Iron corroles and porphyrins as very efficient and highly selective catalysts for the reactions of α-diazo esters with amines

Article abstract of DOI:10.1055/s-2006-939035

Iron corroles and porphyrins catalyze the reactions of amines with ethyl diazoacetate extremely efficiently, leading to complete and rapid conversion into N-substituted glycine ethyl esters by simultaneous addition of the substrates to the catalysts. The selectivity toward activation of the NH bonds is remarkable and quite different from other catalysts. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-939035

LETTER
951
Iron Corroles and Porphyrins as Very Efficient and Highly Selective Catalysts
for the Reactions of a-Diazo Esters with Amines
Iron
Cor
r
roles and
i
Porph
s
yrins Aviv, Zeev Gross*
Department of Chemistry and Institute of Catalysis Science and Technology, Technion – Israel Institute of Technology, Haifa 32000, Israel
Fax +11(972)48295703; E-mail: chr10zg@techunix.technion.ac.il
Received 22 January 2006
efficiency. Complete and rapid conversion into N-sub-
Abstract: Iron corroles and porphyrins catalyze the reactions of
stituted glycine ethyl esters was obtained by adding EDA
and the substrate to the catalyst simultaneously. The
selectivity of the iron corroles toward NH insertion is
amines with ethyl diazoacetate extremely efficiently, leading to
complete and rapid conversion into N-substituted glycine ethyl es-
ters by simultaneous addition of the substrates to the catalysts. The
selectivity toward activation of the NH bonds is remarkable and remarkable and shared only with iron porphyrins.
quite different from other catalysts.
Ar
L
Ar
Ar
Key words: diazo compounds, iron, corroles, porphyrins, catalysis
L
L
N
N
N
N
N
N
N
N
N
N
N
N
Ar Ar
Ar
Ar
Rh
Fe
L'
Fe
Metal-catalyzed transfer of carbene moieties from diazo
compounds to organic substrates, generally considered to
proceed via rate-limiting formation of metallocarbenoid
intermediates, is most frequently applied to additions to
double bonds.1 The same synthetic methodology is also
useful for the related insertion reactions into non-polar C–
H and polar X–H bonds (X = R2N, RO, RS) and this sub-
ject is of significant current interest.1,2 One particularly
important application is the synthesis of protected a-ami-
no acid derivatives from a-diazo esters and amines
(Scheme 1).
Ar = C6F5
Ar
Ar
Ar
2: L = PPh3
3a: L = Cl, Ar = C6F5
3b: L = Cl, Ar = C6H5
1a: L = Cl
1b: L = L' = OEt2
Figure 1
The current investigations started by comparing rhodium
acetate and the corrole complexes 1a and 2 as catalysts for
the following addition and insertion reactions of EDA
(Scheme 2): (a) NH insertion, (b) OH insertion, (c) C=C
addition, and (d) competitive C=C addition and CH in-
sertion.
R1
CO2R4
R3
R1
N2
catalyst
– N2
N
N H
+
R3
CO2R4
R2
R2
H
N
CO2Et
Scheme 1
CO2Et
Cl
Cl
N
+
CO2Et
4a
4b
NH2
The first reported catalysts for such NH insertion
reactions were copper bronze and CuCN,3,4 which were
overshadowed by the 1970s work of Paulissen et al. with
Rh2(OAc)4 as catalyst.5 All of the recently introduced
copper-, silver-, and ruthenium-based catalysts suffer
from one or more of the following limitations: restricted
reaction conditions (dropwise addition of the diazo com-
pound), long reaction times, and low to moderate chemi-
cal yields.2,6,7
a
Cl
N2
Ph
OH
OEt
O
CO2Et
c
H
C
O
b
Ph
CO2Et
8
5
Recent research regarding the potential of metallocorroles
in catalysis revealed that rhodium(III) complexes are
particularly effective for cyclopropanation while iron
complexes are versatile catalysts for epoxidation, cyclo-
propanation, and aziridination of C=C bonds, as well as
C–H hydroxylation.8,9 We found that the easily prepared
iron corroles 1a and 1b (Figure 1)10 catalyze the NH inser-
tion of ethyl diazoacetate (EDA) into amines with high
d
c
6
CO2Et
EtO2C
7
Scheme 2
The results obtained with the different catalysts under oth-
erwise identical reaction conditions revealed remarkable
differences (Table 1). Rhodium acetate is apparently an
excellent catalyst for the reactions of olefins and an alco-
SYNLETT 2006, No. 6, pp 0951–0953
0
4.
