1H NMR spectroscopic studies of the stability of an oxazolidine condensation product

Article abstract of DOI:10.1002/(SICI)1097-458X(199705)35:5<337::AID-OMR86>3.0.CO;2-5

Condensation of (-)-norephedrine with excess formaldehyde under mild conditions leads to formation of the 2:1 condensation product N,N'-methylenebis(4-methyl-5-phenyl)oxazolidine compared with the reaction with 1 mol of formaldehyde, which leads to 4-methyl-5-phenyloxazolidine. ¹H and ¹³C NMR spectroscopy was used to monitor the stability of this compound and its decomposition products. The 2:1 condensation product is found to be stable in CDC1₃ but breaks down rapidly in CD₃OD to yield a 50:50 mixture of 4-methyl-5-phenyloxazolidine and 3-hydroxymethyl-4-methyl-5-phenyloxazolidine. Upon addition of D₂O to this equimolar mixture, the latter compound decomposes to norephedrine and formaldehyde, whereas the former compound is stable.

Full text of DOI:10.1002/(SICI)1097-458X(199705)35:5<337::AID-OMR86>3.0.CO;2-5

MAGNETIC RESONANCE IN CHEMISTRY, VOL. 35, 337È341 (1997)
NOTE
1H NMR Spectroscopic Studies of the Stability of
an Oxazolidine Condensation Product
Gerard P. Moloney,1 Magdy N. Iskander1 and David J. Craik2*
1 School of Pharmaceutical Chemistry, Victorian College of Pharmacy (Monash University), 381 Royal Parade Parkville,
Victoria 3052, Australia
2 Centre for Drug Design and Development, University of Queensland, Brisbane, Queensland 4072 Australia
Condensation of (Ô)-norephedrine with excess formaldehyde under mild conditions leads to formation of the 2 :1
condensation product N,Nº-methylenebis(4-methyl-5-phenyl)oxazolidine compared with the reaction with 1 mol of
formaldehyde, which leads to 4-methyl-5-phenyloxazolidine. 1H and 13C NMR spectroscopy was used to monitor
the stability of this compound and its decomposition products. The 2 :1 condensation product is found to be stable in
CDC1 but breaks down rapidly in CD OD to yield a 50 : 50 mixture of 4-methyl-5-phenyloxazolidine and 3-
3
3
hydroxymethyl-4-methyl-5-phenyloxazolidine. Upon addition of D O to this equimolar mixture, the latter com-
2
pound decomposes to norephedrine and formaldehyde, whereas the former compound is stable. ( 1997 by John
Wiley & Sons, Ltd.
Magn. Reson. Chem. 35, 337È341 (1997) No. of Figures: 6 No. of Tables: 4 No. of References: 14
Keywords: NMR, 1HNMR, 13CNMR, oxazolidine
Received 6 June 1996; revised 15 November 1996; accepted 15 November 1996
properties have not been characterized. It was of inter-
est to characterize its spectra fully and examine its sta-
bility in a similar manner to that of other stability
INTRODUCTION
Oxazolidine derivatives (1) have a variety of uses,
including their application as bactericides,1 fungicides1
and herbicides2 (Scheme 1). Several members of this
class have also undergone trials as mosquito
repellents3 and tanning agents.4
In previous work in this laboratory,5 substituents
R1ÈR5 were varied systematically and their electronic
and steric e†ects on the stability of the oxazolidine
system examined. As part of that work there was inter-
est in synthesizing the compound with R1 \ C H ,
6 5
R2 \ CH and R3 \ R4 \ R5 \ H (2). This compound
3
had previously been reported and characterized by 13C
and 1H NMR.6 In the current study, the synthesis of
this compound was attempted using a route that has
been widely applied for other oxazolidine derivatives,
that is, the condensation of the appropriate b-amino
alcohol with an aldehyde (Scheme 2).
The reaction appeared to proceed smoothly and
yielded a single product. However, on examination of
the 13C and 1H NMR spectra it became apparent that
the product was not 2, as expected, but a 2 :1 conden-
sation product 3 (Scheme 1). Further investigation
revealed that a condensation product would, indeed, be
expected if an excess of formaldehyde were used in the
reaction compared with the single molar equivalent
required for production of 2.
