Catalytic asymmetric addition of alkylzinc and functionalized alkylzinc reagents to ketones

Article abstract of DOI:10.1021/jo048683e

(Chemical Equation Presented) We describe our full report of the catalytic asymmetric addition of simple and functionalized dialkylzinc reagents to a broad range of saturated ketones and enones. The functionalized organozinc reagents contain esters, silyl

Full text of DOI:10.1021/jo048683e

Catalytic Asymmetric Addition of Alkylzinc and Functionalized
Alkylzinc Reagents to Ketones
Sang-Jin Jeon, Hongmei Li, Celina Garc ´ı a, Lynne K. LaRochelle, and Patrick J. Walsh*
P. Roy and Diana T. Vagelos Laboratories, Department of Chemistry, University of Pennsylvania,
2
31 South 34th Street, Philadelphia, Pennsylvania 19104-6323
pwalsh@sas.upenn.edu
Received July 29, 2004
We describe our full report of the catalytic asymmetric addition of simple and functionalized
dialkylzinc reagents to a broad range of saturated ketones and enones. The functionalized organozinc
reagents contain esters, silyl ethers, alkyl chlorides, and alkyl bromides. In general, the resulting
tertiary alcohol products are isolated with high ee’s. With some substrates, yields are low as a
result of the formation of aldol byproducts. Most substrates undergo additions with good yields
reaching as high as 91%.
2
8-15
aryl,16-22
Introduction
ties. Catalytic asymmetric additions of vinyl,
and alkynyl2
3-29
groups to aldehydes have received much
The formation of C-C bonds lies at the very heart of
organic synthesis. As such, the development of new
reactions that result in the formation of C-C bonds is of
fundamental importance. One of the most common
methods to extend the carbon skeleton is through addi-
tion of organometallic reagents to carbonyl compounds.1
Over the past 2 decades, the principal focus in this area
has involved the asymmetric addition of alkyl groups to
aldehydes. Beginning with the seminal works of Oguni3,4
and subsequent mechanistic investigations by Noyori and
(
8) Oppolzer, W.; Radinov, R. N.; Brabander, J. D. Tetrahedron Lett.
1
995, 36, 2607-2610.
(9) Oppolzer, W.; Radinov, R. N. Helv. Chim. Acta 1992, 75, 170-
1
73.
(
1593-1594.
10) Oppolzer, W.; Radinov, R. N. J. Am. Chem. Soc. 1993, 115,
,2
(
11) Lurain, A. E.; Walsh, P. J. J. Am. Chem. Soc. 2003, 125, 10677-
1
0683.
(
12) Chen, Y. K.; Lurain, A. E.; Walsh, P. J. J. Am. Chem. Soc. 2002,
124, 12225-12231.
(
(
(
13) Wipf, P.; Kendall, C. Chem. Eur. J. 2002, 8, 1778-1784.
14) Wipf, P.; Ribe, S. J. Org. Chem. 1998, 63, 6454-6455.
15) Dahmen, S. D.; Br a¨ se, S. Org. Lett. 2001, 3, 4119-4122.
5-7
co-workers, well over 100 catalysts have been reported
that promote the asymmetric addition of organozinc
reagents to aldehydes with very high enantioselectivi-
(16) Dosa, P. I.; Ruble, J. C.; Fu, G. C. J. Org. Chem. 1997, 62, 444-
4
45.
(
17) Bolm, C.; Hermanns, N.; Hildebrand, J. P.; Mu n˜ iz, K. Angew.
Chem., Int. Ed. 2000, 39, 3465-3467.
(
(
(
1) Soai, K.; Niwa, S. Chem. Rev. 1992, 92, 833-856.
2) Pu, L.; Yu, H.-B. Chem. Rev. 2001, 101, 757-824.
3) Oguni, N.; Matsuda, Y.; Kaneko, T. J. Am. Chem. Soc. 1988, 110,
(18) Rudolph, J.; Rasmussen, T.; Bolm, C.; Norrby, P.-O. Angew.
Chem., Int. Ed. 2003, 42, 3002-3005.
(19) Huang, W.-S.; Pu, L. Tetrahedron Lett. 2000, 41, 145-148.
(20) Zhao, G.; Lib, X.-Z.; Wang, X.-R. Tetrahedron: Asymmetry 2001,
12, 399-403.
7
877-7878.
(
4) Oguni, N.; Omi, T. Tetrahedron Lett. 1984, 25, 2823-2824.
5) Kitamura, M.; Suga, S.; Kawai, K.; Noyori, R. J. Am. Chem. Soc.
(
(21) Ko, D.-H.; Kim, K. H.; Ha, D.-C. Org. Lett. 2002, 4, 3759-3762.
(22) Bolm, C.; Hildebrand, J. P.; Mu n˜ iz, K.; Hermanns, N. Angew.
Chem., Int. Ed. 2001, 40, 3284-3308.
1
986, 108, 6071-6072.
(
6) Noyori, R.; Suga, S.; Kawai, K.; Okada, S.; Kitamura, M. Pure
Appl. Chem. 1988, 60, 1597-1606.
7) Kitamura, M.; Okada, S.; Suga, S.; Noyori, R. J. Am. Chem. Soc.
989, 111, 4028-4036.
(23) Pu, L. Tetrahedron 2003, 59, 9873-9886.
