Tandem Transformations Involving Allylic Silanes. 2. Highly Diastereoselective Substitutions Involving [(Trialkylsilyl)methyl]cyclohexene Derivatives with Aldehydes. Synthetic Studies on the Problem of Lewis Acid-Promoted Protodesilylation and Enolization

Article abstract of DOI:10.1021/jo970821v

Diels-Alder cycloaddition of 2-[(trialkylsilyl)methyl]-1,3-butadienes with a variety of dienophiles and substitution reactions between these allylsilane-containing adducts and aldehyde or acid chloride electrophiles have been combined into "tandem sequential reactions." These tandem sequences proceed with equal or greater yield (50-80%) than the reactions performed separately with no decrease in regio- or stereoselectivity. The sequence produces cyclic compounds with three or four stereogenic centers with good to excellent diastereoselectivity from three simple, noncyclic, and achiral reaction partners. Unprecedented levels of diastereoselectivity (de >93%) have been achieved in allylic substitution reactions involving substituted [(trialkylsilyl)methyl]cyclohexene derivatives with aldehyde electrophiles. During the course of these studies, protodesilylation of allylsilanes has been investigated in detail, and a cocatalyst system of TiCl₄ and Me₂AlCl has been developed that has eliminated silicon loss in the substitution reactions studied. Lewis acid-promoted enolization of ester and ketone substrates with chiral centers adjacent to the carbonyl moiety has been studied also. It has been shown that this event in our studies occurs primarily during catalyst quench. This isomerization is prevented by quenching the catalyst with a Lewis base, such as methanol or triethylamine, prior to aqueous workup.

Full text of DOI:10.1021/jo970821v

5254
J . Org. Chem. 1997, 62, 5254-5266
Articles
Ta n d em Tr a n sfor m a tion s In volvin g Allylic Sila n es. 2. High ly
Dia ster eoselective Su bstitu tion s In volvin g
[(Tr ia lk ylsilyl)m eth yl]cycloh exen e Der iva tives w ith Ald eh yd es.
Syn th etic Stu d ies on th e P r oblem of Lew is Acid -P r om oted
P r otod esilyla tion a n d En oliza tion
Michael G. Organ,*,† Derick D. Winkle,† and J ohn Huffmann‡
Department of Chemistry, Indiana University-Purdue University at Indianapolis,
402 North Blackford Street, Indianapolis, Indiana, 46202, and Molecular Structure Center,
Department of Chemistry, Indiana University, Bloomington, Indiana 47405
Received May 7, 1997X
Diels-Alder cycloaddition of 2-[(trialkylsilyl)methyl]-1,3-butadienes with a variety of dienophiles
and substitution reactions between these allylsilane-containing adducts and aldehyde or acid
chloride electrophiles have been combined into “tandem sequential reactions.” These tandem
sequences proceed with equal or greater yield (50-80%) than the reactions performed separately
with no decrease in regio- or stereoselectivity. The sequence produces cyclic compounds with three
or four stereogenic centers with good to excellent diastereoselectivity from three simple, noncyclic,
and achiral reaction partners. Unprecedented levels of diastereoselectivity (de >93%) have been
achieved in allylic substitution reactions involving substituted [(trialkylsilyl)methyl]cyclohexene
derivatives with aldehyde electrophiles. During the course of these studies, protodesilylation of
allylsilanes has been investigated in detail, and a cocatalyst system of TiCl4 and Me2AlCl has been
developed that has eliminated silicon loss in the substitution reactions studied. Lewis acid-promoted
enolization of ester and ketone substrates with chiral centers adjacent to the carbonyl moiety has
been studied also. It has been shown that this event in our studies occurs primarily during catalyst
quench. This isomerization is prevented by quenching the catalyst with a Lewis base, such as
methanol or triethylamine, prior to aqueous workup.
In tr od u ction
reaction has been judged complete. Further, the first
reaction is required to provide the functionality necessary
to allow the second reaction to take place.2e Many
tandem reactions in the literature are of the same
mechanism. Examples include sequential Michael ad-
ditions,4 pericyclic rearrangements,5 and transition-metal-
catalyzed coupling procedures6-8 to name just a few. The
present study is directed at expanding the scope of “key-
step” tandem reactions that can be used in synthesis.
Combining fundamentally different reactions can gener-
The preparation of complex molecules from simple,
cheap, and readily available ones is a strong starting
point to making a synthesis practical. Further, when
such starting materials are incorporated into an approach
that brings about a rapid increase in molecular complex-
ity, the approach is also regarded as efficient.1 Tandem
reactions represent a powerful approach to addressing
overall synthetic efficiency.2 One-pot synthetic sequences
reduce the time, waste, and cost associated with synthetic
chemistry.
(4) (a) Barco, A.; Benetti, S.; Casolari, A.; Pollini, G. P.; Spalluto,
G. Tetrahedron Lett. 1990, 31, 3039-3042. (b) Rosenmund, P.;
Hosseini-Merescht, M. Tetrahedron Lett. 1990, 31, 647- 650. (c)
Schinzer, D.; Kalesse, M. Synlett 1989, 34.
In a recent report, we disclosed our first results in the
area of “tandem reaction sequences” that combine trans-
formations of different fundamental mechanism using
allylsilane substrates.3 Such sequences require the ad-
dition of a second (or third) compound after the first
(5) (a) Burke, S. D.; Buchanan, J . L. Rovin, J . D. Tetrahedron Lett.
1991, 32, 3961-3964. (b) Paterson, I.; Hulme, A. N.; Wallace, D. J .
Tetrahedron Lett. 1991, 32, 7601-7604. (c) J anardhanam, S.; Devan,
B.; Rajagopalan, K. Tetrahedron Lett. 1993, 34, 6761-6764.
(6) For tandem transition-metal-catalyzed cross-coupling procedures,
see: (a) Reiser, O.; Weber, M.; de Meijere, A. Angew. Chem., Int. Ed.
Engl. 1989, 59, 74-79. (b) Albrecht, K.; Reiser, O.; Weber, M.; de
Meijere, A. Synlett 1992, 521-523. (c) Grigg, R.; Sridharan, V.
Tetrahedron Lett. 1992, 33, 7965-7968. (d) Grigg, R.; Kennewell, P.;
Teasdale, A.; Sridharan, V. Tetrahedron Lett. 1993, 34, 153-156. (e)
Harris, G. D., J r.; Herr, R. J .; Weinreb, S. M. J . Org. Chem. 1993, 58,
5452-5464. (f) Brown, D.; Grigg, R.; Sridharan, V.; Tambyrajah, V.
Tetrahedron Lett. 1995, 36, 8137-8140.
(7) For tandem transition-metal-catalyzed cross-coupling procedures
involving carbonylation, see: (a) Crisp, G. T.; Scott, W. J .; Stille, J . K.
J . Am. Chem. Soc. 1984, 106, 7500-7506. (b) Chiba, K.; Mori, M.; Ban,
Y. Tetrahedron 1985, 41, 387-392. (c) Oppolzer, W.; Bienayme, H.;
Genevois-Borella, A. J . Am. Chem. Soc. 1991, 113, 9660-9661. (d)
Crisp, G. T.; Meyer, A. G. J . Org. Chem. 1992, 57, 6972-6975. (e)
Anderson, P. G.; Aranyos, A. Tetrahedron Lett. 1994, 35, 4441-4444.
(f) Grigg, R.; Sridharan, V.; Suganthan, S.; Bridge, A. W. Tetrahedron
1995, 51, 295-306.
* To whom correspondence should be addressed. After August 15,
1997, correspondence should be directed to the following. Phone: (416)
736-2100, ext. 33689. Fax: (416) 736-5936. E-mail: organ@yorku.ca.
† Indiana University-Purdue University at Indianapolis.
‡ Indiana University Bloomington.
X Abstract published in Advance ACS Abstracts, J uly 15, 1997.
(1) Wender, P. A.; Miller, B. L. In Organic Synthesis: Theory and
Applications; Hudlicky, T., Ed.; J AI Press: Greenwich, CT, 1993; Vol.
2, pp 27-66.
(2) For recent reviews on tandem chemistry, see: (a) Ho, T.-L.
Tandem Organic Reactions; Wiley: New York, 1992. (b) Bunce, R. A.
Tetrahedron 1995, 51, 13103-13159. (c) Parsons, P. J .; Penkett, C.
S.; Shell, A. J . Chem. Rev. 1995, 95, 195-206. (d) Tietze, L. F. Chem.
Rev. 1995, 95, 115-136. (e) Denmark, S. E.; Thorarensen, A. Chem.
Rev. 1996, 96, 137-165.
(3) Organ, M. G.; Winkle, D. D. J . Org. Chem. 1997, 62, 1881-1885.
S0022-3263(97)00821-9 CCC: $14.00 © 1997 American Chemical Society

