Stereoselective alkylations of chiral nitro imine and nitro hydrazone dianions. Synthesis of enantiomerically enriched 3-substituted 1-nitrocyclohexenes

Article abstract of DOI:10.1021/jo801790r

(Chemical Equation Presented) Dianions of chiral nitro imines (generated by a combination of LDA and s-BuLi) underwent diastereoselective alkylation with methyl, butyl, isopropyl, allyl, and methallyl iodides. In contrast to the behavior of simple metalloenamines, the most selective auxiliary contained no coordinating groups but did possess a large steric difference between the two substituents. The yield and selectivity of the alkylations were improved by the addition of HMPA or DMPU. The use of (S)-1-naphthylethylamine as the auxiliary afforded the R absolute configuration of the alkylation products. This stereochemical outcome could be rationalized by simple steric approach controlled alkylation in a conformationally fixed, internally coordinated dianion. A SAMP nitro hydrazone gave poorer yields and selectivities.

Full text of DOI:10.1021/jo801790r

Stereoselective Alkylations of Chiral Nitro Imine and Nitro
Hydrazone Dianions. Synthesis of Enantiomerically Enriched
3-Substituted 1-Nitrocyclohexenes†
Scott E. Denmark* and Jeffrey J. Ares
Roger Adams Laboratory, Department of Chemistry, UniVersity of Illinois, Urbana, Illinois 61801
sdenmark@illinois.edu
ReceiVed August 12, 2008
Dianions of chiral nitro imines (generated by a combination of LDA and s-BuLi) underwent dias-
tereoselective alkylation with methyl, butyl, isopropyl, allyl, and methallyl iodides. In contrast to the
behavior of simple metalloenamines, the most selective auxiliary contained no coordinating groups but
did possess a large steric difference between the two substituents. The yield and selectivity of the alkylations
were improved by the addition of HMPA or DMPU. The use of (S)-1-naphthylethylamine as the auxiliary
afforded the R absolute configuration of the alkylation products. This stereochemical outcome could be
rationalized by simple steric approach controlled alkylation in a conformationally fixed, internally
coordinated dianion. A SAMP nitro hydrazone gave poorer yields and selectivities.
Introduction
chiral, nonracemic vinyl ethers are used as dienophiles, the
R-hydroxy lactams (formed by hydrogenolytic unmasking) are
obtained in high enantiomeric purity. This strategy has been
employed in the total syntheses of many representatives of the
pyrrolizidine3 and indolizidine4 classes of natural products,
amino sugars,5 sceletium,6a melodinus,6b and daphniphyllum
alkaloids.6c In addition, the tandem cycloaddition sequence has
been employed to prepare 1-azafenestranes, a novel class of
strained polycyclic compounds.7
Nitroalkenes are an extremely versatile class of organic
compounds, serving as substrates for a wide variety of syntheti-
cally useful reactions.1 One of the major themes in this research
group over the past two decades is the invention, development,
and application of tandem [4 + 2]/[3 + 2] cycloaddition
reactions in which nitroalkenes serve as the 4π components.2
Nitroalkenes undergo smooth cycloaddition reactions with both
activated and unactivated olefins in both an intramolecular and
intermolecular modes. The nitronates generated by this process
serve as useful 1,3-dipoles in [3 + 2] cycloaddition reactions
also in both intermolecular and intramolecular processes. If
In the context of the intramolecular nitroalkene cycloaddition
studies, an efficient and general synthesis of 3-substituted
1-nitrocyclohexenes was required. This was accomplished
(Scheme 1) through the alkylation of the dianion of either a
nitro imine or a nitro hydrazone, followed by reduction and
† This paper is dedicated to the memory of Professor Albert I. Meyers for
his pioneering contributions to asymmetric synthesis and his inspiring example
of passion for organic chemistry.
(1) (a) Berner, O. M.; Tedeschi, L.; Enders, D. Eur. J. Org. Chem. 2002,
1877. (b) The Nitro Group in Organic Synthesis; Ono, N. , Ed.; Wiley: New
York, 2001. (c) Perekalin, V. V.; Lipina, E. S.; Berestovitskaya, V. M.; Efremov,
D. A. Nitroalkenes, Conjugated Nitro Compounds; Wiley: Chichester, 1994. (d)
Barrett, A. G. M. Chem. Soc. ReV. 1991, 20, 95. (e) Kabalka, G. W.; Varma,
R. S. Org. Prep. Proced. Int. 1987, 19, 283. (f) Barrett, A. G. M.; Grabowski,
G. G. Chem. ReV. 1986, 86, 751. (g) Corey, E. J.; Estreicher, H. J. Am. Chem.
Soc. 1978, 100, 6294.
(2) (a) Denmark, S. E.; Thorarensen, A. Chem. ReV. 1996, 96, 137. (b)
Denmark, S. E.; Cottell, J. J. Synthetic Applications of 1,3-Dipolar Cycloaddition
Chemistry toward Heterocycles and Natural Products; John Wiley & Sons: New
Jersey, 2003; Chapter 2.
(3) (a) Denmark, S. E.; Hurd, A. R. J. Org. Chem. 2000, 65, 2875. (b)
Denmark, S. E.; Herbert, B. J. Org. Chem. 2000, 65, 2887, and references cited
therein.
(4) Denmark, S. E.; Martinborough, E. A. J. Am. Chem. Soc. 1999, 121,
3046.
(5) Denmark, S. E.; Juhl, M. HelV. Chim. Acta 2002, 85, 3712.
(6) (a) Denmark, S. E.; Marcin, L. R. J. Org. Chem. 1997, 62, 1675. (b)
Denmark, S. E.; Cottell, J. J. AdV. Synth. Catal. 2006, 348, 2397. (c) Denmark,
S. E.; Baiazitov, R. Y. J. Org. Chem. 2006, 71, 593.
(7) Denmark, S. E.; Montgomery, J. I.; Kramps, L. J. Am. Chem. Soc. 2006,
128, 11620.
10.1021/jo801790r CCC: $40.75
Published on Web 10/15/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 9647–9656 9647

Denmark and Ares
elimination to afford the substituted 1-nitrocyclohexene.8 This
process could also serve as an ideal template for the development
of auxiliary-based asymmetric alkylation reactions, using chiral
nitro imines or nitro hydrazones. Similar reduction and elimina-
tion of the alkylated products would then provide enantiomeri-
cally enriched 3-substituted 1-nitrocyclohexenes. The realization
of this asymmetric alkylation could have numerous implications.
First, it would provide an example of the relatively rare
auxiliary-based stereoselective alkylation of a dianion. Second,
the reduction-elimination sequence would provide access to
versatile nitroalkenes in enantiomerically enriched form. Finally,
through the use of the Nef reaction,9 these 3-substituted
cyclohexenes could be transformed into enantiomerically en-
riched 3-substituted cyclohexanones. This approach to these
compounds would complement the well-established asymmetric
addition of organocuprates to R,ꢀ-unsaturated cyclic ketones.10
demonstrated the feasibility of this process but generally resulted
in low levels of stereoselectivity.14,15
In the mid-1970s, the first major breakthrough in the field of
stereoselective alkylations of chiral metalloenamines was re-
ported by Albert I. Meyers (Scheme 2).16 For example, the
cyclohexanone imine prepared from (S)-phenylalaninol methyl
ether could be metalated with LDA and alkylated to obtain,
after hydrolysis, enantiomerically enriched 2-alkylcyclohex-
anones with enantiomeric ratios (ers) generally greater than 90:
10. The success of the Meyers method was proposed to arise
from the formation of a lithium chelate in the metalloenamine,
the rigidity of which would influence the direction of attack of
the approaching electrophile. The poorer selectivities observed
in the earlier attempts14,15 were, most likely, due to the absence
of an appropriate coordinating group in the chiral amine
precursor. Thus, lacking a chelate and its superior orienting
properties, the metalloenamines were not effective diastereoface
controllers.
