Comparative Diastereoselectivity Analysis of Crotylindium and 3-Bromoallylindium Additions to a-Oxy Aldehydes in Aqueous and Nonaqueous Solvent Systems

Article abstract of DOI:10.1021/jo9613489

The couplings of crotyl bromide (1) and 1,3-dibromopropene (2) to a triad of conformationally unrestricted α-oxy aldehydes in water, aqueous THF (1:1), and anhydrous THF are described. In no example involving 1 was the formation of anti.syn product detected. The proportion of syn isomers reached a maximum (syn/anti = 5.6:1) when the neighboring hydroxyl group was unprotected and water was the reaction medium. Although internal chelation also operates to some degree with 2, considerable erosion of this mechanistic pathway (maximum now only 2:1) in favor of Felkin and "anti-Felkin" transition states is reflected in the product distributions. This trend can be synthetically advantageous, and a utilitarian example is demonstrated. The indium reagents studied here are notably efficient nucleophilic reaction partners in water.

Full text of DOI:10.1021/jo9613489

J . Org. Chem. 1996, 61, 8799-8804
8799
Com p a r a tive Dia ster eoselectivity An a lysis of Cr otylin d iu m a n d
3-Br om oa llylin d iu m Ad d ition s to r-Oxy Ald eh yd es in Aqu eou s a n d
Non a qu eou s Solven t System s
Leo A. Paquette* and Thomas M. Mitzel†
Evans Chemical Laboratories, The Ohio State University, Columbus, Ohio 43210
Received J uly 16, 1996X
The couplings of crotyl bromide (1) and 1,3-dibromopropene (2) to a triad of conformationally
unrestricted R-oxy aldehydes in water, aqueous THF (1:1), and anhydrous THF are described. In
no example involving 1 was the formation of anti,syn product detected. The proportion of syn
isomers reached a maximum (syn/anti ) 5.6:1) when the neighboring hydroxyl group was
unprotected and water was the reaction medium. Although internal chelation also operates to
some degree with 2, considerable erosion of this mechanistic pathway (maximum now only 2:1) in
favor of Felkin and “anti-Felkin” transition states is reflected in the product distributions. This
trend can be synthetically advantageous, and a utilitarian example is demonstrated. The indium
reagents studied here are notably efficient nucleophilic reaction partners in water.
The desirability of producing functionalized acyclic
molecules in a highly diastereoselective manner has
prompted extensive examination of the condensation of
prochiral allylmetal derivatives with aldehydes.1 Many
types of organometallic reagents, most notably those
where M ) B, Al, Cr, Si, Sn, Ti, and Zr, are now
recognized to be capable of delivering a specific stereo-
chemical outcome. The remarkable and highly utilitar-
intermediates formed under these conditions are ame-
nable to regioisomeric equilibration.5 The pre-equilibri-
um allows for E to Z conversion where relevant and
makes provision for steric control of the coupling process
when the allyl R′ or R′′ substituent is sterically bulky.
Otherwise, the reaction is γ-regioselective.
ian interdependence of metal, allylic double-bond geom-
etry and syn or anti configuration of the resultant
homoallylic alcohol has been conveniently classified into
three categories.2 To rationalize the different carbon-
carbon bond-forming stereoselectivities, each type is
characterized by a transition state structure featuring a
unique arrangement of the reacting double bonds.
All of the above transformations are effected in organic
solvent, most often under anhydrous conditions. In light
of the ability of indium to be capable of promoting
allylations in water3 and our specific interest in the levels
of diastereoselectivity attainable in aqueous media,4 we
have presently undertaken an analysis of the diastereo-
selectivitity attending the coupling of the two electroni-
cally distinctive 3-substituted allylic bromides 1 and 2
to R-oxy aldehydes 3-5 in solvents ranging from anhy-
drous tetrahydrofuran to pure water. Very recently,
Isaac and Chan have proposed that the allylindium
In the present investigation, only allyl inversion has
been seen. Chelation control again operates to an ap-
preciable level when the aldehyde carries a free hydroxyl
substituent as in 4. A noteworthy feature associated with
the use of 2 is the option this dibromide offers for
reversing to a significant degree the high syn selectivity
associated with the direct condensation of 4 with allyl
bromide.4a,c
Finally, our findings are rationalized in terms of cyclic
transition states in which the aldehyde carbonyl is
coordinated via the oxygen atom to the indium of the
organometallic. The diastereoselectivity data show that
the chiral reagent/chiral substrate pairs are not particu-
larly conducive to high levels of reaction stereoselectivity
in most cases.
† Permanent address: Department of Chemistry, Trinity College,
Hartford, CT 06106.
X Abstract published in Advance ACS Abstracts, November 15, 1996.
(1) Reviews: (a) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207.
