Photohomolysis and Photoheterolysis in Aryl Sulfonates and Aryl Phosphates

Article abstract of DOI:10.1002/chem.202005426

The photochemical behaviour of selected aryl sulfonates and phosphates (ArOX) in polar and nonpolar media has been investigated by laser flash photolysis (LFP) experiments. Two main pathways have been identified, namely the photohomolysis of the ArO?X bon

Full text of DOI:10.1002/chem.202005426

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
&
Photochemistry
Photohomolysis and Photoheterolysis in Aryl Sulfonates and Aryl
Phosphates
Sergio Bonesi,[a, b] Stefano Protti,[a] and Maurizio Fagnoni*[a]
Abstract: The photochemical behaviour of selected aryl sul-
fonates and phosphates (ArOX) in polar and nonpolar media
has been investigated by laser flash photolysis (LFP) experi-
ments. Two main pathways have been identified, namely the
photohomolysis of the ArOÀX bond or the photoheterolysis
of the ArÀOX bond depending on the nature of the leaving
group (OX) and on the nature of the substituents on the ar-
omatic ring. In nonpolar solvents the esters are quite photo-
stable due to an efficient triplet deactivation. In polar sol-
vents, the homolytic fragmentation of the ArOÀS bond from
the exited singlets was found in aryl sulfonates bearing
moderately electron-donating groups as well as electron-
withdrawing groups. In electron-rich aryl phosphates and
sulfonates photoheterolysis of the ArÀOP/ArÀOS bond took
place as the exclusive pathway.
Introduction
Aryl sulfonates[1] and aryl phosphates[1a,b,2] of general formulae
Ar-O-X (X=SO2R’ or P(O)(OR’)2) were sparsely used as electro-
philic partners in metal-catalysed cross-coupling arylations.
However, in most cases such compounds are known to be
photoreactive where the sulfonyl or the phosphonyl group is
lost upon light irradiation.[3,4] Sparse examples in the literature
suggest that the fate of the photochemical behaviour is
strongly dependent on several factors, including the nature of
the sulfonyl(phosphonyl) group, of the aromatic substituents
and the reaction conditions.[3] The presence of partners able to
interact with the excited states of aryl esters or with the photo-
generated intermediates can also tune the photoreactivity of
the substrates.[3]
Scheme 1. Competing pathways in the photoreactivity of aryl sulfonates
and phosphates ArOX.
The possible pathways involve a monomolecular or a bimo-
lecular process as briefly summarized in Scheme 1. In the latter
case, a photoinduced monoelectronic reduction of the ester
usually occurred and the resulting radical anion can release
the sulfonate(phosphate) anion, to generate an aryl radical
(path a)[3,5] even by an ArSRN1 process.[3,6] When the aromatic is
substituted with an electron-withdrawing group (especially in
nitro substituted aryl phosphates) a photosubstitution took
place in the presence of a suitable nucleophile (Nu, path b).[7,8]
In the last decade we were intrigued to investigate the
photoinduced monomolecular reactions of ArOX. It was ob-
served that in some cases the photohomolysis of the ArOÀX
bond from the singlet state (Scheme 1, path c) resulted in the
formation of a phenoxyl radical and a heteroatom based radi-
cal. Radical coupling may lead to photo-Fries adducts
(path d).[3,9] An alternative pathway can likewise take place
promoted by the population of the triplet state by Intersystem
1
Crossing of ArOX and ensuing photoheterolysis of the ArÀOX
bond to afford an aryl cation (path e).[10] An electron photo-
ejection to generate the corresponding radical cation cannot
be ruled out in the photolysis of electron-rich derivatives
(path f).[11]
[a] Prof. S. Bonesi, Prof. S. Protti, Prof. M. Fagnoni
Department of Chemistry, PhotoGreen Lab
University of Pavia, V. Le Taramelli 12, 27100 Pavia (Italy)
E-mail: fagnoni@unipv.it
Whereas the photoheterolysis of the Ar-OX bond occurs ex-
clusively in aryl phosphates bearing electron-rich substituents
in polar media,[12] a more complex situation was reported in
the case of aryl sulfonates, since the photohomolysis of the
ArOÀS bond (e.g. in aryl nonaflates,[13] imidazylates[14] or tosyl-
ates[15]) was the preferred path. On the other hand, in most
[b] Prof. S. Bonesi
Departamento de Química Orgµnica, CIHIDECAR— CONICET
Facultad de Ciencias Exactas y Naturales, University of Buenos Aires
3er Piso, Pabellón 2, Ciudad Universitaria, Buenos Aires 1428 (Argentina)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/chem.202005426.
