Organocatalytic dimerization of ketoketenes

Full text of DOI:10.1021/jo8024785

drug molecules such as the serotonin 1A receptor antagonist
LY426965, and (-)-phenserine, a candidate for the treatment
of Alzheimer’s disease.7,8 Ketoketene dimers are potential
enzyme inhibitors, and such investigations of biological activity
are currently in progress at our laboratory.4
Some time ago, Elam and, shortly after, Bentrude showed
that dimethylketene could be dimerized using trialkyl phosphites
as nucleophilic catalysts.5 However, following these examples
of catalytic ketoketene dimerization, no general system and no
study of stereoselectivity has emerged in the intervening years.
More recently, Calter showed that a nucleophilic catalyst system
(TMS-quinine or TMS-quinidine) could catalyze the dimeriza-
tion of alkyl-substituted aldoketenes with high enantioselectiv-
ity.3
Organocatalytic Dimerization of Ketoketenes
Ahmad A. Ibrahim, Gero D. Harzmann, and
Nessan J. Kerrigan*
Department of Chemistry, Oakland UniVersity, 2200 North
Squirrel Road, Rochester, Michigan 48309-4477
kerrigan@oakland.edu
ReceiVed NoVember 5, 2008
SCHEME 1. Nucleophile-Catalyzed Dimerization of
Methylphenylketene
A general method for the catalytic dimerization of ke-
toketenes is described. Tri-n-butylphosphine was found to
be the optimal organocatalyst for the racemic reaction. When
lithium iodide was used as an additive, the reaction was rendered
selective for dimer formation (dimer/trimer g16:1). Ring-
opening reactions of the ketoketene dimers as well as
preliminary studies toward the development of an asymmetric
variant are also reported.
We initiated a study of methylphenylketene dimerization by
investigating the alkaloid catalytic systems previously described
by other groups for various [2 + 2]-cycloadditions involving
ketenes.3,9,10 However, only low conversion (<10%) to ke-
toketene dimer was observed with this class of nucleophilic
catalysts (Scheme 1). Inspired by the precedent of Elam’s work,
we proceeded to investigate tricoordinate phosphorus catalysts
for their activity toward methylphenylketene. Tri-n-butylphos-
phine was found to be the most effective of all the catalysts
surveyed (Table 1, entry 1). At this point, the reaction afforded
a mixture of products including the desired ketoketene dimer
product 4a (58% conversion) as well as ketoketene trimer 5a
(42% conversion). Interestingly, no formation of trimer as a
side product had been encountered by Calter’s group when they
reported the TMS-quinine catalyzed dimerization of meth-
ylketene.3
ꢀ-Lactones are interesting targets in synthesis as they are
versatile intermediates and are integral structural features of a
number of biologically active molecules, such as (-)-panclicin
D and 1233A.1,2 While aldoketene dimer ꢀ-lactones have been
used extensively in synthetic activities by Calter and co-workers,
ketoketene dimers have received less attention due to the paucity
of general methods for their preparation.3-5 Ketoketene dimers
are among the most interesting ꢀ-lactones from a reactivity
standpoint due to the presence of an exocyclic double bond with
potential for use as a nucleophile following ring-opening
reactions.6 In addition, ꢀ-lactone ketene dimers have recently
been shown to have activity as enzyme active site inhibitors by
Romo and co-workers.4
As part of our program of studies toward the development
of new reactions involving phosphonium enolates, we sought
to develop an efficient methodology for the dimerization of
ketoketenes. We anticipate that the ketoketene dimers may
provide access to the quaternary stereogenic center of interesting
The lack of dimer selectivity in the PBu3 system may be due
to the enhanced reactivity of the phosphonium enolate inter-
mediates compared to the ammonium enolate species encoun-
tered in TMS-quinine catalyzed reactions.3 We therefore
(1) Yang, H. W.; Romo, D. J. Org. Chem. 1997, 62, 4.
(2) Dymock, B. W.; Kocienski, P. J.; Pons, J.-M. Synthesis 1998, 1655.
(3) Calter, M. A. J. Org. Chem. 1996, 61, 8006. Calter, M. A.; Orr, R. K.
