Novel cobalt-catalyzed carbonylation of 2-aryl-2-oxazolines

Article abstract of DOI:10.1021/ol034380f

(Matrix presented) Catalytic ring-expanding carbonylation of 2-aryl-2-oxazolines is reported as a novel method for the synthesis of 4,5-dihydro-1,3-oxazin-6-ones. Various observations suggest the involvement of cobalt radicals as the catalytically active

Full text of DOI:10.1021/ol034380f

ORGANIC
LETTERS
2003
Vol. 5, No. 9
1575-1577
Novel Cobalt-Catalyzed Carbonylation of
2-Aryl-2-oxazolines
Hongyu Xu and Li Jia*
Department of Chemistry, Lehigh UniVersity, 6 E. Packer AVenue,
Bethlehem, PennsylVania 18015
lij4@lehigh.edu
Received March 4, 2003
ABSTRACT
Catalytic ring-expanding carbonylation of 2-aryl-2-oxazolines is reported as a novel method for the synthesis of 4,5-dihydro-1,3-oxazin-6-ones.
Various observations suggest the involvement of cobalt radicals as the catalytically active species.
Metal-catalyzed carbonylation is a versatile synthetic method
for a broad spectrum of organic carbonyl compounds.1
Significant innovations have continued to occur in this area,
particularly in the carbonylation of heteroatom-containing
substrates.2-5 Along this line, we have focused our research
on the carbonylative polymerization6-8 and demonstrated the
feasibility of synthesizing poly(â-alanine) and its derivatives
via CO-aziridine alternating copolymerization.9 In an effort
to broaden the scope of the copolymerization, we attempted
to synthesize acyclic polyimides by 2-oxazolines-CO co-
polymerization under a mechanistic hypothesis similar to that
of the aziridine-CO copolymerization.10 We discovered, to
our surprise, that the product of the presumed copolymeri-
zation of 2-phenyl-2-oxazoline and CO was 2-phenyl-4,5-
dihydro-1,3-oxazin-6-one. This reaction represents, to the
best of our knowledge, the first catalytic method for the
synthesis of 4,5-dihydro-1,3-oxazin-6-one heterocycles. The
4,5-didydro-1,3-oxazin-6-ones are interesting molecules be-
cause they undergo ring-opening reaction with both nucleo-
philes and electrophiles resulting in â-amino acid derivatives
or â-peptide bonds.10-12 Polymerization of this class of
molecule has been reported to afford acyclic polyimides,
(1) Frohning, C. D.; Kohlpaintner, Ch. W.; Gaub, M.; Seidel, A.;
Torrence, P.; Heymanns, P.; Hohn, A.; Beller, M.; Knifton, J. F.; Klausener,
A.; Jentsch, J.-D.; Tafesh, A. M. In Applied Homogeneous Catalysis with
Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; Wiley-
VCH: Weinheim, Germany, 2000; pp 29-200.
(2) Review on carbonylation of heterocycles: Khumtaveeporn, K.; Alper,
H. Acc. Chem. Res. 1995, 28, 414-422.
(3) Carbonylation of aziridines: (a) Mahadevan, V.; Getzler, Y. D. Y.
L.; Coates, G. W. Angew. Chem., Int. Ed. 2002, 41, 2781-2784. (b) Davoli,
P.; Forni, A.; Moretti, I.; Prati, F.; Torre, G. Tetrahedron 2001, 57, 1801-
1812. (c) Davoli, P.; Moretti, I.; Prati, F.; Alper, H. J. Org. Chem. 1999,
64, 518-521 and references therein. (d) Chamchaang, W.; Pinhas, A. R. J.
Org. Chem. 1990, 55, 2943-2950.
(4) Carbonylation of epoxides: (a) Getzler, Y. D. Y. L.; Mahadevan,
V.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chen. Soc. 2002, 124, 1174-
1175. (b) Lee, T. L.; Thomas, P. J.; Alper, H. J. Org. Chem. 2001, 166,
5424-5426. (c) Drent, E.; Kragtwijk, E. (Shell Oil Company). U.S. Patent
Application 5310948, 1994.
