Synthesis of β-lactams by 4-exo-tet cyclization process induced by electrogenerated cyanomethyl anion, part 2:[1] Stereochemical implications

Article abstract of DOI:10.1002/adsc.200700156

An efficient electrochemically induced synthesis of chiral eis β-lactams has been described, via deprotonation of chiral amides containing an acidic methylene group and a bromine atom as leaving group and bearing a chiral auxiliary or amine function. The electrogenerated base - cyanomethyl anion - is easily obtained by galvanostatic reduction of acetonitrile- tetraethylammonium hexafluorophosphate solutions under very mild conditions. The yields are high and the cis-diastereoselection complete. The use of starting chiral amides has allowed in many cases the preparation of the most abundant isomer in a pure form.

Full text of DOI:10.1002/adsc.200700156

FULL PAPERS
DOI: 10.1002/adsc.200700156
Synthesis of b-Lactams by 4-exo-tet Cyclization Process Induced
by Electrogenerated Cyanomethyl Anion, Part 2:[1]
StereochemicalImpilcations
Marta Ferocia,*
a
Dipartimento di Ingegneria Chimica Materiali Ambiente, Università “La Sapienza”, via Castro Laurenziano 7,
00161 Roma, Italy
Fax : (+39)-06-4976-6749; e-mail: marta.feroci@uniroma1.it
Received: March 26, 2007; Revised: July 2, 2007; Published online: September 11, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: An efficient electrochemically induced yields are high and the cis-diastereoselection com-
synthesis of chiral cis b-lactams has been described, plete. The use of starting chiral amides has allowed
via deprotonation of chiral amides containing an in many cases the preparation of the most abundant
acidic methylene group and a bromine atom as leav- isomer in a pure form.
ing group and bearing a chiral auxiliary or amine
function. The electrogenerated base – cyanomethyl
anion – is easily obtained by galvanostatic reduction Keywords: cathodic reduction; 4-exo-tet cyclization;
of acetonitrile-tetraethylammonium hexafluorophos- electrogenerated cyanomethyl anion; electrosynthe-
phate solutions under very mild conditions. The sis; cis-b-lactams
Introduction
has been carried out using a radical pathway and an
ionic one. The radical synthesis led to trans-isomers[5]
Natural and synthetic b-lactams are compounds of and, when a chiral amine was used, some trans-dia-
great interest due to their well known pharmacologi- stereoselection was obtained. When an oxidative cou-
cal activities.[2] Moreover, they are used also as syn- pling of dianions was used to obtain b-lactams, a mix-
thons for biologically important compounds.[3] It is ture of isomers was attained, with a predominance of
thus obvious that organic chemists are interested in the cis ones.[6] In particular, a good cis-diastereoselec-
finding newer and newer syntheses of these com- tion was obtained. The ionic pathway, starting from
pounds that are stereoselective, easy, proceed with ethyl N-2-iodopropionyl-N-allyl glycinate (after reac-
high yields and, possibly, are green.
tion with t-BuOK in THF) led to the pair of cis-iso-
There are two main pathways for the synthesis of mers.[7]
the azetidin-2-one ring: [2+2]cycloaddition and cycli-
In recent years, the author and collaborators have
zation reactions. Among the [2+2]cycloadditions, the established an electrochemical method for the synthe-
Staudinger reaction between an imine and a ketene[4] sis of azetidin-2-one ring by a cyclization reaction, at
[8]
À
is the most common, due to the predictability of the first following an N C-4 bond formation path, sub-
[1,9]
À
stereochemical output, but also a [2+2]cycloaddition sequently following a C-3 C-4 one. In both cases, a
between an olefin and an isocyanate is used.[3] The very mild methodology has been used, consisting of
À
À
cyclization reaction can be due to an N C-2, an N C- deprotonation and intramolecular halide displacement
4, or a C-3 C-4 bond formation. This method has a to form the four-membered ring. The base used for
less predictable stereochemical output and it is there- the deprotonation is the electrogenerated cyanometh-
fore less used. The stereochemistry of the products is, yl anion (an electrogenerated base, EGB[10]), easily
in fact, very important as it is well known that the obtained by galvanostatic cathodic reduction, at room
spatial structure of the molecules has often a deter- temperature, of a solution of acetonitrile–0.1 mol
mining influence on their biological activity. Thus, dmÀ3 Et4NPF6 (Scheme 1) at a platinum electrode.[11]
À
there is a necessity to develop b-lactam syntheses
This electrogenerated base has a high reactivity,
with a clearly defined stereochemical outcome. In par- being a naked anion as its counterion is a tetraalkyl-
À
ticular, the C-3 C-4 cyclization synthesis of b-lactams ammonium cation. It can deprotonate a nitrogen
Adv. Synth. Catal. 2007, 349, 2177 – 2181
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2177

Marta Feroci
FULL PAPERS
Scheme 1. Electrochemical generation of cyanomethyl
anion.