0
4.
2
0
0
6
Advanced online publication: 14.03.2006
DOI: 10.1055/s-2006-939035; Art ID: G02406ST
© Georg Thieme Verlag Stuttgart · New York

952
I. Aviv, Z. Gross
LETTER
hol, but not for NH insertion. The reactions of EDA with affected by the large excess of amine, in fact, their selec-
ethanol, styrene, and cyclohexene were complete within tivity towards NH activation remained absolute and reac-
minutes, while much of the EDA was not consumed even tions were complete within one minute.
after 26 hours with 4-chloroaniline. A similar trend was
O
revealed for the rhodium corrole 2, except that it is a less
H
Cl
NH2
N
active catalyst in general. Iron corrole 1a displayed very
different catalytic properties. EDA was fully consumed
within one hour regardless of the substrate, but the selec-
tivity varied significantly. The desired products from the
reactions of cyclohexene, ethanol, and styrene were
formed in yields of 0%, 30%, and 91%, respectively, ac-
companied by 100%, 70%, and 9% of EDA coupling
products (diethyl maleate and fumarate). The results ob-
tained with 4-chloroaniline show that 1a is an excellent
NH insertion catalyst; the reaction was complete within
one minute, with 100% selectivity for 4a. Importantly,
practically identical results were obtained with the other
Cl
OEt
+
catalyst
+
EtO2C
Ph EtO2C
+
Ph
+
EDA
syn
anti
Scheme 3
Table 2 Intermolecular Competition between NH and C=C Activa-
tiona
iron corrole 1b and the iron porphyrins 3a and 3b, but the Catalyst
Insertion/
cyclopropanation
Reaction time
iron–salen complex was completely unreactive.11
Fe(tpfc)Cl (1a)
100:0
100:0
100:0
23:77
91:9
1 min, complete
1 min, complete
1 min, complete
24 h, incomplete
24 h, incomplete
Table 1 Reaction Times for the Full Consumption of EDA in the
Reactions Described in Scheme 2a
Fe(tpfpp)Cl (3a)
Fe(tpp)Cl (3b)
Rh(tpfc)PPh3 (2)
Rh2(OAc)4
Catalyst
Reaction time, selectivity
4a
5
8
6 and 7
Fe(tpfc)Cl (1a)
Rh(tpfc)PPh3 (2)
Rh2(OAc)4
1 min,
100%
1 h,
30%
1 h,
91%
1 h,
0%
a Catalyst/EDA/substrate A/substrate B, 1:50:500:500 with 0.4–0.5
mM catalyst at r.t. in CH2Cl2. EDA and the substrates were added to-
gether in one portion. The relative yields of insertion and cyclopropa-
nation products were determined by GC analysis.
26 h,
10 h,
96%
15 min, 30 min,
100%
100%b
68% 6, 7% 7
26 h,
5 min,
100%
5 min,
100%
5 min,
97% 6, 3% 7
100%b
a Catalyst/EDA/substrate, 1:100:1000 with 0.4–0.5 mM of catalyst at
r.t. in CH2Cl2. Reactions were allowed to proceed until TLC indicated
the full consumption of EDA or for 26 hours. The undesired byprod-
ucts formed were maleate and fumarate esters.
To determine the synthetic utility of the reactions, several
ring-substituted anilines were reacted with EDA in the
presence of 0.1 mol% catalyst at >0.5 M concentration of
reagents. The results shown in Table 3 report isolated
yields of the corresponding glycine product obtained by
simultaneous addition of equimolar amounts of amine and
EDA to the pre-dissolved catalyst. Intense emission of ni-
trogen gas started almost instantaneously, reactions were
complete in a few minutes except for the least reactive
amine (4-cyanoaniline), and turnover numbers up to 900
were obtained in all cases. Diethyl ether was found to be
a particularly practical solvent; both the reactants and the
catalyst are very soluble, while in some cases the pure
product precipitated.12 This highly facile catalytic NH ac-
tivation by 1a is also shared by the analogous iron porphy-
rin 3a (Figure 1), but in that case the desired product 4a
was contaminated by 4b. This is reminiscent of results ob-
tained with copper-based catalysts,2d emphasizing the ad-
vantage of iron corroles in being more selective to single
NH insertion. Nevertheless, both iron complexes were
found to catalyze the transformation of isolated 4a into 4b
via the reaction of the former with EDA. This was not the
only example of secondary amines:14 morpholine was also
reactive under the same reaction conditions (reaction time
of 1.5 hours), with the corresponding product being ob-
tained in 92% yield with 0.1% mol 3a as catalyst.
b Complete consumption of EDA was not attained.
