Intramolecular Reaction of (γ-Alkoxyallyl)stannane with Aldehyde: Origin of the Stereoselectivities

Article abstract of DOI:10.1021/jo971309c

The intramolecular cyclization of simple acyclic (γ-alkoxyallyl)stannane aldehydes 1, 2, and 12-15 was investigated to elucidate the relationship between the geometry of the double bond of the allylstannanes, the ring size of cyclic ethers produced by the cyclization, and the procedures for promoting the cyclization. The Lewis acid-mediated cyclization of 1-2, 12-13, and 14-15 gave the trans cyclic ethers 3, 39, and 41, respectively, either predominantly or exclusively irrespective of the geometry of the double bond and of the ring size of the cyclic ethers. The relationship in the thermal cyclization of 1 and 2, which gave the 6-membered cyclic ethers 4 and 3, was straightforward; the Z isomer 1 gave the cis product 4, and the E isomer 2 afforded the trans product 3. However, the relationship in the thermal cyclization of 12 and 13 which afforded the 5-membered cyclic ethers 39 and 40 was different from that expected from the cyclization via the well-accepted cyclic transition state, as observed in the case of 1 and 2. Both the Z (12) and E (13) isomers gave the cis cyclic ether 40 either predominantly or exclusively. The protic acid-mediated (or -catalyzed) cyclization of 12-13 and 14-15 gave the trans cyclic ethers 39 and 41, respectively, regardless of the geometry of the double bonds. On the other hand, the protic acid-promoted cyclization of 1 and 2 was very strange; the Z isomer 1 gave the cis isomer 4, and the E isomer 2 afforded the trans isomer 3. The mechanisms for these cyclization reactions are proposed.

Full text of DOI:10.1021/jo971309c

J . Org. Chem. 1997, 62, 7439-7446
7439
In tr a m olecu la r Rea ction of (γ-Alk oxya llyl)sta n n a n e w ith
Ald eh yd e: Or igin of th e Ster eoselectivities
†
Isao Kadota, Miho Kawada, Vladimir Gevorgyan, and Yoshinori Yamamoto*
Research Center for Organic Resources and Materials Chemistry, Institute for Chemical Reaction Science,
and Department of Chemistry, Graduate School of Science, Tohoku University, Sendai 980-77, J apan
Received J uly 16, 1997X
The intramolecular cyclization of simple acyclic (γ-alkoxyallyl)stannane aldehydes 1, 2, and 12-
1
5 was investigated to elucidate the relationship between the geometry of the double bond of the
allylstannanes, the ring size of cyclic ethers produced by the cyclization, and the procedures for
promoting the cyclization. The Lewis acid-mediated cyclization of 1-2, 12-13, and 14-15 gave the
trans cyclic ethers 3, 39, and 41, respectively, either predominantly or exclusively irrespective of
the geometry of the double bond and of the ring size of the cyclic ethers. The relationship in the
thermal cyclization of 1 and 2, which gave the 6-membered cyclic ethers 4 and 3, was
straightforward; the Z isomer 1 gave the cis product 4, and the E isomer 2 afforded the trans product
3
. However, the relationship in the thermal cyclization of 12 and 13 which afforded the 5-membered
cyclic ethers 39 and 40 was different from that expected from the cyclization via the well-accepted
cyclic transition state, as observed in the case of 1 and 2. Both the Z (12) and E (13) isomers gave
the cis cyclic ether 40 either predominantly or exclusively. The protic acid-mediated (or -catalyzed)
cyclization of 12-13 and 14-15 gave the trans cyclic ethers 39 and 41, respectively, regardless of
the geometry of the double bonds. On the other hand, the protic acid-promoted cyclization of 1
and 2 was very strange; the Z isomer 1 gave the cis isomer 4, and the E isomer 2 afforded the
trans isomer 3. The mechanisms for these cyclization reactions are proposed.
In tr od u ction
predominantly the trans isomer 3 regardless of the
double-bond geometry (eq 1).6 This stereoselectivity was
In recent years there has been an explosion of interest
1
in biologically active natural products of marine origin.
Owing to their structural novelty and toxicity, polycyclic
ethers are particularly attractive targets for synthetic
chemists.2 In view of the structural feature, synthetic
approach requires an efficient method for the synthesis
of â-hydroxy cyclic ether as a constituent unit. We
developed a new strategy for the construction of cyclic
ether via the intramolecular condensation of allylstan-
3
nane with aldehyde. As demonstrated by the total
4
synthesis of hemibrevetoxin B and related polycyclic
ethers,5 this route is efficient for the stereocontrolled
construction of 7-membered cyclic ethers.
In addition, the formation of 6-membered cyclic ethers
was studied by the intramolecular condensation method.
The Lewis acid-mediated cyclization of 1 and 2 gave
quite reasonable from the well-accepted theory for the
Lewis acid-mediated allylic stannane-aldehyde conden-
sation.7 On the other hand, the protic acid-promoted
reaction of 1 and 2 afforded 4 and 3, respectively. Similar
stereoselectivities were observed in the thermal cycliza-
tion. The stereoselectivities via the thermal cyclization
were able to be explained by the well-accepted cyclic
†
Institute for Chemical Reaction Science.
Abstract published in Advance ACS Abstracts, October 1, 1997.
X
7a,8
transition state, but those via the protic acid-promoted
(
1) For recent reviews, see: (a) Shimizu, Y. Chem. Rev. 1993, 93,
cyclization were very difficult to explain on the basis of
the ordinary mechanisms. Accordingly, we proposed the
push-pull mechanism.
During this study, a related cyclization was reported
by Keck and co-workers.9 The reaction of the Z-allylic
stannane 5 in the presence of Lewis or protic acids
afforded the cis product 8 predominantly. On the other
1
1
685-1698. (b) Yasumoto, T.; Murata, M. Chem. Rev. 1993, 93, 1897-
909.
(2) For recent review, see: Alvarez, E.; Candenas, M.-L.; P e´ rez, R.;
Ravelo, J . L.; Mart ´ı n, J . D. Chem. Rev. 1995, 95, 1953-1980 and
references cited therein.
(
3) (a) Yamamoto, Y.; Yamada, J .; Kadota, I. Tetrahedron Lett. 1991,
3
1
2, 7069-7072. (b) Yamamoto, Y.; Kadota, I. Main Group Met. Chem.
994, 17, 269-289.
(4) (a) Kadota, I.; Park, J .-Y.; Koumura, N.; Pollaud, G.; Matsukawa,
Y.; Yamamoto, Y. Tetrahedron Lett. 1995, 36, 5777-5780. (b) Kadota,
I.; Yamamoto, Y. Main Group Met. Chem. 1996, 19, 361-366.
(6) Gevorgyan, V.; Kadota, I.; Yamamoto, Y. Tetrahedron Lett. 1993,
34, 1313-1316.
(5) (a) Suzuki, T.; Sato, O.; Hirama, M.; Yamamoto, Y.; Murata, M.;
Yasumoto, T.; Harada, N. Tetrahedron Lett. 1991, 32, 4505-4508. (b)
Ravelo, J . L.; Regueiro, A.; Mart ´ı n, J . D. Tetrahedron Lett. 1992, 33,
(7) For recent reviews, see: (a) Yamamoto, Y.; Asao, N. Chem. Rev.
1993, 93, 2207-2293. (b) Nishigaichi, Y.; Takuwa, A.; Naruta, Y.;
Maruyama, K. Tetrahedron 1993, 49, 7395-7426.
3
389-3392. (c) Alvarez, E.; D ´ı az, M. T.; P e´ rez, R.; Ravelo, J . L.;
Regueiro, A.; Vera, J . A.; Zurita, D.; Mart ´ı n, J . D. J . Org. Chem. 1994,
(8) For examples, see: (a) Marshall, J . A.; Dehoff, B. S.; Crooks, S.
L. Tetrahedron Lett. 1987, 527-530. (b) Hull, C.; Mortlock, S. V.;
Thomas, E. J . Tetrahedron Lett. 1987, 5343-5346.
