Ruthenium-catalyzed cycloisomerizations of diynols

Article abstract of DOI:10.1021/ja043097o

A wide variety of diynols containing tertiary, secondary, and primary propargylic alcohols undergo a cycloisomerization reaction to form dienones and dienals in the presence of a catalytic amount of [CpRu(CH₃CN) ₃]PF₆. The

Full text of DOI:10.1021/ja043097o

Published on Web 03/10/2005
Ruthenium-Catalyzed Cycloisomerizations of Diynols
Barry M. Trost* and Michael T. Rudd
Contribution from the Department of Chemistry, Stanford UniVersity, Stanford, California 94305
Received November 16, 2004; E-mail: bmtrost@stanford.edu
Abstract: A wide variety of diynols containing tertiary, secondary, and primary propargylic alcohols undergo
a cycloisomerization reaction to form dienones and dienals in the presence of a catalytic amount of [CpRu-
(CH3CN)3]PF6. The formation of five- and six-membered rings is possible using this methodology. Secondary
diynols react to form single geometrical isomeric dienones and -als. Primary diynols undergo a
cycloisomerization as well as a hydrative cyclization process. The utility of primary diynol cycloisomerization
is demonstrated in a synthesis of (+)-R-kainic acid.
Introduction
In contrast to the sometimes-difficult synthesis and protection
of carbonyls, alkynes are known to be synthetically robust, and
the synthesis of substituted alkynes and propargylic alcohols
can be quite simple. A wide variety of methods are known for
the construction of propargylic and homopropargylic alcohols
and amines.16-27 Unfunctionalized alkynes are less readily
accessible than ones containing proximal heteroatoms, but there
are still many straightforward methods of synthesis, including
various isomerization,28-32 homologation,33-38 and metal-
catalyzed cross-coupling39 methods.
Preparation of Diyne Substrates. The most readily available
diynols can be made starting from the commercially available
1,7-octadiyne and 1,6-heptadiyne. Mono- and bis-deprotonation
followed by trapping of the anion(s) with aldehydes and ketones
leads to various diynols. More functionalized diynols 3 can be
Utilization of easily accessible starting materials to create
highly functionalized ring structures through addition reactions
allows the creation of molecular complexity1,2 while preserving
atom economy.3-5 The use of transition metal-catalyzed cy-
clization reactions often allows one to achieve synthetic
efficiency in ways not normally attainable through traditional
means. The intramolecular aldol condensation6-11 is a classic
method to construct 1-acyl-cyclopentenes and 1-acyl-cyclo-
hexenes. However, some of the disadvantages of this reaction
are the difficult or lengthy synthesis of the keto-aldehyde starting
materials and the potential complications inherent when more
than a single enolate can be formed.11 While some methods
have been developed to carry out chemoselective aldol con-
densations,6,7 there is not a general solution to this problem.
Additionally, the inherent reactivity of ketones and aldehydes
makes the use of protecting groups frequently necessary in the
total synthesis of complex molecular targets. We have recently
developed a series of ruthenium-catalyzed propargylic alcohol
dimerization12 and diynol cycloisomerization reactions which
form dienones and dienals with excellent chemoselectivity.13-15
The products of these reactions formally result from a chemose-
lective intramolecular aldol condensation of aldehydes or
ketones with enals.
(16) Wada, M.; Sakurai, Y.; Akiba, K. Tetrahedron Lett. 1984, 25, 1083-1084.
(17) Wada, M.; Sakurai, Y.; Akiba, K. Tetrahedron Lett. 1984, 25, 1079-1082.
(18) Gensler, W. J.; Abrahams, C. B. J. Am. Chem. Soc. 1958, 80, 4593-4596.
(19) Yamaguchi, M.; Hirao, I. Tetrahedron Lett. 1983, 24, 391-394.
(20) Fried, J.; Lin, C. H.; Dalven, P.; Sih, J. C. J. Am. Chem. Soc. 1972, 94,
4343.
(21) Fried, J.; Lin, C. H.; Ford, S. H. Tetrahedron Lett. 1969, 1379.
(22) Turner, J. J.; Leeuwenburgh, M. A.; Van Der Marel, G. A.; Van Boom, J.
H. Tetrahedron Lett. 2001, 42, 8713-8716.
(23) Gronquist, M. R.; Meinwald, J. J. Org. Chem. 2001, 66, 1075-1081.
(24) Ding, C. H.; Dai, L. X.; Hou, X. L. Synlett 2004, 1691-1694.
(25) Yamaguchi, M.; Nobayashi, Y.; Hirao, I. Tetrahedron 1984, 40, 4261-
4266.
(26) Tsunoda, T.; Nagino, C.; Oguri, M.; Ito, S. Tetrahedron Lett. 1996, 37,
2459-2462.
(1) Wender, P. A.; Miller, B. L. Org. Synth.: Theory Appl. 1993, 2, 27.
(2) Bertz, S. H.; Sommer, T. J. Org. Synth.: Theory Appl. 1993, 2, 67.
(3) Trost, B. M. Acc. Chem. Res. 2002, 35, 695-705.
(4) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259-281.
(5) Trost, B. M. Science 1991, 254, 1471.
(6) Wolinsky, J.; Gibson, T.; Slabaugh, M. R. J. Org. Chem. 1964, 29, 3740.
(7) Wolinsky, J.; Barker, W. J. Am. Chem. Soc. 1960, 82, 636-638.
(8) Heathcock, C. H. In ComprehensiVe organic synthesis; Trost, B. M., Ed.;
Pergamon Press: New York, 1991; p 133.
(27) Mitsunobu, O. Synthesis 1981, 1-28.
(28) Brown, C. A.; Yamashita, A. J. Am. Chem. Soc. 1975, 97, 891-892.
(29) Midland, M. M.; Halterman, R. L.; Brown, C. A.; Yamaichi, A. Tetrahedron
Lett. 1981, 22, 4171-4172.
(30) Takano, S.; Sekiguchi, Y.; Sato, N.; Ogasawara, K. Synthesis 1987, 139-
141.
(31) Takano, S.; Shimazaki, Y.; Iwabuchi, Y.; Ogasawara, K. Tetrahedron Lett.
1990, 31, 3619-3622.
(32) Thompson, C. F.; Jamison, T. F.; Jacobsen, E. N. J. Am. Chem. Soc. 2001,
123, 9974-9983.
(9) Nielsen, A. T.; Houlihan, W. J. Org. React. 1968, 16, 1.
(10) House, H. O. Modern synthetic reactions, 2nd ed.; W. A. Benjamin: Menlo
Park, CA, 1972; p 629.
(33) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
(34) Gilbert, J. C.; Weerasooriya, U. J. Org. Chem. 1979, 44, 4997-4998.
(35) Brown, D. G.; Velthuisen, E. J.; Commerford, J. R.; Brisbois, R. G.; Hoye,
T. R. J. Org. Chem. 1996, 61, 2540-2541.
(11) Carey, F. A.; Sunberg, R. J. AdVanced organic chemistry, Part B, 3rd ed.;
Plenum Press: New York, 1990.
(12) Trost, B. M.; Rudd, M. T. J. Am. Chem. Soc. 2001, 123, 8862-8863.
(13) Trost, B. M.; Rudd, M. T. J. Am. Chem. Soc. 2002, 124, 4178-4179.
(14) Trost, B. M.; Rudd, M. T. Org. Lett. 2003, 5, 1467-1470.
(15) Trost, B. M.; Rudd, M. T.; Costa, M. G.; Lee, P. I.; Pomerantz, A. E. Org.
Lett. 2004, 5, 4235.
(36) Muller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521.
(37) Colvin, E. W.; Hamill, B. J. J. Chem. Soc., Chem. Commun. 1973, 151-
152.
(38) Miwa, K.; Aoyama, T.; Shioiri, T. Synlett 1994, 107-108.
(39) Negishi, E.; Anastasia, L. Chem. ReV. 2003, 103, 1979-2017.
9
10.1021/ja043097o CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 4763-4776
4763

A R T I C L E S
Trost and Rudd
and tricarbonyl compounds.60 Several alkyne-alkyne dimer-
izations, other than the one described by us (eq 1),12 using related
catalyst systems have also been described.61-65
Scheme 1. Synthesis of Diynols Based on Dimethylmalonate
Scheme 2
We proposed a mechanism for this (eq 1) dimerization based
upon ruthenacycopentadiene formation, elimination of a water
molecule, re-addition of water to the other side of the ruthena-
cycle, â-H elimination, and reductive elimination. It appeared
to us that only a single molecule of propargylic alcohol should
be required, and thus a cross-coupling of alkynes and propargylic
alcohols should be possible (Scheme 5).
Initially, using the optimized conditions developed for the
intermolecular dimerization of propargylic alcohols (eq 1), the
cycloisomerization (eq 2) of diynols proceeded in good yields
(∼70%). However, it was quickly discovered that the high
prepared through sequential alkylation of dimethylmalonate with
various propargylic and homopropargylic halides (Scheme 1).
Subsequent deprotonation and trapping with various electro-
philes leads to a wide variety of diynol substrates in a few simple
steps. Oxygen- and nitrogen-containing diynols can also be
constructed in a related manner starting from propargylic
alcohols and protected amines. More complex diynols can also
be synthesized in a facile manner. A Sonogashira coupling is
the key reaction in the three-step synthesis which leads to diynol
7 (Scheme 2). Starting from cyclohexanone, the functionalized
tertiary diynol 12 is prepared in four simple steps via a sequence
of enolate alkylation, addition of propynylmagnesium bromide,
acetate formation, and alkylation with acetone (Scheme 3).
