2,5-Dimethylphenacyl esters: A photoremovable protecting group for carboxylic acidsi

Article abstract of DOI:10.1021/jp010220e

Irradiation of 2,5-dimethylphenacyl (DMP) esters (la-c) in benzene or cyclohexane solutions produces the corresponding free carboxylic acids (2a-c) in high chemical yields, along with 6-methyl-l-indanone (3). In methanol, 2-(methoxymethyl)-5-methylacetophenone (4) is formed as a coproduct. Quantum yields for the photorelease of the DMP group are higher in nonpolar solvents, 0.2, than in methanol, (f> 0.1. The photoreaction is initiated by efficient photoenolization, 01. Three transient intermediates were identified by laser flash photolysis of la, the triplet-state photoenol and the two ground-state photoenols of Z and E configuration. Release of the acids 2a-c occurs predominantly from the E isomer, which has a lifetime of about 2 ms in methanol and 0.5 s in benzene solution. The present data are essential for potential applications of the DMP moiety as a photoremovable protecting group in organic synthesis or biochemistry ( caged compounds ).

Full text of DOI:10.1021/jp010220e

J. Phys. Chem. A 2001, 105, 10329-10333
10329
2,5-Dimethylphenacyl Esters: A Photoremovable Protecting Group for Carboxylic Acids†
Miroslav Zabadal,‡ Anna Paola Pelliccioli,§ Petr Kla´n,*,‡ and Jakob Wirz*,§
Department of Organic Chemistry, Faculty of Science, Masaryk UniVersity, Kotlarska 2, CZ-611 37 Brno,
Czech Republic, and Institut fu¨r Physikalische Chemie der UniVersita¨t Basel, Klingelbergstrasse 80,
CH-4056 Basel, Switzerland
ReceiVed: January 19, 2001; In Final Form: July 13, 2001
Irradiation of 2,5-dimethylphenacyl (DMP) esters (1a-c) in benzene or cyclohexane solutions produces the
corresponding free carboxylic acids (2a-c) in high chemical yields, along with 6-methyl-1-indanone (3). In
methanol, 2-(methoxymethyl)-5-methylacetophenone (4) is formed as a coproduct. Quantum yields for the
photorelease of the DMP group are higher in nonpolar solvents, φ ≈ 0.2, than in methanol, φ ≈ 0.1. The
photoreaction is initiated by efficient photoenolization, φ ≈ 1. Three transient intermediates were identified
by laser flash photolysis of 1a, the triplet-state photoenol and the two ground-state photoenols of Z and E
configuration. Release of the acids 2a-c occurs predominantly from the E isomer, which has a lifetime of
about 2 ms in methanol and 0.5 s in benzene solution. The present data are essential for potential applications
of the DMP moiety as a photoremovable protecting group in organic synthesis or biochemistry (“caged
compounds”).
Introduction
conditions: traces of base shorten its lifetime because they make
reketonization easier.26 The E isomer may be trapped by dieno-
philes,20 and it may undergo ring closure to form cyclobu-
tenols.27 In addition to this, when leaving groups are present in
an appropriate position, efficient elimination reactions occur,
as in the case of o-(2-substituted ethyl)benzophenone28 or
acetophenone29 and of DMP chloride.30-32 Here we report on
the mechanism, efficiency, and release rates of DMP esters
(1a-c, Scheme 1).
Photoremovable protecting groups are of interest in organic
synthesis and biochemistry.1-3 Several substituted phenacyl
compounds have been studied in the past decade. First reports4-6
were followed by a series of papers by Givens and co-workers
dealing with the p-hydroxyphenacyl chromophore.7-13 Falvey
et al. studied the photorelease of carboxylic acids, alcohols, and
phosphates from the corresponding phenacyl esters in the
presence of excited-state electron donor or hydrogen donor
molecules.14-18 Some of us recently proposed the 2,5-dimeth-
ylphenacyl (DMP) chromophore as a new photoremovable
protecting group for carboxylic acids.19 Direct photolysis of
DMP esters at 254-366 nm led to the formation of the free
acids in high yields. Moreover, the side products should be
biologically benign so that the group may be useful for
biochemical applications when a proper substitution allows a
reasonable water solubility.
The primary photochemistry is related to that of other ortho-
substituted aromatic ketones, which is now well understood.20-25
In these systems, the excited triplet ketone undergoes fast
intramolecular hydrogen abstraction to yield a biradicaloid
species, the triplet enol. Decay of the latter results in the
formation of two ground-state intermediates, the xylylenols of
Z and E configurations, that reketonize with different kinetics.