À
atom (starting the N C-4 pathway), or a carbon atom
(starting the C-3 C-4 one).
Scheme 3. General reaction Scheme for the electrochemical-
ly induced synthesis of b-lactams.
À
In particular, the deprotonation of a suitable bro-
moamide by electrogenerated cyanomethyl anion led
to the formation of a b-lactam predominantly as the chiral auxiliary (like Oppolzerꢁs camphor sultam or
cis-isomer in good yields (Scheme 2, cis/trans ratio menthol) is used in the Y position, the deprotection
from 87/13 to 93/07).[1,9]
of the produced b-lactam 2 (for example, by hydroly-
Of course, the starting amide being a racemic com- sis) would be quite easy; besides, if a chiral benzyl-
amine is used in the synthesis of the starting amide 1
(X position), the resulting b-lactam 2 (due to a 4-exo-
tet cyclization) would incorporate this chiral part as a
protective group that could be removed (due to the
reactivity of the benzyl group) by many means: using
potassium persulfate,[12,13] lithium[14] or sodium in
liquid ammonia.[8] Moreover, chiral benzylamines are
readily available and relatively cheap, and their use in
Scheme 2. Electrochemically induced synthesis of b-lactams
by C-3 C-4 bond formation.
the synthesis of b-lactams has been suggested by
À
some authors.[13]
I have taken into account both possibilities (X and
pound, a pair of cis-enantiomers was obtained and, Y chiral, but not at the same time), not being able to
obviously, it was not possible to separate them by or- predict which of these two ways would lead to the
dinary means. Starting from these, however, encour- best asymmetric induction. At first, substrates 1a–e
aging results and as a logical continuation of the pre- (Scheme 3, R=H) have been taken into consideration
vious work, I wanted to further study the stereochem- to evaluate the aymmetric induction at the C-3 posi-
ical outcome of this electrochemically induced cycli- tion (Table 1).
zation of amides to b-lactams. As in the reaction re-
The cyanomethyl anion was generated by galvano-
ported in Scheme 2, we obtained a pair of cis- static electrochemical reduction of a solution of aceto-
enantiomers as main products, I thought to incorpo- nitrile, containing tetraethylammonium hexafluoro-
rate a chiral auxiliary or a chiral reagent into the phosphate as supporting electrolyte, at 08C[15] and
starting amide, attempting in this way to obtain a dia- under a nitrogen atmosphere. The electrolysis was
stereomeric mixture in which the two cis-b-lactams carried out in a two-compartment cell with a sintered
could be separated with ordinary means. This work glass/agar gel separator, equipped with platinum
was divided in two parts, the first concerning the dia- anode and cathode, and was stopped after two equiva-
stereoselectivity of this cyclization related to the con- lents of electricity (2 Faradays per mol of amide)[16]
trol of C-4 configuration (substrates 1a–e, Table 1), had passed through the solution. Amide 1 (as a sum
the second concerning the diastereoselectivity related of two rotamers) was then added and the mixture
to the configuration of both C-3 and C-4 atoms (sub- stirred at 08C for 3 h. Usual work-up gave the corre-
strates 1f–j, Table 2) with particular reference to the sponding b-lactam 2. The results are reported in
cis/trans-diastereoselection.