Chem. Eur. J. 2021, 27, 6315 –6323
6315
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
cases triplet sensitization diverts the photoreactivity towards
an aryl cation chemistry.[12a–f,16]
Table 1. Emission maxima (lem) and measured quantum yield fluores-
cence (Fem) of compounds 1–12.
A detailed investigation on the mechanism occurring is im-
portant since these aryl esters may be used as innovative non-
ionic photoacid generators (PAGs).[17] Indeed the photoheterol-
ysis of the Ar-OX bond under deaerated conditions released di-
rectly the strong sulfonic (or phosphoric) acid HOX[12a–f,16]
whereas the sulfonyl radical released in the photohomolysis of
the ArOÀX bond is converted into the corresponding sulfonic
acid upon dioxygen addition.[15]
[a]
Substrates
lem [nm]
Fem
291
0.055
[b]
[b]
–
–
Moreover, the identification of the factors that affected the
photocleavage step is also crucial in devising molecules of po-
tential antitumoral activity activated by the photoheterolytic
cleavage of the Ar-OX bond.[18]
[b]
[b]
–
–
308
317
295
0.03
In view of these premises, we embarked in a detailed photo-
physical investigation (by fluorescence and laser flash photoly-
sis measurements) of selected aryl sulfonates (1–8) and phos-
phates (9–12, see Figure 1) to evaluate the effect of the com-
bined action of the substituents on the aromatic ring and on
the leaving group OX on the photochemistry of these deriva-
tives.
0.097
0.045
[b]
[b]
–
–
370
371
312
0.101
0.183
0.136
294
300
0.078
0.051
Figure 1. Structures of the substituted aryl sulfonates (1–8) and phosphates
(9–12) examined in the present work.
Results and Discussion
[a] 4-Chloroanisole was used as standard (see ref. [12g]) [b] Not detect-
able.
Fluorescence of compounds 1–12
Aryl sulfonates 1–8 exhibited a weak or a negligible fluores-
cence (Table 1 and Figures S1–S9), with the only exception of
derivative 5 (which emission efficiency is in accordance with
other biphenyl derivatives[19]) and of p-N,N-dimethylamino-
phenyl mesylate 8 (Fem =0.101). Phosphates 9–12 exhibited a
low quantum yield emission values reaching Fem =0.183 when
the ester is substituted again with an amino group (com-
pound 9, Table 1).
phate 10 (Figure 2a), which exhibits two characteristic absorp-
tion bands at 320 nm and 390 nm, with a calculated half-time
of 6.1 ms (see Figure 2b, black line) and a decay characterized
by a first-order rate constant (kT) value of 1.7”105 sÀ1 (Table 2).
The presence of oxygen apparently quenched the triplet signal
(Figure 2b, blue line). The triplet lifetimes (tT), the triplet rate
constant (kT) values and the labs of the triplets of the esters are
summarized in Table 2. Accordingly, the exclusive detection of
the triplet signal in aprotic media means that the photocleav-
age of neither the ArÀOX bond nor the ArOÀX bond took
place in compounds 1–12. It should be noticed, however, that
a triplet excited state was also detected in polar media in com-
pounds 5, 8, 10–12 (see Table 2 and following sections).
Laser flash photolysis studies in a nonpolar medium (cyclo-
hexane)
Time-resolved spectroscopy analyses were carried out on a
representative set of aryl esters in cyclohexane. In all the exam-
ined cases, an oxygen sensitive transient species was observed,
nicely displaying a first-order decay (Figures S10–S18). These
species were safely assigned to the triplet states of the exam-
ined compounds. As an exemplary case, we report herein the
transient absorption spectra of p-methoxyphenyl diethylphos-
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6316
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
Figure 2. (a) Transient absorption spectra obtained after a laser pulse (40 ms; lexc =266 nm) of diethylphosphate 10 in N2-saturated cyclohexane solution
(5.0”10À4 m). (b) Decay traces of 10 at 400 nm after the laser pulse (lexc =266 nm) in cyclohexane solution (5.0”10À4 m) under N2 atmosphere (black line) and
under O2 atmosphere (blue line).