(b) Roush, W. R. In Comprehensive Organic Synthesis; Heathcock, C.,
Ed.; Pergamon Press: Oxford, 1991; Vol. 2, p 1. (c) Hoffmann, R. W.
Angew. Chem., Int. Ed. Engl. 1987, 26, 489. (d) Masamune, S.; Choy,
W.; Petersen, J . S.; Sita, L. R. Angew. Chem., Int. Ed. Engl. 1985, 24,
1. (e) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556. (f) Reetz,
M. T. Acc. Chem. Res. 1993, 26, 462.
Resu lts
When 2-(benzyloxy)propanal (3)6 was stirred with
crotyl bromide and indium powder in water at rt for 4 h,
a mixture of products resulted. Flash chromatography
on silica gel resulted in separation of a 1:1 mixture of
(2) Denmark, S. E.; Weber, E. J . Helv. Chim. Acta 1983, 66, 1655.
(3) (a) Li, C.-J . Tetrahedron 1996, 52, 5643. (b) Li, C.-J . Chem. Rev.
1993, 93, 2023. (c) Lubineau, A.; Auge´, J .; Queneau, Y. Synthesis 1994,
741.
(4) (a) Paquette, L. A.; Mitzel, T. M. Tetrahedron Lett. 1995, 36,
6863. (b) Paquette, L. A.; Lobben, P. C. J . Am. Chem. Soc. 1996, 118,
1917. (c) Paquette, L. A.; Mitzel, T. M. J . Am. Chem. Soc. 1996, 118,
1931.
(5) Isaac, M. B.; Chan, T.-H. Tetrahedron Lett. 1995, 36, 8957.
(6) Heathcock, C. H.; Young, S. D.; Hagen, J . P.; Pirrung, M. C.;
White, C. T.; VanDerveer, D. J . Org. Chem. 1980, 45, 3846.
S0022-3263(96)01348-5 CCC: $12.00 © 1996 American Chemical Society

8800 J . Org. Chem., Vol. 61, No. 25, 1996
Paquette and Mitzel
Ta ble 1. In d iu m -P r om oted Ad d ition of Cr otyl Br om id e (1) to r-Oxy Ald eh yd es a t 25 °Ca
the syn,syn (6) and syn,anti (7) alcohols from the anti,anti
diastereomer 8. None of the anti,syn product could be
assignments were made possible by direct 13C NMR
comparisons with authentic samples.7
detected by either high-field 1H or 13C NMR (Table 1,
entry 1). The stereochemical assignments are based on
direct comparison of the 13C NMR signals with those
assigned earlier by Martin and Li7 (see Experimental
Section). The 0.5:0.5:1 diastereoselectivity of this reac-
tion was observed as well in 1:1 H2O-THF and in THF,
although the reaction rate in pure organic solvent was
notably slower (entry 3).
For convenience, the product mixtures derived from
adding the crotylindium reagent to 56 were directly
hydrolyzed with p-toluenesulfonic acid in methanol and
hydrogenated to 9-11 for characterization purposes
(entries 7-9). We have previously reported that indium-
promoted allylations performed in water become increas-
ingly acidic (to a pH level of approximately 2.9) as they
proceed to completion. The same phenomenon was, of
course, observed here. Where entries 7 and 8 are
concerned, deprotection of the silyl ether can materialize,
particularly if stirring is allowed to proceed for an
extended period of time. The predescribed workup
protocol obviates any complications in determining prod-
uct ratios stemming potentially from this source.
The next series of experiments was performed with the
unprotected 2-hydroxypropanal (4).8 Admixing of this
aldehyde with crotyl bromide and indium in the same
three solvent systems led again to only three diastereo-
meric products (entries 4-6). The polarity differences
of these diols on silica gel allowed for their partial
chromatographic separation. Once these fractions had
been individually subjected to catalytic hydrogenation,
it proved conveniently possible to isolate pure 11 and a
1:1 mixture of 9 and 10. The product ratios given in
In addition to the persistent absence of the anti,syn
diastereomer in entries 1-9, attention is called to the
trend reflected in the product compositions. In the
absence of a hydroxyl protecting group R to the aldehyde
carbonyl as in 4, the syn diols are favored by as much as
5.6:1 (in H2O) over the anti isomer. The loss of this acidic
proton as in 3 and 5 results in a decrease in the relative
proportion of the syn diols. In the most extreme case,
the presence of a tert-butyldimethylsilyl residue favors
anti diol formation by a factor of 3:1.
1
Table 1 were derived from H NMR integrations and from
sample weights determined at this stage. Structural
(7) Martin, S. F; Li, W. J . Org. Chem. 1989, 54, 6129. The
compilations of 13C NMR shifts provided for 6 and 8 in this paper are
incomplete. As shown herein, the syn,syn isomer exhibits a necessary
12th carbon peak at 126.5 ppm. For the anti,anti isomer, the unlisted
carbon signal is seen at 127.0 ppm.
