Nucleophilic attack on the carbon-nitrogen double bond leading to tetrahedral intermediates with conformationally restricted stereochemistry

Article abstract of DOI:10.1071/CH08312

Methoxide ion substitution on 4-chloro-1H-2,3-benzoxazine 4 has been investigated. The rates of cyclization have been measured for (Z)- and (E)-O-(2-hydroxyethyl)benzohydroximoyl halides (PhC(X)≤NOCH ₂CH₂OH; X ≤ Cl or Br) using potassium t-butoxide to generate alkoxide ions from the alcohols. The element effect for the Z and E isomers are k_ZBr/k_ZCl ≤ 3.08 and k_EBr/k _ECl ≤ 2.38. The Z-to-E rate ratios are 0.25 (chlorides) and 0.33 (bromides). These reactions are proceeding by rate-determining nucleophilic attack by the alkoxide ion on the carbon-nitrogen double bond (A_N# + D_N). The restricted stereochemistry of the tetrahedral intermediates suggests that the E-hydroximoyl bromide and chloride may be undergoing unassisted ionization to form a zwitterionic intermediate that leads to product. Alternatively, the chair conformation of the tetrahedral intermediate may be undergoing a ring flip to a boat conformation in which there are antiperiplanar electron pairs that assist in the ionization of the bromide or chloride ion.

Full text of DOI:10.1071/CH08312

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2008, 61, 888–893
www.publish.csiro.au/journals/ajc
Nucleophilic Attack on the Carbon–Nitrogen Double Bond
Leading to Tetrahedral Intermediates with Conformationally
Restricted Stereochemistry
James E. Johnson,A,C Lei Lu,A Yi Li,A Mei Hou,A and Jeffrey E. RoweB
ADepartment of Chemistry and Physics, Texas Woman’s University, Denton, TX 76204-5859, USA.
BDepartment of Chemistry, LaTrobe University, Vic. 3086, Australia.
CCorresponding author. Email: jjohnson@mail.twu.edu
Methoxide ion substitution on 4-chloro-1H-2,3-benzoxazine 4 has been investigated. The rates of cyclization have been
measured for (Z)- and (E)-O-(2-hydroxyethyl)benzohydroximoyl halides (PhC(X)=NOCH2CH2OH; X = Cl or Br) using
potassium t-butoxide to generate alkoxide ions from the alcohols. The element effect for the Z and E isomers are
kZBr/kZCl = 3.08 and kEBr/kECl = 2.38. The Z-to-E rate ratios are 0.25 (chlorides) and 0.33 (bromides). These reactions are
proceeding by rate-determining nucleophilic attack by the alkoxide ion on the carbon–nitrogen double bond (AN# + DN).
The restricted stereochemistry of the tetrahedral intermediates suggests that the E-hydroximoyl bromide and chloride
may be undergoing unassisted ionization to form a zwitterionic intermediate that leads to product. Alternatively, the chair
conformation of the tetrahedral intermediate may be undergoing a ring flip to a boat conformation in which there are
antiperiplanar electron pairs that assist in the ionization of the bromide or chloride ion.
Manuscript received: 22 July 2008.
Final version: 25 September 2008.
CH3O
Ar
OCH3
X
OCH3
Introduction
NaOCH3
C
N
C
N
The mechanism and the stereochemical outcome of nucleophilic
substitution at carbon–nitrogen double bonded systems have
been of interest to us for some time. Our most recent work
involved a detailed experimental and theoretical investigation
of the reactions of hydroximoyl halides (1Z and 1E) with meth-
oxide ion.[1] In that study, and in previous studies, we have found
that the reactions of the methoxide ion with hydroximoyl halides
(1Z) are stereospecific.
