On the relative reactivities of allyl, crotyl, α-oxygenated crotyl, γ-oxygenated α-methylallyl, and allenyl tri-n-butylstannane reagents in Lewis acid promoted additions to aldehydes

Full text of DOI:10.1021/jo9519975

2904
J . Org. Chem. 1996, 61, 2904-2907
as 2.1 and 2.35 and, more recently, with allenic stannanes
On th e Rela tive Rea ctivities of Allyl,
exemplified by 2.5.6
Cr otyl, r-Oxygen a ted Cr otyl, γ-Oxygen a ted
r-Meth yla llyl, a n d Allen yl
Tr i-n -bu tylsta n n a n e Rea gen ts in Lew is
Acid P r om oted Ad d ition s to Ald eh yd es
J ames A. Marshall,* J ill A. J ablonowski, and
Gregory S. Welmaker
Department of Chemistry, University of Virginia,
Charlottesville, Virginia 22901
Received November 10, 1995
The addition of allylmetal reagents to carbonyl com-
pounds, especially aldehydes, has received a great deal
of attention as an important C-C bond forming reaction
in natural product synthesis.1 Of the various possible
combinations of metals and allylic groups, crotylboron
and tin reagents have shown particular promise.2,3 These
reagents are readily prepared, relatively stable, and
typically undergo highly diastereoselective addition reac-
tions with the formation of two contiguous stereocenters.
In the course of this work we noted that a number of
reactions reported for allyl- and crotylstannanes, such
as transmetalations with metal halides, could not be
effected with the oxygenated counterparts. We also
sensed that the oxygenated stannanes were less reactive
in additions to certain aldehydes. Since these issues
directly bear on the potential scope and limitations of the
foregoing methodology, we decided to examine competi-
tion reactions between various allylic stannanes and
representative aldehydes as a means of assessing relative
reactivities.7,8
The studies were conducted by treating an equimolar
solution of crotylstannane (3.1) and the competing stan-
nane 3.3 with the aldehyde and 1.2 equiv of BF3‚OEt2 in
CH2Cl2 at low temperature, allowing the reaction to
proceed until the aldehyde was no longer present. After
In the case of boron, the use of chiral auxiliaries makes
possible the synthesis of highly enantioenriched homoal-
lylic alcohols. This approach is effective because these
reactions proceed through cyclic six-center transition
states in which the chiral ligands interact with the
aldehyde substituent. However, most of the synthetically
useful allylic tin additions to aldehydes proceed by an
acyclic transition state in which the tin substituent
orients anti to the forming carbon-carbon bond. Accord-
ingly, chiral ligands on tin would not be expected to
interact with the aldehyde substituent. In principle, the
use of a chiral Lewis acid could provide a chiral environ-
ment at the carbonyl center, thereby leading to enantio-
face selectivity in the ensuing addition. Several recent
studies confirm this expectation.4
1
quenching, the mixture was analyzed by H NMR and
GC to ascertain the identity and ratio of products. In
each case the competition was run two or three times to
assure reproducibility. Response factors were deter-
mined for the GC analyses to ensure accuracy.
With cyclohexanecarboxaldehyde (eq 3) allyltri-n-bu-
tylstannane (3.3a ) showed comparable reactivity to the
standard, crotylstannane (3.1). As suspected, the (E)-γ-
(silyloxy)stannane 3.3b was significantly less reactive
than stannane 3.1.9 Furthermore, the (Z)-isomer 3.3c
was even less reactive; only the crotyl adduct 3.2 could
be isolated from the competition experiment.10 The (Z)-
γ-MOM 3.3d ,11 the R-OTBS 3.3e,12 and the R-OMOM
stannane 3.3f11 were similarly uncompetitive.
A second approach to enantioselective additions entails
the design of reagents in which the tin substituent
resides at a stereogenic allylic center. In view of the well-
established stereoelectronic requirements of allyltin SE2′
reactions, this approach should be highly effective for
enantio- and diastereoselective synthesis.3
(5) Marshall, J . A. Chemtracts: Org. Chem. 1992, 5, 75. Marshall,
J . A. Chem. Rev. 1996, 96, 31.
