Tandem Pd(II)-catalyzed vinyl ether exchange-claisen rearrangement as a facile approach to γ,δ-unsaturated aldehydes

Article abstract of DOI:10.1021/jo062548f

(Chemical Equation Presented) A sequential allyl vinyl ether formation-Claisen rearrangement process catalyzed by a palladium(II)- phenanthroline complex is reported. The effects of allylic alcohol structure, type of vinylating agent, and palladium catalysts are discussed. This method provides a convenient approach to γ,δ unsaturated aldehydes under mild conditions that avoid the use of toxic Hg(II) catalysts. The new methodology has been successfully demonstrated on the kilogram scale.

Full text of DOI:10.1021/jo062548f

Tandem Pd(II)-Catalyzed Vinyl Ether
Exchange-Claisen Rearrangement as a Facile
Approach to γ,δ-Unsaturated Aldehydes
stabilization of the carbocation intermediate. A Lewis acid-
catalyzed vinyl ether exchange process has been proven to be
mild and specific, but so far only toxic Hg(II) has been
successfully applied in such a tandem vinyl ether exchange-
5
Claisen rearrangement process. Alternative approaches to allyl
Xudong Wei,* Jon C. Lorenz, Suresh Kapadia, Anjan Saha,
Nizar Haddad, Carl A. Busacca, and Chris H. Senanayake
vinyl ethers include the reaction of allyl esters with Tebbe’s
6
7
reagent, preparation from silyl enol ethers, and aldol conden-
8
Department of Chemical DeVelopment, Boehringer Ingelheim
Pharmaceuticals, Inc., 900 Ridgebury Road,
Ridgefield, Connecticut 06877
sation of unsaturated esters. Several elegant vinyl ether
formation-Claisen rearrangement sequences leading to γ,δ-
unsaturated aldehydes are known, including the â-haloethyl allyl
9
ammonium salt approach developed by Laird, the â-halo ethyl
ether elimination approach developed by Dulc e` re,10 the allyl
oxide addition to (E)-carboxyvinyltrimethylamonium betaine
Claisen rearrangement-decarboxylation method developed by
xwei@rdg.boehringer-ingelheim.com
ReceiVed December 12, 2006
1
1
B u¨ chi, the Rh(II)-catalyzed Bamford-Stevens/Claisen rear-
1
2
rangement sequence developed by Stoltz, a copper-catalyzed
C-O coupling-Claisen rearrangement process developed by
1
3
Buchwald, and the diallyl ether isomerization method devel-
1
4
oped by Nelson. While often useful, some of these methods
require harsh conditions, are limited to certain types of substitu-
tion patterns, or need high loadings of catalysts and reagents.
It has long been known that Pd(II) salts such as Pd(OAc)2 or
I are also effective Lewis acid catalysts for the vinyl ether
15
exchange reaction. McKeon and Fitton reported that Pd(OAc)2
can be stabilized by cis-bidentate ligands such as phenanthroline
or 2,2′-bipyridyl while still maintaining its activity.16 These
catalysts have been employed for the synthesis of vinyl ethers
A sequential allyl vinyl ether formation-Claisen rearrange-
ment process catalyzed by a palladium(II)-phenanthroline
complex is reported. The effects of allylic alcohol structure,
type of vinylating agent, and palladium catalysts are dis-
cussed. This method provides a convenient approach to γ,δ-
unsaturated aldehydes under mild conditions that avoid the
use of toxic Hg(II) catalysts. The new methodology has been
successfully demonstrated on the kilogram scale.
1
7
of steroids, glycidol, and sugars. Recently, Schlaf and Bosch
optimized the vinyl ether exchange reaction by using a more
soluble and Lewis acidic complex of palladium trifluoroacetate
18
and 4,7-diphenyl-1,10-phenanthroline (IVb) as catalyst. In
addition, an elegant Ir-catalyzed vinyl ether synthesis with vinyl
19
acetate/Na2CO3 has been developed by Ishii et al. However,
a tandem Pd(II)-catalyzed allyl vinyl ether formation-Claisen
rearrangement process has not yet been reported.