(8) Schlosser, M.; Brich, Z. Helv. Chim. Acta 1978, 61, 1903.

Indium Additions to R-Oxy Aldehydes
J . Org. Chem., Vol. 61, No. 25, 1996 8801
Ta ble 2. In d iu m -P r om oted Ad d ition of 1,3-Dibr om op r op en e (2) to r-Oxy Ald eh yd es a t 25 °Ca
A quick scan of Table 2 will reveal that the reactions
of dibromide 2 with 3 and 5 are more anti selective than
those involving crotyl bromide. In each of the alkylations
represented by entries 10-18, it was necessary to moni-
tor reaction progress carefully, especially when complete
consumption of the aldehyde was being approached, in
order to minimize unwanted reductive debromination of
the halohydrin products.9 With this precaution in place,
each of the processes was found to deliver all four possible
diastereomers. In order to secure proper stereochemical
assignments, attention was initially directed to chemical
correlation with epoxides on the one hand and acetonides
on the other. Through analysis of the oxirane proton
coupling constants in 12, the syn/anti relationship of the
From the results compiled in Table 2, the apparent role
played by the vinylic bromine in 2 would appear to be
significant erosion of the chelation capability of the
neighboring R-hydroxyl substituent in 4. Not surpris-
ingly, the presence of an O-benzyl group as in 3 does not
give rise to any improved stereochemical bias. In con-
trast, the bulky tert-butyldimethylsilyl residue in 5
increases the anti selectivity rather steeply to a level
approaching 1:10. This finding is significant in that a
route complementary to the syn-selective direct allylation
of the R-hydroxy aldehyde is now opened.
Discu ssion
We have previously demonstrated that the stereo-
chemical course of allylindium reactions to R- and â-oxy
aldehydes in water as the reaction medium is strongly
influenced by the protecting group resident on the
neighboring oxygen.4a,c Significantly, free hydroxyl de-
rivatives react with excellent diastereofacial control at
accelerated rates to provide heightened percentages of
syn-1,2-diol and anti-1,3-diol products. These observa-
tions have given rise to the conclusion that the chelated
transition states A and B are adopted in these reactions.
neighboring bromine- and hydroxyl-substituted carbons
would be made apparent. With subsequent scrutiny of
the spectral details of 13, full analysis of the inter-
relationship of all three stereogenic centers would be
realized. Unfortunately, overlapping absorptions in the
1H NMR spectra of 13 and its diastereomers precluded
the use of these derivatives for characterization purposes.
An alternative workable solution consisted of removal
of the allylic bromine via reduction with indium in water.
This protocol operates without migration of the double
bond to deliver known diols,10 thereby permitting direct
1
comparison of H and 13C NMR features.
(9) Chen, D.-L.; Li, C.-J . Tetrahedron Lett. 1996, 37, 295.
(10) (a) Li, C.-J .; Chan, T.-H. Can. J . Chem. 1992, 70, 2726. (b)
Hoffmann, R. W.; Metternich, R.; Lanz, J . W. Liebigs Ann Chem. 1987,
881.

8802 J . Org. Chem., Vol. 61, No. 25, 1996
Paquette and Mitzel
Substitution of the hydroxyl proton by CH3, C6H5CH2,
or CH3OCH211 is predictably ameliorative of the chelation
control pathway as reflected in a modest erosion of the
allylation diastereoselectivity. Larger groups such as
tert-butyldimethylsilyl effectively deter transient binding
of the attached oxygen to the indium, at least in water,
and promote alternative conversion to product via Fel-
kin-Ahn transition states.12 Thus, steric effects appear
to exert a significant influence on the outcome of these
single asymmetric induction processes.
The focus of the present investigation was to determine
to what extent and in which direction the stereochemistry
inherent in the olefin geometry of 1 and 2 would be
transmitted to the newly formed C-C bond in the
product. Although crotyl bromide added to 4 with a
preference for adoption of the cyclic chelated transition
states C and D (syn selectivity as high as 5.6:1), the
tions, the Z form of crotyllithium predominates over the
E form to the extent of 85:15.16b The prevailing hypoth-
esis is that this Z stereoselectivity probably does not have
its origin in the particular binding features of any specific
metal but can be attributed to the intrinsic properties of
the allyl anion itself.15d If so, then ready access to
transition states C and D could be made possible by an
entirely comparable geometric isomerization of the cro-
tylindium reagent. Under these circumstances, the
distribution of syn,syn, and syn,anti diastereomers would
come under the control of the relative rates at which C
and D advance to product. In light of the product
distributions, these rates are closely comparable.
