Competing pathways in the photogeneration of didehydrotoluenes from (Trimethylsilylmethyl)aryl sulfonates and phosphates

Article abstract of DOI:10.1002/chem.201404787

The scope of the photochemical generation of a,n-didehydrotoluene diradicals from aryl sulfonates and phosphates and their chemistry are explored. The thermally inaccessible a,2- and a,4-intermediates are efficiently obtained by irradiation of ortho- and para-(trimethylsilylme-thyl)phenyl triflates through heterolytic splitting of the ester anion from the substrate in the triplet state. Triplet phenyl cations are formed and the loss of trimethylsilyl cation from them affords the desired diradicals (³Me₃SiCH₂C₆H₄-OZ→ ³Me₃SiCH2C₆H4^+→CH₂C₆H4). Triplet sensitization is required, for which acetone is used throughout. Direct irradiation leads, on the contrary, to photo-Fries fragmentation (¹Me₃SiCH₂C₆H₄O-Z→Me₃SiCH₂C₆H₄O· + Z). With mesylates, where ester cleavage is less convenient, a further competition from the triplet is direct desilylation. Didehydrotoluenes are also obtained from the corresponding phosphates, although with poor efficiency.

Full text of DOI:10.1002/chem.201404787

DOI: 10.1002/chem.201404787
Full Paper
&
Photochemistry
Competing Pathways in the Photogeneration of
Didehydrotoluenes from (Trimethylsilylmethyl)aryl Sulfonates and
Phosphates
[
a]
Stefano Crespi, Davide Ravelli, Stefano Protti, Angelo Albini, and Maurizio Fagnoni*
3
+
·
·
Abstract: The scope of the photochemical generation of
a,n-didehydrotoluene diradicals from aryl sulfonates and
phosphates and their chemistry are explored. The thermally
inaccessible a,2- and a,4- intermediates are efficiently ob-
tained by irradiation of ortho- and para-(trimethylsilylme-
thyl)phenyl triflates through heterolytic splitting of the ester
anion from the substrate in the triplet state. Triplet phenyl
cations are formed and the loss of trimethylsilyl cation from
Me SiCH C H ! CH C H ). Triplet sensitization is required,
3
2
6
4
2
6
4
for which acetone is used throughout. Direct irradiation
leads, on the contrary, to photo-Fries fragmentation
1
·
·
( Me SiCH C H O-Z!Me SiCH C H O +Z). With mesylates,
3
2
6
4
3
2
6
4
where ester cleavage is less convenient, a further competi-
tion from the triplet is direct desilylation. Didehydrotoluenes
are also obtained from the corresponding phosphates, al-
though with poor efficiency.
3
them affords the desired diradicals ( Me SiCH C H -OZ!
3
2
6
4
Introduction
involves a “radical” path, presumably via benzyl radicals, and
[8,9]
an “ionic” path.
The photochemistry of aromatic compounds seems to provide
an inexhaustible source of new reactions, although these often
Thus a,n-didehydrotoluenes were formed upon irradiation
of E-substituted aryl chlorides (E=Me Si), such as isomeric
3
[
1]
[7]
defy simple rationalization. A new development in recent
(n-chlorobenzyl)trimethylsilanes, as an alternative to the
[1e,2]
years has been the heterolytic cleavage of aromatic halides.
Myers–Saito cycloaromatization of enyne–allenes (Scheme 1a,
[8,10,11]
[7]
When irradiated in polar, protic solvents, aryl chlorides (provid-
ed that no electron-withdrawing substituent is present) pro-
duce phenyl cations in what appears to be a general reaction
right side).
In the photochemical approach, however,
the process starts from a stable aromatic derivative, rather
than from a less-easily prepared polyunsaturated molecule.
The significance of the photochemical approach obviously also
depends on its versatility. However, knowledge is still limited
concerning the generation and the chemistry of didehydroto-
luenes according to Scheme 1a (left side) and the factors that
affect the points of divergence. This is a key issue when poten-
tial applications of such intermediates are considered. As an
example, these diradicals appear to be formed as the active
moiety in several natural compounds in which their ability to
[
1e,2]
via the triplet state [Eq. (1)].
ArꢀX þ hn ! Ar þ Xꢀ
þ
ð1Þ
However, the efficiency of the process largely depends on
ꢀ
the nature of the leaving anion X , on the other substituents
present on the aromatic ring, and on their relative positions.
The chemistry of the triplet cations thus formed has revealed
interesting facets. In particular, they behave as selective elec-
[12,13]
attack DNA endows them with antineoplastic action.