The Z isomer, which in part is also formed directly via hydrogen
abstraction occurring in the singlet state of the ketone, has a
lifetime similar to that of the triplet enol (<1 µs) and gives the
starting compound via an intramolecular [1,5] H-shift. The E
isomer, on the other hand, may persist up to seconds because
its reketonization requires proton transfer through the solvent.
Accordingly, its kinetics are very sensitive to the experimental
Results
Quantum Yield Measurements. Photolysis of DMP esters
(1a-c) in benzene and methanol provides high chemical yields
(85-95% isolated) of the corresponding acids (2a-c), as
reported in a preliminary communication.19 While 6-methyl-1-
indanone (3) is formed as the only coproduct in benzene at low
conversions, the reaction in methanol provides the solvolysis
product 2-(methoxymethyl)-5-methylacetophenone (4) in addi-
tion to indanone 3. These results parallel the photochemistry
of DMP chloride (1d) reported by Bergmark, where the chloride
ion was the leaving group.30,31 Exhaustive irradiation of the ester
solutions in methanol provided, besides 3 and 4, several
unidentified photodegradation products.
Quantum yields for the disappearance of starting material
were determined using valerophenone as an actinometer33,34
(Table 1). The reactions of compounds 1 in nonpolar benzene
were found to be as efficient (φ ) 0.18-0.25) as that of the
R-phenyl-o-nitrobenzyl protecting group.35 However, an unex-
pected decrease in the quantum yields by a factor of ∼2 was
found in methanol relative to benzene. This differs from the
results of Bergmark,31 who reported a quantum yield φ ) 0.76
for the photolysis of DMP chloride 1d in methanol but only φ
) 0.11 in benzene. An explanation for the different behaviors
of DMP chloride 1d and DMP esters 1a-c will be given in the
Discussion. The acids produced by the photolysis of 1a,b are
poorly soluble in methanol and precipitated from the solution
at higher reaction conversions (>35%). All data in Table 1 were
obtained for low conversions (<20%).
† This work was presented at the PP2000 in Costa do Estoril, Portugal,
honoring Professor Ralph Becker’s contributions.
* To whom correspondence should be addressed. P.K.: E-mail
klan@sci.muni.cz; phone +420-5-41129356; fax +420-5-41211214. J.W.:
E-mail J.Wirz@unibas.ch; phone +41-61-2673842; fax +41-61-2673855.
‡ Masaryk University.
§ Universita¨t Basel.
10.1021/jp010220e CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/20/2001

10330 J. Phys. Chem. A, Vol. 105, No. 45, 2001
Zabadal et al.
Scheme 1a
a X ) -OCOPh (a), -OCOCH2Ph (b), -OCOMe (c), and -Cl (d).
TABLE 1: Photochemistry of DMP Esters at >366 nm
X
solvent
φa
-OCOPh (1a)
benzene
methanol
benzene
methanol
benzene
methanol
0.23 ( 0.02
0.09 ( 0.01
0.18 ( 0.02
0.11 ( 0.01
0.25 ( 0.01
0.14 ( 0.02b
-OCOCH2Ph (1b)
-OCOCH3 (1c)
Figure 1. Difference absorption spectra of 1a in MeOH, (+) 20 ns
and (×) 300 ns after the laser pulse. The inset shows kinetic traces
recorded at 340 and 390 nm.
a Quantum yields for ester degradation. Valerophenone was used as
an actinometer.33,34 b Irradiated at >300 nm. Error margins are standard
errors of the means.
(1.2 × 107 s-1) matched the one of the fast decay at 340 nm.
This also was followed by two slower decays (7.5 µs and 2.2
ms). Upon degassing, the lifetime of the short-lived transient
increased at both wavelengths of observation (decay at 340 nm,
growth at 390 nm; τ ≈ 300 ns), in keeping with its assignment
to a triplet state. The lifetimes of the two longer-lived species
were independent of oxygen concentration. In analogy with
o-methylacetophenone,23 the fast transient is assigned to the
TABLE 2: Photochemistry of DMP Esters with Piperylene
at >300 nm
piperylene
X
solvent
(mol L-1
)
φa
-OCOPh (1a)
benzene
0.00
0.10
0.25
0.50
2.00
0.00
0.10
0.25
0.50
2.00
0.1
0.23 ( 0.01
0.20 ( 0.01
0.16 ( 0.01
0.12 ( 0.02
0.07 ( 0.02
0.09 ( 0.01
0.06 ( 0.01
0.04 ( 0.01
0.03 ( 0.01
0.01 ( 0.01
0.18 ( 0.02 (0.22)
0.11 ( 0.01 (0.15)
3
triplet enol, E, and the two slower ones are assigned to the Z
and E ground-state photoenols. The assignment of 3E was
methanol
confirmed by quenching studies with piperylene.23,36
3
Addition of the quencher did not affect the lifetime of E,
but its amount decreased progressively as the concentration of
3
piperylene was increased. The energy of E is too low for
-OCOCH3 (1c) cyclohexane
quenching by piperylene, whereas the precursor triplet ketone
is readily quenched by it.23 Similarly, the lifetimes of the Z and
E were not changed by piperylene, but the amplitudes of these
transients were reduced. Formation of the enols by the triplet
methanol
0.1
a Quantum yields for ester degradation. The values in the parentheses
are those measured without the piperylene addition (Table 1). Error
margins are standard errors of the means.
