New approaches to olefin cross-metathesis

Article abstract of DOI:10.1021/ja993063u

New methodology for the selective cross-metathesis (CM) of terminal olefins employing ruthenium benzylidene 1 is described. CM with symmetric internal olefins was found to provide a useful means for homologating terminal olefins to protected allylic alcoh

Full text of DOI:10.1021/ja993063u

58
J. Am. Chem. Soc. 2000, 122, 58-71
New Approaches to Olefin Cross-Metathesis
Helen E. Blackwell,† Daniel J. O’Leary,‡ Arnab K. Chatterjee,† Rebecca A. Washenfelder,‡
D. Andrew Bussmann,‡ and Robert H. Grubbs*,†
Contribution from the Arnold and Mabel Beckman Laboratory of Chemical Synthesis, DiVision of
Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, and
Department of Chemistry, Pomona College, Claremont, California 91711
ReceiVed August 23, 1999
Abstract: New methodology for the selective cross-metathesis (CM) of terminal olefins employing ruthenium
benzylidene 1 is described.1 CM with symmetric internal olefins was found to provide a useful means for
homologating terminal olefins to protected allylic alcohols, amines, and esters. Due to the limited commercial
availability of symmetric internal olefins, a two-step CM procedure was developed in which terminal olefins
were first homodimerized prior to the CM reaction. Terminal olefins with allylic methyl substituents were
observed to provide CM products in diminished yield albeit with markedly improved trans-selectivity. Reaction
rates were measured for CM reactions utilizing butenediol and allyl alcohol derivatives, and the results
demonstrated distinct advantages in reaction rate and stereoselectivity for reactions employing the disubstituted
olefins. In the course of studies of substrates with allylic oxygen substituents, a new CM application was
discovered involving the metathesis of acrolein acetal derivatives with terminal olefins. Acrolein acetals,
including asymmetric variants derived from tartaric acid, proved to be exceptionally robust and trans-selective
CM substrates. In related work, a pinacol-derived vinyl boronate was also found to be a reactive CM partner,
providing a novel means for converting terminal olefins into precursors for the Suzuki coupling reaction.
Introduction
In contrast, the application of olefin metathesis to the synthesis
of complex organic molecules and natural products was limited
due to the incompatibility of ill-defined, “classical” catalysts
with the diverse functionality encountered in organic synthesis.2a
Recently, however, ring-closing olefin metathesis (RCM) of
acyclic dienes has received considerable attention as a highly
efficient methodology for the synthesis of functionally diverse
carbocycles and heterocycles.5 This is primarily due to the
development of well-defined transition metal catalysts over the
past decade. The two olefin metathesis catalysts that have seen
the most extensive use are the ruthenium benzylidene 1
developed by Grubbs et al.6 and the molybdenum alkylidene 2
developed by Schrock et al.7 The relatively high activities and
functional group tolerance of both catalysts 1 and 2, coupled
with their commercial availability, has dramatically increased
their application in organic synthesis.
Olefin Metathesis. Carbon-carbon bond forming reactions
are among the most important family of reactions in organic
synthesis. One particularly interesting carbon-carbon bond
forming reaction is olefin metathesis, which is the metal-
catalyzed exchange of alkylidene moieties between alkenes (eq
1).2
Historically, olefin metathesis has been studied both from a
mechanistic standpoint3 and in the context of polymer synthesis
(i.e., in ring opening metathesis polymerization, or ROMP).2a,4
Olefin Cross-Metathesis. The volume of work reported in
the areas of RCM, ROMP, and novel combinations thereof has
dramatically overshadowed that reported for olefin cross-
metathesis (CM). This unique method for the intermolecular
* Address correspondence to this author.
† California Institute of Technology.
‡ Pomona College.
(5) For recent reviews of RCM in organic synthesis, see: (a) Grubbs,
R. H.; Chang, S. Tetrahedron 1998, 54, 4413-4450. (b) Schuster, M.;
Blechert, S. Angew. Chem., Int. Ed. Engl. 1997, 36, 2036-2056. (c)
Schmalz, H.-G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1833-1836. (d)
Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28, 446-
552.
(6) PCy3 ) tricyclohexylphosphine. For the preparation and characteriza-
tion of catalyst 1, see: (a) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs,
R. H. Angew. Chem., Int. Ed. Engl. 1995, 34, 2039-2041. (b) Schwab, P.;
Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996, 118, 100-110. (c)
Belderrain, T. R.; Grubbs, R. H. Organometallics 1997, 16, 4001-4003.
(7) (a) Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; DiMare,
M.; O’Regan, M. J. Am. Chem. Soc. 1990, 112, 3875-3886. (b) Bazan, G.
C.; Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V. C.; O’Regan, M.
B.; Thomas, J. K.; Davis, W. M. J. Am. Chem. Soc. 1990, 112, 8378-
8387. (c) Bazan, G. C.; Oskam, J. H.; Cho, H.-N.; Park, L. Y.; Schrock, R.
R. J. Am. Chem. Soc. 1991, 113, 6899-6907.
(1) For preliminary accounts of this work, see: (a) O’Leary, D. J.;
Blackwell, H. E.; Washenfelder, R. A.; Grubbs, R. H. Tetrahedron Lett.
1998, 39, 7427-7430. (b) O’Leary, D. J.; Blackwell, H. E.; Washenfelder,
R. A.; Miura, K.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 1091-1094.
(c) Blackwell, H. E. Ph.D. Thesis, California Institute of Technology, 1999.
(2) For general olefin metathesis references, see: (a) Ivin, K. J.; Mol, J.
C. Olefin Metathesis and Metathesis Polymerization, 2nd ed.; Academic:
San Diego, 1997. (b) Grubbs, R. H.; Pine, S. H. In ComprehensiVe Organic
Synthesis; Trost, B. M., Ed.; Pergamon: New York: 1991; Vol. 5, Chapter
9.3.
(3) For details of the present accepted mechanism of olefin metathesis
involving formation of a metallocyclobutane intermediate, see: Herrison,
J. L.; Chauvin, Y. Makromol. Chem. 1971, 141, 161-176.
(4) For leading references, see: (a) Feldman, J.; Schrock, R. R. Prog.
Inorg. Chem. 1991, 39, 1-74. (b) Grubbs, R. H.; Tumas, W. Science 1989,
243, 907-915.
10.1021/ja993063u CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/21/1999

