Photoinduced electron-transfer reaction of α-bromomethyl-substituted benzocyclic β-keto esters with amines: Selective reaction pathways depending on the nature of the amine radical cations

Article abstract of DOI:10.1007/s11164-012-0646-2

Photoinduced electron-transfer reaction of α-bromomethyl-substituted benzocyclic β-keto esters with tertiary amines was investigated. Debrominated β-keto esters and ring-expanded γ-keto esters were obtained as major products. On the basis of mechanistic experiments it was concluded that these products are formed via a reaction sequence of selective carbon-bromine bond cleavage and subsequent competitive hydrogen abstraction and Dowd-Beckwith ring-expansion of the resulting primary alkyl radicals. The characteristic product distribution observed for the type of amine used is rationalized on the basis of selective reaction pathways of generated radical intermediates that depend on the nature of the amine radical cations.

Full text of DOI:10.1007/s11164-012-0646-2

Res Chem Intermed
DOI 10.1007/s11164-012-0646-2
Photoinduced electron-transfer reaction of
a-bromomethyl-substituted benzocyclic b-keto esters
with amines: selective reaction pathways depending
on the nature of the amine radical cations
•
Eietsu Hasegawa Emi Tosaka Akira Yoneoka
•
•
•
Yukinobu Tamura Shin-ya Takizawa
•
•
Masaaki Tomura Yoshiro Yamashita
Received: 11 October 2011 / Accepted: 7 January 2012
Ó Springer Science+Business Media B.V. 2012
Abstract Photoinduced electron-transfer reaction of a-bromomethyl-substituted
benzocyclic b-keto esters with tertiary amines was investigated. Debrominated b-keto
esters and ring-expanded c-keto esters were obtained as major products. On the basis
of mechanistic experiments it was concluded that these products are formed via a
reaction sequence of selective carbon–bromine bond cleavage and subsequent com-
petitive hydrogen abstraction and Dowd–Beckwith ring-expansion of the resulting
primary alkyl radicals. The characteristic product distribution observed for the type of
amine used is rationalized on the basis of selective reaction pathways of generated
radical intermediates that depend on the nature of the amine radical cations.
Keywords Photoinduced electron-transfer Á a-Bromomethyl-substituted
benzocyclic b-keto esters Á Tertiary amines Á Radical ions Á Dowd–Beckwith
ring-expansion
Introduction
Electron-transfer (ET) is a fundamental reaction process operating in a variety of
reduction and oxidation (redox) reactions [1–5], and single electron-transfer (SET)
E. Hasegawa (&) Á E. Tosaka Á A. Yoneoka Á Y. Tamura Á S. Takizawa
Department of Chemistry, Faculty of Science, Niigata University, Ikarashi-2, 8050,
Niigata 950-2181, Japan
e-mail: ehase@chem.sc.niigata-u.ac.jp
M. Tomura
Institute for Molecular Science, Myodaiji, Okazaki 444-8585, Japan
Y. Yamashita
Department of Electronic Chemistry, Interdisciplinary Graduate School of Science and Engineering,
Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8502, Japan
123

E. Hasegawa et al.
of neutral organic molecules generates their radical ions [6–10]. Among methods
used to promote ET reactions are: redox reagent ET, electrochemical ET, and
photoinduced electron-transfer (PET). PET has unique features not usually
associated with other ground-state ET reactions [11–24]. First, PET between
electron-acceptor (A) and electron-donor (D) molecules produces a pair of radical
anions and radical cations (Scheme 1). Therefore, radical ions generated under the
PET conditions undergo a variety of reactions depending on their inherent nature
and on co-existing counter radical ions. Second, radical intermediates formed by
fragmentation of radical ions become more persistent under PET conditions than
under other ground state ET conditions because steady state concentrations of
PET-generated redox reagents, for example short-lived excited molecules, are low
(Scheme 2). Therefore, PET provides a unique opportunity to investigate how
radical intermediates react under the ET conditions.
We have investigated PET reactions of different carbonyl and halogenated
compounds [25–39]. Tertiary amines are recognized as effective electron donors [40–
45], and are frequently used to photochemically reduce these substrates. The radical
cations of tertiary amines have different chemical reactivity; among the reactions are
deprotonation, release of electrofuges, and hydrogen atom abstraction (Scheme 3). We
havedemonstratedsuchbehaviorfor1,3-dimethyl-2-phenylbenzimidazoline(DMPBI),
N,N-diethyltrimethylsilylmethylamine (DETMSA), and 1,4-diazabicyclo[2.2.2]octane
(DABCO) in PET reactions of epoxy ketones (Scheme 4) [28, 29, 31].
In our early work, we discovered that PET with amines is a useful means of
promoting a ring-expansion reaction of a-halomethyl-substituted cyclic b-keto
esters [30, 31]. Dowd and Beckwith [46–51] originally investigated this type of
rearrangement under tributyltin-based free radical conditions. Since then, related
+
A
D*
D
SET
h
+
A
D
or
+
+
A
D
A*
Scheme 1 Photoinduced electron-transfer of electron acceptor (a) and donor (d) molecules
reductive ET conditions
+e
oxidative ET conditions
e
R E
R
Nu
R E
R Nu
Nu
E
e
+e
R
R
R
R
reductant / oxidant
(concentration)
ET-method
favorable for
redox reagents
electrochemical
photochemical
redox reagents (high)
electrodes (high)
ions
ions
excited state molecules (low)
radicals
Scheme 2 Comparison of fragmentation reaction pathways of organic radical ions generated by three
representative ET methods in solution
123

Photoinduced electron-transfer reaction
e
Scheme 3 Representative
reaction pathways of tertiary
amine radical cations
+
+
R
R
N
N
or
cleavage
(releasing of cations or radicals)
R = electrofuge or H
e
N
N
R
R
H
D
hydrogen
atom
abstraction
H
N
radical
addition
+
D
N
R
R
O
OH
R2
R1
O
O
+H
O
O
H
PET / amine
R1
R1
R1
R2
R2
R2
C
O
O
H
O
H
O
O
cleavage
R1
R2
Me
N
Et
Et
Ph
N
N
amine:
N
H
SiMe3
N
Me
DMPBI
DABCO
DETMSA
Scheme 4 Reaction patterns of epoxy ketone radical anions under PET conditions
O
O
O
OH
O
Br
Me
CO2Et
CO2Et
2
CO2Et
CO2Et
CO2Et
n
n
n
3a (n = 1)
3b (n = 0)
3c (n = 2)
(n = 1)
(n = 0)
(n = 2)
1a
1b
1c
4 (meso)
4 (dl)
2a (n = 1)
2b (n = 0)
2c (n = 2)
5
Me
N
Me
R'
N
R'
Et
N
Ar
Ph
N
N
Et
Et
SiMe3
H
H
N
N
X
Me
Me
DETMSA (R' = Et)
DPTMSA (R' = CH2CH2Ph)
TEA
DABCO (X = N)
ABCO (X = CH)
DMPBI (Ar = Ph)
ADMBI (Ar = p-MeOC6H4)
DMPI
Chart 1 Substrates 1, products 2, 3, 4, and 5, and amine donors
free radical reactions and mechanistic investigations have been reported [52–58].
Miscellaneous ET conditions have also been used to promote similar transforma-
tions [59–62]. However, until recently [63, 64], no example of PET-promoted
123

