Simple approach to β-lactam derivatives from N-acylimidazoles

Article abstract of DOI:10.1021/jo0607010

Reaction of N-acylimidazoles possessing an electron-withdrawing group in the α position with diarylimines produces β-lactams in high yields.

Full text of DOI:10.1021/jo0607010

3-sulfonyl-substituted â-lactams have been obtained by free
radical cyclization of enamides,7 acid-catalyzed rearrangement
of spiro[cyclopropane-1,5′-isooxazolidines],8 and through ru-
thenium9 and rhodium10 catalyzed intramolecular carbene inser-
tion of tertiary 2-diazocarboxamides. However, these methods
are based on multistep protocols and are problematic for
substrates bearing functionalities sensitive to carbenes or
oxidants. The regioselectivity or stereoselectivity of ring closure
in many cases is unacceptably low.10a,b,h,j
Simple Approach to â-Lactam Derivatives from
N-Acylimidazoles
Moshe Nahmany and Artem Melman*
Hebrew UniVersity of Jerusalem, GiVat Ram,
Jerusalem 91904, Israel
amelman@chem.ch.huji.ac.il
Recently, a rhodium-catalyzed Wolf rearrangement of thioesters
was used for preparation of 3-alkoxycarbonyl â-lactams.11
However, this method necessitates an additional desulfurization
step and provides an ambiguous stereochemical outcome.
ReceiVed April 2, 2006
We have published a method for generation of ketenes
through reaction of R-EWG substituted carboxylic acids with
carbodiimides.12,13 Ketenes of type 2 generated in situ by this
reaction were found to react with sterically hindered alcohols,
allowing their acylation in the presence of phenol or thiol
functionalities (Scheme 1).14 In the absence of alcohols, trapping
of ethoxycarbonylketene in situ with a second equivalent of a
carbodiimide was found to produce oxazines through [4 + 2]
cycloaddition.
Reaction of N-acylimidazoles possessing an electron-
withdrawing group in the R position with diarylimines
produces â-lactams in high yields.
On the basis of the efficient trapping of the ketene intermedi-
ates with alcohols and carbodiimides, we expected that similar
trapping with imines could serve as an attractive method for
preparation of â-lactams that are difficult to synthesize by
conventional methods. Our initial approach involved reactions
of carboxylic acids of type 1 with 1 equiv of dicyclohexylcar-
bodiimide (DCC) in the presence of an excess of N-benzylide-
neaniline (Scheme 1).
In contrast to our expectations, no traces of â-lactams of type
4 have been observed. Carboxylic acids of type 1 produced a
complex mixture of products. These results can be attributed
to fast reactions of corresponding ketenes with other nucleo-
philes that proceed faster than the relatively slow [2 + 2]
cycloaddition.
Substituted â-lactams attract considerable research interest
because of their biological activity1 and their ability to serve as
versatile building blocks in synthesis.2 Building â-lactam rings
is a crucial step in synthesis of new â-lactam antibiotics,3 and
development of new preparative approaches toward â-lactam
cycles is highly important.
Preparation of â-lactams through cycloaddition of ketenes
to imines allows convergent syntheses of substituted â-lactams.4
Although a number of methods for preparation of ketenes have
been introduced,5 reaction of acyl halides with tertiary amines
remains the most preparatively useful approach because of the
availability of starting material. However, in a number of cases,
application of acyl halides produces inadequate results. For
example, [2 + 2] cycloaddition to imines with ketenes that were
generated from acyl halides possessing strong electron-
withdrawing groups (EWG) such as carbonyl, phosphoryl, or
sulfonyl produces low yields of corresponding â-lactams.6
For preparation of EWG-substituted â-lactams a number of
useful alternatives to [2 + 2] cycloaddition of ketenes to imines
have been proposed. 3-Alkoxycarbonyl-, 3-phosphoryl-, and
(7) (a) D’Annibale, A.; Resta, S.; Trogolo, C. Tetrahedron Lett. 1995,
36, 9039-9042. (b) D’Annibale, A.; Pesce, A.; Resta, S.; Trogolo, C.