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5255
Sch em e 1
ate products with unique structure and functionality
patterns from a tandem procedure that are not readily
obtainable from sequences involving reactions of the
same mechanism.3,4,9 Therefore, these reactions are
intended to complement tandem reactions of the same
mechanism.
The tandem reaction strategy described in this report
focuses on functional groups that are capable of support-
ing a wide range of chemical transformations. Reaction
conditions are then developed to allow such transforma-
tions to be conducted on a substrate in a predictable
sequence. Allylsilanes are well suited for this purpose
for a number of reasons.10 They have been used widely
in reactions of radical,11 cationic,12 and anionic mecha-
nism.13 Silanes are among the most stable compounds
that contain a carbon-metal bond, especially allylmetal
moieties.14 Therefore, silane moieties can be incorporated
into the structure of starting materials and carried
through a number of steps before they are activated
selectively.15 This is critical to the success of this
chemistry. It is a long-term goal of our program to
investigate the combination of any of the above-men-
tioned mechanisms into sequential transformations to
produce complex compounds from relatively simple pre-
cursors in one operation.
In the short term, we have developed a tandem
pericyclic/ionic reaction sequence involving 2-[(trialkyl-
silyl)methyl]-1,3-butadienes and a variety of dienophile
and electrophile partners, all of which can be reacted
selectively in a stepwise fashion using the same or
different Lewis acid catalyst. The retrosynthetic se-
quence in Scheme 1 demonstrates how this methodology
could be used to prepare the core of sonomolide A (1), a
compound of interest to us. Recently isolated and
characterized, sonomolide A has demonstrated pro-
nounced antifungal activity.16
This report also discusses deleterious side reactions
that occur with allylsilanes in allylic substitution reac-
tions promoted by TiCl4. Despite the aforementioned
stability of the carbon-silicon bond and the tunable
reactivity of allylsilanes, Brønsted acid will protonate the
olefin, and this is accompanied by loss of silicon.17
Protodesilylation can also occur as a side reaction in
Lewis acid-catalyzed substitutions with carbon electro-
philes.18 In the course of our studies, we encountered
this problem and developed a catalyst system that
minimizes this side reaction.
Lewis acid-promoted enolization of ketones and esters
is an unfortunate process that can lead to epimerization
of these sites. This results in a significant reduction of
diastereomeric purity for compounds possessing multiple
chiral centers. Further, if the solution containing the
compound is optically enhanced, loss in enantiomeric
purity can also result. This issue is also addressed in
this study.
(8) For tandem transition-metal-catalyzed cycloisomerization pro-
cedures, see: (a) Oppolzer, W.; DeVita, R. J . J . Org. Chem. 1991, 56,
6256-6257. (b) Meyer, F. E.; Brandenburg, J .; Parsons, P. J .; de
Meijere, A. J . Chem. Soc., Chem. Commun. 1992, 390-392. (c) Trost,
B. M.; Shi, Y. J . Am. Chem. Soc. 1992, 114, 791-792. (d) Trost, B. M.;
Dumas, J . J . Am. Chem. Soc. 1992, 114, 1924-1925. (e) Trost, B. M.;
Shi, Y. J . Am. Chem. Soc. 1992, 114, 791-792. (f) Trost, B. M.; Shi, Y.
J . Am. Chem. Soc. 1993, 115, 12491-12509.
(9) For examples of tandem reactions where each step in the
sequence is of different mechanism, see: (a) Tietze, L.-F.; von Kied-
rowski, G.; Berger, B. Angew. Chem., Int. Ed. Engl. 1982, 21, 221-
224. (b) Tietze, L. F.; von Kiedrowski, G.; Berger, B. Synthesis 1982,
683-687. (c) Xu, Y.-C.; Cantin, M.; Deslongchamps, P. Can. J . Chem.
1990, 68, 2137-2143. (d) Davies, H. W.; Saikali, E.; Young, W. B. J .
Org. Chem. 1991, 56, 5696-5700. Also see ref 2.
(10) For general reviews concerning the use of allylsilanes in organic
synthesis, see: (a) Fleming, I. Chem. Soc. Rev. 1981, 10, 83-111. (b)
Parnes, Z. N.; Bolestova, G. I. Synthesis 1984, 991-1008. (c) Blumen-
kopf, T. A.; Overman, L. E. Chem. Rev. 1986, 86, 857-873. (d) Schinzer,
D. Synthesis 1988, 263-273. (e) Fleming, I.; Dunogues, J .; Smithers,
R. Org. React. 1989, 31, 57-575. (f) Chan, T. H.; Wang, D.; Pellon, P.;
Lamothe, S.; Wie, Z. Y.; Li, L. H.; Chen, L. M. Enantioselective
synthesis using silicon compounds. Frontiers of Organosilicon Chem-
istry; The Royal Society of Chemistry: London, 1991; p 344-355. (g)
Colvin, E. W.; Loreto, M. A.; Montieth, M.; Tommasini, I. Frontiers of
Organosilicon Chemistry; The Royal Society of Chemistry, London,
1991; p 356. (h) Langkopf, E.; Schinzer, D. Chem. Rev. 1995, 95, 1375-
1408.
(11) For discussion related to the use of silyl-stabilized radicals in
synthesis, see: (a) Auner, N.; Walsh, R.; Westrup, J . J . Chem. Soc.,
Chem. Commun. 1986, 207-208. (b) Wilt, J . W.; Lusztyk, J .; Peeran,
M.; Ingold, K. U. J . Am. Chem. Soc. 1988, 110, 281-287.
(12) For discussions regarding the â-effect, see: (a) Hanstein, W.;
Berwin, H. J .; Traylor, T. G. J . Am. Chem. Soc. 1970, 92, 829-836.
(b) Eaborn, C. J . Chem. Soc., Chem. Commun. 1972, 1255. (c) Fleming,
I. Frontier Orbitals and Organic Chemical Reactions; Wiley: London,
1976; p 81. (d) Wierschke, S. G.; Chandrasekhar, J .; J orgensen, W. L.
J . Am. Chem. Soc. 1985, 107, 1496-1500. (e) Lambert, J . B. Tetrahe-
dron 1990, 46, 2677-2689.
(13) For uses of allylsilanes in anion-mediated transformations,
see: (a) Yamamoto, Y.; Saito, Y.; Naruyam, K. J . Org. Chem. 1980,
45, 195-196. (b) Tsai, D. J . S.; Matteson, D. S. Tetrahedron Lett. 1981,
22, 2751-2752. (c) Tamao, K.; Nakajo, E.; Ito, Y. Tetrahedron 1988,
44, 3997-4007. (d) Chan, T. H.; Pellon, P. J . Am. Chem. Soc. 1989,
111, 8737- 8738. (e) Chan, T. H.; Horvath, R. F. J . Org. Chem. 1989,
54, 317-327. (f) Chan, T. H.; Wang, D. Chem. Rev. 1995, 95, 1279-
1292.
Resu lts a n d Discu ssion
Diels-Ald er Rea ction . The first step in the general
sequence investigated was a [4 + 2] cycloaddition be-
(15) Au-Yeung, B. W.; Fleming, I. J . Chem. Soc., Chem. Commun.
1977, 79, 81.
(16) Morris, S. A.; Curotto, J . E.; Zink, D. L.; Dreikorn, S.; J enkins,
R.; Bills, G. F.; Thompson, J . R.; Vincente, F.; Basilio, A.; Liesch, J .
M.; Schwartz, R. E. Tetrahedron Lett. 1995, 36, 9101-9104.
(17) For reports dealing with acid-catalyzed protodesilylation, see:
(a) Coughlin, D.J .; Salomon, R. G. J . Org. Chem. 1979, 44, 3784-3790.
(b) Pennanen, S. I. Synth. Commun. 1980, 10, 373-379. (c) Hosomi,
A.; Iguchi, H.; Sasaki, J .-I.; Sakurai, H. Tetrahedron Lett. 1982, 551-
554. (d) Ireland, R. E.; Varney, M. D. J . Am. Chem. Soc. 1984, 106,
3668-3670. (e) Shibasaki, M.; Fukasawa, E.; Ikegami, S. Tetrahedron
Lett. 1983, 24, 3497-3500. (f) Corey, E. J .; Letavic, M. A. J . Am. Chem.
Soc. 1994, 117, 9616-9617.
(14) (a) Reynolds, W. F.; Hamer, G. K.; Bassindale, A. R. J . Chem.
Soc., Perkin Trans. 2 1977, 971-974. (b) Walsh, R. Acc. Chem. Res.
1981, 14, 246-252. (c) Giordan, J . C. J . Am. Chem. Soc. 1983, 105,
6544-6546.
(18) (a) Trost, B. M.; Self, C. R. J . Am. Chem. Soc. 1983, 105, 5942-
5944. (b) Majetich, G.; Casares, A. M.; Chapman, D.; Behnke, M. J .
Org. Chem. 1986, 51, 1745-1753. (c) Majetich, G.; Desmond, R. W.;
Soria, J . J . J . Org. Chem. 1986, 51, 1753-1769.

5256 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
Sch em e 2
Ta ble 1. Rea ction Con d ition s for Diels-Ald er Cycloa d d ition betw een 2-[(Tr ia lk ylsilyl)m eth yl]-1,3-bu ta d ien es 5 a n d
Va r iou s Dien op h iles 6 (See Sch em e 2 for Str u ctu r es)
entry
diene
dienophile
R
R1
R2
catalysta
T (°C)
solvent
product no.
yieldb
1
2
3
4
5
6
7
8
9
5a
5a
5a
5a
5a
5b
5b
5b
5b
5b
5b
5b
6a
6a
6a
6a
6a
6a
6b
6c
6d
6d
6ef
6fg
-CH3
-CH3
-CH3
-CH3
-CH3
-Ph
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-COCH3
H
H
H
H
H
H
H
AlCl3
40
40
40
40
40
40
rt
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
toluene
CH2Cl2
CH2Cl2
CH2Cl2
7a
7a
7a
7a
7a
7b
7c
7d
7e
7e
7f
38c
88
52
15
38d
84
97
95
84
90
59
89
Me2AlCl
EtAlCl2
BF3‚OEt2
TiCl4
Me2AlCl
Me2AlCl
Me2AlCl
no catalyst
Me2AlCle
Me2AlCl
no catalyst
-Ph
-Ph
-COCH2CH3
-CO2CH3
-CO2CH3
-CH3
-CO2CH3
-CO2CH3
rt
-Ph
100
40
40
rt
10
11
12
-Ph
-Ph
-(C(O)CH2CH2CH2)-
-(C(O)OC(O))-
-Ph
7g
a
b
0.2 equiv of catalyst was used relative to 5 unless stated otherwise. All reported yields are based on purified material following
silica gel chromatography. c Thirty-five percent of the protodesilylated side product 8 was also formed. Thirty-six percent of the
d
protodesilylated side product 8 was also formed. e One equiv of Lewis acid was required. f Dienophile 6e was 2-cyclohexen-1-one. g Dienophile
6f was maleic anhydride.
tween 2-[(trialkylsilyl)methyl]-1,3-butadienes and a va-
riety of dienophiles (see Scheme 2 and Table 1). Such
reactions have been carried out using 2-[(trimethylsilyl)-
methyl]-1,3-butadiene (5a )3 with R,â-unsaturated carbo-
nyl and ester substrates, both with and without cataly-
sis.19 Diene 5a and methyl acrylate were used to
optimize the reaction conditions. Of all the catalysts
tried, Me2AlCl provided the best yield (Table 1, entry 2).
The principal reason for lower yield with other catalysts
is protodesilylation of 7a producing 8. It was discovered
subsequently that adducts derived from 2-[(dimethyl-
phenylsilyl)methyl]-1,3-butadiene (5b)3 were less sus-
ceptible to protodesilylation, so all subsequent reactions
were done with this diene. Addition of 5b to a variety of
dienophiles under Me2AlCl catalysis provided all adducts
in very good yield (Table 1, entries 6, 7, 8 and 10). The
para adduct was isolated in all cases with unsymmetrical
dienophiles when the reaction was performed using Lewis
acid catalysis.
These selective and high-yielding results are critical
for the success of the tandem sequence. It is imperative
that the first reaction goes to completion and that it
proceeds regioselectively and stereoselectively where
applicable. The number of potential isomers produced
in subsequent step(s) doubles for each additional adduct
or suitably reactive byproduct generated in the first step.
Obviously, this would have a negative effect on yield, but
product isolation from a complex mixture of isomers could
make such a tandem approach impractical.
When the reaction indicated in entry 8 (Table 1) (i.e.,
product 7d ) was worked up using the standard conditions
of saturated NaHCO3
quench followed by ether extrac-
tion, a 2.4:1 trans:cis ratio of adducts was observed. The
1H NMR spectrum of the reaction mixture at the comple-
tion of the Diels-Alder reaction revealed all trans isomer
prior to workup. This indicated that isomerization
occurred as a result of the heterogeneous conditions of
the aqueous quench. To avoid this, a nonaqueous quench
was performed by adding neutral alumina directly to the
reaction mixture followed by solvent removal in vacuo.
The crude product/alumina mixture was then loaded to
the top of a prepacked silica gel column and eluted off,
providing the adduct, now with only a trace of the cis
isomer (21:1, trans:cis). This epimerization is discussed
further in the next section.
Electr op h ilic Su bstitu tion Rea ction s. Cycload-
ducts 7 were then reacted in a series of electrophilic
substitutions with a variety of aldehydes and acid
chlorides. In doing so, a number of Lewis acids and
reaction conditions have been probed to optimize these
substitutions.
i. Ad d ition s to Ald eh yd es. The first allylic substi-
tution reactions were attempted with adduct 7a and
propionaldehyde (9a ) to optimize reaction conditions, i.e.,
Lewis acid, solvent, and temperature (see Scheme 3 and
Table 2). Although Me2AlCl catalysis worked very nicely
for the Diels-Alder study, it failed to promote the
substitution reaction, and starting materials were recov-
ered quantitatively. Stronger Lewis acids such as AlCl3
and TiCl4 promoted conversion of 7a and 9a to 10a and
11a (see method A, Table 2). However, this was ac-
companied by substantial protodesilylation of 7a leading
to the formation of 8a . It was shown subsequently that
(19) (a) Hosomi, A.; Saito, M.; Sakurai, H. Tetrahedron Lett. 1979,
429-432. (b) Sakurai, H.; Hosomi, A.; Saito, M.; Sasaki, K.; Iguchi,
H.; Sasaki, J -i.; Araki, Y. Tetrahedron 1983, 39, 883-894. (c) Hosomi,
A.; Sakata, Y.; Sakurai, H. Tetrahedron Lett. 1985, 26, 5175-5178.
(d) Hosomi, A.; Otaka, K.; Sakurai, H. Tetrahedron Lett. 1986, 27,
2881-2884.

