Rate-determining nitrogen inversion in the isomerisation of isoimides to imides and azides to tetrazoles: Direct observation of intermediates stabilized by trifluoroethyl groups

Article abstract of DOI:10.1039/b202005j

Reaction of N-alkyl- and N-trifluoroalkylbenzimidoyl chlorides 2 in the presence of either acetate or azide ion leads initially to the formation of the corresponding isoimides 4(Z) and imidoyl azides 8(Z) (both of which could be observed spectroscopically

Full text of DOI:10.1039/b202005j

View Article Online / Journal Homepage / Table of Contents for this issue
Rate-determining nitrogen inversion in the isomerisation of
isoimides to imides and azides to tetrazoles: direct observation
of intermediates stabilized by trifluoroethyl groups
2
Anthony F. Hegarty,* Nuala M. Tynan and Suzanne Fergus
Chemistry Department, University College Dublin, Dublin 4, Ireland
Received (in Cambridge, UK) 26th February 2002, Accepted 8th May 2002
First published as an Advance Article on the web 30th May 2002
Reaction of N-alkyl- and N-trifluoroalkylbenzimidoyl chlorides 2 in the presence of either acetate or azide ion
leads initially to the formation of the corresponding isoimides 4(Z) and imidoyl azides 8(Z) (both of which could
be observed spectroscopically). These are formed via nucleophilic trapping of the nitrilium cations 3 in aqueous
organic solutions. Subsequently these imidoyl acetates and azides rearrange to the more stable imides 5 or tetrazoles
9. These rearrangements are characterised by a low dependence on solvent polarity and insensitivity to added salts,
indicating that the rate-determining step is the isomerisation of the initially formed Z-isomer of the imine to the
E-isomer (imine nitrogen inversion) rather than the subsequent N O acyl group transfer or cyclisation. The
Hammett ρ value (Ϫ0.4), obtained for the rearrangement of the imidoyl azides to the tetrazoles, compares well to
other systems where the rate-determining step (nitrogen inversion) was similar. Nitrogen inversion in these imine
systems is therefore significantly slower (ca. 10-fold relative to an ethyl group) in the presence of the trifluoroethyl
group on nitrogen.
Introduction
Imines 1 exist in two stereoisomeric forms (Scheme 1), in direct
Scheme 2
Scheme 1
analogy to alkenes, with the cis (Z) and trans (E) forms. Unlike
most alkyl- or aryl-substituted alkenes, the ground state inter-
conversion of imines may be observed spectroscopically in
some cases at room temperature in the absence of added
catalysts. However, the rates of interconversion vary a great
deal, depending on the substituents.1 The influence of substitu-
ents on the imino nitrogen is markedly greater than that of
substituents at carbon; this is especially pronounced for
heteroatom substituents2 whose presence can lead to a very
significant reduction in the isomerisation rates. Typically, for
aryl-substituted compounds, the reaction constant at carbon is
less than Ϫ0.5 and at nitrogen greater than 1.0, that is, electron-
withdrawing substituents in direct conjugation with the imino
nitrogen have an accelerating effect, whereas substituents on the
aryl groups on the imino carbon have the opposite effect.1
Bulky substituents attached to nitrogen may also have a strong
accelerating effect in these reactions.3
The detailed mechanism of interconversion has been the sub-
ject of considerable debate. The reaction is relatively insensitive
to solvent polarity,4 which strongly favours the lateral shift
mechanism, involving a nitrogen lone pair inversion, as shown
in Scheme 2. The alternative torsional mechanism which
Scheme 3
intermediates could not be observed in most cases, these
reactions show behaviour which is typical of other C᎐N bond
isomerisation reactions, including the observation of acid
᎐
catalysis.2b,8
The use of heteroatoms, such as nitrogen or oxygen, attached
directly to the imine nitrogen in order to slow nitrogen inversion
has been successful in many cases. However this strategy does
have the disadvantage that the N–heteroatom bond may be
weak and subsequent cleavage of this bond may occur readily,
leading to the loss of the useful stereochemical information
incorporated in the original reaction.
involves C᎐N bond rotation, as shown in Scheme 3, may be
᎐
important as a contributor or even the dominant pathway in
We now report that the introduction of the N-trifluoroethyl
group permits the observation of the intermediates formed on
trapping of simple N-alkylnitrilium ions by both azide and
acetate ions and that their subsequent reactions can also
be followed; the key factor in allowing the observation of the
some cases.5
Previous investigations demonstrated that Z to E inter-
conversion could be the rate-determining step in a number of
reactions such as intramolecular acyl transfer.6,7 Although the
1328
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b202005j

View Article Online
Table 1 First-order rate constants at 25 ЊC for rearrangement of iso-
imides 4(Z) and azides 8(Z) formed upon reaction of imidoyl chlorides
2 with acetate or azide nucleophiles, in 6 : 4 water–dioxane
individual steps in the formation of tetrazoles and imides in
these systems is the significant slowing of nitrogen inversion by
the trifluoroethyl group, which is quantified by this study.
