Asymmetric allylation, crotylation, and cinnamylation of N-heteroaryl hydrazones

Article abstract of DOI:10.1021/ol9026864

(Chemical Equation Presented) A new class of N-heteroaryl hydrazones has been developed as an alternative to N-acylhydrazones and 2-amlnophenol-derlved lmlnes In asymmetric allylatlon, crotylatlon, and clnnamylatlon reactions with chlral allylchlorosllanes. The hydrazones are readily and Inexpensively prepared, perform well In the allylatlon chemistry, and more Importantly, the product hydrazldes may be smoothly reduced by Pd(OH)2catalyzed hydrogenation to reveal the corresponding amine products.

Full text of DOI:10.1021/ol9026864

ORGANIC
LETTERS
2010
Vol. 12, No. 4
688-691
Asymmetric Allylation, Crotylation, and
Cinnamylation of N-Heteroaryl
Hydrazones
Miriam Inbar Feske, Alexander Buitrago Santanilla, and James L. Leighton*
Department of Chemistry, Columbia UniVersity, New York, New York 10027
leighton@chem.columbia.edu
Received November 21, 2009
ABSTRACT
A new class of N-heteroaryl hydrazones has been developed as an alternative to N-acylhydrazones and 2-aminophenol-derived imines in
asymmetric allylation, crotylation, and cinnamylation reactions with chiral allylchlorosilanes. The hydrazones are readily and inexpensively
prepared, perform well in the allylation chemistry, and more importantly, the product hydrazides may be smoothly reduced by Pd(OH)2-
catalyzed hydrogenation to reveal the corresponding amine products.
The asymmetric allylation of imines is one of the more
versatile and general methods for the synthesis of nonracemic
chiral carbinamines,1 as this approach can produce both
secondary and tertiary carbinamines, as well as a second
stereocenter in the allylic position of the product through
the use of substituted allyl fragments. Although we have
documented the utility of allylsilane 1a in reactions with
aldehyde- and ketone-derived acylhydrazones to provide
chiral carbinamine products with good to excellent levels of
enantioselectivity (Scheme 1),2 two important limitations in
scope were uncovered during the course of those studies:
(1) most aliphatic aldehyde-derived hydrazones reacted with
poor enantioselectivity and (2) acylhydrazones;with only
a few exceptions;did not perform well with substituted
allylsilanes 1b and 1c. While both limitations were success-
fully addressed by the use of 2-aminophenol-derived imines
(Scheme 1),3 this directing/activating group was not without
its own limitation: the oxidative cleavage of the 2-hydrox-
yphenyl group (typically with ceric ammonium nitrate (CAN)
or PhI(OAc)2) from the homoallylamine products can be an
Scheme 1
(1) For two recent reviews, see: (a) Ding, H.; Friestad, G. K. Synthesis
2005, 2815. (b) Friestad, G. K.; Mathies, A. K. Tetrahedron 2007, 63, 2541.
(2) (a) Berger, R.; Rabbat, P. M. A.; Leighton, J. L. J. Am. Chem. Soc.
2003, 125, 9596. (b) Berger, R.; Duff, K.; Leighton, J. L. J. Am. Chem.
Soc. 2004, 126, 5686.
10.1021/ol9026864  2010 American Chemical Society
Published on Web 01/19/2010