The surprising differences in catalyst selectivity were fur-
ther investigated by carrying out competition reactions
between styrene and 4-chloroaniline (Scheme 3), with
limited amounts of EDA. The advantage of this investiga-
tion is that it takes into account that coordination of the
amine to the metal center could be a major factor in affect-
ing reactivity and/or selectivity, which was indeed proven
to be the case for both rhodium complexes (Table 2). One
significant result is that EDA was not fully consumed
even after 24 hours when both the olefin and the amine
were present (Table 2). As rhodium-catalyzed cyclopro-
panation of styrene required only 5–15 minutes in the ab-
sence of 4-chloroaniline (Table 1) this clearly shows that
the amine reduces the reactivity of the system. The selec-
tivity of NH vs. C=C activation under the conditions de-
scribed in Table 2 were about 1:3 and 10:1 for 2 and
Rh2(OAc)4, respectively. However, both catalysts are of
no practical utility under these conditions, as the reactions
were not complete even after 24 hours. On the other hand,
the reactivity of the iron complexes 1a, 3a, and 3b was not
Synlett 2006, No. 6, 951–953 © Thieme Stuttgart · New York

LETTER
Iron Corroles and Porphyrins
953
Table 3 NH-Activation of Ring-Substituted Anilines and Morpho- References and Notes
line by EDAa
(1) (a) Doyle, M. P.; McKervey, M. A.; Ye, T. In Modern
Catalytic Methods for Organic Synthesis with Diazo
Compounds; John Wiley & Sons: New York, 1998.
(b) Singh, G. S. Curr. Org. Synth. 2005, 2, 377.
Catalyst
Substrate
Reaction time Yield
(min)
Fe(tpfc)(OEt2)2 (1b)
Fe(tpfc)Cl (1a)
4-Cl-C6H4NH2
3-CN-C6H4NH2
4-CN-C6H4NH2
C6H4NH2
3
5
92%
94%
93%
93%
95%
90%
94%
95%
92%
(2) (a) Burtoloso, A. C. B.; Correia, C. R. D. J. Organomet.
Chem. 2005, 690, 5636. (b) Huang, H.; Wang, Y.; Chen, Z.;
Hu, W. Adv. Synth. Catal. 2005, 347, 531. (c) Davis, F. A.;
Wu, Y.; Xu, H.; Zhang, J. Org. Lett. 2004, 6, 4523.
(d) Fructos, M. R.; Belderrain, T. R.; Nicasio, M. C.; Nolan,
S. P.; Kaur, H.; Diaz-Requejo, M. M.; Perez, P. J. J. Am.
Chem. Soc. 2004, 126, 10846.
(3) Yates, P. J. Am. Chem. Soc. 1952, 74, 5376.
(4) (a) Saegusa, T.; Ito, Y.; Kobayashi, S.; Hirota, K.; Shimizu,
T. Tetrahedron Lett. 1966, 6131. (b) Nicoud, J.-F.; Kagan,
H. B. Tetrahedron Lett. 1971, 2065.
(5) Paulissen, R.; Hayez, E.; Hubert, A. J.; Teyssie, P.
Tetrahedron Lett. 1974, 15, 607.
(6) (a) Morilla, M. E.; Diaz-Requejo, M. M.; Belderrain, T. R.;
Nicasio, M. C.; Trofimenko, S.; Perez, P. J. Chem. Commun.
2002, 2998. (b) Bachmann, S.; Fielenbach, D.; Jorgensen,
K. A. Org. Biomol. Chem. 2004, 2, 3044.
(7) (a) Galardon, E.; Le Maux, P.; Simonneaux, G. J. Chem.
Soc., Perkin Trans. 1 1997, 2455. (b) Galardon, E.; Le
Maux, P.; Simonneaux, G. Tetrahedron 2000, 56, 615.
(8) (a) Simkhovich, L.; Mahammed, A.; Goldberg, I.; Gross, Z.
Chem. Eur. J. 2001, 7, 1041. (b) Simkhovich, L.; Goldberg,
I.; Gross, Z. J. Porphyrins Phthalocyanines 2002, 6, 439.
(c) Saltsman, I.; Simkhovich, L.; Balazs, Y. S.; Goldberg, I.;
Gross, Z. Inorg. Chim. Acta 2004, 357, 3038.