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5257
Sch em e 3
Ta ble 2. Su bstitu tion Rea ction Con d ition s w ith Cyclic Allylsila n e Diels-Ald er Ad d u cts 7 a n d Ald eh yd e Electr op h iles 9
(See Sch em e 3 for Str u ctu r es)
starting
adduct
product
numbers
yieldd (%)/
10:11
entry
aldehyde
R
R1
R2
R3
catalysta
AlCl3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
7a
7a
7a
7a
7b
7b
7b
7b
7b
7b
7b
7c
7c
7c
7c
7c
7c
7d
7d
7e
7f
9a
9a
9a
9a
9a
9a
9b
9b
9b
9c
9c
9a
9a
9c
9c
9b
9b
9a
9b
9a
9a
9a
-CH3
-CH3
-CH3
-CH3
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-COCH3
-COCH3
-COCH3
-COCH3
-COCH3
-COCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
-CH2CH3
-CH2CH3
-CH2CH3
-CH2CH3
-CH2CH3
-CH2CH3
-CH(CH3)2
-CH(CH3)2
-CH(CH3)2
-CH2CH2Ph
-CH2CH2Ph
-CH2CH3
-CH2CH3
-CH2CH2Ph
-CH2CH2Ph
-CH(CH3)2
-CH(CH3)2
-CH2CH3
-CH(CH3)2
10a a n d 11a
10a a n d 11a
10a a n d 11a
10a a n d 11a
10a a n d 11a
10a a n d 11a
10b a n d 11 b
10b a n d 11 b
10b a n d 11 b
10c a n d 11 c
10c a n d 11 c
10d a n d 11 d
10d a n d 11 d
10e a n d 11 e
10e a n d 11 e
10f a n d 11 f
10f a n d 11 f
10g a n d 11g
10h
40e/4:1
no reaction
trace
60e/4.4:1
61e/4.3:1
65/4.9:1
60e/1.1:1
70/1.3:1
82f/2:1
Me2AlCl
BF3‚OEt2
b
TiCl4
TiCl4
b
c
Me2AlCl/TiCl4
b
TiCl4
c
c
Me2AlCl/TiCl4
Me2AlCl/TiCl4
b
TiCl4
54e/5.8:1
76/7.4:1
55e/3:1
72/3.6:1
53e/4.4:1
62/4.4:1
54e/2.3:1
69/1.8:1
65/4.6:1
53/trans only
c
c
c
Me2AlCl/TiCl4
b
TiCl4
Me2AlCl/TiCl4
b
TiCl4
Me2AlCl/TiCl4
b
TiCl4
c
c
c
c
c
c
Me2AlCl/TiCl4
Me2AlCl/TiCl4
Me2AlCl/TiCl4
Me2AlCl/TiCl4
Me2AlCl/TiCl4
Me2AlCl/TiCl4
-COCH2CH3 CH3
-COCH2CH3 CH3
-CO2CH3
-CO2CH3 -CH2CH3
NRg
NR
NR
-(C(O)CH2CH2CH2)-
-(C(O)OC(O))-
-CH2CH3
-CH2CH3
7g
a
b
One equiv of active catalyst was used in these reactions. Method A: a solution of allylsilane 7 and aldehyde 9 in CH2Cl2 was cooled
to -60 °C and TiCl4 added dropwise. c Method B: a solution of dimethylaluminum chloride (0.2 equiv) and TiCl4 (1.0 equiv) in CH2Cl2
d
was cooled to -60 °C and then added to a second flask containing 7, 9, and 0.2 equiv of Me2AlCl, also at -60 °C. All reported yields are
based on purified material following silica gel chromatography. e Up to 20% of the desilylated compound 8 was obtained in addition to the
substitution products 10 and 11. f 2.0 equiv of the aldehyde was used. No substitution product was produced in this reaction, and most
g
of the starting allylsilane was recovered.
Sch em e 4
when this reaction was performed using TiCl4 catalysis
in the presence of Me2AlCl (see method B, Table 2),
protodesilylation was eliminated and yield was enhanced
considerably (compare entries 5 vs 6, 7 vs 8, 10 vs 11, 12
vs 13, and 14 vs 15 in Table 2). This effect is discussed
below. All substitutions involving aldehyde electrophiles
proceeded with allyl inversion. Of the four possible
diastereomers (eight for addition to 7d ) that can arise
from this substitution, only two were observed, i.e., 10
and 11 (one in the case of 10h ).
Str u ctu r a l Elu cid a tion of Su bstitu tion P r od u cts
w ith Ald eh yd e Su bstr a tes. It has been confirmed that
the substituents at C1 and C3 of 10d and 11d are cis
and trans as indicated. Independent oxidation of these
two compounds with PCC provided different products,
i.e., 12 and 13 (Scheme 4). This proved that the two
isomers are cis and trans. However, at the time, it
provided no information to distinguish between them or
any indication as to the relative stereochemistry at the
carbinol center.
The structure of the major isomer from the reaction of
7b and isobutyraldehyde, i.e., 10f, has been unambigu-
ously established by single-crystal X-ray analysis (Figure
1). The structure demonstrates that the alcohol-contain-
ing side chain occupies the axial position and is trans to
the acetyl moiety that is equatorial. This conformation,
albeit in the solid state, proved to be important in relating
the connectivity of this structure to the structure of the

5258 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
F igu r e 3. Proposed conformations of substitution products
to account for differences between the 1H NMR chemical shifts
of the exocyclic methylene protons.
signals for these two protons might be expected to be
separated by a larger chemical shift than those of 10.
The 1H NMR spectra of 10f and 11f clearly illustrate this
difference. Similar structural assignments have been
suggested for cyclohexane compounds possessing an
exocyclic methylene adjacent to a (1-oxoalkane) side
chain.20
F igu r e 1. Single crystal X-ray structure of 10f.
Ster eoselectivity in Su bstitu tion Rea ction w ith
Ald eh yd e Su bstr a tes. That 7d was epimerized by Me2-
AlCl in the Diels-Alder study raised the possibility that
TiCl4 might be causing some epimerization of the prod-
ucts from the substitution reaction. This would occur by
enolization of the position on the ring next to the ketone
or ester functionalities. Thus, “observed stereoselectiv-
ity” in the substitution reaction would be lowered. To
test this possibility, trans isomer 10a was treated with
1 equiv of TiCl4 in CH2Cl2 and worked up as if it were a
regular substitution reaction using saturated NaHCO3
to quench the catalyst (Scheme 5).
1
The H NMR spectrum of this mixture revealed that
20% of 10a had been converted to 11a . The structure of
11a drawn in Scheme 5 is the enantiomer of that drawn
for 11a in Scheme 3, and they represent the same
compound in this study (i.e., the study is on racemic
material). Coincidentally, this was the same ratio ob-
tained when the substitution reaction was performed and
worked up in the usual way. This result provided two
separate pieces of information. First, it confirmed the
possibility that at least some of 11 could have been
formed by Lewis acid-catalyzed isomerization of 10.
Second, it confirmed the structure of the minor isomer
in this study. Conversion of 10a to 11a , rather than some
other diastereomer, can only happen if the carbinol
center, in addition to C3, has the same relative stereo-
chemistry in the major and minor isomers.
With this result in hand, a substitution reaction was
performed with 14 (Scheme 6). This compound was
similar to 7c except that it lacks the ketone moiety on
the two-carbon side chain. Any isomerization occurring
as a direct result of the carbonyl moiety would not have
been possible with any substitution product derived from
14. The ethyl group would serve the same role as the
acetyl group as the conformational anchor of the cyclo-
hexene ring. Thus, any axial/equatorial selectivity seen
in a substitution reaction should be similar to that seen
for the ketone parent, providing coordination to that
carbonyl group was not involved in any observed stereo-
selectivity.
F igu r e 2. 1H NMR spectra of trans adduct 10f (A) and cis
adduct 11f (B).
minor isomer 11f and other products in this sequence by
NMR spectroscopy.
Examination of the 1H NMR spectra of products 10 and
11 reveal that the signal for each exocyclic proton is a
singlet for both isomers. Further, the signals for these
protons overlap for one of the isomers, while they are
separated by at least 0.2 ppm in the case of the second
isomer. These values are characteristic and diagnostic
1
for all adducts in this study. The H NMR spectrum of
10f (confirmed by X-ray analysis) and the minor isomer
11f are shown in Figure 2. From Figure 2A, we extrapo-
lated the structure of the major isomer for all electrophilic
substitution reactions, i.e., they are all trans products
resulting from an axial transition state.
The chemical shift values of the vinyl protons can be
understood by examining the conformation of the major
and minor products (Figure 3). The trans adducts 10
place the alcohol sidechain axial, which reduces the A1,3
-
strain between this group and the exocyclic methylene
(see Figure 1). The result is that both Ha and Hb are in
remotely similar environments resulting in closer chemi-
cal shifts for these protons. The cis adducts 11 place both
groups equatorial, which results in an interaction be-
tween the alcohol side chain and Hb. Therefore, the
Substitution under identical conditions as for 7c with
propionaldehyde electrophile (saturated NaHCO3 work-
up) yielded the axial addition product 15 almost exclu-
sively (93% diastereomeric excess (de)). The same addi-
tion with 7c provided a ratio of 3.6:1 of 10d :11d (56%
de). Although it cannot be said with certainty that the
initial selectivity observed in substitutions involving 7
were all equal to that observed with 14, it indicates that
(20) Pillot, J .-P.; De´le´ris, G.; Dunogue`s, J .; Calas, R. J . Org. Chem.
1979, 44, 3397-3400.

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5259
F igu r e 4. Proposed transition state structures leading to the products of axial, i.e., trans, and equatorial, i.e., cis addition.
Sch em e 5
Sch em e 6
Sch em e 7
1
such substitutions warrant further investigation. Com-
pounds 10 may have been the principal initial products
formed in the substitution step, and subsequent isomer-
ization lowered the observed selectivity.
The H NMR spectrum of the crude mixture revealed
that the equatorial substitution product 11b was barely
detectable (93% de). The ratio of axial to equatorial
products was now 28:1 in stark contrast to the 1:1
mixture obtained previously. A similar result was ob-
served when 7b was reacted with propionaldehyde (9a )
(Table 2, entries 5 and 6). In the absence of the methanol
quench, the observed selectivity for 10a :11a was ap-
proximately 4.5:1 (63% de). This ratio with the methanol
quench improved to 9:1 (80% de).
In order to confirm that enolization of the substitution
products (10) was indeed linked to observed stereoselec-
tivity, it had to be determined if enolization occurred
during the reaction at all or if it happened exclusively
upon quench. In light of the epimerization observed with
7d in the Diels-Alder study, attention was focused on
the nature of the quench of the substitution reaction. The
substitution reaction involving 7b and isobutyraldehyde
(9b) using either method A or B produced a ratio of axial
to equatorial substitution products (i.e., 10b:11b) of
approximately 1:1 following workup and purification.
This reaction was chosen for these experiments because
any change in diastereoselectivity in this reaction would
be readily apparent as there was no appreciable selectiv-
ity displayed previously. Despite the success in control-
ling enolization using neutral alumina in the Diels-Alder
reaction (7d ), quenching TiCl4 with triethylamine or
methanol was viewed to be easier to carry out and more
likely to become a general protocol if successful.
These results confirm suspicions that enolization oc-
curs to a large extent during aqueous workup. Although
these experiments do not discount that enolization occurs
during the substitution reaction itself, they indicate that
it is a minor event if it does occur at all. Further, the
active form of the titanium catalyst that promotes this
isomerization is presumably TiCl3OR. The excess metha-
nol is thought to form Ti(OCH3)4, which is then too weak
of a Lewis acid to promote enolization.
It is highly likely that the majority of electrophilic
substitution reactions performed in this study using
aldehyde electrophiles are highly stereoselective for the
axial product. To our knowledge, the results reported
here are the most highly selective electrophilic substitu-
tions reported using cyclic allylsilanes of this type.
Further, these results are important for any researcher
working in the area of Lewis acid-promoted transforma-
tions that employ such sensitive substratessparticularly
The substitution reaction was performed following
method B, and when the reaction was judged to be
complete by TLC analysis, 5 equiv of dry methanol was
added at -5 °C and the mixture stirred for 20 min (see
Scheme 7). Then, the mixture was diluted with ether
and washed with saturated NaHCO3 in the usual way.