9:1 DMSO/MeOH
Ar
2Z
ϳ100%
1Za–c for X ꢀ F, Cl, Br
X
CH3O
Ar
NaOCH3
ꢁ
2Z
C
N
C
N
The kinetics of the displacement of chloride, bromide, and
fluoride leaving groups (the element effect) by methoxide ion
from hydroximoyl halides (1 in Scheme 1) have been studied,
along with the stereochemistry of the products. The 1Z ele-
ment effect of (Ar = C6H5) Br/Cl/F = 2.21:1.0:79.7 along with
the similar 1E element effect of (Ar = C6H5) Cl/F = 1.0:18.3
and (Ar = 4-CH3OC6H4) Br/Cl/F = 1.97:1.0:12.1 suggested an
addition–elimination pathway where the methoxide ion under-
goesrate-determiningattackatthecarbon–nitrogendoublebond,
leading to a charged tetrahedral intermediate (AN# + DN). The
Z isomers of the starting materials led entirely to the Z product
(2Z), whereas the E starting materials led to a mixture of the E
and Z substitution products with the E isomer (2E) predominat-
ing (the E fluoride gave 85% E and 15% Z). This was explained
by considering the stereoelectronic effects in the charged tetra-
hedral intermediates (Scheme 2 for the 1-fluoro compound).The
tetrahedral intermediates A and C are formed by nucleophilic
attack perpendicular to the plane of the Z and E hydroximoyl
halide functional group. Tetrahedral intermediate A from the Z
isomer has an electron pair antiperiplanar to the leaving group
(fluoride ion) and does not require rotation in order to form a
π-bond between carbon and nitrogen.
9:1 DMSO/MeOH
Ar
OCH3
ϳ20%
OCH2
2E
1Ea–c for X ꢀ F, Cl, Br
ϳ80%
Scheme 1.
Intermediate C resulting from methoxide attack on the E
isomer requires a conformational change in order to have a non-
bonded electron pair on nitrogen antiperiplanar to the leaving
group (halide ion). Ab initio calculations were carried out on
the Z and E isomers of the hydroximoyl fluoride–methoxide ion
reaction. The lowest barrier for rotation in the tetrahedral inter-
mediate C (from the E fluoride) leads to tetrahedral intermediate
B. Conformation B leads to the E product (2E) in agreement
with the observed product distribution. Although there is evi-
dence that nucleophilic substitution at N-alkoxyimines may
proceed through a variety of mechanistic pathways (DN# +AN,
AN + DN#),[2–6] it is believed that alkoxide substitution of
alkoxyimidoyl halides proceeds by a stepwise rate-determining
addition followed by elimination.
© CSIRO 2008
10.1071/CH08312
0004-9425/08/110888

Nucleophilic Attack on the C=N Double Bond
889
F
OCH3
F
C
C
N
N
Ar
OCH3
Ar
1Zb
1Eb
CH3Oꢂ
9:1 DMSO/MeOH
CH3Oꢂ
9:1 DMSO/MeOH
Slow
Slow
OCH3
ꢂ
ꢂ
ꢂ
N
N
N
F
CH3O
C
OCH3
Ar
F
Ar
F
Ar
C
C
OCH3
OCH3
OCH3
B
C
A
ꢂFꢂ
ꢂFꢂ
Fast
Fast
OCH3
CH3O
CH3O
Ar
C
N
C
N
Ar
OCH3
2Z
2E
Scheme 2.
OCH3
OH
H2N
Ph
OCH2CH2OH
H2N
Ph
Cl
O
BrCH2CH2OH
NaOH
NaOCH3
PCl5
N
N
N
O
N
O
NH
O
7
6
5
4
3
NaNO2/HX
CH3
O
OMe
Cl
O
OCH2CH2OH
X
X
Cl
hν
N
ꢂ
N
ꢁ
8Z
N
ꢂ
N
Ph
OCH2CH2OH
Ph
8Z
O
8E
D
8Za: X ꢀ Cl
b: X ꢀ Br
8Ea: X ꢀ Cl
b: X ꢀ Br
ꢂ
OMe
N
N
H
O
Cl
X
N
OCH2CH2OH
N
ꢁ
Scheme 3.