SCHEME 1
SCHEME 2
In 1988, we communicated our initial results on the alkylation
of chiral nitro imine dianions and the conversion of these
alkylated products into enantiomerically enriched nitroalkenes.11
Remarkably, in the intervening 20 years, no new methods for
the enantioselective synthesis of such nitroalkenes have been
described. Because the foundation of this work is rooted in the
pioneering studies of Professor Albert I. Meyers, we felt it a
fitting tribute to this great chemist to provide the full details of
this work, as well as the results of alkylation studies on the
dianion of a chiral nitro hydrazone.
17
2. Stereoselective Alkylations of Hydrazones. In 1976,
Enders reported his initial work on the asymmetric alkylation
of hydrazones derived from (S)-1-amino-2-methoxymethylpyr-
rolidine (SAMP).18,19 SAMP hydrazones prepared from ketones
and aldehydes can be metalated and alkylated with broad
generality, high yield, and selectivity (Scheme 3). Removal of
the auxiliary from the product hydrazones by hydrolysis or
ozonolysis afforded R-alkylated ketones in high enantiomeric
purity. As in the case of metalloenamines, the coordination of
lithium with the methoxy oxygen is believed to play a major
role in controlling the stereoselectivity observed. Further
development of the use of SAMP and its enantiomer (RAMP)
by the Enders group has resulted in a wide variety of asymmetric
alkylation reactions that still in 2008 represent the methods of
choice for the asymmetric synthesis of alkylated ketones and
aldehydes.17,20 Furthermore, modified versions of SAMP have
been employed in stereoselective reactions.21 Remarkably,
Background
1. Stereoselective Alkylations of Metalloenamines. 12 The
discovery of metalloenamines as enolate equivalents in the early
1960s represented a landmark advance in the development of
site-selective carbon-carbon bond forming reactions.13 In
addition to sharing many of the same advantages over classical
enolates as their enamine analogues, metalloenamines offered
the additional benefit of greater chemical reactivity compared
to enamines.13a Moreover, the use of chiral amines as precursors
to the metalated imines introduced the ability to carry out these
alkylations in a stereoselective manner. Early efforts in this area
(8) Denmark, S. E.; Sternberg, J. A.; Lueoend, R. J. Org. Chem. 1988, 53,
1251.
(9) For a review on the Nef reaction, see: Noland, W. E. Chem. ReV. 1955,
55, 137.
(14) Mea-Jacheet, D.; Horeau, A. Bull. Soc. Chim. Fr. 1968, 4571.
(15) Kitomoto, M.; Hiroi, K.; Terashima, S.; Yamada, S. Chem. Pharm. Bull.
1974, 22, 459.
(10) (a) Feringa, B. L. Acc. Chem. Res. 2000, 33, 346. (b) Perlmutter, P.
Conjugate Addition Reactions in Organic Synthesis; Pergamon: Oxford, 1992.
(c) Schmalz, H.-G. ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: Oxford, 1991; Vol. 4, Chapter 1.5. (d) Rossiter, B. E.;
Swingle, N. M. Chem. ReV. 1992, 92, 771.
(16) (a) Meyers, A. I.; Williams, D. R.; Druelinger, M. J. Am. Chem. Soc.
1976, 98, 3032. (b) Meyers, A. I.; Williams, D. R.; Erickson, G. W.; White, S;
Druelinger, M. J. Am. Chem. Soc. 1981, 103, 3081. For a personal, historical
account, see: (c) Meyers, A. I. J. Org. Chem. 2005, 70, 6137.
(17) For reviews on the chemistry of hydrazone anions, see: (a) Enders, D.
In Asymmetric Synthesis; Morrison, J. D. , Ed.; Academic Press: New York,
1984; Vol 3, Chapter 4. (b) Bergbreiter, D. E.; Momongan, M. In ComprehensiVe
Organic Synthesis, Additions to C-X Bonds, Part 2; Heathcock, C. H., Ed.;
Pergamon Press: Oxford, 1991; Vol. 2, Chapter 1.17. (c) Fey, P. In StereoselectiVe
Synthesis, Houben-Weyl, Methods of Organic Chemistry; Helmchen, G., Hoffman,
R. W., Schaumann, E., Eds.; Theime: Stuttgart, 1995; Vol. E21a, pp 994-1015.
(d) Enders, D.; Peters, R.; Wortmann, L. Acc. Chem. Res. 2000, 33, 157. (e)
Job, A.; Janeck, C. F.; Bettray, W.; Peters, R.; Enders, D. Tetrahedron 2002,
58, 2253.
(11) Denmark, S. E.; Ares, J. J. J. Am. Chem. Soc. 1988, 110, 4432.
(12) For reviews of metalloenamine chemistry see: (a) Bergbreiter, D. E.;
Newcomb, M. In Asymmetric Synthesis; Morrison, J. D., Ed.; Academic Press:
New York, 1983; Vol. 2, Chapter 9. (b) Fraser, R. R. In ComprehensiVe
Carbanion Chemistry; Buncel, E., Durst, T., Eds.; Elsevier: Amsterdam, 1984;
Part B, Chapter 2. (c) Martin, S. F. In ComprehensiVe Organic Synthesis,
Additions to C-X Bonds, Part 2; Heathcock, C. H., Ed.; Pergamon Press: Oxford,
1991; Vol. 2, Chapter 1.16. (d) Mangelinckx, S.; Guibellina, N.; De Kimpe, N.
Chem. ReV. 2004, 104, 2353.
(13) (a) Stork, G.; Dowd, S. J. Am. Chem. Soc. 1963, 85, 2178. (b) Wittig,
G.; Frommeld, H. D.; Suchanek, P. Angew. Chem. 1963, 75, 978.
(18) Enders, D.; Eichenauer, H. Angew. Chem., Int. Ed. Engl. 1976, 15, 549.
(19) Enders, D.; Kipphardt, H.; Fey, P. Org. Synth. 1987, 65, 183.
9648 J. Org. Chem. Vol. 73, No. 24, 2008

Chiral Nitro Imine and Nitro Hydrazone Dianions
asymmetric alkylation remains one of the truly useful synthetic
reactions for which a general asymmetric, catalytic process has
yet to be developed.22
derivatives. Finally, Myers29 employed pseudoephedrine as a
chiral auxiliary in the dianionic diastereoselective alkylation of
hydroxy amides. Removal of the auxiliary affords enantiomeri-
cally enriched carboxylic acids, primary alcohols, aldehydes,
or ketones.
SCHEME 3
CHART 1
3. Stereoselective Alkylations of Dianions. The alkylation
of dianions has long been known as a versatile and important
way to construct carbon-carbon bonds.23 In recent years,
asymmetric variants of dianion alkylations have become more
commonplace. Such asymmetric dianion alkylations can be
divided into two main categories depending on whether remov-
able chiral auxiliaries are used. Those dianion alkylations that
are simply diastereoselective by nature are much more com-
mon.24
In contrast, employment of a chiral auxiliary for asymmetric
induction in a dianion alkylation, followed by auxiliary removal
to leave behind the newly created stereogenic center, is rare.
Following our initial disclosure,11 Bartoli25 reported the alky-
lation of chiral ꢀ-enamino ketones via their dianions (Chart 1).
Substrates of this type are derived from 1-phenylethylamine.
With the exception of the methyl electrophiles, alkylations
generally proceed with diastereoselectivities (drs) greater than
85:15, and these reactions ultimately afforded enantiomerically
enriched γ-alkylated chiral 1,3-diketones. Two groups have used
8-phenylmenthol and related compounds as dianionic chiral
auxiliaries in the synthesis of R-alkyl amino acid derivatives.
Fukumoto26 uses carboxylic acids, while Berkowitz27 employs
amides to effect the dianionic asymmetric alkylation at the group
next to the carboxylic acid or amide nitrogen, respectively.
