Kinetics of acetonitrile-assisted oxidation of tertiary amines by hydrogen peroxide

Article abstract of DOI:10.1039/b102066h

The rate of oxidation of tertiary amines by aqueous hydrogen peroxide is increased in the presence of acetonitrile due to the formation of a reactive intermediate. The active oxidant, presumably peroxyacetimidic acid, was quantified by a photometric method. Activation parameters of the acetonitrile-assisted and non-assisted oxidations are given in the temperature range 20 to 40 °C. The increased rate of the assisted oxidation is explained by the low enthalpy of activation although the entropy of activation is more negative due to a highly ordered transition state.

Full text of DOI:10.1039/b102066h

View Online / Journal Homepage / Table of Contents for this issue
Kinetics of acetonitrile-assisted oxidation of tertiary amines by
hydrogen peroxide
Gerhard Laus
2
Immodal Pharmaka GmbH, Bundesstrasse 44, A-6111 Volders, Austria
Received (in Cambridge, UK) 6th March 2001, Accepted 23rd March 2001
First published as an Advance Article on the web 17th May 2001
The rate of oxidation of tertiary amines by aqueous hydrogen peroxide is increased in the presence of acetonitrile due
to the formation of a reactive intermediate. The active oxidant, presumably peroxyacetimidic acid, was quantified by
a photometric method. Activation parameters of the acetonitrile-assisted and non-assisted oxidations are given in the
temperature range 20 to 40 ЊC. The increased rate of the assisted oxidation is explained by the low enthalpy of
activation although the entropy of activation is more negative due to a highly ordered transition state.
sulfoxides and alkenes in alkaline solution.5 In the present work
Introduction
it was found that admixture of small amounts of acetonitrile
(mole fraction of water χ from 1.0 to 0.8) to buffered solutions
(pH 7) of pteropodine 1 accelerated the observed rate of oxida-
In the course of studies on alkaloidal constituents of the South
American vines Uncaria tomentosa (Willd.) DC. and Uncaria
guianensis (Aubl.) Gmel. (Rubiaceae),1 N-oxides of oxindole
tion by a factor of approximately 200. Oxidations were carried
alkaloids† were required as reference compounds. A simple pro-
cess was desired to yield pure N-oxide samples in aqueous solu-
out with an alkaloid concentration of 10Ϫ5 to 10Ϫ4 mol dmϪ3
and an excess of hydrogen peroxide from 0.02 to 0.2 mol dmϪ3
.
tions which can be used directly for reversed-phase HPLC.
Hydrogen peroxide was chosen as the oxidant. Because of the
known tendency of spiro oxindole alkaloids to isomerize in
aqueous solution2 and due to their poor solubility in water, the
effect of organic co-solvents was to be examined. Of course
these co-solvents had to be compatible with the requirements of
HPLC. During this work the rate-enhancing effect of
acetonitrile on oxidations in neutral aqueous solutions was dis-
covered and shown to be a general phenomenon.
Over a wide range of mixtures (χ from 0.8 to 0.2) the observed
rate coefficients were independent of χ. Addition of still more
acetonitrile (χ < 0.2) resulted in a sudden drop of the rate of
reaction. This unusual behaviour is illustrated in Fig. 1a. The
results with the weaker base isopteropodine 2 were similar,
although this alkaloid is oxidized more slowly than the 7-
epimer 1 (Fig. 1b). Fortunately, this alkaloid also isomerizes
more slowly. The polarity of the solutions at a mole fraction of
χ = 0.5 is sufficiently low to inhibit the undesired isomerization.
Thus, oxidation of the oxindole alkaloids was 4 to 5 orders of
magnitude faster than isomerization at χ = 0.5.
Results
Effect of solvent polarity
The rate of isomerization of oxindole alkaloids is controlled by
solvent polarity. Therefore, admixture of organic solvents to the
buffered aqueous medium (pH 7) is a logical approach to avoid
or minimize the problem of isomerization. Oxidations were
carried out with an alkaloid concentration of 10Ϫ5 mol dmϪ3
and an excess of hydrogen peroxide from 0.2 to 2.4 mol dmϪ3
.