OCH2CH2OH
Ph
N
8E
Ph
In the present paper, we extended our investigation of nucleo-
philicsubstitutionatthecarbon–nitrogendoublebondtosystems
in which the conformations of the tetrahedral intermediates are
conformationally restricted. The first system we investigated
was the cyclic hydroximoyl chloride 4 where the leaving group
is bonded to a six-membered ring. Nucleophilic attack by the
methoxide ion on 4 leads to a tetrahedral intermediate con-
strained to the cyclic equivalent of C. We then investigated 8Z
and 8E, where intramolecular nucleophilic alkoxide ion attack
at the carbon–nitrogen double bond leads to a cyclic tetrahedral
intermediate.
9Z
8Ea: X ꢀ Cl
b: X ꢀ Br
Scheme 4.
The benzamidoxime 6 was alkylated with 2-bromoethanol
to give Z-O-2-hydroxyethylbenzamidoxime 7 followed by dia-
zotization in the presence of hydrochloric acid to give the
Z-O-(2-hydroxyethyl)benzohydroximoyl chloride (8Za) or in
the presence of hydrobromic acid to give the equivalent bromide
8Zb (Scheme 4). Irradiation of benzene solutions of 8Za or 8Zb
led to a photostationary state mixture of the Z and E isomers in
a 1:1 ratio. These isomers could not be separated by chromato-
graphy. In order to obtain pure samples of the E isomers, the Z
isomers in these mixtures were selectively reacted with pyrro-
lidine to give Z-pyrrolidyl-O-(2-hydroxyethyl)benzamidoxime
9Z. Under the reaction conditions, the E isomers did not react.
Results and Discussion
Preparation of Starting Materials
1H-2,3-Benzoxazine-4(3H)-one 3 and 4-chloro-1H-2,3-benz-
oxazine 4 were prepared by a modification of the method of
Pifferi et al.[7,8] The reaction of 4 with methoxide ion led to
4-methoxy-1H-2,3-benzoxazine 5 as shown in Scheme 3.

890
J. E. Johnson et al.
OCH2CH2OH
Cl
OCH2CH2OH
CH3O
Ph
O
NaOCH3
C
N
C
N
ꢁ
O
DMSO
MeOH
Ph
Ph
N
10
8Za
11Z
Scheme 5.
X
ꢂO
X
ꢂXꢂ
Ph
O
O
O
O
Ph
O
N
N
Ph
ꢂ
N
8Za: X ꢀ Cl
b: X ꢀ Br
10
E
O
O
ꢂXꢂ
ꢂ
ꢁ
N
Ph
G
ꢂXꢂ
X
Ph
ꢂO
O
Ph
O
N
ꢂ
O
O
X
X
N
O
N
ꢂ
F
H
Ph
8Ea: X ꢀ Cl
b: X ꢀ Br
Scheme 6.
This method was based on our previous observation that the Z
isomers of hydroximoyl chlorides react almost completely with
pyrrolidine at 32◦C in 24 h, whereas under the same conditions,
the E isomers do not react.[6] Separation of Z-pyrrolidyl-O-
(2-hydroxyethyl)benzamidoxime 9Z from the E-hydroximoyl
halide (8Ea or 8Eb) was easily accomplished by column
chromatography.
for clarity. We have also previously proposed[6] that only one
antiperiplanar electron pair is necessary for the stereochemically
controlled elimination of chloride ion from such an inter-
mediate, and that this electron pair can come from the incoming
nucleophile. In the intermediate D, the methoxy group can freely
rotate, allowing one of the electron pairs on the oxygen to be
antiperiplanar to the chloride leaving group.
