Photochemistry of o-vinylbenzaldehyde: Formation of a ketene methide intermediate and its trapping with secondary amines

Article abstract of DOI:10.1021/ja954111n

Irradiation of o-vinylbenzaldehyde (1) in the presence of primary and secondary amines results in N-H addition across the two chromophores of 1, to give o-ethylbenzamides 6. Photoreaction of deuterium-labeled aldehyde 15 with piperidine gave an amide (17)

Full text of DOI:10.1021/ja954111n

J. Am. Chem. Soc. 1996, 118, 4361-4365
4361
Photochemistry of o-Vinylbenzaldehyde: Formation of a
Ketene Methide Intermediate and Its Trapping with Secondary
Amines
,†
†
,‡
‡,§
,⊥
S. V. Kessar,* A. K. S. Mankotia, J. C. Scaiano,* M. Barra, J. Gebicki,* and
K. Huben
⊥
Contribution from the Department of Chemistry, Panjab UniVersity, Chandigarh 160 014, India,
Department of Chemistry, UniVersity of Ottawa, Ottawa, Ontario K1N 6N5, Canada, and Institute
of Applied Radiation Chemistry, Technical UniVersity, 90-924 Lodz, Poland
X
ReceiVed December 7, 1995
Abstract: Irradiation of o-vinylbenzaldehyde (1) in the presence of primary and secondary amines results in N-H
addition across the two chromophores of 1, to give o-ethylbenzamides 6. Photoreaction of deuterium-labeled aldehyde
1
5 with piperidine gave an amide (17) carrying a deuterium at the â-carbon of the ethyl side chain whereas use of
N-deuteriopiperidine led to deuterium incorporation at the R-carbon (14). These results are explained on the basis
of a 1,5 hydrogen shift in the excited state of 1 to give a ketene methide intermediate (7) which becomes trapped
with amines. Upon 308 nm laser excitation of 1 in acetonitrile or benzene solution, a weak transient absorption
having λmax at 380 nm was observed. Irradiation of 1 isolated in an argon matrix with 313 nm light also revealed
formation of an intermediate with a UV absorption maximum around 380 nm. Its IR spectrum displayed characteristic
-
1
ketene stretching vibrations at 2086 and 2098 cm , providing definitive support for the ketene methide structure 7.
Introduction
As reported earlier,1,2 irradiation of 1 in the presence of
secondary amines did not afford products characteristic of
benzaldehyde or styrene. Instead, an amidic adduct entailing
N-H addition across the two chromophores of 1 was obtained
in good yield (6; 80-86%). It was suggested that, on excitation
of 1, a 1,5 hydrogen shift occurs to give a ketene methide
intermediate which becomes trapped by the amine (1 f 7 f
In the context of the interest in novel transformations of
bichromophoric molecules, we have reported preliminary results
on the photochemistry of o-vinylbenzaldehyde (1).1
-3
This
substrate combines features of two very extensively studied
chromophores, i.e., benzaldehyde (2) and styrene (3). Its
4
photoreactions with amines are of particular interest since on
irradiation in the presence of amines, both arylcarbonyl and
arylalkenyl compounds are known to form adducts, albeit of
different types. In the case of benzaldehydes, R-C-H adducts
6
). This proposal was supported by deuterium-labeling experi-
1,2
ments. Although ketene methide intermediacy is well estab-
lished for some elimination and ring-opening reactions,
6-8
formation of 7 from 1 involves an unprecedented migration of
an acyl hydrogen to a vinyl terminus. Therefore, additional
spectroscopic evidence on this point was sought. Meanwhile,
Wilson and co-workers reported that low-intensity irradiation
(2 f 4) are obtained, whereas with styrenes, linkage through
the nitrogen atom is the predominant reaction course (3 f 5)
5
(see Scheme 1). The mechanism proposed in both cases
involve electron transfer quenching of the substrate excited state
by ground state amine to form an exciplex which undergoes
proton transfer to yield a radical pair which, in turn, can
combine, disproportionate, or diffuse apart.5 However, in
carbonyl compounds the photoreaction occurs largely from the
triplet manifold, whereas a singlet reaction has been proposed
for styrenes.5
of o-(diphenylvinyl)benzaldehyde (8) does not afford products,
such as 10, expected from a ketene methide intermediate (9).9
Transient spectroscopic studies also indicated that no strongly
absorbing transients were produced on such irradiation of 8.