Table 2. Spectroscopic data (tT and labs) and triplet deactivation rate constants (kT) of selected aryl sulfonates and diethyl phosphates.[a]
Substrates
Solvent
tT
kT
[sÀ1
labs
[ms]
]
[nm]
cyclohexane
10.3
9.7”104
390
cyclohexane
cyclohexane
11.8
8.0
8.5”104
1.3”105
400
400
cyclohexane
MeCN
MeOH
cyclohexane
MeCN
2.0
2.4
1.6
10.9
1.8
5.0”105
4.2”105
6.3”105
9.2”104
5.6”105
340
340
350
320
325
cyclohexane
0.3
3.7”106
450
cyclohexane
MeCN
MeOH
cyclohexane
MeCN
6.1
2.3
2.8
1.8
0.9
1.7”105
4.3”105
3.6”105
5.6”105
1.1”106
400
400
400
405
410
cyclohexane
5.4
1.87”105
310
[a] Values measured with a laser pulse at lexc =266 nm under N2 atmosphere; Concentration of substrates=5.0”10À4 M.
Laser flash photolysis (LFP) studies of sulfonates 1–8 in
polar solvents
similar behaviour was observed for p-tert-butylphenyl triflate 2
and tosylate 3 (see Figures S21–S24) where, even in this case,
no influence of the oxygen was observed on the generated
species. These signals were unequivocally assigned to the cor-
responding phenoxyl radicals by the comparison made with
the spectra of these species previously reported.[20] However,
traces of an oxygen sensitive signal were found for com-
pound 3 in MeOH and assigned to the triplet excited state (33).
The nonlinear fitting applied on the decay traces measured at
p-tert-Butylphenyl sulfonates 1–3
The LFP analysis (lexc =266 nm) of p-tert-butylphenyl mesylate
1 in MeCN (Figure 3a) and MeOH (Figure 3b) pointed out the
formation of three absorption bands at 290, 380 and 410 nm,
which decay (measured at 410 nm in MeCN) was not affected
by the presence of oxygen (Figure 3c, Figures S19 and S20). A
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6317
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
Figure 3. UV/Vis transient spectra of p-tert-butylphenyl mesylate 1 in (a) nitrogen saturated MeCN and (b) nitrogen saturated MeOH recorded in the range 8–
80 ms after the laser pulse (c) Normalized transient decays of 1 at 410 nm in nitrogen (black) and oxygen (blue) saturated MeCN.
the maximum absorption wavelengths of the transients,
showed a bi-exponential behaviour with r2 values >0.998, in-
dependent on the nature of the solvent. Thus, two lifetime
values appointed as tE and tR were measured in the range of 3
and 30 ms, respectively (see Table 3 for further details).
Aryl mesylates 4–6
The irradiation of p-methoxyphenyl mesylate 4 with a laser
pulse of 266 nm led again to the formation of a transient ab-
sorption spectrum, not sensitive to oxygen with absorption
maxima at 290, 340 and 400 nm (Figure 4a, Figures S25 and
S26). Analogously, p-cyanophenyl mesylate 5, exhibited two
Scheme 2. Transient species formed in polar media from compounds 1–6
after the laser pulse (lexc =266 nm).
bands at 330 nm and 440 nm, respectively (Figure 4b, Figur-
es S27–S30). As in the previous cases, the spectra obtained
from 4 and 5 were assigned to the corresponding phenoxyl
radicals. In the case of 6, a further oxygen sensitive signal was
also present at ca 370 nm (t=2.4 ms and 1.6 ms in nitrogen sa-
turated MeCN (Table 3 and Figure 4c) and MeOH, respectively).
This oxygen sensitive signal (see Figure 4d) was again assigned
to the triplet excited state (Figures S31 and S32).
measured at the maximum absorption wavelengths of the
transients showed a bi-exponential behaviour with r2 values
>0.998 independent on the nature of the solvent as well as
on the type of atmosphere used.
The observed behaviour of aryl sulfonates 1–6 in polar and
protic media is fully compatible with the mechanism suggest-
ed in Scheme 2. Thus, excitation of the aromatics caused the
homolytic cleavage of the ArOÀS bond forming a phenoxyl/
sulfonyl radical pair (I/II, path a) that could undergo either an
out-of-cage hydrogen atom abstraction process (path b, char-
acterized by a kE rate constant) or an in-cage recombination
(path c, kR). The nonlinear fitting applied on the decay traces
As hinted above, two half lifetime values were measured
and appointed as tE and tR. The short lifetime (tE) was assigned
to the phenoxyl radical out-of-cage process while the longer
lifetime (tR) was assigned to the in-cage recombination process
of the radical pair. The bi-exponential behaviour can be de-
C
scribed according to Equation (1) where ArO represents the p-
C
substituted phenoxyl radical and RSO2 represents the sulfonyl
radical.
