A simple, nontoxic iron system for the allylation of zinc enolates

Article abstract of DOI:10.1021/ol200059u

Diiron nonacarbonyl in combinationwith triphenylphosphine has been identified as a low-cost and environmentally benign catalyst system for the allylation of zinc enolates generated in situ from copper-catalyzed asymmetric conjugate addition reactions. The

Full text of DOI:10.1021/ol200059u

ORGANIC
LETTERS
2011
Vol. 13, No. 8
1904–1907
A Simple, Nontoxic Iron System for the
Allylation of Zinc Enolates
Gopala K. Jarugumilli and Silas P. Cook*
Department of Chemistry, Indiana University, 800 East Kirkwood Avenue, Bloomington,
Indiana 47405-7102, United States
sicook@indiana.edu
Received January 8, 2011
ABSTRACT
Diiron nonacarbonyl in combination with triphenylphosphine has been identified as a low-cost and environmentally benign catalyst system for the
allylation of zinc enolates generated in situ from copper-catalyzed asymmetric conjugate addition reactions. The catalyst system provides the
allylated product in modest to good yields at room temperature with unprecedented diastereoselectivity in cyclic enone systems. While
triphenylphosphine was uniquely effective among the investigated ligands, the exact nature of the active catalytic species remains unknown.
Allylic nucleofuges typify the more reactive alkyl electro-
philes available to organic chemists today. However, while
allylic alkylation represents one of the more facile carbon-
carbon bond-forming processes, additional control of reac-
tivity and selectivity has been realized through the use of
transition-metal catalysis.1 Although allylic alkylation reac-
tions catalyzed by palladium have been the most thoroughly
investigated, various procedures involving ruthenium,2
iridium,3 molybdenum,4 tungsten,5 copper,6 nickel,7 and
Figure 1. Natural products recently synthesized from
intermediates 5a and 5b (the substructures of which are
iron8 have also been reported. Unfortunately, these pro-
highlighted in red).9
cedures suffer from the use of costly transition metals,2-4
exotic and/or expensive ligands,4,6,7 elevated reaction tem-
peratures,3,5,8 or multistep catalyst preparation.8
New routes to enantiomerically pure carbocyclic ketones
need continued innovation since these compounds function as
important intermediates in the synthesis of biologically rele-
vant molecules. Specifically, building blocks such as 5-7 (see
abstract) have been used in the synthesis of potential drug
candidates and numerous natural products (Figure 1).9 Dur-
ing a recent foray into the synthesis of another natural
product, we sought an inexpensive, simple transition-metal
complex to catalyze allylic alkylations. Early work with iron
catalysts utilized pyrophoric Na[Fe(CO)3(NO)] thatwaslater
(1) Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921.
(2) (a) Morisaki, Y.; Kondo, T.; Mitsudo, T. A. Organometallics
1999, 18, 4742. (b) Kondo, T.; Morisaki, Y.; Uenoyama, S.; Wada, K.;
Mitsudo, T. J. Am. Chem. Soc. 1999, 121, 8657. (c) Trost, B. M.; Fraisse,
P. L.; Ball, Z. T. Angew. Chem., Int. Ed. 2002, 41, 1059.
(3) (a) Takeuchi, R.; Shiga, N. Org. Lett. 1999, 1, 265. (b) Bartels, B.;
Helmchen, G. Chem. Commun. 1999, 741.
(4) Trost, B. M.; Dogra, K. J. Am. Chem. Soc. 2002, 124, 7256.
(5) (a) Trost, B. M.; Hung, M.-H. J. Am. Chem. Soc. 1983, 105, 7757.
(b) Trost, B. M.; Hung, M.-H. J. Am. Chem. Soc. 1987, 109, 2176. (c)
Lloyd-Jones, G. C.; Pfaltz, A. Angew. Chem., Int. Ed. 1995, 34, 462.
(6) Persson, E. S. M.; Vanklaveren, M.; Grove, D. M.; Backvall, J. E.;
Vankoten, G. Chem.;Eur. J. 1995, 1, 351.
(9) (a) Schmitt, M. H.; Blechert, S. Angew. Chem., Int. Ed. 1997, 36,
1474. (b) Akashi, M.; Sato, Y.; Mori, M. J. Org. Chem. 2001, 66, 7873.
(c) Dijk, E. W.; Panella, L.; Pinho, P.; Naasz, R.; Meetsma, A.;
Minnaard, A. J.; Feringa, B. L. Tetrahedron 2004, 60, 9687. (d) Bettati,
M.; Churcher, I.; Doughty, V. A.; Harrison, T.; Nizi, E.; Smith, A.
World Patent 2005108362; June 5, 2005.
(7) Didiuk, M. T.; Morken, J. P.; Hoveyda, A. H. J. Am. Chem. Soc.
1995, 117, 7273.
(8) (a) Roustan, J. L.; Merour, J. Y.; Houlihan, F. Tetrahedron Lett.
1979, 39, 3721. (b) Xu, Y.; Zhou, B. J. Org. Chem. 1987, 52, 974. (c)
Zhou, B.; Xu, Y. J. Org. Chem. 1988, 53, 4419.
r
10.1021/ol200059u
Published on Web 03/10/2011
2011 American Chemical Society

