Requirement for an oxidant in Pd/Cu co-catalyzed terminal alkyne homocoupling to give symmetrical 1,4-disubstituted 1,3-diynes

Article abstract of DOI:10.1021/jo048428u

(Chemical Equation Presented) Palladium-catalyzed terminal alkyne dimerization, through oxidative homocoupling, is a useful approach to the synthesis of symmetrical 1,4-diynes. Recent investigations have suggested that this reaction might be accomplished in the absence of intentionally added stoichiometric oxidants (to reoxidize Pd(0) to Pd(II)). In this paper, we have fully addressed the question of whether oxygen (or added oxidant) is required to facilitate this process. The presence of a stoichiometric quantity of air (or added oxidant such as I₂) is essential for alkyne dimerization. Excess PPh₃ inhibits alkyne dimerization to enyne, which only occurs to a significant extent when the reaction is starved of oxidant. Theoretical studies shed more light on the requirement for an oxidant in the homocoupling reaction in order for the process to be theromodynamically favorable. The employment of I₂ as the stoichiometric oxidant appears to be the method of choice. The dual role of Cu both in transmetalation of alkynyl units to Pd(II) and in assisting reoxidation of Pd-(0) to Pd(II) is suggested.

Full text of DOI:10.1021/jo048428u

SCHEME 1. Homocoupling of Terminal Alkynes
Requirement for an Oxidant in Pd/Cu
Co-Catalyzed Terminal Alkyne
Homocoupling To Give Symmetrical
1,4-Disubstituted 1,3-Diynes
Andrei S. Batsanov,† Jonathan C. Collings,†
Ian J. S. Fairlamb,*,‡ Jason P. Holland,‡
Judith A. K. Howard,† Zhenyang Lin,§
Todd B. Marder,*,† Alex C. Parsons,†
Richard M. Ward,† and Jun Zhu§
Department of Chemistry, University of Durham,
Durham DH1 3LE, U.K., Department of Chemistry,
University of York, Heslington, York YO10 5DD, U.K., and
Department of Chemistry, The Hong Kong University of
Science and Technology, Clear Water Bay, Kowloon,
Hong Kong, P. R. China
and molecular recognition processes.4 Here, palladium
catalysis has played a key role.5 Traditional methods for
oxidative homocoupling of terminal alkynes 1 to give
symmetrical 1,4-disubsituted diynes 2 (Scheme 1) include
Cadiot-Chodkiewicz,6 Eglington,7 Glaser,8 Hay,9 and
Sonogashira couplings.10 Catalytic systems mediated by
palladium, Pd(0) or Pd(II), are arguably the most mild,
efficient, and selective for the oxidative homocoupling
reactions of 1 f 2.11 Common side products in these
reactions are enynes E,Z-3 and 4, the former produced
through head-to-head coupling and the latter through
head-to-tail coupling.12
ijsf1@york.ac.uk; todd.marder@durham.ac.uk
Received September 7, 2004
Reported protocols for the oxidative homocoupling
include (1) use of (Ph3P)4Pd(0), CuI, Et3N, and chloroac-
etone (as the reoxidant) in benzene;13 (2) (Ph3P)2PdCl2,
CuI, Et3N, or DABCO and bromoacetate (as the reoxi-
dant) in THF;14 (3) (Ph3P)2PdCl2, CuI, and molecular
iodine (as the reoxidant) in i-Pr3N;15 (4) Pd(0)(dba)2 (dba
Palladium-catalyzed terminal alkyne dimerization, through
oxidative homocoupling, is a useful approach to the synthesis
of symmetrical 1,4-diynes. Recent investigations have sug-
gested that this reaction might be accomplished in the
absence of intentionally added stoichiometric oxidants (to
reoxidize Pd(0) to Pd(II)). In this paper, we have fully
addressed the question of whether oxygen (or added oxidant)
is required to facilitate this process. The presence of a
stoichiometric quantity of air (or added oxidant such as I2)
is essential for alkyne dimerization. Excess PPh3 inhibits
alkyne dimerization to enyne, which only occurs to a
significant extent when the reaction is starved of oxidant.
