A solvent effect that influences the preparative utility of N- (silylalkyl)phthalimide and N-(silylalkyl)maleimide photochemistry

Article abstract of DOI:10.1021/jo990087a

The photochemistry of selected N-silylalkyl-substituted phthalimides and maleimides has been investigated with the aim of exploring the generality and preparative consequences of an intriguing solvent effect on excited-state reaction chemoselectivities and quantum efficiencies. An example of this effect is found in the photochemistry of N- [(trimethylsilyl)butyl]phthalimide 10, where irradiation in MeCN leads to production of a mixture of four products that arise by excited-state intramolecular hydrogen-atom abstraction. In contrast, the benzoindolizidine 15 is the sole product produced by a single electron transfer (SET)- desilylation pathway upon irradiation of 10 in 35% H₂O-MeCN. Another example of this solvent effect is found in the photochemistry of the N-silylpropyl- maleimide 17. Irradiation in MeCN results in the production of the 2+2-dimer 19 whereas the pyrrolizidine 18 is generated exclusively by irradiation of 17 in 35% H₂O-MeCN. The results of fluorescence and triplet sensitization experiments suggest that the solvent effect has multiple sources including the control of the nature, reactivity, and intrinsic lifetimes of singlet and triplet excited states of the phthalimide and maleimide systems. The exploratory studies have clearly demonstrated the generality of this solvent effect and how it can be used to enhance the preparative utility of the photochemistry of N-(silylalkyl)phthalimides and N-(silylalkyl)maleimides.

Full text of DOI:10.1021/jo990087a

J . Org. Chem. 1999, 64, 4411-4418
4411
A Solven t Effect Th a t In flu en ces th e P r ep a r a tive Utility of
N-(Silyla lk yl)p h th a lim id e a n d N-(Silyla lk yl)m a leim id e
P h otoch em istr y
,
†
†
†
‡
‡
Ung Chan Yoon,* Sun Wha Oh, Soo Min Lee, Sung J u Cho, J anet Gamlin, and
Patrick S. Mariano*,
‡
Department of Chemistry, College of Natural Sciences, Pusan National University, Pusan 609-735, Korea,
and Department of Chemistry, University of New Mexico, Albuquerque, New Mexico 87131
Received J anuary 19, 1999
The photochemistry of selected N-silylalkyl-substituted phthalimides and maleimides has been
investigated with the aim of exploring the generality and preparative consequences of an intriguing
solvent effect on excited-state reaction chemoselectivities and quantum efficiencies. An example of
this effect is found in the photochemistry of N-[(trimethylsilyl)butyl]phthalimide 10, where
irradiation in MeCN leads to production of a mixture of four products that arise by excited-state
intramolecular hydrogen-atom abstraction. In contrast, the benzoindolizidine 15 is the sole product
produced by a single electron transfer (SET)-desilylation pathway upon irradiation of 10 in 35%
H
2
O-MeCN. Another example of this solvent effect is found in the photochemistry of the
N-silylpropyl-maleimide 17. Irradiation in MeCN results in the production of the 2+2-dimer 19
whereas the pyrrolizidine 18 is generated exclusively by irradiation of 17 in 35% H O-MeCN. The
2
results of fluorescence and triplet sensitization experiments suggest that the solvent effect has
multiple sources including the control of the nature, reactivity, and intrinsic lifetimes of singlet
and triplet excited states of the phthalimide and maleimide systems. The exploratory studies have
clearly demonstrated the generality of this solvent effect and how it can be used to enhance the
preparative utility of the photochemistry of N-(silylalkyl)phthalimides and N-(silylalkyl)maleimides.
In tr od u ction
Sch em e 1
Phthalimides have been the subject of a number of in-
depth photochemical1 and photophysical2 studies in
recent years due to the developing applications of the
3
excited-state chemistry of these substances. Exploratory
investigations of preparative aspects of phthalimide
photochemistry have uncovered the operation of two
major excited-state reaction pathways. These include the
sometimes indistinguishable processes initiated by inter-
or intramolecular hydrogen-atom abstraction and single
electron transfer (SET) (Scheme 1). By virtue of our joint
4
interests in phthalimide photochemistry and organosi-
5
lane SET chemistry, we became attracted to photochemi-
Sch em e 2
cal studies with linked alkylsilane-phthalimide systems.
Our work in this area, thus far, has demonstrated the
existence of interesting excited-state reactivity profiles
of substances in this family. An example is found in the
azomethine ylide 2 forming, excited-state transforma-
tions of N-(silylmethyl)phthalimides 1 and their male-
imide analogues that initiate novel and useful photocy-
cloaddition reactions (Scheme 2).6
More pertinent to the studies described below are
7
observations made in our preliminary investigations of
†
Pusan National University.
University of New Mexico.
‡
(
1) Coyle, J . D. In Synthetic Organic Photochemistry; Horspool, W.
M., Ed.; Plenum Press: New York, 1984.
2) Aveline, B. M.; Matsuga, S.; Redmond, R. W. J . Am. Chem. Soc.
997, 119, 11785.
3) Brana, M. F.; Castellano, J . M.; Roldan, C. M.; Santos, A.;
Vasquez, D.; J imenez, A. J . Med. Chem. 1981, 16, 207.
4) Yoon, U. C.; Kim, H. J .; Mariano, P. S. Heterocycles 1989, 29,
041; Yoon, U. C.; Oh, J . H.; Lee, S. J .; Kim, D. U.; Lee, J . G.; Kang,
K. T.; Mariano, P. S. Bull. Korean Chem. Soc. 1992, 13, 166. Yoon, U.
C.; Lee, S. J .; Cho, S. J .; Lee, C. W.; Mariano, P. S. Bull. Korean Chem.
Soc. 1994, 15, 154. Yoon, U. C.; Kim, J . W.; Ryu, J . Y.; Cho, S. J .; Oh,
S. W.; Mariano, P. S. J . Photochem. Photobiol. 1996, 106, 145.
the photochemistry of higher homologues of 1, specifically
the N-(silylethyl)- and N-(silylpropyl)phthalimides 3 and
8. Particularly intriguing was the finding that irradiation
(
1
(
of 3 in MeCN followed by the addition of D
production of the benzazepinedione-d 4. In contrast, the
benzazepinedione-d 7 is generated exclusively by pho-
toreaction of 3 in 35% D O-MeCN. These observations
2
O leads to
(
2
1
1
2
were rationalized by invocation of a solvent-dependent
competition between two mechanistic pathways for ben-
(5) Yoon, U. C.; Mariano, P. S. Acc. Chem. Res. 1992, 25, 233.
1
0.1021/jo990087a CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/18/1999

4
412 J . Org. Chem., Vol. 64, No. 12, 1999
Yoon et al.
Sch em e 3
Sch em e 4
lective photoreaction of 8 into one that has both a high
chemical yield and quantum efficiency. Driven by a desire
to determine if this solvent effect is general and ap-
plicable in controlling the excited-state reactivity of
phthalimides and other conjugated imides, we have
conducted exploratory photochemical studies with a
variety of N-(silylalkyl)phthalimides and -maleimides.