Rate constant k/sϪ1
R1
R2
R3
Imides 4(Z)a
2.77 × 10Ϫ4
Azides 8(Z)b
5.86 × 10Ϫ3
Results and discussion
Formation of nitrilium ions 3 from imidoyl halides 2
–CF3
–H
The N-trifluoroethyl- and N-isopropylimidoyl chlorides 2a to
2e were shown to undergo hydrolysis to the amide 7 in aqueous
dioxane (1,4-dioxane) via rate-determining loss of chloride ion
to form the nitrilium cation 3.9 The observation of strong
common ion10 and solvent effects confirms the presence of the
nitrilium cation as the initial intermediate in these reactions.
–CH3
–CH3
–H
2.62 × 10Ϫ3
4.55 × 10Ϫ3
—
–CH3
2.91 × 10Ϫ2
(a) Common ion effect. The rates of hydrolysis of the imidoyl
chlorides 2 were found to be sensitive to their initial concen-
tration; even when 2 was reacted at 5 × 10Ϫ4 M, the chloride
ion released in the course of the reaction was sufficient to slow
the reaction appreciably. At lower concentrations (5 × 10Ϫ5 M),
the rate was, however, independent of the concentration
of substrate. The presence of the common ion effect con-
firms that the reaction proceeds through an intermediate
with the loss of chloride ion to give the nitrilium cation 3 in the
rate-determining step.
–CF3
–CF3
–H
–H
—
—
6.22 × 10Ϫ3
8.24 × 10Ϫ3
a Ionic strength 1.0 M (NaClO4). b Ionic strength 0.1 M (NaN3). The
substrate concentrations were 10Ϫ4 M; the acetate-trapped products
were all measured at 300 nm while the azides were measured at different
wavelengths as follows: (c) 270, (d) 260 and (e) 300 nm. Rate constants
were based on the mean of three values.
(b) Solvent effect. Studies of the hydrolysis of these imidoyl
chlorides provides additional evidence for the presence of the
nitrilium cation 3 in the course of these reactions. Rate con-
stants were obtained in various water–dioxane solvent mixtures
at 25 ЊC. Plotting log kobs against the corresponding Y values of
Fainberg and Winstein11 gave the following m values: 1.22
(for 2a); 1.61 (2c); and 1.11 (2e).
Table 2 Rate constants for the rearrangement of the isoimide 4(Z)a to
the imide 5a at 25 ЊC in 80 : 20 dioxane–water
[NaOAc]/M
104k/sϪ1
The m value measures the susceptibility of the substrate to
the ionising power of the solvent mixture. The high m values
reflect the fact that the positive charge formed is not highly
0.8
0.5
0.1
0.04
4.28
4.25
4.17
4.32
ϩ
᎐
delocalised, but is concentrated mainly on the –C᎐N – group
᎐
itself.
of (a) changing the concentration of acetate ion and (b) water
content (Table 2). In both cases the reaction being studied [con-
version of 4(Z) to 4(E)] was essentially independent of acetate
ion and on changing the solvent content from 80 : 20 dioxane–
water to 60 : 40 dioxane–water a modest 2-fold rate decrease
was observed.
Trapping of 3 by acetate anion to form isomides 4(Z)
Having established the presence of nitrilium ions as intermedi-
ates we have found that, in the presence of acetate as a nucleo-
phile, the nitrilium cations 3 undergo selective trapping.12 We
now report that, in the presence of carefully selected concen-
trations of sodium acetate, the nitrilium ions 3a–3e are trapped
by acetate ion to give intermediates 4(Z), which we have now
found can be detected when the nitrogen carries a trifluoroethyl
group. The solvent used in these trapping reactions contained a
high percentage of water in order to ensure that the ionisation
of 2 to 3 is rapid. Although these isoimide intermediates 4(Z)
have not been isolated, they are stable enough such that their
appearance and further reactions can be monitored using
conventional UV and FTIR spectroscopy and HPLC analysis
(see Scheme 4).
The trapping of the nitrilium cations 3 with acetate anion
was also monitored using FTIR spectroscopy (see Fig. 1).
