Structural systematics of the anhydrous 1:1 proton-transfer compounds of 3,5-dinitrosalicylic acid with aniline and monosubstituted anilines

Article abstract of DOI:10.1007/s10870-011-0153-0

The crystal structures of the proton-transfer compounds of 3,5-dinitrosalicylic acid (DNSA) with a series of aniline-type Lewis bases (aniline, 2-hydroxyaniline, 2-methoxyaniline, 3-methoxyaniline, 4-fluoroaniline, 4-chloroaniline and 2-aminoaniline) have been determined and their hydrogen-bonding systems analysed. All are anhydrous 1:1 salts: [(C ₆H₈N)⁺(C₇H₃N ₂O₇)^-] (1), [(C₆H ₈NO)⁺(C₇H₃N₂O ₇)^-] (2), [(C₇H₁₀NO) ⁺(C₇H₃N₂O₇)^-] (3), [(C₇H₁₀NO)⁺(C₇H ₃N₂O₇)^-] (4), [(C₆H ₇FN)⁺(C₇H₃N₂O ₇)^-] (5), [(C₆H₇ClN) ⁺(C₇H₃N₂O₇)^-] (6), and [(C₆H₉N₂)⁺(C ₇H₃N₂O₇)^-] (7), respectively. Crystals of 1 and 6 are triclinic, space group P-1 while the remainder are monoclinic with space group either P2₁/n (2, 4, 5 and 7) or P2₁ (3). Unit cell dimensions and contents are: for 1, a = 7.2027(17), b = 7.5699(17), c = 12.9615(16) A, α = 84.464(14), β = 86.387(15), γ = 75.580(14)°, Z = 2; for 2, a = 7.407(3), b = 6.987(3), c = 27.653(11) A, β = 94.906(7)°, Z = 4; for 3, a = 8.2816(18), b = 23.151(6), c = 3.9338(10) A, β = 95.255(19)°, Z = 2; for 4, a = 11.209(2), b = 8.7858(19), c = 15.171(3) A, β = 93.717(4)°, Z = 4; for 5, a = 26.377(3), b = 10.1602(12), c = 5.1384(10) A, β = 91.996(13)°, Z = 4; for 6, a = 11.217(3), b = 14.156(5), c = 4.860(3) A, α = 99.10(4), β = 96.99(4), γ = 76.35(2)°, Z = 2; for 7, a = 12.830(4), b = 8.145(3), c = 14.302(4) A, β = 102.631(6)°, Z = 4. In all compounds at least one primary linear intermolecular N⁺-H...O(carboxyl) hydrogen-bonding interaction is present which, together with secondary hydrogen bonding results in the formation of mostly two-dimensional network structures, exceptions being with compounds 4 and 5 (one-dimensional) and compound 6 (three-dimensional). In only two cases (compounds 1 and 4), are weak cation-anion or cation-cation π-π interactions found while weak aromatic C-H...O interactions are insignificant. The study shows that all compounds fit the previously formulated classification scheme for primary and secondary interactive modes for proton-transfer compounds of 3,5-dinitrosalicylic acid but there are some unusual variants.

Full text of DOI:10.1007/s10870-011-0153-0

J Chem Crystallogr (2011) 41:1649–1662
DOI 10.1007/s10870-011-0153-0
ORIGINAL PAPER
Structural Systematics of the Anhydrous 1:1 Proton-Transfer
Compounds of 3,5-Dinitrosalicylic Acid with Aniline
and Monosubstituted Anilines
•
•
•
Graham Smith Urs D. Wermuth Peter C. Healy
Jonathan M. White
Received: 22 December 2009 / Accepted: 9 June 2011 / Published online: 25 June 2011
Ó Springer Science+Business Media, LLC 2011
Abstract The crystal structures of the proton-transfer
compounds of 3,5-dinitrosalicylic acid (DNSA) with a
series of aniline-type Lewis bases (aniline, 2-hydroxyani-
line, 2-methoxyaniline, 3-methoxyaniline, 4-fluoroaniline,
4-chloroaniline and 2-aminoaniline) have been determined
and their hydrogen-bonding systems analysed. All are anhy-
drous 1:1 salts: [(C6H8N)?(C7H3N2O7)-] (1), [(C6H8NO)?
(C7H3N2O7)-] (2), [(C7H10NO)?(C7H3N2O7)-] (3), [(C7H10
NO)?(C7H3N2O7)-] (4), [(C6H7FN)?(C7H3N2O7)-] (5),
[(C6H7ClN)?(C7H3N2O7)-] (6), and [(C6H9N2)?(C7H3N2
O7)-] (7), respectively. Crystals of 1 and 6 are triclinic, space
group P-1 while the remainder are monoclinic with space
group either P21/n (2, 4, 5 and 7) or P21 (3). Unit cell
dimensions and contents are: for 1, a = 7.2027(17),
91.996(13)°, Z = 4; for 6, a = 11.217(3), b = 14.156(5),
˚
c = 4.860(3) A, a = 99.10(4), b = 96.99(4), c = 76.35(2)°,
˚
Z = 2;for7, a = 12.830(4), b = 8.145(3), c = 14.302(4) A,
b = 102.631(6)°, Z = 4. In all compounds at least one pri-
mary linear intermolecular N?–H_O(carboxyl) hydrogen-
bonding interaction is present which, together with secondary
hydrogen bonding results in the formation of mostly two-
dimensional network structures, exceptions being with com-
pounds 4 and 5 (one-dimensional) and compound 6 (three-
dimensional). In only two cases (compounds 1 and 4), are
weak cation–anion or cation–cation p–p interactions found
while weak aromatic C–H_O interactions are insignificant.
The study shows that all compounds fit the previously for-
mulated classification scheme for primary and secondary
interactive modes for proton-transfer compounds of
3,5-dinitrosalicylic acid but there are some unusual variants.
˚
b = 7.5699(17), c = 12.9615(16) A, a = 84.464(14), b =
86.387(15), c = 75.580(14)°, Z = 2; for 2, a = 7.407(3),
˚
b = 6.987(3), c = 27.653(11) A, b = 94.906(7)°, Z = 4;
˚
for 3, a = 8.2816(18), b = 23.151(6), c = 3.9338(10) A,
Keywords Hydrogen bonding ꢀ Proton-transfer
compounds ꢀ 3,5-Dinitrosalicylic acid ꢀ Anilines
b = 95.255(19)°, Z = 2; for 4, a = 11.209(2), b =
˚
8.7858(19), c = 15.171(3) A, b = 93.717(4)°, Z = 4; for 5,
˚
a = 26.377(3), b = 10.1602(12), c = 5.1384(10) A, b =
Introduction
Because of its acid strength (pKa = 2.2), 3,5-dinitrosali-
cylic acid (DNSA) is capable of protonating most nitrogen
Lewis bases giving compounds with significantly enhanced
crystallinity through hydrogen bonding involving the
aminium group and the oxygen acceptors of the substituent
groups (carboxyl, phenol, nitro). We have previously
crystallographically characterized and reported the struc-
tures of ca. 40 of these proton-transfer compounds
(including series with aliphatic amines [1], monocyclic
aromatic and heteroaromatic amines [2] and polycyclic
aromatic and heteroaromatic amines [3]). Only one
example of a non-transfer Lewis base–DNSA complex has
G. Smith (&)
Faculty of Science and Technology, Queensland University
of Technology, G.P.O. Box 2434, Brisbane, QLD 4001,
Australia
e-mail: g.smith@qut.edu.au
U. D. Wermuth ꢀ P. C. Healy
School of Biomolecular and Physical Sciences, Griffith
University, Nathan, QLD 4111, Australia
J. M. White
BIO-21, Molecular Science and Biotechnology, University
of Melbourne, Parkville, VIC 3010, Australia
123