However, under high-intensity argon laser-jet irradiation, 10 was
obtained in 12-16% yield. It was argued that, in the lowest
excited state of 8, a [1,5s] sigmatropic shift of hydrogen, to give
1
0
9
, is symmetry disallowed. On the other hand, some of the
†
Panjab University, coordinating center for this study.
‡
§
higher excited states of 8 have a symmetry suitable for an
University of Ottawa.
Present address: Department of Chemistry, University of Waterloo,
Waterloo, Ontario N2L 3G1, Canada.
(6) (a) Krantz, A. J. Am. Chem. Soc. 1974, 96, 4992. (b) Bally, T.;
Michalak, J. J. Photochem. Photobiol., 1992, 69, 185.
(7) (a) Hacker, N. P.; Turro, N. J. J. Photochem. 1983, 22, 131. (b)
Schiess, P.; Eberle, M.; Huys- Francotte, M.; Wirz, J. Tetrahedron Lett.
1984, 25, 2201.
(8) For a general discussion of the reactivity of ketene methide
intermediates see: (a) Kessar, S. V. Pure Appl. Chem. 1987, 59, 381. (b)
Kessar, S. V.; Singh, P.; Vohra, R.; Kaur, N. P.; Venugopal, D. J. Org.
Chem. 1992, 57, 6716. (c) Kuzuya, M.; Miyake, F.; Okuda, T. Tetrahedron
Lett. 1980, 21, 2185.
⊥
Technical University of Lodz.
Abstract published in AdVance ACS Abstracts, April 1, 1996.
1) Kessar, S. V.; Mankotia, A. K. S.; Gujral, G. J. Chem. Soc., Chem.
X
(
Commun. 1992, 840.
2) Kessar, S. V.; Mankotia, A. K. S. J. Chem. Soc., Chem. Commun.
993, 1828.
3) In bichromophoric molecules novel photoreactions not manifest in
(
1
(
individual chromophores can arise due to mutual interaction: Morrison,
H. Acc. Chem. Res. 1979, 12, 383.
(
4) Gilbert, A.; Baggot, J. Essentials of Molecular Photochemistry, 1st
ed.; Blackwell Scientific Publications: Oxford, 1991; Chapters 6 and 7.
5) (a) Cohen, S. G.; Parola, A.; Parsons, G. H., Jr. Chem. ReV. 1973,
3, 141. (b) Lewis, F. D. Acc. Chem. Res. 1986, 19, 401. (c) Lewis, F.
(9) (a) Wilson, R. M.; Schnapp, K. A. Chem. ReV. 1993, 93, 223. (b)
Wilson, R. M.; Patterson, W. S.; Austen, S. C.; Douglas, M. Ho.; Bauer, J.
A. K. J. Am. Chem. Soc. 1995, 117, 7820.
(
7
(10) It was pointed out by Wilson and co-workers that while intramo-
lecular hydrogen abstraction by vinylic carbon is known, it is rarely observed
with olefins that are not conjugated to heteroatoms or constrained in small
D.; Bassani, D. M.; Burch, E. L.; Cohen, B. E.; Engleman, J. A.; Reddy,
G. D.; Schneider, S.; Jaeger, W.; Hartman, P. G.; Scheliga, P. J. Am. Chem.
Soc. 1995, 117, 660. (d) Photoinduced Electron Transfer; Fox, M. A.,
Chanon, M., Eds.; Elsevier: Amsterdam, 1988; Part C. (e) Chanon, M.;
Rajzmann, M.; Chanon, F. Tetrahedron 1990, 18, 6193.
