Preparation of β- and γ-lactams from carbamoyl radicals derived from oxime oxalate amides

Article abstract of DOI:10.1039/b315223e

A general synthetic route to oxime oxalate amides was developed and applied to the preparation of molecules incorporating N-benzyl-N-alkenyl amides linked with acetone oxime or benzaldoxime units. In addition, 2-substituted-thiazolidine-4-carboxylic acid methyl ester amides of oxalyl benzaldoxime were also prepared. It was shown by EPR spectroscopy that the oxalyl benzaldoxime amides dissociated to produce benziminyl and carbamoyl (aminoacyl) radicals when photolysed with 4-methoxyacetophenone as a photosensitizer. Carbamoyl radicals derived from N-alk-3-enyl oxime oxalate amides underwent ring closure to afford pyrrolidin-2-ones. The analogous N-alk-2-enyl precursors afforded azetidin-2-ones. Reactions of the cyclohexenyl and cinnamyl oxime oxalate amides afforded a bicyclic β-lactam and a 3-benzyl-substituted β-lactam respectively. Interestingly, both products were isolated as hydroxylated compounds. A thiazolidine-derived oxime oxalate amide containing an isobutenyl side chain also dissociated with production of the corresponding thiazolidinyl-carbamoyl radical, as shown by EPR spectroscopy. GC-MS evidence indicated that this radical cyclised to afford some of the corresponding penicillin derivative.

Full text of DOI:10.1039/b315223e

View Article Online / Journal Homepage / Table of Contents for this issue
Preparation of ꢀ- and ꢁ-lactams from carbamoyl radicals derived
from oxime oxalate amides†
Eoin M. Scanlan, Alexandra M. Z. Slawin and John C. Walton*
University of St. Andrews, School of Chemistry, St. Andrews, Fife, UK KY16 9ST.
E-mail: jcw@st-and.ac.uk; Fax: 01334 463808; Tel: 01334 463864
Received 25th November 2003, Accepted 19th December 2003
First published as an Advance Article on the web 3rd February 2004
A general synthetic route to oxime oxalate amides was developed and applied to the preparation of molecules
incorporating N-benzyl-N-alkenyl amides linked with acetone oxime or benzaldoxime units. In addition,
2-substituted-thiazolidine-4-carboxylic acid methyl ester amides of oxalyl benzaldoxime were also prepared. It was
shown by EPR spectroscopy that the oxalyl benzaldoxime amides dissociated to produce benziminyl and carbamoyl
(aminoacyl) radicals when photolysed with 4-methoxyacetophenone as a photosensitizer. Carbamoyl radicals derived
from N-alk-3-enyl oxime oxalate amides underwent ring closure to afford pyrrolidin-2-ones. The analogous N-alk-2-
enyl precursors afforded azetidin-2-ones. Reactions of the cyclohexenyl and cinnamyl oxime oxalate amides afforded
a bicyclic β-lactam and a 3-benzyl-substituted β-lactam respectively. Interestingly, both products were isolated as
hydroxylated compounds. A thiazolidine-derived oxime oxalate amide containing an isobutenyl side chain also
dissociated with production of the corresponding thiazolidinyl-carbamoyl radical, as shown by EPR spectroscopy.
GC-MS evidence indicated that this radical cyclised to afford some of the corresponding penicillin derivative
expected to extrude CO2 very easily. Thus 5 appeared promising
Introduction
as a precursor type for aminoacyl radicals (carbamoyl radicals).
The advantages of free-radical based synthetic methods have
In this paper we report the synthesis of a range of oxime oxal-
been ably described in several recent books1 and reviews.2 Many
ate amides, a study of radical generation by EPR spectroscopy
free radical chain syntheses rely on organotin compounds but,
and some applications of these compounds to the preparation
because of their toxicity, alternative reagents are very desirable.3
of β- and γ-lactams. Part of this research was published in a
Oxime esters [R1R2C᎐NOC(O)R] contain weak N–O bonds
preliminary communication.6
᎐
and were shown to release C-centred radicals under certain
circumstances. Hasebe and co-workers used benzophenone
Results and discussion
oxime esters as photochemical sources of alkyl radicals in
preparations of alkyl aromatics,4a,b alkyl chlorides4c,d and
alkanes.4c,d In our group we used EPR spectroscopy to show
that the efficiency of radical release from oxime esters could be
improved by using p-methoxyacetophenone (MAP) as a photo-
sensitiser and by incorporating methoxy substituents into an
aromatic group R1. In this way good yields of cyclized products
(e.g. 3) were obtained from suitably unsaturated di- and
tri-methoxy-substituted aldoxime esters such as 1 (Scheme 1).5
Preparation and characterisation of oxime oxalate amides
In the only previous synthesis of an oxime oxalate amide,
Jochims et al. treated acetone O-(chlorooxalyl)oxime 4 (R1᎐R2᎐
᎐
᎐
Me) with 2 eq. of aniline to afford the corresponding amide
(5a).7
Not only did this method ‘waste’ one eq. of amine, but we
also found that it failed for other amines. We attempted to make
mono-amides of oxalyl chloride with N-benzyl-N-n-butyl-
amine, for subsequent condensation with oximes, but without
success. However, when 4 was treated with 1 eq. of a primary or
secondary amine, in the presence of 1 eq. of pyridine at 0 ЊC,
good yields of the corresponding oxime oxalate amides were
obtained. As Scheme 2 shows, the method worked well for benz-
aldehyde oxalyl oximes as well as the acetone analogues. The
oxime oxalate amides derived from benzaldehyde, 5d–h, are
Scheme 1 Cyclization of a C-centred radical derived from an oxime
ester.
1
2
ؒ
Interference from the iminyl radicals (R R C᎐N ) produced
᎐
was minimal. They simply abstracted a hydrogen atom from the
solvent and the resulting imine hydrolysed during work-up to
afford the corresponding aldehyde or ketone. It was evident that
there was wide scope for radical generation from this class of
compounds. These results suggested that other oxime deriv-
atives could be useful as ‘clean’ sources of various radical types.
Oxime oxalate amides 5 (OOAs) retain the weak N–O bond of
1. The acyloxyl radical generated on fission of this bond was
† Electronic supplementary information (ESI) available: Gaussian98
ARC files for model carbamoyl radicals. See http://www.rsc.org/
suppdata/ob/b3/b315223e/
Scheme 2 Preparation of oxime oxalate amides.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
716

View Article Online
believed to be the syn isomers (i.e. E-isomers). The synthesis
of benzaldoxime was specific to the syn isomer.8 Reported
chemical shifts of the oximyl hydrogen of various substituted
benzaldoximes were dramatically different for the syn and anti
isomers.9 All the syn isomers had protons with δ > 8, while the
anti isomers (Z-isomers) had δ < 7.5. The δ values for 5d–h were
all in the range 8.38–8.50 ppm, i.e. as expected for syn isomers.
This was confirmed for 7b by an X-ray structure determination
(see below). However, the 1H NMR spectra of 5e,f showed two
resonances for the oximyl and other H-atoms. This probably
indicated the presence of two isomers due to restricted rotation
about the N–C(O) amide bond. The isomer proportions were
close to 1 : 1 in each case.
We also investigated if the radical methodology developed
for the preparation of simple bicyclic lactams could be applied
to the preparation of penicillins. The general synthetic strategy
envisaged preparation of penicillins via 4-exo cyclisations
of thiazolidine-containing OOAs 7. Thus, thiazolidines with
double bonds suitably placed for ring closure, e.g. 6b, were
required. Only one thiazolidine of this type had previously been
prepared.10 The synthetic strategy we employed was based on
the well known condensation reaction of penicillamine esters
with appropriate aldehydes.11 This condensation reaction had
never been applied to the preparation of a vinylic thiazolidine
and it was anticipated that there might be a mixture of products
arising from the competing Michael addition, whereby nucleo-
philic attack would occur at the “softer” Michael centre of the
double bond rather than at the desired aldehyde carbon. To bias
the equilibrium towards the desired thiazolidine condensation
product, the Michael addition site was hindered by using the
dimethyl-substituted double bond of 3-methylbut-2-enal. In
this way functionalised thiazolidines 6 were prepared from con-
densation of -cysteine, ,-penicillamine and -penicillamine
with aldehydes (Scheme 3). Because the yields of 6b were low
(25–28%) we tried to circumvent this poor step, via conden-
sations with glyoxal, and protected glyoxal. The thiazolidine
could then have been elaborated via a Wittig reaction. However,
condensation of penicillamine with glyoxal failed and, although
condensation with OCHCH(OMe)2 succeeded, deprotection
failed.