Reaction of 4-Chloro-1H-2,3-benzoxazine 4
with Methoxide Ion
Intramolecular Nucleophilic Substitution from E
and Z-O-(2-Hydroxyethyl)benzohydroximoyl
Chlorides and Bromides
4-Chloro-1H-2,3-benzoxazine 4 was treated with methoxide ion
in 10% methanol/90% dimethyl sulfoxide and gave a second-
order rate constant of 0.39 M−1 s−1 at 44.6◦C. Under the same
conditions, theopen-chaincompound 1Eb[1] gavearateconstant
of 1.42 × 10−2 M−1 s−1.Thus the cyclic hydroximoyl chloride 4
reacts 27 times faster than the open-chain hydroximoyl chloride
1Eb. This suggests that 4 also reacts by rate-determining attack
of the methoxide ion leading to the charged tetrahedral inter-
mediate, but in the cyclic case, there is a smaller steric effect to
the approach of the methoxide as compared with the open-chain
compound 1Eb, leading to the faster rate. The electron pairs on
nitrogen cannot initially be antiperiplanar to the leaving group in
the cyclic tetrahedral intermediate (D, Scheme 3). However, the
barrier to ring inversion of the cyclohexene-like system in inter-
mediate D would only be 20–24 kJ mol−1, so this intermediate
could easily invert to give an intermediate where the lone pair
on nitrogen is trans to the leaving chloride (Scheme 3).[9] These
structures are shown in Scheme 3 with the benzene ring removed
Initial work on the reaction of 8Za with sodium methoxide ion as
the base indicated (by gas chromatography–mass spectroscopic
(GC/MS)analysis)thattwoproductswereformedinthereaction.
One of the products was the desired cyclic hydroximate 10, but
there was a competing reaction where methoxide ion attacked
the carbon–nitrogen double bond to give the substitution product
11Z (Scheme 5). In order to avoid formation of 11Z, the base
for the reaction was changed to potassium t-butoxide.
t-Butoxide ion is a strong base but a sterically congested
nucleophile, and under the conditions of the experiments
reported here, it converted the alcohol starting materials to the
equivalent alkoxide ions. The rate constants for the intramole-
cular nucleophilic substitution from 8Ea and b and 8Za and
b were measured in solutions containing 40 or 60% dimethyl
sulfoxide in t-butyl alcohol (Scheme 6). Different concentration
ratios of the potassium t-butoxide over the halides confirmed
that these reactions were following first-order kinetics. The

Nucleophilic Attack on the C=N Double Bond
891
Table 1. Rate constants for the intramolecular substitution of
O-(2-hydroxyethyl)benzohydroximoyl halides in DMSO/t-butyl alcohol
Conditions: 0.01 M halide; 0.05 M potassium t-butoxide
material should lead to intermediate F, which would need to
undergo a ring flip to intermediate E to get the non-bonded elec-
tron pairs on nitrogen and oxygen antiperiplanar to the leaving
halide ion. We estimate that the barrier to this ring flip is greater
T [◦C]
103k [s−1
]
than 46 kJ mol .
−1 [10] The rates of E isomers were faster than the
Compound
% DMSO
corresponding Z isomers, indicating that a high-energy ring flip
was not taking place. One possible explanation is that the prod-
uct is obtained by unassisted loss of halide from intermediate F
to form a zwitterionic intermediate G, which undergoes rotation
to form the π-bond in 10 (Scheme 6). It is also possible that
the unassisted ionization takes place concurrently with rotation
about the carbon–nitrogen bond, thus avoiding the high-energy
zwitterionic intermediate G.
Another possible route to product could be a ring flip of inter-
mediate F to a boat conformation H, which has two electron pairs
antiperiplanar to the leaving group. It is possible that the ring-flip
barrier to the boat is less than the 46 kJ mol−1 that we estimated
for the conversion F into E.
