Probing for a leaving group effect on the generation and reactivity of phenyl cations

Article abstract of DOI:10.1021/jo300290v

Phenyl cations are smoothly generated by the photoheterolytic cleavage of an Ar-LG bond (LG = leaving group). With the aim of evaluating the scope of the method, a series of 4-methoxy-2-(trimethylsilyl)phenyl derivatives (sulfonic, LG = MeSO₃ and CF₃SO₃, phosphate, LG = (EtO) ₂(O)PO esters and the corresponding chloride) have been compared as probes for evaluating the leaving group ability. The photocleavage was a general reaction, with the somewhat surprising order (EtO)₂(O)PO ～ Cl > CF₃SO₃ > MeSO₃ (φ = 0.50 to 0.16 in CF₃CH₂OH and lower values in MeCN-H₂O). The ensuing reactions did not depend on the LGs but only on the structure of the phenyl cation (the silyl group tuned the triplet to singlet intersystem crossing and the electrophilicity) and on the medium (formation of a complex with water slowed the electrophilic reactions).

Full text of DOI:10.1021/jo300290v

Article
pubs.acs.org/joc
Probing for a Leaving Group Effect on the Generation and Reactivity
of Phenyl Cations
Erika Abitelli, Stefano Protti, Maurizio Fagnoni,* and Angelo Albini
PhotoGreenLab, Department of Chemistry, University of Pavia, Viale Taramelli 12, 27100 Pavia, Italy
S
* Supporting Information
ABSTRACT: Phenyl cations are smoothly generated by the photoheterolytic
cleavage of an Ar−LG bond (LG = leaving group). With the aim of evaluating the
scope of the method, a series of 4-methoxy-2-(trimethylsilyl)phenyl derivatives
(sulfonic, LG = MeSO3 and CF3SO3, phosphate, LG = (EtO)2(O)PO esters and
the corresponding chloride) have been compared as probes for evaluating the
leaving group ability. The photocleavage was a general reaction, with the
somewhat surprising order (EtO)2(O)PO ∼ Cl > CF3SO3 > MeSO3 (Φ = 0.50 to
0.16 in CF3CH2OH and lower values in MeCN−H2O). The ensuing reactions did
not depend on the LGs but only on the structure of the phenyl cation (the silyl group tuned the triplet to singlet intersystem
crossing and the electrophilicity) and on the medium (formation of a complex with water slowed the electrophilic reactions).
phenyl cation. Phenyl cations11,16 are highly electrophilic
INTRODUCTION
Modern arylation methods are generally based on a metal-
■
species17,18 that exist in two spin configurations, viz. the singlet
(1Ar+ having a π6σ0 structure)17 or the triplet (3Ar+ having a
mediated activation of an Ar−X bond,1 most often in aryl
π5σ1 structure). These two intermediates show a different
iodides, bromides, and sulfonates.1,2 Esters (sulfonates,
phosphates) are convenient intermediates for the conversion
of the phenol function into different groups. Actually, the
reactivity of aryl triflates toward the oxidative insertion of
chemistry, namely, solvolysis for the singlets and addition onto
π bond nucleophiles (Nu-H) or homolytic hydrogen
abstraction from the solvent for the triplets (Scheme 1).18
transition metal complexes rivals that of the widely used
halides. Triflates, however, are both base-sensitive and
Scheme 1. Photogeneration and Reactivity of Phenyl Cations
thermally labile, and alternative sulfonate leaving groups such
as mesylates, nonaflates,3 and tosylates4 have been recently
adopted. In particular, mesylates are significantly less expensive
and a better choice from the point of view of atom economy.
Both Ni-catalyzed5 and, though less frequently, Pd-catalyzed6
cross-coupling reactions have been reported. In contrast, the
rarely employed aryl phosphates react sluggishly under metal
catalysis7 only with aggressive nucleophiles, such as Grignard or
organoaluminum reagents, besides the recently reported
reaction with boranes under Suzuki−Miyaura conditions.8
Skrydstrup and co-workers proposed a reactivity scale for
the Heck reaction, where aryl triflates lay between aryl iodides
and bromides, while aryl phosphates lay far below aryl chlorides
(I > OTf ≥ Br > Cl > OP(O)(OPh)2).9 In the same way,
The relative stability of the two spin states depends on the
ring substituents FG, with the singlet, in most cases, being the
the leaving ability of sulfonates was found to be OTf >
most stable species.17,18 The multiplicity of the initially photo-
benzenesulfonate > OTs > OMs, with mesylates as the least
generated phenyl cation Ar+ is that of the excited state (II)
where the cleavage of the Ar−LG bond takes place. With
electron-rich halides and esters, phenyl cations are generated
in the triplet state, and chemoselective arylations have been
developed.11 Indeed, phenyl cations have been efficiently
generated by detachment of a wide variety of LGs with a
marked difference in nucleofugal ability.
active leaving group (LG),10 a result that could be roughly
related to the pKa value of the conjugate acid HX.
In past years, an alternative activation of Ar−Cl and Ar−O
bonds has been devised by photochemical means and involves
the photoheterolytic cleavage of such bonds.11 Thus, the
irradiation in polar solvents of aryl chlorides, sulfonates
(triflates and mesylates), and diethyl phosphates substituted
with strong (e.g., NMe2, OMe,12 OH13) or mild (e.g., t-Bu14 or
n-alkyl15) electron-donating groups caused the heterolytic
cleavage of the Ar−LG bond and ultimately gave a triplet
Received: February 9, 2012
Published: March 8, 2012
© 2012 American Chemical Society
3501
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
It would be useful to establish how the leaving group affects
both the efficiency in phenyl cation generation and the ensuing
chemoselectivity of this intermediate and whether the
nucleofugal ability of the LGs is related to the pKa of the
conjugate acid liberated, as in thermal reactions. Sparse data
support such a relationship also upon irradiation in the case of
7-methoxycoumarin-4-yl-methyl carboxylate, sulfonate, and
phosphate esters.19 Moreover, in the related photogeneration
of vinyl cations, it has been reported that leaving group abilities,
at least in the halogen series, were related to the acidity of the
conjugate acids HX,20 and in the photogeneration of a
naphthylmethyl cation, the photofugacity order for the ionic
process showed a preference for the chloride anion with respect
to the diethyl phosphate anion.21
Chart 1. Preparation of Silylated Methoxyphenyl Esters
1a−c
a
a
Reagents and conditions: (a) BuLi, Me3SICl, −78 → 25 °C, then H+,
76%; (b) CH3SO2Cl, Et3N, DCM, 72%; (c) Tf2O, Py, 0 → 50 °C,
88%; (d) NaH, (EtO)2POCl, THF, 54%.
