Mechanistic studies of the phosphine-catalyzed homodimerization of ketoketenes

Article abstract of DOI:10.1016/j.tetlet.2010.10.026

The mechanism of PBu₃-catalyzed homodimerization of ketoketenes has been explored and compared with that of the previously reported trialkylphosphite-mediated reactions. NMR studies of the PBu₃- catalyzed reaction implicated the involvement of tetravalent phosphonium intermediates. Phosphonium intermediates in the catalytic cycle were trapped through reaction with trimethylsilyl chloride and 4-chlorobenzaldehyde, and the resulting products were characterized. A method for the stoichiometric generation of phosphonium enolates was developed as a result of these studies. No evidence was obtained for the involvement of pentacovalent phosphorane intermediates in trialkylphosphine-catalyzed ketoketene homodimerization reactions, in contrast with the mechanism of the trialkylphosphite-mediated homodimerization of dimethylketene. An X-ray crystal structure analysis of methylphenylketene dimer showed that it possesses Z-geometry about the exocyclic olefin.

Full text of DOI:10.1016/j.tetlet.2010.10.026

Tetrahedron Letters 51 (2010) 6690–6694
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Mechanistic studies of the phosphine-catalyzed homodimerization of
ketoketenes
Pei-Hsun Wei a, Ahmad A. Ibrahim a, Mukulesh Mondal a, Divya Nalla a, Gero D. Harzmann a,
Frank A. Tedeschi a, Kraig A. Wheeler b, Nessan J. Kerrigan a,
⇑
a Department of Chemistry, Oakland University, 2200 N. Squirrel Rd, Rochester, MI 48309-4477, USA
b Department of Chemistry, Eastern Illinois University, 600 Lincoln Avenue, Charleston, IL 61920-3099, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The mechanism of PBu3-catalyzed homodimerization of ketoketenes has been explored and compared
with that of the previously reported trialkylphosphite-mediated reactions. NMR studies of the PBu3-cat-
alyzed reaction implicated the involvement of tetravalent phosphonium intermediates. Phosphonium
intermediates in the catalytic cycle were trapped through reaction with trimethylsilyl chloride and 4-
chlorobenzaldehyde, and the resulting products were characterized. A method for the stoichiometric gen-
eration of phosphonium enolates was developed as a result of these studies. No evidence was obtained
for the involvement of pentacovalent phosphorane intermediates in trialkylphosphine-catalyzed ketoke-
tene homodimerization reactions, in contrast with the mechanism of the trialkylphosphite-mediated
homodimerization of dimethylketene. An X-ray crystal structure analysis of methylphenylketene dimer
showed that it possesses Z-geometry about the exocyclic olefin.
Received 12 September 2010
Accepted 6 October 2010
Available online 20 October 2010
Keywords:
Disubstituted ketenes
Ketoketenes
Ketoketene dimers
b-Lactones
Diastereoselectivity
Ketenes
Ó 2010 Elsevier Ltd. All rights reserved.
Phosphonium enolate
Enolates
Phosphines
Dimerization of ketenes is an important reaction for accessing
b-lactones.1–3 Elam and Bentrude independently showed that dim-
ethylketene could be homodimerized using trialkylphosphites as
nucleophilic catalysts.1c–f
In 1996 Calter showed that a nucleophilic catalyst system
(TMS-quinine or TMS-quinidine) could catalyze the homodimer-
ization of aldoketenes with high enantioselectivity (Scheme 1).2
While aldoketene dimer b-lactones have been used extensively in
the synthesis of polyketides (polypropionates) by Calter et al.,
ketoketene dimers have received less attention due to the paucity
of general methods for their preparation.3–5
O
O
•
1 mol%
P(OEt)
O
3
OTMS
93%
Me
Me
Me
Me
Me
N
Me
1a
2a
N
O
•
O
5 mol%
TMS-quinine
OMe
O
79%
97% ee
Me
TMS-quinine
Me
Me
2b
Scheme 1. Literature examples for homodimerization of ketenes.