Acyclic diynols are also available via similar steps. An ester-
tethered diynol can be prepared in a direct fashion starting from
butynoic acid (Scheme 4). DCC-mediated ester formation
followed by alkylation with acetone yields diynol 15 in two
steps.
concentration/mixed-solvent system required for the dimeriza-
tion (eq 1) to yield a single product was not required for the
cycloisomerization. The catalyst loadings could also be lowered
in many of the cases. For example, the substrate which is a
direct conjugate to a propargylic alcohol dimerization, diynol
substrate 26, cycloisomerized in nearly quantitative yield with
only 1 mol % ruthenium complex 16 (eq 3).
Cycloisomerization of Tertiary Diynols. The utility of the
ruthenium precatalyst [CpRu(CH3CN)3]PF6 (16) in various
alkene-alkyne coupling reactions has been reported extensively
in the literature.40-49 This and related [CpRu] complexes are
also known to promote several alkyne-alkyne coupling reac-
tions, notably including alkyne trimerization50-52 and diyne
cycloadditions with 1,3-dienes,53 allylic ethers,54 other alkenes,55
nitriles,56,57 isocyanates,58 isothiocycanates,59 carbon disulfide,59
The success of this reaction led us to explore the scope of
the cycloisomerization reaction with various tertiary diynols
(Table 1). Terminal alkynes (entries 1 and 3) as well as internal
alkynes (entries 2 and 4) are well tolerated. Geminal diester
groups in the tether are not required, as the substitution of ether
and sulfonamide functionality does not result in any decrease
in yield (entries 2 and 3). Bicyclic unsaturated ketones can be
(40) Trost, B. M.; Surivet, J. P.; Toste, F. D. J. Am. Chem. Soc. 2004, 126,
15592.
(41) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 2002, 124, 5025-5036.
(42) Trost, B. M.; Pinkerton, A. B.; Toste, F. D.; Sperrle, M. J. Am. Chem.
Soc. 2001, 123, 12504-12509.
(54) Yamamoto, Y.; Kitahara, H.; Ogawa, R.; Itoh, K. J. Org. Chem. 1998, 63,
9610-9611.
(43) Trost, B. M.; Toste, F. D.; Pinkerton, A. B. Chem. ReV. 2001, 101, 2067-
(55) Yamamoto, Y.; Kitahara, H.; Ogawa, R.; Kawaguchi, H.; Tatsumi, K.; Itoh,
K. J. Am. Chem. Soc. 2000, 122, 4310-4319.
2096.
(44) Trost, B. M.; Surivet, J. P.; Toste, F. D. J. Am. Chem. Soc. 2001, 123,
2897.
(56) Yamamoto, Y.; Okuda, S.; Itoh, K. Chem. Commun. 2001, 1102-1103.
(57) Yamamoto, Y.; Ogawa, R.; Itoh, K. J. Am. Chem. Soc. 2001, 123, 6189-
6190.
(45) Trost, B. M.; Toste, F. D.; Shen, H. J. Am. Chem. Soc. 2000, 122, 6138-
6138.
(58) Yamamoto, Y.; Takagishi, H.; Itoh, K. Org. Lett. 2001, 3, 2117-2119.
(59) Yamamoto, Y.; Takagishi, H.; Itoh, K. J. Am. Chem. Soc. 2002, 124, 28-
29.
(46) Trost, B. M.; Brown, R. E.; Toste, F. D. J. Am. Chem. Soc. 2000, 122,
5877-5878.
(47) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 2000, 122, 714-715.
(48) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 9728-9729.
(49) Trost, B. M.; Toste, F. D. Tetrahedron Lett. 1999, 40, 7739-7743.
(50) Yamamoto, Y.; Ishii, J. I.; Nishiyama, H.; Itoh, K. J. Am. Chem. Soc. 2004,
126, 3712-3713.
(60) Yamamoto, Y.; Takagishi, H.; Itoh, K. J. Am. Chem. Soc. 2002, 124, 6844-
6845.
(61) Le Paih, J.; Monnier, F.; Derien, S.; Dixneuf, P. H.; Clot, E.; Eisenstein,
O. J. Am. Chem. Soc. 2003, 125, 11964-11975.
(62) Le Paih, J.; Derien, S.; Bruneau, C.; Demerseman, B.; Toupet, L.; Dixneuf,
P. H. Angew. Chem., Int. Ed. 2001, 40, 2912-2915.
(63) Le Paih, J.; Derien, S.; Dixneuf, P. H. Chem. Commun. 1999, 1437-1438.
(64) Dixneuf, P. H.; Bruneau, C.; Derien, S. Pure Appl. Chem. 1998, 70, 1065-
1070.
(51) Yamamoto, Y.; Arakawa, T.; Ogawa, R.; Itoh, K. J. Am. Chem. Soc. 2003,
125, 12143-12160.
(52) Yamamoto, Y.; Hata, K.; Arakawa, T.; Itoh, K. Chem. Commun. 2003,
1290-1291.
(53) Varela, J. A.; Castedo, L.; Saa, C. Org. Lett. 2003, 5, 2841-2844.
(65) Yoshida, M.; Sugimoto, K.; Ihara, M. ArkiVoc 2003, 35-44.
9
4764 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
Scheme 3
Scheme 4. Synthesis of Ester-Tethered Diynol
Table 1. Ruthenium-Catalyzed Cycloisomerization of Tertiary
1,6-Diynols
Scheme 5. Mechanistic Proposal for Diynol Cycloisomerization
Scheme 6. Rationale for Formation of Cyclopentadiene
a All reactions at 0.1 M concentration in acetone with ∼1 equiv of H2O
at room temperature for 1 h. E ) CO2Me.
complete disappearance of the acetate group; no side products
contained an acetate residue either. A mechanistic rationale
for the formation of these unexpected products is given in
Scheme 6.
It is envisioned that elimination of the “internal” leaving group
on 38 would form 39, which following a proton abstraction gives
40. Functionalized cyclopentadiene 35 is then produced fol-
lowing the previously discussed sequence of events (Scheme
5). The difficulty of acetate elimination after cycloisomerization
likely precludes this possibility. Additionally, the elimination
of an “internal” hydroxyl group has precedence in the dimer-
ization of propargylic alcohols using [Cp*Ru(CH3CN)3]PF6 as
formed easily as well (entry 4). This example also demonstrates
compatibility with aromatic alkynes. In general, 1-5 mol %
catalyst is all that is required to achieve complete conversion
of this class of substrates. In entries 5 and 6, not only is the
cycloisomerzation taking place, but the tertiary propargylic
leaving groups on the inside of the forming rings are eliminated
as well, thus realizing a synthesis of functionalized cyclopen-
tadienes. The corresponding free alcohol of 12 also cyclizes,
although in slightly diminished yield. The crude NMR revealed
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4765

A R T I C L E S
Trost and Rudd
Scheme 7. [Cp*Ru(CH3CN)3]PF6-Catalyzed Dimerization of Propargylic Alcohols
Scheme 8. Rationale for Observed Isomer of 69
isomerization conditions, the product isolated is not the expected
unsaturated ketone 52, but instead is the isomeric [6H]-pyran
53 (Scheme 9).
It is assumed that the initial product is 52 and pyran 53 is
the isolated product on the basis of the instability of the initial
R-keto ester relative to the pyran. This type of isomerization is
well known66 for R,â,γ,δ-unsaturated ketones, and the presence
of electron-withdrawing groups is known to stabilize the pyran
isomer. On the other hand, an internal carbonyl group (15) does
not affect the product outcome, and the expected dienone 54 is
isolated in excellent yield (Scheme 9). While 15 readily
participates in the cycloisomerization, changing the location of
the carbonyl group such that the leaving group is in conjugation
with the carbonyl (55) results in a very poor cycloisomerization
substrate (Figure 1). Having demonstrated a range of tertiary
propargylic alcohol 1,6-diynes in the cycloisomerization reac-
tion, we sought to test the feasibility of 1,7-diynes to function
in this reaction (Table 2).
the catalyst (Scheme 7)12 and with substituted cyclobutylpro-
pargylic alcohols.65
Substrates containing only internal propargylic alcohol func-
tionality can undergo cycloisomerization as well (eq 4).
Cyclization of 47 was accomplished under slightly modified
conditions with the more sterically hindered [Cp*Ru(CH3CN)3]-
PF6 to produce a single geometrical isomeric cross-conjugated
aldehyde (48) whose configuration was assigned on the basis
of a 2% NOE between the allylic methyl group and the
cyclopentene methylene. The somewhat low yield may be
attributable to the volatility of the product. The isolation of 48
as the only isomer supports the proposed ruthenacyclopentadiene
mechanism, as the observed olefin geometry agrees with the
prediction of a â-hydride elimination from an intermediate like
50 (Scheme 8).
While the yields shown in Table 2 are good, some optimiza-
tion was required for each compound. It was found that 7 mol
% of catalyst 16 was required to achieve full conversion of
geminal diester compound 62. The terminal alkyne 1,7-diyne
58 needed to be heated to 60 °C in 10 vol % water/acetone
with 1 equiv of malonic acid in order to reach full conversion
with 10 mol % 16. To realize good yield in the cyclization of
the more sterically hindered internal diyne 60, raising the
concentration to 1 M was required. However, with these
modifications, 1,7-diynes can participate in the ruthenium-
catalyzed cycloisomerization to yield cyclic unsaturated ketones
and aldehydes. Submitting two 1,8-diynes (56, 57) to the
cycloisomerization conditions led to very low conversion, and
thus this was not further pursued (Figure 1). Other larger ring
systems were not explored.