E. Hasegawa et al.
Dowd–Beckwith ring-expansion has been reported except in our previous publi-
cations [30, 31].
In this full paper we report details of PET-promoted Dowd–Beckwith ring-
expansion reaction of a-bromomethyl-substituted benzocyclic b-keto esters with
tertiary amines. Characteristic product distributions were observed, depending on
the kind of amine used. Factors determining the reaction pathway of generated
radical intermediates are discussed. The substrates, amines, and products that were
investigated are shown in Chart 1.
Experimental
General procedures
NMR spectra were recorded in CDCl3, with tetramethylsilane (Me4Si) as internal
standard, at 200 and 500 MHz for 1H NMR, and 50 MHz for 13C NMR.
Photoreactions were conducted in a Pyrex test tube (1.5 cm diameter) immersed
in a water bath at room temperature with a 500 W Xe–Hg lamp as light source.
Column chromatography was performed with silica gel (Wakogel C-200).
Preparative thin-layer chromatography (TLC) was performed on 20 cm 9 20 cm
plates coated with silica gel (Wakogel B-5F). Acetonitrile (MeCN) was distilled
from P2O5 and subsequently from K2CO3. N,N-dimethylformamide (DMF) was
˚
dried with 4-A molecular sieves and distilled under reduced pressure. Benzene
(PhH) was treated with H2SO4, water, 5 % NaOH, water, CaCl2, and then distilled
from CaH2. Tetrahydrofuran (THF) was distilled from sodium–benzophenone under
N2. Other reagents and solvents were used without further purification.
Substrates, amine donors, and sensitizers
Substrates 1a and 1c are known compounds [52]. Preparation of 1b is described below.
To a mixture of NaH (ca 60 % in oil, 1.19 g, 29.7 mmol) and hexamethylphosphoric
triamide (HMPA; 5.0 mL, 28.7 mmol) in THF (20 mL) was slowly added a THF
solution (10 mL) of ethyl 1-indanone-2-carboxylate (4.88 g, 23.9 mmol) obtained by
NaH-promoted reaction of 1-indanone with diethyl carbonate. After stirring for 1 h
under N2, CH2Br2 (8.4 mL, 120 mmol) was added, and the resulting mixture was
heated at 75 °C for 22 h. Then, extraction with Et2O was performed after addition of
H2O. The extract was treated with sat. aqueous Na2S2O3, sat. aqueous NaHCO3, and
sat. aqueous NaCl, dried over anhydrous MgSO4, and then concentrated. Column
chromatography with dichloromethane (CH2Cl2) gave 1b (5.28 g, 17.8 mmol, 74 %).
Further purification was performed by distillation: pale yellow oil; bp 149–150 °C
(0.60 mmHg). IR (neat) 1736, 1712 cm-1 1H NMR (200 MHz) d 1.23 (t,
.
J = 7.0 Hz, 3H), 3.39 (d, J = 18.0 Hz, 1H), 3.79 (d, J = 18.0 Hz, 1H), 3.80 (d,
J = 10.4 Hz, 1H), 4.01 (d, J = 10.4 Hz, 1H), 4.19 (q, J = 7.0 Hz, 2H), 7.42–7.67 (m,
3H), 7.79 (d, J = 7.7 Hz, 1H). 13C NMR (50 MHz) d 14.0 (q), 34.4 (t), 36.7 (t), 61.4
(s), 62.3 (t), 125.0 (d), 126.4 (d), 127.9 (d), 134.8 (s), 135.8 (d), 153.2 (s), 168.5 (s),
199.0 (s). Anal. calcd for C13H13O3Br C, 52.55; H, 4.41, found C, 52.55; H, 4.19.
123

Photoinduced electron-transfer reaction
DETMSA [65], DMPBI [33, 37], 2-anisyl-1,3-dimethylbenzimidazoline (AD-
MBI) [37], and 1,3-dimethyl-2-phenylimidazoline (DMPI) [31], depicted in Chart 1,
were prepared by literature procedures. N,N-diphenethyltrimethylsilylmethylamine
(DPTMSA) (Chart 1) was prepared similarly to DETMSA as described below. To a
mixture of diphenethylamine (588 mg, 2.61 mmol), obtained by reaction of
phenethylamine with phenethylbromide, and triethylamine (TEA) (0.75 mL,
5.38 mmol) in MeCN (3 mL) was slowly added an MeCN solution (1 mL) of
iodotrimethylsilane (838 mg, 3.92 mmol). The resulting mixture was heated at
88 °C for 17 h. After addition of 0.5 M NaOH, extraction with Et2O was performed.
The extract was treated with water, sat. aqueous NaCl, dried over anhydrous MgSO4,
and then concentrated. Column chromatography (ethyl acetate (EtOAc)–n-C6H14
1:3) gave DPTMSA (668 mg, 2.14 mmol, 82 %). Further purification was performed
by distillation: pale yellow oil; bp *100 °C (0.25 mmHg). IR (neat) 1030,
856 cm-1. 1H NMR (200 MHz) d 0.05 (s, 9H), 2.12 (s, 2H), 2.74 (s, 8H), 7.16–7.32
(m, 10H). 13C NMR (50 MHz) d 1.2 (q, 3C), 33.4 (t), 45.7 (t, 2C), 59.4 (t, 2C), 125.9
(d, 2C), 128.3 (d, 4C), 128.8 (d, 4C), 140.8 (s, 2C). Anal. calcd for C20H29NSi C,
77.10; H, 9.38; N, 4.50, found C, 77.10; H, 9.00; N, 4.49. HRMS (EI) calcd for
C20H29NSi 311.2069, found 311.2089; calcd for C17H20N (M?-SiMe3) 238.1596,
found 238.1564.
Sensitizers 9,10-bis(dimethylamino)anthracene (BDMAA), 9,10-dimethoxyan-
thracene (DMA), 1,6-bis(dimethylamino)pyrene (BDMAP), and 1,6-dimethoxypy-
rene (DMP), depicted in Scheme 5, are all known, and their preparations have been
reported [38].
General procedure for photoreaction of 1 with amine donors
A solution of 1 (0.50–0.64 mmol) and an appropriate quantity of amine donor in a
solvent (MeCN, DMF, PhH, THF or methanol (MeOH)) with or without additives
was purged with N2 for 5 min before irradiation. This solution was irradiated with
light, k [ 280 nm for direct irradiation of 1 and k [ 360 nm using a cut-off glass-
filter for sensitization reactions using a sensitizer (0.025–0.032 mmol), for an
appropriate irradiation time. Photolysates obtained from the reaction in MeCN,
DMF, PhH, THF, or MeOH were usually concentrated, and extraction was
performed if necessary. A photolysate obtained from reaction in DMF was extracted
with EtOAc. The extract was treated with water, sat. aqueous NaCl, dried over
anhydrous MgSO4, and then concentrated. The residue was subjected to column
substrate
sensitizer
1sensitizer
*
R
substrate
R
products
sensitizer
amine
h
R
R
sensitizer
BDMAA (R = Me2N) BDMAP (R = Me2N)
DMA (R = MeO) DMP (R = MeO)
amine
Scheme 5 Photosensitization pathways using arene sensitizers and amines
123