1b
In 2008, our group reported a versatile trialkylphosphine cata-
lytic system which provided a general method for ketoketene
(disubstituted ketene) homodimerization (Scheme 2).4 A range of
ketoketene dimer b-lactone products were obtained in good to
excellent yields and with excellent diastereoselectivity (Scheme
2).4
We have recently demonstrated the synthetic potential of
ketoketene dimers by showing that they can undergo highly
diastereoselective ring-opening reactions to afford, for example,
1,3-diketones (dr up to 90:10).4c
O
O
•
O
0.1 equiv PBu3 or PMe3
R1
R1
R1
R2
R2
0.3 equiv LiI
CH2Cl2/Et2O
0º C
R2
60-90% yield
dr 16:1
8 examples
≥
Scheme 2. Phosphine-catalyzed homodimerization of ketoketenes.
⇑
Corresponding author. Tel.: +1 2483702085.
E-mail address: kerrigan@oakland.edu (N.J. Kerrigan).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.10.026

P.-H. Wei et al. / Tetrahedron Letters 51 (2010) 6690–6694
6691
Table 1
_
Major 31P NMR resonances observed in the nucleophile-catalyzed homodimerization
O
R1
of diphenylketene 1c to form dimer 2ca
+
O
•
PBu3
O
•
R2
E-3
O
O
•
O
0.1 equiv catalyst
R1
R2
R1
R2
R1
R2
R1
1
CH Cl
R1
R2
2
2
R2
0 ºC to r t
R1 > R2
Entry
R1
R2
Nucleophile
31P NMR signals (ppm)
Yield of 2cb
O
_
+
PBu3
O
PBu3
1
2
3
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
PBu3
PMe3
PPh3
P(OEt)3
P(OEt)3
13.6
—
67
61
0
0
0
R2
R1
R1
À5.1
R2
4
7.9
4
5c
À55
a
All reactions were conducted at 0.5 M concentration of ketene in CH2Cl2 unless
stated otherwise.
O
b
O
Yields are isolated yields.
Reaction conducted with 0.5 equiv of P(OEt)3 at 5.5 M concentration of ketene
c
R2
R1
R1
in Et2O by Baldwin’s group (see Ref. 11).
R2
Z-2
Table 2
Scheme 3. Mechanism A for catalytic homodimerization of ketoketenes.
Major 31P NMR resonances observed in the nucleophile-catalyzed homodimerization
of ketoketenes 1a–fa
O
O
•
In this Letter, we describe our studies on the mechanism of the
phosphine-catalyzed reaction with a range of ketoketene sub-
strates, and compare it to the mechanism of Bentrude’s phos-
phite-mediated dimethylketene dimerization reaction.1
O
0.1 equiv catalyst
R1
R1
R1
R2 CH2Cl 2 or as specified below
0 ºC
R2
R2
We originally postulated that ketoketene dimer 2 was formed
through mechanism A presented in Scheme 3.4b Nucleophilic at-
tack of the trialkylphosphine catalyst on the less sterically hin-
Entry
R1
R2
Catalyst
31P NMR signals
(ppm)
Yield
of 2a–f
1
Et
Et
Et
Et
i-Bu
Et
Me
Me
Ph
Ph
Ph
Ph
PBu3
PBu3/LiI
PPh3
P(OEt)3
PBu3
PBu3/LiI
PBu3
34.2
34.3
À5.1
98
77
0
dered side of the ketoketene
1
(where R2 = less sterically
2b
3
demanding substituent) would result in the stereoselective forma-
tion of phosphonium enolate E-3.6 Nucleophilic addition of 3,
through C, to a second molecule of 1 would give rise to a second
enolate intermediate 4 (stabilized through a six-membered chelate
when LiI is used as an additive). 4-Exo-trig cyclization and elimina-
tion of the phosphine would result in the generation of ketoketene
dimer product Z-2.
4
5
À51.6, À53.3, À55.5, À56.6
0
Ph
34.3c, 10.9
—
84
87d
73
>99f
6b
7e
8
c-Hexyl
Me
Me
33.2, 31.5c
À53.8
P(OMe)3
a
b
c
d
e
f
Yields are isolated yields.