Heating of the reaction mixture was required to obtain
significant conversion with either the [CpRu] or the [Cp*Ru]
complex, and addition of malonic acid resulted in an increase
in yield from 20% to 55%. The standard ruthenium complex
16 does indeed promote the cyclization, but the [Cp*Ru] catalyst
carries out the transformation more efficiently. This is in
agreement with the fact that the “internal” elimination was also
seen with this catalyst in the dimerization of propargylic alcohols
to form hemiacetal products (Scheme 7). This again demon-
strates that elimination of the “internal” hydroxyl group requires
more bulky, donating ligands. This may indicate that coordina-
tion of the alcohol to ruthenium (not feasible with “internal”
alcohols) is important to achieve a rapid and high-yielding
reaction with the [CpRu] catalyst 16.
Cycloisomerization of Secondary Diynols. Attempts to
utilize secondary propargylic alcohols in the simple dimerization
reaction (eq 1) led to very low conversions and yields of the
corresponding dimeric products. It is assumed that the ease with
which tertiary propargylic alcohol diynes cyclize is indicative
of the increased rate of formation and stability of the proposed
ruthenacyclopentadiene intermediates, and that this might enable
secondary propargylic alcohols to participate in the cycloi-
somerization as well. Gratifyingly, submitting these compounds
While all the previous examples utilized electronically neutral
alkynes, electron-poor alkynes can also participate quite satis-
factorily in the cyclization. When alkynoate 51, with an ester
group external to the forming ring, is submitted to the cyclo-
(66) Kluge, A. F.; Lillya, C. P. J. Org. Chem. 1971, 36, 1977.
9
4766 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
Table 3. Ruthenium-Catalyzed Cycloisomerization of Secondary
Diynols
Figure 1. Poor substrates for cycloisomerization.
Table 2. Ruthenium-Catalyzed Cycloisomerization of Tertiary
1,7-Diynols
a Reactions at 0.1 M concentration in acetone with ∼1 equiv of H2O at
room temperature for 1 h. E ) CO2Me.
Scheme 10. Steric Effect in Determination of Olefin Geometry
a 0.1 M concentration in acetone with ∼1 equiv of H2O at room
temperature. b 0.1 M concentration in 10 vol % water/acetone at 60 °C, 1
equiv of malonic acid. c 1 M concentration in acetone with ∼1 equiv of
H2O at room temperature. E ) CO2Me
Another possibility is that the elimination/1,2 shift of the
hydroxyl group occurs in a stereoselective fashion based upon
a preference to minimize steric interaction between the R group
and the cyclopentane ring (70 vs 71, Scheme 10).
to the standard cyclization conditions led to formation of the
expected products as single geometrical isomers in good yields
(Table 3).
Cycloisomerization of Primary Diynols. Even though the
dimerization of primary propargylic alcohols does not proceed
at all,12 on the basis of the success of secondary propargylic
alcohols in the cycloisomerization reaction, primary propargylic
alcohol diynes were prepared and examined in the ruthenium-
catalyzed cycloisomerization as well. When 72 was submitted
to the standard cycloisomerization conditions (10 mol % 16,
acetone, ∼1 equiv of H2O, room temperature), we were gratified
to see that some cycloisomerized product 73 was formed.
However, the conversion was very low. Addition of malonic
acid and heating of the reaction to 60 °C led to complete
conversion of the starting material, but the isolated yield was
only 30% even though the reaction was nearly spot-to-spot. The
crude mass recovery was also only a little over 30%, and it
seemed that decomposition to baseline material must have been
occurring. Various other conditions were tried in order to
The cyclizations of secondary propargylic alcohols to form
five- and six-membered rings are possible; however, the six-
membered-ring formation is somewhat lower yielding, and
various other decomposition products are formed as well. In
general, these reactions are not as clean as the tertiary alcohol
examples, but good yields can still be obtained with alkyl
(entries 1 and 3) and aromatic (entry 2) secondary propargylic
alcohol diynes. Also, the only isolable products have the E-olefin
configuration at the γ,δ-double bond. This olefin geometry could
be the result of the thermodynamic stability of this isomer as
compared the corresponding Z-olefin. It was demonstrated in
the context of propargylic alcohol dimerization12 that the R,â-
alkene can isomerize under the reaction conditions (16 and
aqueous acetone), and thus it is reasonable that the γ,δ-alkene
may as well.
Scheme 9. Cycloisomerization of Electron-Poor Diynols
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4767

A R T I C L E S
Trost and Rudd
Table 4. Temperature, Concentration, and Malonic Acid Effect on
Ratio of 75 to 76
Table 5. Cycloisomerization/Hydrative Cyclization of Ethyl Diynol
concn
(M)
reaction
time (h)
entrya
temp (°C)
additive
ratio 75:76
1
2
60
60
0.1
0.1
<1
1
1.5:1
2.5:1
malonic acid
(1 equiv)
3
4
5
rt
rt
0-rt
0.1
1
0.1
15
3
>30
3:1
2.2:1
3.4:1
entry
temp (
°
C)
ratio 78:79
total yield (%)
1
2
3
rt
60
100
1:1.5
1:2.0
1:2.0
77
78
83
a Reactions with 10 mol % 16, in 10 vol % water/acetone. Total yield
of 75 and 76 is approximately 90%.
optimize the yield, and it was discovered that the addition of
more (up to 10 vol %) water increased the isolated yield of 73
to 45% (eq 5).
reactions, then conversion to both products was very low.
Importantly, it was demonstrated that the two products do not
interconvert. Submission of either isolated compound to the
reaction conditions did not lead to any formation of the other
product. This indicates that the mechanism of this reaction is
diverging into two different products or that two different
mechanisms are operative. It is unclear whether the rate of
reaction is the cause of the change in product ratio, or if the
rate is a byproduct of some more fundamental change. The
product ratio between the “normal” and the “hydrated” products
was also very sensitive to substrate identity.
A seemingly minor change to an ethyl alkyne (77) leads to
a reversal in the major product formed (Table 5). Under
conditions where methyl alkyne substrate 74 cyclizes in a 1.5:1
ratio of normal to hydrated products, ethyl alkyne 77 cyclizes
in a 1:1.5 ratio (entry 1). Raising the temperature (entries 2
and 3) increases the relative amount of the hydrated product,
as was seen in the previous example. In this example, it is quite
obvious that the ethyl alkyne becomes the propyl group, while
the propargylic alcohol carbons are the R-hydroxy ketone portion
in 79. If the tether between the two alkyne moieties is lengthened
to four atoms, then the product ratio is pushed even further (1:
3.3) toward the formation of the hydrated product 82 (eq 7).
At this point, it was assumed that the low yield and mass
recovery were results of the instability of the product because
the reaction appeared very “clean” by TLC, and that ketone
products, resulting from internal alkynes, should be more stable.
In fact, when 74 was treated with 10 mol % 16 in 10 vol %
water/acetone (0.1 M) at 60 °C, nearly a quantitative amount
of the expected mass was recovered. Unexpectedly, two different
products (75 and 76), in a ratio of 1.5:1, were isolated (eq 6).
While 75 is the expected methyl ketone product, 76 is a
formally hydrated form of this compound (the molecular weight
is that of 75 plus water). The mechanism of this transformation
is clearly not a simple hydration of either 74 or 75, but for
convenience, 76 and related products will be referred to as
hydrated products. It also noteworthy that the carbons in 75
and 76 are pictured in the same orientation; the methyl alkyne
is transformed into the ethyl group. It is not obvious that this is
the case at this juncture, but this will become clear with the
next example. The ratio of the two products was then found to
be quite sensitive to temperature, concentration, and malonic
acid (Table 4).
Diynol 80 does not cyclize to a significant extent at room
temperature and must be heated in order to obtain good
conversion. These examples demonstrate that the more steric
hindrance present in a cyclization reaction (ethyl alkyne vs
methyl alkyne), the more “hydrated” product that is formed.
Mechanistic Considerations in Primary Diynol Cycliza-
tions. While the mechanism presented in Scheme 5 nicely
accounts for the results with tertiary and secondary propargylic
alcohols, to rationalize the formation of both products in the
primary alcohols case, the following mechanism is proposed
(Scheme 11).
Basically, the longer amount of time required for full
conversion of the starting material, the higher percentage of
“normal” product 75 that was obtained. Inclusion of malonic
acid (entry 2) or lower temperatures (entries 2-5) reduced the
reaction rate, thus leading to more 75. Higher concentration
(entry 4) increased the rate of reaction, and thus the 3:1 ratio
obtained at room temperature was reduced to 2.2:1. The reaction
conducted at 0 °C (entry 5) did not go to completion even after
30 h, but upon raising the temperature to room temperature,
the remaining starting material was consumed in a few hours.
All these conditions resulted in a combined yield of ∼90%. If
a small amount of water (∼1 equiv) was used in any of these
The key component to the proposed catalytic cycles is the
invocation of the ruthenacyclopentatriene (84) resonance in
9
4768 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
Scheme 11. Mechanistic Rationale for the Formation of Hydrative and Normal Cyclization Products with Primary Diynols
analogy to Kirchner88-90 and Dixneuf’s61,63 proposal for phos-
phine and carboxylate addition to ruthenacyclopentadienes.