E. Hasegawa et al.
chromatography and/or TLC separations, using appropriate mobile phases (EtOAc–
n-C6H14, EtOAc–PhH, CH2Cl2, etc.) to give recovered 1 and the products.
Photoproducts were identified on the basis of their NMR and IR spectra.
Photoproducts 2a, 2c, 3a, and 3c are known [52]. Ring-expansion product 2b: IR
1
(neat) 1726, 1684, 1182 cm-1. H NMR (200 MHz) d 1.25 (t, J = 7.2 Hz, 3H),
2.74–3.21 (m, 5H), 4.17 (q, J = 7.2 Hz, 2H), 7.26–7.37 (m, 2H), 7.51 (td, J = 7.4,
1.5 Hz, 1H), 8.00–8.04 (m, 1H). 13C NMR (50 MHz) d 14.1 (q), 32.0 (t), 40.2 (d),
40.7 (t), 61.0 (t), 127.1 (d, 2C), 128.8 (d), 131.8 (s), 133.8 (d), 141.5 (s), 173.0 (s),
195.9 (s). Anal. calcd for C13H14O3 C, 71.54; H, 6.47, found C, 71.56; H, 6.18.
Reduced product 3b: IR (neat) 1740, 1712 cm-1. 1H NMR (200 MHz) d 1.19 (t,
J = 7.1 Hz, 3H), 1.52 (s, 3H), 3.00 (d, J = 17.5 Hz, 1H), 3.71 (d, J = 17.5 Hz,
1H), 4.15 (q, J = 7.1 Hz, 2H), 7.37–7.50 (m, 2H), 7.60–7.68 (m, 1H), 7.80 (d,
J = 7.5 Hz, 1H). 13C NMR (50 MHz) d 14.0 (q), 21.0 (q), 40.0 (t), 56.0 (s), 61.5 (t),
124.9 (d), 126.4 (d), 127.8 (d), 134.7 (s), 135.3 (d), 152.6 (s), 172.0 (s), 203.5 (s).
Anal. calcd for C13H14O3 C, 71.54; H, 6.47, found C, 72.05; H, 6.32. HRMS (EI)
calcd for C13H14O3 218.0943, found 218.0936.
Ring-expanded dimers 4: meso form, mp 217–218 °C, recrystallized from
1
CH2Cl2–ethanol (EtOH). IR (neat) 1712, 1672 cm-1. H NMR (200 MHz) d 1.05
(t, J = 7.2 Hz, 6H), 2.18–2.30 (m, 2H), 2.65–2.96 (m, 8H), 3.39–3.78 (m, 6H), 7.14
(d, J = 7.4 Hz, 2H), 7.23–7.44 (m, 4H), 7.68 (dd, J = 7.7, 1.7 Hz, 2H). 13C NMR
(50 MHz) d 13.5 (q, 2C), 31.0 (t, 2C), 31.2 (t, 2C), 45.6 (t, 2C), 52.4 (s, 2C), 61.6 (t,
2C), 127.0 (d, 2C), 128.6 (d, 2C), 129.5 (d, 2C), 132.8 (d, 2C), 137.2 (s, 2C), 140.4
(s, 2C), 172.5 (s, 2C), 200.8 (s, 2C). Anal. calcd for C28H30O6 C, 72.71; H, 6.54,
found C, 72.85; H, 6.49. dl form, mp 175–176 °C, recrystallized from EtOH. IR
(neat) 1726, 1676 cm-1. 1H NMR (200 MHz) d 0.99 (t, J = 7.2 Hz, 6H), 2.16–2.28
(m, 2H), 2.64–2.97 (m, 8H), 3.35–3.66 (m, 6H), 7.15 (d, J = 7.4 Hz, 2H),
7.22–7.43 (m, 4H), 7.62 (dd, J = 7.6, 1.5 Hz, 2H). 13C NMR (50 MHz) d 13.4 (q,
2C), 30.3 (t, 2C), 31.0 (t, 2C), 45.3 (t, 2C), 52.3 (s, 2C), 61.5 (t, 2C), 127.1 (d, 2C),
128.5 (d, 2C), 129.5 (d, 2C), 132.8 (d, 2C), 137.4 (s, 2C), 139.4 (s, 2C), 172.4 (s,
2C), 201.1 (s, 2C). Anal. calcd for C28H30O6 C, 72.71; H, 6.54, found C, 72.76; H,
6.39.
Naphthol 5: mp 141–143 °C, recrystallized from CH2Cl2–n-C6H14. IR (neat)
1
3416, 1684, 1250 cm-1. H NMR (200 MHz) d 1.45 (t, J = 7.1 Hz, 3H), 4.46 (q,
J = 7.1 Hz, 2H), 6.68 (broad s, 1H), 7.50–7.66 (m, 3H), 7.88–7.93 (m, 1H),
8.20–8.29 (m, 2H). 13C NMR (50 MHz) d 14.4 (q), 61.5 (t), 107.7 (d), 122.1 (d),
123.4 (d), 126.9 (s), 127.1 (d), 127.5 (d), 127.6 (s), 129.1 (s), 133.8 (s), 152.1 (s),
167.4 (s). Anal. calcd for C13H12O3 C, 72.21; H, 5.59, found C, 72.50; H, 5.67.
Determination of the percentage of deuterium incorporation into photoproducts
DMPBI-D [66, 67] was synthesized in the same manner as DMPBI [37], in which
deuterated benzaldehyde (PhCDO) obtained by deprotection of 2-deutero-2-phenyl-
1,3-dithiane was protected with N,N-dimethyl-o-phenylenediamine. 1H NMR
analysis revealed no H peak at C2 of DMPBI. D2O (Aldrich, 99.96 D%) was
purchased and used for photoreactions. The percentage of deuterium incorporated
(D%) into photoproducts was determined by use of the equations: for 2a,
123

Photoinduced electron-transfer reaction
D% = 100 9 [1 - (H intensity at C3)/(H intensity at C4)]; for 2b,
D% = 100 9 {[1 - (H intensities at C3 and C4) - (H intensity of methylene
protons of CO2CH2CH3)]/(H intensity at C4)}; for 3a, H intensity of CH3 was
compared to CH2D of 3a-D using H intensity at C3 as a standard peak.
Triethylborane (Et3B) and air-promoted reaction of 1a with DMPBI
To a mixture of DMPBI (173.1 mg, 0.772 mmol) and 1a (200.1 mg, 0.643 mmol)
in PhH (20 mL) was added 1.0 M THF solution of Et3B (0.16 mL, 0.16 mmol). The
resulting mixture was stirred for 5 min under N2 followed by introduction of air
(2 mL) by use of a syringe. After stirring for an additional 4 h, the reaction mixture
was concentrated. The residue was subjected to column chromatography (EtOAc–
n-C6H14) and subsequent TLC separation (EtOAc–PhH) gave 2a (11.6 mg,
0.0499 mmol, 8 %), 3a (21.3 mg, 0.0917 mmol, 14 %), and recovered 1a
(134.8 mg, 0.433 mmol, 67 %).
Results and discussion
Photoreaction of 1a with silyl-substituted amines (DETMSA, DBTMSA)
and TEA
Irradiation of 1a with DETMSA produced 2a, 3a, and 4. Compounds 4 were
obtained as a mixture of two diastereomers, the meso and dl isomers. The individual
stereoisomers of 4 were characterized by means of their spectral data, and finally
determined by X-ray crystallographic analysis (Fig. 1). Results of reactions in
different solvents are summarized in Table 1. Both irradiation and DETMSA are
necessary to obtain the observed products; this was confirmed by stirring the same
solution in the dark for 2 h, which resulted in recovery of 1a and no 2a, 3a, and 4 on
irradiation of 1a for 2 h in the absence of DETMSA; approximately 8 % 1a was
consumed. Several interesting observations can be pointed out from Table 1. Yields
of ring-expanded products 2a and 4 in MeCN and MeOH were greater than in PhH
Fig. 1 ORTEP plots of X-ray crystallographic data of 4 (meso and dl forms) [68]
123

E. Hasegawa et al.
Table 1 Photoreaction of 1a with DETMSA in different solvents
Entry
Solvent
H2O (%)
Irrad.
time (h)
Conv.
of 1a (%)
Yields (%)a
2a
3a
4
1
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeOH
PhH
1
65
54
76
75
75
87
88
83
87
93
55
71
61
68
75
69
68
53
44
33
10
9
21
13
20
16
10
18
17
28
13
19
2
10
5
0.5
1
3
10
7
4
10
30
10
10
1
5
1
6
6
7b
1.5
2
6
6
8
1
7
9
1
9
10
PhH
10
1
13
1a (0.64 mmol), DETMSA (5 equiv. vs 1a). Total volume of the mixed solvent was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Average from two independent experiments
(entries 1, 8 and 9). As the quantity of water was increased, the yield of 2a
significantly increased whereas those of both 3a and 4 gradually decreased (entries
1, 3, 4 and 5). The product ratio did not depend significantly on irradiation time
(entries 2, 4, 6 and 7), which suggests that the products do not interconvert and that
no significant degradation occurs during the course of the experiment. We therefore
decided to conduct further mechanistic experiments mainly in MeCN.
Table 2 Photoreaction of 1a with different amine donors in dry or aqueous MeCN
Entry
Donor
(equiv. vs 1a)
H2O (%)
Irrad.
time (h)
Conv.
of 1a (%)
Yields (%)a
2a
3a
4
1
2
3
4
5
6b
7
8
9
DETMSA (1.5)
DETMSA (5)
DETMSA (1.5)
DETMSA (5)
DETMSA (10)
DETMSA (5)
DPTMSA (5)
TEA (5)
1
1
1
1
1
2
2
5
5
56
65
61
75
73
88
40
87
91
42
55
56
68
73
68
61
34
33
8
10
7
30
21
24
16
12
17
28
16
16
10
10
10
10
10
7
10
6
10
8
TEA (5)
10
10
1a (0.64 mmol), solvent (10 mL). Total volume of MeCN and water was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Average from two independent experiments
123