Org. Lett. 2003, 5, 4745.
(4) Purohit, V. C.; Richardson, R. D.; Smith, J. W.; Romo, D. J. Org. Chem.
2006, 71, 4549.
(7) Rasmussen, K.; Calligaro, D. O.; Czachura, J. F.; Dreshfield-Ahmad, L. J.;
Evans, D. C.; Hemrick-Luecke, S. K.; Kallman, M. J.; Kendrick, W. T.; Leander,
J. D.; Nelson, D. L.; Overshiner, C. D.; Wainscott, D. B.; Wolff, M. C.; Wong,
D. T.; Branchek, T. A.; Zgombick, J. M.; Xu, Y.-C. J. Pharmacol. Exp. Ther.
2000, 294, 688.
(8) Huang, A.; Kodanko, J. J.; Overman, L. E. J. Am. Chem. Soc. 2004,
126, 14043.
(9) Taggi, A. E.; Hafez, A. M.; Wack, H.; Young, B.; Ferraris, D.; Lectka,
T. J. Am. Chem. Soc. 2002, 124, 6626.
(10) Zhu, C.; Shen, X.; Nelson, S. G. J. Am. Chem. Soc. 2004, 126, 5352.
(5) Hasek, R. H.; Clark, R. D.; Elam, E. U.; Martin, J. C. J. Org. Chem.
1962, 27, 60. Elam, E. U. J. Org. Chem. 1967, 32, 215. Bentrude, W. G.; Johnson,
W. D. J. Am. Chem. Soc. 1968, 90, 5924. Aronov, Y. E.; Cheburkov, Y. A.;
Knunyants, I. L. IzV. Akad. Nauk SSSR, Ser. Khim. 1967, 8, 1758. Moore, H. W.;
Duncan, W. G. J. Org. Chem. 1973, 38, 156.
(6) Calter, M. A.; Song, W.; Zhou, J. J. Org. Chem. 2004, 69, 1270.
10.1021/jo8024785 CCC: $40.75
Published on Web 01/16/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 1777–1780 1777

TABLE 1. Effect of Reaction Conditionsa on Yield and Selectivity
of Methylphenylketene Dimerization
SCHEME 2. Postulated Effect of Lithium Iodide on
Product Distribution in PBu3-Catalyzed Ketoketene
Dimerization
LiI
(equiv)
yieldb (%) dimer/
(conv, %) trimerc
entry
catalyst (equiv)
solvent
PhCH3
PhCH3
CH2Cl2
Et2O
1
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.2)
PBu3 (0.1)
0
0
0
- (>99)
- (>99)
- (>99)
45
1.4:1.0
2.5:1.0
4:1
>16:1
>16:1
4:1
2d
3
4
5
6
7
1.0
1.0
1.0
1.0
0.2
0.3
0.3
0.3
0.3
0.3
0.3
PhCH3/Et2O 41
THF 50
CH2Cl2 /Et2O 50
>16:1
10:1
8
CH2Cl2 /Et2O - (>99)
9
CH2Cl2/Et2O
0
10
11
12
13
PBu3 (0.1)
PPh3(0.1)
CH2Cl2/Et2O 60
CH2Cl2 /Et2O 47
CH2Cl2 /Et2O 10
CH2Cl2 /Et2O <10
CH2Cl2/Et2O 78
16:1
16:1
P(OEt)3 (0.1)
TMS-quinine (0.1)
14e PBu3 (0.1)
16:1
a Reactions carried out at a concentration of 0.5 M ketoketene in
solvent. b Isolated yield of Z-dimer after flash column chromatography
c
1
on neutral silica. Determined by H NMR analysis of Me signals in the
1H NMR of crude product mixture. d Reaction conducted at a con-
centration of 0.1 M of ketoketene in solvent. e Purified by passing
through a plug column of neutral silica (iatrobeads).
TABLE 2. Yields and Selectivities for the Formation of 4a-h
envisaged that the addition of a Lewis acid to this reaction
system would lead to a selective reaction through stabilization
of enolate intermediate 3, which would possess lower reactivity
and hence favor dimer formation over trimer formation (Scheme
2).10-12 Indeed, this change to the protocol proved successful
in enhancing dimer formation (Table 1, entries 4, 5, and 7).
