Zwitterionic phosphonium sulfonates as easily phase-separable ion-tagged wittig reagents

Article abstract of DOI:10.1021/jo801418a

(Chemical Equation Presented) Zwitterionic phosphonium sulfonates 3, conveniently derived from TPPMS (1), can be used as Wittig reagents in solution. The excess reagents and byproduct TPPMSO (6) can be easily separated from the product alkenes by simple precipitation with a less polar solvent. The alkenes thus obtained were often sufficiently pure without chromatographic purification. A one-pot protocol for the synthesis of α,β-unsaturated esters has been developed and appears to be convenient.

Full text of DOI:10.1021/jo801418a

SCHEME 1. Synthesis of Zwitterionic Phosphonium
Sulfonates
Zwitterionic Phosphonium Sulfonates as Easily
Phase-Separable Ion-Tagged Wittig Reagents
Congde Huo, Xun He, and Tak Hang Chan*
Department of Chemistry, McGill UniVersity, Montreal,
Quebec, Canada H3A 2K6
tak-hang.chan@mcgill.ca
ReceiVed June 30, 2008
SCHEME 2. Wittig Reactions of Zwitterionic Phosphonium
Sulfonates 3a-d
Zwitterionic phosphonium sulfonates 3, conveniently derived
from TPPMS (1), can be used as Wittig reagents in solution.
The excess reagents and byproduct TPPMSO (6) can be
easily separated from the product alkenes by simple pre-
cipitation with a less polar solvent. The alkenes thus obtained
were often sufficiently pure without chromatographic puri-
fication. A one-pot protocol for the synthesis of R,ꢀ-
unsaturated esters has been developed and appears to be
convenient.
The Wittig reaction is an important reaction in organic
synthesis. The separation of the product alkene from the
byproduct triphenylphosphine oxide (Ph3PO) is a classical
problem, requiring tedious chromatography or recrystallization.
To overcome this problem, polymer-bound10 or fluorous-
tagged11 phosphines have been developed. We report here our
study on the use of ion-tagged Wittig reagents.
Most ion tag-supported reagents used imidazolium8 or phos-
phonium9 ions as the support. Because the sodium salt of
triphenylphosphine-m-sulfonate (TPPMS, 1) is used industri-
ally,12 we first examined the ionic salt 1,2-dimethyl-3-butyl-
imidazolium triphenylphosphine-m-sulfonate (2), prepared from
the reaction of 1 with 1,2-dimethyl-3-butylimidazolium bromide.
Reaction of 2 with benzyl tosylate gave the zwitterionic
phosphonium salt 3a together with 1,2-dimethyl-3-butylimida-
zolium tosylate. It became obvious to us that the imidazolium
moiety is redundant as the same zwitterionic compound 3a can
be prepared from 1 and benzyl bromide (Scheme 1). Compounds
3b-d were prepared similarly from the corresponding bro-
mides.13
The Wittig reaction of 3a with various carbonyl compounds
was evaluated in different base/solvent conditions (Scheme 2),
and the results are summarized in Table 1. While NaOH in water
was able to effect the reaction between 3a and 4-nitrobenzal-
dehyde (4a) in good yield; NaOH in methanol was generally
more effective for all aldehydes tested. The separation of the
In the past few decades, considerable efforts have been
devoted toward the development of supports to bind catalysts,
reagents, or scavengers to facilitate the purification process after
a synthetic reaction. Following the success of Merrifield peptide
synthesis,1 insoluble solid polymer resin was soon used as a
support for reagents and catalysts.2,3 Recent efforts have been
directed toward the use of soluble polymers4-6 or fluorous-
phase synthesis7 to restore the homogeneous reaction conditions.
In these cases, the phase separation depends on the difference
in the molecular weight of the support from the product or on
the affinity of the fluorous tag for fluorous solvents. Recently,
the use of ion tag as soluble support for organic synthesis has
been explored.8,9 phase separation depends on the differential
solubility of the ionic moiety in polar versus nonpolar solvents.
(1) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149.
(2) (a) Guillier, F.; Orain, D.; Bradley, M. Chem. ReV. 2000, 100, 2091. (b)
James, I. W. Tetrahedron 1999, 55, 4855. (c) Ley, S. V.; Baxendale, I. R. Chem.
Rec. 2002, 2, 377.
