organic compounds
was found in the difference Fourier synthesis midway between
N7 and N15, and was freely re®ned. The details of this
hydrogen bond are: N7Á Á ÁH15 1.35 (3), N15Á Á ÁH15 1.45 (3),
ethyl 2-[(E)-5-chloro-2-hydroxy-4-nitrophenylazo]-3-(E)-am-
ino-2-butenoate [(III); Rodrigues et al., 1996]. The two
compounds, (II) and (III), differ mainly in that (III) contains a
Ê
ꢁ
N7Á Á ÁN15 2.479 (3) A and N7Á Á ÁH15Á Á ÁN15 125 (2) . Of the
substituents, the acetyl group and the phenyl ring at N7 are
approximately coplanar with the plane of the central six-
primary amine (R = H) instead of the 4-methoxyphenylamine
4
group. The principal characteristics of the two structures are
the same with the exception that in (III), the H atom remains
bonded to the amine nitrogen, i.e. the compound exists
predominantly as the azoenamine tautomer, (Ia). Thus, based
on the results obtained from (II) and (III), it may be
concluded that the position of the tautomeric equilibria is a
sensitive function of the substituents bonded to the terminal
azo and amine N atoms (R and R ).
ꢁ
membered ring, the dihedral angles being 3.5 (2) and 7.3 (2) ,
respectively. By contrast, the 4-methoxyphenyl group bonded
ꢁ
to the aminic nitrogen (N15) is rotated by 67.7 (1) from the
central molecular plane. Thus, the planarity of the molecule as
a whole is perturbed by the 4-methoxyphenyl group.
As noted above, the main purpose of the present work was
to identify the tautomer in which the title compound, (II),
exists in the solid state [or the percentage of the possible forms
3
4
As the only hydrogen-bond donor of the molecule is
involved in the intramolecular hydrogen bond, the inter-
molecular packing is governed by van der Waals interactions.
(
Ia±c) if it exists as an equilibrium mixture]. This can be
performed crystallographically by determining the bond
orders from the bond-length±bond-order curves proposed by
Burke-Laing & Laing (1976). The C9ÐC13 and C13ÐN15
Ê
bond lengths of 1.473 (4) and 1.271 (3) A (Table 1), respec-
2
2
tively, are close to the values reported for a pure Csp ÐCsp
Experimental
Ê
Compound (II) was prepared by the following procedure: aniline
single bond [1.487 (5) A; Shmueli et al., 1973] and a pure
Ê
N double bond (1.27 A; Burke-Laing & Laing, 1976).
These results are consistent with those obtained from NMR
(0.93 g, 10 mmol) in tetra¯uoroboric acid (10 ml of ca 30% HBF
was diazotized by adding a solution of sodium nitrite (0.69 g,
0 mmol) in water (5 ml). After several minutes, sodium tetra-
uoroborate (1 g, 9 mmol) was added, the suspension of benzene-
4
)
C
1
spectroscopy (Mach a cÏ ek et al., 2001), which have indicated
that (II) exists in CDCl solution as a 90:10 mixture of (Ib):(Ia)
¯
3
diazonium tetra¯uoroborate formed was mixed, and the precipitated
solid collected by suction and thoroughly pressed on a small sintered-
glass ®lter. The almost dry product was added portion-wise to a
solution of 4-(4-methoxyphenylamino)-3-penten-2-one (2.26 g,
11 mmol) in diisopropyl ether (15 ml). The mixture was stirred at
273 K for 2 h, whereupon the solid was collected by suction, resus-
pended in a chloroform±ethyl acetate mixture (1:1, ca 10 ml) and
transferred onto a silica-gel column. The product was eluted by the
same solvent mixture. The fraction containing the product along with
the unreacted enaminone was subjected to vacuum distillation to
remove the solvent, and the residue was repeatedly triturated with
hexane. The less hexane-soluble residue was again submitted to
chromatography on a silica-gel column with a chloroform±ethyl
acetate mixture (1:1). Finally, the product was recrystallized from
cyclohexane (m.p. 411±413 K).
[
with no contribution of (Ic)]. On the other hand, the N7ÐN8
Ê
and N8ÐC9 bond distances [1.316 (3) and 1.309 (3) A,
respectively] are both intermediate between single and double
bonds (bond orders ca 1.4 and 1.7), assuming values of 1.41,
Ê
1
.23, 1.45 and 1.27 A for NÐN, N N, CÐN and C N bonds,
respectively (Burke-Laing & Laing, 1976). Obviously, if only
forms (Ia) and (Ib) contribute to the electronic structure of
the molecule, then there is a discrepancy in the indications
concerning the position of the imine-enamine and azohy-
drazone tautomerisms. The discrepancy does not seem to be
caused by some contribution of the enol form, (Ic), as indi-
cated by a pure single-bond and a pure double-bond character
of the C9ÐC10 and C10ÐO11 bonds, respectively (Table 1).
Instead, the inconsistency most likely originates from the
existence of the symmetrical hydrogen bond [i.e. the H atom is
not bonded to N7 as required by formula (Ib)] which should
promote an accumulation of electron density on N7 and
subsequently its transfer to the adjacent phenyl ring. Indeed,
some degree of conjugation of the hydrazone moiety (atoms
N7, N8 and C9) with the phenyl ring (which should result in
the lowering and increasing of the C9ÐN8 and N8ÐN7 bond
orders, respectively) is clearly seen in (i) the coplanarity of the
hydrazone group with the phenyl ring, (ii) a partial double-
bond character of C1ÐN7, and (iii) a non-equivalency of the
phenyl-ring CÐC bonds (Table 1).
Crystal data
C
18
H
19
N
3
O
2
m
D measured by ¯otation in
bromoform/cyclohexane
r
M = 309.36
Monoclinic, P2
1
Ê
/c
Mo Kꢂ radiation
Cell parameters from 15
re¯ections
a = 11.525 (5) A
Ê
b = 14.465 (6) A
Ê
c = 9.761 (3) A
ꢁ
ꢃ = 7±18
ꢁ
� 1
ꢀ
= 92.82 (4)
V = 1625.3 (11) A
ꢄ = 0.08 mm
T = 293 (2) K
Ê
3
Z = 4
Plate, yellow
0.30 Â 0.25 Â 0.10 mm
�
3
D
D
x
= 1.264 Mg m
= 1.27 (1) Mg m
� 3
m
Another purpose of this work was to compare the present
results with those of similar structures in order to explore the
effects of substituents R ±R bonded to the azoenamine
skeleton on the position of the tautomeric equilibria in the
solid state. However, a search of the Cambridge Structural
Database (Allen & Kennard, 1993) for structures containing
the azoenamine substructure revealed only one compound,
Data collection
Syntex P2
ꢃ/2ꢃ scans
1
diffractometer
h = � 13 ! 0
1
4
k = 0 ! 17
3054 measured re¯ections
l = � 11 ! 11
2
1
890 independent re¯ections
319 re¯ections with I > 2ꢅ(I)
2 standard re¯ections
every 98 re¯ections
intensity decay: 4%
R
int = 0.049
ꢁ
ꢃ
max = 25.1
ꢀ
738 Viktor Kettmann et al.
18
C H
N O
19 3 2
Acta Cryst. (2001). C57, 737±739