Subsequent improvements to the methodology involving chang-
ing the solvent system to CH2Cl2/Et2O, reducing the catalyst
(PBu3) loading to 0.1 equiv, and reducing the lithium iodide
loading to 0.3 equiv led to an improved yield (60%) in the
reaction (Table 1, entry 10). Finally, purification by passing the
crude product mixture through a plug of neutral silica (iatro-
beads), rather than performing standard flash column chroma-
tography, facilitated an improvement to a synthetically useful
isolated yield of 78% (Table 1, entry 14).
LiI was found to be the most effective Lewis acid. LiClO4
also proved successful as an additive in modulating dimer
selectivity, although a lower yield (44% compared to entry 10,
Table 1) of ketoketene dimer was obtained in this case. Other
lithium salts, such as LiOTf, and stronger Lewis acids such as
TiCl4, AlCl3, and B(cyclohexyl)2Cl proved less successful
leading to poor selectivity or poor conversion/decomposition
of starting ketoketene.
entry PR3
R1
R2 yielda (%) Z/Eb product
1
PBu3 Ph
Me
Et
Me
78
77
90
>16:1
37:1
39:1
4a
4b
4c
4d
4e
4f
2
PBu3 Ph
3
PBu3 4-MePh
4
PBu3 6-MeO-2-naphthyl Me
PMe3 Ph
PBu3 Ph
PBu3 3-thienyl
62
>16:1
5c
6c
7
Ph
iBu
Et
61/78d
84
28:1
20:1
3:1
78
60
4g
4h
8c,d PMe3 cyclopentyl
Ph
a Isolated yield after passing through a plug column of neutral silica
(iatrobeads). Purity g95% in all cases with the exception of 4 h (90%
purity) as determined by GCMS and HPLC analysis. b Determined by
1H NMR analysis of Me signals in 1H NMR of crude reaction mixture
or by HPLC analysis on an AD column. c No LiI was used. d 0.2 equiv
of PMe3 was used.
precedence in the work of Tidwell on the addition of
organolithiums to ketoketenes and in the work of Calter on
the cinchona alkaloid-catalyzed dimerization of aldoketenes
to support this analysis.3,6,13
Having optimized the conditions for methylphenylketene
dimerization, we then examined the substrate tolerance of the
methodology. Variation of the alkyl substituent proved suc-
cessful under the present conditions (entries 2, 6, and 8, Table
2). The methodology also proved to be highly tolerant of
variation in the aryl group (entries 3, 4, and 7, Table 2).
Sterically demanding substrates (entries 5 and 8, Table 2) were
dimerized efficiently when the smaller trimethylphosphine was
used instead of tri-n-butylphosphine as the dimerization catalyst.
The reaction times (see the Supporting Information for full
The major olefin isomer of methylphenylketene dimer 4a
is presumed to be the Z-isomer. This is the isomer that would
be expected on the basis of an analysis of the reaction
mechanism. A nucleophile would be expected to add to the
side of the ketene that is less sterically hindered in order to
minimize steric interactions in the transition state leading to
2 and 3 (Scheme 2). Arising from this situation, A1,3 strain
would also be minimized in product 4. There is significant
(11) France, S.; Wack, H.; Hafez, A. M.; Taggi, A. E.; Witsil, D. R.; Lectka,
T. Org. Lett. 2002, 4, 1603.
(12) Calter, M. A.; Tretyak, O. A.; Flaschenriem, C. Org. Lett. 2005, 7, 1809.
(13) Baigrie, L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc. 1985,
107, 5391.
1778 J. Org. Chem. Vol. 74, No. 4, 2009

SCHEME 3. Ring-Opening Reactions of Ketoketene Dimers
details) ranged from 50 min (for formation of 4a) to 3 days
(for formation of 4e).
The dimerization of cyclopentylphenylketene (1h) showed
poor selectivity for the Z-isomer, presumably because of the
similar steric size of the phenyl and the cyclopentyl substituents
and the resulting lack of steric bias encountered by the
approaching nucleophilic catalyst.