Contrary to a recent report,13 structures [such as 4(Z)] analo-
gous to the acetate-trapped products of these nitrilium cations
have been previously reported and documented extensively.12
Rates of rearrangement of 4(Z) to 4(E). The isoimides 4(Z)
then undergo a slower subsequent reaction (identified as the
conversion to the imides 5). The rate constants for this reaction
could be measured either from HPLC relative peak areas for
4, 5 and 7 or from the change in optical density using UV
spectroscopy and are reported in Table 1. The final products
formed from 4 were confirmed by isolation, and also from UV
and HPLC analysis of authentic samples, as the imides 5.
Confirmation that the reaction being studied was rearrange-
ment of 4(Z) to 4(E) rather than, say, formation of the nitrilium
ion 3 or the isoimide 4(Z), was obtained by studying the effect
Fig. 1 IR spectrum of the Z-isomer 4a, formed by trapping the
nitrilium cation 3a using acetate ion.
Formation of the Z-isomer 4a was observed at 1774 cmϪ1
,
which is in the same region as that observed for vinylic
acetates14 and O-acylisoureas.15 Absorbances at 1720 and 1683
cmϪ1, assigned to the imide product 5a, were also observed.
The IR spectrum of the Z-isomer 4a and its rearrangement
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334
1329

View Article Online
Scheme 4
Table 3 Observed rate constants for the reaction of the isoimide 4(Z)
to 5a as a function of time are shown in Figs. 1 and 2,
respectively.
formed from 2a (1 × 10Ϫ4 M) at 25 ЊC, 60 : 40 water–dioxane and in the
presence of acetic acid–acetate buffer (ionic strength 1.0 M) measured
at 300 nm
Having established the structures of these intermediate
acetates 4(Z), the less stable intermediates carrying an N-alkyl
group were also successfully examined (see Table 1). Overall
the presence of an N-trifluoroethyl group reduces the rate of
isomerisation ca. 10-fold (relative to an N-ethyl group).
[NaOAc]/M
[AcOH]/M
104kobs/sϪ1
0.66
0.50
0.33
0.33
0.50
0.66
7.60
14.8
26.5
Effect of acid on trapping and isomerisation. In the presence
of acid the rate of reaction of the isoimide 4(Z) increases (see
Table 3). Subsequent HPLC product analysis showed that
under these conditions the amide 7 (retention time 6.91 min)
was formed in addition to the imide 5 (retention time 9.79 min),
from the intermediate isoimide 4 (retention time 14.16 min)
(see Fig. 2).
As the overall rate of reaction of 4 increases with increasing
acid, the amide 7 becomes correspondingly the major product,
formed from the isoimide by acid-catalysed ester hydrolysis
(giving as final products up to 90% amide, 10% imide). In
theory, acid-catalysed rearrangement of the isoimide (4) to
1330
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334

View Article Online
Trapping by azide to form imidoyl azides 8(Z)
Using azide ion as nucleophile, the nitrilium ions 3 were selec-
tively trapped to give the imidoyl azides 8(Z) as intermediates.
The appearance of these intermediates was identified by IR
spectroscopy with absorbances at 2136 and 2038 cmϪ1 when a
2,2,2-trifluoroethyl group was present on nitrogen (8a). The
intensities of these absorbances were shown to decrease with
time (Fig. 4) as the azide 8a was converted to the corresponding
Fig. 2 Graph showing the relationship between the imide 5a, isoimide
4a and amide 7a with time in 0.04 M sodium acetate, at an ionic
strength of 1.0 M (NaClO4).
imide 5 is possible since acid catalysis of imine isomerisation
is known. However, the product analysis shows that the
formation of 7 rather than 5 is enhanced by acid, which rules
this out.
Fig. 4 IR spectrum of the Z-isomer 8a formed when the nitrilium ion
3a is trapped using azide ion. The initial 16 scans are shown and also
the final scan.
Below pH 4 the trapping of the nitrilium ion 3 cannot be
carried out successfully with HOAc–AcOϪ since the concen-
tration of the acetate ion (which is the active trap) becomes
much reduced relative to acetic acid concentration. Similarly,
at very low acetate concentrations (<0.02 M) larger percen-
tages of amide 7 were formed; this is presumably because
under these conditions there is not enough acetate present
to trap the nitrilium ion formed, relative to the reaction
of 3 with water (giving the amide 7). At much higher concen-
trations of acetate (1.0 M), even at pH 7, the amide 7 was
also produced on reaction with 3, probably by attack of
the acetate ion on the isoimide 4, which was formed initially.