refined to the air-stable, soluable, and nonpyrophoric nBu4N-
[Fe(CO)3(NO)] by Zhou and co-workers.8 Further refine-
ments came when Plietker et al. introduced phosphines and
N-heterocyclic carbene (NHC) ligands to improve regioselec-
tivity and catalytic activity while allowing the use of N- and
S-nucleophiles.10 This system was extended to decarboxyla-
tive etherification reactions by Tunge and co-workers.11 While
these important studies demonstrate the versatility of the
nBu4N[Fe(CO)3(NO)] system, the requisite multistep catalyst
preparation and elevated reaction temperatures were unsui-
table for our needs. To overcome these issues, we began
evaluating potential catalysts in the context of zinc enolates,
which generally behave as relatively weak nucleophiles.12 To
that end, we were intrigued by reports detailing the use of
diiron nonacarbonyl (Fe2(CO)9) in the allylic alkylation of
malonate derivatives.13
(Table 1). In accord with previous observations,9c,14
tetrakis(triphenylphosphine)-palladium provided the de-
sired product in excellent yield (90%) and decent dr (9:1
trans/cis) (Table 1, entry 3). Employing the iron complex
nBu4N[Fe(CO)3(NO)] resulted in a modest 25% yield at
room temperature (Table 1, entry 4) and 50% yield at
60 °C with only a 3:1 dr (Table 1, entry 5). Interestingly, the
nBu4[Fe(CO)3(NO)]-catalyzed reaction was inhibited in
the presence of triphenylphosphine. Subjecting zinc eno-
late 4a to allyl acetate in the presence of Fe2(CO)9 pro-
duced 15% of desired product 5a after 4 days at room
temperature (Table 1, entry 6). Furthermore, the same
reaction conducted at elevated temperatures lacked any
detectable product 5a, likely due to product decomposi-
tion.15 When the allylation of zinc enolate 4a was pro-
moted with Fe2(CO)9 (10 mol %) in the presence of
triphenylphosphine (10 mol %), however, a remarkable
acceleration in reaction rate was accompanied with an
80% yield and 41:1 dr (Table 1, entry 7). We believe this to
be the highest trans/cis ratio reported for a conjugate
addition/allylation sequence reported to date. Lowering
the catalyst loading to 2.5 mol % Fe2(CO)9 resulted in
slightly diminished yield and dr (Table 1, entry 9).
Table 1. Allylation with Palladium and Iron Catalystsa
Table 2. Ligand Influence on Catalyst Systema
a NR = no reaction; ND = not determined.
To begin our studies, a small panel of transition-metal
complexes were evaluated for their ability to effect the
allylation of zinc enolates produced from the asymmetric
copper-catalyzed conjugate addition of dimethylzinc
a NR = no reaction; ND = not determined; NA = not applicable;
Mes = mesityl group. b 2 M solution in THF.
(10) (a) Dieskau, A.; Tabassam, M.; Plietker, B. Angew. Chem., Int.
Ed. 2009, 48, 7251. (b) Plietker, B. Angew. Chem., Int. Ed. 2006, 45, 6053.
(c) Plietker, B. Angew. Chem., Int. Ed. 2006, 45, 1469. (d) Plietker, B.;
Andre, D.; Kartin, M.; Anja, J. Angew. Chem., Int. Ed. 2008, 47, 198. (e)
Jegelka, M.; Plietker, B. Org. Lett. 2009, 11, 3462.
(11) Trivedi, R.; Tunge, J. A. Org. Lett. 2009, 11, 5650.
(12) (a) Guo, H. C.; Ma, J. A. Angew. Chem., Int. Ed. 2006, 45, 354.
(b) Rathgeb, X.; March, S.; Alexakis, A. J. Org. Chem. 2006, 71, 5737.
(13) (a) Ladoulis, S. J.; Nicholas, K. M. J. Organomet. Chem. 1985,
285, C13. (b) Silverman, G. S.; Strickland, S.; Nicholas, K. M. Organo-
metallics 1986, 5, 2117. (c) Akermark, B.; Sjogren, M. P. T. Adv. Synth.
Catal. 2007, 349, 2641.
(14) Naasz, R.; Arnold, L. A.; Minnaard, A. J.; Feringa, B. L. Chem.
Commun. 2001, 735.
(15) Extended reaction times under these conditions routinely led to
lower yields of 5a.
Org. Lett., Vol. 13, No. 8, 2011
1905

