L-Prolinoyl chiral picket iron porphyrins evaluated for the enantioselective epoxidation of alkenes

Article abstract of DOI:10.1039/b212480g

Four atropisomers of an L-prolinoyl picket porphyrin were synthesised from tetra-o-aminophenyl porphyrin (TAPP) and were evaluated as alkene epoxidation catalysts after incorporation of iron in the porphyrin core. In the case of the αααα atropisomer bearing the four amino groups on the same side, a bulky base was employed in order to suppress the eventual reaction on the non-functionalised side of the porphyrin. The resulting enantioselectivities were compared with either other chiral motifs or with the corresponding strapped porphyrins. The enantioselectivities obtained with picket porphyrins are as high as those for strapped porphryins, and in some cases, even higher.

Full text of DOI:10.1039/b212480g

View Article Online / Journal Homepage / Table of Contents for this issue
L-Prolinoyl chiral picket iron porphyrins evaluated for the
enantioselective epoxidation of alkenes
a
Bernard Boitrel* and Valerie Baveux-Chambenoıt
b
´
ˆ
a
´
´
Laboratoire Organometalliques et Catalyse: Chimie et Electrochimie Moleculaire
(CNRS UMR 6509), Institut de Chimie de Rennes, Universite de Rennes 1,
´
Campus de Beaulieu, 35042, Rennes cedex, France. E-mail: Bernard.Boitrel@univ-rennes1.fr;
Fax: +33 (0)2 23 23 56 37; Tel: +33 (0)2 23 23 63 72
b
`
`
´
Laboratoire Synthese et Electrosynthese Organometalliques (FRE2595),
Universite de Bourgogne, 6, Bd Gabriel, 21000, Dijon, France
´
Received (in Montpellier, France) 17th December 2002, Accepted 30th January 2003
First published as an Advance Article on the web 13th May 2003
Four atropisomers of an L-prolinoyl picket porphyrin were synthesised from tetra-o-aminophenyl
porphyrin (TAPP) and were evaluated as alkene epoxidation catalysts after incorporation of iron in the
porphyrin core. In the case of the aaaa atropisomer bearing the four amino groups on the same side, a bulky
base was employed in order to suppress the eventual reaction on the non-functionalised side of the porphyrin.
The resulting enantioselectivities were compared with either other chiral motifs or with the corresponding
strapped porphyrins. The enantioselectivities obtained with picket porphyrins are as high as those for strapped
porphryins, and in some cases, even higher.
state.10 In three previous reports, such picket porphyrins bear-
Introduction
ing chiral amino acids have been synthesised. In two of them,
N-Boc-L-alanine and N-Boc-L-phenylalanine were condensed
on the abab atropisomer of tetra-o-aminophenyl porphyrin
(TAPP) by the isobutyl chloroformate method, which allows
no epimerisation during the synthesis.13,14 In the third report,
N-Boc-L-proline has been attached on the same porphyrin via
the dicyclohexylcarbodiimide (DCC) procedure, yielding zinc
porphyrins useful for the chiral recognition of amino esters.15
Our own results with L-prolinoyl strapped porphyrins pre-
pared from the abab and aabb atropisomers of TAPP
prompted us to enlarge this investigation to the four analogous
picket porphyrins based on the four atropisomers of TAPP.16
Such a study has been carried out only once9 in the case of the
epoxidation of styrene. In this study, the chiral centre came
from a binaphthyl aldehyde condensed with pyrrole and the
aabb atropisomer was shown to be the most enantioselective
with 57% ee, a result consistent with the catalyst of Collman,
Rose and coworkers.17
In our case, the four atropisomers of TAPP were primarily
separated by low pressure silica gel chromatographies and then
functionalised with the N-Boc amino acid (Fig. 1). For the
aaaa atropisomer, Lindsey’s method for the equilibrium dis-
placement was applied.18 The typical coupling reaction was
carried out using mixed anhydride activation, a classical proce-
dure in peptide synthesis.19 The formation of the amide bond
between the porphyrin and the N-Boc-L-proline was accom-
plished in dry tetrahydrofuran (THF) in the presence of N-
methylpiperidine at room temperature after activation of the
carboxylic acid of the amino acid with isobutylchloroformate
at ꢀ20 ꢁC, in the same solvent.20 An excess of N-protected
amino acid (100 equiv.) was required. This coupling reaction
proceeds with excellent yields of ca 85%.
In the permanent quest for selective and efficient enantioselec-
tive porphyrinoid catalysts for the epoxidation of alkene, a very
large number of different chiral superstructures has already
been reported.1–3 Unfortunately, and although some remark-
able catalysts are known, systematic studies dealing with very
subtle structural variations and leading to complete series of
new catalysts are still lacking. So far, only rare illustrations
of these studies have been reported and obviously, what has
been realised for the salen type ligand4 remains to be underta-
ken in the context of the porphyrin core.5,6 Nevertheless, some
examples of detailed studies have recently been published.5–11
We have scrutinised the influence of different chiral straps in
D2- or C2-symmetrical superstructures. Although the enantio-
meric excesses (ee) were low (31% at best), we were able to reach
two conclusions.12 Firstly, no general structure-activity rela-
tionship was to be found between the steric hindrance and
observed enantioselectivity. Secondly, the chiral strap, if not
flexible enough or too close to the metal centre, has a negative
effect on the enantioselectivity. An alternative to these too-con-
gested chiral porphyrins could be illustrated by the analogous
catalysts in which the straps do not exist. In this work, the cor-
responding—with the same chiral inducer—picket porphyrins
have been studied in exactly the same conditions. Additionally,
to compare with already described results, two other chiral
amino acids have also been employed. To do so, we chose
two amino acids with an alcohol function that can be easily
protected with the tosyl group, thereby increasing the size of
the lateral chain of the amino acid.