Recently in our research work we needed a series of low
molecular weight γ,δ-unsaturated aldehydes as building blocks
for SAR studies. Although many of these compounds are known,
surprisingly few practical laboratory methods suitable for their
preparation in the gram to kilogram scale are available. Synthesis
of the requisite γ,δ-unsaturated ester or ketone from an allylic
alcohol and an ortho ester or enol ether catalyzed by protic acid
(
4) H+ catalyzed: Thomas, A. F. J. Am. Chem. Soc. 1969, 91, 3281.
Dauben, W. G.; Dietsche, T. J. J. Org. Chem. 1972, 37, 1212.
2+
(
5) Hg catalyzed: (a) Watanabe, W. H.; Conlon, L. E. J. Am. Chem.
Soc. 1957, 79, 2828. Church, R. F.; Ireland, R. E.; Marshall, J. A. J. Org.
Chem. 1966, 31, 2526. (b) Saucy, G.; Marbet, R. HelV. Chim. Acta 1967,
5
0, 2091.
(6) Kinney, W. A.; Coghlan, M. J.; Paquette, L. A. J. Am. Chem. Soc.
1
,2
(the Johnson-Claisen ortho ester Claisen rearrangement) is
1985, 107, 7352-7360.
a well-established reaction, and the corresponding aldehydes
are often prepared from the esters by further reduction/
(7) Maeda, K.; Shinokubo, H.; Oshima, K.; Utimoto, K. J. Org. Chem.
1
996, 61, 2262-2263.
(
8) Hiersemann, M. Synthesis 2000, 1279-1290.
1
b,3
reoxidation steps. Drawbacks associated with this three-step
sequence are not only the use of moisture-sensitive reductants
and oxidants, which is undesirable in large-scale synthesis, but
also the lengthy and sometimes problematic workup operations.
The difficulty behind a direct Claisen approach to aldehydes
(9) Laird, T.; Ollis, W. D.; Sutherland, I. O. J. Chem. Soc., Perkin Trans.
1 1980, 1477-1486.
(
(
(
10) Dulc e` re, J.-P.; Rodriguez, J. Synthesis 1993, 399.
11) B u¨ chi, G.; Vogel, D. E. J. Org. Chem. 1983, 48, 5406-5408.
12) May, J. A.; Stoltz, B. M. J. Am. Chem. Soc. 2002, 124, 12426.
(13) Nordmann, G.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125, 4978-
4
4979.
is, in many cases, the formation of allyl vinyl ethers from acetal
(14) Nelson, S. G.; Wang, K. J. Am. Chem. Soc., 2006, 128, 4232-
elimination, presumably due to the lack of additional oxygen
4
233 and references cited therein.
(
15) McKeon, J. E.; Fitton, P.; Griswold, A. A. Tetrahedron 1972, 28,
(
1) For Ireland-Claisen and the Johnson ortho ester Claisen rearrange-
ments: (a) Ireland, R. E.; Mueller, R. H. J. Am. Chem. Soc. 1972, 94, 5897.
b) Johnson, W. S.; Werthemann, L.; Bartlett, W. R.; Brocksom, T. J.; Li,
T.-T.; Faulkner, D. J.; Petersen, M. R. J. Am. Chem. Soc. 1970, 92, 741.
2) For a most recent review of Claisen rearrangement: Castro,
A. M. M. Chem. ReV. 2004, 104 (6), 2939 -3002.
3) For example: Valentine, J. C.; McDonald, F. E.; Neiwert, W. A.;
227-232.
(16) McKeon, J. E.; Fitton, P. Tetrahedron 1972, 28, 233-238.
(
(17) (a) Weintraub, P. M.; King, C.-H. J. Org. Chem. 1997, 62, 1560-
1562. (b) Tachibana, T.; Aihara, T. CAN: 112:138900; JP 01272577; Seimi
Chemical Co., Ltd. Japan, 1989. (c) Handerson, S.; Schlaf, M. Org. Lett.
2002, 4, 407-409.
(
(
(18) Bosch, M.; Schlaf, M. J. Org. Chem. 2003, 68, 5225-5227.
(19) Okimoto, Y.; Sakaguchi, S.; Ishii, Y. J. Am. Chem. Soc. 2002, 124,
1590.