(9) (a) Keck, G. E.; Dougherty, S. M.; Savin, K. A. J . Am. Chem.
Soc. 1995, 117, 6210-6223. For a related work on intermolecular
reactions, see: (b) Keck, G. E.; Savin, K. A.; Cressman, E. N. K.; Abbott,
D. E. J . Org. Chem. 1994, 59, 7889-7896.
5
9, 2848-2870. (d) Kadota, I.; Matsukawa, Y.; Yamamoto, Y. J . Chem.
Soc., Chem. Commun. 1993, 1638-1641. (e) Yamamoto, Y.; Kadota, I.
Bull. Soc. Chim. Belg. 1994, 103, 619-629. (f) Alvarez, E.; Candenas,
M.-L.; P e´ rez, R.; Ravelo, J . L.; Mart ´ı n, J . D. Chem. Rev. 1995, 95, 1953-
1
1
980. (g) Oguri, H.; Hishiyama, S.; Oishi, T.; Hirama, M. Synlett 1995,
252-1254.
S0022-3263(97)01309-1 CCC: $14.00 © 1997 American Chemical Society

7
440 J . Org. Chem., Vol. 62, No. 21, 1997
Kadota et al.
Ta ble 1. Su m m a r y of th e Ster eoselectivities in th e
Cycliza tion of 1, 2, 5, a n d 6
product
substrate
Lewis acid
protic acid
∆
1
2
(Z)
(E)
trans
trans
cis
trans
cis
trans
5
6
(Z)
(E)
cis
trans
cis
trans
cis
cis
F igu r e 1. Proposed secondary orbital overlap in the syn-
9
synclinal transition state.
Sch em e 1. P r ep a r a tion of
Z)-(γ-Alk oxya llyl)sta n n a n e 1
a
(
hand, the E substrate 6 afforded the trans isomer 7 as a
major product (eq 2). On the basis of these results, the
a
(a) Allyl bromide, NaH, THF; (b) s-BuLi, TMEDA, THF; (c)
n-Bu3SnCl; (d) SO3‚py, DMSP, Et3N, CH2Cl2.
acyclic substrates 1, 2, and 12-15 which would produce
authors concluded that the stabilization by a secondary
orbital overlap between the oxygen of the carbonyl and
the stannyl methylene carbon is the most important
factor for the determination of stereochemical outcome
5
-, 6-, and 7-membered cyclic ethers.
(Figure 1). The stereoselectivities observed in the cy-
clizations of 1, 2, 5, and 6 are summarized in Table 1.
Remarkable differences between our results and Keck’s
observations can be found in the Lewis acid-promoted
cyclization of Z substrates (1 and 5) and the thermal
cyclization of E substrates (2 and 6). More recently,
Denmark et al. reported that the protic acid-promoted
reaction of a conformationally rigid cyclic system (9)
provided a bridged compound (11) with high stereose-
lectivity (eq 3).10 It was concluded that the reaction
Resu lts
P r ep a r a tion of th e Su bstr a tes 1, 2, a n d 12-15.
The (γ-alkoxyallyl)stannanes 1, 2, and 12-15, having an
aldehyde at the terminus of the carbon chain, were
prepared as shown in Schemes 1 and 2. The Z isomer 1
3
was easily synthesized by the usual method (Scheme 1).
Selective monoallylation of 1,4-butanediol (16) with allyl
bromide/NaH gave 17. Treatment of 17 with 2.2 equiv
of s-BuLi/TMEDA followed by trapping the resulting
3
allylic anion 18 with n-Bu SnCl afforded 19 in 68%
yield.11 Oxidation of the alcohol 19 produced 1 in 66%
yield. In contrast to preparation of (Z)-(γ-alkoxyallyl)-
stannanes, there was great difficulty in the synthesis of
the E isomers. The lithium anion-tin trapping method
produced exclusively Z-allylic stannanes, as shown in
Scheme 1, because a chelate (Z)-lithium derivative such
as 18 is more stable than the corresponding E isomer.
1
2
After several unfruitful attempts, we found that the
proceeded via an acyclic transition state (10) and there
was no interaction between the tin atom and oxygen
atom. Although the mechanisms proposed by Keck and
Denmark are reasonable, at least in their own systems,
and explain well their stereochemical results (eqs 2 and
0
Pd -catalyzed hydrostannylation of allenyl ether provided
an allowable route to the desired E isomers (Scheme 2).
The propargyl ether alcohol 21, prepared from 20 and
propargyl bromide/NaH, was converted into the TBDMS
1
3
3
), it was difficult for us to explain the origin of the
(
11) (a) Keck, G. E.; Abbott, D. E.; Wiley, M. R. Tetrahedron Lett.
stereoselectivities in our cyclic ether synthesis based
upon their mechanisms (Table 1). To elucidate the origin
of the observed stereoselectivities in our previous studies,
we examined the intramolecular cyclization of the simple
1
2
987, 28, 139-142. (b) Koreeda, M.; Tanaka, Y. Tetrahedron Lett. 1987,
8, 143-146.
(12) For example, hydrostannation of the corresponding propargyl
ether gave an inseparable mixture of regioisomers.
13) (a) Koerber, K.; Gore, K.; Vatele, J .-M. Tetrahedron Lett. 1991,
2, 1187-1190. (b) Gevorgyan, V.; Liu, J .-X.; Yamamoto, Y. J . Org.
Chem. 1997, 62, 2963-2967.
(
3
(10) Denmark, S. E.; Hosoi, S. J . Org. Chem. 1994, 59, 5133-5135.

Reaction of (γ-Alkoxyallyl)stannane with Aldehyde
Sch em e 2. P r ep a r a tion of 1, 2, a n d 12-15a
J . Org. Chem., Vol. 62, No. 21, 1997 7441
Ta ble 2. Cycliza tion s of 12 a n d 13a
time temp
ratio
yield
b
b
entry substrate reagent equiv (min) (°C) (39:40) (%)
1
2
3
4
5
6
12
13
BF3‚OEt2
c
CF3CO2H
BF3‚OEt2
c
1.2
30 -78 90:10
>95
d
120
100
2:98< >95
2.0
1.2
10 -78 71:29
>95
79
>95
87
30 -78 93:7
120
100 30:70
CF3CO2H
2.0
10 -78 83:17
a
All reactions were carried out with 0.1 M substrate in CH2Cl2
under the conditions indicated in the table and quenched with
Et3N at the reaction temperature. b Ratios and yields were
determined by capillary GLC analyses. Propiophenone was used
as an internal standard. c Benzene was used as a solvent. Trans
isomer 39 was not detected.
d
Ta ble 3. Cycliza tion s of 1 a n d 2a
a
(a) Propargyl bromide, NaH, THF; (b) TBDMSCl, imidazole,
DMF; (c) t-BuOK; (d) (i) TBAF, THF, (ii) HMDS, TMSCl; (e) (i)
n-Bu3SnH, Pd(PPh3)4; (ii) K2CO3, MeOH; (f) SO3‚py, DMSO, Et3N,
CH2Cl2.
time temp
ratio
yield
b
b
entry substrate
reagent
equiv (min) (°C)
(3:4)
(%)
ether 22. Isomerization of the propargyl ether 22 was
performed by heating it in the presence of a catalytic
amount of t-BuOK, affording the corresponding allenyl
ether 23.14 Replacement of the TBDMS group of 23 with
a TMS group by the traditional method gave 24. Treat-
1
2
3
4
5
6
7
8
9
0
1
2
3
1
TiCl
4
2.0
2.0
30 -70
30 -70
60 120
83:17
68:32
2:98<
13:87
11:89
7:93
20:80
10:90
11:89
25:75
29:71
18:82
84:16
87:13
98:2
86
80
78
BF
3
‚OEt
2
c
CF SO H
3
3
3
3
3
3
3
3
3
2.0
d
d
5
-70
>95
>95
94
>95
>95
>95
>95
66
CF
CF
CF
CF
SO
SO
SO
H
H
H
15 -70
15 -90
30 -70
ment of 24 with n-Bu SnH in the presence of a catalytic
3 4
amount of Pd(PPh ) at 60 °C followed by desilylation
3
0.1
2.0
2.0
2.0
2.0
2.0
2.0
2.0
CO
2
H
5
5
-70
-70
0
25
-70
gave a 5.5:1 mixture of 25 and 26 in 58% yield. The yield
of the E isomer 26 was decreased at lower temperatures.