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 59
catalysts 1 and 2.14,15 The novel ruthenium-catalyzed homolo-
gation of homoallylglycine derivatives Via CM has been
reported by Gibson et al.16 Efficient crossed yne-ene17 and ring-
opening cross-metathesis (ROM) reactions18,19 using catalysts
1 and 2 have also been demonstrated. Finally, CM is being
employed with increasing frequency in the synthesis of solution-
phase combinatorial libraries of highly functionalized dimeric
molecules.20
formation of carbon-carbon double bonds has not yet found
widespread application in organic synthesis because general
reaction conditions that give high product and/or trans/cis
selectivity have not been developed. The simplified CM reaction
between two terminal olefins is depicted in eq 2.
Outlined herein are several new approaches for the selective
CM of unhindered terminal olefins. Our approach, using
symmetric disubstituted olefins as coupling partners, was
inspired in part by the synthesis of telechelic polymers21 Via
tandem ROMP coupled with the CM of disubstituted internal
olefins (eq 3). Blechert et al. have also used this approach in
Generally, this reaction proceeds to yield three unique
products: one desired heterodimeric product and two undesired
homodimeric products, each as a mixture of olefin isomers. The
majority of the work reported to date in the area of CM has
focused upon terminal olefin substrates, because employing
asymmetrically-substituted internal olefins as starting materials
can add further unwanted complexity to the final product
mixture. A predominance of the early reports of CM employing
“classical” catalysts8 involved the synthesis of insect pheromone
natural products: these compounds are frequently isolated from
natural sources as a specific ratio of cis and trans isomers, and
therefore CM proved to be a moderately effective route toward
synthesizing these product mixtures.9 However, for application
to synthetic organic chemistry in general, control of trans/cis
ratios and product selectivity is essential.
The advent of well-defined ruthenium and molybdenum
metathesis catalysts 1 and 2 has generated renewed interest in
developing methods for the selective CM of terminal olefins.
Crowe et al. have demonstrated that π-substituted terminal
olefins such as styrene10 and acrylonitrile11 can be used to
efficiently functionalize terminal olefins employing molybdenum
catalyst 2. Crowe has also reported a useful terminal olefin cross-
coupling procedure utilizing nucleophilic alkenes such as
allyltrimethylsilane.12,13 Recently, Blechert et al. have shown
that certain sterically hindered terminal olefins do not undergo
self-metathesis, but rather can be selectively functionalized with
a variety of commercially available terminal olefins using both
the ROM of strained cyclic olefins with symmetrically disub-
stituted olefins.18b To further probe the viability of this approach
for applications in organic synthesis, we have explored the Ru-
catalyzed homologation of unhindered terminal alkenes Via CM
with functionally diverse, disubstituted internal olefins.
(14) Bru¨mmer, O.; Ru¨ckert, A.; Blechert, S. Chem. Eur. J. 1997, 3, 441-
446.
(15) Highly functionalized silsesquioxanes and spherosilicates have been
prepared Via the CM of various alkenes with vinyl-substituted silsesquioxane
and spherosilicate frameworks employing molybdenum alkylidene 2. The
lack of homodimerization of the vinyl-substituted silicon frameworks was
attributed to steric bulk. See: Feher, F. J.; Soulivong, D.; Eklund, A. G.;
Wyndham, K. D. J. Chem. Soc., Chem. Commun. 1997, 1185-1186.
(16) (a) Gibson, S. E.; Gibson, V. C.; Keen, S. P. Chem. Commun. 1997,
1107-1108. (b) Baigini, S. C. G.; Gibson, S. E.; Keen, S. P. J. Chem.
Soc., Perkin Trans. 1 1998, 16, 2485-2499.
(17) For solution-phase yne-ene metathesis employing alkylidene 1,
see: (a) Stragies, R.; Schuster, M.; Blechert, S. Angew. Chem., Int. Ed.
Engl. 1997, 36, 2518-2520. For solid-phase yne-ene metathesis employing
catalyst 1, see: (b) Schu¨rer, S. C.; Blechert, S. Synlett 1998, 166-168. (c)
Schuster, M.; Blechert, S. Tetrahedron Lett. 1998, 39, 2295-2298.
(18) For recent ROM references, see: (a) Randall, M. L.; Tallarico, J.
A.; Snapper, M. L. J. Am. Chem. Soc. 1995, 117, 9610-9611. (b) Schneider,
M. F.; Blechert, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 411-412. (c)
Schneider, M. F.; Lucas, N.; Velder, J.; Blechert, S. Angew. Chem., Int.
Ed. Engl. 1997, 36, 257-259. (d) Snapper, M. L.; Tallarico, J. A.; Randall,
M. L. J. Am. Chem. Soc. 1997, 119, 1478-1479. (e) Tallarico, J. A.;
Bonitatebus, P. J.; Snapper, M. L. J. Am. Chem. Soc. 1997, 119, 7157-
7158. (f) Tallarico, J. A.; Randall, M. L.; Snapper, M. L. Tetrahedron 1997,
53, 16511-16520. (g) Cuny, G. D.; Cao, J.; Hauske, J. R. Tetrahedron
Lett. 1997, 38, 5237-5240. (h) Cao, J.; Cuny, G. D.; Hauske, J. R. Mol.
DiVers. 1998, 3, 173-179.
(8) Issues of poor selectivity plagued early CM efforts employing
“classical” catalysts. For a recent review of CM using “classical” catalysts,
see: Finkel’shtein, E. S.; Bykov, V. I.; Portnykh, E. B. J. Mol. Catal. 1992,
76, 33-52.
(9) (a) Rossi, R. Synthesis 1977, 817-836. (b) Banasiak, D. S. J. Mol.
Catal. 1985, 28, 107-115. (c) Crisp, G. T.; Collis, M. P. Aust. J. Chem.
1988, 41, 935-942. (d) Bykov, V. I.; Butenko, T. A.; Finkel’shtein, E. S.;
Henderson, P. T. J. Mol. Catal. 1994, 90, 111-116. (e) Bykov, V. I.;
Finkel’shtein, E. S. J. Mol. Catal. A 1998, 133, 17-37.
(10) Crowe, W. E.; Zhang, Z. J. J. Am. Chem. Soc. 1993, 115, 10998-
10999.
(19) For a novel variant of ROM including a tandem RCM reaction,
see: Stragies, R.; Blechert, S. Synlett 1998, 169-170.
(20) (a) Boger, D. L.; Chai, W.; Ozer, R. S.; Anderson, C.-M. Biorg.
Med. Chem. Lett. 1997, 7, 463-468. (b) Boger, D. L.; Chai, W. Tetrahedron
1998, 54, 3955-3970. (c) Boger, D. L.; Chai, W.; Jin, Q. J. Am. Chem.
Soc. 1998, 120, 7220-7225. (d) Giger, T.; Wigger, M.; Aude´tat, S.; Benner,
S. A. Synlett 1998, 688-691. (e) Bra¨ndli, C.; Ward, T. R. HelV. Chim.
Acta 1998, 81, 1616-1621.
(11) Crowe, W. E.; Goldberg, D. R. J. Am. Chem. Soc. 1995, 117, 5162-
5163.
(12) Crowe, W. E.; Goldberg, D. R.; Zhang, Z. J. Tetrahedron Lett. 1996,
37, 2117-2120.
(13) For the ruthenium-catalyzed CM of functionalized terminal olefins
with allyldimethylsilyl-derivatized polystyrene resin, see: Schuster, M.;
Lucas, N.; Blechert, S. Chem. Commun. 1997, 823-824.
(21) For a recent report from these laboratories, see: Hillmyer, M. A.;
Nguyen, S. T.; Grubbs, R. H. Macromolecules 1997, 30, 718-721 and
references therein.

60 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
Table 1. CM Reactions with Symmetrically Disubstituted Olefins
contrast, the use of neat allyl acetate provided only a marginal
amount (10%) of the desired cross-product (data not shown),
presumably due to the statistically favored dimerization of allyl
acetate by 1 dominating the catalytic cycle, and the likely
formation of a less stable ruthenium methylidene species.25 The
cis-2-butene-1,4-diol bis-trifluoroacetate26 (entry 6) afforded a
reduced yield of the homologated allylic trifluoroacetate 6 (63%,
2.8:1 E/Z), yet with an E/Z ratio approximating that of the allylic
acetate 5.
Direct reaction of benzoate 3 with 1,4-butenediol (entry 7,
Table 1) occurred in dichloromethane at room temperature to
yield allylic alcohol 7 (54%, 5:1 E/Z), despite the limited
solubility of the diol. Elevating the temperature led to apparent
decomposition of alkylidene 1. No improvement in the isolated
yield of 7 was observed when the reaction was conducted as a
homogeneous mixture in chloroform. Several diether derivatives
of cis-1,4-butenediol (entries 8-11) were found to provide better
CM yields and improved trans selectivity. For example, the bis-
tert-butyl,27 bis-trityl,28 bis-benzyl ether,29 and bis-TBS30 sub-
strates provided CM products with E/Z ratios ranging from 7:1
to 10:1. While no attempts were made to separate the olefin
stereoisomers in the present study, the increased trans selectivity
observed in the CM of benzoate 4 with the bis-TBS diol now
represents a synthetically useful protocol for the direct instal-
lation of E-allylic alcohol functionality.
product:
%
a
entry
substrate
equiv
E/Zb
1
2
3
4
5
6
7
8
9
R1 ) R2 ) CH2OAc (cis)
R1 ) R2 ) CH2OAc (cis)
R1 ) CH2OAc, R2 ) H
R1 ) CH2OAc, R2 ) H
R1 ) CH2OAc, R2 ) H
R1 ) R2 ) CH2OC(O)CF3 (cis)
R1 ) R2 ) CH2OH (cis)
R1 ) R2 ) CH2OtBu (cis)
R1 ) R2 ) CH2Otrityl (cis)
2
1
4
2
1
4
2
2
2
2
2
2
4
2
4
2
5: 89 4.7:1
5: 77
5: 81
5: 80
5:1
3:1
4:1
5: 59 5.7:1
6: 63c 2.8:1
7: 56d
8: 90
5:1
7:1
8:1
9:1
9: 75e
10: 71f
10 R1 ) R2 ) CH2OCH2Ph (cis)
11 R1 ) R2 ) CH2OTBS (cis)
12 R1 ) R2 ) CH2CH2CH3 (cis)
13 R1 ) R2 ) CH2NHBoc (cis)
14 R1 ) R2 ) CH2C(O)OMe (trans)
15 R1 ) R2 ) CH2C(O)NMe(OMe) (trans)
16 R1 ) R2 ) CH2CH2OTBS (trans)
11: 77g 10:1
12: 72
13: 71
3:1
3:1
14: 74 3.3:1
15: 17 1.9:1
16: 49g 2.8:1
a Isolated product yields. b Determined by 1H NMR integration.
c Yield determined after NEt3 deprotection of the allyl trifluoroacetate
ether (to afford allylic alcohol 7). d Reaction run at room temperature.
e Yield determined after the formic acid deprotection of the allyl trityl
ether to afford 7. f Yield determined after H2/Pd-C hydrogenation-
hydrogenolysis of allyl benzyl ether. g Yield determined after TBAF
deprotection of allyl TBS ether.
Results and Discussion
Purely aliphatic functionality could be readily incorporated
employing this CM methodology: for example, the CM of cis-
3-hexene with 4 (entry 12) yielded the ethyl functionalized
internal olefin cross-product 12 in good yield (72%, 3:1 E/Z).
The compatibility of nitrogen-containing substrates was next
probed through the CM of Boc-protected cis-1,4-diamino-
butene31 (entry 13). Boc-protected allylic amine 13 was isolated
in good yield (71%, 3:1 E/Z), which demonstrates CM as a
straightforward route to the introduction of nitrogen functional-
ity.
Initial Results. 9-decen-1-yl benzoate (3)22 was chosen as a
model terminal olefin substrate because of its low volatility and
its UV chromophore significantly aided synthetic manipulations.
Treatment of benzoate 3 with 1-2 equiv of a symmetric internal
olefin and 5 mol % ruthenium benzylidene 1 in refluxing
dichloromethane provided the desired CM products in good
yields (eq 4). The CM reactions proceeded largely to completion
All of the CM reactions discussed up to this point involved
cis-disubstituted internal olefins. We chose to employ cis olefins
at the outset because it had been observed previously that
ruthenium alkylidene 1 is more reactive toward the more
sterically accessible cis olefin.25 However, trans-disubstituted
internal olefins were also found to be reactive coupling partners
for CM with terminal olefins.32 Dimethyl trans-3-hexene-1,6-
dioate33 (entry 14) provided the desired homoallylic ester cross
product (14) as the major product (74%, 3:1 E/Z; recovered
over 12 h, and any benzoate homodimer side-product (4,
5-10%) could be easily recovered and recycled in a subsequent
cross-metathesis step. In all of the cases examined thus far, the
reaction has favored the formation of the trans olefin isomer.
(24) This is the exact opposite effect that Blechert et al. observed in the
ROM of cyclic olefins with monosubstituted olefins versus disubstituted
olefins. A large excess (up to 10-fold) of the less reactive disubstituted
olefin was required to suppress the ROMP of the strained cyclic olefin
substrates, while only 1 equiv of the corresponding monosubstituted olefin
was required to effect analogous yields. See ref 18b.
Our initial efforts focused upon elaborating benzoate 3 to
the corresponding allylic alcohol derivatives (Table 1).23 The
commercially available cis-2-butene-1,4-diol diacetate (entry 1)
provided the homologated allylic acetate 5 in excellent yield
(89%, 4.7:1 E/Z) using 2 equiv of internal olefin. When only 1
equiv of diacetate was employed, the yield of 5 decreased (77%)
and no significant change in the trans/cis ratio was observed
(entry 2). Interestingly, the use of 2 equiv of diacetate was found
to be more efficient than simply using 1, 2, or 4 equiv of allyl
acetate (entries 3-5).24 Employing the diol acetate as solvent
(55 equiv, 45 °C, 12 h) increased the isolated yield of 5 to 91%,
although with diminished trans olefin content (3:1 E/Z). In
(25) Ullman, M.; Grubbs, R. H. Organometallics 1998, 17, 2484-2489.
(26) Prepared according to a standard literature procedure: Lardon, A.;
Reichstein, T. HelV. Chim. Acta 1954, 37, 443-450.
(27) Prepared using a general method: Alexakis, A.; Gardette, M.; Colin,
S. Tetrahedron Lett. 1988, 29, 2951-2954.
(28) Prepared using a general method: Chaudary, S. K.; Hernandez, O.
Tetrahedron Lett. 1979, 2, 95-98.
(29) Prepared by a general procedure: Forster, R. C.; Owen, L. N. J.
Chem. Soc., Perkin Trans. 1 1978, 822-829.
(30) Prepared using a general method: Corey, E. J.; Venkateswarlu, A.
J. Am. Chem. Soc. 1972, 94, 6190-6191.
(31) Prepared according to a modified literature procedure. See: Zuwen,
H.; Nadkarni, D. V.; Sayre, L. M.; Greenaway, F. T. Biochim. Biophys.
Acta 1995, 1253, 117-127.
(32) This corroborated well with the observation of Blechert et al. that
trans-disubstituted internal olefins are reactive substrates for ROM. See
ref 18b.
(22) Prepared according to a general literature procedure: Schlessinger,
R. H.; Lopes, A. J. Org. Chem. 1981, 46, 5252-5253.
(23) Full experimental details of the CM and self-metathesis reactions
and full characterization of the products (1H NMR, 13C NMR, and HRMS)
can be found in the Supporting Information.
(33) Prepared according to a literature procedure: Gassman, P. G.;
Bonser, S. M.; Mlinaric-Majerski, K. J. Am. Chem. Soc. 1989, 111, 2652-
2662.