5260 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
Sch em e 8
Ta ble 3. Su bstitu tion Rea ction Con d ition s w ith Cyclic Allylsila n e Diels-Ald er Ad d u cts 7 a n d Acid Ch lor id e
Electr op h iles 17 (See Sch em e 8 for Str u ctu r es)
yieldg (%)
18:19:20
entry adduct
acid chloride
R1
R2
R4
catalysta
Me2AlClb
product no.
1
2
3
4
5
6
7
8
9
7c
7c
7c
7c
7c
7c
7b
7b
7b
7d
17a
17a
17a
17a
17a
17b
17b
17b
17c
17c
-COCH3
-COCH3
-COCH3
-COCH3
-COCH3
-COCH3
-CO2CH3
-CO2CH3
-CO2CH3
H
H
H
H
H
H
H
H
H
-CH3
-CH3
-CH3
-CH3
-CH3
18a , 19a , a n d 20a
18a , 19a , a n d 20a
18a , 19a , a n d 20a
18a , 19a , a n d 20a
18a , 19a , a n d 20a
18b, 19b, a n d 20b
18c, 19c, a n d 20c
18c, 19c, a n d 20c
18d , 19d , a n d 20d
18e, 19e, a n d 20e
no reaction
24h/1.5:1:0
78/2.3:1.6:1
67/4.0:3.0:1
85/2:1:7.6
78/1.6:0.8:1
96/10.5:10:1
87/2:1.7:1
68/1:1:1
c
d
e
Me2AlCl/TiCl4
Me2AlCl/AlCl3
Me2AlCl/AlCl3
f
AlCl3
d
d
-CHdCH-Ph (E) Me2AlCl/AlCl3
-CHdCH-Ph (E) Me2AlCl/AlCl3
f
-CHdCH-Ph (E) AlCl3
d
-Ph
Me2AlCl/AlCl3
Me2AlCl / AlCl3
d
10
-COCH2CH3 -CH3 -Ph
78/1:1.3:0
a
b
In all cases, the silane was the limiting reagent and 1.5 equiv of the acid chloride was used. Two equiv of catalyst was used. c Two
equiv of TiCl4 was used with 0.2 equiv of Me2AlCl. Two equiv of AlCl3 were used with 0.2 equiv of Me2AlCl. e Two equiv of AlCl3 was
d
used with 1.2 equiv of Me2AlCl. f Two equiv of catalyst was used. All reported yields are based on purified material following silica gel
g
h
chromatography. In addition to these products, 10% of 18 and 19 isomerized from the exocyclic methylene to the conjugated enone.
Sch em e 9
those in the area of asymmetric synthesis (e.g., Diels-
Alder cycloaddition).
transition states would result in opposing stereochem-
istry at the carbinol site in both the major and minor
isomers.
Tr an sition States for Su bstitu tion Reaction . Tran-
sition states that would lead to the axial (i.e., 10) and
equatorial (i.e., 11) substitution products obtained in this
study are shown in Figure 4.21 Although other transition
states are possible, we do not believe that they are
operative. We propose that the transition states leading
the major (axial) product and the minor (equatorial)
product, if in fact it is operative at all, both proceed via
the antiperiplanar arrangements indicated. This places
the bulky R-groups away from the ring in either case.22
Further, we believe that reaction takes place mainly via
the axial route shown (vide supra). If we assume that
the bulky R-group never sits on top of the cyclohexene
ring in any favorable transition state, favorable synclinal
ii. Ad d it ion s t o Acid Ch lor id es. Electrophilic
substitution with 7 and a range of acid chlorides (17)
provided adducts 18-20 in good yield (see Scheme 8 and
Table 3). The best results here were achieved using AlCl3
catalysis (2.0 equiv) with 0.20 equiv of Me2AlCl cocata-
lyst. In almost all cases, the ratio of cis:trans substitution
adducts (18:19) was approximately 1:1. Similar stereo-
selectivity was observed by Pillot and co-workers in
analogous substitutions with acid chlorides.20 We believe
that isomerization of these adducts, now containing two
enolizable sites, is responsible to a large degree for the
poor observed stereoselectivity. Elucidation of the struc-
ture of the products in this section was done in a manner
similar to that discussed for the aldehyde substitution
products (vide supra). The chemical shifts of exocyclic
protons in the 1H NMR spectra once again serve as a
diagnostic marker for cis and trans substitution products.
(21) For early discussions concerning the origin of axial vs equatorial
selectivity in additions to cyclohexene derivatives, see: (a) Allinger,
N. L.; Riew, C. K. Tetrahedron Lett. 1966, 1269-1272. (b) House, H.
O.; Terfertiller, B. A.; Olmstead, H. D. J . Org. Chem. 1968, 33, 935-
942. (c) Chamberlain, P.; Witham, G. H. J . Chem. Soc, Perkin Trans.
2 1972, 130-135.
(22) For discussion concerning synclinal and antiperiplanar transi-
tion states in allylsilane substitution reactions, see: (a) Hayashi, H.;
Ito, H.; Kumada, M. Tetrahedron Lett. 1982, 23, 4605-4606. (b)
Hayashi, H.; Konishi, M.; Kumada, M. J . Org. Chem. 1983, 48, 281-
282. (c) Denmark, S. E.; Weber, E. J . Helv. Chim. Acta 1983, 66, 1655-
1660. (d) Denmark, S. E.; Almstead, N. G. J . Org. Chem. 1994, 59,
5130-5132. (e) Denmark, S. E.; Hosoi, S. J . Org. Chem. 1994, 59,
5133-5135.
Ta n d em Diels-Ald er /Electr op h ilic Su bstitu tion
Rea ction s. i. Step Tw o:Ald eh yd e Ad d ition s. For the
majority of tandem reactions, the Diels-Alder reaction
was performed using the optimal conditions of 0.2 equiv
of Me2AlCl in CH2Cl2 at either rt or at reflux (see Scheme
9 and Table 4). When cycloaddition was judged complete
by TLC, the solution was cooled to -60 °C, and 1.0 equiv
of the aldehyde was added followed by dropwise addition

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5261
Ta ble 4. Ta n d em Diels-Ald er Cycloa d d ition /Ald eh yd e Su bstitu tion Sequ en ce (See Sch em e 9 for Str u ctu r es)
Diels-Alder
T (°C)
second
step
yieldb/
10:11
entry
dienophilea
R1
R2
aldehyde
R3
product no.
1
2
3
4
5
6
7
8
9
6a
6a
6a
6a
6a
6b
6b
6b
6b
6c
6c
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-CO2CH3
-COCH3
H
H
H
H
H
H
H
H
H
40
40
40
40
40
rt
rt
rt
rt
rt
9a
9b
9b
9c
9c
9a
9c
9b
9b
9a
9b
-CH2CH3
10a a n d 11a
10b a n d 11b
10b a n d 11b
10c a n d 11c
10c a n d 11c
10d a n d 11d
10e a n d 11e
10f a n d 11f
10f a n d 11f
10g
73/4.3:1
-CH(CH3)2
-CH(CH3)2
-CH2CH2Ph
-CH2CH2Ph
-CH2CH3
41/2:1
66c/2:1
59d/7:1
66/7:1
75/3:1
56/4.2:1
51/2.1:1
57%c/3.1:1
65/3:1
-COCH3
-CH2CH2Ph
-CH(CH3)2
-COCH3
-COCH3
-COCH2CH3
-COCH2CH3
)
-CH(CH3 2
10
11
-CH3
-CH3
-CH2CH3
rt
-CH(CH3)2
10h a n d 11h
53/trans only
a
b
Diene 5b was used for all tandem reactions. All reported yields are based on purified material following silica gel chromatography.
c Two equiv of the aldehyde was used in the second step in the sequence. One equiv of TiCl4 was used in the first step with no Me2AlCl
d
followed by addition of the aldehyde upon complete formation of 7b.
Sch em e 10
Ta ble 5. Ta n d em Diels-Ald er Cycloa d d ition /Acid Ch lor id e Su bstitu tion Sequ en ce (See Sch em e 10 for Str u ctu r es)
Diels-Alder second
acid
chloride
yieldb/
18:19:20
entry dienophilea
R1
R2
-H
-H
-H
-H
T (°C)
step
R4
product no.
1
2
3
4
5
6b
6b
6a
6a
6c
-COCH3
-COCH3
-CO2CH3
-CO2CH3
40
40
rt
rt
rt
17a
17b
17b
17c
17v
-CH3
18a , 19a , a n d 20a
71/1.9:1.5:1
-CHdCHPh (E) 18b, 19b, a n d 20b 70%/ 1.1:0.6:1
73/20:17:1
18d , 19d , a n d 20d 54/1.1:1.1:1
18e, 19e, a n d 20e 79/ 1:1.1:0
-CHdCHPh (E) 18c, 19c, a n d 20c
-Ph
-Ph
-COCH2CH3 -CH3
a
b
Diene 5b was used for all tandem reactions. All reported yields are based on purified material following silica gel chromatography.
of TiCl4 (1.1 equiv). In each case, the yield of the tandem
sequence was higher than the overall yield of the peri-
cyclic and ionic reactions performed independently.
Clearly, the reduction in product handling is playing a
major role in this improvement in synthetic efficiency.
Within experimental limits, the ratios of cis and trans
products from the independent substitution reactions and
tandem sequences are the same. Of course, this would
be expected on the basis of the epimerization studies
outlined above.
operations separately, indicating that this is still a highly
efficient one-pot sequence.
P r otod esilyla tion Stu d ies. The principal cause of
reduced yields in both the Diels-Alder and substitution
reactions with aldehydes is protodesilylation of the
allylsilane moiety of cycloadducts 7 (see Tables 1 and 2).
The presence of Me2AlCl in either the Diels-Alder or
substitution reaction mixtures dramatically increases the
yield of these transformations. This can be seen when
the yields of the tandem procedure employing aldehyde
electrophiles are compared with the results of the sub-
stitution reaction itself. In most cases, the overall yield
of the tandem procedure was superior to the yield of the
corresponding substitution reaction with 7 and the same
electrophiles. This is seen upon inspection of Table 2,
entries 5, 10, and 12 vs Table 4, entries 1, 5, and 6,
respectively. If the Diels-Alder reactions are quantita-
tive with 1:1 stoichiometry of the diene and dienophile
as indicated by the high yields shown in Table 1, the only
possible difference between the tandem run following the
cycloaddition and the independent substitution reaction
is the presence of Me2AlCl. To verify any involvement
of this catalyst in yield enhancement, 0.2 equiv of Me2-
AlCl was added to substitution reaction mixtures and
then they were allowed to proceed as usual. In most
cases, the yield increased significantly (see Table 2,
entries 8, 11, 13, 15, and 17).
ii. Step Tw o:Acid Ch lor id e Ad d ition s. In light of
the fact that Me2AlCl was used in the Diels-Alder
reaction, all tandem runs therefore used the cocatalyst
system (see Scheme 10 and Table 5). When the cycload-
dition was complete, the mixture was cooled and the
electrophile solution added. This mixture was prepared
by combining AlCl3 (2.0 equiv) with Me2AlCl (0.3 or 1.2
equiv) in CH2Cl2 at rt, stirring for 20 min, and then
adding the requisite acid chloride. This slurry was cooled
to -65 °C, stirred until homogeneity was reached, and
then transferred to the flask containing the cycloaddition
intermediate 7. Unlike the tandem runs with aldehyde
electrophiles, there was not a profound increase in yield
over the product of the two steps performed indepen-
dently. The protodesilylated cycloadduct 8 will react also
with the acylium ion, thus contributing to overall yield.
In the aldehyde case, once protodesilylation occurred, the
olefin is no longer nucleophilic enough to promote sub-
stitution. Therefore, there is no reason to expect a major
difference in yield for substitutions with acid chloride
electrophiles. That being said, all tandem yields are
quite comparable to those obtained by performing the
Allylsilanes are protodesilylated by Brønsted acid as
either a planned procedure17
or as an undesired side
reaction.18 It is possible that Me2AlCl is serving as a
proton scavenger in the present study and is essentially
cleaning the TiCl4 in situ as has been suggested previ-

