Addition of alkynes to aldehydes and activated ketones catalyzed by rhodium-phosphine complexes

Article abstract of DOI:10.1021/jo701643h

(Chemical Equation Presented) A mixture of 2-(di-tert-butylphosphino) biphenyl and dicarbonylacetonato rhodium(I) provides an effective catalyst system for the addition of alkynes to aldehydes and activated ketones. In contrast to the more common zinc-cat

Full text of DOI:10.1021/jo701643h

Addition of Alkynes to Aldehydes and Activated Ketones Catalyzed
by Rhodium-Phosphine Complexes
Pawan K. Dhondi, Patrick Carberry, Lydia B. Choi, and John D. Chisholm*
Department of Chemistry, 1-014 Center for Science and Technology, Syracuse UniVersity,
Syracuse, New York 13244
jdchisho@syr.edu
ReceiVed July 26, 2007
A mixture of 2-(di-tert-butylphosphino)biphenyl and dicarbonylacetonato rhodium(I) provides an effective
catalyst system for the addition of alkynes to aldehydes and activated ketones. In contrast to the more
common zinc-catalyzed processes, enolizable 1,2-dicarbonyls are excellent substrates for these rhodium-
catalyzed additions. This reaction allows for the formation of propargylic alcohols under mild conditions,
tolerating many functional groups (such as carboxylic acids) that are incompatible with other methods.
Little selectivity was observed in cases of unsymmetrical 1,2-diketones. Addition of alkynes to aldehydes
with an adjacent chirality center usually provides the Felkin addition product with excellent selectivity
in some cases. Studies on the catalyst structure show that both the â-diketonate and a carbon monoxide
ligand appear to be bound to the active catalyst. The use of chiral phosphines to induce asymmetry in the
propargyl alcohol products provided low enantioselectivity, which may be due to the phosphine having
a distal relationship to the reacting centers. Modification of other ligands, such as the â-diketonate, appears
to be a more promising avenue for the development of an enantioselective variant.
Introduction
ketones,14,22-27 imines,28-39 and R,â-unsaturated ketones40-45
have been the subject of numerous studies from many investiga-
tors. Several investigators have also published case studies in
which several of these different methods were compared and
evaluated.46,47 These developments have focused primarily on
The area of catalytic nucleophilic addition reactions of alkynes
has seen significant development in the past decade, with these
results being summarized in several excellent reviews.1-4
Transition metal catalyzed additions of alkynes to aldehydes,5-21
(1) Pu, L. Tetrahedron 2003, 59, 9873.
(2) Frantz, D. E.; Fa¨ssler, R.; Tomooka, C. S.; Carreira, E. M. Acc. Chem.
Res. 2000, 33, 373.
(3) Cozzi, P. G.; Hilgraf, R.; Zimmermann, N. Eur. J. Org. Chem. 2004,
4095.
(4) Garcia, C.; Martin, V. S. Curr. Org. Chem. 2006, 10, 1849.
(5) Frantz, D. E.; Fa¨ssler, R.; Carreira, E. M. J. Am. Chem. Soc. 2000,
122, 1806.
(12) Takita, R.; Yakura, K.; Ohshima, T.; Shibasaki, M. J. Am. Chem.
Soc. 2005, 127, 13760.
(13) Harada, S.; Takita, R.; Ohshima, T.; Matsunaga, S.; Shibasaki, M.
Chem. Commun. 2007, 948.
(14) Cozzi, P. G.; Rudolph, J.; Bolm, C.; Norrby, P.-O.; Tomasini, C. J.
Org. Chem. 2005, 70, 5733.
(15) Xu, Z.; Wang, R.; Xu, J.; Da, C.-S.; Yan, W.-J.; Chen, C. Angew.
Chem., Int. Ed. 2003, 42, 5747.
(6) Anand, N. K.; Carreira, E. M. J. Am. Chem. Soc. 2001, 123, 9687.
(7) Gao, G.; Moore, D.; Xie, R.-G.; Pu, L. Org. Lett. 2002, 4, 4143.
(8) Xu, M.-H.; Pu, L. Org. Lett. 2002, 4, 4555.
(9) Gao, G.; Wang, Q.; Yu, X.-Q.; Xie, R.-G.; Pu, L. Angew. Chem.,
Int. Ed. 2006, 45, 122.
(10) Rajaram, A. R.; Pu, L. Org. Lett. 2006, 8, 2019.
(11) Takita, R.; Fukuta, Y.; Tsuji, R.; Ohshima, T.; Shibasaki, M. Org.
Lett. 2005, 7, 1363.
(16) Kang, Y.-F.; Liu, L.; Wang, R.; Yan, W.-J.; Zhou, Y.-F. Tetrahe-
dron: Assymmetry 2004, 15, 3155.
(17) Kang, Y.-F.; Wang, R.; Liu, L.; Da, C.-S.; Yan, W.-J.; Xu, Z.-Q.
Tetrahedron Lett. 2005, 46, 863.
(18) Jiang, B.; Si, Y.-G. Tetrahedron Lett. 2002, 43, 8323.
(19) Jiang, B.; Chen, Z.; Xiong, W. Chem. Commun. 2002, 1524.