Scheme 1. Structures of compounds 1–5.
The 2 :1 condensation product has also been reported
previously,7 although its 13C and 1H NMR spectral
* Correspondence to: D. J. Craik
Scheme 2.
( 1997 by John Wiley & Sons, Ltd.
CCC 0749È1581/97/050337È05 $17.50

338
G. P. MOLONEY, M. N. ISKANDER AND D. J. CRAIK
studies on oxazines and oxazolidines recently carried
out in this laboratory.5 In the course of this work it
became apparent that in some solvents the conden-
sation compound is stable, whereas in others it rapidly
decomposes to yield two compounds, the desired oxa-
zolidine 2 and the hydroxymethyl derivative 4 (Scheme
1), which on addition of water, break down at substan-
tially di†erent rates, 2 being essentially stable and 4
breaking down readily.
Table 2. 13C NMR spectral data for 2 and
4 in CD OD
3
Chemical shift (ppm)
Carbon
2
4
C-4a
C-4
15.9
60.2
81.1
81.6
15.6
58.3
C-5
82.3
C-2
85.9
C-3a
C-2¾, C-6¾
84.2
126.9
127.6
128.3
140.0
126.9
127.6
128.3
139.9
C-4¾
RESULTS
C-3¾, C-5¾
C-1¾
Following the reaction of norephedrine (5) with formal-
dehyde under the conditions described in the Experi-
mental section, a single product was obtained. The
compound was dissolved in CDC1 and its 13C NMR
3
4. However, after the NMR solution had been left to
stand for 15 min, residual water in the solvent resulted
in complete decomposition of 3 to the oxazolidine
hydrolysis products 2 and 4. It appears that traces of
spectrum recorded. The nine observed peaks have shifts
consistent with structure 3. The peak which identiÐes
the compound as a condensation product is that at 72.6
ppm, characteristic of a methylene group between two
nitrogens. Likewise in the 1H NMR spectrum, a unique
peak integrating to two protons occurs at 3.64 ppm.
The remaining1H and 13C shifts are consistent with the
proposed structure for 3, as are the mass spectral data,
which show a molecular ion at m/z 338. Infrared data
also support the proposed structure.
water in the CD OD are sufficient to decompose the
3
condensation product almost instantaneously when it is
present at a concentration of ca 0.03 M, but that break-
down occurs over the course of minutes at a concentra-
tion of 0.25 M.
Following initial breakdown to a mixture of two
compounds when 3 is added to methanol, the solution
remained stable for several days, there being no change
in the relative intensities of peaks in the spectrum.
In contrast to the NMR data reported in the Experi-
mental section for a CDC1 solution of 3, two sets of
3
resonances of similar intensity are observed in CD OD.
3
However, on addition of an excess (700 ll) of D O, one
This suggests that the compound breaks down into two
2
of the compounds decomposed. The compound identi-
di†erent species in CD OD. As noted below, it was sub-
3
sequently determined that it was residual water in the
CD OD solvent that initiated the breakdown of 3.
3
Analysis of the observed 1H and 13C chemical shifts
shows that the two compounds are 2 and 4. These
NMR spectral data are summarized in Tables 1 and 2.
Peaks in the1H NMR spectrum were attributed to one
or other of the compounds in the mixture by analysis of
the two-dimensional DQF-COSY spectrum and by
Table 3. 1H NMR spectral data for 3 in CD OD
3
Chemical shift
(ppm)
Coupling constant
(Hz)
Proton
Multiplicity
CH
0.68
3.59
4.50
4.90
5.09
7.31
3.66
d
dq
d
6.8
6.6, 6.8
4.5
3
H-4
monitoring peak intensities in
a stability study,
H-2
H-2
H-5
described below, in which there is selective decomposi-
B
A
d
4.4
tion of 4.