8Za
8Ea
8Zb
8Eb
8Zb
8Eb
8Za
8Zb
8Eb
8Za
8Zb
8Eb
8Za
60
60
60
60
40
40
40
40
40
40
40
40
40
33.4
33.4
33.4
33.4
33.4
33.4
33.4
39.3
38.4
39.3
25.0
25.0
25.0
1.81
7.18
5.58
17.1
1.18
4.03
0.50
2.10
6.20
0.90
0.48
1.82
0.20
In summary, it is possible that elimination of chloride ion
or bromide ion from the tetrahedral intermediates generated in
these experiments does not require assistance from the nitrogen
lone pair or any other lone pair on a heteroatom (oxygen in our
case) bonded to the carbon–nitrogen double bond.An alternative
pathway would involve conversion of the first formed intermedi-
ate into a boat conformation that then undergoes elimination of
the halide ion. These conclusions apply only to reactions where
the leaving group is chloride or bromide ion. In the case of elim-
ination of fluoride ion, which is a much poorer leaving group
than chloride or bromide ion, we have shown with our previ-
ous kinetic work and theoretical calculations that the tetrahedral
intermediates are likely to require a lone pair from a heteroatom
to assist in the elimination (Scheme 2).[1]
Table 2. Activation parameters calculated from the rate constants in
Table 1 for 40% DMSO
Compound
ꢀH‡ [kJ mol−1
]
ꢀS‡ [mol−1 K−1] (298 K)
8Zb
8Eb
8Za
77.2
68.2
77.5
−49
−68
−55
kinetic results are reported in Table 1 and the activation para-
meters calculated from the kinetic results are in Table 2. The
element effects of kZBr/kZCl = 3.08 and kEBr/kECl = 2.38 are
similar to the results from the open-chain compounds reported
previously,[1] where kZBr/kZCl = 2.21 (Ar = C6H5) for the
Z-isomers and kEBr/kECl = 1.97 (Ar = 4-CH3OC6H4) for the
E-isomers.The activation parameters (Table 2) measured in 40%
dimethyl sulfoxide in t-butyl alcohol are also very similar to
those reported for the open-chain compounds[1] (ꢀH‡ for 1Zb
was 75 kJ mol−1 and ꢀS‡ (298 K) was −42 J mol−1 K−1).These
results rule out any mechanism involving a direct displacement
of the halogen (ANDN) and indicate that the mechanism of the
intramolecular substitution must be an addition–elimination pro-
cess (AN# + DN), where the attack of the alkoxide ion leading
to the cyclic tetrahedral intermediate (Scheme 6) is the rate-
determining step.The relative rate constant for the Z-isomer over
the E-isomer was calculated to be 0.25 for the chloro compounds
(8Ea and 8Za) and 0.33 for the bromo compounds (8Eb and
8Zb). These ratios are smaller than the ratio of 0.89[1] obtained
for intermolecular reactions of methoxide ion with Z and E
O-methylbenzohydroximoyl chloride. Space-filling molecular
models do not indicate much difference in the ability of the
Z or E alkoxide ions to attack the imine carbon. Indeed, there
may be a slightly greater relief of steric strain in the E isomer on
going from the starting material to the tetrahedral intermediate.
This may explain the higher reactivity of the E isomers in the
intramolecular substitution.
Experimental
General Methods
Photochemical isomerization reactions were carried out in
quartz tubes placed in a Rayonet reactor, model RPR-100
(Southern New England Ultraviolet Co.), fitted with 15 low-
pressure lamps (254 nm). Elementary analyses of some com-
pounds described in the present work were carried out by
Midwest Microlab (Indianapolis, IN, USA). 3-Phenyl-5H-1,4,2-
dioxazine was prepared by the method of Johnson.[11] t-Butanol
was heated under reflux over calcium hydride and distilled.
Dimethyl sulfoxide was stored over dried 4 Å molecular sieves.