We have found that the best region for maximising the
degree of formation of the isoimide 4 and minimising
the subsequent competitive side reactions is 0.04–0.1 M sodium
acetate.
tetrazole 9a. D2O was used for the IR measurements since H2O
interfered with the spectral range being studied.
A rate constant of 5.6 × 10Ϫ3 sϪ1 was calculated from the UV
spectroscopy measurements (measured in 6 : 4 H2O–dioxane).
Both reactions (formation and subsequent rearrangement of
8a) were measured at 25 ЊC. A plot of absorbance versus time is
shown in Fig. 5.
Stability of the imides 5. It was also observed that the imide
products 5 were not stable in the reaction medium. Further
reaction of the imide with acetate ion occurs giving the amide 7
as product when it is left standing for long periods of time. For
example, only 15% of the imide 5a remains when a pure sample
is left standing in 1 M sodium acetate for 5 h. Fig. 3 shows a plot
of a time course of this reaction for 5a, using HPLC to analyse
the products.
Fig. 5 Plot of absorbance versus time for the isomerisation of
Z-imidoyl azide 8a to the tetrazole 9a.
A three-point Hammett plot for the conversion of 8(Z) to 9
(obtained using substituents p-NO2, H, p-MeO) in 80 : 20
dioxane–water gave a ρ value of Ϫ0.4 for variation of an aryl
substituent located at R1. This value indicates that the conver-
sion of 8(Z) to 9 is only slightly susceptible to electronic effects,
and the negative sign suggests that the reaction is aided by elec-
tron donation, which agrees with previous literature reports on
other reactions where nitrogen inversion (by the lateral shift
mechanism) is thought to be rate-determining.7b The effect of
the water content of the solvent on the rate of isomerisation of
8(Z) is shown in Fig. 6; as the water content increases the rate of
isomerisation decreases slightly. Since the ground state of 8(Z)
is relatively more polar than the transition state, an increase in
the solvent polarity should stabilise the former thus increasing
the barrier to inversion.
Fig. 3 Plot of the reaction of 5a in the presence of 0.08 M sodium
acetate at 25 ЊC.
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334
1331

View Article Online
JEOL JMN-GX270 FT spectrometer. 13C NMR spectra were
recorded at 67.80 MHz on the high field instrument and were
all completely decoupled, unless otherwise stated. In all 13C
spectra a 135 DEPT was carried out. IR spectra were recorded
on a Perkin Elmer 1710 FT spectrometer or on a Mattson
Instruments Galaxy Series FTIR 3000 spectrometer. A VG
Analytical 7070 mass spectrometer, with attached INCOS 2400
data system, in the EI mode, was used for recording mass spec-
tra. Ultraviolet and visible spectra were recorded on a Philips
PU8700 series UV spectrophotometer. Thin layer chromato-
graphy (TLC) was performed on Merck precoated silica gel
60F254 slides. Merck silica 60 (Art 7748) was used for prepar-
ative layer chromatography (PLC), and Merck silica 9385, par-
ticle size 0.04–0.063 mm, for flash column chromatography.
Solvents were dried according to standard literature pro-
cedures. Microanalyses were carried out by the Microanalytical
Laboratory, University College Dublin.
Fig. 6 First-order rate constants for the isomerisation of imidoyl
azides 8a, 8d, 8e and 8c in dioxane–water mixtures.
Kinetic method
Comparison of reactions of 4(Z) and 8(Z)
Reactions were initiated by injecting 10–20 µL of the substrate
solution (ca. 0.01 M) in dry, peroxide-free dioxane into a UV
quartz cuvette containing 2 cm3 of the reaction medium. All
reactions were monitored spectrophotometrically covering
several half lives, at the appropriate wavelength. Pseudo-first-
order rate constants were obtained as the slope from plots of
ln(At Ϫ A) versus the time expressed in seconds, where At is the
absorbance of the solution at any time t, and A is the absorb-
ance at “infinite” time, optimised using an iterative linear least
squares program. Solvent mixtures were made up by mixing the
appropriate volumes. Solvents were HPLC grade, and were
dried when required using standard techniques. HPLC was per-
formed using a C18-u-bondupack column, and a standard UV
detector, 60 : 40 acetonitrile–water was used as solvent at a flow
The observed rate constants for rearrangement of the pre-
formed Z-isomers 4 and 8 were measured over a wide range of
salt and solvent compositions. Rate constants obtained for
these reactions measured in a single solvent, 6 : 4 water–
dioxane, are listed together in Table 1 for easy comparison. The
rate constants for the isomerisation of the azides 8(Z) to the
tetrazoles 9 are approximately an order of magnitude faster
than the corresponding rate constant for rearrangement of the
acetates 4(Z) to 5.
