Reactions of Cobalt-Complexed Acetylenic Aldehydes with Chiral (γ-Alkoxyallyl)boranes: Enantioselective Synthesis of 3,4-Dioxy 1,5-Enynes and Stereoselective Entry to Polyfunctional Building Blocks

Article abstract of DOI:10.1021/jo9619387

Toward the goal of developing enantioselective routes to polyfunctional building blocks using Co-based technology, the reactions of (Z)- and (E)-(γ-alkoxyallyl)diisopinocampheylboranes (8, 13) with acetylenic aldehydes and their -Co₂(CO)₆ complexes (10a-c) have been investigated. Whereas phenylpropynal reacts with (Z)-8 with poor efficiency (< 10percent yield) and moderate enantioselectivity, the corresponding Co derivatives 10 undergo high-yielding (65-76percent), diastereoselective (88- >95percent de), and enantioselective (>96percent ee) reactions, producings syn-(3,4-dioxy 1,5-enyne)Co₂(CO)₆ 11a-f. Similarly, the newly prepared (-E)-boranes 13 also react efficienctly with the complexed aldehydes, affording the corresponding (anti-3,4-dioxy 1,5-enyne)Co₂(CO)₆ complexes 15a-c with virtually complete diastereo- and enantioselectivity, while phenylpropynal itself reacts inefficiently and with lower enantioselectivity. Oxidative demetalation of the complexed adducts affords the 3,4-dioxy 1,5-enynes 17 with complete maintenance of stereochemical integrity. Osmium-catalyzed dihydroxylation of a representative dioxy enyne, i.e., 17c, proceeds with high diastereoselectively (9:1), favoring addition anti to the allylic alkoxy group. Although epoxidation of enynediol 25 by t-BuOOH/VO(acac)₂ occurs stereoselectively, it is accompanied by epimerization of the diol unit.

Full text of DOI:10.1021/jo9619387

J . Org. Chem. 1997, 62, 1737-1747
1737
Rea ction s of Coba lt-Com p lexed Acetylen ic Ald eh yd es w ith Ch ir a l
(γ-Alk oxya llyl)bor a n es: En a n tioselective Syn th esis of 3,4-Dioxy
1,5-En yn es a n d Ster eoselective En tr y to P olyfu n ction a l Bu ild in g
Block s
P. Ganesh and K. M. Nicholas*
Department of Chemistry and Biochemistry, University of Oklahoma, Norman, Oklahoma 73019
Received October 16, 1996X
Toward the goal of developing enantioselective routes to polyfunctional building blocks using Co-
based technology, the reactions of (Z)- and (E)-(γ-alkoxyallyl)diisopinocampheylboranes (8, 13) with
acetylenic aldehydes and their -Co2(CO)6 complexes (10a -c) have been investigated. Whereas
phenylpropynal reacts with (Z)-8 with poor efficiency (<10% yield) and moderate enantioselectivity,
the corresponding Co derivatives 10 undergo high-yielding (65-76%), diastereoselective (88->95%
de), and enantioselective (>96% ee) reactions, producing syn-(3,4-dioxy 1,5-enyne)Co2(CO)6 11a -f.
Similarly, the newly prepared (E)-boranes 13 also react efficienctly with the complexed aldehydes,
affording the corresponding (anti-3,4-dioxy 1,5-enyne)Co2(CO)6 complexes 15a -c with virtually
complete diastereo- and enantioselectivity, while phenylpropynal itself reacts inefficiently and with
lower enantioselectivity. Oxidative demetalation of the complexed adducts affords the 3,4-dioxy
1,5-enynes 17 with complete maintenance of stereochemical integrity. Osmium-catalyzed dihy-
droxylation of a representative dioxy enyne, i.e., 17c, proceeds with high diastereoselectively (9:1),
favoring addition anti to the allylic alkoxy group. Although epoxidation of enynediol 25 by t-BuOOH/
VO(acac)2 occurs stereoselectively, it is accompanied by epimerization of the diol unit.
In tr od u ction
excellent syn diastereoselectivity has been observed in
the reactions of [(Z)-γ-alkoxyallyl]boranes5 and bor-
onates6 with aldehydes. Additionally, good to excellent
enantioselectivity has been demonstrated in reactions of
aldehydes with homochiral [(Z)-alkoxyallyl]boronates7
and diisopinocampheylboranes8
The development of stereocontrolled methods for the
construction of acyclic structures containing multiple
asymmetric centers has received much attention because
of its importance for the efficient synthesis of highly
functionalized, biologically active molecules.1 Methodol-
ogy that leads to the stereoselective formation of syn- or
anti-1,2-diol units with the simultaneous generation of
carbon-carbon bonds has been very attractive in this
context. Among the most important methods for produc-
ing such functionality are pinacol coupling2 and R-alkoxy-
allylation of aldehydes by (γ-alkoxyallyl)metal reagents,
with boron3 and tin4 derivatives receiving the most
attention (eq 1 a,b). Among the boron-based reagents,
The stereocontrolled synthesis of anti diol derivatives
by this approach (eq 1b) has been less widely studied, in
part because of the configurational instability of (E)-γ-
alkoxyallyl anions.9 This problem was overcome by
Hoffmann’s group,10 who synthesized achiral (E)-γ-alkoxy-
substituted allylboronates (3) at low temperature and
found that they add to aldehydes, giving anti-1,2-diols
with excellent diastereoselectivity. Prior to the present
report, enantioselective variants of this methodology were
not available.
Recently, our group has been interested in the potential
utility of 3,4-dioxy enyne derivatives 6 and 7 as building
blocks for the stereoselective synthesis of complex mol-
ecules possessing multiple adjacent stereocenters. The
inherently unsymmetrical, chiral, and multifunctional
nature of these compounds offers promise for chemo-,
regio-, and enantioselective functionalization of the un-
saturated units. Moreover, the opportunity to compare
in one system the relative reactivity of the double and
triple bonds and to assess the stereodirecting effects of
the hydroxy versus alkoxy groups is also of interest. At
the inception of this project we envisioned that these
compounds could be prepared via the addition of (γ-
* To whom correspondence should be addressed. Tel.: (405) 325-
3696. Fax: (405) 325-6111. E-mail: knicholas@uoknor.edu.
X Abstract published in Advance ACS Abstracts, February 15, 1997.
(1) Nogradi, M. Stereoselective Synthesis, 2nd ed.; VCH: Weinheim,
Germany, 1995. Control of Acyclic Stereochemistry. Mukaiyama, T.,
Ed. Tetrahedron, Symposium-in-Print 1984, 40(12).
(2) Robinson, G. M. Pinacol Coupling Reactions. In Comprehensive
Organic Chemistry; Trost, B. M., Ed.; Pergamon: Oxford, 1991; Vol.
3, Chapter 2.6, pp 563-610.
(3) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207.
(4) (a) Keck, G. E.; Abbott, D. E.; Wiley, M. R. Tetrahedron Lett.
1987, 28, 139. (b) Marshall, J . A. Chem. Rev. 1996, 96, 31.
(5) Hoffmann, R. W. Pure Appl. Chem. 1988, 60, 123.
(6) Hoffmann, R. W.; Kemper, B. Tetrahedron Lett. 1982, 23, 845.
(7) Roush, W. R.; Grover, P. T. Tetrahedron Lett. 1992, 48, 1981.
(8) Brown, H. C.; J adhav, K. P.; Bhat, K. S. J . Am. Chem. Soc. 1988,
110, 1535.
(9) Hoffmann, R. W.; Angew. Chem., Int. Ed. Engl. 1982, 21, 555.
(10) Hoffmann, R. W.; Kemper, B.; Metternich, R.; Lehmeier, T.
Liebigs Ann. Chem. 1985, 2246.
S0022-3263(96)01938-X CCC: $14.00 © 1997 American Chemical Society

1738 J . Org. Chem., Vol. 62, No. 6, 1997
Ganesh and Nicholas
Sch em e 1
Ta ble 1. P r ep a r a tion of (syn -3,4-Dioxy
alkoxyallyl)metal reagents to acetylenic aldehydes (Scheme
1). Prior studies in our laboratory11 and others12 have
demonstrated that cobalt complexes of acetylenic alde-
hydes and acetals, (RCtCCHO)Co2(CO)6 and [RCtCCH-
(OR)2]Co2(CO)6, undergo highly diastereoselective aldol
reactions with silyl enol ethers. In a related development
Roush and Park found enhanced enantioselectivity in the
asymmetric allyl- and crotylboration of complexed acety-
lenic and propargylic aldehydes relative to the uncom-
plexed aldehydes.13
We report herein the results of a study of the reactions
of (Z)- and (E)-(γ-alkoxyallyl)diisopinocampheylboranes
8 and 13 with acetylenic aldehydes and their -Co2(CO)6
complexes. Whereas the free acetylenic substrates react
with poor efficiency and limited enantioselectivity with
8 and 13, the corresponding cobalt derivatives undergo
efficient regio-, diastereo-, and enantioselective reactions,
producing 3,4-dioxy 1,5-enynes (after demetalation); a
preliminary study of the reactions with (Z)-8 has been
reported.14 Also described is an investigation of two
representative functionalization reactions of the dioxy
enyne adducts, i.e., dihydroxylation and epoxidation.
High chemo- and diastereoselectivity has been found in
both cases, underscoring the potential of these intermedi-
ates for the selective synthesis of molecular targets
containing multiple adjacent stereocenters. Unexpect-
edly, however, epoxidation by VO(acac)2 is accompanied
by epimerization of the diol unit.
1,5-en yn e)Co2(CO)6 11
aldehyde
complex
adduct
%
%
%
ee
borane
complex yielda deb
10a (R ) Ph) 8a (R ) Me)
11a
11bd
11c
11d
11e
11f
75
75
62
73
65
76
>95 >98c
>95 >98c
>95 >96e
>95 >96e
>95 >96e
88 >96e,f
10a
10a
8b (R ) Me)d
8c (R ) CH2OMe)
10b (R ) Me) 8a
10b
8c
8a
10c (R ) H)
Yield after chromatography. Determined by 1H and 13C
a
b
d
d
NMR. Determined by chiral GC on cyclodextrin. Enantiomer
of 8a and 11a . e Determined by 19F NMR of Mosher ester deriva-
tive. f % ee of major diastereomer.
indicated only moderate enantioselectivity (65% ee).
Reactions of the corresponding -Co2(CO)6 complexes
of the acetylenic aldehydes provided a dramatic improve-
ment in efficiency and selectivity. Compounds 10 were
prepared in good yield by complexation of the readily
available acetylenic acetals (Co2(CO)8/20 °C) followed by
Amberlyst-catalyzed hydrolysis (acetone/H2O) at room
temperature.11,17 The reactions of 10 with (alkoxyallyl)-
boranes 8, when carried out as for the free aldehyde, gave
dark red (3-alkoxy-4-hydroxy-1-en-5-yne)Co2(CO)6 com-
plexes 11 in good yield (eq 3, Table 1).14
Resu lts a n d Discu ssion
Addition s of [(Z)-γ-Alkoxyallyl]diisopin ocam ph eyl-
b or a n es t o Alk yn yl Ald eh yd es a n d Th eir Cob a lt
Com plexes. [(Z)-γ-Alkoxyallyl]diisopinocampheylboranes
8 were prepared by metalation of the corresponding allyl
ethers and subsequent treatment with B-methoxy-(+)-
or -(-)-diisopinocampheylborane.8 The reactions of 2-bu-
tynal and 3-phenylpropynal with borane 8 were carried
out by treatment of the borane with 1.3 equiv of BF3‚
Et2O in THF at -78 °C followed by addition of the
aldehyde. After workup, the 3,4-dioxy 1,5-enyne deriva-
tives 9 were isolated in poor yields (5-10%), apparently
the result of polymerization of the acetylenic aldehydes
(formation of a gelatinous, insoluble residue) under the
reaction conditions. NMR and GC/MS analysis suggested
the presence of a single diastereomer in each case, but
chiral GC15 and Mosher ester16 analysis of the products
The H and 13C NMR spectra of the product complexes
1
11 indicated the presence of a single diastereomer (>97%)
that was tentatively assigned the syn stereochemistry on
the basis of literature precedent and the characteristic
chemical shift of the -OH proton (vide infra).10 This
assignment was confirmed by an X-ray structure deter-
mination of the major isomer of 11f.14 Only in the
addition of 8a to the parent complex 10c (R ) H) was
any of the anti-diastereomer detected.