Theoretical studies shed more light on the requirement for
an oxidant in the homocoupling reaction in order for the
process to be theromodynamically favorable. The employ-
ment of I2 as the stoichiometric oxidant appears to be the
method of choice. The dual role of Cu both in transmetalation
of alkynyl units to Pd(II) and in assisting reoxidation of Pd-
(0) to Pd(II) is suggested.
(2) (a) Tour, J. M. Chem. Rev. 1996, 96, 537. (b) Martin, R. E.;
Diederich, F. Angew. Chem., Int. Ed. 1999, 38, 1350.
(3) (a) Bohlmann, F.; Burkhardt, T.; Zdero, C. In Naturally Occur-
ring Acetylenes; Academic Press: New York, 1973. (b) Jones, E. H. R.;
Thaller, V. In Handbook of Microbiology; Laskin, A. I., Lechevalier,
H. A., Eds.; CRC: Cleveland, 1973; Vol. 3, p 63. (c) Yamaguchi, M.;
Park, H.-J.; Hirame, M.; Torisu, K.; Nakamura, S.; Minami, T.;
Nishihara, H.; Hiraoka, T. Bull. Chem. Soc. Jpn. 1994, 67, 1717.
(4) (a) Breitenbach, J.; Boosfeld, J.; Vo¨gtle, F. In Comprehensive
Supramolecular Chemistry; Vo¨gtle, V., Ed.; Pergamon: Oxford, 1996;
Vol. 2, Chapter 2, pp 29-67. (b) Lehn, J.-M. In Supramolecular
Chemistry: Concepts and Perspectives; VCH: Weinheim, 1995.
(5) Siemsen, P.; Livingston, R. C.; Diederich F. Angew. Chem., Int.
Ed. 2000, 39, 2632.
(6) Cadiot, P.; Chodkiewicz, W. In Chemistry of Acetylene; Viehe,
H. G., Ed.; Marcel Dekker: New York, 1969; p 597.
(7) Eglington, G.; Galbraith, R. J. Chem. Soc. 1959, 889.
(8) Taylor, R. J. K. In Organocopper Reagents: A Practical Approach;
Oxford University Press: New York, 1994.
(9) Hay, A. S. J. Org. Chem. 1962, 27, 3320.
(10) (a) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett.
1975, 50, 4467. (b) Sonogashira K. In Metal-Catalyzed Cross-Coupling
Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH: Weinheim,
Germany, 1998; pp 203-230. (c) Sonogashira, K. In Comprehensive
Organic Synthesis; Pergamon Press: 1990; Vol. 3, p 521.
(11) For a review on Pd-catalyzed homocoupling of alkynes, see:
Negishi E.; Alimardanov, A. In Handbook of Organopalladium Chem-
istry for Organic Synthesis; Negishi, E., Ed.; Wiley-Interscience: New
York, 2002; Vol. 1, p989.
(12) For exclusive formation of 1,3- and 1,4-disubsitituted enynes
mediated by Pd, see: (a) Yang, C.; Nolan, S. P. J. Org. Chem. 2002,
67, 591. (b) Lu¨cking, U.; Pfaltz, A. Synlett 2000, 1261. (c) Trost, B. M.;
Sorum, M. T.; Chan, C.; Harms, A. E.; Ru¨ther, G. J. Am. Chem. Soc.
1997, 119, 698 and references therein. (d) Trost, B. M.; Chan, C.;
Ru¨ther, G. J. Am. Chem. Soc. 1987, 109, 3486.
Palladium-catalyzed cross-coupling reactions are widely
applied to the formation of C-C bonds.1 Alkyne dimer-
ization, through oxidative homocoupling, to give 1,3-
diynes is important for a number of applications, par-
ticularly in the construction of linearly π-conjugated
acetylenic oligomers and polymers,2 natural products,3
† University of Durham. Fax: +44 (0)1913844737. Tel: +44
(0)1913342037.