Jenkins28 reported the stereoselective alkylation of cyclohexane-
1,2-dicarboxylic acid monomenthyl ester to afford, after aux-
iliary removal, optically active bishydroxymethyl cyclohexane
Results
1. Nitro Imines.30
1.1. Preparation and Alkylation of Nitro Imines. Follow-
ing the Meyers’ insight for metalloenamine monalkylation,16a
we began our studies by focusing on chiral auxiliaries that
contained coordinating groups. Chiral nitro imines were syn-
thesized as indicated in Scheme 4, using the procedure
developed in these laboratories for the synthesis of achiral nitro
imines.8 Accordingly, cyclohexanone was converted into the
corresponding enol acetate, which was then transformed into
2-nitrocyclohexanone according to the method of Zajac.31
Treatment of this nitro ketone with the acetate salt of the
appropriate chiral amine (benzene, reflux) afforded the desired
nitro imines. The initial set of alkylation substrates is indicated
in Table 1. Chiral methoxy amines used for the synthesis of
nitro imines 1 and 2 were derived from (S)-phenylalaninol and
(S)-phenylglycinol,respectively,usingtheprocedureofMeyers.16a,32
The methoxy amine used in the synthesis of 3 was prepared
from L-norpseudoephedrine by methyl iodide alkylation of the
potassium alkoxide anion. All chiral nitro imines were found
to exist exclusively in the aci-nitro form as judged by their
infrared spectra.8,33 For the neutral nitro enamine (nitro ene
hydrazine) structure, the CdC (1630-1660 cm-1) and NO2 (νs
1250-1280 cm-1) IR stretches are most diagnostic.33a,b The
dipolar (aci-nitronate) structure is characterized by the strong
(20) (a) Enders, D.; Schubert, H.; Nubling, C. Angew. Chem., Int. Ed. Engl.
1986, 25, 1109. (b) Enders, D.; Bhushan, V. Tetrahedron Lett. 1988, 29, 2437.
(c) Enders, D.; Schafer, T.; Piva, O.; Zamponi, A. Tetrahedron 1994, 50, 3349.
(d) Enders, D.; Gatzweiler, W.; Jegelka, U. Synthesis 1991, 1137. (e) Enders,
E.; Zamponi, A.; Raabe, G. Synlett 1992, 897.
(21) (a) Lohray, B. B.; Enders, D. Synthesis 1993, 1092. (b) Enders, D.;
Bettray, W.; Raabe, G.; Runsink, J. Synthesis 1994, 1322.
(22) For notable exceptions, see: (a) List, B. J. Am. Chem. Soc. 2004, 126,
450. (b) Doyle, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127, 62. (c)
Beeson, T. D.; Mastracchio, A.; Hong, J.-B.; Ashton, K.; MacMillan, D. W. C.
Science 2007, 316, 582. Asymmetric phase transfer alkylations must also be
included. See: (d) Asymmetric Phase Transfer Reactions; Maruoka, K., Ed.;
Wiley-VCH: Weinheim, Germany, 2007.
(23) (a) Kaiser, E. M.; Detty, J. D.; Knutson, P. L. A. Synthesis 1977, 509.
(b) Bates, R. B. In ComprehensiVe Carbanion Chemistry, Part A; Buncel, E.,
Durst, D., Eds.; Elsevier: Amsterdam, 1980; Chapter 1. (c) Seebach, D.;
Pohmakotor, M. Tetrahedron 1981, 37, 4047. (d) Thompson, C. M.; Green,
D. L. C. Tetrahedron 1991, 47, 4223. (e) Thompson, C. M. Dianion Chemistry
in Organic Synthesis; CRC Press: Boca Raton, FL, 1994.
(24) (a) Neeland, E. G.; Ounsworth, J. P.; Sims, R. J.; Weiler, L. J. Org.
Chem. 1994, 59, 7383. (b) Micouin, L.; Schanen, V.; Riche, C.; Chiaroni, A.;
Quirion, J.-C.; Husson, H.-P. Tetrahedron Lett. 1994, 35, 7223. (c) Wittenberger,
S. J.; Baker, W. R.; Donner, B. G. Tetrahedron 1993, 49, 1547. (d) Avery, M. A.;
Chong, W. K. M.; Jennings-White, C. J. Am. Chem. Soc. 1992, 114, 974. (e)
Dugger, R. W.; Ralbovsky, J. L.; Bryant, D.; Commander, J.; Massett, S. S.;
Sage, N. A.; Selvidio, J. R. Tetrahedron Lett. 1992, 33, 6763. (f) Rao, A. V. R.;
Singh, A. K.; Varaprasad, C. V. N. S. Tetrahedron Lett. 1991, 32, 4393.
(25) Bartoli, G.; Bosco, M.; Cimarelli, C.; Dalpozzo, R.; De Munno, G.;
Palmieri, G. Tetrahedron: Asymmetry 1993, 4, 1651.
(28) Hulme, S. J.; Jenkins, P. R.; Fawcett, J.; Russell, D. R. Tetrahedron
Lett. 1994, 35, 5501.
(29) Myers, A. G.; Yang, B. H.; Chen, H.; Gleason, J. L. J. Am. Chem. Soc.
1994, 116, 9361.
(30) For a review on the chemistry of nitroenamines, see: Rajappa, S.
Tetrahedron 1981, 37, 1453.
(26) Ihara, M.; Takahashi, M.; Taniguchi, N.; Yasui, K.; Niitsuma, H.;
Fukumoto, K. J. Chem. Soc., Perkin Trans. 1 1991, 525.
(27) Berkowitz, D. B.; Smith, M. K. J. Org. Chem. 1995, 60, 1233.
(31) Dampawan, P.; Zajac, W. W. Synthesis 1983, 545.
(32) Meyers, A. I.; Poindexter, G. S.; Brich, Z. J. Org. Chem. 1978, 43,
892.
J. Org. Chem. Vol. 73, No. 24, 2008 9649

Denmark and Ares
CdN (1590-1605 cm-1) and N-O (1215-1260 and 1120-1180
cm-1) stretches33c also observed in R-keto nitronate salts.33d
6 in 51% yield using the standard procedure. Treatment of 6
with 2 equiv of s-BuLi in THF-HMPA followed by addition
of allyl iodide afforded the allylated products in 48% chemical
yield with a dr of 71.0:29.0 (Scheme 5). The diastereomeric
products could be easily separated by silica gel chromatography,
and both were highly crystalline in the pure state. Importantly,
the diastereoselectivity achieved in this reaction was higher than
any that had been achieved with auxiliaries containing coordi-
nating oxygen atoms. This result confirmed that the monoanion
and dianion alkylations were fundamentally different and
prompted a reexamination of early literature on the stereose-
lective alkylations of chiral metalloenamines that did not contain
coordinating ligands.12a Importantly, in nearly every instance
described in this early work, LDA was employed as the base.
Accordingly, the alkylation of 6 was repeated (Scheme 5) using
1 equiv of LDA (0 °C) and 2 equiv of s-BuLi at -78 °C (the
first equivalent of s-BuLi would be needed to deprotonate the
diisopropylamine formed). We were delighted to discover that
the yield increased to 73%, but more importantly, the dr rose
to 93.5:6.5, more than double anything seen previously.
SCHEME 4
The dianions of these nitro imines were generated using the
optimized protocol from prior studies (2.0 equiv of s-BuLi in
THF containing 5.0 equiv of HMPA at -78 °C).8 Following
dianion formation, the gold-yellow solutions were cooled to -90
°C and treated with various electrophiles. The initial results were
disappointing (Table 1). Phenylalaninol-derived substrate 1
provided the highest level of diastereoselectivity in the alkylation
reaction (70.5:29.5 with butyl iodide), but this thick oily
substrate was operationally difficult to use, and in the alkylations
with methyl and allyl halides, the diastereomeric mixture could
not be separated chromatographically. On the other hand, the
phenylglycinol-derived substrate 2 was crystalline, and both the
methylation and allylation products were separable by chroma-
tography. Unfortunately, diastereoselectivity in these cases was
equally poor. Even the norpseudoephedrine-derived substrate,
where presumably the second stereogenic center could align
itself much more closely to the alkylation sight than in the case
of either 1 or 2, provided low stereoselectivity in the alkylation
step.
SCHEME 5
Fraser34 has reported that stereoselectivity in certain monoan-
ion metalloenamine alkylations can be improved through the
use of magnesium as a counterion. Hence, we treated 3
sequentially with 1 equiv of methyl magnesium bromide, 1 equiv
of s-BuLi, and allyl iodide. The allylated nitro imine product
mixture was obtained in 57% yield, with a dr of only 60:40.