We reported, however, that the efficiency in the photo-
generation of the 4-methoxyphenyl cation depends both on
the nature of LG and on the reaction medium used and did
not necessarily follow the acidity of the HX. In fact, in 2,2,2-
trifluoroethanol (TFE), the Φ−1 value was lower for
4-chloroanisole (0.0722) than for the corresponding 4-methox-
yphenyl diethyl phosphate (0.2823), and in a nonprotic solvent
such as MeCN, both Φ−1 markedly decreased.22,23 Moreover,
sulfonate and phosphate esters of p-methoxyphenol were found
to be particularly suitable for arylation reactions and exhibited a
Φ−1 comparable to that of 4-chloroanisole24 (see below). This
may be related to the formation of either an ion couple (Ar+X−)
or a free Ar+ cation, a fact that may also affect the course of the
ensuing arylation reactions.
Table 1. Fluorescence Quantum Yield (ΦF) and τF of
Methoxyphenyl Derivatives 1a−d
a
CF3CH2OH
CH3OH
ΦF ×
λmax
(nm)
Ar-SiMe3
102
ΦF × 102
λmax (nm)
τF (ns)
b
1a
1b
1c
1d
3.8
0.29
19
310
313
307
310
4.6 (5.2)
0.32 (<0.1)
18 (4.2)
310 (305)
b
0.28
b
312 (311)
311
0.92 (0.94)
c
c
0.7
1.3 (1.9)
<0.2 (1.6)
a
b
In parentheses are the data of the silicon-free analogues. Too weak
c
to be measured. See ref 23.
The irradiations (up to >80% conversion) were carried out
both in neat solvents and in the presence of π bond
nucleophiles, viz. either an aromatic compound (benzene) or
an alkene (allyltrimethylsilane, ATMS; see Table 2, Schemes 2
and 3). This was done under acetone (0.9 M) sensitization in
order to prevent any undesirable homolytic cleavage of the
O−S bond known to occur from the singlet state14 and to
ensure an optimized absorption at the wavelength used. An
equivalent amount of a base (e.g., Cs2CO3) was added in the
experiments in TFE in order to buffer the acidity liberated in
the reaction14 (no base required in water−acetonitrile).
Irradiations at shorter times (not reported) were also carried
out in order to check for the occurrence of secondary processes,
but the product distribution was found to change little with the
progress of the reaction.
Photochemistry. Irradiation (7−16 h) of a 0.05 M solution
of 4-methoxy-2-trimethylsilylphenyl methanesulfonate (1a,
entries 1−6) in each of the solvents investigated caused
extensive conversion (88−100%; see Table 2). In TFE in the
presence of Cs2CO3, three products were formed, viz. anisole 2
(24%), the trifluoroethyl ether 3 (56%), along with a small
amount of methoxy phenol 4 (5%; see Scheme 2). In the
presence of benzene (1 M), photosubstitution remained the
main path but some biphenyl 5 (19%) was obtained at the
expense of both 2 and 3 (Scheme 3). With allyltrimethysilane
(ATMS, 0.5 M), photoallylation to give 6 was efficient (67%
yield) though again accompanied by the formation of 2 and 3.
When the solvent was changed to a water−acetonitrile (1:5)
mixture, acetanilide 7 (72% yield) became the main product
(along with 2 and 4). The product distribution was marginally
affected by 1 M benzene, although traces of 5 have been
detected (Scheme 3). The allylation was again efficient with
ATMS, even if the yield of allylated 6 was lower than that
obtained in TFE. In the last case, however, an excellent mass
balance (98%) was reached.
Some years ago, we demonstrated that both the efficiency of
the Ar−Cl cleavage and the chemoselectivity of the photo-
generated phenyl cation can be tuned by inserting one (or two)
trimethylsilyl group(s) (TMS) in ortho-position with respect to
the leaving group LG.25a This appeared to be a case of a β-silyl
stabilization and made intersystem crossing (ISC) from the
11a,25
1
triplet to the more stable singlet Ar+ viable (Scheme 1).
We now reasoned that such silylated methoxyphenyl cations
could be used as probes to give insight on the role of the
leaving group in the photogeneration of phenyl cations. In fact,
3
ISC to the singlet competes with the typical reactions of Ar+,
viz. arylation and reduction, and the product distribution
obtained from the aryl cation precursors having different
leaving groups could be informative. Toward this aim, the
photoreactivity of sulfonate and phosphate esters of 3-methoxy-
2-trimethysilylphenol 1a−c and of aryl chloride 1d has been
investigated and the results are illustrated below. The com-
pounds were chosen in order to ensure a marked difference on
the pKa of the conjugate acid liberated (ca. 15 pKa units from
the triflic acid to the diethyl phosphoric acid; see below).
RESULTS
■
Silylated esters 1a−c were easily prepared in a few steps starting
from 2-bromo-4-methoxyphenol, as illustrated in Chart 1. For
the sake of comparison, the experiments were compared to
those with 4-chloro-3-trimethylsilylanisole 1d.
Photophysical Data of Compounds 1a−d. Esters 1a−c
exhibited a weak to moderate fluorescence, with triflate 1b
being less emitting (ΦF < 0.05) than chloride 1d and phosphate
1c much more (ΦF ∼ 0.2), which is rather independent from
the alcohol used as the solvent (Table 1).