‡ University of York. Fax:
0044(0)1904434091.
+44 (0)1904432515. Tel:
§ The Hong Kong University of Science and Technology. Fax: +852
23581594. Tel: +853 23587379.
(1) Diederich, F.; Stang, P. J.; Eds. In Metal-Catalyzed Cross-
coupling Reactions; Wiley-VCH: Weinheim, Germany, 1998.
(13) Rossi, R.; Carpita, A.; Bigelli, C. Tetrahedron Lett. 1985, 26,
523.
(14) Lei, A.; Srivastava, M.; Zhang, X. J. Org. Chem. 2002, 67, 1969.
(15) Liu, Q.; Burton, D. J. Tetrahedron Lett. 1997, 38, 4371.
10.1021/jo048428u CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/09/2004
J. Org. Chem. 2005, 70, 703-706
703

SCHEME 2. General Mechanism for the Pd(II) Homocoupling of Alkynes
SCHEME 3. Homocoupling of Phenylacetylene 1a
to Diyne 2a
) E,E-dibenzylidene acetone), n-Bu4NBr, NaOH, and
allyl bromide (as the oxidant) in CH2Cl2.16 Furthermore,
the reaction may be conducted in the presence of air to
give certain types of diynes.17
In all of the above cases, a stoichiometric amount of a
reoxidant is required for successful homocoupling. The
proposed mechanism for the Pd(II)-catalyzed homocou-
pling reaction is shown in Scheme 2. The formation of
the alkynylcuprate I, by base-assisted deprotonation of
1 in the presence of CuI, and subsequent transmetalation
with Pd(II), gives a dialkynylpalladium(II) species II.
Final reductive elimination generates Pd(0) and diyne
2. The Pd(0) species is reoxidized to Pd(II) with various
oxidants to complete the catalytic cycle.
In a recent paper, a member of our group (I.J.S.F.) has
reported the homocoupling of 1 to give 2 using a combi-
nation of (Ph3P)2PdCl2 (3 mol %) and CuI (3 mol %) in a
CH3CN/Et3N (1.5/2.5, v/v) solvent mixture, with ad-
ditional Ph3P (9 mol %) at room temperature.18 The
presence of phosphine was essential in eliminating the
production of enyne byproducts E/Z-3 and 4. However,
although an added oxidant was not required such as I2
or chloroacetone, the question of whether air was pro-
moting the homocoupling was unresolved. In this paper,
we show conclusive evidence that stoichiometric quanti-
ties of air are required under the previously reported
conditions. This is consistent with our (T.B.M.) previous
findings.17a Importantly, theoretical calculations support
the requirement for an oxidant in this reaction system.
In its absence, H2 would formally be produced, which we
demonstrate is thermodynamically unfavorable. After
screening several procedures for the oxidative homocou-
pling, vide supra, we have determined that the use of
the I2 system, developed by Burton and co-workers,15 is
the most versatile at the present time for the homocou-
pling of electron-deficient arylacetylenes, facilitating
access to novel 1,4-diaryl-1,3-diynes.
out under identical conditions to those reported, but in
a nitrogen-filled glovebox (levels of O2 were < 5 ppm)
(Scheme 3).
Specifically, phenylacetylene 1a (2.0 mM) was added
to a reaction vial containing (Ph3P)2PdCl2 (0.06 mM, 3
mol %) and Ph3P (0.18 mM, 9 mol %) in an CH3CN/Et3N
mixture (1.5/2.5, v/v, total volume ) 8 mL). The cocata-
lyst, CuI (0.06 mM, 3 mol %), was added last to initiate
the reaction (there is not a Pd-catalyzed background
reaction in the absence of CuI). Two reactions were
carried out at temperatures of 25 and 60 °C.