(24) Frantz, D. E.; F a¨ ssler, R.; Carreira, E. M. J. Am. Chem. Soc.
2000, 122, 1806-1807.
(
1
10.1021/jo048683e CCC: $30.25 © 2005 American Chemical Society
448
J. Org. Chem. 2005, 70, 448-455
Published on Web 12/16/2004

Asymmetric Addition of Alkylzinc Reagents to Ketones
less attention. Recent studies, however, have resulted in
the development of a variety of catalysts for these
important processes.
Despite significant activity in asymmetric carbonyl
addition reactions, a long-standing problem in organic
synthesis had been the absence of efficient and highly
enantioselective methods to prepare tertiary alcohols.
ketones (eq 2). Although certain substrates underwent
addition with enantioselectivities as high as 89%, most
substrates gave only moderate to low enantioselectivities.
The most significant drawback to this system, however,
was not the low enantioselectivity but the poor catalyst
turnover frequency. At 20 mol % catalyst loading, the
reactions required between 4 and 14 days to reach
completion.4
Although several examples involving organolithium and
Grignard reagents have been reported,6,30-34 these usu-
ally require greater than stoichiometric amounts of chiral
ligands. In addition, the high reactivity of these reagents
precludes the presence of many functional groups and,
therefore, reduces the attractiveness of such strongly
basic reagents. In contrast, organozinc reagents are mild
and exhibit excellent functional group compatibility,35
making them ideal for use in the synthesis of complex,
highly functionalized targets. Organotitanium reagents
and, to a lesser degree, organoaluminum complexes are
also tolerant of functional groups and have been applied
to the asymmetric addition to aldehydes.36-42
6,47
The first examples of the asymmetric addition of
organozinc reagents to ketones43,44 were reported from
4
5
two groups in 1998. Dosa and Fu employed Noyori’s
5
DAIB ligand in the asymmetric addition of diphenylzinc
to ketones (eq 1). Enantioselectivities as high as 91%
were realized with up to 91% yield. Catalysts derived
from amino alcohols such as DAIB will promote the
addition of the more reactive diphenylzinc to ketones, but
not dialkylzinc reagents, because of their reduced reac-
tivity.
We have been actively investigating the mechanisms
of titanium-catalyzed asymmetric additions of alkyl
groups to aldehydes for several years.4
8-53
The results of
these works enabled us to develop the first practical and
highly enantioselective catalyst for the asymmetric ad-
Also in 1998 Ramon and Yus,46,47 reported the first
example of asymmetric addition of alkyl groups to
dition of alkyl,4
4,54,55
phenyl,
56,57
and vinyl groups to
58
ketones. The enantioenriched tertiary alcohol products
of this reaction are useful chiral building blocks and have
been used in natural product synthesis.59 A shortcoming
associated with the use of dialkylzinc reagents, however,
is that very few are commercially available. Furthermore,
like most main-group organometallics, dialkylzinc re-
agents are sensitive to air and moisture. To circumvent
the limited availability of organozinc reagents, Knochel
and co-workers have developed several excellent methods
(
25) Boyall, D.; L o´ pez, F.; Sasaki, H.; Frantz, D.; Carreira, E. M.
Org. Lett. 2000, 2, 4233-4236.
26) Gao, G.; Moore, D.; Xie, R.-G.; Pu, L. Org. Lett. 2002, 4, 4143-
(
4
146.
(
27) Xu, M. H.; Pu, L. Org. Lett. 2002, 4, 4555-4557.
28) Xu, Z.; Wang, R.; Xu, J.; Da, C.-S.; Yan, W.-J.; Chen, C. Angew.
(
Chem., Int. Ed. 2003, 42, 5747-5749.
(
29) Lu, G.; Li, X.; Jia, X.; Chan, W. L.; Chan, A. S. C. Angew. Chem.,
Int. Ed. 2003, 42, 5057-5058.
30) Seebach, D.; Dorr, H.; Bastani, B.; Ehrig, V. Angew. Chem., Int.
Ed. 1969, 8, 982-983.
31) Inch, T. D.; Lewis, G. J.; Sainsbury, G. L.; Sellers, D. J.
Tetrahedron Lett. 1969, 41, 3657-3660.
(
to prepare dialkylzinc compounds bearing various func-
(
tional groups.3
9,60-62
These reagents can be synthesized
(
32) Nozaki, H.; Aratani, T.; Toraya, T.; Noyori, R. Tetrahedron
971, 27, 905-913.
33) Weber, B.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1992, 31,
1
(
(48) Balsells, J.; Walsh, P. J. J. Am. Chem. Soc. 2000, 122, 1802-
8
4-86.
1803.
(
(
34) Weber, B.; Seebach, D. Tetrahedron 1994, 50, 6117-6128.
35) Erdik, E. Organozinc Reagents in Organic Synthesis; CRC
(49) Pritchett, S.; Woodmansee, D. H.; Gantzel, P.; Walsh, P. J. J.
Am. Chem. Soc. 1998, 120, 6423-6424.
Press: Boca Raton, 1996.
(50) Balsells, J.; Betancort, J. M.; Walsh, P. J. Angew. Chem., Int.
Ed. 2000, 39, 3428-3430.
(
36) Takahashi, H.; Kawakita, T.; Yoshioka, M.; Kobayashi, S.;
Ohno, M. Tetrahedron Lett. 1989, 30, 7095-7098.