Fig. 1 X-Ray structure of oxime oxalate amide 7b in the crystal.
EPR spectroscopic study of radical generation and cyclisation
Photolyses (500 W Hg arc, unfiltered) of deoxygenated tert-
butylbenzene solutions of acetone ketoxime oxalate ester
amides 5a–c in the resonant cavity of a 9 GHz EPR spectro-
meter gave rise to spectra of the propan-2-iminyl radical
ؒ
(Me C᎐N ) with EPR parameters identical to those reported in
᎐
2
the literature.13
Addition of one or more eq. of 4-methoxyacetophenone
(MAP) as photosensitiser14 to samples prior to degassing,
dramatically improved the spectral intensity. No signals from
the corresponding carbamoyl radicals were observed for any
of these acetone–oxime derived precursors. However, MAP
photosensitised reactions of the benzaldoxime oxalate ester
amides 5c–h and 7a,b gave spectra containing signals from the
ؒ
iminyl radical PhCH᎐N [g = 2.0034, a(N) = 9.9, a(1H) = 79.9 G
᎐
at 220 K]15 accompanied by 3-line spectra having a(N ) values
of 22 1 G (Table 1 and Fig. 2). A few archetype carbamoyl
radicals have previously been characterised by EPR spectro-
scopy,16,17 and comparison with this data supported identifi-
cation of our 3-line spectra as carbamoyl radicals 10. Similarly,
DFT computations18 of the hyperfine splittings (hfs) of several
model radicals (UB3LYP with a 6-311ϩG(d,p) basis set, Table
1) gave N- and H-atom hfs satisfyingly close to the experi-
mentally measured data. These aminoacyl radicals are all
capable of existing as E- and Z-isomers. Separate spectra
for the two forms were not, however, observed in any case.
Individual components of the aminoacyl N-triplets were often
rather broad (∆Hpp ∼ 2 G) and this might be a consequence of
overlap of E- and Z-species with similar a(N ) values. In the
case of the thiazolidinyl containing radicals a small doublet hfs,
probably from one of the thiazolidine ring H-atoms, was resol-
ved (Fig. 2). Judging by the EPR spectra, 10 are σ-radicals with
high barriers to rotation about their N–C(O) amide bonds.
The results indicated that oxime ester amides 5 underwent
homolysis of their weak N–O bonds on direct and photo-
sensitised UV photolysis to afford mixtures of iminyl and acyl-
oxyl radicals 9. The latter rapidly lost CO2 to release the corre-
sponding carbamoyl radicals 10 (Scheme 4). The ratio of the
concentrations of the iminyl and aminoacyl radicals depended
on temperature and on the nature of R4. Scission of a N–O
bond in 5 should initially produce equal amounts of an iminyl
and an acyloxyl radical 9 (Scheme 4). However, the experi-
mental ratio of aminoacyl to iminyl radicals [10]/[Im] (Table 1)
Scheme 3 Preparation of oxime oxalate amides from thiazolidines.
Secondary amines 6a,b were converted to the corresponding
oxime oxalate amides 7 in high yields on treatment with 4 in the
presence of pyridine. An analogous oxazolidine 8 was also pre-
pared, but this failed to give an oxime oxalate amide; probably
because of steric hindrance from the phenyl group adjacent to
the amine group.
The oxime oxalate amide derived from ,-penicillamine
i.e. 7b was a crystalline solid and hence its structure was
obtained by X-ray diffraction (Fig. 1). As might be expected,
the two oxalyl carbonyl groups were trans to one another. As a
consequence, the molecule exists in an all-trans extended con-
formation and, except for the substituents in the thiazolidine
ring, is almost planar. The N–O bond length of 1.453 Å
is long for an oxime, the usual range being12 1.38–1.43 Å.
This supported our premise that N–O homolysis would be
comparatively facile.
Scheme 4 Photodissociation of oxime oxalate amides.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
717

View Article Online
Table 1 EPR parameters for carbamoyl radicals at 220 K in PhBu-t solutiona
Carbamoyl radical
g-factor/theoretical method
hfs/G
[10]/[Im]b
10e
10h
10f
2.0018
2.0017
2.0016
a(N) = 22.3
0.8
a(N) = 21.7
a(1H) = 0.8
0.8c
0.7d
a(N) = 21.2
a(N) = 21.9
10g
2.0017
2.0016
0.2
1.0
a(N) = 21.5
a(1H) = 1.5
16
2.0018
a(N) = 21.0
a(1H) = 1.6
1.2
ؒ
ؒ
(O)C NMe2
UB3LYP 6-311ϩG(d,p)
UB3LYP 6-311ϩG(d,p)
a(N) = 24.2
a(N) = 26.5
a(1H) = 24.0
a(2H) = Ϫ0.7
a(N) = 23.4
a(3H) = Ϫ0.7
a(2H) = Ϫ0.7
(O)C NHCH2Ph
ؒ
(O)C NMeCH2Ph
UB3LYP 6-311ϩG(d,p)
a
In each case the iminyl radical PhCH᎐N (Im) with g = 2.0034, a(N) = 9.9, a(1H) = 79.9 G at 220 K was also detected. b Ratio of the concentrations
ؒ
᎐
of the carbamoyl and iminyl radicals determined by double integration of suitable EPR lines. c The ring closed radical 11h [g = 2.0025, a(2H) = 22.3,
a(1H) = 31.6 G at 220 K] was also detected. d The bicyclic radical 11f was also detected and is included with 10 in the ratio.
will depend on the completeness of decarboxylation of 9 as well
as the radical termination efficiencies. For most of the oxime
oxalate amides the observed ratio was within a factor of two or
three of unity. Decarboxylations of 9 were, therefore, rapid and
the observed differences can be attributed to minor variations
in the rates of their termination reactions.
2-oxopyrrolidinylmethyl radicals 11h or 4-exo-cyclisations to
produce 2-oxoazetidinylalkyl radicals 11f. noteworthy
A
spectrum containing three radicals, the iminyl, aminoacyl 10h
and the ring-closed 2-oxopyrrolidinylmethyl radical 11h was
described previously.6 The EPR spectra from 5e and 5g showed
minor signals that might have been due to cyclised radicals, in
addition to those of the iminyl and aminoacyl radicals, but they
were not intense enough for definitive analysis.
Previous spectroscopic work showed that oxime esters of
type 1 were susceptible to addition by C- and O-centred radicals
5
ؒ
to produce alkoxyaminyl radicals PhCH(X)N OC(O)R.
When 5d, containing benzyl and H-atom substituents on the
amide N-atom, was photolysed with MAP at T > 320 K a
second spectrum accompanied that of the iminyl radical. The
EPR parameters of this species [g = 2.0046, a(N) = 14.6,
a(1H) = 19.7 G at 320 K] were very similar to those reported
for the oxyaminyl radicals referred to above and it is probable
that this spectrum belongs to an adduct radical of similar
type. Oxyaminyl adducts were not detected for any other
oxime oxalate amides, probably because other reaction path-
ways supervened.
Preparations of ꢀ- and ꢁ-lactams
Preparative scale photolyses were carried out for oxime oxalate
amides 5e–h. A short series of experiments with 5h established
that the γ-lactam, N-benzyl-3-methylpyrrolidin-2-one (12h),
could be obtained in >80% yield from photolyses of dilute
solutions in toluene with a three-fold excess of MAP. Similar
conditions were used for the other substrates. GC-MS analysis
showed that the N-formyl compound 13, resulting from direct
H-atom transfer to the aminoacyl radical, was negligible (<1%)
(Scheme 5). β-Lactam formation was expected to be more
difficult because of the disfavoured 4-exo-ring closures [C4x].