As expected,3,4 the reaction was pseudo-first-order, and the
second-order rate coefficients k2 were calculated from the slope
of a linear graph of k1 vs. [H2O2]. When the oxidation of ptero-
podine 1 was conducted in water–methanol mixtures in order
to suppress the isomerization, a decrease of the rate of reaction
was observed with a linear correlation between the free energy
of activation and the mole fraction of water χ in the range
χ = 1.0 to 0.45 (four points, eqn. (1)). The use of less polar
log (k2/dm3 molϪ1 sϪ1) = 1.13 χ Ϫ 4.83 r2 = 0.989 (1)
In acetonitrile–water mixtures the assisted and non-assisted
oxidations take their course as parallel reactions, and the form-
ation of peroxyacetimidic acid or its anion peroxyacetimidate
in a pre-equilibrium is assumed (Scheme 1). Peracids typically
have pKa values of approximately 8 and, therefore, are mostly
undissociated in neutral solution.6 According to this mech-
anistic concept, the rate law of the total reaction can be
described by eqn. (2), where kW is the rate coefficient for the
non-assisted reaction in water, and k2 that for the reaction of
the assumed active oxidant peroxyacetimidate with the amine.
co-solvents like dioxane or tert-butyl alcohol suppressed the
isomerization but decreased the rate of oxidation even more.
Therefore, these solvents do not appear to be advantageous.
Effect of acetonitrile
Nitriles promote oxidations by hydrogen peroxide of sulfides,
† The numbering system is based on that customarily used for the
hetero-yohimbinoid alkaloids.
864
J. Chem. Soc., Perkin Trans. 2, 2001, 864–868
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b102066h

View Online
Fig. 1 Dependence of the observed second-order rate coefficients of the oxidation of (a) pteropodine 1, (b) isopteropodine 2, (c) pyridine 3 and (d)
2-morpholinoethyl N-phenylcarbamate 4 by hydrogen peroxide at 30 ЊC on the mole fraction χ of water in acetonitrile.
(7)
From the second-order rate coefficients k2 and kW at 20, 30
and 40 ЊC, the enthalpy and entropy of activation, ∆H‡ and
∆S‡, for both the assisted (χ = 0.5) and non-assisted oxidation
(χ = 1.0) were calculated and are summarized in Table 1. While
a negative entropy of activation is expected for a bimolecular
mechanism due to the loss of translational and rotational
degrees of freedom, the presumed cyclic activated complex
(Scheme 2) involving peroxyacetimidic acid would be highly
Scheme 1
From eqns. (3) and (4), eqn. (5) follows and thus the practical
rate law in eqn. (6) is derived, where kW << k2Ј, and kW may be
assumed to decrease further with decreasing solvent polarity,
and therefore eqn. (7) holds.
Scheme 2
ordered and therefore possess still less vibrational entropy.
Unexpectedly, the assisted oxidation of pteropodine 1 has a less
negative entropy of activation and is exceptional in this respect
compared with the other amines. The activated complex is
obviously perturbed in this case, which is possibly due to inter-
action with the hydrogen bond donor carbonyl group on the
same side of the molecule as the lone pair of the nitrogen atom.