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 61
Scheme 1
Scheme 2
homodimer 4: 23%). However, in an attempt to introduce
Weinreb amide functionality through CM, we found that trans-
1,6-bis[methyl(methoxy)amido]hex-3-ene34 (entry 15) was a
poor substrate for CM, affording 15 in only 17% yield and with
poor trans selectivity (1.9:1 E/Z). As substantial homodimeric
cross-product 4 was not generated, we speculate that the
coordination of the amide to the catalyst was inhibiting the
catalytic reaction. Trans-1,6-bis(tert-butyldimethylsilyloxy)hex-
3-ene35 (entry 16) was likewise unreactive. These results are
consistent with the observations of Crowe et al., which indicated
that certain homoallylic substituents on terminal olefins can
deactivate catalytic CM reactions.10,11
Two-Step Procedure for Terminal Olefin Cross-Metath-
esis. These initial results suggested that either cis- or trans-
disubstituted olefins could be employed as efficient coupling
partners in CM reactions. Accordingly, we investigated the use
of a two-step procedure21,36 for terminal olefin CM as outlined
in Scheme 1. First, a terminal olefin was self-metathesized by
treatment with ruthenium alkylidene 1. The mixture of disub-
stituted olefin isomers generated was then subjected to CM with
another terminal olefin employing the methodology described
above. The synthesis of a large pool of functionally diverse,
homodimeric internal olefins Via this first self-metathesis
procedure is shown in Scheme 2.37
For the majority of the terminal olefin substrates studied,
homodimerization with 0.3 mol % 1 in Vacuo (25 °C, 24 h)
provided predominantly trans-disubstituted olefins in good to
excellent yields (Scheme 2). The solvent-free conditions, low
catalyst loading, and high yields make homodimerization Via
self-metathesis employing ruthenium alkylidene 1 an exceptional
methodology for the synthesis of high molecular weight,
symmetrical disubstituted olefins. Furthermore, most of the
homodimeric products were crystalline solids, which expedited
their purification from alkylidene 1.
Performing self-metathesis under vacuum has the benefit of
removing the stoichiometric gaseous byproduct of the reaction,
ethylene, and therefore pushes the self-metathesis reaction
toward completion. Employing a static vacuum that is periodi-
cally refreshed, the solvent-free method can also be used to
homodimerize more volatile substrates such as allylbenzene (27).
a Homodimers synthesized in solution (0.1 M, 5 mol % 1, 45 °C).
b 1.0 mol % 1 used.
Self-metathesis of the protected derivatives of 9-decen-1-ol (3,
19, 21, and 2338) afforded the corresponding homodimers (4,
20, 22, and 24) in excellent yields. Methyl 10-undecylenate (21)
and the ethylene glycol acetal of 10-undecenal (25) were also
found to undergo facile homodimerization reactions. In contrast,
homodimerization of neat, unprotected 9-decen-1-ol (17) gener-
ated only a modest yield of diol 18 with low trans selectivity.
This result is indicative of alcohol 17 potentially sequestering
catalyst 1 by chelation, and effectively shutting down the
catalytic cycle before a thermodynamic trans/cis ratio of
products was achieved.39
(34) Prepared Via a DCC coupling between trans-â-hydromuconic acid
and N,O-dimethylhydroxylamine hydrochloride. See: Nahm, S.; Weinreb,
S. M. Tetrahedron Lett. 1981, 22, 3815-3818.
(35) Prepared according to a general literature procedure: Zhdanov, R.
I.; Zhenodarova, S. M. Synthesis 1975, 222-245.
(36) Nubel, P. O.; Yokelson, H. B.; Lutman, C. A.; Bouslog, W. G.;
Behrends, R. T.; Runge, K. D. J. Mol. Catal. A 1997, 115, 43-50.
(37) The self-metathesis of terminal olefins employing “classical” olefin
metathesis catalysts has been utilized previously in the synthesis of
symmetrically disubstituted olefins. The majority of these applications
involved the synthesis of structurally simple, aliphatic internal alkenes.
See: (a) Marciniec, B.; Gulinski, J. J. Organomet. Chem. 1984, 266, C19-
C21. (b) Marciniec, B.; Maciejewski, H.; Gullinski, J.; Rzejak, Z. J.
Organomet. Chem. 1989, 362, 273-279. (c) Marciniec, B.; Pietraszuk, C.;
Foltynowicz, Z. J. Organomet. Chem. 1994, 474, 83-87.
Aromatic, organometallic, and sulfone-containing homodimer-
ic products (28, 30, 32, 34, and 36) could be prepared in
moderate to good yield Via self-metathesis (Scheme 2).40
(38) Prepared Via a DCC coupling between N-Boc-glycine-OH and
9-decen-1-ol.
(39) Moderate trans selectivity has been consistently observed in almost
all intermolecular metathesis accounts employing benzylidene 1 to date.
This selectivity is consistent with preferential formation of trans-R,â-
disubstituted metallocyclobutane intermediates. We made the assumption
that the predominant olefin regioisomer for these symmetrical homodimers
was trans in our NMR spectroscopic analyses.

62 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
Scheme 3a
Scheme 4a
1
a E/Z ratio determined by H and 13C NMR analyses.39
dimers by self-metathesis (Schemes 3 and 4). These experiments
further confirmed the exceptional functional group tolerance of
ruthenium alkylidene 1. Terminal olefin functionality can be
readily installed into amino acid side chains by the incorporation
of allyl ethers. (O)-Allyl ethers of protected L-serine (37),
L-homoserine (39), and L-tyrosine (41) derivatives were straight-
forward to prepare (Scheme 3), and upon treatment with
alkylidene 1 afforded good yields of their respective homodimers
with moderate trans selectivity (ca. 3:1 E/Z).43 As amino acid
derivatives 37 and 39 were low viscosity oils, self-metathesis
was performed in Vacuo as described above (Scheme 2); the
high viscosity of protected tyrosine derivative 41 required the
self-metathesis reaction to be performed in solvent for optimal
yield of homodimer 42 (71%). While side chain-bridged amino
acids 38, 40, and 42 could be generated Via self-metathesis,
treating Boc-L-allylglycine-OMe under the analogous reaction
conditions (in Vacuo or in CH2Cl2) yielded less than 5% of the
respective homodimer (data not shown); these data44 are
consistent with the observations of Gibson et al. that greater
separation between the olefin and amino acid is required for
efficient CM.16 Finally, in extending CM methodology to
carbohydrate substrates, we observed the crystalline 2,3,4,6-
tetra-O-benzyl-1-R-C-allylglucoside45 43 to undergo facile self-
1
a E/Z ratios determined by H and 13C NMR analyses.39
Volatile substrates such as allylpentafluorobenzene (29) and
1-ferrocene methanol (O)-allyl ether (33)41 can also be efficiently
homodimerized using the standard solution-phase CM condi-
tions introduced above (0.1 M, 5 mol % 1, 45 °C, ca. 12 h).
Notably, treatment of allylpentafluorobenzene (29) with ben-
zylidene 1 generated homodimeric product 30 with 16:1 E/Z
olefin content42 (by 1H NMR); reasons for this increased trans
selectivity remain to be discerned.
Due to the ongoing interest in our laboratory of employing
olefin metathesis in the context of peptide and carbohydrate
synthesis, we next turned our attention to the synthesis of a
series of novel amino acid, carbohydrate, and peptide homo-
(42) Evidence for assigning the major isomer as trans is based upon
comparison of 1H NMR and mp data (30: 91 °C, lit. mp 94-94.5 °C) with
the known E-1,4-bis(pentafluorophenyl)-2-butene: Filler, R.; Choe, E. W.
Can. J. Chem. 1975, 53, 1491-1495.
(43) (O)-Allyl ethers of L-serine, L-homoserine, and L-tyrosine have been
previously employed in the synthesis of peptide macrocycles Via RCM.
See: (a) Miller, S. J.; Blackwell, H. E.; Grubbs, R. H. J. Am. Chem. Soc.
1996, 118, 9606-9614. (b) Blackwell, H. E.; Grubbs, R. H. Angew. Chem.,
Int. Ed. 1998, 37, 3281-3284.
(44) O’Leary, D. J.; Miller, S. J.; Grubbs, R. H. Tetrahedron Lett. 1998,
39, 1689-1690.
(45) For the synthesis of C-allylglucoside 43, see: Lewis, M. D.; Cha,
J. K.; Kishi, Y. J. Am. Chem. Soc. 1982, 104, 4976-4978.
(40) Allyl benzene (27) has been previously reported by Benner et al.
to be an excellent substrate for the synthesis of combinatorial libraries Via
CM. See ref 20d.
(41) The allyl ether was introduced into 1-ferrocene methanol using a
standard literature procedure: Corey, E. J.; Suggs, J. W. J. Org. Chem.
1973, 38, 3224.