Photosensitised reaction of the allyl OOA-precursor 5e yielded
40% of N-benzyl-3-methylazetidin-2-one (12e). Although this is
a modest yield it compares favourably with previous radical
EPR spectra from photosensitised reactions of 5e–h and 7a,b
containing benzyl and unsaturated substituents, or thiazol-
idine-2-alkenyl substituents, displayed iminyl and aminoacyl
radicals (see Fig. 2 and Table 1). These aminoacyl radicals
10 can potentially undergo 5-exo-cyclisations to produce
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
718

View Article Online
Fig. 2 EPR spectrum obtained on photolysis of 7b and MAP in PhBu-t at 220 K. The carbamoyl radical appears as a triplet of doublets in the
ؒ
central region. The two N-triplets from PhCH᎐N appear in the wings.
᎐
Scheme 5 Preparation of β- and γ-lactams.
routes to azetidin-2-ones involving aminoacyl radicals gener-
ated from Co-salophens19 or amidocyclohexadienes.20
Reactions of the cyclohexenyl (5f ) and cinnamyl (5g)
precursors afforded the bicyclic β-lactam 12f and the 3-benzyl-
substituted β-lactam 12g respectively. In both cases the ring-
closed radicals were more stabilised (one was secondary and the
other benzylic) than primary radicals 11e,h and this favoured
cyclisation; as was attested by the good yields of β-lactams.
Interestingly, however, for both 5f and 5g the products isolated
were the hydroxylated compounds 12f (5 : 1 mixture of dia-
stereomers) and 12g (3 : 1 mixture of enantiomers). A plausible
mechanistic interpretation is outlined in Scheme 6. Because the
cyclised radicals 11f and 11g are thermodynamically stabilised
they couple with dissolved dioxygen more rapidly than they
abstract hydrogen from the solvent. The resulting peroxyl
radicals, (14f,g) couple in pairs, extrude dioxygen, and generate
alkoxyl radicals 15f,g that are very good H-atom abstractors
and hence afford the hydroxylated products. Analogous
Scheme 6 Formation of hydroxylated β-lactams.
We also studied the thiazolidine-based OOAs as radical pre-
cursors. The photolysis of oxime oxalate amides 7a,b were first
investigated using EPR spectroscopy. Fig. 2 shows the EPR
spectrum obtained at 220 K for MAP sensitised photolysis of
7b. The EPR parameters of the carbamoyl radicals 10 obtained
from 7a and 7b are in Table 1. The ratios of iminyl to carbamoyl
were close to 1.0, as expected. The ring closed azetidinylalkyl
radical 17 was not, however, observed in the accessible temper-
ature range.
A product study was also carried out in order to determine if
a simple penicillin could be prepared via oxime oxalate amide
methodology (Scheme 7). OOA 7b in the presence of MAP (3 M
eq.) was photolysed with light from a 400 W UV lamp at 80 ЊC
in toluene for 8 h. GC-MS analysis showed a peak having the
correct molecular ion (257) for the cyclised penicillin derivative
18. This was, in fact, the sole product derived from 7b because
all the other components in the chromatogram stemmed from
the solvent or from the MAP. The fragmentation pattern was
consistant with that of the expected penicillin. The formula
weight of the formamide that would be obtained from H-trans-
fer to carbamoyl radical 16 is also 257. However, the MS
did not show the (M Ϫ H)ϩ ion expected for a formamide.
hydroxylation sequences are well known.21
ؒ
Iminyl radicals (PhC(H)᎐N ) were produced as co-inter-
᎐
mediates with the aminoacyls 10. The major product derived
from this species was benzaldehyde. It is probable that the
iminyl radicals abstract hydrogen from the solvent to afford the
corresponding imine which was hydrolysed to benzaldehyde
during work-up. In a few instances traces (<1%) of the azine
(PhCH᎐NN᎐CHPh), formed by combination of two iminyls,
᎐
᎐
was detected by GC-MS.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
719

View Article Online
VG Autospec spectrometer. Ultra Violet irradiation was carried
out in quartz apparatus using a 400 W medium pressure Hg
lamp.
Ether refers to diethyl ether. THF was distilled under nitro-
gen from sodium benzophenone ketyl prior to use. Where dry
DCM was used, it was distilled over CaH2. DMSO was dried,
distilled from CaH2, and stored over 4 Å molecular sieves.
Other organic compounds were used as received. TLC was
carried out using either Polygram silica plates (0.2 mm with
254 nm fluorescent dye) or Fluka alumina plates (0.2 mm with
254 nm fluorescent dye). Fisher silica gel and Fisher neutral
alumina were used for column chromatography. Nitrogen gas
was dried (NaOH, CaCl2, 4 Å molecular sieves) prior to use.
N-Benzylformamide, N-benzyl-3-buten-1-ylamine,23 N-benzyl-
prop-2-enylamine, N-cinnamylbenzylamine, N-benzyl-2-cyclo-
hexenylamine24 and penicillamine methyl ester hydrochloride25
were made by literature methods.
EPR spectra were obtained with a Bruker EMX 10/12
spectrometer operating at 9.5 GHz with 100 kHz modulation.
Samples of the substrate (0.3 to 40 mg) and MAP (1 eq.), in
tert-butylbenzene (up to 0.5 cm3), in 4 mm od quartz tubes,
were de-aerated by bubbling nitrogen for 20 min, and photo-
lysed in the resonant cavity by light from a 500 W super
pressure mercury arc lamp. For reactions performed in cyclo-
propane, the solution was degassed on a vacuum line using the
freeze–pump–thaw technique, and the tube was flame sealed. In
all cases where spectra were obtained, hfs were assigned with
the aid of computer simulations using the Bruker Simfonia
software package. For quantitative measurements signals were
double integrated using the Bruker WinEPR software.
Scheme
7 Photosensitised reactions of thiazolidine-containing
precursors.
Furthermore, negligible amounts of formamides were obtained
from other OOAs. Several attempts were made to isolate the
penicillin by column chromatography and by preparative
HPLC but without success. Column chromatography gave only
MAP and benzaldehyde and HPLC produced a large collection
of minor unidentified components. We conclude that some 18
was formed but that it degraded during photolysis and work-
up.
Conclusions
Benzaldoxime oxalate amides and acetone oxime oxalate
amides can be prepared in high yields from both primary and
secondary amines. The former dissociate efficiently on sensi-
tised photolysis to generate carbamoyl radicals, accompanied
by iminyl radicals. In toluene solution the iminyl radicals
abstract H-atoms to afford, after hydrolysis, benzaldehyde and
hence do not interfere. Carbamoyl radicals with N-alkenyl sub-
stituents ring-closed efficiently to afford substituted pyrrolidin-
2-one or azetidin-2-one derivatives, depending on the position
of the double bond. An interesting feature of the process was
that when the cyclised azetidinylalkyl radicals were secondary
or benzylic [11f, 11g] the corresponding β-lactams contained
hydroxyl groups at the site of the original radical centre. We
explained this hydroxylation by a mechanism involving peroxyl
and alkoxyl radicals (Scheme 6). Good yields of the α-hydroxy-
β-lactams were obtained. Of course, such hydroxy substituents
are an advantage for further functional group elaboration.
The 7-benzyl-2-hydroxy-7-azabicyclo[4.2.0]octan-2-one (12f )
had previously been obtained, as a mixture with the 6- and
7-hydroxylated analogues, by biohydroxylation of the parent
azabicycle with Beauveria sulfurescens.22
2-Substituted-thiazolidine-4-carboxylic acid methyl ester
amides of benzaldehyde oxalyl oxime were also prepared. It was
shown by EPR spectroscopy that they also dissociated to pro-
duce carbamoyl and iminyl radicals when photolysed with
MAP. A thiazolidine-derived oxime oxalate amide containing
an isobutenyl side chain released the corresponding thiazol-
idinyl-carbamoyl radical, (confirmed by EPR spectroscopy).
GC-MS evidence indicated that this radical cyclised to afford
some of the corresponding penicillin derivative 18. However,
none of this product could be isolated and it seems probable
that the cyclisation step was inhibited by the 5-member thiazol-
idine ring and that the sulfur-containing bicyclic β-lactam
degraded during photolysis and work-up.