(20) Trost, B. M.; Weiss, A. H.; Jacobi von Wangelin, A. J. Am. Chem.
Soc. 2006, 128, 8.
10.1021/jo701643h CCC: $37.00 © 2007 American Chemical Society
Published on Web 11/14/2007
9590
J. Org. Chem. 2007, 72, 9590-9596

Addition of Alkynes to Aldehydes and ActiVated Ketones
TABLE 1. Phosphine Ligand Screen
alkynes because they require no prior modification for use in
these nucleophilic additions, as the acidic alkyne C-H bond
readily undergoes activation in the presence of catalytic amounts
of transition metals. Since no activating group is required, these
processes are inherently environmentally friendly and show
excellent atom economy.48 Addition reactions of alkynes both
create a new carbon-carbon bond and establish a new stereo-
center in a single operation. In many cases both high yields
and enantioselectivity can be realized from these transformations
under catalytic conditions. Such transformations have attracted
considerable interest due to the versatility of the alkyne addition
products, which are useful intermediates in the synthesis of
complex molecules such as pharmaceuticals and natural
products.49-53
While the development of catalytic methods for the addition
of alkynes to aldehydes and ketones has received significant
attention, the catalytic addition of alkynes to 1,2-dicarbonyl
compounds has been explored in a more cursory fashion.
Catalytic zinc conditions are compatible with 1,2-dicarbonyls,
but they are limited to nonenolizable systems,54 with enolizable
1,2-dicarbonyls providing very low yields. The high Lewis
acidity of the zinc reagents promotes enolization of the reactive
ketones, resulting in intermolecular Aldol reactions that limit
the effectiveness of zinc catalysts in these cases.
entry
ligand
yield (%)
1
2
none
0
triphenylphosphine (4)
13
trace
9
24
26
11
12
71
83
95
3
4
5
6
tris(pentafluorophenyl)phosphine (5)
tris(2-methoxyphenyl)phosphine (6)
tris(2-methylphenyl)phosphine (7)
dppfa (8)
7
dppba (9)
8
tri-n-butylphosphine (10)
9
10
11
2-(dicyclohexylphosphino)biphenyl (11)
2-(di-tert-butylphosphino)biphenyl (12)
tri-tert-butylphosphine (13)
a Only 5 mol % of the phosphine was used.
One solution to this limitation of catalytic zinc chemistry was
the use of other transition metals to activate the alkyne under
more neutral conditions. Recently our group has reported that
the use of catalytic amounts of Rh(acac)(CO)2 in the presence
of phosphine ligands leads to acetylides with nucleophilic
properties.43,55 These rhodium acetylides act as selective nu-
cleophiles under mild, neutral conditions. Alkyne addition
reactions catalyzed by rhodium complexes tolerate functional
groups (such as unprotected alcohols and carboxylic acids) that
are not compatible with many other metal catalyzed alkyne
addition reactions. Additional studies reported herein on these
rhodium-catalyzed reactions include the results of alkyne
additions to unsymmetrical 1,2-diketones, diastereoselective
additions to chiral aldehydes, the exploration of chiral phos-
phines as ligands as well as a more complete account of
optimization studies and studies on functional group tolerance.
(21) Shahi, S. P.; Koide, K. Angew. Chem., Int. Ed. 2004, 43, 2525.
(22) Cozzi, P. G. Angew. Chem., Int. Ed. 2003, 42, 2895.
(23) Cozzi, P. G.; Alesi, S. Chem. Commun. 2004, 2448.
(24) Liu, L.; Kang, Y.-F.; Wang, R.; Zhou, Y.-F.; Chen, C.; Ni, M.;
Gong, M.-Z. Tetrahedron: Assymmetry 2004, 15, 3757.
(25) Liu, L.; Wang, R.; Kang, Y.-F.; Chen, C.; Xu, Z.-Q.; Zhou, Y.-F.;
Ni, M.; Cai, H.-Q.; Gong, M.- Z. J. Org. Chem. 2005, 70, 1084.
(26) Wang, Q.; Zhang, B.; Hu, G.; Chen, C.; Zhao, Q.; Wang, R. Org.
Biomol. Chem. 2007, 5, 1161.
(27) Liu, L.; Wang, R.; Kang, Y.-F.; Cai, H.-Q.; Chen, C. Synlett 2006,
1245.
(28) Fischer, C.; Carreira, E. M. Org. Lett. 2004, 6, 1497.
(29) Fa¨ssler, R.; Frantz, D. E.; Oetiker, J.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2002, 41, 3054.
(30) Fischer, C.; Carreira, E. M. Synthesis 2004, 1497.
(31) Wei, C.; Li, Z.; Li, C.-J. Org. Lett. 2003, 5, 4473.
(32) Wei, C.; Li, C.-J. J. Am. Chem. Soc. 2003, 125, 9584.
(33) Zhang, J.; Wei, C.; Li, C.-J. Tetrahedron Lett. 2002, 43, 5731.
(34) Zani, L.; Alesi, S.; Cozzi, P. G.; Bolm, C. J. Org. Chem. 2006, 71,
1558.
(35) Gommermann, N.; Koradin, C.; Polborn, K.; Knochel, P. Angew.
Chem., Int. Ed. 2003, 42, 5763.