9
rings: (a) Mawhorter, L. G.; Carlson, S. E.; Bedont, R. A. J. Org. Chem.
1979, 44, 3698. (b) McCullough, J. J. Acc. Chem. Res. 1980, 13, 270. (c)
Hornback, J. M.; Barrows, R. D. J. Org. Chem. 1982, 47, 4285.
S0002-7863(95)04111-4 CCC: $12.00 © 1996 American Chemical Society

4
362 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Kessar et al.
Scheme 1
dienophiles were unsuccessful. Thus, from photolysis of a
solution of 1 in acetonitrile containing excess (∼10 M)
cyclohexene, methyl acrylate, acrylonitrile, dimethyl fumarate,
1
1
or maleic anhydride, no adducts could be isolated.
Quantum Yields. The efficiency of the reaction of 1 with
amines was investigated by measuring quantum yields for
disappearance of 1. The quantum yield for disappearance of 1
in the presence of an equimolar concentration (0.0034 M) of
piperidine in acetonitrile was found to be ∼0.06. According
to the proposed reaction path, this quantum yield should be
dependent on the formation of 7 as well as on its trapping with
the added amine. The quantum yield did, in fact, increase with
increasing amine concentration (φ ≈ 0.17 at 0.1 M piperidine)
but not in a linear manner. The standard double reciprocal plot
of quantum yield vs amine concentration showed a downward
curvature which was irregular and not easily amenable to kinetic
1
2
analysis. A similar dependence on amine concentration was
observed in decay of the transient 7 generated by laser flash
photolysis (vide infra).
Photoaddition of amines to 1 was not appreciably sensitive
to solvent polarity, and disappearance quantum yields in benzene
were similar to those obtained in acetonitrile. Addition of triplet
quenchers (0.0034 M) such as trans-piperylene or cyclohexa-
1
,3-diene did not depress the quantum yield. Similarly, when
prior nitrogen purging of the photolysate was omitted, the
oxygen, dissolved in the solvent under air-saturated conditions,
had no adverse effect on the reaction efficiency. These
experiments suggest involvement of a singlet or a very short
lived triplet excited state.
Deuterium-Labeling Studies. Irradiation of an equimolar
mixture of 1 and N-deuteriopiperidine afforded an amide, the
1
H NMR spectrum of which revealed deuterium incorporation
at the R-carbon of the side chain (14). On the other hand,
photoreaction of deuterium-labeled aldehyde 15 with piperidine
gave an amide (17) carrying a â-deuterium atom in the side
chain. These results are in conformity with ketene methide
intermediacy as shown in Scheme 2. Evidence for an isotopic
effect in the hydrogen transfer step, leading to the intermediate
7
, could be obtained from photolysis of partially deuterated
o-vinylbenzaldehyde. Thus, irradiation of a mixture of 15 (78%)
and 1 (22%) in the presence of an equimolar quantity of
1
13
piperidine revealed, by H NMR spectral analysis, acyl
deuterium enrichment in the recovered o-vinylbenzaldehyde
(84% 15 after 20 min and 96% 15 after 65 min). It is clear
that, under these conditions, 15 is consumed at a slower rate
than 1 (k5 > k6). Interestingly, irradiation of partially deuterated
aldehyde in the absence of amine led to depletion of the acyl
deuterium. Thus, when a mixture of 15 (78%) and 1 (22%)
was irradiated for 75 min, the H NMR spectrum of the
recovered aldehyde showed the presence of 46% deuterium and
allowed [1,5s] sigmatropic shift, and this course is probably
followed under high-intensity irradiation conditions. In this
context, our results on the photochemistry of 1 are even more
interesting, and are now discussed in detail along with additional
flash photolysis and matrix isolation evidence.
