Photochemical 1,3-stannyl rearrangement of allylic stannanes

Article abstract of DOI:10.1039/a707609f

The photochemical 1,3-stannyl rearrangement of allylic stannanes has been investigated. The photorearrangement of (E)-cinnamyl(triphenyl)stannane is not observed in benzene under anaerobic conditions, while the photoinduced 1,3-stannyl migration takes place in the same solvent under aerobic conditions, or in the presence of organic halides or a radical-trapping agent to give a photoequllibrium mixture of the cinnamylstannane and its branched regioisomer, 1-phenylprop-2-enyl(triphenyl)stannane, with the latter predominating. Cinnamyl(trialkyl)stannanes and their homologues also afford the corresponding branched allylstannanes under similar photochemical conditions. These 1,3-stannyl migrations proceed intramolecularly via cinnamyl π-π* excitation in competition with homolytic (cinnamyl)C-Sn bond fission. In contrast, the 1,3-stannyl migration of crotyl- and prenyl-(tributyl)-stannanes is not efficient, but their triphenyl or dibutylphenyl derivatives undergo the 1,3-rearrangement via excitation of the phenyl group(s) on the tin atom to give a regioisomeric mixture of the starting linear tin compounds and the branched ones with the former predominating.

Full text of DOI:10.1039/a707609f

View Article Online / Journal Homepage / Table of Contents for this issue
Photochemical 1,3-stannyl rearrangement of allylic stannanes
Akio Takuwa,* Takashi Kanaue, Koichi Yamashita and Yutaka Nishigaichi
Department of Chemistry, Faculty of Science, Shimane University, Matsue 690, Japan
The photochemical 1,3-stannyl rearrangement of allylic stannanes has been investigated. The
photorearrangement of (E)-cinnamyl(triphenyl)stannane is not observed in benzene under anaerobic
conditions, while the photoinduced 1,3-stannyl migration takes place in the same solvent under aerobic
conditions, or in the presence of organic halides or a radical-trapping agent to give a photoequilibrium
mixture of the cinnamylstannane and its branched regioisomer, 1-phenylprop-2-enyl(triphenyl)stannane,
with the latter predominating. Cinnamyl(trialkyl)stannanes and their homologues also afford the
corresponding branched allylstannanes under similar photochemical conditions. These 1,3-stannyl
migrations proceed intramolecularly via cinnamyl ð-ð* excitation in competition with homolytic
(cinnamyl)C᎐Sn bond fission. In contrast, the 1,3-stannyl migration of crotyl- and prenyl-(tributyl)-
stannanes is not efficient, but their triphenyl or dibutylphenyl derivatives undergo the 1,3-rearrangement
via excitation of the phenyl group(s) on the tin atom to give a regioisomeric mixture of the starting linear
tin compounds and the branched ones with the former predominating.
Introduction
Ph3Sn
Ph
It is well known that most allyl-metal reagents are subjected to
more or less rapid 1,3-allylic rearrangement from 2 to thermo-
dynamically more stable 1.1 Among allyl-metal reagents the
branched allylstannanes 2 (R = Me or Ph, M = SnBu3) are
relatively stable, but they have also been found to undergo some
isomerization to the corresponding linear allylstannane 1 at a
temperature above ~100 ЊC in nonpolar solvents or at an ambi-
ent temperature in polar solvents.2,3 In contrast, interestingly, it
has been reported that allylic silanes4 and germanes5 reversely
rearrange from species 1 to isomers 2 under photochemical
3a
i
ii
iii
3a
3a
3a
+
+
+
SnPh3
Ph
SnPh3
Ph
Ph3SnSnPh3
+
4a
4a
+
R
R
Ph
Ph3SnCl
M
M
Ph
1
2
+
Ph
Ph
conditions. However, unambiguous evidence for such photo-
chemical rearrangement of allylic stannanes from species 1 to
isomers 2 has not been reported.6
Scheme 1 Conditions: i, hν (>320 nm), 1 h, C6H6, N2; ii, hν (>320 nm),
1 h, C6H6, O2; iii, hν (>320 nm), 1 h, CHCl3, N2
Recently, we have communicated that cinnamylstannanes 1
(R = Ph, M = SnBu3 or SnPh3) isomerize photochemically to
the branched regioisomers 2.7 In this paper, we describe in
detail the photochemical 1,3-stannyl migration from carbon to
carbon in allylstannanes. This novel photorearrangement is
observed in a solvent containing organic halides or radical-
trapping agents. Some mechanistic features of this photo-
rearrangement are also discussed.
diphenylhexa-1,5-dienes as by-products clearly indicated that
the homolytic (cinnamyl)C᎐Sn bond cleavage8 took place par-
tially, because they are the self-coupling products of triphenyl-
tin radical and cinnamyl radical, respectively.