Results and Discussion
(36) Koradin, C.; Polborn, K.; Knochel, P. Angew. Chem., Int. Ed. 2002,
41, 2535.
(37) Koradin, C.; Gommermann, N.; Polborn, K.; Knochel, P. Chem.
Eur. J. 2003, 9, 2797.
(38) Jiang, B.; Si, Y.-G. Angew. Chem., Int. Ed. 2004, 43, 216.
(39) Jiang, B.; Si, Y.-G. Tetrahedron Lett. 2003, 44, 6767.
(40) Kno¨pfel, T. F.; Zarotti, P.; Ichikawa, T.; Carreira, E. M. J. Am.
Chem. Soc. 2005, 127, 9682.
(41) Kno¨pfel, T. F.; Boyall, D.; Carreira, E. M. Org. Lett. 2004, 6, 2281.
(42) Chang, S.; Na, Y.; Choi, E.; Kim, S. Org. Lett. 2001, 3, 2089.
(43) Lerum, R. V.; Chisholm, J. D. Tetrahedron Lett. 2004, 45, 6591.
(44) Chen, L.; Li, C.-J. Chem. Commun. 2004, 2362.
(45) Nishimura, T.; Washitake, Y.; Nishiguchi, Y.; Maeda, Y.; Uemura,
S. Chem. Commun. 2004, 1312.
(46) Marshall, J. A.; Bourbeau, M. P. Org. Lett. 2003, 5, 3197.
(47) Kirkham, J. E. D.; Courtney, T. D. L.; Lee, V.; Baldwin, J. E.
Tetrahedron 2005, 61, 7219.
(48) Trost, B. M. Science 1991, 254, 1471.
(49) Shang, S.; Iwadare, H.; Macks, D. E.; Ambrosini, L. M.; Tan, D.
S. Org. Lett. 2007, 9, 1895.
Initial attempts at alkyne addition with rhodium catalysis
focused on the addition of 1-octyne to ethyl pyruvate as a test
reaction (Table 1). Use of Rh(acac)(CO)2 as a precatalyst and
triphenylphosphine 4 as a ligand produced trace amounts of 1,2-
addition products (Table 1, entry 2). Only when the phosphine
was changed to the bulky, electron-rich 2-(di-tert-butylphos-
phino)biphenyl (12)56,57 and tri-tert-butylphosphine (13) did the
yields become noteworthy (Table 1, entries 10 and 11). More
highly functionalized triaryl phosphines were much less effective
ligands for this transformation (Table 1, entries 2-5). Bidentate
phosphines were also poor ligands for the addition reaction
(Table 1, entries 6 and 7). Control experiments showed that
both the phosphine ligand (Table 1, entry 1) and the rhodium
complex were required for the reaction to proceed. These
experiments implicate a rhodium-phosphine complex as the
active catalyst. The formation of rhodium acetylides with
(50) Fettes, A.; Carreira, E. M. J. Org. Chem. 2003, 68, 9274.
(51) Reber, S.; Kno¨pfel, T. F.; Carreira, E. M. Tetrahedron 2003, 59,
6813.
(52) Maezaki, N.; Tominaga, H.; Kojima, N.; Yanai, M.; Urabe, D.;
Tanaka, T. Chem. Commun. 2004, 406.
(55) Dhondi, P. K.; Chisholm, J. D. Org. Lett. 2006, 8, 67.
(56) Tomori, H.; Fox, J. M.; Buchwald, S. L. J. Org. Chem. 2000, 65,
5334.
(53) Crimmins, M. T.; She, J. J. Am. Chem. Soc. 2004, 126, 12790.
(54) Jiang, B.; Chen, Z.; Tang, X. Org. Lett. 2002, 4, 3451.
(57) Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem.
Soc. 2000, 122, 1360.
J. Org. Chem, Vol. 72, No. 25, 2007 9591

Dhondi et al.
TABLE 2. Variation of Phosphine-Transition Metal Ratio
TABLE 3. Variation of Solvent
mol % of
mol % of
entry
solvent
yield (%)
entry
Rh(acac)(CO)2
phosphine 12
yield (%)
1
2
3
4
5
6
7
8
9
1,4-dioxane
THF
1,2-DME
benzene
toluene
1,2-DCE
DMF
DMSO
MeOH
EtOH
10% H2O/1,4-dioxane
i-PrOH
90
94
62
75
76
36
60
0
trace
27
49
74
1
2
3
4
5
6
7
8
5
3
3
3
3
3
1
1
10
6
3
80
84
59
90
86
90
22
52
9
12
15
3
5
10
11
12
rhodium-phosphine complexes has been observed several times
in the organometallic literature,58-60 supporting the hypothesis
of a rhodium-phosphine complex as the active catalyst.
% rhodium catalyst loading did not further increase the yield
of the reaction. Adding more phosphine at a 1 mol % catalyst
loading did lead to a significant increase in yield (Table 2, entry
8), indicating that adventitious oxidation of the phosphine may
limit the lifetime of the catalyst at very low catalyst loadings.