(3) Smith, C. D.; Baxendale, I. R.; Tranmer, G. K.; Baumann, M.; Smith,
S. C.; Lwethwaite, R. A.; Ley, S. V. Org. Biomol. Chem. 2007, 5, 1562.
(4) Bergbreiter, D. E. Chem. ReV. 2002, 102, 3345.
(9) (a) Poupon, J.-C.; Boezio, A. A.; Charette, A. B. Angew. Chem., Int. Ed.
2006, 45, 1415. (b) Stazi, F.; Marcoux, D.; Poupon, J.-C.; Latassa, D.; Charrette,
A. B. Angew. Chem., Int. Ed. 2007, 46, 5011.
(10) Bolli, M. H.; Ley, S. V. J. Chem. Soc. Perkin Trans. 1 1998, 2243, and
references cited therein.
(11) Galante, A.; Lhoste, P.; Sinou, D. Tetrahedron Lett. 2001, 42, 5425.
(12) (a) Cornils, B.; Kuntz, E. G. J. Organomet. Chem. 1995, 502, 177. (b)
Baricelli, P. J.; Baricelli, D.; Lujano, E.; Melean, L. G.; Borusiak, M.; Lopez-
Linares, F.; Bastidas, L. J.; Rosales, M. J. Mol. Catal. A 2007, 271, 180.
(13) While NaBr was formed in the formation of 3, its presence appeared to
be innocuous as it had no significant influence to the Wittig reaction.
(5) Toy, P. H.; Janda, K. D. Acc. Chem. Res. 2000, 33, 546.
(6) Charette, A. B.; Boezio, A. A.; Janes, M. K. Org. Lett. 2000, 2, 3777.
(7) (a) Horva´th, I. T.; Ra´bai, J. Science 1994, 266, 72. (b) Studer, A.; Hadida,
S.; Ferritto, S. Y.; Kim, P. Y.; Jeger, P.; Wipf, P.; Curran, D. P. Science 1997,
275, 823. For reviews, see: (c) Zhang, W. Tetrahedron 2003, 59, 4475; Chem.
ReV. 2004, 104, 2531.
(8) (a) Fraga-Dubreuil, J.; Bazureau, J. P. Tetrahedron Lett. 2001, 42, 6097.
(b) Miao, W.; Chan, T. H. Org. Lett. 2003, 5, 5003. (c) Miao, W.; Chan, T. H.
Acc. Chem. Res. 2006, 39, 897.
10.1021/jo801418a CCC: $40.75
Published on Web 09/27/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 8583–8586 8583

TABLE 1. Wittig Reactions of 3a with Various Aldehydes
SCHEME 3. Synthesis of Methylated Resveratrol
TABLE 2. Wittig Reactions of 3c with Various Aldehydesa
a Base/solvent: K2CO3/MeOH.
as benzophenone, acetophenone, cyclohexanone, and acetone,
were found to be unreactive under the conditions and were
recovered quantitatively. 4-acetylbenzaldehyde (4i) reacted with
3c to give the alkene 5q in excellent yield, demonstrating high
chemoselectivity. Again, the separation of the product alkenes
5 from the byproduct phosphine oxide 6 (TPPMSO) was
achieved by adding diethyl ether to allow precipitation of the
phosphine oxide 6. Furthermore, it was not necessary to prepare
the phosphonium salt 3c as a separate step in this reaction. By
simply mixing TPPMS (1), methyl bromoacetate, potassium
carbonate, and the aldehyde 4 together in methanol and stirring
the mixture at room temperature, the R,ꢀ-unsaturated ester 5
was obtained in excellent yield with high purity containing little
byproduct according to their 1H NMR without chromatography
(see the Supporting Information). This one-pot reaction appears
to be as convenient as the well established Horner-Wads-
worth-Emmons (HWE) modification.15 On the other hand, the
HWE reaction is stereoselective in giving the thermodynamically
more stable E-R,ꢀ-unsaturated esters, whereas the present
reaction gives a mixture of E- and Z-isomers. This lack of
stereoselectivity could be useful if the Z-isomer is the desired
product or if a mixture of stereoisomers is required for the
biological activity.16 Furthermore, the mixture of stereoisomers
could be isomerized to the thermodynamically more stable
E-isomer.17We demonstrated this ease of conversion with the
synthesis of E-5l. After the one-pot reaction of TPPMS, methyl
bromoacetate and benzaldehyde was completed, the phosphine
