Direct Photolysis of Phenacyl Protecting Groups Studied by Laser Flash Photolysis: An Excited State Hydrogen Atom Abstraction Pathway Leads to Formation of Carboxylic Acids and Acetophenone

Full text of DOI:10.1021/ja971431t

J. Am. Chem. Soc. 1998, 120, 2965-2966
2965
Direct Photolysis of Phenacyl Protecting Groups
Scheme 1. Homolytic Bond Scission Mechanism
Studied by Laser Flash Photolysis: An Excited State
Hydrogen Atom Abstraction Pathway Leads to
Formation of Carboxylic Acids and Acetophenone
Anamitro Banerjee and Daniel E. Falvey*
Department of Chemistry and Biochemistry
UniVersity of Maryland,College Park, Maryland 20742
ReceiVed May 5, 1997
Scheme 2. Mechanism in Presence of H-Atom Donors
Photochemically removable protecting groups (PRPGs) are
molecular entities that can be covalently attached to a functional
group of a molecule for the purposes of rendering it inert to some
1
particular set of reaction conditions. The PRPG is then released
when the protected compound is exposed to light. Such protecting
2
groups find uses not only in multistep syntheses but also in
biophysical studies where they are often called “caged molecules”
or “phototriggers”.3
,4
Investigations in several laboratories have focused on various
aromatic R-keto groups as PRPGs.5 Among the earliest R-keto
derivatives to be studied were those based on the phenacyl (i.e.,
the benzoylmethyl, PhCOCH
2
) chromophore. Sheehan6 and
others7 reported that photolysis of 4-methoxyphenacyl esters in
,8
1,4-dioxane solutions gave good yields of the corresponding acids
along with 4-methoxyacetophenone. These workers proposed a
mechanism involving excited-state C-O bond homolysis giving
phenacyl radical along with an acyloxy radical (Scheme 1).
•
Although elimination of CO
2
from acyloxy radicals (RCO
2
) is
9
•
an extremely fast process, especially when the product (R ) is a
resonance-stabilized species,10 it was assumed that abstraction of
an H-atom was occurring competitively with decarboxylation.
In the course of our studies on the photoinduced electron-
transfer behavior of phenacyl esters, we had occasion to undertake
Figure 1. Effect of H-atom donors on the photolysis (150 W Hg-Xe
lamp) of 1 (3.5) mM, N -purged solutions) in various solvents. Panel A
illustrates the effects of adding 6.20 mM n-Bu SnH to benzene (b) and
7.6 mM CHD to benzene (2) compared with neat benzene (O). In these
direct irradiation of phenacyl phenylacetate 1 in CH
suming the mechanism in Scheme 1, direct homolysis of this
3
CN. As-
2
3
•
2 2
compound would lead to phenylacetoxy radical (PhCH CO ).
1
Pincock11 has estimated the lifetime of this species to be 200 ps.
experiments the solutions were 10.3 mM in 1. Panel B illustrates the
Consequently, we anticipated rapid decarboxylation would give
effect of adding 435 mM 2-PrOH to CH CN (b) compared with neat
CHCN (O) along with the effect of adding 2-PrOH to H O/CH CN (ca.
2 3
3
•
the benzyl radical (PhCH
2
) which would dimerize to give
bibenzyl (1,2-diphenylethane) as a stable product. Contrary to
our expectations, we found that photolysis of phenacyl ester gave
no detectable bibenzyl (<0.5% by HPLC analysis), but instead
high yields of acetophenone and phenylacetic acid (eq 1).12
Again, when phenacyl (2-methoxyphenyl)acetate was subjected
to similar photolysis conditions, acetophenone and the corre-
sponding acid were detected as products.
3:2) (2) compared with the same solvent containing no additives.
These results caused us to reconsider the previously proposed
mechanism for phenacyl ester photolysis. The experiments below
lead to the conclusion that H-atom transfer,13 rather than C-O
bond homolysis, is the initial photochemical step in the decay
mechanism (Scheme 2). The resulting ketyl radical 2 decays
through a second-order process, perhaps involving a net dispro-
portionation with the donor radical via a relatively long-lived
coupling product of structure 3.
This conclusion is supported by the following observations:
(1) The quantum yield for decomposition of 1 is greatly
increased when efficient H-atom donors (1,4-cyclohexadiene,
CHD; tributyltin hydride or 2-propanol) are added to the sample
(Figure 1). For example, 1 is stable for over several hours of
irradiation in neat benzene, yet it is consumed in <10 min when
(
1) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis,
nd ed.; Wiley: New York, 1991.
2) Pillai, V. N. R. In Organic Photochemistry; Padwa, A., Ed.; Marcel
Dekker: New York, 1987; Vol. 9, p 225.
3) Willner, I.; Willner, B. In Biological Applications of Photochemical
Switches; Morrison, H., Ed.; Wiley-Interscience: New York, 1993; p 1.
4) Corrie, J. E. T.; Trentham, D. R. In Biological Applications of
Photochemical Switches; Morrison, H., Ed.; Wiley: New York, 1993; p 243.