On the other hand, irradiation of compound 3a in benzene
under oxygen gave a mixture of regioisomers 3a and 4a (ratio of
1:9) in 50% yield. Thus, the photochemical 1,3-rearrangement
from the cinnamylstannane 3a to the benzylstannane 4a could
be observed in benzene under aerobic conditions. These results
suggest that the homolytic (cinnamyl)C᎐Sn bond cleavage and
the 1,3-stannyl migration process occur competitively (Scheme
2). Even though the branched allyltin 4a was formed under
anaerobic conditions, it would be reconverted to the starting
linear allyltin 3a by free-radical chain reaction involving
an SH2Ј process9 between compound 4a and the triphenyltin
radical. Such an SH2Ј reaction, however, would be inhibited
under aerobic conditions because of the triphenyltin radical
reacting rapidly with oxygen.10 These two competitive pro-
cesses were also supported by the result from the photolysis in
chloroform. Irradiation of compound 3a in chloroform under
anaerobic conditions gave again a regioisomeric mixture of 3a
and 4a (ratio of 1:13) in 55% yield together with triphenyltin
Results and discussion
Although the photochemical 1,3-allylic rearrangement 1 → 2
of allylic silanes4 and germanes5 has been observed in a
benzene or hexane solution, the photochemical 1,3-stannyl
rearrangement of (E)-cinnamyl(triphenyl)stannane 3a was not
observed in these solvents. For example, when irradiation of
compound 3a was carried out in benzene under nitrogen with a
300-W high-pressure mercury arc lamp through saturated aq.
CuSO4 (>320 nm), no 1,3-rearrangement product, 1-phenyl-
prop-2-enyl(triphenyl)stannane 4a, not obtained but a small
amount of hexaphenyldistannane and diphenylhexa-1,5-dienes
were produced together with recovery of substrate 3a (78%)
(Scheme 1). The formation of hexaphenyldistannane and
J. Chem. Soc., Perkin Trans. 1, 1998
1309

View Article Online
Table 1 Thermal reactions of compounds 3a and 4a in the presence of
of organic halides towards triphenyltin radical; the reactivity
AIBNa
increases in the order primary < secondary < tertiary < allyl for
different alkyl groups in the alkyl halides, and the ease of
abstraction of halogens increases in the order Cl < Br.8d,11c,d,12
Unexpectedly, iodoform was not an effective organic halide for
the photorearrangement probably due to its absorption of the
light (>320 nm; data not shown). Thus, the 4a:3a ratios were
very high in benzene solution containing CCl4, CHCl3, CH2Br2,
Run Allyltin Conditions
Ratio of 3a:4ab Total yield (%)
1
2
3
4
3a
4a
4a
4a
AIBN–C6H6–N2
AIBN–C6H6–N2
AIBN–C6H6–O2
AIBN–CHCl3–N2
100:0
100:0
7:93
96
97
91
73
1:99
CH ᎐CHCH Br or PriBr. Similar ratios were obtained when the
᎐
2
2
a The reactions were carried out under reflux for 6 h in the presence of
10 mol% AIBN under the conditions described in the Table. b Ratios
were determined by 1H NMR spectra.
branched allyltin 4a was subjected to photolysis in chloroform
or in benzene containing CH2Br2 (runs 14,15). Thus, the photo-
equilibrium should be attained between compound 3a and 4a
under these conditions, in which the sterically more crowded
compound 4a was the major product. The 1,3-rearrangement
was also observed when compound 3a was irradiated in ben-
zene containing radical scavengers such as 2,2,6,6-tetramethyl-
piperidine-N-oxyl (TEMPO) or galvinoxyl (runs 11,12).
SnPh3
hν
Ph3Sn
Ph
Ph
3a
4a
hν
In addition, we further confirmed the photochemical
(cinnamyl)C᎐Sn bond fission in competition with 1,3-stannyl
migration by a crossover experiment between compound 3a and
allyltributylstannane 5 (Scheme 3). When an equimolar mixture
•
Ph
+
RX
Ph3Sn•
+
Bu3Sn
ii
Ph3Sn
Ph
i
3a
5
R•
Ph3SnX
Ph3Sn
Ph
3a
3a
Scheme 2
+
+
5
5
chloride (30%), where the triphenyltin radical could abstract a
chlorine atom8c–e from the solvent (Scheme 2).†
+
+
The SH2Ј processes were confirmed by the following experi-
ments under free-radical conditions and the results are sum-
marized in Table 1. The linear allyltin 3a was recovered intact
after being refluxed in benzene containing a catalytic amount of
2,2Ј-azoisobutyronitrile (AIBN) under nitrogen, whereas the
branched allyltin 4a isomerized completely to compound 3a
under the same conditions. However, the allyltin 4a was
recovered without isomerization to the cinnamyltin 3a when
refluxed with AIBN in benzene solution under oxygen or in
chloroform solution (runs 3 and 4). Since it is known that
tributyltin radical is generated by heating of allyl(tributyl)-
stannane in the presence of AIBN,8d the results of the present
radical reactions clearly indicate that triphenyltin radical con-
verts benzyl isomer 4a into cinnamyl isomer 3a but it is trapped
efficiently with oxygen10 or with chloroform8c–e before select-
ively attacking the sterically less crowded terminal olefinic
carbon in compound 4a rather than the sterically more con-
gested internal one in isomer 3a. This is a reason why photo-
chemical 1,3-rearrangement of compound 3a was not observed
in benzene or hexane under anaerobic conditions, but was
observed under aerobic conditions or in chloroform solution.