Photoinduced electron-transfer reaction
We next investigated the effects of amines (Table 2). Product distributions are
comparable for DETMSA and TEA but longer irradiation was required for TEA
than for DETMSA (compare entry 6 with 9). As the quantity of DETMSA was
increased, the yield of 2a increased and that of 4 decreased (compare entry 1 with
entry 2; see also entries 3, 4, and 5). It should be also noted that irradiation of 1a
with DABCO in MeCN gave a complicated mixture in which no 2a, 3a, and 4 were
detected irrespective of the presence or absence of water (not shown in Table 2).
On the basis of these results, plausible reaction pathways giving the observed
products are presented in Scheme 6. PET produces a radical anion of 1a (1aÁ2, ketyl
form), and subsequent intramolecular SET (dissociative) to the Cb–Br bond gives a
primary alkyl radical 6a, as discussed in previous communications on PET reactions
of b-bromo-substituted benzocycloalkanones including 1a [30, 34]. This primary
alkyl radical undergoes competitive hydrogen atom abstraction and Dowd–
Beckwith ring-expansion to give 3a and 7a, respectively. Ring-expanded radical
7a abstracts a hydrogen atom to become 2a; 7a also undergoes dimerization to give
4. Under ordinary free radical conditions using trialkyltin hydrides, formation of a
dimeric product like 4 has never been reported [46–58]. This would suggest that
trapping of 6a by a hydrogen atom is slow under these conditions compared with
when an effective hydrogen-atom donor, for example an alkyltin hydride, is present.
a-Silyl-substituted amines, for example DETMSA and DPTMSA, are recognized
as silicon-activated amines that are more effective electron donors than non-silyl-
substituted amines, for example TEA [43, 65, 69, 70]. As shown in Scheme 7, their
radical cations undergo efficient desilylation assisted by nucleophiles, for example
specific anions and solvents. On the other hand, basic species abstract a proton from
the position adjacent to the silyl-substituted carbon, generating other a-amino alkyl
radicals [65]. Therefore, increasing the quantities of H2O and DETMSA,
respectively, assist the desilylation and deprotonation processes of the radical
cation of DETMSA (DETMSAÁ?). Consequently, these effects should assist the
progress of PET reaction of 1a; this is, indeed, observed in Tables 1 and 2. Notably,
such a water effect was not observed for reactions with TEA (entries 8 and 9 in
Table 2). In the reaction with DPTMSA, the oxidative dealkylation product,
diphenethylamine, was indeed obtained (entry 7 in Table 2).
To obtain more information about the intermediates involved, we conducted
deuterium-labeling experiments with D2O, expecting that anionic species, if
generated, should be deuterated (Table 3). A significant amount of deuterium was
O
O
O
O
O
O
Br
Br
7a
1) h / DETMSA
2) SET
intra SET
R
2
R
R
R
R
C
Br cleavage
6a
1a
R = CO2Et
7a
1a
4
Br
H
H
O
Me
R
H
R
3a
2a
Scheme 6 Radical pathways in PET reaction of 1a with DETMSA
123

E. Hasegawa et al.
Table 3 Deuterium-labeling experiment with D2O in photoreaction of 1a with DETMSA
Entry Solvent Additive Irrad. time
(h)
Conv. of 1a
(%)
Yields (%)a
2a (D% at
C3)b
3a (D% at
CH3)b
4
1
2
3
4
MeCN D2O
MeCN D2O
MeCN H2O
1
2
2
2
70
90
90
46
64 (81)
60 (82)
59
11 (*0)
7 (*0)
7
Obsdc
Obsdc
Obsdc
25
DMF
D2O
42 (87)
13c
1a (0.64 mmol), DETMSA (5 equiv. vs 1a). Total volume of organic solvent and water (5 %) was kept to
10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
c
Not determined
introduced to the b-carbon of 2a whereas deuterium incorporation at the methyl of
3a was negligible (entries 1, 2 and 4).
These observations could be rationalized by assuming that 7a is reduced to
become anion 8a that is protonated (deuterated) by H2O (D2O) (Scheme 8), in
which most plausible reductants are a-amino alkyl radicals, because of their strong
electron-donating ability [71]. In contrast, the primary alkyl radical 6a in Scheme 6
is not deuterated by D2O.
Photoreaction of 1a with imidazolines (DMPBI, DMPI)
We next conducted photoreaction of 1a with imidazolines DMPBI and DMPI
(Table 4). In contrast with use of the silylamines, formation of 3a became
significant whereas the yield of 4 dropped dramatically, becoming negligible in the
reactions with DMPBI. Although a sizeable solvent effect on product yield was not
observed with DMPBI (entries 1, 3, and 4), addition of water significantly reduced
the yield of 2a and increased that of 3a (entry 2). This tendency is opposite to that
for DETMSA (Table 1). The yield of 3a with DMPBI was greater than that with
Nu
H2O
e
R'
R'
R'
R'
R'
H
H
N
N
NH
+
O
R'
[NuSiMe3]+
R'
R'
N
nucleophiles (Nu): Br , MeCN, MeOH, H2O, etc
SiMe3
B
H2O
SiMe3
e
R'
R'
R'
H
R'
R'
O
N
NH
+
N
SiMe3
R'
SiMe3
BH+
bases (B): ketyl (carobonyl radical anion), amine
Scheme 7 Desilylation and deprotonation of a-silylamine radical cations
123

Photoinduced electron-transfer reaction
Table 4 Photoreaction of 1a with DMPBI or DMPI in different solvents
Entry
Donor (equiv. vs 1a)
Solvent
Conv. of 1a (%)
Yields (%)a
2a
3a
4
1
2
3
4
5
6
7
DMPBI (1.2)
DMPBI (1.2)
DMPBI (1.2)
DMPBI (1.2)
DMPI (1.2)
DMPI (5)
MeCN
aqMeCNb
DMF
PhH
93
62
85
89
70
82
67
31
11
35
36
46
45
32
19
33
22
24
6
4
0
6
6
PhH
14
11
14
PhH
11
34
DMPI (15)
PhH
1a (0.64 mmol), irradiation time: 2 h. Total volume of the mixed solvent was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Aqueous condition: 10 % (v/v) H2O was added
O
O
H2O (D2O)
H(D)
R
R'
R'
R
N
O
2a(D)
8a
R
R'
R'
R'
H2O
N
NH
7a
R = CO2Et
R'
Scheme 8 Ionic pathway in PET reaction of 1a with a-silylamines
DMPI (compare entry 5 with entries 1, 3, and 4). The absence of a significant
solvent effect suggests that solvents do not act as effective hydrogen-atom donors.
In these reactions, plausible hydrogen-atom donors are radical cations and neutral
forms of imidazolines. As the quantity of DMPI was increased, the yield of 3a
increased and that of 2a decreased, although not substantially. This suggests that
DMPI is a less effective hydrogen donor than DMPBI for conversion of the primary
alkyl radical 6a to 3a (entries 5–7).
A deuterium-labeling experiment with D2O further revealed the difference
between the electron and hydrogen-donating properties of DMPBI and DMPI
(Table 5). Significant deuterium incorporation at the b-carbon of 2a was observed
with DMPI but minimum incorporation was observed with DMPBI. Increasing the
quantity of DMPI reduced the yield of 2a and increased that of 3a, which is same as
the tendency reported in Table 4, and the percentage of deuterium incorporated
significantly increased (entries 3 and 4). The latter observation could be rationalized
by assuming more of the a-aminoalkyl radical is generated by deprotonation of the
DMPI radical cation by excess DMPI. It should be also noted that the yield of 2a
was again low and that of 3a was high in aqueous MeCN (compare entry 1 of
123