Results and discussion
The chiral integrity of the N-Boc-L-proline was conserved
after the coupling reaction as shown by the 1H NMR spectrum
of the porphyrin in which the Boc group was removed with
CF3CO2H–CH2Cl2 (1:10) at room temperature.21 The result-
ing porphyrin exhibits a sharp and well-resolved 1H NMR
Synthesis and characterisation of the free ligands
The picket porphyrins described in this work were synthesised
as precursors of different strapped analogues that have been
1
studied in solution by H NMR spectroscopy and in the solid
942
New J. Chem., 2003, 27, 942–947
DOI: 10.1039/b212480g
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
Fig. 1 Synthetic pathway for the preparation of the four L-prolinoyl chiral atropisomers (M ¼ 2H or FeCl).
spectrum.y Particularly noteworthy are the protons bound to
the b-pyrrolic and chiral carbons, which appear as 2 singlets
(4H) at 8.76 and 8.83 ppm and a triplet (3J ¼ 6.8 Hz) at
3.16 ppm, respectively. Such an observation was not possible
with picket porphyrins having Boc group, because of broad
signals most certainly resulting from the rotation of the bulky
Boc residue, as reported for the L-alanine22 and L-phenylala-
nine analogues.14,23
Different observations must be underlined:
–For both olefins, it is striking that the aabb isomer leads to
the lowest ee. Surprisingly, it has been shown that this atropi-
somer, bearing binaphthyl residues, exhibits very high enan-
tioselectivies.17
Asymmetric epoxidation of olefins
The four L-prolinoyl iron(III) chloro derivatives 1–4 were eval-
uated for the epoxidation of two quite different olefins: p-
chlorostyrene and 1,2-dihydronaphthalene. The former is a
terminal olefin, the latter disubstituted and more encumbered.
It has already been shown that the enantioselectivities obtained
with these two olefins can be significantly different. Addition-
ally, in the case of the abab atropisomer, two other chiral
amino acids were employed: O-p-toluenesulfonyl-L-trans-4-
hydroxyproline (TsO-L-HOPro) and O-p-toluenesulfonyl-L-
tyrosine (TsO-L-Tyr), which are structural analogues of the
native amino acids but more sterically hindered (Fig. 2).
Iodosylbenzene was used as the single oxygen donor and a
1000:100:1 ratio was maintained for the substrate, oxidant
and catalyst, respectively. These were dissolved in degassed
CH2Cl2 in a Schlenk tube. The mixture was stirred at room
temperature for 30 min and then quenched with 2% PPh3 in
CH2Cl2 to consume the excess iodosylbenzene. Enantiomeric
excesses were determined by GC using a Chirasil-Dex CB col-
umn. The results are listed in Table 1. For comparison pur-
poses, the ee obtained with two strapped porphyrins in the
abab and aabb geometries are also reported in this table
(entries 7 and 8).12
y In the L-prolinoyl series, and in opposition with the L-alanine and
L-phenylalanine cases, the porphyrins bearing the deprotected amino
acid, in our hands, were stable enough to allow NMR data to be
recorded but not enough to allow further reactions after their purifi-
cation.
Fig. 2 Different picket and strapped chiral porphyrins (M ¼ 2H or
FeCl) built from L-proline, L-trans-4-O-tosylhydroxyproline or L-O-
tosyltyrosine.
New J. Chem., 2003, 27, 942–947
943

View Article Online
Table 1 Enantiomeric excesses (%) obtained with different L-amino
acyl porphyrins for the epoxidation of p-chlorostyrene and 1,2-di-
hydronaphthalene
Conclusion
With a common chiral moiety, we have investigated the enan-
tioselectivities of the four possible atropisomers of a chiral
picket porphyrin. The results are compared with those
obtained for two strapped porphyrin geometries, prepared
with the same chiral motif, evidently. Three clear conclusions
can be drawn. (1) For the picket structures, the abab confor-
mation is the best one. (2) In the case of strapped porphyrins,
the aabb geometry is confirmed to lead to the best enantio-
selectivity. (3) Independently of the nature of the olefin, the
correct balance between steric hindrance and chiral induction
cannot be achieved with this type of porphyrin. New strategies
targeting more versatile and efficient catalysts are under inves-
tigation in our laboratories.
Compound
p-Chlorostyrene
1,2-Dihydronaphthalene
1Fe
2Fe
16.5
4
34
9
3Fe
10
22
4
14.5
10
28
7
4Fea
5Fe
6Fe
1DiisoFeb
2DiisoFeb
1
5
28
19
14
a
b
With 1-tert-butyl-5-phenylimidazole. See ref. 12.
Experimental
–The ee with the abab atropisomer reaches 34% in the case
of the disubstituted olefin, a better enantioselectivity than the
one reported for the analogous strapped porphyrin 1DiisoFe,
for which only 19% was obtained.