GC/MS analysis of the reaction conducted at 25 °C
after 0.25 h indicated the disappearance of a small
amount of 1a and the formation of 2a (0.058 mM), but
critically, no further change was seen after 24 h (con-
centration of 2a ) 0.058 mM ( 0.005 mM; samples
withdrawn at 4, 8, 18, and 24 h). The total formation of
2a can be accounted for by reduction of Pd(II) to Pd(0)
in the first step of the catalytic cycle shown in Scheme 2
(theoretical amount of 2a produced through the first
turnover ) 0.06 mM, overall theoretical conversion of 1a
f 2a ) 1 mM). A similar result was observed at 60 °C.
Conducting the reaction in the presence of freshly
prepared (Ph3P)4Pd (0.06 mM, 3 mol %), instead of
(Ph3P)2PdCl2 and additional Ph3P, resulted in no turnover
at all under the glovebox conditions.
The effect of air on the reaction of 1a f 2a was
assessed using identical reactant/catalyst/cocatalyst con-
centrations as for the above glovebox reactions. Careful
kinetic experiments were performed to monitor the
disappearance of 1a and appearance of 2a with time,
using the (Ph3P)2PdCl2 precatalyst, in the presence and
absence of additional Ph3P ligand at 60 °C (Figures 1 and
2, respectively). A small quantity of air was introduced
by opening the reaction vessel for 2 min, once all the
components of the reaction mixture had been added (the
reaction mixture was prepared under a stream of dried
To establish whether trace quantities of air were
promoting the conversion of 1 f 2, reactions were carried
(16) Vlassa, M.; Ciocan-Tarta, I.; Margineanu, F.; Oprean, I. Tet-
rahedron 1996, 52, 1337.
(17) (a) Nguyen, P.; Yuan, Z.; Agocs, L.; Lesley G.; Marder, T. B.
Inorg. Chim. Acta 1994, 220, 289. (b) Takahashi, A.; Endo, T.; Nozoe,
S. Chem. Pharm. Bull. 1992, 40, 3181. (c) Jung, F.; Burger, A.;
Biellmann, J. F. Org. Lett. 2003, 5, 383.
(18) Fairlamb, I. J. S.; Ba¨uerlein, P. S.; Marrison, L. R.; Dickinson,
J. M. Chem. Commun. 2003, 632.
704 J. Org. Chem., Vol. 70, No. 2, 2005

TABLE 1. Isolated Yields of
p-R-C6H4-CtC-CtC-C6H4-p-R Using O2 or Anaerobic I2 as
Oxidant
entry
R
% yield (O2)
% yield (I2)
1
2
3
4
5
6
7
8
H, 2a
80
22
81
31
73
66
52
CN, 2b
CO2Me, 2c
CF3, 2d
Me, 2e
SMe, 2f
OMe, 2g
NMe2, 2h
41
42
85
69
80
78
59
(Ph3P)2PdCl2, to give (Ph3P)nPd(0) (where n ) 1-4) and
diyne 2a, through the intermediate species, (Ph3P)nPd-
(σ-alkynyl)2 (n ) 1 or 2). Subsequent turnovers will be
dependent on the efficiency of reoxidation of Pd(0) to Pd-
(II) by O2 vide infra. Note that in this reaction we do not
observe the formation of enynes E/Z-3a and 4a and that
the formation of Ph3PdO is only observed toward the
latter stages of the reaction (by GC/MS).
FIGURE 1. Concentration versus time plot for the conversion
of 1a (0, 2.0 mM) to 2a ([) in the presence of additional
Ph3P: reaction vessel opened to air for 2 min at t0, where the
relative concentrations of 1a and 2a refer to [1a]t/[1a]0 and
[2a]t/[1a]0, respectively.
In the absence of Ph3P, one can see that the rate of
enyne formation is faster in the initial stages (<2 min
during which air is being introduced) as compared to the
remaining course of the reaction (Figure 2). The O2
concentration in solution is expected to be lower at this
stage, and hence, the Pd(0) concentration will be higher,
facilitating generation of the enyne byproducts (E/Z-3a
and 4a) probably via alkyne C-H oxidative addition,
alkyne insertion into Pd-H, and C-C reductive elimina-
tion. This suggests that excess phosphine slows down/
inhibits enyne generation, probably by tying up needed
coordination sites on Pd(0) (Figure 1).
