer process or as hydrogen transfer (see
Scheme 1a).[12,14d–f] The case of alkylbenzenes offers
a suitable opportunity for distinguishing this alterna-
tive and to class decatungstate in comparison with
other compounds known to participate into the two
processes, in particular aromatic ketones, the bench-
mark of organic photochemistry.
Scheme 5. Competitive photocatalyzed benzylation of 2a
with toluene and toluene-d8.
Some key data about the photochemical reaction of
The data above demonstrate that the benzylation of the decatungstate anion with alkylbenzenes are com-
electrophilic alkenes directly from alkylbenzenes is pared in Table 4 with the corresponding data for rep-
possible under TBADT photocatalyzed conditions, resentative ketones. For the last compounds, Wagner
indeed it is a rather general and clean reaction, at et al.[22] concluded that where deuteration at the ben-
least when the trap is a good electron acceptor (an zylic position led to a negligible isotopic effect, a rate-
alkene bearing two electron withdrawing substituents; determining, irreversible complexation took place,
actually, the only case where a yield lower than ~50% whereas the occurrence of an isotopic effect indicated
has been obtained is the reaction with enone 2h with either a reversible complexation prior to hydrogen
a single activating group). The reaction is tolerant to- transfer, or an independent atom hydrogen transfer
wards ring substitution (although a strongly electron- path.[22] In a later study, Scaiano et al.[23] found it to
withdrawing substituent such as the cyano group de- be more appropriate to depict the main reaction path-
creases somewhat the alkylation yield) and steric hin- way as the irreversible formation of charge transfer
dering or substitution degree at the reacting benzyl (CT) encounter complexes followed by hydrogen
carbon, as demonstrated by the results in Table 3. Bi- atom transfer in competition with non-radiative decay
benzyls are formed only in a few cases and at any via reverse charge transfer.[23] Thus, different shades
rate in tiny amounts. Such a general scope, however, of a single mechanism accounted for the results, and
Adv. Synth. Catal. 2013, 355, 2891 – 2899
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2895

Hisham Qrareya et al.
FULL PAPERS
Figure 1. Transient absorption at 780 nm observed upon flashing (355 nm) a 2ꢁ10À4 M TBADT solution: a) in MeCN (bold
black line) and in the presence of toluene (4.5ꢁ10À3; 9.0ꢁ10À3; 1.5ꢁ10À2; 2.2ꢁ10À2; 3.0ꢁ10À2 M); b) same in 5:1 MeCN/
H2O; c) same in 5:1 MeCN/H2O, 0.5M LiClO4; d) in MeCN (bold black line) and in the presence of benzyltrimethylsilane
(9.3ꢁ10À3; 1.9ꢁ10À2; 3.8ꢁ10À2; 6.6ꢁ10À2; 9.4ꢁ10À2; 1.4ꢁ10À1; 9.4ꢁ10À2; 1.9ꢁ10À1 M); e) same in 5:1 MeCN/H2O; f) same in
5:1 MeCN/H2O, 0.5M LiClO4.
the intervention of direct hydrogen abstraction or former exhibits a low rate of reaction with toluene,
complete electron transfer as competitive processes 4.6ꢁ105 LmolÀ1 sÀ1,[22] close to that with cyclohexane
was not required.
(7.5ꢁ105 LmolÀ1 sÀ1),[22] a well defined deuterium
Comparing the values in Table 4, the difference be- effect and a large preference for attack at the tertiary
tween the np* triplet state of benzophenone and the position with p-cymene,[22] while none of these charac-
pp* trifluoroacetophenone triplet is apparent.[22] The teristics is observed with the latter ketone, where the
2896
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2891 – 2899

Decatungstate Photocatalyzed Benzylation of Alkenes with Alkylaromatics
ly supports the role of electron transfer. On the other
hand, the characteristics of the chemical reaction are
those of hydrogen transfer, viz. deuterium effect on
the quenching, selectivity for the tertiary position
with p-cymene and competitive reaction at the ben-
zylic and ethereal positions in 4-methylanisole (only
the first reaction takes place when the electron trans-
fer mechanism is involved).[10]
Although the results should be compared with care,
pending an unambiguous determination of the
[14e]
ERED
and of the electronic structure of the
TBADT reactive species, as well as of the geometric
parameters of the decatungstate anion in the reacting
configuration, which certainly differ from those of
both np* and pp* ketone states, an economical ra-
tionalization is offered through Scheme 6.