1
9
5
4% protium at the acyl position. The deuterium lost from the
acyl carbon was incorporated into the side chain by formation
of 18 (∼16%) + 19 (∼14%). These results can be understood
in terms of reversion of 16 to 15 or, after rotation about the
C-C single bond, to 18 and 19. The latter process results in
replacement of the acyl deuterium of 15 with protium from the
â-carbon of the side chain. It seems that in the presence of
added piperidine this reversion is suppressed to a large extent
4
Results and Discussion
Preparative Experiments. Irradiation of an equimolar
solution of aldehyde 1 and piperidine (0.0034 M) in acetonitrile
with Pyrex-filtered UV light from a medium-pressure mercury
lamp gave the amide 6a in 85% isolated yield. Similarly,
amides 6b (86%) and 6c (85%) were obtained from reaction of
(k [amine] . k1-3).
(
11) On irradiation of 1 in acetonitrile, without any additives, slow
dimerization of 1 was observed (refs 1 and 2).
12) For a similar observation in photoreactions of carbonyl compounds
1
with diethylamine and benzylamine, respectively. Photolysis
of dimethoxy aldehyde 11 in the presence of piperidine produced
the amide 12 (80%). Irradiation of an equimolar solution of 1
and methanol in acetonitrile did not afford the corresponding
methanol adduct 13. However, when methanol was used in 10
M excess or as a solvent, the ester 13 was obtained in low yield
(
with amines see: Parola, A. H.; Rose, A. W.; Cohen, S. G. J. Am. Chem.
Soc. 1975, 97, 6202.
(13) In experiments with deuterated substrate 15, the extent of acyl
deuterium incorporation was estimated from the ratio of the acyl hydrogen
1
signal at δ 10.28 to an aryl proton signal at δ 7.85 in the H NMR spectrum
(30%). Efforts to trap the proposed intermediate 7 with added
recorded on a 300 MHz Bruker instrument.

Photochemistry of o-Vinylbenzaldehyde
J. Am. Chem. Soc., Vol. 118, No. 18, 1996 4363
Figure 2. Transient absorption spectrum recorded immediately after
308 nm laser excitation of o-vinylbenzaldehyde in acetonitrile solution.
Inset: calculated electronic spectrum for an intermediate ketene.
Figure 1. Transient decay traces (monitored at 380 nm) recorded upon
308 nm laser excitation of o-vinylbenzaldehyde in acetonitrile solution
in the absence (A) and in the presence (B) of piperidine (0.1 M).
Scheme 2
Figure 3. Observed transient decay constant monitored at 380 nm in
acetonitrile (b) and methanol (O) solutions as a function of piperidine
concentration.
Figure 4. Difference electronic absorption spectrum of o-vinylben-
zaldehyde isolated in an argon matrix, obtained by subtraction of the
spectrum recorded before irradiation from the spectrum recorded after
irradiation for 180 min through a 313 nm interference filter.
Irradiation of 1 isolated in an argon matrix (10 K) with 313
nm light revealed the formation of an intermediate with a UV
absorption maximum around 380 nm (Figure 4). Its IR spectrum
displayed characteristic ketene stretching vibrations at 2086 and
Laser Flash Photolysis and Matrix Isolation. Upon 308
nm laser excitation of 1 in acetonitrile or benzene solution, a
weak transient absorption (τ > 100 µs) having λmax at 380 nm
was observed (Figures 1 and 2). Neither the decay nor the yield
of this transient was affected by oxygen. This absorption was
tentatively assigned to ketene methide 7. The electronic
spectrum generated for this intermediate through MOPAC/
ZINDO calculations showed a single absorption band, in the
range 300-700 nm, with a maximum at 390 (Figure 2, inset),
in good agreement with experimental observation.
-
1
2098 cm , providing definitive support for the proposed
intermediate 7 (Figure 5). The UV and IR data obtained in the
present work are similar to those reported for ketene methide
-
1
intermediate 21 (λmax ) 390 nm, νCdCdO ) 2075-2115 cm )
formed on matrix isolation or flash photolysis of benzocy-
6,7
clobutenone (20). Generated in an argon matrix, 7 undergoes
secondary photochemical reaction with 365 nm light to form
2-methylbenzocyclobutenone (22) (νCdO ) 1776 and 1811
-
1
cm ) (Scheme 3).