Typical Kinetic Method
Sufficient potassium t-butoxide to make a 0.055 M solution
was weighed into a stoppered flask and dissolved in 10 mL
of the appropriate DMSO/t-butyl alcohol solution. This solu-
tion was thermally equilibrated for 15 min. A solution of the
halide (200 µL, 10−4 mol) in the same DMSO/t-butyl alco-
hol solution was added and the solution was well mixed and
allowed to thermally re-equilibrate for a further 2 min before
the first 0.5-mL aliquot was taken and quenched with 1 mL
of 0.034 M HCl. Further aliquots were taken after appropri-
ate time intervals. These solutions were analyzed by high-
performance liquid chromatography (HPLC) on an octadecyl
column using 40% acetonitrile and 60% water as the eluting
solvent. Rate constants were obtained from an average of at
least two measurements. Normalization factors for peak areas
were previously determined by analysis of solutions contain-
ing known amounts of reactants and product. Reactions were
The Z hydroximoyl halides 8Za and 8Zb should lead to
intermediate E, in which non-bonded electron pairs on nitro-
gen and oxygen are antiperiplanar to the axial halide leaving
group, which should lead to a rapid conversion into the prod-
uct. Conversely, the E isomer (8Ea and 8Eb) of the starting

892
J. E. Johnson et al.
followed for at least 85% of the reaction. The rate constants
were determined using infinity values calculated by a computer
program designed to give the best linear regression analy-
sis of the data. The rate data reported in Table 1 are usually
the average of two independent experiments with agreement
between the experiments better than 4%. Activation parameters
were obtained from these kinetic data from plots with excel-
lent correlations (r > 0.9997). A similar procedure was used for
the methoxide substitution on 4-chloro-1H-2,3-benzoxazine 4.
Compound 4 was dissolved in dry dimethyl sulfoxide and stan-
dardized sodium methoxide solution was added to make up the
90% DMSO/10% methanol solution. Again aliquots were taken
after appropriate time intervals and quenched before analysis
by HPLC.
(20 mL) cooled in an ice–salt bath. Stirring was continued in the
ice bath for 1 h and then at room temperature for a further 4 h.
The product was extracted into dichloromethane. Removal of the
solvent gave the desired product, which was further purified by
vacuum distillation (0.75 g, 68%). δH (300 MHz, CDCl3) 2.15
(s, 1H), 3.95 (m, 2H), 4.40 (m, 2H), 7.40 (m, 3H), 7.81 (m, 2H).
δC (75.47 MHz, CDCl3) 61.73, 76.51, 127.03, 128.35, 130.59,
132.20, 138.53. Calc. for C9H10ClNO2: C 54.2, H 5.1, Cl 17.8,
N 7.0. Found: C 53.9, H 5.2, Cl 17.5, N 6.8%.
(Z)-O-(2-Hydroxyethyl)benzohydroximoyl bromide 8Zb
was prepared in a similar way using 40% HBr (25 mL) instead
of the 20% HCl (Yield: 0.68 g, 50%). δH (300 MHz, CDCl3) 2.0
(s, 1H), 3.94 (m, 2H), 4.43 (m, 2H), 7.39 (m, 3H), 7.79 (m, 2H).
δC (75.47 MHz, CDCl3) 61.82, 76.47, 128.20, 128.36, 130.62,
131.88, 133.67. Calc. for C9H10BrNO2: C 44.3, H 4.1, Br 32.7,
N 5.7. Found: C 44.2, H 4.2, Br 32.5, N 5.6%.