These results revealed substantial differences between the
metalloenamine monoanion alkylation reactions and these nitro
imine dianion alkylations. Whereas Meyers had found that a
coordinating moiety was required for high stereoselectivity for
the monoanions, our results with similar auxiliaries suggested
that lithium coordination was not playing a major role in the
dianion alkylations. To evaluate the requirement for a coordinat-
ing ligand, chiral nitro imine 6 derived from (S)-1-(naphthyl)-
ethylamine was synthesized and evaluated. The commercially
available chiral amine was readily transformed into nitro imine
This result immediately prompted the question as to whether
a similar improvement in yield and selectivity might be obtained
by using LDA with a chiral nitro imine containing a coordinating
group. Thus, phenylglycine-derived substrate 2 was transformed
into its dianion using the new LDA (1 equiv)/s-BuLi (2 equiv)
protocol, followed by alkylation with allyl iodide. The yield
did improve substantially from 60 to 83%, but the diastereo-
selectivity improved only marginally from 67.5:32.5 to 70.5:
29.5. Taken together, these results suggest that the difference
in size between the groups on the chiral auxiliary is playing
the key role in these LDA/s-BuLi alkylation reactions.
1.2. Optimization of Alkylation with 6. To gain further
insight into the role of the amide base in these asymmetric
alkylation reactions, as well as to understand the effect of
TABLE 1. Alkylation of Dianions from Nitro Imines 13-15
entry
substrate
R1
R2
R3
product
yielda, %
drb
1
2
3
4
5
6
1
1
1
2
2
3
CH2OCH3
CH2OCH3
CH2OCH3
CH2OCH3
CH2OCH3
CH(OCH3)Ph
CH2Ph
CH2Ph
CH2Ph
Ph
Ph
CH3
CH3
4a
4b
4c
5a
5c
5c
53
55
66
47
60
60
59.0:41.0c
70.5:29.5
65.0:35.0c
54.0:46.0
67.5:32.5
67.5:32.5
CH3(CH2)3
CH2CHdCH2
CH3
CH2CHdCH2
CH2CHdCH2
a Combined yield of chromatographically homogeneous diastereomers. b Determined by yields of separated isolated products. c Determined by
1
integration in the H NMR spectrum of an unresolved mixture.
9650 J. Org. Chem. Vol. 73, No. 24, 2008

Chiral Nitro Imine and Nitro Hydrazone Dianions
experimental parameters on outcome, a detailed study of the
effect of additives, solvent, amide base, and base stoichiometry
was carried out with substrate 6, using allyl iodide as the test
electrophile. The results are collected in Table 2. The use of
coordinating additives was examined first since we wished to
remove the toxic HMPA from the reaction protocol. Elimination
of HMPA from the reaction mixture led to recovery of
significant amounts of unreacted starting material (about 40%)
along with diminished selectivity (entry 1). The use of warming
cycles, up to -20 °C, in the metalation step failed to lead to
stoichiometric dianion formation. Substitution of TMEDA for
HMPA led to low yields and reduced selectivity (entry 3). We
were gratified to discover that the less toxic dimethylpropylene
urea (DMPU)35 provided a higher yield and better diastereo-
selectivity than HMPA (entry 4). The only drawback associated
with DMPU was that a small amount of starting material usually
remained after the reaction, a situation not generally observed
with HMPA. The use of warming cycles in the metalation did
not eliminate this problem but did reduce diastereoselectivity.
We therefore decided to employ DMPU in the standard
metalation protocol since the small amount of starting material
could be readily removed by chromatography.
(entry 9) suggest that it may be too large to quantitatively
deprotonate 6 in the time given, thus obscuring its role in the
overall dianion structure.
Finally, base stoichiometry was examined. Unexpectedly, the
use of 1.0 equiv of LDA and 1.0 equiv of s-BuLi (entry 10)
provided a good yield and high selectivity. In this case,
diisopropylamine, and not the lithium amide, must be serving
as ligand for the dianion. This implies either that the second
pKa of 6 is lower than that of diisopropylamine or that the
monoanion of 6 is kinetically more acidic than diisopropylamine
and proton transfer is slow. Favorable results in the generation
of the dianion of a different nitro imine system (vide infra)
prompted us to examine the results of using LDA alone to
generate the dianion. Thus, treatment of 6 with 2-3 equiv of
LDA alone at -20 °C, followed by alkylation at -78 °C,
afforded the desired products in 76-83% yield but with
diminished selectivity (entries 11 and 12).
The scope of this asymmetric alkylation was surveyed with
respect to the electrophile, and the results are collected in Table
3. Diastereomeric ratios were determined both by measuring
yields of isolated products and by HPLC analysis of the purified
but unresolved reaction mixtures, the latter providing a direct
comparison of diastereoselectivities. The reaction worked well
for allyl, methallyl, and butyl alkylations (entries 3, 4, and 2),
both in terms of yield and selectivity. Alkylation with methyl
iodide occurred in good yield (entry 1), but the diastereoselec-
tivity was lower than with any other electrophile. To address
this, dimethyl sulfate was employed, but the yield was unac-
ceptably low (<10%). In contrast, isopropyl iodide (entry 5)
gave good selectivity but low yield. In each instance, the
diastereomeric products were crystalline and easily separated
by chromatography with the exception of the isopropyl adduct
7e wherein the low yield precluded isolation of a minor
diastereomer. A pure sample of the minor isomer of the methyl
adduct 7a was obtained by conducting the alkylation with
HMPA instead of DMPU since the minor isomer of 7a coeluted
with the starting material on chromatography. Unexpectedly,
both benzyl halides and benzyl tosylates gave poor results;
benzyl bromide afforded a reasonable dr (95.5:4.5) but low yield
(<40%).
TABLE 2. Optimization of Allylation of 6
amide
amide base, s-BuLi,
entry base equiv equiv additive solvent yielda b
%
drc
,
1
2
3
4
5
6
7
8
9
LDA
LDA
LDA
LDA
LDA
LDA
LDEA
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
0
none
HMPA
TMEDA THF 19 (27)
THF 34
THF 73
(58) 89.5:10.5
93.5:6.5
82.0:18.0
95.0:5.0
74.0:26.0
90.5:9.5
89.0:11.0
91.0:9.0
91.5:8.5
91.5:8.5
87.0:13.0
89.0:11.0
DMPU
DMPU
DMPU
DMPU
DMPU
DMPU
DMPU
DMPU
DMPU
THF 74 (82)
Et2O 15
DME 56 (60)
THF 78 (82)
THF 59 (68)
THF 41 (64)
THF 68 (77)
THF 76
LDCA 1.0
LiTMP 1.0
TABLE 3. Survey of Electrophile in Alkylations of 6
10 LDA
11 LDA
12 LDA
1.0
2.0
3.0
0
THF 83
a Combined yield of chromatographically homogeneous diastereomers.
b Yields in parentheses are based on converted starting material.
c Determined by yields of separated isolated products.
In a brief survey of different solvents, the reduced solubility
of the starting material in ether (entry 5) and DME (entry 6)
led to poorer results. Because the amide base was essential for
good stereoselectivity, a number of amide bases were surveyed
(entries 7-9). Lithium diisopropylamide (LDA) was uniformly
superior to lithium diethylamide (LDEA), lithium dicyclohexy-
lamide (LDCA), and lithium tetramethylpiperidine (LiTMP) in
both yield and selectivity. The diastereoselectivity was not
affected by the choice of base as both LDA and LDCA gave
comparable results (entries 4 and 8). The low yield and large
amount of unreacted starting material seen with the LiTMP
product yielda b, %
drc d, %
,
,
entry
R
1
2
3
4
5
CH3
7a
7b
7c
7d
7e
63(70)
59(62)
74(82)
56(64)
16(34)
80.5:20.5 (80.5:20.5)
98.0:2.0 (99.0:1.0)
95.0:5.0 (90.0:10)
92.5:7.5 (93.0:7.0)
97.5:2.5 (95.0:5.0)
CH3(CH2)3
CH2CHdCH2
CH2C(CH3)dCH2
(CH3)2CH
a Combined yield of chromatographically homogeneous diastereomers.
b Yields in parentheses are based on converted starting material.
c Determined by yields of separated isolated products. d Ratios in
parentheses were determined by HPLC analysis of purified but unresolved
mixtures.