Oxidation of each isomer afforded the aldehydes 12 and
3 2
CCl CO H
1
1
1
1
CH
CH
2
ClCO
2
H
600
3600
5
3
CO
2
H
HCl
TiCl
90
1
3, respectively. Other starting materials (2, 14, and 15)
2
4
30 -70
30 -70
60 120
>95
>95
>95
>95
>95
86
1
5
were prepared by the same procedure as above.
14
BF
c
3
‚OEt
2
1
1
1
1
5
6
7
8
Cycliza tion of 12 a n d 13. The results of the cycliza-
tion reactions of 12 and 13 are summarized in Table 2.
CF
CF
3
3
SO
CO
3
H
2.0
2.0
2.0
5
5
5
-70 >98:2
-70
-70
2
H
97:3
97:3
The BF
3
2
‚OEt -mediated reaction gave the cyclized prod-
HCl
ucts 39 and 40 in good to very high yields, and high trans
selectivity was obtained regardless of the geometry of the
double bond of the starting materials (entries 1 and 4).
Good trans selectivity was observed in the reaction in the
a
All reactions were carried out with 0.1 M substrate in CH2Cl2
under the conditions indicated in the table and quenched with
aqueous satd NaHCO3 at the reaction temperature. b Ratios and
yields were determined by capillary GLC analyses. Tetradecane
was used as an internal standard. Toluene was used as a solvent.
The substrate was added very slowly to the reagent, and thus
at the initial stage of the reaction [CF3SO3H]:[1] >> 2:1. The final
ratio was 2:1.
c
presence of CF
3
CO
2
H (entries 3 and 6). The thermal
d
cyclization of 12 proceeded with very high cis stereose-
lectivity, and the cis isomer 40 was obtained as the sole
product (entry 2). The cis selectivity was decreased when
stereoselectivity of the thermal reaction strongly de-
pended upon the substrate geometry, i.e., the reaction
of the Z isomer 1 gave cis product 4 exclusively, whereas
that of the E isomer 2 afforded 3 exclusively (entries 3
and 15). In each case, only one stereoisomer was
detected. Notable stereoselectivities were observed in the
protic acid-promoted cyclization. The reaction of 1 in the
1
3 was used as a substrate (entry 5). In all cases, the
cyclization products were obtained in good to nearly
quantitative yields.
Cycliza tion of 1 a n d 2. The results on the reaction
of 1 and 2 are summarized in Table 3. Both substrates
cyclized in the presence of Lewis acids to give the trans
isomer 3 predominantly in good to very high yields
presence of CF
3 3 3 2
SO H, CF CO H, and other protic acids
(entries 1, 2, 13, and 14). On the other hand, the
proceeded smoothly to give 3 with good to high diaste-
reoselectivities, the stereoselectivity being similar to that
of the reaction via the Lewis acids (entries 4-12).
However, the treatment of 2 with the protic acids afforded
4 either predominantly or exclusively (entries 16-18).
The mode of the stereoselectivity in the protic acid-
mediated reaction agrees with that in the thermal
cyclization and is not in accord with that in the Lewis
acid-mediated reaction. Interestingly, the use of a cata-
(
14) Brandsma, L.; Verkruijsse, H. D. Synthesis of Acetylenes, Allenes
and Cumulenes, a Laboratory Manual; Elsevier: New York, 1981; pp
2-93.
15) Although most of the early stage starting materials prepared
9
(
1
1,12
according with known procedures
H-NMR spectroscopic analyses only, the Z- and E-allylic stannanes
2, 13, 21, 27, 28, 39, and 40 were fully characterized by IR, H-NMR,
and elemental analyses or high-resolution mass spectra. Because of
their low stability, the aldehydes 14-17 could not be fully character-
ized. See the Experimental Section.
were characterized by IR and
1
1
1

7
442 J . Org. Chem., Vol. 62, No. 21, 1997
Kadota et al.
Sch em e 3. Acyclic Tr a n sition Sta te Mod el
Ta ble 4. Cycliza tion s of 14 a n d 15a
time
equiv (min)
temp yield
(°C)
entry substrate
reagent
(%)b
1
2
3
4
5
6
7
8
9
0
1
2
3
14
BF3‚OEt2
TiCl4
c
CF3SO3H
CF3CO2H
CCl3CO2H
HCl
BF3‚OEt2
c
CF3SO3H
CF3CO2H
CCl3CO2H
HCl
1.2
2.0
40
40
1200
40
-78
-78
150
-78
-78
25
-78
-78
150
-78
-78
25
45
28
trace
trace
27
2.0
2.0
2.0
2.0
1.2
30
d
900
210
30
1200
40
30
900
210
0
12
79
mined by the coupling constant of J ab ) 2.5 Hz, and the
cis stereochemistry of 43 was confirmed by that of J ab
15
)
trace
trace
3
16,17
4
.0 Hz.
The stereochemistries of the other compounds
1
1
1
1
2.0
2.0
2.0
2.0
(
3, 4, and 41) were determined by the comparison of their
H-NMR spectra with those of authentic materials.
1
18
d
0
-78
trace
Tr a n sit ion St a t e St r u ct u r es a n d Or igin of t h e
a
Ster eoselectivities. 1. Lew is Acid -P r om oted Cy-
All reactions were carried out with 0.1 M substrate in CH2Cl2
under the conditions indicated in the table and quenched with
Et3N at the reaction temperature. Yields were determined by
cliza tion . It is widely accepted that the Lewis acid-
b
mediated reaction of allylic stannanes with aldehydes
capillary GLC analyses. Propiophenone was used as an internal
proceeds via an acyclic transition state.7 The Lewis acid-
c
standard. No cis isomer was detected. Benzene was used as a
promoted cyclization of 1, 2, and 12-15 gave the trans
isomers either predominantly or exclusively, regardless
of the geometry of the double bond of the allylic stan-
nanes. The trans preference can be explained by the
d
solvent. Total decomposition of the starting material was ob-
served.
7
well-accepted acyclic transition state model (Scheme 3).
To avoid the 1,3-diaxial repulsion, the allylic stannane
moiety and the carbonyl oxygen of the substrates are
oriented to a pseudoequatorial position as depicted by the
synclinal transition state 44, which leads to the trans
cyclic ethers 3, 39, and 41 regardless of the double-bond
geometry and of the size of the rings produced.
F igu r e 2. Comparison of the coupling constants of 42 and
4
3.
2
. Th er m a l Rea ction . The substrates 14 and 15
gave only trace amounts of cyclized products under the
thermal conditions, and therefore the stereochemical
discussion on the thermal reaction of 14 and 15 can not
be made at the present time. The thermal reaction of 1,
lytic amount of CF
the substrate 1, giving a 20:80 mixture of 3 and 4 in an
essentially quantitative yield (entry 7).