oxide 6 was removed by filtration as before. The crude product
product alkenes 5 from the byproduct phosphine oxide 6
(TPPMSO) was very easy. After the reaction was completed, a
less polar solvent such as diethyl ether was added to the reaction
mixture to allow precipitation of the phosphine oxide 6. After
simple filtration, the organic layer was then free of 3a and 6, as
evident from the absence of signal in the 31P NMR. In most
cases, the product alkene 5 needed no further purification on
the basis of their 1H NMR. Under such conditions, trans-
cinnamic aldehyde (entry 6) was converted to the corresponding
diene in high yield, and aliphatic aldehyde such as hydrocin-
namaldehyde (entry 7) also worked well. There was no reaction
at all between 3a and ketones such as benzophenone (entry 8),
acetophenone, cyclohexanone, or acetone in MeOH/NaOH and
the ketones were recovered quantitatively. This reaction has
therefore a high selectivity for aldehydes. Thus, 4-acetylben-
zaldehyde (4i) reacted with 3a chemoselectively (entry 9) in
quantitative yield. Using compound 3b and 3, 5-dimethoxy-
benzaldehyde under the same reaction conditions, the methylated
resveratrol 5j was obtained in 90% yield (Scheme 3). Methylated
resveratrol 5j had been converted to the bioactive resveratrol.14
With the more acidic compound 3c, potassium carbonate
could be used to effect the Wittig reaction. As shown in Table
2, various aromatic and aliphatic aldehydes were converted to
the corresponding alkenes 5 in excellent yields. Ketones, such
(14) Several syntheses of 5j using the classical Wittig reaction of the
corresponding triphenylphosphonium salt of 3b have been reported. The reaction
required strong base (n-BuLi or t-BuOK) in THF at low temperature and had
yields ranging from 60% to 92% with E/Z ratios between 1:1 and 1:2.8. (a) Ali,
M. A.; Kondo, K.; Tsuda, Y. Chem. Pharm. Bull. 1992, 40, 1130–6. (b) Zhang,
W.; Go, M. L. Eur. J. Med. Chem. 2007, 42, 841. (c) Petitt, G. R.; Grealish,
M. P.; Jung, M. K.; Hamel, E.; Petitt, R. K.; Chapuis, J. C.; Schmidt, J. M.
J. Med. Chem. 2002, 45, 2534. (d) Yu, J.; Gaunt, M. J.; Spencer, J. B. J. Org.
Chem. 2002, 67, 4624.
(15) Kurti, L., Czako, B., Eds. Strategic Applications of Named Reactions
in Organic Synthesis; Elsevier Academic Press: Burlington, MA, 2005; pp 212-
3, and references cited therein.
(16) See, for example: Chan, T. H.; Koumaglo, K. Tetrahedron Lett. 1986,
27, 883.
8584 J. Org. Chem. Vol. 73, No. 21, 2008

SCHEME 4. Reduction of TPPMSO (6)
can be used as Wittig reagents. The excess reagents and
byproduct TPPMSO (6) can be easily separated from the product
alkenes by simple precipitation with a less polar solvent such
as ether. The alkene products thus obtained were often suf-
ficiently pure according to their 1H NMR spectra to require no
chromatographic purification. The sulfonate moiety is robust
toward strongly basic conditions and resists reduction by silanes
and alanes. The TPPMSO (6) can be recycled back to TPPMS
(1) in high yield.
was dissolved in THF and 25 mol % of diphenyl disulfide16
was added and the mixture was refluxed overnight to give pure
E-5l.
With the less acidic phosphonium salt 3d, stronger base
was required to effect the Wittig reaction. Using lithium
hexamethyldisilazide in THF, 3d reacted with 4-nitroben-
zaldehyde (1a) to give 1-(4-nitrophenyl)pent-1-ene (5r) in
90% isolated yield with E/Z ) 2.1:1.0. Again, the separation
of the product alkene from the byproduct phosphine oxide
was achieved by ether precipitation.