Preparative Irradiation. Protic solvents (TFE and water/
acetonitrile 1:5) were chosen for the irradiation (λirr centered
at 310 nm) of silyl esters 1a−c (0.05 M), as these had be
shown to be convenient with the silicon-free analogues.12
3502
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
a
Table 2. Photolysis of 1a−d (0.05 M) in the Presence of Various π Bond Nucleophiles
b
products (yield %)
entry
Ar-SiMe3
solvent
nucleophile
tirr, h (conversion, %)
RD
SL
AR
DP
c
1
2
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
1b
1c
1c
1c
1c
1c
1c
1d
1d
1d
1d
1d
1d
TFE
TFE
TFE
none
12 (100)
7 (100)
7 (95)
2 (24)
2 (19)
2 (3)
3 (56)
3 (42)
3 (20)
7 (72)
7 (54)
7 (22)
3 (44)
3 (42)
3 (26)
7 (47)
7 (22)
7 (38)
3 (63)
3 (47)
3 (31)
7 (40)
7 (56)
7 (52)
3 (58)
3 (46)
3 (43)
7 (73)
7 (79)
7 (36)
4 (5)
c
c
C6H6 1 M
ATMS 0.5 M
none
5 (19)
6 (67)
3
4
H2O/MeCN 1:5
H2O/MeCN 1:5
H2O/MeCN 1:5
16 (92)
16 (96)
12 (88)
7 (91)
2 (6)
4 (13)
4 (16)
4 (22)
5
C6H6 1 M
ATMS 0.5 M
none
2 (4)
5 (2)
6
2 (4)
6 (50)
c
7
TFE
2 (22)
2 (4)
c
8
TFE
C6H6 1 M
ATMS 0.5 M
none
12 (94)
12 (88)
24 (85)
24 (91)
24 (92)
7 (95)
5 (21)
6 (50)
c
9
TFE
2 (6)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
H2O/MeCN 1:5
H2O/MeCN 1:5
H2O/MeCN 1:5
2 (19)
2 (15)
2 (17)
2 (8)
C6H6 1 M
ATMS 0.5 M
none
5 (3)
6 (8)
c
TFE
c
TFE
C6H6 1 M
ATMS 0.5 M
none
7 (93)
2 (6)
5 (27)
6 (60)
c
TFE
12 (86)
24 (80)
24 (85)
24 (82)
6 (100)
6 (100)
6 (100)
25 (92)
25 (91)
25 (87)
2 (3)
H2O/MeCN 1:5
H2O/MeCN 1:5
H2O/MeCN 1:5
2 (9)
C6H6 1 M
ATMS 0.5 M
none
2 (10)
2 (6)
5 (2)
6 (2)
c
TFE
2 (21)
2 (11)
2 (15)
2 (23)
2 (16)
2 (22)
c
TFE
C6H6 1 M
ATMS 0.5 M
none
5 (17)
6 (18)
c
TFE
H2O/MeCN 1:5
H2O/MeCN 1:5
C6H6 1 M
ATMS 0.5 M
5 (tr)
H2O/MeCN 1:5
6 (31)
a
b
Reaction sensitized by acetone (0.9 M). Yields were based on consumed aryl silane and determined by GC. According to the discussion section,
c
the observed pathways have been classified as follows: RD = Reduction, SL = Solvolysis, AR = Arylation, DP = Deprotection Cesium carbonate (0.025 M)
added.
Scheme 2. Product Distribution in the Irradiation of
Aromatics 1a−d in Neat Solvents
after 7 h irradiation in neat TFE, showing a product distribution
quite similar to 1a. Thus, silylated anisole 2 (22%) and trifluo-
roethyl ether 3 (44%) were exclusively formed (Table 2).
Differently from 1a, no photodeprotection was observed in all
of the photolysis experiments of triflate 1b. Arylation products
have been likewise formed in variable amounts when either
benzene or ATMS were added to the reaction mixture (21 and
50% yields, respectively) though solvolysis remained the main
process. When using aqueous acetonitrile as the solvent, the
Ar−C bond formation was significantly diminished (5 and 6
formed in less than 10% amount) and compound 7 was again
by far the main product. Notice that no desilylation took place
in all of the experiments performed.
Photolysis of 4-methoxy-2-trimethylsilylphenyl diethyl phos-
phate (1c, entries 13−18) in TFE (Table 2) caused an efficient
consumption of the substrate (<12 h irradiation time), and
photosubstitution and photoreduction were the preferred
processes. The mass balance (>80%) was high in the experi-
ments in the presence of the nucleophiles, where the yield of
biphenyl 5 (27%) was much lower than that of allylated 6
(60%). The arylation strongly decreased in MeCN/water
mixture, where formation of solvolytic amide 7 was favored.
4-Chloro-3-trimethylsilylanisole (1d, entries 19−24) was
completely consumed after 6 h irradiation in TFE and almost
completely (>85%) in 25 h in aqueous acetonitrile (Table 2).
In neat solvents, photosubstitution was by far the main process
observed. Arylation yields were modest, particularly when using
benzene as the nucleophile, and contrary to the previous cases,
the yield of allylated 6 was larger in MeCN/water mixture
(31%) than in TFE (18%). Photosubstitution products were
Scheme 3. Product Distribution in the Irradiation of
Aromatics 1a−d in the Presence of π Bond Nucleophiles
4-Methoxy-2-trimethylsilylphenyl trifluoromethanesulfonate
(1b, 0.05 M, entries 7−12) was almost completely consumed
3503
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
consistently the main compounds in each experiment, and
reduction was also present in a 10−20% yield.
Quantum Yield Measurements. The decomposition
quantum yield (Φ−1) of esters 1a−c (5 × 10−3 M) was
measured by irradiation in TFE and H2O/MeCN (1:5) at 254
nm. The results are shown in Table 3 in comparison with
Tf (pKa = −14.9)10 > Cl (pKa = −7) > Ms (pKa = −1.9)10
>
phosphate (pKa = 0.71).26 Table 3 shows that this is not the
case and that the order depends on the reaction medium. The
phosphate is the most efficient leaving group in the series, and
noteworthy, there is no competing cleavage of the ArO−P
bond. The reactivity order is little affected in passing from
silicon-free (phosphate > Cl = Tf > Ms) to silicon-containing
anisole derivatives (phosphate ∼ Cl > Tf > Ms). The effective-
ness of the LGs level in mixed aqueous medium, where at any
rate the phosphate group is again the best one and the sulfonic
esters perform better than the chloride (phosphate > Ms =
Tf ∼ Cl). The mesylate group appeared to be the LG less
affected by the solvent since the Φ−1 value of 1a is the same in
the two protic media considered.