Metallacyclopentatrienes, which can have a folded or planar
geometry, have been isolated for a variety of metals.91-96 In
particular, a [CpRu] fragment has been shown to exist as a
ruthenacyclopentadiene or -pentatriene, depending on the ligand
environment.96 In the case of primary propargylic alcohol diynes,
addition of water to one of the two carbene carbons would
produce either 85 or 89, depending on the chemoselectivity of
the addition. Rearrangement of 85 would lead to 86 and,
following hydride shift and protonation, would result in forma-
tion of the “hydrated” product 88. Rearrangement of 89 would
lead to 90, which results in formation of 92 following hydride
shift and â-hydroxide elimination or protonation followed by
water elimination. The differences in product ratios may be a
result of subtle factors contributing to which carbene carbon is
attacked. For example, a hydroxy methylene is slightly larger
than a methyl group, and thus water attack is favored at the
methyl carbon. Conversely, the use of a larger ethyl group leads
to favored attack at the smaller hydroxymethylene. It is also
possible that this mechanism and the “elimination first” mech-
anism presented in Scheme 5 are both operative. It is believed
that the reason for this “water addition first” mechanism
(Scheme 11) becoming active with primary alcohols is related
to the poor leaving group ability of primary propargylic alcohols
relative to secondary and tertiary alcohols.
To probe the effect of the free hydroxyl group on the
cyclization, several protected tertiary and primary diynols were
submitted to the reaction conditions. As Table 6 reveals, the
presence of a free propargylic alcohol is not critical for
reactivity. Bispropargylic alcohol 93 cyclizes to form a single
product 94, the cycloisomerized “normal” product (entry 1).
While it may initially appear that no hydrative product is formed
and that something about the substrate must be preventing this
pathway, the expected hydrative product has a facile elimination
which could take place, and thus both pathways may still be
operative (eq 8).
(67) Noyori, R.; Tomino, I.; Yamada, M.; Nishizawa, M. J. Am. Chem. Soc.
1984, 106, 6717-6725.
(68) Noyori, R.; Tomino, I.; Tanimoto, Y.; Nishizawa, M. J. Am. Chem. Soc.
1984, 106, 6709-6716.
(69) Prepared from cyclohexene and BH3-SMe2.
(70) Prepared from 1,5-COD and BH3-SMe2 in DME.
(71) Fleming, I.; Winter, S. B. D. J. Chem. Soc., Perkin Trans. 1 1998, 2687.
(72) Fleming, I.; Roberts, R. S.; Smith, S. C. J. Chem. Soc., Perkin Trans. 1
1998, 1209.
(73) See Supporting Information for preparation.
(74) Crabtree, R. H.; Davis, M. W. J. Org. Chem. 1986, 51, 2655-2661.
(75) Crabtree, R. H.; Davis, M. W. Organometallics 1983, 2, 681-682.
(76) Brown, J. M.; Hall, S. A. Tetrahedron Lett. 1984, 25, 1393-1396.
(77) Evans, D. A.; Morrissey, M. M. J. Am. Chem. Soc. 1984, 106, 3866-
3868.
(78) Evans, D. A.; Morrissey, M. M. Tetrahedron Lett. 1984, 25, 4637-4640.
(79) Stork, G.; Kahne, D. E. J. Am. Chem. Soc. 1983, 105, 1072-1073.
(80) Van Staden, L. F.; Gravestock, D.; Ager, D. J. Chem. Soc. ReV. 2002, 31,
195-200.
(81) Ager, D. J. Synthesis 1984, 384-398.
(82) Smitrovich, J. H.; Woerpel, K. A. J. Org. Chem. 1996, 61, 6044-6046.
(83) Tamao, K.; Kumada, M.; Maeda, K. Tetrahedron Lett. 1984, 25, 321.
(84) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organometallics 1983, 2,
1694-1696.
(85) Fleming, I.; Henning, R.; Plaut, H. J. Chem. Soc., Chem. Commun. 1984,
29-31.
(86) Miyata, O.; Ozawa, Y.; Ninomiya, I.; Naito, T. Tetrahedron 2000, 56,
6199-6207.
(87) Yoo, S. E.; Lee, S. H. J. Org. Chem. 1994, 59, 6968-6972.
(88) Becker, E.; Mereiter, K.; Puchberger, M.; Schmid, R.; Kirchner, K.
Organometallics 2003, 22, 2124-2133.
(89) Pavlik, S.; Gemel, C.; Slugovc, C.; Mereiter, K.; Schmid, R.; Kirchner, K.
J. Org. Chem. 2001, 617, 301-310.
The monoprotected analogue 95 also behaves similarly and
generates a 1:1.7 mixture of products, favoring the addition of
water to the free propargylic alcohol alkyne (entry 2). Entry 3
demonstrates the exclusive elimination of a tertiary propargylic
alcohol under either low water (∼1 equiv of H2O) or high water
(10 vol % H2O) conditions. Protection of the primary propargylic
alcohol also has no effect (entry 4). Protection of the simple
primary propargylic alcohol substrate as a methyl ether or as
an acetate does not affect the ratio of products at all (entries 5
(90) Becker, E.; Ruba, E.; Mereiter, K.; Schmid, R.; Kirchner, K. Organome-
tallics 2001, 20, 3851-3853.
(91) Pu, L.; Hasegawa, T.; Parkin, S.; Taube, H. J. Am. Chem. Soc. 1992, 114,
2712-2713.
(92) Pu, L.; Hasegawa, T.; Parkin, S.; Taube, H. J. Am. Chem. Soc. 1992, 114,
7609-7610.
(93) Gemel, C.; Lapensee, A.; Mauthner, K.; Mereiter, K.; Schmid, R.; Kirchner,
K. Montash. Chem. 1997, 128, 1189-1199.
(94) Kerschner, J. L.; Fanwick, P. E.; Rothwell, I. P. J. Am. Chem. Soc. 1988,
110, 8235-8238.
(95) Hirpo, W.; Curtis, M. D. J. Am. Chem. Soc. 1988, 110, 5218-5219.
(96) Albers, M. O.; Dewaal, D. J. A.; Liles, D. C.; Robinson, D. J.; Singleton,
E.; Wiege, M. B. J. Chem. Soc., Chem. Commun. 1986, 1680-1682.
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4769

A R T I C L E S
Trost and Rudd
Table 6. Effect of Protecting Groups on Cycloisomerization/
Hydrative Cyclization
Scheme 12. Possible Mechanism of Tertiary Propargylic Alcohol
Cycloisomerization
rated into the product 29. Raising the amount of H2O18 to 100
equiv only resulted in 42% of the product containing O18 (37
times the background level), even though every attempt was
made to exclude adventitious H2O.16 This indicates that the
carbonyl oxygen can come from both the propargylic alcohol
and exogenous water; the eliminated water from the propargylic
alcohol can be transferred to the other alkyne, but in the presence
of a large excess of water, exchange may occur as well. While
this explanation is reasonable, there is a possibility that O18 can
be incorporated after the aldehyde is formed via a Lewis-acid-
catalyzed hydration/elimination process. For primary propargylic
alcohol substrates that participate in the hydrative process,
clearly exogenous water is being incorporated, although the
“normal” products were not checked for O18 incorporation. At
this point, we still believe that for tertiary and secondary
propargylic alcohol diynes, the mechanism depicted in Scheme
5 is largely correct. It can be proposed that all the reactions
presented in this article proceed by water addition followed by
a facile δ-elimination of water to yield the observed R,â,γ,δ-
unsaturated aldehydes and ketones, and that only one product
is isolated because addition next to the hindered tetrasubstituted
center is disfavored (Scheme 12).
However, the remarkable difference between reactions with
tertiary and primary propargylic alcohols suggests otherwise.
For example, tertiary alcohol diynes react nearly instantaneously
with ∼1 equiv of water, while the corresponding primary
substrates require excess water and heat to react quickly. While
elimination of the tertiary δ-hydroxyl may be faster than that
of the corresponding primary hydroxyl, if the elimination (110
to 25) is the rate-limiting step, then a buildup of the δ-hydroxy
ketone should be seen. However, if this type of elimination
occurs within the catalytic cycle (91 to 92, Scheme 11) and is
the rate-limiting step, then it is possible that both tertiary and
primary propargylic alcohol diynes react via the same mecha-
nism. However, it is unclear why the amount of water present
would have such an effect on primary alcohols, but none for
tertiary. Therefore, from the data at hand, it seems likely that
tertiary/secondary alcohol substrates react by one mechanism,
and primary alcohols primarily react by another. Finally, it is
important to note that the preceding discussion is based upon
empirical examples and not upon kinetic and mechanistic
studies.
a Reactions at 0.1 M concentration in 10 vol % water/acetone at 60 °C
for 1 h. b Ratio determined by 1H NMR integration on crude samples. E )
CO2Me.
and 6; a 1.5:1 ratio is obtained with the free alcohol as well).
Entry 7 reveals that a free alcohol is not required for cyclization
of tertiary alcohol substrates either. On the other hand, a tertiary
acetate does not participate in the cycloisomerization at all. It
is possible that the presence of the highly activated tertiary
propargylic acetate leads to deactivation of the catalyst (entry
8).