The reaction was also evaluated with respect to concentration
and solvent. Concentrations between 0.6 and 1.0 M were found
to be optimal. A variety of solvents were evaluated in the alkyne
addition reaction (Table 3). Ethereal solvents consistently gave
excellent yields with THF providing the highest yield of 94%
(Table 3, entry 2). Benzene and toluene also gave reasonable
yields. Halogenated solvents, such as 1,2-dichloroethane, gave
much lower yields. While DMSO was completely incompatible
with the reaction (this may be due to its ability to compete as
a ligand for the transition metal, or due to oxidation as DMSO
has been shown to oxidize some transition metal complexes61)
other polar aprotic solvents like DMF performed adequately.
Polar protic solvents, like methanol and ethanol, proved
detrimental to the reaction, as did the addition of significant
amounts of water (Table 3, entries 9-11). As the stability of
rhodium acetylides in polar protic solvents like water has been
well documented,62 this may indicate a lack of solubility of the
rhodium-phosphine complex in polar protic media, as the
phosphine ligand is quite hydrophobic. Isopropanol proved to
be an adequate solvent as well, supporting the hypothesis that
the lower yields are due to poor catalyst solubility (Table 3,
entry 12).
With the experimental conditions optimized, the scope of the
reaction with regard to electrophile was explored (Table 4).
Acyclic 1,2-diketones like 2,3-butanedione (16) and 3,4-
hexanedione (18) proved to be excellent substrates for the
addition reaction. Cyclic 1,2-diketones like 1,2-cyclohexanedi-
one (20) were not good substrates for the addition reaction
(Table 4, entry 3). Inspection of the 1H NMR of 1,2-
cyclohexanedione showed that one of the ketones was com-
pletely enolic, unlike acyclic 1,2-diketones. This imparts
significantly lower reactivity in comparison to acyclic 1,2-
diketones. 1,2-Ketoesters like ethyl pyruvate (1) are also reactive
Further optimization of the reaction conditions was performed
with 2-(di-tert-butylphosphino)biphenyl 12 instead of tri-tert-
butylphosphine 13. Due to its propensity to undergo oxidation,
tri-tert-butylphosphine can be difficult to handle and store over
long periods, whereas phosphine 12 is much more stable and
less prone to oxidation, being a white solid instead of a liquid.
A number of rhodium sources were also examined for their
ability to function as precatalysts in the reaction. While Rh-
(acac)(CO)2 provided an active catalyst, use of other rhodium-
(I) complexes (including Rh(acac)(C2H4)2, [RhCl(COD)]2, and
[RhCl(CO)2]2) did not provide any alkyne addition products.
Evidently both the carbon monoxide and the â-diketonate ligand
are required for catalytic activity. The ratio of alkyne to
electrophile was also evaluated. Using only 2 equiv of ethyl
pyruvate resulted in a 50% yield, while use of only 1 equiv
gave a 26% yield of propargyl alcohol 3. The alkyne addition
appears to require a high concentration of electrophile for high
yields. Decomposition of the alkyne was minimal, as significant
amounts of unreacted alkyne were recovered from these
reactions.
A number of other reaction parameters were varied during
the optimization of the reaction by using 4-pentyn-1-ol and ethyl
pyruvate as the test substrates. The reaction was found to
proceed to completion in 24 h at temperatures as low as 40 °C
with no reduction in yield. The reaction proceeded at room
temperature but after 24 h the yield was only 29%. Allowing
the reaction to proceed for 48 h increased the yield to 57%,
indicating that the catalyst was viable for extended periods at
room temperature but the addition reaction was sluggish.
Variation of the ratio of phosphine to transition metal indicated
that 3 equiv of phosphine ligand to rhodium precatalyst provided
the best yield of alkyne addition product (Table 2, entry 4).
Catalyst loadings of 3 mol % were easily tolerated, but the use
of 1 mol % led to a significant decrease in the yield of the
reaction. Increasing the amount of phosphine ligand at 3 mol
(58) Esteruelas, M. A.; Lahoz, F. J.; Onate, E.; Oro, L. A.; Rodriguez,
L.; Steinert, P.; Werner, H. Organometallics 1996, 15, 3436.
(59) Zargfarian, D.; Chow, P.; Taylor, N. J.; Marder, T. B. J. Chem.
Soc., Chem. Commun. 1989, 9, 540.
(60) Garcia Alonso, F. J.; Ho¨hn, A.; Wolf, J.; Otto, H.; Werner, H.
Angew. Chem., Int. Ed. Engl. 1985, 97, 406.
(61) Ahn, Y. M.; Yang, K.; Georg, G. I. Org. Lett. 2001, 3, 1411.
(62) Marder, T. B.; Zargarian, D.; Calabrese, J. C.; Herskovitz, T. H.;
Milstein, D. J. Chem. Soc., Chem. Commun. 1987, 1484.
9592 J. Org. Chem., Vol. 72, No. 25, 2007

Addition of Alkynes to Aldehydes and ActiVated Ketones
TABLE 4. Addition of 4-Pentyn-1-ol to Aldehydes, 1,2-Diketones,
and 1,2-Ketoesters
SCHEME 1
94% yield (Table 4, entry 13). More reactive alkyl aldehydes
provided good yields of the propargylic alcohol product except
for pivaldehyde 34, which was too sterically hindered and gave
only a 6% yield of addition product (Table 4, entry 11).