Reaction carried out in CH2Cl2/Et2O.
Major signal.
Alternatively in mechanism B (Scheme 4), addition of 3, through
oxygen, to a second molecule of 1 would give rise to enolate 5.
Zwitterion 5 would add to a third molecule of 1 to give enolate
6, which would undergo cyclization to afford 2 and regenerate 3.
In this alternative proposal mechanism B (Scheme 4), which is
related to the mechanism proposed by Bentrude for the P(OMe)3-
mediated dimerization of dimethylketene, 3 (generated through
addition of PBu3 to ketene) would act as the dimerization catalyst.1
Bentrude and co-workers obtained evidence for the involvement of
a pentacovalent phosphorane species 7 (PR3 = P(OMe)3) in the
P(OMe)3-mediated dimerization of dimethylketene by carrying
out 31P NMR analysis (d À53.8 ppm for 7) of the reaction.1 The
phosphorane was stable enough to be isolated and, indeed, only
on heating at >60 °C did Bentrude’s group observe decomposition
of the pentacovalent species to generate dimethylketene dimer
and P(OMe)3. They suggested a mechanism which would involve
attack of the O (rather than C) of enolate 3 on a second molecule
of ketoketene leading to the formation of a pentacovalent phos-
phorus species 7.1 Mechanism B deviates significantly from Ben-
trude’s original proposal in that 5 (with PR3 = PBu3) would not
cyclize to give 7, but instead would add, through C of enolate 5,
to another molecule of ketene to give 6. The electron donating alkyl
substituents (of PBu3) on the phosphonium center of 5 would be
expected to stabilize the positively charged phosphorus of 5 to a
greater degree than when P(OMe)3 is used, hence disfavoring
cyclization to a pentacovalent phosphorane intermediate 7.7a This
critical difference would enable catalytic turnover in our system,
unlike in Bentrude’s system.1e,4 Interestingly, when Elam carried
1,3-Cyclobutanedione regioisomer.
Reaction carried out in THF at À25 °C.
Reaction performed by Bentrude’s group using 0.5 equiv P(OMe)3 in EtOAc,
yield after heating to >60 °C, see Ref. 1e.
out the P(OEt)3-catalyzed dimerization of dimethylketene under
solventless conditions, less than 0.01 equiv of catalyst was requir-
ed.1d Presumably, catalytic turnover is observed in this situation
due to the exothermic nature (90–100 °C) of the neat reaction,
resulting in decomposition of 7.1d
In order to elucidate information about which mechanism (A or
B) is operative in our system, and about the intermediates in-
volved, we carried out extensive 31P NMR spectroscopic studies
of the nucleophile-catalyzed homodimerization reaction of keto-
ketenes 1a–f (Tables 1 and 2), as well as performing intermediate
trapping experiments. In all of these studies, a catalytic amount
(0.1 equiv) of the nucleophilic catalyst was used, unless otherwise
stated. We initially investigated the dimerization of diphenylke-
tene 1c (Table 1). PBu3 and PMe3 were found to be the only suc-
cessful homodimerization catalysts. Careful 31P NMR monitoring
of the PBu3-catalyzed reaction (Table 1) revealed a strong signal
at 13.6 ppm. We propose that this signal represents a phospho-
nium enolate based on the region (10–40 ppm) in the 31P NMR
spectrum where the signal appears, which is characteristic for a
tetravalent phosphonium species.7 Phosphonium enolates 3–6
were considered to be possible structures for the signal at
13.6 ppm (Scheme 3 and 4). To help distinguish between the

6692
P.-H. Wei et al. / Tetrahedron Letters 51 (2010) 6690–6694
31P NMR d À5.1 ppm).