Below are presented the results of these investigations,
which demonstrate the ubiquitous nature of the solvent
effect in controlling the chemoselectivities and efficiencies
of the (silylalkyl)phthalimide and -maleimide photore-
actions.
zazepinedione formation (Scheme 3). We proposed that
the reactive excited state of 3 in MeCN undergoes H-atom
abstraction, giving eventually the equilibrating phthal-
imido-cyclobutanol 5-R-silylbenzazepinedione 6. Deute-
riolysis of the labile Si-C bond in 6 upon workup then
provides 4. In the more polar/protic/silophilic solvent, 35%
D O-MeCN, excitation of 3 is followed by an SET-
2
desilylation process (either concerted or sequential) to
yield the mono-N-deuteriobenzazepinedione 7.
Resu lts
N-(Silyla lk yl)p h th a lim id e a n d N-(Silyla lk yl)m a -
leim id e P h otoch em istr y. Several N-(silylalkyl)phthal-
imides and N-(silylalkyl)maleimide analogues were pre-
pared (see Supporting Information for details about the
preparation of the substrates) and subjected to explor-
atory studies in order to probe the magnitude and scope
of the solvent effects on the chemoselectivities of their
photochemical reactions. As shown in Scheme 5 and
Table 1, the nature and efficiencies of the photoreactions
of (silylbutyl)phthalimide 10 are dramatically altered by
Observations made in our early studies7 with the
silylpropyl derivative 8 suggested that the solvent control
of mechanistic preferences might have an impact on the
preparative potential of phthalimide photochemistry.
Specifically, we found that irradiation of an MeCN or
acetone solution of 8 promotes nonselective reaction
leading to production of a mixture of four photoproducts
(Scheme 4) that arise by pathways stemming from the
diradical intermediates generated by γ-H-atom abstrac-
tion in the excited state of 8. In contrast, benzopyrrolizi-
dine 9 is the exclusive (96%) product formed in the
a change in solvent from MeCN to 35% H O-MeCN. In
2
MeCN, 10 is photochemically transformed to a mixture
of products, including 12, 14, and 15. In contrast,
photoreaction of 8 in 30% H
solvent from MeCN to 30% H
2
O-MeCN. The change in
O-MeCN also profoundly
2
irradiation of a 35% H O-MeCN solution of 10 leads to
2
alters the quantum efficiency of this process. Thus,
irradiation in MeCN for 6 h leads to only 55% conversion
the efficient and exclusive production of the benzoin-
dolizidine 15 (94%). Although this solvent change has a
less dramatic effect on the excited-state reactivity of the
silylpentylphthalimide 11, the trend remains in the
direction of higher chemoselectivity for reaction in the
aqueous solvent system (Scheme 5, Table 1). The influ-
ence of solvent on the nature and efficiencies of conju-
gated imide photoreactions is also seen in the photo-
chemistry of the silylpropyl-maleimide 17. While irradia-
tion of an MeCN solution of 17 results in quantitative
conversion to the 2+2-photodimer 19, similar treatment
of 8, while a 20 min irradiation of 8 in 30% H O-MeCN
2
is sufficient to promote 100% conversion to the benzopy-
rrolizidine 9.
The preliminary observations described above suggest
that a simple change in solvent can transform an unse-
(6) Yoon, U. C.; Kim, D. U.; Kim, J . C.; Lee, J . G.; Mariano, P. S.;
Lee, Y. J .; Ammon, H. L. Tetrahedron Lett. 1993, 34, 5859. Yoon, U.
C.; Kim, D. U.; Lee, C. W.; Choi, Y. S.; Lee, Y. J .; Ammon, H. L.;
Mariano, P. S. J . Am. Chem. Soc. 1995, 117, 2689. Yoon, U. C.; Cho,
S. J .; Lee, Y. J .; Mancheno, M. J .; Mariano, P. S. J . Org. Chem. 1995,
of 17 in 35% H O-MeCN cleanly produces the function-
2
6
0, 2353.
7) Lee, Y. J .; Ling, R.; Mariano, P. S.; Yoon, U. C.; Kim, D. U.; Oh,
S. W. J . Org. Chem. 1996, 61, 3304.
alized pyrrolizidine 18 (64%) not contaminated with 19
(
(Scheme 6, Table 1).

Silylalkyl-Substituted Phthalimides and Maleimides
J . Org. Chem., Vol. 64, No. 12, 1999 4413
Sch em e 5
H
2
O-MeCN results in generation of the deceptively
complex, dimeric product 29 (unknown stereochemistry)
(
Scheme 8). Formation of 29 likely occurs via the same
type of sequence operating in the photochemistry of
silylalkyl)phthalimides and -maleimides, except in this
(
case the initially formed cyclic amidol undergoes dehy-
dration to form the corresponding enamide. This is
followed by enanmide-acyliminium ion coupling and
water addition. Interestingly, 29 is also produced by
irradiation of an MeCN solution of naphthalimide 28, but
under these conditions the reaction is much less efficient.
In a similar manner, the nature of the photoreaction
of (silylpropyl)-N-acetylbenzamide 30 is not altered ap-
preciably by a change in the solvent from MeCN to 30%
H
2
O-MeCN. For example, preparative irradiation of 30
in either MeCN or 30% H O-MeCN provides the same
2
products, γ-acetamidopropiophenone 31, N-silylallyl-
amide 32, and N-acetylbenzamide (33), albeit in slightly
different ratios (Scheme 9). To elucidate the mechan-
ism(s) responsible for generation of the propiophenone
derivative 31, 30 was irradiated in 35% D O-MeCN.
2
This leads to formation of 32 and 33, along with the
R-monodeuteriopropiophenone 31D. Consequently, it is
reasonable to conclude that 31 is formed by a mechanistic
route involving initial excited-state δ-H-atom abstraction
followed by cyclization of the resulting diradical 34,
amidol ring opening, and R-silyl ketone protodesilylation
The magnitude of the solvent effect is somewhat
diminished in photoreactions of the allylsilane-containing
maleimides 20 and 21. Photoreactions of these substrates
(Scheme 10). Plausible mechanisms for formation of the
in either MeCN or 35% H
the respective pyrrolizidine 22 or indolizidines 23 and
4 (Scheme 7). The change to the aqueous solvent system
2
O-MeCN lead to formation of
unsaturated acetamide 32 and imide 33 involve respec-
tive sequential acyl cleavage-diradical disproportion-
ation and Norrish Type-II cleavage reactions of the
excited state of 30.
2
promotes only a slight increase in the yields of both of
these photocyclization reactions (Scheme 7, Table 1).
Similar observations are made in studies with the N-[2-
F lu or escen ce Stu d ies. Information about the pos-
sible contribution of excited-state multiplicity to the
solvent effects summarized above has come from fluo-
rescence studies with the (silylpropyl)naphthalimide 28
and its N-methyl analogue 37. Unlike simple phthalim-
ides and maleimides, 1,8-naphthalimides are highly
fluorescent substances (i.e., they have longer lived and
efficiently emitting singlet excited states). As is seen by
inspection of the spectra in Figure 1 and data in Table
2, the quantum yields for fluorescence of 28 and 37
increase upon changing the solvent from MeCN to 35%
(silylmethyl)benzyl]maleimide 25 (Scheme 7, Table 1)
where photoreaction in both MeCN and 35% H O-MeCN
2
leads to the benzoindolizidines 26 and 27.