Table 1 and are consistent with a reaction pathway in-
volving deprotonation of the acidic methylene in the
a-position to the nitrogen atom, followed by internal
halide displacement with a 4-exo-tet cyclization.
Results and Discussion
As can be seen from Table 1, the yields in b-lactams
Considering the structure of the amide used in the 2 are generally good (69 to 79%), irrespective of the
previous paper,[1] I recognized two positions with position of the chiral part of the molecule. Moreover,
which
a
chiral reagent could be linked (see except for the case of amide 1b, the two diastereo-
Scheme 3): the nitrogen atom (X is in this case a meric b-lactams are obtained with a fairly good dia-
chiral group) and the carbonyl group in the b-position stereoselection [from 35/65 to 23/77, this last result
to the nitrogen atom (Y is a chiral auxiliary).
being achieved with the more hindered 1-(2-naphthyl)-
The choice of the chiral reagent or auxiliary has ethylamine]. Unfortunately, any attempt to separate
been done in view of an easy removal. In fact, when a the diastereoisomers with ordinary means failed. In
2178
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2177 – 2181

Synthesis of b-Lactams by 4-exo-tet Cyclization Process
FULL PAPERS
Table 1. Electrochemically promoted diastereoselective synthesis of monosubstituted b-lactams.
Entry
Reagent
Product[a]
Yield[b]
79%
dr[c]
1
1a
1b
2a
2b
33/67
2
71%
50/50
3
4
1c
2c
71%
69%
26/74
35/65
1d
2d
5
1e
2e
76%
23/77
[a]
As sum of two epimers at the asterisked position.
Yields in isolated product, as sum of two epimers at the asterisked position.
[b]
[c]
Diastereomeric ratio between the two diastereoisomeric products. The first number of the ratio refers to the diastereoiso-
mer in which the proton marked with the asterisk resonates at a lower field with respect to the corresponding proton of
1
the other diastereoisomer in the H NMR spectrum.
the case of amide 1b (entry 2), the bulky isopropyl of our electrogenerated cyanomethyl anion as a
group of the chiral auxiliary is perhaps too far from highly reactive base.
the reaction centre and it fails to influence the stereo-
chemical output of the reaction (attaining a 50/50 dia- eoselection with these substrates, we have introduced
stereomeric ratio). a methyl group (R=CH3 in Scheme 3) to try to repro-
Having obtained good results in yields and diaster-
b-Lactam 2c has been previously synthesized by duce the good cis-diastereoselectivity obtained in our
Cordero and co-workers[17] in 61% yield as an insepa- previous work with racemic amides (Scheme 2). The
rable 1:1 mixture of isomers starting from a 1:1 mix- same chiral reagents or auxiliaries reported in Table 1
ture of chiral spiro[cyclopropane-1,5’-isoxazolidine] have been used (in this case, starting amides 1 are
isomers treated with p-TsOH in CH3CN at 508C (we used as mixtures of diastereoisomers and rotamers)
have synthesized this b-lactam in 71% yield with a and the results are reported in Table 2.
diastereomeric ratio of 26/74).
The two main features of this Table are the high
A similar reaction was carried out by Gonzµlez- yields (except for amide 1f, entry 1, in which the
MuÇiz and co-workers[18] starting from a halogenated chiral auxiliary is Oppolzerꢁs camphor sultam[19]), be-
amide 1 (Scheme 3, in which R=H, X=p-methoxy- tween 78 and 83%, obtained under very mild condi-
benzyl, Y=OCH3 and Cl instead of Br) in MeCN as tions, and the complete cis-diastereoselection of the
solvent and using Cs2CO3 as base; in this case the re- produced b-lactams 2f–j. In fact, in no case the trans-
action took 10 days to obtain the corresponding b- isomers of 2f–j have been detected.
lactam 2 in 54% yield. This fact confirms the validity
Adv. Synth. Catal. 2007, 349, 2177 – 2181
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2179

Marta Feroci
FULL PAPERS
Table 2. Electrochemically promoted diastereoselective synthesis of cis-disubstituted b-lactams.
