Microwave-assisted carbonyl-carbonyl coupling route for the preparation of a useful intermediate in the synthesis of carbapenems

Article abstract of DOI:10.1080/00397910802638586

Microwave-assisted organic synthesis has been used for the cyclization of a monocyclic oxoamide to the corresponding bicyclic derivative via a Wittig-type coupling by means of phosphite reagents. The parameters influencing the reaction yields under microw

Full text of DOI:10.1080/00397910802638586

This article was downloaded by: [UTSA Libraries]
On: 07 October 2014, At: 16:48
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
Synthetic Communications: An
International Journal for Rapid
Communication of Synthetic
Organic Chemistry
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/lsyc20
Microwave-Assisted
Carbonyl–Carbonyl Coupling
Route for the Preparation of
a Useful Intermediate in the
Synthesis of Carbapenems
Zhining Xia a , Sha Long a , Shangyou Xiao a
,
Xiaoping Zou a , Antonio D'Aurizio b & Mauro
Panunzio b
a Institute of Chemistry and Chemical Engineering ,
Chongqing University , Chongqing, China
b ISOF-CNR , Dipartimento di Chimica , G.
Ciamician, Bologna, Italy
Published online: 12 May 2009.
To cite this article: Zhining Xia , Sha Long , Shangyou Xiao , Xiaoping Zou , Antonio
D'Aurizio & Mauro Panunzio (2009) Microwave-Assisted Carbonyl–Carbonyl Coupling
Route for the Preparation of a Useful Intermediate in the Synthesis of Carbapenems,
Synthetic Communications: An International Journal for Rapid Communication of
Synthetic Organic Chemistry, 39:12, 2151-2160, DOI: 10.1080/00397910802638586
To link to this article: http://dx.doi.org/10.1080/00397910802638586
PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness,
or suitability for any purpose of the Content. Any opinions and views
expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the
Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the
Content.
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions

Synthetic Communications1, 39: 2151–2160, 2009
Copyright # Taylor & Francis Group, LLC
ISSN: 0039-7911 print=1532-2432 online
DOI: 10.1080/00397910802638586
Microwave-Assisted Carbonyl–Carbonyl
Coupling Route for the Preparation of a Useful
Intermediate in the Synthesis of Carbapenems
Zhining Xia,1 Sha Long,1 Shangyou Xiao,1 Xiaoping Zou,1
Antonio D’Aurizio,2 and Mauro Panunzio2
1Institute of Chemistry and Chemical Engineering, Chongqing University,
Chongqing, China
2ISOF-CNR, Dipartimento di Chimica, G. Ciamician, Bologna, Italy
Abstract: Microwave-assisted organic synthesis has been used for the cyclization
of a monocyclic oxoamide to the corresponding bicyclic derivative via a Wittig-
type coupling by means of phosphite reagents. The parameters influencing the
reaction yields under microwave (dielectric heating) and classical conditions
(convective heating) have been evaluated.
Keywords: Carbapenem, imipenem, MAOS (microwave-assisted organic synth-
esis), phosphite-mediated cyclization, Wittig intramolecular reaction
INTRODUCTION
A valuable route to the synthesis of the bicyclic scaffold of penem,
independently developed by the pioneering work of two research
groups,[1–3] is based on the carbonyl–carbonyl coupling via a phosphite-
mediated reductive olefination–cyclization of a suitable oxoamide and
has been demonstrated to be very useful for the preparation of other
b-lactam carbapenem antibiotics[4,5] including bicyclic meropenem[6]
and tricyclic sanfetrinem (Scheme 1).[7,8]
Received November 5, 2008.
Address correspondence to Mauro Panunzio, ISOF-CNR, Department of
Chemistry, G. Ciamician, via Selmi 2, Bologna I-40126, Italy. E-mail: mauro.
panunzio@unibo.it
2151