These considerations carry important implications for
allylindium reagents where E/ Z equilibration does not
operate due to structural circumstances. Under condi-
tions where double-bond configuration is not compro-
mised as in the predescribed scenario, high stereoselec-
tivitiy should be observed during coupling to aldehydes.
The stereochemical outcome of these considerations will
depend as usual on whether chelation control is operative
or not. These aspects of the problem are currently under
active investigation.
At the present time, it is unclear whether the γ-bro-
mine outcome is due to an electronic effect, to the large
steric size of this atom, to the longer C-Br bond, or to a
combination of these influences. The indication exists
that E/ Z equilibration is also facile in this instance. The
replacement of methyl by a γ-bromine is accompanied by
an unexpected dropoff in the level of syn product (now
only 2:1). Bromine substitution also has the effect of
eroding the diastereofacial selectivity of the reactions.13
Whereas the crotyl bromide experiments provided no
detectable evidence for the formation of anti,syn products,
those involving 2 are fully nonselective with generation
of approximate 1:1 mixtures in each example.
Recognition must necessarily be given to the fact that
crotyl organometallics such as the Grignard,14 potas-
sium,15 and lithium derivatives16 undergo E/ Z equilibra-
tion with considerable ease. Higher homologs behave
analogously.17 In TMEDA under equilibrating condi-
notable preference exhibited by 2 for reaction via the
transition states E and F or G is sufficiently elevated
when R′ is TBS to favor formation of the anti isomer to
the extent of 10:1! This appreciable kinetic selectivity
holds synthetic utility. Following reductive debromina-
tion and desilylation, anti diol 15 is obtained predomi-
nantly. This diastereoselection is opposite to that real-
ized upon direct condensation of 4 with allylindium in
water, a process which is highly syn-selective.
Lastly, we point out the greater efficiency of indium-
promoted crotylation and bromoallylation in water and
aqueous THF. A 10-20% enhancement in combined
product yield was universally observed. Although these
C-C bond-forming reactions happen not to be particu-
larly diastereoselective when conducted in aqueous en-
vironments, each transformation occurs with remarkably
few, if any, side reactions. The higher reaction rate in
water than in aqueous THF may be a result of the
(11) (a) Still, W. C; McDonald, J . H. Tetrahedron Lett. 1980, 21,
1031. (b) Still, W. C.; Schneider, J . A. Tetrahedron Lett. 1984, 25, 729.
(12) (a) Cherest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968,
2199. (b) Ahn, N. T.; Eisenstein, O. Nouv. J . Chim. 1977, 1, 61.
(13) Selected examples include the following: (a) Williams, D. R.;
Klingler, F. D. Tetrahedron Lett. 1987, 28, 869. (b) Roush, W. R.;
Peseckis, S. M.; Walts, A. C. J . Org. Chem. 1984, 49, 3429. (c) Roush,
W. R.; Harris, D. J .; Lesur, B. M. Tetrahedron Lett. 1983, 24, 2227. (d)
Wuts, P. G. M.; Bigelow, S. S. J . Org. Chem. 1983, 48, 3489. (e)
Hoffmann, R. W.; Zeiss, H.-J .; Ladner, W.; Tabche, S. Chem. Ber. 1982,
115, 2357. (f) Fujita, K.; Schlosser, M. Helv. Chim. Acta 1982, 65, 1258.
(g) Yamamoto, Y.; Yatagai, H.; Maruyama, K. J . Am. Chem. Soc. 1981,
103, 3229. (h) Yamamoto, Y.; Yatagai, H.; Maruyama, K. J . Chem. Soc.,
Chem. Commun. 1980, 1072.
(14) (a) Nordlander, J . E.; Roberts, J . D. J . Am. Chem. Soc. 1959,
81, 1769. (b) Nordlander, J . E.; Young, W. G.; Roberts, J . D. J . Am.
Chem. Soc. 1961, 83, 494. (c) Whitesides, G. M.; Nordlander, J . E.;
Roberts, J . D. J . Am. Chem. Soc. 1984, 84, 2010.
(15) (a) Schlosser, M.; Hartmann, J . Angew. Chem., Int. Ed. Engl.
1973, 12, 439. (b) Hartmann, J .; Muthukrishnan, R.; Schlosser, M.
Helv. Chim. Acta 1974, 57, 2261. (c) Schlosser, M. Angew. Chem., Int.
Ed. Engl. 1974, 13, 701. (d) Schlosser, M.; Hartmann, J . J . Am. Chem.
Soc. 1976, 98, 4674.
(17) (a) Sandel, V. R.; McKinney, S. V.; Freedman, H. H. J . Am.
Chem. Soc. 1968, 90, 495. (b) Schlosser, M. Chimia 1974, 28, 395. (c)
Schlosser, M.; Hartmann, J .; David, V. Helv. Chim. Acta 1974, 57,
2261. (d) Rauchschwalbe, G.; Schlosser, M. Helv. Chim. Acta 1975,
58, 1094.