3
3
3
pathway, 1a f E f Z + E, is reduced by quenching of 1.
Steady-state analysis leads to the Stern-Volmer relation A0/Aq
Bergmark reported a product ratio of 3/4 ) 0.68 for photolysis
of 1d in methanol solution.30,31 In the course of a reinvestigation
of this compound,36 we found that HCl release affects the
product ratio but not the quantum yield of disappearance of 1d.
We obtained a product ratio of 3/4 ) 0.7 after exhaustive
irradiation of a 5 × 10-3 M solution of 1d, but that ratio
increased to >0.9 when the irradiation was stopped at lower
conversions. Bergmark had noted30 that the formation of 4 was
reduced when methanol was saturated with NaHCO3. We
support this finding, though not an associated reduction in the
quantum yield. When 1 equiv of HCl gas is dissolved in
methanol prior to irradiation, the ratio 3/4 is restored to 0.7 at
very low conversions (<5%). Saturation of methanolic solutions
with NaHCO3 or HCl prior to irradiation had no significant
influence on the quantum yields of 1a. A ratio < 1.8 of the
photoproducts (3/4) was obtained with each of the DMP esters
1a-c at >366 nm. Quantum yields for the degradation of esters
1a,c in the presence of piperylene are given in Table 2.
Laser Flash Photolysis. Transient absorption spectra of 1a
in air-saturated methanol solution showed the formation of three
intermediates, a short-lived one, λmax ≈ 340 nm, and two longer-
lived ones, λmax ≈ 390 nm, which are the wavelengths of
maximum absorption of the expected triplet enol 3E and of the
photoenols, respectively (Figure 1).23,26 In air saturated solution
a fast decay (τ ) 79 ns) at 340 nm was followed by two slower
first-order decays (τ ) 8.3 µs and 1.7 ms). At 390 nm there
was a resolved initial increase in absorbance, whose rate constant
) 1 + kq τ0[q]. A0 and Aq are the absorbances of the enol
3
transients in the absence and presence of quencher q, respec-
tively, kq is the second-order rate constant for quenching of 31a
by piperylene, and τ0 is the lifetime of 1a in the absence of
3
3
quencher. The amplitude ratio A0/Aq of the E isomer increased
linearly with piperylene concentration (6 points up to 1.5 M)
with a slope of kq τ0 ) 14 ( 1 M-1 37
.
Thus, the long-lived E
3
isomer forms by the triplet pathway. Assuming a quenching
3
rate constant of kq ) 5 × 109 M-1 s-1, the lifetime of 1a is
estimated as 3τ0 ≈ 2.8 ns. The Stern-Volmer plot for the Z-enol
had about the same initial slope but reached a plateau at
A0/Aq ≈ 2. Thus, formation of the Z-xylylenol is not fully
quenched at high piperylene concentrations, because a large
fraction (∼1/2) is generated by the singlet pathway.
Addition of sodium azide (NaN3) also reduced the lifetimes
of the photoenols E and Z. Plots of the rate constants versus
the concentration of N3- (0-0.03 M, 5 points for E; 0-0.1 M,
4 points for Z) were linear with slopes of (9.5 ( 0.2) × 105
and (1.62 ( 0.04) × 106 M-1 s-1, respectively. Upon addition
of NaOH (5 × 10-4 M) the decay of the xylylenols was no
longer biexponential. The decay of the two enols increased
progressively upon addition of greater amounts of base ((0.5-
5) × 10-3 M, 4 points). A Stern-Volmer plot yielded a
quenching rate constant of (2.0 ( 0.1) × 109 M-1 s-1. This
corroborated our assignment of the enols, because sensitivity
of photoenols to bases is to be expected.23,26

2,5-Dimethylphenacyl Esters
J. Phys. Chem. A, Vol. 105, No. 45, 2001 10331
SCHEME 2
Two independent methods were used to estimate the absolute
quantum yield for total enol formation (E + Z) from 1a in
methanol. First, the initial transient absorbance at 390 nm was
compared with that at 525 nm of a matched solution of
benzophenone in acetonitrile. The triplet state of benzophenone,
λmax ) 525 nm and ꢀ ) 7000 M-1 cm-1, is formed with unit
quantum yield.38 The extinction coefficient of the xylylenols is
not known. Assuming ꢀ ≈ 3000 M-1 cm-1, which is the
extinction coefficient of 2,2-dimethylisoindene,39 we obtain φZ+E
≈ 2. Another estimate of φZ+E ≈ 2 was obtained by comparing
the initial transient absorbances of 1a,d in matched methanol
solutions. Here, we assumed that the quantum yield for enol
formation from 1d is unity, because that of HCl release from
1d is 0.76. Each of these estimates relies on an unproven
assumption, and the result of both, φZ+E ≈ 2, is in fact
unreasonably high. Nevertheless, the results do indicate that the
total yield of enols from 1a is on the order of unity.