(9) (a) Gross, Z.; Simkhovich, L.; Galili, N. Chem. Commun.
1999, 599. (b) Simkhovich, L.; Gross, Z. Tetrahedron Lett.
2001, 42, 8089.
(10) (a) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem. Int. Ed.
1999, 38, 1427. (b) Simkhovich, L.; Galili, N.; Saltsman, I.;
Goldberg, I.; Gross, Z. Inorg. Chem. 2000, 39, 2704.
(c) Simkhovich, L.; Goldberg, I.; Gross, Z. Inorg. Chem.
2002, 41, 5433. (d) Simkhovich, L.; Gross, Z. Inorg. Chem.
2004, 43, 6136.
Fe(tpfc)Cl (1a)
40
3
Fe(tpfc)(OEt2)2 (1b)
Fe(tpfc)(OEt2)2 (1b)
Fe(tpfc)(OEt2)2 (1b)
Fe(tpfpp)Cl (3a)
Fe(tpp)Cl (3b)
4-OMe-C6H4NH2
4-Me-C6H4NH2
4-Cl-C6H4NH2
4-Cl-C6H4NH2
morpholine
2
2
3
3
Fe(tpfpp)Cl (3a)
90
a Catalyst/EDA/substrate, 1:1000:1000, with 0.5–0.75 mM catalyst at
r.t. in Et2O. EDA and the substrate were added together in one portion
and the reported yields are of isolated products.13
This work demonstrates that iron corroles and porphyrins
catalyze the NH insertion of EDA into anilines very effi-
ciently, leading to their full conversion into N-substituted
glycine ethyl esters within minutes upon addition of the
substrates together in one portion to the catalyst at room
temperature. Only the copper-based catalysts developed
by Perez et al. come close to these results, but gradual ad-
dition of primary amines was still required and reaction
times were longer, despite the 4 mol% of catalyst em-
ployed.2d,6a The most outstanding features of the current
catalytic system are: a) very large selectivity toward NH
insertion, b) operation at 0.1 mol% catalyst loading, c)
very high yields with very short reaction times, and d)
non-poisoning by a very large excess and high concentra-
tion of amines (0.5–0.75 M). These remarkable differ-
ences relative to other catalysts suggest the operation of a
different reaction mechanism in the iron-based catalysts.
Detailed mechanistic aspects are out of the scope of this
report and will certainly require further investigation, but
several observations suggest that metallocarbenoid inter-
mediates might not be involved. The strongest indications
are the very fast emission of nitrogen gas only when
amines are used as substrates and that Fe(tpp)Cl (3b) is an
excellent NH insertion catalyst, although it does not cata-
lyze cyclopropanation of olefins by EDA.15 Future inves-
tigations will focus on mechanistic aspects, applications
to more complex systems, and attempts to solve the open
challenge of developing catalysts for asymmetric inter-
molecular NH insertion reactions.6b
(11) Bryliakov, K. P.; Talsi, E. P. Angew. Chem. Int. Ed. 2004,
43, 5228.
(12) Quantitative reactions of amines with EDA: To a solution
of catalyst (1 mg, 1–1.5 mmol) in argon-purged Et2O (1.5
mL), a solution of EDA and substrate (1–1.5 mmol each) in
Et2O (0.5 mL) was added at once. The 4-Cl-, 4-CN-, and
3-CN-substituted products precipitated from the reaction
mixture, while the products of other amines were obtained
by evaporating the solvent and treatment of the solid
material with cold Et2O. Reported yields are of isolated
products (0.2–0.3 g); their purity was checked by GC
analysis (> 99% for all the products) and their identity was
confirmed by NMR spectroscopy and comparison with
published data.2d,6a,13
(13) Gawinecki, R.; Kolehmainen, E.; Kucybala, Z.;
Osmialowski, B.; Kauppinen, R. Magn. Reson. Chem. 1998,
36, 848.
(14) All other amines that were examined so far (piperidine, N-
methylaniline, propylamine) reacted similarly and these and
many other results will soon be published.
(15) (a) Wolf, J. R.; Hamaker, C. G.; Djukic, J. P.; Kodadek, T.;
Woo, L. K. J. Am. Chem. Soc. 1995, 117, 9194. (b) Gross,
Z.; Galili, N.; Simkhovich, L. Tetrahedron Lett. 1999, 40,
1571.
Acknowledgment
This research was supported by the German Israel Cooperation
(DIP).
Synlett 2006, No. 6, 951–953 © Thieme Stuttgart · New York