Tetrahedron 1997, 53, 13129-13138. (c) Attenni, B.; Cerreti, A.; D’Annibale,
A.; Resta, S.; Trogolo, C. Tetrahedron 1998, 54, 12029-12038.
(8) Cordero, F. M.; Salvati, M.; Pisaneschi, F.; Brandi, A. Eur. J. Org.
Chem. 2004, 2205-2213.
(9) (a) Choi, M. K. W.; Yu, W. Y.; Che, C. M. Org. Lett. 2005, 7, 1081-
1084.
(10) (a) Brown, D. S.; Elliott, M. C.; Moody, J. C.; Mowlem, T. J.;
Marino, J. P.; Padwa, A. J. Org. Chem. 1994, 59, 2447-2455. (b) Moody,
C. J.; Miah, S.; Slawin, A. M. Z.; Mansfield, D. J.; Richards, I. C.
Tetrahedron 1998, 54, 9689-9700. (c) Anada, M.; Watanabe, N.; Hash-
imoto, S. I. Chem. Commun. 1998, 1517-1518. (e) Basak, A.; Mandal, S.
Tetrahedron Lett. 2002, 43, 4241-4243. (e) Wee, A. G. H.; Duncan, S. C.
Tetrahedron Lett. 2002, 43, 6173-6176. (f) Yoon, C. H.; Nagle, A.; Chen,
C.; Gandhi, D.; Jung, K. W. Org. Lett. 2003, 5, 2259-2262. (g) Padwa,
A.; Straub, C. S. J. Org. Chem. 2003, 68, 227-239. (i) Wee, A. G. H.;
Duncan, S. C. J. Org. Chem. 2005, 70, 8372-8380. (h) Yoon, C. H.; Nagle,
A.; Chen, C.; Gandhi, D.; Jung, K. W. Org. Lett. 2003, 5, 2259-2262. (i)
Ridley, D. D.; Simpson, G. W. Aust. J. Chem. 1986, 39, 687-698. (j) Gois,
P. M. P.; Afonso, C. A. Eur. J. Org. Chem. 2003, 3798-3810.
(11) Jiao, L.; Zhang, Q.; Liang, Y.; Zhang, S.; Xu, J. J. Org. Chem.
2006, 71, 815-818.
(1) (a) Rolinson, G. N. J. Antimicrob. Chemother. 1998, 41, 589-603.
(b) The Organic Chemistry of â-Lactams; Georg, G. I., Ed.; Verlag
Chemie: New York, 1993. (c) Fisher, J. F.; Meroueh, S. O.; Mobashery,
S. Chem. ReV. 2005, 105, 395-424.
(2) (a) Alcaide, B.; Almendros, P. Curr. Med. Chem. 2004, 11, 1921-
1949. (b) Palomo, C.; Aizpurua, J. M.; Ganboa, I.; Oiarbide, M. Curr. Med.
Chem. 2004, 11, 1837-1872. (c) Deshmukh, A. R. A. S.; Bhawal, B. M.;
Krishnaswamy, D.; Govande, V. V.; Shinkre, B. A.; Jayanthi, A. Curr.
Med. Chem. 2004, 11, 1889-1920.
(3) See: Antimicrobial Agents: Antibacterials and Antifungals; Bryskier,
A., Ed.; ASMPress: Washington, DC, 2005; Chapters 5-14, pp 113-446.
(4) (a) Palomo, C.; Aizpurua, J. M.; Ganboa, I.; Oiarbide, M. Eur. J.
Org. Chem. 1999, 3223, 3-3235. (b) Tidwell, T. T. Eur. J. Org. Chem.
2006, 563-576.
(5) Tidwell, T. T. Ketenes II; John Wiley & Sons: Hoboken, 2006; pp
55-192.