4-Chloro-1H-2,3-benzoxazine 4
Phosphoruspentachloride(4.18 g, 0.02 mol)wasaddedtoanice-
cold solution of 1H-2,3-benzoxazine-4(3H)-one (1 g, 6.7 mmol)
in chloroform (100 mL) and the mixture stirred at 0◦C for 4 h and
at room temperature for 30 min. Inorganic salts were removed by
filtration and the chloroform solution was washed with saturated
sodium bicarbonate solution before removing the chloroform by
evaporation. The crude product was purified by chromatography
(silica gel; 55:45 hexane/diethyl ether) giving 0.32 g (28%) of
4-chloro-1H-2,3-benzoxazine 4 as a yellow oil. δH (300 MHz,
CDCl3)5.03(s, 2H), 7.06–7.70(m, 4H). δC (75.47 MHz, CDCl3)
67.5, 121.7, 123.7, 124.3, 128.7, 132.1, 133.3, 154.4. Calc. for
C8H6ClNO: C 57.3, H 3.6, Cl 21.2, N 8.4. Found: C 57.4, H 3.7,
Cl 21.3, N 8.2%.
(E)-O-(2-Hydroxyethyl)benzohydroximoyl Chloride 8Ea
(Z)-O-(2-Hydroxyethyl)benzohydroximoyl chloride (1.0 g) in
benzene (50 mL) was placed in 13-cm quartz tubes and irra-
diated in a Rayonet Photochemical reactor (254 nm) for 4 h.
Analysis of the benzene solution by GC/MS indicated a mix-
ture of the Z and E isomers in the ratio 1:1. The benzene solution
was extracted with 10% sodium carbonate solution before dry-
ing the benzene solution (MgSO4) and removing the solvent by
evaporation.This mixture of isomers was stirred with pyrrolidine
(50 mL) at room temperature for 2 h. Water (250 mL) was added
andthepHofthesolutionadjustedto1with6 MHCl.Thedesired
product was extracted into ether. Drying the ether solution and
removing the solvent by evaporation gave a yellowish oil (0.38 g,
38%).The pure (E)-O-(2-hydroxyethyl)benzohydroximoyl chlo-
ride was obtained after chromatography and vacuum distillation.
δH (90 MHz, CDCl3) 2.0 (s, 1H), 3.7 (m, 2H), 4.1 (m, 2H), 7.5
(m, 5H). Calc. for C9H10ClNO2: C 54.2, H 5.1, Cl 17.8, N 7.0.
Found: C 54.0, H 5.1, Cl 17.6, N 7.1%.
4-Methoxyl-1H-2,3-benzoxazine 5
4-Chloro-1H-2,3-benzoxazine 4 (350 mg, 2.10 mmol) was
added to a stirred solution of sodium metal (0.084 g, 3.6 mmol)
in 3 mL of methanol. The resulting solution was heated under
reflux for 30 min before removing the solvent by evaporation.
The residue was taken up in ether. Removal of the solvent gave
0.28 g (82%) of the desired product. The product was further
purified by column chromatography. δH (90 MHz, CDCl3) 3.92
(s, 3H), 4.85 (s, 2H), 7.00–7.69 (m, 4H). δC (50.31 MHz, CDCl3)
54.0, 66.8, 119.8, 122.7, 123.2, 127.8, 131.7, 134.4, 163.4. Calc.
for C9H9NO2: C 66.3, H 5.6, N 8.6. Found: C 66.2, H 5.6,
N 8.5%.
(E)-O-(2-Hydroxyethyl)benzohydroximoyl bromide 8Eb
was obtained by a similar method (21%). δH (300 MHz, CDCl3)
2.8 (s, 1H), 3.95 (m, 2H), 4.38 (m, 2H), 7.63 (m, 5H). Calc. for
C9H10BrNO2: C 44.3, H 4.1, Br 32.7, N 5.7. Found: C 44.3,
H 4.2, Br 32.9, N 5.7%.
Acknowledgements
(Z)-O-(2-Hydroxethyl)benzamidoxime 7
Acknowledgement is made to the Minority Biomedical Research Support
Program of the National Institutes of Health (NIH-MBRS Grant GM08256),
The Robert A. Welch Foundation, and the Texas Woman’s University
Research Enhancement Program for support of the present work. J.E.R.
thanks the Texas Woman’s University for hospitality during the completion
of part of this work.