R2
observed, with only a signal for free PPh3
(
R1
From this, we suggest that PPh3 is too sterically hindered and not
sufficiently nucleophilic to generate a significant amount of phos-
phonium enolate at equilibrium. P(OEt)3 is less sterically hindered
and functions as a better nucleophile than PPh3 but the resulting
enolate 3 (31P NMR d 7.9 ppm) is presumably too stabilized, due
to the electron withdrawing alkoxy groups, to undergo further
reaction with diphenylketene under our conditions (0.5 M concen-
tration of ketene in CH2Cl2). However, it should be noted that Bald-
win’s group found that when P(OEt)3 (1 equiv) was used under
more concentrated conditions (5 M concentration of ketene in
Et2O) it could react with diphenylketene (2 equiv) to form a penta-
covalent phosphorane which gave a 31P NMR signal at À55 ppm
(Table 1, entry 5).11
O
_
O
R1
O
R1
R2
P
R3
R2
O
O
•
7
R1
+
O
•
PR3
R2
R1
R2
R1
R2
5
1
R1 > R2
PR3 = PBu3
or P(OMe)3
_
_
O
R1
R2
O
R1
+
O
PR3
R1
R2
O
R2
R1
E-3
+
We next examined the nucleophile-catalyzed dimerization of
ethylphenylketene 1d (Table 2, entries 1–4). We had previously
found it necessary to use LiI as an additive in order to reduce trimer
formation and favor dimer formation in the PBu3-catalyzed reac-
tion.4b However, this strategy was found to be only necessary for
alkylarylketenes possessing a –CH3 substitutent due to the greater
reactivity of these ketenes vis-à-vis more sterically hindered ket-
enes (R1 = Et, i-Bu, Ph). The dimerization of ethylphenylketene pro-
ceeded efficiently when PBu3 was used as the nucleophilic catalyst,
both in the presence and in the absence of lithium iodide (Table 2,
entry 1 vs entry 2) to provide the desired dimer in good to excel-
lent yields (77–98%). The same resting state of the catalyst was ob-
served in both cases, that is, a tetravalent phosphonium species
giving rise to a strong signal at 34 ppm. The first enolate 3d was
not observed under the catalytic reaction conditions, as judged
by the absence of a signal between 10 and 20 ppm. It appears in
these cases that a tetravalent phosphonium intermediate other
than 3 (Schemes 3 and 4) is giving rise to the signal at 34 ppm. This
is presumed to be due to the enhanced reactivity and lower stabil-
ity of 3d relative to diphenylketene-derived enolate 3c, and so the
resting state of the catalyst is strongly dependent upon the struc-
ture of the ketoketene. To clarify the structure of the observed
intermediate (4, 5 or 6), we attempted to prepare and characterize
it through a modification of the reaction conditions. At À78 °C,
PBu3 (ca. 0.6 equiv) was added quickly to a dilute solution of eth-
ylphenylketene (1 equiv) in CH2Cl2 (0.05 M concentration of ke-
tene in CH2Cl2). Monitoring by 31P NMR showed the formation of
a species giving a major signal at 34 ppm. TMSCl (2 equiv) was
then added at À78 °C to trap the intermediate. The resulting prod-
uct was characterized and was determined to be silylketene acetal
9d, a derivative of 5d (Schemes 4 and 6), which lends support to
the proposal of mechanism B as being operative in the dimeriza-
tion of ketoketenes.12 To test this mechanistic proposal we
PR3
R2
6
O
O
R2
R1
R1
R2
Z-2
Scheme 4. Mechanism B for catalytic homodimerization of ketoketenes.
possible enolate structures (3–6) for the signal at 13.6 ppm, we
carried out a number of control experiments. First, we investigated
where the 31P NMR resonance of an acylphosphonium derived
from diphenylacetyl chloride and PBu3 would appear. After stirring
PBu3 (1 equiv) and diphenylacetyl chloride (1 equiv) in CH2Cl2 for
50 min, a signal was observed at 31.8 ppm. Indeed, based on liter-
ature precedent, an acylphosphonium signal would be expected to
appear between 28 and 32 ppm, and so structure 4 could be ruled
out as a possibility for the signal at 13.6 ppm.8 Second, when PBu3
(1 equiv) was added to diphenylketene (1 equiv) in CH2Cl2 at
À78 °C, the signal for PBu3 (À30.7 ppm) disappeared completely
and was cleanly replaced by a signal at 13.5 ppm (94% of all 31P
NMR signals by integration). 13C and 1H NMR analysis of the result-
ing phosphonium species in CD2Cl2 supported our assignment of
structure 3 to the signal at 13.6 ppm (in the 31P NMR spectrum).9
The species giving rise to a signal at 13.6 ppm was also converted
into enol ester derivative 8c and fully characterized, a structure
which supports the assignment of structure 3 to the signal at
13.6 ppm (Scheme 5).10 Therefore, phosphonium enolate 3 appears
to be the catalyst resting state in the PBu3-catalyzed dimerization
of diphenylketene. This is presumably due to the attenuated reac-
tivity of 3c (R1 = R2 = Ph). The electronically stabilized and steri-
cally hindered nature of 3c would be expected to slow its
reaction with another molecule of diphenylketene.