The epimer 2 shows the expected behaviour. In general, it can
be seen that, although the loss of entropy in the activation step
is large, the enthalpies of activation of the reactions in water–
acetonitrile mixtures are much lower than those in water, which
suggests an early (reactant-like) transition state and causes the
overall higher rates of the assisted reactions. Surprisingly, at
χ = 0.1 and 20 ЊC the oxidation of pteropodine 1 becomes
pseudo-zero-order (k1/sϪ1 = 10Ϫ9), the rate being independent of
substrate concentration, whereas the oxidation of isopteropo-
dine 2 is immeasurably slow under these conditions. A likely
(2)
[CH C(᎐NH)OOH] = K[CH CN][H O ]
(3)
(4)
(5)
᎐
3
3
2
2
k2K[CH3CN] = k2Ј
k [CH C(᎐NH)OOH] = k Ј[H O ]
᎐
2
3
2
2
2
(6)
J. Chem. Soc., Perkin Trans. 2, 2001, 864–868
865

View Online
Table 1 Second-order rate coefficients and activation parameters
χ = 1.0 kW/10Ϫ5 dm3 molϪ1 sϪ1
Amine
T = 293 K
303 K
313 K
∆H‡/kJ molϪ1
∆S‡/JmolϪ1K Ϫ1
Ϫ129
Ϫ139
Ϫ254
Ϫ95
Pteropodine 1
7.6
1.0
1.2
1.6
3.2
0.30
20
36
57
59
25
71
52
36
Isopteropodine 2
2.3
1.8
4.5
6.5
0.50
5.1
2.4
11
13
0.83
Pyridine 3
2-Morpholinoethyl N-phenylcarbamate 4
N,N-Dimethylaniline 5
N,N-Dimethylbenzylamine 6
Ϫ153
Ϫ229
χ = 0.5 k2/dm3 molϪ1 sϪ1
Pteropodine 1
Isopteropodine 2
Pyridine 3
2-Morpholinoethyl N-phenylcarbamate 4
N,N-Dimethylaniline 5
N,N-Dimethylbenzylamine 6
4.7
0.037
0.052
0.40
0.68
2.27
8.3
11.6
0.091
0.056
1.15
1.34
2.47
43
31
0.76
37
23
Ϫ83
Ϫ165
Ϫ267
Ϫ125
Ϫ169
Ϫ235
0.065
0.054
0.65
1.01
2.33
0.77
explanation could be that (1) the low solvent polarity and (2)
the low temperature decrease the rate of formation of the
reactive intermediate. Acetamide, which arises as a secondary
product, was identified and quantified by GC-MS and shown to
be present in equimolar amounts based on amine consumed.
peroxide.7 Thermostatted mixtures of acetonitrile and aqueous
buffer pH 7 (χ = 0.5) were equilibrated with two different con-
centrations of hydrogen peroxide. Then a solution of potassium
iodide was added, and the iodine liberated was determined
by measuring the absorbance at 352 nm every 10 seconds.
Peroxyacetimidic acid concentration was calculated from the
absorbance at zero time by second-order extrapolation. It was
confirmed that, without the equilibration period prior to the
measurement, no peroxy acid species could be detected. 3-
Chloroperbenzoic acid was used for calibration. It was also
established that the resulting 3-chlorobenzoic acid does not
interfere with the measurement at 352 nm. The accuracy of the
method was estimated as 2.5% using known amounts of 3-
chloroperbenzoic acid in the presence of a 20-fold excess of
hydrogen peroxide. The determinations were carried out at 20,
30 and 40 ЊC. The equilibrium constants (Scheme 1) K × 104 at
these temperatures were calculated as 2.0, 3.3 and 5.2, respect-
ively. Thus, the molar heat change of reaction ∆HR between
acetonitrile and hydrogen peroxide was found to be 35 kJ molϪ1
under these conditions. Overnight equilibration was neces-
sary for the mixtures containing methanol or tert-butyl alcohol
to form the peroxyimidic acid, and K × 104 at 30 ЊC was
approximately 1.
Other tertiary amines
In order to prove that the acetonitrile-assisted oxidation of ter-
tiary amines in neutral solution is a general phenomenon, the
oxidations of some other tertiary amines were studied. The
substrates were selected to represent as wide a scope as possible.