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 63
Scheme 5
Scheme 6
a Using 2.7 mol % 1. b Using 2.5 mol % 1.
metathesis in solution, affording the novel R,R-linked dimer 44
in high yield (93%, 4:1 E/Z).46
In an attempt to probe the general applicability of olefin self-
metathesis for the generation of more complex molecular
architectures, we introduced terminal olefin functionality into
a hydrophobic pentapeptide framework (45) through incorpora-
tion of L-serine (O)-allyl ether.43 Treatment of pentapeptide
alkene 45 with ruthenium alkylidene 1 under standard solution-
phase self-metathesis conditions generated the side chain-bridged
homodimer 46 in good yield. Interestingly, the trans/cis
selectivity appeared to approximate that for the CM of the free
amino acid (37). The facile synthesis of 46 demonstrates the
utility of self-metathesis methodology for the synthesis of
unique, acyclic peptidic architectures containing non-native
C-C linkages; cyclic peptide olefin counterparts have been
previously generated employing the intramolecular metathesis
variant, RCM.47
Finally, to demonstrate the efficacy of our two-step CM
protocol, selected symmetrical disubstituted olefins prepared by
self-metathesis (Schemes 2-4) were further processed in the
CM with terminal olefins. We selected internal and terminal
olefin substrates which exhibited varied functionalities and
sterics to explore the scope and limitations of this two-step
procedure (and selective CM overall). Accordingly, many of
the substrates were based upon the structurally diverse amino
acid, carbohydrate, and peptide structures shown in Schemes 3
and 4.48 Furthermore, several of the terminal olefins in Schemes
3 and 4 were utilized as substrates for CM, either with self-
metathesized homodimers or simple cis-1,4-butenediol substrates
from Table 1. Representative examples of this second CM
processing step are shown in Schemes 5 and 6. Overall, the
heterodimeric products were formed selectively and with
moderate to good trans selectivity. Standard solution-phase CM
conditions were employed throughout (0.1 M in terminal olefin,
2 equiv disubstituted olefin, CH2Cl2, 5 mol % 1, 45 °C), and
the reactions were generally complete in 12 h.
Allyl benzene homodimer 28 was reacted with 9-decen-1-yl
benzoate 3 (Scheme 5) to yield the benzyl-functionalized internal
olefin 47 in good yield (68%, 3.7:1 E/Z). Allyl benzene (27)
itself was also demonstrated to undergo facile CM with
butenediol diacetate (87%, 3:1 E/Z). 9-Decen-1-yl Boc-glycinate
23 was observed to react with 9-decen-1-yl acetate homodimer
20 to afford the differentially functionalized 9-eicosene (49) in
good yield (72%, 3.5:1 E/Z). Finally, the homodimer of allyl
phenyl sulfone was found to be a very reactive CM partner,
providing heterodimer 50 in high yield (90%, 7:1 E/Z).
As illustrated in Scheme 6, treatment of Boc-L-serine(O-allyl)-
OMe (37) with bis-acetate 20 generated lipophilic amino acid
derivative 51 in high yield with improved trans selectivity (86%,
6:1 E/Z). The related lipophilic sugar 52 could be prepared in
similar fashion through the CM of C-allylglucoside 43 with bis-
acetate 2 (73%, 2.8:1 E/Z). CM of glucoside 43 with Boc-L-
serine(O-allyl)-OMe homodimer 38 generated a low yield of
the amino acid/sugar heterodimer 53 (37%, 5:1 E/Z).49 In
analogy to the synthesis of TBS-ether 11 (Table 1), silyl ether
derivatized sugar 54 was generated in good yield with pro-
nounced trans selectivity (70%, 9:1 E/Z).
(46) For recent CM and RCM applications in carbohydrate synthesis,
see: (a) Feng, J.; Schuster, M.; Blechert, S. Synlett 1997, 129-130. (b) El
Sukkari, H.; Gesson, J.-P.; Renoux, B. Tetrahedron Lett. 1998, 39, 4043-
4046. (c) Fu¨rstner, A.; Mu¨ller, T. J. Org. Chem. 1998, 63, 424-425. (d)
Calimente, D.; Postema, M. H. D. J. Org. Chem. 1999, 64, 1770-1771.
(e) Postema, M. H. D.; Calimente, D. Tetrahedron Lett. 1999, 40, 4755-
4759.
To study the scope of CM in the functionalization of more
complex substrates, we chose to investigate the CM of penta-
peptide 45 with disubstituted internal olefins (Scheme 7).
Treatment of 45 with the 9-decen-1-yl Boc-glycinate dimer 24
under standard solution-phase CM conditions afforded the
(47) For leading references to peptide RCM, see refs 5a and 43.

64 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
Scheme 7
Scheme 8
glycinate functionalized peptide 55 in moderate yield with
modest trans selectivity (66%, 5.5:1 E/Z). This reaction further
demonstrated the general applicability of CM as mild methodol-
ogy for the introduction of diverse functionality tethered through
a carbon-carbon bond. In summary, it appeared that our two-
step procedure for the CM of selected internal olefins was not
only effective but also, in view of the wide variety of substrates
studied, substantially broad in scope.
how the steric bulk of the allylic-substituted terminal olefin
effected the regioselectivity of CM, we prepared 3-buten-2-O-
tert-butyldiphenylsilyl ether (60).51 CM of ether 60 with the
bis-acetate generated a low yield of cross-product 61 with
reduced trans selectivity (61, 7.5:1 E/Z). This result suggested
that the increased steric bulk of silyl ether 60 in comparison to
benzoate 56 had reduced its reactivity with catalyst 1; the
concomitant loss of trans selectivity, however, indicated that
steric bulk was not the only factor governing stereoselective
CM.
Cross-Metathesis of Terminal Olefins with Allylic Methyl
Substituents. To further study the influence of steric effects
on the trans/cis selectivity of CM employing catalyst 1, a series
of CM reactions was conducted on terminal olefins with allylic
methyl substituents (Scheme 8).50 At the outset of our studies,
it was anticipated that allylic substitution would introduce steric
hindrance close to the olefin group, and potentially direct CM
toward the less sterically hindered trans product. This was
consistently observed in the small group of substrates studied.
The increased steric hindrance of the terminal olefin, however,
was also observed to significantly reduce the CM yields.
3-Buten-2-yl benzoate (56)22 was selected as the initial model
terminal olefin substrate. CM of 56 with 2 equiv of cis-1,4-
butenediol bis-acetate under standard solution-phase conditions
generated cross-product 57 with high trans selectivity, albeit
in modest yield (30%, 16:1 E/Z). This result was in direct
contrast to that observed for the CM of 9-decen-1-yl benzoate
(3) and the bis-acetate (Table 1, entry 1), where the cross-
product 5 was generated in considerably higher yield with lower
trans selectivity (89%, 4.7:1 E/Z), indicating that allylic methyl
substituents do direct trans selective CM. Notably, the E/Z ratio
for heterodimer 57 was almost four times greater than that for
5.
Reactivity of Disubstituted Olefins versus Monosubstituted
Olefins. The results presented thus far demonstrate that both
cis- and trans-disubstituted olefins are effective substrates for
CM. Employing an excess of the disubstituted olefin relative
to the terminal olefin component in CM was observed to lead
to good yields of the desired heterodimeric product. In certain
cases, we observed higher yields employing the disubstituted
olefins instead of the monosubstituted counterpart. While using
an excess of one olefin component should statistically push the
reaction toward the heterodimeric product (if both olefins have
comparable reactivities), we believed at the outset that employ-
ing an excess of the disubstituted olefin component in CM would
statistically favor formation of an alkyl-substituted ruthenium
alkylidene over the unsubstituted ruthenium methylidene.
Because the methylidene formed from 1 had been shown to
decompose considerably faster than other ruthenium alkylidene
species,25 we anticipated that preferential formation of an
alkylidene species employing substituted olefins would extend
the metathesis activity of 1 and potentially lead to higher yields
of the desired heterodimeric product. As illustrated in Scheme
9, when one reactant is a disubstituted alkene it is possible to
have a CM catalytic cycle that does not involve a methylidene
intermediate. Alternatively, when both reactants are terminal
olefins (as indicated by the (H) adjacent to the second R2 in
Scheme 9), the catalytic cycle generates a ruthenium meth-
ylidene for each productive CM reaction.
CM of allyl-substituted benzoate 56 with the bulky cis-1,4-
butenediol bis-OTBS gave similar results, affording the coupling
product 58 in improved yield with dramatic trans selectivity
(54%, 47:1 E/Z). Again, the E/Z ratio for 58 was approximately
four times greater than that observed for the CM of 9-decen-
1-yl benzoate (3) with the bis-OTBS compound (11, 10:1 E/Z,
Table 1, entry 11). This combination of the allylic methyl
substitution on 56 and the bulky TBS protecting groups
generated a very selective CM reaction. In attempting to discern
To further probe the use of disubstituted olefins in CM
reactions, we conducted four side-by-side CM reactions with
methyl 10-undecenylate (21), 5 mol % 1, and 2 equiv each of