_
O
•
O
OTMS
1.0 equiv PBu
-
3
The use of less basic phosphorus-centered catalysts proved to
be an ineffective strategy for catalyzing the dimerization of diph-
enylketene (Table 1, entries 3 and 4). To understand why this
was so, we analyzed these reactions by 31P NMR spectroscopy. In
the case of PPh3, no signal for a phosphonium enolate was
Cl
TMSCl
-78 ºC
+
+
Ph
Ph
Bu P
3
Bu P
3
CD Cl
2
2
Ph
Et
Et
3d
Et
or CH Cl
2
2
O
-78 ºC
10d
Ph
Cl
>99%
1, 0.5 equiv PBu
-78 ºC
3
E:Z = 95:5
4-ClPhCHO
-78 ºC
Et
2. TMSCl, -78 ºC
-78 ºC
O
O
Ph
Et
TMSO
Ph
O
O
Ph
O
Ph
Et
Cl
_
-
O
O
+
Cl
-
Et
Ph
O
•
-
Cl
O
Ph
O
Cl
1.0 equiv PBu
Et
+
3
Ph
Et
Bu P
+
+
Bu P
3
Ph
Ph
3
Bu P
Bu P
3
3
CH Cl
Et
2
2
Et
0 ºC to rt
Ph
Ph
Cl
Ph
3c
P NMR δ 13.5 ppm
Ph
8c
-78 ºC
9d
8d
11d
>99%
>99%
65%
31
dr = 88:12
Scheme 5. Trapping of phosphonium enolate 3c with an acyl chloride.
Scheme 6. Synthetic potential of phosphonium enolate 3d.

P.-H. Wei et al. / Tetrahedron Letters 51 (2010) 6690–6694
6693
prepared 3c in quantitative fashion, as before, through the reaction
of PBu3 (1 equiv) with diphenylketene (1 equiv). The resulting rel-
atively stable enolate 3c was then added to a solution of ethyl-
phenylketene. The dimerization of ethylphenylketene was
observed to proceed efficiently (yield of 2c = 95%). 31P NMR moni-
toring indicated that 3c was present in high concentration for
much of the reaction, while another species (5, 31P NMR d
34 ppm) was gradually formed. While such a result supports the
notion of 3 being the true dimerization catalyst (mechanism B),
we cannot rule out the possibility that PBu3 is being formed in a
reversible fashion from 3 or 5, and that PBu3 itself is actually the
active catalyst.