2-Bromoethanol (19 g, 0.15 mol) was added dropwise to a
stirred solution of benzamidoxime (10 g, 0.075 mol) and sodium
hydroxide (8 g, 0.2 mol) in water (350 mL) cooled in an ice bath.
Aftertheadditionofthebromoethanolwascomplete, themixture
was heated under reflux for 5 h. On cooling, the aqueous solu-
tion was saturated with sodium chloride and the organic product
extracted into dichloromethane. Removal of the solvent gave the
desired product after recrystallization from 1:6 acetone/hexane
(9.0 g, 67%), mp 87–88◦C. δH (300 MHz, CDCl3) 3.26 (s, 1H),
3.90 (m, 2H), 4.19 (m, 2H), 4.91 (s, 2H), 7.36 (m, 3H), 7.55
(m, 2H). δC (75.47 MHz, CDCl3) 63.29, 74.16, 125.69, 128.50,
129.96, 131.93, 152.24. Calc. for C9H12N2O2: C 60.0, H 6.7,
N 15.6. Found: C 59.9, H 6.8, N 15.6%.
References
[1] J. E. Johnson, D. D. Dolliver, L. Yu, D. C. Canseco, M. A. McAl-
lister, J. E. Rowe, J. Org. Chem. 2004, 69, 2741. doi:10.1021/
JO030299E
[2] J. E. Johnson, E. A. Nalley, C. Weidig, M. Arfan, J. Org. Chem. 1981,
46, 3623. doi:10.1021/JO00331A008
[3] J. E. Johnson, S. L. Todd, J. L. Gardner, T. M. Gardner, P. Buck,
A. Ghafouripour, W. Zimmerman, J. Phys. Org. Chem. 1994, 7, 352.
doi:10.1002/POC.610070704
(Z)-O-(2-Hydroxyethyl)benzohydroximoyl Chloride 8Za
A solution of sodium nitrite (0.7 g, 0.01 mol) in water
(6 mL) was added to a rapidly stirred solution of (Z)-O-(2-
hydroxethyl)benzamidoxime (1.0 g, 5.5 mmol) in 20% HCl
[4] J. E. Johnson, I. Jano, M. A. McAllister, J. Phys. Org. Chem.
1999, 12, 240. doi:10.1002/(SICI)1099-1395(199903)12:3<240::
AID-POC123>3.0.CO;2-B

Nucleophilic Attack on the C=N Double Bond
893
[5] J. E. Johnson, A. Ghafourapour, M. Arfan, S. L. Todd, D. A. Sitz,
J. Org. Chem. 1985, 50, 3348. doi:10.1021/JO00218A021
[6] J. E. Johnson, S. L. Todd, S. M. Dutson, A. Ghafouripour,
R. M. Alderman, M. R. Hotema, J. Org. Chem. 1992, 57, 4648.
doi:10.1021/JO00043A022
[7] G. Pifferi, E. Testa, Tetrahedron 1966, 22, 2107. doi:10.1016/S0040-
4020(01)82130-X
[8] G. Pifferi, P. Consonni, E. Testa, Tetrahedron 1968, 24, 4923.
doi:10.1016/S0040-4020(01)98690-9
[9] The barrier for inversion of cyclohexene is 22 kJ mol−1: E. L. Eliel,
S. H. Wilen, L. N. Mander, Stereochemistry of Organic Compounds
1994, p. 727 (Wiley-Interscience: New York, NY).
[10] The ring flip barrier in 1,4,2-dioxazine is 45.6 kJ mol−1: F. G. Riddle,
The Conformational Analysis of Heterocyclic Compounds 1980,
pp. 96–97 (Academic Press: New York, NY).
[11] J. E. Johnson, J. R. Springfield, J. S. Hwang, L. J. Hayes,
W. C. Cunningham, D. L. McClaugherty, J. Org. Chem. 1971, 36,
284. doi:10.1021/JO00801A010