(11) (a) J u, J .; Reddy; B. R.; Khan, M. A.; Nicholas, K. M. J . Org.
Chem. 1989, 54, 5426. (b) Tester, R.; Varghese, V.; Montana, A. M.;
Khan, M.; Nicholas, K. M. J . Org. Chem. 1990, 55, 186.
(12) Schreiber, S. L.; Klimas, M. T.; Sammakia, T. J . Am. Chem.
Soc. 1986, 108, 3128.
(13) (a) Roush, W. R.; Park, J . C. J . Org. Chem. 1990, 55, 1143. (b)
Roush, W. R.; Park, J . C. Tetrahedron Lett. 1991, 32, 6285.
(14) Ganesh, P.; Nicholas, K. M. J . Org. Chem. 1993, 58, 5587.
(15) Chiral GC analysis was performed on a cyclodextrin B column
(30 m), 70 °C (5 min) f 0 °C 5°/min.
(17) Coppola, G. M. Synthesis 1984, 1021.
(18) Roush, W. R.; Michaelides, M. R.; Tai, D. F.; Lesur, B. M.;
Chong, W. K. M.; Harris, D. J . J . Am. Chem. Soc. 1989, 111, 2984 and
references therein.
(19) Saha, M.; Muchmore, S; Van der Helm, D.; Nicholas, K. M. J .
Org. Chem. 1986, 51, 1960 and references therein.
(16) Dale, T. A.; Mosher, H. S. J . Am. Chem. Soc. 1973, 95, 512.

Reactions of Co-Complexed Acetylenic Aldehydes
J . Org. Chem., Vol. 62, No. 6, 1997 1739
Sch em e 2
Ta ble 2. P r ep a r a tion of (a n ti-3,4-Dioxy
Moreover, analysis of 11a -f using chiral NMR shift
reagent Eu(hfacam)3 pointed to the formation of a single
enantiomer (g95%) in each case as well. This conclusion
was confirmed by a combination of Mosher ester 19F NMR
characterization and chiral GC (cyclodextrin B) studies
of the demetalated compounds (vide infra). Furthermore,
reaction of (+)- or (-)-borane 8 with 10a (Table 1, entries
1 and 2) gave opposite enantiomers of comparable optical
purity. Although the reactions with MOM derivative 8c
resulted in somewhat lower yields, the high syn selectiv-
ity and enantioselectivity were still preserved. It is
noteworthy that the enantioselectivities found in the
present study are somewhat higher than those reported
by Roush and Park for the corresponding reactions of 5
with tartrate-derived crotylboranes.13 However, whether
this modest effect is derived from differences in the chiral
group of the borane, the γ-alkoxy vs alkyl group, or the
differing acetylenic substituents is unclear.
1,5-en yn e)Co2(CO)6 (15)
aldehyde
borane
product % yielda % deb % ee
10a (R ) Ph) 13 (R ) Me)
10b (R ) Me) 13
15a
15b
15c
50
44
20
>95
>95
>95
>95c
>95c
>95c
10c (R ) H)
13
Yield after chromatography. Determined by 1H and 13C
NMR. c Determined by 19F NMR of Mosher ester derivative and
Eu(hfacam)3 analysis.
a
b
Sch em e 3
P r ep a r a tion a n d Ad d ition s of [(E)-γ-Alk oxya llyl]-
d iisop in oca m p h eylb or a n es t o Alk yn yl Ald eh yd es
a n d Th eir Coba lt Com p lexes. We then sought access
to the unknown [(E)-γ-methoxyallyl]diisopinocampheyl-
boranes 13, which could lead to the anti adducts 15.
Hoffmann’s procedure for synthesizing (E)-γ-alkoxyallyl-
boronates was adapted to gain entry to 13. (E)-1-
Methoxy-3-(phenylthio)propene (12) was reduced with 2
equiv of potassium naphthalenide at -120 °C followed
by reaction with B-methoxydiisopinocampheylborane; the
resulting [(E)-methoxyallyl]-diisopinocampheylborane so-
lution was treated first with BF3‚Et2O at -78 °C followed
by addition of phenylpropynal (5) and gradual warming
to 25 °C (Scheme 2). After ethanolamine workup and
chromatography, the adduct 14 was obtained in low yield
(15%), as with the additions of the [(Z)-γ-alkoxyallyl]-
boranes to the uncomplexed aldehyde 5. Isolation of 14
also was complicated by the similarity of its Rf value to
the coproduct (E)-1-methoxy-3-(phenylthio)propene. This
problem was overcome by complexation of crude 14 with
cobalt octacarbonyl; separation of the cobalt complex 15a
by flash chromatography and subsequent demetalation
(Ce(IV)) gave pure 14. Although the yield of 14 was low,
a single diastereomer apparently was produced, judged
to be anti by 1H NMR by comparing the complexed
product of 14 and 15a . Its enantiomeric purity was
determined to be 89% ee by 19F NMR analysis of its
Mosher ester.
plexes 16 presumably arise from competing â-hydride
transfer from the borane reagent. 1H and 13C NMR
analysis of the crude reaction products detected only a
single diastereomer (>97% de) in each case, assigned the
anti stereochemistry. The hydroxy proton resonances of
the isolated 15 were uniformly found 0.7-1.0 ppm upfield
from those of the syn adducts 11, a feature seen for a
related series of compounds.10 The enantiomeric purity
of 15a -c was determined by chiral NMR shift reagent
1
analysis and by 19F and H NMR analysis of the Mosher
esters of the demetalated adducts. In all cases, high
enantiomeric excesses were observed (>95%).
The diastereoselectivities observed in these reactions
can be explained satisfactorally by the familiar six-
membered chelated transition state model for allylmetal
additions to carbonyl compounds10,20 (Scheme 3). Thus,
of the transition states involving the [(Z)-alkoxyallyl]-
borane, that leading to the syn adduct (i.e., A) avoids a
1,3-diaxial interaction between the bulky (alkynyl)Co2-
(CO)6 and isopinocampheyl units. Conversely, with the
[(E)-alkoxyallyl]borane additions the more stable transi-
tion state (A′) is that leading to the anti product. The
somewhat lower diastereoselectivity obtained from the
addition of (Z)-8 to the parent complex 10c (R ) H) may
reflect its lesser steric requirement derived from the bent
Corresponding reactions between (E)-13 and the com-
plexed acetylenic aldehydes 10a -c proved more efficient,
affording moderate yields (Table 2) of the complexes
15a -c, accompanied by variable amounts (25%) of alco-
hol complexes 16 derived from aldehyde reduction. Ac-
cordingly, the yields obtained in the E-borane series were
somewhat lower than from the corresponding [(Z)-γ-
methoxyallyl]diisopinocampheylboranes. The alcohol com-
(20) Bhat, K. S.; Brown, H. C. J . Am. Chem. Soc. 1986, 108, 5919.

1740 J . Org. Chem., Vol. 62, No. 6, 1997
Ganesh and Nicholas
Ta ble 3. Decom p lexa tion of En yn e Com p lexes by
(NH4)2Ce(NO3)2
complex
R
X
Y
product % yield
% ee
11a
11b
11c
11d
11e
11f
15a
15b
15c
Ph
Ph
Ph
OMe
OMe
OCH2OMe
H
H
H
H
H
H
OMe
OMe
OMe
17a
17b
17c
17d
17e
17f
18a
18b
18c
72
70
70
64
60
50
70
65
40
>98a,b
g98a
g98b
g98b
g98b
g98b
g98b
g98b
g98b
CH3 OMe
CH3 OCH2OMe
H
Ph
CH3
H
gained prominence for the diastereo- and enantioselective
elaboration of C-C double bonds.21 In the former cases,
Kishi22 and later others23 have found that high diaste-
reoselectivity is obtained with substrates possessing
allylic alkoxy or hydroxy groups, with preferential intro-
duction of the hydroxyl groups anti to the oxygenated
group. It thus appeared that the dioxy enyne intermedi-
ates would be excellent candidates for stereoselective
dihydroxylation of the C-C double bond. It has been
observed that triple bonds have lower reactivity than
double bonds; the dihydroxylations of conjugated enynes
occur exclusively at the double bond giving rise to
ynediols.24
The osmylation of the selected enyne adduct 17c was
carried out by adding OsO4 (5 mol %) and 2 equiv of
N-methylmorpholine N-oxide to 17c in acetone/water (5:
1) with stirring at 0-25 °C for 12 h.25 After complete
conversion was confirmed by TLC, chromatographic
isolation gave crude triol 19 as an oil (75%, eq 5). The
Me
H
H
H
GC analysis on cyclodextrin B. b 19F NMR analysis of Mosher
a
ester.
geometry of the coordinated alkyne.19 This effect has
been noted in the reactions of silyl enol ethers with 1011
and the related [(RCtCCR2)Co2(CO)6]+ complexes.12 Al-
though the low yield obtained in the addition to the
uncomplexed acetylenic aldehyde clouds the issue, it is
perhaps surprising that the diastereoselectivity is so high
in this case given the limited steric demand of the
“skinny” linear alkynyl unit.
The origin of the increased enantioselectivity in the
reactions of the complexed aldehydes is less apparent.
Indeed, a detailed analysis of the transition states for
additions of the (γ-alkoxyallyl)diisopinocamphenylborane
additions to aldehydes is lacking.20 Once again, assum-
ing chairlike transition states, the basis for enantiose-
lection presumably lies in the asymmetry created by the
conformational preferences of the bulky, chiral isocam-
pheyl units. Examination of hand-held and computer-
generated (MMX) models suggest that by minimizing the
steric interactions between the isocampheyl units a chiral
pocket is created on the side of the boron at which the
reacting allyl and aldehyde fragments are assembled (A,
B below). The methano bridge of the pseudoequatorial
1H NMR spectrum of 19 indicated an 85:15 mixture of
diastereomers; decreasing the reaction temperature to 0
°C increased the diastereoselectivity to 90:10. Unfortu-
nately, the isomeric triols 19a ,b could not be separated
1
by TLC. The H NMR spectrum of the mixture showed
an apparent triplet at 4.85 for the propargylic proton and
multiplets at 4.02 ppm (1H) and at 3.8 ppm (3H) for the
methylene and methine protons of the diol unit, respec-
tively, of the major isomer.
Since the relative stereochemistry of 19a ,b was un-
certain on the basis of the NMR analysis, separable,
crystalline derivatives were sought. Acetylation of 19a ,b
(Ac2O, pyridine) occurred quantitatively to give a 9:1
mixture of oily triacetates 20a ,b, but these, too, were
inseparable by TLC/flash chromatography (Scheme 4).