(33) (a) Ostercamp, D. L.; Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1985,
1021. (b) Fetell, A. I.; Feuer, H. J. Org. Chem. 1978, 43, 497. (c) Feuer, H.;
McMillan, R. M. J. Org. Chem. 1979, 44, 3410. (d) Feuer, H.; Sarides, C.; Rao,
C. N. R. Spectrochim. Acta 1963, 19, 431.
1.3. Assignment of Enantiomeric Purity and Absolute
Configuration. The conversion of these alkylated products into
J. Org. Chem. Vol. 73, No. 24, 2008 9651

Denmark and Ares
enantiomerically enriched nitroalkenes, though obviously the
end objective, was also required to determine the absolute
configuration and enantiomeric purity of the products. These
assessments were accomplished in two independent ways using
the major diastereomer from the butyl alkylation 7b and the
methyl alkylation 7a. The results for 7b are shown in Scheme
6. Treatment of an ethanol solution of 7b (major) with 4.0 equiv
each of sodium borohydride and cerium trichloride,8 added in
portions over 8 h, led to formation of nitroalkene 8 in 53%
yield ([R]27D -20.7 (CHCl3, c 3.21)). Basification of the acidic
washings from this reaction, followed by extraction and Kugel-
rohr distillation, allowed for recovery of the (S)-1-(naphthyl)-
ethylamine auxiliary in 52% yield. Conversion of this recovered
amine into its 3,5-dinitrobenzoyl amide, followed by analysis
on a CSP-HPLC column,36 showed that the amine had an er of
99.8:0.2. This is noteworthy since the starting (S)-1-(naphthyl)-
ethylamine used to form the nitro imine substrate 6 had an er
of 95.5:4.5 when similarly analyzed. Apparently, recrystalliza-
tion of 6 enhanced its enantiomeric purity, and the entire cycle
served as a means of purifying the chiral auxiliary!
are shown in Scheme 7. Treatment of 7a (major) with 3 equiv
of sodium borohydride and cerium trichloride8 afforded ni-
troalkene 11 in 61% yield. Similar transformation into ketone
12 was followed by treatment with semicarbazide hydrochloride
to afford the solid semicarbazone 13 ([R]27 -18.8 (EtOH, c
D
0.42)). A comparison of the optical rotation data for 13 to that
in the literature which had been chemically correlated to (R)-
pulegone indicated that the semicarbazone was of the R absolute
configuration, with an ee of 91%.41
SCHEME 7
SCHEME 6
1.4. Preparation and Alkylation of Nitro Imines 16, 18,
and 21. The results from the reactions of nitro imine 6 clearly
indicated that the diastereoselectivity of the dianion alkylations
was related to the size difference between the two non-hydrogen
substituents on the stereogenic center of the amine. Thus, to
improve the stereodifferentiation in dianion alkylations, nitro
imines 16 and 18 were devised in which the groups on the
stereogenic center are sterically quite disparate. These substrates
would be synthesized initially in racemic form to evaluate the
diastereoselectivities in the alkylation. In those instances where
promising results were obtained, the enantiomerically pure
amine would be prepared by resolution or asymmetric synthesis.
The synthesis of racemic 16 is shown in Scheme 8. The
known amine 1542 was prepared in two steps, starting from
9-anthrylnitrile. Addition of methylmagnesium bromide to the
nitrile in a benzene-ether mixture provided methylimine 1443
in 78% yield after silica gel chromatography. Reduction of 14
with sodium cyanoborohydride according to the method of
Ciganek42 afforded the required chiral amine in 66% yield.
Simply stirring the acetate salt of this amine with 1.0 equiv of
2-nitrocyclohexanone in benzene provided the crystalline nitro
imine 16 in 76% yield. Unfortunately, metalation of 16 by either
the usual LDA/s-BuLi protocol or through the use of 3.0 equiv
of LDA, followed by addition of allyl iodide, resulted in the
formation of multiple products and low mass recovery.
Treatment of nitroalkene 8 with L-Selectride,37 followed by
nitronate hydrolysis, afforded 3-butylcyclohexanone 9 in 35%
yield (Scheme 6). This ketone was subsequently treated with
(2R,3R)-butanediol (er >99.7:0.3)38 according to the method
of Corey39 to generate a diastereomeric mixture of dioxolane
ketals 10. Carbon-13 NMR analysis of this mixture (125 MHz)
indicated only the presence of the 3R epimer (C(3) ) 43.16
ppm).39a None of the C(3)S epimer was detected at 44.01 ppm39a
with a S/N ratio of 33.2. Hence the R enantiomer was formed
with an er of >97:3.40
This conclusion was confirmed by studies with the major
diastereomer from the methyl alkylation 7a, and these results
Racemic nitro imine 18 was synthesized by the standard
procedure from pinacolylamine acetate and cyclohexanone
(benzene, reflux) in 56% yield (Scheme 9). Allylation experi-
ments with this compound were also disappointing. Use of the
LDA (1.0 equiv)/s-BuLi (2.0 equiv) protocol provided an
(34) Fraser, R. R.; Akiyama, F.; Banville, J. Tetrahedron Lett. 1979, 2929.
(35) Mukhopadhyay, T.; Seebach, D. HelV. Chim. Acta 1982, 65, 385.
(36) Pirkle, W. H.; Mahler, G.; Hyun, M. H. J. Liquid Chromatogr. 1986, 9,
443.
(37) Mourad, M. S.; Varma, R. S.; Kabalka, G. W. Synthesis 1985, 654.
(38) The enantiomeric purity of this diol was determined in the following
way: Treatment of the diol with 1.0 equiv of 3,5-dinitrophenylisocyanate
(generated in situ from the corresponding acyl azide) produced the mono-
carbamate, which was analyzed by CSP-HPLC; tR (R,R), 8.82 min; tR (S,S),
12.68 min (Pirkle L-naphthylleucine, hexane/isopropanol, 85/15; 0.75 mL/
min, 254 nm).
(40) Seebach has correlated 3-butyl- and 3-methylcyclohexanone (both
dextrorotatory isomers are R) by chiroptical methods, the latter of which has
41
been chemically correlated to (R)-pulegone: Langer, W.; Seebach, D. HelV.
Chim. Acta 1979, 62, 1710.
(41) Allinger, N. L.; Riew, C. K. J. Org. Chem. 1975, 40, 1316.
(42) Ciganek, E. U.S. Patent 4,076,830; 1978.
(43) Martynoff, M.; Chauvin, M.; Grumez, M.; Lefevre, N. Bull. Soc. Chim.
Fr. 1958, 164.
(39) (a) Corey, E. J.; Naef, R.; Hannon, F. J. J. Am. Chem. Soc. 1986, 108,
7114. (b) Hiemstra, H.; Wynberg, H. Tetrahedron Lett. 1977, 2183.
9652 J. Org. Chem. Vol. 73, No. 24, 2008

Chiral Nitro Imine and Nitro Hydrazone Dianions
SCHEME 10
inseparable mixture of diastereomeric products in 44% chemical
yield with a 76.0:24.0 dr. Use of 3.0 equiv of LDA provided
the same chemical yield, but the dr dropped to 67.0:33.0.
To extend the scope of the dianions derived from the
1-(naphthyl)ethylamine auxiliary, substrate 21 was constructed
to assess the feasibility of conducting asymmetric dianion
alkylations in the presence of an aromatic ring (Scheme 10).
The synthesis of 21 began with the nitration of 5-methoxy-2-
tetralone44 using the anionic nitration method of Feuer.45
Treatment of the tetralone with potassium t-butoxide in THF at
-50 °C produced a green anion that was treated with isoamyl
nitrate to afford the solid potassium nitronate salt 20 in 85%
yield. For this transformation to be successful, the potassium
t-butoxide must be of high purity; otherwise, the potassium salt
would not precipitate out of solution. Attempts to isolate the
nitro ketone directly led to very low yields (15%) due to its
instability. However, potassium salt 20 could be transformed
into nitro imine 21 by first treating the acetate salt of the
1-naphthylethylamine with an additional equivalent of acetic
acid in benzene and then adding the solid potassium nitronate
salt. This procedure, although low yielding, did provide suf-
ficient quantities of 21 for trial alkylations.
from the reaction sight can influence the dianion structure and
affect the stereochemical outcome of the reaction.
