An efficient preparation of β-aryl-β-ketophosphonates by the TFAA/H3PO4-mediated acylation of arenes with phosphonoacetic acids

Article abstract of DOI:10.1021/jo800973e

(Chemical Equation Presented) β-Aryl-β-ketophosphonates can be efficiently prepared in good yield by using a TFAA/85% H₃PO ₄-mediated acylation of electron-rich arenes with phosphonoacetic acids. The conditions offer advantages over

Full text of DOI:10.1021/jo800973e

An Efficient Preparation of
ꢀ-Aryl-ꢀ-ketophosphonates by the TFAA/
H3PO4-Mediated Acylation of Arenes with
Phosphonoacetic Acids
generally utilize the same basic strategy of elaborating a benzoyl
derivative to the targeted ketophosphonate (Scheme 1). Ad-
ditionally, many of the literature methods require the use of
strong base and cryogenics, which can limit their utility in
certain instances.
George P. Luke,* Christopher K. Seekamp, Zhe-Qing Wang,
and Bertrand L. Chenard
Process Chemistry, Neurogen Corporation,
Branford, Connecticut 06405
As part of a recent development project, we required supplies
of several novel ꢀ-aryl-ꢀ-ketophosphonates (1, Scheme 2),
which were to be used in subsequent HWE reactions in the
preparation of key acrylophenone intermediates. These closely
related 2,3-dimethylphenol derivatives differ primarily in the
length of the hydroxyalkyl chain appended to the phenol oxygen.
In considering a strategy for the preparation of 1, we recognized
that the only commercially available derivative of 2,3-dimeth-
ylphenol that appeared to be a suitable substrate for the existing
methods of preparing ꢀ-aryl-ꢀ-ketophosphonates was 2,3-
dimethylanisaldehyde, which was rather expensive. Preparing
related benzoyl derivatives to serve as suitable substrates for
existing methods would only add steps to our process. Further-
more, the reaction conditions used in some of the existing
methods were not compatible with some of the functionality in
our system and/or were not attractive from a process chemistry
standpoint. We therefore began to consider alternative methods
of preparing ꢀ-aryl-ꢀ-ketophosphonates. The goal was to
identify an efficient process that utilizes relatively mild reaction
conditions. We knew that the Friedel-Crafts acylation of 2,3-
dimethylanisole (2a) with simple acid chlorides was a regiose-
lective, high-yielding reaction,5 so we considered an approach
whereby we would use a phosphonoacetic acid to acylate an
appropriate 2,3-dimethylphenol derivative (2, Scheme 2), di-
rectly affording a ꢀ-aryl-ꢀ-ketophosphonate. This successful
strategy is the subject of this paper.
At the outset, we considered using the known dieth-
ylphosphonoacetyl chloride,6 prepared from commercially avail-
able diethylphosphonoacetic acid (3a), in the Friedel-Crafts
acylation; however, we experienced some difficulties in isolating
the acid chloride. We therefore quickly turned our attention to
the more direct approach of using diethylphosphonoacetic acid
(3a) itself in the acylation. 2,3-Dimethylanisole (2a) was chosen
as the model substrate for surveying various reaction conditions.
We had little success using some of the more common promoters
of Friedel-Crafts acylations with carboxylic acids, such as
MeSO3H, H2SO4, trifluoroacetic anhydride (TFAA), or POCl3.
gluke@nrgn.com
ReceiVed May 6, 2008
ꢀ-Aryl-ꢀ-ketophosphonates can be efficiently prepared in
good yield by using a TFAA/85% H3PO4-mediated acylation
of electron-rich arenes with phosphonoacetic acids. The
conditions offer advantages over existing methods of prepar-
ing these useful compounds by not requiring the use of strong
base, cryogenics, or an anhydrous and inert atmosphere.
Furthermore, some functional groups not tolerated with the
reaction conditions used in existing methods are compatible
with the herein described conditions.