A greater understanding of how the solvent controls
the chemoselectivities of these photoreactions has come
from the results of triplet sensitization studies. Ben-
zophenone (50 mM) sensitized irradiation (λ > 310 nm)
of 17 (5 mM) in either MeCN or 35% H O-MeCN leads
2
to exclusive formation of the dimer 19. In addition,
acetone triplet sensitized photoreactions of 17 in either
H O-MeCN. Since this solvent change does not ap-
2
neat acetone or 35% H
2
O-acetone likewise yield the 2+2-
preciably alter the absorption spectrum (λmax and extinc-
dimer 19. Thus, the maleimide triplet excited-state
undergoes 2+2-photodimerization independent of the
solvent system used. By default, pyrrolizidine 18 must
arise by an SET-desilylation reaction of a singlet excited
state of 17. Contrastingly different results arise in triplet
sensitization studies with the phthalimide analogue 8.
Acetophenone and acetone sensitized reactions of this
substrate in MeCN and neat acetone, respectively, lead
to production of the same four photoproducts (se Scheme
tion coefficient) of the substrates, the enhanced fluores-
cence efficiencies in the aqueous solvent system are most
likely a result of the increased lifetime (slower rate of
decay) of the naphthalimide singlet excited states. The
data in Table 2 also show that the presence of the
silylpropyl tether in 1,8-naphthalimide 28 causes a
significant reduction in the emission quantum yield (as
compared to 37), which implicates quenching of the
naphthalimide singlet by the Si-C bond containing side
chain.
4
) as is obtained by direct irradiation in MeCN. However,
the benzopyrrolizidine 9 is generated exclusively in
triplet sensitized photoreactions of 8 by both acetophe-
Discu ssion
none in 35% H
2
O-MeCN and acetone in 35% H O-
2
Mech a n istic Differ en ces. Observations made in this
study demonstrate that solvent has a pronounced impact
on both the efficiencies and chemical selectivities of
acetone. Unlike that of the maleimide derivative 17, the
triplet excited-state reactivity of 8 is altered by a change
in solvent.
(silylalkyl)phthalimide and (silylalkyl)maleimide photo-
Stu d ies w ith th e N-(Silylp r op yl)n a p h th a lim id e 28
a n d Ben zim id e 30. Additional information about the
mechanistic origin(s) of the preparatively significant
solvent effect described above has come from studies with
the naphthalimide derivative 28. Irradiation of 28 in 30%
reactions. The most dramatic effects are seen in reactions
of substrates that contain simple (trimethylsilyl)alkyl
tethers. In these systems, a change in solvent from MeCN
to 35% H
2
O-MeCN alters their excited-state reactivity
from inefficient and often nonselective H-atom abstrac-

4
414 J . Org. Chem., Vol. 64, No. 12, 1999
Yoon et al.
Ta ble 1. P r od u ct Distr ibu tion s a n d Rela tive Qu a n tu m Efficien cies of P h otor ea ction s of Silyla lk yl-P h th a lim id e a n d
Ma leim id e Su bstr a tes
-
product yields
percent conversion (time)
MeCN 35% HOH-MeCN
substrates
MeCN
35% HOH-MeCN
15 (94%)
ΦHOH-MeCN/ΦMeCN
1
0
12 (14%)
70% (6 h)
76% (0.3 h)
22
1
1
4 (38%)
5 (16%)
1
1
13 (33%)
14 (35%)
16 (58%)
75% (2.5 h)
84% (0.5 h)
6
1
1
4 (37%)
6 (4%)
1
2
2
7
0
1
19 (98%)
22 (68%)
23 (66%)
18 (64%)
22 (71%)
23 (82%)
84% (2 h)
63% (27 h)
100% (3 h)
71% (1.6 h)
57% (7 h)
100% (2 h)
1.1
3.5
1.5
2
4 (11%)
2
5
26 (34%)
26 (88%)
27 (3%)
61% (5 h)
54% (4 h)
1.1
2
7 (37%)
Sch em e 6
Sch em e 8
Sch em e 7
Sch em e 9
the silylindolizidine 41. It is known8 that substances
related to 40, owing to the extremely high efficiencies of
their Norrish-Type II photocleavage reactions, undergo
ready secondary photoconversion to benzazepinedione 14.
As a result, it is often difficult to detect these substances
as primary photoproducts in phthalimide excited-state
reactions. In addition, conversion of the initially formed
TMS-substituted indolizidine 41 to 15 occurs by water-
induced protodesilylation of the dione tautomer 42 during
workup of the photolysate.
In contrast, excited-state reactions of the (silylalkyl)-
phthalimides in the aqueous solvent system appear to
occur either exclusively or predominantly by an SET-
desilylation mechanism. In this route, the phthalimide
excited state serves as an electron acceptor and the σC-Si
bond as the donor. Silyl transfer to water can occur either
in concert with or following the SET step (Scheme 12).
tion and 2+2-dimerization processes into fast and high-
yielding SET-promoted photocyclization reactions. Al-
though less pronounced, this solvent change also influences
the photoreactivity of substrates in which the TMS
moiety is part of an allyl- or benzylsilane function or
where the chromophore is a 1,8-naphthalimide or N-
acetylbenzamide.
The solvent effect on photoreactions of simple (silyl-
alkyl)phthalimides (e.g., 3, 8, 10, and 11) is associated
with a change in reaction mechanism. Excited-state
reactions of these substrates in MeCN, like those of other
members of the N-alkylphthalimide family, follow well-
known H-atom abstraction pathways. As illustrated for
phthalimide 10 (Scheme 11), competitive γ- and ω-H-
atom transfer from the side chain to the excited carbonyl
oxygen results in generation of the respective diradical
intermediates, 38 and 39, which serve as precursors of
olefins 12 (disproportionation), benzazepinedione 40, and
(8) Kubo, Y.; Tanabe, S. Bull. Chem. Soc. J pn, 1992, 65, 2875.
Maruyama, K.; Kubo, Y. J . Org. Chem. 1985, 50, 1426.

Silylalkyl-Substituted Phthalimides and Maleimides
J . Org. Chem., Vol. 64, No. 12, 1999 4415
Sch em e 10
Sch em e 11
Sch em e 12
that the role played by water is to control the energetic
ordering and rates of interconversion of two differently
reactive excited states (e.g., n-π vs π-π*, or singlet vs
triplet). It is well-known that H-bonding characteristics
of the solvent can influence the relative energies of n-π*
F igu r e 1. Fluorescence spectra of naphthalimides 28 and 37
in MeCN and 35% H
MeCN; (b) 28 in 35% H
MeCN. Excitation is at 311 nm and concentrations (ca. 1 ×
2
O-MeCN at 25 °C: (a) 37 in 35% H
2
2
O-
O-MeCN; (c) 37 in MeCN; (d) 28 in
9
and π-π* excited states of aromatic ketones. These
-5
1
0
M) of all solutions are adjusted to give equal absorbances
energetic priorities, in turn, govern the interrelated
singlet excited-state lifetimes and triplet formation ef-
ficiencies of carbonyl compounds,10 since singlet-triplet
intersystem crossing (ISC) rates are often strongly de-
pendent on excited-state configurations (S(n-π*) f T(π-
π*) fast, while S(π-π*) f T(π-π*) slow).11
at this wavelength.
Ta ble 2. F lu or escen ce Qu a n tu m Yield s for
1
,8-Na p h th a lim id es 28 a n d 37 a t 25 °C
Φfa
naphthalimide
MeCN
35% HOH-MeCN
Some information about the origin(s) of the solvent
effect has come from our fluorescence studies with the
2
3
8
7
0.01
0.02
0.08
0.15
1
,8-naphthalimides 28 and 37. The data listed in Table
2 show that the fluorescence quantum efficiencies of these
substances are considerably larger in 35% H O-MeCN
a
Determined on the basis of the use of 37 (refs 19 and 20) as
actinometer.