Surprisingly given the highly reactive nature of ethylphenylke-
tene, the first phosphonium enolate 3d was formed with good con-
version through the quick addition of 1 equiv of PBu3 to 1 equiv of
ethylphenylketene at À78 °C (Scheme 6). In this way a tetravalent
phosphonium enolate species, which gave a 31P NMR (CD2Cl2) d at
13.4 ppm, was generated. Phosphonium enolate 3d was trapped
cleanly through reaction with TMSCl at À78 °C to give the corre-
sponding silyl enol ether 10d quantitatively with an olefin isomer
ratio = 95:5. The major olefin is presumed to be the E-isomer by
analogy with the stereochemical outcome of the addition of nucle-
ophiles to disubstituted ketenes reported by Tidwell and co-work-
ers.6 Compound 10d gave a 31P NMR (CD2Cl2) d at 32.6 ppm (major
isomer).13 Alternatively 3d, generated quantitatively, could be en-
gaged in a formal [2 + 2] cycloaddition with 4-chlorobenzaldehyde
to give 11d (65%, dr = 88:12), with the trans-isomer being the ma-
jor isomer.14 Therefore, this stoichiometric method of phospho-
nium enolate generation represents a promising new avenue for
ketene reaction development, given the lack of a stoichiometric
variant available to other catalytic systems, for example, alkaloid
and ferrocenylamine catalytic systems.2,15
izations involve tetravalent phosphonium enolate intermediates,
which enable catalytic turnover in our system, even under mild
conditions (0 °C or À25 °C). No evidence was observed for the
involvement of pentacovalent phosphorus intermediates in the
PBu3-catalyzed reaction. Knowledge of the structures of the inter-
mediates is expected to have implications for the design of chiral
phosphine catalysts suitable for imparting asymmetric induction
in reactions of ketenes. Furthermore, we have shown that phos-
phonium enolate intermediates involved in PBu3-catalyzed ketene
dimerizations may be trapped through reaction with various elec-
trophiles, hinting at the potential of these enolates for the develop-
ment of new synthetic methodologies. Future work will focus on a
kinetics investigation of the catalytic system, and on the applica-
tion of the catalytic system to complex molecule synthesis.16
Acknowledgments
We thank the National Science Foundation (Grant Nos. CHE-
0911483 to N.J.K., CHE-0722547 to K.A.W., and CHE-0821487 for
NMR facilities at Oakland University), and Oakland University
(N.J.K.) for financial support of this work.
Supplementary data
Supplementary data (characterization data and procedures for
the preparation of 2a, 2f, 3c, 8c, 8d, 9d, 10d, and 11d, and X-ray
crystal structure data for methylphenylketene dimer 2g) associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2010.10.026.
References and notes
Phosphonium enolate intermediates were also observed when
the PBu3-catalyzed dimerization of other ketenes, such as iso-
butylphenylketene and dimethylketene, were monitored using
31P NMR analysis (Table 2, entry 5 and 7). Indeed, pentacovalent
phosphorane intermediates were only ever detected when trial-
kylphosphites were used as reaction promoters (Table 2, entries
4 and 8). Dialkylketenes gave the b-lactone or 1,3-cyclobutanedi-
one regioisomer when exposed to PBu3 catalysis, depending upon
the alkyl substituents chosen. Dimethylketene was efficiently
dimerized to give the b-lactone when tri-n-butylphosphine was
used as the catalytic system. On the other hand, cyclohexylethylke-
tene was dimerized to give 1,3-cyclobutanedione exclusively,
rather than the b-lactone regioisomer (Table 2, entry 6). The switch
in regioselectivity may be due to the higher reactivity of cyclo-
hexylethylketene-derived enolate 6f at C, compared to the lower
reactivity of dimethylketene- or alkylarylketene-derived enolates
6 at C.
The major olefin isomer of the ketoketene dimer b-lactones
was determined to be the Z-isomer by an X-ray crystal structure
analysis of methylphenylketene dimer 2g (see Supplementary
data). This is the isomer that would be expected based on 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 transi-
tion state leading to 3, 5, and 6 (Scheme 4). Arising from this
situation A1,3 strain would be minimized in product 2. Diastere-
oselectivity (Z-isomer: E-isomer >16:1 in most cases) in ketene
dimerization is extremely important given the need to access
enolates stereoselectively, following ring-opening, in order for
them to be broadly useful in synthesis.3,4
1. (a) Tidwell, T. T. Ketenes, 2nd ed.; Wiley: New York, 2006; (b) Sauer, J. C. J. Am.
Chem. Soc. 1947, 69, 2444–2448; (c) Hasek, R. H.; Clark, R. D.; Elam, E. U.;
Martin, J. C. J. Org. Chem. 1962, 27, 60–64; (d) Elam, E. U. J. Org. Chem. 1967, 32,
215–216; (e) Bentrude, W. G.; Johnson, W. D. J. Am. Chem. Soc. 1968, 90, 5924–
5926; (f) Bentrude, W. G.; Johnson, W. D.; Khan, W. A. J. Am. Chem. Soc. 1972, 94,
3058–3067; (g) Aronov, Y. E.; Cheburkov, Y. A.; Knunyants, I. L. Izv. Akad. Nauk
SSSR, Ser. Khim. 1967, 8, 1758–1768; (h) Moore, H. W.; Duncan, W. G. J. Org.