–In the aaaa geometry, although the enantiogenic face is
sterically crowded, and provided a very bulky axial base is
employed (Table 2), the best ee among the four atropisomers
is obtained for the terminal olefin. Incidentally, it can be noted
that the effect on the enantioselectivity of 1-tBu-5-phenylimi-
dazole24 is significant in comparison with that of 4,5-diphenyl-
imidazole.
–The substitution of L-Pro with either TsO-L-HOPro or
TsO-L-Tyr does not improve the enantioselectivity. This con-
firms that, even in a flexible structure, too much hindrance
implies a negative consequence on the selectivity of the epoxi-
dation reaction.
The choice of a chiral inducer with a cyclic structure, namely
L-proline, induces a significant increase in the ee in comparison
with other motifs such as phenylalanine or related structures.25
In the latter, although the ee does not reach 10%, an inverted
structure-activity relation is found, according to the number of
pickets. Additionally, an increase in the steric hindrance at the
periphery of the cycle is totally vain in the case of the terminal
olefin and has no effect for the disubstituted one, as shown
with catalyst 5Fe (Table 1).
The combined results obtained from both L-prolinoyl picket
and L-prolinoyl strapped porphyrins show that the abab-
strapped structures are not very selective, independently of
the nature of the olefin. In our case, the analogous picket cat-
alyst leads to much better results. In the aabb geometry, the
opposite conclusion seems to be reached: the strapped struc-
tures are more selective than the picket moieties. This cannot
be explained by the possible mobility of the pickets, which
should be at least equivalent in the abab conformation.
Finally, in light of our results, the aaab conformation repre-
sents the most versatile catalyst with respect to the nature of
the olefin. Indeed, the porphyrin 3Fe is the only catalyst for
which the ee obtained for both olefins are quite similar. We
believe that, if the structural problem of the single picket on
one side of the macrocycle can be solved, this conformation
could deserve more investigation.
General considerations
´
The analytical facilities were provided by the Universite
de Bourgogne (C.S.M.) in Dijon. 1H (500.13 MHz) and 13C
(125.05 MHz). NMR spectra were recorded on a Bruker
Avance DRX 500 spectrometer and referenced to the residual
protonated solvents. Mass spectra were acquired on a MS/MS
´
ZABSpec TOF spectrometer at the Universite de Rennes 1
(C.R.M.P.O.). UV-vis spectra were recorded on a Varian Cary
1E spectrometer. IR spectra were recorded on a Bruker IFS 66
spectrometer. All solvents (ACS for analysis) were purchased
from Carlo Erba. THF was distilled from potassium metal.
CH2Cl2 was used as received. Triethylamine and N-methylpi-
peridine were distilled on CaH2 . The starting materials were
generally used as commercially available (Acros, Aldrich)
without any further purification. All reactions were performed
under an argon atmosphere and monitored by TLC (silica,
CH2Cl2–CH3OH). Column flash chromatographies were per-
formed on silica gel (Merck TLC-Kieselgel 60H, 15 mm).
Synthesis of free ligands
a-5,15:b-10,20-Tetrakis[2-(N-tert-butoxycarbonyl-L-prolinoyl-
amido)phenyl]porphyrin, 1. N-Boc-L-proline (19.46 g, 90.4
mmol, 100 equiv.) were dissolved in dry THF (120 mL) under
an argon atmosphere, at ꢀ20 ꢁC. N-methylpiperidine (16.5
mL, 135 mmol, 150 equiv.) and then isobutylchloroformate
(11.15 mL, 85.9 mmol, 95 equiv.) were added. Immediately,
a white precipitate appeared. Then a solution of 610 mg of
abab-TAPP (0.9 mmol) in 50 mL of THF, maintained at
ꢀ20 ꢁC, was added. The reaction mixture was stirred 3 h at this
temperature and then was warmed to room temperature. The
mixture was filtered and the precipitate washed with diethyl
ether. The solution was evaporated under vacuum and the
residue was chromatographed on silica gel (elution with
a CH2Cl2–MeOH (98:2)). The pure product was obtained in
1
85% yield (1.15 g). H NMR 500 MHz (d ppm, CDCl3 , 300
K): ꢀ2.58 (s. 2H, NHpyr), 0.35 (m, 12H, Pro), 0.95 (s, 36H,
Boc), 1.28 (s, 8H, Pro), 1.53 (s, 4H, Pro), 3.56 (s, 4H, Pro*),
7.57 (s, 8H, aro), 7.88 (s, 4H, aro), 8.05 (s, 4H, NHCO), 8.69
(s, 4H, aro), 8.75 (s, 4H, b-pyr), 8.78 (s, 4H, b-pyr). 13C
NMR 125 MHz (d ppm, CDCl3 , 300 K): 28.2, 30.7, 45.3,
46.4, 61.7, 79.5, 80.1, 115.3, 121.4, 122.3, 123.6, 130.4, 131.9,
132.0, 134.8, 138.7, 153.2, 154.6, 170.7, 171.3. UV-vis
[CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1 cmꢀ1)]: 419 (357.2), 513 (22.3),
546 (5.2), 588 (6.3), 645 (1.9). MS (FAB): m/z ¼ 1463.0
Table 2 Enantiomeric excesses (%) obtained with 4Fe and different
axial bases for the epoxidation of p-chlorostyrene and 1,2-dihydro-
naphthalene
ꢃ
[M]+ . IR (KBr, u/cmꢀ1): 3380 (NH), 1698 (CO).