2
(
(
6
3
.20 mM n-Bu SnH was added. This appears to be a general
(
effect that does not require a particular solvent or H-atom donor.
Similar increases in photolysis efficiency are observed when CHD
is added to CH CN and when 2-propanol is added to water. In
3
each case, the only products are acetophenone and the carboxylic
acid.
(5) Givens, R. S.; Kueper, L. W., III Chem. ReV. 1993, 93, 55.
(6) Sheehan, J. C.; Umezawa, K. J. Org. Chem. 1973, 38, 3771.
(7) Epstein, W. W.; Garrossian, M. J. Chem. Soc., Chem. Commun. 1987,
5
32.
8) Baldwin, J. E.; McConnaughie, A. W.; Moloney, M. G.; Pratt, A. J.;
(
If homolytic bond scission were the first irreversible step in
the reaction (i.e., following Scheme 1), then we would expect no
Shim, S. B. Tetrahedron 1990, 46, 6879.
(
(
(
(
9) Pincock, J. A. Acc. Chem. Res. 1997, 30, 43.
10) Falvey, D. E.; Schuster, G. B. J. Am. Chem. Soc. 1986, 108, 7419.
11) DeCosta, D. P.; Pincock, J. A. J. Am. Chem. Soc. 1989, 111, 8948.
12) Typical photolysis runs employed λ > 295 nm.
(13) Lissi, E. A.; Encinas, M. V. In Handbook of Organic Photochemistry;
Scaiano, J. C., Ed.; CRC Press: Boca Raton, FL, 1989; Vol. II, p 111.
S0002-7863(97)01431-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/17/1998

2966 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Communications to the Editor
2
2
decays a second-order fashion (following d[2]/dt ) k
c
[2] where
5
-1 -1
k
c
) 3.48 × 10 ꢀ s ). If ꢀ320 is estimated using the corres-
16
9
-1
ponding value for acetophenone ketyl, then k
c
) 7 × 10 M
-
1
s . This value is near the diffusion limit and thus consistent
with a radical-radical reaction. Second, the long-lived absorption
spectrum differs from that of acetophenone or its enol (270
nm).1
8,19
Third, UV and H NMR spectra taken immediately after
steady-state photolysis show only formation of PhCOCH and
PhCH CO H. Thus, the long-lived species is not stable, and
1
3
Figure 2. (A) Transient absorption spectrum taken 400 ns after laser
flash photolysis (308 nm, 10 ns, 50 mJ) of 10.4 mM solutions of 1 in
benzene. The inset illustrates the absorption decay (to 7 µs) at 320 nm.
This spectrum is assigned to the excited triplet state of 1. (B) Transient
2
2
moreover, it must decay cleanly to give these products. Finally,
extensive photochemical studies on benzophenone in 2-propanol
have revealed a metastable species, termed the “light absorbing
transient” or LAT.2
0-23
This has been assigned to a coupling
spectra taken 4.0 and 90.0 µs after pulsed laser photolysis of 1 in N
2
-
purged 2-propanol. Under the latter conditions, decay of the triplet is
rapid and only ketyl 2 and coupling product 3 are detected. The inset
illustrates absorption decay (to 450 µs) at 320 nm.
product where the 2-hydroxy-2-propyl radical binds to the ortho
or para ring positon on the benzophenone ketyl radical. By
analogy to the earlier work, we tentatively assign the long-lived
transient to analogous coupling product exemplified by 3 in
Scheme 2.
Unlike the benzophenone LAT, 3 has available to it a facile
decay pathway involving elimination of carboxylate from the
allylic position of an electron-rich double bond coupled with
generation of the oxidized donor species. (For 2-propanol this
or little dependence of photolysis times on the presence of an
H-atom donor. On the other hand, these results follow readily
from the mechanism in Scheme 2. In this mechanism, H-atom
abstraction competes with photon-wasting nonradiative decay of
the excited triplet state of phenacyl ester 1.
Results from earlier work on similar systems are also consistent
would correspond to acetone, for CHD, benzene. For n-Bu SnH,
3
6
with this observation. Sheehan reported that the photolysis of
nucleophilic attack at the Sn atom of the corresoponding adduct
4
-methoxyphenacyl esters was much faster in dioxane (a H-atom
would allow for net reduction of 1 and formation of Sn-Nuc. It
is also possible that two molecules of 2 would couple in a similar
fashion and then disproportionate to give 1 along with the
observed products.) While this mechanism is consistent with the
available data, further spectroscopic information is necessary to
confirm the structure of 3 and to verify the final step in the
mechanism. We are currently attempting to generate and
characterize 3 at low temperatures. Results from these experi-
ments will be reported in due course.
These experiments identify hydrogen abstraction as an impor-
tant pathway in the direct photolysis of R-keto protecting groups.