[
3]
[4]
trophiles, typically adding to alkenes, alkynes, and (hetero)-
[
5]
aromatics, thus leading to convenient metal-free phenylation
[
1e,2]
procedures.
ent in the starting molecule, it may split away directly giving
However, when an electrofugal group E is pres-
[
6,7]
a highly unsaturated neutral species.
Thus, a s,p-diradical,
a didehydrotoluene (DHT), is formed when E is at the benzylic
[
7]
position (Scheme 1a, left side). In turn, DHTs undergo a chem-
ical reaction that, in accordance with previous investigations,
[a] S. Crespi, Dr. D. Ravelli, Dr. S. Protti, Prof. A. Albini, Prof. M. Fagnoni
PhotoGreen Lab, Department of Chemistry
University of Pavia, Viale Taramelli 12, 27100 Pavia (Italy)
E-mail: fagnoni@unipv.it
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404787.
Scheme 1. a) Thermal (right) and photochemical (left) generation of a,n-
DHTs. b) Photochemistry of aryl sulfonates and phosphates
Chem. Eur. J. 2014, 20, 1 – 8
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
We (and others) have found that inorganic esters (phos-
phates, sulfonates), of general formula ArꢀOZ, operate as nu-
cleofugal leaving groups in the place of chlorides (path a in
Table 1. Photophysical data about phenyl esters 1–3.
ArOZ Stoke’s shift (MeOH)
l
F
[nm]
l
F
[nm]
F
ꢀ1
3
ꢀ1
[
4,14]
[10 cm
]
(F , MeOH) (F , C H ) (MeOH)
Scheme 1b),
in competition with ArOꢀZ bond cleavage
F
F
6
12
[
15]
(path b in Scheme 1b). The easy availability of such com-
1a
4.25
3.64
4.83
3.74
3.84
5.14
3.73
3.24
3.63
305 (0.035) 300 (0.036) 0.61
297 (0.003) 296 (0.009) 0.25
309 (0.006) 304 (0.005) 0.14
299 (0.003) 295 (0.003) 0.65
300 (0.003) 294 (0004) 0.27
309 (0.02) 304 (0.03) 0.35
304 (0.23) 302 (0.28) 0.18
294 (0.21) 296 (0.27) 0.02
298 (0.12) 298 (0.13) 0.07
1
1
2
2
b
c
a
b
pounds makes the choice appealing, although the scope is
little known. As a contribution to the clarification of the
matter, we decided to carry out a systematic study of some
esters of (trimethylsilylmethyl)phenols, more precisely, mesy-
lates (1a–c), triflates (2a–c), and diethylphosphates (3a–c;
Figure 1), and to supplement product studies with a computa-
tional investigation.
2c
3
3
3
a
b
c
magnitude, whereas the medium (cyclohexane and methanol)
had a much smaller influence. The fluorescence quantum yield
was found to be very low for the sulfonates 1 and 2, but up to
28% for the phosphates 3 (Table 1).
The photochemical reactions were carried out both by direct
irradiation at 254 nm and by acetone sensitization. Previous
studies on related compounds suggested that decomposition
[
7a]
Figure 1. (Trimethylsilylmethyl)phenyl mesylates (1), triflates (2), and phos-
phates (3).
is faster in water–methanol mixed solvents,
and this was
confirmed for 1a, but since some of the esters were not suffi-
ciently soluble in that medium, the irradiations (Table 2) were
carried out in neat methanol, in order to obtain a sensible
comparison. All of the compounds considered were quite pho-
Results
tosensitive, with a disappearance quantum yield (F ) higher
ꢀ1
than 0.6 for the para-mesylate and triflate, somewhat lower for
Experimental studies
the meta and ortho isomers of the same esters, and much
The esters used were prepared by known procedures (see the
Supporting Information). All of the compounds examined have
a large absorption envelope with some vibrational structure
lower for the phosphates (F 1 =0.02–0.18; Table 1). The prod-
ꢀ
uct(s) obtained at elevated (ꢂ80%) conversion are summar-
ized in Table 2. Upon direct irradiation, a series of products
conserving the trimethylsilyl moiety predominated. These in-
cluded phenol 4, formed through detachment of the sulfonyl
group, and the sulfonylphenol 4’, (only from triflate 2a), as
(
lmax ꢁ270 nm; Figure 2 for 2a, Figures S1–S9 in the Support-
ing Information for the others), and a moderate Stoke’s shift
3
ꢀ1
(20–40 nm or 3–5ꢁ10 cm ).
The physical properties of the excited states depended on
well as products in which the SO group had been lost, namely
2
the structure of the ester, but by no more than an order of
the trifluoromethylphenol 4’’ and the trifluoromethyl ether 4’’’
(
Scheme 2).
