Synthesis, characterisation and properties of bismuth(III) ester pendant arm picket porphyrins

Article abstract of DOI:10.1039/b308776j

A series of porphyrins bearing ester pendant arms has been synthesised. The number of potential coordinating groups was varied from one to four. In the case of the four-picket compounds, the arm length was also investigated. Structural properties towards

Full text of DOI:10.1039/b308776j

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, characterisation and properties of bismuth(III) ester
pendant arm picket porphyrins†
Zakaria Halime, Lydie Michaudet, Mathieu Razavet, Christian Ruzié and Bernard Boitrel*
Institut de Chimie, UMR CNRS 6509, Université de Rennes 1, Av. du Général Leclerc,
Campus de Beaulieu, 35042 Rennes cedex, France.
E-mail: Bernard.Boitrel@univ-rennes1.fr; Fax: 02 2323 5637; Tel: 02 2323 6372
Received 28th July 2003, Accepted 8th September 2003
First published as an Advance Article on the web 19th September 2003
A series of porphyrins bearing ester pendant arms has been synthesised. The number of potential coordinating
groups was varied from one to four. In the case of the four-picket compounds, the arm length was also investigated.
Structural properties towards the coordination of the bismuth() cation were studied. In the particular case of the
two-picket molecules, the substitution isomers (5,10 or 5,15) were also prepared. An improvement of the rate of
metallation was detected in the case of the single-picket porphyrin.
four-picket porphyrin, although no full characterisation of the
Introduction
product was supplied. We reasoned on that if we were able
to obtain comparable increase in metallation rate with the
bismuth() cation, the porphyrin could become a compatible
ligand for a use in α-radiotherapy; reminding that the half-life
time of 213Bi is only of 46 min.3 Study of the same type of
porphyrins was therefore investigated, with ours bearing ester
pendant arms as depicted in Scheme 1. Assuming that if the
pickets were indeed flexible enough, they could coordinate and
presumably stabilize the bismuth within the macrocycle. It is
evident that the picket length is a decisive structural parameter.
Thus, we prepared three different porphyrins by the reaction of
commercially available acyl chlorides with αααα-TAPP. The
acylation reagents were the single acyl chloride of ethyl malonyl
(EM) acid, ethyl succinyl (ES) acid and methyl glutaryl (MG)
acid. Such compounds are readily synthesised and the yields
almost quantitative. Metal insertion was achieved by heating
with a solution of bismuth nitrate (10 equiv.) in pyridine at
50 ЊC for 2 h. It is worth noting that these conditions are
relatively gentle compared to the usual preparation of bismuth
porphyrins, e.g. overnight reflux in pyridine or benzonitrile.
The different complexes were purified by chromatography on
silica-gel and eluted with MeOH–CH2Cl2. During this process,
formation of free-base porphyrin was scarcely observed from
bismuth release unlike previously reported for unfunctionalized
porphyrins.7
Bismuth is commonly used in antiulcer1 and antibacterial drugs
and it has recently revealed potential interest in cancer therapy.2
212Bi and 213Bi isotopes, which have short half-lives of 61 and 46
min, respectively, are α and β-particle and γ-ray emitters. Due
to its high linear energy transfer, α-particle decay is particu-
larly promising for tumour destruction. 213Bi produces α and
β-particles along with low energy γ-rays and this latter property
makes 213Bi a rather preferred element in radiotherapy.3
BiIII and BiV are the two common oxidation states of the
element. The chemistry of BiIII has been largely more developed
and has led to synthesis of numerous complexes with co-
ordination number varying from three to nine.4 Its affinity with
nitrogen and oxygen atoms makes bismuth a good candidate
for complexation with chelators and macrocycles such as di-
ethylenetriaminepentaacetic acid (DTPA) and 1,4,7,10-
tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA).5 The
high thermodynamic constants obtained for these complexes
have led us to investigate the reactivity of bismuth with
porphyrins.