37) Takahashi, H.; Kawakita, T.; Ohno, M.; Yoshioka, M.; Koba-
yashi, S. Tetrahedron 1992, 48, 5691-5700.
38) Seebach, D.; Beck, A. K.; Heckel, A. Angew. Chem., Int. Ed.
001, 40, 92-138.
39) Knochel, P.; Vettel, S.; Eisenberg, C. Appl. Organomet. Chem.
995, 9, 175-188.
40) Chan, A. S. C.; Zhang, F.-Y.; Yip, C.-W. J. Am. Chem. Soc. 1997,
19, 4080-4081.
41) Lu, J.-F.; You, J.-S.; Gau, H.-M. Tetrahedron: Asymmetry 2000,
1, 2531-2535.
42) Abbott, D. E.; You, J.-S.; Hsieh, S.-H.; Gau, H.-M. J. Chem. Soc.,
Chem. Commun. 2001, 1546-1547.
(51) Balsells, J.; Walsh, P. J. J. Am. Chem. Soc. 2000, 122, 3250-
3251.
(
(52) Balsells, J.; Costa, A. M.; Walsh, P. J. Isr. J. Chem. 2001, 41,
251-261.
(
2
1
1
1
(53) Walsh, P. J. Acc. Chem. Res. 2003, 36, 739-749.
(54) Garc ´ı a, C.; LaRochelle, L. K.; Walsh, P. J. J. Am. Chem. Soc.
2002, 124, 10970-10971.
(
(
(55) Jeon, S.-J.; Walsh, P. J. J. Am. Chem. Soc. 2003, 125, 9544-
9545.
(
(56) Garc ´ı a, C.; Walsh, P. J. Org. Lett. 2003, 5, 3641-3644.
(57) Li, H.; Garc ´ı a, C.; Walsh, P. J. Proc. Nat. Acad. Sci. 2004, 101,
5425-5427.
(
(58) Li, H.; Walsh, P. J. J. Am. Chem. Soc. 2004, 126, 6538-6539.
(59) Yus, M.; Ram o´ n, D. J.; Prieto, O. Eur. J. Org. Chem. 2003,
2745-2748.
(
43) Ram o´ n, D. J.; Yus, M. Angew. Chem., Int. Ed. 2004, 43, 284-
2
7
87.
(
44) Betancort, J. M.; Garc ´ı a, C.; Walsh, P. J. Synlett 2004, 749-
(60) Rozema, M. J.; Eisenberg, C.; L u¨ tjens, H.; Ostwald, R.; Belyk,
K.; Knochel, P. Tetrahedron Lett. 1993, 34, 3115-3118.
(61) Langer, F.; Devasagayaraj, A.; Chavant, P.-Y.; Knochel, P.
Synlett 1994, 410-412.
60.
(
(
(
45) Dosa, P. I.; Fu, G. C. J. Am. Chem. Soc. 1998, 120, 445-446.
46) Ram o´ n, D. J.; Yus, M. Tetrahedron Lett. 1998, 39, 1239-1242.
47) Ram o´ n, D. J.; Yus, M. Tetrahedron 1998, 54, 5651-5666.
(62) Knochel, P. Chemtracts: Org. Chem. 1995, 8, 205-221.
J. Org. Chem, Vol. 70, No. 2, 2005 449

Jeon et al.
and used without isolation and have been successfully
added to aldehydes with high enantioselectivity.
ing a reduction in the catalyst loading of up to a factor
of 10 compared to the Yus system (eq 2).4 Additionally,
the reaction generally required 24-48 h to reach comple-
tion, a significant decrease in the reaction time from the
previously reported best catalyst system. Furthermore,
our method for the asymmetric addition of alkyl groups
to ketones is operationally very simple and has been
verified in other laboratories.6
6,47
In this study we have successfully applied our catalyst
to the asymmetric addition of dialkylzinc and function-
alized organozinc reagents to a broad range of ketones.
We have found, for the first time, that organozinc
compounds bearing esters, chlorides, bromides, and silyl
ethers can be added to ketones to provide functionalized
tertiary alcohols with high levels of enantioselectivity.
A portion of this research has been previously re-
8,69
5
4,55
ported.
Results and Discussion
Asymmetric Addition of Diethyl- and Dimethyl-
zinc to Simple Ketones. The development of our
catalyst for the enantioselective addition of alkyl groups
to ketones was based on mechanistic observations on
related catalysts.44 In these studies we have found that
bis(sulfonamide) diol ligands react with titanium tetra-
isopropoxide to liberate 4 equiv of 2-propanol as il-
lustrated in Scheme 1.44 This compound and its ligand
adduct have been characterized by X-ray crystallography.
We believe that the bis(sulfonamide) diol ligand 1 binds
to titanium in a multidentate similar fashion.
Although electron-donating or electron-withdrawing
substituents on acetophenone derivatives had little effect
on the ee of the tertiary alcohol product from diethylzinc
addition (Table 1, entries 1-6), the substituents did have
a significant impact on the turnover frequencies of the
catalyst (entries 2-5). Electron-deficient 3′-(trifluoro-
methyl)acetophenone reacted readily at 2 mol % catalyst
loading, generating the product in 14 h at room temper-
ature. At the other extreme, 4′-methoxyacetophenone
underwent addition more slowly, requiring over 4 days
for complete conversion, even at 10 mol % ligand loading.