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 65
Scheme 9
but steady rate in the CM with allyl OTBS. Analogous results
for homodimer formation 22 in the CM of the acetate series
were also obtained (data not shown). From these results, it
appeared that secondary metathesis of the heterodimers 62 and
63 and homodimer 22 was not occurring at an appreciable rate.
Interestingly, the trans/cis ratio of heterodimer 63 was observed
to be higher throughout the course of the reaction with the bis-
OTBS substrate relative to allyl OTBS (Figure 2b). A higher
trans/cis ratio was also observed in the formation of acetate
heterodimer 62. While definitive reasons for the latter ratio were
not realized, the qualitative GC-MS analysis of these four
reactions strongly suggested that employing these disubstituted
olefins provided more chemo- and stereoselective CM relative
to reactions employing monosubstituted olefins.
Catalyst Initiation Rates. After the bis-OTBS substrate was
found to provide enhanced CM reaction rates and yields, we
next became interested in determining a qualitative measure of
catalyst initiation rates for this and several other olefin substrates.
Although a study of the initiation rates of ruthenium alkylidene
1 by a variety of alkyl-substituted olefins has been published,25
we were uncertain as to the effect of allylic oxygen functionality
on these rates.
(1) allyl acetate, (2) allyl OTBS, (3) cis-1,4-butenediol bis-
acetate, and (4) cis-1,4-butenediol bis-OTBS (eq 5). The
1
Three reactions (Scheme 10) were examined by H NMR
spectroscopy, and the results are shown in Figure 3. For these
experiments, the temperature and concentration of catalyst and
substrate was adjusted to provide reasonable rates and signal-
to-noise ratios for NMR quantitation. Benzylidene 1 was
initiated by terminal olefin 21 and allyl-OTBS at approximately
the same rate, with 90% of the catalyst initiated within 5 min
at room temperature. In contrast, the bis-OTBS substrate reacted
with benzylidene 1 at a slower rate, requiring approximately
90 min to reach 90% initiation. The gradual increase in
concentration of the methylidene species 65 was largely a
consequence of the experimental conditions: the gaseous
reaction byproduct ethylene was not being purged as it normally
would under open reflux conditions.52 We conclude from these
studies that terminal olefins bearing allylic oxygen functionality
initiated the catalyst with a rate comparable with that of
“isolated” terminal olefins (i.e., olefins far removed from other
functional groups). Therefore, we believed the benefit of using
disubstituted olefins in certain CM applications was the result
of minimizing the available pool of terminal olefins which can
give rise to the less stable ruthenium methylidene species.
advantage of employing certain internal olefins in CM reactions
is evident from the graphs shown in Figures 1 and 2.
Reaction profiles for the formation of heterodimeric CM
products 62 and 63 and subsequent disappearance of starting
material 21 are shown in Figure 1. For both the acetate (Figure
1a,b) and the OTBS (Figure 1c,d) series, CM with the
disubstituted olefin afforded higher yields of the respective
heterodimer product. In the case of the bis-acetate substrate,
almost all of olefin 21 is consumed in 2 h, while consumption
takes almost 6 h for the allyl acetate reaction. For the silyl ether
substrates, the starting material 21 was consumed considerably
faster but the reactivity pattern was similar: 21 was consumed
in under 15 min for the bis-OTBS reaction, while it took almost
2 h for the reaction of olefin 21 with allyl OTBS to proceed to
completion.
The enhanced reaction rates observed with the disubstituted
olefins could be due to a statistical effect, in that 2 equiv of a
disubstituted olefin has twice the number of potential alkylidenes
to transfer when compared with 2 equiv of a terminal olefin.
To examine this possibility, a reaction employing 4 equiv of
the allyl OTBS substrate was subjected to a rate study (Figure
1c). This reaction was observed to proceed at a slower overall
rate and provide a slightly diminished yield when compared
with the reactions employing 2 equiv of allyl OTBS or 2 equiv
of the bis-OTBS substrate. This former effect is most likely a
consequence of the catalytic ruthenium species being engaged
in the unproductive self-metathesis of the excess allyl OTBS
starting material.
CM Reactions Involving Two “Isolated” Olefins. We
discovered in later studies with more complex substrates that
the benefit of employing disubstituted olefins in CM did not
appear to be general. While a systematic study was not
conducted, we observed that as the number of carbon units
between the olefin and any functional or sterically bulky group
was increased, CM of the two monosubstituted olefins, instead
of one monosubstituted and one disubstituted olefin, afforded
competitive yields of the desired heterodimeric cross-product.
For example, cross-coupling reactions using 9-decen-1-yl N-Boc
glycinate (23) and various equivalents of 9-decen-1-yl acetate
(19) or the internal olefin homodimer 20 demonstrated no benefit
to employing the disubstituted olefin for CM instead of the
corresponding terminal olefin (Scheme 11). Using 1 or 2 equiv
of olefin 19 or 20 afforded similar yields of heterodimer 49.
Furthermore, employing 0.5 equiv of disubstituted olefin 20 was
not analogous to 1 equiv of monosubstituted olefin 19 (28% vs
45% yield of heterodimer 49), which indicated the lower overall
reactivity of the disubstituted olefin. While these data did not
invalidate our two-step CM procedure described above, it did
suggest that, in the case of structurally “isolated” olefins, the
In monitoring the formation of methyl ester homodimer 22
for the OTBS reaction series (Figure 2a), a negligible amount
of 22 was formed in the reaction with the bis-OTBS substrate,
while the formation of homodimer 22 was occurring at a slow,

66 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
Figure 1. GC-MS reaction profiles for the CM of alkene 21 with mono- and disubstituted olefins. (a) Concentration of heterodimeric product 62
formed vs reaction time in the CM of 21 and allyl acetate and cis-1,4-butenediol bis-acetate. (b) Disappearance of 21 vs reaction time in the CM
of 21 and allyl acetate and cis-1,4-butenediol bis-acetate. (c) Concentration of heterodimeric product 63 formed vs reaction time in the CM of 21
and allyl OTBS (2 and 4 equiv) and cis-1,4-butenediol bis-OTBS (2 equiv). (d) Disappearance of 21 vs reaction time in the CM of 21 and allyl
OTBS and cis-1,4-butenediol bis-OTBS. Data obtained from GC-MS analysis of reaction aliquots (1,4-dichlorobenzene as internal standard, data
corrected for relative response).
first self-metathesis step was not essential. Finally, the compa-
rable yields of 49 obtained when equivalent amounts of either
the mono- or disubstituted olefin (19 or 20) were employed
suggested that preferential formation of a ruthenium alkylidene
over the methylidene may not be governing the reaction outcome
in the CM of structurally “isolated” olefins.
These results were in contrast to those observed in the GC-
MS and 1H NMR analysis of CM with protected allylic alcohols
above (Figures 1-3), indicating that functionality allylic to the
olefin could be influencing metathesis. We speculate that the
heightened heterodimer yields employing disubstituted olefins
with electron-donating allylic functionality are related to the
alkylidenes generated upon metathesis. Relatively bulky, electron-
donating substituents on alkylidenes have been shown previously
to accelerate metathesis processes;25 therefore, the preferential
formation of a more active alkylidene employing an excess of
the disubstituted olefin could be the reason for the observed
higher CM yields.53-55
CM with terminal olefins yielding protected R,â-unsaturated
aldehydes. The preparation of R,â-unsaturated aldehydes has
been accomplished previously by Wittig56 homologation of
aldehydes employing reagents such as Ph3PdCHCHO57 or with
acetal58 or imine59 protected two-carbon ylides. Addition-
elimination methods have also been used to homologate
aldehydes.60-62 In cases where a terminal olefin is serving as
(48) Terminal olefin derived sugars and amino acids have previously
been employed in CM with other terminal olefins. See refs 13 and 46b.
(49) For recent syntheses of carbon-carbon linked glycosyl amino acids,
see: (a) Dondoni, A.; Marra, A.; Massi, A. J. Chem. Soc., Chem Commun.
1998, 1741-1742. (b) Dondoni, A.; Massi, A.; Marra, A. Tetrahedron Lett.
1998, 39, 6601-6604. (c) Hu, Y.-J.; Roy, R. Tetrahedron Lett. 1999, 40,
3305-3308.
(50) Racemic allylic-substituted terminal olefins were employed in these
CM studies.
(51) Prepared according to a literature procedure: Hanessian, S.; Lavallee,
P. Can. J. Chem. 1975, 53, 2975-2977.
(52) The methylidene content was found to diminish when the NMR
tubes were periodically purged with argon.
(53) Chelation of functionality in the allylic position of the metal
alkylidene to the metal center should be disfavored because this would form
a strained, four-membered ring.
(54) The poorer performance of disubstituted “isolated” olefins in CM
reactions could also be due to a slower rate of CM, a rate that becomes
competitive with the intrinsic rate of catalyst decomposition.
Cross-Metathesis of Terminal Olefins with Acrolein
Acetals. In the course of examining the activity of substrates
for CM with allylic oxygen functionality, we discovered that
certain acrolein acetals were particularly robust substrates for