5262 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
ously.23,24 Although this appears to be a reasonable
explanation for the substitution results, it is unlikely that
the trace amount of acid present in the reaction mixtures
in this study (presumably from TiCl4) could be respon-
sible for the amount of protodesilylation observed in the
Diels-Alder reaction (Table 1, entry 5). However, sev-
eral equivalents of HCl would be produced upon aqueous
catalyst quench with saturated NaHCO3. Perhaps the
biphasic nature of the quench prevents the acid from
being consumed by the base before it effects protodesi-
lylation of the allylsilane moiety in 7.
A series of experiments was conducted to try to
pinpoint when protodesilylation was occurring to better
understand the process. One tandem sequence was
performed using TiCl4 to compare with the standard
conditions that use Me2AlCl during the cycloaddition
followed by TiCl4 for the substitution. The yield with
Me2AlCl was approximately 7% better than without it
(Table 4, entries 4 and 5), which is a significant differ-
ence, but not a large one. This result is in stark contrast
to the 38% yield obtained for the Diels-Alder result
catalyzed with TiCl4 (Table 1, entry 5). It was not clear
how the two-step tandem yield with TiCl4 (59%) could
be so much higher than the Diels-Alder result by itself.
To this end, the course of the TiCl4-catalyzed Diels-Alder
reaction leading to the formation of 7b was followed by
1H NMR spectroscopy, and the spectra are shown in
Figure 5.
With 20 mol % TiCl4, the cycloaddition required several
hours to complete, and only a trace amount of 8b was
visible at the end (Figure 5D, T ) 18 h). A similar result
was found when 1.0 equiv of TiCl4 was used. In this case,
the reaction was complete in 13 min. However, when
the catalytic reaction was worked up in the usual way
(NaHCO3 quench and ether extraction), the 1H NMR
spectrum of the crude material revealed that 52% of 7b
had protodesilylated (Figure 5E). These results cast
serious doubt that protodesilylation occurs during the
cycloaddition, but rather occurs on aqueous workup, even
though the work-up conditions are basic.
F igu r e 5. 1H NMR spectra from 3.5 to 7.0 PPM of the Diels-
Alder cycloaddition between 5b and 6a (1.5:1) using 20 mol %
TiCl4 in CD2Cl2. Spectra were taken at 3 min (A), 10 min (B),
60 min (C), 18 h (D), and following quench with saturated
NaHCO3 and aqueous workup (i.e., crude; E). The peak denoted
as “1” is the signal for the vinyl proton on 7b, while the peak
denoted as “2” is the signal for the equivalent exocyclic
methylene protons of the protodesilylated product 8b. Peaks
centered at 5.98, 6.35, and 6.52 are vinyl signals from 6a , while
those at 4.78, 4.92, 5.04, and 6.35 (overlapping with those of
6a ) are vinyl signals from 5b.
In the tandem experiments using aldehyde electro-
philes, allylsilanes 7 were both formed and reacted in
situ, thus minimizing protodesilylation, which illustrates
one clear advantage of this tandem strategy. This was
1
confirmed in a separate H NMR experiment. Using 1.0
equiv of TiCl4, 7b was both formed and subsequently
reacted with propionaldehyde in a single NMR tube,
providing the tandem products 10c and 11c with minimal
formation of 8c. This is consistent with the results
discussed above for the tandem reaction using TiCl4
catalysis alone that provided results similar to those
obtained when the cocatalyst Me2AlCl was used (entries
4 and 5 in Table 4).
inverted this ratio to 4:1. When 1.2 equiv of Me2AlCl
was employed, the ratio widened to 7:1.
These results suggest that substitution occurs by two
different pathways of similar mechanism. Brønsted acid
present in the reaction flask could have protodesilylated
7c, providing 8c, and formation of 20 arises by Friedel-
Crafts acylation on the trisubstituted olefin.25 The
regioselectivity of the substitution with 8c is determined
by the stability of the tertiary cation intermediate formed
on the way to 20c, the apparent product of retention. In
the absence of Brønsted acid, which has been removed
by the Me2AlCl prior to silane addition, “normal” allyl-
silane attack on the electrophile occurs providing the
In another protodesilylation study, the effect of Me2-
AlCl in a substitution reaction with acetyl chloride was
studied (see Scheme 8 and Table 3, entries 3-5). When
the reaction was catalyzed with AlCl3 alone, the ratio of
products resulting from inversion:retention, i.e., 18c +
19c:20c, was 1:2.5. Pretreating the AlCl3/acid chloride
mixture with 0.2 equiv of Me2AlCl prior to silane addition
(23) The role of the proton scavenger in such reactions was initially
proposed by Snider; see: (a) Snider, B. B.; Rodini, D. J .; Conn, R. S.
E.; Sealfon, S. J . Am. Chem. Soc. 1979, 101, 5285-5293. (b) Snider,
B. B.; Rodini, D. J . Tetrahedron Lett. 1980, 21, 1815-1818.
(24) Monti, H.; Audran, G.; Monti, J .-P.; Leandri, G. J . Org. Chem.
1996, 61, 6021-6023.
(25) A similar rationale for analogous substitutions involving allyl-
silanes and acid chlorides has been proposed by other groups; see: (a)
Polla, M.; Frejd, T. Acta Chem. Scand. 1993, 47, 716- 720. (b) Also see
ref 20.

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5263
product of inversion (i.e., 18c and 19c). The presence of
acid is no doubt highly variable in Friedel-Crafts reac-
tions because it can enter the reaction flask from a
number of sources including the catalyst (i.e., AlCl3) and
the acid chloride reagent that inevitably contains some
amount of the carboxylic acid. The amount of acid
present dictated which of these pathways dominates.
The cocatalyst system developed herein provides a
convenient and reliable method for controlling the mech-
anism of allylsilane substitution with acid chloride elec-
trophiles. Pillot had reported that when the acid chloride/
AlCl3 complex was added to the allylsilane, rather than
the reverse, the product of allyl inversion was isolated
as the major product and in good yield.20 In our hands,
we could not reproduce this effect by altering the order
of addition of reaction components. With our substrates,
the yield and ratio of products of inversion and retention
did not vary greatly regardless of the order of addition.
However, use of the cocatalyst system has provided
nearly complete control over this reaction.
Alder step using either chiral catalysts or a chiral
auxiliary on the dienophile to provide useful optically
enhanced (pure) building blocks.26 In fact, some excellent
enantiomeric excesses have been achieved already using
5a with a chiral copper catalyst.26c
In addition to continuing to explore the all-carbon
system in these tandem studies using allylsilanes, we are
currently exploring a number of projects stemming from
these initial studies. Diels-Alder reactions employing
glyoxylic acid derivatives27,28 as dienophiles followed by
electrophilic substitution with aldehyde electrophiles has
already yielded promising results in the pursuit of
substituted pyran-based targets. The use of aza dienes29
to produce pyridine- and piperidine-based molecules are
similarly being pursued. These results will be reported
in due course.
Exp er im en ta l Section
Gen er a l P r oced u r e. All reactions were carried out under
a positive atmosphere of dry argon unless otherwise indicated.
Solvents were distilled prior to use: Et2O was distilled from
sodium benzophenone; CH2Cl2 was distilled from CaH2. All
TiCl4 used in these studies was drawn from a sample that had
been distilled and stored in an air-free storage flask equipped
with a glass 14/20 joint (sealed with fluorine-based grease)
through which needles were inserted while the flask was under
a positive pressure of dry argon that was supplied through a
side arm equipped with a Teflon needle valve. A distilled
sample of TiCl4 was used over a maximum period of 3 weeks,
at which time it was redistilled. Melting points are uncor-
rected. Unless otherwise indicated, 1H NMR spectra were
recorded in CDCl3 at 300 MHz. 13C NMR spectra were
recorded at 75 MHz in CDCl3. Chemical shifts are listed
relative to CHCl3 (δ 7.24) for 1H NMR and (δ 77.00) for 13C
NMR.
Con clu sion s a n d F u tu r e Con sid er a tion s
Results presented in this report indicate that pericyclic
[4 + 2] cycloaddition and ionic allylsilane substitution
reactions can be combined effectively in a sequential
fashion. Further, these sequences are highly efficient
when compared to the same reactions performed inde-
pendently in terms of yield. That is, yield is at least as
good in the tandem sequences as it is the for the overall
yield of the reactions performed independently. Further,
this is accompanied by no significant changes in regio-
or stereoselectivity. The products obtained in this study
also illustrate a rapid buildup of molecular complexity
from very simple starting materials.
Preparations and spectral data not included here have been
published previously.3 However, the general procedures for
all procedures have been included here along with the spectral
data for previously unreported structures.
Meth yl 4-[(Tr im eth ylsilyl)m eth yl]-3-cycloh exen eca r -
boxyla te (7a ). The following general procedure was used for
all cycloadditions unless noted otherwise. To a flame-dried
flask equipped with a stir bar was added 3 mL of dry CH2Cl2,
42 mg of 5a (0.30 mmol), and 48 mg of methyl acrylate (6a )
(0.56 mmol). This was followed by dropwise addition of
dimethylaluminum chloride (60 µL of a 1 M solution, 0.06
mmol). The mixture was heated to reflux for 3 h, cooled to rt,
and quenched with saturated NaHCO3. The layers were
A cocatalyst system employing Me2AlCl and TiCl4 has
been developed that exhibits dramatic effects on yields
in substitution reactions employing allylsilane nucleo-
philes and aldehyde electrophiles. The proposed role of
Me2AlCl is that of a proton scavenger that is “cleaning”
the stronger TiCl4 catalyst that is necessary to promote
the electrophilic substitutions with aldehyde electro-
philes.23
The presence of Me2AlCl in the reaction medium of
Friedel-Crafts acylation reactions employing allylsilanes
has demonstrated remarkable effects on regioselectivity.
The cocatalyst is presumably consuming Brønsted acid
that otherwise would be involved in protodesilylating the
starting silane, thus altering the reaction pathway.
Further, we believe that more experimentation with this
cocatalyst system will result in reliable regioselectivity
in these acylation reactions.
The selectivity demonstrated in the substitution reac-
tions involving aldehyde electrophiles is, to the best of
our knowledge, the most selective of any substitution
reaction employing such cyclic allylsilane nucleophiles.
It has also been demonstrated that this result is only
observable when care is taken to decompose the highly
reactive titanium species present following substitution
to prevent isomerization during aqueous workup. This
was accomplished by the addition of anhydrous methanol
prior to aqueous workup. These results demonstrate that
cyclic allylsilane substrates of this type, which, by-in-
large, have been ignored synthetically, could be used in
a reliable fashion to set the stereochemistry of contiguous
centers off the carbocycle.
(26) (a) Evans, D. A.; Chapman, K. T.; Bisaha, J . J . Am. Chem. Soc.
1988, 110, 1238-1256. (b) Evans, D. A.; Miller, S. J .; Lectka, T. J .
Am. Chem. Soc. 1993, 115, 6460-6461. (c) Corey, E. J .; Letavic, M. A.
J . Am. Chem. Soc. 1995, 117, 9616-9617.
(27) For Diels-Alder reactions using aldehyde dienophiles, see: (a)
Weinreb, S. M.; Staib, R. R. Tetrahedron 1982, 38, 3087-3128. (b)
Boger, D. L.; Weinreb, S. M. Hetero Diels-Alder Methodology in
Organic Synthesis; Academic Press: San Diego, CA, 1987.
(28) (a) J ung, M. E.; Shishido, K.; Light, L.; Davis, L. Tetrahedron
Lett. 1981, 4607-4610. (b) Schmidt, R. R.; Abele, W. Angew. Chem.,
Int. Ed. Engl. 1982, 21, 302-304. (c) Schmidt, R. R.; Wagner, A.
Tetrahedron Lett. 1983, 24, 4661-4664. (d) J urczak, J .; Golebiowski,
A.; Rahm, A. Tetrahedron Lett. 1986, 27, 853-856. (e) Hosomi, A.;
Sakata, Y.; Sakurai, H. Tetrahedron Lett. 1985, 26, 5175-5178. (f)
Hosomi, A.; Otaka, K.; Sakurai, H. Tetrahedron Lett. 1986, 27, 2881-
2884.
(29) (a) Fitton, A. O.; Frost, J . R.; Houghton, P. G.; Suschitzky, H.
J . Chem. Soc., Perkin Trans. 1 1977, 1450-1463. (b) Moore, H. W.;
Hughes, G. M. Tetrahedron Lett. 1982, 23, 4003- 4006. (c) Brady, W.
T.; Shieh, C. H. J . Org. Chem. 1983, 48, 2499-2502. (d) Boger, D. L.
Chem. Rev. 1986, 86, 781-793. (e) Boger, D. L. Tetrahedron 1983, 39,
2869-2939. (f) Ohshiro, Y.; Komatsu, M.; Uesaka, M.; Agawa, T.
Heterocycles 1984, 22, 549-559. (g) Boger, D. L.; Honda, T.; Dang, Q.
J . Am. Chem. Soc. 1994, 116, 5619-5630. (h) Gonza´lez, C.; Pe´rez, D.;
Guitia´n, E.; Castedo, L. J . Org. Chem. 1995, 60, 6318-6326. (i) Pe´rez,
D.; Bure´s, G.; Guitia´n, E.; Castedo, L. J . Org. Chem. 1996, 61, 1650-
1654.
These results give a strong indication that this tandem
sequence can be carried out asymmetrically in the Diels-