2. Nitro Hydrazones. The successes achieved in SAMP
hydrazone asymmetric alkylation reactions,17e coupled with the
general and efficient alkylations of achiral nitro hydrazone
dianions described previously,8 made evaluation of the 2-ni-
trocyclohexanone SAMP hydrazone (23) a logical next step
(Scheme 12). Like nitro imines, nitro hydrazones can be easily
doubly deprotonated, and their subsequent alkylations proceed
in generally higher yields than those of nitro imines. The
disadvantage of nitro hydrazones lies in the fact that their
transformations to nitroalkenes require harsher conditions
(sodium borohydride, then acetic anhydride at 120 °C) than nitro
imines.8 We sought initially to evaluate alkylations of the
dianion of 23. If higher yields and selectivities could be achieved
than in the nitro imine cases, we would reevaluate methods of
achieving mild conversion to nitroalkenes.
SCHEME 8
SCHEME 11
SCHEME 9
The synthesis of SAMP nitro hydrazone 23 is shown in
Scheme 12 and begins with N-formyl prolinol.46 Direct ap-
plication of the Enders sequence of methylation, hydrolysis, urea
formation, and Hofmann degradation afforded crude SAMP,
which was treated directly with 2-nitrocyclohexanone/acetic acid
in benzene to provide pure 23 in 15% overall yield from formyl
prolinol.
Results of the alkylation experiments with 21 are shown in
Scheme 11. Use of the established protocol of LDA (1 equiv)
and s-BuLi (2 equiv) led to a complex mixture, with low yields
of the desired products. We hypothesized that the highly reactive
s-BuLi might be reacting with the methoxy-containing aromatic
ring; therefore, we repeated the alkylation using exclusively
LDA as the base (entries 2 and 3). In both instances, much
cleaner reactions ensued, with best results being obtained using
2 equiv of LDA (entry 2). However, in this instance, diaste-
reoselectivity was less than half of that observed in the simple
cyclohexyl system, indicating that structural changes distant
SCHEME 12
(44) Aristoff, P. A.; Johnson, P. D.; Harrison, A. W. J. Am. Chem. Soc. 1985,
107, 7967.
(45) Feuer, H.; Pivawer, P. M. J. Org. Chem. 1966, 31, 3152.
J. Org. Chem. Vol. 73, No. 24, 2008 9653

Denmark and Ares
The results of alkylation experiments with 23 are collected
in Scheme 13. The dianions were formed using the optimized
protocol from the achiral nitro hydrazone series,8 namely, by
forming with s-BuLi in THF/HMPA. Unlike the nitro imines,
however, a reasonable level of stereoselection was achieved in
the alkylation despite the absence of amide bases in the
deprotonation. The two diastereomeric products could not be
separated by chromatography, so drs were determined by
integration of the aci-proton signals in the 1H NMR spectrum.
DMPU did not serve as a suitable replacement additive for
HMPA in the nitro hydrazone alkylations. The fact that a
reasonable level of stereoselectivity was achieved despite the
absence of an amide base suggested that, unlike the nitro imines,
an internal chelation of the lithium ion by the methoxy oxygen,
as suggested by Enders,17a may be of critical importance in the
dianion structure. However, if this is the case, it might be
unreasonable to expect any significant improvement with lithium
amide bases. Indeed, the use of the LDA (1.0 equiv)/s-BuLi
(2.0 equiv) protocol produced only a marginal increase in
selectivity with a concurrent reduction in yield. The use of LDA
alone (3.0 equiv) led to a decrease in both yield and selectivity.
without additional anion stabilizing groups. Because these
structures described herein are secondary nitro compounds, only
R,ꢀ-dianions are possible. The R,ꢀ-dianion of 1-nitrocyclohex-
ane is generated at -90 °C with n-BuLi and t-BuLi in the
presence of HMPA,50a conditions similar to those employed for
nitro imines. Thus, the question of gross dianion structure is
understood in terms of the relative kinetic acidifying effects of
a hydrazone/imine moiety (by loss of Ha) to form vi versus the
nitronate moiety (by loss of Hb) to form v from the monoanion
iv. In all cases studied, the products arose from alkylation of
the dianion v exclusively. However, we cannot rule out partial
formation of vi since the documented instability of such species
may account for lower yields with R-nitro imines.
The detailed structures of the R-nitro imine (2Li+•62-) and
R-nitro hydrazone (2Li+•232-) dianions have not been eluci-
dated, but it is nonetheless instructive at this point to speculate
on reasonable possibilities based on the known structures of
the isolated nitro, hydrazone and imine anions. The solution
structures of metalated hydrazones and imines derived from both
ketones and aldehydes have been extensively investigated by
the groups of Newcomb and Bergbreiter,12a Enders,17aFraser,12b
Knorr,51 Meyers,52 and Collum.53 Direct observation of lithioe-
namines and lithiohydrazones coupled with stereochemical
analysis of kinetic trapping experiments has provided a wealth
of information on the carbon-carbon and carbon-nitrogen
double bond geometries. These studies are primarily concerned
with the origin of the “syn effect”12c and effects of deprotonation
conditions on anion structure in acyclic and macrocyclic
frameworks. In 2Li+•62- and 2Li+•232-, the geometries of the
carbon-carbon and carbon-nitrogen bonds are assured (E
CdC/Z C-N, vi, Scheme 14) by both structural constraints and
the observed syn disposition of the nitrogen substituent in the
products.
SCHEME 13
Discussion
1. Dianion Generation and Structure. The use of poly-
metalated derivatives of functionalized organic molecules has
been extensively developed23 since the pioneering work of
Hauser and Harris.47 Di- and even trianions have been generated
from hydrazones, again following initial studies by the Hauser
group.48 These polyanions, however, have additional stabilizing
groups on the terminal nitrogen (aryl, arylSO2, or arylCO)
wherein NH deprotonation constitutes one of the anions. In
contrast, the dianions described herein possess the general
structure i which have been reported on only two previous
occasions, first by Hauser in the alkylation the phenylimine of
acetylacetophenone (i: Z ) PhCO, R ) Ph).48 In a second, more
related example, Fuchs49 has reported the formation and trapping
of ꢀ-keto ester tosylhydrazone trianions, ii.
SCHEME 14
The more relevant question is the location and role of the
counterion and overall aggregation state. These questions have
been addressed in an ongoing series of elegant studies by Collum
using a combination of 6Li/15N NMR spectroscopy, X-ray
crystallography, molecular weight determinations, and reaction
kinetics.53 For the structure of 2Li+•62-, the closest analogies
are found in the X-ray structures of lithio cyclohexanone
phenylimine (25)53c and lithio cyclohexanone cyclohexylimine
By comparison, there exists considerable precedent in the
chemistry of polymetalated derivatives of nitro aliphatic com-
pounds from the work of Seebach.50 Both, R,R′-dianions and
R,ꢀ-dianions (superenamines) iii have been generated with and
(46) Enders, D.; Fey, P.; Kipphardt, H. Org. Prep. Proced. Int. 1985, 17, 1.
(47) (a) Hauser, C. R.; Harris, T. M. J. Am. Chem. Soc. 1958, 80, 6360. (b)
Harris, T. M.; Harris, C. M. Org. React. 1969, 17, 155. (c) Henoch, F. E.;
Hampton, K. G.; Hauser, C. R. J. Am. Chem. Soc. 1967, 89, 463. (d) Beam,
C. F.; Sandifer, R. M.; Foote, R. S.; Hauser, C. R. Synth. Commun. 1976, 6, 5.
(e) Sandifer, R. M.; Davis, S. E.; Beam, C. F. Synth. Commun. 1976, 6, 339.
(48) Boatman, S.; Hauser, C. R. J. Org. Chem. 1966, 31, 1758.
(50) (a) Brandli, U.; Eyer, M.; Seebach, D. Chem. Ber. 1986, 119, 575. (b)
Seebach, D.; Henning, R.; Mukhopadhyay, T. Chem. Ber. 1982, 115, 1705. (c)
Seebach, D.; Henning, R.; Gonnermann, J. Chem. Ber. 1979, 112, 234. (d)
Seebach, D.; Henning, R.; Lehr, F. Angew. Chem., Int. Ed. Engl. 1978, 17, 458.