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6318
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
Table 3. Spectroscopic data (tE, tR and labs), out-of-cage escape (kE) and in-cage coupling (kR) rate constants of p-substituted phenoxyl radicals.[a]
Substrates
Solvent,
conditions
tE
tR
labs
kE
[sÀ1
kR
[mÀ1 sÀ1
[ms]
[ms]
[nm]
]
]
MeCN
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
3.5
3.8
3.1
3.6
3.9
3.1
4.4
4.2
1.5
3.7
2.9
3.0
29.7
27.9
32.4
21.0
37.4
24.2
25.0
21.4
31.4
35.4
37.4
29.1
410
410
405
405
405
405
405
405
405
405
405
405
2.9”105
2.6”105
3.2”105
2.8”105
2.6”105
3.2”105
2.3”105
2.4”105
6.6”105
2.7”105
3.4”105
3.4”105
5.4”109
1.1”1010
7.6”109
1.1”1010
5.2”109
1.5”1010
8.6”109
1.2”1010
4.1”109
6.2”109
6.9”109
5.2”109
MeOH
MeCN
MeOH
MeCN
MeOH
MeCN
MeOH
MeCN
MeOH
MeCN
MeOH
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
3.8
4.1
3.5
3.6
2.8
9.6
1.5
1.8
2.8
8.1
5.0
8.0
33.0
30.3
35.3
22.0
40.9
53.0
25.4
23.3
42.7
53.8
63.2
45.4
410
410
410
410
340; 500
340; 500
350; 520
340; 500
440
2.6”105
2.4”105
2.9”105
2.8”105
3.6”105
1.0”105
6.7”105
5.6”105
3.5”105
1.2”105
2.0”105
1.2”105
9.1”109
1.6”1010
7.4”109
3.1”1010
9.6”109
2.2”1010
3.6”109
1.2”1010
1.4”1010
7.6”109
1.0”1010
1.2”1010
440
440
440
[a] Values measured by laser flash photolysis (lexc =266 nm) in MeCN and MeOH; Concentration of substrates=5.0”10À4 m.
Figure 4. Transient absorption spectra obtained after a laser pulse (lexc =266 nm) of a solution of (a) 4 and (b) 5 in nitrogen saturated MeCN. Transient ab-
sorption spectra obtained for compound 6 in (c) nitrogen- and (d) oxygen-saturated MeCN.
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6319
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
p-N,N-Dimethylaminophenyl sulfonates 7 and 8
The photochemical behaviour observed for p-N.N-dimethylami-
nophenyl tosylate 7 deserves a separate comment. Figure 5a
shows the time-resolved spectrum of 7 in MeCN under N2 at-
mosphere in a time scale of 400 ms. Interestingly, the absorp-
tion profile comparable to those recorded for previously exam-
ined aryl sulfonates evolved quantitatively into a persistent,
oxygen-insensitive signal with maxima located at ca. 470 and
510 nm (see Figure 5b and Figures S33 and S34). An analogous
behaviour was pointed out by our group when analysing p-
methoxyphenyl tosylate ester.[15] On the other hand, the LFP
analyses of p-N,N-dimethylaminophenyl mesylate 8 led to a
very different photochemical scenario. Indeed, the time-re-
solved spectra showed, along with the initial formation of a
triplet state, a broad absorption band in N2-saturated media
(lMAX =465 nm), as illustrated in Figure 5c and Figures S35 and
S36. The intensity of both signals was strongly lowered by the
presence of oxygen (see Figure 5d). The latter transient ab-
The out-of-cage escape rate is a unimolecular pathway
showing a mono-exponential decay, and the corresponding kE
values were easily calculated from the reciprocal of the life-
time, kE =1/tE. On the other hand, the in-cage coupling rate
constants (kR) were obtained by plotting the reciprocal of the
concentration of the p-substituted phenoxyl radicals versus
time (see Section 4 in Supporting Information). The kR values
were thus obtained from the slopes of the linear regressions. It
is apparent from the data collected in Table 3 that kE values
are quite similar in MeCN and in MeOH at a fixed atmosphere
(in the 2.0–6.6”105 sÀ1 range) but a dependence on the reac-
tion atmosphere was observed. Indeed, the kE values measured
under oxygen atmosphere are systematically slightly lower
than those measured under nitrogen. This observation is com-
C
patible with the reaction of sulfonyl radical RSO2 with molecu-
lar oxygen to afford sulfonic acid (RSO3H, path d) via the perox-
osulfonyl radical III, as already observed in previous stud-
ies.[15,21]
sorption spectrum and the kinetic profile are comparable to
+
C
that of the radical cation 8 generated by irradiation of a solu-
The kE values measured for compounds 1–3, were little af-
fected by the nature of the R group of the sulfonyl moiety (R-
SO2, see Table 3). On the other hand, the second-order rate
constant (kR) of the bimolecular in-cage recombination of p-
substituted phenoxyl and sulfonyl radicals was in all cases near
the diffusion-controlled limit (109–1010 mÀ1 sÀ1) both under N2-
and O2-saturated conditions independently on the substituents
present on the aromatic ring.