Axial base
p-Chlorostyrene 1,2-Dihydronaphthalene
a-5,10:b-15,20-Tetrakis[2-(N-tert-butoxycarbonyl-L-prolinoyl-
amido)phenyl]porphyrin, 2. 2 was synthesised according to the
above-mentioned procedure, fully described for compound 1.
The pure compound was obtained in 89% yield (1.18 g). 1H
Pyridine
4,5-Diphenylimidazole
5
6
8
8
1-tBu-5-phenylimidazole 22
10
944
New J. Chem., 2003, 27, 942–947

View Article Online
NMR 500 MHz (d ppm, CDCl3 , 300 K): ꢀ2.68 (s, 2H, NHpyr),
ꢀ1.96 (s, 1H, Pro), ꢀ1.63 (m, 1H, Pro), ꢀ0.99 (m, 1H, Pro),
ꢀ0.44 (m, 2H, Pro), ꢀ0.12 (m, 4H, Pro), 0.63 (m, 4H, Pro),
0.89 (m, 6H, Pro), 1.17 (m, 36H, Boc), 1.61 (s, 3H, Pro),
2.66 (m, 2H, Pro), 3.49 (m, 4H, Pro*), 7.18 (s, 2H), 7.64 (d,
Jo ¼ 7.0 Hz, 2H), 7.90 (m, 4H), 8.00 (d, Jo ¼ 7.5 Hz, 4H),
8.10 (s, 2H), 8.16 (m, 4H), 8.61 (d, Jo ¼ 6.5 Hz, 2H), 8.71
(s, 2H), 8.80 (s, 4H, NHCO), 8.87 (s, 2H). 13C NMR 125
MHz (d ppm, CDCl3 , 300 K): 28.7, 28.9, 30.1, 43.9, 46.4,
59.5, 61.7 114.5, 117.7, 120.6, 123.7, 124.6, 125.0, 127.3,
127.5, 130.2, 130.6, 131.8, 133.2, 134.0, 136.09, 139.1, 145.9,
(s, 2H, Pro), 3.45 (m, 4H, Pro), 4.72 (s, 2H, Pro), 7.17 (s,
8H, aro + aroTs), 7.41 (s, 4H, aroTs), 7.51 (s, 8H, aro), 7.87
(s, 8H, aro), 8.39 (s, 4H, NHCO), 8.63 (s, 4H, aro), 8.75 (s,
8H, b-pyr). 13C NMR 125 MHz (d ppm, CDCl3 , 300 K):
21.9, 27.4, 28.1, 34.2, 37.2, 52.1, 58.7, 59.9, 79.0, 80.6, 81.0,
115.1, 122.4, 124.1, 127.9, 130.3, 132.5, 133.6, 135.7, 138.4,
145.4, 152.9, 154.5, 169.3, 170.3. UV-vis [CH2Cl2 , l/nm
(10ꢀ3ꢂe/Mꢀ1 cmꢀ1)]: 421 (392.4), 515 (20.0), 547 (4.9), 589
(6.5), 647 (1.8). MS (FAB): m/z ¼ 2145.3 [M + H]+. IR
(KBr, u/cmꢀ1): 3475 (NH), 1699 (CO).
152.9, 154.3, 170.5, 171.5. UV-vis [CH2Cl2 , l/nm (10ꢀ3
ꢂ
a-5,15:b-10,20-Tetrakis[2-(N-tert-butoxycarbonyl-O-para-toluene-
sulfonyl-L-tyrosinoylamido)phenyl]porphyrin, 6. 6 was synthe-
sised according to the above-mentioned procedure, fully
described for compound 1, but with N-tert-butoxycarbonyl-
O-para-toluenesulfonyl-L-tyrosine (8) as the amino acid. The
pure compound was obtained in 29% yield (0.11 g). 1H
NMR 500 MHz (d ppm, CDCl3 , 300 K): ꢀ2.50 (s, 2H, NHpyr),
0.82 (m, 36H, Boc), 2.32 (m, 4H, CH2Tyr), 2.37 (m, 12H,
CH3Ts), 2.46 (m, 4H, CH2Tyr), 3.67 (m, 4H, H*), 4.70 (s,
4H, NHTyr), 5.76 (m, 4H, aro), 6.12 (s, 4H, aro), 6.25 (m,
4H, aro), 7.13 (m, 12H, aro), 7.26 (m, 4H, aro), 7.31 (m, 4H,
aro), 7.42 (m, 4H, aro), 7.58 (m, 4H, aro), 7.69 (m, 4H, b-
pyr), 7.83 (m, 4H, aro), 8.65 (m, 4H, aro), 8.69 (m, 8H,
NHCO + b-pyr). 13C NMR 125 MHz (d ppm, CDCl3 , 300
K): 22.0, 28.0, 28.2, 28.6, 37.9, 51.2, 55.7, 79.9, 122.0, 122.3,
124.0, 128.6, 128.9, 130.0, 130.2, 132.1, 135.2, 136.1, 137.8,
145.6, 154.8, 169.9. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 422 (254.9), 516 (14.8), 549 (4.7), 591 (4.7), 649 (2.0).
e/Mꢀ1 cmꢀ1)]: 419 (308.2), 513 (16.8), 546 (3.9), 587 (4.9),
642 (1.7). HR-MS (LSI-MS): calcd. m/z ¼ 1485.7012
[M + Na]+ for C84H94N12NaO12 , found 1485.6994. IR (KBr,
u/cmꢀ1): 3383 (NH), 1698 (CO).
a-5,10,15:b-20-Tetrakis[2-(N-tert-butoxycarbonyl-L-prolinoyl-
amido)phenyl]porphyrin, 3. 3 was synthesised according to the
above-mentioned procedure, fully described for compound 1.