Addition of 0.015 M of acetic acid did not affect the lifetime
of the Z-xlylenol in methanol, but that of the E isomer increased
about 30-fold. Further addition of acetic acid (up to 0.1 M) had
no effect. This indicates that acid inhibits the decay of E back
to starting material by rapid reprotonation (at oxygen) of any
deprotonated E enol.
hypothesis. Elimination restores the aromaticity of the unstable
photoenols. On the other hand, an adiabatic reaction from E
to form the products 3 or 4 in their excited triplet state would
3
be energetically unfavorable because of the low triplet energy
3
of E (ca. 100 kJ mol-1).39,40
LFP of 1a in benzene solution afforded analogous transients
as in methanol. The lifetime of 3E in air-saturated solution was
74 and 63 ns at 340 (decay) and 390 nm (growth), respectively.
As expected, Z decayed more rapidly in benzene (420 ns at
390 nm), whereas E was much longer-lived in this solvent, with
a lifetime of about 0.5 s. A Stern-Volmer plot of the amplitude
The results we obtained with the DMP esters 1a-c studied
here are surprisingly different from those reported for the DMP
chloride 1d. Quantum yields of DMP ester degradation were
significantly lower in methanol than in nonpolar solvents, and
the biexponential decay kinetics observed for the photoenol
transients (λmax ) 390 nm) clearly indicated formation of two
xylylenol isomers. In contrast, Bergmark reported a quantum
yield for the photoelimination of HCl from 1d of 0.76 in
methanol and of 0.11 in benzene solution. Netto-Ferreira and
Scaiano observed formation of only one photoenol from 1d in
methanol, which they assigned to the E isomer.
ratio A0/Aq of the E isomer versus piperylene concentration (6
3
points up to 1.5 M) gave a slope of kq τ0 ) 2.18 ( 0.06 M-1 37
.
Assumption of a rate constant of kq ) 5 × 109 M-1 s-1 yields
3τ0 ≈ 0.4 ns for the lifetime of the ketone triplet.
The LFP results on 1c in aerated methanol solution were
similar to those obtained on 1a. The photoenol triplet 3E decayed
with a lifetime of 71 ns at 340 nm. This corresponded to the
initial growth at 390 nm (78 ns), which was followed by the
decays of Z (6.8 µs) and E (2 ms). A plot of the amplitude ratio
In an effort to elucidate the origin of these qualitative
differences, we reinvestigated 1d and found that the reaction
mechanism proposed earlier requires substantial revision.36
Photorelease of HCl from 1d in methanol to products 3 and 4
(φ ) 0.76) proceeds by efficient heterolytic elimination of
chloride ion from the Z-xylylenol, which is the only photo-
enol formed in this solvent. The E isomer is not formed in
methanol, because photoenolization from the singlet state is
A0/Aq for the E-photoenol versus quencher concentration (4
points up to 0.6 M) gave a value of kq τ0 ) 7.4 ( 0.4 M-1 37
.
3
Discussion
Irradiation of DMP esters 1a-c in methanol leads to efficient
formation of the corresponding free acids 2a-c along with 3
and 4 as coproducts. The transient intermediates observed by
LFP of DMP ester 1a in degassed methanol are shown in
Scheme 2. The structural assignments are based on quenching
studies with piperylene and previous work on the related
transients observed by LFP of 2-methylacetophenone.23
A central issue of the present investigation is at which stage
in the photokinetic scheme the elimination reaction actually takes
place. Previous studies by Bergmark and Netto-Ferreira de-
scribed the photochemistry of 1d. Although Bergmark31 did not
exclude that product formation could take place directly from
3
highly efficient, and the triplet pathway (reaction via E) does
not compete measurably. In benzene solution both the E-
and Z-photoenols are formed, but only the E isomer gives
product 3 by HCl elimination whereas the Z isomer decays
predominantly by intramolecular reketonization to 1d in this
solvent.