d
6.6
In the course of preparing samples in CD OD for
H-2¾–H-6¾
m
s
—
3
13C NMR studies, it was found that the use of a higher
CH OH
—
2
concentration of 3 (0.25 M) than was used in the initial
1H experiments allowed temporary observation of 3
itself as well as the two hydrolysis products. Spectral
assignments for 3 in CD OD are given in Tables 3 and
3
Table 4. 13C NMR spectral data for
3 in CD OD
Table 1. 1H NMR spectral data for 2 and 4 in CD OD
3
3
Carbon
Chemical shift (ppm)
Chemical shift
(ppm)
Coupling constant
(Hz)
C-4a
C-4
15.9
61.4
Proton
2
4
Multiplicity
2
4
CH
0.72
3.52
4.42
4.87
4.88
7.28
0.65
3.58
4.58
4.84
5.03
7.28
4.23
d
dq
d
6.7
6.7, 6.8
5.7
6.9
6.6, 6.8
4.2
C-5
81.4
3
H-4
C-2
85.5
H-2
H-2
H-5
C-3a
C-2¾, C-6¾
81.4
B
A
d
5.6
4.1
126.9
127.1
128.3
139.9
d
6.6
6.5
C-4¾
H-2¾–H-6¾
d
—
—
C-3¾, C-5¾
C-1¾
CH OH
s
2
( 1997 by John Wiley & Sons, Ltd.
MAGNETIC RESONANCE IN CHEMISTRY, VOL. 35, 337È341 (1997)

STABILITY OF AN OXAZOLIDINE CONDENSATION PRODUCT
339
Ðed as 2 apparently remains stable in solution under
these conditions, whereas 4 breaks down over the
course of several hours to yield norephedrine and form-
aldehyde. ConÐrmation that the product obtained from
the breakdown in the presence of water is norephedrine
is seen in Fig. 1, which shows the Ðnal spectrum from
the decomposition experiment in the lower trace, while
the upper trace shows the same sample to which an
additional amount of pure norephedrine has been
added. Peaks attributed to norephedrine in the decom-
posed mixture clearly increase in intensity in the spiked
sample.
structures of the open-chain norephedrine derivatives
and closed oxazolidine are similar, it is unlikely that the
di†erence in apparent stability for this series is related
to thermodynamic control of the reaction. In other
words, the stabilization of the closed oxazolidines by
the R3 substituent is likely to be similar to that produc-
ed in the corresponding open compounds. Therefore,
the di†erences in breakdown most likely arise from dif-
ferential stabilization of intermediates or transition
states.
It appears that the basicity of the oxazolidine nitro-
gen is the driving force causing the preferential hydro-
lysis of the N-hydroxymethyloxazolidine compound 4.
The oxazolidine nitrogen of 4 is more basic than that in
the secondary amine 2 and this assists in the oxazoli-
dine nitrogen of 4 donating electron density to form the
cationic imine intermediate I shown in the proposed
mechanism in Scheme 3. Further hydrolysis of I would
result in the formation of the N,N@-dihydroxymethyl
derivative II. The nitrogen in this intermediate would be
susceptible to protonation followed by cleavage of
formaldehyde to yield the N-hydroxymethyl-b-amino
alcohol derivative III, which undergoes further decom-
position to give norephedrine and another molecule of
formaldehyde.
DISCUSSION
In previous studies of oxazolidine hydrolysis, it has
been suggested that the breakdown occurs via a Schi†
base intermediate.8h12 If this mechanism also applies in
the current case, it is of interest to determine why the
hydrolysis occurs much more readily for 4 (R3 \
CH OH) relative to 2 (R3 \ H). Examination of trends
2
in analogous compounds5 suggests that intermediate
breakdown rates occur for R3 \ CH and C H . As the
3
6 5
Figure 1. 300 MHz 1H NMR spectra of 3 in CD OD. (a) Spectrum recorded 60 min after addition of 700 ll of D O; (b) spectrum recorded
after addition of 20 mg of norephedrine to mixture (a).
3
2
( 1997 by John Wiley & Sons Ltd.