Acetone O-(chlorooxalyl)oxime7 4a
A solution of acetone oxime (2.92 g; 40 mmol) in diethyl ether
(10 cm3) was added dropwise under stirring to a cold (Ϫ20 ЊC)
solution of oxalyl chloride (7.60 g; 60 mmol) in diethyl ether
(10 cm3). The mixture was stirred at Ϫ20 ЊC for 30 min followed
by 10 min at 23 ЊC (after which time the initial ppt had com-
pletely dissolved). The solvent was removed at 23 ЊC/13 torr to
yield a colourless volatile oil, which started to turn yellow on
standing at rt (6.07 g; 94%); δH (2 isomers) 2.11, 2.12 (3H, s,
CH3).
Acetone O-(anilinooxalyl)oxime7 5a
A solution of aniline (1.86 g; 20 mmol) in CHCl3 was added
dropwise under stirring to a cold (Ϫ40 ЊC) solution of 4a
(1.62 g, 10 mmol) in CHCl3 (20 cm3). After stirring for 1 h at
23 ЊC, anilinium chloride was filtered off. Evaporation of the
filtrate yielded a colourless powder which was purified via
recrystallisation from CHCl3 (25 cm3) at Ϫ20 ЊC (1.42 g; 65%,
lit.7 = 69%), mp = 175–178 ЊC; νmax(NaCl)/cmϪ1 1746, 1703
(C᎐O); δ 2.16, 2.21 (6H, s, CH ), 7.28–7.69 (5H, m, ArH), 8.95
᎐
H
3
(1H, br, N–H), δC 17.5, 21.9 (CH3), 136.3, 125.6, 129.2, 119.8
(ArC), 153.3, 158.8, 167.6 (C᎐N, C᎐O).
᎐
᎐
Acetone O-(benzylaminooxalyl)oxime 5b
A solution of benzylamine (2.14 g; 20 mmol) in CHCl3 (10 cm3)
was added dropwise under stirring to a cold (Ϫ40 ЊC) solution
of 4a (1.63 g; 10 mmol). After stirring for 1 h at rt the reaction
mixture was left standing overnight. The mixture was filtered,
and the filtrate evaporated to yield a colourless powder. The
product was then washed with hexane–diethyl ether and ethyl
acetate, and the solution was filtered to furnish 5b as a colour-
less powder (3.14 g; 67%); mp. 198–202 ЊC; δH 2.11, 2.12 (3H, s,
CH3), 4.55 (2H, d, Ar–CH2), 7.35 (5H, s, Ar–H), 7.52–7.68 (1H,
br, NH), 2.11, 2.14 (6H, s, CH3-dioxime); δC 17.9, 22.4 (CH3),
Experimental
1H NMR spectra were recorded at 300 MHz and 13C NMR
spectra at 75 MHz, in CDCl3 solutions with tetramethylsilane
as reference. Coupling constants are expressed in Hz. EI mass
spectra were obtained with 70 eV electron impact ionisation
and CI spectra were obtained with isobutane as target gas on a
44.4 (CH ), 158.7, 167.9 (C᎐N), (C᎐O), 137.1, 128.5, 128.4,
᎐
᎐
2
129.3 (ArC); CIMS m/z (%), 235 (MHϩ, 50), 180 (100); found:
235.1092, C12H15N2O3 requires 235.1086.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
720

View Article Online
Acetone O-(N-benzyl-N-butylaminooxalyl)oxime 5c
and N-benzyl-2-cyclohexenylamine (0.98 g; 5.3 mmol) in DCM
(5 cm3). The mixture was allowed to reach rt and then stirred at
rt for 3 h. After this time, pentane (ca. 10 cm3) was added in
order to promote formation of the pyridine hydrochloride ppt.
The solution was filtered and then evaporated to dryness. The
crude product was purified via flash column chromatography to
A solution of N-benzyl-N-butylamine (1.86 g; 11.3 mmol) was
added dropwise to a cold (Ϫ40 ЊC) solution of 4a (1.63 g;
10 mmol) in CHCl3 (20 cm3). After stirring for 1 h at Ϫ20 ЊC,
the solvent was removed at 0 ЊC to yield a colourless solid (1.57
g; 54%). The product was then recrystallised from dry CHCl3;
δH 0.96 (3H, t, J 7, CH3), 1.33 (2H, m, CH2), 1.55 (2H, m, CH2),
2.01 (6H, s, 2 CH3), 3.22 (2H, t J 7, NCH2), 4.49 (2H, s,
NCH2Ph), 7.23–7.64 (5H, m, ArH); δC 16.9 (CH3), 21.5 (2 ×
CH3). 19.7 (CH2), 29.7 (CH2), 47.3 (NCH2), 51.1 (NCH2Ph),
give a colourless oil (1.80 g; 95%); νmax(NaCl)/cmϪ1 1744 (C᎐O),
᎐
1612 (C᎐N); δH 1.56 (1H, m, CH), 1.63 (1H, m, CH), 1.74 (1H,
᎐
m, CH), 1.97 (3H, m, CH), 4.43 (1H, d, J 15, ArCH), 4.77 (1H,
d, J 15, ArCH), 5.50 (1H, m, ᎐CH), 5.94 (1H, m, ᎐CH), 7.24–
᎐
᎐
7.76 (10H, m, ArH), 8.49 (1H, s, HC᎐N); δ 21.7 (CH2), 24.7
᎐
C
127.9, 128.5, 127.4, 135.5 (ArC), 161.5, 165.5, 165.7 (C᎐O,
᎐
(CH2), 28.9 (CH2), 45.5 (ArCH2), 56.8 (CH), 126.9, 127.5,
127.8, 127.9, 128.9, 129.0, 129.4, 129.5, 132.6, 133.8, (ArC,
C᎐N); CIMS m/z (%), 291 (MHϩ 89), 164 (100).
᎐
᎐CH, ᎐CH), 157.9, 158.0, 158.9 (C᎐O, C᎐O, C᎐N); CIMS m/z
᎐
᎐
᎐
᎐
᎐
Benzaldehyde O-(chlorooxalyl)oxime 4d7
(relative intensity), 363 (MHϩ, 11%), 104 (100); found: MHϩ
363.1713, C22H23N2O3 requires MHϩ 363.1710.
A solution of benzaldoxime (1.21 g; 10 mmol) in diethyl ether
(10 cm3) was added dropwise to a cold (Ϫ40 ЊC) solution of
oxalyl chloride (1.90 g; 15 mmol) in diethyl ether (10 cm3). After
stirring for 1 h at Ϫ20 ЊC the solvent was evaporated at Ϫ10 ЊC/
13 torr to leave a colourless temperature sensitive powder which
was dried under vacuum for 2 h to remove residual diethyl
Benzaldehyde O-(N-cinnamyl-benzylaminooxalyl)oxime 5g
To a stirred solution of 4d (1.3 g; 6.0 mmol) in DCM (20 cm3)
was added dropwise a solution of pyridine (0.62 g; 6 mmol) in
DCM (5 cm3) followed by N-cinnamylbenzylamine (1.34 g;
6 mmol) in DCM (5 cm3). The solution was allowed to reach rt
and then stirred at rt for 3 h. After this time a small amount of
pentane (ca. 10 cm3) was added in order to promote formation
of the pyridine hydrochloride salt. The ppt was filtered off and
the filtrate collected. The filtrate was evaporated to dryness to
give the crude product as a colourless oil. The product
was purified via flash column chromatography to give 5g as a
ether. (2.01 g; 96%, lit.7 = 97%); νmax(NaCl)/cmϪ1 1791 (C᎐O);
᎐
δH 7.2–7.49 (5H, m, ArH), 8.55 (1H, s, CH); δC 128.2, 128.8,
129.2, 132.9 (Ar) 159.4, 160.4 (C᎐N, C᎐O); m/z (%) 211 (Mϩ,
᎐
᎐
20%).
Benzaldehyde O-(N-benzylaminooxalyl)oxime 5d
To a stirred solution of benzaldehyde O-(chlorooxalyl)oxime
(1.2 g; 6 mmol) in DCM (10 cm3) at 0 ЊC was added a solution
of pyridine (0.5 g; 6 mmol) in DCM (5 cm3), followed by a
solution of benzylamine (0.64 g; 6 mmol) in DCM (5 cm3). The
mixture was stirred at Ϫ10 ЊC for 10 min and then at rt for 3 h.