Overall, these reactions demonstrate that air (O2) does
promote the homocoupling process and that the role of
excess phosphine is to thwart enyne generation. It is
important to note, however, that if O2 is bubbled through
a solution of 1a f 2a at room temperature, under the
standard conditions given above, we find that only low
conversions to 2a are seen (∼20%). Furthermore, a
number of other products are observed by GC/MS analy-
sis, not least the formation of Ph3PdO in the early stages
of the reaction, the production of other tri- and tetrameric
acetylenic products, and the rapid production of pal-
ladium black. The separation of the minor side products
from 2a has so far proven unsuccessful.
To compare the use of O2 and I2 as oxidants, a series
of symmetric para-R-substituted 1,4-diphenyl-1,3-buta-
diynes (2b-h) was prepared. Details of the synthesis and
characterization are provided in the Supporting Informa-
tion and the results are summarized in Table 1. The
reactions with O2 as the oxidant were easier to set up
and the product was easier to isolate, but these often gave
lower yields and incomplete reactions.
FIGURE 2. Concentration versus time plot for the conversion
of 1a (0, 2.0 mM) to 2a ([) and E/Z-3a+4a (O) in the absence
of additional Ph3P: reaction vessel opened to air for 2 min at
t0, where the relative concentrations of 1a, 2a, and E/Z-3a+4a
refer to [1a]t/[1a]0, [2a]t/[1a]0, and [E/Z-3a+4a]t/[1a]0, respec-
tively.
N2 using standard Schlenk techniques; dried solvents
were degassed using three freeze-thaw-pump cycles).
The reactions were then sampled using a gastight
syringe, and all GC samples were quenched through a
small plug of silica (analysis aliquot ) 50 µL, 0.69 g of
Kieselgel 60 silica; particle size 0.035-0.070 mm, 220-
440 mesh), eluted with CH2Cl2 (∼ 3 mL), and then stored
at -30 °C prior to analysis by GC. The time-dependence
of post-quenched samples showed no change after 72 h
at this temperature, proving to be a reliable quenching
procedure in this case.19
I2 oxidation, in the absence of O2, proved to be higher
yielding in several cases (entries 1, 2, 4, 5, and 7;
compounds 2a, 2b, 2d, 2e, and 2g). Here, an extra
synthetic step involving a sodium thiosulfate wash to
remove any unreacted iodine gave purer products. The
R ) CF3 product (entry 4, 2d) was further characterized
by single-crystal X-ray diffraction (see the Supporting
Information), and the bond distances and angles are quite
The initial rate seen in Figure 1 is somewhat faster
than the overall rate, which is presumably associated
with rapid reaction of the alkynylcuprate, derived from
reaction of 1a with CuI in the presence of Et3N, with
(19) Niemela, E. H.; Lee, A. F.; Fairlamb, I. J. S. Tetrahedron Lett.
2004, 45, 3593.
J. Org. Chem, Vol. 70, No. 2, 2005 705

TABLE 2. Reaction Energies and Free Energies (kcal
mol-1) of Reactions (1) (∆E1 and ∆G1), (2) (∆E2 and ∆G2),
and (3) (∆E3 and ∆G3) Calculated with Different
Theoretical Methods
for R ) Me suggest that the B3LYP calculations slightly
underestimate the endothermicity of reactions (1) by 2.8
kcal/mol. The experimental formation enthalpy and free
energy of H2O(g) are -57.8 and -54.6 kcal/mol, respec-
tively.22 Both the B3LPY and CCSD(T) calculations give
comparable values. From these calculations, we conclude
that reactions (1) are expected to be slightly endothermic
and are thus not particularly favorable thermodynami-
cally. The presence of oxidants, e.g., O2, provides the
thermodynamic driving force for alkyne homocoupling
reactions.
∆E1 ∆G1
0.9 -0.8 -57.9 -51.3 -58.3 -50.6
R ) Ph -1.2 -1.3 -60.0 -51.8
CCSD(T) R ) Me 3.7 -56.4
∆E2
∆G2
∆E3
∆G3
B3LYP
R ) Me
-60.1
similar to those we recently found for the R ) CO2Me
derivative.20
Comments on the Reoxidation of Pd(0) to Pd(II).