(51) (a) Knorr, R.; Low, P. J. Am. Chem. Soc. 1980, 102, 3241. (b) Knorr,
R.; Weiss, A.; Low, P.; Rapple, E. Chem. Ber. 1980, 113, 2462.
(52) Meyers, A. I.; Williams, D. R.; White, S.; Erickson, D. W. J. Am. Chem.
Soc. 1981, 103, 3088.
(49) (a) Bunnell, C. A.; Fuchs, P. L. J. Am. Chem. Soc. 1977, 99, 5184.
9654 J. Org. Chem. Vol. 73, No. 24, 2008

Chiral Nitro Imine and Nitro Hydrazone Dianions
(26),53i which both exist as solvated dimers in the solid state.
These structures represent the limits of η1 and η3 bonding motifs
that are characteristic of these species. Moreover, Collum has
clearly shown that these different modes (and aggregation states)
are strongly solvent dependent; in pure THF, 26 exists as a
highly solvated monomer, and in THF containing HMPA, the
analogous lithio cyclohexanone isopropyl imine (27) also exists
as a solvated monomer.53i
lithio nitronate moiety in such away as to fix the position of
the C-N bond to the 1-naphthylethyl group synplanar to the
enamine C-C double bond. This feature provides the primary
diastereoface differentiation for the approach of the alkylating
agent to the metalloenamine carbon. The difference between
these two structures is the η3 or η1 coordination motifs for the
second lithium atom, which is suggested to be controlled by
the solvating power of the medium. The η3 coordination of the
second lithium atom in vii could be preferred in pure THF, that
is, wherein the enamine C-C double bond can compete with
the solvent for the lithium ion. The η1 coordination of both
lithium atoms in viii would be preferred in a medium containing
stronger donor solvents (HMPA or DMPU). The slight increase
in diastereoselectivity seen with HMPA or DMPU could arise
from removing the η3-coordinated lithium atom from the Re
face of the dianion (assuming that is its preferred location on
steric grounds). Neither of these models explains the enhanced
selectivity afforded by the use of amide bases in combination
with s-BuLi. Presumably, the amide plays an important role as
a ligand for the one of the lithium atoms by influencing
aggregation state or the conformational preferences of the
auxiliary.56
2. Rationalization of Stereoselectivity. Any attempt to
rationalize the stereochemical outcome of the dianion alkylations
must be tempered with the caveat that the solution structures
and transitions structure stoichiometries are not known. That
said, from empirical observation of the stereochemical outcome,
three significant contributions to the alkylation selectivity can
be identified: (1) the difference in steric bulk between the
substituents on the chiral auxiliary, (2) the use of an amide base
in the deprotonation, and (3) the use of a strongly coordinating
additive (HMPA or DMPU). The lack of a requirement for a
coordinating appendage on the auxiliary contradicts the lessons
from Meyers, Koga, and Enders who introduced oxygen and
nitrogen groups to improve the selectivity of metalloenamine
alkylations. Apparently, the conformation-controlling features
of those coordinating groups in the monoanion structures are
not needed in the dianions. This behavior suggests that the
dianions take up defined conformations in structures that are
dependent on the presence of an amide base54 and a coordinating
ligand. Curiously, however, the size of base has little influence
on the selectivity of alkylation, but both HMPA and DMPU
improved the selectivity.
These empirical observations, taken together with the absolute
stereochemical course of the reaction, allow the formulation of
hypothetical structures for the chiral dianions (Scheme 15).
Under the assumption that the dianion 2Li+•62- is monomeric
in solution, two limiting structures are suggested that incorporate
the structural features of individual lithio metalloenamines (vide
supra) and nitronates.55 To explain the high selectivity without
chelating groups on the auxiliary implies that the metalloe-
namines adopts a highly preferred conformation that provides
significant levels of facial differentiation. Structures vii and viii
feature a metalloenamine that is internally coordinated to the
SCHEME 15
Conclusion
The dianions of nitro imines derived from chiral amines could
be generated by the combination an amide base (LDA) and
s-BuLi. These dianions underwent alkylation in moderate yields
and good diastereoselectivities with methyl, butyl, isopropyl,
allyl, and methallyl iodides. DMPU improved the yield and
selectivity of the alkylation. The nitro imine 6 derived from
(S)-1-(naphthyl)ethylamine afforded the R configuration of the
alkylation products. The nitro hydrazone gave poorer results
compared to the nitro imines including inferior yields and
selectivities. Also, HMPA was required with the nitro hydra-
zones, while the safer DMPU could be used with nitro imines.
Although empirical models could rationalize the stereochemical
outcome, the well-established structural variability of simple
metalloenamines discourages a too literal interpretation of these
models.
(53) (a) Collum, D. B.; Kahne, D.; Gut, S. A.; DePue, R. T.; Mohamadi, R.;
Wanat, R. A.; Clardy, J.; VanDuyne, G. J. Am. Chem. Soc. 1984, 106, 4865. (b)
Wanat, R. A.; Collum, D. B. J. Am. Chem. Soc. 1985, 107, 2078. (c) Wanat,
R. A.; Collum, D. B.; VanDuyne, G.; Clardy, J.; DePue, R. T. J. Am. Chem.
Soc. 1986, 108, 3415. (d) Bernstein, M. P.; Romesberg, F. E.; Fuller, D. J.;
Harrison, A. T.; Williard, P. G.; Liu, Q. Y.; Collum, D. B. J. Am. Chem. Soc.
1992, 114, 5100. (e) Bernstein, M. P.; Collum, D. B. J. Am. Chem. Soc. 1993,
115, 789. (f) Bernstein, M. P.; Collum, D. B. J. Am. Chem. Soc. 1993, 115,
8008. (g) Romesberg, F. E.; Collum, D. B. J. Am. Chem. Soc. 1995, 117, 2166.
(h) Liao, S.; Collum, D. B. J. Am. Chem. Soc. 2003, 125, 15114. (i) Zuend,
S. J.; Ramirez, A.; Lobkovsky, E.; Collum, D. B. J. Am. Chem. Soc. 2006, 128,
5839. (j) Ma, Y.; Collum, D. B. J. Am. Chem. Soc. 2007, 129, 14818.
(54) The special role of LDA has already been demonstrated in selective
alkylations of chiral oxazolines. Meyers, A. I.; Knaus, G.; Kamata, K.; Ford,
M. E. J. Am. Chem. Soc. 1976, 98, 567.
Experimental Section
General Methods. See Supporting Information.
General Procedure 1. Preparation of Chiral Nitro Imines.
2-[S-(1-Naphthyl)ethylimino]-aci-1-nitrocyclohexane (6). Freshly
distilled (130-140 °C (ABT), 0.5 mmHg) (S)-(naphthyl)ethylamine
(7.66 g, 44.8 mmol) was dissolved in benzene (150 mL) and cooled
in ice. Acetic acid (2.68 g, 2.55 mL, 44.8 mmol) was added
(55) Klebe, G.; Bohn, K. H.; Marsch, M.; Boche, G. Angew. Chem., Int. Ed.
Engl. 1987, 26, 78.
J. Org. Chem. Vol. 73, No. 24, 2008 9655

Denmark and Ares
dropwise. After cooling in ice 10 min, the salt precipitated from
solution, stopping the stir bar. 2-Nitrocyclohexanone (6.1 g, 42.6
mmol) in benzene (75 mL) was added dropwise over 25 min with
external shaking required in the beginning until the stir bar was
shaken free of the precipitated salts. After addition, the mixture
was heated to reflux with water removal (Dean-Stark trap) for
1 h. The cooled solution was poured into 1% aqueous HOAc (220
mL) and extracted with EtOAc (1 × 200 mL, then 2 × 100 mL).