2152
Z. Xia et al.
Scheme 1. Popular nontraditional b-lactam antibiotics.
From an experimental point of view, one equivalent of b-lactam-
oxoamide of type 2 (Scheme 2) is treated in a high-boiling hydrocarbon
solvent with several equivalents (from four to eight) of a trialkylpho-
sphite for several hours at high temperature. Usually yields are from
satisfactory to good. [Triethylphosphite (TEP) is the most popular one
because of its low cost, relative low toxicity, good stability, and commer-
cial availability, although alkyldiethoxyphosphite have been shown to be
more reactive.[4,8]
]
In the course of our studies on the synthesis of carbapenem antibio-
tics, we became interested in devising rapid cyclization of an intermediate
oxalimide 2, synthetized in our laboratory, to a bicyclic intermediate 1
(Scheme 2), taking advantage of our experience in the use of MAOS
(microwave-assisted organic synthesis).[9]
The final goal of our research was to establish an easy, rapid proce-
dure that could be adopted for a liquid-phase combinatorial synthesis of
different intermediates of type 1 and a scale-up application of the result-
ing MAOS technology to the synthesis of Imipenem[10] and related
b-lactam antibiotics (Scheme 1). As a matter of fact, MAOS has been
proved to be an efficient and rapid method: dramatically reduced reac-
tion times, increased product yields, and enhanced product purities[11,12]
Scheme 2. Cyclization mediated by microwaves.

Microwave-Assisted Wittig Cyclization
2153
are among the several advantages generated by this technique. On the
other hand, after the discovery of thienamycin, a prototype of the well-
known commercial carbapenems such as imipenem and meropenem,
research on such compounds is needed to test the new compounds with
molecular diversity against strains of antibiotic-resistant bacteria, includ-
ing the very dangerous MRSA (methicillin-resistant Staphylococcus
aureus).
RESULTS AND DISCUSSION
Compound 2 was prepared according to the synthetic path in the litera-
ture with slight modifications in reagents and conditions (Scheme 3).[13]
As the starting point of this study to establish reliable experimental
conditions for those we were going to test by MAOS, compound 1 was
prepared by classical heating (CH) using commercially available TEP
as the cyclization agent. After several attempts, the best conditions in
term of yields and ratios of the reagents were as follows.
Preparation of 1 from 2 by Classical Heating in a Single Step
A dry, three-necked flask, equipped with a magnetic stirrer and a reflux
condenser and under a nitrogen atmosphere, was charged with the oxo-
amide 2 (5.83 g, 8.9 mmol) and 264 ml of anhydrous xylene. TEP [P(OEt)3]
(12.3 ml, 71.7 mmol) and q-dihydroxybenzene (0.29 g, 2.6 mmol) were
Scheme 3. Reagents and conditions: (i) Allyl bromide, Zn (2 eq), Y% ¼ 98; (ii)
KmnO4, CH3COOH, acetone, H2O, Y% ¼ 97%; (iii)
DMAP (cat.),
DCC, toluene Y% ¼ 93; and (iv) , pyridine, toluene, Y% ¼ 81.

2154
Z. Xia et al.
added, and the mixture was heated at 139 ꢀC. The reaction mixture was
stirred for 3.5 h and monitored by thin-layer chromatography (TLC)
until the starting material disappeared. The reaction mixture was cooled
and washed with saturated aqueous NaHCO3 solution, KHSO4 solution
(1 M), and finally brine. The organic layer was dried, and the solvent was
removed under vacuum. The crude product was purified by flash chro-
matography on silica gel (CH2Cl2–acetone 98:2) to give 1 (1.77 g,
32.1%) as an amorphous solid (>95% purity).
Mp 54–58ꢀC. IR (KBr) t (cmꢁ1): 1778, 1709, 1607, 1523, 1333. 13C
NMR d, 400 MHz (CDCl3): ꢁ5.005, ꢁ4.311, 17.916, 22.462, 25.667,
28.319, 32.389, 40.216, 40.925, 52.170, 64.977, 65.743, 67.508, 76.684,
77.00, 77.316, 79.858, 123.623, 124.008, 127.956, 143.152, 147.558, 155.72,
1
160.754, 176.08. H NMR d, 400 MHz (CDCl3): ꢁ0.084 (s, 6H), 0.873
(s, 9H), 1.25 (d, 3H), 1.441 (s, 9H), 2.9 (m, 1H), 3.0–3.1 (m, 2H), 3.16 (dd,
1H), 3.34 (m, 3H), 4.24 (m, 2H), 4.92 (bs, 1H), 5.20, 5.40 (AB, 2H), 7.65
(d, 2H), 8.2 (d, 2H). Anal. calcd. for C29H43N3O8SSi: C, 56.01; H, 6.97;
N, 6.76. Found: C, 56.03; H, 6.95; N, 6.78.
Preparation of 1 from 2 by Classical Heating in a Two-Step Procedure
As for the previous procedure, the flask was charged with the oxoamide 2
(0.330 g, 0.5 mmol) and 5 ml of anhydrous toluene. TEP [P(OEt)3]
(0.450 ml, 3.7 mmol) was added, and the mixture was heated at 90ꢀC
under stirring for 4 h. After cooling at rt, the liquid phase was evaporated
at reduced pressure by a rotavapor, the residue was dissolved in xylene
(5 ml), catalytic p-dihydroxybenzene (0.04 g) was added, and the resulting
mixture was heated at reflux for 1 h. The reaction mixture was cooled and
washed with saturated aqueous NaHCO3 solution, KHSO4 solution
(1 M), and finally brine. The organic layer was dried, and the solvent
was removed under vacuum. The crude product was purified by flash
chromatography on silica gel (CH2Cl2–acetone 98:2) to give 1 (0.125 g,
40.3%) and 12 (0.022 g, 7%).
Compound 12: oil. IR (CHCl3) t (cmꢁ1): 1750, 1708; 1H NMR
(400 MHz, CDCl3): d ¼ 8.22 (d, J ¼ 8.8 Hz, 2H), 7.52 (d, J ¼ 8.8 Hz, 2H),
5.23 (s, 2H), 4.78 (bs, 1H), 4.12 (m, 3H), 3.24 (m, 1H), 2.95 (m, 5H),
1.43 (s, 9H), 1.20 (d, J ¼ 6 Hz, 3H), 0.85 (s, 9H), 0.05 (s, 3H), 0.03 (s, 3H).
Microwave-Assisted Wittig Cyclization
To assess the ability of microwave heating to drive the cyclization as good
as or possibly better than under CH conditions, we performed a series of