Likewise, complexation of triols 19a ,b with Co2(CO)8 also
produced an inseparable mixture. Finally, the cobalt
carbonyl complexes of triacetate 20 were prepared and
separated (21a ,b Scheme 4). 1H NMR analysis of crude
21a ,b reconfirmed the diastereomeric ratio (9:1). Im-
isocampheyl unit appears to be the primary stereodif-
ferentiating element by its relative steric and electronic
interactions with the nearby allyl methylene group
(greater in B) vs the aldehyde oxygen (lesser in A). The
greater enantioselectivities found for the cobalt complexes
relative to the free aldehydes may derive from the
stronger gauche interaction between the bulky (alkynyl)-
Co2(CO)6 group of the complexed aldehyde and the alkoxy
group of the allyl unit compared to the smaller alkynyl
unit; this would cause distortion of the chair, magnifying
the stereodifferentiating interactions with the campheyl
units.
Decomplexation of the adducts 11 and 15 was ac-
complished conveniently and with no loss in enantiomeric
purity upon treatment with (NH4)2Ce(NO3)6/Et3N/acetone
(-78 to 20 °C, Table 3).11 In most cases, enantiomeric
excesses were determined by NMR analysis of the cor-
responding Mosher esters.
(21) Cha, J . K.; Kim, S. N. Chem. Rev. 1995, 95, 1761.
(22) Cha, J . K.; Christ, W. J .; Kishi, Y. Tetrahedron 1984, 40, 2247.
(23) (a) Evans, D. A.; Kaldor, S. W. J . Org. Chem. 1990, 55, 5424.
(b) Smith, D. A.; Wang, Z.; Schreiber, S. L. Tetrahedron 1990, 46, 4793.
(c) Seletsky, B. M.; Luke, G. P.; Marshall, J . A. J . Org. Chem. 1994,
59, 3413. (d) Brimacombe, J . S.; Kabir, A. K. M. S. Carbohydr. Res.
1988, 179, 21. (e) Saito, S.; Morikawa, Y.; Moriwake, T. J . Org. Chem.
1990, 55, 5424.
(24) J eong, K. S.; Sjo, P.; Sharpless, K. B. Tetrahedron Lett. 1992,
27, 1992.
(25) Cha, D. Y.; Kelly, R. C.; Van Rheenen, V. Tetrahedron Lett.
1976, 23, 1973. Sharpless, K. B.; Schroder, G.; Mungall, W. S.; Marko,
I.; J acobsen, E. R. J . Am. Chem. Soc. 1988, 110, 1968.
Rea ctivity Stu d ies of 3,4-Dioxy 1,3-En yn es. Di-
h yd r oxyla tion . Osmium-promoted dihydroxylation has

Reactions of Co-Complexed Acetylenic Aldehydes
J . Org. Chem., Vol. 62, No. 6, 1997 1741
Sch em e 4
dioxy enynes. The following epoxidizing systems were
tested: (1) m-chloroperbenzoic acid (m-CPBA);31 (2) VO-
(acac)2/t-BuOOH;32 (3) magnesium monoperphthalate;33
and (4) Mo(CO)6/tert-butyl hydroperoxide.34 m-CPBA was
found to be ineffective for epoxidizing 17d between 25
and 80 °C in dichloromethane. Similarly, magnesium
monoperphthalate also failed to produce any desired
product from substrate 17d as indicated by TLC. Reac-
tion of 18a with Mo(CO)6/t-BuOOH at 25-60 °C led to
polymerized material.
tert-Butyl hydroperoxide/vanadyl acetylacetonate did
successfully epoxidize 3,4-alkoxy en-yne 17d ; however,
the conversion was less than 20% at room temperature
after 48 h; increasing the temperature to 70 °C produced
no significant improvement, instead causing extensive
portantly, the isomers 21a ,b were separable by prepara-
tive TLC and obtained as red oils. The stereochemistry
assigned to the major product 19a was based on the
vicinal coupling constants between the hydrogens on the
MOM-bearing carbon (C3) and on the adjacent acetate-
bearing carbon (C2) of 21a ,b, these being 4.0 Hz (for
major isomer 21a ) and 3.0 Hz (minor isomer 21b),
respectively. NMR data for a number of reported related
compounds26,27 reveal a higher coupling constant for the
anti diastereomers relative to the syn derivatives. PC
Model MMX calculations on 20a ,b also support this
assignment, the major conformation of syn 20b being
predicted to have a coupling constant of 2 Hz, that for
anti 20a calculated to be 7 Hz.
1
decomposition. The H NMR spectrum of the resulting
epoxide 24 suggested formation of a single diastereomer,
but complete characterization was deferred in search of
increased reactivity. To test whether the relative ster-
eochemistry of the hydroxy and alkoxy groups was a
major factor in determining the reactivity toward epoxi-
dation, anti-3,4-dioxy enyne 18a was subjected to typical
epoxidation conditions with VO(acac)2/tert-butyl hydro-
peroxide, but only starting material was recovered.
The observed preferred dihydroxylation anti to the
preexisting allylic alkoxy group is consistent with the
empirical rule formulated by Kishi for osmylation of
allylic-substituted substrates.22,23 This transition-state
model, illustrated for substrate 17c in structure C, is
reactant like, closely resembling the ground-state con-
formation of allylic ethers A, which minimizes allylic
strain. The presence of the syn homoallylic hydroxyl
group apparently does not override this preference.
Since monosubstituted double bonds are relatively
unreactive toward metal-catalyzed epoxidation, we sought
to increase the reactivity of the substrate by unmasking
the allylic hydroxy group.35 The MOM derivative 17c
was therefore deprotected using trimethylsilyl bromide
at room temperature, providing diol 25, albeit in low yield
(25%).36 After considerable experimentation, deprotec-
tion of 17c by HCl/H2O/THF37 was found to be superior
(90%, eq 7). Diol 25 successfully underwent epoxidation
Ep oxid a tion . Epoxidation of allylic alcohols is a
generally useful approach to stereocontrolled generation
of multiple stereocenters.28 Although the peracid epoxi-
dation of cyclic allylic alcohols is well-known to give high
syn stereoselectivity, the selectivity for acyclic allylic
substrates is much more variable.29 Similarly, metal-
catalyzed epoxidation with peroxides generally exhibit
high syn stereoselectivity with cyclic allylic alcohols but
the selectivity for acyclic substrates is more complex,
depending considerably on both the catalyst (e.g., Mo vs
V) and the substrate.30 Generally, moderate to high anti
selectivity relative to an allylic or syn relative to a
homoallylic hydroxyl group has been observed. However,
few substrates having both allylic and homoallylic OH-
(R) groups as in the dioxy enynes 17 and 18 have been
epoxidized. Accordingly, we examined the reactivity of
various epoxidizing reagents with representative 3,4-
by VO(acac)2-t-BuOOH (eq 8) at 0-20 °C over several
hours; chromatography afforded a 7:3 mixture of diaster-
eomeric epoxides 26 (45% yield) as indicated by NMR;
reaction at 0 °C for 48 h enhanced the stereoselectivity
to g90% de. Thus, the diol 25, in contrast to the MOM-
protected adduct 17c, undergoes epoxidation with high
1
conversion and acceptable rate. Although the H NMR
(31) Eustache, J .; Bernardon, M.; Shroot, B. Tetrahedron lett. 1987,
28, 4861.
(26) Lewinski, K.; Beaudin, S.; Marshall, J . A. J . Org. Chem. 1993,
58, 5876.
(32) Mihelich, E. D.; Daniel, K.; Eickhoff, D. J . J . Am. Chem. Soc.
1981, 103, 7690.
(33) Brougham, P.; Looper, M. S.; Cummerson, D. A.; Heaney, H.;
Thompson, N. Synthesis 1987, 1015.
(34) Duke, R. K.; Richards, R. W. J . Org. Chem. 1984, 49, 1898.
(35) Kanemoto, S.; Nonaka, K.; Oshima, K.; Utimoto, K. Tetrahedron
Lett. 1986, 27, 3387.
(36) Hanessian, S.; Delorme, D.; Dufresne, Y. Tetrahedron Lett.
1984, 25, 2515.
(27) Schmidt, R. R.; Kufner, U. Liebigs Ann. Chem. 1986, 1610.
(28) Rao, A. S. In Comprehensive Organic Synthesis; Trost, B. M.,
Ed.; 1993; Vol. 7, Chapter 3.1, pp 357-386. J ohnson, R. A.; Sharpless,
K. B. Ibid. Chapter 3.2, pp 390-436.
(29) Tomioka, H.; Suzuki, T.; Oshima, K.; Nozaki, H. Tetrahedron
Lett. 1982, 23, 3387.
(30) Hoveyda, A.; Evans, D. Chem. Rev. 1993, 93, 1310 and refer-
ences therein.
(37) Panek, J .; Cirillo, C. J . Org. Chem. 1994, 59, 3057.

1742 J . Org. Chem., Vol. 62, No. 6, 1997
Ganesh and Nicholas
Sch em e 5
F igu r e 1. X-ray structure of 27a .
and mass spectra of 26 supported its formulation as a
diol epoxide(s), spectral overlap and our inability to
separate the isomers prompted us to derivatize it.
For the purpose of stereochemical determination, we
sought a crystalline derivative of 26 by the following
methods: acetylation (Ac2O, pyridine);38 formation of an
acetonide (2,2-dimethoxypropane/p-toluenesulfonate);39
conversion to a tert-butyldimethylsilyl derivative;40 and
complexation with Co2(CO)8. The first three approaches
resulted in decomposition of the sensitive epoxide but the
last proved successful. Since dicobalt octacarbonyl is
known to deoxygenate epoxides at room temperature,41
the crude epoxide 26 was treated with an equimolar
quantity of dicobalt octacarbonyl at 0 °C (eq 9). Chro-
matography on neutral alumina gave the cobalt complex
27a ,b, but separation of the isomers by regular silica,
alumina, or reversed-phase TLC methods failed. The 1H
NMR spectrum of the major epoxide showed a doublet
of doublets at 5.0 ppm (J ) 7.5, 4.9 Hz) for the propargylic
protons, an apparent triplet at 3.55 ppm (J ) 3.4, 3.3
Hz) for the homopropargylic protons, and a multiplet at
3.34 ppm for the methylenic protons of the oxirane, but
an unambiguous stereochemical assignment was not
possible on the basis of the NMR spectrum. Finally,
fractional crystallization of 27a ,b by pentane vapor
diffusion into a CH2Cl2 solution afforded an X-ray quality
crystal of the major epoxide isomer 27a . Examination
of the X-ray structure of 27a (Figure 1) led to two
important conclusions: (1) epoxidation had occurred syn
to the allylic hydroxy group; and, remarkably, (2) the two
hydroxyl groups were disposed anti to each other, ap-
parently having inverted from the original adduct 25.
leading to 27 along with the yields are shown in Scheme
5; stereochemistry in question is indicated by wiggly
lines. The syn stereochemistry at C3,4 assigned to 11c
was supported by the 3.2 ppm chemical shift of the
hydroxyl proton of 11c in the region characteristic of the
syn-adducts10 and by the X-ray proven structure of the
corresponding methoxy derivative 11f. To establish that
decomplexation of 11c by ceric ammonium nitrate did
not cause epimerization, the alkyne 17c was recomplexed
with Co2(CO)8 to produce complex 11c having an Rf value
1
and H NMR spectrum identical to those of the original
complex.
In order to rule out the possibility that epimerization
occurred during the MOM deprotection step, the stereo-
chemistry of diol 25 was established by its conversion to
the corresponding acetonide 28 (90%) using dimethoxy-
propane/pyridinium p-toluenesulfonate (PPTs, eq 11).
1
The H NMR spectrum of 28 indicated that it had the
trans geometry, as was expected being derived from the
syn diol. This assignment is based on the nearly coin-
cident methyl resonances of the acetonide (∆δ ) 0.015
ppm), indicative of approximate C2 symmetry. Ample
literature data42,43 show a much smaller chemical shift
difference for the methyl groups of trans acetonides (ca.