1650
J Chem Crystallogr (2011) 41:1649–1662
been described, that with the very weak base phenazine [4],
although the 1:1 adducts with urea [5], 1,1-diethylurea [6],
phenylurea [7], trans-1,4-dithiane-1,4-dioxide [8] and the
pseudopolymorphic DNSA solvate compounds, which may
be also be loosely considered as adducts of this acid. These
solvates are the two monohydrate polymorphs [9, 10], four
1,4-dioxane solvates and a t-butyl alcohol solvate [10].
However, solvates are uncommon among the DNSA pro-
ton-transfer compounds with only seven known examples:
five hydrates and two ethanol solvates. The five hydrates
are ethylenediaminium 3,5-dinitrosalicylate monohydrate
[1], salicylaldoximium 3,5-dinitrosalicylate dihydrate
[11], brucinium 3,5-dinitrosalicylate monohydrate [12],
1-ammonio-5-hydroxynaphthalene monohydrate [13] and
the unusual mixed piperidine DNSA–picrate (a monohy-
drate) [14]. The two ethanol solvates are considered more
novel, but possible because crystals of most of these DNSA
compounds are obtained from ethanol–water solvent. They
quinoline, quinaldic acid and 1,10-phenanthroline [3],
1-amino-5-hydroxynaphthalene [13] and benzidene [15]).
Furthermore, with the same set of aromatic bases, there is
an increased incidence of weaker aromatic ring C–H_O
hydrogen bonding, e.g. in the structure of the 1:1 DNSA
compound with 2,20-bipyridine [3] where there is only one
formal inter-species N?–H_O hydrogen bond, there are
seven C–H_O associations.
Associative Modes in the DNSA Proton-Transfer
Compounds
From a consideration of the structural features of the large
number of examples of DNSA–Lewis base proton-transfer
compounds characterized crystallographically, we have
previously systematized the modes of hydrogen-bonding
association on the basis of primary and secondary inter-
molecular interactions [3]. With all orders of both aliphatic,
aromatic and heteroaromatic amines, protonation is usually
followed by primary N?–H(aminium)_O (carboxylate)
cation–anion association. Secondary structure-building
hydrogen-bonding associations then occur through other
available donor and acceptor sites to expand the basic
cation–anion units into one-dimensional chains, two-
dimensional networks or three-dimensional framework
structures. There is, not surprisingly a decrease in the
associative efficacy of the protonated functional group of
the Lewis base proceeding from primary to secondary to
tertiary amines. There are only four examples in which the
carboxyl-O is not involved in the primary cation–anion
interaction but instead usually participates in only weaker
secondary O_H–C interactions. They are the compounds
of DNSA with hexamethylenetetramine [1, 21], 4-cyano-
pyridine [2] and strychnine [22] (in both of these examples
there is a primary cyclic bidentate N?–H_O,O0(phenol,
nitro) interaction [graph set R21(6)] [23], but with 1-amino-
5-hydroxynaphthalene, a carboxyl-O acts as an acceptor in
a strong hydrogen bond with the hydroxyl group of the
cation [13]). This R21(6) interaction is common in second-
ary structure extensions among proton-transfer compounds
of DNSA [3] (Scheme 1).
are
a partial ethanolate, 2-aminopyridinium-3,5-dini-
trosalicylate-ethanol (4/4/1) [2] and benzidinium1?-benz-
idinium2?-3,5-dinitrosalicylate-ethanol (3/1/4/2) [15]. The
stoichiometry for these DNSA compounds is predomi-
nantly 1:1, usually formed as a preferred product irre-
spective of reactant ratios, which is consistent with
observations for carboxylic acid-2-aminopyrimidine com-
pounds [16]. Only two DNSA examples are known in
which adduct formation is found, one with 4-aminobenzoic
acid [4-carboxyanilinium 3,5-dinitrosalicylate-4-amino-
benzoic acid (1:1:1)] [17] and one with pyridine [pyridini-
um 3,5-dinitrosalicylate-3,5-dinitrosalicylic acid (1:1:1)]
[2], the latter differing from the former in having the DNSA
proton-donor molecule rather than the base-acceptor mol-
ecule as the adduct molecule. With the bifunctional exam-
ples, the utility of the protonated amino group in enhancing
secondary associative effects, through the formation of
hydrogen-bonding networks, is well recognized [1].
A number of compounds of DNSA with a series of
substituted primary anilines [18], N-substituted anilines
[19] and phenylenediamines [20] have been synthesized
and characterized spectroscopically to ascertain the pres-
ence of proton transfer and/or electron transfer. It was
found, as expected on the grounds of pKa differences
between the Lewis base and DNSA, that proton-transfer
compounds predominate, with electron transfer restricted
to compounds not only considered possible on pKa grounds
but those prepared from aprotic solvents rather than ethanol
or aqueous ethanol. We have also observed that among the
crystallographically characterized proton-transfer DNSA
compounds (all prepared using ethanol or aqueous ethanol
as solvent), there is a low incidence of what can be con-
sidered secondary p–p interactive effects involving the
aromatic cations and the DNSA anions (e.g. almost
exclusively those with the polycyclic Lewis bases, e.g.
The primary N?–H_O associative modes for the
DNSA proton-transfer compounds [3] are as follows:
O
O
O
N
+
H
N
-
Scheme 1 Proximal R21(6) interaction
123