Thus, in order to observe directly the photochemical 1,3-
rearrangement of compound 3a it was found necessary to sup-
press the SH2Ј chain reaction between the 1,3-stannyl-migrated
product 4a and the triphenyltin radical formed by homolytic
cleavage of (cinnamyl)C᎐Sn bond. Therefore, we examined the
photolysis of compound 3a in benzene solution containing
various organic halides, from which the halogen atom is known
to be readily transferred to trialkyltin radical.8c–e,11 The 1,3-
rearrangement was again observed under the conditions shown
in Table 2. The ratio of products 3a and 4a obtained varied
from 1:0 to 1:19 depending on the organic halides used as an
additive. The amount of 4a increased with increasing reactivity
SnPh3
Ph
Bu3Sn
Ph
4a
3c
+
Ph3Sn
6
Scheme 3 Conditions: i, hν (>320 nm), 1 h, C6H6, N2; ii, hν (>320 nm),
1 h, CHCl3, N2
of allylstannanes 3a and 5 in benzene was irradiated (>320 nm)
for 1 h under nitrogen, cinnamyl(tributyl)stannane 3c (15%)
and allyltriphenylstannane 6 (22%) were obtained together with
starting tin reagents, 3a (76%) and 5 (48%). Since compound
5 does not absorb light of longer wavelength than 280 nm,
(cinnamyl)C᎐Sn bond fission of compound 3a occurred select-
ively under these conditions to produce the triphenyltin radical,
which attacks the terminal olefinic carbon in compound 5 to
give the cross-coupling product 6 and tributyltin radical via
SH2Ј reaction [equation (1)]. The resulting tributyltin radical
reacts with photorearranged product 4a from substrate 3a
to afford another cross-coupling product 3c [equation (2)]. On
Ph3Sn•
–Bu3Sn•
(1)
5
Bu3Sn
SnPh3
6
•
Ph
Bu3Sn•
–Ph3Sn•
(2)
4a
3c
Ph3Sn
SnBu3
•
the other hand, the cross-coupling reaction was negligible
when the photolysis was carried out in chloroform, and 1,3-
rearrangement product 4a (3a:4a = 6:94) was again observed
without any contamination of the cross-coupling products
(Scheme 3).
† Cinnamyl radical was also scavenged by dimerization or coupling
with chloromethyl radicals to afford small amounts of diphenyl-1,5-
hexadienes, 4,4-dichloro-1-phenylbut-1-ene and 3-trichloromethyl-1-
phenylprop-1-ene.
1310
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Table 2 Photochemical isomerization of [(E)-3-arylbut-2-enyl]tin compounds 3a–d and 4aa
Run
Allyltin
Solvent
Additive
Ratio of 3:4b
E:Z ratio of 3b
Total yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
4a
4a
3b
3c
3c
3de
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Chloroform
Chloroform
Benzene
[2H]Chloroform
Benzene
Benzene
Benzene
CH2Br2
CH2Cl2
CHCl3
6:94
96:4
14:86
5:95
13:87
6:94
99:1
95:5
8:92
78:22
6:94
64:36
7:93
10:90
7:93
1:99
3:97
7:93
12:88
40:60
82:18
47:53
20:80
75:25
27:73
93:7
86:14
67:33
81:9
42:58
98:2
53:47
29:71
38:62
54
71
47
24
59
55
80
73
57
65
64
84
55
42
52
53
58
68
76
CCl4
BunBr
PriBr
BunCl
ButClc
Allyl bromide
Allyl chloride
TEMPOd
Galvinoxyle
CH2Br2
CH2Br2
TEMPOd
CH2Br2
50:50
30:70
38:62
a Irradiation was carried out in a solvent containing 3 equivalents of an additive under nitrogen with a 300-W high-pressure mercury lamp (>320 nm)
1
for 1 h. b Ratios were determined by H NMR spectra. c Butyl chloride (10 equiv.) was added. d TEMPO (1.0 equiv.) was added. e Galvinoxyl (0.2
equiv.) was added. e Irradiation was carried out with >400 nm light for 40 min.
gradually with irradiation time due to C᎐Sn bond fission of
both compounds 3a and 4a occurring in parallel with the 1,3-
stannyl migration. The photochemical 1,3-stannyl rearrange-
ment was also tested by using allyltin compounds bearing
substituents on the tin center or on the allylic framework
(Scheme 4). When all the phenyl groups on the tin in compound
3a were replaced by methyl (in 3b) or butyl groups (in 3c), the
1,3-stannyl migration also proceeded under the same photo-
chemical conditions used for compound 3a, to give a regio-
isomeric mixture of products 3 and 4 in which compound 4b
or 4c was obtained as the major regioisomer, respectively
(runs 16–18 in Table 2; Scheme 4). Thus, the substituents (R) on
hν
Ar
R3Sn
R3Sn
Ar
3
4
a; R = Ar = Ph; b; R = Me, Ar = Ph;
Fig. 1 Photoisomerization of (E)-3a to 4a in a nitrogen purged ben-
zene solution in the presence of dibromethane (3 equiv.) under irradi-
ation with >320 nm light. Ratio of 4a:3a (᭹); Z:E ratio of recovered
3a (᭡); total yield of 3a and 4a (᭺).
c; R = Bu, Ar = Ph; d; R= Ph, Ar = pyren-1-yl
Scheme 4
Next, attention was focused on the E:Z ratio of compound
3a. While starting compound 3a was the E-form, the recovered
3a was a mixture of E/Z-isomers (Table 2). For example, the
E:Z ratio of recovered substrate 3a was 1:1.5 after irradiation
of a benzene solution of (E)-3a in the presence of dibromo-
methane (Table 2, run 1). Then the variation of both the E:Z
ratio of compound 3a and the ratio of products 3a:4a was
followed as a function of irradiation time under the same con-
ditions. The results are shown in Fig. 1, which also includes the
total yield of compounds 3a and 4a.