tion of 8 with a laser pulse in the presence of persulfate anion
(see Figure S45).[22]
On the basis of the obtained results and on the available
data, we postulated a photoreactivity dependent on the
nature of the sulfonyl group. Thus, tosylate 7 underwent ho-
molysis of the ArOÀS bond (Scheme 3, path a) to afford the
corresponding phenoxyl/p-toluensulfonyl radical pair (I/II). Re-
combination of the radicals results in the exclusive formation
Figure 5. (a) Transient absorption spectra obtained after a laser pulse (400 ms; lexc =266 nm) of tosylate 7 (5.1”10À4 m) in N2-saturated MeCN solution.
(b) Transient decay at 500 nm after the laser pulse (lexc =266 nm) in MeCN solution (5.1”10À4 m). (c) Transient absorption spectra of p-N,N-dimethylamino-
phenyl mesylate 8 in nitrogen-saturated MeCN (d) Transient decay at 465 nm in nitrogen- and oxygen saturated MeCN at 465 nm.
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6320
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
phosphate (10) are depicted in Figures 6a,b (see also Figures
S39 and S40 for further details). Thus, Figure 6a displays the
transient absorption spectrum of 10 in N2-saturated MeCN so-
lution, whose maximum absorption wavelength is centred at
440 nm. Interestingly, no signals were observed during the
analysis of p-cyanophenyl phosphate 12.
As in the case of sulfonate 8, the decay traces show a
mono-exponential behaviour and after applying a nonlinear re-
gression fitting the half-lifetime values were easily obtained
and are collected in Table 4. It is apparent from the table that
the lifetime values depended on the atmosphere used in each
solvent, being the value measured under oxygen atmosphere
lower than those measured under nitrogen. The transient ab-
sorption spectra recorded from 9–11 have the kinetic profile
superimposable to that of the corresponding radical cations 9–
+
C
11 generated by irradiating a solution of 9–11 with a laser
pulse in the presence of persulfate anion (Figures S46–S48).
The photochemistry of aryl phosphates bearing electron-do-
nating groups was previously investigated by our research
group[12a,21] and the feasibility of the heterolytic cleavage of
the ArÀOP bond from the triplet state has been proved.
Scheme 4 illustrates the fragmentation pathway that provides
the triplet aryl cation transients VII and the subsequent mono-
Scheme 3. Transient species formed in polar media from compounds 7–8
after the laser pulse (lexc =266 nm).
of the Fries rearranged product V via the long lived cyclohexa-
dienone intermediate IV (paths b,c).[15] On the other hand, irra-
diation of mesylate 8 induced a heterolysis of the Ar-OS bond
from the triplet state, leading to the triplet aryl cation VI
(paths d,e). As previously observed in the case of 4-chloroani-
sole,[22] this intermediate may react with 8 in a bimolecular
electron transfer process (path f) to give 8·+ and the aryl radi-
cal VII (prone to abstract a hydrogen atom from the solvent).
3
Oxygen quenching of 8 hindered the formation of the corre-
+
C
sponding radical cation 8
.
Laser flash photolysis studies of phosphates 9–12 in polar
solvents
The LFP analysis of phosphate esters 9–12 (lexc =266 nm)
pointed out the presence of a broad absorption profile in both
solvents (see Figure 6 and Figures S37–S41). Some representa-
tive transient absorption spectra of p-methoxyphenyl diethyl-
Scheme 4. Transient species formed in polar media from compounds 9–12
after the laser pulse (lexc =266 nm).
Figure 6. Transient absorption spectra obtained after a laser pulse (100 ms; lexc =266 nm) of 5.1”10À4 m MeOH solution of p-methoxyphenyl diethylphosphate
(10) under N2 atmosphere; (b) Decay traces of 10 (5.1”10À4 m in MeOH) at 440 nm after the laser pulse (lexc =266 nm).