Bismuth porphyrins are not very widespread but a few exist
and have been fully characterised. First examples reported were
from Treibs in 1969 and Buchler in 1974.6 Since then, Brothers
and co-workers7 as well as Guilard and co-workers8 have
investigated the complexation of bismuth with numerous por-
phyrins. In parallel to the work of Brothers and co-workers, we
have developed a route with porphyrins bearing potentially co-
ordinating pendant arms.9 These arms are directed in a per-
pendicular fashion with respect to the porphyrin macrocycle
and offer the possibility of bending in order to satisfy the metal
coordination sphere. In a recent communication we discussed
the crystal structure of a four-ester arm bismuth porphyrin in
which one carbonyl oxygen belonging to an identical neigh-
bouring porphyrin binds to the bismuth atom. Herein we report
the synthesis and characterisation of a series of ester arm
porphyrins along with their bismuth derivatives.
Results and discussion
This study was originally motivated by
a report from
Buckingham et al. dealing with the rapid insertion of different
metals such as copper in a porphyrin.10 This publication
described the reaction of meso(tetra-o-aminophenyl)porphyrin
(atropisomer αααα, TAPP) with maleic anhydride, leading to a
† Electronic supplementary information (ESI) available: characteris-
ation data for the remainder of the compounds. See http://www.rsc.org/
suppdata/dt/b3/b308776j/
Scheme 1 Bismuth coordination to the four-ester porphyrins.
D a l t o n T r a n s . , 2 0 0 3 , 4 2 5 0 – 4 2 5 4
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
4250

View Article Online
Additionally, as shown by the proton NMR data, the C4
symmetry of the different complexes is retained as no atropiso-
merisation occured. The metallation rate was monitored by
UV-visible spectroscopy. The Soret band of the bismuth por-
phyrin appears around 470 nm as shown on Fig. 1 for the form-
ation of 1BiNO3, the latter possessing four pickets of the ES
type. After 90 min, metal insertion is complete as indicated by
the loss of free-base Soret band at 420 nm. Although certainly
not sufficient for an effective use in therapy as 75% of the activ-
ity would be lost after such a period, this result represents a
significant improvement compared to porphyrins such as tetra-
phenylporphyrin (TPP) which requires elevated temperatures
and 6 h of stirring.
Fig.
3
Top view of the 1BiNO3 dimer illustrating clearly the
antiprismatic conformation around the bismuth centre and also the
displacement of the two macrocycles to each other.
belonging to an arm attached to a symmetrically related
macrocycle, forming in this way, a centrosymmetric dimer in
the solid state. In contrast to the previously reported X-ray
structure of (OEP)BiSO3CF3,8 the bridge between the two
porphyrins is not built by coordination of the counter-anion
but by the intramolecular coordinating group, namely the
carbonyl of the ester group. The Bi atom lies 1.125 Å above
the four-nitrogen plane and the mean Bi–N bond length is
2.34(2) Å; these values are similar to those observed in
Fig. 1 Monitoring of bismuth insertion in 1.
On the other hand, the same study applied to 2 and 3 did not
lead to significant variation in time of metallation. This observ-
ation seems inconsistent with our first hypothesis that folding
of the pickets helps to stabilize the metal. In the case of the
ethyl succinyl picket, we were able to obtain single crystals suit-
able for X-ray study. These were obtained by slow diffusion of
H2O–MeOH onto a saturated THF solution of 1BiNO3. The
ORTEP11 plot of the X-ray structure is shown in Fig. 2.9
(OEP)Bi(SO3CF3) which also adopts
a centrosymmetric
dimeric form (∆4N = 1.07 Å and <Bi–N> = 2.31(1) Å). The
Bi–O bond lengths are larger than the sum of their ionic radii
(1.13 ϩ 1.40 = 2.53 Å);12 2.706(5) and 2.789(4) Å for the bonds
with the nitrate anion (respectively with O13 and O15), 2.816(5)
Å with the oxygen of the water molecule (Ow1) and 3.018(5) Å
with the oxygen atom of the carbonyl group (O8#). A second
water molecule (Ow2) is present in the cage and as shown in
Fig. 2, both water molecules are engaged in an intra- and
inter-molecular hydrogen bonding network, contributing to
the stability of the dimer. One arm of the molecule does
not participate in any interaction and thus exhibits a regular
conformation. As mentioned above, one arm participates to the
coordination sphere of the Bi atom belonging to the second
molecule of the dimer, and the two remaining arms involved
in the hydrogen bonding network with Ow1 and Ow2,
adopt folded conformations. The porphyrin core is not planar
and adopts a dome distortion and this is especially reflected
by the deviation of the eight β-pyrrolic carbon atoms.13
These can be separated by up to 0.312 Å from the porphyrin
mean base plane, with an average deviation of 0.138 Å. This is
the result of the out-of-plane coordination of the metallic
cation.