In some cases (entries 5, 6, and 8, Table 2), it was found
that a high preference for dimer formation could be obtained
even in the absence of lithium iodide. This was ascribed to the
lower reactivity of enolate 3e/3f/3h compared to 3a. In the case
of enolate 3e, this would be due to electron withdrawal by
resonance by two phenyl substituents, while in the cases of 3f
and 3h, increased steric bulk of the enolate subsitutents (iBu or
Ph vs Me) presumably decreases their reactivity. This would
lead to a preference for intramolecular O-enolate ring closure
over the trimer forming intermolecular reaction.
We postulate that the ketoketene dimer 4 is formed through
the mechanism presented in Scheme 2. Nucleophilic attack of
the trialkylphosphine catalyst to the less sterically hindered side
of the ketoketene 1 (where R2 ) less sterically demanding
substituent) would result in the stereoselective formation of
phosphonium enolate 2.3,6,13 Nucleophilic addition of 2 to a
second molecule of 1 would give rise to a second lithium enolate
intermediate 3 stabilized through a 6-membered chelate. 4-Exo-
trig cyclization and elimination of the phosphine results in
generation of the ketoketene dimer product 4. Some support
for this mechanism was provided by 31P NMR spectroscopy.
31P monitoring of the dimerization of 1f (in the absence of LiI)
revealed a strong signal at 34.5 ppm which is a resonance
consistent with the structure of lithium-free phosphonium ion
3.14 This suggests that 3 is the resting state of the catalyst (for
formation of 4f).
SCHEME 4. Asymmetric Variant
An alternative mechanism would involve attack of the O
(rather than C) of phosphonium enolate 2 on a second molecule
of ketoketene leading to the formation of a pentacovalent
phosphorane species, as was observed by Bentrude and co-
workers in the P(OMe)3-catalyzed dimerization of dimeth-
ylketene.5 However, this mechanism was considered less likely
due to the absence of any observed 31P signals upfield from
external 85% H3PO4 (apart from one for PBu3 at -30.7 ppm).15
The enhanced stability of enolate intermediates 3 derived from
alkylarylketenes relative to those derived from dimethylketene
should slow cyclization to a pentacovalent intermediate. In
addition, electron-donating alkyl substituents (of PBu3) on the
phosphonium center of 3 would be expected to stabilize the
positively charged phosphorus of 3 to a greater degree than when
P(OMe)3 is used as a catalyst, hence disfavoring cyclization to
a pentacovalent phosphorane intermediate.14 Taking these factors
into account, it is perhaps not surprising that a pentacovalent
phosphorane intermediate was not observed in this study.
ring opened using MeOLi to afford ꢀ-ketoester 8 in an excellent
yield of >99% as a 1.6:1.0 mixture of diastereomers. ꢀ-Lactone
4a was also converted to Weinreb amide 9 in good yield (90%,
dr ) 1.4:1.0) by reaction with the Weinreb amine under
conditions developed by Calter and co-workers.3 Surprisingly,
4a was converted to 1,3-diketone 10 in high yield (93%) and
good diastereoselectivity (8:1) through reaction with excess
n-BuLi, rather than undergoing double addition to give the
expected ꢀ-hydroxyketone.16 The latter reaction underscores the
potential of ketoketene dimers to provide exciting opportunities
in synthetic activities, highlighting as it does their unusual
reactivity.
Finally, studies toward the development of an asymmetric
variant of this reaction have been initiated. We have investigated
phosphines possessing axial chirality as these have shown
promise in a variety of organocatalytic reactions.17 Thus far,
proof of concept has been demonstrated as ethylphenylketene
dimer has been obtained in a good enantiomeric excess of 80%
using (R,R)-BINAPHANE 11 (0.1 equiv) as catalyst (Scheme
4).18
Having demonstrated the tolerance of the methodology for a
variety of ketoketenes, we then investigated the potential utility
of the ketoketene dimer products for synthesis by carrying out
a variety of ring-opening reactions to access other useful
functionality (Scheme 3). ꢀ-Lactone 4b was reduced using
LiAlH4 or DIBAL to give the desired ꢀ-hydroxyketone 7 in
74% yield as a 1:1 mixture of diastereomers. ꢀ-Lactone 4a was
Future studies will involve extending the substrate scope of
the asymmetric variant of this reaction as well as exploring a
number of ring-opening reactions of the ketoketene dimers for
applications in drug molecule and natural product synthesis.