E. Hasegawa et al.
Table 5 Deuterium-labeling experiments with D2O in the photoreaction of 1a with DMPBI or DMPI
Entry
Donor (equiv. vs 1a)
Solvent
Conv. of 1a (%)
Yields (%)a
2a (D% at C3)b
3ac
4
1
2
3
4
DMPBI (1.2)
DMPBI (1.2)
DMPI (1.2)
DMPI (15)
MeCN
DMF
64
62
60
47
11 (1)
29
21
5
0
4
37 (2)
MeCN
MeCN
41 (63)
27 (80)
30
23
19
1a (0.64 mmol), irradiation time: 2 h. Total volume of organic solvent and D2O (5 %) was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
c
D% at CH3 of 3a was not determined
Table 5 with entry 2 of Table 4). In contrast, such an unexpected water effect was
not observed in aqueous DMF (compare entry 2 of Table 5 with entry 3 of Table 4).
These results suggest that DMPBI directly donates a hydrogen atom to the radical
intermediates 6a and 7a. We then conducted photoreactions of 1a with 2-deuterated
DMPBI (DMPBI-D) (Table 6). Conversion and product yields were substantially
lower than those of reactions using DMPBI (Table 4), which would be consistent
with the different bond-dissociation energy between C–D for DMPBI-D and C–H
for DMPBI. As expected, significant incorporation of deuterium by the b-carbon of
2a and the methyl carbon of 3a were both observed. Particularly noteworthy is that
the percentage of deuterium incorporation was not affected by addition of D2O
(compare entries 4 with 5).
Table 6 Deuterium-labeling experiments with DMPBI-D in the photoreaction of 1a
Entry Solvent Additive Irrad. time
(h)
Conv. of 1a
(%)
Yields (%)a
2a (D% at
C3)b
3a (D% at
CH3)b
4
1
2
3
4
5
6
MeCN
1
4
2
6
6
2
53
49
34
44
41
51
4 (57)
10 (71)
8 (71)
0
MeCN H2O
DMF
Tracec
11 (63)
16 (75)
11 (74)
39 (25)
0
14 (36)
12 (41)
14 (40)
6 (60)
Trace
3
DMF
DMF
PhH
H2O
D2O
Trace
7
1a (0.64 mmol), DMPBI-D (1.2 equiv. vs 1a). Total volume of organic solvent and water (5 %) was kept
to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
c
Not determined
123

Photoinduced electron-transfer reaction
O
O
O
Br
1) PET / DMPBI
R
R
R
2) C -Br cleavage
Me
N
1a
6a
7a
Br
Ph
R = CO2Et
H(D)
N
Me
Me
N
Ph
N
Me
O
O
DMPBI
H(D)
H(D)
R
R
3a
2a(D)
Scheme 9 PET reaction pathways of 1a with DMPBI
On the basis of these results, plausible reaction pathways are proposed in
Scheme 9. Both alkyl radicals 6a and 7a abstract hydrogen atom from the radical
cation of DMPBI (DMPBIÁ?). It has recently been reported that DMPBIÁ? is a more
effective hydrogen-atom donor than its neutral form, because the former gives a
stable imidazolium cation and the latter gives an imidazoline radical [72].
Therefore, DMPBI-D significantly deters these steps to reduce the yields of 2a
and 3a. Indeed, marked deuterium incorporation into these products was observed.
As described above, addition of water (H2O and D2O) significantly changed the
product ratio in MeCN whereas such a water effect was not observed in DMF. At
present, it seems difficult to rationalize these unexpected observations. On the other
hand, water decelerated the reactions to some extent in both solvents (Tables 4; 5).
Water solvation to the initially formed radical anion of 1a (Scheme 6) may interfere
with intramolecular SET between the ketyl part and Cb–Br bond of 1aÁ2. As already
discussed, water accelerates desilylation of DETMSAÁ?, which assists the overall
progress of the reaction, whereas deprotonation of DMPBIÁ? may not be assisted by
water.
As suggested in Scheme 7, deprotonation of amine radical cations could be
achieved by co-existing amines because the acidity of amine radical cations is
known to be comparable with that of alkyl ammonium ions [40–42]. We then
conducted deuterium-labeling experiments of 1a with D2O in the presence of
different amine bases (Table 7). When tert-butylamine (tBuNH2), n-butylamine
(nBuNH2), and diethylamine (Et2NH) were used, deuterium incorporation into 2a
was significant. On the other hand, when the less basic pyridine and 2,6-di-tert-
butylpyridine were used, the percentage of deuterium incorporation into 2a was low.
Because competitive PET reaction between DMPBI and the amine bases added was
considered, photoreaction of 1a with Et2NH was performed to find that Et2NH
indeed promoted the reaction (entries 8 and 12). Interestingly, sizable amounts of 4
were obtained, similar to the reactions with TEA and DMPI (Tables 2, 4, 5).
123

E. Hasegawa et al.
Table 7 Deuterium-labeling experiments with D2O in the photoreaction of 1a with DMPBI
Entry
Base (equiv. vs 1a)
Solvent
Conv. of 1a (%)
Yields (%)a
2a (D% at C3)b
3ac
4
1d
2
MeCN
MeCN
MeCN
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
64
99
91
62
40
44
45
29
47
50
48
34
70
41
11 (1)
29
1
0
0
0
4
4
4
0
Et2NH (12)
18 (58)
47 (5)
3
4e
Pyridine (12)
29
21
23
24
23
13
23
22
25
17
19
22
37 (2)
5
tBuNH2 (12)
nBuNH2 (12)
Et2NH (1.8)
35 (39)
32 (39)
33 (80)
37 (88)
32 (68)
28 (65)
35 (48)
29 (54)
51 (5)
6
7
8f
Et2NH (1.8)
12
0
9
Et2NH (6)
10
11
12f
13
14
Et2NH (12)
4
Et2NH (24)
0
Et2NH (24)
7
Pyridine (12)
2,6-Di-tert-butylpyridine (12)
2
46 (2)
0
1a (0.64 mmol), DMPBI (1.2 equiv vs 1a), irradiation time: 2 h. Total volume of organic solvent and
D2O (5 %) was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
c
D% at CH3 of 3a was not determined
d
Same as entry 1 of Table 5
e
Same as entry 3 of Table 5
f
No DMPBI was added
In contrast, inspection of the quantity of DMPBI remaining in the presence of amine
bases (entries 5, 6, 10, 13 and 14) revealed that more DMPBI was consumed than
the amount of 1a reacted (base: tBuNH2, nBuNH2, Et2NH, pyridine, and 2,6-di-tert-
butylpyridine; reacted 1a: 0.243, 0.285, 0.332, 0.431, and 0.266 mmol; consumed
DMPBI: 0.326, 0.381, 0.424, 0.560, and 0.482 mmol). Therefore, we concluded that
DMPBI more efficiently reacts with 1a than co-existing amine bases.
These results enable us to propose a mechanism for PET reaction of 1a and
DMPBI in the presence of Et2NH (Scheme 10). Et2NH abstracts a proton from
DMPBIÁ? to produce an imidazolyl radical (DMPBIÁ) that reduces radical 7a to
anion 8a. In situ-generated ammonium Et2N?H2 may protonate 8a more efficiently
than D2O, which would be compatible with the fact that the amount of deuterium
incorporated into 2a decreased as more of Et2NH was added (entries 7, 9–11).
One may consider the possibility of an ET-initiated radical chain mechanism
originally proposed by Tanner [73–77]. In this mechanism, initial ET between 1a
and DMPBI generates DMPBIÁ?, and then DMPBIÁ? loses a proton to become
DMPBIÁ, which reduces 1a to the radical anion of 1a. Both 6a and 7a abstract
DMPBI to give DMPBIÁ that promotes the subsequent chain process. We then
123