Anisole 5 from the meta-phosphate 3b and benzyltrimethyl-
silane 6 (from reductive elimination of the ester), mostly in
traces, were likewise formed. Irradiations of triflate 2a carried
out in other solvents (cyclohexane, CH Cl , ethyl acetate, aceto-
2
2
nitrile) led to only TMS-conserving products (Table 2).
In contrast, under acetone sensitization, products resulting
from both desilylation and reductive elimination of the ester
group predominated (Scheme 2). These included toluene 7, di-
phenylethane 7’, the two alcohols 7’’ and 7’’’, and benzyl ether
8
. Some of the compounds underwent desilylation only,
whether accompanied by hydrolysis of the ester group or not
esters 9aMs, 9aPhos, 9c, and phenols 9’ and 9’’). Separate ex-
(
periments showed that alcohol 9’ resulted from the secondary
photoreaction of 4a. Notably, the presence of oxygen in the
direct irradiation of 2a in neat methanol affected only slightly
the product distribution, whereas the photoreactivity of this
triflate was completely suppressed under oxygenated sensi-
tized conditions.
Figure 2. Absorption (solid line; left axis) and emission (dotted line; right
axis) spectra of ester 2a in methanol.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[
a]
Table 2. Products from the irradiation of phenyl esters 1–3 (0.025m).
ArOZ Conditions Silylated products (% yield)
Silicon-free products (% yield)
ArꢀO moiety not maintained ArꢀO moiety not maintained ArꢀO moiety maintained
ArꢀO moiety maintained
Direct irradiation
Sensitized irradiation
4a (38)
4a (4)
1
1
1
a
b
c
[
b]
6 (tr)
7’’ (8), 7’’’ (8)
9aMs (23), 9’ (6), 9’’ (8)
Direct
4b (44)
4b (93)
[
b]
Sens.
Direct
4c (28)
4c (28)
[
b]
Sens.
Direct
Direct
Direct
Direct
Direct
Direct
4a (16), 4’ (9), 4’’ (13), 4’’’ (33)
4’ (7), 4’’ (7), 4’’’ (26)
4a (7), 4’ (9), 4’’ (9), 4’’’ (44)
4’ (13), 4’’ (9), 4’’’ (23)
4’ (5), 4’’ (7), 4’’’ (20)
6 (tr)
6 (tr)
9’ (11)
[
[
[
[
[
c,d]
e]
f,d]
g,d]
d,h]
2
a
4’ (11), 4’’ (16), 4’’’ (59)
[
b]
Sens.
Sens.
6 (tr)
—
7 (8), 7’ (tr), 7’’ (42), 7’’’ (25)
[b,c,i]
—
—
—
Direct
Sens.
4b (27)
4b (tr)
6 (6)
6 (tr)
8 (4)
2
2
3
3
3
b
c
[
b]
7’’ (21), 7’’’ (15), 8 (51)
Direct
4c (8), 4’’ (22)
6 (tr)
6 (tr)
[
b]
Sens.
7’’ (10), 7’’’ (15)
Direct
7’ (1), 7’’ (9)
7’’ (21), 7’’’ (37)
a
b
c
[
b]
Sens.
9aPhos (6)
Direct
4b (23), 5 (3)
6 (5)
—
[b,i]
Sens.
—
—
—
Direct
4c (13)
6 (tr)
7’ (tr), 7’’ (5)
7’’ (30), 7’’’ (28)
9c (5)
[
b]
Sens.
[a] A 0.025m solution of 1–3 in MeOH (direct irradiation at l=254 nm and sensitized irradiation at l=310 nm) using N
2
-purged solution. Conversion
ꢂ80%; [b] acetone 10% v/v was used as sensitizer; [c] irradiation carried out in O
2
-saturated solution; [d] conversion between 50 and 70%; [e] irradiation
2 2
in cyclohexane; [f] in CH Cl ; [g] in MeCN; [h] in ethyl acetate; [i] no significant reagent consumption was observed.
Computational studies
(
ꢀ0.23; Figure 3b). The question was whether heterolytic
Support was sought in computation, by using DFT at the
theory level CPCM-B3LYP/6-31G(d), adopting an unrestricted
cleavage of the arylꢀO bond was a possibility.