2
as compared to MeCN, a phenomenon that appears to
be associated with naphthalimide π-π* singlet-state
lifetimes and not the rate constants for fluorescence (see
Both pathways lead to formation of the diradical precur-
sor of the non-TMS-containing primary photoproducts
(e.g., 10 f 15).
P ossible Sou r ces of th e Solven t Effects. A number
(
(
9) Brealey, G. J .; Kasha, M. J . Am. Chem. Soc. 1955, 77, 4462.
10) Caldwell, R. A. Tetrahedron Lett. 1969, 26, 2121. Biszok, L.;
of plausible arguments can be offered to explain the
solvent-dependent mechanistic selectivities of (silylalkyl)-
phthalimide photoreactions. For example, it is possible
Berces, L.; Linschitz, H. J . Am. Chem. Soc. 1997, 119, 11071.
(11) El-Sayed, M. A. J . Chem. Phys. 1963, 38, 2834.

4
416 J . Org. Chem., Vol. 64, No. 12, 1999
Yoon et al.
above). In addition, fluorescence from the naphthalimide
π-π* singlet is modestly quenched by intramolecular
SET when the σC-Si bond containing the silylpropyl side
chain is present (i.e., 28 vs 37). Thus, the enhancement
seen in the efficiency of photoproduction of 29 in changing
Calculations based on ground-state reduction potentials
and excited-state energies15 indicate that phthalimide
singlet and triplet excited-state reduction potentials in
MeCN fall between +2.4 and +1.7 V, respectively, and
1
6
cyclic voltammetric measurements made on alkylsilanes
the solvent from MeCN to 35% H
consequence of a medium effect on the lifetime (longer
in 35% H
O-MeCN) of the SET reactive singlet of 28.20
2
O-MeCN is likely a
suggest that these substances are oxidized in MeCN at
ca. +2.8 V. On the basis of these values, one can imagine
how the aqueous solvent systems, by promoting increases
in phthalimide excited-state reduction potentials and
decreases in alkylsilane oxidation potentials could cause
intramolecular SET to be favorable (i.e., ∆GSET < 0) in
the singlet excited states of (silylalkyl)phthalimides. But
it is difficult to see how these effects overcome the high
(ca. 25 kcal/mol) thermodynamic barrier to SET that
exists in the triplet states of these substances.
2
It has been proposed that n-π* singlet and not triplet
excited states of N-alkylphthalimides participate in in-
1
2
tramolecular hydrogen-atom abstraction reactions. In
addition, studies by Coyle1 and Davidson have pro-
vided convincing evidence for the singlet excited-state
origin of intramolecular SET reactions of phthalimides
containing electron donor substituted N-substituents.
Adding to these observations are the results that show
2c
13
The results from studies with the (silylpropyl)male-
imide 17 clearly demonstrate that the triplet of this
substrate undergoes 2+2-photodimerization independent
of the solvent used. This suggests that the SET-desilyl-
ation reaction occurs in the singlet manifold as a conse-
quence of the control by water of the relative rates of
singlet SET-desilylation vs intersystem crossing.
Investigations with the allyl- and benzylsilane-substi-
tuted maleimides, 20, 21, and 25, show that the chemical
selectivities and efficiencies of these photoreactions are
indeed controlled by factors which govern the rates of
SET. Accordingly, SET-promoted cyclizations occur ex-
clusively in photoreactions of 20, 21, and 25, independent
of the solvent system used. The major difference between
these substrates and the (silylpropyl)maleimide 17 re-
sides in the oxidation potentials of their C-Si-containing
electron donor group. In contrast to simple alkylsilanes,
T
that 8 reacts by a hydrogen-atom abstraction pathway
in MeCN or acetone and an SET-desilylation route in
3
5% H
2
O-MeCN or 35% H O-acetone. Together, these
2
observations provide a basis to understand one source
of the solvent effects described above. Water may modu-
late the nature (n-π* vs π-π*), interconversion rates
(kISC), and thus, selective reactivity (H-atom abstraction
vs SET-desilylation) of (silylalkyl)phthalimide excited
states. (Silylalkyl)phthalimide n-π* singlets and triplets
will likely be of lowest energy in MeCN, while their
singlet π-π* states could be competitively populated in
the aqueous solvent system. If the phthalimide π-π*
singlets undergo selective SET-desilylation reactions, it
would be easy to see how their longer lifetimes (owing to
slower ISC) would result in higher reaction quantum
yields.
Yet another factor contributing to the observed solvent
effect might be associated with alterations in the rates
of SET and desilylation. Accordingly, water could func-
tion to control the selectivity of these processes by
enhancing the rates of intramolecular SET in (silylalkyl)-
phthalimide π-π* singlets through hydrogen. In addi-
tion, if desilylation occurs in concert with one-electron
oxidation of the C-Si bond, the greater silophilicity of
water would increase the rate14 of conversion of singlet
1
5
which have oxidation potentials in the range of +2.8 V,
the π-electron donating allyl- and benzylsilanes are more
readily oxidized (ca. 1.6-2.0 V).17 Consequently, ther-
modynamically driven SET dominates over 2+2-photo-
cycloaddition, H-atom abstraction, and other decay pro-
cesses available to the excited states of these substrates.
Su m m a r y. Although complex in its origin(s), the
solvent effects summarized above have an important
impact on the preparative utility of (silylalkyl)phthalim-
ide and maleimide excited-state chemistry. Based on the
examples provided in this study, it appears to be gener-
ally true that a simple change in solvent from MeCN to
(silylalkyl)phthalimides to the key diradical intermedi-
ates in the cyclization pathways.
A consideration of the energetics of the alkylsilane to
phthalimide SET step is germane to this discussion.
35% H O-MeCN can be used to alter the photochemical
2
reactivity of these substrates. In the systems probed in
the current effort, a change to more protic-silophilic
solvent systems transforms less efficient and in some
cases less selective hydrogen atom abstraction and 2+2-
dimerization type processes into selective, high-yielding
SET-desilylation reactions that are potentially appli-
cable to the synthesis of functionalized nitrogen hetero-
cycles.
(
12) (a) Griesbeck, A. G.; Hirt, J .; Peters, K.; Peters, E. M.;
vonSchering, H. G. Chem. Eur. J . 1996, 2, 1388. (b) Kanaoka, K.;
Flippen. J . L.; Karle, I.; Witkop, B. J . Am. Chem. Soc, 1974, 96, 4719.
(
c) Coyle, J . D.; Harriman, A.; Newport, G. L. J . Chem. Soc., Perkin
Trans. 2 1979, 799.
13) (a) Barlow, J . H.; Davidson, R. S.; Lewis, A.; Russell, D. R. J .
(
Chem. Soc., Perkin Trans. 2 1979, 1103. (b) Results supporting a
different view have been presented (see ref 12a).
(
14) Zhang, X. M.; Yeh, S. R.; Hong, S.; Freccero, M.; Albini, A.;
Falvey, D.; Mariano, P. S. J . Am. Chem. Soc. 1994, 116, 4211.
15) Reddy, D. W.; Muck, D. L. J . Am. Chem. Soc. 1971, 93, 4264.