Microwave-Assisted Wittig Cyclization
2155
experiments under microwave irradiation, changing different parameters.
Generally speaking, for the sake of simplicity, we decided to use a single
solvent (the higher boiling xylene) and skip the removal of excess TEP (a
dedicated experiment showed no significant variation in the final yield).
Accordingly, a mixture of TEP (excess) and oxoamide 2 in xylene was
exposed to irradiation in a sealed tube using the monomode CEM Dis-
cover and the multimode Milestone MicroSYNTH microwave ovens.
The reaction of oxamides of type 2 and excess TEP is supposed to
take place[1,14] via an initial formation of the betaine 8 by a two-electron
transfer from the phosphorous atom to the imides carbonyl oxygen
(Scheme 4). The betaine 8, under the thermal conditions of the reaction,
generates the transition carbene 9 by elimination of triethylphosphate. In
turn, carbene 9 is trapped by the excess of TEP to furnish the phosphite
ylide 10, which generates the bicyclic compound 1 via a classical intramo-
lecular Wittig reaction. On the other end, decomposition of unreacted
phosphite ylide 10 with water presumably results in the formation of
the reduced product 12 (Scheme 4, path 10 ) 11 ) 12). It has been
reported that the formation of the ylide 10 takes place at a relatively cool
temperature,[13] whereas the Wittig cyclization step needs much great
temperature. With this information, we started our research studies, plan-
ning a two-step, one-pot reaction as previously described for CH on the
base of different operational temperatures during the reaction cycle.
Scheme 4. Proposed mechanism for Wittig-type cyclization.