(6) Marshall, J . A.; Perkins, J . F.; Wolf, M. A. J . Org. Chem. 1995,
60, 5556.
(7) Kinetics have been reported for reactions of allylic silanes,
stannanes, and germanes with carbocations. Hagen, G.; Mayr, H. J .
Am. Chem. Soc. 1991, 113, 4954.
(8) These studies were initiated at the University of South Carolina.
Preliminary results are reported in the Ph.D. Thesis of G. S. Welmaker,
University of South Carolina, 1992.
(9) For leading references to the synthesis of these stannanes, see:
Marshall, J . A.; J ablonowski, J . A.; Elliott, L. M. J . Org. Chem. 1995,
60, 2662.
(10) Keck and co-workers have also noted that trans-crotyl tri-n-
butylstannane is more reactive than the cis isomer in additions to
various aldehydes. Keck, G. E.; Savin, K. A.; Cressman, E. N. K.;
Abbott, D. E. J . Org. Chem. 1994, 59, 7889.
(11) Marshall, J . A.; Welmaker, G. S.; Gung, B. W. J . Am. Chem.
Soc. 1991, 113, 647.
(12) Marshall, J . A.; Welmaker, G. S. Tetrahedron Lett 1991, 32,
2101.
For several years we have been pursuing such a
strategy first with R- and γ-oxygenated stannanes such
(1) For an overview, see: Yamamoto, Y.; Asao, N. Chem. Rev. 1993,
93, 2207.
(2) Boron: Roush, W. R. In Comprehensive Organic Synthesis; Trost,
B. M., Ed.; Pergamon Press: New York, 1991; Vol. 2, Chapter 1.1; pp
1-49.
(3) Tin: ref 2 and Yamamoto, Y.; Sheda, N. Advances in Detailed
Reaction Mechanisms; J AI Press, 1994; Vol. 3, pp 1-44.
(4) (a) Marshall, J . A.; Tang, Y. Synlett 1992, 653. (b) Keck, G. E.;
Krishnamurthy, D.; Grier, M. C. J . Org. Chem. 1993, 58, 6543 and
references cited therein.
(13) Marshall, J . A.; Hinkle, K. W. J . Org. Chem. 1995, 60, 1920.
0022-3263/96/1961-2904$12.00/0 © 1996 American Chemical Society

Notes
J . Org. Chem., Vol. 61, No. 8, 1996 2905
with (E)-2-heptenal than with cyclohexanecarboxalde-
hyde, possibly due in part to steric effects.
As a cross check on the foregoing results, we conducted
studies in which the (Z)-γ-(silyloxy)allylic stannane 3.3c
was allowed to compete in turn with the R-isomer 3.3e,
the (Z)-γ-OMOM 3.3d , and the R-OMOM analogue 3.3f
(eq 5). The ratios of adducts from the latter two competi-
tions 4.2b :5.1b and 4.2b :5.1c showed good agreement
with those predicted from eq 4 (81:19 vs 12:3 and 88:12
vs 12:2), but the ratio of adducts 4.2b:5.1a (49:51) from
3.3c and the R-isomer 3.3e was less than would be
expected from the results of eq 4 (12:6). Conceivably,
selective decomposition of stannane 3.3c could account
for the discrepancy. However, when this reaction was
monitored by quenching at various times, the ratios of
the two products and the unreacted stannanes were each
found to be essentially constant.
When these experiments were performed with (E)-2-
heptenal, the ratios of crotyl to competitor stannane
adducts 4.1:4.2 decreased. Also, with this aldehyde
allyltributylstannane (3.3a ) was somewhat less reactive
than crotylstannane 3.1. Interestingly, the (E)- and (Z)-
γ-(silyloxy)stannanes 3.3b and 3.3c now showed compa-
rable reactivity to each other, but they were still consid-
erably less reactive than the crotylstannane 3.1. The γ-
and R-OMOM stannanes 3.3d and 3.3f were the least
competitive of the series. Both the γ-OTBS stannane
3.3c and the γ-OMOM stannane 3.3d showed higher
reactivity than their R-isomers 3.3e and 3.3f. However,
in all cases the crotylstannane 3.1 consumed the lion’s
share of the aldehyde.