The overall observed behaviour may be summarised using
Scheme 4. Both formation and trapping of the nitrilium cation
3 by the incoming nucleophile are rapid when a highly aqueous
solvent is used; this leads to the formation of the Z-isomer 4 or
8,16 which then isomerises via rate-determining nitrogen inver-
sion to the E-form [4(E) or 8(E)], which then undergoes either
rapid O N acyl group migration or cyclisation.
rate of 0.6 cm3 minϪ1
.
IR analytical method
The trapping reactions of the nitrilium ion 3a were monitored
using a Nicolet Impact 410 FTIR Spectrometer. A solvent mix-
ture of 6 : 4 D2O–CH3CN was employed, as its absorption in
the infrared spectrum did not interfere with the spectral range
being studied. A solution of N-(2,2,2-trifluoroethyl)-p-nitro-
benzimidoyl chloride 2a, in dry CH3CN was added dropwise to
a stirred solution of sodium azide–sodium acetate in D2O–
CH3CN. After the addition was complete the solution (0.1 M
concentration) was injected via a syringe into the IR cell
consisting of CaF2 cell windows with Teflon spacers.
Conclusion
The rate of nitrogen inversion in the imines 4 and 8 is reduced
by ca. 10-fold, relative to an N-ethyl or -isopropyl group, by
the introduction of an N-trifluoroethyl group. This permits the
observation of the intermediate Z-imidoyl azides and acetates
by HPLC and FTIR spectroscopy for the first time. Con-
sequently direct measurement of the rates of nitrogen inversion
can be performed. We conclude that these steps are rate-
determining† in the conversion of 4(Z) to 5 and of 8(Z) to 9,
rather than the subsequent O N acyl group migration or
tetrazole formation, respectively. The trifluoroethyl group,
which slows nitrogen inversion significantly, is therefore a useful
substituent in controlling the rates of reactions that are limited
by this process.
Substrates
The imidoyl chlorides 2 were synthesised by treatment of the
corresponding amides 7 with thionyl chloride or phosphorus
pentachloride as already described.9 The synthesis and physical
properties of a number of the imidoyl chlorides 2a–2d have also
been reported.9
Experimental
N-(2,2,2-Trifluoroethyl)-p-methoxybenzamide (7e). A solu-
tion of freshly distilled triethylamine (4.86 g, 0.048 mol) and
2,2,2-trifluoroethylamine (4.77 g, 0.048 mol) in dry diethyl ether
(100 cm3) was added dropwise, over 30 min, to a vigorously
stirred solution of p-anisoyl chloride (8.22 g, 0.048 mol) in dry
diethyl ether. The reaction mixture was refluxed for 3 h and
upon cooling; the solids were dissolved in water. The ethereal
layer was washed with 1 M hydrochloric acid and brine. After
drying (MgSO4) and concentration of the ethereal layer, the
residual solid was recrystallised from ethanol. A white solid was
obtained, yield, 8.92 g (80%), mp 129–130 ЊC (Found: C, 51.27;
H, 4.33; N, 5.99; F, 24.90. C10H10F3NO2 requires C, 51.51; H,
General
Melting points were determined using either a Gallenkamp
melting point block or a Büchi 530 melting point apparatus and
are uncorrected. 1H NMR spectra were recorded at 60 MHz on
a JEOL JNM-PMX60 spectrometer and at 270 MHz on a
† As noted by a referee, in order to establish that the rate-determining
step is nitrogen inversion, the free energy barrier (∆G) should be evalu-
ated from the rate constants and compared with ∆G values previously
reported for planar inversion in comparable imidates or amidines, see
review by Jennings and Wilson.17 However accurate ∆G values of nitro-
gen inversion for O-acyl imidates do not exist and we also had specific
difficulty in making measurements at variable temperatures, which
would be required in order to estimate activation barriers.