Photoinduced electron-transfer reaction
O
O
O
Br
1) PET / DMPBI
R
R
R
2) C Br cleavage
1a
6a
7a
Br
Me
N
R = CO2Et
DMPBI
Ph
N
Et2NH Et2NH2
DMPBI
Me
DMPBI
O
O
H2O (D2O)
H(D)
R
R
2a(D)
8a
Scheme 10 Effect of diethylamine on PET reaction pathways of 1a with DMPBI
applied an Et3B–air-initiated reaction, which is known to promote free radical
reaction at room temperature [78, 79], to the above system. Under free radical
conditions, although DMPBI and Et3B (0.2 equiv. vs DMPBI)–air in PhH at room
temperature for 4 h promoted the reaction, conversion of 1a (33 %) and yields of 2a
Table 8 Photoreactions of 1b and 1c with amine donors in different solvents
Entry Substrate
1
Donor (equiv. vs
1)
Solvent
Irrad. time
(h)
Conv. of 1
(%)
Yields (%)a
2
3
5
1
1b
1b
1b
1b
1b
1b
1b
1b
1b
1b
1b
1b
1b
1b
1c
1c
DETMSA (5)
DETMSA (5)
DETMSA (5)
DETMSA (5)
DETMSA (5)
DETMSA (5)
DPTMSA (5)
DMPBI (1.2)
DMPBI (1)
MeCN
aqMeCN
aqMeCN
PhH
4
2
4
4
4
4
4
2
4
4
4
4
4
4
2
2
100
55
80
98
77
83
69
98
88
97
62
47
51
88
59
45
10
16
13
5
8
6
2
6
47
44
52
54
2
3b
4
5
THF
14
8
25 30
17 38
6
aqTHF
PhH
7
7
6
22
18
17
39
4
8
PhH
49
34
43
14
0
9
aqMeCN
aqMeCN
aqMeCN
MeCN
PhH
7
10
11
12
13
14c
15b
16
DMPBI (1.5)
DMPBI (2)
8
27 Trace
DABCO (5)
DABCO (5)
ABCO (5)
0
0
0
2
5
51
50
67
–
0
PhH
0
DETMSA (5)
DMPBI (1.2)
aqMeCN
PhH
40
Trace
–
1 (0.64 mmol). Total volume of organic solvent and H2O (10 %) was kept to 10 mL
a
Isolated yields based on conversion of 1. The yield of 3 was determined by 1H NMR analysis of the
mixture of 1 and 3
b
Average from two independent experiments
c
Conditions: 1b (0.32 mmol), PhH (5 mL)
123

E. Hasegawa et al.
(8 %) and 3a (14 %) were low, and no 4 was obtained (compare with entry 4 of
Table 4). Further information comes from comparison of electrochemical data. The
oxidation potential of DMPBIÁ, which is same as the reduction potential of DMPBI?
(-1.62 V vs SCE, estimated from literature data, -2.06 V vs ferrocene/ferroce-
nium) [72] is lower than the reduction potential of 1a (-1.85 V vs SCE) [30], which
suggests that ET between 1a and DMPBIÁ would not be feasible. Thus, even if it is
operating, its contribution to the reaction pathways should not be significant.
Photoreactions of 1b and 1c with amines
PET reactions of 1b using silylamines, DMPBI, and bicyclic amines were also
investigated (Table 8). Contrasting product distributions are apparent for silylam-
ines and DMPBI. Whereas naphthol 5 was a major product in reactions with
silylamines (entries 1–7), production of 2b and 3b was more significant in reactions
with DMPBI (entries 8–11). Another notable observation is that 5 was exclusively
obtained in reactions using DABCO and 1-azabicyclo[2.2.2]octane (ABCO) (entries
12–14). Finally, experiments revealed that DETMSA was effective and DMPBI
was ineffective for PET-promoted Dowd–Beckwith ring-expansion reaction of 1c
(entries 15 and 16). We have previously observed that 1c is a less effective substrate
than 1a and 1b for free radical Dowd–Beckwith ring-expansion reactions using
tributylstanane and tris(trimethylsilyl)silane [58].
Deuterium-labeling experiments using D2O were conducted for reactions of 1b,
and essentially the same results were obtained as those obtained for reactions of 1a
(Table 9) (see above), i.e. significant deuterium incorporation into the ring-
expansion product 2b was observed with DETMSA whereas no deuterium labeling
of 2b was observed when DMPBI was used.
Selective PET reaction pathways of 1b, depending on three types of tertiary
amine, are presented in Scheme 11. With DETMSA, 2b and 3b are formed via a
sequence similar to reaction of 1a. In contrast, the major product 5, a tautomer of 9,
must be formed via either hydrogen atom elimination from 7b or hydride
elimination from 8b, giving 9, in which hydrogen atom and hydride acceptors could
not be specified. When DMPBI is used as donor, both 6b and 7b efficiently abstract
a hydrogen atom from DMPBIÁ?. When DABCO is used, hydrogen atom abstraction
Table 9 Deuterium-labeling experiment with D2O in photoreaction of 1b with DETMSA or DMPBI
Entry
Donor (equiv. vs 1a)
Solvent
Conv. of 1a (%)
Yields (%)a
2b (D% at C3)b
3bc
5
1
2
DETMSA (5)
DMPBI (1.2)
MeCN
DMF
100
94
18 (80)
32 (0)
11
23
48
3
1a (0.64 mmol), irradiation time: 4 h. Total volume of organic solvent and D2O (5 %) was kept to 10 mL
a
Isolated yields based on conversion of 1b. The yield of 3b was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
c
D% at CH3 of 3b was not determined
123

Photoinduced electron-transfer reaction
from 7b by DABCOÁ? is mainly responsible for formation of 5. DABCOÁ? is known
to be stable [80], so it does not undergo deprotonation but abstracts a hydrogen atom
from 7b to give 9.
Photosensitized reaction of 1a with amines
Photosensitization is an alternative means of promoting PET reactions, and also
attractive from the perspective of organic synthesis [11–13, 20, 23]. By use of
photosensitization, a sensitizer is selectively excited by the light that is not absorbed
by the substrates. As a result, undesired photoreactions arising from the excited
states of the substrates do not occur. We have previously demonstrated that PET
sensitization using electron-donating arene sensitizers and amines is effective in
Table 10 Deuterium-labeling experiment in the BDMAP photosensitized reaction of 1a in DMF
Entry Donor
Additive
Irrad. time
(h)
Conv. of 1a
(%)
Yields (%)a
2a (D% at
C3)b
3a (D% at
CH3)b
4
1
2
3
DETMSA D2O
8
7
7
52
45
68
40 (89)
28 (3)
13 (*0)
18 (*0)
26 (*0)
16
DMPBI
DMPBI
D2O
Trace
Trace
D2O,
41 (83)
Et2NH
4
5
DMPBI-
D
8
42
56
2 (75)
7 (74)
14 (39)
10 (35)
4
DMPBI-
D
H2O
12
12
1a (0.64 mmol), BDMAP (0.05 equiv. vs 1a), DETMSA (5 equiv. vs 1a), DMPBI and DMPBI-D (1.2
equiv. vs 1a), Et2NH (1.8 equiv. vs 1a). Total volume of DMF and water (5 %) was kept to 10 mL
a
Isolated yields based on conversion of 1a. The yield of 3a was determined by 1H NMR analysis of the
mixture of 1a and 3a
b
Determined by 1H NMR analysis (see ‘‘Experimental’’ section)
N
NH
O
OH
O
O
O
Br
1) PET / amine
2) C -Br cleavage
Br
R
R
R
R
R
DABCO
Et2N
1b
R = CO2Et
6b
5
7b
9
Nu
DMPBI
DMPBI
DETMSA
[NuSiMe3]+
Et2N
H2O (D2O)
H
O
O
O
Me
R
R
H(D)
R
8b
3b
2b
Scheme 11 PET reaction pathways of 1b with three types of tertiary amine
123