This experiment also showed that as long as unreacted
aldehyde was present, no isomerization of the R-OTBS
stannane 3.3e to the γ-isomer 3.3c occurred. However,
upon complete consumption of aldehyde the isomerization
was complete within 10-20 min. Thus it would appear
that the unreacted aldehyde somehow retards R to γ
isomerization. Clearly this phenomenon makes possible
the use of R-oxygenated allylic stannanes as SE2′ reagents
except with relatively unreactive aldehydes.5,14
We also monitored the competition reaction between
crotylstannane 3.1 and the (Z)-γ-OTBS stannane 3.3c for
2-heptenal as a function of time (see eq 4). Within 2 min,
all of the aldehyde was consumed. Furthermore, during
the course of this reaction the intial 85:15 ratio of (E)-
and (Z)-crotyl isomers decreased as the faster reacting
(E)-isomer was consumed.10,15 Thus, most of the adduct
4.1 arises from the (E)-crotyl isomer over a time period
in which negligable decomposition of the competitor
stannanes 3.3b-f would be expected to occur. In light
of these findings, we believe that the trends reflected in
eq 4 are valid, but clearly care must be exercised in their
extrapolation.
In the second phase of this study we examined the
relative reactivity of crotylstannane 3.1 vs the (R)- and
(S)-γ-OTBS and the (S)-OMOM allylic stannanes 3.3c-
(R), 3.3c(S), and 3.3d (S) in MgBr2-promoted additions
to (S)-2-(benzyloxy)propanal (6.1) (eq 6). All of these
A competition experiment in which the crotylstannane
3.1 (1 equiv) was allowed to react with 1 equiv each of
cyclohexanecarboxaldehyde and (E)-2-heptenal yielded a
1:1 mixture of adducts 3.2 and 4.1. Thus it may be
concluded that stannanes 3.3b-f react relatively faster
(14) Cf. Marshall, J . A.; Gung, W. Y. Tetrahedron 1989, 45, 1043.
Marshall, J . A.; Gung, B. W. Israel J . Chem. 1991, 31, 199. Marshall,
J . A.; Yashunsky, D. V. J . Org. Chem. 1991, 56, 5493.
(15) The crotylstannane used in the study was prepared from (E)-
crotyl chloride and Bu3SnLi. Matarasso-Tchiroukhine, E.; Cadiot, P.
J . Organomet. Chem. 1976, 121, 155, 169.

2906 J . Org. Chem., Vol. 61, No. 8, 1996
Notes
Ta ble 1. Rela tive Rea ctivities of Allylic a n d Allen ic
oxygenated stannanes were significantly less reactive
than crotyl.10 As we have previously shown, both the (R)-
and (S)-OTBS stannanes 3.3c(R) and 3.3c(S) give rise
to single adducts differing in double bond geometry with
the former reacting at the faster rate.16
Sta n n a n es w ith (E)-2-Hep ten a l P r om oted by BF 3‚OEt2 in
CH2Cl2 a t -78 °C
R
a/b
As a consequence of these reactivity differences, a
number of prototype addition reactions reported for allyl-
and crotylstannanes may not be applicable to oxygenated
stannanes and allenylstannanes.4 Included in this cat-
egory are reactions that require harsh Lewis acid pro-
moters and those that proceed slowly at room tempera-
ture.17 Considering the enormous potential of these
latter reagents in asymmetric synthesis, it is of interest
to search for more compatible and effective promoters
and, ideally, true catalysts that will accommodate the less
reactive allenic and oxygenated allylic stannanes.
In the final phase of this survey we pitted the allenyl-
stannanes 7.1a and 7.1b6 against crotylstannane 3.1, an
85:15 mixture of (E)- and (Z)-isomers, in additions to
2-heptenal with BF3‚OEt2 as the Lewis acid promoter.
Both allenyl stannanes were markedly less reactive than
crotyl.