After this time the ppt which had formed was filtered off and
the solvent was removed. The product was filtered through a
pad of silica and was recrystallised from DCM–hexane at
Ϫ20 ЊC to give 5d as colourless platelets (1.38 g; 82%) mp 20–
22 ЊC; δH 4.57 (2H, d, J 6, ArCH2), 7.31–7.40 (5H, m, ArH),
7.43–7.55 (3H, m, ArH), 7.75–7.78 (2H, m, ArH), 8.62 (1H, s,
colourless oil (2.0 g; 85%); νmax(NaCl)/cmϪ1 1764 (C᎐O), 1665
᎐
(C᎐N); δH 3.92 (1H, d, J 6, N–CH2), 4.03 (1H, d, J 6, N–CH2),
᎐
4.46 (1H, s, ArCH2), 4.62 (1H, s, ArCH2), 5.98–6.14 (1H, m,
᎐CH), 6.41 (1H, d, J 15, ᎐CH), 7.16–7.60 (15H, m, ArH), 8.38
᎐
᎐
(1H, s); δC 45.7, 47.1, 49.6, 51.1 (CH2), 127.4, 128.3, 128.5,
128.7, 128.9, 129.0, 129.1, 129.3, 129.4, 129.5 (ArC), 135.3
(CH᎐CH), 158.1 (C᎐O), 161.9 (C᎐N).
᎐
᎐
᎐
Benzaldehyde O-(N-benzyl-3-buten-1-ylaminooxalyl)oxime 5h
To a stirred solution of 4d (1.43 g; 0.8 mmol) in DCM (15 cm3)
at 0 ЊC was added pyridine (0.54 g; 6.8 mmol) in DCM (8 cm3)
followed by N-benzyl-3-buten-1-ylamine (1.1 g; 6.8 mmol) in
DCM (8 cm3). The mixture was allowed to reach rt and then
stirred at rt for 2 h. After this time pentane (ca. 3 cm3) was
added to promote formation of the pyridine hydrochloride
ppt. The ppt was filtered off and the filtrate was evaporated
to dryness. The resulting oil was purified via flash column
chromatography to give the product as a white solid (1.60 g;
70%) mp 0–5 ЊC; νmax(NaCl)/cmϪ1 1739 (C᎐O), 1636 (C᎐N);
HC᎐N); δ 44.7 (ArCH2), 128.3, 128.4, 128.5, 129.1, 129.2,
᎐
C
129.3, 129.4, 132.8 (ArC), 156.0, 158.7, 159.5 (C᎐O, C᎐O, C᎐
᎐
᎐
᎐
N); CIMS m/z (%), 283 (MHϩ, 29%), 180 (100); found: MHϩ
283.1082, C16H15N2O3 requires MHϩ 283.1083).
Benzaldehyde O-(N-benzyl-prop-2-enylaminooxalyl)oxime 5e
To a stirred solution of 4d (2.11 g; 10 mmol) in DCM (25 cm3)
at 0 ЊC was added pyridine (1.0 g; 13 mmol) in DCM (5 cm3)
followed by N-benzylprop-2-enylamine (1.5 g; 10 mmol) in
DCM (10 cm3). The mixture was allowed to reach rt and then
stirred at rt for 2 h. After this time pentane (ca. 3 cm3) was
added to promote formation of the pyridine hydrochloride
ppt. The ppt was filtered off and the filtrate was evaporated to
dryness to give a yellow oil. The oil was further purified via
flash column chromatography to give the pure target molecule
as a colourless oil which solidified on cooling (1.40 g; 44%)
mp = 0–5 ЊC; νmax(NaCl)/cmϪ1 1744 (C᎐O), 1656 (C᎐N);
᎐
᎐
δH (2 isomers) 2.34 (1H, q, J 7), 2.40 (1H, q, J 7) (apparent
sextet), 2.32 (1H, t, J 7, N–CH2), 2.42 (1H, t, J 7, N–CH2), 4.55
(1H, s, Ar–CH2), 4.69 (1H, s, Ar–CH2), 5.00–5.13 (2H, m,
᎐CH ), 5.63–5.82 (1H, m, ᎐CH), 7.30–7.70 (10H, m, ArH), 8.42
᎐
᎐
2
(1/2H, s), 8.50 (1/2H, s); δC (2 isomers) 31.1, 32.5, 43.4, 46.5,
47.1, 51.8 (CH ), 117.3 (᎐CH ), 127.9, 128.3, 128.3, 128.6,
᎐
2
2
128.6, 128.8, 128.9, 132.3, (ArC), 134.9, 135.7 (᎐CH), 157.7,
᎐
157.9 (C᎐O), 160.9, 161.2 (C᎐N); CIMS m/z (%) MHϩ (337,
᎐
᎐
᎐
᎐
24%), 234 (92), 190 (25), 104 (77), 57 (100); found: MHϩ
337.1562, C20H21N2O3 requires MHϩ 337.1552.
δH (2 isomers) 3.83 (1H, d, J 7, CH2), 3.88 (1H, d, J 7, CH2),
4.46 (1H, s, ArCH2), 4.64 (1H, s, ArCH2), 5.18–5.27 (2H, m,
᎐CH ), 5.66–5.84 (1H, m, ᎐CH), 7.22–7.68 (10H, m, ArH), 8.41
᎐
᎐
2
1-Benzyl-3-methylazetidin-2-one26 12e
(1/2H, s), 8.44 (1/2H, s); δC (2 isomers) 47.8, 48.8, 50.3, 51.1
(CH ), 119.2 (᎐CH ), 127.9, 128.0, 128.1, 128.3, 128.6, 128.8,
᎐
2
2
A solution of benzaldehyde O-(N-benzylprop-2-enylamino-
oxalyl)oxime 5e (800 mg; 2.5 mmol) and MAP (1.13 g;
7.5 mmol) in toluene (400 cm3) was photolysed at 100 ЊC by
light from a 400 W medium pressure Hg lamp for 5 h. After this
time the reaction mixture was evaporated to dryness. The result-
ing oil was evaporated to dryness and purified via column
chromatography (DCM; MeOH) to give the pure product as
a colourless oil (173 mg; 40%); δH 1.27 (3H, d J 7, Me), 2.73
(1H, dd J 5 and 2, CH), 3.15 (1H, m, CH), 3.26 (1H, m, CH),
128.9, 129.0, 129.1, 130.4, 130.6, 132.1 (ArC), 132.7 (᎐CH),
᎐
157.7, 157.9 (C᎐O), 163.3 (C᎐N); CIMS m/z (relative intensity),
᎐
᎐
(CI), 323 (MHϩ, 36%), 220 (100); found: MHϩ 323.1392,
C19H19N2O3 requires MHϩ 323.1396.
Benzaldehyde O-(N-benzyl-2-cyclohexenylaminooxalyl)oxime 5f
To a stirred solution of 4d (1.13 g; 5.3 mmol) in DCM (20 cm3)
at 0 ЊC was added pyridine (0.42 g; 5.3 mmol) in DCM (5 cm3)
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
721

View Article Online
4.30 (1H, d, J 15, CH), 4.35 (1H, d, J 15, CH), 7.30 (5H, m,
ArH); δC 13.7 (CH3), 44.5 (CH), 45.8 (CH2), 46.7 (CH2–Ph),
was refluxed under Dean–Stark conditions for 24 h. After this
time the pentane was removed and the thiazolidine was iso-
lated as a colourless oil via column chromatography (hexane :
EtOAc, 8 : 2),(2.43 g, 92%). The amine was further purified by
recystallisation from hexane and the thiazolidine was isolated
as colourless platelets (mp 18–20 ЊC). The thiazolidine was iso-
lated as a mixture of isomers in a ratio of 2 : 1. Major isomer:
δH 1.06 (3H, d, J 7, CHCH3), 1.11 (3H, d, J 7, CHCH3), 1.99
(1H, m, CH(CH3)2), 2.24 (1H, br, NH), 2.76 (1H, t, J 10,
5-HCH), 3.30 (1H, dd, J 7, 10, 5-HCH), 3.87 (1H, dd, J 7 and
10, 4-CH), 4.36 (1H, d, J 7, 2-CH), 5.21 (2H, AB, CH2–Ph),
127.7, 128.3, 128.8, 135.9 (Ar), 171.4 (C᎐O).