E. Hasegawa et al.
promoting reductive transformation of a variety of organic substrates (Scheme 5)
[25, 31, 33–39].
We then conducted deuterium-labeling experiments for BDMAP photosensitized
reaction of 1a using DETMSA and DMPBI (Table 10). The results showed a
tendency of deuterium incorporation at the appropriate positions of 2a and 3a
similar to that for direct irradiation of 1a with amines. Namely, the percentage of
deuterium labeling of 2a when D2O was added was high for the reaction with
DETMSA (entry 1) and low for the reaction with DMPBI (entry 2). By contrast,
addition of Et2NH increased both conversion of 1a and deuterium labeling of 2a
(compare entry 3 with entry 2), which is rationalized on the basis of effective
deprotonation of the DMPBI radical cation by Et2NH discussed above. When
DMPBI was replaced by DMPBI-D, deuterium labeling of 2b became significant,
and that of 3a was observed, but the reaction required longer irradiation time and the
yields of 2a and 3a were relatively low in the presence or absence of water.
We next conducted photosensitization reaction of 1a using ADMBI (Chart 1),
which has been reported to be a more effective reagent than DMPBI [38].
Dimethylamino or methoxy-substituted anthracenes and pyrenes were used as
sensitizers (Scheme 5). As shown in Table 11, although all sensitizers effectively
promoted the reaction, the yields of desired 2a were relatively low and less than
those of 3a, and no 4 was obtained.
Conclusion
We have investigated photoinduced electron-transfer (PET) reaction of a-bromo-
methyl-substituted benzocyclic b-keto esters with tertiary amines. In the reactions,
primary alkyl radicals derived from selective carbon–bromine bond cleavage
undergo competitive hydrogen atom abstraction and Dowd–Beckwith ring-expan-
sion. On the basis of detailed mechanistic experiments including deuterium labeling
by use of D2O and DMPBI-D, effects of external amine bases, factors governing the
reaction pathway of generated radical intermediates are discussed from the
perspective of the nature of amine radical cations. Three types of reaction pattern
Table 11 Photosensitized reaction of 1a using ADMBI in DMF
Entry
Sensitizer
Conv. of 1a (%)
Yields (%)a
2a
3a
1
2
3
4
BDMAA
DMA
51
64
77
74
24
16b
19
24
25
31
23
BDMAP
DMP
14b
1a (0.50 mmol), sensitizer (0.05 equiv. vs 1a), ADMBI (1.2 equiv. vs 1a), DMF (5 mL), irradiation time:
5 h
a
Isolated yields based on conversion of 1a
Determined by 1H NMR
b
123

Photoinduced electron-transfer reaction
of amine radical cations are suggested. DETMSAÁ? undergoes desilylation to form
an a-amino alkyl radical that acts as a SET reductant. In contrast, DMPBIÁ? donates
a hydrogen atom to radical intermediates derived from substrate radical anions.
When specific amine bases are added, deprotonation of DMPBIÁ? occurs and the
DMPBIÁ formed then acts as a SET reductant. Finally, DABCOÁ? undergoes neither
deprotonation nor hydrogen atom donation, but abstracts a hydrogen atom from the
radical intermediates to give unsaturated products. A similar reaction pattern of
DABCOÁ? has previously been suggested in PET reaction of epoxy ketones [29].
Photosensitization reactions were also investigated, and essentially the same
reaction patterns and product distributions as those of direct irradiation reactions
were observed. Unfortunately, effective photosensitization conditions promoting
Dowd–Beckwith rearrangement to give the desired ring-expansion products in high
yields have not yet been found; this is certainly a challenge for future investigation.
Acknowledgments This work was partly supported by Grants-in-Aid for Scientific Research from the
Ministry of Education, Culture, Sports, Science and Technology, Japan. We thank Dr Yukiko Matsumura
(Tokyo Institute of Technology) for her assistance with elemental analysis and high-resolution mass
spectroscopy of some of the compounds reported. We thank Professor James M. Tanko (Virginia
Polytechnic Institute and State University, USA) for his help with editing of the manuscript. The
assistance of Mr Tsuneaki Hoshi with preparation of the manuscript is also acknowledged.
References
1. L. Eberson, Electron transfer reactions in organic chemistry (Springer, Berlin, 1987)
2. P.S. Mariano (ed.), Advances in Electron Transfer Chemistry, vol. 1–6 (JAI Press, Greenwich,
1991–1999)
3. V. Balzani (ed.), Electron Transfer in Chemistry, vol. 1–5 (Wiley, Weinheim, 2001)
4. H. Lund, O. Hammerich (eds.), Organic Electrochemistry, 4th edn. (Marcel Dekker, New York,
2001)
5. D.J. Procter, R.A. Flowers, (ed.), Tetrahedron Symposium-in Print, Electron Transfer Reagents in
Organic Synthesis, vol. 65; 52 (2009)
6. M. Schmittel, A. Burghart, Angew. Chem. Int. Ed. Engl. 36, 2550 (1997)
7. D.J. Berger, J.M. Tanko, in The Chemistry of Double-Bonded Functional Groups, ed. by S. Patai
(Wiley, New York, 1997), pp. 1281–1354
8. H.D. Roth, in Reactive Intermediate Chemistry, Chapter 6, ed. by R.A. Moss, M.S. Platz, M. Jones Jr.
(Wiley, Hoboken, 2004), pp. 205–272
9. P.E. Floreancig (ed.), Tetrahedron Symposium-in Print, The Chemistry of Radical Ions, vol 62; 21
(2006)
10. Z.V. Todres, Ion-Radical Organic Chemistry Principles and Applications, 2nd edn. (CRC Press,
Boca Raton, 2009)
11. M.A. Fox, M. Chanon (eds.), Photoinduced Electron Transfer, Parts A–D (Elsevier, Amsterdam,
1988)
12. G.J. Kavarnos, Fundamental of Photoinduced Electron Transfer (VCH Press, New York, 1993)
13. W.M. Horspool, F. Lenci (eds.), CRC Handbook of Organic Photochemistry and Photobiology, 2nd
edn. (CRC Press, Boca Raton, 2004)
14. K. Mizuno, T. Hayamizu, H. Maeda, Pure Appl. Chem. 75, 1049 (2003)
15. J. Cossy, D. Belloti, Tetrahedron 62, 6459 (2006)
16. A.G. Griesbeck, N. Hoffmann, K.D. Warzecha, Acc. Chem. Res. 40, 128 (2007)
17. P.E. Floreancig, Synlett 191 (2007)
18. P.A. Waske, N.T. Tzvetkov, J. Mattay, Synlett 669 (2007)
19. N. Hoffmann, Chem. Rev. 108, 1052 (2008)
20. N. Hoffmann, J. Photochem. Photobio. C 9, 43 (2008)
21. D. Ravelli, D. Dondi, M. Fagnonia, A. Albini, Chem. Soc. Rev. 38, 1999 (2009)
123