The pure compound was obtained in 90% yield (1.19 g). 1H
NMR 500 MHz (d ppm, CDCl3 , 300 K): ꢀ2.61 (s, 2H, NHpyr),
ꢀ0.88 (s, 2H, Pro), ꢀ0.75 (s, 2H, Pro), 0.55 (s, 10H, Pro), 1.13
(s, 6H, Pro), 1.30 (s, 27H, Boc), 1.45 (s, 9H, Boc), 1.58 (s, 4H,
Pro), 3.53 (s, 4H, Pro*), 7.64 (s, 4H), 7.89 (s, 4H), 8.01 (s, 4H),
8.31 (s, 2H), 8.78 (m, 10H). 13C NMR 125 MHz (d ppm,
CDCl3 , 300 K): 26.3, 28.0, 28.9, 30.1, 30.2, 40.9, 43.8, 45.8,
61.7, 79.8, 112.5, 114.4, 121.8, 123.7, 125.0, 130.4, 131.8,
133.6, 134.7, 138.9, 153.7, 154.5, 169.3, 170.5. UV-vis
[CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1 cmꢀ1)]: 419 (372.7), 513 (19.8),
547 (5.1), 588 (6.0), 644 (2.2). HR-MS (LSI-MS): calcd.
m/z ¼ 1485.7012 [M + Na]+ for C84H94N12NaO12 , found
1485.7015. IR (KBr, u/cmꢀ1): 3382 (NH), 1698 (CO).
ꢃ
MS (EI): m/z ¼ 2343.0 [M]+ . IR (KBr, u/cmꢀ1): 3414
(NH), 1694 (CO).
N-tert-Butoxycarbonyl-L-tyrosine, 7. In a 500 mL round-bot-
tom flask equipped with a stir bar, an addition funnel, a ther-
mometer and a reflux condenser, in an ice bath, 6.1 g (152.5
mmol) of NaOH were dissolved in 150 mL of water. L-Tyro-
sine (25.12 g, 138.6 mmol) in suspension in 100 mL of tert-
butanol are added. Di-tert-butyl dicarbonate (30.26 g, 138.6
mmol) in solution in 50 mL of tert-butanol was carefully added
dropwise (with control of temperature). After the addition
period, the mixture was allowed to react overnight at room
temperature. The two layers were separated in a separatory
funnel and the aqueous layer washed twice with 35 mL of pent-
ane. The resulting organic layer was extracted three times with
10 mL of a saturated solution of sodium bicarbonate. The aqu-
eous layer was cooled down in an ice bath and acidified with 2
M HCl (around 110 mL) until pH 2 was reached. The resulting
precipitate was dissolved in 100 ml of ethyl acetate, washed
twice with 25 mL of water and dried over magnesium sulfate.
Evaporation of the solvents yielded a yellow oil, from which
white crystals were obtained by addition of hexane, overnight
a-5,10,15,20-Tetrakis[2-(N-tert-butoxycarbonyl-L-prolinoyl-
amido)phenyl]porphyrin, 4. 4 was synthesised according to the
above-mentioned procedure, fully described for compound 1.
The pure compound was obtained in 79% yield (1.05 g). 1H
NMR 500 MHz (d ppm, CDCl3 , 300 K): ꢀ2.63 (s, 2H, NHpyr),
0.47 (m, 2H, Pro), 0.98 (m, 9H, Pro), 1.49 (m, 9H, Pro), 1.53
(m, 36H, Boc), 3.01 (m, 2H, Pro), 3.41 (m, 2H, Pro), 3.80
(m, 4H, Pro*), 7.44 (m, 2H), 7.54 (m, 4H), 7.64 (m, 2H),
7.84 (m, 2H), 7.89 (t, Jo ¼ 7.5 Hz, 4H), 8.04 (s, 1H), 8.10
(m, 1H), 8.21 (m, 1H), 8.25 (d, Jo ¼ 7.5 Hz, 2H), 8.64 (t,
Jo ¼ 7.5 Hz, 2H), 8.73 (m, 2H), 8.77 (d, Jo ¼ 4.5 Hz, 2H,
b-H), 8.79 (d, Jo ¼ 4.5 Hz, 2H, b-H), 8.83 (m, 2H), 8.87 (m,
4H), 9.01 (d, Jo ¼ 4.0 Hz, 2H, b-H), 9.05 (d, Jo ¼ 4.0 Hz,
2H, b-H), 9.12 (m, 1H), 9.22 (m, 1H). 13C NMR 125 MHz
(d ppm, CDCl3 , 300 K): 21.3, 21.5, 22.3, 23.2, 24.1, 28.9,
29.1, 29.7, 30.2, 30.7, 31.1, 43.9, 45.4, 46.7, 46.8, 59.4, 60.6,
61.6, 79.5, 79.9, 81.0, 81.9, 82.4, 114.8, 115.1, 117.7, 118.1,
120.7, 120.9, 124.1, 124.7, 125.0, 127.9, 128.4, 129.7, 130.2,
130.6, 131.9, 134.7, 135.0, 135.4, 135.6, 135.8, 136.1, 136.5,
137.6, 137.8, 138.2, 153.7, 154.2, 154.5, 170.1, 170.2, 171.1,
172.2. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1 cmꢀ1)]: 419
(443.1), 514 (20.1), 549 (5.6), 587 (6.3), 644 (4.4). HR-MS
1
in a refrigerator. Yield (29.25 g, 75%). H NMR 500 MHz (d
ppm, CDCl3 , 300 K): 1.42 (s, 9H, Boc), 3.04 (s, 2H, CH2),
4.55 (d, 1H, CH), 6.73 (d, 2H, aro), 6.97 (d, 2H, aro). 13C
NMR 125 MHz (d ppm, CDCl3 , 300 K): 28.7, 37.5, 54.9, 81.2,
116.1, 127.7, 130.5, 155.4, 156.3, 176.2. MS (FAB): m/z ¼ 281
ꢃ
(LSI-MS):
calcd.