Fig. 3 illustrates the displacement of the two macrocycles
relative to one another. Reported X-ray structures of bismuth
porphyrins have always shown dimers with close super-
imposition of the macrocycles.7 In our case the rings are hori-
zontally separated by a significant 7.240 Å. The figure shows
also the coordinating ester arms are positioned right through
the anisotropic current of the porphyrin. The natural edged
position of the ester arms on each porphyrin undoubtedly
favours this displacement.
Fig. 2 Molecular structure of the 1BiNO3 dimer showing the two
macrocycles related to each other by the coordination of their
respective ester arm. The dotted lines illustrate the hydrogen bonding
mode between coordinated and inclusion water molecules. Hydrogen
atoms have been omitted for clarity.
In an attempt to uncover the possiblity for this dimeric
assembly to exist in solution, 1H NMR spectroscopy for
1BiNO3 was studied. Variable-temperature NMR spectra are
shown in Fig. 4. At 300 K, the methylene protons of the
ester groups appear at 3.37 ppm as a broad singlet whereas at
290 K, they become magnetically non-equivalent with
two singlets at 3.31 and 3.39 ppm. The same observation is
The BiIII is eight-coordinate with a slightly distorted square-
antiprismatic geometry, Fig. 3. The four nitrogen atoms of the
macrocycle form a square, with the other distorted square being
formed by four oxygen atoms described as follows: two oxygen
atoms of the nitrate anion, the oxygen atom of a water mole-
cule and a carbonyl oxygen atom of the terminal ester group
D a l t o n T r a n s . , 2 0 0 3 , 4 2 5 0 – 4 2 5 4
4251

View Article Online
The metal insertion in these four new compounds was
achieved using the same procedure to that of the preparation of
1BiNO3. Products were also purified by chromatography on
silica gel with MeOH–methylene chloride and this occurred
without any significant demetallation. This indicates that the
stabilities of these bismuth picket porphyrins are not seriously
affected by the number of ethyl succinyl pickets although
isomer 10BiNO3 seems slightly more stable than the other
complexes during chromatography. Metallation kinetics were
followed by UV-visible spectroscopy and again, we applied the
same strategy to that of 1BiNO3. For all ligands but one, the
metal insertion was found to be complete after 2 h in pyridine at
50 ЊC. Surprisingly, the reaction was found to be faster in the
case of porphyrin 11 bearing one picket only, Fig. 5. In the same
conditions of solvent and temperature, the metallation reaction
was complete after only 45 min. This increase in rate may be
attributed to the lack of steric hindrance above the macrocycle
making metal insertion faster. Indeed, in the X-ray structure of
1BiNO3, only one picket binds directly to the bismuth. If we
assume that the same binding mode exists also in 11BiNO3, the
same stabilization of the metal occurs but with much faster
rates.
Fig. 4 Variable-temperature 1H NMR for 1BiNO3 in the 0–4 ppm
region.
made for the methyl group at 0.85 ppm. It is worth mentioning
that a 1 : 3 intensity ratio is observed instead of an expected
1 : 2 : 1.
As the temperature decreases, one can clearly assess that a
single pendant arm of the porphyrin is in a different magnetic
environment than the three others, as shown by the integration
of the signal. The same temperature dependence is observed for
the aromatic protons of the phenyl rings, with the exception of
the apparent triplet at 7.90 ppm.14 These observations have led
us to suspect an unsymmetrical geometry around the BiIII ion
involving one arm of the porphyrin. This is somewhat com-
patible with the X-ray crystallographic structure of 1BiNO3
although it is not clear whether or not geometries are identical
in solution and in solid state. However, it seems more probable
that the dimer does not exist in solution. Indeed, the protons of
the pendant arm bound to the bismuth via the carbonyl oxygen,
would be expected to be strongly shifted upfield because of
their localization in the anisotropic current of the porphyrin
and this shift is not observed. All the aliphatic protons are
normally shifted between 0.5 and 3.5 ppm and therefore the 1H
NMR spectrum is in agreement with a monomeric complex in
which only one of the picket is capable of interacting with the
metal.
Fig. 5 Monitoring of bismuth insertion in 11.