In order to investigate this unexpected stereochemical
result, we probed in detail the stereochemistry of each
of the steps leading to 27a . The sequence of reactions
(38) Reactions and Synthesis in the Organic Chemistry Lab; Tietze
L. F., Eicher, Th., Eds.; University Science Books: Mill Valley, CA,
1988.
(39) Mori, K.; Maemoto, S. Liebigs Ann. Chem. 1987, 863.
(40) Gao, Y.; Hanson, R. M.; Klunder, J . M.; Sharpless K. B. J . Am.
Chem. Soc. 1987, 109, 5765.
(42) Hoffmann, R. W.; Ladner, W. Chem. Ber. 1983, 116, 1631.
(43) Miyaura, N.; Suzuki, A.; Moriya, T. Tetrahedron Lett. 1995, 36,
1889.
(41) Dowd, P.; Kang, K. J . Chem. Soc., Chem. Commun. 1974, 385.

Reactions of Co-Complexed Acetylenic Aldehydes
J . Org. Chem., Vol. 62, No. 6, 1997 1743
0.01 ppm) compared to those of cis acetonides (ca. 0.17
ppm). Additionally, since the protons of the stereocenters
of 28 had very similar chemical shifts, the acetonide was
complexed with dicobalt octacarbonyl (90%) in order to
resolve the stereogenic proton resonances and to facilitate
NOE and NOESY experiments. The trans structure
assigned to 29 (and hence to 28) was thus supported by
difference NOE and NOESY experiments with weak
NOE effects (less than 1%) being observed for the two
stereogenic ring protons. Hence, epimerization at C3/
C4 does not occur during removal of the MOM group.
Finally, to assess the possibility that epimerization of
the diol 25 occurred in the epoxidation reaction itself, a
control experiment was performed by stirring 25 with
VO(acac)2 and 4 Å molecular sieves under typical epoxi-
dation conditions. Indeed, preparative TLC afforded a
diastereomeric mixture of diols 25a and 25b (eq 12),
first time, likewise react moderately efficiently and highly
stereoselectively with the complexed aldehydes producing
the corresponding anti-3,4-dioxy enyne adducts. De-
metalation of these complexes provides access to all four
stereoisomeric families of 3,4-dioxy 1,5-enynes, attractive
building blocks for the synthesis of target molecules
possessing multiple adjacent stereocenters. Dihydroxy-
lation of a representative syn-dioxy enyne occurs selec-
tively at the carbon-carbon double bond and preferen-
tially (9:1) anti to the allylic alkoxy group. When the
highly enantio- and diastereoselective addition of γ-alkoxy-
allyl boranes to cobalt-complexed acetylenic aldehydes is
combined with the stereoselective dihydroxylation of the
resulting dioxy enynes, a highly enantioselective entry
to molecules with four adjacent tetraoxygenated centers
is provided, of considerable potential in carbohydrate
synthesis. Epoxidation of a representative syn-dioxy
enyne also occurs selectively but is complicated by
epimerization of the starting material, affording an anti-
diol epoxide, with epoxidation occurring syn- to the allylic
hydroxyl group. In conclusion, we note the key roles
played by the -Co2(CO)6 moiety in these transformations.
First, it serves a protective function for the triple bond,
blocking potentially competing addition/polymerization
reactions. Second, there is a considerable stereocontrol-
ling effect of complexation, presumably reflecting the
greatly increased steric demand of the bulky, bent
-(alkynyl)Co2(CO)6 moiety relative to the linear, “skinny”
alkyne. Efforts to expand the scope of this methodology
and to demonstrate the synthetic potential of the derived
3,4-dioxy 1-en-5-ynes in natural products synthesis are
underway.
1
apparent in both the H and 13C NMR spectra of 25a ,b.
The doublet at 4.60 and apparent triplet at 4.32 ppm in
the original spectrum of 25 now appeared as multiplets
at 4.60 ppm and at 4.32 ppm; two sets of stereogenic 13
C
NMR resonances were also observed. Thus, the diol
undergoes epimerization under the epoxidation condi-
tions. We have found no previous reports of OV(acac)2-
promoted epimerization of allylic or propargylic alcohols,
though few epoxidations have been carried out on poly-
oxygenated substrates with less reactive monosubstituted
double bonds.44,45 VO(acac)2 presumably functions as a
Lewis acid, promoting reversible ionization of an OH
group (via chelation?). It is not a simple acid-catalyzed
epimerization since deprotection of 17d under aqueous
acidic conditions did not cause epimerization.
If epimerization procedes epoxidation, selective forma-
tion of epoxide 26 from the anti diol is unexpected given
the general preference for an allylic OH to direct anti
epoxidation and a homoallylic OH to favor syn addition.
However, as noted earlier, very few diols that are both
allylic and homoallylic have been epoxidized.46 Because
of the uncertainty introduced by epimerization, it is
premature to speculate about a detailed transition state
to account for the observed selectivity. Additional studies
that eliminate this complication and that explore a wider
variety of substrates are needed.
Exp er im en ta l Section
Gen er a l Meth od s/Ma ter ia ls. Deuterated solvents were
dried over 4 Å molecular sieves and stored and handled under
N2. 1H NMR spectra were obtained at 300 MHz using CDCl3
or C6D6 as solvent; 13C NMR spectra were measured at 75.4
MHz. Preparative TLC was carried out on silica gel E. Merck
(G-60PF254-366) with 20 × 20 cm glass plates (1 mm). Analyti-
cal TLC was performed on silica gel IB-F plates. Flash
chromatography was conducted by the method of Still using
E. Merck silica gel (230-400 mesh) under 20 psi of N2.46
Organolithium reagents were titrated before use.47 Melting
points (uncorrected) of solid cobalt complexes were determined
in sealed capillaries. All reactions were conducted in oven-
or flame-dried glassware under an atmosphere of nitrogen.
Transfers were made by syringe or cannula. Methylene
chloride, dichloroethane, pyridine, and triethylamine were
distilled from CaH2; benzene, toluene, THF, and diethyl ether
were distilled from sodium benzophenone ketyl; acetic anhy-
dride was distilled from P2O5; ethanolamine was azeotropically
distilled using benzene. Elemental analyses were performed
by Midwest Microlabs, Indianapolis, IN. Those new com-
pounds that were not characterized by elemental analysis
(typically oils) were judged to be g95% pure by 1H and 13C
NMR analysis. X-ray diffraction was carried out by Dr. M.
Khan, the O.U. staff crystallographer.
Con clu sion s/Su m m a r y
Allyl methyl ether,48 methoxymethyl allyl ether,49 methoxy-
allene,50 R-methoxy-3,3,3-trifluoropropanoyl chloride,51 (acety-
lenic acetal)Co2(CO)6,11 and Mosher esters of the alcohols52
were prepared using literature procedures. (E)-1-Methoxy-3-
The addition of [(Z)-alkoxyallyl]diisopinocampheyl-
boranes to cobalt-complexed acetylenic aldehydes occurs
efficiently and with excellent regio-, diastereo- (syn), and
enantioselectivity in contrast to the corresponding poorly
yielding reactions of free acetylenic aldehydes. [(E)-
Alkoxyallyl]diisopinocampheylboranes, prepared for the
(47) Kofron, W. G. Baclawski, L. M. J . Org. Chem. 1976, 41, 1879.
(48) Benedict, D. R.; Bianchi, T. A.; Cate, L. A. Synthesis 1979, 428.
(49) Gras, J . L.; Chang, K. W.; Guerin, A. Synthesis 1985, 74.
(50) Hoff, S.; Brandsma, L.; Arens, J . F. Recl. Trav. Chim. Pays-
Bas 1968, 87, 916.
(51) Dale, T. A.; Mosher, H. S. J . Am. Chem. Soc. 1973, 95, 512.
(52) (a) Dale, T. A.; Dull, L. D.; Mosher, H. S. J . Org. Chem. 1969,
34, 2543. (b) Kabuto, K.; Yasuhara, F.; Yamaguchi, S. Tetrahedron
Lett. 1980, 21, 307.
(44) Mihelich, E. D.; Daniel, K.; Eickhoff, D. J . J . Am. Chem. Soc.
1981, 103, 7690.
(45) Roush, W. R.; Michaelides, M. R. Tetrahedron Lett. 1986, 27,
3353.
(46) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923.

1744 J . Org. Chem., Vol. 62, No. 6, 1997
Ganesh and Nicholas
(phenylthio)propene was obtained using a modification of
Hoffmann’s method by conducting the reaction at -30 °C.10
(3R*,4R*)-4-Meth oxy-1-p h en yl-5-h exen yn -3-ol (9). To
a flame-dried sidearm flask equipped with a stir bar was added
0.40 mL (3.8 mmol) of allyl methyl ether using a syringe
followed by 50 mL of freshly distilled THF. The solution was
cooled to -78 °C, 3.4 mL (3.1 mmol) of sec-butyllithium was
added slowly through a syringe, and the resulting yellow
solution was stirred at -78 °C for 15 min to 30 min. In a
flame-dried flask 0.970 g (3.1 mmol) of (-)-B-methoxydiisopi-
nocampheylborane was weighed inside the drybox. Freshly
distilled THF (15 mL) was added to the (-)-B-methoxydiiso-
pinocampheylborane, and the solution was cooled to -78 °C
before canulation into the solution of lithiated allyl methyl
ether. The yellow color was immediately discharged, and the
solution was stirred at -78 °C for 1 h. To the solution of the
ate complex was added 0.50 mL (4.1 mmol) of BF3‚Et2O to form
[(Z)-γ-methoxyallyl]diisopinocampheylborane (8). Meanwhile,
0.400 g (3.07 mmol) of phenylpropargyl aldehyde dissolved in
20 mL of THF at -78 °C was slowly added to the solution of
8 and stirred at -78 °C for 3 h and then 12 h at room
temperature. The solution was concentrated under vacuum,
and the residue was dissolved in 150 mL of freshly distilled
diethyl ether and then filtered into a flame-dried side-arm
flask with a magnetic stirrer using a Teflon tube. The solution
was cooled to 0 °C, and 0.19 mL (3.1 mmol) of distilled
ethanolamine was added. The mixture was stirred at 0 °C
for 1 h and at 20 °C for 24 h and then filtered. The product
was subjected to Kugelrohr distillation (bp 100 °C; 5 mm) and
then purified by flash column chromatography on silica (3:1
hexane/ethyl acetate) to yield 50 mg (0.25 mmol) of 9 (8%):
1H NMR (CDCl3, 300 MHz) δ 7.49 (m, 2H), 7.29 (m, 3H), 5.83
(m, 1H), 5.46 (dd, J ) 17.6, 8.14 Hz, 1H), 4.51 (d, J ) 7.1 Hz,
1H), 3.77 (pseudo triplet, J ) 7.5, 7.2 Hz, 1H), 3.41 (s, 3H);
13C NMR (CDCl3, 75 MHz) δ 134.3, 132.2, 132.2, 128.9, 128.8,
128.6, 86.2, 86.0, 66.0, 57.5; MS (DIP, 70 eV) m/e 171.0 (M -
31, 3), 131.0 (M - 71, 43), 115 (M - 87, 12), 103 (M - 99, 27),
77 (M - 125, 48), 71 (M - 131, 100).
chromatography (95:5 petroleum ether/ether) to obtain 9.60 g
(27.2 mmol) of the aldehyde complex 10b (84%).
10a : 1H NMR (CDCl3) δ 10.5 (s, 1H), 7.35 (m, 2H), 7.25 (m,
3H); IR (CH2Cl2) 3060, 3980, 2980, 2880, 2090, 2050, 1660,
1480, 1060, 1010 cm-1
.