MAGNETIC RESONANCE IN CHEMISTRY, VOL. 35, 337È341 (1997)

340
G. P. MOLONEY, M. N. ISKANDER AND D. J. CRAIK
Scheme 5.
Hydrolysis of the methylene bridge in this case14 gave a
related N-hydroxymethyl derivative 7 and the corre-
sponding unsubstituted derivative 8.
EXPERIMENTAL
Spectra
1H and 13C NMR spectra were recorded in 5 mm tubes
at 25 ¡C on a Bruker AM 300 WB spectrometer. The
deuterium signal of the solvent was used as the lock and
tetramethylsilane (TMS) was the internal standard.
One-dimensional NMR experiments were carried out
with a spectral width of 3000 Hz, a 45¡ pulse angle, 16
384 data points and a repetition delay of 2.0 s. Sixteen
scans were accumulated prior to Fourier transform-
ation.
The DQF-COSY spectrum was recorded using a
spectral width of 3000 Hz, a 90¡ pulse of 7.8 ls and a
repetition delay of 2.0 s. For each FID, 32 scans were
accumulated. The two-dimensional data were collected
as a 512 ] 1024-word matrix and were zero-Ðlled to
1024 ] 2048 prior to Fourier transformation. A sine-
bell window function was applied in both dimensions.
Mass spectra and high-resolution mass spectra were
obtained on a Jeol JMX DX-300 double-focusing
instrument. Melting points were determined on a Gal-
lenkamp melting point apparatus and are uncorrected.
Methanol was distilled from iodine and magnesium and
stored over 3 Ó molecular sieves. The solution was con-
centrated on a Buchi rotary evaporator. Infrared
spectra were measured KBr disks on a Bruker IFS66
FTIR spectrometer.
Scheme 3.
The current study has demonstrated the reactive
nature of the N,N@-methylene bridge in the 2 :1 conden-
sation product in the presence of H O and has shown
2
that there is an interesting and signiÐcant di†erence in
the rate of breakdown of the two hydrolysis products,
which di†er only in their N-substituent. Previous
workers7 have utilised the unstable and reactive nature
of the N,N@-methylene bridge in the synthesis of a
number of N-substituted norephedrine derivatives by
reacting the title compound with various acetate deriv-
atives, as shown in Scheme 4. Other compounds of
similar structure containing an N,N@-methylene bridge
have also been shown to be extremely reactive. Taurolin
(6), which is claimed to chemically inactivate bacterial
endotoxins,13 has been examined in aqueous solution
and an equilibrium established, as outlined in Scheme 5.
(4R,5S-4ºR,5ºS)
N,Nº-Methylene-bis-(4-methyl-5-phenyloxazolidine) (3)
A 38% formalin solution (2.17 m1, 27.5 mmo1) was
added dropwise to a stirred solution of ([)-nor-
ephedrine (2.0 g, 13.0 mmo1) in methanol (20 ml). After
reÑuxing for 2 h, the solution was cooled, Ðltered and
the methanol reduced under vacuum. The white powder
was recrystallized from methanol to give the 2 :1 con-
densation product 3 as small, white needles (1.30 g,
61.4%), m.p. 100È101 ¡C (lit. m.p. 98È99 ¡C). Found:
Scheme 4.
M` 338.197; C
H
N O requires M` 338.200. V
21 26 2 2
max
( 1997 by John Wiley & Sons, Ltd.
MAGNETIC RESONANCE IN CHEMISTRY, VOL. 35, 337È341 (1997)

STABILITY OF AN OXAZOLIDINE CONDENSATION PRODUCT
341
3040, 2932, 1492, 1452, 1382, 1294, 1184, 1086, 1060,
85.2 (C-2), 126.5 (C-2@, C-6@) 127.2 (C-4@) 128.0 (C-3@,
C-5@) 139.7 (C-1@).