In conclusion, we have explored possibilities of bismuth
complexation by ester arm picket porphyrins. We have shown
that tethering at least one ester picket above the porphyrin
plane stabilizes the complex enough and enables us to fully
characterise products. For the four-picket porphyrin, this
stabilization may be explained by the dimeric edifice in
which a reciprocal coordination of the metals is ensured by one
carbonyl from an ester function. In the case of the two-picket
porphyrins, without an X-ray structure, it is more difficult to
rationalise the greater stability of the 5,10 isomer. Additionally,
the rate of metallation is neither affected by the length of the
picket nor by their relative positions but most certainly by the
lack of steric hindrance. With the one-ester arm porphyrin 11,
it seems that we have found a good balance between the
thermodynamic stability and the rate of bismuth insertion.
As mentioned earlier, the length of the pickets has no notable
effect on both the metallation rates by Bi(NO3)3 and the stabil-
ity of the complex. It was therefore then logical to investigate
the influence of the number of ester pickets.
To do so, we have undertaken the synthesis and characteris-
ion of four other picket porphyrins 8–11 bearing respectively
three, two, two and one ester picket on the same side of the
porphyrin plane (Scheme 2). The synthesis is preceded by one
step consisting of the protection of the undesired amino func-
tions. Thus, αααα-TAPP was allowed to react with 1.8 equiv.
acetyl chloride and all posssible products were obtained, from
the single acetylated to the three acetylated. The four-acetylated
porphyrin is only present as a trace on TLC plates. The problem
with this synthesis is that the two acetylated porphyrins 5 and 6
are not isolable separately at this stage and acylation by the
ethyl succinyl chloride is needed in order to be able to separate
Experimental
General considerations. 1H (500.13 MHz, 300.13 MHz) NMR
spectra were recorded on Bruker Avance spectrometers and
referenced to the residual protonated solvent. Mass spectra were
performed on a MS/MS ZABSpec TOF spectrometer at the
University of Rennes I (C.R.M.P.O.). UV-visible spectra were
recorded on a Varian Cary 1E and an Uvikon XL spectro-
meters. Infrared spectra were recorded on Bruker IFS 66 and 28
spectrometers. All solvents (ACS for analysis) were purchased
from Carlo Erba. THF was distilled from potassium metal
whereas methanol was distilled from magnesium turnings;
CH2Cl2 was used as received. Triethylamine was distilled from
CaH2. The starting materials were generally used as received
(Acros, Aldrich) without any further purification. All reactions
were performed under an argon atmosphere and monitored by
1
products. Unfortunately, even on a 500 MHz H NMR spec-
trum, the expected patterns for the 5,15 and 5,10 isomers 9 and
10 are not observed. Indeed, for the 5,15 diacetylated ligand 9,
one would expect to obtain two doublets, each integrating for
4H and in the case of the 5,10 disubstituted porphyrin 10, four
doublets integrating for 2H each.
In order to clarify the ambiguity surrounding the two iso-
mers, porphyrin 9 was synthesized by an alternative route
1
depicted in Scheme 3. This route is unambiguous as the H
NMR spectrum of 13 exhibits two expected doublets at 8.99
and 8.84 ppm with the typical β-pyrrolic coupling constant of
4.7 Hz.15
D a l t o n T r a n s . , 2 0 0 3 , 4 2 5 0 – 4 2 5 4
4252

View Article Online
Scheme 2 Synthetic pathway to the desired acetyl-ester porphyrins.
TLC (silica, CH2Cl2–MeOH). Column flash chromatography
M
Ϫ1 cmϪ1): 419 (337), 514 (20.9), 546 (5.8), 588 (6.7), 643 (3.7).
was performed on silica gel (Merck TLC-Kieselgel 60H, 15
µm). Elemental analyses were obtained on an EA 1108 Fisons
Instrument.
1H NMR (CDCl3, 323 K, 500 MHz): δH 8.86 (8H, s, β-pyr), 8.59
(4H, d, J = 7.1 Hz, arom), 8.00 (4H, d, J = 7.5 Hz, arom), 7.86
(4H, t, J = 7.6 Hz, arom), 7.55 (4H, t, J = 7.2 Hz, arom),
7.19 (4H, s, –NHCO), 3.44 (8H, br s, –CH2CH3), 2.18 (8H, t,
J = 6.7 Hz, –CH2CH2), 1.70 (8H, br s, –CH2CH2), 0.79 (12H, br
s, –CH2CH3), Ϫ2.59 (2H, s, –NH pyr). 13C NMR (CDCl3, 300
K, 125 MHz): δC 172.6, 170.1, 138.7, 135.4, 132.1, 130.3, 123.9,
122.9, 115.5, 60.7, 31.5, 29.0, 14.1.