5264 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
separated, and the aqueous layer was twice extracted with
ether. The pooled organic fraction was washed with water and
dried over anhydrous MgSO4. Following solvent removal in
vacuo, the product was purified by flash chromatography (2%
ether in hexanes), providing 60 mg of 7a as a clear colorless
liquid (88% yield). All spectra correspond to that previously
reported.19
mixture was allowed to warm to -10 °C, at which point the
reaction was quenched with saturated NaHCO3. The layers
were then separated, and the aqueous layer was twice ex-
tracted with ether. The pooled organic layers were then dried
over anhydrous MgSO4. Following solvent removal in vacuo,
the product was purified by flash chromatography (35% ether
in hexanes), providing 27 mg of combined products (10b:11b,
1.3:1, 70% yield).
Gen er a l P r oced u r e for Ta n d em Rea ction Usin g a n
Ald eh yd e in th e Secon d Step . The following procedure was
used for all tandem reactions involving diene 5b, dienophiles
6, and aldehydes 9.
Meth yl 3-(1-Hyd r oxy-3-p h en ylp r op yl)-4-m eth ylen ecy-
cloh exa n eca r boxyla te (10c a n d 11c). To a flame-dried
flask equipped with a stir bar was added 3.0 mL of dry CH2-
Cl2, 100 mg of 5b (0.49 mmol), and 43 mg of methyl acrylate
(6a ) (0.49 mmol) followed by dropwise addition of 9.3 mg of
dimethylaluminum chloride (100 µL of a 1 M solution, 0.10
mmol). The mixture was heated to reflux for 4 h, at which
time the cycloaddition was judged complete and the solution
was cooled to -60 °C. Seventy-two mg of hydrocinnamalde-
hyde (9c) (0.54 mmol) was added, followed by 102 mg of TiCl4
(0.54 mmol), producing a red mixture.
Dim eth yl 4-[(Dim eth ylp h en ylsilyl)m eth yl]-4-cycloh ex-
en e-1,2-d ica r boxyla te (7e). Following the general cycload-
dition procedure, 75 mg of 5b (0.37 mmol), 36 mg of dimethyl
fumarate (6d ) (0.25 mmol), and 23 mg of dimethylaluminum
chloride (250 µL of a 1 M solution, 0.25 mmol) provided,
following flash chromatography (20% ether in hexanes), 77 mg
of allylsilane 7e as a clear colorless liquid (89% yield): IR (thin
film) 3070, 3000, 1736 cm-1 1H NMR (CDCl3, 300 MHz) δ
;
7.50-7.43 (m, 2 H), 7.37-7.30 (m, 3 H), 5.15 (s, 1 H), 3.66 (s,
3 H), 3.63 (s, 3 H), 2.83-2.66 (m, 2 H), 2.41-2.29 (m, 1 H),
2.21-1.91 (m, 3 H), 1.64 (s, 2 H), 0.28 (s, 3 H), 0.27 (s, 3 H);
13C NMR (CDCl3, 75 MHz, APT pulse sequencessevens up (+),
odds down (-)) δ 175.5 (+), 175.2 (+), 138.6 (+), 133.5 (-),
133.1 (+), 129.0 (-), 127.7 (-), 117.2 (-), 51.8 (-), 41.8 (-),
41.1 (-), 33.3 (+), 28.1 (+), 26.4 (+), -2.9 (-), -3.0 (-); HRMS
calcd for C19H26O4Si 346.1601, found 346.1598.
cis-4-[(Dim et h ylp h en ylsilyl)m et h yl]-4-cycloh exen e-
1,2-d ica r boxylic a n h yd r id e (7g). To a flame-dried flask
equipped with a stir bar was added 5.0 mL dry CH2Cl2, 100
mg of 5b (0.49 mmol), and 66 mg of maleic anhydride (6e) (0.67
mmol). The reaction mixture was heated to reflux for 5 h and
cooled to rt. Following solvent removal in vacuo, the product
was purified by flash chromatography (60% ether in hexanes),
providing 132 mg of 7f as a clear colorless liquid (89% yield):
After being stirred for 1 h, the mixture was allowed to warm
to -10 °C, at which point the reaction was quenched with
saturated NaHCO3. The layers were then separated, and the
aqueous layer was twice extracted with ether. The pooled
organic layers were then dried over anhydrous MgSO4. Fol-
lowing solvent removal in vacuo, the product was purified by
flash chromatography (35% ether in hexanes), providing 93
mg of combined products (10c:11c, 7:1, 66% yield).
1
IR (thin film) 3070, 3048, 1843, 1778 cm-1; H NMR (CDCl3,
t r a n s-1-(5-Ace t yl-2-m e t h yle n e cycloh e xyl-2-oxop r o-
p a n e (12). To a flame-dried flask equipped with a stir bar
was added 1.5 mL of dry CH2Cl2, 23 mg of 10d (0.12 mmol),
and 76 mg of pyridinium chlorochromate (PCC) (0.35 mmol).
This dark mixture was stirred at room temperature for 30 h
and quenched with saturated NH4Cl. The layers were sepa-
rated and the aqueous layer twice extracted with ether. The
pooled organic phase was dried over anhydrous MgSO4, and
solvent was removed in vacuo. Flash chromatography (30%
ether in hexanes) provided 12 mg of 12 as a clear oil (51%
yield): IR (thin film) 3080, 1716, 1714 cm-1; 1H NMR (CDCl3,
300 MHz) δ 4.79 (s, 1 H), 4.35 (s, 1 H), 3.13 (dd, J ) 11.8, 3.7
Hz, 1 H), 2.67-2.34 (m, 4 H), 2.17-2.05 (m, 1 H), 2.13 (s, 3
H), 2.01-1.91 (m, 2 H), 1.78 (q, J ) 11.8 Hz, 1 H), 1.48 (qd, J
) 11.8, 3.7 Hz, 1 H), 1.04 (t, J ) 7.4 Hz, 3 H); 13C NMR (CDCl3,
75 MHz, APT pulse sequencesevens up (+), odds down (-)) δ
212.01 (+), 210.33 (+), 146.47 (+), 108.84 (+), 54.75 (-), 49.97
(-), 35.71 (+), 35.16 (+), 30.80 (+), 29.56 (+), 27.77 (-), 7.57
(-); HRMS calcd for C12H18O2 194.1307, found 194.1312.
cis-1-(5-Acet yl-2-m et h ylen ecycloh exyl-2-oxop r op a n e
(13). Following the protocol for the preparation of 12, 25 mg
of 11d (0.13 mmol) and 82 mg of PCC (0.38 mmol) in 1.5 mL
of CH2Cl2 gave following flash chromatography (20% ether in
hexanes) 9 mg of 13 as a clear oil (36% yield): IR (thin film)
3075, 1713, 1705 cm-1; 1H NMR (CDCl3, 300 MHz) δ 4.94 (m,
1 H), 4.90 (m, 1 H), 3.35 (m, 1 H), 2.87 (tt, J ) 11.8, 3.7 Hz, 1
H), 2.61-2.38 (m, 2 H), 2.34-2.24 (m, 1 H), 2.15 (s, 3 H), 2.05-
1.89 (m, 2 H), 1.51-1.15 (m, 3 H), 0.99 (t, J ) 7.4 Hz, 3 H);
13C NMR (CDCl3, 75 MHz, APT pulse sequencesevens up (+),
odds down (-)) δ 211.62 (+), 211.31 (+), 145.27 (+), 113.22
(+), 54.94 (-), 46.45 (-), 33.40 (+), 32.03 (+), 29.27 (+), 29.14
(+), 28.43 (-), 7.84 (-); HRMS calcd for C12H18O2 194.1307,
found 194.1313.
300 MHz) δ 7.49-7.44 (m, 2 H), 7.38-7.31 (m, 3 H), 5.40 (m,
1 H), 3.21 (m, 2 H), 2.50 (dd, J ) 14.7, 6.6 Hz, 1H), 2.32-1.97
(m, 3 H), 1.89 (d, J ) 13.2 Hz, 1 H), 1.72 (d, J ) 13.2 Hz, 1 H),
0.28 (s, 6 H); 13C NMR (CDCl3, 75 MHz) δ 174.43, 174.22,
137.99, 136.39, 133.42, 129.12, 127.77, 117.89, 39.98, 39.36,
29.48, 28.06, 24.05, -2.95, -3.24; HRMS calcd for C17H20O3Si
300.1182, found 300.1188.
Gen er a l P r oced u r e for Electr op h ilic Su bstitu tion
w ith Ald eh yd es a n d Diels-Ald er Ad d u cts 7. To avoid a
very lengthy section here, one experimental procedure for each
set of electrophilic substitution conditions is discussed in detail
along with the appropriate spectral data for those compounds.
Every substitution indicated on Table 2 was done using either
method A or B on the same scale. Also included is the general
procedure for the tandem Diels-Alder/electrophilic substitu-
tion sequence using aldehyde electrophiles.
Met h od A: Met h yl 3-(1-H yd r oxyp r op yl)-4-m et h yl-
en ecycloh exa n eca r boxyla te (10a a n d 11a ). To a flame-
dried flask equipped with a stir bar was added 3.0 mL of dry
CH2Cl2, 100 mg of 7b (0.35 mmol), and 24 mg of propional-
dehyde (9a ) (0.39 mmol). After the solution was cooled to -60
°C, 66 mg of TiCl4 (0.35 mmol) was added dropwise, producing
an orange solution. After being stirred for 1 h, the mixture
was allowed to warm to -10 °C, at which point the reaction
was quenched with saturated NaHCO3. The layers were then
separated, and the aqueous layer was twice extracted with
ether. The pooled organic layers were then dried over anhy-
drous MgSO4. Following solvent removal in vacuo, the product
was purified by flash chromatography (45% ether in hexanes),
providing 45 mg of combined products (10a :11a , 4.3:1, 61%
yield).
Meth od B: Meth yl 3-(1-Hyd r oxy-2-m eth ylp r op yl)-4-
m eth ylen ecycloh exa n e ca r boxyla te (10b a n d 11b). To a
flame-dried flask equipped with a stir bar was added 0.75 mL
of dry CH2Cl2, 50 mg of 7b (0.17 mmol), and 14 mg of
isobutyraldehyde (9b) (0.19 mmol). After the mixture was
cooled to -60 °C, 3.2 mg of dimethylaluminum chloride (35
µL of a 1.0 M solution, .035 mmol) was added dropwise. In a
separate flame-dried flask equipped with a stir bar was added
0.5 mL of dry CH2Cl2, 3.2 mg of dimethylaluminum chloride
(35 µL of a 1 M solution, 0.35 mmol), and 36 mg of TiCl4 (0.19
mmol). This yellow solution was stirred for 20 min at rt, cooled
to -60 °C, and added by cannula to the allylsilane solution,
generating an orange solution. After being stirred for 1 h, this
4-E t h yl-1-[(d im et h ylp h en ylsilyl)m et h yl]-1-cycloh ex-
en e (14). To a flask equipped with a stir bar was added 1
mL of dry ethanol, 539 mg of 5b (1.98 mmol), and 369 mg of
p-toluenesulfonylhydrazide (1.98 mmol). The mixture was
heated to reflux for 2 h and allowed to gradually cool to rt, at
which time the formation of white crystals was observed. The
slurry was cooled further in an ice bath, and the crystals were
collected by vacuum filtration. These crystals were dried
under high vacuum and used directly in the reduction. To a
flame-dried flask was added 6 mL of dry THF and 300 mg of
the crude hydrazone (0.68 mmol). The solution was cooled to
0 °C and 52 mg of LiAlH4 (1.36 mmol) added slowly. The