Pyridine 3 was chosen as an aromatic heterocyclic amine. In
industry, N-alkylmorpholine N-oxides are important solvents
for cellulose. Therefore a representative of this class of com-
pounds was of interest, and the phenylcarbamate of 2-
morpholinoethanol 4 was used to enable UV detection. Again,
the typical dependence of the rate coefficient on the mole frac-
tion χ was observed (Fig. 1c, d). In addition, the oxidations of
N,N-dimethylaniline 5 and N,N-dimethylbenzylamine 6 as
further examples of distinct structural classes were examined
at χ = 0.5 and χ = 1.0. All these substrates showed a similar
behaviour. Rate coefficients and activation parameters are given
in Table 1. In all cases, the enthalpy of activation is consider-
ably lower for the nitrile-assisted oxidation which explains the
increased rate.
Discussion
Peroxybenzimidic acid or its anion peroxybenzimidate has
been proposed as a reactive intermediate in the conversion of
benzonitrile to benzamide by alkaline hydrogen peroxide.8 More
recently, spectroscopic evidence has been put forward which
supports the existence of peroxyacetimidate as an intermediate
in the reaction of hydrogen peroxide, acetonitrile and alkali.9,10
Admittedly, the photometric assay used in this work does not
prove the structure but it does prove the presence of a peroxy
acid species. Little doubt remains that the active oxidant is
indeed peroxyacetimidic acid.
It is noteworthy that the oxidations described in this report
were carried out in phosphate-buffered solution at pH 7 (phos-
phate ions reportedly suppress the Radziszewski hydrolysis of
nitriles,11 and eliminate metal ion involvement12), whereas the
literature so far describes nitrile-assisted oxidations of amines
only in alkaline solutions. For instance, pyridine was oxidized
by hydrogen peroxide in the presence of benzonitrile in meth-
Effects of co-solvents and pH
The rates of the assisted oxidations showed a dependence on
solvent polarity similar to that of the non-assisted reactions
when methanol or tert-butyl alcohol was added. When the
oxidation of pteropodine 1 was performed in a mixture of
acetonitrile and methanol (75:25, same molar acetonitrile con-
centration as in χ = 0.5) at 30 ЊC it became pseudo-zero-order
(k1/sϪ1 = 10Ϫ7). This implies that the formation of peroxyacet-
imidic acid becomes rate-determining under these conditions.
N,N-Dimethylbenzylamine 6 showed a similar behaviour,
whereas isopteropodine 2, pyridine 3, 2-morpholinoethyl N-
phenylcarbamate 4 and N,N-dimethylaniline 5 did not react at
all in this solvent mixture. It was ascertained that the rate of
oxidation of 1 and 2 at 30 ЊC (χ = 1.0) is not significantly differ-
ent at pH 6 and 8. The assisted oxidation of 1 at 30 ЊC (χ = 0.5)
did not show any pH-dependence when pH 6, 7, 8 or 9
buffers were used, either. It was also demonstrated that the
buffer concentration (0.005, 0.010 M) does not affect the rate.
anol only after the addition of sodium hydroxide solution.13
A
likely explanation for the observations in the literature13 is that
methanol was used as the solvent. Solvents which are less polar
than water have now been shown (1) to decrease the rate of
formation of the active oxidant and (2) to decrease the rate of
oxidation in general. In fact, the oxidation of pyridine by hydro-
gen peroxide in acetonitrile is started by addition of water.
Detection and quantification of peroxyacetimidic acid
The presumed peroxyacetimidic acid was detected using a
modified spectrophotometric method which allows determin-
ation of peracids in the presence of a large excess of hydrogen
866
J. Chem. Soc., Perkin Trans. 2, 2001, 864–868

View Online
It has also been reported that the effect of pH on the rate of
epoxidation of styrenes with a mixture of nitrile and hydrogen
peroxide is less than expected.11 In the present work it has been
found that the oxidation of tertiary amines, assisted or not, is
not affected by pH in the range 6 to 9, either.