(16) (a) West, P.; Purmont, J . I.; McKinley, S. V. J . Am. Chem. Soc.
1968, 90, 797. (b) Bates, R. B.; Beavers, W. A. J . Am. Chem. Soc. 1974,
96, 5001.

Indium Additions to R-Oxy Aldehydes
J . Org. Chem., Vol. 61, No. 25, 1996 8803
bromide (107 mg, 0.798 mmol), and indium powder (67 mg,
0.585 mmol) in water (5.9 mL) was stirred for 4.5 h. After
the customary workup and flash chromatography, a colorless
oil (118 mg, 91%) was obtained. This material was dissolved
in anhydrous methanol (20 mL) containing 5 mg of p-
toluenesulfonic acid and stirred for 6 h, at which time ethyl
acetate (20 mL) and water (20 mL) were introduced. The
separated aqueous phase was extracted with ethyl acetate (4
× 10 mL), and the combined organic layers were dried and
evaporated. Purification via flash chromatography (silica gel,
elution with 1:1 hexanes/ethyl acetate) gave two fractions in
a 1:3 ratio (61 mg, 97%). Independent hydrogenation of each
fraction gave a 1:1 mixture of 9 and 10 (15 mg) and pure 11
(46 mg). These diols were identified by means of their 13C
NMR spectra as described above.
hydrostatic pressure that materializes around the small
globules that make their appearance in purely aqueous
environments. This phase separation is not seen in 1:1
aqueous THF.
Exp er im en ta l Section 18
Ad d ition s In volvin g Cr otyl Br om id e a n d Ald eh yd e 3.
A. In H2O. To a magnetically stirred mixture of 3 (100 mg,
0.617 mmol) and water (6.7 mL) were added crotyl bromide
(123 mg, 0.914 mmol) and indium powder (77 mg, 0.670 mmol).
After 4 h, ethyl acetate (10 mL) was introduced and the layers
were separated 45 min later. The aqueous phase was ex-
tracted with ethyl acetate (3 × 10 mL), and the combined
organic solutions were dried and concentrated. Chromatog-
raphy of the residue on silica gel (elution with 20:1 hexanes/
ethyl acetate) gave a fraction containing a 1:1 mixture of two
syn diastereomers (61 mg) and a second pure anti fraction (60
mg, 89% combined yield). The components were identified on
B. In H2O-THF (1:1). From 300 mg (1.59 mmol) of 5,
321 mg (2.39 mmol) of crotyl bromide, and 201 mg (1.75 mmol)
of indium powder in 17.6 mL of 1:1 H2O-THF that had been
stirred for 4 h, chromatographed (340 mg, 88%), and hydro-
genated, there was isolated via chromatography 45 mg of a
1:1 mixture of 9 and 11 in addition to 133 mg of 11.
7
the basis of their 13C NMR chemical shifts in CDCl3 (at 75
MHz): syn,syn, 141.6, 138.4, 128.4 (2 C), 127.8 (2 C), 127.6,
126.5, 114.5, 78.3, 70.9, 40.8, 16.1, 15.0 ppm; syn,anti, 139.8,
138.4, 128.4 (2 C), 127.8 (2 C), 127.7, 115.2, 78.2, 76.5, 70.9,
40.2, 17.6, 15.3 ppm; anti,anti, 140.5, 138.4, 128.4 (2 C), 127.6
(2 C), 127.0, 115.4, 76.4, 75.7, 70.6, 39.5, 16.3, 13.7 ppm.
B. In H2O-THF (1:1). A mixture of 3 (100 mg, 0.617
mmol), crotyl bromide (123 mg, 0.914 mmol), and indium
powder (77 mg, 0.670 mmol) in 1:1 H2O-THF (6.6 mL) was
stirred for 4 h. Following product isolation in the predescribed
manner, there was isolated 60 mg of a 1:1 mixture of the syn
diastereomers together with 58 mg of the anti,anti diol (87%
combined). The products were identified on the basis of their
13C NMR chemical shifts.
C. In THF . A mixture of 3 (100 mg, 0.617 mmol), crotyl
bromide (127 mg, 0.914 mmol), and indium powder (77 mg,
0.670 mmol) in THF (6.7 mL) was stirred at rt for 12 h before
being processed in the usual way. The first chromatographic
fraction (50 mg) consisted of equal amounts of the syn,syn and
syn,anti diols. To elute subsequently was a pure fraction of
the anti,anti product (51 mg, 74% combined).
Ad d ition s In volvin g Cr otyl Br om id e a n d Ald eh yd e 4.