Due to the lack of effect of the propargylic functionality for
most substrates, these data seem to indicate that a propargylic
alcohol is not required for cyclization of certain substrates. Some
understanding of the mechanism of both reactions can be gained
from this information along with the data from a reaction carried
out with O18-water.97 When the reaction in Table 1, entry 1,
was performed in the presence of 3 equiv (relative to 28) of
H2O18, only 2 times the background level of O18 was incorpo-
Synthesis of r-Kainic Acid. The factors which determine
the ratio of “normal” to “hydrated” cyclization products were
(97) 75% isotopic purity as measured by mass spectrometric analysis carried
out by Dr. David Walthall in the Brauman group at Stanford University
on Oct 31, 2001.
9
4770 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
Scheme 13. Retrosynthesis of R-Kainic Acid
One of the most straightforward methods to introduce
oxygenation at the δ-carbon is a chemoselective hydroboration-
oxidation of the γ,δ-olefin in the presence of the R,â-olefin and
the ketone. While unprecedented in a conjugated system like
this, it would seem to be reasonable. Using simple borane
reagents (BH3-SMe2), mainly 1,2-reduction products were
isolated in low yield along with extensive decomposition under
various oxidation conditions. The use of dicyclohexylborane69
or 9-BBN dimer70 led to predominantly 1,6-reduction; however,
low yields were obtained uniformly. Use of rhodium or nickel
catalysts also did not allow any isolation of the desired product.
It is believed that the presence of the conjugated ketone was
the problem in the reaction, and thus the ketone was masked as
an ethylene glycol ketal (Scheme 16).
then further explored in the context of a synthesis of R-kainic
acid (Scheme 13).14
We planned to introduce the isopropylidene fragment through
an olefination of the ketone (111). We envisioned hydroxyl-
directed hydrogenation would set the required relative stereo-
chemistry present in the natural product. Introduction of a
hydroxyl group or oxygen equivalent would immediately follow
the ruthenium-catalyzed cycloisomerization of the diyne sub-
strate 112. Based upon the hypothesis that less of the hydrated
product is formed when there is greater steric hindrance near
the propargylic alcohol, the cycloisomerization of 112 should
be favored over the corresponding hydration/cyclization.
The key cyclization precursor 112 can be made in two distinct
ways. First, aldimine 113 can be alkynylated with a protected
propargylic alcohol (116) to produce diyne 117. Protection of
the free amine with tosyl chloride and removal of the THP group
reveals the desired racemic cyclization substrate 118 (Scheme
14). Alternatively, a Mitsunobu reaction of a chiral propargylic
alcohol (-)-121 and a protected propargylic amine provides a
similar protected amino diynol 124 in a more convergent fashion
(Scheme 15). Asymmetric reduction of the ynone 122 using
the LiAlH4/BINOL/MeOH system developed by Noyori67,68
provided rapid, high-yielding, and nearly enantiopure access to
propargylic alcohol (-)-121. The sense of chirality of this
alcohol eventually leads to the unnatural enantiomer of R-kainic
acid; however, the natural enantiomer is readily available by
simply using (S)-BINOL in the asymmetric reduction.
Chemoselective hydroboration-oxidation to form primary
alcohol 128 was then readily accomplished using dicyclohexy-
lborane and a standard basic peroxide workup. The resulting
free alcohol was then protected as the nonligating TBS-silyl
ether (129). The homoallylic benzyl ether then needed to be
deprotected in order to carry out the planned directed hydro-
genation. Alternatively, the required oxygenation at the δ-car-
bon can be introduced through a 1,6-addition of an oxygen
surrogate group (-SiPhMe2) (Scheme 17). While much work
has been done on the 1,4-addition of silyl cuprates,71,72 the 1,6-
addition was unknown. When the phenyldimethylsilyl cuprate73
was added to (+)-125 at -78 °C no reaction occurred; however,
when the temperature was raised to 0 °C, a mixture of 1,6-
addition diastereomers 131 (in a 2.8:1 ratio) was isolated. The
olefin could be easily isomerized into conjugation with DBU
in refluxing benzene to give the desired product 132 in 87%
yield over the two steps. It is also of note that no racemization
occurred in the cyclization or conjugate addition steps. The
benzyl group was then removed using Pd/C in formic acid/
methanol to provide free alcohol 133.
Upon submitting diyne 118 to the standard cycloisomerization
conditions developed for primary propargylic alcohols (10 mol
% ruthenium catalyst 16, 10 vol % water/acetone, 60 °C), the
cycloisomerization product 125 can be isolated in good yield
(75%). The ratio of cycloisomerized to “hydrated” products (5:
1) was much greater in this case as compared to the simple
primary diynol substrate (Table 7). This is in agreement with
the observation that the more hindered the propargylic alcohol
is, the less hydrated product that is obtained.
The relative stereochemistry was then set by a directed
hydrogenation of the sterically hindered tetrasubstituted olefin
(130, 133). Various functional groups are known to participate
in a directed reduction including methyl ethers; however, no
reaction occurred with benzyl ether compounds 129 or 132 using
Crabtree’s catalyst,74,75 [Ir(cod)Py(PCy3)]PF6. On the basis of
the hindered nature of the olefin, the most active catalysts used
76-78
in hydroxyl-directed hydrogenations ([Rh(nbd)dppb]BF4
and [Ir(cod)py(PCy3)]PF674,75,79 (Crabtree’s catalyst)) were tested
first with the free alcohol compounds 130 and 133. Under all
conditions attempted, the [Rh] catalyst failed to give any
reduction product. Crabtree’s catalyst is known to be even more
active, and that proved to be the case here as well. Under 1
atm of hydrogen, no reaction occurred with either substrate,
but ketal 130 reacted utilizing a higher pressure of hydrogen
(8.2 atm) and 20 mol % [Ir]; reduction occurred to give a
quantitative yield of the expected reduced product (()-134 as
As was observed for simple substrates, lowering the temper-
ature increased the ratio favoring cycloisomerized product 125.
In this example (entries 3 and 4), the formation of the “hydrated”
product 126 can be completely eliminated; however, a slightly
higher isolated yield of the cycloisomerized product 125 can
be obtained if the reaction is performed at 45 °C in the presence
of 1 equiv of malonic acid, even though a small amount of 126
is formed as well. These two products are easily separable, and
thus the conditions from entry 2 are considered superior.
Scheme 14. Synthesis of Racemic Cyclization Precursor
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4771

A R T I C L E S
Trost and Rudd
Scheme 15. Enantioselective Synthesis of Cyclization Precursor
Table 7. Optimization of the Ruthenium-Catalyzed
Cycloisomerization
has even demonstrated that a ketone directs hydrogenation more
efficiently than a free alcohol, while ethylene glycol ketals
function poorly as directing groups.75 This may indicate that
the ketone is coordinating to the iridium catalyst and reducing
the rate of the desired hydroxyl-directed hydrogenation. Using
lower catalyst loadings, lower conversions were obtained, and
at higher catalyst loadings (50%), somewhat higher conversions
were obtained (0.7:1 133 to 135). Attempts to form the dimethyl
ketal in situ with trimethylorthoformate led to no reaction. Even
the isolated dimethyl ketal of 133 failed to react to any
significant extent. It was envisioned that if another Lewis acid
could break up this ketone-iridium coordination, then greater
catalyst turnover could be achieved. Use of Ti(OiPr)4 led to no
reaction, and acetic acid also produced no product. However,
use of 1 equiv of B(OiPr)3 along with 20 mol % Crabtree’s
catalyst led to increased catalyst turnover such that the desired
hydrogenated product (135) was isolated in 65% yield (95%
based on recovered 133) (Scheme 18). Use of excess B(OiPr)3
led to decreased yields.
ratio
isolated yield
entry
temp (
°
C)
additive (equiv)
125:126
of 125 (%)
1
2
3
4
60
45
rt
5:1
10:1
1:0
75
80
75
75
malonic acid (1)
malonic acid (1)
rt
1:0
a single diastereomer. It was subsequently found that 5% [Ir]
was sufficient (under 100 atm of hydrogen gas) to convert the
ketal 130 completely into a single reduced product (()-134 in
quantitative yield (Scheme 18).
The R,â-unsaturated ketone 133 was even more difficult to
reduce than the ketal-protected analogue 130; with 2000 psi
hydrogen and 20% [Ir], a 1:1 ratio of 133 to 135 was isolated
after 24 h. Similar to the reduction of 130, only a single
diastereomer resulting from directed hydrogenation was isolated.
Tetrasubstituted olefins and R,â-unsaturated alkenes are known
to be difficult to reduce with Crabtree’s catalyst, and Crabtree
The racemic synthesis of R-kainic acid was then completed
in five steps (Scheme 19). The methyl ketone was olefinated to
form diol (()-138 following removal of the ketal protecting
group and reprotection of the resulting diol as a bis-TBS ether.
Oxidation of diol (()-138 to bis-acid (()-139 was accomplished
Scheme 16. Synthesis of Directed Hydrogenation Precursor via Hydroboration
Scheme 17. Synthesis of Directed Hydrogenation Precursor via 1,6-Silylcuprate Addition
9
4772 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
Scheme 18. Hydroxyl-Directed Reduction of Tetrasubstituted
Olefins
138. Higher yields were obtained if the elimination was carried
out with HF/acetonitrile first, followed by the Woepel oxidation
to diol (+)-138 without any prior purification (Scheme 20). Both
primary alcohols were then oxidized simultaneously with Jones’s
reagent to give a quantitative yield of the diacid (-)-139. The
tosyl group was then removed in a known step86,87 with Li/
NH3 to give the unnatural enantiomer of R-kainic acid in 80%
yield over two steps (Scheme 20).