᎐
7-Benzyl-2-hydroxy-7-aza-bicyclo[4.2.0]octan-2-one 12f
A solution of 5f (800 mg; 2.2 mmol) and MAP (1.0 g; 6.6 mmol)
in toluene (400 cm3) was photolysed at 100 ЊC by light from a
400 W medium pressure Hg lamp for 5 h. After this time the
reaction mixture was evaporated to dryness. The resulting oil was
purified via column chromatography (DCM; MeOH) to give a
mixture of isomers as a colourless oil (330 mg, 70%). The product
exists as a mixture of anti and syn isomers (5 : 1) and these were
separated by column chromatography (EtOAc; hexane) (35 : 65);
anti δH 1.21–1.80 (5H, m), 1.95 (1H, m), 3.23 (1H, m), 3.78 (1H,
dd), 4.13 and 4.53 (2H, d, J 15), 4.27 (1H, m, HC–OH), 7.29 (5H,
m, Ar); δC 15.0 (C-7), 22.9 (C-8), 28.8 (C-6), 44.4 (CH2–Ar), 50.8
(C-1), 54.9 (C-4), 65.5 (C-OH), 127.7, 128.3, 128.8, 135.8 (Ar),
7.36 (5H, s, ArH); δC 20.9, 21.1 (CH ), 34.0 (᎐CH), 38.0
᎐
3
(C-5), 65.9 (C-4), 67.6 (CH2–Ph),78.5 (C-2), 128.7, 128.8, 129.0,
129.1 (ArC), 171.7 (C᎐O); minor isomer: δ 0.98 (3H, d, J 7,
᎐
H
CHCH3), 1.04 (3H, d, J 7, CHCH3), 1.79 (1H, m, CH(CH3)2),
2.24 (1H, br, NH), 3.02 (1H, dd, J 7 and 10, 5-HCH), 3.20 (1H,
dd, J 7 and 10, 5-HCH), 4.13 (1H, t, J 7, 4-CH), 4.44 (1H, d,
J 7, 2-CH), 5.21 (2H, AB, CH2–Ph), 7.36 (5H, s, ArH); δC 20.2,
168.6 (C᎐O); CIMS m/z (relative intensity), 232 (MHϩ, 100%),
᎐
20.8 (CH ), 35.5 (᎐CH), 37.7 (C-5), 64.8 (C-4), 67.5 (CH –
᎐
found MHϩ 232.1344, C14H18NO2 requires MHϩ 232.1338.
Syn δH 1.15–2.12 (5H, m), 2.53 (1H, m), 3.81 (1H, d, J 4), 3.95
(1H, m), 4.03 ( 1H, m, HC-OH), 4.09 and 4.53 (2H, d, J 15,
CH2Ar), 7.22 (5H, m, Ar); δC (CDCl3) 15.9 (C-7), 24.4(C-8), 40.2
(C-6), 45.0 (CH2–Ar, C-9), 51.1 (C-1), 61.7 (C-4), 77.2 (C-OH),
3
2
Ph),77.0 (C-2), 128.7, 128.8, 129.0, 129.1 (ArC), 171.7 (C᎐O);
᎐
m/z (relative intensity)(ElectroSpray) 288 (100, Mϩ ϩ Na);
found Mϩ 266.1219, C14H20NO2S requires Mϩ 266.1215.
5,5–Dimethyl-2-(2-methylpropenyl)-thiazolidine-4-carboxylic
acid methyl ester 6b
128.1, 128.4, 129.0, 135.0 (Ar), 160.7 (C᎐O).
᎐
To a stirred solution of the methyl ester of ,-penicillamine
hydrochloride (4.5 g; 22.5 mmol) in MeOH (40 cm3) at 0 ЊC
was added a solution of triethylamine (2.3 g; 22.5 mmol) in
MeOH (10 cm3) and a solution of 3-methylbut-2-enal (1.9 g;
22.5 mmol) in MeOH (10 cm3). The mixture was allowed to
reach rt and then stirred at rt for 24 h. The solvent was removed
and the crude product was purified via column chromatography
(hexane; EtOAc) (8 : 1). The product was recrystallised from
hexane at Ϫ20 ЊC to give colourless needles mp = 52–54 ЊC
(1.3 g; 25%); δH 1.24 (3H, s, CH3), 1.65 (3H, s, CH3), 1.70 (3H, s,
CH3), 1.74 (3H, s, CH3), 2.91 (1H, br, NH), 3.63 (1H, s, 3-H),
1-Benzyl-3-(hydroxy-phenyl-methyl)-azetidin-2-one 12g
A solution of 5g (400 mg; 0.97 mmol) and MAP (452 mg;
3.0 mmol) in toluene (400 cm3) was photolysed at 100 ЊC by
light from a 400 W medium pressure Hg lamp for 5 h. After this
time the reaction mixture was evaporated to dryness. The result-
ing oil was purified via column chromatography (DCM; MeOH)
to give a mixture of isomers (3 : 1; anti : syn) as a colourless oil
(169 mg; 69%). The isomers were separated by further column
chromatography (EtOAc; hexane) anti δH 3.04 (1H, t, J 5, N–
CH), 3.20 (1H, dd, J 5 and 3, N–CH), 3.59 (1H, m, COCH), 3.81
(1H, br, OH), 4.39 (2H, AB, ArCH2), 5.24 (1H, d, J 3, ArCH),
7.14–7.38 (10H, m, ArH); δC (CDCl3) 40.2 (N–CH2), 45.9
(ArCH2), 56.7 (CH), 69.6 (CHOH), 127.6, 127.7, 127.9, 128.3,
3.77 (1H, s, OMe), 5.24 (1H, m, ᎐CH), 5.39 (1H, m, ᎐CH);
᎐
᎐
δC 18.9 (CH3), 26.0 (CH3), 28.9 (CH3), 29.5 (CH3), 52.5 (OMe),
60.0 (q, C3), 65.0 (C2), 74.8 (C1), 123.7 (᎐CH), 137.9 (᎐MeMe),
᎐
᎐
170.2 (C᎐O); m/z (relative intensity) 229 (52, Mϩ), 214 (11), 196
128.4, 128.5, 128.7, 128.8 (ArC), 168.3 (C᎐O); EIMS m/z (relative
᎐
᎐
intensity) 267 (Mϩ, 10%), 176 (17), 133 (38), 105 (62), 91 (100), 84
(62), 77 (55); found Mϩ 267.1267, C17H17NO2 requires Mϩ
267.1259; syn δH 2.99 (1H, dd, J 2 and 6, N–CH), 3.14 (1H, t, J 6,
N–CH), 3.61 (1H, m, COCH), 3.81 (1H, br, OH), 4.35 (2H, AB,
ArCH2), 5.02 (1H, d, J 7, ArCH2), 7.03–7.44 (10H, m, ArH); δC
41.6 (N–CH2), 45.8 (ArCH2), 56.1 (CH), 73.0 (CHOH), 127.6,
(15), 170 (20), 155 (86), 95 (100); EIMS found Mϩ 229.1137,
C11H19NO2S requires 229.1129. X-Ray diffraction showed this
to be the 4S stereoisomer.
Attempted preparation of 2-formyl-5,5-dimethylthiazolidine-4-
carboxylic acid methyl ester
127.8, 127.9, 128.1, 128.2, 128.6, 128.8, 128.9 (ArC), 168.3 (C᎐O);
᎐
EIMS m/z (relative intensity) 267 (Mϩ, 11%), 176 (37), 133 (59),
105 (47), 91 (100), 77 (41); found Mϩ 267.1259, C17H17NO2
requires Mϩ 267.1259.