4.32; N, 6.00; F, 24.44%); νmax (KBr)/cmϪ1 1646 (C᎐O), 1156
᎐
(CF3, st); λmax (dioxane)/nm 250 (ε = 13400); δH (270 MHz,
1332
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334

View Article Online
CD3COCD3) 3.98 (3H, s, OCH3), 4.42 (2H, q, J = 8.97 Hz,
CH2), 7.34 (2H, d, J = 6.5 Hz, H3, H5), 8.22 (2H, d, J = 6.5 Hz,
H2, H6); δC (68 MHz, CDCl3) 40.94 (q, JFC = 34.4 Hz, CH2),
55.44 (1C, q, CH3), 124.57 (1C, q, JFC = 278.3 Hz, CF3), 113.56
(2C, d, CH, C3, C5), 129.42 (2C, d, CH, C2, C6), 125.66 (1C, s,
(0.45 g, 0.0017 mol) in Analar acetone (1 cm3) was added to a
stirred solution of sodium acetate (2.78 g, 0.035 mol) and acetic
acid (0.5 cm3) in 1 : 1 acetone–water (12 cm3). After stirring at
room temperature overnight, the acetone was removed and the
aqueous residue was extracted with chloroform. After drying
(MgSO4), the organic layer was removed and the crude product
was purified by column chromatography. Yield, 0.35 g, 72%
(Found: C, 45.80; H, 3.29; N, 9.62; F, 19.67. C11H9F3N2O4
requires C, 45.53; H, 3.12; N, 9.65; F, 19.64%); νmax (KBr)/cmϪ1
C1), 162.67 (1C, s, C4), 167.54 (1C, s, C᎐O); m/z 233 (Mϩ, 10%),
᎐
135 (100), 107 (15), 77 (48).
N-(2,2,2-Trifluoroethyl)-p-methoxybenzimidoyl chloride (2e).
Phosphorus pentachloride (0.13 g, 0.006 mol) was added to a
solution of the amide 7e (0.125 g, 0.006 mol) in dry benzene
(25 cm3). After refluxing for 12 h, the benzene and phosphorus
oxychloride were removed by distillation to yield a yellow oil
which was purified by passing through a layer of Merck silica
7734, particle size 0.063–0.2 mm, using dry CH2Cl2 as eluant.
Yield, 0.098 g (65%). Found: C, 47.48; H, 3.58; N, 5.48; F,
22.76; Cl, 13.98. C10H9F3ClNO requires C, 47.73; H, 3.60; N,
1720 (C᎐O), 1681 (C᎐O), 1535 (NO , as stretch), 1150 (CF, st);
᎐
᎐
2
λmax (dioxane)/nm 260 (ε = 11,700); δH (270 MHz, CDCl3) 2.21
(3H, s, CH3), 4.56 (2H, q, J = 8.61 Hz, CH2), 7.81 (2H, d, H2,
H6), 8.36 (2H, d, H3, H5); δC (68 MHz, CDCl3) 25.66 (1C, q,
2
1
CH3), 45.39 (q, JFC = 35.4 Hz, CH2), 123.69 (q, JFC = 280.5
Hz, CF3), 124.34, 129.25 (4C, d, CH, aromatic), 140.27 (1C, s,
C1), 150.01 (1C, s, C4), 171.35, 171.29 (2C, s, C᎐O); m/z 290
᎐
(Mϩ, 5%), 150 (60), 104 (20), 76 (15), 43 (100).
5.57; F, 22.69, Cl, 14.12%); νmax (KBr)/cmϪ1 1662 (C᎐N), 1155
᎐
(CF3, st), 689 (CF3, st); λmax (dioxane)/nm 274 (ε = 177000);
δH (270 MHz, CDCl3) 3.85 (3H, s, CH3), 4.14 (2H, q, J = 9.34
Hz, CH2), 6.90 (2H, d, J = 9.16 Hz, H3, H5), 7.99 (2H, d,
J = 9.16 Hz, H2, H6).
N-Acetyl-N-ethyl-p-nitrobenzamide (5b). Compound 5b was
similarly formed in 67% yield (Found: C, 56.18; H, 5.22; N,
11.73. C11H12N2O4 requires C, 55.95; H, 5.08; N, 11.86%);
νmax (KBr)/cmϪ1 1710 (C᎐O), 1679 (C᎐O); ν
(dioxane)/nm
᎐
᎐
max
237 (ε = 7900); δH (270 MHz, CDCl3) 1.28 (3H, t, J = 7.35 Hz,
CH3), 2.15 (3H, s, CH3), 3.49 (2H, q, J = 7.35 Hz, CH2), 7.90
(2H, d, J = 8.98 Hz, H3, H5), 8.25 (2H, d, J = 8.98 Hz, H2, H6);
m/z 236 (Mϩ, 20%), 150 (55), 76 (52), 43 (100).