The photoreaction of DMP esters 1a-c parallels that of 1d
in benzene solution. Formation of 3 exclusively via the
E-photoenol of 1a in benzene is strongly supported by quenching
experiments with piperylene. The E-photoenol forms only by
the triplet pathway (Scheme 2). Quenching of the triplet ketone
precursor with piperylene gives a linear Stern-Volmer relation
for the amplitudes of the E-photoenol absorption determined
3
the triplet enol, E, both he and Netto-Ferreira32 favored the
hypothesis of elimination occurring from the ground-state
photoenols. They suggested E-xylylenol to be the most likely
intermediate forming indanone 3. Bergmark also proposed that
the solvolysis reaction occurs via a SN1 displacement on either
Z or E, whereas Netto-Ferreira concluded that the Z isomer
reketonizes rapidly (e20 ns) and does not play a role in the
formation of the products.
3
3
by LFP, A0/Aq ) 1 + kq τ0[q]. The slope kq τ0 of 2.18 ( 0.06
M-1 is in satisfactory agreement with the value of kq τ0 ) 1.8
( 0.1 that is derived from the dependence of the quantum yield
of photoelimination on piperylene concentration (Table 2).
3
The scenario changes in methanol solution. Carboxylate ions
are much poorer leaving groups than the chloride ion, and the
enols of the DMP esters do not dissociate as easily as that of
1d. Quantum yields of DMP ester degradation in methanol are
That the elimination reaction takes place from either or both
isomers of the xylylenols appears to be the most reasonable

10332 J. Phys. Chem. A, Vol. 105, No. 45, 2001
Zabadal et al.
SCHEME 3
Experimental Section
Materials and Methods. Benzene and methanol were puri-
fied by distillation through a vacuum-sealed column (70 cm)
packed with glass detritus. 2,5-Dimethylphenacyl chloride (R-
chloro-2′,5′-dimethylacetophenone) was prepared according to
the standard procedure described elsewhere.19,31 NMR spectra
were recorded on a Bruker 300 MHz spectrometer. 1H and 13
C
NMR data were measured in the solvent indicated with
tetramethylsilane as an internal standard. Mass spectra were
obtained using EI (30 eV). Gas chromatography was performed
on a Shimadzu GC 17A gas chromatograph equipped with a
DB1 column. Melting points were determined on a Kofler hot-
stage Rapido 79-2106 and were not corrected.
General Method for Synthesis of 2,5-Dimethylphenacyl
Esters. 2,5-Dimethylphenacyl chloride (32 mmol) in acetone
(25 mL) was added to a stirred mixture of the corresponding
acid (1.3-2 equiv) or its sodium salt (1.2 equiv), sodium iodide
(1.2 equiv), and triethylamine (1.1 equiv) in 250 mL of acetone.
The mixture was stirred and heated under reflux for ∼8 h, and
the product formation was monitored by TLC. The solids were
removed, the solvent was distilled off, and the residue was
extracted with benzene. The benzene layer was washed three
times with 10% HCl (50 mL), 1 M aqueous sodium hydroxide
(50 mL), and water. The organic layer was dried over MgSO4,
and the solvent was removed to obtain the crude products, which
were purified by crystallization or by flash chromatography
(petroleum ether-diethyl ether mixture, 4:1).
substantially lower than those found in nonpolar solvents (Table
1). Therefore, a mechanism involving heterolytic cleavage is
unlikely. If heterolysis took place, its rate should be similar for
both isomers. In fact, both Z and E are formed and decay with
very different lifetimes. On this basis, heterolysis of the shorter-
lived Z isomer can be excluded. Contrary to 1d, where the
photoelimination of HCl proceeds from the Z-xylylenol in
methanol solution,36 the present results for esters 1a-c indicate
reaction predominantly via the E-xylylenol.
Flash photolysis data show that about half of the Z isomer is
formed through the singlet pathway. We estimate the total
quantum yield of enol formation (E + Z) to be close to unity.