2156
Z. Xia et al.
The results obtained, reported in Table 1, warrant some comments:
1. The increase of yields and the shorter reaction time from CH to DH is
evident from entries 1, 2, 3, and 6 respectively. Once again, MAOS has
displayed the advantages claimed.
2. There are no dramatic differences in the reaction time using mono-
mode or multimode ovens (entries 3–5 and 6–8). However, significant
differences in yields are observed. No dedicated studies on this aspect
have been performed at this time.
3. Hydrocarbon solvents have been shown to be more efficient (as already
known using CH) than other solvents, and xylene is better than toluene,
probably because of the greater boiling point (entries 3, 6, and 9).
Table 1. Synthesis of 1 from 2 with different temperature, time, and solvent
Yieldb
(%)
Entry
Solvent
Temperaturea
Heating
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Toluene=xylene 90ꢀC = 240 min then 139ꢀC= 60 min
CH
CH
40.3c
Xylene
Xylene
Xylene
Xylene
Xylene
Xylene
Toluene
Toluene
Toluene
Toluene
Cl-benzene
CH2Cl2
CHCl3
139ꢀC = 210 min
32.1
61.8
57.9
42.6
50.5
40.2
115ꢀC = 70 min then 150ꢀC = 95 min
115ꢀC = 35 min then 150ꢀC = 60 min
115ꢀC = 15 min then 150ꢀC = 30 min
115ꢀC = 30 min then 150ꢀC = 95 min
115ꢀC = 30 min then 150ꢀC = 60 min
80ꢀC = 30 min then 115ꢀC = 70 min
115ꢀC = 40 min then 150ꢀC = 70 min
150ꢀC = 90 min
DHd
DHd
DHd
DHe
DHe
DHe
DHd
DHd
DHd
DHd
DHd
DHd
DHd
DHd
Trace
53.4
29.5
6.3
45.6
0
0
0
0
115ꢀC = 90 min
115ꢀC = 65 min then 135ꢀC = 95 min
50ꢀC = 210 min
65ꢀC = 60 min
102ꢀC = 180 min
1,4-Dioxane
DMF
153ꢀC = 30 min
Note. A ratio of 1=8=0.20 (in mmol) of oxoamide 2, TEP, and dihydroxyben-
zene was used in all the experiments reported.
aThe reaction temperatures were reached by applying a power of 280 W for
CEM monomode and 300 W for MicroSYNTH.
bYields have been calculated for entries 1, 2, and 6 on pure isolated products.
The yields for entries 3–5 and 6–12 were calculated by HPLC analysis by the
support of a standard calibration curve. For entries 13–16, no traces of 1 were
detected in HPLC analyses.
cThe product was fully identified by standard analytical data.
dMonomode CEM-Discover was used as the microwave oven model.
eMultimode MicroSYNTH (Milestone) was used as the microwave oven model.

Microwave-Assisted Wittig Cyclization
2157
4. The best ratio (1=8) between oxoamide 2 and TEP has been demon-
strated to be identical to that used in the CH experiment.
5. Better yields are achieved with a two-step reaction (entries 8 and 9) in
accordance with Scheme 4 and related discussion.
Preparation of 1 from 2 by Dielectrical Heating (Multimode
MicroSYNTH Milestone) (Table 1, Entry 6)
A mixture of 2 (0.4 mmol, 0.26 g), TEP (3.2 mmol, 0.390 ml), a catalytic
amount of p-dihydroxybenzene, and xylene (0.1 mmol, 0.01 g) was irra-
diated in a closed system for 70 min at a temperature of 115 ꢀC. After
this period, the temperature rose to 150 ꢀC, and irradiation was contin-
ued for an extra 95 min until high-performance liquid chromatography
(HPLC) showed the disappearance of the starting material. The crude
reaction mixture was cooled at rt, and the solvent was removed by rota-
vapor. The residue was dissolved in ethyl acetate (30 ml) and washed
with an aqueous solution (5%) of 5 ml of NaHCO3, 5 ml of a solution
of KHSO4 (1 M), and brine. The organic phase was dried, the solvent
was removed, and the residue was purified by flash chromatography
(methylene chloride=acetone 95=5) to give pure 1 (0.125 g, 50.5% yields).
Determination of Content/Concentration of 1 in the Crude Reaction
Mixture by the Standard Curve Method
General
A Shimadzu LC-10AT HPLC system, HPLC columm (250 ꢂ 4.6 mm
i.d.), and Kromasil C18 5 micrometer (Dalian Elite Analytical Instru-
ment Co., Ltd., China) were used. All the samples were analyzed under
the same conditions: mobile phase: 85% acetonitrile, 15% water; flow
rate: 1 mL=min; detection wavelength: 254 nm.
Determination of Standard Curve
1. Flash-chromatography-purified 1 was crystallized and dissolved into
mobile-phase solution (85% acetonitrile and 15% water). A stock stan-
dard solution of 1.00 mg=ml was prepared.
2. The series of standard solutions with different concentrations of 1
(Table 1) were obtained by diluting the stock solution into mobile-
phase solution. After injection of 20-ml standard samples into the

2158
Z. Xia et al.
Table 2. Standard curve calibrations
Concentration (mg=mL)
Peak area (mVa s)
0.08
0.09
0.10
0.20
0.40
0.60
0.80
1484073
1588699
1811760
3485764
7023536
10581910
13908490
amV is millivolts.
HPLC, the relative peak areas were automatically calculated by work
station (Table 2).
3. The standard curve was made according to the peak areas against the
relative concentrations (Fig. 1).
4. From the peak area of 1, its amount can be calculated through the
standard curve (Fig. 1).
Preparation of 1 from 2 by Dielectrical Heating (Monomode CEM)
(Table 1, Entry 3 as Typical Example)
A mixture of 2 (0.1 g), TEP (0.08 ml), a catalytic amount of p-dihydroxy-
benzene, and xylene (2 ml) was irradiated in a closed system for 70 min at
Figure 1. Standard curve calibration of 1: y ¼ 2E þ 07x; correlation coefficient
R2 ¼ 0.9998.