(6) (a) Motoyoshiya, J.; Hirata, K. Chem. Lett. 1988, 211-214. (b)
Lozinskii, M. O.; Bodnar, V. N.; Pelkis, P. S. Zh. Org. Khim. 1980, 16,
228-229.
(12) Nahmany, M.; Melman, A. Org. Lett. 2001, 3, 3733-3735.
(13) Shelkov, R.; Nahmany, M.; Melman, A. J. Org. Chem. 2002, 67,
8975-8982.
(14) Shelkov, R.; Nahmany, M.; Melman, A. Org. Biomol. Chem. 2004,
2, 397-492.
10.1021/jo0607010 CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/29/2006
5804
J. Org. Chem. 2006, 71, 5804-5806

SCHEME 1
SCHEME 2
The initial failure to get cycloaddition products prompted us
to examine an alternative way to generate ketenes that would
exclude concomitantly generated nucleophiles that might result
in reactions competing with [2+2] cycloadditions. Although
generation of ketenes from carboxylic acids of type 1 can be
achieved with a number of coupling reagents,14 each of them
should produce an equimolar amount of a nucleophile that will
compete with the relatively slow [2 + 2] cycloaddition with
imines.
Consequently, we adopted another strategy to generate
ketenes of type 2. Instead of attempting to trap ketene
intermediates that are quantitatively produced through treatments
of carboxylic acids of type 1 with carbodiimides, it might be
possible to generate a small concentration of ketenes of type 2
through a reversible dissociation of derivatives of type 5 (eq
1). In this case the equilibrium concentration of ketenes of type
2 would be small but constant, thus allowing their [2 + 2]
cycloaddition with an imine.
N-acylimidazoles through E1cb elimination reaction. The situ-
ation is different for N-acylimidazoles that possess highly acidic
R hydrogens that can be abstracted with relatively weak bases
such as imidazole itself.19
In line with our predictions, treatment of carboxylic acids
1a-g with 1.1 equiv of CDI followed by reaction with imines
produced the corresponding â-lactams 4a-g with essentially
1
quantitative yield (Scheme 2). According to H NMR data all
â-lactams 4a-g possessed trans stereochemistry as evidenced
by coupling constants of endocyclic protons (2.2-2.9 Hz).20
Exclusive trans stereoselectivity for these compounds is most
likely related to easy epimerization of the highly acidic R proton
that produced the much more thermodynamically stable trans
isomer.
In addition to known â-lactams 4a,c,d, similar products were
obtained for phenylsulfinylacetic acid 1b. Steric induction by
the chiral sulfoxide function was found to be moderate, thus
providing a 2.5:1 mixture of diastereomers. Despite the high
yield of â-lactams 1b, both tert-butyl- and methylsulfinylacetic
acids failed to provide even a trace amount of â-lactams. Most
probably, the lack of reactivity of alkylsulfinylacetic acids is
due to the lower acidity21 of the R protons that precludes E1cb
elimination of imidazole in the corresponding N-acylimidazoles
and formation of ketenes of type 2.
This would be possible if carboxylic derivatives of type 5
possess a relatively weak X-CO bond that would provide a
sizable equilibrium concentration of ketene of type 2, and there
is a possibility for fast elimination of HX in the absence of
basic catalysts (i.e., tertiary amines) that can produce undesirable
zwitter-ionic adducts.15 Both of these requirements can be
achieved if imidazole is employed as the leaving group.