Conclusion
In this article, the extension of the propargylic alcohol
dimerization to an intramolecular reaction has been detailed.
Not only are the yields for the cycloisomerization very high
for a range of five- and six-membered ring-forming reactions,
but the reaction represents a novel chemoselective alternative
to aldol condensation. Tertiary and secondary propargylic
alcohol diynes cycloisomerize in the presence of a catalytic
amount of ruthenium complex 16 to yield R,â,γ,δ-unsaturated
aldehydes and ketones. Secondary propargylic alcohol diynes
cyclize to form the more thermodynamically stable E-dienone,
a single geometrical isomer. Primary propargylic alcohol diynes
cyclize as well, but in addition to the expected products, formally
hydrated cyclized products are isolated. A mechanistic proposal
(Scheme 11) has been put forth which supports the formation
of these products. It is reasonable that the primary alcohols react
via this alternate pathway because of reduced leaving group
ability. The product ratios presented can be explained by
proposing that attack of a molecule of water occurs at the least
hindered carbene carbon. This proposal was reinforced, and the
utility of the methodology was demonstrated, through the
synthesis of (+)-R-kainic acid. It is also evident that the presence
of a propargylic alcohol is not critical for the hydrative
using Jones’s reagent, and the tosyl group was removed using
Li/NH3 to yield racemic R-kainic acid.
The synthesis of (+)-R-kainic acid was also completed via a
related olefination, oxidation, and deprotection sequence. After
the initial step of the Peterson olefination,80,81 it was envisioned
that the elimination and the oxidation of the phenyldimethylsilyl
group could be carried out simultaneously. This in fact was
possible using the Woerpel modification82 of the Tamao-
Fleming oxidation.83-85 The standard Tamao-Fleming condi-
tions require strong acids or electrophiles to oxidize phenyldi-
methylsilyl groups, which were expected to be incompatible
with the olefin present in the molecule. The addition of
(trimethylsilylmethyl)lithium proceeded smoothly to give 140,
which could be directly converted to (+)-138 using the Woerpel
KH/t-BuOOH/TBAF conditions (Scheme 20).
However, some amount of a protodesilyated product 141 was
also isolated under these conditions, and modifications to
minimize this undesired product resulted in lower yields of (+)-
Scheme 19. Completion of Racemic Synthesis of R-Kainic Acid
Scheme 20. Completion of (+)-R-Kainic Acid
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4773

A R T I C L E S
Trost and Rudd
crude product which was further purified on silica (20% ether/petroleum
ether) to yield 29 mg (90%) of 59, which agreed with the H NMR
data for this known compound.
cyclization pathway to function, and thus the possibility of a
hydrative diyne cyclization on simple diynes is raised. This
avenue has been explored and was discussed in a previous
publication.98
1
Rf (40% Et2O/petroleum ether): 0.65. IR (neat): 2932, 2858, 1671,
1615, 1447, 1383, 1361, 1273, 1225, 1193, 1154, 1127, 1043, 965,
1
828, 745 cm-1. H NMR (300 MHz, CDCl3): 9.69 (s, 1H), 5.74 (s,
Experimental Section
1H), 2.23 (m, 2H), 2.16 (m, 2H), 1.83 (s, 3H), 1.63 (m, 4H), 1.59 (s,
3H). 13C NMR (75.4 MHz, CDCl3): 194.1, 157.0, 139.1, 135.0, 123.1,
32.1, 25.9, 22.2, 21.9, 21.8, 19.7. HRMS (EI, [M]+): calcd for C11H16O,
164.1201; found, 164.1194 (1.0), 163.1137(1.1), 150.1022 (10.6),
149.0990 (100), 147.1187 (2.4), 97.0647 (18.0), 91.0537 (16.3), 83.0849
(10.3).
3-Formyl-4-(2-methyl-propenyl)-cyclopent-3-ene-1,1-dicarboxy-
lic Acid Dimethyl Ester (29). To a test tube containing 28 (53 mg,
0.2 mmol) were added acetone (2.0 mL), water (5 µL, 0.27 mmol),
and catalyst 16 (1 mg, 0.002 mmol) under argon. The resulting yellow-
orange solution was then stirred for 1 h, after which it was filtered
through a pad of silica gel with Et2O as the eluent. The solvent was
then removed in vacuo to give the crude product which was further
purified by silica gel chromatography (60% Et2O/petroleum ether) to
yield 52 mg (98%) of 29.
2-(3-Methyl-but-1-enyl)-cyclohex-1-enecarbaldehyde (69). To a
test tube containing 68 (32 mg, 0.18 mmol), acetone (1.8 mL), and
water (5 µL, 0.3 mmol) was added catalyst 16 (8 mg, 0.018 mmol)
under argon. The resulting yellow solution was stirred at room
temperature for 2 h, after which it was filtered through a pad of silica
with ether containing 2% Et3N as the eluent. The solvent was then
removed in vacuo to give the crude product which was purified on
silica (10% ether/petroleum ether) to yield 29.5 mg (60%) of 69.
Rf (40% Et2O/petroleum ether): 0.80. IR (neat): 2931, 2858, 1664,
Rf (80% Et2O/petroleum ether): 0.65. Mp ) 86-87 °C. IR (neat):
2955, 2848, 1736, 1659, 1629, 1579, 1434, 1268, 1202, 1076 cm-1
.
1H NMR (300 MHz, CDCl3): 9.86 (s, 1H), 6.27 (s, 1H), 3.74 (s, 6H),
3.41 (s, 2H), 3.24 (s, 2H), 1.91 (s, 3H), 1.86 (s, 3H). 13C NMR (75.4
MHz, CDCl3): 188.1, 171.8, 155.5, 144.17, 135.0, 117.3, 57.3, 53.2,
45.7, 37.9, 27.7, 20.8. HRMS (EI, [M]+): calcd for C14H18O5, 266.1154;
found, 266.1164 (0.8), 252.0940 (100), 191.0721 (55.9), 147.0811
(16.3), 119.0859 (10.1).
1
1631, 1582, 1460, 1373, 1274, 1234, 1154, 961 cm-1. H NMR (300
MHz, CDCl3): 10.32 (s, 1H), 6.95 (d, J ) 16.0 Hz, 1H), 5.99 (dd, J
) 16.0 Hz, J ) 7.0 Hz, 1H), 2.47 (m, 1H), 2.42 (m, 2H), 2.29 (m,
2H), 1.7-1.6 (m, 4H), 1.08 (d, J ) 7.0 Hz, 6H). 13C NMR (75.4 MHz,
CDCl3): 191.2, 153.2, 144.1, 134.4, 122.6, 32.3, 28.2, 23.1, 22.5, 22.2,
21.8. HRMS (EI, [M]+): calcd for C12H18O, 178.1358; found, 178.1345
(3.8), 163.1124 (2.7), 161.1335 (13.4), 145.1014 (11.2), 135.0801 (100).
1-[5-Benzyloxymethyl-1-(toluene-4-sulfonyl)-4-vinyl-2,5-dihydro-
1H-pyrrol-3-yl]-ethanone (125). (()-125: To a test tube containing
118 (29 mg, 0.07 mmol), acetone (1 mL), and water (0.1 mL) were
added 16 (3 mg, 0.007 mmol) and malonic acid (7.3 mg, 0.07 mmol)
under argon. The resulting yellow solution was sealed and stirred at
45 °C for 1 h and filtered through a pad of silica with ether as the
eluent. The solvent was then removed in vacuo to yield a yellow oil
which was further purified on silica (60% ether/petroleum ether) to
yield 23 mg (80%) of (()-125.
(+)-125: To a flask containing 124 (1.4 g, 2.66 mmol), acetone
(27 mL), and water (0.54 mL) were added 16 (116 mg, 0.266 mmol)
and malonic acid (280 mg, 2.66 mmol) under argon. The resulting
yellow solution was sealed and stirred at 40 °C for 3 h and filtered
through a pad of silica with ether as the eluent. The solvent was then
removed in vacuo to yield a yellow oil which was further purified on
silica (60% ether/petroleum ether) to yield 870 mg (80%) of (+)-125.
Rf (60% ether/petroleum ether): 0.35. IR (neat): 3089, 3031, 2922,
2867, 1732, 1681, 1652, 1622, 1597, 1580, 1495, 1454, 1343, 1262,
1163, 1099, 1049, 1017, 925, 816, 740, 698, 665 cm-1. 1H NMR (300
MHz, CDCl3): 7.74 (d, J ) 8.1 Hz, 2H), 7.3-7.2 (m, 7H), 7.15 (dd,
J ) 17.8 Hz, J ) 11.4 Hz, 1H), 5.53 (d, J ) 11.4 Hz, 1H), 5.47 (d, J
) 17.8 Hz, 1H), 5.12 (bs, 1H), 4.56 (d, J ) 12.3 Hz, 1H), 4.45 (s,
2H), 4.48 (d, J ) 12.3 Hz, 1H), 3.38 (m, 2H), 2.44 (s, 3H), 2.24 (s,
3H). 13C NMR (75.4 MHz, CDCl3): 194.6, 144.8, 143.9, 138.0, 135.2,
132.8, 129.9, 128.6, 128.3, 127.6, 127.4, 127.1, 123.3, 73.5, 71.7, 67.9,
55.9, 30.6, 21.6. HRMS (EI, [M]+): calcd for C23H25NO4S, 411.1504;
found, 411.1509 (0.9), 290.0851 (45.3, -CH2OBn), 248.0738 (59.4),
155.0170 (25.4). [R]25D ) 69.64 (c ) 0.1, MeOH). The ee was evaluated
using a chiral HPLC: OD column, flow rate ) 1 mL/min, solvent )
90/10 IPA/heptane, retention times ) 17.30 (+), 20.66 (-) min.