Table 3. Quantum Yield of Photodecomposition (Φ−1) of
Phenyl Trimethylsilyl Esters 1a−c and Chloride 1d in Protic
Media Measured at 254 nm
a
quantum yields (Φ−1
)
Ar-SiMe3
CF3CH2OH
H2O/MeCN (1:5)
1a
1b
1c
1d
0.16 (0.055)
0.3 (0.07)
0.11
0.11
0.24
As a matter of fact, TFE was confirmed as the elective solvent
for the phenyl cation generation and thus the best medium for
carrying out arylation reactions. The processes occurring were
reduction (RD, to give the dechlorinated analogue 2),
substitution (SL, yielding 3 or 7), and photodeprotection of
the ester group (DP; see compound 4). Arylation (AR) took
place in the presence of π nucleophiles, leading to
functionalized benzenes 5 and 6 (Scheme 3 and Table 2).
Under the present conditions (irradiation in protic solvents
under acetone sensitization), no secondary desilylation took
place as previously observed in related compounds.25a
c
0.50 (0.28)
b
d
b
d
0.43 (0.07)
0.068 (0.01)
a
In parentheses is the quantum yield of the silicon-free analogue.
From ref 25a. From ref 23. From ref 22.
b
c
d
silicon-free analogues and with 1d. The Φ−1 values ranged from
0.16 to 0.5 in TFE but did not exceed 0.24 in 1:5 H2O/MeCN.
The values in aqueous acetonitrile were consistently lower than
those in TFE. The order of photoreactivity in TFE is 1c > 1b >
1a, which is the same as that observed with silicon-free aryl
esters. The introduction of a silyl group made the photo-
heterolysis of the Ar−O bond more efficient and strongly
enhanced (up to a factor of 4 with 1b) the values of Φ−1. The
aqueous medium levels the photoreactivity of 1a and 1b.
To summarize, the effect caused by introducing an ester
group in the place of a chloro atom is not uniform since this
made the aromatic more reactive in 1:5 H2O/MeCN, whereas
in TFE, only phosphate 1c has a Φ−1 comparable with 1d. The
Φ−1 values have been likewise measured in TFE at 310 nm
under acetone-sensitized conditions (not shown), and com-
parable values to that measured at 254 nm were observed
except for the case of 1c where Φ−1 was halved.
The product distribution observed is consistent with a triplet
phenyl cation intermediate (3III; see Scheme 4). Hydrogen
Scheme 4. Photochemistry of Arylsilanes 1a−c
DISCUSSION
■
As previously demonstrated for the case of compound 1d,25 the
introduction of a trimethylsilyl group onto an aromatic
(whether chloroanisole or a sulfonate or phosphate ester of
4-methoxyphenol) did not significantly affect either the
fluorescence efficiency or the singlet lifetime (see Table 1).
Thus, one can assume that ISC in silylated compounds 1a−c is
comparable to that of the parent phenyl esters (ΦISC ≥ 0.7)12
and the Ar−O bond cleavage in esters 1a−c has to arise from
the triplet state. The formation of triplet phenyl cations from
these esters was feasible as it was for the desilylated
analogues.12 Photoheterolysis of the Ar−O bond took place
as the exclusive path, except for the case of mesylate 1a, where a
partial deprotection, probably arising from the singlet, induced
the cleavage of the ArO−S bond.14 Notice further that no
Photo-Fries products from the sulfonyl radical liberated in the
process have been detected. In analogy with 1d, the
photodecomposition quantum yield (Φ−1) in TFE was
consistently higher for trimethylsilyl-substituted esters 1a−c
in comparison with the silicon-free analogues, confirming that
the presence of a R3Si group favors the photocleavage of
substituted anisoles (Table 3).
abstraction from the solvent (to afford 2) or addition onto π
nucleophiles is a diagnostic reaction of these species. Computa-
tional investigations have previously demonstrated that,
because of the β-stabilizing effect of the SiMe3 group in these
cations, the singlet (1III) lies lower in energy and it is accessi-
ble via ISC.25 This unselective intermediate undergoes
photosolvolysis reactions and adds to the reaction medium
(Scheme 4).
The addition of the triplet cations onto π nucleophiles is
slowed by the silyl group both by electronic (β-stabilization)25
and steric effects. This made the addition more selective, with
consistently better results with ATMS than with benzene, in
agreement with the nucleophilicity order.27 Indeed, the
arylation of benzene was always dwarfed by solvolysis. Notice
further that the arylation of ATMS was more efficient with
If the photolability of the Ar−X bond follows the order
of the HX pKa values, the reactivity order would be
3504
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
silylated esters than with the corresponding chloride, exactly as
observed for the nonsilylated compounds.12,28 At any rate,
complexation of phenyl cations by a nucleophilic solvent such
as water18 suppressed arylation reactions (Table 2) in all but a
couple of cases (including the allylation of 1d).
Finally, we attempted to recognize whether the leaving anion
affected the chemistry of the photogenerated phenyl cations.