The practice of using 1,1-carbonyldiimidazole (CDI) as
coupling reagent is extensive. It has been used for preparation
of esters, amides, thioesters, and other carboxylic acid deriva-
tives.16 To the best of our knowledge, there are no examples of
use of N-acylimidazoles for generation of ketenes. The only
relevant, similar examples involve generation of ethyl ketenes
from N-acylbenzotriazoles by thermolysis17 or treatment with
NaH.18 These harsh conditions are understandable, since the low
basicity of imidazole is not sufficient for producing ketenes from
Carboxylic acids 1h and 1i possessing sufficiently acidic R
protons also failed to provide â-lactams. Lack of reactivity of
monoethyl 2-methyl is most probably related to the substantially
(15) (a) Broxton, T. J.; Duddy, N. W. J. Org. Chem. 1981, 46, 1186-
1191. (b) Cheong, D. Y.; Yoh, S. D.; Choi, J. H.; Shim, K. T. J. Korean
Chem. Soc. 1992, 36, 446-452.
(19) Tokutake, N.; Miyake, M.; Kirisawa, M. Synthesis 1983, 66-67.
(20) Palomo, C.; Cossio, F. P.; Arrieta, A.; Odriozola, J. M.; Oiarbide,
M.; Ontoria, J. M. J. Org. Chem. 1989, 54, 5736-5745.
(21) There is a difference of approximately 2.0 in Bordwell pKa between
aryl and alkyl sulfoxides. See: (a) Matthews, W. S.; Bares, J. E.; Bartmess,
J. E.; Bordwell, F. G.; Cornforth, F. J.; Drucker, G. E.; Margolin, Z.;
McCallum, R. J.; McCollum, G. J.; Vanier, N. R. J. Am. Chem. Soc. 1975,
97, 7006-7014. (b) Bordwell, F. G.; Branca, J. C.; Johnson, C. R.; Vanier,
N. R. J. Org. Chem. 1980, 45, 3884-3889.
(16) Handbook of Reagents for Organic Synthesis: ActiVating Reagents
and Protecting Groups; Pearson, A. J., Roush, W. R., Eds.; John Wiley &
Sons: Chichester, 1999; pp 93-96.
(17) Katritzky, A. R.; Huang, T. B.; Voronkov, M. V.; Steel, P. J. J.
Org. Chem. 2000, 65, 8069-8973.
(18) Katritzky, A. R.; Soleiman, M.; Yang, B. Heteroatom. Chem. 1996,
7, 365-367.
J. Org. Chem, Vol. 71, No. 15, 2006 5805

lower reactivity of disubstituted versus monosubstituted ketenes.22
Lack of reactivity of the imidazolide of cyanoacetic acid can
be attributed to the lower stability of the corresponding ketene,23
which can result in its very low equilibrium concentration in
the reaction mixture.
solid, mp 156-157 °C, 96%; IR (CDCl3) 3019, 1764, 1215, 757
1
cm-1; H NMR (300 MHz) 4.36 (s, 1H), 5.45 (s, 1H), 6.97-7.80
(m, 15H); 13C NMR (100.62 MHz) 56.02, 77.5, 117.2, 124.9, 125.9,
128.7, 129.1, 129.2, 129.3, 134.6, 134.6, 136.5, 137.6, 155.5. Anal.
Calcd for C21H17NO3S: C, 69.40, H, 4.71, N, 3.85. Found: C,
69.00, H, 4.69, N, 3.80.
Attempts to conduct similar [2 + 2] cycloadditions using
carboxylic acids of type 1 and substituted alkenes such as
ethoxyethylene, cyclopentadiene, or 2,3-dihydrofurane did not
provide any trace of the corresponding cyclobutanes. Similarly,
no cycloaddition was observed with substituted alkynes such
as phenylacetylene. Reactions with other imines such as
N-propylidene-2-phenylethylamine and ethyl N-benzylideneg-
lycinate did not provide â-lactams, most probably because of
the low stability of the imines in the presence of imidazole,
which is an unavoidable component of the reaction mixture.