1-[5-Benzyloxymethyl-4-[2-(dimethyl-phenyl-silanyl)-ethylidene]-
1-(toluene-4-sulfonyl)-pyrrolidin-3-yl]-ethanone (131). The solution
of Li(SiMe2Ph) was prepared as follows: To a cooled, dry flask with
a Schlenk fitting on top was added Li granules (0.208 g, 30 mmol).
The flask was then cooled to 0 °C, and THF (15 mL) and phenyldi-
methylsilyl chloride (1.66 mL, 10 mmol) were added under argon. Over
30 min the solution turned a dark red color, and the reaction was stirred
a further 4 h at 0 °C. The flask was then placed in a -15 °C freezer
1-[3-(2-Methyl-propenyl)-1H-inden-2-yl]-ethanone (34). To a test
tube containing 7 (41 mg, 0.19 mmol) were added acetone (1.9 mL),
water (5 µL, 0.27 mmol), and catalyst 16 (4 mg, 0.0095 mmol) under
argon. The resulting yellow-orange solution was then stirred for 1 h,
after which it was filtered through a pad of silica gel with Et2O as the
eluent. The solvent was then removed in vacuo to give the crude product
which was further purified by silica gel chromatography (40% Et2O/
petroleum ether) to yield 36 mg (89%) of 34.
Rf (40% Et2O/petroleum ether): 0.35. Mp ) 57-60 °C. IR (neat):
2967, 2922, 2858, 1649, 1556, 1446, 1370, 1288, 1247, 1184, 1147,
1
1016, 948 cm-1. H NMR (200 MHz, CDCl3): 7.51 (m, 1H), 7.33-
7.42 (m, 3H), 6.24 (m, 1H), 3.74 (d, J ) 2.6 Hz, 2H), 2.44 (s, 3H),
2.02 (d, J ) 1.4 Hz, 3H), 1.61 (d, J ) 0.8 Hz, 3H). 13C NMR (50.3
MHz, CDCl3): 197.1, 150.2, 144.9, 143.7, 140.7, 140.2, 128.1, 126.8,
124.4, 123.4, 118.5, 38.6, 29.7, 25.7, 20.5. HRMS (EI, [M]+): calcd
for C15H16O, 212.1201; found, 212.1198 (1.8), 197.0948 (100),
129.0699 (10.5), 128.0625 (10.9).
1-[2-(2-Methyl-propenyl)-4,5,6,7-tetrahydro-3H-inden-1-yl]-etha-
none (35). To a test tube containing 12 (61 mg, 0.22 mmol) were added
acetone (2.2 mL), water (5 µL, 0.27 mmol), and catalyst 16 (9.5 mg,
0.022 mmol) under argon. The resulting yellow-orange solution was
then stirred for 1 h, after which it was filtered through a pad of silica
gel with Et2O as the eluent. The solvent was then removed in vacuo to
give the crude product which was further purified by silica gel
chromatography (20% Et2O/petroleum ether) to yield 30 mg (63%) of
35.
Rf (40% Et2O/petroleum ether): 0.60. IR (neat): 2926, 2854, 1696,
1668, 1633, 1558, 1540, 1506, 1436, 1374, 1255, 1203, 1167 cm-1
.
1H NMR (300 MHz, CDCl3): 6.39 (s, 1H), 3.16 (broad s, 2H), 2.35
(s, 3H), 2.32 (m, 4H), 1.88 (s, 3H), 1.81 (s, 3H), 1.68 (m, 4H). 13C
NMR (75.4 MHz, CDCl3): 200.4, 146.9, 142.9, 138.9, 137.7, 136.4,
120.9, 46.7, 31.2, 28.1, 25.5, 24.8, 23.1, 22.8, 20.5. HRMS (EI, [M]+):
calcd for C15H20O, 216.1514; found, 216.1518 (2.9), 201.1291 (14.7),
82.9459 (100).
2-(2-Methyl-propenyl)-cyclohex-1-enecarbaldehyde (59). To a test
tube containing 58 (33 mg, 0.2 mmol), malonic acid (21 mg, 0.2 mmol),
acetone (3.6 mL), and water (0.4 mL) was added catalyst 16 (8.6 mg,
0.02 mmol) under argon. The resulting yellow solution was stirred at
60 °C for 1 h, after which it was filtered through a pad of silica with
ether as the eluent. The solvent was then removed in vacuo to give the
(98) Trost, B. M.; Rudd, M. T. J. Am. Chem. Soc. 2003, 125, 11516.
9
4774 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005

Ruthenium-Catalyzed Cycloisomerizations of Diynols
A R T I C L E S
overnight. The concentration was then measured by the Gillman double
titration method: A 0.5 mL aliquot was quenched with 2 mL of water
and titrated with standardized 0.1 M HCl using phenolphthalein to give
the total base. A second 0.5 mL aliquot was then quenched with 2 mL
of 1,2-dibromoethane in a dry test tube. After mixing for 30 s, 2 mL
of water was added and titrated with standardized 0.1 M HCl using
phenolphthalein to give the amount of alkoxide base. The amount of
alkoxide is subtracted from the total base to give the concentration of
silyllithium. The concentration was generally ∼0.5 M.
1-[4-[2-(Dimethyl-phenyl-silanyl)-ethyl]-5-hydroxymethyl-1-(tolu-
ene-4-sulfonyl)-pyrrolidin-3-yl]-ethanone (135). To a cooled, dry flask
under hydrogen containing 133 (121 mg, 0.265 mmol) were added
degassed (freeze-pump-thaw 3×) and hydrogen-purged (15 min H2
bubble through) DCM (13.2 mL) and B(O-iPr)3 (61 µL, 0.265 mmol).
Crabtree’s catalyst [Ir(cod)Py(PCy3)]PF6 (43 mg, 0.053 mmol) was then
added quickly under hydrogen. The flask was then added to a Parr
apparatus, sealed quickly under hydrogen, purged 2 times with 1000
psi hydrogen, and then sealed under 2000 psi hydrogen. The reaction
was stirred for 24 h, the pressure was released, and petroleum ether
was added to precipitate most of the iridium. The crude product was
then filtered through silica using ether as the eluent. The solvent was
then removed in vacuo to yield a yellow oil which was further purified
on silica (80% ether/petroleum ether) to yield 135 (79 mg, 65%) and
recovered 133 (40 mg).
To a cooled, dry flask containing CuCN (0.142 g, 1.58 mmol) was
added THF (5.5 mL) at 0 °C. The 0.52 M solution of Li(SiMe2Ph)
(8.0 mL, 4.1 mmol) was then added and stirred 25 min at this
temperature and then cooled to -78 °C. (+)-125 (0.62 g, 1.51 mmol)
in THF (5.5 mL, washed with 2 mL) was then added slowly. After 1
h, the reaction was warmed to 0 °C and stirred an additional 4 h.
Saturated aqueous NH4Cl was then added to quench the reaction, which
was then filtered through Celite to remove most of the copper residues.
The biphasic mixture was then extracted with ether and more NH4Cl.
The organic layers were then dried over MgSO4 and filtered, and the
solvent was removed in vacuo to yield a yellow oil which was further
purified on silica (50% ether/petroleum ether) to yield 0.82 g of a 2.8:1
mixture of diastereomers 131. The diastereomers can be separated on
silica (40% ether/petroleum ether) but were generally carried on as a
mixture.
Rf (50% ether/petroleum ether): 0.10. IR (neat): 3516, 3058, 2958,
2922, 2849, 1710, 1596, 1426, 1343, 1248, 1162, 1113, 1091, 1055,
835, 816, 731, 702, 667 cm-1. 1H NMR (400 MHz, C6D6): 7.70 (d, J
) 8.4 Hz, 2H), 7.28 (m, 2H), 7.13 (m, 3H), 6.72 (d, J ) 8.4 Hz, 2H),
3.82 (m, 2H), 3.6 (m, 3H), 3.05 (m, 1H), 2.99 (bs, 1H), 2.10 (m, 1H),
1.87 (s, 3H), 1.49 (s, 3H), 0.64 (m, 1H), 0.41 (m, 1H), 0.17 (m, 2H),
-0.06 (s, 3H), -0.11 (s, 3H). 13C NMR (100 MHz, C6D6): 204.5,
143.3, 138.6, 133.8, 133.6, 129.8, 129.2, 128.0, 127.9, 66.1, 65.6, 53.2,
48.2, 46.6, 28.6, 22.4, 21.1, 13.2, -3.06, -3.90. LRMS (CI, [M +
1]+): calcd for C24H33NO4SSi, 460.1; found, 460.1. HRMS (EI, [M -
CH2OH]+): calcd for C23H30NO3SSi, 428.1716; found, 428.1730 (34.4).
Data for major diastereomer of 131: Rf (50% ether/petroleum
ether): 0.2. IR (neat): 3067, 3029, 2955, 2922, 2856, 1715, 1598, 1495,
1454, 1427, 1403, 1347, 1249, 1161, 1113, 1093, 1027, 835, 734, 700,
306.1008 (6.4), 135.0627 (100). [R]25 ) -15.83 (c ) 0.75, MeOH).