If reaction via Z would contribute to the overall quantum yield
of photodegradation in methanol, we would expect a higher
quantum yield in methanol than in benzene. This is the case
for 1d but opposite to what we find for 1a-c. Quenching
experiments with piperylene support this conclusion. Photo-
degradation is largely suppressed by addition of 2 M quencher
concentrations. If Z were reactive and responsible for the
formation of 3, then we would observe a saturation of the
2,5-Dimethylphenacyl benzoate (1a): yield 4.5 g (95%);
white solid; mp 90-91 °C (from diethyl ether); IR (KBr) 2944,
1720, 1700, 1597, 1450, 1372, 1283, 1229, 1122, 1019, 958
3
quenching effect. The quantitative agreement between the kq τ0
1
cm-1; H NMR (300 MHz, CDCl3) δ 2.37 (s, 3 H), 2.48 (s, 3
value of 14 ( 1 M-1 determined by LFP for the amount of E
with that of 4.6 ( 0.3 M-1 determined from the reaction
quantum yields (Table 2) is less than perfect. The discrepancy
may be due to a small unquenchable contribution in the reaction
quantum yields from reaction via Z-photoenol or, more likely,
to experimental error in the determination of the latter. The data
for the very slow conversion at high piperylene concentrations
had to be determined near the detection limit for GC analysis.
H), 5.40 (s, 2 H), 7.15-7.25 (m, 2 H), 7.43-7.48 (m, 3 H),
7.58 (t, J ) 7.6 Hz, 1 H), 8.11-8.14 (m, 2 H); 13C NMR (75.5
MHz, CDCl3) δ 20.8, 21.1, 67.9, 128.6, 128.9, 129.7, 130.1,
132.3, 133.1, 133.5, 134.7, 135.6, 136.1, 166.3, 196.0; MS (EI)
m/z 268 (M+), 147, 133, 105, 77. Anal. Calcd for C17H16O3:
C, 76.10; H, 6.01. Found: C, 76.53; H, 6.11.
2,5-Dimethylphenacyl phenylacetate (1b): yield 3.9 g
(84%); yellowish liquid; IR (KBr) 3030, 2927, 2867, 1744,
1702, 1497, 1453, 1239, 1145, 969 cm-1; 1H NMR (300 MHz,
CDCl3) δ 2.28 (s, 3 H), 2.41 (s, 3 H), 3.74 (s, 2 H), 5.10 (s, 2
H), 7.07-7.31 (m, 8 H); 13C NMR (75.5 MHz, CDCl3) δ 20.4,
20.8, 40.7, 67.7, 127.1, 128.5, 129.4, 129.9, 131.9, 132.0, 132.2,
132.7, 133.6, 134.5, 135.3, 135.6, 170.9, 195.9; MS (EI) m/z
282 (M+), 133, 105, 91, 77.
We conclude that for the DMP esters 1a-c the only
productive photoenol is that of E configuration, and we withdraw
an opposite proposal made in a preliminary communication.19
The Z-photoenols revert to starting material by an intramolecular
1,5-hydrogen shift. Release rates for the acids 2a-c are on the
order of 103 s-1 in methanol. The lower quantum yield in
methanol compared to benzene indicates that reketonization of
the E-xylylenol by proton transfer through solvent competes
efficiently with product formation. Competition by the reke-
tonization reaction is also supported by the increased lifetime
of E in the presence of small amounts of acetic acid that was
observed by LFP.
2,5-Dimethylphenacyl acetate (1c): yield 3.3 g (84%);
yellow liquid; IR (KBr) 3023, 2930, 2870, 1750, 1702, 1422,
1372, 1229, 1176, 1080, 967 cm-1; 1H NMR (300 MHz, CDCl3)
δ 2.20 (s, 3 H), 2.35 (s, 3 H), 2.45 (s, 3 H), 5.15 (s, 2 H),
7.13-7.26 (m, 2 H), 7.39 (s, 1 H); 13C NMR (75.5 MHz, CDCl3)
δ 20.6, 20.7, 21.0, 67.5, 128.7, 132.3, 133.0, 134.6, 135.5, 135.9,
170.5, 196.0; MS (EI) m/z 206 (M+), 165, 133, 105, 91, 77.
6-Methyl-1-indanone (3): yellow solid; mp 48-51 °C (from
petroleum ether); IR (KBr) 2959, 2921, 2855, 1708, 1491, 1282,
The arguments given above indicate that heterolytic cleavage
of the xylylenols from 1a-c is inefficient. Formation of the
solvolysis product 4 in methanol may then be attributed to a
concerted mechanism, where nucleophilic attack of the solvent
facilitates the elimination, as shown in Scheme 3. Such a
mechanism is supported by the quenching effect of added azide
on both xylylenols, which suggests that quantum yields >80%
could be achieved by addition of 1 M azide (kq ≈ 1 × 106 M-1
s-1, τ0 ) 7.5 µs).
Further experiments with better leaving groups such as
phosphates will be required to explore whether the full potential
of DMP as a photoremovable protecting group can be exploited
to give quantum yields of elimination approaching unity.