A cursory ligand survey revealed triphenylphosphine to
be unusually effective at enhancing the reactivity (Table 2,
entry 5). Even relatively minor changes to tri-p-tolylphos-
phine or tri-p-methoxyphenylphosphine resulted in longer
reaction times, lower yields, and compromised dr’s (Table 2,
entries 6 and 7). Substantially more electron-donating
phosphines such as tri-n-butyl- or tricyclohexylphosphine
or electron-deficient phosphines such as tri(2-furyl)phos-
phine did not support efficient catalysis (Table 2, entries 3, 4,
and 8). In contrast to previous work with Fe2(CO)9 and
malonates, amine-based ligands demonstrated a relatively
small effect on the reaction (Table 2, entries 1 and 2).13c
The unusually high dr was consistent throughout the
course of the reaction as verified by samples obtained at
low reaction conversion. Additionally, reaction times ex-
tended well beyond full conversion led to an erosion in the
trans/cis ratio. The thermodynamic preference for dr appears
to be about 2.4:1 trans/cis as determined by base-catalyzed
epimerization studies. Furthermore, a variety of allylic
electrophiles such as allylic iodides, bromides, sulfonyl esters,
and different carbonates all proved inferior to allyl acetate.16
We further probed the scope of this reaction with
substrates such as cyclopentenone (3) and cycloheptenone
(2). In the context of cyclohexenone (1), dimethylzinc
resulted in a higher yield and dr when compared to
diethylzinc (Table 3, entries 1 and 2). Interestingly, while
(Table 3, entry 6).17 The use of diethylzinc, however, was
successful and producedthe desired product 7bin a modest
50% yield and an 18:1 dr. Although the yields are not
uniformly high, the diastereoselectivity of the overall
reaction represents a significant improvement over current
methods. While for some substrates the low yields could be
attributed to incomplete conjugate addition (Table 3, en-
tries 4 and 6), in general longer than necessary reaction
times led to product decomposition and lower yields.
The extension of this methodology to acyclic systems
met with limited success. While the reaction of zinc en-
olates generated from acyclic enones provided products in
modest yields, the diastereoselectivity was significantly
diminished (Scheme 1). Additionally, the move to more
substituted allyl acetates proved synthetically unviable. A
reexamination of the iron source and additives led to only
limited yields of the desired product.
Scheme 1. Allylation of Zinc Enolates Generated from the
Asymmetric Conjugate Addtion to Acyclic Enones
Table 3. Allylation of Various Enones with Iron Catalysta
The narrow requirement for triphenylphosphine led to
the preparation of several potential active catalyst species
(Table 4). Interestingly, no combination of Fe(CO)5 and
Table 4. Examination of Various Iron Sources for Their Ability
To Catalyze the Allylation of Zinc Enolatesa
a NR = no reaction; ND = not determined. b Isolated yields. c 15 h
for conjugate addition.
a NR = no reaction; ND = not determined.
dimethylzinc resulted in somewhat diminished yields when
compared to diethylzinc with cycloheptenone, the dr was
well over 99:1 (Table 3, entries 3 and 4). In line with
previous results, the copper-catalyzed conjugate addtion
of dimethylzinc into cyclopentenone was problematic
PPh3 could reproduce the rate and yield observed for
Fe2(CO)9 and PPh3 (Table 4, entries 3 to 7). Furthermore,
phosphine loadings of less than a 1:2 molar ratio of
(16) See Supporting Information for a table of representative allylic
electrophiles.
(17) (a) Degrado, S. J.; Mizutani, H.; Hoveyda, A. H. J. Am. Chem.
Soc. 2001, 123, 755 and references therein.
1906
Org. Lett., Vol. 13, No. 8, 2011

phosphine/Fe produced low (<15%) yields while molar
ratios up to 2:1 had little effect on the yield or dr.18 These
data suggest a heterogeneity among active iron species
produced from Fe2(CO)9 and PPh3. To compare the
activity of potential active iron catalysts, both Fe(PPh3)-
(CO)4 and Fe(PPh3)2(CO)3 were prepared according to
known literature procedures.19 Investigation of these com-
plexes in the allylation of zinc enolates revealed them to
be nearly inactive (Table 4, entries 1 and 2). Additionally,
photolysis of Fe2(CO)9 to generate Fe(CO)5 and Fe(CO)4
failed to produce a viable catalyst for the reaction. Cur-
rently, the specie(s) generated from Fe2(CO)9 and PPh3
responsible for the observed activity remain elusive.
nontoxic catalytic system for the allylation of zinc enolates
in reasonable yields and high trans to cis ratios. The
efficiency of the system allows the reactions to be run
under mild conditions at room temperature. While the
exact nature of the active catalytic specie(s) remains un-
known, the operational simplicity and affordability of this
system should result in the widespread adoption of this
technology. Current work is aimed at elucidating the active
catalytic specie(s) and expanding the reaction scope of
both the nucleophile and electrophile.
Acknowledgment. Financial support for this work was
provided by Indiana University.
In summary, the equimolar combination of diiron
nonacarbonyl and triphenylphosphine produces a cheap,
Supporting Information Available. Experimental pro-
cedures describing the synthesis of compounds 5-7 and 9,
as well as the characterization data, are provided. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(18) See Supporting Information for a table of PPh3 loading with
Fe2(CO)9.
(19) (a) Brunet, J. J.; Kindela, F. B.; Neibecker, D. J. Organomet.
Chem. 1989, 368, 209. (b) Cotton, F. A.; Parish, R. V. J. Chem. Soc. 1960,
1440.
Org. Lett., Vol. 13, No. 8, 2011
1907