(
Wintgens, V.; Valat, P.; Kossanyi, J .; Biczok, L.; Demeter, A.; Berces,
T. J . Chem. Soc., Faraday Trans. 1994, 90, 411.
Exp er im en ta l Section
(
16) Klinger, R. J .; Kochi, J . K. J . Am. Chem. Soc. 1980, 102, 4790.
Wong, C. L.; Kochi, J . K. J . Am. Chem. Soc. 1979, 101, 5593.
17) Taylor, T. G.; Berwin, H. J .; J erkunica, J .; Hall, M. L. Pure Appl.
Gen er a l In for m a tion . All reported reactions were run
under a dried nitrogen atmosphere. Unless otherwise noted,
all reagents were obtained from commercial sources and used
without further purification. All compounds were isolated as
(
Chem. 1972 30, 599. Brown, R. S.; Eaton, D. F.; Hosomi, A.; Traylor,
T. G.; Wright, J . M. J . Organomet. Chem. 1974, 66, 249. Weidner, U.;
Scweig, A. Angew. Chem., Int. Ed. Engl. 1972, 11, 4429.
1
13
oils and shown to be >90% pure by H and/or C NMR unless
(
18) Kanaoka, Y.; Migita, Y.; Mizoguchi, T. Tetrahedron Lett. 1973,
4, 1193.
19) Takahashi, Y.; Miyashi, T.; Yoon, U. C.; Oh, S. W.; Mancheno,
M.; Su, Z.; Falvey, D. E.; Mariano, P. S. Unpublished results.
20) For a discussion of this same effect, see also: Wintgens, V.;
otherwise noted.
1
1H NMR and 13C NMR spectra were recorded by using
(
CDCl
3
solutions unless otherwise specified, and chemical shifts
at 7.24 ppm
for H NMR) and 77.0 ppm (for C NMR). C NMR resonance
assignments were aided by the use of the DEPT technique to
are reported in ppm relative to residual CHCl
(
3
(
1
13
13
Valat, P.; Kossanyi, J .; Biczok, L.; Demeter, A.; Berces, T. J . Chem.
Soc., Faraday Trans. 1994, 90, 411.

Silylalkyl-Substituted Phthalimides and Maleimides
J . Org. Chem., Vol. 64, No. 12, 1999 4417
determine numbers of attached hydrogens, and 1H NMR
coupling constants (J values) are reported in Hz. Infrared
spectra were obtained using neat liquids unless otherwise
specified, and data are reported in units of cm . Low (MS)
and high (HRMS) mass spectra, reported as m/z (relative
intensity), were recorded by using electron impact ionization
5.43 (m, 1H, NCH
HNCH CHdCH), 7.37-7.61 (m, 4H, aromatic); C NMR -1.2
(SiMe ), 16.5 (CH SiMe ), 26.9 (CH CH SiMe ), 41.2 (NCH ),
CHdCH), 5.66-5.79 (m, 2H, C(OH)-
2
13
2
3
2
3
2
2
3
2
-
1
81.5 (COH), 123.6 and 123.8 (CH, alkenic), 122.9, 130.1, 132.6,
and 138.4 (CH, aromatic), 131.9 and 144.5 (C, aromatic), 167.6
(CdO); IR (KBr) 3100-3550 (br OH), 1680 (CdO); MS (EI)
+
(EI) or chemical ionization (CI).
m/z (rel intensity) 289 (M , 20), 73 (100); HRMS (EI) m/z
Photochemical reactions were conducted by using an ap-
paratus consisting of a 450-W medium-pressure mercury lamp
surrounded by a glass filter and within a quartz, water-cooled
289.1496 (C16
Ir r a d ia tion of N-[5-(Tr im eth ylsilyl)p en tyl]p h th a lim -
id e (11) in 35% H O-CH CN. A solution of 11 (250 mg, 0.86
mmol) in 65 mL of CH CN and 35 mL of H O was irradiated
with Vycor-filtered light under N for 30 min (84% conversion
of 11). Concentration of the photolysate gave a residue that
was subjected to column chromatography (silica, ethyl acetate:
hexane ) 1:3), yielding 98 mg (58%) of 16 and 44 mg (35%) of
14.
2
H23NO Si requires 289.1498).
2
3
well that was purged with O
2
-free N
2
both before and during
3
2
irradiation. Photochemical reaction progress was monitored
2
1
by gas chromatography, TLC, or H NMR.
Ir r a d ia tion of N-[4-(Tr im eth ylsilyl)bu tyl]p h th a lim id e
(
10) in CH
phthalimide (10) (500 mg, 1.82 mmol) in 200 mL of CH
was irradiated with Vycor-filtered light under N for 6 h (70%
3
CN. A solution of N-[4-(trimethylsilyl)butyl]-
3
CN
2
Ir r a d ia tion of N-[3-(Tr im eth ylsilyl)p r op yl]m a leim id e
conversion of 10). Concentration of the photolysate gave a
residue that was subjected to column chromatography (silica,
ethyl actate:hexane ) 1:3), yielding 47 mg (14%) of 12, 41 mg
(17) in CH
maleimide (17, 90 mg, 0.43 mmol) in 50 mL of CH
irradiated with Pyrex-filtered light under N for 2 h (84%
conversion of 17). Concentration of the photolysate gave a
residue that was crystallized (CH Cl ), yielding 74 mg (98%)
3
CN. A solution of N-[3-(trimethylsilyl)propyl]-
3
CN was
2
(16%) of 15 and 86 mg (38%) of 14
1
1
2: mp 96-97 °C (CH
.25 (s, 1H, OH), 1.44 (d, 2H, J ) 8.1, CH
H, J ) 8.4 and J ) 14.6, NCH ), 4.13 (dd, 1H, J ) 4.9 and J
14.7, NCH ), 5.13-5.26 (m, 1H, NCH CHdCH), 5.62-5.74
m, 2H, C(OH)HNCH CHdCH), 7.36-7.62 (m, 4H, aromatic);
), 23.3 (CH SiMe ), 41.5 (NCH ), 81.4
COH), 123.6 and 123.8 (CH, alkenic), 122.7, 130.0, 132.5, and
32.6 (CH, aromatic), 132.0 and 144.5 (C, aromatic), 167.6 (Cd
2
Cl
2
; H NMR -0.02 (s, 9H, SiMe
3
),
2
2
1
1
1
2
SiMe ), 3.64 (dd,
3
of 19: mp 244-246 °C (CH
2
Cl
), 0.43-0.52 (m, 4H, CH
CH
2
); H NMR -0.12 (s, 18H,
SiMe ), 1.50-1.66 (m, 4H,
);
), 41.4
), 174.8 (CdO); IR (KBr) 1700 (CdO); MS
2
SiMe
NCH
3
2
3
)
2
2
2
2
), 3.38 (s, 4H, 4CH), 3.56 (t, 4H, J ) 7.8, 2NCH
), 13.7 (CH SiMe ), 22.3 (NCH CH
2
1
3
(
2
C NMR -1.8 (SiMe
(CH), 42.5 (NCH
(EI) m/z (rel intensity) 422 (M , 3), 105 (100); HRMS (EI) m/z
422.2059 (C20 Si requires 422.2057).