m/z ¼ 1485.7012
[M + Na]+
for
[M]+ . Anal. calcd for C14H19NO5 (%): C, 59.78, H, 6.81, N,
C84H94N12NaO12 , found 1485.7024. IR(KBr, u/cmꢀ1): 3481
(NH), 1697 (CO).
4.98; found (%): C, 59.47, H, 7.13, N, 4.88. IR (KBr,
u/cmꢀ1): 3413 (OH), 1765 (CO), 1690 (CO).
a-5,15:b-10,20-Tetrakis[2-(N-tert-butoxycarbonyl-O-para-toluene-
sulfonyl-L-trans-4-hydroxyprolinoylamido)phenyl]porphyrin, 5.
5 was synthesised according to the above-mentioned proce-
dure, fully described for compound 1, but with N-tert-butoxy-
carbonyl-O-para-toluenesulfonyl-L-trans-4-hydroxyproline(10)
as the amino acid. The pure compound was obtained in 49%
N-tert-Butoxycarbonyl-O-para-toluenesulfonyl-L-tyrosine, 8.
In a 500 mL round-bottom flask equipped with a stir bar, an
addition funnel, a thermometer and a reflux condenser, in an
ice bath, 1.82 g (45.5 mmol) of NaOH were dissolved in 50
mL of water. 7 (11.64 g, 41.4 mmol) in solution in 50 mL of
diethyl ether was added. One added dropwise 7.89 g (41.4
mmol) of tosyl chloride in solution in 50 mL of diethyl ether.
After the addition period, the mixture was allowed to react over-
night at room temperature. After separation of the two layers,
the aqueous layer was washed twice with 35 mL of pentane.
1
yield (0.95 g). H NMR 500 MHz 0(d ppm, CDCl3 , 300 K):
ꢀ2.56 (s, 2H, NHpyr), 0.13 (m, 18H, Boc), 0.43 (s, 8H, Pro),
0.84 (s, 8H, Boc), 0.91 (m, 10H, Boc), 1.41 (s, 2H, Pro), 1.65
(m, 4H, Pro), 2.33 (m, 12H, CH3Ts), 2.83 (s, 2H, Pro), 3.04
New J. Chem., 2003, 27, 942–947
945

View Article Online
The resulting organic layer was extracted three times with 10
mL of a saturated solution of sodium bicarbonate. The aqu-
eous layer was cooled down in an ice bath and acidified with
2 M HCl (around 110 mL) until pH 1 was reached. The result-
ing precipitate was dissolved in 50 ml of ethyl acetate, washed
twice with 25 mL of water and dried over magnesium sulfate.
Evaporation of the solvents yielded a white solid, which was
recrystallised from petroleum ether, overnight in a refrigerator.
Yield (15.32 g, 85%). 1H NMR 500 MHz (d ppm, CDCl3 , 300
K): 1.44 (s, 9H, Boc), 2.47 (s, 3H, CH3Ts), 3.06 (m, 1H, CH2),
3.18 (m, 1H, CH2), 4.58 (m, 1H, CH), 6.95 (d, 2H, Jo ¼ 8.5
Hz, aroTyr), 7.13 (d, 2H, Jo ¼ 8.5 Hz, aroTyr), 7.33 (d, 2H, Jo ¼
8.1 Hz, aroTs), 7.72 (d, 2H, Jo ¼ 8.1 Hz, aroTs). 13C NMR 125
MHz (d ppm, CDCl3 , 300 K): 22.1, 28.7, 37.6, 54.5, 81.0,
122.9, 128.9, 130.2, 131.0, 132.7, 135.4, 145.8, 149.1, 155.7,
175.5. MS (EI): m/z ¼ 435 [M]+. Anal. calcd for C21H25NO7S
(%): C, 57.92, H, 5.79, N, 3.22, S, 7.36; found (%): C, 57.82,
H, 5.74, N, 3.10, S, 6.64. IR (KBr, u/cmꢀ1): 1746 (CO),
1672 (CO).
cmꢀ1)]: 417 (102.1), 574 (6.1). MS (FAB): m/z ¼ 1514.3
[M ꢀ Cl ꢀ 2H]+. IR (KBr, u/cmꢀ1): 3387 (NH), 1699 (CO).