J Chem Crystallogr (2011) 41:1649–1662
1651
Type 1: Single linear interaction. This set of structures
involves a single primary aminium N?–H_O(carboxyl)
interaction (graph set D), with secondary propagating
interactions giving one-, two- and three-dimensional
extension. Examples of one- and two-dimensional
extension are mostly confined to the tertiary aliphatic
examples, e. g. with triethylamine [1], although aromatic
examples are known e.g. with 8-aminoquinoline [24] and
8-hydroxyquinoline [25]. Three-dimensional extension
is common among the primary amines where strong
secondary interactions are present, e.g. in the isomeric
monoaminobenzoic acid compounds [16] but com-
pounds with the polycyclic heteroaromatics such as
1,10-phenanthroline [3] and 2,20-bipyridine [3, 26],
where significant aromatic C–H_O interactions are
present together with a minor incidence of cation–anion
p–p association.
anilines chosen were the parent aniline (AN) and the mono-
substituted analogs, 2-hydroxyaniline (o-aminophenol =
OAP), 2-methoxyaniline (o-anisidine = OAN), 3-methoxy-
aniline (m-anisidine = MAN), 4-chloroaniline (FLAN),
4-chloroaniline (CLAN) and 2-aminoaniline (o-phenyl-
enediamine = OPDA) (pKa range for the set: 4.00–4.74).
We have previously described the compounds with the
monocarboxy-substituted aniline compounds (the isomeric
monoaminobenzoic acids), 2-aminobenzoic acid and
3-aminobenzoic acid (both 1:1), and the previously men-
tioned 1:1:1 adduct example with 4-aminobenzoic acid [9].
In all of these (as with primary amines generally), pro-
tonation of the amine group results in a Type 1 linear
hydrogen-bonding interaction with the carboxyl group of
DNSA, together with secondary extension through strong
N?–H_O hydrogen bonds involving the other aminium
protons. These interactions, which may number up to six
per aminium group are largely responsible for the forma-
tion of stable two- and three-dimensional network or
framework structures, usually giving crystals with rela-
tively high melting points and good morphology. The
DNSA compounds reported here are all anhydrous 1:1
salts, as follows: with aniline, [(C6H8N)?(C7H3N2O7)-]
(1); with 2-hydroxyaniline, [(C6H8NO)?(C7H3N2O7)-] (2);
with 2-methoxyaniline, [(C7H10NO)?(C7H3N2O7)-] (3);
with 3-methoxyaniline, [(C7H10NO)?(C7H3N2O7)-] (4);
with 4-fluoroaniline, [(C6H7FN)?(C7H3N2O7)-] (5); with
4-chloroaniline, [(C6H7ClN)?(C7H3N2O7)-] (6), and with
2-aminoaniline, [(C6H9N2)?(C7H3N2O7)-] (7). With the
exception of 5, the 1:1 compounds of DNSA with the set of
anilines used in this work have previously been reported
(the anilines in 1–4 and 6 [18] and o-phenylenediamine 7
[20]) but only two of them (2 and 4) appear not to be the
same as the compounds reported in this work (Scheme 3).
Type 2: Cyclic R21(4) interaction. This category involves
the aminium N?–H group in a cyclic three-centred
bidentate interaction with both carboxylate oxygens
[graph set R21(4)] and is quite common, e.g. the
compounds with 3-amino-1,2,4-triazole [27] and ben-
zylamine [28].
Type 3: Cyclic R22(N) interaction. This category involves
two proton donors of the cation including one aminium
proton and both carboxylate-O acceptors in a cyclic
R22(N) association, where N is usually 8, e.g. the
compounds with 2-aminopyrimidine [2], cytosine [29],
guanidine [30], 2-aminothiazole [31] and trimethoprim
[32], but N may also be 7 (the compounds with
benzimidazole [3] and salicylaldoxime [11]). Another
R22(8) variant involves the aminium proton and an
adjacent aromatic ring H atom donor in an asymmetric
cyclic association, such as is found in the structure of
quinolinium 3,5-dinitrosalicylate [3] (Scheme 2).
The object of the current work was to prepare and
structurally characterize a series of compounds of DNSA
with a number of primary anilines which, although rela-
tively weak bases, have pKa values which would result in
formation of proton-transfer compounds. The structures of
these could then be used to test the assembly mode types on
the basis of the previously described classifications [3] and
possibly allow a systematization of the structural aspects of
the DNSA compounds with the primary anilines. The
Experimental
Preparation of Compounds
Compounds 1–7 were prepared using equimolar quantities
(0.10 mmol) of 3,5-dinitrosalicylic acid with the appro-
priate Lewis base. All preparations involved heating
50 cm3 of a solution of the two reactants in 80:20% etha-
nol–water under reflux for ca. 10 min. Volume reduction to
Scheme 2 Type 1–3
interactions
-
-
+
-
H
+
N
O
N
H
O
O
+
N
H
O
O
O
H
[R2 (7)]
Type 2 [R2 (4)]
2
[D]
Type 1
Type 3
1
123