To a stirred solution of the methyl ester of ,-penicillamine
hydrochloride (2 g; 10 mmol) in MeOH (20 cm3) at 0 ЊC was
added a solution of triethylamine (1.0 g; 10 mol) in MeOH
(5 cm3) and dimethoxyacetaldehyde (1.17 g; 10 mmol). The mix-
ture was stirred at room temperature for 24 h and after this time
the mixture was filtered and the solvent removed. The yellow oil
which resulted was separated by column chromatography and
the thiazolidine was isolated as a mixture of isomers in 81%
yield. The dimethoxythiazolidine was dissolved in MeOH and
treated with solutions of 0.1 M, 1.0 M, 5 M and 6 M HCl. Each
solution was heated to reflux for up to 3 h but the expected
thiazolidine was never observed
1-Benzyl-3-methylpyrrolidin-2-one 12h
A solution of 5h (800 mg; 2.38 mmol) and MAP (1.07 g; 7.14
mmol) in toluene (400 cm3) was photolysed at 100 ЊC by light
from a 400 W medium pressure Hg lamp for 5 h. After this time
the reaction mixture was evaporated to dryness. The resulting
oil was purified via column chromatography (DCM; MeOH) to
give the product 12h as a colourless oil (378 mg; 84%); δH 1.25
(3H, d, J 17, CH3), 1.6 (1H, m, CH), 2.25 (1H, m, CH2), 4.42
(1H, AB, CH), 4.48 (1H, AB), 7.21–7.37 (5H, m, ArH); δC 16.4
(CH3), 27.1 (CH2), 36.8 (CH), 44.7 (CH2), 46.8 (CH2), 127.5,
Benzaldehyde O-(2-isopropyl thiazolidine-4-carboxylic acid
benzyl ester oxalyl) oxime 7a
128.1, 128.6 (5 × CH), 136.7 (C), 177.4 (C᎐O). EIMS m/z
᎐
To a stirred solution of 4d (633 mg; 3 mmol) at 0 ЊC in DCM
(10 cm3) was added pyridine (237 mg; 3 mmol) in DCM (5 cm3)
and 6a (800 mg; 3 mmol) in DCM (5 cm3). The mixture was
allowed to stir at 0 ЊC for 10 min and then at rt for 3 h. After
this time a small quantity of pentane was added in order to
promote formation of the pyridine hydrochloride salt. The
mixture was filtered and the solvent removed under vacuum.
The oxime oxalate amide was purified by flash column chrom-
(relative intensity) 189 (100, Mϩ), 174 (11), 161 (12), 91(90);
found Mϩ 189.1153, C12H15NO requires Mϩ 189.1154.
2-Isopropyl thiazolidine-4-carboxylic acid benzyl ester 6a
To a stirred solution of the benzyl ester of cysteine (2.0 g;
10 mmol) in pentane (50 cm3) was added triethylamine (2.0 g;
20 mmol) and isobutyraldehyde (1.44 g; 20 mmol). The mixture
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
722

View Article Online
atography to give the product as a colourless oil (1.18 g; 90%);
The compound was isolated as a mixture of isomers in a ratio
of 2 : 1, νmax(NaCl)/cmϪ1 1747 (C᎐O), 1659 (C᎐N); δ 0.97–1.04
(6H, m, 2 × CH3), 1.97 (1/3H m, CH(CH3)2), 2.15 (2/3H, m,
CH(CH3)2), 3.23–3.42 (2H, m, 5-HCH), 5.07–5.37 (2H, m,
4-CH and 2-CH), 5.30 (2H, s, OCH2), 7.26–7.53 (8H, m, ArH),
photolysed by light from a 400 W medium pressure UV lamp at
100 ЊC for 8 h. After this time the mixture was allowed to cool
and the solvent was removed. GC-MS: peak no. 142, PhCHNH
(9%) m/z (%) 103 (Mϩ 100), 76 (26), 50 (15), peak no. 174, (9%)
PhCH2OH 108 (Mϩ 87), 107 (66), 79 (100), 77 (56), peak no.
217, OMeC6H4CHCH2 (8%) 134 (Mϩ 100), 119 (32), 91 (28),
peak no. 332, MAP (56%) peak no. 373, PhCH2CH2Ph (52%),
182 (Mϩ 26), 91 (100), 65 (12), peak no. 435, PhCHOHCH2Ph
(25%), 198 (Mϩ 3), 107 (82), 92 (100), 79 (60), 77(35), peak no.
459, penicillin derivative 17 (18%), 257 (Mϩ 78), 242 (16), 228
(94), 225 (86), 214 (100), 198 (31), 187 (46), 170 (54), 155 (24),
142 (46), 115 (80), 114 (46), 95 (97), 84 (43), 67 (39) (several
additional unidentified compounds derived from MAP were
also present).
᎐
᎐
H
7.69–7.77 (2H, m, arH), 8.40 (1H, s, HC᎐N); δ 19.3, 19.5, 20.4,
᎐
C
20.5 (CH3), 31.9, 33.4 (CH2), 63.5, 64.5 (CH), 68.5, 68.8
(OCH2), 71.9, 72.0 (CH), 128.9, 129.1, 129.1, 129.4, 132.7,
132.8 (ArC), 158.7, 165.6, 169.0 (C᎐O, C᎐N); CIMS found
᎐
᎐
MHϩ 441.1489, C23H25N2O5S requires MHϩ 441.1485.
Benzaldehyde O-(5,5–dimethyl-2-(2-methylpropenyl)-thiazol-
idine-4-carboxylic acid methyl ester oxalyl) oxime 7b
To a stirred solution of 4d (700 mg; 3.2 mmol) in DCM (10 cm3)
was added dropwise a solution of pyridine (0.3 g; 3.2 mmol) in
DCM (5 cm3) followed by 6b (0.8 g; 3.2 mmol) in DCM (5 cm3).
The solution was allowed to reach rt and then stirred at rt for
3 h. After this time a small amount of pentane (ca. 10 cm3) was
added in order to promote formation of the pyridine hydro-
chloride salt. The ppt was filtered off and the filtrate collected.
The filtrate was evaporated to dryness and the resulting oil was
filtered through a pad of silica with DCM. The product was
again evaporated to dryness and recrytallised from DCM–
hexane at Ϫ20 ЊC to give the title compound as colourless plate-
lets, (1.2 g; 92%); mp = 110–112 ЊC; (found C, 58.9; H, 6.4;
N, 6.0%; C20H24N2O5S requires C, 59.2; H 6.2; N, 6.9%);
νmax(NaCl)/cmϪ1 1758 (C᎐O), 1739 (C᎐O), 1658 (C᎐N); δH 1.44
2-(2-Methylpropenyl)-4-phenyl-oxazolidine 8
To a stirred solution of (S)-β-aminophenethyl alcohol (1.37 g;
10 mmol), in DCM (15 cm3) at 0 ЊC and in the presence of 4 Å
molecular sieves was added a solution of 3-methylbut-2-enal
(840 mg; 10 mmol) in DCM (5 cm3). The mixture was stirred at
rt for 3 h after which time the solution was filtered and concen-
trated to give a yellow solid. The solid was recrystallised from
EtOAc–hexane at Ϫ20 ЊC to give the title compound as colour-
less needles (1.85 g; 91%) mp = 70–72 ЊC; δH 1.88 (3H, s, CH3),
1.90 (1H, s, CH3), 3.29 (1H, br, NH), 3.83 (1H, dd, J 11 and 4,
CH), 3.96 (1H, dd, J 9 and 4, CH), 4.32 (1H, dd, J 9 and 4, CH),
6.08 (1H, d, J 9, CH), 7.21–7.37 (5H, m, ArH), 8.28 (1H, d, J 9,
CH); δC 19.1 (CH3), 27.1 (CH3), 68.0 (CH2),77.3 (C–Ph), 125.4,
127.6, 127.7, 129.0 (ArC), 149.0 (᎐C), 161.9 (᎐CH).