Microwave-Assisted Wittig Cyclization
2159
a temperature of 115 ꢀC in a CEM Discover oven. After this period, the
temperature was rose to 150 ꢀC, and irradiation was continued for an
extra 95 min until HPLC showed the disappearance of the starting mate-
rial. The yields (61.8%) were calculated by HPLC on the crude reaction
mixture.
CONCLUSION
In this article, we have demonstrated that the MAOS technique shows
better results, in terms of yields and reaction time, than the classical
CH method. The efficiency of this technique on a highly functionalized
substrata confers extra value, in our opinion, to the MAOS.
ACKNOWLEDGMENTS
The authors thank the Ministry of Science and Technology (China) for
the financial support (Grant No. 2006DFA43520). M. P. thanks
Ministero Affari Esteri (MAE) for the financial support within the
12th Joint Commission on S & T Cooperation Between the People’s
Republic of China and the Italian Republic for the years 2006–2009.
A. D. thanks Novachem Aromatici, Gallarate, Italy, for financial
support.
REFERENCES
1. Afonso, A.; Hon, F.; Weinstein, J.; Ganguly, A. K. New synthesis of penems,
the oxalimide cyclization reaction. J. Am. Chem. Soc. 1982, 104, 6138–6139.
2. Battistini, C.; Scarafile, C.; Foglio, M.; Franceschi, G. A new route to penems
and carbapenems. Tetrahedron Lett. 1984, 2395–2398.
3. Yoshida, A.; Hayashi, T.; Takeda, N.; Oida, S.; Ohki, E. New synthesis of
penems via reductive cyclization reaction of oxalimides with trialkyl
phosphite. Chem. Phar. Bull. 1983, 31, 768–771.
4. Mori, M.; Oida, S. A short-step synthesis of orally active carbapenem
antibiotic CS-834. Chem. Pharm. Bull. 2000, 48, 126–130.
5. Bitha, P.; Lin, Y.-I. Efficient synthesis of oral THF 1b-methylcarbapenem
(OCA-983) via the wittig process. Synth. Commun. 2001, 31, 587–594.
6. Sunagawa, M.; Matasumura, H.; Inoue, T.; Fukasawa, M.; Kato, M. A novel
carbapenem antibiotic, Sm-7338 structure-activity-relationships. J. Antibiotics
1990, 43, 519–532.
7. Tamburini, B.; Perboni, A.; Rossi, T.; Donati, D.; Gaviraghi, G.; Tarzia, G.
Recent Advances in the Chemistry of Anti-infective Agents; RSC: Cambridge,
1992.

2160
Z. Xia et al.
8. Panunzio, M.; Camerini, R.; Pachera, R.; Donati, D.; Marchioro, C.;
Perboni, A. Highly stereocontrolled synthesis of enantiomeric 4-methoxy
trinems via resolution of scalemic enol phosphates. Tetrahedron: Asymmetry
1996, 7, 2929–2938.
9. Panunzio, M.; Campana, E.; Martelli, G.; Vicennati, P.; Tamanini, E.
Microwave in organic synthesis: Myth or reality?. Mat. Res. Innov. 2004, 8,
27–32.
10. Leanza, W. J.; Windonger, K. J.; Miller, T. M.; Christensen, B. G.
N-Acetimidoyl- and N-formimidoylthienamycin derivatives: Antipseudomo-
nal b-lactam antibiotics. J. Med. Chem. 1979, 22, 1435–1436.
11. Loupy, A. Microwaves in Organic Synthesis; Wiley-VCH: Weinheim, 2002.
12. Kappe, C. O. The use of microwave irradiation in organic synthesis: From
laboratory curiosity to standard practice in twenty years. Chimia 2006, 60,
308–312.
13. Yoshida, A.; Tajima, Y.; Takeda, N.; Oida, S. An efficient carbapenem
synthesis via an intramolecular Wittig reaction of new trialkoxyphosphorane-
thiolesters. Tetrahedron Lett. 1984, 25, 2793–2796.
14. Perrone, E.; Alpegiani, M.; Bedeschi, A.; Giudici, F.; Franceschi, G. The
carbonyl–carbonyl coupling route to penems: A stepwise analysis. Tetra-
hedron Lett. 1984, 25, 2399–2402.