Equilibrium of the 1,3-rearrangement between 3a and 4a was
accomplished after irradiation for 50 min (ratio of compound
3a:4a 1:19). Interestingly, the Z-isomer was little observed in
recovered compound 3a until equilibrium was attained, but the
amount of Z-isomer increased dramatically soon after that
and (Z)-3a became the main geometrical isomer with further
continuous irradiation. This phenomenon implies that the
photoexcited isomer (E)-3a rearranges more rapidly to the
regioisomer 4a than to the geometrical isomer (Z)-3a. Since
compound (Z)-3a could be formed alternatively by the reverse
photochemical 1,3-stannyl migration (4a → 3a) (Table 2,
runs 14, 15), the proportion of (Z)-3a increases steeply around
the equilibrium state, in which isomer 4a becomes the main
regioisomer. The total yields of isomers 3a and 4a decreased
the tin have no influence on the photorearrangement. In add-
ition, [3-(pyren-1-yl)prop-2-enyl]triphenylstannane 3d having
an α-pyrenyl group in the allylic moiety also gave a regio-
isomeric mixture of products 3d and 4d in the ratio 1:7 even
under irradiation with light of longer wavelength than 400 nm
(run 19 in Table 2). These results revealed clearly that the pres-
ent photochemical 1,3-rearrangement of cinnamyl(trialkyl)-
stannanes should involve direct photoexcitation of the
cinnamyl π system. This was further supported by an inspection
of electronic absorption spectra of these allylic stannanes. (E)-
Cinnamyl(triphenyl)stannane 3a and (E)-cinnamyl(tributyl)-
stannane 3c absorb light at nearly the same wavelength [λmax
(log ε) for 3a: 276 nm (4.38); for 3c: 273 nm (4.36)], whereas
compound 3d absorbs light of longer wavelength [λmax = 366
nm (log ε = 4.54)] than do analogues 3a and 3c. Therefore, these
absorption maxima are not due to a triphenyltin or tributyltin
moiety but are due to allylic moieties in these allylstannanes.
Intramolecularity of the present photochemical 1,3-stannyl
migration was examined by a crossover experiment (Scheme 5).
When an equimolar mixture of cinnamyl(methyl)diphenyl-
stannane 7 and [3-(p-tolyl)prop-2-enyl]triphenylstannane 8 in
chloroform was photolysed for 1 h, intramolecular rearrange-
ment products, 1-phenylprop-2-enyl(methyl)diphenylstannane
9 (38%) and [1-(p-tolyl)prop-2-enyl]triphenylstannane 10 (46%)
J. Chem. Soc., Perkin Trans. 1, 1998
1311

View Article Online
Table 3 Photoisomerization of crotyl- and prenyl-tins 11
Run
Allyltin
Irradn (λ/nm)
Solvent (additive)
Irradn time (t/h)
Ratio of 11:12a
Total yield (%)b
1
2
3
4
5
6
7
8
9
11a
11b
11b
11b
11e
11e
11e
11c
11d
11e
280
280
280
254
280
280
280
280
280
280
CHCl3
CHCl3
CHCl3
CHCl3
0.5
0.5
1
1
0.5
1
1.5
3
3
100:0
92:8
92:8
88:12
84:16
78:22
77:23
100:0
94:6
70
53
30
49
72
44
32
65
54
48
C6H6 (CH2Br2)c
C6H6 (CH2Br2)c
C6H6 (CH2Br2)c
C6H6 (TEMPO)d
C6H6 (TEMPO)d
C6H6 (TEMPO)d
10
3
72:28
a Ratios were determined by 1H NMR spectra. b Yields were determined by 1H NMR spectra using p-cyanobenzaldehyde as internal standard.
c Dibromomethane (3 equiv.) was added. d TEMPO (0.2 equiv.) was added.
and prenyl-stannanes involves not direct photoexcitation of the
+
Ph2MeSn
Ph2MeSn
Ph3Sn
Ph
Ph3Sn
Tol-p
allylic π system but should involve an aromatic π-π* excitation
of phenyl substituents at the tin atom as reported in photo-
chemical 1,3-silyl migration of allylic silanes with aromatic
substituents at the silicon centre.4 Further work should be re-
quired to solve the origin of the different mechanism between
cinnamylstannanes and crotyl- or prenyl-stannanes.
7
8
Ph
+
+
Ph2MeSn
Ph
7
9
+
+
Experimental
Tol-p
Mps were determined on a Yanaco melting point apparatus
and are uncorrected. The 1H NMR spectra were recorded on a
JEOL JNM GX270 FT spectrometer with tetramethylsilane
(TMS) as internal standard: coupling constants (J) are given in
Hz. The IR spectra were recorded on a Hitachi 260-50 instru-
ment. The UV spectra were measured on a JASCO UVICEC-
510 or on a Shimadzu UV3100 spectrometer. Preparative TLC
(silica gel 60 PF254) was used after being dried in an air oven at
120 ЊC for 2 h. Flash chromatography was performed on silica
gel 60 (E. Merck). All solvents and organic halides were dis-
tilled and dried before use by standard procedures.
Tol-p
Ph2MeSn
10
8
Scheme 5 Conditions: hν (>320 nm), 1 h, CHCl3, N2
were obtained together with the recovery of reactants 7 (2.4%)
and 8 (4%). No crossover products were detected in the reaction
mixture. Thus, the photochemical 1,3-stannyl migration of
cinnamylstannanes involves photoexcitation of the cinnamyl π
system and proceeds intramolecularly via an orbital symmetry
allowed concerted [1,3]-sigmatropic mechanism5 accompanying
a side reaction of (cinnamyl)C᎐Sn bond fission. In contrast to
the reported thermal rearrangement,2 the present photo-
rearrangement proceeds towards the opposite direction, giving
sterically more crowded allylstannanes as the major product at
the photostationary state.