Entry
Reagent
Product[a]
Yield[b]
dr[c]
1
1f
1g
2f
2g
52%
79%
50/50
2
49/51
3
4
1h
1i
2h
2i
78%
80%
46/54
47/53
5
1j
2j
83%
42/58
[a]
As sum of two cis-diastereoisomers.
[b]
Yields in isolated product, as sum of two cis-diastereoisomers. The cis-configuration was established on the basis of the
1
coupling constant (in the H NMR spectrum) of the proton marked with the asterisk, as reported in the literature. See,
for example, ref.[6]
[c]
Diastereomeric ratio between the two cis-diastereoisomeric products. The first number of the ratio refers to the diaste-
reoisomer in which the proton marked with the asterisk resonates at a lower field with respect to the corresponding
proton of the other diastereoisomer in the 1H NMR spectrum.
The two isomers of each cis-b-lactam have been ob- Conclusions
tained in a nearly equimolar amount (from 42/58 to
50/50); nevertheless, in the cases of compounds 2f In conclusion, I have described an easy and conven-
and 2h–j they could be separated and one of the two ient electrochemical strategy to promote the cis-ste-
diastereoisomers isolated in a pure form (see Support- reoselective synthesis of chiral b-lactams in high
ing Information, again l-menthol – entry 2 – seems an yields by 4-exo-tet cyclization of suitable bromo
unsuitable chiral auxiliary for this kind of reaction).
amides bearing a chiral auxiliary or chiral protective
b-Lactam 2h has been previously synthesized by group. The base (cyanomethyl anion) is obtained in a
Shibasaki and co-workers[20] in 86% yield using an stoichiometric amount by simple galvanostatic elec-
À
N C-2 pathway, starting from a cis-b-amino-thiol trolysis of a solution of acetonitrile/tetraethylammoni-
ester, in more drastic conditions and with a transition um hexafluorophosphate. This is a very reactive base,
metal reagent [Cu(I)OTf, CaCO3, toluene, reflux]; in as its counterion is a tetraalkylammonium cation, so it
this case (contrary to our method) the stereochemical can be considered a naked ion. This very efficient
outcome of the reaction was innate in the cis nature electrochemical synthesis leads to a complete diaster-
of the reagent, as the 3 and 4 positions are not eoselection in the products, as only the cis-isomers
touched during the reaction.
are obtained. In most cases, the most abundant cis-
lactam could be isolated in pure form.
2180
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2177 – 2181

Synthesis of b-Lactams by 4-exo-tet Cyclization Process
FULL PAPERS
[5] A. D’Annibale, D. Nanni, C. Trogolo, F. Umani, Org.
Lett. 2000, 2, 401–402 and references cited therein.
[6] T. Kawabata, T. Minami, T. Hiyama, J. Org. Chem.
1992, 57, 1864–1873.
[7] O. Kitagawa, N. Kikuchi, T. Hanano, K. Aoki, T. Yama-
zaki, M. Okada, T. Taguchi, J. Org. Chem. 1995, 60,
7161–7165.
[8] M. Feroci, M. Orsini, L. Palombi, L. Rossi, A. Inesi,
Electrochim. Acta 2005, 50, 2029–2036.