To this aim, we compared the product selectivity with different
LGs, as shown in Table 4. Perhaps surprisingly, no marked
aryl sulfonates found application in photolithography and
biology.31,32
EXPERIMENTAL SECTION
■
NMR spectra were recorded on a 300 MHz spectrometer. The
attributions were made on the basis of H and 13C NMR, as well as
1
DEPT-135 experiments; chemical shifts are reported in parts per
million downfield from Me4Si. The photochemical reactions were
performed by using nitrogen-purged solutions in quartz tubes and a
multilamp reactor fitted with six 15 W phosphor-coated lamps
(maximum of emission 310 nm) for the irradiation. The progress of
the reaction was followed by GC analyses. Photolyzed solutions have
been concentrated in vacuo and purified by chromatographic
separation (Millipore (60 Å, 35−70 μm) silica gel, eluant: cyclo-
hexane/ethyl acetate mixtures). Quantum yields were measured at
254 nm (4 Hg lamps, 15 W). Benzene and allyltrimethylsilane
(ATMS) were commercially available and freshly distilled before
use. 4-Chloro-3-trimethylsilylanisole (1d) was prepared starting from
3-bromoanisole.25a
Table 4. Comparison of the Product Distribution between
1d and Esters 1a−c upon Irradiation in Protic Media
reduction/
substitution
arylation/reduction + substitution
H2O/
H2O/
MeCN
1:5
(neat)
TFE
TFE
H2O/
MeCN 1:5;
TFE
C6H6 ATMS MeCN 1:5; ATMS
Ar-SiMe3 (neat)
1 M
0.5 M C6H6 1 M
0.5 M
1d
1a
1b
1c
27:73
30:70
33:67
11:89
24:76
24:76
29:71
18:82
23:77
31:69
31:69
34:66
24:76
74:26
61:39
64:36
0:100
3:97
8:92
3:97
35:65
2:98
Emission measures have been carried out by means of a
spectrofluorimeter. Fluorescence lifetime measurements have been
carried out by using a single photon counter.
13:87
3:97
Synthesis of 4-Methoxy-2-trimethylsilylphenol (4). For the
preparation of the title compound, we modified the procedure
reported for the synthesis of 2-(trimethylsilyl)phenol.33 To an
anhydrous THF solution of 2-bromo-4-methoxyphenol34 (4 g, 19.5
mmol) kept under nitrogen was added slowly n-BuLi (1.6 M solution
in hexane, 27 mL, 43 mmol) at −78 °C. After 40 min stirring, freshly
distilled trimethylsilyl chloride (6.4 mL, 50 mmol) was added
dropwise. The resulting mixture was warmed to room temperature
and stirred for 4 h, then quenched with 1 N HCl (20 mL). After being
stirred overnight, extraction with ethyl acetate and purification of
the raw material by column chromatography afforded phenol 4 as a
colorless solid (2.9 g, 76% yield, mp 46−48 °C). 4: Spectroscopic data
are in accord with the literature.35 Anal. Calcd for C10H16O2Si: C,
61.18; H, 8.21. Found: C, 61.1; H, 8.1.
Synthesis of 4-Methoxy-2-trimethylsilylphenyl Methanesul-
fonate (1a). Following a literature procedure for the synthesis of
4-methoxyphenyl methanesulfonate,12 4-methoxy-2-trimethylsilylphenol
(4, 300 mg, 1.5 mmol) was treated with methanesulfonyl chloride
(0.14 mL, 1.8 mmol) and triethylamine (0.32 mL, 2.3 mmol) in
dichloromethane (7 mL) at 0 °C. After distillation at reduced pressure,
mesylate 1a was obtained as a colorless oil (302 mg, 72% yield). 1a:
1H NMR (CDCl3) δ 7.45 (d, 1 H, J = 9 Hz), 7.00 (d, 1 H, J = 3 Hz),
difference appeared from these comparisons, and in neat
solvents, the reduction/substitution ratio did not change
significantly for Ar−Cl and Ar−O cleavage, despite the fact
that the quantum yields changed by a factor >5. Likewise, the
arylation proportion in the presence of π traps depended little
on the LG. Thus, the chemistry observed depended on the
nature of the phenyl cation involved rather than from the
structure of the aromatic used as precursor, or in other words,
the reaction is satisfactorily described as a SN1 process.
This work demonstrates the large scope of the photo-
activation of Ar−X bonds, which applies uniformly, though
with varied efficiency, to bonds of different strengths under
similar, mild conditions, contrary to metal catalysis that usually
depends dramatically on the bond energy. The order of
reactivity in the photocleavage of compounds 1a−d is related
neither to the strength of the acid liberated nor to what is
generally observed in metal-mediated cross-coupling reactions.
It would appear that photoheterolysis is effective also with
leaving groups not necessarily pertaining to the family of the
“good leaving group in organic synthesis” (phosphates here are
an example) and possibly may be further extended.
The stabilizing effect of phenyl cations by a silicon atom in
the β position plays the same role with (methoxy)phenyl esters
as in the corresponding chloride. The end products maintain
the SiMe3 function, thus offering an entry to highly
functionalized aromatics, for example, via oxidative coupling
reactions.29 As an alternative, due to the easy removal,30 the
SiMe3 group can be considered as a directing, removable group.
Moreover, the introduction of the TMS group, optionally
coupled with the further stabilization by aqueous media, allows
tuning the triplet versus singlet cation chemistry, shifting to the
formation of the Ar−heteroatom bond in place of Ar−C bonds.
The efficient photoheterolysis of the Ar−O bond in aryl
phosphates and sulfonates could have useful implications. As an
example, the release of stable phosphate-based biologically
interesting molecules from suitable “caged” precursors has been
intensively investigated.31 Compound 1c could be a candidate
for this aim, where aryl phosphates have not yet been used.7 On
the other hand, the photochemical generation of strong
sulfonic acid via photoinduced Ar−O or O−S cleavage14,32 in
6.90 (dd, 1 H, J = 3 and 9 Hz), 3.80 (s, 3 H), 3.20 (s, 3 H), 0.4 (s, 9
H); 13C NMR (CDCl3) δ 157.3, 147.7, 133.6, 121.3 (CH), 120.6
(CH), 114.7 (CH), 55.5 (CH3), 38.1 (CH3), −0.1 (CH3); IR (neat)