Typical procedures
ꢀ-5,10,15,20-Tetrakis{2-[3-(ethoxycarbonyl)propionylamido]-
phenyl}porphyrin 1. TAPP (0.2 g, 0.3 mmol) was dissolved
in 20 mL freshly distilled THF under argon. 10 equivalents
anhydrous Et3N were added and the mixture was cooled to
0 ЊC. Ethyl succinyl chloride (5 equiv.) was then added and the
solution stirred overnight. The solvent was evaporated to dry-
ness and the residue dissolved in a minimum of CH2Cl2 to be
poured on a silica gel chromatography column. The product
was eluted with 0.1% CH3OH–CH2Cl2 and obtained in
93% yield (330 mg, 0.28 mmol). Microanalysis: C68H66N8O12ؒ
CH3OH, found (calc): C, 68.25 (67.97); H, 4.44 (4.79); N, 9.33
(9.19)%. FAB-MS: m/z = 1187.8 [M ϩ H]ϩ. FTIR (KBr, cmϪ1):
1731 (CO)ester, 1694 (CO)amide. UV-VIS (CH2Cl2, λ/nm, 10Ϫ3ε/
1BiNO3. In a 50 mL flask, 1 (26 mg, 0.02 mmol) was dis-
solved in 7 mL pyridine. This was warmed up to 55 ЊC before
Bi(NO3)3ؒ5H2O (100 mg, 0.2 mmol) was added. We note here
that 10 equivalents of the bismuth salt were necessary for all the
following metallation reactions. After 2 h heating, the solvent
was evaporated and dissolved in a minimum of CH2Cl2 to be
poured on a silica gel chromatography column. The product
was eluted with 1.2% CH3OH–CH2Cl2 and obtained in 85%
yield (25 mg, 0.017 mmol). Microanalysis: C68H64BiN9O15ؒ
2H2O, found (calc): C, 55.16 (54.73); H, 4.47 (4.59); N, 8.27
D a l t o n T r a n s . , 2 0 0 3 , 4 2 5 0 – 4 2 5 4
4253

View Article Online
Scheme 3 Alternative route to porphyrin 9.
(8.45)%. FAB-MS: m/z = 1393,8 [M Ϫ NO3]ϩ. UV-VIS
References
(CH2Cl2, λ/nm, 10Ϫ3ε/MϪ1 cmϪ1): 354 (42.4), 472 (176.9), 598
(9.8), 644 (8.3). FTIR (KBr, cmϪ1): 1732 (CO)ester, 1695
(CO)amide, 1384 (NO3), 991 (Bi–N(por)). 1H NMR (CDCl3, 300
K, 500 MHz): δ 9.24 (8H, s, β-pyr), 8.68 (4H, br s, arom), 8.22
(3H, br s, –NHCO), 8.13 (1H, br s, –NHCO), 7.89 (4H, m,
arom), 7.90 (4H, t, arom), 7.56 (4H, br s, arom), 3.46 (2H,
br s, –CH2CH3), 3.32 (6H, br s, –CH2CH3), 2.04 (8H, br s,
–CH2CH2), 1.92 (8H, br s, –CH2CH2), 0.96 (3H, br s,
–CH2CH3), 0.87 (9H, s, –CH2CH3). 13C NMR (CDCl3, 300 K,
125 MHz): δC 173.5, 170.8, 149.3, 140.4, 134.9, 133.5, 131.3,
130.6, 123.6, 123.1, 122.7, 119.3, 60.8, 31.5, 29.5, 14.1.