10b: 1H NMR (CDCl3) δ 10.48 (s, 1H), 2.68 (s, 3H); IR (CH2-
Cl2) 2980, 2930, 2900, 2860, 2010, 1670, 1310, 1010 cm-1
10c: 1H NMR (CDCl3) δ 10.5 (s, 1H), 5.99 (s, 1H); IR (CH2-
Cl2) 2980, 2900, 2860, 2090, 2040, 2010, 1690, 1100 cm-1
.
.
[(3R,4R)- an d (3S,4S)-4-Alkoxy-1-su bstitu ted-5-h exen yn -
3-ol]d icoba lt Hexa ca r bon yl Com p lexes 11a -e. The fol-
lowing procedure is representative: In a flame-dried side-arm
flask equipped with a stirring bar was added 0.49 mL (5.2
mmol) of allyl methyl ether followed by 50 mL of freshly
distilled THF. The solution was cooled to -78 °C, and 3.2 mL
(1.3M in hexane, 4.2 mmol) of sec-butyllithium was added; the
solution immediately turned bright yellow. The solution was
stirred at -78 °C for 15-30 min. Into a flame-dried flask 1.39
g (4.15 mmol) of (-)-B-methoxydiisopinocampheylborane was
weighed inside the drybox. Freshly distilled THF (15 mL) was
added to the borane, and the solution was cooled to -78 °C
before canulation into the solution of lithiated allyl methyl
ether. The yellow color was immediately discharged, and the
solution was stirred at -78 °C for 1 h. To this mixture was
added 0.70 mL (5.5 mmol) of BF3‚Et2O to form [(Z)-γ-meth-
oxyallyl]diisopinocampheylborane (8a ). Meanwhile, 1.73 g
(4.15 mmol) of (phenylpropynal)dicobalt hexacarbonyl was
dissolved in 20 mL of THF, cooled to -78 °C, and slowly added
to the solution of 8a and stirred at -78 °C for 3 h and then 12
h at room temperature. The mixture was concentrated under
vacuum, and the residue was dissolved in 150 mL of freshly
distilled diethyl ether and then filtered into a flame-dried side-
arm flask containing a magnetic stirring bar using a Teflon
tube. The solution was cooled to 0 °C, and 0.70 mL (11 mmol)
of distilled ethanolamine was added. The solution was stirred
at 0 °C for 1 h and then at room temperature for 24 h and
was filtered under nitrogen. The product was purified by flash
column chromatography on silica (3:1 petroleum ether/diethyl
ether) to yield 1.35 g (2.77 mmol, 67%) of 11a as a dark red
oil: 1H NMR (CDCl3, 300 MHz) δ 7.56 (m, 2H), 7.30 (m, 3H),
5.9 (ddd, J ) 17.3, 8.1, 6.8 Hz, 1H), 5.38 (dd, J ) 17.3, 8.1 Hz,
2H), 4.93 (apparent t, J ) 6.4, 4.8 Hz, 1H), 3.76 (apparent t,
Deter m in a tion of th e En a n tiom er ic P u r ity of 9. (a )
By Ga s Ch r om a togr a p h y w ith a Ch ir a l Cyclod extr in
Sta tion a r y P h a se. Temperature program: Ti ) 70 °C; ti )
5 min; 5 °C per min; Tf ) 220 °C, tf ) 20 min. The retention
time of major enantiomer was 40.4 min, integration 1.3; that
of the minor was 40.6 min (0.26); ee ) 65%.
J ) 6.4, 6.8 Hz, 1H), 3.29 (s, 3H), 3.28 (d, J ) 4.8 Hz, 1H); 13
C
NMR (CDCl3, 75 MHz) δ 200, 139.9, 136.8, 131.8, 130.7, 129.7,
122.4, 108.5, 87.7, 76.9, 57.8; IR (cm-1, CHCl3) 3550-3450,
3100, 2090-2000, 1060; MS (FAB, 3-nitrobenzyl alcohol) m/e
432.15 (M+ - 2CO, 100), 404 (M+ - 3CO, 14), 376 (M+ - 4CO,
98), 348 (M+ - 5CO, 96), 320 (M+ - 6CO, 7).
(b) Mosh er Ester An a lysis. In a flame-dried side-arm
flask equipped with a stir bar was added 120.0 mg of freshly
prepared Mosher chloride. Dry methylene chloride (5 mL),
0.90 mL (10 mmol) of dry pyridine, and 8 mg of (dimethylami-
no)pyridine (DMAP) were added to the Mosher chloride. In a
dry round-bottom flask, 68.0 mg (0.340 mmol) of 60 dissolved
in 5 mL of dry CH2Cl2 was canulated into the side-arm flask
containing the Mosher chloride. The solution was stirred at
room temperature for 2 days. 3-(Dimethylamino)propylamine
(0.100 mL, 0.795 mmol) was added to the reaction product and
the mixture stirred for 15 min. The solution was then diluted
with 20 mL of CH2Cl2, washed with saturated Na2CO3 and
brine, and dried over MgSO4. The organic phase was then
concentrated in vacuo to yield the ester. The 19F NMR of the
Mosher ester of 9 showed two resonances for the CF3 groups
of the diastereomers in a 7:3 ratio at 6.38 ppm (major) and
9.10 ppm (minor).
(c) Eu (h fa ca m )3 An a lysis. In the drybox, 30 mg (0.030
mmol) of Eu(hfacam)3 was weighed into a flame-dried flask.
In an oven-dried vial, 10 mg of 9 was dissolved in 2 mL of
CDCl3. The sample was analyzed by NMR after each addition
of solid Eu(hfacam)3. After addition of nearly 15 mg, the
methoxy signals of 9 split into two distinct signals (70:15)
indicating 65% ee.
11c: dark red oil; 1H NMR (CDCl3, 300 MHz) δ 7.56 (m,
2H), 7.30 (m, 3H), 5.90 (ddd, J ) 16.5, 10.4, 6.7 Hz, 1H), 5.38
(dd, J ) 17.1, 10.2 Hz, 2H), 4.98 (apparent t, J ) 6.9, 4.0 Hz,
1H), 4.73 (d, J ) 6.6 Hz, 1H), 4.61 (d, J ) 6.6 Hz, 1H), 4.08
(apparent t, J ) 7.2, 7.4 Hz, 1H), 3.4 (d, J ) 4.0 Hz, 1H), 3.37
(s, 3H); 13C NMR (CDCl3, 75 MHz) δ 200, 136.0, 131.8, 130.6,
129.7, 96.1, 83.1, 76.6, 58.0; IR (cm-1, CHCl3) 3560-3450,
3100, 2090-2000, 1028; MS (FAB, 3-nitrobenzyl alcohol) m/e
434 (M+ - 3CO, 26), 406 (M+ - 4CO, 100), 378 (M+ - 5CO,
33), 350 (M+ - 6CO, 12).
11d : dark red solid; mp ) 52 °C; 1H NMR (CDCl3, 300 MHz)
δ 5.79 (ddd, J ) 17, 9.1, 6.7 Hz, 1H), 5.44 (dd, J ) 17.1, 9.0
Hz, 2H), 4.66 (apparent t, J ) 5.7, 5.4 Hz, 1H), 3.53 (apparent
t, J ) 6.5, 5.5 Hz, 1H), 3.11 (s, 3H), 3.10 (d, J ) 4.8 Hz, 1H),
2.62 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 200, 136.0, 130.8,
129.3, 128.3, 121.1, 86.8, 75.5, 56.5; IR (cm-1, CHCl3) 3560-
3450, 3030, 2333, 2089-2000, 1373; MS (FAB, 3-nitrobenzyl
alcohol) m/e 398 (M+ - CO, 15.7), 370 (M+ - 2CO, 100.0), 342
(M+ - 3CO, 50), 314 (M+ - 5CO, 64), 258 (M+ - 6CO, 30).
Anal. Calcd for C14H12O8Co2: C, 39.46; H, 2.84. Found: C,
39.36; H, 2.94.
11e: dark red oil; 1H NMR (CDCl3, 300 MHz) δ 5.8 (ddd, J
) 16.5, 9.0, 7.8 Hz, 1H), 5.44 (dd, J ) 16.5, 9.0 Hz, 2H), 4.93
(apparent t, J ) 8.5, 3.0 Hz, 1H), 4.75 (d, J ) 6.9 Hz, 1H),
4.62 (d, J ) 6.6 Hz, 1H), 3.92 (apparent t, J ) 8.1, 7.8 Hz,
1H), 3.37 (s, 3H), 3.18 (d, J ) 3.3 Hz, 1H), 2.61 (s, 3H); 13C
NMR (CDCl3, 75 MHz) δ 134.5, 122.6, 96.1, 94.6, 94.5, 82.3,
75.2, 56.6, 22.31; IR (cm-1, CHCl3) 3460-3436, 2950, 2089-
2012, 1366, 1018; MS (FAB, 3-nitrobenzyl alcohol) m/e 428 (M+
(Alk yn yl a ld eh yd e)d icoba lt Hexa ca r bon yl Com p lexes
10a -c. In a typical experiment, a side-arm flask equipped
with a magnetic stirring bar, 13.9 g (32.6 mmol) of (3-butynal
diethyl acetal)Co2(CO)6, 100 mL of acetone, 3 g of Amberlyst-
15 resin, and 2 mL of water was added, and the solution was
stirred at room temperature for 24 h with TLC monitoring
(silica gel, 9:1 petroleum ether/ether). The solvent was
removed by rotary evaporation, and the mixture was then
loaded on a column of silica gel and separated by flash

Reactions of Co-Complexed Acetylenic Aldehydes
J . Org. Chem., Vol. 62, No. 6, 1997 1745
- CO, 9.4), 400 (M+ - 2CO, 100), 372 (M+ - 3CO, 94), 344
(M+ - 4CO, 79), 316 (M+ - 5CO, 32), 288 (M+ - 6CO, 40).
Anal. Calcd for C15H14Co2O9: C, 39.50; H, 3.09. Found: C,
38.99; H, 3.22.
[(3R,4S)-4-Meth oxy-1-su bstitu ted -5-h exen yn -3-ol]d ico-
ba lt Hexa ca r bon yl 15a -c. A solution of potassium naph-
thalenide was prepared from 3.91 g (10.6 mmol) of potassium
and 1.28 g (10.6 mmol) of naphthalene in 40 mL of dry THF
by stirring for 1 h at room temperature. After addition of 10
mL of diethyl ether and 10 mL of pentane, the solution was
cooled to -120 °C (liquid nitrogen and pentane). First, 1.15 g
(6.36 mmol) of (E)-1-methoxy-3-(phenylthio)propene (12) and
then 2.00 g (6.32 mmol) of (-)-B-methoxydiisopinocampheyl-
borane dissolved in 10 mL of dry THF was added by a cannula.
The mixture was stirred at -120 °C for 1 h and at -78 °C for
3 h and then filtered through a sintered glass funnel contain-
ing Celite under nitrogen at -78 °C. The solution of the ate
complex was treated with 1.00 mL (8.41 mmol) of BF3‚Et2O.