Cycliza tion of 14 a n d 15. The results on the cycliza-
tion of 14 and 15 are summarized in Table 4. Very
interestingly, in all cases the trans isomer 41 was
obtained as the sole product regardless of the geometry
of the substrates and the type of the reagents used. The
3
3
SO H led to complete conversion of
2
, 12, and 13 proceeded smoothly to afford the desired
cyclic ethers. As expected from the cyclic transition state
models proposed for the thermal condensation, the ster-
eochemical outcome of the products was strongly depend-
7
a,8
ent upon the double-bond geometry of the substrates.
use of BF
3
‚OEt
2
gave 41 in reasonable to good yields
or protic acids gave
1 in an allowable to very low yields. In particular, the
As shown in Scheme 4, the (Z)-allylstannane 12 was
converted to the cis isomer 40 with very high stereose-
lectivity via a cis-fused cyclic transition state (45). The
(entries 1 and 8). The use of TiCl
4
4
thermal reactions of 14 and 15 did not afford the cyclized
product at all (entries 5 and 9). It was rather surprising
for us to observe the inefficient thermal cyclization of the
acyclic systems 14 and 15, since the cyclization to a
(E)-allylstannane 13 gave also the cis isomer 40 predomi-
nantly, but the cis stereoselectivity in the cyclization of
1
1
3 was lower than that for 12. Perhaps the reaction of
3 proceeds mainly through a cis-fused cyclic transition
7
-membered ring in the cyclic systems proceeded very
state (46), but a trans-fused cyclic transition state (47)
intervenes to a cetain extent, giving the minor trans
isomer 39 along with the major product 40. In the case
of the Z-allylic stannane 12, a cis-fused one (45) is the
only possible transition state, since the interaction
well in respect to the chemical yield and stereo-
selectivity.3
-5
Perhaps, in contrast to 6-membered ring
formation, the cyclization to a 7-membered ring requires
a certain constrained system in order to bring the
γ-carbon of allylic stannane close to the aldehyde.
Str u ctu r a l Assign m en ts of th e P r od u cts. Because
of the volatility of 39 and 40, the stereochemistries of
(
17) Gaudemer, A. In Stereochemistry; Kagan, H. B., Ed.; Georg
Thieme Publishers: Stuttgart, 1977; Vol. 1, p 90.
18) To comfirm the stereochemical assignment, the cis isomer 40
1
these products were confirmed by H-NMR analyses of
(
the benzoate derivatives 42 and 43, prepared from 39
and 40 by the usual benzoylation method (Figure 2). The
trans relationship between H and H of 42 was deter-
a b
was converted to a bicyclic lactone by four steps in good yield. In
contrast, the trans isomer 39 did not give any cyclized product under
the same conditions because of the steric problem of the trans-fused
5
-5 ring system.
(16) (a) Nicolaou, K. C.; Prasad, C. V. C.; Somers, P. K.; Hwang,
C.-K. J . Am. Chem. Soc. 1989, 111, 5330-5334. (b) Nicolaou, K. C.;
Prasad, C. V. C.; Somers, P. K.; Hwang, C.-K. J . Am. Chem. Soc. 1989,
1
11, 5335-5340.

Reaction of (γ-Alkoxyallyl)stannane with Aldehyde
J . Org. Chem., Vol. 62, No. 21, 1997 7443
Sch em e 4. Cyclic Tr a n sition Sta te Mod el for th e
Th er m a l Cycliza tion s of 12 a n d 13
Sch em e 6. Acyclic Tr a n sition Sta te Mod el
Sch em e 7. Cycliza tion w ith Ca ta lytic Am ou n ts of
P r otic Acid s (HX)
Sch em e 5. Cyclic Tr a n sition Sta te Mod el for th e
Th er m a l Cycliza tion s of 1 a n d 2
The remarkable stereoselectivities observed for the
cyclization of 1 and 2 may be explained by the size of
9
protic acids; similar discussions have been made by Keck
and Denmark.10 Proton as the smallest Lewis acid would
coordinate to an aldehyde oxygen in an anti manner, and
the reaction of 1 would proceed through a syn-synclinal
between Sn and the carbonyl oxygen atom is required.
However, in the case of the E isomer 13, both cis- and
trans-fused models 46 and 47 are conceivable for the
cyclic transition states. The preference of 46 over 47 may
be due to the stability of the cis-fused 5-6 ring system in
the transition state. As mentioned below, a trans-fused
model (49) is preferable (as was expected) for the case of
the 6-membered ring system.
transition state (51) in which Bu
3
Sn is replaced in an
anti-S ′ manner and the secondary orbital overlap be-
E
tween the LUMO of the cabonyl and the HOMO of the
allylstannane is possible. Therefore, the protic acid-
promoted cyclization of 1 gives the cis isomer 4. Quite
similarly, the reaction of 2 would proceed via a syn-
synclinal transition state (52), which produces the trans
isomer 3. Perhaps, protic acid (HX) could be regenerated
The high stereoselectivities observed in the thermal
reaction of 1 and 2 can be explained clearly by the cyclic
transition state model (Scheme 5). The reaction of 1 gave
exclusively the cis cyclic ether 4 via a cis-fused model
(48). The transition state model 48 is very similar to the
model 45; only a cis-fused structure is allowable for each
case. The E-allylic stannane 2 gave the trans isomer 3
with very high stereoselectivity via the trans-fused 6-6
system 49.
by the exchange reaction between +SnBu
Scheme 7, 54), and thus, even the use of a catalytic
and O-H
3
3
. P r otic Acid -P r om oted Cycliza tion . The protic
(
acid-mediated reaction of 12 and 13 gave the trans cyclic
ether 39 predominantly, regardless of the double-bond
geometry of the starting allylic stannanes. Although the
yields were very low, only the trans cyclic ether 41 was
obtained in the reactions of 14 and 15. These results can
be explained by the usual acyclic transition state model
as depicted in Scheme 6; instead of Lewis acid, the proton
would coordinate to the carbonyl oxygen of 12-15, and
the cyclization would proceed through an acyclic transi-
tion state (50). On the other hand, it was interesting to
find that the stereoselectivities of the reactions of 1 and
amount of HX could promote and complete the cyclization
reaction (Table 3, entry 7). The syn-synclinal type
transition state models such as 51 and 52 would be
allowable only in the case of very small Lewis acids. In
contrast, sterically demanding Lewis acids such as
BF
3
‚OEt
2 4
and TiCl would prevent bringing the carbonyl
19
oxygen close to the CH
2
carbon of the allylstannane (53).
How can we explain by the syn-synclinal model the
stereoselectivities observed in the protic acid-promoted
processes to 5-membered (12 and 13) and 7-membered
(
14 and 15) formation? Compared with the 6-membered
ring systems 51 and 52, the distance between the
carbonyl oxygen and the CH -Sn in the syn-synclinal
2
were strongly dependent upon the double-bond geom-
etry of the starting materal, as was observed in the
thermal cyclization. Thus, the reaction of 1 or 2 gave 4
or 3, respectively, with very high to good stereoselectivi-
ties.
2
(
19) Even in the case of BF
3 2 4
‚OEt and TiCl , the reaction via 51
and 52 may be involved to a certain extent.

7
444 J . Org. Chem., Vol. 62, No. 21, 1997
Kadota et al.
Ta ble 5. Su m m a r y of th e Ster eoselectivities in th e
230 mesh ASTM), and for flash chromatography, Merck silica
gel 60 (230-400 mesh ASTM) was employed.
Cycliza tion of 1, 2, a n d 12-15
5
-membered
ring
6-membered
ring
7-membered
ring
All solvents were dried immediately before use. Ether and
THF were distilled from sodium/benzophenone ketyl; dichlo-
romethane, hexane, benzene, triethylamine, pyridine, DMF,
product
substrate
Lewis acid trans
12 (Z) 13 (E) 1 (Z)
2 (E)
trans
trans
trans
14 (Z) 15 (E)
DMSO, and TMEDA were distilled from CaH
distilled from Mg(OMe) . All reactions involving air- and/or
moisture-sensitive materials were carried out in an argon
atmosphere. On workup, extracts were dried over MgSO
-(Allyloxy)-1-bu t a n ol (17). To a stirred suspension of
2
; methanol was
trans
cis
trans
trans
cis
cis
trans
trans
2
∆
cis
trans
protic acid
trans
trans
4
.
4
model (56) of 12 becomes longer, whereas that in the syn-
synclinal model (57) of 14 becomes shorter. Similar
argument can be made for 13 and 15. These changes in
the transition states might prevent them from taking the
syn-synclinal geometries, and thus the usual acyclic
transition states such as 50 would become favorable.
Therefore, the syn-synclinal models are applicable for
very limited stereostructures among the allylstannane-
aldehyde substrates (see also footnote 21a of ref 9).