Ir r a d ia tion of N-[3-(Tr im eth ylsilyl)p r op yl]m a leim id e
(17) in 35% H O-CH CN. A solution of N-[3-(trimethylsilyl)-
propyl]maleimide (17) (320 mg, 1.52 mmol) in a solution of
110 mL of CH CN and 60 mL of H O was irradiated with
Pyrex-filtered light under N for 1.6 h (71% conversion of 17).
3
2
3
2
2
1
3
C NMR -1.5 (SiMe
3
2
3
2
2
+
(
1
H
34
N
2
O
4
2
O); IR (KBr) 3100-3550 (br OH), 1680 (CdO); MS (EI) m/z
+
(
rel intensity) 275 (M , 34), 73 (100); HRMS (EI) m/z 275.1337
Si requires 275.1342).
1
2
3
15 2
(C H21NO
1
5: mp 154-155 °C (acetone); H NMR 1.19-1.31 (m, 2H,
NCH CH CH ), 1.70-1.77 (m, 2H, NCH CH ), 2.05 (dd, 1H,
), 2.39 (d, 1H, J ) 13.2, NCH ),
), 3.92 (dd, 1H,
), 4.11 (s, 1H, OH), 7.28-7.48 (m,
3
2
2
2
2
2
2
2
J ) 3.2 and J ) 10.3, C(OH)CH
2
2
Concentration of the photolysate gave a residue that was
subjected to column chromatography (silica, ethyl acetate:
2
.99 (dd, 1H, J ) 3.1 and J ) 12.9, C(OH)CH
2
J ) 4.7 and J ) 12.8, NCH
2
hexane ) 1:5), yielding 96 mg (64%) of 18: mp 122-124 °C
1
3
1
2
(
(
H, aromatic), 7.51-7.54 (m, 2H, aromatic); C NMR 20.1
(acetone); H NMR 1.52-1.68 (m, 1H, NCH
2
CH
), 2.22-2.34 (m, 1H, C(OH)CH
(m, 1H, C(OH)CH ), 2.74 (s, 1H, OH), 3.26-3.55 (m, 2H,
NCH ), 5.93 and 7.10 (two d, 2H, J ) 5.7, CHdCH); C NMR-
(CDCl ) 28.3 (NCH CH ), 33.9 (C(OH)CH ), 41.8 (NCH ), 98.5
2
), 2.04-2.17
2
NCH
NCH
2
CH
2
CH
2
), 25.5 (NCH
2
CH
2
), 36.0 (C(OH)CH
2
), 36.7
(m, 1H, NCH
2
CH
2
), 2.49-2.64
2
), 86.8 (COH), 121.8, 123.6, 129.4, and 132.4 (CH,
2
1
3
aromatic), 130.7 and 149.1 (C, aromatic), 165.8 (CdO); IR
KBr) 3200-3500 (br OH), 1670 (CdO); MS (EI) m/z (rel
2
(
3
2
2
2
2
+
intensity) 203 (M , 100), 174 (74); HRMS (EI) m/z 203.0944
(COH), 126.6 and 150.0 (CHdCH), 174.2 (CdO); IR (KBr)
(
C
12
H
13NO
2
requires 203.0946).
Ir r a d ia tion of N-[4-(Tr im eth ylsilyl)bu tyl]p h th a lim id e
10) in 35% H O-CH CN. A solution of 10 (150 mg, 0.54
CN and 21 mL of H
was irradiated with Vycor-filtered light under N for 20 min
76% conversion of 10). Concentration of the photolysate gave
3150-3500 (br OH), 1680 (CdO); MS (EI) m/z (rel intensity)
+
139 (M , 100); HRMS (EI) m/z 139.0638 (C
7
H
9
NO
2
requires
(
2
3
139.0633).
mmol) in a solution of 39 mL of CH
3
2
O
Ir r a d ia t ion of N-[2-((Tr im et h ylsilyl)m et h yl)p r op -2-
en yl]m a leim id e (20) in CH CN. A solution of N-[(2-((tri-
methylsilyl)methyl)-2-propenyl]maleimide (20) (400 mg, 1.79
mmol) in 200 mL of CH CN was irradiated with Pyrex-filtered
light under N for 27 h (63% conversion of 20). Concentration
2
3
(
a residue that was subjected to column chromatography (silica,
ethyl acetate:hexane ) 1:3), yielding 79 mg (94%) of 15.
Ir r a d ia tion of N-[5-(Tr im eth ylsilyl)p en tyl]p h th a lim -
3
2
of the photolysate gave a residue that was subjected to column
id e (11) in CH
phthalimide (11) (400 mg, 1.38 mmol) in 200 mL of CH
was irradiated with Vycor-filtered light under N for 2.5 h (75%
conversion of 11). Concentration of the photolysate gave a
residue that was subjected to column chromatography (silica,
ethyl actate:hexane ) 1:3), yielding 99 mg (33%) of 13, 9 mg
3
CN. A solution of N-[5-(trimethylsilyl)pentyl)]-
chromatography (silica, ethyl actate:hexane ) 1:2), yielding
1
3
CN
116 mg (68%) of 22: mp 104-105 °C (acetone); H NMR 2.02
2
(s, 1H, OH), 2.40 (d, 1H, J ) 15.5, C(OH)CH
2
), 2.73 (d, 1H, J
), 4.17 (d, 1H,
), 5.97 (dd, 2H, J
) 1.4 and J ) 5.8, CHdCHCO), 7.13 (dd, 1H, J ) 1.6 and J )
) 15.5, C(OH)CH
2
), 3.80 (d, 1H, J ) 15.4, NCH
2
J ) 15.4, NCH ), 5.09-5.10 (m, 2H, CdCH
2
2
1
8
13
(
4%) of 16, and 67 mg (37%) of 14.
1
5.7, CHdCHCO); C NMR 41.8 (C(OH)CH
2 2
), 46.2 (NCH ), 97.1
1
6: mp 136-137 °C (acetone); H NMR 0.59-0.71 (m, 1H,
NCH CH CH ), 1.23-1.72 (m, 5H, NCH (CH ), 1.98-2.11 (m,
H, C(OH)CH ), 2.50 (dd, 1H, J ) 7.7 and J ) 14.6, C(OH)-
), 2.98 (dd, 1H, J ) 2.6 and J ) 10.9, NCH ), 3.27 (d, 1H,
), 4.60 (s, 1H, OH), 7.32-7.58 (m, 4H,
aromatic); C NMR 22.7 (NCH CH CH ), 27.5 (C(OH)-
CH CH ), 30.2 (NCH CH ), 39.1 (C(OH)CH ), 39.2 (NCH ), 91.6
COH), 122.4, 123.3, 129.6, and 132.9 (CH, aromatic), 131.5
and 148.4 (C, aromatic), 168.3 (CdO); IR (KBr) 3200-3550
(COH), 109.4 (CdCH ), 147.8 (CdCH ), 126.7 and 149.9 (CHd
2
2
2
2
2
2
2
)
3
CH), 172.9 (CdO); IR (KBr) 3100-3550 (br OH), 1700 (CdO);
+
1
CH
J
2
MS (EI) m/z (rel intensity) 151 (M , 100), 132 (69); HRMS (EI)
2
2
m/z 151.0636 (C
Ir r a d ia t ion of N-[2-((Tr im et h ylsilyl)m et h yl)p r op -2-
en yl]m a leim id e (20) in 35% H O-CH CN. A solution of
N-[(2-(trimethylsilyl)methyl)-2-propenyl]maleimide (20, 220
mg, 0.99 mmol) in 65 mL of CH CN and 35 mL of H O was
irradiated with Pyrex-filtered light under N for 7 h (57%
8 9 2
H NO requires 151.0633).