1652
J Chem Crystallogr (2011) 41:1649–1662
Scheme 3 The interacting
compounds for 1–7
O2N
NO2
OH
OCH3
OH
NH2
CO2H
DNSA
NH2
NH2
OAP
OANIS
AN
F
Cl
OCH3
NH2
NH2
NH2
NH2
NH2
MANIS
CLAN
FLAN
OPDA
ca. 40 cm3 followed by partial or complete room temper-
ature evaporation of the hot-filtered solutions gave in all
cases, crystals with good morphology. However, the
melting points for these compounds indicate that 2 and 4
are probably different from the Issa et al. [18, 20] com-
pounds [melting points (K) with comparative Issa values
given in parentheses: (1) 451.6–454.9 (453); (2)
480.4–481.1 (433); (3) 452.0–453.3 (446); (4) 426.8–427.5
(463); (5) 453.8–458.0 (not reported); (6) 466.7–468.0
(468); (7) 479.8–480.2 (483)]. The differences in the
products obtained with 2 and 4 are probably due the use of
benzene-ethanol solvent in the original preparations rather
than aqueous ethanol.
others were included at calculated positions and treated as
riding models. A complete listing of cell and structural
refinement data is given in Table 1. In the case of com-
pound 2, crystal twinning was found to be a problem but an
adequate specimen was found and used for data collection.
Nevertheless a problem was evident in the refinement,
resulting in a high refinement residual, considered to be
partially the result of the presence of rotational disorder
about the C7–C1_C4 ring diagonal in the DNSA anion,
which generates two partial sites for the C2 phenolic sub-
stituent group and the attached H (O2A, H2A) (see
Fig. 1b), giving site occupancies of 0.833(10) (O2) and
0.177(10) (O2A). This creates larger than normal dis-
placement parameters for the oxygen atoms in the con-
formationally dissimilar nitro groups at C3 and C5.
Data Collection, Structure Solution and Refinement
X-ray diffraction data were collected at ambient tempera-
ture on either a Bruker SMART CCD-detector diffrac-
tometer (compounds 2, 4 and 7) or a Rigaku AFC 7R
diffractometer (compounds 1, 3, 5 and 6). Graphite crystal
Results and Discussion
The Comparative Structures 1–7
˚
monochromatized Mo Ka X-radiation (k = 0.71073 A)
was used in all cases, with the Rigaku instrument from a
12 kW rotating anode source. With all compounds no
significant crystal decay was in evidence (maximum 2.1%
for 3). Data were corrected for Lorentz and polarization
effects but not for extinction and only in the case of
compound 6, for absorption. The structures were solved by
direct methods and refined using full-matrix least-squares
(on F2) using SHELX97 [33, 34] (in the case of compounds
1, 3, 5 and 6 using the TeXsan system [35]). Anisotropic
thermal parameters were used for all non-hydrogen atoms.
Hydrogen atoms potentially involved in hydrogen-bonding
interactions were located by difference methods and both
positional and thermal parameters were refined while
In all of the structures reported here 1–7 the presence of
anhydrous 1:1 proton-transfer compounds of DNSA has
been confirmed. The protonated nitrogen of the amine
substituent group of the anilinium cation subsequently
gives hydrogen-bonding interactions with the oxygen
acceptors (carboxylate, phenol or nitro) on the DNSA
anion, giving in all cases either two- or three-dimensional
framework structures. A full listing of hydrogen-bonding
interactions is given in Table 2, and comparative structural
parameters in Table 3. The individual structures are con-
sidered below. All structures fit the previously described
general structural motif categories for primary DNSA–
protonated Lewis base interactions. Although the Type 3
123

J Chem Crystallogr (2011) 41:1649–1662
1653
123

1654
J Chem Crystallogr (2011) 41:1649–1662
motif has a high probability of occurrence in general
molecular interactions [38], its incidence is relatively small
because of its functional group specificity. It would
therefore be reasonable to assume that among the anilinium
compounds of DNSA, Type 3 interactions would be
excluded and that Type 1 and Type 2 structural motifs
would be found. This is the case, with the compounds 1–4
being Type 1 while 5–7 are Type 2. Of the other analogous
aniline-type compounds of DNSA, with the isomeric
monoaminobenzoic acids [16], two have Type 1 interactions
while the third is Type 2. These three, together with the
compounds from this series (1–7) have the commonly-
occurring cyclic R21(6) N?–H_O,O0(phenol, nitro) proximal
group interactions among the secondary hydrogen-bonding
extensions.
With the DNSA anion, the intramolecular O(phe-
nol)_O(carboxyl) hydrogen bond [graph set S(6)] with the
acid proton anti-related on the carboxyl group is found in
three of the present examples [3, 4 and 7 (O_O, 2.495(2),
˚
2.482(3) and 2.447(4) A, respectively)] and has a 73%
incidence among the 39 previously reported examples of
proton-transfer compounds of DNSA [3]. With the DNSA
anion, it has been determined that the phenate species has
greater thermodynamic stability cf. the salicylate species
[21] so this observed statistic is not unexpected. However,
the occurrence of the salicylate anion form is certainly
influenced by crystal environment, particularly with respect
to the disposition of the cation donor groups relative to the
facially located proximal acceptor groups of the DNSA
anion (carboxyl, phenolic and nitro). The intramolecular
hydrogen bond in the DNSA salicylate-type anion species
[as found in compounds 1, 2, 5 and 6 (O_O, 2.447(2),
˚
2.469(6), 2.465(3) and 2.435(3) A, respectively)], in which
the proton is located on the phenolic-O, is similar to those
found in salicylic acid [39], in the DNSA acid solvates [9,
10] and its adducts [4–8].
With all compounds reported here the atom numbering
scheme for the DNSA anion and the anilinium cations in 1–7
is as shown in Fig. 1a–g. For DNSA, this scheme is consis-
tent with that used in previous reports on proton-transfer
compounds of DNSA completed in our laboratories [1–3].
The Individual Structures
[(C6H8N)?(C7H3N2O7)-] (1)
With the product from the reaction of aniline (AN) with
DNSA in 80% ethanol–water, [(AN)?(DNSA)-],
1
(Fig. 1a), the anilinium group gives not one but two pri-
mary linear Type 1 hydrogen-bonding interactions with
carboxyl oxygen acceptor groups of separate DNSA anions
i
ii
˚
˚
[N11–H_O71 , 2.861(3) A; N11–H_O72 , 2.797(3) A;
symmetry codes: (i) x ? 1, y, z ? 1; (ii) –x ? 1, -y ? 1,
123

J Chem Crystallogr (2011) 41:1649–1662
1655
Fig. 1 Atom numbering scheme for the individual anilinium cation
and DNSA anion species in compounds 1–7 (a–g). In 2, atoms O2 and
O2A [SOF 0.823(10) and 0.177(10), respectively] represent
rotationally disordered sites for the phenolic group. In all figures,
atoms are shown as 30% probability displacement ellipsoids [36, 37].
Inter-species hydrogen-bonding interactions are shown as dashed lines
123