3Fe(Cl). Prepared from 3, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 418 (99.0), 575 (8.5). MS (FAB): m/z ¼ 1515.6
[M ꢀ Cl ꢀ H]+. IR (KBr, u/cmꢀ1): 3383 (NH), 1699 (CO).
4Fe(Cl). Prepared from 4, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 417 (67.8), 572 (6.5). MS (FAB): m/z ¼ 1513.6
[M ꢀ Cl ꢀ 3H]+. IR (KBr, u/cmꢀ1): 3411 (NH), 1697 (CO).
5Fe(Cl). Prepared from 5, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 417 (101.6), 576 (7.5). MS (FAB): m/z ¼ 2198.0
[M ꢀ Cl + H]+. Anal. calcd for C112H116ClFeN12O24S4ꢂH2O
(%): C, 59.74, H, 5.28, N, 7.46, S, 5.70; found (%): C, 59.64,
H, 5.38, N, 7.48, S, 5.31. IR (KBr, u/cmꢀ1): 3476 (NH),
1701 (CO).
N-tert-Butoxycarbonyl-L-trans-4-hydroxyproline, 9. 9 was
synthesised according to the above-mentioned procedure, fully
described for compound 7. The pure compound was obtained
in 79% yield (25.5 g). 1H NMR 500 MHz (d ppm, CDCl3 ,
300K): 1.40 (s, 9H, Boc), 3.46 (m, 1H, Pro), 3.56 (m, 3H,
Pro), 4.43 (m, 1H, Pro), 4.47 (m, 1H, Pro*), 6.98 (s, 1H,
OH). 13C NMR 125 MHz (d ppm, CDCl3 , 300 K): 14.5,
21.4, 28.1, 28.6, 28.7, 38.2, 39.2, 54.7, 54.9, 58.0, 58.3, 60.9,
67.4, 69.5, 70.1. MS (EI): m/z ¼ 232 [M + H]+. Anal. calcd
for C10H17NO5 (%): C, 51.94, H, 7.41, N, 6.06; found (%):
C, 51.41, H, 7.59, N, 6.15. IR (KBr, u/cmꢀ1): 3412 (OH),
1734 (CO), 1663 (CO).
6Fe(Cl). Prepared from 6, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 418 (68.0), 578 (4.3). MS (FAB): m/z ¼ 2395.8
[Mꢀ Clꢀ H]+. Anal. calcd for C128H124ClFeN12O24S4ꢂCH2Cl2-
(%): C, 61.51, H, 5.04, N, 6.67, S, 5.09; found (%): C, 61.92, H,
5.45, N, 6.68, S, 4.78. IR (KBr, u/cmꢀ1): 3417 (NH), 1692 (CO).
General procedure for asymmetric epoxidation
The catalyst (1 mmol) and the olefin (1000 mmol), as well as the
base (250 mmol) with 4Fe, were dissolved in degassed CH2Cl2
(2 mL) in a Schlenk tube. While stirring under N2 , PhIO (100
mmol) was added in one portion. The mixture was stirred at
room temperature for 30 min and then quenched with 2%
PPh3 in CH2Cl2 . The solvent was removed under reduced
pressure. Pentane was added and the mixture was filtered.
Dodecane (100 mL, GC standard) was added. Enantiomeric
excesses were determined by GC on a Chirasil-Dex CB capil-
larity column. The GC conditions were as follows: (epoxide,
oven temperature, retention time): p-chlorostyrene oxide,
120 ꢁC, 22.5 and 24.3 min; 1,2-dihydronaphthalene oxide,
130 ꢁC, 18.9 and 20.4 min.
N-tert-Butoxycarbonyl-O-para-toluenesulfonyl-L-trans-4-
hydroxyproline, 10. 10 was synthesised according to the above-
mentioned procedure, fully described for compound 8. The
1
pure compound was obtained in 81% yield (4.5 g). H NMR
500 MHz (d ppm, CDCl3 , 300 K): 1.27 (s, 9H, Boc), 2.29 (s,
3H, CH3Ts), 3.43 (m, 4H, Pro), 4.31 (m, 1H, Pro), 4.89 (m,
1H, Pro*), 7.22 (d, Jo ¼ 8.1 Hz, 2H, aroTs), 7.62 (d,
Jo ¼ 8.1 Hz, 2H, aroTs). 13C NMR 125 MHz (d ppm, CDCl3 ,
300 K): 22.1, 35.8, 39.3, 52.3, 55.0, 82.4, 128.2, 130.5, 145.8,
156.4, 178.3. MS (EI): m/z ¼ 386 [M]+. Anal. calcd for
C17H23NO7S (%): C, 52.97, H, 6.01, N, 3.63, S, 8.32; found
(%): C, 53.05, H, 6.08, N, 3.94, S, 7.93. IR (KBr, u/cmꢀ1):
1754 (CO), 1636 (CO).
Acknowledgements
General procedure for iron insertion
This work was supported by the CNRS and the MENRT,
especially for the grant to V. B.-C.
Incorporation of iron in the free-base porphyrins was accom-
plished by the iron(II) bromide method under a controlled
atmosphere. The free-base ligand was dissolved in THF inside
a glove box maintained under 1 ppm of dioxygen. 2,6-Lutidine
and an excess of FeBr2 were added to the mixture. The solu-
tion was heated at 55 ꢁC for 12 h until the reaction was com-
pleted as indicated by UV-vis spectroscopy. Temperature
was carefully monitored so that no atropisomerisation was
to occur. The product was oxidised by air for 1 h. The solvent
was removed under vacuum and the residue dissolved in
CH2Cl2 and washed with brine. The product was chromato-
graphed on silica gel.