This point was answered by calculating the increase in
energy related to the elongation of the ArꢀOZ bond by ap-
proximately 2 ꢂ, from the equilibrium value up to approxi-
mately 3.5 ꢂ (Figure 3c). Relevant parameters for the “stretch-
(
(
U) formalism when triplet states were under consideration
see the Supporting Information for further details). The aro-
matic ring in ground state S of benzylsilanes 1–3, is in every
0
3
case planar, with the ester oxygen lying, as one might expect,
in the same plane (Figure 3a for 4-substituted triflate 2a, Fig-
ures S11–S19 in the Supporting Information for the others). A
ed” structures are reported in Figure 3c for 2a(a) (see
3
ST
2a(a) ), and in Figures S11–S19 (see the Supporting Informa-
tion) for the others. The most telling point was that the nega-
tive charge on the oxygen atom more than doubled upon
stretching, rising to >ꢀ0.60. In contrast, the spin density on
the same atom remained almost unchanged or diminished for
profound change was apparent in the triplet states T , which
1
are puckered at the optimized minima, the ring carbon bearing
the ester group protruding out of the aromatic plane by ap-
proximately 108, while the arylꢀO bond is further tilted by ap-
proximately 308 (Figure 3b). The groups present in these mole-
cules further introduce a variety of conformations. Thus, in
puckered triplets the a conformer has the TMS and the ester
moiety protruding out of the plane on the same side (Fig-
ure 3b), whereas in the b conformer they are on opposite
sides (Figure S10 in the Supporting Information). The ArꢀOZ
bond length was approximately 0.06 ꢂ longer than in the
ground state and the oxygen atom bore some negative charge
[16]
triflates (lower than 5%), whereas it increased for mesylates
and phosphates, reaching a value of 15–25%. The energy pro-
file obtained for each of the compounds considered is shown
in Figure 4, where the energy change (DE) associated with the
elongation of the title bond with respect to the equilibrium
geometry (DI) is plotted. As can be seen, the energy barrier
ꢀ
1
confronted was 25–29 kcalmol
for mesylates, 13–19 kcal
ꢀ
1
ꢀ1
mol
for triflates and 31–34 kcalmol
for phosphates
(1 kcal=4.184 kJ).
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Discussion
The lowest singlet excited state
in esters 1–3 appears to be of
pp* character, as is almost
always the case for aromatic
molecules. Fluorescence is an
important deactivation mecha-
nism for phosphates (F =0.12–
F
0.28; Table 1), which explains the
poor reactivity of these com-
pounds, but not for the sulfo-
nates. The photophysical proper-
ties in methanol and cyclohex-
ane parallel each other closely,
whereas the chemistry changes
completely in the polar solvent
and proceeds via the triplet.
Taken together, these results
support the conclusion that the
singlet excited-state nature (a
pp* state) remains the same in
all of the derivatives studied,
whereas a triplet, also pp*, is
formed by intersystem crossing
(
ISC).
With all of the compounds ex-
Scheme 2. Products obtained upon irradiation of aryl esters 1-3
amined, except for 1a, two main
photochemical processes occur;
homolytic cleavage of the ArOꢀ
Z bond to give a phenoxy and a sulfonyl or phosphonyl radi-
cal, and heterolytic fragmentation of the ArꢀOZ bond. That sul-
[17]
fonate aryl esters, in analogy with acyl aryl esters, undergo
·
·
homolytic cleavage to radical pair 10-O and Z has been previ-
[14c,15]
ously documented in the literature.
Phenol 4 and sulfonyl
phenol 4’ arise from this path. In a variation of it, the triflyl rad-
ical fragments to yield a trifluoromethyl radical, and phenol 4’’,
as well as trifluoromethyl ether 4’’’ from their recombination.
The process involves the singlet excited state, apparent from
Figure 3. Optimized structures and relevant parameters for: a) ground state
3
3
2
a, b) first triplet excited state 2a(a) and c) first triplet excited state 2a(a)
3
ST
upon stretching of the ArꢀO bond up to approximately 3.5 ꢂ ( 2a(a) ). The
arrows indicate the charge (qESP) on the ArꢀO atom (the spin density on the
same atom is listed in parentheses).
Figure 4. Calculated energy profile related to the elongation of the ArꢀOZ
bond by approximately 2 ꢂ, from the equilibrium value up to approximately
3
3
3
.5 ꢂ in compounds 1- 3.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the fact that these compounds predominate by direct irradia-
tion, but not upon acetone sensitization, and further from the
lack of a dependence on the polarity of the medium.
lane 5, resulting from the trapping of singlet phenyl cation
1
+
11 b , was obtained. A further exception to the generalization
above is that of mesylate 1a. This is the fastest reacting sub-
strate, but leads to a complex mixture and degradation. The
main product under sensitized conditions is the desilylated
ester 9aMs. In view of the reaction sequence proposed above,
this is not unreasonable, since the ArꢀOZ cleavage has to con-
front a high barrier in this case (Figure 4) and this opens the
With triflates and mesylates, the fluorescence quantum yield
decreases by one to two orders of magnitude. As mentioned,
the triplet state reactions differ from those of the singlet, as
shown by the complete change in the chemistry under sensi-
tized conditions, particularly with the para derivatives. The first
triplet formed is a pp* state and the computational analysis
above strongly supports the conclusion that in a polar
medium heterolytic cleavage occurs in this case, since stretch-
ing of the ArꢀOZ bond is accompanied by a marked charge
separation. The prediction that this cleavage process is more
efficient for triflates than for the other two families studied
[21]
path to homolytic desilylation to yield radical 14.