In conclusion, we have demonstrated that N-acylimidazoles
possessing an electron-withdrawing group at the R position are
capable of producing â-lactams in high yield. The observed
results are compatible with a reaction mechanism involving
reversible generation of low concentrations of the corresponding
ketenes followed by their trapping with substituted imines.
trans-3-Diethoxyphosphoryl-1,4-diphenyl-azetidin-2-one (4d).6a
Pale yellow crystals, mp 81-83 °C, 99%; IR (CDCl3) 3018, 1756,
1215, 757 cm-1; 1H NMR (300 MHz) 1.31 (dt, J1 ) 6.9 Hz, J2 )
5.4 Hz, 6H), 3.50 (dd, J1 ) 15.4 Hz, J2 ) 2.9 Hz, 1H), 4.19 (dq,
J1 ) 7.7 Hz, J2 ) 7.3 Hz, 4H), 5.18 (dd, J1 ) 15.4 Hz, J2 ) 2.9
Hz, 1H), 7.03-7.39 (m, 10H); 13C NMR (100.62 MHz) 16.1, 16.2,
55.5, 55.5, 56.2, 57.7, 62.5, 62.6, 62.8, 62.9, 116.7, 124.0, 125.6,
128.7, 128.8, 129.0, 136.2, 136.3, 137.0, 137.0, 158.6, 158.7. Anal.
Calcd for C19H22NO4P: C, 63.50, H, 6.17, N, 3.90. Found: C,
63.21, H, 6.12, N, 3.58.
trans-3-Diethoxyphosphoryl-1-phenyl-4-(4-bromophenyl)aze-
tidin-2-one (4e). White crystals, mp 75-76 °C, 98%; IR (CDCl3)
1
3018, 1758, 1215, 754 cm-1; H NMR (300 MHz) 1.28 (dt, J1 )
4.4 Hz, J2 ) 4.0 Hz, 6H), 3.44 (dd, J1 ) 15.8 Hz, J2 ) 2.9 Hz,
1H), 4.15 (dq, J1 ) 7.7 Hz, J2 ) 7.3 Hz, 4H), 5.13 (dd, J1 ) 15.8
Hz, J2 ) 2.9 Hz, 1H), 6.97-7.30 (m, 9H); 13C NMR (100.62 MHz)
16.3, 16.4, 55.1, 55.1, 56.1, 58.1, 62.7, 62.8, 63.1, 63.2, 116.8,
124.4, 129.1, 129.3, 134.7, 134.9, 135.0, 136.9, 137.0, 158.6, 158.7;
HRMS calcd for C19H21BrNO4PNa+ 460.0284, found 460.0268.
Experimental Section
trans-3-Diethoxyphosphoryl-1-phenyl-4-(4-cyanophenyl)aze-
tidin-2-one (4f). Pale yellow crystals, mp 83-84 °C, 94%; IR
General Procedure for Preparation of â-Lactams (4a-g). A
solution of acid of type 1 (1 mmol) and CDI (178 mg, 1.10 mmol)
in dichloromethane (1 mL) was stirred at 25 °C for 1 h. To the
reaction mixture was added 1.0 mL of a corresponding imine in
solution (1 M solution, CH2Cl2). The reaction mixture was stirred
for 1 h, filtered through Celite, and evaporated. The residue was
purified by flash chromatography (neat chloroform) to afford title
compounds.
1
(CDCl3) 3019, 2400, 2233, 1761, 1215, 754 cm-1; H NMR (300
MHz) 1.28 (dt, J1 ) 6.9 Hz, J2 ) 2.2 Hz, 6H), 3.44 (dd, J1 ) 15.8
Hz, J2 ) 2.5 Hz, 1H), 4.15 (dq, J1 ) 7.7 Hz, J2 ) 7.3 Hz, 4H),
5.21 (dd, J1 ) 15.8 Hz, J2 ) 2.5 Hz, 1H), 7.01-7.24 (m, 5H),
7.43 (d, J ) 8.0 Hz, 2H), 7.62 (d, J ) 8.0 Hz, 2H); 13C NMR
(100.62 MHz) 16.4, 16.5, 55.27, 55.19, 56.4, 57.8, 63.0, 63.1, 63.4,
63.5, 113.0, 116.8, 118.1, 124.8, 126.7, 129.3, 133.2, 136.9, 136.9,
142.0, 142.0, 158.3, 158.4; HRMS calcd for C20H21N2O4PH+
385.1313, found 385.1317.