Figure 2. Quenching of the 780 nm transient formed by
flashing (355 nm) a 2ꢁ10À4 M TBADT solution in 5:1
!
MeCN/H2O, 0.5M LiClO4 by toluene ( ) and toluene-d8
~
( ).
Table 4. Reaction parameters for the reaction of aromatic
ketones and TBADT with toluene (and p-cymene) in
MeCN.
[a]
Acceptor (excited state kQ ꢁ107 LmolÀ1 sÀ1 kH/kD
P/T[b]
ERED, V vs. SCE)
Ph2CO (1.17)
0.046 (0.27)[c]
7.3 (19)[c]
2.3
1
1.4
2.38
0.35
4.7
–
Scheme 6. Proposed mechanism for the photocatalyzed ben-
zylation of olefins.
PhCOCF3 (1.65)
1-azaxanthone (1.67)
TBADT (ꢀ2.16)
16
52[d] (71, 160[d])[c]
0.55
[a]
Comparing abstraction rate from C6H5CH3 vs. C6D5CD3.
Primary (P) vs. tertiary (T) hydrogen abstraction in p-
cymene.
kQ with benzyltrimethylsilane.
In 5:1 MeCN/H2O, 0.5M LiClO4.
[b]
TBADT (A in the scheme) forms a polar exciplex
with toluene at a rate faster than ketones in view of
the more favorable DGET. However, in a moderately
polar solvent such as MeCN, proton transfer within
[c]
[d]
À
the exciplex (from the non-polarized benzylic C H
position in the toluene radical cation[27,28] to a reasona-
reaction is two orders of magnitude faster.[22] This sug- bly non-basic species, such as [W10O32]4À, where the
gests that exciplexes with differing degrees of partial partial additional charge is largely delocalized) is too
electron transfer are involved (only partial, because slow to matter and no irreversible reaction takes
DGET from toluene is positive, compare the excited place (path a), differently from the case of ketones. In
state ERED of the ketones with EOX for toluene, 2.2 V LiClO4 containing acetonitrile/water mixture, the in-
vs. SCE).[24] The structure of the reacting triplet has crease of ionic strength and water complexation stabi-
also a role, as shown in the case of 1-azaxanthone lize the exciplex and prolong its lifetime (path b),
with the same ERED as trifluoroacetophenone but of thus offering a chance to reaction. Still, the kinetic
+
np* nature, that exhibits a similar kQ, but shows no acidity of PhCH3C is too low[28] to allow proton trans-
isotopic effect.[23] The reactive state of decatungstate fer to the medium and only hydrogen transfer within
is more easily reduced (ERED ꢀ2.16)[14e] than that of the exciplex takes place (path c). These characteristics
ketones and DGET from toluene is zero or slightly explain that the reaction requires a favorable CT in-
negative. As a matter of fact, quenching under condi- teraction (conditions C), and thus the stabilization of
tions C occurs at a rate (5.2ꢁ108 LmolÀ1 sÀ1) that is the exciplex. Nevetheless, the process maintains the
that expected for an electron transfer process with peculiar selectivity of hydrogen atom transfer (HAT),
[25]
comparable DGET
and indeed similar to that ob- resulting from the preferred orientation in the exci-
served for the quenching of TBADT by aroma- plex. With a good donor such as 4-methylanisole,
tics.[14e,26] The dramatic medium dependence (an effect passing to conditions C involves an increase in the CT
of ionic strength rather than a specific complexation, character, and thus an increased reaction at the ben-
since LiClO4 and NaClO4 have the same effect) clear- zylic rather than the ethereal position. On the contra-
Adv. Synth. Catal. 2013, 355, 2891 – 2899
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2897

Hisham Qrareya et al.