NaH (12 g of a 60% suspension in mineral oil, 0.3 mol,
prewashed with hexane) in THF (150 mL) at 0 °C was added
dropwise a solution of 1,4-butanediol (16) (110 mL, 1.25 mol)
in THF (80 mL). After 1 h, allyl bromide (21.7 mL, 0.25 mol)
was added dropwise, and the mixture was refluxed for 2 h.
The reaction was quenched with water (200 mL). Most of the
organic solvent was evaporated, and the aqueous residue was
extracted with ether (500 mL). The organic layer was washed
with water (200 mL × 2) and concentrated. The crude product
was purified by distillation to give 17 (25.2 g, 80% based on
allyl bromide): colorless oil; bp 67-70 °C/2.0 mmHg; R
f
) 0.39
(
hexane/AcOEt, 1:1); IR (neat) 3600-3100, 3090, 2940, 2860,
-
1
1
1
645, 1420, 1345, 1100, 1060, 1000, 930 cm ; H-NMR (270
) δ 5.92 (dddd, J ) 17.0, 10.5, 5.5, 5.5 Hz, 1H),
.28 (dddd, J ) 17.0, 1.5, 1.5, 1.5 Hz, 1H), 5.19 (dddd, J )
MHz, CDCl
3
5
1
3
1
0.5, 1.5, 1.5, 1.5 Hz, 1H), 3.99 (ddd, J ) 5.5, 1.5, 1.5 Hz, 2H),
.65 (t, J ) 6.0 Hz, 2H), 3.49 (t, J ) 6.0 Hz, 2H), 2.10 (brs,
H), 1.70-1.50 (m, 4H).
4
-[(Z)-[3-(T r ib u t y ls t a n n y l)-1-p r o p e n y l]o x y ]-1-b u -
ta n ol (19). To a solution of 17 (2.60 g, 20 mmol) and TMEDA
(
6.6 mL, 44 mmol) in THF (100 mL) at -78 °C was added
s-BuLi (44 mL of a 1.1 M solution in cyclohexane, 44 mmol),
and the resulting mixture was stirred for 1 h. To this yellow
solution was added n-Bu SnCl (6.5 mL, 24 mmol), and the
3
mixture was allowed to warm to rt. The reaction was
quenched with water (30 mL), then most of the organic
solvents were evaporated, and the aqueous residue was
extracted with ether (250 mL). The organic layer was washed
with brine (100 mL). Concentration and column chromatog-
Con clu sion
The relationship between the substrate geometries and
the stereochemistries of the cyclic ethers is summarized
in Table 5. The noticeable stereoselectivities, which were
strongly dependent upon the double-bond geometry, were
observed in the protic acid-mediated cyclization of 1 and
. In contrast, Keck et al. observed that even in the
BF ‚OEt -mediated reaction of the acyclic systems 5 and
the diastereoselectivity of the cyclization could depend
upon the geometry of the double bond (Table 1). Perhaps,
in the case of allylstannanes 1 and 2, the oxygen atom
at the γ-position may decrease the secondary orbital
overlap interaction. Therefore, the reaction mediated by
a sterically demanding Lewis acid such as BF
TiCl would proceed via the acyclic transition state 44.
The syn-synclinal transition states 51 and 52 would be
allowable only in the case of proton as the smallest Lewis
acid.
raphy gave 19 (5.72 g, 68%): colorless oil; R
IR (neat) 3600-3100, 3030, 2960, 2920, 2870, 1650, 1461,
1417, 1360, 1250, 1180, 1153, 1100, 1068, 958, 870, 768, 730,
f
2 2
) 0.22 (CH Cl );
-
1
1
6
6
3
3
80, 655 cm ; H-NMR (270 MHz, CDCl ) δ 5.78 (ddd, J )
.0, 1.2, 1.2 Hz, 1H), 4.52 (ddd, J ) 8.9, 8.9, 6.0 Hz, 1H), 3.76-
.65 (m, 4H), 1.80-1.20 (m, 16H), 0.96-0.80 (m, 15H); HRMS
2
3
2
m/ z calcd for C19
-[(Z)-[3-(Tr ibu tylsta n n yl)-1-p r op en yl]oxy]bu ta n a l (1).
To a stirred mixture of 19 (5.72 g, 16.6 mmol), Et N (13 mL,
5 mmol), and DMSO (15 mL) in CH Cl (30 mL) at 0 °C was
added SO ‚py (8.66 g, 54.4 mmol). After 2 h, the reaction
40 2
H O Sn 420.2048, found 420.2057.
6
4
3
9
2
2
3
mixture was diluted with ether (200 mL) and washed with
water (100 mL) and brine (100 mL). Concentration and
3
‚OEt
2
or
column chromatography gave 1 (4.59 g, 66%): colorless oil; R
0.67 (hexane/AcOEt, 3:1); IR (neat) 3040, 2960, 2930, 1730,
f
4
)
-
1 1
1
650, 1460, 1100, 960, 870 cm ; H-NMR (270 MHz, CDCl
3
)
δ 9.81 (t, J ) 1.4 Hz, 1H), 5.75 (ddd, J ) 6.0, 1.3, 1.3 Hz, 1H),
4
2
1
.53 (ddd, J ) 8.9, 6.0, 6.0 Hz, 1H), 3.72 (t, J ) 6.0 Hz, 2H),
.58 (ddd, J ) 7.1, 7.1, 1.4 Hz, 2H), 2.00-1.90 (m, 2H), 1.50-
.20 (m, 12H), 1.00-0.80 (m, 15H).
Exp er im en ta l Section
Gen er a l P r oced u r es. 1H- and 119Sn-NMR spectra were
3-(2-P r op yn yloxy)-1-p r op a n ol (21). To a stirred suspen-
sion of NaH (8.3 g of a 60% suspension in mineral oil, 0.21
mol, prewashed with hexane) in THF (200 mL) at 0 °C was
added dropwise a solution of 1,3-propanediol (20) (57 mL, 0.79
mol) in THF (40 mL). After 1 h, propargyl bromide (18 mL,
0.2 mol) was added dropwise, and the mixture was stirred at
rt for 14 h. The reaction was quenched with water (150 mL).
Most of the organic solvent was evaporated, and the residue
was extracted with ether (400 mL). The organic layer was
washed with water (150 mL × 2) and concentrated. The crude
product was purified by distillation to give 21 (9.1 g, 40% based
on propargyl bromide): colorless oil; bp 53-54 °C/1.5 mmHg;
recorded on J EOL GSX-270 spectrometers. Chemical shifts
are reported in delta (δ) units, in part per million (ppm)
downfield from tetramethylsilane or tetramethylstannane or
in ppm relative to the singlet at 7.26 ppm for chloroform.
Coupling constants are reported in hertz (Hz). IR spectra
-
1
(
1
cm ) were measured on neat compounds with a Hitach 260-
0 or Shimadzu FTIR8200A infrared spectrophotometer.
High-resolution mass spectra were obtained with a J EOL J MS-
HX110 spectrometer. Capillary GC analyses were performed
on a Shimadzu GC-14A instrument equiped with a flame
ionization detector and CPB1-M25-025 column. All reactions
were monitored by thin layer chromatography using Merck
precoated aluminum plates (Kieselgel 60 F254, 0.2 mm).
Column chromatography was done on Merck silica gel 60 (70-
R
f
) 0.28 (hexane/AcOEt, 1:1); IR (neat) 3650-3050, 3280,
2940, 2870, 2100, 1474, 1439, 1356, 1268, 1092, 1049, 1019,
-
1 1
962, 922, 900 cm ; H-NMR (270 MHz, CDCl ) δ 4.17 (d, J )
3

Reaction of (γ-Alkoxyallyl)stannane with Aldehyde
J . Org. Chem., Vol. 62, No. 21, 1997 7445
2
2
2
.4 Hz, 2H), 3.78 (t, J ) 5.8 Hz, 2H), 3.71 (t, J ) 5.8 Hz, 2H),
3020, 2940, 2910, 2860, 2835, 2715, 1720, 1642, 1456, 1408,
1369, 1355, 1248, 1146, 990, 952, 866, 760, 720, 676, 650 cm ;
-1
.45 (t, J ) 2.4 Hz, 1H), 1.99 (br, 1H), 1.87 (quint, J ) 5.8 Hz,
1
H).