5-p-Nitrophenyl-1-(2,2,2-trifluoroethyl)tetrazole (9a). A solu-
tion of N-(2,2,2-trifluoroethyl)-p-nitrobenzimidoyl chloride 2a
(0.43 g, 0.0016 mol) in Analar acetone (20 cm3) was added
dropwise to a vigorously stirred solution of sodium azide (2.08
g, 0.032 mol) in 1 : 1 acetone–water (50 cm3). The mixture was
left stirring at room temperature overnight. The acetone was
then removed by distillation. The resulting white precipitate
was filtered and washed with water. The pure product was
obtained by passing through a layer of Merck silica 7734,
particle size 0.063–0.2 mm, using dry CH2Cl2 as eluant. Yield,
0.295 g (68%), mp 118–120 ЊC (Found: C, 39.47; H, 2.17; N,
25.14; F, 20.72. C9H6F3N5O2 requires C, 39.57; H, 2.21; N,
25.64; F, 20.86%); λmax (dioxane)/nm 272 (ε = 12200); νmax
N-Acetyl-N-isopropyl-p-nitrobenzamide (5c). Compound 5c
was formed in 65% yield (Found: C, 56.23; H, 6.64; N, 11.45.
C12H14N2O4 requires C, 56.91; H, 6.76; N, 11.06%); νmax (KBr)/
cmϪ1 1704 (C᎐O), 1685 (C᎐O), 1527 (NO , as stretch);
᎐
᎐
2
λmax (dioxane)/nm 236 (ε = 7800); δH (270 MHz, CDCl3) 1.44
(6H, d, J = 6.78 Hz, (CH3)2), 2.11 (3H, s, CH3), 4.47 (1H, sept,
J = 6.78 Hz, CH ), 7.81 (2H, d, J = 8.97 Hz, H3, H5), 8.31 (2H,
d, J = 8.97 Hz, H2, H6); δC (68 MHz, CDCl3–CD3COCD3)
20.53 (2C, q, CH3), 26.59 (1C, q, CH3), 50.90 (1C, d, CH),
124.11, 129.15 (4C, d, CH, aromatic), 142.30 (1C, s, C1), 149.76
(KBr)/cmϪ1 1537 (conj. cyclic sys.), 1445 (N᎐N), 1171 (CF, st);
᎐
(1C, s, C4), 172.83, 172.95 (2C, s, C᎐O); m/z 250 (Mϩ, 25%), 193
δH (270 MHz, CDCl3) 5.08 (2H, q, J = 7.69 Hz, CH2), 7.91 (2H,
᎐
d, J = 8.98 Hz, H3, H5), 8.48 (2H, d, J = 8.98 Hz, H2, H6);
(10), 150 (45), 76 (20), 43 (100).†
2
δC (68 MHz, CDCl3–CD3COCD3) 48.79 (q, JFC = 36.6 Hz,
1
CH2), 122.25 (q, JFC = 279.4 Hz, CF3), 124.66, 130.61 (4C, d,
CH, aromatic), 129.31 (1C, s, C1), 149.92 (1C, s, C4), 154.34
References
(1C, s, C᎐N); m/z 273 (Mϩ, 60%), 76 (40), 83 (100).
᎐
1 H. O. Kalinowski and H. Kessler, Top. Stereochem., 1973, 7, 295.
2 (a) D. Y. Curtin, E. J. Grubbs and C. G. Mc Carty, J. Am. Chem.
Soc., 1966, 88, 2775; (b) K. J. Dignam and A. F. Hegarty, J. Chem.
Soc., Perkin Trans. 2, 1979, 1437.
3 Z. Rappoport and R. Ta-Shma, Tetrahedron Lett., 1972, 52, 5281;
H. Kessler, Angew. Chem., Int. Ed. Engl., 1970, 9, 219; R. Knorr,
T. P. Hoang, J. Mehlstaubl, M. Hintermeyer-Hilpert, H.-D.
Ludemann, E. Lang, G. Sextl, W. Rattay and P. Bohrer, Chem. Ber.,
1993, 126, 217.
4 H. Kessler, Tetrahedron Lett., 1968, 15, 2041; H. Kessler and
D. Leibfritz, Tetrahedron Lett., 1969, 6, 427; H. Kessler
and D. Leibfritz, Tetrahedron, 1970, 26, 1805; D. Leibfritz and
H. Kessler, J. Chem. Soc., Chem. Commun., 1970, 655; H. Kessler,
P. F. Bley and D. Leibfritz, Tetrahedron, 1971, 27, 1687.
5 M. Raban, J. Chem. Soc., Chem. Commun., 1970, 1415; M. Raban
and E. Carlson, J. Am. Chem. Soc., 1971, 93, 685; G. E. Hall,
W. J. Middleton and J. D. Roberts, J. Am. Chem. Soc., 1971, 93,
4778; T. J. Lang, G. J. Wolber and R. D. Bach, J. Am. Chem. Soc.,
1981, 103, 3275.