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6321
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
Table 4. Spectroscopic data (t and labs) and decay (k) rate constants for aryl phosphates 9–12.[a]
Substrates
Solvent
Atmosphere
t
k
[sÀ1
labs
[ms]
]
[nm]
MeCN
MeOH
MeCN
MeOH
MeCN
MeOH
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
N2
O2
N2 or O2
N2 or O2
137.0
58.8
135.1
55.6
20.0
15.6
22.2
11.0
13.0
7.7
7.3”103
1.7”104
7.4”103
1.8”104
5.0”104
6.4”104
4.5”104
1.1”105
7.7”104
1.3”105
1.8”105
1.8”105
–
465
465
465
465
440
440
440
435
450
450
450
450
–
5.7
5.7
[b]
MeCN
MeOH
–
[b]
–
–
–
[a] Value measured by laser flash photolysis (lexc =266 nm) in MeCN and MeOH; Concentration of substrates: 5.0”10À4 m. [b] No signals detected.
electronic oxidation induced on the starting phosphate to
afford the corresponding radical cations VIII. No photochemi-
cal homolytic cleavage of the ArOÀP bond took place in com-
pounds 9–12 since phenoxyl radical IX has been not detected
during the LFP experiments. A particular case is represented
by the electron-poor p-cyanophenyl diethylphosphate 12, that
is photostable under all the conditions examined.
ing electron-withdrawing group as well as moderately elec-
tron-donating groups (compounds 1–6, path d). The cleavage
generates a phenoxyl/sulfonyl radical couple as demonstrated
by the detection of the former radical by laser flash photolysis
(path d, see also Table 3) and it is independent on the type of
sulfonate used (mesylate, triflate or tosylate). The measure-
ment of two lifetime values after the cleavage points to a radi-
cal recombination (path e) in competition with a radical
escape (path f). The presence of a strong electron-donating
group (e.g. NMe2), however, led to a leaving group dependent
reactivity. Photolysis of N,N-diethylamino tosylate 7 led again
to a radical couple but the electron-rich capability of the aro-
matic ring was successful in trapping the electrophilic sulfonyl
group and a Fries adduct was readily formed via the long lived
cyclohexadienone intermediate (path g, see also Figure 5).
Quite surprisingly, shifting from a tosylate to a mesylate
group in compound 8, changes dramatically the photochemis-
try occurring in a heterolysis of the ArÀOX bond (Scheme 5,
The overall photochemical behaviour of compounds 1–12
has been summarized in Scheme 5. Irradiation of ArOX caused
the initial population of the singlet state (path a). Com-
pounds 1–12 exhibited a very low radiative behaviour from
the singlet witnessed by their low fluorescence quantum yield
(Table 1, Figures S1–S9). This is in part due to the efficient ISC
to give the corresponding triplets 31-12 well detectable in non-
polar solvents (path b) and, in some cases, even in polar sol-
vents albeit the increase of polarity caused a lowering of the
triplet lifetimes (see Table 2). Thus, in cyclohexane, com-
3
pounds 1–12 are rather photostable and deactivation of 1-12
is a favoured process (path c). In polar solvents, however, a
photoreactivity begins to play. Homolytic fragmentation in aryl
esters from the exited singlets is apparent in sulfonates bear-
path h). This is safely demonstrated by the presence of ArÀ
+
C
OX formed via an electron transfer reaction between the aryl
cation and the starting mesylate (path i). The nature of the ArÀ
Scheme 5. Pathways in the photochemistry of esters 1–12.
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6322
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005426
Chemistry—A European Journal
+
[5] a) Q.-Y. Chen, Z.-T. Li, J. Org. Chem. 1993, 58, 2599–2604; b) E. Torti, S.
Protti, S. Bollanti, F. Flora, A. Torre, G. Brusatin, A. Gerardino, L. Businaro,
M. Fagnoni, G. Della Giustina, L. Mezi, ChemPhotoChem 2018, 2, 425–
432.
C
OX was confirmed by persulfate oxidation of ArÀOX thus ex-
cluding the involvement of other radical cation species (e.g. ArÀ
+
C
H
see Figures S49–S51), which can derive by the hydrogen
atom abstraction of the aryl cation from the solvent.[10] The pho-
toheterolytic pathway in aryl sulfonates, however, can be pro-
moted by triplet sensitization as previously demonstrated.[12]
The homolytic ArOÀP(O)(OR)2 bond cleavage did not take
place in phosphates 9–12 probably due to the high BDE
energy of the OÀP bond.[23]A photoheterolytic cleavage may
be however possible providing that the aromatic ring bears
electron-donating groups to stabilize the positive charge in
3Ar+ (e.g. in compounds 9–11). No available monomolecular
pathways are available in the electron-poor aromatic phos-
phate 12 that did not undergo any detectable photoreactivity.