Tandem Transformations Involving Allylic Silanes
J . Org. Chem., Vol. 62, No. 16, 1997 5265
mixture was heated to reflux for 12 h. After the mixture was
cooled to rt, 25 mg of water was added followed by 40 µL of a
1 M NaOH solution. The solution was stirred for 30 min and
then dried over anhydrous MgSO4. Following solvent removal
in vacuo, the product was purified by flash chromatography
(straight hexanes), providing 20 mg of 14 (11% yield) as a clear,
1-Acetyl-4-m eth ylen e-3-((E)-3-ph en yl-1-oxo-2-pr open yl)-
cycloh exa n e (18b a n d 19b) p lu s Regioisom er 20b. To a
flame-dried flask equipped with a stir bar was added 1.5 mL
of dry CH2Cl2, 50 mg of 7c (0.18 mmol), and 4 mg of
dimethylaluminum chloride (54 µL of a 1 M solution, 0.054
mmol). In a separate flame-dried flask equipped with a stir
bar were added 48 mg of AlCl3 (0.50 mmol), 0.75 mL of dry
CH2Cl2, and 2 mg of dimethylaluminum chloride (25 µL of a 1
M solution, 0.025 mmol). This yellow slurry was stirred for
20 min at rt, 46 mg of cinnamoyl chloride (17b) (0.28 mmol)
was added, and both solutions were cooled to -60 °C. The
acid chloride solution was added by cannula to the allylsilane
solution generating a yellow mixture that was allowed to warm
gradually to -10 °C. Following dilution with ether, the
mixture was quenched with saturated NaHCO3. The layers
were separated, and the aqueous layer was twice extracted
with ether. The organic layers were combined and dried over
anhydrous MgSO4. Following solvent removal in vacuo, the
product was purified by flash chromatography (11% ether in
hexanes), providing 53 mg of combined products (18b:19b:20b,
1.6:0.8:1, 78% yield). Compound 18b was cleanly separable
from 19b and 20b, which could only be isolated together as a
mixture. Ratios were determined by 1H NMR spectroscopy
on both the crude mixture and purified sample of 19b and 20b
using the well-resolved vinyl proton signals (see text for
discussion).
18b: white crystals; mp 105.9-107.3 °C; IR (thin film):
3078, 3062, 1710, 1683 cm-1; H NMR (CDCl3, 300 MHz) δ 7.61
(d, J ) 16.2 Hz, 1 H), 7.56-7.48 (m, 2 H), 7.39-7.33 (m, 3H),
6.99 (d, J ) 16.2 Hz, 1 H), 5.03-4.99 (m, 2 H), 3.60-3.55 (m,
1 H), 2.94 (tt, J ) 11.8, 3.7 Hz, 1 H), 2.46-2.36 (m, 1 H), 2.32
(dt, J ) 14.0, 3.7 Hz, 1 H), 2.17 (s, 3 H), 2.15-1.92 (m, 2H),
1.67-1.53 (m, 1 H), 1.40 (qd, J ) 11.8, 4.4 Hz, 1 H); 13C NMR
(CDCl3, 75 MHz) δ 211.30, 199.11, 145.27, 142.80, 134.44,
130.48, 128.85, 128.40, 122.89, 113.40, 54.49, 46.42, 32.16,
29.37, 29.22, 28.40. Anal. Calcd for C18H20O3: C, 80.56; H,
7.51. Found: C, 80.21; H, 7.54.
1
colorless oil: IR (thin film) 3069 cm-1; H NMR (CDCl3, 300
MHz) δ 7.53-7.46 (m, 2 H), 7.37-7.30 (m, 3 H), 5.18 (m, 1 H),
2.11-1.99 (m, 1 H), 1.94-1.51 (m, 4 H), 1.63 (s, 2 H), 1.35-
1.02 (m, 4 H), 0.87 (t, J )7.4 Hz, 3 H), 0.27 (s, 3 H), 0.26 (s, 3
H); 13C NMR (CDCl3, 75 MHz) δ 139.59, 134.58, 133.58, 128.78,
127.62, 119.27, 35.13, 31.90, 31.19, 29.14, 29.08, 26.69, 11.54,
-2.74; HRMS calcd for C17H26Si 258.1805, found 258.1803.
1-(5-Eth yl-2-m eth ylen ecycloh exyl)-1-p r op a n ol (15 a n d
16). Following the protocol for the preparation of 10b and 11b
(i.e., method B), 42 mg of 14 (0.16 mmol), 10 mg of propional-
dehyde (9a ) (0.18 mmol), 34 mg of TiCl4 (0.18 mmol), and 47
mg of dimethylaluminum chloride (50 µL of a 1 M solution,
0.05 mmol) in 2 mL of dry CH2Cl2 gave, after flash chroma-
tography (5% ether in hexanes), 17 mg of combined products
(15:16, 27:1, 60% yield). Compounds 15 and 16 were insepa-
rable, and the ratio of each was determined by 1H NMR
spectroscopy on both the crude mixture and purified sample
using the well-dissolved vinyl proton signals (see text for
discussion). The spectral data for 15, the major isomer, are
included here: IR (thin film) 3362, 3070 cm-1; 1H NMR (CDCl3,
300 MHz) δ 4.68-4.63 (m, 2 H), 3.69 (td, J ) 9.6, 3.0 Hz, 1
H), 2.22-1.00 (m, 13 H), 0.94 (t, J ) 7.4 Hz, 3 H), 0.89 (t, J )
7.4 Hz, 3 H); 13C NMR (CDCl3, 75 MHz, APT pulse sequences
evens up (+), odds down (-)) δ 150.29 (+), 109.28 (+), 71.62
(-), 50.03 (-), 34.63 (+), 34.19 (+), 33.55 (-), 32.19 (+), 29.48
(+), 28.16 (+), 11.57 (-), 9.89 (-); HRMS calcd for C12H22
182.1672, found 182.1665.
O
Gen er a l P r oced u r e for Electr op h ilic Su bstitu tion
w ith Acid Ch lor id es a n d Diels-Ald er Ad d u cts 7 Usin g
AlCl3 Ca ta lysis. Substitution reactions using this catalyst
system, as indicated in Table 3, were carried out using the
following general procedure.
Compounds 19b and 20b: 1H NMR (CDCl3, 300 MHz) δ
7.61(d, J ) 13.2 Hz), 7.56-7.49 (m), 7.40-7.34 (m), 6.84 (d, J
) 15.5 Hz), 6.76 (d, J ) 16.2 Hz), 5.61 (br s), 4.85 (s), 4.51 (s),
3.37 (dd, J ) 11.8, 3.8 Hz), 3.26 (m), 2.63-2.44 (m), 2.16 (s),
2.15 (s), 2.26-1.43 (m).
1,3-Dia cetyl-4-m eth ylen ecycloh exa n e (18a a n d 19a )
p lu s Regioisom er 20a . To a flame-dried flask equipped with
a stir bar was added 2.0 mL of dry CH2Cl2, and 48 mg of AlCl3
(0.36 mmol). The mixture was cooled to -65 °C, and 22 mg of
acetyl chloride (17a ) (0.27 mmol) was added. After the mixture
was stirred for 10 min, 50 mg of 7c was added dropwise, and
the solution was allowed to warm gradually to -10 °C. The
reaction was then diluted with ether and quenched with
saturated NaHCO3. The layers were separated, and the
aqueous phase was extracted twice with ether. The combined
organic layers were washed with water and dried over
anhydrous MgSO4. Following solvent removal in vacuo, the
product was purified by flash chromatography (20% ether in
hexanes), providing 27.5 mg of combined products (18a :19a :
20a , 2:1:7.6, 85% yield). Compound 18a was cleanly separable
from 19a and 20a , which could only be isolated together as a
mixture. Ratios were determined by 1H NMR spectroscopy
on both the crude mixture and purified sample of 19a and 20a
using the well-resolved vinyl proton signals (see text for
discussion).
18a : a clear, colorless oil; IR (thin film) 3075, 1713, 1709
cm-1; 1H NMR (CDCl3, 300 MHz) δ 4.95 (s, 1 H), 4.92 (s, 1 H),
3.33 (m, 1 H), 2.81 (tt, J )11.8, 3.7 Hz, 1 H), 2.34-2.25 (m, 2
H), 2.13 (s, 3 H), 2.12 (s, 3 H), 2.03-1.89 (m, 2 H), 1.50-1.20
(m, 2 H); 13C NMR (CDCl3, 75 MHz, APT pulse sequencesevens
up (+), odds down (-)) δ 211.12 (+), 208.91 (+), 145.08 (+),
113.43 (+), 55.86 (-), 46.35 (-), 31.97 (+), 29.25 (+), 28.83
(+), 28.33 (-), 27.86 (-); HRMS calcd for C11H16O2 180.1151,
found 180.1152.
Gen er a l P r oced u r e for Ta n d em Rea ction s Usin g a n
Acid Ch lor id e in th e Secon d Step : Meth yl 4-Meth ylen e-
3-((E)-3-p h en yl-1-oxo-2-p r op en yl)cycloh exa n eca r b oxy-
la te (18c a n d 19c) p lu s Regioisom er 20c. To a flame-dried
flask equipped with a stir bar was added 2.0 mL of dry CH2-
Cl2, 100 mg of 2-[(dimethylphenylsilyl)methyl]-1,3-butadiene
(5b) (0.49 mmol), and 42 mg of methyl acrylate (6a ) (0.49
mmol). This was followed by dropwise addition of 14 mg of
dimethylaluminum chloride (150 µL of a 1 M solution, 0.15
mmol). The mixture was stirred at rt for 4 h, at which time
the cycloaddition was judged complete and the solution was
cooled to -65 °C. In a separate flame-dried flask equipped
with a stir bar was added 131 mg of AlCl3 (0.98 mmol), 1.5
mL of dry CH2Cl2, and 5 mg of dimethylaluminum chloride
(47 µL of a 1 M solution, 0.047 mmol). This yellow slurry was
stirred for 20 min at rt, and 123 mg of cinnamoyl chloride (17b)
(0.74 mmol) was added. After the solution was cooled to 65
°C, the acid chloride solution was added by cannula, generating
a yellow mixture that was allowed to warm gradually to -10
°C. At that point, the mixture was diluted with ether and
quenched with saturated NaHCO3
. The layers were separated,
and the aqueous layer was twice extracted with ether. The
combined organic phase was then washed with water and dried
over anhydrous MgSO4. Following solvent removal in vacuo,
the product was purified by flash chromatography (11% ether
in hexanes), providing 102 mg of combined product (18c:19c:
20c, 20:17:1, 73% yield).
18c: white needles recrystallized from hexanes; mp 89.2-
90.3 °C; IR (thin film) 3060, 3028, 1732 cm-1; 1H NMR (CDCl3,
300 MHz) δ 7.59 (d, J )16.2 Hz, 1 H), 7.56-7.47 (m, 2 H),
7.40-7.31 (m, 3 H), 6.97 (d, J )16.2 Hz, 1 H), 5.00 (s, 1 H),
4.96 (s, 1 H), 3.65 (s, 3 H), 3.56 (m, 1 H), 2.88 (tt, J )11.8, 3.7
Hz, 1 H), 2.44 (m, 1 H), 2.37-2.24 (m, 1 H), 2.15-1.95 (m, 2
H), 1.80-1.67 (m, 1 H), 1.63-1.47 (m, 1 H); 13C NMR (CDCl3,
Compounds 19a and 20a : 1H NMR (CDCl3, 300 MHz) δ 5.52
(br s), 4.81 (s), 4.42 (s), 3.09 (dd, J ) 11.8, 3.7 Hz), 3.00 (s),
2.59-2.37 (m), 2.18 (s), 2.13 (s), 2.12 (s), 2.08 (s), 2.01-1.89
(m), 181-1.67 (m), 1.63-1.38 (m).
Gen er a l P r oced u r e for Electr op h ilic Su bstitu tion
w ith Acid Ch lor id es a n d Diels-Ald er Ad d u cts 7 u sin g
AlCl3/Me2AlCl Ca ta lysis. Substitution reactions using this
catalyst system, as indicated in Table 3, were carried out using
the following general procedure.