Although the electron-rich 4′-methoxyacetophenone was
expected to react more slowly, we were surprised by the
magnitude of this reactivity difference. Nonetheless,
these acetophenone derivatives gave excellent enantio-
selectivities.
SCHEME 1. Reaction of a Bis(sulfonamide) Diol
with Titanium Tetraisopropoxide; Reaction Can
Be Driven to the Right by Removal of Liberated
Isopropanol
In the case of 2′-methylacetophenone, the increase in
steric hindrance around the carbonyl group resulted in
a dramatic drop in the isolated yield (entry 6). The
shielding of the carbonyl group by the o-methyl substitu-
ent reduces the rate of the addition reaction and results
in competitive and irreversible deprotonation of the
ketone. Once formed, the resulting enolate then under-
goes aldol condensation and dehydration, consuming
another equivalent of substrate. Similar results were
observed with R-tetralone, which gave the tertiary alcohol
in only 35% isolated yield (entry 7). Nonetheless, the
enantioselectivities of 2-methylacetophenone and R-tet-
ralone were over 95%.
Acetophenone derivatives with longer alkyl chains,
such as valerophenone and 1-chloropropiophenone, also
exhibited enantioselectivities just under 90% in the
asymmetric addition (entries 9, 10, and 12).
Enantioenriched allylic alcohols are among the most
useful chiral building blocks in asymmetric synthesis,
because the double bond can be selectively functionalized
by employing hydroxyl-directed reactions. For this reason
alkyl addition to simple enones was examined. As shown
in Table 1 (entries 13 and 14), these substrates exhibited
In previous studies on the (BINOLate)Ti-based cata-
lyzed asymmetric addition of alkylzinc reagents to alde-
hydes, we proved that the carbonyl addition step proceeds
by delivery of an alkyl group from titanium to the
carbonyl carbon rather than involving direct addition of
6
6,67
an alkylzinc reagent to the carbonyl group.
It is
entirely possible that an analogous reaction manifold is
in operation in the addition of alkyl groups to ketones
outlined in this report. In our studies we have found that
the mechanism of titanium-catalyzed asymmetric addi-
tion reactions can be very complex.53 Speculation on the
reaction mechanism or mode of stereochemical induction
is premature at this time.
We first explored the addition of simple, commercially
available diethyl- and dimethylzinc reagents to a variety
of substrates (eq 3). As shown in Table 1, catalyst
loadings as low as 2 mol % could be employed, represent-
(
63) Rozema, M. J.; Sidduri, A.; Knochel, P. J. Org. Chem. 1992,
7, 1956-1958.
64) Langer, F.; Schwink, L.; Devasagayaraj, A.; Chavant, P.-Y.;
Knochel, P. J. Org. Chem. 1996, 61, 8229-8243.
5
(
(
65) Schwink, L.; Knochel, P. Tetrahedron Lett. 1994, 35, 9007-
9
010.
(
66) Davis, T. J.; Balsells, J.; Carroll, P. J.; Walsh, P. J. Org. Lett.
(68) Yus, M.; Ram o´ n, D. J.; Prieto, O. Tetrahedron: Asymmetry
2002, 13, 2291-2293.
(69) Mukaiyama, T.; Shintou, T.; Fukumoto, K. J. Am. Chem. Soc.
2003, 125, 10538-10539.
2
001, 3, 699-702.
(67) Balsells, J.; Davis, T. J.; Carroll, P. J.; Walsh, P. J. J. Am. Chem.
Soc. 2002, 124, 10336-10348.
450 J. Org. Chem., Vol. 70, No. 2, 2005

Asymmetric Addition of Alkylzinc Reagents to Ketones
TABLE 1. Asymmetric Addition of Alkyl Groups to Ketones with Chiral Ligand in eq 3
a
ee’s determined by HPLC or GC; see Supporting Information.
high levels of enantioselectivity in the asymmetric ad-
dition with diethylzinc. The dialkyl ketone 4-phenyl-2-
butanone underwent addition with 70% enantioselectiv-
ity. Although the enantioselectivity with this substrate
was not in the synthetically useful range, it is impressive
that the catalyst is reasonably efficient at differentiating
between the carbonyl lone pairs. In this case, the catalyst
must bind syn to the methyl group or the alkyl chain,
both of which have similar steric environments (Figure
FIGURE 1. Isomeric catalyst-substrate adducts may lead
to decreased enantioselectivity.
1
). Such dialkyl ketones will likely be a challenging class
of substrates for asymmetric addition reactions.
It is interesting to compare trans-4-phenyl-3-butenone
with the saturated analogue 4-phenyl-2-butanone (en-
tries 14 and 15). The presence of the double bond allows
a 5-fold reduction in the catalyst loading and a decrease
in reaction time relative to that of the saturated ketone.
Additionally, the yield and enantioselectivity with
the enone substrate are also significantly higher. Al-
though the enone is less sterically hindered, it is unlikely
that steric effects alone can account for this significant
reactivity and enantioselectivity difference. It is possible
that there is an interaction between the double bond and
either the zinc or titanium. This interaction must be
J. Org. Chem, Vol. 70, No. 2, 2005 451

Jeon et al.
SCHEME 2. (A) Asymmetric Addition of
TABLE 2. Asymmetric Addition of Alkyl Groups to
Cyclic Enonesa
Diphenylzinc to trans-4-Phenyl-3-buten-2-one;
(
B) Asymmetric Vinyl Addition to Acetophenone
weak, however, because titanium(IV) and zinc(II) are
unable to donate significant electron density into the
CdC π* orbital.