D
1
666 cm-1. H NMR (400 MHz, C6D6): 7.78 (d, J ) 8 Hz, 2H), 7.31
1-[4-(2-Hydroxy-ethyl)-5-hydroxymethyl-1-(toluene-4-sulfonyl)-
pyrrolidin-3-yl]-ethanone (136). To a solution of 134 (150 mg, 0.3
mmol) in acetone (15 mL) was added pTSA monohydrate (58 mg, 0.3
mmol). The reaction was stirred for 24 h and filtered through silica
(50% DCM/acetone) with ether as the eluent. The solvent was then
removed in vacuo. The crude product was purified further (50% DCM/
acetone) to yield 89 mg (87%) 136 as a mixture of acetal and open
hydroxy ketone isomers.
(m, 2H), 7.17-7.0 (m, 8H), 6.74 (d, J ) 8 Hz, 2H), 5.13 (t, J ) 7.6
Hz, 1H), 4.56 (bs, 1H), 4.21 (dd, J ) 10 Hz, J ) 2.8 Hz, 1H), 4.18 (d,
J ) 12 Hz, 1H), 4.13 (d, J ) 12 Hz, 1H), 3.57 (m, 2H), 3.17 (dd, J )
10 Hz, J ) 8 Hz, 1 H), 2.54 (d, J ) 4.8 Hz, 1H), 1.83 (s, 3H), 1.67 (s,
3H), 1.50 (t, J ) 7.6 Hz, 2H), 0.058 (s, 3H), 0.051 (s, 3H). 13C NMR
(125 MHz, C6D6): 202.9, 142.9,138.3, 138.0, 136.5, 134.4, 133.7,
129.6, 129.5, 128.4, 128.2, 127.8, 127.6, 73.5, 73.0, 60.6, 57.2, 47.8,
30.1, 26.6, 21.0, 19.5, -3.2. HRMS (EI, [M]+): calcd for C31H37NO4-
SSi, 547.2213; found, 547.2233 (0.1), 504.2039 (0.5), 426.1570 (38.0),
135.0623 (100).
Rf (50% acetone/DCM): 0.55. IR (neat): 3454, 2951, 2882, 1709,
1
1597, 1453, 1341, 1162, 1049, 918, 668 cm-1. H NMR (300 MHz,
CD2Cl2): 7.72 (d, J ) 8.4 Hz, 2H), 7.37 (d, J ) 8.4 Hz, 2H), 3.79-
3.47 (m, 4H), 3.25 (m, 2H), 3.00 (dd, J ) 11.7 Hz, J ) 9 Hz, 1H),
2.43 (s, 3H), 2.30 (m, 2H), 1.24 (s, 3H). 13C NMR (75.4 MHz, CD2-
Cl2): 144.6, 133.5, 130.2, 127.9, 94.6, 68.7, 65.6, 59.0, 49.0, 43.2,
36.6, 28.3, 24.9, 21.7. HRMS (EI, [M - CH2OH]+): calcd for C15H20-
NO4S, 310.1113; found, 310.1123 (97.3).
δ-Kainic Acid. (()-δ-Kainic Acid: To a cooled, dry flask contain-
ing (()-139 (8 mg, 0.022 mmol) was added THF (0.2 mL). The flask
was cooled to -78 °C, and liquid ammonia (∼1 mL) was added. Li
granules (∼1 mg, 0.16 mmol) were then dropped into the flask. A blue
color immediately developed. The reaction was stirred 30 min and
quenched with isoprene (∼1 mL or until the blue color disappeared).
The mixture was warmed to room temperature, and all the volatile
components were blown off under argon. Water (0.2 mL) was then
added, and the pH was adjusted to ∼7. The solution was then passed
through an ion exchange column (Amberlight CG-50) eluting with water
and then 3% ammonium hydroxide. The fractions containing kainic
acid were then collected, and water was removed on a lyophilizer to
yield kainic acid, which was recrystallized from ethanol/water (3.8 mg,
80%).
(+)-δ-Kainic Acid: To a cooled, dry flask containing (-)-139 (10
mg, 0.025 mmol) was added THF (0.3 mL). The flask was cooled to
-78 °C, and liquid ammonia (∼1.5 mL) was added. Li granules (∼1
mg, 0.16 mmol) were then dropped into the flask. A blue color
immediately developed. The reaction was stirred 30 min and quenched
with isoprene (∼1 mL or until the blue color disappeared). The mixture
was warmed to room temperature, and all the volatile components were
blown off under argon. Water (0.2 mL) was then added, and the pH
was adjusted to ∼7. The solution was then passed through an ion
1-[5-Benzyloxymethyl-4-[2-(dimethyl-phenyl-silanyl)-ethyl]-1-
(toluene-4-sulfonyl)-2,5-dihydro-1H-pyrrol-3-yl]-ethanone (132). To
a cooled, dry flask containing 131 (0.82 g, 1.5 mmol) were added
benzene (55 mL) and DBU (45 µL, 0.3 mmol). The solution was then
heated at reflux for 7 h and extracted with ether and dilute HCl. The
organic layers were dried over MgSO4 and filtered, and the solvent
was removed in vacuo to yield a yellow oil which was further purified
on silica (30% ether/petroleum ether) to yield 132 (0.72 g, 87% over
two steps).
Rf (50% ether/petroleum ether): 0.25. IR (neat): 3058, 3030, 2958,
2923, 2854, 1687, 1658, 1626, 1598, 1495, 1454, 1427, 1347, 1259,
.
1163, 1112, 1093, 816, 734, 700, 665 cm-1 1H NMR (400 MHz,
C6D6): 7.69 (d, J ) 8 Hz, 2H), 7.34 (m, 2H), 7.2-7.0 (m, 8H), 6.68
(d, J ) 8 Hz, 2H), 4.69 (m, 1H), 4.43 (d, J ) 14 Hz, 1H), 4.34 (ddd,
J ) 13.6 Hz, J ) 4.4 Hz, J3 ) 2 Hz, 1H), 4.30 (d, J ) 12 Hz, 1H),
4.18 (d, J ) 12 Hz, 1H), 3.76 (dd, J ) 10 Hz, J ) 4.8 Hz, 1H), 3.66
(dd, J ) 10 Hz, J ) 2.4 Hz, 1H), 2.69 (td, J ) 13.2, J ) 4.4 Hz, 1H),
2.0 (td, J ) 13.2, J ) 4 Hz, 1H), 1.76 (s, 3H), 1.45 (s, 3H), 0.69 (td,
J ) 14.2 Hz, J ) 4.4 Hz, 1H), 0.45 (td, J ) 14.2 Hz, J ) 4.4 Hz, 1H),
0.14 (s, 3H), 0.13 (s, 3H). 13C NMR (100 MHz, C6D6): 193.2, 154.2,
143.2, 138.4, 138.3, 135.8, 133.7, 130.2, 129.7, 129.2, 128.5, 128.0,
127.8, 127.6, 73.5, 71.7, 69.4, 56.2, 29.4, 21.3, 20.9, 14.2, -3.5, -3.6.
HRMS (EI, [M]+): calcd for C31H37NO4SSi, 547.2213; found, 547.2224
(0.2), 426.1501 (16.5), 398.1307 (26.2), 348.1115 (16.2), 306.1022
(21.9), 135.0621 (100). [R]25 ) 109.35 (c ) 0.24, MeOH). The ee
D
was evaluated using a chiral HPLC: OD column, flow rate ) 1 mL/
min, solvent ) 90/10 IPA/heptane, retention times ) 15.14 (+), 18.66
(-) min.
9
J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005 4775

A R T I C L E S
Trost and Rudd
exchange column (Amberlight CG-50) eluting with water and then 3%
ammonium hydroxide. The fractions containing kainic acid were then
collected, and water was removed on a lyophilizer to yield kainic acid,
Acknowledgment. We thank the National Institutes of Health
(GM-13598) and the National Science Foundation for their
generous support of our programs. Mass spectra were provided
by the Mass Spectrometry Regional Center of the University
of California-San Francisco, supported by the NIH Division
of Research Resources.
1
which was recrystallized from ethanol/water (4.2 mg, 80%). The H
NMR was consistent with the known spectra, but the exact location of
signals is known to vary somewhat with concentration. The pictured
spectrum has a mis-referenced H2O peak (the values below are
corrected).
Supporting Information Available: Experimental procedures
and spectroscopic characterization (IR, 1H and 13C NMR,
HRMS, or combustion analysis) for all other compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
Mp ) 240-245 °C (lit. mp 243-244 °C). H NMR (500 MHz,
D2O): 5.0 (s, 1H), 4.71 (s, 1H), 4.03 (d, J ) 3.0 Hz, 1H), 3.60 (dd, J
) 12.5 Hz, J ) 7.0 Hz, 1H), 3.41 (t, J ) 11.0 Hz, 1H), 2.91 (m, 2H),
2.23 (dd, J ) 15.5 Hz, J ) 6.5 Hz, 1H), 2.12 (dd, J ) 15.5 Hz, J )
8.0 Hz, 1H), 1.76 (s, 3H). [R]25D ) +13.7 (c ) 0.13, H2O). (lit. [R]25
D
) -14.2 (c ) 0.18, H2O).
JA043097O
9
4776 J. AM. CHEM. SOC. VOL. 127, NO. 13, 2005