6 D. G. McCarthy and A. F. Hegarty, J. Chem. Soc., Perkin Trans. 2,
1977, 1080; D. G. McCarthy and A. F. Hegarty, J. Chem. Soc.,
Perkin Trans. 2, 1977, 1085.
7 (a) A. F. Hegarty, K. Brady and M. Mullane, J. Chem. Soc.,
Chem. Commun., 1978, 871; (b) A. F. Hegarty, K. Brady and
M. Mullane, J. Chem. Soc., Perkin Trans. 2, 1980, 535.
8 W. B. Jennings, S. Al-Showiman, M. S. Tolley and D. R. Boyd,
J. Chem. Soc., Perkin Trans. 2, 1975, 1535.
9 A. F. Hegarty, S. J. Eustace, N. M. Tynan, N. N. Pham-Tran and
M. T. Nguyen, J. Chem. Soc., Perkin Trans. 2, 2001, 1239.
10 I. Ugi, F. Beck and U. Fetzer, Chem. Ber., 1962, 95, 126; A. F.
Hegarty, J. D. Cronin and F. L. Scott, J. Chem. Soc., Perkin Trans. 2,
1975, 429.
5-p-Nitrophenyl-1-isopropyltetrazole (9c). Compound 9c was
similarly prepared in 56% yield, mp 122–124 ЊC. (Found: C,
51.14; H, 4.68; N, 30.24. C10H11N5O2 requires C, 51.52; H, 4.72;
N, 30.03%); λmax (dioxane)/nm 272 (ε = 12,200); δH (270 MHz,
CDCl3) 2.02 (6H, d, J = 6.59 Hz, (CH3)2), 5.27 (1H, sept,
J = 6.59 Hz, CH ), 7.93 (2H, d, J = 8.97 Hz, H3, H5), 8.45 (2H,
d, J = 8.97 Hz, H2, H6); m/z 233 (Mϩ, 25%), 191 (15), 76 (20),
43 (100).
5-Phenyl-1-(2,2,2-trifluoroethyl)tetrazole (9d). Yield 64%
(Found: C, 47.33; H, 3.08; N, 24.21; F, 25.38. C9H7F3N4
requires C, 47.39; H, 3.06; N, 24.55; F, 24.98%); λmax (dioxane)/
nm 243 (ε = 6,900); δH (270 MHz, CDCl3) 5.02 (2H, q, J = 7.75
Hz, CH2), 7.6 (5H, m, aromatic); m/z 228 (Mϩ, 1%), 186 (100),
77 (62).
5-p-Methoxyphenyl-1-(2,2,2-trifluoroethyl)tetrazole
(9e).
Yield, 50%, mp 104–106 ЊC (Found: C, 46.52; H, 3.51; N, 21.70;
F, 22.07. C9H6F3N5O2 requires C, 46.80; H, 3.53; N, 20.49; F,
20.87%); λmax (dioxane)/nm 253 (ε = 13,200); δH (270 MHz,
CDCl3) 3.52 (3H, s, OCH3), 4.43 (2H, q, J = 7.69 Hz, CH2), 6.42
(2H, d, J = 7 Hz, H3, H5), 6.81 (2H, d, J = 7 Hz, H2, H6); m/z
258 (Mϩ, 10%), 215, (100), 108 (20), 90 (40).
N-Acetyl-N-(2,2,2-trifluoroethyl)-p-nitrobenzamide (5a).
A
solution of N-(2,2,2-trifluoroethyl)-p-nitrobenzimidoyl chloride
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334
1333

View Article Online
15 M. T. Mc Cormack, PhD Thesis, National University of Ireland,
11 A. H. Fainberg and S. Winstein, J. Am. Chem. Soc., 1956, 78, 2770.
12 A. F. Hegarty, Acc. Chem. Res., 1980, 13, 448.
1976.
13 R. W. Darbeau, E. H. White, N. Nunez, B. Coit and M. Daigle,
J. Org. Chem., 2000, 65, 1115.
14 D. H. Williams and I. Fleming, Spectroscopic Methods in Organic
Chemistry, McGraw-Hill, London, 1995, 5th edn., p. 46.
16 J. E. Johnson and S. C. Cornell, J. Org. Chem., 1980, 45, 4144.
17 W. B. Jennings and V. E. Wilson, Acyclic Organonitrogen
Stereodynamics, eds. J. B. Lambert and Y. Takeuchi, VCH,
New York, 1992, pp. 219–224.
1334
J. Chem. Soc., Perkin Trans. 2, 2002, 1328–1334