The combined organic layers were washed with water (1 × 200
mL) and brine (1 × 200 mL), dried (Na2SO4), and evaporated to a
solid that was recrystallized from EtOAc to afford 6.38 g of pale
yellow needles (51%). Data for 6: mp 203-204 °C; [R]30D +1032.9
(c 1.02, CH2Cl2); 1H NMR (300 MHz) δ 11.75 (br d, 1 H),
8.02-7.75 (m, 3 H), 7.60-7.40 (m, 4 H), 5.58 (m, 1 H), 2.67-2.40
(m, 3 H), 2.10-1.90 (m, 1 H), 1.75-1.30 (m, 4 H), 1.74 (d, J )
6.8 Hz, 3 H); 13C NMR (75.5 MHz) δ 158.6, 138.6, 133.9, 129.5,
129.3, 128.3, 126.7, 125.9, 125.8, 122.3, 121.6, 119.3, 49.5, 26.9,
25.6, 23.8, 21.8, 21.2; IR (CCl4) 2946 (w), 1593 (s), 1373 (s); MS
(70 eV) 298 (M+ + 2, 2), 297 (M+ + 1, 11), 297 (M+, 57), 250
(100); Rf 0.28 (hexane/EtOAc, 2:1); HPLC tR ) 23.68 min (Supelco
LC-Si, hexane/EtOAc, 9:1). Anal. Calcd for C18H20N2O2 (MW
296.37): C, 72.95; H, 6.80; N, 9.45. Found: C, 72.83; H, 6.91; N,
9.43.
Nitro Imine Alkylations: 1 Equiv LDA/2 Equiv s-BuLi Protocol.
2-[S-(1-Naphthyl)ethylimino]-3S-2-propenyl-aci-1-nitrocyclohex-
ane (7c). The following alkylation of 6, using 1 equiv of LDA and
2 equiv of s-BuLi, is representative of all nitro imine alkylations
using this base combination: n-BuLi (2.4 M, 0.281 mL, 0.675
mmol) was added to a solution of diisopropylamine (68 mg, 0.094
mL, 0.675 mmol) in THF (4 mL) at 0 °C in a 25 mL, three-necked,
round-bottomed flask equipped with a stir bar, thermometer,
nitrogen inlet, and septum. The solution was stirred at 0 °C for 25
min, and then DMPU (518 mg, 0.489 mL, 4.05 mmol) was added.
After stirring for 5 min at 0 °C, a solution of nitro imine 6 (200
mg, 0.675 mmol) in THF (6 mL) was added rapidly, forming a
reddish solution which faded to orange-yellow after the addition
was complete. The solution was stirred at 0 °C for 30 min, cooled
to -78 °C, and s-BuLi (1.3 M, 1.1 mL, 1.42 mmol) was added
dropwise, forming a reddish solution. The solution was stirred at
-78 °C for 2.5 h, cooled to -90 °C (methanol/liquid nitrogen),
and allyl iodide (295 mg, 0.161 mL, 1.75 mmol) was added neat.
A brownish-yellow solution formed rapidly which was allowed to
warm to 0 °C over 1 h. The clear tan solution was poured into 1%
aqueous HOAc (22 mL) and extracted with EtOAc (3 × 20 mL).
The combined organic layers were washed with water and brine,
dried over Na2SO4, and evaporated to a dark oil, which solidified.
Silica gel chromatography (hexane/EtOAc, 3:1) provided the major
diastereomer (158.35 mg) and minor diastereomer (8.65 mg) of
7c, along with starting material (20.8 mg). Yield: 73%; dr 95.0:
5.0.
Data for 7c (major): mp 158-159 °C; [R]28D +1070.6 (c 0.91,
CH2Cl2); 1H NMR (300 MHz) δ 11.75 (br d, 1 H), 8.01-7.77 (m,
3 H), 7.60-7.33 (m, 4 H), 5.80-5.55 (m, 2 H), 5.20-5.00 (m, 2
H), 2.90-2.30 (m, 5 H), 1.80-1.00 (m, 4 H), 1.72 (d, J ) 6.7 Hz,
3 H, CH3); 13C NMR (75.5 MHz, CDCl3) δ 161.4, 139.4, 134.7,
133.8, 129.5, 129.2, 128.1, 126.7, 125.9, 125.8, 122.1, 121.4, 119.3,
117.8, 49.0, 36.9, 34.6, 24.9, 23.6, 23.4, 16.0; IR (CCl4) 2946 (w),
1593 (s), 1450 (w), 1383 (s), 1229 (w), 1171 (m), 1132 (s), 1074
(m), 1016 (w), 920 (w), 816 (s); MS (70 eV) 338 (M+ + 2, 1),
337 (M+ + 1, 11), 336 (M+, 43), 291 (80), 290 (100), 277 (10),
276 (35), 156 (73), 155 (100), 154 (37), 136 (11); Rf 0.57 (hexane/
EtOAc, 2:1); HPLC tR ) 8.57 min (Supelco LC-Si, hexane/EtOAc,
9:1). Anal. Calcd for C21H24N2O2 (MW 336.44): C, 74.97; H, 7.19;
N, 8.33. Found: C, 74.85; H, 7.28; N, 8.25.
2-[S-(1-Naphthyl)ethylimino]-3R-2-propenyl-aci-1-nitrocyclohex-
ane (7c, minor). Data for 7c (minor): mp 132-135 °C; [R]28
D
1
+683.9 (c 0.33, CH2Cl2); H NMR (300 MHz) δ 12.45 (br d, 1
H), 8.10-7.80 (m, 3 H), 7.61-7.42 (m, 4 H), 5.75 (m, 1 H), 5.35
(m, 1 H), 4.80 (br d, 1 H), 4.50 (br d, 1 H), 3.00-2.72 (m, 2 H),
2.60 (m, 1 H), 1.90-1.40 (m, 6 H), 1.73 (d, J ) 6.6 Hz, 3 H); IR
(CCl4) 2946 (w), 1593 (s), 1381 (s), 1123 (s); MS (70 eV) 336
(M+, 1), 155 (100); Rf 0.38 (hexane/EtOAc, 2:1); HPLC tR ) 17.84
min (Supelco LC-Si, hexane/EtOAc, 9:1). Anal. Calcd for
C21H24N2O2 (MW 336.44): C, 74.97; H, 7.19; N, 8.33. Found: C,
75.06; H, 7.34; N, 8.14.
3R-n-Butyl-1-nitrocyclohexene (8). To the pure major diaste-
reomer of 7d (525 mg, 1.49 mmol) in ethanol (23 mL) was added
cerium trichloride heptahydrate (1.11 g, 2.98 mmol), followed by
sodium borohydride (113 mg, 2.98 mmol). After 1.5 h, an additional
0.5 equiv of each reagent was added. This was repeated three times
until, after 8 h, a total of 4 equiv each of NaBH4/CeCl3 had been
added. Acetone (5 mL) was added, and the milky suspension was
poured into a separatory funnel containing 1 N HCl (54 mL) and
hexane (54 mL). The funnel was shaken, and the layers were
separated. The aqueous layer was further extracted with hexane (2
× 50 mL) and saved for recovery of the auxiliary. The combined
organic layers were washed with water and brine, backwashed with
hexane (20 mL), dried (Na2SO4), and evaporated to a residue which
was purified by flash chromatography (hexane/EtOAc, 15:1) to
afford 143 mg of a green oil (53%). Data for 8: bp 100 °C (0.1
1
mmHg); [R]29 -20.7 (c 3.21, CHCl3); H NMR (300 MHz) δ
D
7.22 (s, 1 H), 2.65-2.36 (m, 3 H), 1.95-1.76 (m, 2 H), 1.70-1.16
(m, 8 H), 0.91 (t, 3 H, J ) 7.0 Hz); IR (CCl4) 2932 (s), 2861 (m),
1522 (s), 1337 (s); Rf 0.61 (hexane/EtOAc, 2:1).
Acknowledgment. We gratefully acknowledge the financial
support provided for this project by the National Institutes of
Health (RO1 GM 30938) and the (former) Upjohn Company.
Supporting Information Available: General experimental
1
details and H NMR spectra of all characterized compounds.
(56) A reviewer has suggested that a Li-π-naphthalene coordination might
be responsible for the preferred conformation of the stereocontrolling group. In
a simple calculation, it was found that the cyclohexene ring in the dianion is not
planar and pseudoaxial hydrogens may also contribute to the direction of
electrophilic attack. We thank the reviewer for this suggestion.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO801790R
9656 J. Org. Chem. Vol. 73, No. 24, 2008