For alkene epoxidation in 75% methanol, the oxidation step
is rate-determining and the pre-equilibrium is attained, whereas
the oxidation is fast for the case of dimethyl sulfoxide, a much
more reactive substrate than alkenes, and hence the rate is gov-
erned by the formation of peroxyimidic acid.14 In contrast,
epoxidation in methanol has been reported to be independent
of alkene structure, indicating that the slow step has to be the
formation of the intermediate.13 It has now been demonstrated
that, for the oxidation of tertiary amines, the experimental con-
ditions, i.e. solvent polarity and temperature, determine which
step becomes rate-limiting.
In the absence of a substrate, the intermediate peroxyimidic
acid has been postulated to react with the peroxide anion to
yield amide and oxygen, the classic Radziszewski hydrolysis of
nitriles.13 However, it has been concluded that this cannot be a
major pathway because the overall rate should be second-order
in hydrogen peroxide which is not the case.14 An experiment
with doubly isotope-labelled H218O2 has also refuted this mech-
anism.12 Rather, at high alkalinity (pH > 10), the peroxyimidate
decomposes homolytically, inducing radical decomposition of
hydrogen peroxide.14
Finally, it must be noted that the described kinetic experi-
ments were conducted in very dilute solutions which show a
reasonably ideal behaviour, whereas at higher amine concentra-
tions the kinetics become less transparent; for instance, a 3/2-
order in H2O2 has been observed.15 Similarly, the decrease
of rate of styrene epoxidation with increasing styrene concen-
tration has been attributed to the decrease of polarity of the
reaction mixture.11 Moreover, in acetonitrile-rich mixtures the
hydrogen ion activity is significantly increased16 which may
contribute to the sudden drop of the rate at χ < 0.2.
statted ( 0.1 K) reaction mixture by HPLC analysis, using an
RP-18 (5 µm) column (125 × 4 mm id, Merck). A mixture of
acetonitrile and 0.01 M aqueous phosphate buffer pH 7.0
(45:55) with a flow of 1.3 cm3 minϪ1 at 80 ЊC was used as eluent
for compounds 1 (tR 3.0 min), 2 (tR 4.4 min), 7 (tR 1.3 min), 8
(tR 1.4 min), and a mixture of methanol and buffer (45:55) with
a flow of 1.0 cm3 minϪ1 at 50 ЊC for compounds 3 (tR 2.2 min),
4 (tR 3.3 min), 9 (tR 1.4 min), and 10 (tR 2.0 min) with detection
at 247 nm. A mixture of methanol and buffer (55:45) with a
flow of 1.0 cm3 minϪ1 at 50 ЊC was used for compounds 5 (tR
4.5 min), 6 (tR 6.2 min), 11 (tR 1.5 min), and 12 (tR 1.7 min)
with detection at 205 nm. Aliquots of the reaction mixture
were diluted with buffer (1:10) immediately prior to analysis.
Relative standard deviation (RSD) of the primary HPLC
measurements was 0.8%. In each experiment, 3 to 5 data points
were used to determine a single rate coefficient k1 using the
absolute values of peak areas, and 3 to 4 concentrations of
hydrogen peroxide were used to obtain k2Ј values. The kinetic
experiments were reproduced 3 to 7 times, and RSD of k2Ј was
≤10%.
Photometric determination of peroxyacetimidic acid
2 cm3 of a mixture of acetonitrile and 0.005 M phosphate
buffer pH 7 (75:25 by volume, χ = 0.5) were thermostatted
( 0.1 ЊC) in a quartz cuvette, 20 or 40 µl of 0.3% hydrogen
peroxide added, and the mixture equilibrated for at least 10
min. Then, 1 cm3 thermostatted 0.2 M potassium iodide
solution was added and the photometric program started. The
wavelength was set at 352 nm. The first measurement was
recorded 20 s after mixing, and then every 10 s for as long as
80 s. An identical mixture of acetonitrile, buffer and iodide
solution was used as a blank. From these data the absorbance
at zero time was extrapolated and used for the calculation of
peroxyacetimidate concentration. Six determinations were
made at 20, 30 and 40 ЊC each. An eight-level calibration
in the range 0.5 to 20 µmol dmϪ3 was performed using 3-
chloroperoxybenzoic acid, the potency of which was assayed by
titration. Finally, from the peroxyacetimidate concentration
and the known concentrations of acetonitrile and hydrogen
peroxide, the equilibrium constants K were calculated. Cali-
brations were also performed in methanol and tert-butyl
alcohol mixtures which were used for the determination of K.