Additions were also performed on differentially substituted
1,2-diketones to evaluate the selectivity of the process. Addition
to 2,3-pentanedione 40 gave a 50:50 mixture of ketone regioi-
somers (Scheme 1). Addition to 1-phenyl-1,2-propanedione 43
gave a 62:38 mixture of ketone regioisomers favoring addition
of the alkyne to the ketone next to the phenyl ring. This result
is, in hindsight, not surprising, as the internal ketone cannot
enolize and is flanked by two electron-withdrawing groups,
which increases its electrophilicity.
Different alkynes were also evaluated as nucleophiles in the
addition reaction (Table 5). Simple alkyl alkynes like 1-octyne
(2) performed well, as did aryl alkynes and silyl protected
alkynes (Table 5, entries 1-3). The presence of a propargylic
alcohol was somewhat problematic with propargyl alcohol itself
providing only an 8% yield of product (Table 5, entry 4).
Protection of the alcohol as a silyl ether greatly improved the
yield, as did the use of a more hindered propargyl alcohol like
2,2-dimethylpropargyl alcohol 53 (Table 5, entries 5 and 6).
Moving the hydroxyl group further from the propargylic position
provided much higher yields of product (Table 5, entries 7 and
8). The reaction otherwise showed a remarkable tolerance to
functional groups present on the alkyne. Esters, nitriles, car-
boxylic acids, and imides were all well tolerated and provided
excellent yields of alkyne addition product. In the case of 10-
undecynoic acid 61, the addition product with 2,3-butanedione
was difficult to separate from the starting material, leading to
lower yields due to repeated chromatography. However, the use
of 3,4-hexanedione (18) provided 62 in 94% yield. One
equivalent of triethylamine (with respect to the carboxylic acid)
was also added to the reaction when an unprotected carboxylic
acid was used to keep the conditions neutral. Use of an alkyne
bearing an aldehyde was also explored, and this provided a 59%
yield of product (Table 5, entry 13). Addition of diyne 73 and
ethyl propiolate 71 to 2,3-butanedione was also evaluated.
Diynes47,51 and ethyl propiolate10,20,21,63,64 can be problematic
substrates in zinc-catalyzed systems, requiring specific reaction
conditions to mediate the addition reactions. The rhodium-
catalyzed conditions did not prove to be useful for these systems
a The reaction was performed at 60 °C.
in the rhodium-catalyzed 1,2-addition. Increasing the size of
the alkane next to the ketone did result in lower yields, however
(Table 4, entries 6 and 7). Aldehydes were also investigated as
substrates for the addition reaction. Aromatic aldehydes like
benzaldehyde (36) provided only poor yields of addition product
(Table 4, entry 12) unless the aromatic group was quite electron
poor, as in the case of 4-nitrobenzaldehyde 38, which gave a
(63) Gao, G.; Wang, Q.; Yu, X.-Q.; Xie, R.-G.; Pu, L. Angew. Chem.,
Int. Ed. 2006, 45, 122.
(64) Meta, C. T.; Koide, K. Org. Lett. 2004, 6, 1785.
J. Org. Chem, Vol. 72, No. 25, 2007 9593

Dhondi et al.
TABLE 5. Addition of Alkynes to 2,3-Butanedione
SCHEME 2
the aldehydes have a chirality center next to the carbonyl
(Scheme 2). Addition of phenylacetylene to glyceraldehyde
acetonide 75 gave a 81:19 mixture of diastereomers favoring
the shown diastereomer 76 (Scheme 2, in all cases the major
isomer is shown). The stereochemistry of the major product was
confirmed by Mosher’s acid analysis65 after separation of the
diastereomers by HPLC (see the Supporting Information for
details). This ratio is significantly higher than when the same
addition was performed with the lithium acetylide at -78 °C,66
perhaps due to the larger size of the rhodium complex. Addition
of phenylacetylene to Garner’s aldehyde 77 was even more
selective, providing a single addition product 78. The selectivity
of these additions proved to be highly dependent on the chiral
aldehyde used as a coupling partner, as use of the protected
lactate 79 gave only a 62:38 mixture of diastereomers and
aldehyde 82 provided only a 54:46 mixture. Additionally,
without a nearby electronegative atom to activate the aldehyde,
the yield of the addition reaction decreased to only 32%, as in
the case of aldehyde 82. The selectivity of these additions can
be rationalized by using the Felkin-Anh model67,68 for asym-
metric induction except for aldehyde 82, where the selectivity
is quite poor.
Attempts were made to render the rhodium-catalyzed alkyne
addition process enantioselective through the use of an enan-
tiomerically pure chiral phosphine ligand. With few bulky
electron-rich derivatives of chiral phosphines readily available,
a number of chiral triaryl phosphines were screened for
enantioselectivity, with the synthesis of a dialkyl analogue to
be pursued in the event good enantioselectivity was obtained
(Table 6, see the Supporting Information for a complete list).
Only some of these reactions gave enough product for analysis
and none gave high asymmetric induction (Table 6, entries 1-3).
After an evaluation of 2,3-butanedione 16 as an electrophile,
there was concern that the poor enantioselectivity was related
to an inability to differentiate the methyl and methyl ketone
a 3,4-Hexanedione was used instead of 2,3-butanedione. b One equivalent
of Et3N was added.
(65) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.