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 67
Figure 3. Catalytic species formed by reacting catalyst 1 (10 mM)
with either methyl undecylenate 21, allyl OTBS, or cis-1,4-butenediol
bis-OTBS, each 100 mM in CD2Cl2 at 22 °C. Alkylidene formation in
methyl undecylenate reaction (filled diamonds), allyl OTBS reaction
(filled squares), and cis-1,4-butenediol bis-OTBS (filled circles).
Methylidene formation in methyl undecylenate reaction (diamonds),
allyl OTBS reaction (squares). Percent of total catalytic species
1
determined by H NMR integration of carbene resonances at 18-20
ppm.
Scheme 10a
Figure 2. GC-MS reaction profiles for the CM of alkene 21 with allyl
OTBS and cis-1,4-butenediol bis-OTBS. (a) Concentration of homo-
dimeric product 22 formed vs reaction time in the CM of 21 and allyl
OTBS and cis-1,4-butenediol bis-OTBS. (b) Trans/cis ratios of hetero-
dimeric product 63 vs reaction time in the CM of 21 and allyl OTBS
and cis-1,4-butenediol bis-OTBS. Data obtained from GC-MS analysis
of reaction aliquots (1,4-dichlorobenze as internal standard, data
corrected for relative response).
an aldehyde precursor, a cross-metathesis approach offers a
direct means for homologation (Scheme 12).
Although acrylonitrile has been successfully employed in
molybdenum alkylidene 2 catalyzed CM reactions,11 conjugated
a L ) ligand on metal center.
(55) Another role of the disubstituted olefin may be to limit formation
of a particular metallacycle that reduces catalytic efficiency because it is
either less unreactive or readily decomposes. Because this type of role relates
directly to the nature of the olefin substituent, it appears to be more
consistent with the unique behavior exhibited by olefins containing proximal
polar or sterically bulky groups. Two possibilities for the “unfavored”
metallacycle, requiring a monosubstituted olefin for its formation, are as
follows:
olefins including acrolein were found to be unreactive in
reactions using catalytic ruthenium benzylidene 1. Unconjugated
acrolein acetals, on the other hand, were found to be viable
metathesis substrates.63,64 Our initial investigations employed
(59) Meyers, A. I.; Tomioka, K.; Fleming, M. P. J. Org. Chem. 1978,
43, 3788-3789.
(60) Wollenberg, R. H.; Albizati, K. F.; Peries, R. J. Am. Chem. Soc.
1977, 99, 7365-7367.
(61) Wittig, G.; Reiff, H. Angew. Chem., Int. Ed. Engl. 1968, 80, 8-15.
(62) Meyers, A. I.; Nabeya, A.; Adickes, H. W.; Politzer, I. R.; Malone,
G. R.; Kovelesky, A. C.; Nolen, R. L.; Portnoy, R. C. J. Org. Chem. 1973,
38, 36-56.
(63) For a recent example of an acrolein acetal used in an RCM reaction,
see: (a) Crimmins, M. T.; King, B. W. J. Am. Chem. Soc. 1998, 120, 9084-
9085. For the recent report of the ROM of cyclopropenone ketal with
terminal olefins, see: (b) Michaut, M.; Parrain, J.-L.; Santelli, M. Chem.
Commun. 1998, 2567-2568.
(56) For recent reviews, see: (a) Kelly, S. E. In ComprehensiVe Organic
Synthesis; Trost, B. M., Ed.; Pergamon Press: New York, 1991; Vol. 1,
Chapter 3, pp 755-782. (b) Maryanoff, B. E.; Reitz, A. B. Chem. ReV.
1989, 89, 863-927.
(57) Bestmann, H. J.; Vostrowsky, O.; Paulus, H.; Billman, W.; Stransky,
W. Tetrahedron Lett. 1977, 121-124.
(58) Daubresse, N.; Francesch, C.; Rolando, C. Tetrahedron 1998, 54,
10761-10770.

68 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
Scheme 11a
Scheme 13
Scheme 14
The reactivity of acetal 67 was unexpected, as it was
originally thought that allylic disubstitution would hinder the
CM reaction. Extending this methodology to substrates with
allylic trisubstitution could, in principle, provide access to
additional functional groups such as R,â-unsaturated esters and
methyl ketones. However, attempts at ruthenium-catalyzed CM
of benzoate 3 with ortho ester 7066 or ketal 7167 proved
unsuccessful, potentially due to their relative steric bulk in
comparison to acetal 67 above.
a 5 mol % 1, CH2Cl2, 45 °C, 12 h.
Scheme 12
CM reactions between terminal olefin 3 and 2-vinyl-1,3-
dioxolane (72), a commercially available acrolein acetal with
enhanced acid-stability compared to diethyl acetal 67, gave
excellent yields of the dioxolane-protected R,â-unsaturated
aldehyde 73 (Scheme 14). For example, a 74% isolated yield
of protected aldehyde 73 was obtained with the catalyst loading
reduced to as little as 1 mol % 1. The yield could be improved
to 93% (7:1 E/Z) with a catalyst loading of 2.5 mol % 1 (Scheme
14), with the reaction proceeding to 90% conversion after 3
h.1b
Attempting to build upon our earlier work, which demon-
strated certain advantages to using symmetrically disubstituted
olefins as CM partners (see above), we prepared fumaraldehyde
bis(ethylene glycol acetal) (74)68 by homodimerization of vinyl
dioxolane 72 employing ruthenium benzylidene 1 under standard
solution-phase CM reaction conditions (Scheme 15). However,
bis-acetal 74 did not prove as reactive as vinyl dioxolane 72 in
CM reactions with terminal olefin 3, presumably due to steric
factors. Interestingly, however, the trans/cis ratio improved when
the commercially available acrolein diethyl acetal (67) and our
standard terminal olefin substrate, 9-decen-1-yl benzoate (3)
(Scheme 13). R,â-Unsaturated aldehyde 69 was obtained in 82%
yield using 2.5 mol % 1 and 2 equiv of acetal 67. Although the
acid-sensitive diethyl acetal cross-metathesis product (68) could
be isolated with chromatography employing Et3N-treated silica
gel, it was more convenient to recover the R,â-unsaturated
aldehyde 69 after formic acid hydrolysis. Using a Luche
reduction,65 R,â-unsaturated aldehyde 69 was converted to
allylic alcohol 7 in an efficient and highly trans-selective manner
(Scheme 13). This three-step allylic alcohol synthesis was an
improvement, in terms of both yield and trans selectivity, upon
our CM procedure using protected cis-2-butene-1,4-diols de-
scribed above (Table 1).
(66) Prepared Via a literature procedure. See: Gassman, P. G.; Chavan,
S. J. Chem. Soc., Chem. Commun. 1989, 837-839.
(67) Prepared Via a literature procedure. See: Gassman, P. G.; Burns,
S. J.; Pfister, K. B. J. Org. Chem. 1993, 58, 1449-1457.
(68) Cyclic diacetals of fumaraldehyde have been prepared previously.
See: Sokolov, G. P.; Hillers, S. Khim. Geterotsikl. Soedin. 1969, 1, 32-
35.
(64) Allylbenzene has recently been reported to efficiently undergo CM
reactions with acrolein, acrolein dimethyl acetal, and acrylonitrile catalyzed
by ruthenium benzylidene 1. We have not observed productive CM with
olefin 3 and either acrolein or acrylonitrile in our laboratory. See: Blanco,
O. M.; Castedo, L. Synlett 1999, 557-558.
(65) Gemal, A. L.; Luche, J. L. Tetrahedron Lett. 1981, 4077-4080.