In the asymmetric addition of diethylzinc and dimeth-
ylzinc to aldehydes with catalysts derived from amino
alcohol ligands, Noyori and co-workers have reported that
dimethylzinc is less reactive than diethylzinc by a factor
of 20. We have found that reaction of propiophenone with
dimethylzinc and reaction of acetophenone with dieth-
ylzinc with our catalyst under the same conditions both
yield 2-phenyl-2-butanol with similar rates and enantio-
selectivities between 94% and 96% (entries 1 and 17).
The same configuration of the chiral catalyst provided
2
-phenyl-2-butanol of opposite configuration in these
additions, as expected. R-Tetralone again resulted in poor
yield (26%) in the reaction with dimethylzinc. Nonethe-
less, the enantioselectivity was still high (98%, entry 8).
Addition of dimethylzinc to valerophenone proceeded in
a
b
All reactions performed at 10 mol% catalyst loading. ee’s
determined by GC or HPLC; see Supporting Information. c Reac-
tion conducted at 0 °C.
8
5% ee and in 81% yield with 2 mol % catalyst, and 3′-
chloropropiophenone underwent reaction with 96% ee
and 90% yield under the same conditions (entries 11 and
tion with dimethyl- and diethylzinc to afford tertiary
cyclic allylic alcohols in good to excellent enantioselec-
tivities (Table 2). Reactions were conducted at room
temperature with 10 mol % catalyst and were complete
in less than 40 h in most cases. With the exception of
entries 1-3 and the benzofuranone (entries 12 and 13),
the enantioselectivities were g95%. In the case of entries
1-3, which have no substituent at the 2-position, the
products were formed with reduced enantioselectivity
1
8).
Addition of dimethylzinc to chalcone proceeded slowly
at 10 mol % over 2 d giving 54% yield of the tertiary
alcohol with an enantioselectivity of 46% (entry 16).
Alcohols of this type are better synthesized by either
56
addition of diphenylzinc to trans-4-phenyl-3-buten-2-one
or vinyl addition to acetophenone derivatives,58 as shown
in Scheme 2.
(
52-61%). This result was not unexpected, because the
Asymmetric Additions to Cyclic Conjugated
Enones. We next turned our attention to enantioselec-
tive 1,2-addition reactions of cyclic R,â-unsaturated
ketones. The resulting tertiary allylic alcohols are
valuable synthetic intermediates that can be easily
elaborated to achieve highly fuctionalized chiral building
blocks.
two lone pairs of the carbonyl oxygens are in very similar
steric environments and the catalyst cannot easily dif-
ferentiate between them. The substrate cyclohexenone
exhibited reactivity higher than that of the 2-substituted
enones in the asymmetric additions as a result of the
decreased steric hindrance around the carbonyl group.
We were, therefore, able to run the asymmetric addition
of diethylzinc to cyclohexenone at 0 °C rather than at
room temperature. Unfortunately, however, reducing the
temperature did not significantly improve the enantio-
selectivity.
5
5
In contrast to substrates lacking a substituent at the
-position, enones bearing substituents R to the carbonyl
2
Using the conditions described in eq 4, we found that
conjugated cyclic enones underwent asymmetric alkyla-
group exhibited excellent levels of enantioselectivity. As
shown in Table 2, these include five-, six-, and seven-
452 J. Org. Chem., Vol. 70, No. 2, 2005

Asymmetric Addition of Alkylzinc Reagents to Ketones
membered cyclic enones. Methyl, phenyl, and TBS-
protected hydroxy methyl substituents at the 2-position
in the cyclohexenone series underwent addition with high
levels of enantioselectivity. When the asymmetric addi-
tion to 2-hydroxymethyl-cyclohex-2-enone was conducted
without protection of the hydroxyl, product of 5% ee was
obtained. Gratifyingly, protection as the TBS ether
allowed the asymmetric addition of dimethyl- and dieth-
ylzinc to proceed with 99% enantioselectivity in 81% yield
in both cases (entries 8 and 9). Densely functionalized,
highly enantioenriched products such as those formed in
these addition reactions are expected to be useful chiral
building blocks in enantioselective synthesis. Decreased
yields were observed with an exocyclic enone, which gave
Asymmetric Additions of Functionalized Dialkyl-
zinc Reagents to Ketones. The results in Tables 1 and
2 demonstrate that commercially available dimethyl- and
diethylzinc readily add to a wide range of ketones with
high enantioselectivities. Unfortunately, only a limited
number of dialkylzinc reagents are commodity chemicals.