) 14.2, NCH
2
1
3
2
2
2
2
3
2
2
2
2
2
2
(
3
2
2
+
(
br OH), 1710 (CdO); MS (EI) m/z (rel intensity) 217 (M , 85),
conversion of 20). Concentration of the photolysate gave a
residue that was subjected to column chromatography (silica,
ethyl acetate:hexane ) 1:2), yielding 60 mg (71%) of 22.
Ir r a d ia tion ofN-[3-((Tr im eth ylsilyl)m eth yl)bu t-3-en yl]-
1
2
60 (100); HRMS (EI) m/z 217.1107 (C13
H
15NO
2
requires
17.1103).
1
1
3: mp 101-102 °C (CH
.50-0.58 (m, 2H, CH SiMe
), 3.65 (dd, 1H, J ) 8.0 and J ) 15.0, NCH
H, OH), 4.07 (dd, 1H, J ) 4.9 and J ) 14.9 Hz, NCH
2
Cl
2
); H NMR -0.05 (s, 9H, SiMe
3
),
0
2
3
), 1.94-2.05 (m, 2H, CH
), 3.85 (s,
), 5.29-
2
CH
2
-
m a leim id e (21) in CH
silyl)methyl)but-3-enyl]maleimide (21) (500 mg, 2.11 mmol)
in 200 mL of CH CN was irradiated with Pyrex-filtered light
3
CN. A solution of N-[(3-((trimethyl-
SiMe
1
3
2
2
3

4
418 J . Org. Chem., Vol. 64, No. 12, 1999
Yoon et al.
under N for 3 h (100% conversion of 21). Concentration of
the photolysate gave a residue that was subjected to column
chromatography (silica, ethyl actate:hexane ) 1:2), yielding
2
photolysate was concentrated in vacuo, giving a residue that
was subjected to preparative TLC (silica, ethyl acetate) to yield
41 mg (17%) of 29: 1H NMR 0.90 (m, 2H), 1.28 (m, 1H), 1.70
(s, 1H, OH), 1.95 (m, 1H), 2.60 (m, 1H), 3.13 (m, 2H), 3.47 (m,
2
30 mg (66%) of 23 and 56 mg (11%) of 24.
1
2
3: mp 137-138 °C (acetone); H NMR 2.07 (d, 2H, J )
3.2, C(OH)CH ), 2.32 (d, 1H, J ) 12.7, NCH CH ), 2.70 (d,
H, J ) 13.1, NCH CH ), 2.97 (dd, 1H, J ) 3.9 and J ) 12.6,
), 3.92 (s, 1H, OH), 4.09 (dd, 1H, J ) 5.9 and J ) 12.7,
), 5.02 (d, 2H, J ) 9.7, CdCH ), 6.02 and 7.00 (two d,
), 37.0
), 140.4
), 126.9 and 150.2 (CHdCH), 167.9 (CdO); IR (KBr)
100-3500 (br OH), 1670 (CdO); MS (EI) m/z (rel intensity)
1
H), 3.73 (t, 1H), 4.44 (m 3H), 7.25 (m, 2H), 7.47 (m, 2H), 7.71
1
1
2
2
2
13
(m, 4H), 7.97 (m, 2H), 8.20 (d, 1H), 8.54 (d,1H); C NMR 29.7,
2
2
3
1
0.0, 30.5, 44.3 and 45.9 (CH
24.9, 125.2, 127.6, 131.8, 132.5, 133.9, 134.6, and 137.1 (Ar
2
), 48.3 (CH), 70.1 (C), 118.3,
NCH
NCH
2
2
2
C), 119.7, 123.2, 126.3, 126.4, 127.0, 127.2, 128.0, 128.2, 130.5,
and 132.1 (Ar CH), 162.4 and 159.3 (CdO); MS (FAB) m/z 441
1
3
2
(
(
H, J ) 5.9, CHdCH); C NMR 33.8 (C(OH)CH
NCH CH ), 43.9 (NCH ), 89.2 (COH), 114.5 (CdCH
CdCH
2
2
2
2
2
+
(
(
M
- H
2
O, 71), 440 (49); HRMS (FAB) m/z 441.1633
requires 441.1603).
Ir r a d ia tion of N-Acetyl-N-[3-(tr im eth ylsilyl)p r op yl]-
ben za m id e (30) in 30% H O-MeCN a n d MeCN. A solution
of 30 (200 mg, 0.72 mmol) in 200 mL of 30% H O-MeCN was
2
30 21 2 2
C H N O
3
1
+
65 (M , 64), 110 (100); HRMS (EI) m/z 165.0794 (C
9
H
11NO
2
2
requires 165.0790).
4: mp 60-61 °C (acetone); H NMR 0.04 (s, 9H, SiMe
.27 (d, 1H, J ) 8.3, C(OSiMe )CH
C(OSiMe )CH ), 2.30 (d, 1H, J ) 13.3, NCH
J ) 13.3, NCH CH ), 2.89 (dd, 1H, J ) 3.4 and J ) 12.8,
NCH ), 4.17 (dd, 1H, J ) 6.0 and J ) 12.8, NCH ), 4.93 (d,
H, J ) 16.9, CdCH ), 6.15 and 6.99 (two d, 2H, J ) 5.9, CHd
CH); C NMR 1.9 (SiMe ), 33.8 (C(OSiMe )CH ), 37.2 (NCH
CH ), 46.4 (NCH ), 90.6 (C(OSiMe )), 113.6 (CdCH ), 140.6
CdCH ), 127.2 and 150.3 (CHdCH), 167.5 (CdO); IR (KBr)
1
2
3
),
2
2
irradiated with Corex-filtered light under N for 2 h (72%
1
3
2
), 2.01 (d, 1H, J ) 13.2,
CH ), 2.62 (d, 1H,
2
conversion of 30). The photolysate was concentrated in vacuo,
giving a residue that was subjected to preparative TLC (silica
gel, ethyl acetate:hexane ) 2:1) to yield 30 mg of 31 (28%), 11
mg of 32 (12%), and 7 mg of 33 (7%).
3
2
2
2
2
2
2
2
2
1
3
3
3
2
2
-
A solution of 30 (200 mg, 0.72 mmol) in 200 mL of MeCN
was irradiated with Corex-filtered light under N for 3 h (63%
2
conversion of 30). The photolysate was concentrated in vacuo,
giving a residue that was subjected to preparative TLC (silica
gel, ethyl acetate:hexane ) 2:1) to yield 53 mg of 31 (57%), 22
mg of 32 (29%), and 10 mg of 33 (14%).
2
2
3
2
(
2
+
1
700 (CdO); MS (EI) m/z (rel intensity) 237 (M , 72), 74 (100);
Si requires 237.1185).