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 69
Scheme 15a
Scheme 16. Synthesis of Tartrate-Derived Acetals, Vinyl
Boronate, and Epoxide Substrates via CM
a Reagents and conditions: (a) 0.2 M in acetal, 5 mol % 1, CH2Cl2,
45 °C. (b) 2 equiv of acetal component, 0.2 M 3, 5 mol % 1, CH2Cl2,
45 °C. (c) HCO2H-CH2Cl2 (1:8), 25 °C.
the hindered bis-acetal 74 was employed (73: E/Z ) 9.7:1 from
74; 73: E/Z ) 7:1 from 72).
Application of this CM acetal methodology to the construction
of â,γ-unsaturated aldehydes Via CM was next explored
(Scheme 15). Unfortunately, the homologue of acetal 67,
3-butenal diethyl acetal 75, did not appear to be a promising
substrate for CM. CM of benzoate 3 with 3-butenal diethyl acetal
75 or its homodimer 76, followed by acetal hydrolysis afforded
only low yields of the â,γ-unsaturated aldehyde 78 with poor
trans/cis selectivity. These results could again be rationalized
by the observation that certain homoallylic substituents on
terminal olefins deactivate catalytic CM.10,11 Specifically,
formation of the alkylidene from acetal 75 allows for the
formation of a potential five-membered chelate between one
of the oxygens of 75 and the ruthenium 1 metal center, which
could act to inhibit catalysis.
Given the success of vinyl dioxolanes in CM reactions,
structural variations of the five-membered ring were also
examined. Vinyl cyclopentane (84) was tested as a CM substrate
to compare the carbocyclic five-membered-ring system with the
dioxolane systems above. Olefin 84 was found to provide a CM
product in good yield and with E-selectivity (85: 66%, 7:1 E/Z).
These yields were interesting when compared with results
described earlier for the CM of terminal olefins with allylic
methyl substituents: for example, 3-butene-2-ol derivatives 56
and 60 were observed to undergo CM reactions with difficulty
(Scheme 8). In contrast, the results obtained with cyclic
substrates 72, 79, 82, and 84 revealed that ring-constrained
allylic disubstitution can be accommodated in CM reactions
initiated by ruthenium benzylidene 1.
Extending the scope of the reaction to include asymmetric
acrolein acetals was considered worthwhile because chiral R,â-
unsaturated acetals are useful synthetic intermediates.69 Ac-
cordingly, diethyl vinylidene-L-tartrate (79) was prepared70 and
found to provide a trans selective (6.7:1 E/Z by 1H NMR) CM
product 80 in excellent yield (86% after acetal hydrolysis)
approximating that of vinyl dioxolane CM product 73 (Scheme
16).71 Due to difficulties encountered in product purification
for this combination of substrates, a variation using the dimethyl
vinylidene-L-tartrate (82) and TBS-protected 9-decen-1-ol (81)
was examined to determine an isolated yield for the CM
reaction. Satisfyingly, the CM reaction was found to proceed
in 94% isolated yield of cross product 83 with 6:1 trans/cis
selectivity. The use of asymmetric acrolein equivalents, coupled
with emergent asymmetric metathesis catalysts,72 suggests a
novel means for effecting catalytic kinetic resolutions Via CM.
If ring constraint was an important factor for the success of
these CM reactions, we reasoned that allylic epoxides such as
butadiene monoxide (86) should be viable CM substrates.
Unfortunately, this substrate did not react to any appreciable
(69) For an example of an asymmetric Simmons-Smith reaction using
tartrate-derived R,â-unsaturated acetals, see: Mori, A.; Arai, I.; Yamamoto,
H. Tetrahedron 1986, 42, 6447-6458.
(72) (a) Fujimura, O.; dela Mata, F. J.; Grubbs, R. H. Organometallics
1996, 15, 1865-1871. (b) Fujimura, O.; Grubbs, R. H. J. Am. Chem. Soc.
1996, 118, 2499-2500. (c) Fujimura, O.; Grubbs, R. H. J. Org. Chem.
1998, 63, 824-832. (d) Alexander, J. B.; La, D. S.; Cefalo, D. R.; Hoveyda,
A. H.; Schrock, R. R. J. Am. Chem. Soc. 1998, 120, 4041-4042. (e) La,
D. S.; Alexander, J. B.; Cefalo, D. R.; Graf, D. D.; Hoveyda, A. H.; Schrock,
R. R. J. Am. Chem. Soc. 1998, 120, 9720-9721.
(70) Tsuzuki, T.; Koyama, M.; Tanabe, K. Bull. Chem. Soc. Jpn. 1967,
40, 1008-1013.
(71) The trans/cis ratio was determined by 1H NMR of the crude tartrate
CM product 80. The yield of tartrate CM product 80 was determined after
acid hydrolysis of the acetal to afford aldehyde 69.

70 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
Blackwell et al.
extent. Other epoxide-containing substrates can be employed
in CM reactions, however, as evidenced by the successful
reaction of allyl glycidyl ether (88) with benzoate 3 to form
cross-product 89 in 61% yield (3:1 E/Z).
Experimental Section
General Experimental Details. NMR spectra were recorded on
either a JEOL GX-400, Bruker Avance-400, or Bruker AM-500
spectrometer. Chemical shifts are reported in parts per million (ppm)
downfield from tetramethylsilane (TMS) with reference to internal
solvent. Multiplicities are abbreviated as follows: singlet (s), doublet
(d), triplet (t), quartet (q), quintet (quint), and multiplet (m). The reported
1H NMR data refer to the major olefin isomer unless stated otherwise.
The reported 13C NMR data include all peaks observed and no peak
assignments were made. Optical rotations were recorded on a Jasco
DIP-1000 digital polarimeter at 589 nm and are reported as [R]D
(concentration in grams/100 mL of solvent). Low- and high-resolution
mass spectra were provided by either the Southern California Mass
Spectrometry Facility (University of California, Riverside) or the UCLA
Mass Spectrometry Facility (University of California, Los Angeles).
Analytical thin-layer chromatography (TLC) was performed using
silica gel 60 F254 precoated plates (0.25 mm thickness) with a
fluorescent indicator. Flash column chromatography was performed
using silica gel 60 (230-400 mesh) from EM Science.80 cis-3-Hexene
was purchased from Chemsampco, Gray Court, SC. All other chemicals
were purchased from the Aldrich, Strem, or Nova Biochem Chemical
Companies, and used as delivered unless noted otherwise. CH2Cl2 was
purified by passage through a solvent column prior to use.81 Catalyst
1 was prepared according to a published procedure.6
Finally, in view of the success of vinyl dioxolanes in CM
reactions, a cyclic vinyl boronate was tested as a CM substrate.73
A pinacol-derived vinyl boronate (90)74 was found to react with
terminal olefin 3 to furnish cross-product 91 in good yield (67%)
and with excellent trans-selectivity (>20:1 E/Z). Their use in
CM provides a novel “one-step” method for converting terminal
olefins into substrates for the Suzuki coupling reaction, a
transformation with proven utility in complex natural product
total syntheses.75,76 Efforts to extend the use of vinyl boronates
in CM applications are currently underway in these laboratories.
Summary and Future Prospectives
In conclusion, cross-metathesis reactions involving internal
disubstituted olefins and certain terminal olefins with allylic
disubstituton appear to be a promising method for the direct
homologation of terminal olefins. The desired heterodimeric
cross-products could be generated in good to excellent yields
employing 1 equiv of terminal olefin, a 2-fold excess of the
second component, and 1-5 mol % ruthenium benzylidene 1.
Furthermore, the cross-metathesis reactions were shown to be
systematically more trans selective as the steric bulk at the
allylic position of either the internal olefin or the terminal olefin
was increased. Details of the current rationale behind the
improved chemoselectivity of allylic oxygen functionalized
olefins have been presented. The cross-metathesis methodology
described herein should be of particular use for the function-
alization of advanced intermediates in organic syntheses, for
the synthesis of diverse combinatorial libraries, and for the
construction of dimeric molecules for use as tools in molecular
biology.77 The CM homodimerization procedure employing
benzylidene 1 also allows rapid access to functionally diverse
chain transfer agents for the synthesis of novel telechelic
polymers by ROMP. Future work is directed toward the
installation of other functional groups Via CM such as protected
phosphorus,78 sulfur,79 and alkyne functionality, all of which
allow for further post-metathesis synthetic manipulation. Routes
toward dendritic architectures Via selective CM are also being
pursued in our laboratory. Finally, the simplicity and power of
CM as an intermolecular carbon-carbon bond forming reaction
is only now being appreciated; we anticipate that as selective
CM routes are disclosed, the volume of CM applications in
synthesis will dramatically escalate.
Peptide Synthesis. N-Boc-L-serine(O-allyl) methyl ester (36), N-Boc-
L-homoserine(O-allyl) methyl ester (38), and N-Boc-L-tyrosine(O-allyl)
methyl ester (37) were prepared according to a modified literature
procedure.82 Peptide 44 was synthesized by conventional solution-phase
synthesis methods using a racemization free fragment condensation
strategy. Couplings were mediated by N,N-dicyclohexylcarbodiimide
(DCC)/1-hydroxybenzotriazole (HOBT).83 The Boc group was used to
protect the N-terminus, and the C-terminus was protected as a methyl
ester. Deprotections were performed using 1:1 trifluoroacetic acid/
CH2Cl2 and saponification, respectively. All intermediates were char-
1
acterized by H NMR and TLC, and if necessary purified by column
chromatography on silica gel.
Representative Procedure for Solution-Phase Cross-Metathesis
Reaction. Compound 5. 9-Decen-1-yl benzoate (3) (69 µL, 0.25 mmol)
was added Via syringe to a stirring solution of cis-1,4-bis(acetyloxy)-
but-2-ene (79 µL, 0.5 mmol) and 1 (21 mg, 0.025 mmol, 10 mol %) in
CH2Cl2 (2.5 mL). The flask was fitted with a condenser and refluxed
under nitrogen for 3 h. The reaction mixture was then reduced in volume
to 0.5 mL and purified directly on a silica gel column (2 × 10 cm),
eluting with 9:1, 4:1, and 2:1 hexane/ethyl acetate (100 mL aliquots).
A pale yellow oil was obtained (68 mg, 82% yield, 5:1 trans/cis as
1
determined by integration of peaks at 4.50 and 4.61 ppm in the H
NMR spectrum). 1H NMR (500 MHz, CDCl3, ppm): δ 8.03 (2H, d, J
) 7.2 Hz), 7.53 (1H, t, J ) 7.4 Hz), 7.42 (2H, t, J ) 7.8 Hz), 5.78-
5.72 (1H, broad m), 5.57-5.50 (1H, broad m), 4.50 (2H, d, J ) 6.4
Hz), 4.30 (2H, t, J ) 6.7 Hz), 2.06-2.02 (2H, broad m), 2.03 (3H, s),
1.75 (2H, m), 1.44-1.31 (10H, broad m). 13C NMR (125 MHz, CDCl3,
ppm): δ 170.7, 166.5, 150.5, 136.4, 135.2, 132.6, 130.5, 129.4, 128.2,
123.7, 123.3, 65.1, 64.9, 60.2, 32.1, 29.2, 29.1, 28.9, 28.7, 28.6, 27.4,
25.9, 20.9. Rf ) 0.36 (9:1 hexane/ethyl acetate); HRMS (FAB) calcd
for C20H28O4 [M - H]+ 333.2066, found 333.2067.
(73) For a recent example of the synthesis of cyclic alkenylboronates
Via RCM employing ruthenium catalyst 1, see: Renaud, J.; Ouellet, S. G.
J. Am. Chem. Soc. 1998, 120, 7995-7996.
(74) Prepared according to a literature procedure: Hunt, A. R.; Stewart,
S. K.; Whiting, A. Tetrahedron Lett. 1993, 34, 3599-3602.
(75) Suzuki, A. Pure Appl. Chem. 1986, 58, 629-638.
(76) Converting a terminal olefin to a vinylboronic acid or protected
variant for the Suzuki coupling reaction often requires a three-step procedure
involving (1) oxidative cleavage to the aldehyde, (2) subsequent reaction
with dimethyl diazomethylphosphonate to provide the terminal alkyne, and
(3) finally, conversion to the vinylboronate by hydroboration. For a recent
example, see: Scheidt, K. A.; Tasaka, A.; Bannister, T. D.; Wendt, M. D.;
Roush, W. R. Angew. Chem., Int. Ed. 1999, 38, 1652-1655.
(77) The natural product FK506 was recently homodimerized employing
1 through its endogenous C(28) allyl group to yield a cell-permeable protein
dimerizer, FK1012. See: Diver, S. T.; Schreiber, S. L. J. Am. Chem. Soc.
1997, 119, 5106-5109.
Representative Reduced Pressure Procedure for Self-Metathesis
Reaction. Compound 4. 9-Decen-1-yl benzoate (3)84 (349 mg, 1.34
mmol) and 1 (3.5 mg, 4 µmol, 0.3 mol %) were combined in a 1 dram
(80) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-
2925.
(81) The solvent columns are composed of activated alumina (A-2) and
supported copper redox catalyst (Q-5 reactant). See: Pangborn, A. B.;
Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J. Organo-
metallics 1996, 15, 1518-1520.
(82) For the allylation procedure, see: Sugano, H.; Miyoshi, M. J. Org.
Chem. 1976, 41, 2352-2353. For the methyl ester formation, see: Hirai,
Y.; Aida, T.; Inoue, S. J. Am. Chem. Soc. 1989, 111, 3062-3063.
(83) Bodansky, M. Peptide Chemistry; Springer-Verlag: New York,
1988; pp 55-146 and references therein.
(84) Prepared according to a general literature procedure: Schlessinger,
R. H.; Lopes, A. J. Org. Chem. 1981, 46, 5252-5253.
(78) For the RCM of alkenyl phosphonates employing 1, see: Hanson,
P. R.; Stoianova, D. S. Tetrahedron Lett. 1998, 39, 3939-3942.
(79) Preliminary results from these laboratories show that alkenyl ester
derivatives of cysteine are active substrates for CM.