trans-3-Ethoxycarbonyl-1,4-diphenyl-azetidin-2-one (4a).10a
Colorless oil, 98%; IR (CDCl3) 3019, 1762, 1731, 1215, 757 cm-1
;
1H NMR (300 MHz) 1.23 (t, J ) 7.0 Hz, 3H), 3.88 (d, J ) 2.4 Hz,
1H), 4.19 (q, J ) 7.3 Hz, 2H), 5.25 (d, J ) 2.4 Hz, 1H), 6.95-
7.30 (m, 10H); 13C NMR (100.62 MHz) 14.15, 57.25, 62.12, 63.58,
117.18, 124.41, 126.20, 127.08, 129.05, 129.12, 136.35, 137.20,
1259.27, 166.30.
trans-3-Diethoxyphosphoryl-1-phenyl-4-(4-chlorophenyl)aze-
tidin-2-one (4g). White crystals, mp 93-94 °C, 98%; IR (CDCl3)
1
3018, 1758, 1215, 754 cm-1; H NMR (300 MHz) 1.28 (dt, J1 )
4.4 Hz, J2 ) 4.4 Hz, 6H), 3.44 (dd, J1 ) 15.8 Hz, J2 ) 2.9 Hz,
1H), 4.15 (dq, J1 ) 7.7 Hz, J2 ) 7.3 Hz, 4H), 5.13 (dd, J1 ) 15.8
Hz, J2 ) 2.9 Hz, 1H), 6.98-7.30 (m, 9H); 13C NMR (100.62 MHz)
16.36, 16.44, 55.21, 55.23, 56.1, 58.1, 62.8, 62.9, 63.2, 63.3, 116.9,
122.9, 124.5, 127.5, 129.1, 132.5, 135.7, 137.00, 137.03, 158.6,
158.7. Anal. Calcd for C19H21ClNO4P: C, 57.95, H, 5.37, N, 3.56.
Found: C, 57.66, H, 5.46, N, 3.43.
trans-3-Benzenesulfinyl-1,4-diphenyl-azetidin-2-ones (4b). White
amorphous solid mixture of diastereomers in 2.5:1 ratio, mp 140-
141 °C (major isomer), 97%; IR (CDCl3) 3018, 1757, 1500, 1215,
1
757 cm-1; H NMR (300 MHz) 4.08 (d, J ) 2.4 Hz, 1H, minor
stereoisomer), 4.27 (d, J ) 2.4 Hz, 1H, major stereoisomer), 5.28
(d, J ) 2.4 Hz, 1H, major stereoisomer), 5.41 (d, J ) 2.4 Hz, 1H,
minor stereoisomer), 6.82-7.77 (m, 15H [major] + 15H [minor]);
13C NMR (100.62 MHz) 56.2, 76.4, 117.1, 124.5, 124.9, 126.2,
128.9, 129.1, 129.1, 129.3, 131.9, 135.5, 136.5, 139.0, 157.3. Anal.
Calcd for C21H17NO2S: C, 72.60, H, 4.93, N, 4.03. Found: C,
72.35, H, 5.05, N, 4.00.
Acknowledgment. This research was supported by the Israel
Science Foundation.
trans-3-Benzenesulfonyl-1,4-diphenyl-azetidin-2-one (4c). White
Supporting Information Available: Complete experimental
1
details and copies of H and 13C NMR spectra of products 4a-g.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(22) Allen, A. D.; Kresge; A. J.; Schepp, N. P.; Tidwell, T. T. Can. J.
Chem. 1987, 65, 1719-1723.
(23) McAllister, M. A.; Tidwell, T. T. J. Org. Chem. 1994, 59, 4506-
4515.
JO0607010
5806 J. Org. Chem., Vol. 71, No. 15, 2006