Chem. 1973, 38, 156–158.
2. (a) Calter, M. A. J. Org. Chem. 1996, 61, 8006–8007; (b) Calter, M. A.; Orr, R. K.
Org. Lett. 2003, 5, 4745–4748; (c) Purohit, V. C.; Richardson, R. D.; Smith, J. W.;
Romo, D. J. Org. Chem. 2006, 71, 4549–4558.
3. (a) Calter, M. A.; Guo, X. J. Org. Chem. 1998, 63, 5308–5309; (b) Calter, M. A.;
Song, W.; Zhou, J. J. Org. Chem. 2004, 69, 1270–1275; (c) Calter, M. A.; Liao, W. J.
Am. Chem. Soc. 2002, 124, 13127–13129; (d) Ma, G.; Nguyen, H.; Romo, D. Org.
Lett. 2007, 9, 2143–2146.
4. (a) Kerrigan, N. J.; Ibrahim, A. A.; Harzmann, G. D. Abstracts of Papers, 236th
National Meeting of the American Chemical Society, Philadelphia, PA;
American Chemical Society: Washington, DC, 2008; ORGN 531; (b) Ibrahim,
A. A.; Harzmann, G. D.; Kerrigan, N. J. J. Org. Chem. 2009, 74, 1777–1780; (c)
Ibrahim, A. A.; Smith, S. M.; Henson, S.; Kerrigan, N. J. Tetrahedron Lett. 2009, 50,
6919–6922.
5. Lv, H.; Zhang, Y.-R.; Huang, X.-L.; Ye, S. Adv. Synth. Catal. 2008, 350, 2715–2718.
6. (a) Baigrie, L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc. 1985, 107, 5391–
5396; (b) Dondoni, A.; Fantin, G.; Fogagnolo, M.; Mastellari, A.; Medici, A.;
Pedrini, P. J. Org. Chem. 1984, 49, 3478–3483.
7. (a) Zhu, X.-F.; Henry, C. E.; Kwon, O. J. Am. Chem. Soc. 2007, 129, 6722–6723; (b)
Hudson, H. R.; Dillon, K. B.; Walker, B. J. 31P NMR Data of Four Coordinate
Phosphonium Salts and Betaines. In Handbook of Phosphorus-31 Nuclear
Magnetic Resonance Data; Tebby, J. C., Ed.; CRC Press: Boca Raton, FL, 1991;
pp 181–226; (c) Dennis, L. W.; Bartuska, V. J.; Maciel, G. E. J. Am. Chem. Soc.
1982, 104, 230–235; (d) Chestnut, D. B.; Quin, L. D. Tetrahedron 2005, 61,
12343–12349.
8. (a) Vedejs, E.; Diver, S. T. J. Am. Chem. Soc. 1993, 115, 3358–3359. and references
cited therein; (b) Fan, R.-H.; Hou, X.-L. Tetrahedron Lett. 2003, 44, 4411–4413.
9. Selected characterization data for 3c: 1H NMR (200 MHz, CD2Cl2): d 7.76–6.85
(m, 10H), 1.70–1.26 (m, 18H), 0.88 (t, J = 7.1 Hz, 9H); 13C NMR (50 MHz,
CD2Cl2): d 151.1 (d, J = 47.3 Hz), 143.3, 143.0, 141.6, 135.2, 128.9, 127.8, 126.5,
123.0, 115.1 (d, J = 65.7 Hz), 25.1, 24.5 (d, J = 13.9 Hz), 20.4 (d, J = 40.8 Hz), 13.7;
31P NMR (81 MHz, CD2Cl2): d 13.5.