ν/cm−1 2956, 1366, 840. Anal. Calcd for C11H18O4SSi: C, 48.15; H,
6.61. Found: C, 48.0; H, 6.6.
Synthesis of 4-Methoxy-2-(trimethylsilyl)phenyl Trifluoro-
methanesulfonate (1b). According to the procedure for the
synthesis of 2,5-[bis(trimethylsilyl)]phenyl-1,4-trifluoromethanesulfo-
nate,36 phenol 4 (800 mg, 4 mmol) was dissolved in anhydrous
pyridine, the resulting solution was cooled to 0 °C under nitrogen, and
trifluoromethanesulfonic anhydride was added dropwise. After heating
at 50 °C for 1 h and removing the solvent in vacuo, the residue was
purified by column chromatography to give 1b as a colorless oil (1.14 g,
37
1
88% yield). 1b: H NMR (CD3COCD3) δ 7.35−7.30 (m, 1 H),
7.10−7.05 (m, 2 H), 3.96 (s, 3 H), 0.30 (s, 9 H); 13C NMR
(CD3COCD3) δ 159.9, 149.4, 135.2, 122.7 (CH), 122.4 (CH), 119
(q, CF3, J = 317 Hz) 117.0 (CH), 56.4 (CH3), 0.4 (CH3); IR (neat)
ν/cm−1 2952, 1504, 1121, 1140, 889. Anal. Calcd for C11H15F3O4SSi:
C. 40.23; H. 4.60. Found: C, 40.2; H, 4.5.
Synthesis of 4-Methoxy-2-trimethylsilylphenyl Diethylphos-
phate (1c). Following the method previously described by
Hatano for the synthesis of (R)-1,1′-binaphthalene-2,2′-bis-
(diethoxyphosphinate),38 phenol 4 (400 mg, 2 mmol) was dissolved
in anhydrous THF and treated with NaH at 0 °C. Diethyl
chlorophosphate (0.64 mL, 4.4 mmol) was carefully added, then the
3505
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
mixture was stirred at room temperature for 3 h. After usual workup,
AUTHOR INFORMATION
Corresponding Author
*E-mail: fagnoni@unipv.it. Tel: +390382987316. Fax:
+390382987323.
■
bulb-to-bulb distillation afforded 1c as a colorless oil (360 mg, 54%
1
yield). 1c: H NMR (CDCl3) δ 7.40−7.35 (d, 1 H, J = 9 Hz), 6.95−
6.90 (d, 1 H, J = 3 Hz), 6.85−6.80 (dd, 1 H, J = 3 and 9 Hz), 4.25−
4.05 (m, 4 H), 3.80 (s, 3 H), 1.40−1.30 (m, 6 H), 0.30 (s, 9 H); 13C
NMR (CDCl3) δ 155.7, 149.5, 131.4, 120.7 (CH), 118.5 (CH), 114.6
(CH), 64.3 (CH2), 55.4 (CH3), 16.0 (CH3), 15.9 (CH3), −1.05
(CH3); IR (neat) ν/cm−1 2985, 1248, 1038, 974, 840. Anal. Calcd for
C14H25O5PSi: C, 50.59; H, 7.58. Found: C, 50.6; H, 7.6.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Irradiation of 1a in Neat 2,2,2-Trifluoroethanol (TFE). A
solution of mesylate 1a (412 mg, 1.5 mmol, 0.05 M), cesium carbonate
(244 mg, 0.75 mmol, 0.025 M), and acetone (3 mL, 0.9 M) in TFE
(30 mL) was irradiated for 8 h. After the solvent was removed in
vacuo, column chromatography afforded 64 mg of 3-methoxyphenyl
trimethylsilane 2 (colorless oil, 24% yield) and 234 mg of 2-(2,2,2-
trifluoroethoxy)-5-methoxytrimethylsilylbenzene 3 (colorless oil, 56%
yield). GC analysis also showed the presence of 4-methoxy-2-
trimethylsilylphenol (4, 4% yield). 2: Spectroscopic data are in accord
with the literature.25 Anal. Calcd for C10H16OSi: C, 66.61; H, 8.94.
Found: C, 66.6; H, 8.9. 3: 1H NMR (CDCl3) δ 7.00−6.95 (d, 1H, J =
3 Hz), 6.85−6.80 (dd, 1H, J = 3 and 8 Hz), 6.75−6.70 (d, 1H, J = 8
Hz), 4.30−4.20 (q, 2H, J = 8 Hz), 3.80 (s, 3 H), 0.30 (s, 9 H);
13C NMR (CDCl3) δ 155.5, 154.3, 130.1, 130.0 (q, CF3, J = 275 Hz),
121.6 (CH), 114.2 (CH), 110.5 (CH), 65.5 (q, CH2, J = 35 Hz), 55.6
(CH3), −1.3 (CH3); IR (neat) ν/cm−1 2955, 1482, 1269, 1216, 1165,
839. Anal. Calcd for C12H17F3O2Si: C, 51.78; H, 6.16. Found: C, 51.8;
H, 6.1.
Irradiation of 1b in 5:1 MeCN/H2O Mixture. Triflate 1b (493
mg, 1.5 mmol, 0.05 M) and acetone (3 mL, 0.9M) were dissolved in
30 mL of 5:1 MeCN/H2O mixture and irradiated for 24 h (85%
consumption of 1b). The solvent was removed in vacuo and the
residue purified by silica gel chromatography giving 44 mg of 2 (19%
yield based on the consumption of 1b) and 142 mg of 2-trimethylsilyl-
4-methoxyacetanilide (7, 47% yield based on the consumption of 1b).
7: Spectroscopic data are in accord with the literature.25 Anal. Calcd
for C12H19NO2Si: C, 60.72; H, 8.07; N, 5.90. Found: C, 60.7; H, 8.1;
N, 5.9.
Irradiation of 1a in TFE in the Presence of Benzene. Mesylate
1a (412 mg, 1.5 mmol, 0.05 M), cesium carbonate (244 mg,
0.75 mmol, 0.025 M), acetone (3 mL, 0.9 M), and benzene (27.5 mL,
30 mmol, 1 M) were dissolved in 30 mL of TFE and irradiated for 7 h.
Purification of the raw product by silica gel chromatography afforded
a mixture of 2 (51 mg, 19% yield), 3 (175 mg, 42% yield), and
4-methoxy-2-trimethylsilylbiphenyl (5, 73 mg, 19% yield). 5: 1H NMR
(CDCl3, from the mixture) δ 7.45−7.25 (m, 5H), 7.25−7.15 (m, 1H),
6.95−6.85 (m, 2H), 3.90 (s, 3H), 0.10 (s, 9H); IR (of the mixture)
ν/cm−1 2954, 1587, 1478, 1216, 1057, 1040, 838.