(7) Copolymerization of CO and epoxides: (a) Allmendinger, M.;
Eberhardt, R.; Luinstra, G.; Rieger, B. J. Am. Chem. Soc. 2002, 124, 5646-
5647. (b) Penney, J. M. Diss. Abstr. Int., B 1999, 60, 2688. (c) Furukawa,
J.; Iseda, H.; Saegusa, T.; Fujii, H. Makromol. Chem. 1965, 89, 263-268.
(8) Studies toward copolymerization of CO and imines: (a) Kacker, S.;
Kim, S. J.; Sen, A. Angew. Chem., Int. Ed. 1998, 37, 1251-1254. (b)
Dghaym, R. D.; Yaccato, K. J.; Arndtsen, B. A. Organometallics 1998,
17, 4-6.
(9) (a) Jia, L.; Ding, E.; Anderson, W. R. Chem. Commun. 2001, 1436-
1437. (b) Jia, L.; Sun, H.; Shay, J. T.; Allgeier, A. M.; Hanton, S. D. J.
Am. Chem. Soc. 2002, 124, 7282-7283. (c) Zhao, J.; Ding, E.; Allgeier,
A. M.; Jia, L. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 376-385.
(10) See Supporting Information.
(5) Carbonylation of imines: (a) Dghaym, R. D.; Dhawan, R.; Arndtsen,
B. A. Angew. Chem., Int. Ed. 2001, 40, 3228-3230. (b) Lafrance, D.; Davis,
J. L.; Dhawan, R.; Arndtsen, B. A. Organometallics 2001, 20, 1128-1136.
(c) Davis, J. L.; Arndtsen, B. A. Organometallics 2000, 19, 4657-4659.
(6) For reviews on CO-olefin copolymerization: (a) Drent, E.; Budzelaar
P. H. M. Chem. ReV. 1996, 96, 663-681. (b) Sen, A. Acc. Chem. Res.
1993, 26, 303-310.
(11) Kobayashi, S.; Tsukamoto, Y.; Saegusa, T. Macromolecules 1990,
23, 2609-2612.
(12) Baker, W.; Ollis, W. D. J. Chem. Soc. 1949, 345-349.
10.1021/ol034380f CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/08/2003

which are our original targets.11 Since 2-oxazolines as
common versatile synthetic intermediates are readily avail-
able often in optically pure forms via a variety of synthetic
methods,13 the novel reaction can be an important addition
to the repertoire of â-amino acid/â-peptide synthesis. Mo-
tivated by the above reasons, we followed up on the
discovery and wish to report our preliminary results here.
products in 56, 8, and 83% yields, respectively (entries 7, 8,
and 10). The yield of the carbonylation of 2,5-diphenyl-2-
oxazoline decreases to 52% when the reaction is carried out
under 200 psi (entry 9). The enamide PhCONHCHdCHPh,
which is a simple ring-opened isomer of the substrate, is
the major byproduct that apparently escaped from carbony-
lation.16 The formation of the enamide byproduct is sup-
pressed by high CO pressure. In contrast, the yields of the
carbonylation of the 4-methyl and 5-methyl substrates are
unaffected by CO pressure and are limited by the catalyst
turnovers, as no byproduct was observed in the reaction
mixtures (entries 7 and 8).
The catalytic reactions were typically carried out at 60
°C under 200-1000 psi of CO pressure in THF in the
presence of 5 mol % cobalt complexes BnCo(CO)4 1 and
BnCOCo(CO)4 2.14 Complexes 1 and 2 cannot be separated
because they are at equilibrium under CO.14 The most distinct
characteristics of the reaction is that an aromatic substituent
is required at the C(2) position of the 2-oxazoline substrate.15
A variety of aryls are suitable substituents. The electron-
donating 4-tolyl group renders the highest yield, while the
electron-withdrawing 3-fluorophenyl group and the sterically
encumbering 2-tolyl group decrease the yield (Table 1,
Understanding the mechanism of the carbonylation reac-
tion necessarily requires first perception of the catalytically
active species. One critical clue came from the fate of the
benzyl group in the catalyst precursors. After failing to detect
any benzyl-containing byproducts from the reaction, we
monitored the catalytic carbonylation of 2-phenyl-2-oxazoline
1
in situ by H NMR in a Wilmad high-pressure NMR tube.