Materials
(E)-Cinnamyl(triphenyl)stannane 3a,13 (E)-cinnamyl(trimeth-
yl)stannane 3b,14 (E)-cinnamyl(tributyl)stannane 3c,15 allyl-
(tributyl)stannane 5,15 but-2-enyl(tributyl)stannane 11a,16
but-2-enyl(triphenyl)stannane 11b16 and 3-methylbut-2-enyl-
(tributyl)stannane 11c15 were prepared using previously
reported methods.
Photoisomerization of but-2-enylstannane (crotylstannane)
and 3-methylbut-2-enylstannane (prenylstannane) was also
examined (Scheme 6), and the results are summarized in Table
(E)-[3-(Pyren-1-yl)prop-2-enyl]triphenylstannane 3d. Butyl-
lithium (14.8 ml of a 1.61 mol dmϪ3 solution in hexane, 24
mmol) was added dropwise to a solution of 1-bromopyrene
(5.62 g, 20 mmol) in diethyl ether (80 ml)–benzene (20 ml) at
room temperature under nitrogen. After stirring of the mixture
for 30 min, 3-bromopropene (3.15 g, 26 mmol) was added to the
solution and the mixture was stirred for 20 min before being
quenched by the addition of saturated aq. NH4Cl in an ice-
bath. The mixture was extracted with diethyl ether (50 ml × 3),
and the organic layer was washed with brine, dried (Na2SO4),
and concentrated under reduced pressure. Flash chrom-
atography of the residue on silica gel, using hexane as eluent,
gave 3-(pyren-1-yl)propene (3.2 g, 66%) as a pale yellow oil.
Butyllithium (6.2 ml, 10 mmol) was added dropwise to a
solution of the foregoing 3-(pyren-1-yl)propene (2.42 g, 10
mmol) in THF (20 ml) at Ϫ70 ЊC under nitrogen, followed by a
solution of triphenyltin chloride (3.85 g, 10 mmol) in THF (10
ml) at Ϫ30 ЊC. The mixture was allowed to warm to room tem-
perature, was stirred for 1.5 h at that temperature, and was
quenched by aq. KF (10%). The resulting precipitate was fil-
tered off and the filtrate was extracted with diethyl ether. The
combined organic layer was washed with brine, dried (Na2SO4),
and concentrated under reduced pressure to afford a crude yel-
lowish crystalline product. Recrystallization of the crude prod-
uct from benzene–hexane and then from chloroform–diethyl
hν
R
Me
R
R3Sn
Me
R3Sn
11
12
a; R = Bun, R = H; b; R = Ph, R = H; c; R = Bun, R = Me;
d; R3 = Bun2Ph, R = Me; e; R = Ph, R = Me
Scheme 6
3. Crotyl(tributyl)stannane 11a and prenyl(tributyl)stannane
11c were inert under irradiation with the light through a Pyrex
filter (>280 nm), whereas the corresponding (triphenyl)-
stannanes, 11b and 11e, did undergo 1,3-stannyl rearrangement
under the same conditions to give a mixture of compounds 11b
and 12b or 11e and 12e, respectively, in which the starting linear
tin compound (11b or 11e) was the major regioisomer.
In addition, the photoisomerization of prenyl(dibutyl)phenyl-
stannane 11d was also observed although photoequilibrium
again lay close to the starting prenylstannanes (run 9). Thus, in
contrast to cinnamylstannanes, photoisomerization of crotyl-
and prenyl-stannanes depended on the substituent at the tin
atom, i.e. the 1,3-stannyl migration could be observed when the
stannanes had phenyl group(s) at the tin centre. These results
suggest that the photochemical 1,3-stannyl migration in crotyl-
1312
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
ether gave the title compound 3d (0.9 g, 28%) as yellow crystals,
mp 143 ЊC (Found: C, 75.55; H, 4.43. C37H28Sn requires C,
75.16; H, 4.77%); δH 2.81 (2H, d, J 8.6), 6.69 (1H, dt, J 15.4 and
8.6) and 7.29–8.11 (25H, m, one olefinic and ArH); δC 18.5,
123.5, 123.7, 124.6, 124.9, 125.0, 125.8, 126.6, 126.9, 127.3,
127.5, 128.3, 128.7, 129.2, 129.9, 131.1, 131.6, 132.2, 133.1,
136.9, 137.2, 137.4 and 138.3; λmaxCHCl3/nm 288 (log ε 3.70)
and 366 (3.83).
(E)-Cinnamyl(methyl)diphenylstannane 7. This tin reagent was
prepared by the coupling reaction of cinnamylmagnesium
chloride with methyldiphenyltin iodide17 following a previously
described method.13 The crude tin reagent thus obtained was
purified by flash chromatography on silica gel with hexane as
eluent: oil (Found: C, 65.08; H, 5.66. C22H22Sn requires C,
65.23; H, 5.47%); δH 0.52 (3H, s), 2.37 (2H, d, J 8.3, CH2), 6.28
(1H, d, J 15.6), 6.43 (1H, dt, J 15.9 and 8.3) and 7.10–7.58
(15H, m); νmax/cmϪ1 3060, 3020, 1635, 1600, 1435, 1085, 958,
750, 730 and 700.