Hardcastle, K. I. J. Am. Chem. Soc.; 2005, 127, 4586-4587. Baldwin,
I. R.; Whitby, R. J. Chem. Commun. 2003, 2786-2787.
10.1021/jo062548f CCC: $37.00 © 2007 American Chemical Society
4
250
J. Org. Chem. 2007, 72, 4250-4253
Published on Web 04/21/2007

TABLE 1. Pd(II)-Catalyzed Vinylation-Claisen Rearrangement
FIGURE 1. Pd(II) and Ir(I) catalysts.
In view of the low cost of many vinylating reagents, fair
catalytic turnover numbers, and mild reaction conditions, we
explored the possibility of devising a tandem Pd(II)-catalyzed
vinylation-Claisen rearrangement process for the preparation
of the desired γ,δ-unsaturated aldehydes.20 Herein we report
our results.
Results and Discussion. Several incorporated design features
should be noted here. (1) As many Claisen rearrangements
require a high temperature to proceed, a vinylating agent with
a high boiling point is desired. The large difference between
the boiling points of vinylating agent and product should also
facilitate distillative isolation. (2) The equilibrium nature of the
vinylation reaction requires an excess of vinylating agent. To
achieve atom and volume efficiency, a vinyl ether with low
molecular weight is therefore preferred. (3) A more polar
reaction medium would be expected to increase the reaction
rate. (4) Both the vinylating reagent and the catalyst have to be
commercially available and inexpensive. Commercial triethyl-
eneglycol divinyl ether (TGDV) fulfills all of the above
requirements. It possesses a high boiling point (>120 °C/18
mmHg), relatively low molecular weight (as two vinyl ether
units are available in one molecule), increased polarity, and a
modest price of ca. $90/kg.
Using 1.5 equiv of TGDV as starting material and solvent,
we were delighted to find that a series of primary allylic alcohols
underwent the vinyl ether exchange reaction efficiently at 70-
7
5 °C when the complex of Pd(OAc)2 and 1,10-phenanthroline
was used as catalyst (Table 1, entries a to e). The superior ligand
,7-diphenyl-1,10-phenanthroline, used in ref 18 for better
4
solubility, is not necessary here, because TGDV dissolves the
less soluble 1,10-phenanthroline complex catalyst without any
difficulties. Claisen rearrangement, however, proceeded very
slowly under these conditions, and only trace amounts of alde-
hyde could be observed even after prolonged reaction time.
Reactions at higher temperature were also tried but failed,
most likely due to significant decomposition of Pd(II) catalyst
at temperatures above ∼90 °C giving palladium black. The
Pd(OAc)2 complex (IIIa or IVa) is usually active enough, very
likely benefiting from the high polarity of TGDV. For example,
reaction of 3-methylbutenol reached its theoretical equilibrium
ratio of ca. 75-80% readily within 6 h at 70 °C with Pd(OAc)2
complex IIIa or IVa in TGDV (Table 1 entry d), but the same
vinylation reaction with IVa failed when 20 equiv of butyl vinyl
ether was used as vinylating agent (Table 1, entry 18 in ref
18). The reaction was normally faster when the Pd(tfa)2 complex
(IIIb) was employed, but 10-20 mol % of H u¨ nig’s base
a
When IIIb was used as catalyst, H u¨ nig’s base (i-Pr2NEt) was added
as counter base to inhibit acetal formation. b Condition A: 50 mmol of
allyllic alcohol, 75 mmol of TGDV, and 0.5 mol % catalyst was stirred at
(20) Our initial experiments of 3-methyl-2-buten-1-ol and vinyl acetate
catalyzed by 1% [Ir(COD)Cl]2 (V) at 100 °C in toluene following the method
of Ishii (ref 19) gave 3,3-pentenal in ca. 35% yield (NMR) with some
difficulties in isolation of the volatile product from the toluene reaction
mixture. No reaction was observed with the secondary alcohol 1-penten-
7
0 °C for 5-15 h, then 120-150 °C for 5-9 h. Condition B: 50 mmol of
allyllic alcohol, 75 mmol of TGDV. and 0.5 mol % catalyst was stirred at
c
8
5 °C for 24-48 h, then 120 °C for 0.5 h. Isolated yield. Numbers in
3
-ol. During the preparation of this paper, Ishii’s group published their
1
d
parentheses are yields based on H NMR before isolation. Rearrangement
Ir-catalyzed vinylation-Claisen rearrangement method, yet it was limited
to primary allylic alcohols. Masao, M.; Sakaguchi, S.; Ishii, Y. J. Org. Chem.
was conducted at 190 °C. e About 5-10% impurities remained in the
product even after two distillations.