[6] P. M. Fidelibus, G. F. Silbestri, M. T. Lockhart, S. D. Mandolesi, A. B.
Chopa, J. C. Podestà, Appl. Organomet. Chem. 2007, 21, 682–687.
[7] ”Photonucleophilic Substitution Reactions”: M. Fagnoni, A. Albini, Or-
ganic Photochemistry and Photophysics, Molecular and Supramolecular
Photochemistry (Eds.: V. Ramamurthy, K. Schanze), Dekker, New York,
2006, pp. 131–177.
[8] a) E. Havinga, R. O. de Jongh, W. Dorst, Recl. Trav. Chim. Pays-Bas 1956,
75, 378–383; b) R. O. de Jongh, E. Havinga, Recl. Trav. Chim. Pays-Bas
1968, 87, 1318–1326; c) C. Karapire, S. Icli, Handbook of Organic Photo-
chemistry and Photobiology, 2nd ed. (Eds.: W. H. Horspool, F. Lenci), CRC,
Orlando, 2004, Section 37.
[9] M. I. Quindt, S. M. Bonesi, in Photochemistry, Vol. 48, The Royal Society
of Chemistry, London, 2021, pp. 261–292.
[10] a) S. Lazzaroni, D. Ravelli, S. Protti, M. Fagnoni, A. Albini, C. R. Chimie
2017, 20, 261–271; b) V. Dichiarante, S. Protti, M. Fagnoni, J. Photochem.
Photobiol. A Chem. 2017, 339, 103–113.
[11] a) H.-I. Joschek, L. I. Grossweiner, J. Am. Chem. Soc. 1966, 88, 3261–
3268; b) G. Grabner, G. Kbhier, J. Zechner, N. Getoff, Photochem. Photo-
biol. 1977, 26, 449–458; c) J. Den Heijer, O. B. Shadid, J. Cornelisse, E.
Havinga, Tetrahedron 1977, 33, 779–786; d) G. Kçhler, N. Getoff, J.
Chem. Soc. Faraday Trans. I 1978, 74, 1029–1035; e) G. Kçhler, N. Getoff,
J. Chem. Soc. Faraday Trans. I 1980, 76, 1576–1584; f) J. Zechner, G. Kijh-
ler, N. Getoff, I. Tatischeff, R. Klein, Photochem. Photobiol. 1981, 34, 163–
168.
[12] a) M. De Carolis, S. Protti, M. Fagnoni, A. Albini, Angew. Chem. Int. Ed.
2005, 44, 1232–1236; Angew. Chem. 2005, 117, 1258–1262; b) S. Protti,
M. Fagnoni, A. Albini, Angew. Chem. Int. Ed. 2005, 44, 5675–5678;
Angew. Chem. 2005, 117, 5821–5824; c) V. Dichiarante, M. Fagnoni, A.
Albini, Chem. Commun. 2006, 3001–3003; d) S. Protti, M. Fagnoni, A.
Albini, J. Am. Chem. Soc. 2006, 128, 10670–10671; e) V. Dichiarante, M.
Fagnoni, A. Albini, Angew. Chem. Int. Ed. 2007, 46, 6495–6498; Angew.
Chem. 2007, 119, 6615–6618; f) S. Lazzaroni, S. Protti, M. Fagnoni, A.
Albini, Org. Lett. 2009, 11, 349–352; g) V. Dichiarante, D. Dondi, S.
Protti, M. Fagnoni, A. Albini, J. Am. Chem. Soc. 2007, 129, 5605–5611.
[13] C. Raviola, V. Canevari, S. Protti, A. Albini, M. Fagnoni, Green Chem.
2013, 15, 2704–2708.
Conclusions
In summary, this work demonstrates that the photoreactivity
of aryl esters depends on several factors such as the polarity of
the medium, the electron-donor/acceptor capability of the
substituents tethered to the aromatic ring as well as the
nature of the leaving group. These factors may be taken into
account in view on the possible use of aryl esters in metal-free
arylations and in the design of nonionic PAGs. The formation
of the phenoxyl/sulfonyl radical couple or of an aryl cation
may be however tuned at will by choosing the suitable combi-
nation of the substituents present on the aromatic ring.
Acknowledgements
S.M.B. is a research member of CONICET, Argentina. S.P. and
M.F. thanks the Universitiamo crowfunding project (University
of Pavia) for partial support.
[14] H. Qrareya, S. Protti, M. Fagnoni, J. Org. Chem. 2014, 79, 11527–11533.