Exp er im en ta l Section 18
Gen er a l P r oced u r e for BF3‚OEt2-P r om oted Com p etitive
Ad d ition s. To a 0.1 M solution of the aldehyde and stannanes
(1.0 equiv each) in CH2Cl2 cooled to -78 °C was added BF3‚OEt2
(1.2 equiv). After 1 h, TLC analysis indicated that all of the
aldehyde had been consumed and the reaction was quenched
with saturated NaHCO3 solution. After warming to rt, the
mixture was extracted with CH2Cl2, dried over Na2SO4, and
concentrated under reduced pressure. The adducts were sepa-
rated from residual stannane by flash chromatography on silica
gel with ethyl acetate in hexane. The ratio of adducts was
determined by GC and 1H NMR analysis.
Sta n n a n es a n d P r od u ct Com p osition a s a F u n ction of
Rea ction Tim e. Three individual experiments were performed
simultaneously according to the foregoing protocol. Each of the
three was quenched after varying time intervals from 2 to 30
min. Analysis of the products and unreacted stannane was
performed as described. Known mixtures of stannanes and
products were analyzed by GC to establish relative response
factors. 1H NMR analysis was employed as a cross check on
the structures of components and composition of the mixtures.
Gen er a l P r oced u r e for MgBr 2‚OEt2-P r om oted Com p eti-
tive Ad d ition s. A 0.1 M solution of the aldehyde and stannanes
(1.0 equiv each) in CH2Cl2 was cooled to -23 °C, and MgBr2‚OEt2
(1.2 equiv) was added. After 1 h, the mixture was warmed to 0
°C. After an additional 0.5 h, TLC analysis indicated that all
of the aldehyde had been consumed and the reaction was
quenched with saturated NaHCO3 solution. After warming to
rt, the mixture was extracted with CH2Cl2, dried over Na2SO4,
and concentrated under reduced pressure. The adducts were
separated from residual stannane by flash chromatography on
silica gel with ethyl acetate in hexane. The ratio of adducts was
determined by GC and 1H NMR analysis.
The foregoing studies provide a semiquantitative com-
parison of reactivities for some of the more synthetically
useful stannane reagents (Table 1). In all cases the
parent crotyl- and allylstannanes 3.1 and 3.3a are the
most reactive. The relative reactivity is also dependent
upon the structure of the aldehyde (compare eq 3 with
eq 4). Thus, the R- and γ-OTBS and OMOM allylic
stannanes 3.3b-f react faster with (E)-2-heptenal than
with cyclohexanecarboxaldehyde, but none can effectively
compete with crotylstannane 3.1. We were unable to
examine a range of Lewis acids as the oxygenated
stannanes are decomposed by the stronger of these and
fail to react when weaker Lewis acids are used as
promoters.
P r eviou sly Un ch a r a cter ized Sta n n a n es a n d Ad d u cts:
(r el 1R,2S,Z)-4-((ter t-Bu tyld im eth ylsilyl)oxy)-1-cycloh exyl-
2-m eth yl-3-p en ten -1-ol (3.4e). To a stirred solution of stan-
(17) Cf. Hachiya, I.; Kobayashi, S. J . Org. Chem. 1993, 58, 6958.
Bedeschi, P.; Casolari, S.; Costa, A. L.; Tagliavini, E.; Umani-Ronchi,
A. Tetrahedron Lett. 1995, 36, 7897. Aspinall, H. C.; Browning, A. F.;
Greeves, N.; Ravenscroft, P. Tetrahedron Lett. 1994, 35, 4639. Henry,
K. J .; Grieco, P. A.; J agoe, C. T. Tetrahedron Lett. 1992, 33, 1817.
(18) For typical experimental protocols, see: Marshall, J . A.; Wang,
X-j. J . Org. Chem. 1991, 56, 960. Unless otherwise stated 1H and 13C
NMR spectra were determined at 300 and 75 MHz, respectively, on
dilute solutions in CDCl3.
(16) Marshall, J . A.; J ablonowski, J . A.; Luke, G. P. J . Org. Chem.