FULL PAPERS
ry, with poor donors such as p-cymene, only an over- a more extensive application than organic photocata-
all increase of the reactivity is observed. lysts, exploiting the electron acceptor characteristics,
Previous studies, in particular with benzyl alcohols, while profiting of the selectivity of the hydrogen atom
have clearly supported the HAT nature of the proc- transfer processes. These topics probably deserve
ess[8,10,12,14d–f] and a high ionic strength was not re- a more extended application in synthetic chemistry.
quired. With toluenes, that are both poor hydrogen
and electron donors, the reaction requires medium as-
sistance. Likewise, more active than toluene is benzyl- Experimental Section
trimethylsilane, the preparative benzylation by which
has been previously reported.[9] The laser flash photol- Typical Procedure for the Photocatalyzed
ysis experiments reported here support the mecha- Benzylation of Olefins
nism proposed. The silane is a better donor (EOX
A 0.5M LiClO4 solution in MeCN/H2O 5:1 (30 mL) of the
alkylaromatic (1, 0.5M) and the olefin (2, 0.1M) in the pres-
1.65 V vs SCE)[9] and quenches the decatungstate
ence of 200 mg of TBADT (2ꢁ10À3 M) was poured into two
quartz tubes and purged for 5 min with nitrogen, the tubes
were then septum capped and irradiated for 24 h with ten
15 W phosphor-coated lamps (emission centered at 310 nm).
The solvent was removed under vacuum from the photo-ir-
radiated solution and the product isolated by purification of
the residue by column chromatography (cyclohexane/ethyl
acetate as eluents).
anion (path d in Scheme 6) already in MeCN and,
more strongly, in MeCN/H2O with a small further in-
crease with the lithium salt (see Figures 1d–f). How-
ever, the amount of reduced decatungstate formed is
not larger than that observed in neat MeCN, indeed is
somewhat lower. In this case, DGET is markedly nega-
tive, so that the intermediate is better referred to as
a radical ion pair, and a good electrofugal group such
as the trimethylsilyl cation is present. Fast fragmenta-
tion in competition with back electron transfer leads
to irreversible formation of benzyl radicals (paths
e and g) in competition with back electron transfer
(path f). Apparently, in this case reduced decatung-
state is complexed or at any rate is not revealed at
780 nm in the time-window explored.[14e]
The success of the desired alkylation further de-
mands that a good trap is available for relatively slow
reacting benzyl radicals (e.g., these add to electrophil-
ic alkenes at a rate lower by a factor >102 with re-
spect to aliphatic radicals).[29] In fact, benzylation was
successful with easily reduced olefins such as the
doubly substituted derivatives reported above, but
was ill extended to less reducible traps such as similar
monosubstituted derivatives (see, for example, entry 4
in Table 2). It has thus to be assumed that reduced
decatungstate transfers an electron to the alkene.[9]
This has two favorable effects, first by lowering the
chance of back electron transfer, second by generat-
ing the alkene radical anion, a better trap for the nu-
cleophilic benzyl radical.[9,30] This is in accord with the
incorporation of a H (from the protic medium), not
D, atom in the a-position in the experiment with tolu-
ene-d8.[31]
Acknowledgements
H.Q. acknowledges financial support from the Italian Minis-
try of Foreign Affairs through the E-PLUS project (Enhance-
ment of the Palestinian University System). We thank Dr. S.
Montanaro and Dr. S. Protti (University of Pavia) for the
precious help.
References
[1] a) J. Wencel-Delord, F. Glorius, Nat. Chem. 2013, 5,
369–375; b) D. Y.-K. Chen, S. W. Youn, Chem. Eur. J.
2012, 18, 9452–9474; c) S. Rousseaux, B. Liegault, K.
Fagnou, in: Modern tools for the synthesis of complex
bioactive molecules, (Eds.: J. Cossy, S. Arseniyadis),
John Wiley & Sons, Inc., Hoboken, 2012, pp 1–32;
d) B. G. Hashiguchi, S. M. Bischof, M. M. Konnick,
R. A. Periana, Acc. Chem. Res. 2012, 45, 885–898;
e) R. H. Crabtree, J. Chem. Soc. Dalton Trans. 2001,
2437–2450; f) R. H. Crabtree, J. Organomet. Chem.
2004, 689, 4083–4091; g) A. S. Larsen, K. Wang, M. A.