ꢀ-Aryl-ꢀ-ketophosphonates (I) are frequently used reagents
in Horner-Wadsworth-Emmons (HWE) olefination reactions
with aldehydes or ketones to produce acrylophenones (II, eq
1).1 Furthermore, ꢀ-aryl-ꢀ-ketophosphonates can serve as
precursors in the preparation of γ-aryl-γ-ketophosphonates,2a
certain ꢀ-phosphonic acid R-amino acid derivatives,2b as well
as aryl-substituted 3-furylphosphonates.2c In addition to their
synthetic utility, ꢀ-aryl-ꢀ-ketophosphonates are known to pos-
sess a range of biological activity.3 Some ꢀ-aryl-ꢀ-ketophos-
phonates inhibit the proliferation of tumor cells,3a some possess
bone anabolic activity,3b and others have been identified as
thyroid receptor ligands.3c
Several methods of preparing ꢀ-aryl-ꢀ-ketophosphonates have
been described in the literature.4 All of the literature methods
(b) Arcoria, A.; Fisichella, S. Ann. Chim. 1967, 57, 1228. (c) Sampson, P.;
Hammond, G. B.; Wiemer, D. F. J. Org. Chem. 1986, 51, 4342. (d) Calog-
eropoulou, T.; Hammond, G. B.; Wiemer, D. F. J. Org. Chem. 1987, 52, 4185.
(e) Savignac, P.; Coutrot, P. Synthesis 1978, 682. (f) Kim, D. Y. Bull. Korean
Chem. Soc. 1997, 18, 339. (g) Kim, D. Y.; Kong, M. S.; Lee, K. J. Chem. Soc.,
Perkin Trans. 1 1997, 1361. (h) Kim, D. Y.; Kong, M. S.; Rhie, D. Y. Synth.
Commun. 1995, 25, 2865. (i) Kim, D. Y.; Kong, M. S.; Kim, T. H. Synth.
Commun. 1996, 26, 2487. (j) Lee, K.; Wiemer, D. F. J. Org. Chem. 1991, 56,
5556.
(1) For a review, see: Marynoff, B. E.; Reitz, A. B. Chem. ReV. 1989, 89,
863.
(2) (a) Verbicky, C. A.; Zercher, C. K. J. Org. Chem. 2000, 65, 5615. (b)
Kjærsgaard, A.; Jørgensen, K. A. Org. Biomol. Chem. 2005, 3, 804. (c) Ruzziconi,
R.; Couthon-Gourve`s, H.; Gourve`s, J.-P.; Corbel, B. Synlett 2001, 703.
(3) (a) Nguyen, L. M.; Diep, V. V.; Phan, H. T.; Niesor, E. J.; Masson, D.;
Guyon-Gellin, Y.; Buattini, E.; Severi, C.; Azoulay, R.; Bentzen, C. L. WO
2004026242, 2004. (b) Nguyen, L. M.; Diep, V. V.; Phan, H. T.; Niesor, E. J.;
Masson, D.; Guyon-Gellin, Y.; Buattini, E.; Severi, C.; Azoulay, R.; Bentzen,
C. L. WO 2004026245, 2004. (c) Erion, M. D.; Jiang, H.; Boyer, S. H. U.S.
20060046980, 2006.
(5) Cragoe, E. J.; Woltersdorf, O. W.; Gould, N. P.; Pietruszkiewicz, A. M.;
Ziegler, C.; Sakurai, Y.; Stokker, G. E.; Anderson, P. S.; Bourke, R. S.;
Kimelberg, H. K.; Nelson, L. R.; Barron, K. D.; Rose, J. R.; Szarowski, D.;
Popp, A. J.; Waldman, J. B. J. Med. Chem. 1986, 29, 825.
(4) (a) Corey, E. J.; Kwiatkowski, G. T. J. Am. Chem. Soc. 1966, 88, 5654.
(6) Coutrot, P.; Ghribi, A. Synthesis 1986, 661.
10.1021/jo800973e CCC: $40.75
Published on Web 07/22/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 6397–6400 6397

SCHEME 1. General Approach of Existing Methods to
Prepare ꢀ-Aryl-ꢀ-ketophosphonates
SCHEME 3. Smyth’s Proposed Acylation Mechanism
Applied to Phosphonoacetic Acids
SCHEME 2. Proposed Strategy for the Preparation of
ꢀ-Aryl-ꢀ-ketophosphonates
indeed able to acylate 2c with 3a using TFAA and 85%
phosphoric acid to afford 1b in comparable yield and selectivity.
Unfortunately, however, this material is also a syrup and
therefore could not be purified by crystallization.