Meanwhile, 1.76 g (4.24 mmol) of (phenylpropynal)Co2(CO)6
dissolved in 20 mL of THF was cooled to -78 °C and slowly
added to the solution of [(E)-γ-methoxyallyl]diisopinocampheyl-
borane (13) and stirred at -78 °C for 3 h and then for 12 h at
room temperature. The solution was concentrated under
vacuum, and the residue was dissolved in 150 mL of freshly
distilled diethyl ether and then filtered via a Teflon cannula
into a flame-dried side-arm flask containing a magnetic
stirring bar. The solution was then cooled to 0 °C, and 0.49
mL (8.00 mmol) of dry ethanolamine was added. The mixture
was stirred at 0 °C for 1 h and then at 20 °C for 24 h and then
was filtered under nitrogen. The crude product, obtained by
evaporation of the filtrate, was purified by flash column
chromatography on silica (3:1 petroleum/ether) to yield 1.00 g
(2.04 mmol, 48.1%) of 15a as a dark red oil: 1H NMR (CDCl3,
300 MHz) δ 7.56 m, 2H), 7.30 (m, 3H), 5.9 (m, 1H), 5.48 (dd,
J ) 15.8, 11.7 Hz, 2H), 4.94 (apparent t, J ) 6.4, 4.8 Hz, 1H),
3.7 (apparent t, J ) 7.8, 6.9 Hz, 1H), 3.21 (s, 3H), 2.43. (d, J
) 4.5 Hz, 1H); 13C NMR (CDCl3, 75 MHz) δ 199.6, 137.9, 135.3.
130.0, 128.8, 128.5, 127.6, 122.0, 96.8, 87.3, 73.5, 55.7; IR
(cm-1) (CHCl3) 3550-3450, 3100, 2090-2000, 1060; MS (DIP,
70 eV) m/e 432.15 (M+ - 2CO, 30), 404 (M+ - 3CO, 21), 376
(M+ - 4CO, 70), 348 (M+ - 5CO, 82.0), 320 (M+ - 6CO, 72).
11f: light red solid; mp ) 40 °C; 1H NMR (CDCl3, 300 MHz)
δ 5.93 (s, 1H), 5.65 (ddd, J ) 18, 9, 6.7 Hz, 1H), 5.25 (dd, J )
18, 9 Hz, 2H), 4.49 (apparent t, J ) 6.3, 4.3 Hz, 1H), 3.34
(apparent t, J ) 7.3, 6.8 Hz, 1H), 3.21 (s, 3H), 3.2 (d, J ) 4.21
Hz, 1H); 13C NMR (CDCl3, 75 MHz) 200.5, 134.1, 121.8, 94.7,
87.0, 75.5, 72.9, 56.7, 30.20; IR (cm-1, CHCl3) 3450, 2995,
2090-1982, 1366, 1018; MS (FAB, 3-nitrobenzyl alcohol) m/e
384 (M+ - CO, 19), 356 (M+ - 2CO, 100), 328 (M+ - 3CO,
60), 300 (M+ - 4CO, 44), 272 (M+ - 5CO, 11), 244 (M+ - 6CO,
28). Anal. Calcd for C13H10Co2O8: C, 37.89; H, 2.45. Found:
C, 37.72; H, 2.62.
(3R,4R)- an d (3S,4S)-4-Alkoxy-1-su bstitu ted-5-h exen yn -
3-ol 17a -f by Dem eta la tion of 11a -f. The following
procedure is representative: To a stirred solution of 11a (0.110
g, 0.223 mmol) in 20 mL of acetone at -78 °C was added a
solution of 0.844 g (1.54 mmol) of ceric ammonium nitrate in
10 mL of acetone (until the CO ceased). The reaction was
stirred for 2 h at -78 °C and then warmed to room temper-
ature and allowed to stir for 1 h. The mixture was poured
into 20 mL of saturated NaCl solution and extracted thor-
oughly with ether. The ether layer was washed with brine,
dried over MgSO4, and then evaporated to give a yellow oil.
The oil was subjected to column chromatography over silica
(8:2, petroleum ether/diethyl ether) to give 32.0 mg (0.158
mmol, 72%) of 17a as a yellow oil: 1H NMR (CDCl3, 300 MHz)
δ 7.43 (m, 2H), 7.33 (m, 3H), 5.82 (m 1H), 5.4 (dd, J ) 8.4,
17.9 Hz, 2H), 4.5 (d, J ) 7.0 Hz, 1H), 3.73 (t, J ) 7.3 Hz, 1H),
3.4 (s, 3H); MS (LCMS, 0.1 M ammonium acetate) m/e 220
(M+ + 18, 100), 153 (M+ - 17, 2).
17c: red oil; 1H NMR (CDCl3, 300 MHz) δ 7.43 (m, 5H),
5.90 (m, 1H), 5.44 (dd, J ) 14.1, 10.0 Hz, 2H), 4.55 (d, J ) 6.0
Hz, 1H), 4.74 (d, J ) 6.3 Hz, 1H), 4.59 (d, J ) 5.9 Hz, 1H),
4.19 (apparent triplet, J ) 6.31, 6.56 Hz, 1H), 3.43 (s, 3H),
2.80 (bs, 1H); MS (LCMS, 0.1 M ammonium acetate) m/e 250
1
15b: red oil; H NMR (CDCl3, 300 MHz) δ 5.83 (ddd, J )
13.1, 9.0, 7.5, Hz, 1H), 5.48 (dd, J ) 13.1, 9.0 Hz, 2H), 4.63
(apparent t, J ) 7.5, 4.8 Hz, 1H), 3.54 (apparent t, J ) 7.2,
7.6 Hz, 1H), 3.30 (s, 3H), 2.66 (s, 3H), 2.17 (d, J ) 5.1 Hz,
1H); 13C NMR (CDCl3, 75 MHz) δ 200, 135.63, 128.96, 121.51,
(M+ + 18, 100), 232 (M+ + 1, 1), 215 (M+ - 17, 9), 206 (M+
27, 1), 171 (M+ - 61, 5).
-
17d : yellow oil; 1H NMR (CDCl3, 300 MHz) δ 5.73 (m, 1H),
5.36 (dd, J ) 11.4, 17.6 Hz, 2H), 4.26 (d, J ) 6.0 Hz, 1H), 3.46
(at, J ) 6.5, 6.0 Hz, 1H), 3.33 (s, 3H), 1.8 (s, 3H,); MS (LCMS,
128.33, 97.69, 94.30, 87.1, 73.80, 55.874, 21.276; IR (cm-1
)
(CHCl3) 3560-3450, 3030, 2089-2000, 1373; MS (DIP, 70 eV)
m/e 398 (M+ - CO, 9), 370 (M+ - 2CO, 29), 342 (M+ - 3CO,
26), 314 (M+ - 4CO, 6), 286.0 (M+ - 5CO, 15), 258 (M+ - 6CO,
61), 226 (M+ - 6CO - OMe, 100).
0.1 M ammonium acetate) m/e 158 (M+ + 18, 100), 123 (M+
17, 10).
-
17e: yellow oil; 1H NMR (CDCl3, 300 MHz) δ 5.83 (m, 1H),
5.41 (dd, J ) 9.3, 15.4 Hz, 2H), 4.73 (d, J ) 6.7 Hz, 1H), 4.67
(d, J ) 6.7 Hz, 1H), 4.29 (m, 1H), 4.06 (apparent triplet, J )
7.2, 6.3 Hz, 1H), 3.41 (s, 1H), 1.85 (s, 3H); MS (LCMS, 0.1 M
ammonium acetate) m/e 188 (M+ + 18, 100), 153 (M+ - 17,
10).
15c: red oil; 1H NMR (CDCl3, 300 MHz) δ 5.93 (s, 1H), 5.84
(m, 1H), 5.41 (dd, J ) 10.1, 14.0 Hz, 2H), 4.94 (apparent t, J
) 5.9, 5.9 Hz, 1H), 3.63 (apparent t, J ) 7.0, 6.8 Hz, 1H), 3.26
(s, 3H), 2.41 (d, 1H, J ) 6.4 Hz); 13C NMR (CDCl3, 75 MHz)
200.5, 134.1, 121.8, 94.7, 87.0, 75.5, 72.9, 56.7, 30.2; IR (cm-1
)
17f: yellow oil; 1H NMR (CDCl3, 300 MHz) δ 5.74 (m, 1H),
5.39 (dd, J ) 10.0, 14.3 Hz, 2H), 4.29 (d, J ) 7.0 Hz, 1H), 3.65
(dd, J ) 7.1, 7.0 Hz, 1H), 3.43 (s, 3H), 2.46 (s, 1H); MS (LCMS,
(CHCl3) 3450, 2995, 2090-1982, 1366, 1018; MS (DIP, 70 eV)
m/e 356 (M+ - 2CO, 75), 328 (M+ - 3CO, 46), 300 (M+ - 4CO,
23), 272 (M+ - 5CO, 11), 244 (M+ - 6CO, 100).
0.1 M ammonium acetate) m/e 144 (M+ + 18, 100), 109 (M+
17, 10).
-
4-Meth oxy-1-p h en yl-5-h exen yn -3-ol (15a ) fr om P h en -
ylp r op yn a l (5). A solution of potassium naphthalenide was
prepared from 0.80 g (20.4 mmol) of potassium and 2.62 g (20.4
mmol) of naphthalene in 40 mL of dry THF by stirring for 1 h
at room temperature. After addition of 10 mL of diethyl ether
and 10 mL of pentane, the solution was cooled to -120 °C
(liquid nitrogen and pentane). First, 1.84 g (10.2 mmol) of (E)-
1-methoxy-3-(phenylthio)propene (12) and then 3.23 g (10.2
mmol) (-)-B-methoxydiisopinocampheylborane dissolved in 10
mL of dry THF were added by a cannula. The mixture was
stirred at -120 °C for 1 h and at -78 °C for 3 h and then
filtered through a sintered glass funnel containing Celite under
nitrogen at -78 °C. The solution of the ate complex was
treated with 1.70 mL (13.6 mmol) of BF3‚Et2O to generate [(E)-
γ-methoxyallyl]diisopinocampheylborane. Meanwhile, 1.00
mL (8.18 mmol) of phenyl propynal dissolved in 20 mL of THF
was cooled to -78 °C and slowly added to the solution of the
borane (13) and stirred at -78 °C for 3 h and then for nearly
12 h at room temperature. The mixture was concentrated
under vacuum, and the residue was dissolved in 150 mL of
freshly distilled diethyl ether and then filtered into a flame-
dried side-arm flask containing a magnetic stirring bar using
Det er m in a t ion of E n a n t iom er ic P u r it y of 17a -f.
Ch ir a l GC. The temperature program used in the separation
of the enantiomers is as follows: initial temperature ) 70 °C,
initial time ) 5 min, rate of heating ) 5 °C/min; final
temperature ) 220 °C, final time ) 20 min. The retention
time of the only detected enantiomer of 17a was 40.36 min;
ee g 99%.
Mosh er Ester An a lysis. The Mosher esters of all the
decomplexed alcohols were prepared by the method described
in the synthesis of Mosher ester of 9. 19F NMR of the Mosher
esters of 17a -f (ref CF3COOH, CDCl3, 282 MHz): 17a , 4.01
ppm; 17c, 4.02 ppm; 17d , 3.89 ppm; 17e, 3.90 ppm; 17f: 3.87
ppm.
Eu (h fa ca m )3 An a lysis. In a flame-dried flask, 30 mg (0.03
mmol) of Eu(hfacam)3 was weighed in the drybox. In a dry
vial, 10 mg of the 11a was dissolved in 3 mL of CDCl3 and
used in the analysis of 11a . The sample was analyzed after
each addition of Eu(hfacam)3. After nearly 15 mg was added,
the methoxy signals of 11a showed a single peak, indicating
that the enantiomeric excess was close to 95%.

1746 J . Org. Chem., Vol. 62, No. 6, 1997
Ganesh and Nicholas
a Teflon cannula. The solution was cooled to 0 °C, and 0.49
mL (8.00 mmol) of dry ethanolamine was added. The mixture
was stirred at 0 °C for 1 h and then at 20 °C for 24 h and then
filtered under nitrogen. The crude product (1.48 g) was
dissolved in 25 mL of dry diethyl ether and cannulated into a
flask containing 2.51 g of dicobalt octacarbonyl. The solution
was stirred at room temperature for 2 h. After rotoevaporation
the product was purified by flash chromatography on silica
(75:25 petroleum ethyl/diethyl ether) to yield 0.700 g (1.43
mmol, 19.1%) of 15a (op cit).