Thus, from a synthetic point of view, the optimum conditions
for the oxidation of tertiary amines by hydrogen peroxide to
N-oxides are (1) a solvent system of high polarity, (2) pH higher
than 6 but lower than 10, and (3) the mole fraction χ of
acetonitrile at least 0.2 but not higher than 0.7.
These results shed new light on previous work and provide
more insight into the mechanistic aspects of the nitrile–
hydrogen peroxide–solvent–substrate system.
Synthetic procedures
2-Morpholinoethyl N-phenylcarbamate 4. 2-Morpholino-
ethanol (656 mg, 5.0 mmol) was refluxed with phenyl iso-
cyanate (667 mg, 5.6 mmol) in 50 cm3 hexane for 2 h. The
product crystallized readily upon cooling. Yield: 1.01 g (81%);
mp 73 ЊC, δH 2.52 (4H, AAЈ), 2.67 (2H, t, J 6), 3.73 (4H, BBЈ),
4.30 (2H, t, J 6), 6.76 (1H, br s), 7.06 (1H, tt, J 7, J 2), 7.26–7.40
(4H, m) ppm.
Experimental
Materials
The alkaloids 1 and 2 were obtained by supercritical fluid
extraction of Uncaria tomentosa root with CO2, followed by
conventional acid–base work-up and separation by column
chromatography (silica gel, EtOAc–hexane = 9:1). All other
reagents were of analytical grade (Merck, Darmstadt). Con-
Pteropodine N-oxide 7. Pteropodine 1 (90 mg, 0.24 mmol)
was dissolved in 5 cm3 acetonitrile, 5 cm3 30% hydrogen per-
oxide were added, and the mixture was allowed to stand at
22 ЊC. After 12 h, 30 cm3 water and 20 mg manganese dioxide
were added. The excess hydrogen peroxide was destroyed and
acetonitrile was distilled off under reduced pressure. The
remaining aqueous solution was cooled and extracted five
times with dichloromethane. The extracts were combined,
dried over anhydrous sodium sulfate and evaporated. The resi-
due was chromatographed on silica gel using ethyl acetate–
methanol = 1:1 as the eluent. The respective fraction (Rf = 0.23)
crystallized upon concentrating and yielded 70 mg pteropodine
N-oxide (75%); mp 173 ЊC (dec.), [α]2D1 = Ϫ79Њ (c = 1.02 in
CH2Cl2), δH 1.51 (3H, d, J 6), 1.80 (1H, m), 1.90 (1H, d, J 12),
2.0 (1H, m), 2.18 (1H, m), 2.30 (1H, ddd, J 13, J 13, J 13), 2.53
(1H, m), 3.24 (1H, m), 3.55 (1H, m), 3.62 (3H, s), 3.7 (1H, m),
4.32 (1H, m), 4.38 (1H, m), 4.98 (1H, qd, J 6, J 10), 6.92 (1H, d,
1
centration of H2O2 was determined iodometrically. H-NMR
spectra were recorded in CDCl3 at 200 MHz (J values in Hz).