References
1
2
J. P. Collman, X. M. Zhang, V. J. Lee, E. S. Uffelman and J. I.
Brauman, Science, 1993, 261, 1404.
E. Rose, M. Quelquejeu, R. P. Pandian, A. Lecas-Nawrocka, A.
Vilar, G. Ricart, J. P. Collman, Z. Wang and A. Straumanis,
Polyhedron, 2000, 19, 581.
3
4
5
E. Rose, A. Lecas, M. Quelquejeu, A. Kossanyi and B. Boitrel,
Coord. Chem. Rev., 1998, 180, 1407.
E. N. Jacobsen, in Catalytic Asymmetric Synthesis, ed. I. Ojima,
VCH Publishers, New York, 1993, pp. 159–202.
E. N. Jacobsen and M. H. Wu, in Comprehensive Asymmetric Cat-
alysis I–III, eds. E. N. Jacobsen, A. Pfaltz and H. Yamamoto,
Springer-Verlag, Berlin and Heidelberg, 1999, vol. II, pp. 646.
1Fe(Cl). Prepared from 1, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
cmꢀ1)]: 417 (66.1), 572 (3.7). HR-MS (LSI-MS): calcd. m/z ¼
1116.4210 [M ꢀ 4Boc ꢀ Cl + 4H]+ for C64H60FeN12O4 , found
1116.4212. IR (KBr, u/cmꢀ1): 3382 (NH), 1701 (CO).
´
6
C. Perollier, M. Mazzanti, J.-P. Simonato, F. Launay, R.
Ramasseul and J.-C. Marchon, Eur. J. Org. Chem., 2000, 583.
Z. Gross and S. Ini, J. Org. Chem., 1997, 62, 5514.
7
8
´
C. Perollier, J. Pecaut, R. Ramasseul and J. -C. Marchon, Inorg.
Chem., 1999, 38, 3758.
´
2Fe(Cl). Prepared from 2, according to the above-mentioned
iron insertion method. UV-vis [CH2Cl2 , l/nm (10ꢀ3ꢂe/Mꢀ1
9
G. Reginato, L. D. Bari, P. Salvadori and R. Guilard, Eur. J. Org.
Chem., 2000, 1165.
946
New J. Chem., 2003, 27, 942–947

View Article Online
ˆ
10 B. Boitrel, V. Baveux-Chambenoıt and P. Richard, Eur. J. Org.
amino groups of the isomer abab are alternatively on one side
and on the other one.
Chem., 2001, 4213.
11 R. Zhang, W.-Y. Yu, H.-Z. Sun, W.-S. Liu and C.-M. Che,
Chem.-Eur. J., 2002, 8, 2495.
12 B. Boitrel, V. Baveux-Chambenoˆıt and P. Richard, Eur. J. Inorg.
Chem, 2002, 1666.
13 E. Rose, B. Boitrel, M. Quelquejeu and A. Kossanyi, Tetrahedron
Lett., 1993, 34, 7267.
14 J. -P. Renaud, P. Battioni and D. Mansuy, Nouv. J. Chim., 1987,
11, 279.
15 C. Z. Wang, Z. A. Zhu, Y. Li, Y. T. Chen, X. Wen, F. M. Miao,
W. L. Chan and A. S. C. Chan, New J. Chem., 2001, 25, 801.
16 J. P. Collman, R. R. Gagne, C. A. Reed, T. R. Halbert, G. Lang
and W. T. Robinson, J. Am. Chem. Soc., 1975, 97, 1427–1439.
The rotation of the aminophenyl group leads to four atropi-
somers, stable in solution at room temperature. They have been
named aaaa for the isomer with the four amino groups on the
same side of the porphyrin, aaab for the isomer with three amino
groups on one side and the fourth one on the other side, aabb
when the porphyrin bears two contiguous amino groups on one
face with the two other groups on the other face. Finally, the
17 J. P. Collman, Z. Wang, A. Straumanis, M. Quelquejeu and
E. Rose, J. Am. Chem. Soc., 1999, 121, 460.
18 J. Lindsey, J. Org. Chem., 1980, 45, 5215.
19 A. Lecas, Z. Renko and E. Rose, Tetrahedron Lett., 1985, 26,
1019.
20 B. Boitrel, A. Lecas, Z. Renko and E. Rose, New J. Chem., 1989,
13, 73.
21 B. Boitrel, A. Lecas and E. Rose, J. Chem. Soc., Chem. Commun.,
1989, 349.
22 B. Boitrel, E. Camilleri, Y. Fleche, A. Lecas and E. Rose, Tetra-
hedron Lett., 1989, 30, 2923.
´
23 J. -P. Renaud, Ph.D. Thesis, Universite de Paris VI,Paris, France,
1986,.
24 A. M. van Leusen, F. J. Scharrt and D. van Leusen, Recl.: J. Roy.
Neth. Chem. Soc., 1979, 98, 258.
25 E. Rose, M. Soleilhavoup, L. Christ-Tommasino, G. Moreau,
J. P. Collman, M. Quelquejeu and A. Straumanis, J. Org. Chem.,
1998, 63, 2042.
New J. Chem., 2003, 27, 942–947
947