1H NMR (19b) (CDCl3, 300 MHz) δ 7.1-7.4 (m, 5H), 4.41 (d,
J ) 2.6 Hz, 1H), 4.50 (d, J ) 6.7 Hz, 1H), 4.31 (d, J ) 6.7 Hz,
1H), 3.83 (dd, J ) 5.3, 2.6 Hz, 1H), 3.65 (m, 2H), 3.00 (s, 3H).
Acetyla tion of 19a ,b to 20a ,b. To a side-arm flask
containing 5 mL of pyridine and 25 mL freshly distilled acetic
anhydride was added 0.200 g (0.751 mmol) of 19a ,b, and the
mixture was stirred at room temperature for 10-12 h. The
solution was evaporated, and the syrupy residue was dissolved
in 100 mL of dichloromethane and then washed with water,
saturated sodium bicarbonate, and brine solution. After
drying over MgSO4, evaporation of the organic phase gave the
triacetates 20a ,b as a yellowish oil (0.206 g, 70%). 1H NMR
analysis indicated a 9:1 mixture of isomers that could not be
separated by column or PTLC chromatography: 1H NMR
(CDCl3, 300 MHz) δ 7.45-7.25 (m, 5H), 5.81 (d, J ) 4.7 Hz,
1H), 5.34 (m, 1H), 4.94 (dd, J ) 6.9 Hz, 2H), 4.77 (dd, J ) 6.9
Hz, 1H), 4.54 (dd, J ) 12.2, 3.2 Hz, 1H), 4.27 (dd, J ) 12.0,
6.6 Hz, 1H), 4.13 (dd, J ) 5.8, 4.9 Hz, 1H), 4.13 (dd, J ) 5.8,
4.8 Hz, 1H), 3.43 (s, 3H), 2.05 (s, 3H), 2.15 (s, 6H).
(3R,4S)-4-Meth oxy-1-su bstitu ted-5-h exen -yn -3-ols 18a -c
via Dem et a la t ion of 15a -c. The following procedure is
representative: To a stirred solution of 15a (0.500 g, 1.02
mmol) in 20 mL of acetone at -78 °C was added dropwise a
solution of 1.68 g (3.06 mmol) of ceric ammonium nitrate in
10 mL of acetone (CO evolution). The reaction mixture was
stirred for 2 h at -78 °C, warmed to room temperature, and
then allowed to stir for 1 h with monitoring by TLC (silica,
3:2 petroleum ether/diethyl ether). The mixture was poured
into 20 mL of saturated NaCl solution and extracted thor-
oughly with ether. The ether layer was washed with brine,
dried over MgSO4, and then evaporated to give a yellow oil.
The oil was subjected to column chromatography (4:1, petro-
leum ether/ether) to give 0.109 g (0.539 mmol, 72%) of 18a as
a yellow oil. 1H NMR: (CDCl3, 300 MHz) δ 7.50 (m, 2H), 7.30
(m, 3H), 5.86 (m, 1H), 5.38 (dd, J ) 10.25, 10.3 Hz, 2H), 4.62
(dd, J ) 3.7, 4.0 Hz, 1H), 3.82 (dd, J ) 7.6, 3.8 Hz, 1H), 3.4 (s,
3H); MS (DIP, 70 eV) m/e 189 (M+ - 31, 5), 160 (M+ - 60, 9),
149 (M+ - 71, 54).
Coba lt Com p lex of Tr ia ceta tes 20a ,b (21a ,b). In a
flame-dried side-arm flask equipped with a magnetic stirring
bar 0.144 g (0.418 mmol) of dicobalt octacarbonyl dissolved in
50 mL of dry methylene chloride was cooled to 0 °C. Then
0.150 g (0.382 mmol) of triacetates 20a /b dissolved in cold
methylene chloride was slowly added to the solution of Co2-
(CO)8. The mixture was stirred at 0 °C for 8-10 h until
complete complexation was observed by TLC. The dark red
solution was filtered under N2 (by cannula) through a short
pad of neutral alumina, and solvent was then concentrated to
give a dark red oil. The oil was purified by preparative TLC
(4:1 petroleum ether/ethyl acetate) at 0 °C to give 0.203 g (79%)
of 21a ,b (dark red oil) as a 9:1 isomeric mixture: 1H NMR
(21a ) (CDCl3, 300 MHz) δ 7.47 (m, 2H), 7.32 (m, 3H), 6.59 (d,
J ) 6.7 Hz, 1H), 5.19 (m, 1H), 4.60 (s, 2H), 4.49 (dd, J ) 11.9,
4.2 Hz, 1H), 4.17 (dd, J ) 6.2, 11.9 Hz, 1H), 4.02 (dd, J ) 6.7,
4.0 Hz, 1H), 3.32 (s, 3H), 2.11 (s, 3H), 2.04 (s, 3H), 1.95 (s,
3H); IR (cm-1, CH2Cl2) 2080, 2020, 2005, 1985 (s), 1755 (m);
13C NMR (CDCl3, 75 MHz) δ 198.9, 170.4, 169.4, 137.2, 129.5,
128.9, 128.00, 97.6, 92.1, 93.2, 79.5, 72.6, 70.4, 62.0, 56.1, 20.7,
18b: yellow oil; 1H NMR (CDCl3, 300 MHz) δ 5.82 (m, 1H),
5.37 (dd, J ) 10.1, 9.1 Hz, 2H), 4.37 (dd, J ) 4.3, 3.9 Hz, 1H),
3.69 (t, J ) 3.6, 7.3 Hz, 1H), 3.67 (s, 3H), 1.8 (s, 3H); MS (DIP,
70 eV) m/e 158 (M+ + 18, 100), 123 (M+ - 17, 10).
18c: yellow oil; 1H NMR (CDCl3, 300 MHz) δ 5.74 (m, 1H),
5.39 (dd, J ) 10.1, 9.3 Hz, 2H), 4.29 (t, J ) 4.3 Hz, 1H), 3.65
(1H, m), 3.33 (s, 3H), 2.46 (s, 1H); MS (DIP, 70 eV) m/e 98
(M+ - 28, 12.6), 95 (M+ - 31, 33.8), 71 (M+ - 55, 18.2), 70.12
(M+ - 56, 100).
Deter m in a tion of th e En a n tiom er ic P u r ity of 18a ,b.
Mosher Ester Analysis. The Mosher esters of the decomplexed
alcohols of 18a ,b were prepared as per the method described
for 9. 19F NMR of 18a ,b (CF3COOH ref, CDCl3, 282 MHz):
18a , 3.86 ppm; 18b, 3.85 ppm.
Eu (h fa ca m )3 An a lysis. In a flame-dried flask 30 mg (0.03
mmol) of Eu(hfacam)3 was weighed in the drybox and 300 mL
of CDCl3 added to prepare a 0.05 M solution. In a dry vial 10
mg of 15a -c was dissolved in 2 mL of CDCl3 and used in the
analysis of 15a -c. The sample was analyzed by NMR after
each 0.1 mL addition of the Eu(hfacam)3 solution. After nearly
40 µL addition, the methoxy signals of 15a -c showed only a
single peak, indicating that the enantiomeric excess was at
least 95%.
1
92.1, 93.2; H NMR (21b) (CDCl3, 300 MHz) δ 7.35-7.4 (m,
5H), 6.48 (d, J ) 5.9 Hz, 1H), 5.34 (m, 1H), 4.65 (dd, J ) 6.9,
6 Hz, 2H), 4.37 (dd, J ) 12.0, 6.6 Hz, 1H), 4.12 (dd, J ) 11.5,
6.0 Hz, 1H), 4.04 (dd, J ) 5.8, 3.0 Hz, 1H), 3.38 (s, 3H), 2.0 (s,
3H), 2.05 (s, 3H), 2.10 (s, 3H).
Attem p ted Ep oxid a tion of 18a . Into a flame-dried side-
arm flask equipped with a stir bar and septum was added 20.0
mg (0.100 mmol) of 18a and 5 mL of anhydrous dichlo-
romethane. The mixture was cooled to 0 °C and then charged
with 3.0 mg of VO(acac)2 (0.010 mmol). This was followed by
slow addition of 30.0 µL of 5 M tert-butyl hydroperoxide in
cyclooctane (0.15 mmol); the solution immediately turned dark
purple. It was then stirred at 0 °C for 1 h and then at room
temperature for 48 h. TLC of the reaction mixture indicated
no epoxide formation, and the starting material was recovered
(3:2 petroleum ether/diethyl ether). The reaction mixture was
poured into 50 mL of diethyl ether and washed with saturated
sodium bisulfite. The ether extract was concentrated to give
starting enyne 18a .
Det er m in a t ion of t h e E n a n t iom er ic P u r it y of 14.
Mosh er Ester An a lysis. The Mosher esters of the decom-
plexed alcohols were prepared as per the method described
for 9: 19F NMR of Mosher ester of 14 (CDCl3, vs CF3COOH):
3.86 ppm, 6.91 ppm. (ee ) 89%).
(2R,3S,4R)- a n d (2S,3S,4R)-6-P h en yl-3-(m et h oxy-
m eth oxy)-5-h exyn e-1,2,4-tr iol (19a ,b). In a side-arm flask
equipped with a magnetic stirring bar 0.232 g (1.98 mmol) of
N-methylmorpholine N-oxide, 5 mL of H2O, and 2 mL of
acetone were cooled to 0 °C. To this solution was added 0.230
g (0.99 mmol) of 17d with stirring. OsO4 solution (0.110 mL,
0.45 M in CH2Cl2) was then added by syringe. The reaction
was complete after stirring 15 h at 0 °C. The mixture was
then poured into 20 mL of ethyl acetate, and unreacted OsO4
was reduced by adding 1 g of sodium metabisulfite. The
mixture was diluted with 10 mL of ethyl acetate and dried
over MgSO4. It was then filtered through a pad of Celite and
concentrated to give 0.200 g of a 9:1 mixture of the triols 19a
and 19b (75%). All glassware was cleaned with a saturated
solution of sodium metabisulfite: 1H NMR (19a ) (CDCl3, 300
MHz) δ 7.35-7.4 (m, 5H), 4.91 (d, J ) 6.7 Hz, 1H), 4.80 (d, J
) 6.7 Hz, 1H), 4.84 (d, J ) 3.6 Hz, 1H), 4.02 (m, 1H), 3.8 (m,
3H), 3.45 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 131.7, 128.7,
128.4, 122.2, 98.3, 87.3, 86.2, 81.2, 71.4, 63.4, 56.4; MS (DIP,
70 eV) m/e 267 (M+ + 1, 1) 221 (M+ - 45, 6), 175 (M+ - 91, 9);
(3S,4R)-1,2-Ep oxy-3-m eth oxy-5-h ep tyn -4-ol (24). Into
a flame-dried side-arm flask equipped with a stir bar and
septum was added 70.0 mg (0.520 mmol) of 17d and 5 mL of
anhydrous dichloromethane. The mixture was cooled to 0 °C
and then charged with 7.0 mg of VO(acac)2 (0.26 mmol). This
was followed by slow addition of 0.15 mL of 5 M tert-butyl
hydroperoxide in cyclooctane (0.83 mmol). The solution im-
mediately turned dark purple. It was then stirred at 0 °C for
1 h and then at room temperature for 48 h. The mixture was
poured into 50 mL of diethyl ether and washed with saturated
aqueous sodium bisulfite solution. The ether extract was
concentrated to give a yellow oil that was chromatographed
on a preparative silica plate with petroleum ether/ether eluant
(4:1), giving 30 mg (0.21 mmol, 40%) of epoxide 24 (Rf
)
0.32): 1H NMR (CDCl3, 300 MHz) δ 4.37 (m, 1H), 3.53 (m,
1H), 3.34 (m, 1H), 3.51 (m, 3H), 3.21 (m, 1H), 2.8 (dd, J ) 3.6,
2.4 Hz, 1H), 2.5 (d, J ) 6 Hz, 1H), 2.07 (dd, J ) 11.4, 2.4 Hz,
1H), 1.86 (s, 3H); IR 3600, 1250 cm-1; MS (DIP, 70 eV) m/e
155 (M+ - 1, 6) 125 (M+ - 31, 32), 87 (M+ - 69, 100).