1656
J Chem Crystallogr (2011) 41:1649–1662
˚
Table 2 Hydrogen-bonding associations (A/°) for compounds 1–7
D
H
A
D–H
H_A
D_A
D–H_A
(a) (1) [(AN)?(DNSA)-]a
O2
H2
O72
O71i
O72ii
O2
0.96(4)
0.94(3)
0.98(3)
0.91(3)
0.91(3)
0.91(3)
1.49(4)
1.93(3)
1.82(4)
2.39(3)
2.55(3)
2.47(3)
2.448(2)
2.861(3)
2.797(3)
3.123(3)
3.198(3)
3.074(3)
174(4)
170(2)
173(3)
138(2)
129(2)
125(2)
N11
N11
N11
N11
N11
H11A
H11B
H11C
H11C
H11C
O31
O31iii
(b) (2) [(OAP)?(DNSA)-]b
O2
H2
O72
O71
O71i
O21
O21ii
O72iii
O2iv
O2i
0.96(6)
0.93
1.52(6)
1.60
2.469(6)
2.53
169(5)
175
O2A
O21
N11
N11
N11
N11
N11
N11
H2A
H21
0.90(6)
0.86(6)
0.86(6)
0.90(7)
1.00(7)
1.00(7)
1.00(7)
1.70(6)
2.39(6)
2.14(6)
1.88(7)
2.46(7)
2.36(6)
2.26(6)
2.593(4)
2.689(5)
2.940(5)
2.777(5)
3.270(7)
2.979(6)
2.997(5)
175(4)
101(4)
154(5)
180(9)
138(4)
120(4)
130(4)
H11A
H11A
H11B
H11C
H11C
H11C
O31i
(c) (3) [(OAN)?(DNSA)-]c
O72
N11
N11
N11
N11
H72
O2
O2i
O31i
0.86(5)
0.89(4)
0.89(4)
0.96(3)
0.83(4)
1.66(5)
1.90(4)
2.35(4)
2.11(3)
1.92(4)
2.495(2)
2.735(3)
2.966(4)
2.965(3)
2.750(3)
161(5)
156(3)
126(3)
148(3)
171(4)
H11A
H11A
H11B
H11C
O2
O71ii
(d) (4) [(MAN)?(DNSA)-]d
O72
N11
N11
N11
N11
H72
O2
0.99(3)
0.88(3)
0.88(3)
0.99(3)
0.81(3)
1.53(3)
1.93(3)
2.58(3)
1.80(3)
2.13(3)
2.482(3)
2.802(3)
2.893(4)
2.770(3)
2.939(3)
160(3)
170(2)
102(2)
168(3)
176(4)
H11A
H11A
H11B
H11C
O2
O31
O71i
O32ii
(e) (5) [(FLAN)?(DNSA)-]e
O2
H2
O72
O72i
O71i
O72i
O2
0.98(5)
0.98(5)
0.93(5)
0.93(5)
0.94(5)
0.94(5)
0.94(5)
0.91(6)
1.53(5)
2.47(5)
2.04(5)
2.47(5)
2.26(5)
2.26(5)
2.31(5)
1.93(6)
2.465(4)
2.929(4)
2.965(5)
3.091(5)
3.107(5)
3.041(5)
3.003(5)
2.839(5)
157(5)
108(4)
175(5)
124(4)
125(4)
140(4)
130(4)
179(5)
O2
H2
N11
N11
N11
N11
N11
N11
H11A
H11A
H11B
H11B
H11B
H11C
O31
O32ii
O71iii
(f) (6) [(CLAN)?(DNSA)-]f
O2
H2
O72
0.81(5)
0.81(5)
0.87(5)
0.87(5)
0.98(4)
0.98(4)
0.90(4)
1.66(6)
2.53(5)
2.38(4)
2.32(4)
1.84(4)
2.53(4)
2.09(4)
2.435(3)
2.950(3)
3.140(4)
2.960(4)
2.814(4)
3.208(4)
2.988(4)
161(6)
114(4)
147(4)
130(4)
173(4)
126(3)
180(4)
O2
H2
O72i
O2ii
O31ii
O71iii
O72iii
O71
N11
N11
N11
N11
N11
H11A
H11A
H11B
H11B
H11C
123