(E)-[3-(p-Tolyl)prop-2-enyl]triphenylstannane 8. This tin
reagent was prepared by the coupling reaction of 3-(p-tolyl)-
prop-2-enylmagnesium chloride with triphenyltin chloride fol-
lowing a previously described method:13 mp 86–87 ЊC (from
methanol) (Found: C, 69.87; H, 5.51. C28H26Sn requires C,
69.89; H, 5.45%); δH 2.30 (3H, s), 2.60 (2H, d, J 8.1, CH2), 6.30
(1H, d, J 15.6), 6.42 (1H, dt, J 15, and 8.1), 7.02–7.10 (4H, m)
and 7.25–7.63 (15H, m).
3-Methylbut-2-enyl(dibutyl)phenylstannane 11d. This tin
reagent was prepared by the coupling reaction of 3-methylbut-
2-enylmagnesium chloride with dibutylphenyltin iodide17 fol-
lowing a previously described method.15 The crude tin reagent
thus obtained was purified by column chromatography on
alumina with hexane as eluent: oil (Found: C, 59.82; H, 8.86.
C19H32Sn requires C, 60.19; H, 8.51%); δH 0.66 (6H, t, J 7.1),
1.04–1.09 (4H, m), 1.25–1.39 (4H, m), 1.48–1.60 (4H, m), 1.54
(3H, s), 1.67 (3H, s), 1.88 (2H, d, J 8.8), 5.34 (1H, t, J 8.8) and
7.30–7.47 (5H, m); νmax/cmϪ1 3060, 2960, 2925, 2850, 1460,
1430, 1375, 1075, 725 and 700.
products was also determined by integration of their signal
intensities using p-cyanobenzaldehyde as internal standard. The
product was isolated by PLC, developing with 9:1 hexane–
dichloromethane. The results are given in Tables 2 and 3, and in
Fig. 1.
Typical irradiation procedure for (E)-cinnamyl(triphenyl)-
stannane 3a in benzene under anaerobic or aerobic conditions
A nitrogen-purged solution of compound 3a (46.7 mg, 0.1
mmol) in benzene (5 ml) was irradiated through saturated aq.
CuSO4 with a high-pressure Hg lamp for 1 h. The solvent was
removed under reduced pressure and 1H NMR measurement of
the residue showed no formation of a 1,3-stannyl migration
product. The residue was chromatographed on PLC, develop-
ing with 9:1 hexane–dichloromethane. The Rf 0.18 band con-
tained 4 mg of hexaphenyldistannane: mp 230–232 ЊC (lit.,19
229–232 ЊC). The Rf 0.57 band contained 36.4 mg (78%) of
compound 3a (E/Z = 94:6). The Rf 0.77 band contained 5 mg
of (E)-1,4-diphenylhexa-1,5-diene3b contaminated by 1,6-
diphenylhexa-1,5-diene (ratio 92:8): δH 2.64 (2H, dd, J 7.3 and
7, 3-H2), 3.43 (1H, dt, J 7.3, 4-H), 5.07 (2H, m, 6-H2), 5.96–6.06
(1H, m, 5-H), 6.12 (1H, dt, J 7, 2-H), 6.39 (1H, d, J 15, 1-H)
and 7.06–7.43 (10H, m, ArH); νmax(CHCl3)/cmϪ1 3080, 3000,
2920, 1638, 1600, 1495, 1452, 1075, 1030, 990, 970 and 920.
A benzene solution of compound 3a was purged with oxygen
for 15 min and irradiated under the above conditions. The reac-
tion mixture was washed with saturated aq. NaHCO3, and the
organic layer was dried (Na2SO4). After evaporation of the
mixture, the product was isolated by TLC, developing with
hexane–dichloromethane (9:1) to give a regioisomeric mixture
of compounds 3a and 4a (ratio 3a:4a = 11:89) in 51% yield.
Spectroscopic and physical properties of rearranged allylic
stannanes 4 and 12
Spectroscopic properties of 1-phenylprop-2-enyl(tributyl)-
stannane 4c, but-3-en-2-yl(tributyl)stannane 12a, and 2-methyl-
but-3-en-2-yl(tributyl)stannane 12c agreed with the data
reported in ref. 2.
1-Phenylprop-2-enyl(triphenyl)stannane 4a. Mp 104 ЊC (from
methanol) (Found: C, 69.02; H, 5.19. C27H24Sn requires C,
69.42; H, 5.18%); δH 4.16 (1H, d, J 9.5, 1-H), 4.81–5.10 (2H, m,
3-H2), 6.44 (1H, dt, J 17 and 10, 2-H), 7.02–7.20 (5H, m, ArH)
and 7.25–7.50 (15H, m, ArH); δC 42.9, 111.6, 124.8, 127.2,
128.4, 128.6, 129.0, 137.4, 138.1, 138.8, 142.2.
3-Methylbut-2-enyl(triphenyl)stannane 11e. This tin reagent
was prepared by the reaction of triphenyltin-lithium with 3-
methylbut-2-enyl bromide following a previously described
method:15 mp 61–63 ЊC (from methanol) (Found: C, 65.69; H,
5.55. C23H24Sn requires C, 65.91; H, 5.77%); δH 1.43 (3H, s),
1.64 (3H, s), 2.35 (2H, d, J 8.8), 5.46 (1H, br t, J 8.8) and 7.34–
7.62 (15H, m).
1-Phenylprop-2-enyl(trimethyl)stannane 4b. Oil; δH 0.06 (9H,
s, SnMe3), 3.48 (1H, d, J 10.3, 1-H), 4.79–4.94 (2H, m, 3-H2),
6.33 (1H, dt, J 16.8 and 10.1, 2-H) and 7.03–7.41 (5H, m, ArH).