In an effort to identify a solid product, we prepared the
dimethyl ketophosphonate 1c. Thus, TFAA/H3PO4-mediated
acylation of 2c with dimethylphosphonoacetic acid (3b) pro-
duced 1c with comparable efficiency and selectivity as previ-
ously observed (∼85% yield, 88:12 para/ortho). We were
pleased to find that in this instance the product turned out to be
a solid as hoped. Simple crystallization of the crude product
from n-BuOAc afforded 1c in 66% isolated yield and >98%
purity with <1% of the undesired ortho isomer.
However, employing the rarely used combination of TFAA and
85% phosphoric acidsconditions originally described by Galli7
and later studied in greater detail by Smyth8swe were pleased
to find the desired reaction had occurred, and the corresponding
ꢀ-aryl-ꢀ-ketophosphonate 4 was isolated in very good yield
(87%), albeit as an 88:12 mixture of the desired para isomer
and the undesired ortho isomer (eq 2). While we were certainly
pleased with this initial result, we were somewhat disappointed
the reaction was not as regioselective as the Friedel-Crafts
acylation of 2a with simple acid chlorides.
By analogy to Smyth’s proposed acylation mechanism,8b the
phosphoric acid presumably serves to convert an initially formed
acyl trifluoroacetate (III) to the much more reactive acyl
bis(trifluoroacetyl)phosphate (IV, Scheme 3). The amount of
TFAA required is determined by the amount of phosphonoacetic
acid and phosphoric acid used in the reaction. For every
equivalent of phosphonoacetic acid, 1 equiv of TFAA is
consumed in forming III. For every equivalent of 85%
phosphoric acid used, an additional 3 equiv of TFAA are
requiredsone is hydrolyzed by the ca. 1 molar equiv of water,
and two are consumed in the formation of IV. The carboxylic
acids surveyed by Galli and Smyth were simple, unfunction-
alized carboxylic acids, and no phosphonoacetic acids were
used.9
In the TFAA/H3PO4-mediated acylations of 2a and 2b with
3a, we used 4 equiv of TFAA and 1 equiv of 85% H3PO4. When
2c was used as the substrate, an additional equivalent of TFAA,
which effectively served to protect the alcohol as a TFA ester
in situ, was required. The TFA ester readily hydrolyzed during
the workup. We were concerned with the expense associated
with using 5 equiv of TFAA on scale, so we explored reducing
the stoichiometry of the phosphoric acid and TFAA. Smyth had
demonstrated that substoichiometric amounts of phosphoric acid
are sufficient with some of the more reactive arenes, as his
proposed mechanism would suggest.10 Indeed we were able to
reduce the amounts of H3PO4 and TFAA required to 0.2 and
3.0 equiv, respectively (in theory, only 2.6 equiv of TFAA is
required with 0.2 equiv of 85% H3PO4; we used 3.0 equiv out
of convenience). We found that with the reduced quantities of
We next looked at the acylation of 2b, which incorporates a
hydroxyethyl chain protected as a benzoate ester, with 3a using
the TFAA/H3PO4 conditions. Once again, we were pleased to
find the desired reaction had taken place to deliver 1a in very
good yield; however, no improvement in the regioselectivity
of the reaction was observed. Furthermore, since 1a is a syrup,
it could not be purified by crystallization, which would be the
most preferred method of purification on scale.
Our inability to easily separate the two regioisomers was not
our only concern at this point. Our plan for the HWE reaction
in our system was to use a slight excess of 1 to ensure complete
consumption of our aldehyde, and then simply wash out
unconsumed 1 with an alkaline aqueous extraction during the
workup. Unfortunately 1a was too lipophilic to be easily washed
out in an alkaline aqueous extraction. However, saponification
of 1a with LiOH yielded 1b, which possesses the desired
solubility in aqueous base. This observation prompted us to
consider preparing 1b directly by using 2c, which has no
protecting group, in the acylation reaction. We did not need
the protecting group for any of the planned downstream
chemistry, including the HWE reaction and, prior to identifying
conditions for preparing 1, had simply assumed a protecting
group was needed for the acylation step. Proceeding without a
protecting group would further streamline our process by
eliminating protection-deprotection steps. Gratifyingly, we were
(9) The acylation of indole and 2-methylindole with diethylphosphonoacetic
acid (3a) activated by acetic anhydride has previously been described: Sla¨tt, J.;
Janosik, T.; Wahlstro¨m, N.; Bergman, J. J. Heterocycl. Chem. 2005, 42, 141.