᎐
᎐
᎐
(3H, s, CH3), 1.63 (3H, s, CH3), 1.67 (3H, s, CH3), 1.69 (3H, s,
CH ), 3.80 (1H, s), 3.83 (3H, s, OCH ), 5.60 (1H, m, ᎐CH), 6.08
᎐
᎐
᎐
3
3
(1H, m, ᎐CH), 7.42–7.50 (3H, m, Ar), 7.71–7.75 (2H, m, Ar),
Attempted preparation of benzaldehyde O-(N-2-(2-methyl-
propenyl)-4-phenyloxazolidine)aminooxalayl oxime
᎐
8.43 (1H, s, N᎐CH); δ 18.2 (CH ), 23.9 (CH ), 25.7 (CH ), 32.1
᎐
C
3
3
3
(CH ), 51.6 (Me ), 52.4 (OCH ), 60.1, 71.5, 121.5 (᎐CH), 128.5,
᎐
To a stirred solution of 4d (1.41 g; 6.7 mmol) in DCM (15 cm3)
at 0 ЊC was added a solution of pyridine (530 mg; 6.7 mmol) in
DCM (5 cm3) followed by a solution of 2-(2-methylpropenyl)-4-
phenyl-oxazolidine (1.36 g; 6.7 mmol) in DCM (5 cm3). The
mixture was allowed to stir at 0 ЊC for 10 min and then at rt for
3 h. After this time the solvent was removed to give unreacted
starting materials.
3
2
3
128.7, 129.0, 132.1 (ArC), 140.1 (᎐CMe ), 157.3, 158.4 (C᎐O),
᎐
᎐
2
169.2 (C᎐N); CIMS found MHϩ 405.1488, C20H25N2O5S
᎐
requires MHϩ 405.1491.
Crystal data for 7b‡: C20H24N2O5S, M = 404.47, colourless
platelets, crystal dimensions 0.1 × 0.1 × 0.2 mm, monoclinic,
space group P21/c, a = 8.9072(13), b = 18.141(3), c = 12.9956(19)
Å, β = 99.949(3)Њ, V = 2092.1(5) Å3, Dc = 1.284 Mg mϪ3, T =
125(2) K, R = 0.0341, Rw 0.0831 for 2956 reflections with
I>2σ(I ) and 259 variables. Data were collected on a Bruker
SMART diffractometer with graphite-monochromated Mo–Ka
radiation (λ = 0.71073 Å). The structure was solved by direct
methods and refined using full-matrix least squares methods.
Atomic coordinates and bond lengths and coordinates are
listed in the ESI and the structure is shown in Fig. 1.
Acknowledgements
We thank the EPSRC (grant no. GR/N37674/01) and the Royal
Society of Chemistry for financial support.
References
1 Radicals in Organic Synthesis, eds. P. Renaud and M. P. Sibi, vols.
1 & 2, Wiley-VCH, Weinheim, 2001; A. F. Parsons, An Introduction
to Free Radical Chemistry, Blackwell Science, Oxford, 2000; D. P.
Curran, N. A. Porter and B. Giese, Stereochemistry of Radical
Reactions, VCH, Weinheim, 1995.
2 (a) W. R. Bowman, C. F. Bridge and P. Brookes, J. Chem. Soc.,
Perkin Trans. 1, 2000, 1; (b) A. J. McCarroll and J. C. Walton,
Angew. Chem., Int. Ed., 2001, 40, 2224; (c) A. J. McCarroll and
J. C. Walton, J. Chem. Soc., Perkin Trans. 1, 2001, 3215; (d ) A. G.
Fallis and I. M. Brinza, Tetrahedron, 1997, 53, 17543.
3 (a) A. Studer and S. Amrein, Synthesis, 2002, 835; (b) P. A. Baguley
and J. C. Walton, Angew. Chem., Int. Ed., 1998, 37, 3072.
4 (a) M. Hasebe, K. Kogawa and T. Tsuchiya, Tetrahedron Lett., 1984,
25, 3887; (b) M. Hasebe and T. Tsuchiya, Tetrahedron Lett., 1986,
27, 3239; (c) M. Hasebe and T. Tsuchiya, Tetrahedron Lett., 1987, 28,
6207; (d ) M. Hasebe and T. Tsuchiya, Tetrahedron Lett., 1988, 29,
6287.
5 A. J. McCarroll and J. C. Walton, Chem. Commun., 2000, 351;
A. J. McCarroll and J. C. Walton, J. Chem. Soc., Perkin Trans. 2,
2000, 2399.
Photolysis of benzaldehyde O-(2-isopropylthiazolidine-4-
carboxylic acid benzyl ester oxalyl) oxime 7a
A stirred solution of 7a (400 mg; 0.9 mmol) and MAP (137 mg;
0.91 mmol) in toluene (25 cm3) was photolysed for 3 h at rt by
light from a 400 W UV lamp. After this time the ppt which
had formed was filtered off and the filtrate was evaporated to
dryness to give a yellow oil. The oil was purified by column
chromatography (EtOAc–hexane) to give only benzaldehyde
and MAP. The ppt was found to be insoluble in all deuterated
solvents and was probaly polymer.
Photolysis of benzaldehyde O-(5,5–dimethyl-2-(2-methyl-
propenyl)thiazolidine-4-carboxylic acid methyl ester oxalyl)
oxime 7b
To a solution of 7b (400 mg; 2 mmol) in toluene (400 cm3) was
added 3 M eq. of MAP (890 mg; 6 mmol). The mixture was
6 E. M. Scanlan and J. C. Walton, Chem. Commun., 2002, 2086.
7 J. C. Jochims, S. Hehl and S. Herzberger, Synthesis, 1990, 12, 1128.
8 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and A. R. Tatchell,
Vogel’s Textbook of Organic Chemistry, 5th Edition, Longman,
Harlow, 1989.
‡ CCDC reference numbers 225168. See http://www.rsc.org/suppdata/
ob/b3/b315223e/ for crystallographic data in.cif or other electronic
format.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
723

View Article Online
9 I. Pejkovic´-Tadic´, M. Hranisavljevic´-Jakovljevic´, S. Nesˇic´,
C. Pascual and W. Simon, Helv. Chim. Acta, 1965, 48,
1157.
10 P. G. Sammes, S. Smith and G. T. Woolley, J. Chem. Soc.,
Perkin Trans. 1, 1984, 2603.
11 I. McMillan and R. J. Stoodley, J. Chem. Soc., Chem. Commun.,
1968, 11; H. T. Nagasawa, D. J. W. Goon, W. P. Muldoon and
R. T. Zera, J. Med. Chem., 1984, 27, 591.
12 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer and
A. G. Orpen, J. Chem. Soc., Perkin Trans. 2, 1987, S1.
13 J. W. Cooper, B. P. Roberts and J. N. Winter, J. Chem. Soc., Chem.
Commun., 1977, 320.
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W.
Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres,
C. Gonzalez, M. Head-Gordon, E. S. Replogle and J. A. Pople,
GAUSSIAN 98 (Revision A.11), Gaussian, Inc., Pittsburgh, PA,
2001.
19 G. B. Gill, G. Pattenden and S. J. Reynolds, J. Chem. Soc.,
Perkin Trans. 1, 1994, 369; G. Pattenden and S. J. Reynolds, J. Chem.
Soc., Perkin Trans. 1, 1994, 379.
20 L. V. Jackson and J. C. Walton, Chem. Commun., 2000, 2327.
21 J. A. Howard, in Free Radicals, ed. J. K. Kochi, Wiley, New York,
1973, vol. II, ch. 12, p. 3.
14 D. Griller, K. U. Ingold and J. C. Scaiano, J. Magn. Reson., 1980, 38,
169.
15 R. F. Hudson and K. A. F. Record, J. Chem. Soc., Chem. Commun.,
1976, 539.
16 A. R. Forrester and F. A. Neugebauer in Landolt-Börnstein,
Magnetic Properties of Free Radicals, eds. H. Fischer and
K.-H. Hellwege, Springer, Berlin, 1979, vol. 9c1.
17 A. F. Bella, L. V. Jackson and J. C. Walton, J. Chem. Soc., Perkin
Trans. 2, 2002, 1839.
18 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr.,
R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone,
M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
22 A. Archelas, J. D. Fourneron and J. D. Furstoss, Tetrahedron Lett.,
1988, 29, 6611.
23 J. McCann and S. F. Overman, J. Am. Chem. Soc., 1987, 109(20),
6107.
24 M. Mori, N. Kanda, I. Oda and Y. Ban, Tetrahedron, 1985, 41(23),
5465.
25 I. Buchanan, C. D. Garner and W. Clegg, J. Chem Soc., Dalton
Trans., 1984, 1333.
26 W.-B. Wang and E. J. Roskamp, J. Am. Chem. Soc., 1993, 115, 9417–
9420.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 1 6 – 7 2 4
724