Optical rotation was measured using a Perkin-Elmer 141
polarimeter and is given in units of 10Ϫ1 deg cm2 gϪ1. Peroxy-
acid assays were performed on
spectrophotometer.
a Shimadzu UV-160A
Kinetic procedures
Typically, 10 Ϫ A cm3 reagent mixture (hydrogen peroxide,
0.005 M phosphate buffer pH 7, optional co-solvent, with or
without acetonitrile) were pre-equilibrated for 10 min, and A
cm3 of concentrated amine solution (in buffer, acetonitrile or
co-solvent as required) were added in order to obtain a final
amine concentration of 10Ϫ4 to 10Ϫ5 mol dmϪ3 in a volume of
10 cm3. Oxidations were followed by monitoring the thermo-
J. Chem. Soc., Perkin Trans. 2, 2001, 864–868
867

View Online
J 7), 7.07 (1H, t, J 7), 7.23 (2H, m), 7.56 (1H, s), 9.0 (1H, br s)
was obtained (92%); mp 151 ЊC (lit.: 151–152 ЊC).17 The hydro-
ppm.
chloride had a melting point of 123 ЊC after crystallization
from acetone (lit.: 124–125 ЊC).18
Isopteropodine N-oxide 8. Oxidation of isopteropodine 2 (55
mg, 0.15 mmol) as described for pteropodine afforded 37 mg
isopteropodine N-oxide (65%); mp 210 ЊC (dec.), [α]2D1 = Ϫ44Њ
(c = 0.75 in MeOH), δH 1.51 (1H, m), 1.55 (3H, d, J 6), 1.77 (1H,
td, J 5, J 10), 1.93 (1H, q, J 13), 2.39 (1H, dd, J 9, J 12), 2.69
(1H, td, J 5, J 12), 2.80 (1H, td, J 12, J 7), 3.54 (1H, dd, J 12,
J 5), 3.61 (3H, s), 3.62 (1H, m), 3.75 (1H, dd, J 2, J 12), 4.00
(1H, t, J 9), 4.13 (1H, d, J 12), 5.06 (1H, dq, J 10, J 6), 6.83
(1H, d, J 7), 7.00 (1H, t, J 8), 7.18 (1H, t, J 8), 7.55 (1H, s), 8.01
(1H, d, J 7), 9.84 (1H, s) ppm.
N,N-Dimethylbenzylamine N-oxide 12. N,N-Dimethylbenzyl-
amine 6 (1.35 g, 10 mmol) was dissolved in 50 cm3 acetonitrile
and 40 cm3 water and heated to 70 ЊC. Then, 10 cm3 30%
hydrogen peroxide (87 mmol) were added dropwise with
stirring. After distilling off the acetonitrile under reduced
pressure, the solution was extracted several times with chloro-
form. The extracts were dried over anhydrous sodium sulfate
and evaporated. According to TLC, HPLC and IR, the isolated
compound was identical with the N-oxide prepared by a
literature procedure.19 The residue was redissolved in concen-
trated hydrochloric acid and evaporated again. The N,N-
dimethylbenzylamine N-oxide hydrochloride crystallized from
water–dioxane (1:1) as needles. Yield: 1.21 g (80%); mp 121 ЊC
(lit.: 121–122.5 ЊC).20
Pyridine N-oxide 9. Pyridine 3 (1.0 g, 12.6 mmol) was dis-
solved in 2 cm3 acetonitrile and 8 cm3 water and heated to
75 ЊC. Then, 3 cm3 30% hydrogen peroxide (26 mmol) were
added dropwise with stirring. The reaction mixture was diluted
with 20 cm3 water and extracted three times with 20 cm3
dichloromethane. The extracts were dried over anhydrous
sodium sulfate and evaporated. Pure pyridine N-oxide was
obtained after chromatography on silica gel (ethyl acetate–
methanol = 1:1) in yields from 80 to 90% of the theoretical
amount. According to TLC, HPLC and IR, the product was
identical with commercial pyridine N-oxide; mp 61 ЊC.
References
1 K. Keplinger, G. Laus, M. Wurm, M. P. Dierich and H. Teppner,
J. Ethnopharmacol., 1999, 64, 23.
2 G. Laus, D. Brössner, G. Senn and K. Wurst, J. Chem. Soc., Perkin
Trans. 2, 1996, 1931.
3 C. J. Toney, F. E. Friedli and P. J. Frank, J. Am. Oil Chem. Soc., 1994,
71, 793.