In summary, this work greatly enlarges the scope of our ini-
tial report on chlorobenzylsilanes and clarifies the range of
possibilities for photochemical access to DHTs. Thus, aryl sulfo-
nates and, albeit with a lower efficiency, phosphates can be
used in place of halides for the photochemical generation of
DHTs via phenyl cation intermediates (path a in Scheme 3).
(compare the energies involved in Figure 4) is fully borne out
by experiment and the barrier is very close to that calculated
[
7b]
for the corresponding chlorides.
It is thus reasonable that
the quantum yields should be of the same order of magnitude.
Under sensitized conditions, a triplet state is clearly involved in
the process, as demonstrated by the efficient quenching with
oxygen (Table 2).
3
+
Cleavage produces phenyl cations in the triplet state ( 11 ).
In the presence of hydrogen-donating solvents, such as metha-
nol, such species are known to be reduced. In this case, a rudi-
mentary indication of such a process is given by the formation
of traces of benzyltrimethylsilane (6) in most of the reactions
considered. In the absence of a specific trap, however, the fast-
est reaction is elimination of the trimethysilyl cation to give di-
3
3
+
dehydrotoluenes 12. The optimized structure of cation 11
has the CH ꢀSiMe bond perpendicular to the molecular plane
2
3
1
5 [18]
and an electronic structure (s p ) that is similar to that of
Scheme 3. Photochemistry of benzylsilanes 1–3.
5
the benzyltrimethylsilane radical cation (p ) produced by elec-
tron transfer. For both intermediates, desilylation is facile and
However, this change may introduce side paths, such as fluo-
rescence (important for phosphates, up to 28% for the para
isomer) and chemical reaction from the singlet state (i.e. homo-
lytic fragmentation) important for mesylates, up to 20% for
the para isomer (path b in Scheme 3). Path a occurs exclusively
on the triplet surface, and triplet sensitization is a useful way
to enable this path, as apparent in particular with the o- and
p-triflates, where acetone sensitization shifts the reaction from
almost fully singlet to almost fully triplet. However, in several
ꢀ
1 [7b,19]
overcomes a modest barrier (18–27 kcalmol ).
Previous investigations on DHTs formed from the cycloaro-
matization of enyne-allenes had evidenced two different chem-
istries, such as hydrogen abstraction from the solvent to form
benzyl radicals (13) and an “ionic” path leading to nucleophile
[
8d,9]
3
+
addition.
Didehydrotoluenes formed from cations 11 are
3
of triplet multiplicity ( 12) and in accordance with their radical
nature, hydrogen abstraction is a favored process and products
from benzyl radicals predominate (toluene 7, diphenylethane
ꢀ1
cases the high energy of such triplets (75–78 kcalmol , calcu-
lated; see the Supporting Information, Table S3) compared
[
7]
7
’, phenylethanol 7’’). Particularly convincing are the results
ꢀ
1
[22]
in acetone-sensitized experiments, where 2-methyl-1-phenyl-
with that of acetone (78.9 kcalmol , experimental) makes
this procedure ineffectual under “sensitized conditions”.
The trimethylsilyl group is confirmed as an excellent leaving
propan-2-ol 7’’’ was detected. This arises from the competitive
·
trapping of the Me COH radical (from the photochemical reac-
2
[19]
tion of acetone with MeOH) with radical 13, and further sup-
ports the presence of the latter species.
group. Virtually all of the triplet cations fragmented to give
the didehydrotoluene and only traces of 6 were left to indicate
the role of the phenyl cation as the intermediate.
In the case of meta-triflate 2b, however, benzyl ether 8 is
the main product. This is the expected product from the
The proposed reaction sequence applies regardless of the
relative positions of nucleofugal and electrofugal groups, thus
giving access to a,2-, a,3-, and a,4-DHTs, whereas thermal cy-
cloaromatization is useful only for the a,3-didehydrotoluenes.