Lockwood, G. L. Rice, T.-J. Won, S. Lovell, M. Sadilek,
F. Turecek, J. M. Mayer, J. Am. Chem. Soc. 2002, 124,
10112–10123.
[2] P. R. Schreiner, A. A. Fokin, Chem. Rec. 2004, 3, 247–
257.
Conclusions
[3] For selected recent examples dealing with the direct ac-
Despite the above limitations, the process presented
here is promising for application. In fact, the excellent
electron acceptor property of decatungstate, stronger
than most aromatic ketones, allows one to make H
abstraction from moderate (electron) donors more ef-
ficient than would be from the mere consideration of
the atom transfer process. This photocatalyst thus has
À
tivation of benzylic C H bonds, see: a) G. Pandey, S.
Pal, R. Laha, Angew. Chem. 2013, 125, 5250–5253;
Angew. Chem. Int. Ed. 2013, 52, 5146–5149; b) L.
Meng, J. Su, Z. Zha, L. Zhang, Z. Zhang, Z. Wang,
Chem. Eur. J. 2013, 19, 5542–5545; c) J. Park, Y.-M.
Lee, W. Nam, S. Fukuzumi, J. Am. Chem. Soc. 2013,
135, 5052–5061.
2898
asc.wiley-vch.de
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2891 – 2899

Decatungstate Photocatalyzed Benzylation of Alkenes with Alkylaromatics
[4] Y.-R. Luo, Handbook of bond dissociation energies in
organic compounds, CRC Press LLC, Boca Raton,
2003.
[16] R. L. Huang, H. H. Lee, L.-Y. Wong, J. Chem. Soc.
1965, 6730–6737.
[17] S. Araneo, F. Fontana, F. Minisci, F. Recupero, A.
Serri, J. Chem. Soc. Chem. Commun. 1995, 1399–1400.
[18] J. Metzger, J. Hartmanns, P. Koell, Tetrahedron Lett.
1981, 22, 1891–1894.
[19] H.-Y. Li, L.-J. Xing, T. Xu, P. Wang, R.-H. Liu, B.
Wang, Tetrahedron Lett. 2013, 54, 858–860.
[20] a) C. Qiao, E. N. G. Marsh, J. Am. Chem. Soc. 2005,
127, 8608–8609; b) L. Li, E. N. G. Marsh, J. Am. Chem.
Soc. 2006, 128, 16056–16057.
[5] M. Fagnoni, D. Dondi, D. Ravelli, A. Albini, Chem.
Rev. 2007, 107, 2725–2756.
[6] a) E. Baciocchi, C. Rol, G. C. Rosato, G. V. Sebastiani,
J. Chem. Soc. Chem. Commun. 1992, 59–60; b) L. Cer-
menati, M. Fagnoni, A. Albini, Can. J. Chem. 2003, 81,
560–566.
[7] M. Mella, M. Fagnoni, A. Albini, Eur. J. Org. Chem.
1999, 2137–2142.
[8] a) M. D. Tzirakis, I. N. Lykakis, M. Orfanopoulos,
Chem. Soc. Rev. 2009, 38, 2609–2621; b) I. N. Lykakis,
E. Evgenidou, M. Orfanopoulos, Curr. Org. Chem.
2012, 16, 2400–2414.
[21] M. Ohashi, S. Suwa, Y. Osawa, K. Tsujimoto, J. Chem.
Soc. Perkin Trans. 1 1979, 2219–2223.
[22] P. J. Wagner, R. J. Truman, A. E. Puchalski, R. Wake, J.
Am. Chem. Soc. 1986, 108, 7727–7738.
[23] C. Coenjarts, J. C. Scaiano, J. Am. Chem. Soc. 2000,
122, 3635–3641.
[9] S. Montanaro, D. Ravelli, D. Merli, M. Fagnoni, A.
Albini, Org. Lett. 2012, 14, 4218–4221.
[10] M. D. Tzirakis, M. Orfanopoulos, Org. Lett. 2008, 10,
873–876.
[11] a) D. Attanasio, L. Suber, K. Thorslund, Inorg. Chem.