Amines were found to effectively quench the 380 nm transient
(
Figure 1). Quenching rate constants for piperidine showed a
Formation and Reactivity of Ketene Methide Intermediate
7. Formation of 7 from 1 involves a novel transfer of an acyl
hydrogen to a vinylic carbon. In fact, in view of the ubiquity
non-first-order dependence on amine concentration (Figure 3)
in agreement with the quantum yield studies (vide supra).

4
364 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Kessar et al.
was calculated to be less stable than 23 only by 0.3 kcal).
Moreover, carbonyl and vinyl stretching vibrations of matrix-
isolated aldehyde 1 point toward a mixture of rotamers trapped
from the gas phase. As an alternate explanation, some charge
transfer character in the excited state of 1 favoring hydrogen
(hydride) movement in the observed direction may be consid-
ered, e.g., 26 f 7. However, lack of sensitivity of the
photoreaction to solvent polarity does not support this conten-
tion. It seems that the observed hydrogen shift can be best
rationalized in terms of the large difference in the strength of
acyl (∼87 kcal/mol) and vinyl (∼107 kcal/mol) carbon-
1
5b
hydrogen bonds; i.e., the reaction is driven by the thermo-
9
dynamic factors. The observation of Wilson et al. that low-
intensity irradiation of 8 does not lead to a ketene methide
intermediate is perhaps a consequence of insufficient energy
or a steric constraint to a [1,5a] sigmatropic shift in the lowest
1
6
excited state of this highly conjugated molecule.
Figure 5. Expanded ketene stretching region of the IR spectrum of
o-vinylbenzaldehyde isolated in an argon matrix after irradiation for
Finally, the low reactivity of 7 in comparison to that reported
8
for the ketene methide 21 merits comment. The latter
1
50 min through a 313 nm interference filter.
intermediate, when generated by photolysis of 20, can be readily
Scheme 3
7b
trapped with dienophiles or methanol. In contrast, we failed
4
2
to obtain π + π adducts from 7 (vide supra), and its reaction
with methanol was slow. This could be a consequence of rapid
retro hydrogen transfer (7 f 23) of the type shown by
deuterium-labeling experiments (vide supra). A similar pathway
is not available to 21 as it is devoid of the necessary allylic
hydrogen. The situation is somewhat reminiscent of enols
formed on photolysis of aromatic carbonyl compounds carrying
an o-alkyl substituent. The (Z)-enols rapidly revert back and
1
7
cannot be trapped with dienophiles nor exchange hydroxylic
18
hydrogen, unlike the (E)-enols. However, electronic or steric
factors may also be responsible for the low reactivity of 1
1
9
compared to 21.
Conclusions
We have shown that on UV irradiation of o-vinylbenzalde-
hyde (1) an unprecedented monophotonic shift of the acyl
hydrogen to the â-vinyl carbon occurs. The formed ketene
methide intermediate 7 reverts to 1 by back hydrogen transfer
but can be trapped effectively with secondary amines to afford
o-ethylbenzamides. Failure to observe any transients or an
adverse effect of triplet quenchers indicates involvement of a
rapid, possibly concerted singlet pathway for the formation of
7
from 1. As such, a [1,5a] sigmatropic shift in the first excited
state of 1 may be involved in this transformation.
Experimental Section
Materials. o-Vinylbenzaldehyde was prepared from 3,4-dihydroiso-
quinoline20 (bp 60-65 °C, 1 Hgmm), as reported by Dale et al., and
purified by distillation (bp 113-115 °C, 18 Hgmm) and column
chromatography over silica gel (Acme synthetic chemicals, 100-200
21
of the light-induced hydrogen abstraction by the carbonyl group,
hydrogen transfer in the other direction (24 f 25) could have
been envisaged.1 We have found no chemical or spectroscopic
evidence for the latter reaction course. This could be a
consequence of an unfavorable conformational equilibrium in
the ground state of o-vinylbenzaldehyde, i.e., a low population
of 24 which is the conformer suitable for hydrogen transfer
(
15) (a) Wagner, P. J. In Handbook of Organic Photochemistry and
Photobiology; Horspool, W. M., Song, Pill-Soon, Eds.; CRC Press: Boca
Raton, FL, 1995; p 449. (b) Lowry, T. H.; Richardson, K. S. Mechanism
and Theory in Organic Chemistry, 2nd ed.; Harper and Row: New York,
4
1
981; p 147.