The a,2- and a,4-didehydrotoluenes react exclusively by hy-
drogen abstraction producing the benzyl radical 13. With the
meta isomer, however, >55% of the product comes from the
“ionic” path (ether 8). Such a manifestation of the “meta effect”
“
ionic” path and it is tempting to attribute this change in reac-
1
tivity to fast ISC to 12, actually the ground state in this inter-
mediate.
[
7b]
The different behavior of the meta-DHT may be
viewed as another manifestation of the “meta effect” in aro-
matic compounds that has been evidenced in many photo-
[
20]
chemical reactions. Further notice that in the case of meta-
phosphate, a small amount of (m-methoxybenzyl)trimethylsi-
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
finds a parallel in the small amount of singlet phenyl cation
product (5) from the meta-phosphate.
When homolytic cleavage of the ArOꢀZ bond is not favored,
other processes intervene, such as desilylation in the case of
M. Fagnoni, A. Albini in Experiments in green and sustainable chemistry
(
2
Eds.: H. W. Roesky, D. K. Kennepohl) Wiley-VCH Verlag GmbH & Co
009, pp. 260–265.
[
3] See for instance: a) S. Protti, D. Dondi, M. Mella, M. Fagnoni, A. Albini,
Eur. J. Org. Chem. 2011, 3229–3237; b) V. Dichiarante, M. Fagnoni, M.
Mella, A. Albini, Chem. Eur. J. 2006, 12, 3905–3915; c) M. Mella, P.
Coppo, B. Guizzardi, M. Fagnoni, M. Freccero, A. Albini, J. Org. Chem.
1
a (path c in Scheme 3). This result demonstrates the limits of
the photochemical approach.
2
001, 66, 6344–6352.
[
[
4] a) S. Protti, M. Fagnoni, A. Albini, Angew. Chem. 2005, 117, 5821–5824;
Angew. Chem. Int. Ed. 2005, 44, 5675–5678; b) S. Protti, M. Fagnoni, A.
Albini, J. Org. Chem. 2012, 77, 6473–6479.
Conclusion
5] a) B. Guizzardi, M. Mella, M. Fagnoni, A. Albini, Tetrahedron 2000, 56,
On the basis of the work described above, the scheme for the
generation of didehydrotoluene adducts from the irradiation
9383–9389; b) M. Slegt, H. S. Overkleeft, G. Lodder, Eur. J. Org. Chem.
2
007, 5364–5375; c) V. Dichiarante, M. Fagnoni, A. Albini, Angew. Chem.
3
3
of I in polar media, via triplet I and phenyl cation II
Scheme 1) is thus recognized as a general scheme in aromatic
2007, 119, 6615–6618; Angew. Chem. Int. Ed. 2007, 46, 6495–6498.
[
6] S. Protti, V. Dichiarante, D. Dondi, M. Fagnoni, A. Albini, Chem. Sci. 2012,
(
3
, 1330–1337.
7] a) S. Protti, D. Ravelli, B. Mannucci, A. Albini, M. Fagnoni, Angew. Chem.
012, 124, 8705–8708; Angew. Chem. Int. Ed. 2012, 51, 8577–8580;
b) D. Ravelli, S. Protti, M. Fagnoni, A. Albini, J. Org. Chem. 2013, 78,
814–3820.
8] a) R. Nagata, H. Yamanaka, E. Okazaki, I. Saito, Tetrahedron Lett. 1989,
0, 4995–4998; b) A. G. Myers, E. Y. Kuo, N. S. Finney, J. Am. Chem. Soc.
989, 111, 8057–8059; c) K. Toshima, K. Ohta, T. Kano, T. Nakamura, M.
Nakata, M. Kinoshita, S. Matsumura, Bioorg. Med. Chem. 1996, 4, 105–
13; d) A. G. Myers, P. S. Dragovich, E. Y. Kuo, J. Am. Chem. Soc. 1992,
14, 9369–9386.
photochemistry. Two powerful electrophiles are produced, trip-
let phenyl cations and didehydrotoluenes, the latter showing
distinct diradical reactivity. Applications of these results are un-
derway to exploit (trimethylsilylmethyl)phenyl esters as precur-
sors for the photogeneration of aggressive diradicals for the
[
2
3
[
3
1
[
12,13]
cleavage of DNA.
1
1
Experimental Section
[
9] a) I. Suzuki, A. Shigenaga, A. Manabe, H. Nemoto, M. Shibuya, Tetrahe-
dron 2003, 59, 5691–5704; b) M. E. Cremeens, T. S. Hughes, B. K. Car-
penter, J. Am. Chem. Soc. 2005, 127, 6652–6661.