inefficient transformation. Indeed, it is often necessary to
methylate the phenol and reduce the alkene first, in order to
obtain acceptable yields of the free amine.4 Our ongoing
search for an effective and general directing/activating group
that may be more simply, reliably, and efficiently removed
has uncovered a new class of N-heteroarylhydrazones, and
we report the details herein.
1b or 1c. The corresponding benzothiazole and benzimida-
zole hydrazones were examined as well, but these proved
less effective than the benzoxazole hydrazone.
Scheme 3
Scheme 2
Our investigations focused next on heteroarylhydrazones
typified by N-(2-pyridyl)hydrazone 6 (Scheme 3). In direct
analogy to complex A, we envisioned that treatment of 6
with 1b would lead to the formation of complex B, and we
hoped that a system could be identified that would perform
well in this crotylation process. While 6a (R ) CH2CH2Ph)
did react smoothly with (R,R)-1b, the product was produced
with poor enantioselectivity (<40% ee). Substituted pyridines
and quinolines were examined, and it was eventually found
that 6-chloro-2-quinoline derived hydrazone 7 provided good
results. Indeed, 7 proved more reactive than 3, reacting
smoothly with (R,R)-1b at room temperature to give hy-
drazide 8 as a single diastereomer in 67% yield and 94%
ee. As expected, hydrogenation under the conditions de-
scribed for the reduction of 4 was effective for 8 as well
giving 9 in 98% yield. Alternatively, reduction with SmI2
(the method typically employed with the acylhydrazide
products8) could be carried out following benzoylation of
the hydrazide to give 10 in 74% overall yield.
A brief survey of substrate scope was conducted and the
results are summarized in Table 1. In every case the products
(8, 11-14) were produced as a single diastereomer, and with
good to excellent enantioselectivity (81-94% ee). The
relatively large range in ee values observed is perhaps
surprising in light of the similarity of the substrates, and is
suggestive of one or more isomeric complexes (and ulti-
mately transition states) related to B being relatively close
in energy. Nevertheless the method does provide a useful
alternative to the 2-aminophenol-derived imine crotylation
reactions discussed above for aliphatic aldehyde-derived
imines.
N-Heteroarylhydrazones5 such as 2 (X ) O, S, N-R′)
were considered because it was envisioned that they would
react with our allylchlorosilanes (in analogy to the acylhy-
drazones and N-(2-hydroxyphenyl)imines) to produce a
complex such as A (Scheme 2). Interestingly, this would
require a loss of aromaticity, but as these are heteroarenes,
this was predicted to be not necessarily prohibitive. Further,
there was reason to be optimistic that the N-N bond in the
product N-arylhydrazides would be susceptible to metal-
catalyzed hydrogenation.6 Benzoxazole-derived hydrazone
3 was prepared7 and upon treatment with (R,R)-1a in
refluxing CH2Cl2 gave hydrazide 4 in 64% yield and 91%
ee. Hydrogenation of the N-N bond in 4 did indeed prove
feasible with Pd(OH)2 as the catalyst, and this procedure
produced amine 5 in 90% yield. While these two reactions
represented a potential solution to the allylation of aliphatic
aldehyde-derived imines, 3 did not perform well with either
(3) (a) Rabbat, P. M. A.; Valdez, S. C.; Leighton, J. L. Org. Lett. 2006,
8, 6119. (b) Huber, J. D.; Leighton, J. L. J. Am. Chem. Soc. 2007, 129,
14552. (c) Huber, J. D.; Perl, N. R.; Leighton, J. L. Angew. Chem., Int. Ed.
2008, 47, 3037.
(4) (a) Porter, J. R.; Travers, J. F.; Hoveyda, A. H.; Snapper, M. L.
J. Am. Chem. Soc. 2001, 123, 10409. (b) Sugiura, M.; Bobvieux, F.;
Kobayashi, S. Synlett 2003, 1749.
(5) We could find no examples of the use of such hydrazones in
nucleophilic addition reactions. There are reports on the use of such
hydrazones as ligands in coordination complexes. See, for example: (a)
Tupolova, Y. P.; Lukov, V. V.; Kogan, V. A.; Popov, L. D. Russ. J. Coord.
Chem. 2007, 33, 301. (b) Tang, J.; Costa, J. S.; Pevec, A.; Kozlevcar, B.;
Massera, C.; Roubeau, O.; Mutikainen, I.; Turpeinen, U.; Gamez, P.;
Reedijk, J. Cryst. Growth Des. 2008, 8, 1005.
(6) (a) Hearn, M. J.; Chung, E. S. Synth. Commun. 1980, 10, 253. (b)
Toti, A.; Frediani, P.; Salvini, A.; Rosi, L.; Giolli, C. J. Organomet. Chem.
2005, 690, 3641.
It was of interest as well to pursue an alternative for our
recently reported cinnamylation protocol for 2-aminophenol-
(7) Preparation of hydrazone 3 and all other hydrazones described herein
was straightforward: the commercially available 2-chloroheteroarene is
treated with hydrazine, and the resulting N-heteroarylhydraine is condensed
with the aldehyde. See the Supporting Information for details.
(8) (a) Burk, M. J.; Martinez, J. P.; Feaster, J. E.; Cosford, N.
Tetrahedron 1994, 50, 4399. (b) Ding, H.; Friestad, G. K. Org. Lett. 2004,
6, 637.
Org. Lett., Vol. 12, No. 4, 2010
689