To broaden the scope of our ketone addition chemistry,
we desired to explore the use of functionalized dialkylzinc
reagents. Dialkylzinc reagents have been prepared from
zinc halides by a number of routes, the most popular
being transmetalation with more electropositive organo-
metallic compounds such as organolithium or Grignard
reagents.7
5,76
2
The metal halide products (LiX and MgX ),
however, are sufficiently Lewis acidic to catalyze some
asymmetric addition reactions, resulting in erosion of the
enantioselectivity. To circumvent this potential problem,
Knochel and co-workers developed procedures for the
2
0% and 32% yield with 99% enantioselectivity in each
case (entries 10 and 11). Presumably these low yields are
due to formation of aldol products. The substrate 6,7-
dihydro-5H-benzofuran-4-one participated in the asym-
metric addition reaction with dimethyl- and diethylzinc
with good yields (79% and 88%, entries 12 and 13,
respectively). When compared to other ketones containing
substituents R to the carbonyl, this furan derivative
resulted in reduced enantioselectivities with all orga-
nozinc reagents employed in this study. At this stage, it
is not clear what factors are responsible for the dimin-
ished enantioselectivities. The conjugation of the furanyl
oxygen with the carbonyl gives the substrate vinylogous
ester character. The resulting tertiary alcohol was ex-
pected to display an increased propensity for racemiza-
tion. Consistent with this prediction, the addition prod-
ucts derived from this substrate racemize readily on
standing. Cyclopentenone derivatives were found to give
excellent enantioselectivities and moderate yields in the
addition process (entries 14-17). Finally, the 8,9-dihydro-
benzocyclohepten-5-one gave yields around 70% and
enantioselectivities g96% (entries 18 and 19). This
substrate, like the cyclohexenones in entries 1 and 2, does
not possess a substituent on the double bond R to the
carbonyl group. Nonetheless, the adjacent benzo group
allows the catalyst to effectively differentiate between the
prochiral faces of the ketone.
generation of a variety of functionalized dialkylzinc
reagents that do not produce salt byproducts.3
9,60-62
These
reagents have been successfully employed in asymmetric
additions to aldehydes catalyzed by titanium bis(sulfona-
mide) complexes to provide functionalized secondary
alcohols.6
4,77
Two of Knochel’s methods for the synthesis of dialkyl-
zinc reagents were employed in this study. The first,
method A, involves reaction of functionalized primary
alkyl iodides with diethylzinc, as outlined in Scheme
6
0,63
3.
This procedure was initiated by combining neat
diethylzinc with the functionalized alkyl iodide and
heating to 50 °C for 20 h. Under these conditions, alkyl
exchange between the zinc and alkyl iodide ensues,
establishing an equilibrium between diethylzinc, the
functionalized alkyl iodide, ethyl iodide, and the func-
tionalized organozinc reagents. Removal of the volatile
materials (ethyl iodide and excess diethylzinc) drives the
equilibrium toward the functionalized dialkylzinc re-
agents. We have generated the organozinc reagents from
1-iodooctane, 1-chloro-4-iodobutane, and 4-iodobutyl piv-
alate. It should be noted that the organozinc reagent
derived from 1-chloro-4-iodobutane is light-sensitive and
must be used shortly after generation. In our hands,
storage of this reagent either neat or as a dilute solution
resulted in formation of a precipitate that was unreactive
in the asymmetric addition reactions.
We have assigned the absolute stereochemistry of three
products from Table 1 (entries 1,70 9, and 17 ) and two
products from Table 2 (entries 1171 and 14 ). In each of
these cases the catalyst based on the (R,R)-diamine added
the alkyl group to the si face of the ketone substrate.
Overall, although the yields of the products in the
asymmetric ketone alkylation reactions are variable, they
are generally good and as high as 88%. In those cases
that give low yields, we believe that the cause is irrevers-
ible deprotonation of the substrate under the basic
reaction conditions followed by aldol condensation/
dehydration. In some cases, the aldol products were
isolated and characterized. It is interesting to note that
the oxophilic titanium-based catalyst results in formation
of the 1,2-addition products with no isolation of 1,4-
addition products. In contrast, the softer copper(I) cata-
lysts promote conjugate addition with dialkylzinc re-
46
70
72
SCHEME 3. Synthesis of Dialkylzinc Reagents by
Method A and Method B
(73) Degrado, S. J.; Mizutani, H.; Hoveyda, A. H. J. Am. Chem. Soc.
2002, 124, 13362-13363.
7
3,74
agents.
(
74) Kitamura, M.; Miki, T.; Nakano, K.; Noyori, R. Tetrahedron
Lett. 1996, 37, 5141-5144.
(
70) Archelas, A.; Furstoss, R. J. Org. Chem. 1999, 64, 6112-6114.
71) Fujisawa, T.; Watanaba, M.; Sato, T. Chem. Lett. 1984, 2055-
(75) Seebach, D.; Behrendt, L.; Felix, D. Angew. Chem., Int. Ed.
Engl. 1991, 20, 1008-1009.
(76) Weber, B.; Seebach, D. Tetrahedron 1994, 50, 7473-7484.
(77) Brieden, W.; Ostwald, R.; Knochel, P. Angew. Chem., Int. Ed.
1993, 32, 582-584.
(
2
2
058.
(
72) Lee, S.-F.; Barth, G.; Djerassi, C. J. Am. Chem. Soc. 1981, 103,
95-301.
J. Org. Chem, Vol. 70, No. 2, 2005 453

Jeon et al.
TABLE 3. Functionalized Dialkylzinc Addition to
Ketones using Method Aa
TABLE 4. Asymmetric Addition of Functionalized Alkyl
Groups to Ketones using Method Ba
a
All reactions performed at 10 mol% catalyst loading. b ee’s
a
b
All reactions performed at 10 mol% catalyst loading. ee’s
determined by HPLC or GC; see Supporting Information.
determined by HPLC or GC; see Supporting Information. c ZnR2
was made from BH3‚SMe2 instead of HBEt2.