1991, 30, 590–592; b) A. Molinari, G. Magnacca, G. Pa-
pazzoni, A. Maldotti, Appl. Catal. B: Environ. 2013,
138–139, 446–452.
[12] a) I. N. Lykakis, C. Tanielian, M. Orfanopoulos, Org.
Lett. 2003, 5, 2875–2878; b) C. Tanielian, I. N. Lykakis,
R. Seghrouchni, F. Cougnon, M. Orfanopoulos, J. Mol.
Catal. A: Chem. 2007, 262, 170–175; c) I. N. Lykakis, C.
Tanielian, R. Seghrouchni, M. Orfanopoulos, J. Mol.
Catal. A: Chem. 2007, 262, 176–184; d) M. D. Tzirakis,
I. N. Lykakis, G. D. Panagiotou, K. Bourikas, A. Ly-
courghiotis, C. Kordulis, M. Orfanopoulos, J. Catal.
2007, 252, 178–189.
[24] a) J. O. Howell, J. M. Goncalves, C. Amatore, L. Kla-
sinc, R. M. Wightman, J. K. Kochi, J. Am. Chem. Soc.
1984, 106, 3968–3976; b) S. Fukuzumi, K. Ohkubo, T.
Suenobu, K. Kato, M. Fujitsuka, O. Ito, J. Am. Chem.
Soc. 2001, 123, 8459–8467.
[25] a) D. Rehm, A. Weller, Isr. J. Chem. 1970, 8, 259–271;
b) S. Farid, J. P. Dinnocenzo, P. B. Merkel, R. H. Young,
D. Shukla, G. Guirado, J. Am. Chem. Soc. 2011, 133,
11580–11587.
[26] For the EOX of the other alkylbenzenes studied, see
Table S1 in the Supporting Information.
[27] E. Baciocchi, M. Bietti, O. Lanzalunga, J. Phys. Org.
Chem. 2006, 19, 467–478.
[28] a) N. D’Alessandro, A. Albini, P. S. Mariano, J. Org.
Chem. 1993, 58, 937–942; b) M. Jonsson, in: Recent Re-
search Developments in Physical Chemistry, (Ed.: S. G.
Pandalai), Transworld Research Network, Trivandrum,
1997; c) D. Norberg, P.-E. Larsson, N. Salhi-Benachen-
hou, J. Phys. Chem. A 2008, 112, 4694–4702; d) X.
Zhang, F. G. Bordwell, J. Org. Chem. 1992, 57, 4163–
4168.
[13] B. S. Chhikara, S. Tehlan, A. Kumar, Synlett 2005, 63–
66.
[14] a) D. C. Duncan, T. L. Netzel, C. L. Hill, Inorg. Chem.
1995, 34, 4640–4646; b) D. C. Duncan, M. A. Fox, J.
Phys. Chem. A 1998, 102, 4559–4567; c) C. Tanielian,
Coord. Chem. Rev. 1998, 178–180, 1165–1181; d) T.
Kothe, R. Martschke, H. Fischer, J. Chem. Soc. Perkin
Trans. 2 1998, 503–508; e) I. Texier, J. A. Delaire, C.
Giannotti, Phys. Chem. Chem. Phys. 2000, 2, 1205–
1212; f) C. Tanielian, R. Seghrouchni, C. Schweitzer, J.
Phys. Chem. A 2003, 107, 1102–1111.
[29] M. Walbiner, J. Q. Wu, H. Fischer, Helv. Chim. Acta
1995, 78, 910–924.
[30] L. Cermenati, C. Richter, A. Albini, Chem. Commun.
1998, 805–806.
[31] For an analogy, see: I. Ryu, A. Tani, T. Fukuyama, D.
Ravelli, M. Fagnoni, A. Albini, Angew. Chem. 2011,
123, 1909–1912; Angew. Chem. Int. Ed. 2011, 50, 1869–
1872.
[15] M. Ueda, E. Kondoh, Y. Ito, H. Shono, M. Kakiuchi, Y.
Ichii, T. Kimura, T. Miyoshi, T. Naito, O. Miyata, Org.
Biomol. Chem. 2011, 9, 2062–2064.
Adv. Synth. Catal. 2013, 355, 2891 – 2899
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2899