1
3
H-NMR (270 MHz, CDCl ) δ 9.82 (t, J ) 1.8 Hz, 1H), 5.78
-[(ter t-Bu tyld im eth ylsilyl)oxy]-3-(2-p r op yn yloxy)p r o-
p a n e (22). To a solution of 21 (8.5 g, 74.5 mmol) and imidazole
7.7 g, 113 mmol) in DMF (450 mL) at 0 °C was added
(ddd, J ) 6.0, 1.8, 1.8 Hz, 1H), 4.58 (ddd, J ) 9.1, 9.1, 6.0 Hz,
1H), 4.03 (t, J ) 6.3 Hz, 2H), 2.72 (ddd, J ) 6.3, 6.3, 1.8 Hz,
2H), 1.68-1.20 (m, 16H), 1.00-0.80 (m, 15H);
(CDCl ) δ -15.70; HRMS m/ z calcd for C18
found 404.1748.
3-[(E)-[3-(Tr ib u t ylst a n n yl)-1-p r op en yl]oxy]p r op a n a l
(13): colorless oil; R ) 0.67 (hexane/AcOEt, 3:1); IR (neat)
3050, 3020, 2950, 2920, 2850, 2725, 1728, 1658, 1642, 1461,
1417, 1377, 1338, 1293, 1182, 1123, 1070, 1020, 1000, 960, 920,
3
874, 802, 772, 714, 685 cm ; H-NMR (270 MHz, CDCl ) δ
1
19
(
Sn-NMR
TBDMSCl (13.1 g, 90.4 mmol), and the mixture was stirred
at rt for 1 h. The mixture was diluted with ether (700 mL)
and washed with water (300 mL), saturated aqueous NaHCO
3
36 2
H O Sn 404.1737,
3
(200 mL), and brine (200 mL). Concentration and column
f
chromatography gave 22 (16.3 g, 96%): colorless oil; R
f
) 0.34
(hexane/AcOEt, 20:1); IR (neat) 3320, 2960, 2940, 2870, 2120,
-
1
1
1
8
4
6
2
475, 1465, 1443, 1394, 1365, 1258, 1192, 1100, 1009, 942, 911,
82, 845, 786, 720, 665 cm-1; H-NMR (270 MHz, CDCl
1
) δ
9.80 (t, J ) 1.6 Hz, 1H), 6.10 (ddd, J ) 12, 1.5, 1.5 Hz, 1H),
5.01 (ddd, J ) 8.7, 8.7, 12 Hz, 1H), 3.94 (t, J ) 6.2 Hz, 2H),
3
.13 (d, J ) 2.4 Hz, 2H), 3.70 (t, J ) 6.2 Hz, 2H), 3.60 (t, J )
.4 Hz, 2H), 2.44 (d, J ) 2.4 Hz, 1H), 1.84 (quint, J ) 6.2 Hz,
2.72 (ddd, J ) 6.2, 6.2, 1.6 Hz, 2H), 1.65-1.20 (m, 16H), 1.98-
1.78 (m, 15H); 1 Sn-NMR (CDCl
19
) δ -18.23. Anal. Calcd for
H), 0.90 (s, 9H), 0.06 (s, 6H). Anal. Calcd for C12
H
24
O
2
Si:
3
C, 63.10; H, 10.59. Found: C, 62.86; H, 10.55.
-(Alle n yloxy)-3-[(t er t -b u t yld im e t h ylsilyl)oxy]p r o-
p a n e (23). A mixture of 22 (16.0 g, 70.0 mmol) and t-BuOK
790 mg, 7 mmol) was stirred at 70 °C for 2 h. (The reaction
C
18
H
36
O
2
Sn: C, 53.62; H, 9.00. Found: C, 53.34; H, 8.83.
4-(2-P r op yn yloxy)-1-bu ta n ol (27): colorless oil; R ) 0.47
(hexane/AcOEt, 1:1); IR (neat) 3600-3200, 2950, 2860, 2220,
1
f
-1 1
(
3
1440, 1360, 1100, 940 cm ; H-NMR (270 MHz, CDCl ) δ 4.14
1
was monitored by H-NMR.) The mixture was diluted with
ether (70 mL) and washed with water (50 mL). The organic
layer was filtered through a silica gel pad, and the filtrate was
concentrated under reduced pressure (below 30 °C). The crude
product was purified by flash distillation to give 23 (9.4 g,
5
2
1
(d, J ) 2.8 Hz, 2H), 3.65 (t, J ) 6.0 Hz, 2H), 3.56 (t, J ) 6.0
Hz, 2H), 2.42 (t, J ) 2.8 Hz, 1H), 1.95 (s, 1H), 1.76-1.60 (m,
4H).
1-[(ter t-Bu tyld im eth ylsilyl)oxy]-4-(2-p r op yn yloxy)bu -
ta n e (28): colorless oil; R ) 0.40 (hexane/AcOEt, 10:1); IR
f
9%): colorless oil; bp 68 °C/1.0 mmHg; IR (neat) 3050, 2970,
940, 2870, 1960, 1477, 1450, 1396, 1370, 1356, 1261, 1206,
108, 1043, 1012, 948, 896, 840, 785, 670 cm-1; 1H-NMR (270
(neat) 3320, 2970, 2940, 2870, 2165, 2120, 1476, 1468, 1448,
1393, 1364, 1260, 1100, 1035, 1010, 980, 940, 840, 816, 783,
-1
1
668 cm ; H-NMR (270 MHz, CDCl ) δ 4.13 (d, J ) 2.5 Hz,
3
MHz, CDCl
H), 3.71 (t, J ) 6.3 Hz, 2H), 3.65 (t, J ) 6.3 Hz, 2H), 1.84
quint, J ) 6.3 Hz, 2H), 0.89 (s, 9H), 0.05 (s, 6H).
-(Allen yloxy)-3-[(tr im eth ylsilyl)oxy]p r op a n e (24). A
3
) δ 6.71 (t, J ) 5.8 Hz, 1H), 5.43 (d, J ) 5.8 Hz,
2H), 3.63 (t, J ) 6.2 Hz, 2H), 3.53 (t, J ) 6.2 Hz, 2H), 2.41 (t,
J ) 2.5 Hz, 1H), 1.72-1.50 (m, 4H), 0.89 (s, 9H), 0.04 (s, 6H).
Anal. Calcd for C13
64.03; H, 10.83.
2
(
26 2
H O Si: C, 64.41; H, 10.81. Found: C,
1
mixture of 23 (9.25 g, 40.5 mmol) and TBAF (49 mL of a 1 M
solution in THF, 49 mmol) was stirred at rt for 1.5 h. The
mixture was diluted with ether (100 mL), washed with water
1-(Alle n yloxy)-4-[(t er t -b u t yld im e t h ylsilyl)oxy]b u -
ta n e (29): colorless oil; R
f
) 0.33 (hexane/AcOEt, 10:1); IR
(neat) 3060, 2980, 2950, 2880, 1957, 1738, 1476, 1450, 1392,
1360, 1260, 1206, 1104, 1051, 1012, 986, 944, 894, 841, 780,
(50 mL) and brine (50 mL), and concentrated. The crude
-
1
1
product was directly used for the next reaction.
3
720, 665 cm ; H-NMR (270 MHz, CDCl ) δ 6.72 (t, J ) 5.8
A mixture of the alcohol obtained, HMDS (17.0 mL, 81
mmol), and TMSCl (1 drop) was stirred at 100 °C for 1 h.