When we attempted to prepare 4 from 2a using these same conditions, we saw
no reaction. Upon addition of 85% H3PO4 to the reaction mixture, 2a was
converted to 2,3-dimethyl-4-methoxyacetophenone nearly quantitatively.
(10) Smyth also demonstrated that the TFA generated in the reaction can be
recovered and converted back to TFAA. See ref 8a.
(7) Galli, C. Synthesis 1979, 303.
(8) (a) Smyth, T. P.; Corby, B. W. Org. Process Res. DeV. 1997, 1, 264. (b)
Smyth, T. P.; Corby, B. W. J. Org. Chem. 1998, 63, 8946.
6398 J. Org. Chem. Vol. 73, No. 16, 2008

TABLE 1. ꢀ-Aryl-ꢀ-ketophosphonates (5) Prepared by the
TFAA/H3PO4-Mediated Acylation of Arenes with Phosphonoacetic
Acids (3)a
SCHEME 4. Formation of a ꢀ,ꢀ-Diarylvinylphosphonate
5b and 5d). In certain instances, the dimethylphosphonate
products may be preferred over the diethylphosphonate products
for potentially different physical properties (crystallinity, aque-
ous solubility, etc.). Phosphonoacetic acids bearing R-substit-
uents can also be used in the reaction (5p-r). As already
demonstrated in the case of 2, some functional groups, such as
alcohols and esters (5n and 5o), which may not be compatible
with literature methods of preparing ꢀ-aryl-ꢀ-ketophosphonates,
are tolerated by using the TFAA/H3PO4 conditions.
The workup was designed to minimize or eliminate subse-
quent purification procedures. For most substrates the following
workup was used. Once the reaction was deemed complete
(either because of complete consumption of starting arene or
no further conversion), the reaction mixture was cooled to 0
°C and diluted with water. The pH of the mixture was then
adjusted to 8-9 by the addition of aqueous NaOH. After
warming to rt, the mixture was then extracted with an organic
solvent (typically CH2Cl2, toluene, or MTBE). The phosphate
and TFA salts, as well as any unconsumed phosphonoacetic
acid, remain in the aqueous layer, which is discarded. The
product-containing organic extract is then extracted with 1 N
aq NaOH. The acidic ꢀ-aryl-ꢀ-ketophosphonate partitions into
the aq NaOH layer, while the uncharged components, including
any remaining starting arene and diarylvinylphosphonate (vide
infra), remain in the organic layer.11 The product-containing
aqueous layer is then acidified with concentrated HCl and
extracted with fresh organic solvent (typically CH2Cl2, toluene,
or MTBE). This extract is then concentrated to yield pure
product, which can be used in subsequent chemical transforma-
tions without further purification. This generalized workup
procedure was not optimized for every example, and differences
were noted in the relative solubilities of the various ꢀ-aryl-ꢀ-
ketophosphonates. Therefore some of the yields in Table 1 may
be improved with further optimization in the workup.
In general, the acylations are reasonably clean reactions.
Modest to low yields can generally be attributed to incomplete
conversion or low recovery in the extractive workup due to
solubility issues. The only byproduct of note is a ꢀ,ꢀ-
diarylvinylphosphonate (6) resulting from incorporation of a
second equivalent of the starting arene into the product. In many
cases, diarylvinylphosphonate formation was minimal. One
notable exception was the acylation of 2-methylanisole, where
diarylvinylphosphonate formation was rather significant (1:1
mixture of 5j/6j, Scheme 4). Fortunately, the diarylvinylphos-
a All reactions were run neat with 1.0 equiv of arene, 4.0-4.8 equiv
of TFAA, 1.0-1.2 equiv of phosphonoacetic acid, and 1.0-1.2 equiv of
85% H3PO4. All reactants were combined at 0 °C, and the reaction
mixture then warmed to rt (except where noted) for a period of 2-24 h.
Products were isolated by using the extractive procedure described in
the text, except where noted. In cases where regioisomeric mixtures
were produced, the ratio is noted below the yield. b Run at 50 °C.
c Product extracted at pH 8-9, then purified by column chromatography.
d Product only extracted at pH 8-9
these reagents, heating to 50 °C was necessary to achieve the
desired conversion in a reasonable period of time. With these
modified conditions, 2c was acylated with 3b to provide 1c in
comparable yield and selectivity as in the stoichiometric case.