16) In a footnote Wilson et al. refer to our unpublished flash photolysis
results with 1 and also mention these possibilities (ref 9b).
17) (a) Porter, G.; Tchir, M. F. J. Chem. Soc. A 1971, 3772. (b) Wagner,
(
(
P. J.; Chem, C. P. J. Am. Chem. Soc. 1976, 98, 239. (c) Scaiano, J. C.
Acc. Chem. Res. 1982, 15, 252. (d) Scaiano, J. C.; Netto-Ferreira, J. C. J.
Am. Chem. Soc. 1991, 113, 5800.
(18) (a) Gebicki, J.; Reimschussel, W.; Zurawinska, B. J. Phys. Org.
Chem. 1990, 3, 38. (b) Michalak, J.; Wolszczak, M.; Gebicki, J. J. Phys.
Org. Chem. 1993, 6, 465.
15a
leading to 25.
.0) do not show any appreciable difference in stability of
minimum energy conformations approximating 23 and 24 (24
However, AM1 calculations (MOPAC version
6
(14) Transfer of the side chain alkyl hydrogen in styrenes has been
reported only occasionally (see ref 10). The shift of the acyl hydrogen to
the carbonyl oxygen in phthalaldehyde is well documented: (a) Scaiano,
J. C.; Encinas, M. V.; George, M. V. J. Chem. Soc., Perkin Trans. 2 1980,
(19) It may be noted that 9 is also sluggish in its reaction with methanol
(ref 9). Further, methyl substitution in the ortho position of 21 drastically
depresses its reactivity toward methanol (ref 7).
(20) Pelletier, J. C.; Cava, M. D. J. Org. Chem. 1987, 52, 616.
(21) Dale, W. J.; Starr, L.; Strobel, C. W. J. Org. Chem. 1961, 26, 2225.
7
7
24. (b) Gebicki, J.; Kuberski, S. J. Chem. Soc., Perkin Trans. 2 1990,
65.

Photochemistry of o-Vinylbenzaldehyde
J. Am. Chem. Soc., Vol. 118, No. 18, 1996 4365
mesh). Compound 8 was secured from 6,7-dimthoxy-3,4-dihydroiso-
quinoline (bp 155-160 °C, 1 Hgmm) in a similar way. N-
Deuteriopiperidine was obtained by repeated shaking of piperidine with
308 nm, ∼ 6 ns, e40 mJ/pulse). A flow sample cell (constructed of
2
2
2
7 × 7 mm Suprasil tubing) was employed in order to ensure that fresh
solution was irradiated by each laser pulse. Solutions were deaerated
by bubbling with oxygen-free nitrogen through a container attached to
the photolysis flow cell with Teflon lines. The temperature of the
photolysis samples was 20-22 °C. The system is similar to that
discussed in earlier publications.25
D
2
O, drying over CaH
the acyl hydrogen of l, it was converted into an R-siloxynitrile, bp 105-
08 °C, 4-5 Hgmm, which was exposed to LTMP (2 equiv) at -78
C in THF for 1 h and the reaction mixture quenched with D O.
Representative Procedure for Preparative Scale Reaction of 1
2
, and distillation. For deuterium exchange of
1
°
2
Calculations. Molecular modeling calculations were performed on
a Tektronix CAChe workstation (software version 3.5) using the AM1
semiempirical method, as part of the MOPAC program version 6.1.