(66) Kang, S. H.; Kim, W. J. Tetrahedron Lett. 1989, 30, 5915.
(67) Ahn, N. T. Top. Curr. Chem. 1980, 88, 145.
(68) Cherest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968, 18,
4199.
(Table 5, entries 14 and 15), as conversions were poor and the
reaction mixtures were complicated by numerous side products.
A number of chiral aldehyde substrates was also evaluated
to determine the degree of diastereoselectivity in cases where
9594 J. Org. Chem., Vol. 72, No. 25, 2007

Addition of Alkynes to Aldehydes and ActiVated Ketones
TABLE 6. Use of Chiral Phosphines
Formation of complex 88 was verified by mixing Rh(acac)-
(CO)2 and phosphine 12 together in benzene-d6 and observing
the reaction with 31P NMR and IR. The 31P NMR spectra of
the reaction showed two signals: one for unbound phosphine
(a singlet at 19.4 ppm) and one for complex 88 (a doublet at
79.9 ppm). Multiplicity was observed because 103Rh (the single
naturally occurring isotope of rhodium) is NMR active with a
spin of 1/2, the coupling constant (JRh-P ) 182 Hz) being
consistent with other rhodium-phosphine complexes.70 The IR
spectra of the reaction showed the disappearance of the carbonyl
peaks for complex 87 (two absorbances at 2082 cm-1 and 2011
cm-1) and the appearance of a new signal at 1957 cm-1, also
consistent with 88.70 Infrared absorbances at 1590 cm-1 and
1523 cm-1 showed the â-diketonate ligand was still bound to
the rhodium.
entry electrophile alkyne
liganda
yield (%) ee (%)
1
2
3
4
5
16
16
16
38
38
63
63
63
14
14
84 (6 mol %)
69
13
5
1
12
20
27
25
85 (4.5 mol %)
86 (4.5 mol %)
85 (4.5 mol %)
86 (4.5 mol %)
17
10
a Ligand structures.
After formation of the rhodium-phosphine complex, the
alkyne coordinates leading to complex 89. Evidence for this
coordination was obtained from the 1H NMR of phenylacetylene
taken in the presence of Rh(acac)(CO)2 and phosphine 12. The
1H NMR of phenylacetylene typically shows a peak at 2.72 ppm
for the alkyne C-H proton in benzene-d6. This proton shifts
downfield to 2.96 ppm in the presence of rhodium complex
88. The chemical shift change is indicative of a reversible
binding of the alkyne to the transition metal complex.2,71 Alkyne
complex 89 was the last complex that can be observed
spectroscopically, as addition of an electrophilic carbonyl leads
directly to product formation without any intermediates being
observed by NMR. The lifetime of these intermediates may be
too short or the amount generated under the catalytic conditions
may be too small to be observed with NMR.
groups on either side of the reactive ketone. Additional
experiments were therefore performed with 4-nitrobenzaldehyde
38, where differentiation of the enantiotopic faces should be
less problematic. While the enantioselectivity improved with
use of this substrate, they were still far from useful.
Coordination of the alkyne to the rhodium complex results
in the alkyne C-H becoming much more acidic.72 The excess
phosphine in the reaction can then act as a base, removing the
proton from the alkyne to form a rhodium(I) acetylide such as
90. Deprotonation likely occurs from an equivalent of free
phosphine, followed by dissociation of the phosphine bound to
the rhodium (3 equiv of phosphine with respect to the rhodium
are present). Coordination of the electrophile to the rhodium
complex can then lead to a complex such as 91, in which the
carbonyl is activated by coordination to the rhodium center. The
distal relationship between the phosphine and the electrophile
provides an explanation as to why such low enantioselectivities
are observed with chiral phosphines. The low enantioselectivities
can be rationalized by interactions with the chiral counterion,
which can enforce some stereocontrol.73 Bond formation
between the acetylide and the carbonyl leads to rhodium
alkoxide 92, which is then protonated by the phosphonium salt
to provide propargylic alcohol product 93 and regenerate the
resting state of the catalyst, complex 88. Given the difficulties
associated with engineering a phosphine that can control the
enantioselectivity of the alkyne addition, we have changed our
focus to modifying the â-diketonate ligand, which may be more
fruitful.
FIGURE 1. Mechanistic hypothesis for the alkyne addition reaction.
The poor enantioselectivity resulting from the use of chiral
phosphines prompted further investigation into the mechanism
of the reaction (Figure 1). Displacement of one carbon monoxide
ligand from Rh(acac)(CO)2 by a phosphine is the first step in
the reaction mechanism, forming the rhodium-phosphine
complex 88. This step is a well-precedented reaction of this
rhodium complex in the presence of phosphine ligands.69,70
(70) Pruchnik, F. P.; Smolenski, P.; Wajda-Hermanowicz, K. J. Orga-
nomet. Chem. 1998, 570, 63.
(71) De Koning, A. J.; Van Rijn, P. E.; Boersma, J.; Van der Kerk, G.
J. M. J. Organomet. Chem. 1979, 174, 129.
(72) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. In
AdVanced Inorganic Chemistry; Wiley: New York, 1999; p 681.
(73) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. Science 2007,
317, 496.
(69) Serron, S.; Huang, J.; Nolan, S. P. Organometallics 1998, 17, 534.