5266 J . Org. Chem., Vol. 62, No. 16, 1997
Organ et al.
75 MHz, APT pulse sequencesevens up (+), odds down (-)) δ
198.91 (+), 175.80 (+), 145.32 (+), 142.69 (-), 134.48 (+),
130.42 (-), 128.83 (-), 128.36 (-), 123.15 (-), 113.07 (+), 54.18
(-), 51.58 (-), 38.57 (-), 32.11 (+), 30.06 (+), 29.90 (+). Anal.
Calcd for C18H20O3: C, 76.03; H, 7.09. Found: C, 75.95; H,
7.11.
of AlCl3 (0.98 mmol), and 14 mg of dimethylaluminum chloride
(150 µL of a 1 M solution, 0.15 mmol) in 4 mL of dry CH2Cl2
gave, after flash chromatography (10% ether in hexanes), 104
mg of combined products (18e:19e:20e, 1.0:1.1:0, 79% yield).
18e: white crystals recrystallized from pentane; mp 58.0-
58.9 °C; IR (thin film) 3072, 1707, 1677 cm-1; 1H NMR (CDCl3,
300 MHz) δ 7.91 (d, J ) 7.4 Hz, 2 H), 7.55-7.47 (m, 1 H), 7.40
(t, J ) 7.4 Hz, 2 H), 4.90 (s, 1 H), 4.86 (s, 1 H), 4.31 (m, 1 H),
2.88 (td, J ) 9.6, 3.7 Hz, 1 H), 2.60-2.32 (m, 2 H), 2.18 (m, 2
H), 1.96 (t, J ) 12.5 Hz, 1 H), 1.89-1.64 (m, 2 H), 1.00 (t, J )
7.4 Hz, 3 H), 0.85 (d, J ) 5.9 Hz, 3 H); 13C NMR (CDCl3, 75
MHz, APT pulse sequencesevens up (+), odds down (-)) δ
215.09 (+), 201.01 (+), 144.88 (+), 136.32 (+), 132.84 (-),
128.46 (-), 128.41 (-), 113.12 (+), 52.02 (-), 49.26 (-), 39.94
(+), 36.81 (+), 35.21 (-), 31.38 (+), 20.03 (-), 7.49 (-). Anal.
Calcd for C18H22O2: C, 79.96; H, 8.20. Found: C, 79.84; H,
8.19.
1
19c: colorless oil; IR (thin film) 3081, 3027, 1732 cm-1; H
NMR (CDCl3, 300 MHz) δ 77.60 (d, J ) 16.2 Hz, 1 H), 7.56-
7.49 (m, 2 H), 7.40-7.33 (m, 3 H), 6.83 (d, J ) 16.2 Hz, 1 H),
4.85 (s, 1 H), 4.50 (s, 1 H), 3.66 (s, 3 H), 3.35 (m, 1 H), 2.63-
2.42 (m, 2 H), 2.34-1.51 (m, 5 H); 13C NMR (CDCl3, 75 MHz,
APT pulse sequencesevens up (+), odds down (-)) δ 199.83
(+), 175.08 (+), 146.03 (+), 142.44 (-), 134.44 (+), 130.46 (-),
128.87 (-), 128.36 (-), 125.38 (-), 109.91 (+), 53.73 (-), 51.71
(-), 42.14 (-), 35.37 (+), 31.45 (+), 30.17(+); HRMS calcd for
C18H20O3 284.1413, found 284.1427.
Met h yl 4-Met h ylen e-3-(1-oxo-1-p h en yl)cycloh exa n e-
ca r boxyla te (18d a n d 19d ) p lu s Regioisom er 20d . Fol-
lowing the general tandem procedure with acid chloride
electrophiles, 150 mg of 2-[(dimethylphenylsilyl)methyl]-1,3-
butadiene (5b) (0.74 mmol), 64 mg of methyl acrylate (6a ) (0.74
mmol), 21 mg of dimethylaluminum chloride (220 µL of a 1 M
solution, 0.22 mmol), 156 mg of benzoyl chloride (17c) (1.11
mmol), and 197 mg of AlCl3 (1.48 mmol) in 6 mL of dry CH2-
Cl2 gave, after flash chromatography (10% ether in hexanes),
102 mg of combined products (18d :19d :20d , 1.1:1.1:1, 54%).
Compound 18d was cleanly separable from 19d and 20d ,
which could only be isolated together as a mixture. Ratios
19e: white needles recrystallized from hexanes; mp 111.9-
112.7 °C; IR (thin film): 3083, 1708, 1674 cm-1 1H NMR
;
(CDCl3, 300 MHz) δ 7.92 (d, J )7.4 Hz, 2 H), 7.56-7.49 (m, 1
H), 7.41 (t, J )7.4 Hz, 2 H), 4.80 (s, 1 H), 4.30 (s, 1 H), 3.93-
3.86 (m, 1 H), 2.58-2.32 (m, 4 H), 2.03-1.85 (m, 4 H), 1.03 (t,
J )7.4 Hz, 3 H), 0.88 (d, J )5.9 Hz, 3 H); 13C NMR (CDCl3, 75
MHz, APT pulse sequencesevens up (+), odds down (-)) δ
231.38 (+), 200.69 (+), 147.05 (+), 136.74 (+), 133.10 (-),
128.53 (-), 128.45 (-), 110.43 (+), 57.22 (-), 50.42 (-), 44.33
(+), 35.84 (-), 34.89 (+), 32.34 (+), 20.10 (-), 14.08 (-). Anal.
Calcd for C18H22O2: C, 79.96; H, 8.20. Found: C, 79.87; H,
8.14.
1
were determined by H NMR spectroscopy on both the crude
mixture and purified sample of 19d and 20d using the well-
resolved vinyl proton signals (see text for discussion).
18d : clear, colorless oil; IR (thin film) 3072, 1738, 1683
cm-1; 1H NMR (CDCl3, 300 MHz) δ 7.96-7.90 (m, 2 H), 7.56-
7.38 (m, 3 H), 4.88 (br s, 1 H), 4.72 (br s, 1 H), 4.32 (t, J ) 5.2
Hz, 1 H), 3.65 (s, 3 H), 3.01 (m, 1 H), 2.40-2.27 (m, 3 H), 2.04-
1.91 (m, 2 H), 1.78-1.64 (m, 1 H); 13C NMR (CDCl3, 75 MHz)
δ 201.04, 175.77, 145.93, 136.50, 132.90, 128.54, 128.48,
112.22, 51.65, 48.76, 38.49, 32.18, 31.14, 30.01; HRMS calcd
for C16H19O3 (M+ + 1) 259.1335, found 259.1346.
Ack n ow led gm en t. This work was supported in part
by a research grant from Eli Lilly and Company. D.D.W.
acknowledges The Purdue School of Science for a R.I.F.
Fellowship. The authors are indebted to Professors Scott
Denmark and Ian Fleming for their insightful sugges-
tions during the course of these studies. Dr. Anthony
Shuker is acknowledged for his useful comments in the
production of this manuscript.
Compounds 19d and 20d : 1H NMR (CDCl3, 300 MHz) δ 7.93
(m), 7.59-7.49 (m), 7.46-7.38 (m), 5.55 (m), 4.81 (s), 4.33 (s),
3.92 (dd, J ) 12.5, 3.0 Hz), 3.66 (s), 3.64 (s), 3.59 (br s), 2.66-
2.44 (m), 2.27 (m), 2.23-1.95 (m), 1.78-1.54 (m).
Su p p or tin g In for m a tion Ava ila ble: 1H or 13C NMR
spectra for all compounds prepared in this study and X-ray
crystallographic data for 10f (44 pages). This information 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.
4-Meth yl-6-m eth ylen e-1-(1-oxo-1-p h en yl)-3-(1-oxop r o-
p yl)cycloh exa n e (18e a n d 19e) p lu s Regioisom er 20e.
Following the general tandem procedure with acid chloride
electrophiles, 100 mg of 2-[(dimethylphenylsilyl)methyl]-1,3-
butadiene (5b) (0.49 mmol), 48 mg of 4-hexen-3-one (6c) (0.49
mmol), 103 mg of benzoyl chloride (17c) (0.74 mmol), 131 mg
J O970821V