1 G. Baxter, Chem. Ber., 1992, 28, 445.
2 R. Kozak, T. Waldmann, R. Atcher and O. Gansow, Trends
Biotechnol., 1985, 4, 259; D. E. Milenic, M. Roselli, S. Mirzadeh,
C. G. Pippin, O. A. Gansow, D. Colcher, M. W. Brechbiel and
J. Schlom, Cancer Biother. Radiopharm., 2001, 16, 133.
3 S. Hassfjell and M. W. Brechbiel, Chem. Rev., 2001, 101, 2019.
4 R. Hancock, I. Cukrowski, J. Baloyi and J. Mashishi, J. Chem. Soc.,
Dalton Trans., 1993, 2895; R. Luckay, J. Reibenspie and
R. Hancock, J. Chem. Soc., Chem. Commun., 1995, 2365; R. Luckay,
I. Cukrowski, J. Mashishi, J. Reibenspies, A. Bond, R. Rogers and
R. Hancock, J. Chem. Soc., Dalton Trans., 1997, 901.
5 M. W. Brechbiel, C. G. Pippin, T. J. McMurry, D. Milenic,
D. Roselli, D. Colcher and O. A. Gansow, J. Chem. Soc., Chem.
Commun., 1991, 1169; T. K. Nikula, M. R. McDevitt, R. D. Finn,
C. Wu, K. R. W. K. Garmestani, M. W. Brechbiel, M. J. Curcio,
C. G. Pippin, L. Tiffany-Jones, M. W. Geerlings, C. Apostolidis,
R. Molinet, M. W. Geerlings, O. A. Gansow and D. A. Scheinberg,
J. Nucl. Med., 1999, 40, 166.
Crystallography for 1BiNO3
Crystallographic details for 1BiNO3 have already been reported
in ref. 9. Crystal data: C68H64BiN9O15ؒ2H2O, M = 1492.3, tri-
¯
clinic, space group P1, a = 14.3980(4), b = 14.9790(4), c =
6 A. Treibs, Justus Liebigs Ann. Chem., 1969, 728, 115; J. W. Buchler
and K. L. Lay, Inorg. Nucl. Chem. Lett., 1974, 10, 297.
7 J. Barbour, W. J. Belcher, P. J. Brothers, C. E. F. Rickard and
D. C. Ware, Inorg. Chem., 1992, 31, 746; B. Boitrel, M. Breede,
P. J. Brothers, M. Hodgson, L. Michaudet, C. E. F. Rickard and
N. Al Salim, Dalton Trans., 2003, 1803.
8 L. Michaudet, D. Fasseur, R. Guilard, Z. Ou, K. M. Kadish,
S. Dahoui and C. Lecomte, J. Porphyrins Phtalocyanines, 2000, 4,
261.
9 L. Michaudet, P. Richard and B. Boitrel, Chem. Commun., 2000,
1589.
10 D. A. Buckingham, C. R. Clark and W. S. Webley, J. Chem. Soc.,
Chem. Commun., 1981, 192.
16.5670(4) Å, α = 116.205(1), β = 97.2490(11), γ = 92.6720(12)Њ,
V = 3159.02(14) Å3, Z = 2, Dc = 1.569 g cmϪ3, F(000) = 1516.
Data were collected at 110 K on a Nonius Kappa CCD dif-
fractometer with Mo-Kα radiation (λ = 0.71073 Å). The struc-
ture was solved by a Patterson search program and refined by
full-matrix least-squares method on F 2 (14332 unique reflec-
tions, 1054 parameters). The anisotropic refinement led to final
residuals wR2 = 0.101 for all data and R1 = 0.055 for 10533
intensities with I > 2σ(I ), and GOF = 1.028. The largest ∆(ρ)
residual densities are 1.36 and Ϫ1.69 e ÅϪ3
.
CCDC reference number 149483.
See http://www.rsc.org/suppdata/dt/b3/b308776j/ for crystal-
lographic data in CIF or other electronic format.
11 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
12 R. D. Shannon, Acta Crystallogr., Sect. A, 1976, 32, 751.
13 D. J. Nurco, C. J. Medforth, T. P. Forsyth, M. M. Olmstead and
K. M. Smith, J. Am. Chem. Soc., 1996, 118, 10918.
14 P. Permutter, M. Rose and P. Shenan, Tetrahedron Lett., 1988, 29,
1427; B. Boitrel, E. Camilleri, Y. Fleche, A. Lecas and E. Rose,
Tetrahedron Lett., 1989, 30, 2923.
15 B. Boitrel, A. Lecas and E. Rose, J. Chem. Soc., Chem. Commun.,
1989, 349.
Acknowledgements
The CNRS, La Ligue contre le cancer as well as Région
Bretagne are gratefully acknowledged for financial support.
D a l t o n T r a n s . , 2 0 0 3 , 4 2 5 0 – 4 2 5 4
4254