[1-(Pyren-1-yl)prop-2-enyl]triphenylstannane 4d. This com-
pound isomerized to compound 3d with partial decomposition
on TLC. It was isolated as follows: After irradiation of com-
pound 3d in benzene containing 3 equivalent of dibromo-
methane with light of wavelength >400 nm for 40 min, the
solvent was removed under reduced pressure. Diethyl ether was
added to the residue and the resulting ethereal solution was
washed with aq. KF (10%), dried over Na2SO4 Evaporation of
the solution gave an oil, which was triturated with chloroform–
hexane and left overnight in a refrigerator (~ Ϫ20 ЊC) to give
compound 4d as crystals: mp 116–118 ЊC (Found: C, 74.97; H,
4.52. C37H28Sn requires C, 75.16; H, 4.77%); δH 4.96–5.12 (2H,
m, 3-H2), 5.24 (1H, d, J 8.5, 1-H), 6.62–6.80 (1H, d, 2-H), 7.18–
7.45 (15H, m, Ph3Sn) and 7.85–8.25 (9H, m, ArH); δC 39.6,
112.3, 123.2, 124.5, 124.7, 125.1, 126.0, 126.3, 126.9, 127.4,
127.6, 128.1, 128.4, 128.7, 128.8, 129.0, 130.9, 131.6, 136.5,
136.8, 137.2, 137.5, 138.3 and 139.9; νmax(CHCl3)/cmϪ1 3060,
3050, 2920, 1620, 1600, 1585, 1480, 1430, 1075, 1020, 1000, 900
and 854.
General procedure for irradiation
Irradiation was carried out with a 300-W high-pressure Hg arc
lamp through a saturated aq. CuSO4 filter about 1 cm in thick-
ness (>320 nm), or through a Pyrex filter (>280 nm), or through
a saturated solution of CuSO4 and NaNO2 in aq. NH4OH ~1
cm in thickness (>400 nm).18 Irradiation with light of 254 nm
was carried out with a 60-W low-pressure Hg arc lamp in a
quartz vessel. An allyltin compound in benzene or in benzene
containing
2,6-di-tert-butyl-α-(3,5-di-tert-butyl-4-oxocyclo-
hexa-2,5-dien-1-ylidine)-p-tolyloxyl (galvinoxy) or TEMPO
was purged with nitrogen and irradiated for 1 h. After evapor-
ation off of the solvent, the residue was analysed by means of
1H NMR spectroscopy, from which the product ratio was
determined by integration of signal intensities. The products
were isolated by preparative TLC (PLC) developing with 9:1
hexane–dichloromethane.
Irradiation of an allyltin compound in chloroform or in ben-
zene solution containing organic halides was carried out under
similar conditions to those described above, but the work-up
procedure was different from that above. After irradiation, the
reaction mixture was washed with saturated aq. NaHCO3, and
the organic phase was dried (Na2SO4). After evaporation off of
But-3-en-2-yl(triphenyl)stannane 12b. Mp 74–75 ЊC; δH 1.52
(3H, d, J 7.3, CH3), 2.88 (1H, dq, J 7.2 and 7.1, 2-H), 4.84 (1H,
d, J 10, 4-H), 4.89 (1H, d, J 17.1, 4-H), 6.28 (1H, ddd, J 17.1, 10
and 7.1, 3-H) and 7.36–7.62 (15H, m, ArH); νmax(KBr)/cmϪ1
1
the solvent, the residue was analysed by means of H NMR
spectroscopy, from which the product ratio was determined by
integration of signal intensities. In some cases, the yield of
J. Chem. Soc., Perkin Trans. 1, 1998
1313

View Article Online
2 (a) V. J. Jephcote and E. J. Thomas, J. Chem. Soc., Perkin Trans. 1,
1991, 429; (b) I. Fleming and M. Rowley, J. Chem. Soc., Perkin
Trans. 1, 1987, 2259.
3 (a) J. A. Verdone, J. A. Mangravite, N. M. Scarpa and H. G. Kuivila,
J. Am. Chem. Soc., 1975, 97, 843; (b) B. M. Trost and E. Keinan,
Tetrahedron Lett., 1980, 21, 2595.
4 M. Kira, T. Taki and H. Sakurai, J. Org. Chem., 1989, 54, 5647.
5 M. Kobayashi and M. Kobayashi, Chem. Lett., 1986, 385.
6 Y. Watanabe, T. Yoneda, T. Okumura, Y. Ueno and T. Toru, Bull.
Chem. Soc. Jpn., 1993, 66, 3030.
3050, 2950, 2900, 1620, 1480, 1430, 1140, 1070, 1020, 1000, 960,
850, 730 and 700.
2-Methylbut-3-en-2-yl(dibutyl)phenylstannane
12d.
Oil
(Found: C, 59.82; H, 8.86. C19H32Sn requires C, 60.19; H,
8.51%); δH 0.85–1.80 (24H, m, Me2 ϩ Bu2), 4.62 (1H, dd, J 17.3
and 1.5, 4-H), 4.73 (1H, dd, J 10.5 and 1.5, 4-H), 6.07 (1H, dd,
J 17.3 and 10.5, 3-H) and 7.28–7.57 (5H, m, ArH).
2-Methylbut-3-en-2-yl(triphenyl)stannane 12e. Mp 67–69 ЊC;
δH 1.49 (6H, s, Me2), 4.75–4.87 (2 H, m, 4-H2), 6.26 (1H, dd,
J 17.2 and 10.6, 3-H), 7.35–7.56 (15H, m, ArH); νmax/cmϪ1 3045,
3015, 2930, 2850, 1620, 1475, 1424, 1115, 1072, 995, 890, 723
and 700.