[15] E. Torti, G. Della Giustina, S. Protti, D. Merli, G. Brusatin, M. Fagnoni, RSC
Adv. 2015, 5, 33239–33248.
[16] S. Protti, D. Dondi, M. Fagnoni, A. Albini, Green Chem. 2009, 11, 239–
249.
[17] a) C. J. Martin, G. Rapenne, T. Nakashima, T. Kawai, J. Photochem. Photo-
biol. C 2018, 34, 41–51; b) N. A. Kuznetsova, G. V. Malkov, B. G. Gribov,
Russ. Chem. Rev. 2020, 89, 173–190.
Conflict of interest
The authors declare no conflict of interest.
[18] S. Crespi, D. Ravelli, S. Protti, A. Albini, M. Fagnoni, Chem. Eur. J. 2014,
20, 17572–17578.
Keywords: aryl
photoheterolysis
spectroscopy
phosphates
·
aryl
sulfonates
·
[19] J. W. Bridges, P. J. Creaven, R. T. Williams, Biochem. J. 1965, 96, 872–878.
[20] a) P. K. Das, M. V. Encinas, S. Steenken, J. C. Scaiano, J. Am. Chem. Soc.
1981, 103, 4162–4166; b) E. J. Land, G. Porter, E. Strachan, Trans. Fara-
day Soc. 1961, 57, 1885–1893; c) E. J. Land, G. Porter, Trans. Faraday
Soc. 1963, 59, 2016–2026; d) E. J. Land, M. Ebert, Trans. Faraday Soc.
1967, 63, 1181–1190; e) Z. B. Alfassi, R. H. Schuler, J. Phys. Chem. 1985,
89, 3359–3363; f) M. R. Ganapathi, R. Hermann, S. Naumov, O. Brede,
Phys. Chem. Chem. Phys. 2000, 2, 4947–4955; g) G. Siano, S. Crespi,
S. M. Bonesi, J. Org. Chem. 2020, 85, 14012–14025.
[21] M. Terpolilli, D. Merli, S. Protti, V. Dichiarante, M. Fagnoni, A. Albini, Pho-
tochem. Photobiol. Sci. 2011, 10, 123–127.
[22] I. Manet, S. Monti, G. Grabner, S. Protti, D. Dondi, V. Dichiarante, M. Fag-
noni, A. Albini, Chem. Eur. J. 2008, 14, 1029–1039.
·
photohomolysis
· time-resolved
[1] For selected reviews, see: a) B. M. Rosen, K. W. Quasdorf, D. A. Wilson, N.
Zhang, A. M. Resmerita, N. K. Garg, V. Percec, Chem. Rev. 2011, 111,
1346–1416; b) B. J. Li, D. G. Yu, C. L. Sun, Z. J. Shi, Chem. Eur. J. 2011,
17, 1728–1759; c) C. M. So, F. Y. Kwong, Chem. Soc. Rev. 2011, 40, 4963–
4972; d) S. Chassaing, S. Specklin, J.-M. Weibel, P. Pale, Curr. Org. Synth.
2012, 9, 806–827; e) W. K. Chow, O. Y. Yuen, P. Y. Choy, C. M. So, C. P.
Lau, W. T. Wong, F. Y. Kwong, RSC Adv. 2013, 3, 12518–12539.
[2] For selected reviews, see: a) S. Protti, M. Fagnoni, Chem. Commun.
2008, 3611–3621; b) H. Chen, Z. Huang, X. Hu, G. Tang, P. Xu, Y. Zhao,
C.-H. Cheng, J. Org. Chem. 2011, 76, 2338–2344; c) F.-S. Han, Chem. Soc.
Rev. 2013, 42, 5270–5298.
[23] K. Hemelsoet, F. Van Durme, V. Van Speybroeck, M.-F. Reyniers, M. Waro-
quier, J. Phys. Chem. A 2010, 114, 2864–2873.
[3] ”The photochemistry of phosphate and sulfonate esters”:D. Ravelli, M.
Fagnoni, CRC Handbook of Organic Photochemistry and Photobiology
(Eds.: A Griesbeck, M. Oelgemoeller, F. Ghetti), 3rd ed., CRC, Boca Raton,
2012, pp. 393–417.
[4] A. Albini, M. Fagnoni, Photochemically-Generated Intermediates in Syn-
thesis, Wiley, Hoboken 2013.
Manuscript received: December 21, 2020
Accepted manuscript online: January 22, 2021
Version of record online: March 5, 2021
Chem. Eur. J. 2021, 27, 6315 –6323
www.chemeurj.org
6323
ꢀ 2021 Wiley-VCH GmbH