New Approaches to Olefin Cross-Metathesis
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 71
vial. A magnetic stir bar was added to the vial, which was placed inside
a vacuum chamber and held under vacuum (60-100 mTorr) ac-
companied by stirring for 36 h at room temperature. The thick
burgundy-colored oil was observed to steadily produce gas during the
course of the reaction. The reaction mixture was dissolved in 1.0 mL
of CH2Cl2 and applied to a silica gel column (2 × 10 cm, eluting with
CH2Cl2 (375 mL)). Pure fractions were concentrated to give a clear,
colorless, viscous oil which formed a white solid over time (312 mg,
94% yield, 3.8:1 trans/cis as determined by integration of peaks at 5.38
and 5.35 ppm in the 1H NMR spectrum). 1H NMR (500 MHz, CDCl3,
ppm): δ 8.03 (4H, d, J ) 7.2 Hz), 7.53 (2H, t, J ) 7.3 Hz), 7.42 (4H,
t, J ) 7.2 Hz), 5.38 (2H, m), 4.30 (4H, t, J ) 6.7 Hz), 2.10-1.90 (4H,
m), 1.75 (4H, quint, J ) 7.2 Hz), 1.50-1.20 (20H, m). 13C NMR (125
MHz, CDCl3, ppm): δ 166.3, 132.5, 130.4, 130.2, 129.7, 129.4, 128.1,
64.9, 32.4, 29.6, 29.4, 29.2, 29.1, 28.9, 28.6, 27.0, 25.9. Rf ) 0.50 (9:1
hexane/ethyl acetate); HRMS (FAB) calcd for C32H44O4 [M + H]+
493.3239, found 493.3318.
simultaneously added Via syringe to a stirring solution of 1 (12 mg,
0.014 mmol, 2.9 mol %) in CH2Cl2 (2.5 mL). The flask was fitted
with a condenser and refluxed under nitrogen for 12 h. The reaction
mixture was then reduced in volume to 0.5 mL and purified directly
on a silica gel column (2 × 10 cm), eluting with 5:1 hexane/ethyl acetate
(200 mL). A clear oil was obtained (214 mg, 94% yield, 9:1 trans/cis
as determined by the relative intensities of the peaks at 125.3 and 124.8
1
ppm in the 13C NMR spectrum). H NMR (300 MHz, CDCl3, ppm):
δ 6.00 (1H, m), 5.55 (2H, m), 4.82 (1H, d, J ) 3.7 Hz), 4.73 (1H, d,
J ) 3.7 Hz), 3.80 (6H, s), 3.57 (2H, t, J ) 6.6 Hz), 2.07 (2H, m),
1.50-1.21 (12H, m), 0.87 (9H, s), 0.02 (6H, s). 13C NMR (75 MHz,
CDCl3, ppm): δ 170.6, 170.2, 141.1, 125.3, 124.8, 108.1, 102.7, 63.8,
53.4, 53.3, 33.4, 32.6, 30.0, 29.9, 29.7, 29.0, 26.5, 26.3, 18.9, 14.8. Rf
) 0.23 (9:1 hexane/ethyl acetate); HRMS (FAB) calcd for C23H42O7Si
[M + H]+ 459.2778, found 459.2776. Calcd elemental analysis: C,
60.23; H, 9.23. Found: C, 59.98; H, 9.15.
Representative Solution-Phase Self-Metathesis Reaction. Com-
pound 46. Pentapeptide 45 (100 mg, 0.10 mmol) was added to a stirring
solution of 1 (1.2 mg, 1.0 µmol, 1 mol %) in CH2Cl2 (0.5 mL). The
flask was fitted with a condenser and refluxed under nitrogen for 20 h.
The reaction mixture was then reduced in volume to 0.25 mL and
purified directly on a silica gel column (2 × 10 cm), eluting with 3:1,
5:1, and 9:1 ethyl acetate/hexane (100 mL aliquots), and finally with
100% ethyl acetate (200 mL). An off-white crystalline solid was
obtained (61 mg, 62% yield, 2.8:1 trans/cis as determined by the relative
intensities of peaks at 129.1 and 129.0 ppm in the 13C NMR spectrum).
1H NMR (500 MHz, CDCl3, ppm): δ 7.47 (2H, d, J ) 6.6 Hz), 7.27
(2H, d, J ) 9.1 Hz), 7.11 (2H, d, J ) 6.3 Hz), 6.83 (2H, s), 5.66 (2H,
m), 5.37 (2H, m), 4.66 (2H, m), 4.54 (2H, m), 4.24 (2H, m), 4.05-
3.93 (6H, br m), 3.81 (2H, m), 3.70 (8H, s), 2.42 (2H, m), 1.70-1.59
(12H, br m), 1.46 (30H, m), 0.99-0.89 (36H, br m). 13C NMR (125
MHz, CDCl3, ppm): δ 175.7, 175.6, 173.6, 173.5, 172.9, 171.7, 170.4,
170.3, 156.6, 129.1, 129.0, 80.8, 71.1, 69.7, 66.8, 60.4, 57.0, 55.0, 54.1,
52.3, 51.1, 40.7, 40.3, 29.8, 29.3, 28.4, 27.1, 24.8, 24.6, 24.0, 23.0,
22.97, 22.0, 21.8, 19.4, 17.3. Rf ) 0.08 (3:1 ethyl acetate/hexane); [R]D
) -13.67 (CH2Cl2, c 0.34); LRMS (FAB) calcd for C64H114N10O18
[(M) - Boc]+ 1211.8, found 1211.9.
Acknowledgment. Work at Caltech was generously sup-
ported by grants from the National Institutes of Health (NIH)
and Zeneca Pharmaceuticals. Work at Pomona College was
supported by grants from the National Science Foundation
(NSF). H.E.B. thanks the ACS Division of Organic Chemistry
for a Graduate Fellowship (supported by Pfizer, Inc.). D.J.O.
thanks Pomona College for provision of a Steele junior faculty
leave. R.A.W. thanks the Caltech SURF Program and the
Pomona College Chemistry Department for a summer fellow-
ship. Dr. Katsukiyo Miura is acknowledged for his assistance
in the acrolein acetal cross-metathesis portion of this work. We
thank the referees for their insightful comments.
Supporting Information Available: Experimental proce-
dures and full characterization (1H and 13C NMR, HRMS/
elemental analysis, mp, [R]D) of the cross- and self-metathesis
1
products; H NMR spectra for 4, 5, 7, 8, 10, 12-16, 18, 20,
22, 24, 26, 28, 30, 32, 34, 38, 40, 42, 44, 46-49, 51-55, 57,
59, 61, 73, 74, 76, 78, 85, 89, and 91 (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
Representative Acrolein Acetal Cross-Metathesis Reaction. Com-
pound 83. Dimethyl vinylidene-L-tartrate70 (215 µL, 1.0 mmol) and
9-decen-1-(tert-butyldimethylsilane)-yl (165 µL, 0.5 mmol) were
JA993063U