The in situ generated diorganozinc reagents were then
employed in the asymmetric addition to a variety of
ketones under the conditions outlined in eq 4. In the
initial stages of our investigations, we found it easiest
to generate organozinc reagents without functional groups.
Once we had mastered the synthesis and asymmetric
addition of the newly formed organozinc reagents, we
focused on the use of functionalized derivatives. As
outlined in the Experimental Section, care must be
exercised in the preparation of the organozinc reagents
using these methods. Examination of the results of these
reactions (Table 3) indicates that the additions to ac-
etophenone derivatives proceeded with enantioselectivi-
ties g96% (entries 1-6). In these cases, the yields were
best with addition of octyl or 4-chlorobutyl groups (70-
ketones to provide access to functionalized tertiary alco-
hols with good to excellent levels of enantioselectivity and
moderate to high yields.
The second procedure for the in situ preparation of
dialkylzinc reagents, method B, is based on a boron-zinc
exchange reaction (Scheme 3).6
4,65
Initial hydroboration
‚SMe
of the olefin employed either diethylborane or BH
3
2
with olefin/borane ratios of 1:1 and 3:1, respectively at
0 °C. Combination of the resulting borane with diethyl-
zinc at 0 °C initiated the alkyl group redistribution. Like
method A, the resulting equilibrium must be driven
toward the functionalized organozinc reagent by removal
of the volatile material, including the remaining dieth-
ylzinc and triethylborane. Caution must again be exer-
cised in quenching the excess diethylzinc and triethyl-
borane (see the Experimental Section).
Results for the asymmetric ketone alkylation reactions
using these reagents are shown in Table 4. Using the
conditions outlined in eq 4, several classes of ketones
were examined with dialkylzinc reagents containing a
branched alkyl group, TBS ether, pivalyl ester, and
bromide functional groups. Additions to acetophenone
derivatives and acyclic and cyclic conjugated enones
provided functionalized alcohols with high levels of
enantioselectivity (90-99%). As observed in method A,
however, the benzofuranone derivative resulted in slightly
9
5%). Reactions of the dipivalyl derivative were slower
and resulted in yields between 53% and 58% (entries 3
and 6). The reduced reactivity of the pivalyl derivative
could be due to competitive binding of the ester carbonyl
group to the catalyst or intra- or intermolecular coordina-
tion to zinc. The functionalized organozinc reagents were
then examined in additions to cyclic enones (entries
7
-16). Very high enantioselectivities were observed
g96%) except for additions to the benzofuranone sub-
strate, where enantioselectivities ranged from 76% to
5% (entries 10-12). The results in Table 3 indicate that
(
8
the alkyl iodide/diethylzinc exchange method of Knochel
can be readily employed in the asymmetric additions to
454 J. Org. Chem., Vol. 70, No. 2, 2005

Asymmetric Addition of Alkylzinc Reagents to Ketones
lower enantioselectivity (entry 18). In our hands, yields
were significantly better when the hydroboration was
performed with diethylborane than with borane dimethyl
sulfide (compare entries 1 and 2 as well as 13 and 14).
There was essentially no impact of the choice of hydrobo-
rating reagent on the enantioselectivities. On the basis
of these results, no further reactions were initiated with
borane dimethyl sulfide.
In the addition of protected alcohols to 3′-methylac-
etophenone, the TBS ether and the pivalate both exhib-
ited high enantioselectivities (entries 3 and 4). The yield
of the tertiary alcohol with the TBS ether was much
higher, however. The TBS ether additions showed vari-
ability in the yields ranging from 51% to 89% (entries 3,
collected in a cold trap cooled by liquid nitrogen. Quenching
these reagents should be performed under a nitrogen atmo-
sphere by dilution with hexanes followed by dropwise addition
of a solution of 2-propanol in hexanes.
General Procedure for Asymmetric Addition Reac-
tion. 2-(tert-Butyl-dimethyl-silanyloxymethyl)-1-methyl-
cyclohex-2-enol. The bis(sulfonamide) diol ligand 1 (11 mg,
0.02 mmol, 10 mol %) was weighed into a 10-mL flame-dried
Schlenk flask and purged with dinitrogen. Dimethylzinc (0.6
mL of a 1 M toluene solution, 0.6 mmol, 3 equiv) was added
followed by addition of the titanium(IV) isopropoxide (0.18 mL
of a 1.4 M hexanes solution, 0.24 mmol, 1.2 equiv) at room
temperature. After 5 min, 2-(tert-butyl-dimethyl-silanyloxy-
methyl)-cyclohex-2-enone (50 µL, 0.2 mmol) was added neat.
The reaction mixture was stirred at room temperature. After
4
0 h the reaction was quenched with a few drops of water.
Ethyl acetate (3-4 mL) was added, and the resulting solution
was dried over MgSO , filtered, and concentrated under
9
, 11, and 15), and yields with the pivalate derivative
were, in general, lower than those with other zinc
reagents (entries 4, 7, 17, and 18). We were also able to
employ an alkylbromide in the addition reaction with 3′-
methylacetophenone, 2-acetonaphthone, and trans-4-
phenyl-3-buten-2-one (entries 5, 12, and 16). The enan-
tioselectivities were g90% with yields between 48% and
4
reduced pressure. The crude product was purified by column
chromatography on silica gel (hexanes/EtOAc 95/5) to give the
20
product (41.5 mg, 81% yield, >99% ee) as an oil: [R]
D
) +54.3
, 500 MHz) δ 0.07 (s, 3H), 0.10
s, 3H), 0.91 (s, 9H), 1.36 (s, 3H), 1.52-1.61 (m, 2H), 1.73-
1
(
3 3
c 1.27, CHCl ); H NMR (CDCl
(
1
1
.80 (m, 2H), 1.91-1.97 (m, 1H), 2.06-2.13 (m, 1H), 3.38 (br,
H), 4.04 (d, J ) 11.7 Hz, 1H), 4.50 (m, 1H), 5.67 (t, J ) 3.6
8
9%.