General procedure for the irradiation of compounds 1–3: A solu-
tion of the substrate (1–3, 0.025m) in the chosen medium (2 mL)
was poured into quartz tubes, purged with nitrogen, and irradiated
using a multilamp reactor fitted with four low pressure Hg lamps
[
[
10] M. Abe, Chem. Rev. 2013, 113, 7011–7088.
11] a) J. J. Li, P. W. Peterson, R. K. Mohamed, I. V. Alabugin, Eur. J. Org. Chem.
2
013, 2505–2527; b) R. K. Mohamed, P. W. Peterson, I. V. Alabugin,
(
emission maximum at 254 nm) or ten 15 W phosphor-coated
Chem. Rev. 2013, 113, 7089–7129; c) K. K. Wang, Chem. Rev. 1996, 96,
lamps (emission maximum at 310 nm) for the acetone-sensitized
photoreactions. The solution was then analyzed by GC/MS. The
photoproducts were identified by comparison with authentic sam-
ples and quantified by using calibration curves.
207–222; d) E. Kraka, D. Cremer, WIREs Comput. Mol. Sci. 2014, 4, 285-
324.
[
12] See for instance: a) C.-F. Lin, M.-J. Wu, J. Org. Chem. 1997, 62, 4546–
548; b) I. Suzuki, Y. Tsuchiya, A. Shigenaga, H. Nemoto, M. Shibuya, Tet-
4
rahedron Lett. 2002, 43, 6779–6781; c) I. Suzuki, S. Uno, Y. Tsuchiya, A.
Shigenaga, H. Nemoto, M. Shibuya, Bioorg. Med. Chem. Lett. 2004, 14,
2959–2962.
Acknowledgements
[
13] a) A. G. Myers, C. A. Parrish, Bioconjugate Chem. 1996, 7, 322–331;
b) A. V. Kuzmin, V. V. Popik, Chem. Commun. 2009, 5707–5709.
14] a) M. De Carolis, S. Protti, M. Fagnoni, A. Albini, Angew. Chem. 2005, 117,
This work was supported by the Fondazione Cariplo (grant
no. 2011–1839). S.P. acknowledges MIUR, Rome (FIRB-Futuro in
Ricerca 2008 project no. RBFR08J78Q) for financial support. We
thank Prof. P. Hoggard (Santa Clara University) for fruitful dis-
cussion. We are grateful to B. Mannucci (Centro Grandi Stru-
menti-Pavia) for the valuable assistance. This work was funded
by the CINECA Supercomputer Center, with computer time
granted by ISCRA projects (codes HP10CZEHG6, HP10C6PCC2,
and HP10C3CPWN).
[
1258–1262; Angew. Chem. Int. Ed. 2005, 44, 1232–1236; b) E. Abitelli, S.
Protti, M. Fagnoni, A. Albini, J. Org. Chem. 2012, 77, 3501–3507; c) D.
Ravelli, M. Fagnoni in CRC Handbook of Organic Photochemistry and
Photobiology, 3rd ed. (Eds.: A. Griesbeck, M. Oelgemoeller, F. Ghetti),
CRC Press, 2012, pp. 393–417.
[
15] a) J. L. Stratenus, E. Havinga, Rec. Trav. Chim. Pays-Bas 1966, 85, 434–
436; b) Y. Ogata, K. Takagi, S. Yamada, J. Chem. Soc. Perkin Trans. 2 1977,
1629–1632; c) K. Pitchumani, M. C. D. Manickam, C. Srinivasan, Indian J.
Chem. B 1993, 32, 1074–1076; d) J. Andraos, G. G. Barclay, D. R. Me-
deiros, M. V. Baldovi, J. C. Scaiano, R. Sinta, Chem. Mater. 1998, 10,
1694–1699; e) L. K. Crevatꢃn, S. M. Bonesi, R. Erra-Balsells, Helv. Chim.
Acta 2006, 89, 1147–1157; f) Y. Kageyama, R. Ohshima, K. Sakurama, Y.
Fujiwara, Y. Tanimoto, Y. Yamada, S. Aoki, Chem. Pharm. Bull. 2009, 57,
Keywords: aryl sulfonates
cations · photochemistry · radicals
· cleavage reactions · phenyl
1257–1266; g) M. Terpolilli, D. Merli, S. Protti, V. Dichiarante, M. Fagnoni,
A. Albini, Photochem. Photobiol. Sci. 2011, 10, 123–127.
[
1] a) M. Fagnoni, A. Albini in Molecular and Supramolecular Photochemistry,
Vol. 14: Organic Photochemistry and Photophysics (Eds.: V. Ramamurthy,
K. Schanze), Dekker, New York 2006, pp. 131–177; b) Handbook of Syn-
thetic Photochemistry (Eds.: A. Albini, M. Fagnoni) Wiley-VCH, Weinheim
[16] By comparison, in 4-chlorobenzyltrimethylsilane triplet state, the CꢀCl
bond is tilted out of the plane by 288 with a partial negative charge of
ꢀ0.21, which increases to ꢀ0.78 upon stretching, implying an energy
ꢀ
1
[7b]
increase of 14 kcalmol . See reference
.