Irradiation of 1b in TFE in the Presence of Allyltrimethylsi-
lane (ATMS). Triflate 1b (493 mg, 1.5 mmol, 0.05 M), cesium
carbonate (244 mg, 0.75 mmol, 0.025 M), acetone (3 mL, 0.9 M), and
ATMS (2.4 mL, 15 mmol, 0.5 M) were dissolved in TFE (30 mL) and
irradiated for 12 h (88% consumption of 1b). Purification by column
chromatography yielded 3 (95 mg, 26% yield based on the
consumption of 1b) and 2-allyl-5-methoxyphenyltrimethylsilane 6
(colorless oil, 145 mg, 50% yield based on the consumption of 1b). 6:
1H NMR (CDCl3), δ: 7.15−7.10 (d, 1 H, J = 8.5 Hz), 7.05−7.00 (d, 1
S.P. acknowledges MIUR, Rome (FIRB-Futuro in Ricerca 2008
project RBFR08J78Q), for financial support. We thank Dr. V.
Dichiarante for her help.
REFERENCES
■
(1) Modern Arylation Methods; Ackermann, L., Ed.; Wiley-VCH
Verlag GmbH & Co: Weinheim, Germany, 2009.
(2) Rosen, B. M.; Quasdorf, K. W.; Wilson, D. A.; Zhang, N.;
Resmerita, A.-M.; Garg, N. K.; Percec, V. Chem. Rev. 2011, 111, 1346−
1416.
́
(3) (a) Barluenga, J.; Florentino, L.; Aznar, F.; Valdes, C. Org. Lett.
2011, 13, 510−513. (b) Peng, A.-Y.; Chen, B.-T.; Wang, Z.; Wang, B.;
Mo, X.-B.; Wang, Y.-Y.; Chen, P.-J. J. Fluorine Chem. 2011, 132,
982−986.
(4) Limmert, M. E.; Roy, A. H.; Hartwig, J. F. J. Org. Chem. 2005, 70,
9364−9370.
(5) (a) Percec, V.; Golding, G. M.; Smidrkal, J.; Weichold, O. J. Org.
Chem. 2004, 69, 3447−3452. (b) Kobayashi, Y.; Mizojiri, R.
Tetrahedron Lett. 1996, 37, 8531−8534.
(6) Zhang, L.; Qing, J.; Yang, P.; Wu, J. Org. Lett. 2008, 10, 4971−
4974.
(7) Protti, S.; Fagnoni, M. Chem. Commun. 2008, 3611−3621.
(8) Chen, H.; Huang, Z.; Hu, X.; Tang, G.; Xu, P.; Zhao, Y.; Cheng,
C.-H. J. Org. Chem. 2011, 76, 2338−2344.
(9) Ebran, J.-P.; Hansen, A. L.; Gøgsig, T. M.; Skrydstrup, T. J. Am.
Chem. Soc. 2007, 129, 6931−6942.
(10) Ming, So, C.; Zhou, Z.; Po Lau, C.; Kwong, F. Angew. Chem., Int.
Ed. 2008, 47, 6402−6406.
(11) For selected references on the aryl cation chemistry, see:
(a) Dichiarante, V.; Fagnoni, M. Synlett 2008, 787−800. (b) Slegt, M.;
Hermen, S.; Overkleeft, H. S.; Lodder, G. Eur. J. Org. Chem. 2007,
5364−5375. (c) Penenory, A. B.; Arguello, J. E. In Handbook of
́
̃
̃
̈
Synthetic Photochemistry; Albini, A., Fagnoni, M., Eds.; Wiley-VCH:
Weinheim, Germany, 2010; pp 319−352. (d) Dichiarante, V.;
Fagnoni, M.; Albini, A. In Modern Arylation Methods; Ackermann, L.,
Ed.; Wiley-VCH Verlag GmbH & Co: Weinheim, Germany, 2009; pp
513−535.
(12) De Carolis, M.; Protti, S.; Fagnoni, M.; Albini, A. Angew. Chem.,
Int. Ed. 2005, 44, 1232−1236.
(13) (a) Grabner, G.; Richard, C.; Koehler, G. J. Am. Chem. Soc.
1994, 116, 11470−11480. (b) Manet, I.; Monti, S.; Fagnoni, M.;
Protti, S.; Albini, A. Chem.Eur. J. 2005, 11, 140−151. (c) Canevari,
V.; Fagnoni, M.; Bortolus, P.; Albini, A. ChemSusChem 2011, 4,
98−103.
(14) Terpolilli, M.; Merli, D.; Protti, S.; Dichiarante, V.; Fagnoni, M.;
Albini, A. Photochem. Photobiol. Sci. 2011, 10, 123−127.
(15) (a) Hori, K.; Sonoda, T.; Harada, M.; Yamazaki-Nishida, S.
Tetrahedron 2000, 56, 1429−1436. (b) Protti, S.; Fagnoni, M.; Albini,
A. J. Am. Chem. Soc. 2006, 128, 10670−10671. (c) Lazzaroni, S.; Protti,
S.; Fagnoni, M.; Albini, A. Org. Lett. 2009, 11, 349−352.
(d) Dichiarante, V.; Fagnoni, M.; Albini, A. Green Chem. 2009, 11,
942−945.
H, J = 2.7 Hz), 6.90−6.85 (dd, 1 H, J = 2.7 and 8.5 Hz), 6.05−5.90
(m, 1 H), 5.15−4.95 (m, 2 H), 3.80 (s, 3 H), 3.50 (d, 2 H, J = 5.56
Hz), 0.35 (s, 9 H); 13C NMR (CDCl3) δ 157.0, 139.8, 138.2 (CH),
137.3, 130.2 (CH), 120.3 (CH), 115.5 (CH2), 113.8 (CH), 55.0
(CH3), 39.1 (CH2), 0.1 (CH3); IR (neat) ν/cm−1 2953, 1478, 1247,
837. Anal. Calcd for C13H20OSi: C, 70.85; H, 9.15. Found: C, 70.9;
H, 9.1.