(14) Vedejs, E.; Diver, S. T. J. Am. Chem. Soc. 1993, 115, 3358. Zhu, X.-
F.; Henry, C. E.; Kwon, O. J. Am. Chem. Soc. 2007, 129, 6722.
(15) Dennis, L. W.; Bartuska, V. J.; Maciel, G. E. J. Am. Chem. Soc. 1982,
104, 230. Chestnut, D. B.; Quin, L. D. Tetrahedron 2005, 61, 12343.
(16) Ferna´ndez-Meg´ıa, E.; Iglesias-Pintos, J. M.; Sardina, F. J. J. Org. Chem.
1997, 62, 4770.
(17) Wurz, R. P.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 12234.
(18) Xiao, D.; Zhang, Z.; Zhang, X. Org. Lett. 1999, 1, 1679.
J. Org. Chem. Vol. 74, No. 4, 2009 1779

ꢀ-Lactone 4a (Table 2, Entry 1). Methylphenylketene (197 mg,
1.48 mmol), stirred for 1 h at 0 °C, was isolated as a colorless oil
Experimental Section
General Procedure A for Dimerization of Ketoketenes.
Ketoketene (1.39 mmol, 1.0 equiv) was dissolved in CH2Cl2 (1.4
mL). LiI (0.42 mmol, 0.3 equiv) was dissolved in Et2O (1.4 mL)
and was then transferred to the flask containing the ketoketene
solution, and the resulting solution (0.5 M of ketoketene in solvent)
was cooled to 0 °C. Tri-n-butylphosphine (0.14 mmol, 0.1 equiv)
was added in one portion, and the mixture was stirred for the
indicated time at the indicated temperature. The reaction was then
diluted with CH2Cl2 (5 mL) and quenched by adding deionized
water (10 mL). The layers were separated, the aqueous layer was
extracted with CH2Cl2 (2 × 5 mL), and the combined organics were
dried over anhydrous sodium sulfate. The solvent was removed
under reduced pressure to provide the crude product for 1H NMR/
GCMS analysis. 10% EtOAc/hexane (20 mL) and dichloromethane
(5 mL) were added to the crude residue, which was passed through
a plug column of neutral silica (iatrobeads, 2 × 2 cm, 10 g) and
was eluted with 10% EtOAc/hexane (100 mL). Finally, solvent was
removed under reduced pressure to yield the desired ketoketene
dimer in high purity (g95%), in most cases, as determined by
1
(152 mg, 78%). The Z/E ratio was determined to be >16:1 by H
1
NMR analysis: IR (thin film) 1881, 1844, 1699, 1140 cm-1; H
NMR (200 MHz, CDCl3, TMS) δ 7.52-7.16 (m, 10H), 1.92
(s, 3H), 1.86 (s, 3H); 13C NMR (50 MHz, CDCl3) δ 171.4, 146.9,
136.2, 135.2, 129.4, 128.6, 128.6, 127.6, 127.4, 126.3, 108.6, 64.4,
19.6, 15.6; MS (EI 70 eV) m/z 264, 132, 104, 78; (M+ + Na)
HRMS m/z calcd for C18H16O2Na 287.1043, found 287.1039.
Acknowledgment. We gratefully acknowledge Oakland
University for financial support (startup funds and scholarship
award) and Degussa for providing phosphepines. We also thank
Professors Arthur W. Bull and Roman Dembinski for helpful
discussions.
Supporting Information Available: Experimental proce-
dures and product characterization data for compounds 4a-h
and 7-10. This material is available free of charge via the
Internet at http://pubs.acs.org.
GCMS and HPLC analysis, and confirmed by H and 13C NMR
1
spectroscopy.
JO8024785
1780 J. Org. Chem. Vol. 74, No. 4, 2009