1
828, 805 cm-1; H NMR (300 MHz, CDCl3) δ 2.44 (s, 3 H),
2.65-2.69 (m, 2 H), 3.08 (dd, J ) 5.8 Hz, 2 H), 7.35 (d, J )
7.9 Hz, 1 H), 7.40 (dd, J1 ) 7.9 Hz, J2 ) 1.5 Hz, 1 H), 7.55
(broad s, 1 H); 13C NMR (75.5 MHz, CDCl3) δ 21.1, 25.4, 36.6,
123.7, 126.4, 135.9, 137.2, 137.3, 152.5, 207.1; MS (EI) m/z
146 (M+), 118, 115, 103, 91. Anal. Calcd for C10H10O: C,
82.16; H, 6.89. Found: C, 82.39; H, 6.92.
2-(Methoxymethyl)-5-methylacetophenone (4): yellowish
solid; mp 31-32 °C; IR (KBr) 2994, 2928, 2831, 1723, 1437,
1288, 1089, 996 cm-1; 1H NMR (300 MHz, CDCl3) δ 2.39 (s,

2,5-Dimethylphenacyl Esters
J. Phys. Chem. A, Vol. 105, No. 45, 2001 10333
(7) Conrad, P. G.; Givens, R. S.; Hellrung, B.; Rajesh, C. S.; Ramseier,
M.; Wirz, J. J. Am. Chem. Soc. 2000, 122, 9346-9347.
(8) Conrad, P. G.; Givens, R. S.; Weber, J. F. W.; Kandler, K. Org.
Lett. 2000, 2, 1545-1547.
(9) Givens, R. S.; Athey, P. S.; Matuszewski, B.; Kueper, L. W.; Xue,
J. Y.; Fister, T. J. Am. Chem. Soc. 1993, 115, 6001-6012.
(10) Givens, R. S.; Jung, A.; Park, C. H.; Weber, J.; Bartlett, W. J. Am.
Chem. Soc. 1997, 119, 8369-8370.
3 H), 2.57 (s, 3 H), 3.42 (s, 3 H), 4.70 (s, 2 H), 7.29 (d, J1 )
7.9 Hz, J2 ) 8.2 Hz, 1 H), 7.50 (s, J ) 5.8 Hz, 2 H); 13C NMR
(75.5 MHz, CDCl3) δ 21.0, 29.2, 58.6, 72.6, 128.0, 129.9, 132.5,
136.3, 136.6, 136.7, 201.6; MS (EI) m/z 178 (M+), 163, 147,
117, 91. Anal. Calcd for C11H14O2: C, 74.13; H, 7.92. Found:
C, 73.91; H, 7.96.
Quantum Yield Measurements. Solutions of the DMP ester
(∼5 × 10-3 M) in benzene, methanol, or cyclohexane were
irradiated simultaneously at >366 or > 300 nm with valerophe-
none solutions used as an actinometer in a “merry-go-round”
apparatus immersed in a water bath. Samples in Pyrex tubes
were degassed in three freeze-pump-thaw cycles before being
sealed. The >366 and >300 nm bands from a medium-pressure
400 W Teslamp mercury lamp were isolated by filtration with
Corning CS 0-52 and CS 7-37 filters or a CS 7-37 filter,
respectively. The photoproduct yields were analyzed by GC
calibrated with the authentic compounds. All quantum yields
reported are calculated for the ester degradation.
Laser Flash Photolysis. Experiments were carried out by
exciting the sample solutions with absorbances of <0.5/cm-1
at 308 nm with the 20 ns, ∼100 mJ pulses of a XeCl excimer
laser. A pulsed Xenon arc was used as the monitoring beam
(4.5 cm path length, orthogonal to the excitation pulse). The
detection system allowed monitoring of either the kinetics at a
single wavelength using a transient digitizer or the whole
transient spectrum, with an Optical Multichannel Array. Metha-
nol stock solutions were prepared so that absorption at the
irradiation wavelength was around 0.2. To avoid interference
from the photoproducts, samples were discarded after each laser
shot and replaced by fresh ones.
(11) Givens, R. S.; Park, C. H. Tetrahedron Lett. 1996, 37, 6259-6262.
(12) Givens, R. S.; Weber, J. F. W.; Conrad, P. G.; Orosz, G.; Donahue,
S. L.; Thayer, S. A. J. Am. Chem. Soc. 2000, 122, 2687-2697.
(13) Park, C. H.; Givens, R. S. J. Am. Chem. Soc. 1997, 119, 2453-
2463.
(14) Banerjee, A.; Falvey, D. E. J. Am. Chem. Soc. 1998, 120, 2965-
2966.
(15) Banerjee, A.; Falvey, D. E. J. Org. Chem. 1997, 62, 6245-6251.
(16) Banerjee, A.; Lee, K.; Falvey, D. E. Tetrahedron 1999, 55, 12699-
12710.