2
010; c) N. J. Turro, V. Ramamurthy, J. C. Scaiano, Modern Molecular Pho-
[17] M. A. Miranda, F. Galindo in CRC Handbook of Organic Photochemistry
and Photobiology, 2nd ed. (Eds.: F. Lenci, W. Horspool), CRC Press, Boca
Raton, United States, 2004, pp. 42-1–42-11.
[18] a) S. Lazzaroni, D. Dondi, M. Fagnoni, A. Albini, J. Org. Chem. 2008, 73,
206–211; b) S. Lazzaroni, D. Dondi, M. Fagnoni, A. Albini, J. Org. Chem.
2010, 75, 315–323; c) S. V. Bondarchuk, B. F. Minaev, Chem. Phys. 2011,
389, 68–74; d) S. V. Bondarchuk, B. F. Minaev, J. Phys. Chem. A 2014, 118,
3201–3210.
tochemistry of Organic Molecules, University Science Books, 2010; d) N.
Hoffmann, Photochem. Photobiol. Sci. 2012, 11, 1613–1641; e) A. Albini,
M. Fagnoni, Photochemically-Generated Intermediates in Synthesis John
Wiley & Sons, Hoboken 2013.
2] a) V. Dichiarante, M. Fagnoni, Synlett 2008, 787–800; b) V. Dichiarante,
M. Fagnoni, A. Albini in Modern Arylation Methods (Ed.: L. Ackermann)
Wiley-VCH Verlag GmbH & Co 2009, pp. 513–535; c) S. Protti, D. Dondi,
[
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[
19] a) J. P. Dinnocenzo, S. Farid, J. L. Goodman, I. R. Gould, S. L. Mattes, W. P.
Todd, J. Am. Chem. Soc. 1989, 111, 8973–8975; b) K. Taylor, K. Miura, F.
Akinfaderin, A. J. Fry, J. Electrochem. Soc. 2003, 150, D85–D86.
[21] The direct photolysis of the CꢀSi bond in benzyltrimethylsilanes has
been rarely described. See for instance: a) M. Kira, H. Yoshida, H. Sakur-
ai, J. Am. Chem. Soc. 1985, 107, 7767–7768; b) H. Hiratsuka, Y. Kadokura,
H. Chida, M. Tanaka, S. Kobayashi, T. Okutsu, M. Oba, K. Nishiyama, J.
Chem. Soc. Faraday Trans. 1996, 92, 3035–3041.
[
20] a) E. Havinga, R. O. de Jongh, W. Dorst, Recl. Trav. Chim. Pays-Bas 1956,
7
5, 378–383; b) H. E. Zimmerman, V. R. Sandel, J. Am. Chem. Soc. 1963,
5, 915–922; c) H. E. Zimmerman, S. Somasekhara, J. Am. Chem. Soc.
8
[22] CRC Handbook of Organic Photochemistry, Vol. 1 (Ed.: J. C. Scaiano), CRC
Press, Boca Raton, United States, 1989.
1963, 85, 922–927; d) H. E. Zimmerman, J. Am. Chem. Soc. 1995, 117,
8988–8991; e) V. Dichiarante, D. Dondi, S. Protti, M. Fagnoni, A. Albini,
J. Am. Chem. Soc. 2007, 129, 5605–5611; f) V. Dichiarante, D. Dondi, S.
Protti, M. Fagnoni, A. Albini, J. Am. Chem. Soc. 2007, 129, 11662.
Received: August 8, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Photochemistry
S. Crespi, D. Ravelli, S. Protti, A. Albini,
M. Fagnoni*
&
& – &&
The trouble with triplets: a,2-, a,3-,
and a,4-didehydrotoluene diradicals
were photochemically generated from
suitable (trimethylsilylmethyl)phenyl
mesylates, triflates, and phosphates. The
initial photoheterolytic splitting of the
ester anion from the substrate in the
triplet state formed a triplet phenyl
cation that upon loss of trimethylsilyl
cation afforded the desired diradicals.
Triplet sensitization by acetone was re-
quired since direct irradiation led mainly
to photo-Fries fragmentation.
Competing Pathways in the
Photogeneration of
Didehydrotoluenes from
(Trimethylsilylmethyl)aryl Sulfonates
and Phosphates
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!