10. Characterization data for 8c: IR
t ;
max (thin film): 1765, 1718 cmÀ1 1H NMR
In conclusion, we have carried out mechanistic investigations of
the PBu3-catalyzed homodimerization of ketoketenes, and this has
led us to consider mechanism B as the most likely mechanism. We
have demonstrated that trialkylphosphine-catalyzed homodimer-
(400 MHz, CDCl3, TMS): d 7.80–7.12 (m, 15H), 3.32 (t, J = 7.6 Hz, 1H), 2.21–1.18
(m, 20H), 0.92 (t, J = 7.2 Hz, 3H), 0.86 (t, J = 7.3 Hz, 9H); 13C NMR (100 MHz,
CDCl3): d 170.0, 156.6 (d, J = 17 Hz), 136.9, 136.9, 135.1, 130.9, 129.8, 129.3,
129.1 (d, J = 7 Hz), 128.7, 128.6, 128.4, 128.1, 128.0, 127.5, 53.1, 25.7, 23.8 (d,

6694
P.-H. Wei et al. / Tetrahedron Letters 51 (2010) 6690–6694
13. Characterization data for 10d: IR m ;
max (thin film): 2960, 1690, 1465, 1253 cmÀ1
J = 8 Hz), 23.5 (d, J = 17 Hz), 20.7 (d, J = 46 Hz), 13.2, 11.3; 31P NMR (162 MHz,
85% H3PO4, CDCl3): d 35.4; (M++Na) HRMS Anal. Calcd for C36H48O2P: 543.3386
(m/z). Found: 543.3387.
1H NMR (400 MHz, CD2Cl2, TMS at À78 °C) d 7.67–7.56 (m, 3H), 7.42–7.31 (m,
2H), 2.74 (q, J = 7.5 Hz, 2H), 2.28–2.17 (m, 4H), 1.89–1.48 (m, 14H), 1.21–1.04
(m, 12H), 0.59 (s, 9H); 13C NMR (100 MHz, CD2Cl2 at À78 °C) d 145.2 (d,
J = 28.2 Hz), 135.7 (d, J = 11.1 Hz), 128.2, 128.0, 127.9, 127.0, 25.3, 23.3 (d,
J = 14.7 Hz), 23.1, 19.3 (d, J = 47.1 Hz), 12.9, 11.8, À0.7; 31P NMR (162 MHz, 85%
H3PO4 at À78 °C, CD2Cl2) d 32.5 (major: 95% of all 31P species), 32.2 (minor: 5%
of all 31P species).
11. Baldwin, J. E.; Swallow, J. C. J. Org. Chem. 1970, 35, 3583–3584.
12. Characterization data for 9d: 1H NMR (400 MHz, CDCl3, TMS) for major product:
d 7.71–7.34 (m, 10H), 2.79 (dq, J = 1.8, 7.5 Hz, 2H), 2.42–2.31 (m, 6H), 1.97–
1.44 (m, 14H), 1.46 (t, J = 7.3 Hz, 3H), 1.24 (t, J = 7.2 Hz, 3H), 1.16 (t, J = 7.2 Hz,
9H), 0.65 (s, 9H); 13C NMR (100 MHz, CDCl3) for major product: d 170.7, 147.6
(d, J = 23.9 Hz), 129.8, 129.3 (d, J = 11.8 Hz), 129.1, 129.1, 128.4, 128.4, 128.0,
127.4, 126.2, 116.4, 28.2 (d, J = 8.0 Hz), 24.1 (d, J = 4.4 Hz), 23.8, 23.6, 20.5, 13.5,
11.5, 11.5, 1.3; 31P NMR (162 MHz, 85% H3PO4, CH2Cl2): d 34.1 (minor, 5% of all
31P species), 33.7 (major, 85% of all 31P species), 33.5 (minor, 10% of all 31P
species).
14. Spectroscopic data for 11d agreed with that previously reported in: Mondal,
M.; Ibrahim, A. A.; Wheeler, K. A.; Kerrigan, N. J. Org. Lett. 2010, 12, 1664–1667.
15. Hodous, B. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 1578–1579.
16.
A
preliminary kinetics investigation has shown that the reaction rate is
dependent upon catalyst concentration.