The NMR reaction was carried out at 20 mol % catalyst
loading under 200 psi of CO pressure in THF-d8 at 60 °C.
The concentration of the substrate was chosen so that only
10 psi of CO, i.e., 5% of the total initial pressure in the NMR
tube, would be consumed at the completion of the reaction.
Toluene was produced during the reaction. Although we are
not certain as to the source of the hydrogen atom (not
deuterium, the product is toluene-d0), it seems that only
abstraction of a hydrogen atom by a benzyl radical provides
a reasonable explanation for the formation of toluene. The
benzyl radical logically would arise from the homolytic
dissociation of the Bn-Co bond,17 which in turn would leave
the 17-electron Co(CO)4 as the most likely catalytically active
species. We have performed several experiments to test the
involvement of Co radical species using the parent 2-phenyl-
Table 1. Ring-Expanding Carbonylation of
2-Aryl-2-oxazolinesa
entry
precatalyst
1 and 2
Ar
R1 R2
yieldb
1
2
3
4
5
6
7
8
9
Ph
H
H
H
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Ph
Ph
H
H
H
H
H
H
95% (92%)
>98% (88%)
30% (9%)
74% (72%)
45% (32%)
85% (78%)
56% (48%)
8%
52%
83% (60%)
0%
0%
4%
1 and 2
1 and 2
1 and 2
1 and 2
1 and 2
1 and 2
1 and 2
1 and 2
1 and 2
3
4-tolyl
2-tolyl
3-fluorophenyl
2-furanyl
2-thiophenyl
Ph
Ph
Ph
Ph
Ph
Ph
Ph
2-oxazoline as the substrate. First, we tested Na+Co(CO)4
-
10c
11d
12d
13d
and CH3COCo(CO)3(PPh3) (3) as precatalysts (entries 11 and
12). No reaction was observed as expected. Next, when Co2-
(CO)8 was used as the precatalyst, 4% of the substrate was
converted to the product as compared to 95% conversion
with 1 and 2 as the catalyst under the identical reaction
conditions (entry 13 vs entry 1). In comparison, when an
equimolar mixture of AIBN and Co2(CO)8 was used as the
precatalyst, the conversion was improved to 85% (entry 14),
while AIBN alone does not catalyze the carbonylation
reaction (entry 15). The catalytic activity of 1 and 2 was
completely inhibited by 1 equiv of TEMPO (entry 16). The
above observations as well as the fact that the ring-expanding
carbonylation requires a 2-aryl substituent ubiquitously
prompt a mechanistic hypothesis based on cobalt radicals.
We thus propose that the catalytic cycle is composed of
2-oxazoline coordination, intramolecular oxidative addition
(cleavage of the O-C(5) bond), CO insertion and coordina-
tion, and finally reductive elimination to extrude the product
(Scheme 1). The formally 19-electron intermediate A is likely
an (18 + δ)-electron species, i.e., the spin-delocalized
-
Na+Co(CO)4
Co2(CO)8
14d,e Co2(CO)8 with AIBN Ph
85%
0%
0%
15e
16a,f 1 and 2 with TEMPO Ph
AIBN
Ph
a Reaction conditions unless otherwise specified: 5 mol % Co; 7.60 mmol
of 2-oxazoline, 50 mL of THF, 200 psi of CO; 60 °C; 48 h. b Yield
determined by 1H NMR, and the isolated yield is given in parentheses.
c 1000 psi of CO. d 10 mol % Co. e 5 mol % AIBN. f Co/TEMPO molar
ratio ) 1:1.
entries 2-4). Substrates with heteroaryls at the C(2) position
can also be carbonylated in good to moderate yields (entries
5 and 6). Substitution at the C(4) and C(5) positions
significantly influences the reactivity of the substrates (entries
7-10). The 4-methyl, 5-methyl, and 5-phenyl derivatives
of 2-phenyl-2-oxazoline were converted to the corresponding
(13) (a) Gant, T. G.; Meyers, A. I. Tetrahedron 1994, 50, 2297-2360.