J Chem Crystallogr (2011) 41:1649–1662
1657
Table 2 continued
D
H
A
D–H
H_A
D_A
D–H_A
(g) (7) [(OPDA)?(DNSA)-]g
O72
N11
N11
N11
N11
N11
N21
N21
H72
O2
0.97(3)
0.98(2)
0.89(3)
0.89(3)
0.95(2)
0.95(2)
0.84(3)
0.91(3)
1.49(4)
1.93(2)
2.13(3)
2.56(3)
1.86(2)
2.33(3)
2.41(3)
2.08(3)
2.447(2)
2.909(3)
2.914(3)
3.050(3)
2.800(2)
2.822(3)
3.245(3)
2.973(3)
165(3)
175(2)
146(2)
116(2)
170(2)
112(2)
175(2)
167(2)
H11A
H11B
H11B
H11C
H11C
H21A
H21B
N21i
O71i
O72i
O2ii
O31ii
O51iii
O71i
a
b
c
d
e
f
Symmetry codes: (i) = x ? 1, y - 1, z; (ii) = -x, -y ? 1, -z ? 1; (iii) = -x ? 1, -y ? 1, -z ? 1
Symmetry codes: (i) = x, y ? 1, z; (ii) = -x ? 1, -y ? 2, -z ? 1; (iii) = -x ? 1, -y ? 1, -z ? 1; (iv) = x ? 1, y, z
Symmetry codes: (i) = x, y, z - 1; (ii) = x ? 1, y ? 1, z
Symmetry codes: (i) = x, y - 1, z; (ii) = -x, -y ? 1, -z ? 1
Symmetry codes: (i) = -x ? 1, -y ? 2, -z ? 1; (ii) = -x ? 1, -y ? 1, -z ? 1; (iii) = -x ? 1, -y ? 2, -z
Symmetry codes: (i) = -x ? 2, -y ? 1, -z ? 3; (ii) = -x ? 2, -y ? 1, -z ? 2; (iii) = x, y, z - 1
Symmetry codes: (i) = -x ? 1, -y, -z ? 2; (ii) = x ? 1/2, -y ? 1/2, z ? 1/2; (iii) = x ? 1/2, -y ? 1/2, z - 1/2
g
˚
Table 3 Comparative structural features (A/°) for the DNSA moiety in the compounds 1–7
Compound
–Op_H–Oc
–Op–H_Oc
C2–C1–C7–O71
C2–C3–N3–O32
C4–C5–N5–O52
Reference
[(AN)?(DNSA)-] (1)
[(OAP)?(DNSA)-] (2)
[(OAN)?(DNSA)-] (3)
[(MAN)?(DNSA)-] (4)
[(FLAN)?(DNSA)-] (5)
[(CLAN)?(DNSA)-] (6)
[(OPDA)?(DNSA)-] (7)
a-(DNSA)ꢀH2O
–
2.447(2)
2.469(6)
–
-178.2(2)
-179.8(4)
-178.0(2)
-172.9(2)
-171.0(3)
-171.5(2)
-173.8(2)
178.3(4)
-174.8(2)
-177.5(4)
-168.9(2)
175.0(2)
142.8(3)
148.7(3)
179.7(2)
170.8(3)
160.0(3)
-172.2(2)
-163.8(4)
174.6(3)
This work
This work
This work
This work
This work
This work
This work
[9]
–
2.495(2)
2.482(3)
–
-179.3(2)
-179.3(4)
-179.1(2)
177.6(2)
–
2.465(4)
2.435(3)
–
2.447(4)
–
–
2.566(3)
2.549(4)
177.9(3)
b-(DNSA)ꢀH2O
178.3(4)
176.6(4)
[10]
a- and b-represent the two polymorphs of DNSA monohydrate
AN aniline, OAP 2-hydroxyaniline, OAN 2-methoxyaniline, MAN 3-methoxyaniline, FLAN 4-fluoroaniline, CLAN 4-chloroaniline, OPDA
2-aminoaniline, Op phenolic-O, Oc carboxyl-O
-z] (Fig. 2). This type of duplex interaction is unusual
among proton-transfer compounds of DNSA. In addition, a
symmetric cyclic R21(6) proximal group extension is pres-
ent, as well as an R22(4) association across an inversion
centre (Table 2a) result in a two-dimensional framework
structure, which extends down the a cell direction in the
unit cell. In addition there are significant cation–anion
p–p aromatic ring interactions with inter-ring separation
[Cg(C1–C6)_Cg(C11–C61)] and inter-ring dihedral angle
minor twinning problem as well as rotational disorder about the
C7–C1_C4 ring diagonal of the DNSA anion. This disorder
phenomenon results in the C2 phenolic group rotating into two
partial sites [O2 (SOF = 0.823(10)) and O2A (SOF = 0.177
(10))] and is not unknown in DNSA species, being found in
its compounds with nicotinamide (SOF 0.76:0.24) [40],
2,6-diaminopyridine (0.90:0.10) [2] and quinaldic acid
(0.59:0.41) [3] as well as in the adduct with urea (0.80:0.20) [5].
However the structure of 2 is quite well defined as
proton-transfer having extensive hydrogen bonding
(Fig. 3). The aminium group of the OAP cation molecule
forms a primary Type 1 hydrogen-bonding association with
˚
(a1,2) of 3.53(1) A and 2.1(5)°, respectively.
[(C6H8NO)?(C7H3N2O7)-] (2)
˚
a DNSA carboxylate-O [N11–H_O72, 2.777(5) A], while
the other two protons give three further associations, two to
The structure of the compound of 2-hydroxyaniline with
DNSA, [(OAP)?(DNSA)-], 2 (Fig. 1b) suffers from both a
proximal phenolic (O2) and nitro (O31) oxygen acceptors
123

1658
J Chem Crystallogr (2011) 41:1649–1662
Type 1, N?–H_O(carboxyl) interaction [N11–H_O71ii,
˚
2.750(3) A; symmetry code: (ii) x ? 1, y ? 1, z]. This,
together with secondary linkages, including an asymmetric
R21(6) association with DNSA phenolate- and nitro-O
acceptors (Table 2c) link the cation–anion pairs into sheets
perpendicular to the b axis of the cell, giving a two-
dimensional network structure which involves no cation–
anion p–p associations (Fig. 4). No interactions involving
the oxygen of the substituent methoxy group are found and
this group is close to co-planar with the parent ring [torsion
angle C31–C21–O21–C211, -2.4(3)°].
[(C7H10NO)?(C7H3N2O7)-] (4)
The proton-transfer compound from the reaction of
m-anisidine (MAN) with DNSA, [(MAN)?(DNSA)-], 4
(Fig. 1d) shows the primary Type 1 N?–H_O(carboxyl)
i
˚
interaction [N11–H_O71 , 2.770(3) A: symmetry code:
(i) x, y - 1, z]. Three other interactions (Table 2d) involving
the MAN aminium protons and oxygen acceptors of the
DNSA anions (including a symmetric three-centre R21(6)
proximal extension) link the cations and anions into cen-
trosymmetric heterotetramer units about an R44(12) ring
system, which is extended into one-dimensional ribbons
along the b axial direction (Fig. 5). The alternating stacks of
DNSA anions forming down the approximate c axial direc-
tion are separated by distances which suggest anion–anion
p–p interactions [ring centroid separation Cg_Cg (C1–C6),
Fig. 2 A perspective view of the packing of 1 in the unit cell viewed
down a. Hydrogen-bonding associations are shown as dashed lines. For
symmetry codes in this figure and in Figs. 3, 4, 5, 6, 7, 8, see Table 2
[R21(6)] and the third to an OAP phenolic-O (O21). The
OAP phenolic group is also strongly linked to the second
DNSA carboxylate-O extending the structure in the b-cell
˚
3.480(2) A]. The cation–anion separations are larger
˚
[3.892(2) A] but these associations result in the structure
being layered. As with the OAN cations in 3 the methoxy
groups are not involved in hydrogen bonding and the meth-
oxy-O is essentially coplanar with the benzene ring.
i
˚
direction [O21–H_O71 , 2.593(4) A; symmetry code:
(i) x, y ? 1, z] (Fig. 3). This gives a two-dimensional
hydrogen-bonded network structure which does not involve
the second nitro group nor is there any evidence of any p–p
stacking effects. Also present in the OAP cation is an
intramolecular hydrogen bond between the aminium group
[(C6H7FN)?(C7H3N2O7)-] (5)
The compound of 4-fluoroaniline with DNSA, [(FLAN)?
(DNSA)2], 5 (Fig. 1e) has a primary symmetric Type 2
anilinium N?–H_O,O0(carboxyl) association [N11–
˚
and the phenolic-O [2.689(5) A].
[(C7H10NO)?(C7H3N2O7)-] (3)
ii
ii
˚
˚
H_O71 , 2.965(5) A; _O72 , 3.091(5) A: symmetry
code: (ii) -x ? 1, -y ? 1, -z ? 1] and is propagated
through a number of secondary interactions (Table 2e),
including a symmetric R21(6) proximal association with
The structure of the DNSA compound with 2-methoxyan-
iline, [(OAN)?(DNSA)-], 3 (Fig. 1c) is based on a primary
Fig. 3 A view of the packing of
2 in the unit cell viewed down
b. The minor disordered
phenolic group is not shown
123