It is further argued that direct homolysis occurs inefficiently, if
at all. Of course, this does not exclude the existence of other
decay mechanisms. Indeed compelling evidence has been
presented for a photohomolysis/neighboring group assisted electron-
transfer pathway in the decay of 4-hydroxyphenacyl phosphate
esters.24 Likewise, it appears that the benzoin derivatives can
14
donor of moderate reactivity) than it was in benzene. Givens
showed a deuterium isotope effect in the photolysis of 4-meth-
oxyphenacyl phosphate esters in CH OH. The latter workers also
3
proposed a pathway involving formation of a ketyl radical. In
that case they suggested that unimolecular decay of the latter
would give an enol along with a phosphate-based radical.
(
2) Laser flash photolysis (LFP) experiments show the excited
triplet state of the ester is formed upon direct photolysis. Figure
A shows the transient UV-vis spectrum obtained following
2
pulsed laser (308 nm, 10 ns, 50 mJ/pulse) photolysis of 1 in
benzene. The spectrum consists of a broad absorption band with
a maximum near 340 nm. This species decays following clean
first-order kinetics fitting to a lifetime of 750 ns (benzene, Ar-
purged). This species is assigned to the excited triplet state of
the phenacyl group because its lifetime is diminished in the
presence of O
state spectrum of acetophenone.
3) The excited triplet state abstracts a H-atom from various
2
and because of its resemblance to the excited triplet
15,16
decay through a mechanism involving photocleavage coupled with
cyclization to form a benzofuran derivative.1
4,25,26
The relative
(
donors to form the ketyl radical 2. For example, addition of
increasing amounts of CHD to solutions of 1 increases the decay
rate of the excited triplet state. A pseudo-first-order analysis of
efficiencies for each of the possible pathways is undoubtedly
controlled by factors such as substitution on the PRPG, the nature
of the leaving group, the solvent polarity, and the presence of
H-atom donors. Future research directed at understanding how
these various factors determine the decay mechanism actually
followed will be helpful in designing improved PRPGs.
8
-1
27
the decays provides second-order rate constants of 5.2 × 10 M
-
1
8
-1 -1
s
for CHD, 6.2 × 10 M
s
for tributyltin hydride, and 8.8
6
-1 -1
×
10 M
s
for 2-propanol. The latter compares favorably to
6
-1 -1
17
the reported value of 2.13 × 10 M
s
for acetophenone. In
Acknowledgment. We are grateful to the NSF and the Graduate
Research Board of the University of Maryland for partial support.
addition to quenching the triplet absorption, the H-atom donors
cause formation of a new, longer-lived transient species having
a maximum near 310 nm and a much weaker tail extending to
JA971431T
4
25 nm (Figure 2B). Such an absorption spectrum is character-
(18) Haspra, P.; Sutter, A.; Wirz, J. Angew. Chem., Int. Ed. Engl. 1979,
18, 617.
istic of ketyl radicals formed from other acetophenone deriva-
tives.16 This spectrum is therefore assigned to 2.
(19) Chiang, Y.; Kresge, A. J.; Capponi, M.; Wirz, J. HelV. Chim. Acta
1
986, 69, 1331.
The ketyl radical 2 decays to a long-lived (τ >1 ms) species
that absorbs in a similar region of the spectrum as illustrated in
Figure 2B. The precise nature of this long-lived transient is still
under investigation. However, several facts are relevant. First,
(20) Demeter, A.; L a´ szl o´ , B.; B e´ rces, T. Ber. Bunsen-Ges. Phys. Chem.
988, 92, 1478.
1
(21) Demeter, A.; B e´ rces, T. J. Photochem. Photobiol., A 1989, 46, 27.
22) Chilton, J.; Giering, L.; Steel, C. J. Am. Chem. Soc. 1976, 98, 1865.
(
(23) Pitts, J. N.; Letsinger, R. L.; Taylor, R. P.; Patterson, J. M.;
Recktenwald, G.; Martin, R. B. J. Am. Chem. Soc. 1959, 81, 1068.
(24) Park, C.-H.; Givens, R. S. J. Am. Chem. Soc. 1997, 119, 2453.
(25) Corrie, J. E. T.; Trentham, D. R. J. Chem. Soc., Perkin Trans. 1 1992,
2409.
(26) Sheehan, J. C.; Wilson, R. M. J. Am. Chem. Soc. 1964, 86, 5277.
(27) The desired cleavage pathway would be dictated by the particular
application. For example, while the chemistry described here provides high
yields of carboxylic acids, it does not result in prompt release of the acids.
(
14) Givens, R. S.; Athey, P. S.; Matuszewski, B.; Kueper, L. W., III; Xue,
J.-y. J. Am. Chem. Soc. 1993, 115, 6001.
15) Ledger, M. B.; Porter, G. J. Chem. Soc., Faraday Trans. 1972, 68,
(
5
39.
(
16) Lutz, H.; Breheret, E.; Lindqvist, L. J. Phys. Chem. 1973, 77, 1758.
17) Berger, M.; McAlpine, E.; Steel, C. J. Am. Chem. Soc. 1978, 100,
(
5
147.