1038, 714 cm~1. MS: m/z 338 (M`, 0.1%), 176 (100),
163 (2), 146 (80), 105 (53), 57 (43). 1H NMR (CDC1 ); d
3
0.69 [d, 6H, 2 ] CH (4a), J \ 6.9 Hz], 3.48 (dq, 2H,
3
2 ] H-4, J \ 6.5, 6.9 Hz), 3.64 [s, 2H, CH (3a)], 4.52 (d,
2
2H, 2 ] H-2 , J \ 4.7 Hz), 4.91 (d, 2H, 2 ] H-2 ,
Acknowledgement
B
A
J \ 4.7 Hz), 5.08 (d, 2H, 2 ] H [ 5, J \ 6.5 Hz), 7.30
(m, 10H, 2 ] H-2@, H-3@, H-4@, H-5@, H-6@). 13C NMR
D. J. C. is grateful for the support of an Australian Research Council
Senior Fellowship.
(CDC1 ): d 15.3 (C-4a) 60.2 (C-4), 72.7 (C-3a), 80.3 (C-5),
3
REFERENCES
1. (a) B. Steiner, M. Repas, M. Ciha, D. Mlynarak and V. Sasink-
ova, Czech. Pat. 240 393 (1987); Chem. Abstr. 110, 57651g
(1989); (b) B. Steiner, M. Repas, V. Sasinkova and D. Pavlia-
kova, Czech. Pat. 247 581 (1988); Chem. Abstr. 110, 95211w
(1989); (c) B. Steiner, M. Repas, M. Ciha and V. Sasinkova,
Czech. Pat. 240 400 (1987); Chem. Abstr. 110, 135225n
(1989).
2. E. G. Teach (Stanfer Chemical Co.) US Pat. 4 723 986
(1988); Chem. Abstr. 108, 217796x (1988).
3. (a) S. C. Das, M. Bhuyan and B. C. Chakaborty, Indian J. Med
Res. 87, 176 (1988); (b) W.G. Taylor and D.E. Schreck, J
.Pharm. Sci. 78, 109 (1989).
Chem. Abstr. 97, 127237r (1982); (c) J. Engel, B. Kutscher,
G. Oepen, G. Niebch and P. Metzenauer, Ger. Pat. 3 622 839
(1987); Chem. Abstr. 106, 155945t (1987); (d) J. Engel, V.
Jakovlev and K. Theimer, Ger. Pat. 3 143 950 (1982); Chem.
Abstr. 97, 215900g (1982); (e) J. Engel, Braz. Pat. 8 107 374
(1983); Chem. Abstr. 100, 6206b (1983).
8. M. Johansen and H. Bundgaard, J. Pharm. Sci. 72, 1294
(1983).
9. H. Bundgaard and M. Johansen, Int. J. Pharm. 10, 165
(1982).
10. T. H. Fife and L. Hagopian, J. Am. Chem. Soc. 90, 1007
(1986).
4. M. Yang, X. Xu and L. Li, Xibei Qinggongye Xueyuan Xuebao
1, 11 (1988).
11. T. H. Fife and J. E. C Hutchins, J. Org. Chem. 45, 2099
(1980).
5. G. P. Moloney, PhD Thesis, Victorian College of Pharmacy,
Melbourne (1990).
12. A. Buur and H. Bundgaard, Int. J. Pharm. 18, 325 (1984).
13. R. W. Pfirrman and G. B Leslie, J. Appl. Bacteriol. 46, 97
(1979).
14. E. Myers, M. C. Allwood, M. J. Gidley and J. K. M. Sanders, J.
Appl. Bacteriol. 48, 89 (1980).
6. (a) M. Baudet and M. Gelbcke, Anal. Lett. 12, 325 (1979);
(b) M. Baudet and M. Gelbcke, Anal. Lett. 12, 641 (1979);
(c) M. Gelbcke and M. Baudet, Nouv. J. Chim. 7, 41 (1983).
7. (a) J. Engel, Br. Pat. 2 096 132 (1983); Chem. Abstr. 98,
178944x (1982); (b) J. Engel, Br. Pat. 2 087 397 (1982);
( 1997 by John Wiley & Sons Ltd.
MAGNETIC RESONANCE IN CHEMISTRY, VOL. 35, 337È341 (1997)