4 A. A. Oswald and D. L. Guertin, J. Org. Chem., 1963, 28, 651.
5 D. Bethell, A. E. Graham, J. P. Heer, O. Markopoulou, P. C. B. Page
and B. K. Park, J. Chem. Soc., Perkin Trans. 2, 1993, 2161.
6 W. E. Parker, L. P. Witnauer and D. Swern, J. Am. Chem. Soc., 1957,
79, 1929.
7 D. M. Davies and M. E. Deary, Analyst, 1988, 113, 1477.
8 K. B. Wiberg, J. Am. Chem. Soc., 1953, 75, 3961.
9 V. Vaque, N. Dupuy, B. Sombret, J. P. Huvenne and P. Legrand,
Appl. Spectrosc., 1997, 51, 407.
10 V. Vaque, N. Dupuy, B. Sombret, J. P. Huvenne and P. Legrand,
J. Mol. Struct., 1997, 555.
2-Morpholinoethyl N-phenylcarbamate NЈ-oxide 10. 2-
Morpholinoethyl N-phenylcarbamate 4 (100 mg, 0.40 mmol)
was dissolved in 2 cm3 acetonitrile, 2 cm3 water and 0.1 cm3 30%
hydrogen peroxide (0.87 mmol) were added, and the mixture
was stirred at 70 ЊC for 5 h. The mixture was diluted with 10 cm3
water and extracted three times with dichloromethane. The
extracts were dried and concentrated under reduced pressure.
The product was precipitated with diethyl ether and dried.
Yield: 80%; mp 190 ЊC (dec.), δH (CDCl3) 3.22 (2H, d, J 11),
3.38 (2H, td, J 11, J 3), 3.60 (2H, AAЈ, J 5), 3.76 (2H, dd, J 11,
J 1), 4.44 (2H, td, J 11, J 1), 4.89 (2H, BBЈ, J 5), 7.06 (1H, t,
J 7), 7.31 (2H, td, J 7, J 1), 7.53 (2H, d, J 7), 9.18 (1H, br s)
ppm.
11 Y. Ogata and Y. Sawaki, Bull. Chem. Soc. Jpn., 1965, 38, 194.
12 J. E. McIsaac, R. E. Ball and E. J. Behrman, J. Org. Chem., 1971, 36,
3048.
13 G. B. Payne, P. H. Deming and P. H. Williams, J. Org. Chem., 1961,
26, 659.
N,N-Dimethylaniline N-oxide 11. N,N-Dimethylaniline 5
(1.21 g, 10 mmol) was dissolved in 50 cm3 acetonitrile and 40
cm3 water and heated to 70 ЊC. Then, 10 cm3 30% hydrogen
peroxide (87 mmol) were added dropwise with stirring. The
reaction was complete after 2 h, as indicated by HPLC. After
distilling the acetonitrile under reduced pressure, the solution
was extracted several times with chloroform. The extracts were
dried and evaporated. After chromatography on silica gel (ethyl
acetate–methanol = 1:1) 1.26 g N,N-dimethylaniline N-oxide
14 Y. Sawaki and Y. Ogata, Bull. Chem. Soc. Jpn., 1981, 54, 793.
15 J. C. Gee and R. C. Williamson, J. Am. Oil Chem. Soc., 1997, 74,
65.
16 C. Kalidas, G. Hefter and Y. Marcus, Chem. Rev., 2000, 100, 819.
17 J. P. Ferris, R. D. Gerwe and G. R. Gapski, J. Org. Chem., 1968, 33,
3493.
18 O. Shigeru, T. Kitao and Y. Kitaoka, J. Am. Chem. Soc., 1962, 84,
3366.
19 A. C. Cope and P. H. Towle, J. Am. Chem. Soc., 1949, 71, 3423.
20 H. Hoffmann, Arch. Pharm., 1971, 304, 614.
868
J. Chem. Soc., Perkin Trans. 2, 2001, 864–868