Electronic spectra were calculated using the ZINDO program.
with Amines. A solution of o-vinylbenzaldehyde (1) (135 mg, 1.02
mmol) and piperidine (100 mg, 1.15 mmol) in 300 mL of acetonitrile
was irradiated under N with Pyrex-filtered light from a 450 W medium-
2
pressure Hanovia lamp, for 5 h. Concentration of the photolysate gave
a residual oil which was dissolved in solvent ether (30 mL). The
ethereal solution was washed successively with 10% hydrochloric acid
Matrix Isolation Studies. Since the sample pressure was insuf-
ficient to prepare a gaseous mixture, argon was passed over a solid
substrate placed in the thermostated tube connected to the deposition
line behind the needle valve. The tube was kept at -8 °C during
(
2 × 10 mL) and saturated aqueous sodium bicarbonate (2 × 10 mL).
After drying over anhydrous sodium sulfate, the solvent was evaporated
in vacuo to obtain an oil (206 mg) which was subjected to silica gel
-
1
deposition. Argon was deposited with a flow rate of 30 µmol min
,
and the total amount deposited varied from 0.9 to 1.8 mmol. The sample
vapor dilued by the argon carrier stream was condensed onto a cesium
iodide window mounted on the expander stage of a Displex DE-202
column chromatography (petroleum ether (60-80 °C)-ethyl acetate,
1
4
:1), yielding 191 mg (86%) of amide 6a as an oil: H NMR (90 MHz,
CCl
.86-3.33 (m, 2H, NCH
ArH); LRMS m/e 217, 216, 133, 132; HRMS for C14
17. 1467, found 217. 1461. The spectral data of 6a completely
matched those of an authentic sample prepared by reaction of
4
) δ -1.18 (t, 3H, CH
3
), 1.60 (br s, 6H, 3CH
2
), 2.60 (q, 2H, CH
), 6.96-7.40 (m, 4H,
19NO, m/e calcd
2
),
(APD Cryogenics) cryostat. During deposition the window support
2
2
), 3.70 (br s, 2H, NCH
2
was kept at 20 K. The matrix ratio was estimated to be 600-700.
H
2
Photochemical reactions were carried out with a high-pressure
mercury lamp, HBO-200 (Narva). Interference filters FS 10-50 (Oriel)
were used. IR spectra of matrix-isolated species were obtained wth
an MB 100 (Bomen) FT-IR spectrometer equipped with a wide-band
2
-ethylbenzoyl chloride with piperidine.
Quantum Yield Measurements. All quantum yield measurements
-1
were carried out in matched quartz tubes in a merry-go-round apparatus
using 313 nm light from a medium-pressure mercury lamp, filtered
DTGS detector working at a resolution of 1 cm . UV-vis absorption
spectra were measured with a PU 8710 (Philips) spectrophotometer.
The irradiation and measurements were performed at 10 K.
-3
through a 2.0 × 10 M solution of potassium chromate in 5% aqueous
2
3
-3
potassium carbonate. A concentration of 3.4 × 10 M 1 was used,
and its disappearance, up to 15% conversion, was monitored by its
UV absorption at 308 nm. A benzophenone-benzhydrol actinometer
Acknowledgment. This work was supported in part by the
DST India and Polish Committee for Scientific Research. J.
C. S. is supported by a research grant from the Natural Sciences
and Engineering Research Council of Canada and by a Killam
Fellowship from the Canada Council.
2
4
was used for light intensity measurement. φ values are based on an
average of at least three runs (φ, (10%).
Laser Flash Photolysis Studies. Laser experiments were carried
out using the pulses from a Lumonics EX-510 excimer laser (Xe/HCl,
JA954111N
(
(
(
22) Popp, F. D.; Mcewen, E. W. J. Am. Chem. Soc. 1957, 79, 3773.
23) Wagner, P. J.; Ersfeld, D. A. J. Am. Chem. Soc. 1976, 98, 4515.
24) Moore, W. M.; Hammond, G. S.; Foss, R. P. J. Am. Chem. Soc.
(25) (a) Scaiano, J. C. J. Am. Chem. Soc. 1980, 102, 7747. (b) Scaiano,
J. C.; Tanner, M.; Weir, D. J. Am. Chem. Soc. 1985, 107, 4396.
1
961, 83, 2789.