2
006, 71, 6285-6286.
J. Org. Chem, Vol. 72, No. 11, 2007 4251

SCHEME 1. Substitution Effect on Claisen Rearrangement
SCHEME 2. Iterative Vinylation-Claisen Rearrangementa
ortriethylamine had to be used to inhibit acetal formation caused
by the higher acidity of trifluoroacetic acid, as described in ref
a
Reagents and conditions: (a) CH2dCHMgBr, -78 to 0 °C; (b) 1.5
equiv of TGDV, 0.5% Pd(OAc)2-phenanthroline, 75-80 °C, 48 h.
1
8. When the vinyl ether exchange step reached equilibrium
1
(easily monitored by H NMR), the temperature was raised to
(entry m) and carveol (entry n) can actually rearrange at slightly
lower temperature (150 °C), although the rationale for this is
unclear. In the case of verbenol (entry o), no rearrangement
was observed. The reaction of cyclic allylic alcohols is well-
known to be problematic since adoption of the chairlike
transition state suffers from syn-pentane interactions and the
boat transition state is inherently higher in energy.22
ca. 120-150 °C and the Claisen rearrangement initiated
immediately, again easily tracked by 1H NMR. When the
rearrangement reaction was complete, the product was readily
isolated by flash distillation under vacuum (ca. 100 mmHg).
The product thus obtained normally contained small amounts
of remaining allylic alcohol or other impurities. A second
distillation produced highly pure aldehyde suitable for our
synthetic purposes.
The reaction of acyclic secondary allylic alcohols (Table 1,
entries f to k) showed interesting differences from those of
primary alcohols. Vinyl ether exchange can also be catalyzed
by either IIIa or IIIb efficiently, usually requiring less than 24
h to finish. However, the Claisen rearrangement step itself seems
to proceed at a faster rate, as rearranged product can be observed
even at a temperature as low as 80 °C, within the same
temperature range for the vinyl ether exchange step. This
facilitates the overall reaction significantly, as the irreversible
Claisen rearrangement consumes the allyl vinyl ether, therefore
driving the equilibrium further to the right-hand side, so a higher
conversion may be achieved with the same or even lower
amount (such as 1 equiv) of TGDV (Table 1, entries f-k).
Reactions of secondary allylic alcohols all proceeded smoothly
at 85 °C, with the formation of allyl vinyl ether intermediates
followed by in situ Claisen rearrangement, giving >90%
conversion (NMR) and usually over 70% yield after isolation.
The faster Claisen rearrangement observed here is in agree-
ment with results reported in literature that radical stabilizing
The potential of this methodology to access more complicated
compounds has been explored as shown in Scheme 2. Starting
from a rearranged product 3-methyl-4-hexenal (3p of Table 1,
entry p), a diene aldehyde 4 was easily obtained in 70% overall
yield by a standard vinyl magnesium bromide addition followed
by a tandem vinyl ether exchange-Claisen rearrangement of
the resulting allylic alcohol. Aldehyde 4 was again subjected
to the same iterative reaction sequence and, to our delight, the
above result was almost exactly reproducible, generating the
desired triene aldehyde 5 in ∼50% overall yield in just 4 steps
from 3-methyl-4-hexenal (3p). Polyene aldehydes are useful
building blocks, e.g., in the synthesis of polyene type natural
1
b
3
products and brevetoxin-type polycyclic ethers. By directly
generating the aldehyde functionality instead of the esters, this
method avoids the reduction-oxidation steps, and is therefore
superior in terms of simplicity and convenience, particularly
when several skipped alkene units need to be constructed in a
linear sequence. The mild reaction conditions should also
tolerate more functional groups in complex molecules sensitive
to acid or severe oxidation/reduction conditions.
One of the future directions to optimize this method would
be to design more stable catalysts that tolerate higher temper-
atures so that the Claisen rearrangement can proceed at a faster
rate. As a result, the equilibrium may be driven to the right
side even more efficiently, improving yield, particularly for the
reactions of primary alcohols. The catalysts used in Table 1
are stable only up to ca. 90 °C. 2,2-Dipyridyl has been proposed
to be an effective ligand for this transformation as well,1 but
reaction conditions and results are not available. For comparison
purposes, reactions with complex IIb and its methoxy-
substituted version VI were examined (Table 2). However, none
of them was found to be superior in terms of stability and
activity. For the reaction of a primary alcohol (Table 2, entry
2
1
substituents at C-4 can accelerate reaction rates significantly.