A. In H2O. A mixture of 4 (80 mg, 1.08 mmol), crotyl bromide
(218 mg, 1.62 mmol), and indium powder (136 mg, 1.18 mmol)
in water (11.8 mL) was stirred at rt for 3.5 h and worked up
in the predescribed manner. The mixture was subjected to
flash chromatography (silica gel, elution with 2:1 hexanes/ethyl
acetate) to give two fractions in a 5.5:1 ratio. Each fraction
was separately hydrogenated at 600 psi in ethanol over 5%
palladium on carbon. Following removal of the catalyst by
filtration and concentration in vacuo, there was isolated 95
mg of a mixture of 9 and 10 as well as 17 mg of 11 (79%
combined yield). The saturated diols were identified on the
basis of their 13C NMR chemical shifts in CDCl37 (at 75 MHz):
9, 79.9, 67.9, 36.8, 23.5, 19.9, 15.9, 11.5 ppm; 10, 78.6, 68.8,
36.3, 26.9, 19.2, 12.7, 11.7 ppm; 11, 78.2, 68.2, 36.7, 25.3, 15.3,
14.4, 10.6 ppm.
B. In H2O-THF (1:1). A mixture of 4 (80 mg, 1.08 mmol),
crotyl bromide (218 mg, 1.62 mmol), and indium powder (136
mg, 1.18 mmol) in 1:1 H2O-THF (11.8 mL) was stirred for
3.5 h before being processed in the manner just described.
After catalytic hydrogenation, there was obtained 84 mg of a
mixture of 9 and 10 together with 28 mg of 11 (78% combined).
C. In THF . A mixture of 4 (200 mg, 2.70 mmol), crotyl
bromide (546 mg, 4.05 mmol), and indium powder (341 mg,
2.97 mmol) in THF (29.7 mL) was stirred at rt for 7.5 h before
being processed in the usual way and hydrogenated. The two
chromatography fractions (5.5:1 ratio) were hydrogenated to
give 193 mg of a mixture of 9 and 10 together with 35 mg of
11 (64% combined yield).
C. In THF . A mixture of 5 (300 mg, 1.59 mmol), crotyl
bromide (321 mg, 2.39 mmol), and indium powder (201 mg,
1.75 mmol) was stirred in THF (17.5 mL) for 11 h before being
processed in the usual way to give two fractions in a 1:3 ratio
(combined weight of 272 mg or 71%). Independent hydrogena-
tion of each fraction gave 35 mg of the 9/10 mixture and pure
11 (106 mg). These diols were identified by the 13C NMR
spectra.
Ad d ition s In volvin g Dibr om id e 2 a n d Ald eh yd e 3. A.
In H2O. A mixture of 3 (135 mg, 0.834 mmol) and water (8.7
mL) was treated with 1,3-dibromopropene (334 mg, 1.668
mmol) and indium powder (101 mg, 0.876 mmol) and stirred
at rt for 2.5 h. Following the addition of ethyl acetate (15 mL)
and additional stirring (30 min), the separated aqueous layer
was extracted with ethyl acetate and the combined organic
phases were dried and evaporated. The residue was subjected
to flash chromatography on silica gel (elution with 30:1
hexanes/ethyl acetate) to give a colorless oil (209 mg, 88%).
An 80 mg (0.280 mmol) sample of this bromohydrin was
dissolved in dry THF (20 mL), treated with sodium hydride
(13 mg, 0.54 mmol), and stirred at rt for 7.5 h. After the
addition of CH2Cl2 (15 mL) and water (10 mL), the aqueous
phase was separated and extracted with CH2Cl2 (3 × 15 mL).
The combined organic layers were washed with water (2 × 20
mL), dried, and concentrated. The residue was purified by
flash chromatography on silica gel (elution with 20:1 hexanes/
ethyl acetate) to yield the stereoisomeric epoxides as a colorless
oil (76 mg, 95%).
The presence of all four possible diastereomers was readily
discerned by 1H NMR spectroscopy at 300 MHz (in CDCl3).
The two trans epoxides exhibited key signals at δ 3.16-3.12
(m, 0.11 H) and δ 3.04 (dd, J ) 4.0, 8.2 Hz, 0.2 H), while the
corresponding absorptions of the cis epoxides appeared at δ
2.97 (dd, J ) 2.2, 6.7 Hz, 0.1 H) and 2.86 (dd, J ) 2.1, 5.3 Hz,
0.2 H). On the basis of the ensuing experiment, these epoxides
are recognized to be a -d , respectively. The 13C NMR spec-
trum (75 MHz, CDCl3) of this mixture exhibited nonoverlap-
ping sets of signals at (138.4, 138.3, 138.2, 138.0), (137.6, 135.1,
134.9, 132.2), (120.6, 120.4, 119.7, 119.5), (75.5, 73.9, 71.5,
71.2), (71.1, 71.0, 70.7, 69.8), (63.3, 62.2, 61.9, 60.7), (58.0, 57.4,
54.9, 54.7), and (21.1, 18.7, 17.7, 17.4) ppm.