1994, 59, 7825.

Notes
J . Org. Chem., Vol. 61, No. 8, 1996 2907
nane 3.3e12 (128 mg, 0.27 mmol) and cyclohexanecarboxaldehyde
(27 mg, 0.24 mmol) in 1.5 mL of CH2Cl2 was added BF3‚OEt2
(31 µL, 0.30 mmol). After 1.5 h, TLC analysis indicated that
all the aldehyde had been consumed and the reaction was
quenched with saturated NaHCO3 solution. After warming to
rt, the mixture was extracted with CH2Cl2, dried over Na2SO4,
and concentrated under reduced pressure. The crude residue
was purified by flash chromatography on silica gel. Elution with
5% ethyl acetate in hexane provided 71.5 mg (65%) of a 45:55
mixture of (E)- and (Z)-isomers of adduct 3.4e. Data for the (Z)-
isomer: 1H NMR δ 6.16 (dd, J ) 5.8, 0.8 Hz, 1H), 4.33 (dd, J )
9.2, 5.8 Hz, 1H), 3.19 (dd, J ) 6.0, 6.0 Hz, 1H), 2.86 (dm, J )
6.9 Hz, 1H), 1.90 (m, 1H), 1.83-1.35 (m, 6H), 1.36-1.02 (m, 4H),
0.98 (d, J ) 6.9 Hz, 3H), 0.92 (s, 9H), 0.12 (s, 6H) ppm; IR (film)
(r el 4S,5R,E)-1-((ter t-Bu tyld im eth ylsilyl)oxy)-4-m eth yl-
6-u n d ecen -2-yn -5-ol (7.2a ). The procedure described for 3.4e
was employed with stannane 7.1a 19 (91 mg, 0.19 mmol), (E)-2-
heptenal (21 mg, 0.19 mmol), and BF3‚OEt2 (22 µL, 0.22 mmol).
The product was purified by flash chromatography on silica gel.
Elution with 5% ethyl acetate in hexane provided 40 mg (69%)
of a 65:35 mixture of syn and anti isomers of adduct 7.2a . Data
for the syn isomer: 1H NMR δ 5.71 (ddd, J ) 15.3, 7.2, 6.5 Hz,
1H), 5.45 (ddd, J ) 15.3, 7.0, 1.4 Hz, 1H), 4.32 (dd, J ) 5.8, 1.9
Hz, 2H), 3.89 (dd, J ) 9.9, 6.8 Hz, 1H), 2.54 (dq, J ) 9.9, 7.2 Hz,
1H), 2.03 (m, 2H), 1.91-1.48 (m, 3H), 1.47-1.22 (m, 4H), 1.15
(d, J ) 7.0 Hz, 3H), 0.12 (s, 6H) ppm; 13C NMR δ 134.3, 19.9,
85.8, 81.7, 75.8, 51.9, 33.7, 31.9, 31.2, 25.8, 22.1, 18.3, 17.0, 13.9,
-5.1 ppm; IR (film) ν 3415, 2200 cm-1
18H34O2Si: C, 69.62; H, 11.04. Found: C, 69.35; H, 11.10.
(r el 4S,5R,E)-1-(P iva loyloxy)-4-m eth yl-6-u n d ecen -2-yn -
. Anal. Calcd for
ν 3415 cm-1
.
C
(r el 3S,4R,E,E)-1-((ter t-Bu tyld im eth ylsilyl)oxy)-3-m eth -
yl-1,5-d eca d ien -4-ol (4.2e). The above procedure was em-
ployed with stannane 3.3e12 (300 mg, 0.63 mmol), (E)-2-heptenal
(64 mg, 0.56 mmol), and BF3‚OEt2 (77 µL, 0.75 mmol). The crude
residue was purified by flash chromatography on silica gel.
Elution with 5% ethyl acetate in hexane provided 139 mg (83%)
of a 79:21 mixture of (E)- and (Z)-isomers of silyl enol ether 4.2e.