In addition, the R-alkyl group may also help the allyl vinyl ether
to reach the concerted chairlike transition state for rearrangement
by a steric effect (Scheme 1). Along this line, reaction of two
tertiary allylic alcohols 2-methyl-3-buten-2-ol and linalool were
examined (Table 1, entries p and q). Under the same conditions,
we found both of these reactions proceeded smoothly with
catalyst IIIb, generating the corresponding aldehyde directly
5,17
1
and cleanly in ∼90% yields ( H NMR). None of the allyl vinyl
ether intermediate was observed, suggesting a much faster
Claisen rearrangement is occurring at this temperature, in
agreement with the trend observed above. Due to its nonvolatile
nature, the rearranged product from linalool was isolated easily
in 80% yield by flash column chromatography. It should also
be noted that vinylation of the sterically demanding tertiary
1
), 1.2 mol % of IIb or VI had to be used to achieve a 60%
yield, likely due to decomposition of the catalyst. Vinylation
of a secondary alcohol (1-penten-3-ol) failed completely (entry
1
2
). Instead, 3-pentanone was identified by H NMR, suggesting
allylic alcohol linalool failed in the literature with IVb as catalyst
and butyl vinyl ether as vinylating agent.18
an isomerization reaction of 3-pentenol predominated. Interest-
ingly, the enolization seemed to be sensitive to an increased
steric hindrance on the double bond such as that imposed by
an additional methyl group (entry 3). In this case, slow vinyl
ether exchange was observed and the aldehyde was obtained in
ca. 40% yield after 48 h. These results suggest that dipyridyls
are an inferior ligand class when compared to the phenanthro-
lines, and further efforts at optimizing the ligand must continue
in the future.
Finally, cyclohexenols (Table 1, entries l to o) can be
vinylated readily as well in ca. 5 h, but the rate of the
rearrangement was found to be strongly dependent on the
structure. Vinyl cyclohexenyl ether (entry l) rearranges only at
1
90 °C. The sterically more hindered 3-methylcyclohexenol
(21) For discussion of substituent effects in the Claisen rearrangement,
see: Burrows, C. J.; Carpenter, B. K. J. Am. Chem. Soc. 1981, 103, 6983-
6
984. Gajewski, J. J.; Gee, K. R.; Jurayj, J. J. Org. Chem. 1990, 55, 1813-
1
822.
(22) Bartlett, P. A.; Pizzo, C. F. J. Org. Chem. 1981, 46, 3896-3900.
4
252 J. Org. Chem., Vol. 72, No. 11, 2007

TABLE 2. Reaction Catalyzed by Dipyridinyl Type Complexesa
used as catalyst, N,N-diisopropylethylamine (6.4 g, 0.05 mol) was
added to inhibit acetal formation as a side reaction. The reaction
condenser should be opened to air and NOT inerted by N or argon.
2
The mixture was heated gently to 70 °C. (An oil bath or water
bath is preferred to a heating mantle for better temperature control.
Decomposition of catalyst was often observed due to some “hot
spots” when heated by a heating mantle.) The reaction mixture
became a clear yellow solution. After the solution was stirred for
5-7 h, NMR showed that the allylic alcohol was converted to its
vinyl ether to the extent of 75-80%. The mixture was then heated
to ca. 120 °C and reflux started. The temperature was raised
gradually to 140-145 °C and stirring was continued at this
1
temperature for 5-10 h until H NMR showed the rearrangement
was complete. For some reaction of cyclohexenols, the rearrange-
ment was carried out at a higher temperature, up to 190 °C. When
the reaction was finished, the reaction mixture was distilled with
an oil bath temperature of ca. 110-140 °C and under ca. 300 mmHg
pressure to give the crude product that was normally subjected to
a second distillation to produce high-purity product. For compounds
with a high boiling point (3l-o,q), distillation was not practical,
and flash column chromatography on silica gel (hexanes) was used,
giving pure product after evaporation of solvent.