(17) Banerjee, A.; Lee, K.; Yu, Q.; Fang, A. G.; Falvey, D. E.
Tetrahedron Lett. 1998, 39, 4635-4638.
(18) Lee, K.; Falvey, D. E. J. Am. Chem. Soc. 2000, 122, 9361-9366.
(19) Kla´n, P.; Zabadal, M.; Heger, D. Org. Lett. 2000, 2, 1569-1571.
(20) Lutz, H.; Bre´he´ret, E.; Lindqvist, L. J. Chem. Soc., Faraday Trans.
1 1973, 69, 2096-2102.
(21) Sammes, P. G. Tetrahedron 1976, 32, 405-422.
(22) Findlay, D. M.; Tchir, M. F. J. Chem. Soc., Faraday Trans. 1 1976,
72, 1096-1100.
(23) Haag, R.; Wirz, J.; Wagner, P. J. HelV. Chim. Acta 1977, 60, 2595-
2607.
(24) Das, P. K.; Encinas, M. V.; Small, R. D., Jr.; Scaiano, J. C. J. Am.
Chem. Soc. 1979, 101, 6965-6970.
(25) Wagner, P. J.; Park, B.-S. Org. Photochem. 1991, 11, 227-366.
(26) Scaiano, J. C.; Wintgens, V.; Netto-Ferreira, J. C. Tetrahedron Lett.
1992, 33, 5905-5908.
(27) Wagner, P. J.; Subrahmanyam, D.; Park, B.-S. J. Am. Chem. Soc.
1991, 113, 709-710.
(28) Tseng, S.-S.; Ullman, E. F. J. Am. Chem. Soc. 1976, 98, 541-
544.
(29) Pelliccioli A. P.; Wirz J. Unpublished results.
(30) Bergmark, W. R. J. Chem. Soc., Chem. Commun. 1978, 61-62.
(31) Bergmark, W. R.; Barnes, C.; Clark, J.; Paparian, S.; Marynowski,
S. J. Org. Chem. 1985, 50, 5612-5615.
(32) Netto-Ferreira, J. C.; Scaiano, J. C. J. Am. Chem. Soc. 1991, 113,
5800-5803.
(33) Wagner, P. J.; Kochevar, I. E.; Kemppainen, A. E. J. Am. Chem.
Soc. 1972, 94, 7489-7494.
(34) Wagner, P. J.; Thomas, M. J.; Harris, E.. J. Am. Chem. Soc. 1976,
98, 7675-7679.
Acknowledgment. We gratefully acknowledge the Czech
Ministry of Education, Youth, and Sport (CEZ: J07/98:
143100005) and the Swiss National Science Foundation (Grant
No. 20-61746.00) for the support of this research. We thank J.
Kla´nova´ for some gas chromatography measurements, and we
are grateful to two anonymous reviewers who have contributed
substantially by their constructive critical comments.
References and Notes
(35) Zhu, Q. Q.; Schnabel, W.; Schupp, H. J. Photochem. 1987, 39,
317-332.
(1) Wuts, P. G. M.; Green, T. W. ProtectiVe Groups in Organic
Chemistry; Wiley: New York, 1999.
(2) Marriott, G., Ed. Methods in Enzymology; Academic Press: New
York, 1998; Vol. 291.
(3) Morrison, H. Biological Applications of Photochemical Switches;
Wiley: New York, 1993.
(4) Sheenan, J.; Umezawa, J. J. Org. Chem. 1973, 38, 3771-3774.
(5) Epstein, W. W.; Garrossian, M. J. Chem. Soc., Chem. Commun.
1987, (8), 532-533.
(6) Baldwin, J. E.; Mcconnaughie, A. W.; Moloney, M. G.; Pratt, A.
J.; Shim, S. B. Tetrahedron 1990, 46, 6879-6884.
(36) Pelliccioli, A.-P.; Kla´n, P.; Zabadal, M.; Wirz, J. J. Am. Chem.
Soc. 2001, 123, 7931-7932.
3
(37) To ensure homodispersity of experimental errors, kq τ0 was
3
determined by nonlinear fit to the inverse function Aq/A0 ) {1 + kq τ0[q]}-1
.
(38) Bonneau, R.; Carmichael, I.; Hug, G. L. Pure Appl. Chem. 1991,
63, 289-299.
(39) Allan, M.; Asmis, K. R.; El houar, S.; Haselbach, E.; Capponi,
M.; Urwyler, B.; Wirz, J. HelV. Chim. Acta 1994, 77, 1541-1548.
(40) Kumar, C. V.; Chattopadhyay, S. K., Das, P. K. J. Am. Chem. Soc.
1983, 105, 5143-5144.