Recently, the development of supported reagents for the
facilitation of the purification of organic reaction products is
questioned partly due to the fact that most of the reagents
reported in the literature are not commercially available and
require multistep synthetic sequences for their preparation. This
drawback limits the general practicability of these supported
reagents. In our method, the TPPMS is commercially available
and also can be prepared from TPP by one step. Relative to the
polymer-supported Wittig reagent,10 because of the low nominal
molecular weight of the sulfonate group (SO3, MW ) 80)
relative to that of most polymer support (MW g 2000), the
loading capacity of the ion tag is much higher. Relative to the
fluorous Wittig reagents,11 the zwitterionic phosphonium sul-
fonates are easily prepared and their Wittig reactions can be
conducted in a range of common solvents instead of perfluoro
solvents. While the use of water-soluble phosphonium salts with
carboxylic acid or phenol group for Wittig reactions had been
previously explored, it met with limited success.22 The Wittig
reaction appeared to be confined to benzyl-containing phos-
phonium bromides and there was no demonstration of the
recycling of the product phosphine oxides.21 It is our belief that
the current approach is a considerable improvement over these
related strategies. Finally, the one-pot protocol for the synthesis
of R,ꢀ-unsaturated esters appears to be convenient.
A key issue is whether the recovered TPPMSO (6) can be
recycled back to the phosphine sulfonate 1. Normally, triph-
enylphosphine oxide is reduced to the corresponding phosphine
with either (1) trichlorosilane (SiHCl3) in combination with or
without triethylamine or (2) LiAlH4 or its derivative such as
alane.18 In the present case, considerable difficulties were met
in the reduction of TPPMSO, presumably because of the electron
withdrawing effect of the sulfonate group. SiHCl3, SiHCl3/NEt3,
SiHCl3/PhNMe2 in toluene, xylene, and PhCN were tried, and
all gave poor conversion (about 10-20% with 1 equiv of
reduction reagent, based on the integrated signals in 31P NMR
of the crude product). If excess reducing reagent was used, the
conversion was increased somewhat but never reached more
than 50% and the product became complicated. LiAlH4, LiAlH4/
CeCl3 systems were also examined, and they did not offer any
improvement. Wyatt et al. reported19 the use of alane generated
in situ from LiAlH4 and H2SO4. Curran et al. had used a similar
reagent, an alane-N,N-dimethylamine complex, for reduction
of fluorous phosphine oxide.20 In our hands, none of these alane
reagents were suitable as they all gave poor yield of 1. Finally,
Spencer et al. recently reported the quantitative reduction of
arylphosphine oxides by using SiHCl3 in combination with
triphenylphosphine as oxygen acceptor.21 The addition of
triphenylphosphine allows the reduction to be performed under
milder conditions even for very electron deficient phosphine
oxides. This method was found to work for TPPMSO (6) giving
TPPMS (1) in excellent yield (Scheme 4). The reaction mixture
was quenched by NaOH solution followed by addition of
methanol. The solid silica gel derived from hydrolysis of the
chlorosilanes was removed by filtration. The filtrate was
concentrated and washed with ether. The TPPMS (1) was
obtained as white solid, identical with the authentic compound.
Triphenylphosphine was easily recovered from the ether solu-
tion. In this respect, the drawback mentioned by Spencer20
regarding the difficulty of separating triphenylphospine from
the product arylphosphine is avoided because of the insolubility
of TPPMS in ether.
Experimental Section
Typical Procedure for Preparation of Phosphonium Salts
3a-d. A mixture of TPPMS (1, 728 mg, 2 mmol) and a slight
excess of the corresponding bromide (2.4 mmol) was stirred at 50
°C overnight. Ether was added, and the precipitate was filtered to
give the target phosphonium salts as white solid.
1
3a. H NMR (400 MHz, DMSO-d6): δ 8.05 (d, J ) 7.2 Hz,
1H), 7.91-7.83 (m, 3H), 7.76-7.58 (m, 10H), 7.28-7.19 (m, 3H),
6.96 (d, J ) 7.2 Hz, 2H), 5.19 (d, J ) 16 Hz, 2H). 31P NMR (81
MHz, DMSO-d6): δ 23.3 (s). HRMS: m/z calcd for C25H22PO3S+
433.1022, found 433.1025.
1
3b. H NMR (400 MHz, DMSO-d6): δ 8.05 (d, J ) 7.2 Hz,
1H), 7.91-7.60 (m, 13H), 6.88 (d, J ) 7.2 Hz, 2H), 6.78 (d, J )
7.2 Hz, 2H), 5.11 (d, J ) 14.8 Hz, 2H), 3.67 (s, 3H). 31P NMR (81
MHz, DMSO-d6): δ 23.7 (s). HRMS: m/z calcd for C26H24PO4S+
463.1127, found 463.1125.