[9] M. Feroci, M. Orsini, L. Rossi, G. Sotgiu, A. Inesi,
ExperimentalSection
General Procedure for Electrochemically Promoted
Diastereoselective Synthesis of b-Lactams
A
constant current electrolysis was carried out (I=
75 mAcmÀ2) in an MeCN-Et4NPF6 0.1 mol dmÀ3 (20 cm3)
solution as catholyte (anolyte: 5 cm3), at 08C, under an
argon atmosphere, in a divided glass cell separated through
a porous glass plug filled up with a layer of agar gel (i.e.,
methyl cellulose 0.5% vol dissolved in DMF-Et4NClO4 1.0
mol dmÀ3); Pt spirals (apparent areas 0.8 cm2) were used
both as cathode and anode. After 193 C were passed, the
current was switched off and amide 1 (1 mmol) was added
to the catholyte and the solution was allowed to stand under
stirring at 08C for 3 h. The solvent was then evaporated
under reduced pressure and the residue extracted with di-
ethyl ether (3”30 cm3). The extracts were analyzed by thin
layer chromatography, GC-MS and 1H NMR; all products
were purified by flash chromatography, using n-hexane-ethyl
acetate 95:5 to 7:3 as eluent. The two b-lactams produced
are named b1 and b2, b1 being the first isomer reported in
the dr in Table 1 and Table 2, and b2 the second one.
Electrochim. Acta 2006, 51, 5540–5547.
[10] J. H. P. Utley, M. F. Nielsen, in: Organic Electrochemis-
try, 4th edn., (Eds.: H. Lund, O. Hammerich), Marcel
Dekker, New York, 2001, pp 1227–1257.
[11] M. Feroci, M. Orsini, G. Sotgiu, L. Rossi, A. Inesi, J.
Org. Chem. 2005, 70, 7795–7798 and references cited
therein.
[12] I. Ojima, S. Suga, R. Abe, Chem. Lett. 1980, 853–856.
[13] a) R. C. Thomas, Tetrahedron Lett. 1989, 30, 5239–
5242; b) J. Aszodi, A. Bonnet, G. Teutsch, Tetrahedron
1990, 46, 1579–1586; c) S. G. Davies, D. R. Fenwick,
Chem. Commun. 1997, 565–566.
[14] D. A. Evans, E. B. Sjogren, Tetrahedron Lett. 1985, 26,
3783–3786 and 3787–3790.
[15] Electrolyses carried out at
a lower temperature
Acknowledgements
(À208C) led to lower chemical yields in b-lactams, with
no improvement in the dr.
The author thanks PRIN 2006, MIUR and CNR (Rome) for
financial support.
[16] An increase in the number of Faradays per mol had no
effect, see ref.[1]
[17] F. M. Cordero, M. Salvati, F. Pisaneschi, A. Brandi,
Eur. J. Org. Chem. 2004, 2205–2213.
[18] G. Gerona-Navarro, M. A. Bonache, R. Herranz, M. T.
Garcꢂa-Lꢃpez, R. Gonzµlez-MuÇiz, J. Org. Chem. 2001,
66, 3538–3547.
[19] The low yield in cis-b-lactam 2f can be due both to the
high steric hindrance of this chiral auxiliary and to the
fact that under these experimental conditions part of
the auxiliary is removed (either from starting amide 1f
or from product 2f). In fact, from the cathodic solution
Oppolzerꢁs camphor sultam has been isolated.
[20] N. Miyachi, F. Kanda, M. Shibasaki, J. Org. Chem.
1989, 54, 3511–3513.
References
[1] Part 1: M. Feroci, J. Lessard, M. Orsini, A. Inesi, Tetra-
hedron Lett. 2005, 46, 8517–8519.
[2] a) W. Dürckheimer, J. Blumbach, R. Lattrell, K. H.
Scheunemann, Angew. Chem. Int. Ed. Engl. 1985, 24,
180–202; b) G. S. Singh, Tetrahedron 2003, 59, 7631–
7649.
[3] A. R. A. S. Deshmukh, B. M. Bhawal, D. Krishnaswa-
my, V. V. Govande, B. A. Shinkre, A. Jayanthi, Curr.
Med. Chem. 2004, 11, 1889–1920.
[4] a) D. A. Hart, D.-C. Ha, Chem. Rev. 1989, 89, 1447–
1465; b) C. Palomo, J. M. Aizpurua, I. Ganboa, M.
Oiarbide, Eur. J. Org. Chem. 1999, 3223–3235.
Adv. Synth. Catal. 2007, 349, 2177 – 2181
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2181