In a separate experiment, a sample of the bromohydrin
mixture (256 mg, 0.900 mmol) and indium powder (103 mg,
0.900 mmol) was stirred in H2O (9 mL) at rt for 8 h, at which
time 1 N HCl (10 mL) was added with stirring. Ethyl acetate
(15 mL) was next introduced, and the separated aqueous layer
was extracted with ethyl acetate (3 × 15 mL). The combined
organic phases were dried and evaporated, and the residual
oil was purified via flash chromatography (silica gel, elution
with 20:1 hexanes/ethyl acetate) to give a mixture of two
diastereomeric alcohols (174 mg, 95%). The syn and anti
configurations were elucidated by comparison of 1H NMR
spectra with those of known compounds.10b Integration of the
distinctive methyl doublets (δ 15.3 for syn, δ 13.7 for anti)
showed the syn/anti ratios to be 1:2.
Ad d ition s In volvin g Cr otyl Br om id e a n d Ald eh yd e 5.
A. In H2O. A mixture of 5 (100 mg, 0.532 mmol), crotyl
(18) For general experimental details, consult ref 4b.
(19) Heathcock, C. H.; Kiyooka, S.; Blumenkopf, T. A. J . Org. Chem.
1984, 49, 4214.

8804 J . Org. Chem., Vol. 61, No. 25, 1996
Paquette and Mitzel
B. In H2O-THF (1:1). Reaction of 3 (200 mg, 1.21 mmol)
with 1,3-dibromopropene (484 mg, 2.42 mmol) and indium
powder (153 mg, 1.33 mmol) in 1:1 H2O-THF (13.2 mL) for
2.5 h gave 291 mg (88%) of the bromohydrin diastereomers.
Following cyclization to epoxides a -d and independent in-
dium-promoted reduction, the product ratio was determined
to be closely comparable to that observed in part A.
C. In THF . Reaction of 3 (200 mg, 1.21 mmol) with 1,3-
dibromopropene (484 mg, 2.42 mmol) and indium powder (153
mg, 1.33 mmol) in THF (13.2 mL) for 7 h afforded 268 mg
(75%) of bromohydrin isomers. Subsequent to chemical cor-
relation in the predescribed manner, the product distribution
was found to be identical to that established in experiments
A and B.
Ad d ition s In volvin g Dibr om id e 2 a n d Ald eh yd e 4. A.
In H2O. A mixture of 4 (80 mg, 1.08 mmol), 1,3-dibromopro-
pene (432 mg, 2.16 mmol), and indium powder (130 mg, 1.13
mmol) in water (11.3 mL) was stirred at rt for 4 h and
processed as described above to give 168 mg (80%) of the
bromohydrin as a colorless oil.
To a solution of this material (249 mg, 1.28 mmol) in
anhydrous methanol (25 mL) was added potassium carbonate
(270 mg, 1.92 mmol), and stirring was maintained for 12 h
prior to solvent evaporation. The solid was separated by
filtration, and the filtrate was concentrated to give the epoxy
alcohol (127 mg, 87%). This oil was dissolved in CH2Cl2 (25
mL), admixed with tert-butyldimethylsilyl chloride (300 mg,
2.00 mmol) and imidazole (330 mg, 4.85 mmol), and stirred
for 12 h. The solvent was removed in vacuo at 20 °C, and the
product was purified by flash chromatography on silica gel
(elution with 50:1 hexanes/ethyl acetate) to give a colorless
oil (353 mg, 78% overall).
C. In THF . A mixture of 4 (148 mg, 2.0 mmol), 1,3-
dibromopropene (800 mg, 4.0 mmol), and indium powder (252
mg, 2.2 mmol) in THF (22 mL) was stirred at rt for 9 h and
processed as described above to give 265 mg (68%) of the
bromohydrin mixture as a colorless oil. This sample was
transformed into e-h as before for product analysis.
Ad d ition s In volvin g Dibr om o 2 fr om Ald eh yd e 5. In
H2O. A mixture of 5 (190 mg, 1.0 mmol), 1,3-dibromopropene
(400 mg, 2.0 mmol), and indium powder (126 mg, 1.1 mmol)
in water (11 mL) was stirred at rt for 2.5 h and worked up in
the predescribed manner to give 272 mg (88%) of bromohydrin
mixture as a colorless oil. To a solution of this material (270
mg, 0.870 mmol) in THF (10 mL) was added sodium hydride
(24 mg, 0.979 mmol). After 10 h of stirring, water was added,
and the separated aqueous layer was extracted with CH2Cl2
(3 × 15 mL). The combined organic layers were washed with
water, dried, and evaporated. The residue was purified via
flash chromatography on silica gel (elution with 20:1 hexanes/
ethyl acetate) to give epoxides e-h as a colorless oil (182 mg,
92%). The relative percentages of the four epoxides were
ascertained by integration of the characteristic peaks denoted
earlier.