3c. 1H NMR (400 MHz, DMSO-d6): δ 8.06-7.72 (m, 14H), 5.40
(d, J ) 14.4 Hz, 2H), 3.59 (s, 3H). 31P NMR (81 MHz, DMSO-
d6): δ 25.4 (s). HRMS m/z calcd for C21H20PO5S+ 433.0764, found
433.0767.
In conclusion, we have demonstrated that zwitterionic phos-
phonium sulfonates 3, conveniently derived from TPPMS (1),
1
3d. H NMR (400 MHz, DMSO-d6): δ 8.05 (d, J ) 7.6 Hz,
1H), 7.91-7.73 (m, 13H), 3.06 (m, 2H), 1.47 (m, 4H), 0.87 (t, J )
6.4 Hz, 3H). 31P NMR (81 MHz, DMSO-d6): δ 23.3 (s). HRMS:
m/z calcd for C22H24PO3S+ 399.1178, found 399.1181.
(17) For examples of Z- to E-isomerization, see: (a) Miyata, O.; Shinada,
T.; Ninomiya, I.; Naito, T. Synthesis 1990, 1123. (b) Baag, M.; Kar, A.; Argade,
N. P. Tetrahedron 2003, 59, 6489. (c) Kim, I. S.; Dong, G. R.; Jung, Y. H. J.
Org. Chem. 2007, 72, 5424.
(18) Leyva, A.; Garcia, H.; Corma, A. Tetrahedron 2007, 63, 7097.
(19) (a) Bootle-Wilbraham, A.; Head, S.; Longstaff, J.; Wyatt, P. Tetrahedron
Lett. 1999, 40, 5267. (b) Griffin, S.; Heath, L.; Wyatt, P. Tetrahedron Lett. 1998,
39, 4405.
Typical Procedure for Wittig Reactions Using Phosphonium
Salts 3a and 3b. To phosphonium salt 3a or 3b (0.2 mmol)
(22) The sulfonate group is more robust than, for example, the carboxylic
acid group to reduction by alane or silane. See: (a) Russell, M. G.; Warren, S.
Tetrahedron Lett. 1998, 39, 7995. (b) Russell, M. G.; Warren, S. J. Chem. Soc.,
Perkin Trans. 1 2000, 505.
(20) Dandapani, S.; Curran, D. P. Tetrahedron 2002, 58, 3855.
(21) Wu, H.-C.; Yu, J.-Q.; Spencer, J. B. Org. Lett. 2004, 6, 4675.
J. Org. Chem. Vol. 73, No. 21, 2008 8585

suspended in methanol (1 mL) was added NaOH (0.25 mmol). After
5 min, the appropriate aldehyde (0.2 mmol) was added. The mixture
was stirred at room temperature overnight. Ether (3 mL) was added
to precipitate 6. After filtration, the filtrate was evaporated to give
the alkene product 5.
The alkenes 5a-j (Table 1 and Scheme 3) are known com-
pounds: 5a;23 5b-e;24 5f;25 5g;26 5i;27 5j.28 They were properly
characterized and found to be in agreement with literature reports.
Typical Procedure for Wittig Reaction Using Phosphonium Salt
3c. Method 1. To phosphonium salt 3c (104 mg, 0.2 mmol)
suspended in methanol (1 mL) was added K2CO3 (0.25 mmol).
After 5 min, the appropriate aldehyde (4, 0.2 mmol) was added.
The reaction mixture was stirred at room temperature overnight.
Ether (3 mL) was added to precipitate 6. After filtration, the filtrate
was evaporated to give the alkene product 5.
was completed. The mixture was purified by chromatography to
give the pure E-5l in 90% yield.