Comparison of the zinc-alkyl iodide exchange (method
A) with the hydroboration-transmetalation protocol
method B) indicates that, in general, the zinc-alkyl
13
1
Hz, 1H) ppm; C{ H} NMR (CDCl
3
, 125 MHz) δ -5.14, -5.02,
1
8.6, 19.6, 26.0, 26.3, 28.4, 39.0, 66.6, 70.0, 127.0, 140.0 ppm;
-1
(
IR (NaCl) 3452, 2950, 2739, 1664, 1471, 1388, 1256, 1140 cm ;
+
iodide method gives better yields with our catalyst in the
asymmetric additions to ketones. Method B, however,
allows addition of alkyl bromides, whereas such an
addition with method A would be more difficult. The
enantioselectivities with these methods are very similar.
As observed in Tables 1-4, the levels of enantioselectivity
are very high for a wide range of ketones and dialkylzinc
reagents.
HRMS calcd for C14
H27OSi (M-OH) 239.1831, found 239.1838.
General Procedure for the Generation of Function-
alized Organozinc Reagents From Alkyl Iodides.6
0,63
A
1
0-mL flame-dried Schlenk flask was equipped with a nitrogen
inlet, a septum, and a stirring bar. It was then charged with
the functionalized iodide PivO(CH I (2 g, 7.1 mmol) and
2 4
)
diethylzinc (3.6 mL, 36 mmol). The reaction mixture was
briefly evacuated, warmed to 50 °C, and stirred for 12 h. The
flask was then cooled to room temperature and subjected to
vacuum (0.4 mmHg). The volatile materials, including the
ethyl iodide and the remaining diethylzinc, were removed
under reduced pressure (ca. 3 h). The resulting organozinc
reagent, (PivO(CH ) ) Zn, was dissolved in toluene to make 1
2 4 2
M solution and used in the asymmetric addition reaction as
described in the General Procedure above.
Conclusions
We report the first catalytic asymmetric addition of
functionalized organozinc reagents to ketones. The or-
ganozinc reagents are generated from alkyl iodides or
alkenes using the protocols of Knochel. These reactive
dialkylzinc reagents are not isolated but can be used
directly in the asymmetric addition reactions. The enan-
tioselectivities generally exceed 90%, providing access to
a broad range of tertiary alcohols and tertiary allylic
alcohols. Also included are asymmetric additions of the
commercially available dimethyl- and diethylzinc to a
series of substituted ketones. Enantioselectivities in these
reactions are excellent for the most part. In our hand,
yields with the commercially available dialkylzinc re-
agents are higher and more consistent than with the in
situ generated functionalized derivatives.
These investigations indicate that our catalyst for the
asymmetric addition of organozinc reagents to ketones
is tolerant of a variety of functionalized zinc reagents.
The methods introduced here provide a solution to a long-
standing problem in enantioselective synthesis: the
preparation of highly enantioenriched, functionalized
tertiary alcohols.
General Procedure for the Generation of Function-
alized Organozinc Reagents From Alkenes via Hy-
6
4,65
droboration.
5 mmol) was placed in a Schlenk flask and cooled to 0 °C,
and Et BH (15 mL, 15 mmol, 1 equiv) was added via syringe
3 2
Et BH was prepared by combining BH ‚SMe (0.48 mL, 5
2 2 2
The alkene CH dCHCH CH OPiv (2.34 g,
1
2
(
2
mmol), BEt (1.41 mL, 10 mmol), and hexanes (13.1 mL)). After
3
3 h at 0 °C the volatile materials were removed under vacuum
(0.5 mmHg), affording the diethyl(alkyl)borane, (PivO(CH
)
2 4
)-
Zn (20 mL of a 1 M
hexanes solution, 20 mmol) slowly at 0 °C. After 0.5 h the
volatile materials, including the excess of Et Zn and the Et B,
2 2
BEt . To this Schlenk flask was added Et
2
3
were removed under reduced pressure at 0.5 mmHg for 3 h at
room temperature. The resulting oil was dissolved in toluene
to make a 1 M solution that was used in the General Procedure
for the asymmetric addition outlined above.
Acknowledgment. This work was supported by the
NIH (GM58101) and the NSF (CHE-0315913). We are
grateful to Akzo Nobel for the gift of alkyl zinc reagents.
Supporting Information Available: General methods,
synthesis of organozinc species, full characterization of new
compounds, conditions for determination of ee, and determi-
nation of absolute stereochemistry. This material is available
free of charge via the Internet at http://pubs.acs.org.
Experimental Section
Cautionary Note. Caution must be used in the synthesis
of functionalized dialkylzinc reagents. The techniques em-
ployed herein require removal of volatile dialkylzincs and
trialkylboranes under reduced pressure. These materials are
JO048683E
J. Org. Chem, Vol. 70, No. 2, 2005 455