Fractional distillation of the reaction mixture afforded 24 (6.66
Hz, 1H), 5.42 (d, J ) 5.8 Hz, 2H), 3.63 (t, J ) 6.2 Hz, 1H),
3.57 (t, J ) 6.2 Hz, 1H), 1.78-1.52 (m, 4H), 0.90 (s, 9H), 0.05
(s, 6H).
g, 88%): colorless oil; bp 50 °C/2.0 mmHg; R
AcOEt, 20:1); IR (neat) 3060, 3030, 2950, 2740, 1956, 1474,
f
) 0.38 (hexane/
1-(Allen yloxy)-4-[(tr im eth ylsilyl)oxy]bu ta n e (30): col-
orless oil; IR (neat) 2970, 2870, 1950, 1730, 1252, 1200, 1100,
-
1
1
1
7
448, 1396, 1356, 1300, 1260, 1210, 1190, 1060, 946, 790, 760,
3
840, 778 cm ; H-NMR (270 MHz, CDCl ) δ 6.71 (t, J ) 5.7
-
1
1
23, 688 cm ; H-NMR (270 MHz, CDCl
3
) δ 6.71 (t, J ) 5.9
Hz, 1H), 5.41 (d, J ) 5.7 Hz, 2H), 3.58 (ddd, J ) 8.6, 6.3, 6.1
Hz, 4H), 1.75-1.53 (m, 4H), 0.09 (s, 9H). Anal. Calcd for
Hz, 1H), 5.43 (d, J ) 5.9 Hz, 2H), 3.68 (t, J ) 6.2 Hz, 2H),
3
9
.63 (t, J ) 6.2 Hz, 2H), 1.85 (quint, J ) 6.2 Hz, 2H), 0.11 (s,
H).
10 20 2
C H O Si: C, 59.95; H, 10.06. Found: C, 59.87; H, 10.06.
4-[(E )-[3-(T r ib u t y ls t a n n y l)-1-p r o p e n y l]o x y ]-1-b u -
ta n ol (31): colorless oil; R ) 0.13 (CH Cl ); IR (neat) 3600-
3100, 2970, 2935, 2885, 2870, 1660, 1645, 1462, 1380, 1194,
Hyd r osta n n yla tion of 24. To a stirred mixture of 24 (6.3
f
2
2
g, 34 mmol) and Pd(PPh
3 4
) (400 mg, 0.35 mmol) was added
-
1
1
n-Bu SnH (18.5 mL, 68 mmol) dropwise. During this addition
3
1127, 1074, 963, 924, 872, 660 cm
; H-NMR (270 MHz,
an exothermic reaction started, and the temperature was
maintained between 50 and 60 °C with a cooling bath if
necessary. Stirring at rt was continued for 1 h, then the crude
CDCl ) δ 6.10, (ddd, J ) 12.4, 1.4, 1.4 Hz, 1H), 4.98 (ddd, J )
3
12.4, 8.5, 8.5 Hz, 1H), 3.64 (m, 4H), 1.80-1.20 (m, 18H), 0.90-
0.80 (m, 15H); HRMS m/ z calcd for C19
40 2
H O Sn 420.2048,
product was diluted with MeOH (50 mL), and K
added to the mixture. After the mixture was stirred at rt for
h, the solvent was removed. Flash column chromatography
gave 25 (6.8 g, 49%) and 26 (1.2 g, 8.7%).
-[(Z)-[3-(Tr ib u t ylst a n n yl)-1-p r op e n yl]oxy]-1-p r o-
p a n ol (25): colorless oil; R ) 0.25 (CH Cl ); IR (neat) 3650-
100, 3050, 2970, 2930, 2880, 1650, 1460, 1416, 1370, 1295,
260, 1192, 1180, 1155, 1100, 1000, 960, 930, 873, 735, 686,
2
CO
3
(2 g) was
found 420.2037.
4-[(E )-[3-(Tr ib u t ylst a n n yl)-1-p r op e n yl]oxy]b u t a n a l
(2): colorless oil; IR (neat) 1730, 1645 cm
-
1
1
5
; H-NMR (270
MHz, CDCl ) δ 9.79 (t, J ) 1.4 Hz, 1H), 6.08 (ddd, J ) 12.1,
3
3
1.3, 1.3 Hz, 1H), 4.95 (ddd, J ) 12.1, 8.6, 8.6 Hz, 1H), 3.63 (t,
J ) 6.1 Hz, 2H), 2.55 (ddd, J ) 7.7, 7.7, 1.4 Hz, 2H), 1.95 (m,
2H), 2.00-0.80 (m, 27H).
f
2
2
3
1
6
4
1
5-(2-P r op yn yloxy)-1-p en ta n ol (33): colorless oil; R
f
)
-
1
1
60 cm ; H-NMR (270 MHz, CDCl ) δ 5.58-5.46 (m, 1H),
3
0.32 (hexane/AcOEt, 1:1); bp 89-91 °C/3 mmHg; IR (neat)
3600-3100, 3320, 3950, 2880, 3130, 1368, 1100 cm ; H-NMR
(270 MHz, CDCl
-1 1
.28 (ddd, J ) 9.2, 9.2, 6.2 Hz, 1H), 3.62-3.50 (m, 4H), 1.95-
.20 (m, 16H), 1.00-0.90 (m, 15H). Anal. Calcd for
3
) δ 5.55 (brs, 1H), 4.14 (d, J ) 2.5 Hz, 2H),
C
18
H
38
O
2
Sn: C, 53.35; H, 9.45. Found: C, 53.17; H, 9.39.
3.65 (t, J ) 6.4 Hz, 2H), 3.53 (t, J ) 6.4 Hz, 2H), 2.42 (t, J )
3
-[(E )-[3-(Tr ib u t ylst a n n yl)-1-p r op e n yl]oxy]-1-p r o-
p a n ol (26): colorless oil; R ) 0.14 (CH Cl ); IR (neat) 3600-
100, 2970, 2930, 2880, 2860, 1530, 1662, 1644, 1466, 1420,
380, 1343, 1296, 1251, 1191, 1128, 1075, 1074, 964, 620, 874,
2.5 Hz, 1H), 1.70-1.40 (m, 6H).
f
2
2
1-[(ter t-Bu tyld im eth ylsilyl)oxy]-5-(2-p r op yn yloxy)p en -
3
1
7
ta n e (34): colorless oil; R
(neat) 3325, 2950, 2880, 2130, 1728, 1479, 1470, 1396, 1368,
1260, 1105, 1012, 862, 780, 669 cm
f
) 0.36 (hexane/AcOEt, 20:1); IR
-1
1
-1
1
14, 660 cm ; H-NMR (270 MHz, CDCl
3
) δ 6.09 (m, 1H), 5.00
; H-NMR (270 MHz,
(
ddd, J ) 8.7, 8.7, 13 Hz, 1H), 3.80-3.70 (m, 4H), 1.95-1.20
CDCl ) δ 4.13 (d, J ) 2.2 Hz, 2H), 3.61 (t, J ) 6.4 Hz, 2H),
3
(m, 16H), 1.00-0.80 (m, 15H); HRMS m/ z calcd for C18
H
38
O
2
Sn
3.51 (t, J ) 6.5 Hz, 2H), 2.41 (t, J ) 2.3 Hz, 1H), 1.70-1.30
(m, 6H), 0.89 (s, 9H), 0.06 (s, 6H).
4
06.1894, found 406.1892.
-[(Z)-[3-(Tr ib u t ylst a n n yl)-1-p r op en yl]oxy]p r op a n a l
12): colorless oil; R ) 0.36 (hexane/AcOEt, 10:1); IR (neat)
3
1-(Alle n yloxy)-5-[(t er t -b u t yld im e t h ylsilyl)oxy]p e n -
ta n e (35): colorless oil; bp 81-83 °C/0.8 mmHg; R ) 0.83
(
f
f

7
446 J . Org. Chem., Vol. 62, No. 21, 1997
Kadota et al.
(
hexane/AcOEt, 3:1); IR (neat) 2980, 2910, 2370, 1968, 1483,
1H), 3.60 (t, J ) 5.9 Hz, 2H), 2.47 (ddd, J ) 7.0, 7.0, 1.8 Hz,
-1
119
1
473, 1458, 1399, 1360, 1263, 1218, 1112, 1058, 854, 792 cm
;
2H), 1.80-1.20 (m, 16H), 0.98-0.70 (15H); Sn-NMR (CDCl
δ -18.83.