E. Hasegawa et al.
22. R.E. Galian, J.P. Prieto, Energy Environ. Sci. 3, 1488 (2010)
23. J.M.R. Narayanam, C.R.J. Stephenson, Chem. Soc. Rev. 40, 102 (2011)
24. D.W. Cho, U.C. Yoon, P.S. Mariano, Acc. Chem. Res. 44, 204 (2011)
25. E. Hasegawa, K. Ishiyama, T. Horaguchi, T. Shimizu, J. Org. Chem. 56, 1631 (1991)
26. E. Hasegawa, Y. Tamura, T. Horaguchi, K. Isogai, T. Suzuki, Tetrahedron Lett. 35, 8643 (1994)
27. E. Hasegawa, H. Katagi, D. Nakagawa, T. Horaguchi, S. Tanaka, Y. Yamashita, Tetrahedron Lett.
36, 6915 (1995)
28. E. Hasegawa, T. Kato, T. Kitazume, K. Yanagi, K. Hasegawa, T. Horaguchi, Tetrahedron Lett. 37,
7079 (1996)
29. E. Hasegawa, K. Ishiyama, T. Fujita, T. Kato, T. Abe, J. Org. Chem. 62, 2396 (1997)
30. E. Hasegawa, Y. Tamura, E. Tosaka, Chem. Commun. 1895 (1997)
31. E. Hasegawa, A. Yoneoka, K. Suzuki, T. Kato, T. Kitazume, K. Yanagi, Tetrahedron 55, 12957
(1999)
32. E. Hasegawa, Y. Tamura, K. Suzuki, A. Yoneoka, T. Suzuki, J. Org. Chem. 64, 8780 (1999)
33. E. Hasegawa, N. Chiba, A. Nakajima, K. Suzuki, A. Yoneoka, K. Iwaya, Synthesis 1248 (2001)
34. E. Hasegawa, S. Takizawa, K. Iwaya, M. Kurokawa, N. Chiba, K. Yamamichi, Chem. Commun.
1966 (2002)
35. E. Hasegawa, J. Photosci. 10, 61 (2003)
36. E. Hasegawa, N. Chiba, T. Takahashi, S. Takizawa, T. Kitayama, T. Suzuki, Chem. Lett. 18 (2004)
37. E. Hasegawa, T. Seida, N. Chiba, T. Takahashi, H. Ikeda, J. Org. Chem. 70, 9632 (2005)
38. E. Hasegawa, S. Takizawa, T. Seida, A. Yamaguchi, N. Yamaguchi, N. Chiba, T. Takahashi,
H. Ikeda, K. Akiyama, Tetrahedron 62, 6581 (2006)
39. E. Hasegawa, H. Hirose, K. Sasaki, S. Takizawa, T. Seida, N. Chiba, Heterocycles 77, 1147 (2009)
40. S.F. Nelsen, J. Am. Chem. Soc. 108, 4879 (1988)
41. J.P. Dinnocenzo, T.E. Banach, J. Am. Chem. Soc. 111, 8646 (1989)
42. V.D. Parker, M. Tilset, J. Am. Chem. Soc. 113, 8778 (1991)
43. U.C. Yoon, P.S. Marinao, R.S. Givens, B.W. Atwater, in Advances in Electron Transfer Chemistry,
vol. 4, ed. by P.S. Mariano (JAI Press, Greenwich, 1994), pp. 117–205
44. F.D. Lewis, in Advances in Electron Transfer Chemistry, vol. 5, ed. by P.S. Mariano (JAI Press,
Greenwich, 1996), pp. 1–39
45. S. Das, V. Suresh, in Electron Transfer in Chemistry, vol. 2, ed. by V. Balzani (Wiley-VCH,
Weinheim, 2001), pp. 379–456
46. P. Dowd, S.C. Choi, J. Am. Chem. Soc. 109, 3493 (1987)
47. P. Dowd, S.C. Choi, J. Am. Chem. Soc. 109, 6548 (1987)
48. A.L.J. Beckwith, D.M. O’Shea, S.W. Westwood, J. Chem. Soc. Chem. Commun. 666 (1987)
49. A.L.J. Beckwith, D.M. O’Shea, S.W. Westwood, J. Am. Chem. Soc. 110, 2565 (1988)
50. P. Dowd, S.C. Choi, Tetrahedron 45, 77 (1989)
51. P. Dowd, W. Zhang, Chem. Rev. 93, 2091 (1993). and references cited therein
52. W.R. Bowman, P.J. Westlake, Tetrahedron 48, 4027 (1992)
53. I. Ryu, H. Fukushima, T. Okuda, K. Matsu, N. Kambe, N. Sonoda, M. Komatsu, Synlett 1265 (1997)
54. C. Chatgilialoglu, C. Ferreri, M. Lucarini, A. Ventrini, A.A. Zavitsas, Chem. Eur. J. 3, 376 (1997)
55. C. Chatgilialoglu, V.I. Timokhin, M. Ballestri, J. Org. Chem. 63, 1327 (1988)
56. M. Sugi, H. Togo, Tetrahedron 58, 3171 (2002)
57. T. Yoshizumi, H. Miyazoe, Y. Sugimoto, H. Takahashi, O. Okamoto, Synthesis 1593 (2005)
58. E. Hasegawa, Y. Ogawa, K. Kakinuma, H. Tsuchida, E. Tosaka, S. Takizawa, H. Muraoka,
T. Saikawa, Tetrahedron 64, 7724 (2008)
59. T. Shono, N. Kse, N. Uematsu, S. Morimoto, E. Okazaki, J. Org. Chem. 55, 5037 (1990)
60. E. Hasegawa, T. Kitazume, K. Suzuki, E. Tosaka, Tetrahedron Lett. 39, 4059 (1998)
61. S.H. Chung, M.S. Cho, J.Y. Choi, D.W. Kwon, Y.H. Kim, Synlett 1266 (2001)
62. M. Sugi, D. Sakuma, H. Togo, J. Org. Chem. 68, 7629 (2003)
63. H. Shimakoshi, M. Abiru, S. Izumi, Y. Hisaeda, Chem. Commun. 6427 (2009)
64. K. Tahara, Y. Hisaeda, Green Chem. 13, 558 (2011)
65. E. Hasegawa, W. Xu, P.S. Mariano, U.C. Yoon, J.U. Kim, J. Am. Chem. Soc. 110, 8099 (1988)
66. H. Chikashita, K. Itoh, Bull. Chem. Soc. Jpn. 59, 1747 (1986)
67. I.-S.H. Lee, E.H. Jeoung, M.M. Kreevoy, J. Am. Chem. Soc. 123, 7492 (2001)
68. CCDC 853453 and 853454 contain the supplementary crystallographic data for this paper. These data
can be obtained free of charge via http://www.ccdc.cam.ac.uk/data_request/cif, by e-mailing
123

Photoinduced electron-transfer reaction
data_request@ccdc.cam.ac.uk or by contacting The Cambridge Crystallographic Data Centre, 12,
Union Road, Cambridge CB2 1EZ, UK; fax: ?44-1223-336033
69. X. Zhang, S.R. Ye, S. Hong, M. Freccero, A. Albini, D.E. Falvey, P.S. Mariano, J. Am. Chem. Soc.
116, 4211 (1994)
70. Z. Su, P.S. Mariano, D.E. Falvey, U.C. Yoon, S.W. Oh, J. Am. Chem. Soc. 120, 10676 (1998)
71. D.D.M. Wayner, D.J. McPhee, D. Griller, J. Am. Chem. Soc. 110, 132 (1988)
72. X.-Q. Zhu, M.-T. Zhang, A. Yu, C.-H. Wang, J.-P. Cheng, J. Am. Chem. Soc. 130, 2501 (2008)
73. J.J. Chen, D.D. Tanner, J. Org. Chem. 53, 3897 (1988)
74. D.D. Tanner, J.J. Chen, J. Org. Chem. 54, 3842 (1989)
75. D.D. Tanner, J.J. Chen, L. Chen, C. Luelo, J. Am. Chem. Soc. 113, 8074 (1991)
76. D.D. Tanner, J.J. Chen, J. Org. Chem. 57, 662 (1992)
77. D.D. Tanner, J.J. Chen, C. Luelo, P.M. Peters, J. Am. Chem. Soc. 114, 713 (1992)
78. K. Nozaki, K. Oshima, K. Utimoto, J. Am. Chem. Soc. 109, 2547 (1987)
79. K. Nozaki, K. Oshima, K. Utimoto, Bull. Chem. Soc. Jpn. 64, 403 (1991)
80. Y.L. Chow, W.C. Danen, S.F. Nelsen, D.H. Rosenblatt, Chem. Rev. 78, 243 (1978)
123