Data for the (E)-isomer: 1H NMR δ 6.24 (dd, J ) 5.9, 1.0 Hz,
1H), 5.62 (ddd, J ) 15.3, 6.7, 6.7 Hz, 1H), 5.43 (ddd, J ) 15.4,
6.9, 1.2 Hz, 1H), 4.30 (dd, J ) 9.0, 6.0 Hz, 1H), 3.93 (dd, J )
5.5, 5.5 Hz, 1H), 2.85 (m, 1H), 2.01 (m, 2H), 1.47-1.20 (m, 4H),
1.85-0.75 (m, 16H), 0.13 (s, 6H) ppm; 13C NMR δ 132.5, 130.1,
111.5, 76.4, 34.8, 31.8, 31.2, 25.6, 25.5, 21.9, 18.0, 16.4, 13.7,
5-ol (7.2b). The procedure described for 3.4e was employed with
stannane 7.1b18 (89 mg, 0.20 mmol), (E)-2-heptenal (21 mg, 0.19
mmol), and BF3‚OEt2 (22 µL, 0.22 mmol). The product was
purified by flash chromatography on silica gel. Elution with 10%
ethyl acetate in hexane provided 28 mg (60%) of a 68:32 mixture
of syn and anti isomers of adduct 7.2b. Data for the syn
isomer: 1H NMR δ 5.72 (ddd, J ) 15.4, 8.3, 6.2 Hz, 1H), 5.45
(dd, J ) 15.7, 7.2 Hz, 1H), 4.66 (s, 2H), 3.89 (dd, J ) 6.6, 6.4
Hz, 1H), 2.58 (m, 1H), 2.08 (m, 3H), 1.58-1.22 (m, H), 1.22 (s,
9H), 1.16 (d, J ) 7.2, 3H), 0.91 (d, J ) 6.9, 3H) ppm; IR (film) ν
3475, 2200 cm-1
.
-5.6, -5.7 ppm; IR (film) ν 3450 cm-1
17H34O2Si: C, 68.39; H, 11.48. Found: C, 68.48; H, 11.43.
(r el 3S,4R,E,E)-1-(Meth oxym eth oxy)-3-m eth yl-1,5-d eca -
. Anal. Calcd for
Ack n ow led gm en t. We are grateful to the National
Science Foundation for support of this study through a
research grant (CHE 9220166). We thank Drs. Shiping
Xie and Mark Wolf and Ms. Michelle Elliott for provid-
ing the various stannanes used in this study.
C
d ien -4-ol (4.2f). The above procedure was employed with
stannane 3.3f11 (255 mg, 0.63 mmol), (E)-2-heptenal (64 mg, 0.56
mmol), and BF3‚OEt2 (77 µL, 0.75 mmol). The product was
purified by flash chromatography on silica gel. Elution with 20%
ethyl acetate in hexane provided 88 mg (69%) of a 60:40 mixture
of (Z)- and (E)-isomers of enol ether 4.2f. Data for the (E)-
isomer: 1H NMR δ 6.18 (d, J ) 6.4 Hz, 1H), 5.66 (ddd, J ) 15.3,
6.7, 2.2 Hz, 1H), 5.35 (ddd, J ) 15.4, 5.8, 1.4 Hz, 1H), 4.80 (s,
2H), 3.66 (dd, J ) 9.5, 6.5 Hz, 1H), 3.94 (dd, J ) 6.0, 6.0 Hz,
1H), 3.40 (s, 3H), 2.90 (m, 1H), 2.02 (m, 2H), 1.40-1.20 (m, 4H),
0.98 (d, J ) 7.0 Hz, 3H), 0.90 (d, J ) 6.8 Hz, 3H), 0.87 (d, J )
7.2 Hz, 3H) ppm; 13C NMR (100 MHz) δ 143.0, 133.0, 130.0,
110.2, 96.4, 76.8, 55.8, 35.4, 32.0, 31.4, 22.2, 16.7, 13.9 ppm; IR
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
previously unknown adducts (5 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O9519975
(film) ν 3450 cm-1
.
(19) Marshall, J . A.; Xie, S. J . Org. Chem. 1995, 60, 7230.