In similar fashion, 2d and 2e were converted to ketophospho-
nates 1d and 1e, respectively. In all cases, simple crystallization
delivered the desired para isomer in very pure form.
We have explored the generality of the TFAA/H3PO4-
mediated acylations of electron-rich arenes with phosphonoacetic
acids as a means of preparing ꢀ-aryl-ꢀ-ketophosphonates. Table
1 lists several of the ꢀ-aryl-ꢀ-ketophosphonates we prepared.
In this survey, we used one standard set of conditions that may
not be optimal for all substrates. Briefly, the reactions were run
by combining 1.0 equiv of the arene, 4.0-4.8 equiv of TFAA,
and 1.0-1.2 equiv of the phosphonoacetic acid, followed by
adding 1.0-1.2 equiv of 85% H3PO4 at 0 °C. The reaction
mixtures were then stirred at rt or, in some cases, 50 °C for a
period of 2-24 h. Better yields were typically obtained with
the more reactive substrates. With less reactive substrates such
as toluene, biphenyl, 2-chloroanisole, and 4-fluoroanisole lower
yields were observed, primarily due to lower conversion (5e,
5f, 5k, and 5l). With the less reactive substrates, heating to 50
°C was beneficial. Most reactions were simply run by using
commercially available diethylphosphonoacetic acid (3a). How-
ever, other phosphonoacetic acids can be used. Dimeth-
ylphosphonoacetic acid (3b) worked just as well (5a and 5c vs
(11) A similar purification procedure has previously been described. See ref
4e.
J. Org. Chem. Vol. 73, No. 16, 2008 6399

with 1,4-dimethoxybenzene (1.38 g, 10 mmol) and TFAA (6.7 mL,
48 mmol). The mixture was cooled to 0 °C with stirring and treated
with dimethylphosphonoacetic acid (3b) (2.02 g, 12.0 mmol)
followed by 85% H3PO4 (1.38 g, 12.0 mmol). After 5 min, the ice
bath was removed, and the orange reaction mixture was stirred at
rt. After 24 h, the reaction mixture was cooled to 0 °C and diluted
with water (30 mL). Next, the pH was adjusted to ∼9 by the careful
addition of 10 N aq NaOH (∼12 mL). The ice bath was removed,
and the mixture was stirred for 10-15 min before it was extracted
with toluene (2 × 30 mL). The combined toluene extracts were
then extracted with 1 N aq NaOH (1 × 40 mL, then 1 × 20 mL).
The combined NaOH extracts were acidified to pH <2 by the
addition of concd HCl (∼5 mL). This acidified aqueous layer was
then extracted with dichloromethane (1 × 50 mL, then 1 × 25
mL). The combined dichloromethane extracts were then concen-
trated in vacuo to afford 5c (1.57 g, 54%) as a pale yellow oil. 1H
NMR (300 MHz, CDCl3) δ 7.31 (d, J ) 3.0 Hz, 1H), 7.07 (dd, J
) 9.1, 3.3 Hz, 1H), 6.03 (d, J ) 8.8 Hz, 1H), 3.90 (s, 3H), 3.86 (d,
phonate can be separated from the desired ꢀ-aryl-ꢀ-ketophos-
phonate in the extractive workup. At this point it is not
completely clear how the diarylphosphonate is formed nor why
it is more favorable in certain cases versus others. However,
we believe its formation may involve an intermediate vinyl
trifluoroacetate. In some instances we have seen mass spectral
evidence (from LCMS analysis of the reaction mixture prior to
workup) for the presence of low levels of this intermediate.12
The vinyl trifluoroacetate is simply hydrolyzed during the
workup and not detected in the crude product or the neutral
extracts. Separately, we have found that isolated 5j can be
subsequently treated with 2-methylanisole in the presence of
TFAA and 85% H3PO4 to form 6j in 79% isolated yield.13
We have attempted to expand the scope of the TFAA/H3PO4-
mediated acylations with phosphonoacetic acids to other electron-
rich systems such as furan, 2-methylthiophene, and ferrocene.
However the reaction mixtures in these cases have been much
more complex, with only low yields of the desired ketophos-
phonates being produced. We have also just begun to extend
the utility of the TFAA/H3PO4 acylation conditions to the
preparation of ꢀ-aryl-ꢀ-ketoesters with malonic acid monoesters;
those results will be reported in due course.