In a separate experiment, 300 mg (0.971 mmol) of the
bromohydrin mixture in water (10 mL) was treated with
indium powder (115 mg, 0.971 mmol) and stirred for 8 h. A
white precipitate was formed during this time. Hydrochloric
acid (10 mL of 1 N) was added, and after 10 min of stirring
the reaction mixture was extracted with ethyl acetate (3 × 15
mL), the combined organic phases were dried and concen-
trated, and the residue was purified by flash column chroma-
tography (silica gel, eluton with 50:1 hexanes/ethyl acetate).
The colorless oil so obtained (198 mg, 89%) was dissolved in
anhydrous methanol (10 mL), treated with a crystal of p-
toluenesulfonic acid, and stirred at rt for 5 h prior to solvent
evaporation. Purification of the residue by flash chromatog-
raphy on silica gel (elution with 2:1 hexanes/ethyl acetate) gave
a 1:9 mixture of the syn and anti diols (85 mg, 85%), as
The presence of all four epoxides was readily discerned by
1H NMR spectroscopy at 300 MHz (in CDCl3). The two trans
epoxides exhibited key signals at δ 3.10 (dd, J ) 4.1, 7.8 Hz,
0.35 H) and δ 2.96 (dd, J ) 4.1, 7.9 Hz, 0.18 H), while the
corresponding absorptions of the cis epoxides appeared at δ
2.85 (dd, J ) 2.2, 4.5 Hz, 0.28 H) and δ 2.77 (dd, J ) 2.2, 4.5
Hz, 0.18 H). On the basis of the ensuing experiment, these
epoxides are recognized to be e-h , respectively.
1
determined by H NMR analysis (see above).
B. In H2O-THF (1:1). From a mixture of 5 (190 mg, 1.0
mmol), 1,3-dibromopropene (400 mg, 2.0 mmol), and indium
powder (126 mg, 1.1 mmol) in 1:1 H2O-THF (11.0 mL), there
was isolated after 2 h 278 mg (90%) of the bromohydrin
mixture as a colorless oil. This unpurified material was
transformed into the epoxides as described above and compa-
rably analyzed.
The 13C NMR spectrum (75 MHz, CDCl3) of this mixture
exhibited nonoverlapping sets of signals at (135.4, 135.2, 132.4,
132.3), (120.4, 120.2, 119.3, 119.1), (67.4, 67.0, 65.5, 65.1),
(63.5, 63.1, 62.7, 62.3), and (57.7, 57.3, 56.6, 56.5) ppm.
In a separate experiment, a sample of the original bromo-
hydrin mixture (168 mg, 0.864 mmol) was stirred with indium
powder (100 mg, 0.864 mmol) in water (8.6 mL) at rt for 7 h,
at which time the reaction mixture was worked up in the usual
manner to afford a mixture of two diols as a viscous, colorless
oil (90 mg, 90%). Distinction between the syn and anti isomers
was accomplished by 1H NMR spectroscopy. Their methyl
shifts appear at δ 1.16 (d, J ) 6.3 Hz) and δ 1.13 (d, J ) 6.4
Hz), respectively.13 Integration of these peaks showed the syn/
anti ratio to be 2:1.
B. In H2O-THF (1:1). A mixture of 4 (148 mg, 2.0 mmol),
1,3-dibromopropene (800 mg, 4.0 mmol), and indium powder
(252 mg, 2.2 mmol) in 1:1 H2O-THF (22 mL) was stirred at
rt for 4.5 h. Comparable processing gave 311 mg (80%) of the
bromohydrin mixture as a colorless oil. This unpurified
material was taken to e-h as described above and analyzed
in this manner.
C. In THF . Stirring 5 (190, 1.0 mmol) and 1,3-dibro-
mopropene (400 mg, 2.0 mmol) with indium powder (126 mg,
1.1 mmol) in THF (11 mL) for 10 h gave 244 mg (79%) of the
bromohydrin mixture. Conversion to the epoxides in the
predescribed manner allowed for analysis of the product ratios
reported in Table 2.
Ack n ow led gm en t. Financial support from the En-
vironmental Protection Agency and the Emissions Re-
duction Research Center at the New J ersey Institute
of Technology is greatfuly acknowledged. The authors
thank Prof. S. F. Martin for an authentic sample of the
anti,syn diol analogue of 9-11 which permitted verifi-
cation of its absence from the crotyl bromide additions
to 3-5, and Dr. Methvin Isaac for relevant discussions.
J O9613489