J Chem Crystallogr (2011) 41:1649–1662
1659
Fig. 4 A perspective view of
the packing of 3 in the unit cell
viewed down c
Fig. 5 A perspective view of
the packing of 4 in the unit cell
viewed down c showing partial
anion ring superimposition
O(phenol) and O(nitro) acceptors, giving a one-dimen-
sional hydrogen-bonded chain structure which extends
down the c cell direction (Fig. 6). No cation–anion p–p
interactions are present.
acceptors (Table 2f), including an asymmetric R21(6) asso-
ciation with O(phenol) and O(nitro) acceptors as well as
other minor C–H_O interactions. No cation–anion p–p
interactions are present
[(C6H7ClN)?(C7H3N2O7)-] (6)
[(C6H9N2)?(C7H3N2O7)-] (7)
The compound of 4-chloroaniline with DNSA, [(CLAN)?
(DNSA)2], 6 (Fig. 1f) gives a three-dimensional hydrogen-
bonded framework structure (Fig. 7). The primary anilini-
um N?–H_O(carboxyl) association is asymmetric Type 2
[N11–H_O71 , 2.814(4) A; _O72 , 3.208(4) A: sym-
metry code: (ii) x, y, z - 1]. This unit is propagated
through five secondary N?–H_O,O0 hydrogen-bonding
interactions to carboxyl, phenolic and nitro oxygen
The 1:1 compound of 2-aminoaniline with DNSA,
[(OPDA)?(DNSA)-], 7 (Fig. 1g) appears to be the same
compound as that reported by Issa et al. [20], on the basis
of its melting point. This would confirm the observation
that unlike benzidene which forms both 1:1 and 1:2 com-
pounds with DNSA, 2-aminoaniline favours a 1:1 stoichi-
ometry. With 7, only one of the substituent amine nitrogens
(N11) is protonated and this and the second amine group
iii
iii
˚
˚
123

1660
J Chem Crystallogr (2011) 41:1649–1662
Fig. 6 A view of the packing of
5 in the unit cell viewed down
the c cell direction
Fig. 7 A perspective view of
the packing of 6 in the unit cell
viewed down the approximate
c cell direction
are involved in a total of seven hydrogen-bonding associ-
ations. Five of these are with protons from N11 and two
with those of the second group (N21), and all but one of
these are with DNSA oxygen acceptors. The exception is
an intermolecular N?–H_N(amine) association [N11–
association, give a two-dimensional sheet structure (Fig. 8)
which shows no p–p interactions.
Structural Systematics of the DNSA–Aniline
Compounds
i
˚
H_N21 , 2.909(3) A: symmetry code: (i) -x ? 1, -y,
-z ? 2]. The primary aminium-carboxylate interaction is
Considering the overall structural features of the com-
pounds 1–7 and the previously reported DNSA structures
with the analogous isomeric monoamino-substituted ben-
zoic acids, in the light of the general systematics previously
derived for the general series of proton-transfer com-
pounds, it may be predicted that all anilinium salts of
i
i
˚
symmetric Type 2 [N11–H_O71 , 2.914(3) A; _O72 ,
˚
3.050(3) A] while the unprotonated amine group involves a
single direct interaction to the same carboxylate group.
These, together with peripheral extensions (Table 2g),
including
a
symmetric cyclic R21(6) N?–H_O,O0
123

J Chem Crystallogr (2011) 41:1649–1662
1661
Fig. 8 A view of the packing of
7 in the unit cell viewed down b
DNSA should fall into the standard Type 1 and Type 2
modes of primary heteromeric cation–anion interaction. An
unusual Type 1 variant is found with 1 where the duplex
N?–H_O interaction involving two different anilinium
protons to two separate carboxyl groups (found only in the
analogous compound with 2-aminobenzoic acid [9]). Due
to the nature and disposition of the substituent functional
groups on the DNSA anion species, particularly the prox-
imal carboxyl, phenol and nitro groups, the secondary
propagation in the molecular assembly process will most
commonly involve either symmetric or asymmetric three-
centre cyclic N?–H_O,O0(phenolic, nitro) [R21(6)] inter-
actions. This is present in the six compounds of this series,
as well as the three compounds with the isomeric monoa-
minobenzoic acids [9]. There is also a strong tendency for
the generation of two- rather than three-dimensional
hydrogen-bonded structures, such as those usually found
with the aliphatic primary amine salts, while only a minor
incidence of aromatic C–H_O and p–p interactions con-
tribute to molecular assembly. However, the presence of
the aromatic ring does affect the dimensionality of the
structural extension in which layering is common.
This is because of the presence of the proximally located
carboxylic acid/carboxyl group, the phenolic/phenate group
and the N3 nitro group. All are commonly involved in het-
eromolecular hydrogen-bonding interactions. The N5 nitro
group is potentially available for hydrogen bonding but
participates to a much lesser extent in molecular assembly, in
the present series only in one structure (7). These important
structural parameters are seen in Tables 2 and 3.
The carboxylate and phenolic groups are, with two
exceptions (the DNSA dianions in [2(EN)2?2(DNSA)2-
ꢀ
(H2O)] [1] and the mixed piperidinium DNSA–picrate
compound [3(PIP)?(DNSA)2-0.79(DNSA)-ꢀ0.21(PA)-ꢀ
H2O] [14]), involved in an intramolecular hydrogen bond
(the range for the 48 members of the overall series,
˚
including 1–7 is 2.409(3) A (the compound with hexa-
˚
methylenetetramine [1]) to 2.540(3) A (the compound with
8-aminoquinoline [24])). The mean for the same series
˚
(2.463 A) is very similar to the values for the current set
˚
(1–7) [range: 2.435(3)–2.495(2) A]. It is subsequently
expected that the carboxylate group should be essentially
coplanar with the benzene ring and this is reflected in the
C2–C1–C7–O71 torsion angle (Table 3 for the 1–7 set and
Refs. [1–3] for the other reported structures), where the
maximum deviation from coplanarity (compound 5, ca. 9°)
is typical.
Structural Systematics of the DNSA Anions
The stereochemical features of the of the DNSA anion
molecules in both proton-transfer compounds and
co-crystalline adducts contribute significantly to the par-
ticular efficacy of this molecule in structural assembly.
The proximal C3 nitro group is commonly involved in
cyclic R21(6) extension interactions and is consequently
found to be rotated out of the plane of the benzene ring
with a higher incidence and generally to a greater extent
123