J. Org. Chem, Vol. 72, No. 25, 2007 9595

Dhondi et al.
provided propargyl alcohol 13 (62 mg) in 91% yield. TLC Rf 0.31
Conclusions
(50% ethyl acetate/CH2Cl2). IR (neat) 3415, 2936, 2243, 1723 cm-1
.
1H NMR (300 MHz, CDCl3) δ 4.16 (s, 1H), 3.69 (t, J ) 6.1 Hz,
2H), 2.36 (s, 3H), 2.33 (t, J ) 7.0 Hz, 2H), 2.08 (s, 1H), 1.73 (p,
J ) 6.6 Hz, 2H), 1.56 (s, 3H). 13C NMR (75 MHz, CDCl3) δ 206.2,
85.5, 79.6, 72.2, 60.3, 30.5, 26.4, 23.0, 14.7. Anal. Calcd for
C9H14O3: C, 63.51; H, 8.29. Found: C, 63.51; H, 8.22.
In summary, a rhodium-catalyzed method for the addition of
alkynes to aldehydes, 1,2-diketones, and 1,2-ketoesters under
mild conditions has been described. The reaction tolerates the
presence of a variety of functional groups, including unprotected
alcohols and carboxylic acids. Additions to R-chiral aldehydes
usually provide the Felkin product in moderate to high selectivity
depending on the substrate. Further studies to elucidate the
reaction mechanism, the development of an enantioselective
variant, and application of this reaction to the synthesis of
bioactive substances are currently underway.
1-(4-Nitrophenyl)hex-2-yne-1,6-diol (39). A solution of 4-pen-
tyn-1-ol (14) (34.7 mg, 0.4 mmol) in 0.5 mL of THF was added to
a mixture of 4-nitrobenzaldehyde 38 (222 mg, 1.2 mmol), Rh(acac)-
(CO)2 (0.012 mmol, 3.1 mg, 3 mol %), and 2-(di-tert-butylphos-
phino)biphenyl 12 (0.036 mmol, 10.9 mg, 9 mol %) under an argon
atmosphere. The reaction tube was then placed in a preheated oil
bath at 40 °C and stirred for 24 h. The reaction mixture was then
preadsorbed on silica gel and purified by silica gel chromatography
(6-18% acetone/CH2Cl2). This sequence provided propargyl
alcohol 39 (88 mg) in 94% yield. Mp 53.5-54.5 °C (CHCl3); TLC
Rf 0.32 (50% ethyl acetate/CH2Cl2). IR (neat) 3347, 2949, 2227,
1606, 1519 cm-1. 1H NMR (300 MHz, CDCl3) δ 8.22 (d, J ) 8.7
Hz, 2H), 7.70 (d, J ) 8.9 Hz, 2H), 5.53 (s, 1H), 3.75 (t, J ) 6.1
Hz, 2H), 2.95 (s, 1H), 2.40 (dt, J ) 7.0, 2.1 Hz, 2H), 1.78 (p, J )
6.5 Hz, 2H). 13C NMR (75 MHz, CDCl3) δ 148.4, 147.7, 127.5,
123.8, 87.7, 79.9, 63.6, 61.5, 30.9, 15.5. Anal. Calcd for C12H13-
NO4: C, 61.27; H, 5.57; N, 5.95. Found: C, 61.22; H, 5.72; N,
6.11.
Experimental Section
2-Hydroxy-2-methyl-dec-3-ynoic Acid Ethyl Ester (3). A
solution of octyne (2) (45 mg, 0.4 mmol) and ethyl pyruvate 1
(142 mg, 1.2 mmol) in 0.5 mL of 1,4-dioxane was added to a
mixture of Rh(acac)(CO)2 (5.2 mg, 0.02 mmol, 5 mol %) and 2-(di-
tert-butylphosphino)biphenyl 12 (12 mg, 0.04 mmol, 10 mol %)
under argon. The reaction was then heated to 90 °C. After 24 h at
90 °C the reaction mixture was allowed to cool to room temperature,
diluted with CH2Cl2, preadsorbed on silica gel, and purified by silica
gel chromatography (80%CH2Cl2/hexane-0.7%acetone/CH2Cl2) pro-
viding 75 mg of 3 (83% yield) as a thick oil. TLC Rf 0.23 (0.7%
acetone/CH2Cl2). IR (neat) 3502, 2934, 2860, 2250, 1739 cm-1
.