Typical Procedure for Wittig Reaction Using Phosphonium Salt
3d. To phosphonium salt 3d (100 mg, 0.2 mmol) suspended in THF
(1 mL) was added LiHMDS (0.2 mmol in THF). After 5 min, the
appropriate aldehyde (0.2 mmol) was added. The reaction mixture
was stirred at room temperature overnight. Ether (3 mL) was added
to give precipitate. The mixture was filtered, and the filtrate was
evaporated to give corresponding alkene product 5r, with properties
in agreement with literature report.33
Reduction of TPPMSO (6) Using Trichlorosiland and Triphenyl-
phosphine. In a 50 mL pressure tube, the phosphine oxide 6 (200
mg, 0.52 mmol) and triphenylphosphine (274 mg, 1.05 mmol) were
suspended in toluene (10 mL) under argon atmosphere. To the
mixture was added trichlorosilane (1 mL, 10 mmol) at room
temperature. The mixture was stirred at 110 °C overnight. The
mixture was cooled to ambient temperature. The mixture was
quenched with NaOH (2 mL, 20 wt %), and MeOH (25 mL) was
added subsequently. The mixture was filtrated by a thin pad of
Celite. The liquid phase was concentrated, and MeOH (25 mL)
was added again. The solution was dried (Na2SO4) and concentrated
under reduced pressure. The residue was washed with ether (3 ×
20 mL) to give the TPPMS (1) as white solid (170 mg, 90% yield).
TPPMS (1): 1H NMR (400 MHz, CD3OD) δ 7.85-7.81 (m, 2H),
7.43-7.39 (m, 1H), 7.37-7.26 (m, 11H); 31P NMR (81 MHz,
CD3OD) δ -4.07 (s).
Method 2. TPPMS (1, 73 mg, 0.2 mmol), methyl bromoacetate
(31 mg, 0.2 mmol), K2CO3 (0.25 mmol), and the appropriate
aldehyde (4, 0.2 mmol) were all added to methanol (1 mL) and the
mixture stirred overnight. The reaction was worked up as above to
give alkene product 5.
The alkenes 5k-q (Table 2) are known compounds: 5k-n;29
5o;30 5p;31 5q.32
Isomerization for the synthesis of E-5l. A mixture of E- and Z-5l
was prepared according to method 2. After addition of ether and
filtration to remove the precipitated 6, the filtrate was concentrated
and the crude product dissolved in anhydrous THF (2 mL). Diphenyl
disulfide (11 mg, 25 mol%) was added, and the mixture was
refluxed under argon overnight. NMR showed that the isomerization
1
TPPMSO (6): H NMR (400 MHz, CD3OD) δ 8.13-8.07 (m,
2H), 7.81-7.75 (m, 1H), 7.69-7.62 (m, 7H), 7.58-7.53 (m, 4H);
31P NMR (81 MHz, CD3OD) 32.6 (s).
(23) Yao, Q.; Zabawa, M.; Woo, J.; Zheng, C. J. Am. Chem. Soc. 2007,
129, 3088.
(24) Cella, R.; Stefani, H. A. Tetrahedron 2006, 62, 5656.
(25) Alacid, E.; Najera, C. AdV. Synth. Catal. 2006, 348, 2085.
(26) Habrant, D.; Stengel, B.; Meunier, S.; Mioskowski, C. Chem.sEur. J.
2007, 13, 5433.
Acknowledgment. We thank Merck Frosst Canada and the
Natural Science and Engineering Research Council (NESERC)
of Canada for financial support.
(27) Olodenko, W.; Mennecke, K.; Vogt, C.; Gruhl, S.; Kirschning, A.
Synthesis 2006, 1873.
(28) Heynekamp, J. J.; Weber, W. M.; Hunsaker, L. A.; Gonzales, A. M.;
Orlando, R. A.; Deck, L. M.; Vander Jagt, D. L. J. Med. Chem. 2006, 49, 7182.
(29) Zhang, Z.; Wang, Z. J. Org. Chem. 2006, 71, 7485.
(30) Xiong, Z.; Wang, N.; Dai, M.; Li, A.; Chen, J.; Yang, Z. Org. Lett.
2004, 6, 3337.
Supporting Information Available: Copies of NMR spectra
of reagents and product alkenes. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO801418A
(31) Nunez, M. T.; Martin, V. S. J. Org. Chem. 1990, 55, 1928.
(32) Yang, D.; Chen, Y.-C.; Zhu, N.-Y. Org. Lett. 2004, 6, 1577.
(33) Kabalka, G. W.; Li, N.-S.; Tejedor, D.; Malladi, R. R.; Trotman, S. J.
Org. Chem. 1999, 64, 3157.
8586 J. Org. Chem. Vol. 73, No. 21, 2008