(16) (a) Stang, P. J. In Dicordinate Carbocations; Rappoport, Z.,
Stang, P. J., Eds.; Wiley: New York, 1997; p 461. (b) Hanack, M.;
Subramanian, L. R. In Methoden der Organischen Chemie; Hanck, M.,
Ed.; Thieme: Stuttgart, Germany, 1990; Vol. E19C, p 249.
(c) Himeshima, Y.; Kobayashi, H.; Sonoda, T. J. Am. Chem. Soc.
1985, 107, 5286−5288. (d) Hanack, M.; Rieth, R. Chem. Ber. 1987,
120, 1659−1666. (e) Slegt, M.; Overkleeft, H. S.; Lodder, G. Eur. J.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR and 13C NMR spectra for compounds 1a−d and 2−7.
This material is available free of charge via the Internet at
http://pubs.acs.org.
3506
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507

The Journal of Organic Chemistry
Article
Org. Chem. 2007, 5364−5375. (f) Steenken, S.; Askokkuna, M.;
Maruthamuthu, P.; McClelland, R. A. J. Am. Chem. Soc. 1998, 120,
11925−11931.
(17) Lazzaroni, S.; Dondi, D.; Fagnoni, M.; Albini, A. J. Org. Chem.
2008, 73, 206−211.
(18) Lazzaroni, S.; Dondi, D.; Fagnoni, M.; Albini, A. J. Org. Chem.
2010, 75, 315−323.
(19) Schade, B.; Hagen, V.; Schmidt, R.; Herbrich, R.; Krause, E.;
Eckardt, T.; Bendig, J. J. Org. Chem. 1999, 64, 9109−9117.
(20) Verbeek, J.-M.; Stapper, M.; Krijnen, E. S.; van Loon, J.-D.;
Lodder, G.; Steenken, S. J. Phys. Chem. 1994, 98, 9526−9536.
(21) (a) Arnold, B.; Donald, L.; Jurgens, A.; Pincock, J. A. Can. J.
Chem. 1985, 63, 3140−3146. (b) However, diethyl phosphate is a
better leaving group than chloride in the photoheterolytic cleavage of
benzoin derivatives. See: Cheruvallil, S. R.; Givens, R. S.; Wirz, J. J. Am.
Chem. Soc. 2000, 122, 611−618.
(22) Manet, I.; Monti, S.; Grabner, G.; Protti, S.; Dondi, D.;
Dichiarante, V.; Fagnoni, M.; Albini, A. Chem.Eur. J. 2008, 14,
1029−1039.
(23) Dichiarante, V.; Dondi, D.; Protti, S.; Fagnoni, M.; Albini, A.
J. Am. Chem. Soc. 2007, 129, 5605−5611; 2007, 129, 11662 (correction).
(24) For arylation via 4-methoxyphenyl cation, see for instance:
(a) Protti, S.; Fagnoni, M.; Mella, M.; Albini, A. J. Org. Chem. 2004, 69,
3465−3473. (b) Protti, S.; Dondi, D.; Fagnoni, M.; Albini, A. Eur. J.
Org. Chem. 2008, 2240−2247.
(25) (a) Dichiarante, V.; Salvaneschi, A.; Protti, S.; Dondi, D.;
Fagnoni, M.; Albini, A. J. Am. Chem. Soc. 2007, 129, 15919−15926.
(b) Protti, S.; Dichiarante, V.; Dondi, D.; Fagnoni, M.; Albini, A.
Chem. Sci. 2012, 3, 1330−1337.
(26) Geissssler, D.; Antonenko, Y. N.; Schmidt, R.; Keller, S.;
Krylova, O. O.; Wiesner, B.; Bendig, J.; Pohl, P.; Hagen, V. Angew.
Chem., Int. Ed. 2005, 44, 1195−1198.
(27) (a) Dichiarante, V.; Fagnoni, M.; Albini, A. J. Org. Chem. 2008,
73, 1282−1289. (b) Mayr, H; Kempf, B.; Ofial, A. R. Acc. Chem. Res.
2003, 36, 66−77.
(28) Protti, S.; Fagnoni, M.; Albini, A. Org. Biomol. Chem. 2005,
2868−2871.
(29) Brenzovich, W. E.; Brazeau, J.-F.; Toste, F. D. Org. Lett. 2010,
12, 4728−4731 and references cited therein.
(30) (a) Zafrani, Y.; Gershonov, E.; Columbus, I. J. Org. Chem. 2007,
72, 7014−7017. (b) Radner, F.; Wistrand, L.-G. Tetrahedron Lett.
1995, 36, 5093−5094.
(31) (a) Dynamic Studies in Biology; Goeldner, M., Givens, R., Eds.;
Wiley-VCH Verlag: Weinheim, Germany, 2005. (b) Givens, R. S.;
Kueper, L. W. III. Chem. Rev. 1993, 93, 55−66.
(32) Andraos, J.; Barclay, G. G.; Medeiros, D. R.; Baldovi, M. V.;
Scaiano, J. C.; Sinta, R. Chem. Mater. 1998, 10, 1694−1699.
(33) Nishide, K.; Ohsugi, S.; Miyamoto, T.; Kumar, K.; Node, M.
Monatsh. Chem. 2004, 135, 189−200.
(34) Kavala, V.; Naik, S.; Patel, B. K. J. Org. Chem. 2005, 70, 4267−
4271.
(35) Ikawa, T.; Nishiyama, T.; Nosaki, T.; Takagi, A.; Akai, S. Org.
Lett. 2011, 13, 1730−1733.
(36) Duong, H. M.; Bendikov, M.; Steiger, D.; Zhang, Q.; Sonmez,
G.; Yamada, J.; Wud, F. Org. Lett. 2003, 5, 4433−4436.
(37) Lakshmi, B. V.; Wefelscheid, U. K.; Kazmaier, U. Synlett 2011,
345−348.
(38) Hatano, M.; Miyamoto, T.; Ishihara, K. J. Org. Chem. 2006, 71,
6474−6484.
3507
dx.doi.org/10.1021/jo300290v | J. Org. Chem. 2012, 77, 3501−3507