Reactions of Co-Complexed Acetylenic Aldehydes
J . Org. Chem., Vol. 62, No. 6, 1997 1747
(3R,4R)-1-Hexen -5-yn e-3,4-d iol (25). By Me3SiBr . To
a cooled (-40 °C) solution of 2.00 g (8.61 mmol) of 17c in 10
mL of dry dichloromethane containing 4 Å molecular sieves
was added slowly 4.50 mL (34.0 mmol) of trimethylsilyl
bromide. The solution was stirred for 8-10 h at -30 °C until
disappearance of the starting material was complete as
indicated by TLC. The reaction mixture was poured into a
solution of saturated sodium bicarbonate, extracted with ether,
dried over anhydrous magnesium sulfate, and evaporated, and
the residue was chromatographed over flash silica with
petroleum ether/water (2/1) eluant to give 1.61 g (3.30 mmol)
of 25 (35% yield): 1H NMR (CDCl3, 300 MHz) δ 7.45 (m, 2H),
7.25 (m, 3H), 6.02 (ddd, J ) 17.1, 10.7, 5.8 Hz, 1H), 5.49 (d, J
) 17.1 Hz, 1H), 5.33 (d, J ) 10.5 Hz, 1H), 4.8 (pseudo triplet,
of crude diol 25 were added 2,2-dimethoxypropane (0.31 mL,
2.5 mmol) in dry methylene chloride (10 mL) and 6.2 mg (2.4
mmol) of pyridinium p-toluene sulfonate. The mixture was
stirred for 9 h at 20 °C until the complete disappearance of
starting material was indicated by TLC. The mixture was
then diluted with 20 mL of diethyl ether, washed with
saturated sodium bicarbonate solution and brine, dried with
anhydrous magnesium sulfate, and concentrated. The residue
was chromatographed over silica gel (9:1 petroleum ether/
diethyl ether) producing 0.12 g (90%) of 28: 1H NMR (C6D6,
300 MHz) 7.45 (m, 2H), 7.0 (m, 3H), 5.83 (m, 1H), 5.43 (d, J )
17.1, Hz, 1H), 5.09 (d, J ) 10.5 Hz, 1H), 4.68 (pseudo triplet,
J ) 7.8, 6.3 Hz, 1H), 4.61 (d, J ) 7.8 Hz, 1H), 1.53 (s, 3H),
1.48 (s, 3H); MS (DIP, 70 eV) m/e 213 (M+ - 15, 9), 182 (M+
46, 2), 172 (M+ - 56, 47).
-
J ) 5.6, 6.7 Hz, 1H), 4.4 (d, J ) 6.6 Hz, 1H), 2.6 (bs, 2H,); 13
C
NMR (CDCl3, 75 MHz) δ 135.5, 131.8, 128.7, 128.3, 122.1,
118.0, 86.8, 86.6, 75.9, 66.6; MS (DIP, 70 eV) m/e intensity
170 (M+ - 18, 8), 154 (M+ - 34, 8), 131 (M+ - 57, 76), 102
(M+ - 86, 100).
[(4S,5R)-2,2-Dim eth yl-4-eth en yl-5-(p h en yleth yn yl)-1,3-
d ioxola n e]d icoba lt Hexa ca r bon yl (29). In a flame-dried
side-arm flask containing a magnetic stirring bar 82.5 mg
(0.241 mmol) of Co2(CO)8 was dissolved in 50 mL of dry
methylene chloride under nitrogen. Then 0.050 g (0.219 mmol)
of 28 dissolved in methylene chloride was added. The solution
was stirred at room temperature for 2-3 h until complete
complexation was indicated by TLC. The dark red solution
was filtered under N2 (by cannula) through a short pad of
neutral alumina, and the solvent was concentrated to give
0.101 g of dark red oily 29 (90%): 1H NMR (C6D6, 300 MHz)
δ 7.6 (m, 2H), 7.4 (m, 3H), 5.9 (ddd, J ) 17.3, 10.2, 7.3 Hz,
1H), 5.5 (d, J ) 17.1 Hz, 1H), 5.35 (d, J ) 10.2 Hz, 1H), 5.1 (d,
J ) 7.8 Hz, 1H), 4.3 (pseudo triplet, J ) 7.7, 7.7 Hz, 1H), 1.49
(s, 3H), 1.51 (s, 3H).
By HCl/H2O. Into a side-arm flask equipped with a stir
bar and septum was added 0.130 g (0.56 mmol) of 17c
dissolved in 1.2 mL of THF:H2O:HCl (concd) solution (8:1:1).
The solution was stirred at room temperature for 9 h until
the starting material completely disappeared as indicated by
TLC. The reaction mixture was then poured into a solution
of saturated sodium bicarbonate, extracted with ether, and
dried over anhydrous magnesium sulfate, the solvent was
evaporated, and the residue was chromatographed over flash
silica (2:1 petroleum ether/ether) to afford 0.100 g (0.53 mmol)
of 25 (94%).
(2R*,3S*,4S*)-1,2-Epoxy-6-ph en yl-5-h exyn e-3,4-diol (26).
To a flame-dried, jacketed, side-arm flask containing a stir bar
were added 0.520 g (3.03 mmol) of 25, 5 mL of dry dichlo-
romethane, and 500 mg of predried 4 Å molecular sieves. The
mixture was cooled to 0 °C, and 80.0 mg (0.304 mmol) of VO-
(acac)2 was added followed by 0.85 mL of 5.5 M t-BuOOH
solution in isooctane. The mixture was stirred for 48 h at 0
°C until complete disappearance of the starting material was
indicated by TLC (1:1 petroleum ether/diethyl ether). The
reaction mixture was then quenched with Na2S2O3, and MgSO4
was added. The mixture was filtered, the solvent rotary
evaporated, and the residue was chromatographed (2:3 petro-
leum ether/ether, silica gel) to give 26 as a colorless oil (0.400g,
65%, 9:1 isomeric ratio): 1H NMR (CDCl3, 300 MHz) δ 7.35-
7.4 (m, 5H), 4.75 (d, J ) 3.9 Hz, 1H), 3.39 (apparent triplet, J
) 3.6, 3.3 Hz, 1H), 3.34 (m, 1H), 2.8 (d, J ) 3.5 Hz, 1H); 13C
NMR (CDCl3, 75 MHz) δ 131.8, 129.0, 87.5, 85.7, 72.2, 65.9,
51.4, 43.6, 29.7; IR 3600, 1250 cm-1; MS (DIP, 70 eV) 187 (M+
- OH, 7), 170 (M+ - 2OH, 59), 131 (M+ - OH (CHOCH2),
100).
VO(a ca c)2-In d u ced Ep im er iza tion of 25. To a flame-
dried flask with a jacket and side arm equipped with a stir
bar were added 0.020 g (0.160 mmol) of 25 and 5 mL of
anhydrous dichloromethane and 50 mg of predried molecular
sieves. The solution was cooled to 0 °C, and 42.0 mg (0.016
mmol) of VO(acac)2 was added to reaction mixture. The
mixture was stirred for 48 h at 0-5 °C. The solution was then
filtered and concentrated, and the residue was separated on
preparative TLC giving 10 mg (0.080 mmoles) of 25a ,b: 1H
NMR (CDCl3, 300 MHz) δ 7.42 (m, 4H), 7.31 (m, 6H), 6.02.
(m, 2H), 5.5 (dd, J ) 6.73, 17.2 Hz, 2H), 5.36 (dd, J ) 10.4,
8.9 Hz, 2H) 4.62 (dd, J ) 3.6, 6.8 Hz, 1H), 4.42 (pseudo triplet,
J ) 6.0 Hz, 1H), 4.33 (m, 1H), 4.26 (m, 1H); 13C NMR (CDCl3,
75 MHz) δ 135.5, 131.8, 128.7, 128.3, 118.0, 86.8, 86.6, 75.8,
75.2, 66.6, 66.5.
X-r a y Str u ctu r e Deter m in a tion of 11f a n d 27a . X-ray
quality crystals of 11f were grown at 0 °C from pure pentane.
Crystals of 27a were obtained when pentane vapor was slowly
diffused into a solution of 27a in CH2Cl2 at -20 °C over 6-7
days. Details of the X-ray data collection and molecular and
crystal data for 11f and 27a have been deposited with the
Cambridge Crystallographic Data Centre.53
[(2R*,3S*,4S*)-1,2-Ep oxy-6-p h en yl-5-h exyn e-3,4-d iol]-
d icoba lt Hexa ca r bon yl (27). In a flame-dried side-arm flask
equipped with a magnetic stirring bar 0.670 g (1.96 mmol) of
dicobalt octacarbonyl was dissolved in 50 mL of dry methylene
chloride under nitrogen and then cooled to 0 °C. In a round-
bottom flask 0.400 g (1.96 mmol) of 26 dissolved in 5 mL of
ice cold methylene chloride was slowly added. The mixture
was stirred at 0 °C for 8-10 h until complete complexation
was observed by TLC. The dark red solution was filtered
under N2 (by cannula) through a short pad of neutral alumina,
and solvent was then rotary evaporated to give a dark red oil.
The oil was purified by preparative TLC at 0 °C to give 0.535
g (0.981 mmol, 50%) of 27 (diastereomeric ratio 9:1): 1H NMR
(CDCl3, 300 MHz) δ 7.6-7.4 (m, 5H), 5.0 (dd, J ) 7.5, 4.9 Hz,
1H), 3.62 (m, 1H), 3.55 (apparent triplet, J ) 3.4, 3.3 Hz, 1H),
3.34 (m, 1H), 2.8 (d, J ) 3.5 Hz, 2H); 13C NMR (CDCl3, 75
MHz) δ 199.2, 129.6, 128.9, 128.7, 128.0, 74.1, 73.6, 52.4, 45.1,
29.7; MS (DIP, 70 eV) m/e 434 (M+ - 2CO, 21) 406 (M+ - 3CO,
9), 378 (M+ - 4CO, 27), 350 (M+ - 5CO, 29), 322 (M+ - 6CO,
10), 304 (M+ - H2O, 35); IR (CH2Cl2) 3600, 2090, 2080, 1250
Ack n ow led gm en t. We are grateful for financial
support provided by the National Institutes of Health
(GM 34799) and the Oklahoma Center for the Advance-
ment of Science and Technology. Determination of the
X-ray structures of 11f and 27a by Dr. Masood Khan,
O.U. Analytical Services Center, is also acknowledged.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
all new compounds reported (22 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O9619387
(53) The author has deposited atomic coordinates for 11f and 27a
with the Cambridge Crystallographic Data Centre. The coordinates
can be obtained on request, from the Director, Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
cm-1
.
[(4R,5R)-2,2-Dim eth yl-4-eth en yl-5-(p h en yleth yn yl)-1,3-
d ioxola n e (28). To a stirred solution of 0.11 g (0.58 mmol)