12-Ethyl-12-hydroxy-13-oxo-pentadec-10-ynoic Acid (62). A
solution of triethylamine (81 mg, 0.8 mmol) in 0.5 mL of THF,
followed by a solution of 3,4-hexanedione (18) (288.4 mg, 2.4
mmol, in 0.5 mL of THF) was added to a mixture of 10-undecynoic
acid (61) (153.5 mg, 0.8 mmol), Rh(acac)(CO)2 (6.3 mg, 0.024
mmol, 3 mol %), and 2-(di-tert-butylphosphino)biphenyl 12 (21.7
mg, 0.072 mmol, 9 mol %) under an argon atmosphere. The reaction
was then warmed to 40 °C. The reaction was quenched after 24 h
by the addition of 0.5 mL of glacial acetic acid. The reaction mixture
was then preadsorbed on silica gel and purified by silica gel
chromatography (5% acetone/CH2Cl2) providing alcohol 51 (224
mg) in 94% yield. TLC Rf 0.20 (4% acetone/CH2Cl2). IR (neat)
1H NMR (300 MHz, CDCl3) δ 4.28 (q, J ) 7.1 Hz, 2H), 3.42 (s
1H), 2.18 (t, J ) 7.0 Hz, 2H), 1.64 (s, 3H), 1.53-1.42 (m, 2H),
1.31 (t, J ) 7.1 Hz, 3H), 1.40-1.19 (m, 2H), 0.87 (t, J ) 6.7 Hz,
3H). 13C NMR (75 MHz, CDCl3) δ 173.4, 85.2, 80.0, 68.1, 62.9,
31.5, 28.6, 28.5, 27.5, 22.7, 18.8, 14.2. Anal. Calcd for C13H22O3:
C, 68.99; H, 9.80. Found: C, 68.67; H, 9.74.
2,7-Dihydroxy-2-methyl-hept-3-ynoic Acid Ethyl Ester (15).
A solution of 4-pentyn-1-ol (14) (34 mg, 0.4 mmol) and ethyl
pyruvate (1) (142 mg, 1.2 mmol) in 0.5 mL of THF was added to
a mixture of Rh(acac)(CO)2 (3.1 mg, 0.012 mmol, 3 mol %) and
2-(di-tert-butylphosphino)biphenyl 12 (11 mg, 0.036 mmol, 9 mol
%) under argon. The reaction mixture was then warmed to 40 °C.
After 24 h at 40 °C the mixture was allowed to cool to room
temperature, diluted with CH2Cl2, preadsorbed on silica gel, and
purified by silica gel chromatography (30-50% ethyl acetate/CH2-
Cl2) providing 75 mg of propargyl alcohol 15 (94% yield). TLC Rf
0.39 (50% ethyl acetate/CH2Cl2). IR (neat) 3449, 2985, 2940, 2248,
1
3438, 2936, 2857, 2234, 1716, 1646 cm-1. H NMR (300 MHz,
CDCl3) δ 2.86 (dq, J ) 18.0, 7.3 Hz, 1H), 2.48 (dq, J ) 18.0, 7.3
Hz, 1H), 2.25 (t, J ) 7.4 Hz, 2H), 2.13 (t, J ) 7.0 Hz, 2H), 1.87
(sextet, J ) 7.3 Hz, 1H), 1.67 (sextet, J ) 7.0 Hz, 1H), 1.56-1.49
(m, 2H), 1.46-1.37 (m, 2H), 1.22 (m, 8H), 1.05 (t, J ) 7.3 Hz,
3H), 0.85 (t, J ) 7.4 Hz, 3H) (the protons for the alcohol and the
carboxylic acid had exchanged with deuterium in the spectra for
this compound, and are therefore not reported). 13C NMR (75 MHz,
CDCl3) δ 209.7, 179.7, 86.8, 79.0, 76.0, 34.0, 33.2, 29.1, 29.0,
28.9, 28.8, 28.7, 28.3, 24.6, 18.6, 8.1, 7.8. Anal. Calcd for
C17H28O4: C, 68.89; H, 9.52. Found: C, 68.77; H 9.46.
1
1741 cm-1. H NMR (300 MHz, CDCl3) δ 4.16 (q, J ) 7.1 Hz,
2H), 3.59 (t, J ) 6.2 Hz, 2H), 2.21 (t, J ) 7.0 Hz, 2H), 1.63 (p, J
) 6.6 Hz, 2H), 1.53 (s, 3H), 1.21 (t, J ) 7.1 Hz, 3H) (the protons
for both alcohols had exchanged with deuterium in the spectra for
this compound, and are therefore not reported). 13C NMR (75 MHz,
CDCl3) δ 172.9, 84.2, 80.3, 67.9, 62.5, 60.9, 30.8, 37.4, 15.1, 13.9.
Anal. Calcd for C10H16O4: C, 59.98; H, 8.05. Found: C, 60.22; H,
8.11.
Acknowledgment. The authors thank the donors of the
Petroleum Research Fund, administered by the American
Chemical Society, and Syracuse University for partial support
of this research. L.B.C. thanks Pfizer for funding during the
summer of 2005.
3,8-Dihydroxy-3-methyl-oct-4-yn-2-one (17). A solution of 2,3-
butanedione (16) (104.4 mg, 1.2 mmol) and 4-pentyn-1-ol (14) (34.7
mg, 0.4 mmol) in 0.5 mL of THF was added to a mixture of Rh-
(acac)(CO)2 (0.012 mmol, 3.1 mg, 3 mol %) and 2-(di-tert-
butylphosphino)biphenyl 12 (0.036 mmol, 10.9 mg, 9 mol %) under
an argon atmosphere. The reaction tube was then placed in a
preheated oil bath at 40 °C and stirred for 24 h. The reaction mixture
was then preadsorbed on silica gel and purified by silica gel
chromatography (30-50% ethyl acetate/CH2Cl2). This sequence
Supporting Information Available: Experimental procedures
and spectroscopic data for new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO701643H
9596 J. Org. Chem., Vol. 72, No. 25, 2007