7 A. Takuwa, T. Kanaue, Y. Nishigaichi and H. Iwamoto, Tetrahedron
Lett., 1995, 36, 575.
8 (a) N. Soundararajan and M. S. Platz, Tetrahedron Lett., 1987, 28,
2813; (b) P. J. Barker, A. G. Davies and M.-W. Tse, J. Chem. Soc.,
Perkin Trans. 2, 1980, 946; (c) P. J. Barker, A. G. Davies, J. A.-A.
Hawari and M.-W. Tse, J. Chem. Soc., Perkin Trans. 2, 1980, 1488;
(d) T. Migita, K. Nagai and M. Kosugi, Bull. Chem. Soc. Jpn., 1983,
56, 2480; (e) K. Mizuno, M. Ikeda, S. Toda and Y. Otsuji, J. Am.
Chem. Soc., 1988, 110, 1288.
Photochemical crossover reaction between cinnamyl(methyl-
diphenyl)stannane 7 and [3-(p-tolyl)prop-2-enyl]triphenyl-
stannane 8
An equimolar mixture of compounds 7 (20.3 mg, 0.05 mmol)
and 8 (24 mg, 0.05 mmol) in a chloroform solution was photo-
lysed with >320 nm light under nitrogen for 1 h. The reaction
mixture was washed with saturated aq. NaHCO3, and the
organic phase was dried over Na2SO4. After evaporation of
the mixture, careful chromatographic separation of the residue
on TLC developing with hexane gave a mixture of compound
7 (E:Z = 33:67) and 1-phenylprop-2-enyl(methyl)diphenyl-
stannane 9 (ratio 7:9 = 6:94) and a mixture of compounds 8
(E:Z = 29:71) and [1-(p-tolyl)prop-2-enyl]triphenylstannane
10 (ratio 8:10 = 8:92) in total yields of 40 and 50%, respect-
9 A. G. Davies, in Chemistry of Tin, ed. P. G. Harrison, Blackie, New
York, 1989, p. 294.
10 B. Malliard, K. U. Ingold and J. C. Scaiano, J. Am. Chem. Soc.,
1983, 105, 5095.
11 (a) H. G. Kuivila, L. W. Menapace and C. R. Warner, J. Am. Chem.
Soc., 1962, 84, 3584; (b) H. G. Kuivila, Acc. Chem. Res., 1968, 1, 299;
(c) B. Mahiou and G. J. Gleicher, J. Org. Chem., 1990, 55, 4466; (d)
D. J. Carlsson and K. U. Ingold, J. Am. Chem. Soc., 1968, 90, 7047;
(e) K. W. Krosley, G. J. Gleicher and G. E. Clapp, J. Org. Chem.,
1992, 57, 840; ( f ) M. Kosugi, K. Kurino, K. Takayama and
T. Migita, J. Organomet. Chem., 1973, 56, C11; (g) J. Orignon and
M. Pereyre, J. Organomet. Chem., 1973, 61, C33.
12 (a) J. Cooper, A. Hudson and R. A. Jackson, J. Chem. Soc., Perkin
Trans. 2, 1973, 1056; (b) L. W. Menapace and H. G. Kuivila, J. Am.
Chem. Soc., 1964, 86, 3047.
13 Y. Tanigawa, I. Moritani and S. Yoshida, J. Organomet. Chem.,
1971, 28, 73.
14 Y. Naruta, H. Uno and K. Maruyama, Nippon Kagaku Kaishi, 1981,
835 (Chem. Abstr., 1981, 95, 114972k).
15 Y. Naruta, Y. Nishigaichi and K. Maruyama, Org. Synth., 1993, 71,
118.
1
ively. H NMR data for compound 9: δH 0.43 (3H, s, SnMe),
3.91 (1H, d, J 9.8, 1-H), 4.82–5.05 (2H, m, 3-H2), 6.38 (1H, dt, J
16.8 and 10, 2-H) and 7.03–7.53 (15H, m, ArH); for compound
10: δH 2.28 (3H, s, p-Me), 4.13 (1H, d, J 9.5, 1-H), 4.83–5.05
(2H, m, 3-H2), 6.35–6.49 (1 H, m, 2-H) and 7.01–7.68 (19H, m,
ArH).
16 E. A. Abel and R. J. Rowley, J. Organomet. Chem., 1975, 84, 199.
17 (a) M. Lequan, F. Meganem and Y. Besace, J. Organomet. Chem.,
1977, 131, 231; (b) F. B. Kipping, J. Chem. Soc., 1928, 2365.
18 J. Mattay, J. Gersdorf and K. Buchkremer, Chem. Ber., 1987, 120,
307.
Acknowledgements
This work was supported in part by a Grant-in-Aid for
Scientific Research (No. 09640636) from the Ministry of
Education, Science and Culture, Japan. We thank Ms M.
Egawa of our department technical office for performing the
microanalysis of some new compounds.
19 D. Blake, G. E. Coates and J. M. Tate, J. Chem. Soc., 1961, 618.
References
Paper 7/07609F
Received 21st October 1997
Accepted 29th January 1998
1 (a) K. W. Wilson, J. D. Roberts and W. G. Young, J. Am. Chem. Soc.,
1950, 72, 218; (b) D. A. Hutchison, K. R. Beck, R. A. Benkeser and
J. B. Grutzner, J. Am. Chem. Soc., 1973, 95, 7075.
1314
J. Chem. Soc., Perkin Trans. 1, 1998