(B) Typical Procedure for Secondary and Tertiary Allylic
Alcohols: Method B. To a solution of triethyleneglycol divinyl
ether (303 g, 1.5 mol) and the corresponding allylic alcohol (1 mol)
in a 1 L flask equipped with an efficient condenser was added the
a
A mixture of 25 mmol of alcohol and 75 mmol of vinylating agent
TGDV was stirred with 0.5 mol % catalyst and 5 mol % H u¨ nig base
i-Pr2NEt) at 75-85 °C. NMR yield reported.
b
(
catalyst 1,10-phenathroline-Pd(OAc)
phenathroline-Pd(CF CO (2.6 g, 0.005 mol). When Pd(CF
complex was used as catalyst, N,N-diisopropylethylamine (6.4 g,
.05 mol) was added to inhibit acetal formation as a side reaction.
The condenser should be opened to air and NOT inerted by N or
2
(2.0 g, 0.005 mol) or 1,10-
In summary, we describe the first tandem Pd(II)-catalyzed
3
)
2 2
CO )
3 2 2
vinyl ether formation-Claisen rearrangement process using the
readily available palladium acetate-phenanthroline catalyst in
combination with TGDV as the vinyl ether starting material.
This provides a convenient approach to a series of γ,δ-
unsaturated aldehydes under mild conditions that avoids the use
of toxic Hg(II). This method is very practical and volume
efficient, and we have prepared some volatile aldehydes on >20
kg scale successfully using this methodology. In addition, remote
alkene units are intact under these conditions, which allows the
efficient synthesis of polyene-type compounds. Further opti-
mization of the reaction system by exploring better designed
catalyst, as well as its application to more elaborate allylic
alcohols, is ongoing in our laboratories. Results will be reported
in due course.
0
2
argon. The mixture was heated gently to 75-80 °C. (An oil bath
or water bath is preferred to a heating mantle for better temperature
control. Decomposition of catalyst was often observed due to some
“hot spots” when heated by a heating mantle.) The reaction mixture
became a clear yellow solution. Reaction was monitored by NMR,
and formation of vinyl ether and aldehyde was usually >90% after
24 to 48 h. The mixture was then heated to ca. 100-120 °C and
stirred at this temperature until NMR showed all of the remaining
vinyl ether had rearranged to aldehyde. When the reaction was
finished, the reaction mixture was distilled with an oil bath
temperature of ca. 110-140 °C and under ca. 300 mmHg pressure
to give the crude product that was normally subjected to a second
distillation to produce high-purity product. For compounds with a
high boiling point (3q, 4, 5), distillation was not practical, and flash
column chromatography on silica gel (hexanes) was used, giving
pure product after evaporation of solvent.
Experimental Section
Preparation of Catalysts (II, III, IV, and VI). All catalysts
used in this study were prepared in almost quantitative yield by
stirring a 1:1 (mol) mixture of palladium salt (Pd(OAc)
2
or
Pd(tfa)
2
) and the corresponding organic ligand in DME (∼5
Acknowledgment. The authors thank Prof. Erick M. Carreira
at ETH, Prof. Peter Wipf at the University of Pittsburgh, and
Prof. Scott Denmark at the University of Illinois, Urbana for
helpful discussions. Dr. Vittorio Farina is thanked for construc-
tive discussion in the preparation of the paper.
mL/mmol of complex). A yellowish solid precipitated out and was
isolated by filtration, rinsed with diethyl ether, and dried in an oven
at 50 °C under reduced pressure.
Tandem Pd(II)-Catalyzed Vinyl Ether Exchange-Claisen
Rearrangement. (A) Typical Procedure for Primary Allylic
Alcohols and Cyclic Cyclohexenols: Method A. To a solution
of triethyleneglycol divinyl ether (303 g, 1.5 mol) and the
corresponding allylic alcohol (1 mol) in a 1 L flask equipped with
an efficient condenser was added the catalyst 1,10-phenathroline-
Supporting Information Available: Characterization of new
compounds and copies of NMR spectra of all products. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Pd(OAc)
Pd(CF CO
2
complex (2.0 g, 0.005 mol) or 1,10-phenathroline-
(2.6 g, 0.005 mol). When Pd(CF CO complex was
3
)
2 2
3
)
2 2
JO062548F
J. Org. Chem, Vol. 72, No. 11, 2007 4253