Ir r a d ia tion ofN-[3-((Tr im eth ylsilyl)m eth yl)bu t-3-en yl]-
m a leim id e (21) in 35% H O-CH CN. A solution of N-[(3-
trimethylsilyl)methyl)-3-butenyl]maleimide (21) (300 mg, 1.26
HRMS (EI) m/z 237.1175 (C12H19NO
2
2
3
A solution of 30 (200 mg, 0.72 mmol) in 200 mL of rigorously
(
dried MeCN was irradiated with Corex-filtered light under N
for 3 h (80% conversion of 30). After irradiation, 1.0 mL of
O was added. The photolysate was concentrated in vacuo,
2
mmol) in 65 mL of CH
with Pyrex-filtered light under N
2
3
CN and 35 mL of H
for 2 h (100% conversion of
1). Concentration of the photolysate gave a residue that was
2
O was irradiated
2
D
2
giving a residue that was subjected to preparative TLC (silica
gel, ethyl acetate:hexane ) 2:1) to yield 14 mg of 31 (12%), a
trace of 32, 31 mg of a substance that rapidly transforms to
31D, and 7 mg of 33 (8%).
subjected to column chromatography (silica, ethyl acetate:
hexane ) 1:2), yielding 172 mg (82%) of 23.
Ir r a d ia tion of N-[2-((Tr im eth ylsilyl)m eth yl)ben zyl]-
m a leim id e (25) in CH
lyl)methyl)benzyl]maleimide (25) (500 mg, 2.11 mmol) in 200
mL of CH CN was irradiated with Pyrex-filtered light under
for 5 h (61% conversion of 25). Concentration of the
3
CN. A solution of N-[2-((trimethylsi-
A solution of 30 (200 mg, 0.72 mmol) in 70 mL of D
30 mL of MeCN was irradiated with Corex-filtered light
under N for 2 h. The photolysate was concentrated in vacuo,
2
O and
3
1
N
2
2
photolysate gave a residue that was subjected to column
chromatography (silica, ethyl acetate:hexane ) 1:2), yielding
giving a residue that was subjected to preparative TLC (silica
gel, ethyl acetate:hexane ) 2:1) to yield 30 mg of 30 (15%), 29
mg of 32 (12%), 14 mg of 31D (11%), and 8 mg of 33 (9%).
1
02 mg (37%) of 27 and 69 mg (34%) of 26.
1
2
6: mp 132-133 °C (acetone); H NMR 2.87 (d, 1H, J )
3
1: 1H NMR 1.95 (m, CH
.5, 2H, CH
), 3.31 (ABq, J ) 5.3 and 4.6, CH1
1H, NH), 7.48 (m, 3H, Ar), 7.91 (m, 2H, Ar); C NMR 22.9
CH Si), 23.5 (CH ), 35.9 (CH ), 39.4 (CH N), 128.0, 128.6,
33.2, 136.6, 170.8 (CdO), 200.0 (CdO); IR (CHCl ) 1674, 1643;
MS (EI) m/z 205 (M , 0.2), 105 (100); HRMS (EI) m/z 205.1099
2
(C12H15NO requires 205.1103).
3
CO and PhCOCH
2
), 3.03 (t, J )
2
NH), 6.48 (s,
1
1
2
3
5.9, C(OH)CH
H, J ) 16.8, NCH
2
), 3.22 (d, 1H, J ) 15.9, C(OH)CH
), 4.94 (d, 1H, J ) 16.8, NCH
2
), 4.28 (d,
), 6.04 (d,
5
2
2
2
3
1
3
H, J ) 5.9, CHdCH), 7.10-7.26 (m, 4H, aromatic); C NMR
(
1
2
3
2
2
8.4 (C(OH)CH ), 40.0 (NCH ), 87.7 (COH), 126.9, 127.0, 127.4
2
2
and 127.6 (CH, aromatic), 130.2 (CHdCH), 130.3 and 131.1
3
+
(
C, aromatic); IR (KBr) 3100-3450 (br OH), 1670 (CdO); MS
+
(EI) m/z (rel intensity) 210 (M , 65), 104 (100); HRMS (EI)
m/z 201.0788 (C12
7: mp 101-102 °C (acetone); H NMR 0.00 (s, 9H, SiMe
.83 (d, 1H, J ) 15.5, C(OSiMe )CH ), 3.18 (d, 1H, J ) 15.6,
C(OSiMe )CH ), 4.27 (d, 1H, J ) 16.9, NCH ), 5.05 (d, 1H, J
16.9, NCH ), 6.20 (d, 2H, J ) 5.9, CHdCH), 7.11-7.25 (m,
H
11NO
2
requires 201.0799).
31D:
1
H NMR 1.97 (m, CH CO and PhCOCH ), 3.03 (m,
3
2
1
2
3
),
1
7
H, CHD), 3.33 (ABq, 2H, J ) 6 Hz, CH
.44 (m, 3H, Ar), 7.94 (d, 2H, Ar).
2
NH), 5.71 (s, 1H, NH),
2
3
2
3
2
2
2: 1H NMR 0.03 (s, 9H, SiMe
H, CdCCH ), 5.52 (s, 1H, NH), 5.77 (d, J ) 19, 1H, HCdC),
.98 (dt, J ) 19, 5, 1H, CdCH); C NMR -1.4 (SiMe
), 44.0 (CH ), 131.3 (HCd), 141.5 (dCH), 168.9 (CdO);
IR (CHCl ) 1744, 1691.
3
3
), 2.00 (s, 3H, Me), 3.89 (m,
)
2
2
5
(
2
1
3
4
H, aromatic); C NMR 1.8 (SiMe
NCH ), 88.9 (C(OSiMe )), 126.7, 127.1, 127.3, and 127.5 (CH,
aromatic), 130.0 (CHdCH), 131.0 and 131.4 (C, aromatic); IR
3 3 2
), 40.4 (C(OSiMe )CH ), 41.1
1
3
3
), 23.3
(
2
3
CH
3
2
+
3
(KBr) 1710 (CdO); MS (EI) m/z (rel.intensity) 273 (M , 75),
1
2
2
04 (100); HRMS (EI) m/z 273.1188 (C15H19NO Si requires
73.1185).
Ack n ow led gm en t. The authors appreciate the fi-
nancial support generously provided by NSF (P.S.M.,
CHE-9707401 and INT-9796064), KOSEF (U.C.Y., 965-
Ir r a d ia tion of N-[2-((Tr im eth ylsilyl)m eth yl)ben zyl]-
m a leim id e (25) in 35% H O-CH CN. A solution of N-[2-
(trimethylsilyl)methyl)benzyl]maleimide (64) (220 mg, 0.81
2
3
(
0
300-002-2, CBM Postech 1997 and 1998), and the
mmol) in 39 mL of CH
with Pyrex-filtered light under N
6
3
CN and 21 mL of H
for 4 h (54% conversion of
4). Concentration of the photolysate gave a residue that was
2
O was irradiated
Korea Research Foundation (U.C.Y., 1998-15-000177).
2
subjected to column chromatography (silica, ethyl acetate:
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
hexane ) 1:2), yielding 4 mg (3%) of 27 and 77 mg (88%) of
cedures are given for preparation of the substrates for the
2
6.
1
photochemical reactions. Spectroscopic data in the form of H
Ir r a d ia tion of N-[3-(Tr im eth ylsilyl)p r op yl]-1,8-n a p h -
th a lim id e (28) in 30% H O (or D O)-MeCN. A solution of
N-[3-(trimethylsilyl)propyl]-1,8-naphthalimide (28) (169 mg,
.54 mmol) in 40 mL of H O and 100 mL of MeCN was
irradiated with Vycor-filtered light under N for 6.5 h. The
13
and C NMR spectra for all previously unreported compounds
2
2
(
10-13 and 15-32) are provided. This material is available
free of charge via the Internet at http://pubs.acs.org.
0
2
2
J O990087A