3
)
1
H-NMR (270 MHz, CDCl ) δ 6.72 (t, J ) 6.0 Hz, 1H), 5.42 (d,
3
J ) 6.0 Hz, 2H), 3.61 (t, J ) 6.4 Hz, 2H), 3.55 (t, J ) 6.4 Hz,
H), 1.70-1.30 (m, 6H), 0.89 (s, 9H), 0.04 (s, 6H).
-(Allen yloxy)-5-[(tr im eth ylsilyl)oxy]bu ta n e (36): col-
orless oil; bp 68-70 °C/1.5 mmHg; IR (neat) 3041, 3953, 2864,
Gen er a l P r oced u r e for th e Cycliza tion of 1, 2, a n d 12-
15. To a stirred solution of the substrate (0.025 mmol) in
2
1
CH
2
Cl
2
(0.1 mL) at -78 °C was added a Lewis or protic acid
Cl ). Thermal reactions were performed
solution (1.0 M, in CH
2
2
472, 1447, 1352, 1251, 1096, 1043, 840 cm-1; H-NMR (270
MHz, CDCl ) δ 6.72 (t, J ) 6.0 Hz, 2H), 5.42 (d, J ) 6.0 Hz,
H), 3.58 (t, J ) 6.3 Hz, 2H), 3.55 (t, J ) 6.3 Hz, 2H), 1.70-
.30 (m, 6H), 0.11 (s, 9H).
-[(Z)-[3-(Tr ib u t ylst a n n yl)-1-p r op e n yl]oxy]-1-p e n -
ta n ol (37): colorless oil; R ) 0.30 (CH Cl ); IR (neat) 3600-
200, 3050, 2950, 2850, 1660, 1470, 1370, 1260, 1169, 1120,
1
in toluene. Reaction times varied with conditions and reagents
as shown in Tables 1-3. The reactions were quenched with
3
Et N and the mixtures allowed to warm to room temperature.
The mixtures were filtered through a silica gel column. The
yields and isomer ratios were determined by capillary GC
analysis.
1
3
2
1
5
f
2
2
3
9
(2R*,3S*)-3-Ben zoyl-2-vin yltetr a h yd r ofu r a n (42): col-
75, 890, 750, 680 cm-1; H-NMR (270 MHz, CDCl
1
) δ 5.77
3
f
orless oil; R ) 0.32 (hexane/AcOEt, 10:1); IR (neat) 3070, 2980,
(
1
1
ddd, J ) 6.2, 1.2, 1.2 Hz, 1H), 4.50 (ddd, J ) 9.0, 9.0, 6.2 Hz,
2950, 2875, 1720, 1644, 1603, 1588, 1494, 1452, 1402, 1368,
1317, 1273, 1180, 1114, 1100, 1073, 1029, 989, 930, 712, 688
H), 3.72-3.62 (m, 4H), 1.70-1.20 (m, 18H), 1.00-0.90 (m,
-
1 1
5H). Anal. Calcd for C20 Sn: C, 55.44; H, 9.77.
H
42
O
2
3
cm ; H-NMR (270 MHz, CDCl ) δ 8.02-8.08 (m, 2H), 7.62-
Found: C, 55.29; H, 9.75.
-[(E )-[3-(Tr ib u t ylst a n n yl)-1-p r op e n yl]oxy]-1-p e n -
ta n ol (38): colorless oil; R ) 0.18 (CH Cl ); IR (neat) 3600-
100, 3050, 2950, 2850, 1658, 1640, 1450, 1370, 1330, 1280,
190, 1120, 1070, 957, 920, 870, 712, 670 cm-1; 1H-NMR (270
) δ 6.10 (ddd, J ) 12, 1.2, 1.2 Hz, 1H), 4.95 (ddd,
7.54 (m, 1H), 7.49-7.41 (m, 2H), 5.95 (ddd, J ) 17.0, 10.4, 5.0
Hz, 1H), 5.41 (ddd, J ) 17.0, 1.8, 1.8 Hz, 1H), 5.29 (ddd, J )
6.0, 2.2, 2.2 Hz, 1H), 5.22 (ddd, J ) 10.3, 1.8, 1.8 Hz, 1H),
4.57-4.52 (m, 1H), 4.20-4.00 (m, 2H), 2.39-2.05 (m, 2H);
5
f
2
2
3
1
HRMS m/ z calcd for C13
(2R*,3R*)-3-Ben zoyl-2-vin yltetr a h yd r ofu r a n e (43): col-
orless oil; R
14 3
H O 218.0943, found 218.0922.
MHz, CDCl
3
J ) 8.4, 8.4, 12 Hz, 1H), 3.70-3.55 (m, 4H), 1.70-1.20 (m,
f
) 0.21 (hexane/AcOEt, 10:1); IR (neat) 3080, 2980,
1
4
8H), 0.90-0.80 (m, 15H); HRMS m/ z calcd for C20
34.2205, found 434.2210.
H
42
O
2
Sn
2950, 2890, 1735, 1650, 1605, 1584, 1490, 1450, 1375, 1354,
1313, 1270, 1178, 1113, 1068, 1027, 989, 927, 880, 805, 769,
-
1
1
5
-[(Z)-[3-(Tr ib u t ylst a n n yl)-1-p r op en yl]oxy]p en t a n a l
14): colorless oil; R ) 0.11 (hexane/AcOEt, 50:1); IR (neat)
045, 2950, 2850, 1730, 1660, 1468, 1380, 1260, 1110, 963, 879,
3
709, 685 cm ; H-NMR (270 MHz, CDCl ) δ 8.06-8.00 (m,
(
f
2H), 7.60-7.52 (m, 1H), 7.48-7.40 (m, 2H), 5.94 (ddd, J ) 17.3,
11.6, 6.2 Hz, 1H), 5.61 (ddd, J ) 6.2, 4.6, 2.6 Hz, 1H), 5.42
(ddd, J ) 17.3, 1.6, 1.6 Hz, 1H), 5.23 (ddd, J ) 17.3, 1.7, 1.6
Hz, 1H), 4.43 (m, 1H), 4.16 (q, J ) 7.8 Hz, 1H), 3.94 (ddd, J )
17.0, 17.0, 5.3 Hz, 1H), 2.50-2.35 (m, 1H), 2.25-2.15 (m, 1H);
3
7
1
6
7
-
1
1
42 cm ; H-NMR (270 MHz, CDCl ) δ 9.77 (t, J ) 1.6 Hz,
3
H), 5.75 (ddd, J ) 6.0, 1.2, 1.2 Hz, 1H), 4.51 (ddd, J ) 9.1,
.0, 6.0 Hz, 1H), 3.69 (t, J ) 6.1 Hz, 2H), 2.48 (ddd, J ) 7.3,
.3, 1.6 Hz, 2H), 1.80-1.20 (m, 16H), 0.98-0.80 (m, 15H);
HRMS m/ z calcd for C13
14 3
H O 218.0943, found 218.0945.
1
19
Sn-NMR (CDCl
-[(E)-[3-(Tr ib u t ylst a n n yl)-1-p r op en yl]oxy]p en t a n a l
15): colorless oil; R ) 0.08 (hexane/AcOEt, 50:1); IR (neat)
960-2550, 1714, 1648, 1630, 1446, 1368, 1177, 1112, 909
3
) δ -16.64.
5
Ack n ow led gm en t. This work was financially sup-
ported by a Grant-in-Aid for Scientific Reserch from the
Ministry of Education, Science and Culture, J apan.
(
f
2
-
1
1
cm ; H-NMR (270 MHz, CDCl
3
) δ 9.76 (t, J ) 1.8 Hz, 1H),
6
.09 (brd, J ) 12.6 Hz, 1H), 4.95 (ddd, J ) 12.6, 8.6, 8.6 Hz,
J O971309C