1662
J Chem Crystallogr (2011) 41:1649–1662
12. Smith G, Wermuth UD, Healy PC, White JM (2006) Aust J Chem
59:320
13. Smith G, Wermuth UD, Young DJ, Healy PC (2007) Acta
Crystallogr E 63:o2517
14. Smith G, Wermuth UD, Healy PC (2006) Acta Crystallogr E
62:o610
15. Smith G, Wermuth UD, White JM (2006) Acta Crystallogr C
62:o402
than the less interactive C5 nitro group. This is reflected in
the C2–C3–N3–O32 torsion angle which within the overall
series is commonly 20° from planarity (maximum ca. 37°
in 5 in this series), compared to the C4–C5–N5–O52 tor-
sion angle where the maximum deviation for this series is
ca. 16° (compound 2), but is commonly ca. 10°.
16. Etter MC, Adsmond D (1990) J Chem Soc Chem Commun, 589
17. Etter MC, Frankenbach GM (1989) Materials 1:10
18. Issa YM, Hindawey AM, Issa RM, Nassar AMG (1980) Rev
Roum Chim 25:1535
Supplementary Material
19. Hindawey AM, Nassar AMG, Issa RM, Issa YM (1980) Ind J
Chem A 19:615–619
20. Issa YM, Hindawey AM, El-Kholy AE, Issa RM (1981) Gazz
Chim Ital 111:27
21. Ng SW, Naumov P, Drew MGB, Wojciechowski G, Brzezinski B
(2001) J Mol Struct 595:29
22. Smith G, Wermuth UD, White JM (2005) Acta Crystallogr C
61:o464
CCDC depositions 752776–752782 contains supplemen-
tary crystallographic data for compounds 1–7 respectively
for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/datarequest/cif by e-mailing data-
request@ccdc.cam.ac.uk, or contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ.
23. Etter MC, MacDonald JC, Bernstein J (1990) Acta Crystallogr B
46:256
24. Smith G, Wermuth UD, Bott RC, White JM, Willis AC (2001)
Aust J Chem 54:165
25. Smith G, Wermuth UD, White JM (2001) Aust J Chem 54:171
26. Song W-D, Guo X–X, Yu L (2007) Acta Crystallogr E 63:o1890
27. Smith G, Lynch DE, Byriel KA, Kennard CHL (1996) Acta
Crystallogr C 52:231
Acknowledgments The authors acknowledge financial support
from the Faculty of Science and Technology, Queensland University
of Technology, the School of Biomolecular and Physical Sciences,
Griffith University and the School of Chemistry, University of
Melbourne.
28. Smith G, Wermuth UD, Healy PC (2002) Acta Crystallogr E
58:o845
References
29. Smith G, Wermuth UD, White JM (2005) Acta Crystallogr E
62:o746
30. Smith G, Bott RC, Wermuth UD (2001) Acta Crystallogr E
57:o640
1. Smith G, Wermuth UD, Bott RC, Healy PC, White JM (2002)
Aust J Chem 55:349
31. Mohamed HA, El-Medani SM, Ramadan RM (2005) J Ind Chem
Soc 82:799
2. Smith G, Wermuth UD, Healy PC, White JM (2002) Aust J Chem
56:707
32. Subashini A, Samuel E, Muthiah PT, Bocelli G, Cantoni A (2007)
Acta Crystallogr E 63:o4049
3. Smith G, Wermuth UD, Healy PC, White JM (2007) Aust J Chem
60:264
33. Sheldrick GM (2008) Acta Crystallogr A 64:112
34. Sheldrick GM (2008) SHELX97, program for crystal structure
4. Kumar VSS, Kuduva SS, Desiraju GR (2002) Acta Crystallogr E
58:o865
¨
determination. University of Gottingen
5. Smith G, Baldry KE, Byriel KA, Kennard CHL (1997) Aust J
Chem 50:727
6. Smith G, Coyne MG, White JM (2000) Aust J Chem 53:203
7. Bott RC, Smith G, Wermuth UD, Dwyer NC (2000) Aust J Chem
53:767
35. Molecular Structure Corporation (1999) TeXsan for Windows.
Version 1.06. MSC, The Woodlands
36. Spek AL (2003) J Appl Crystallogr 36:7
37. Spek AL (2003) PLATON: A crystallographic computing suite
38. Allen FH, Raithby PR, Shields GP, Taylor R (1998) J Chem Soc
Chem Commun, 1034
8. Kumar VSS, Nangia A, Katz AK, Carrell HL (2002) Cryst
Growth Des 2:313
39. Sundaralingam M, Jensen LH (1965) Acta Crystallogr 18:1053
40. Koman M, Martiska L, Valigura D, Glowiak T (2003) Acta
Crystallogr E 59:o441
9. Smith G, Lynch DE, Byriel KA, Kennard CHL (1995) Aust J
Chem 48:1133
10. Kumar VSS, Kuduva SS, Desiraju GR (1999) J Chem Soc Perkin
Trans 2:1069
11. Smith G, Wermuth UD, White JM (2002) Acta Crystallogr E
58:o1315
123