(b) Reuman, M.; Meyers, A. I. Tetrahedron 1985, 41, 837-860.
(14) Galamb, V.; Palyi, G.; Ungvary, F.; Marko, L.; Boese, R.; Schmid,
G. J. Am. Chen. Soc. 1986, 108, 3344-3351.
(15) When the C(2) substitutent is an alkyl group, the reaction product
is a polymer that appears to contain an acyclic imide structure and the
uncarbonylated amide.
(16) A second minor byproduct was also present but was not character-
ized.
(17) Halpern, J. Pure Appl. Chem. 1986, 58, 575-584 and references
therein.
1576
Org. Lett., Vol. 5, No. 9, 2003

be either a concerted event or a stepwise process in the
present cases depending on the substituent at the C(5)
position of the oxazoline substrate. Insertion and coordination
of CO are common organometallic reactions. The reductive
elimination may involve the first formation of an N-aryl-â-
lactam intermediate, which then isomerizes to the final
product, but we have not observed an N-aryl-â-lactam during
the reaction by high-pressure 1H NMR. The combination of
two Co(CO)4 radicals to form Co2(CO)8 is the most reason-
able pathway for the catalyst deactivation, which limits the
conversion of some of the substrates (entries 4 and 6-8).
In summary, we have shown that 2-oxazolines are a new
class of organic molecules that undergo the metal-catalyzed
carbonylation reaction. All evidence points to the mechanism
involving the cobalt radicals as the catalytically active
species. The catalysts for the transformation are almost
certainly not limited to the cobalt radicals. We expect that a
diverse variety of catalysts will be discovered for the novel
and potentially useful reaction.
Scheme 1. Plausible Mechanism of 2-Oxazoline
Carbonylation
Acknowledgment. This work was supported by the
National Science Foundation (CHE-0134285). L.J. gratefully
acknowledges the DuPont Company for a DuPont Young
Professor Award.
canonical structure B is an important contributor to the
bonding of the intermediate A.18 The stability of the odd-
electron species is rendered by the ability of the ligand to
accept the spin density. The extended π-conjugation of a
2-aryl group in the 2-oxazoline undoubtedly is favorable for
the spin delocalization. Note that the addition of organic
radicals to the nitrogen atom of NdC double bond has
recently been reported when an aromatic substitutent at C is
present to stabilize the resulting C-centered radical.19 The
same explanation is applicable to the requirement of a 2-aryl
substituent for the carbonylation. There are also precedents
for C-O bond rupture of the C-OCR2‚ structural moiety.20,21
The addition of the radical activated C-O bond to Co might
Supporting Information Available: Synthesis and char-
acterization of 2-aryl-2-oxazolines and 4,5-dihydro-1,3-
oxazin-6-ones and experimental procedure for the carbon-
ylation reaction. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL034380F
(18) Tyler, D. R.; Mao, F. Coord. Chem. ReV. 1990, 97, 119-140 and
references therein.
(19) (a) Viswanathan, R.; Prabharan, E. N.; Plotkin, M. A.; Johnston, J.
N. J. Am. Chem. Soc. 2003, 125, 163-168. (b) Prabharan, E. N.; Nugent,
B. M.; Williams, A. L.; Nailor, K. E.; Johnston, J. M. Org. Lett. 2002, 4,
4197-4200. (c) Johnston, J. N.; Plotkin, M. A.; Viswanathan, R.;
Prabhaakaran, E. N. Org. Lett. 2001, 3, 1009-1011. We thank Professor
K. S. Feldman of the Pennsylvania State University for making us aware
of the above references.
Scheme 2
(20) Hiraguri, Y.; Endo, T. J. Am. Chem. Soc. 1987, 109, 3779-3780.
(21) Feldman, K. S.; Vong, A. K. K. Tetrahedron Lett. 1990, 31, 823-
826.
Org. Lett., Vol. 5, No. 9, 2003
1577