It being clear that an oxidant is required to carry out
the homocoupling, it remains to consider the process by
which the Pd(0) is reoxidized to Pd(II) under the reaction
conditions. While direct reoxidation of Pd is feasible, it
may be rather slow. Indeed, Figure 1 indicates a zero-
order dependence of the reaction rate on alkyne concen-
tration which would be consistent with the rate-deter-
mining step being Pd reoxidation. However, it is important
to point out that Cu is widely used to accelerate the
reoxidation of Pd(0) to Pd(II) with O2 as in the industri-
ally important Wacker process for the aerobic oxidation
of alkenes to aldehydes.23 Thus, Cu may well have a dual
role in the alkyne homocoupling reaction, first being
utilized to mediate alkynyl transfer to Pd(II) via trans-
metalation from a Cu-alkynyl species, and second in the
redox sequence shown in eqs 4 and 5. Further studies of
the role of Cu in these and related systems are in
progress.
Computational Results
To investigate the thermodynamics of the homocou-
pling process, a series of density functional theory cal-
culations were performed at the B3LYP level to optimize
all the species in reactions (1)-(3). Frequency calcula-
tions at the same level of theory have also been performed
to identify all stationary points as minima (zero imagi-
nary frequencies). The 6-311++G** basis set was used
in the calculations. On the basis of the optimized struc-
tures, single-point calculations at the CCSD(T) level,
using the same basis set, were performed to calculate the
reaction energies for R ) Me. All calculations were
performed using the Gaussian 98 package.21
The energetics of the following reactions have been
theoretically evaluated for R ) Me and Ph at the B3LYP
level of density functional theory.
2RC≡CH f RC≡C-C≡CR + H2
(1)
2RC≡CH + 1/2O2 f RC≡C-C≡CR + H2O (2)
2Cu(I) + O2 f 2Cu(II)
(4)
(5)
H2 + 1/2O2 f H2O
(3)
2Cu(II) + Pd(0) f 2Cu(I) + Pd(II)
Acknowledgment. I.J.S.F. and T.B.M. are grateful
to Johnson-Matthey (U.K.) for generous loans of palla-
dium(II) salts. We thank Pfizer (David Alker) for a
summer research bursary to support J.P.H. and Mr.
Anant R. Kapdi (York) for some GC analysis. T.B.M.
and J.A.K.H. acknowledge One NorthEast for research
support through the Nanotechnology UIC program.
R.M.W. thanks EPSRC for a postgraduate scholarship.
Z.L. thanks the Research Grants Council of Hong Kong
(HKUST6023/04P) for financial support.
Table 2 gives the results of our energetic calculations.
At the B3LYP level, reactions (1) are found to be almost
energetically neutral. The high exothermicity of reactions
(2) is clearly driven by the formation of H2O because eq
2 is the sum of eqs 1 and 3. The CCSD(T) calculations
(20) Fasina, T. M.; Collings, J. C.; Burke, J. M.; Batsanov, A. S.;
Ward, R. M.; Albesa-Jove, D.; Porre`s, L.; Beeby, A.; Howard, J. A. K.;
Scott, A. J.; Watt, S. W.; Viney, C.; Marder, T. B. J. Mater. Chem. in
press, DOI: 10.1039/b413514h.
(21) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A.9; Gaussian, Inc.: Pittsburgh, PA,
1998.
Supporting Information Available: Experimental pro-
cedures and characterization data for the diyne compounds,
crystallographic data (CIF), and details of the kinetic studies
by GC and GC/MS. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO048428U
(22) Atkins, P. W. Physical Chemistry, 5th ed.; Oxford University
Press: Oxford, 1994; p C12.
(23) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals, 3rd ed.; John Wiley and Sons: New York, 2001; pp 207-209.
706 J. Org. Chem., Vol. 70, No. 2, 2005