In summary, we have described a simple and efficient method
of preparing ꢀ-aryl-ꢀ-ketophosphonates using the TFAA/H3PO4-
mediated acylation of arenes with phosphonoacetic acids. The
conditions offer advantages over existing methods in that they
do not require the use of strong base, cryogenics, or an
anhydrous and inert atmosphere. Furthermore, some functional
groups not tolerated with the reaction conditions used in existing
methods are compatible with the herein described conditions.
This methodolgy was used successfully in the efficient prepara-
tion of over 80 kg of a key ꢀ-aryl-ꢀ-ketophosphonate intermedi-
ate for a development program at Neurogen.
3
2JHP ) 21.4 Hz, 2H), 3.79 (s, 3H), 3.76 (d, JHP ) 11.0 Hz, 6H);
31P NMR (121 MHz, CDCl3) δ 25.34; 13C NMR (75 MHz, CDCl3)
δ 192.4 (d, 2JCP ) 7.8 Hz), 153.3, 153.2, 127.1, 121.3, 113.9, 113.1,
56.0, 55.6, 52.7 (d, 2JCP ) 5.9 Hz), 41.4 (d, 1JCP ) 133 Hz). Exact
mass (C12H17O6P + H) calculated 289.0841, measured 289.0845.
The toluene extract was separately concentrated to yield 749 mg
of a ∼7:1 mixture of dimethyl [2,2-bis(2,5-dimethoxyphenyl)vi-
nyl]phosphonate (6c) and 1,4-dimethoxybenzene as a yellow syrup.
1
Characterization of 6c: H NMR (300 MHz, CDCl3) δ 6.99 (d, J
) 2.8 Hz, 1H), 6.86-6.76 (m, 4H), 6.67 (d, J ) 3.0 Hz, 1H), 6.37
2
(d, JHP ) 18.1 Hz, 1H), 3.80 (s, 3H), 3.70 (s, 3H), 3.69 (s, 3H),
3.60 (s, 3H), 3.52 (d, J ) 11.3 Hz, 6H); 31P NMR (121 MHz,
2
CDCl3) δ 20.40; 13C NMR (75 MHz, CDCl3) δ 154.6 (d, JCP
)
3
5.9 Hz), 153.1, 152.9, 151.4, 150.7, 131.2 (d, JCP ) 23.4 Hz),
129.6 (d, 3JCP ) 7.8 Hz), 119.8, 117.3, 116.6, 116.4, 114.7, 114.5,
114.1, 113.2, 112.1, 56.5, 56.2, 55.7, 55.6, 51.9 (d, 2JCP ) 5.9 Hz)
(1JCP not determined due to complexity of the spectrum at >100
ppm). Exact mass (C20H25O7P + H) calculated 409.1416, measured
409.1394.
Experimental Section
Acknowledgement. We gratefully acknowledge the contribu-
tions of Mark T. Kershaw, Neurogen Corporation, who per-
formed the HRMS analyses. We also thank Dr. Andrew Staab,
Neurogen Corporation, for helpful discussions.
Representative Acylation Procedure: Preparation of Dimethyl
[2-(2,5-Dimethoxyphenyl)-2-oxoethyl]phosphonate (5c). A 50 mL
round-bottomed flask equipped with a magnetic stir bar was charged
Supporting Information Available: Experimental proce-
dures for the preparation of compounds 3b, 3c, 5j, and 6j,
(12) A vinyl trifluoroacetate product has previously been observed when
arylaceticacidswereusedinTFAA/H3PO4-mediatedacylationsofarenes:Veeramaneni,
V. R.; Pal, M.; Yeleswarapu, K. R. Tetrahedron 2003, 59, 3283.
(13) It has been reported that ferrocene will react with diethyl (2-oxo-2-
phenylethyl)phosphonate and diethyl (2-oxopropyl)phosphonate in the presence
of triflic or methanesulfonic acid to form the corresponding ꢀ-ferrocenylvi-
nylphosphonates: Plaøuk, D.; Rybarczyk-Pirek, A.; Zakrzewski, J. J. Organomet.
Chem. 2004, 689, 1165.
1
characterization data for compounds 1c-e, and copies of H,
31P, and 13C NMR spectra for all new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
JO800973E
6400 J. Org. Chem. Vol. 73, No. 16, 2008