Scheme 4
Table 1. Enantioselective Crotylation of Heteroarylhydrazones
Several examples of this one-pot cross-metathesis/cin-
namylation reaction were demonstrated and the results are
compiled in Table 2. The reaction is tolerant of a range of
vinylarenes (entries 1-5) as well as vinylcyclohexane (entry
6). The reasons for the drop in ee observed in entry 5 and
the drop in dr observed in entry 6 are not readily apparent,
but may again be suggestive of one or more isomeric
transition states that are relatively close in energy. As with
the crotylation reactions described above, this methodology
provides a viable alternative to the 2-aminophenol-derived
imine cinnamylation reactions.
derived imines.3b Unfortunately, none of the substituted
pyridine- and quinoline-derived hydrazones examined in the
development of the crotylation reaction provided acceptable
levels of enantioslectivity in reactions with (R,R)-1c. Our
survey of heterocyclic motifs thus expanded to include
pyrazines, pyrimidines, and pyridazines. While several such
heteroaryl hydrazones provided promising leads, it was
difficult to identify a system that provided both smooth
conversion/good yields and high levels of enantioselectivity.
After extensive experimentation, the 6-chloro-3-pyridazine-
derived hydrazone 15 was found to provide good results upon
treatment with cinnamylsilane (R,R)-1c, giving 16 in 82%
yield (12:1 dr) and 85% ee (Scheme 4). With this promising
result in hand, we next investigated the performance of the
pyridazine activator in the one-pot cross-metathesis/cin-
namylation procedure.3c While this did work, what was
isolated was a ∼2:1 mixture of 16 and the oxidized product
17. This oxidation was also observed to a lesser extent in
the reaction of 15 with 1c, but it proved possible to suppress
by rigorous exclusion of oxygen. Similar attempts to suppress
the oxidation in the cross-metathesis/cinnamylation procedure
were unsuccessful, and we therefore decided to develop a
procedure in which the crude reaction product was oxidized
with H2O2, allowing the more convenient isolation of one
product instead of two. Upon optimization, this procedure
allowed the one-pot synthesis of 17 (from 1a, styrene, and
15) in 62% yield (>20:1 dr) and 90% ee.
Table 2. Tandem Cross-Metathesis/Cinnamylation Reactions
entry
R
product
yield (%)
dr
ee (%)
1
2a
3b
4a
5
Ph
17
18
19
20
21
22
62
71
68
62
66
72
>20:1
>20:1
>20:1
>20:1
>20:1
5:1
90
90
86
86
82
91
o-CH1-C6H4
p-CH3-C6H4
m-CF3-C6H4
p-F-C6H4
6b
c-Hex
a This reaction was run in CHCl3. b Part two of this reaction (the
cinnamylation reaction) was conducted at 23 °C.
The NdN bond in the oxidized cinnamylation products
may be hydrogenated as above (with Pd(OH)2 as the
catalyst), but, perhaps surprisingly, the chloride on the
pyridazine ring is reduced only very slowly. Because of this
the desired amine product is produced in the presence of
6-chloropyridazin-3-amine, and these two compounds react
690
Org. Lett., Vol. 12, No. 4, 2010

to give the substitution product. This problem is avoided by
a two-stage process wherein 17 is first subjected to hydro-
genation with Pd/C as the catalyst (this results in reduction
of the chloride, the alkene, and reduction of RsNdNsR′
to RsNHsNHsR′), and then to hydrogenation with
Pd(OH)2 as the catalyst (Scheme 5). This procedure allows
the reduction of 17 to give amine 23 in 89% yield.
Scheme 6
Scheme 5
displacement of chloride from silicon, and protonation of
the pseudoephedrine amino group with the liberated HCl.
The stereochemical details are more difficult to discuss with
any authority and we note that the models for asymmetric
induction implied by structures A and B are speculative.
A new class of N-heteroarylhydrazones has been developed
for enantioselective allylation, crotylation, and cinnamylation
reactions. The hydrazones are trivial to prepare, and the N-N
bonds in the product hydrazides are easily cleaved by
Pd(OH)2-catalyzed hydrogenation. This discovery enhances
the practicality and applicability of these synthetically
powerful reactions, and these new hydrazones may find utility
in other nucleophilic addition reactions.
Finally, we have proposed that these hydrazones react with
the chlorosilanes in direct analogy to the reactions of
acylhydrazones with the chlorosilanes,2b and we have secured
spectroscopic evidence for this notion (Scheme 6). Specif-
ically, when hydrazone 15 is mixed with phenylsilane 24, a
1
single new species (assigned as C) is observed by H and
28Si NMR spectroscopy that grows in over the course of
several hours (consistent with the idea we have previously
advanced that the fact that the silanes are ∼2:1 mixtures of
diastereomers is inconsequential;they react by way of a
common intermediate/complex). The chemical shift of the
peak in the 28Si spectrum (-92 ppm) is consistent with the
formation of a five-coordinate complex, and importantly,
the N-CH3 peak is a doublet in the proton spectrum,
consistent with the loss of a proton from the hydrazone,
Acknowledgment. This work was supported by a grant
(CHE-08-09659) from the National Science Foundation. MIF
is a National Science Foundation Graduate Research Fellow.
Supporting Information Available: Experimental pro-
cedures, characterization data, and stereochemical proofs.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL9026864
Org. Lett., Vol. 12, No. 4, 2010
691