Rare stoichiometry of carboxyl-carboxylate benzbetaine complexes: In vitro versus in silico

Article abstract of DOI:10.1039/c5ce00054h

The first betaine-mineral acid complex of 3:2 stoichiometry, as well as a 1:1 complex common for this class of compounds, has been obtained. Crystalline [di-(3-trimethylammonium-benzoic acid)-(3-trimethylammonium-benzoate)]diiodide (1) is the only betaine

Full text of DOI:10.1039/c5ce00054h

CrystEngComm
PAPER
Rare stoichiometry of carboxyl–carboxylate
benzbetaine complexes: in vitro versus in silico†
Cite this: CrystEngComm, 2015, 17,
143
4
Kinga Ostrowska, Iwona Kowalczyk, Michał Kaźmierczak, Andrzej Katrusiak,*
Mirosław Szafran, Anna Komasa and Zofia Dega-Szafran
The first betaine–mineral acid complex of 3 : 2 stoichiometry, as well as a 1 : 1 complex common for this
class of compounds, has been obtained. Crystalline ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
trimethylammonium-benzoate)]diiodide (1) is the only betaine–mineral acid 3: 2 complex obtained so far.
Two concomitant polymorphs of this 3: 2 complex, 1-I and 1-II, have been isolated and characterized by
1
X-ray diffraction and FTIR spectroscopy. Polymorphs 1-I, of monoclinic space group P2 /c, and 1-II, of
orthorhombic space group Pccn, have very similar lattices and aggregates of one zwitterionic molecule
and two cations O–H⋯O bonded into a 3-membered catemeric carboxyl–carboxylate–carboxyl interval,
but polymorph 1-II dissolves better in methanol than 1-I. Alternative hypothetical aggregates of
Received 9th January 2015,
Accepted 19th April 2015
3-trimethylammonium-benzoate hydroiodide 1 : 1, 2 : 1 and 3: 2 complexes optimized at the B3LYP/6-31G-
DOI: 10.1039/c5ce00054h
IJd,p) level of theory suggest that the betaine-to-acid 3: 2 stoichiometry is favoured at the stage of small
www.rsc.org/crystengcomm
cluster aggregation and not due to the complex crystal packing preferences.
groups: neutral organic acids, their salts, and zwitterionic
molecules, like amino acids and betaines. In the salts of
Introduction
7
The carboxylic acid aggregates have been extensively investi-
gated for decades due to their basic significance in chemical
1
sciences, practice and manifold applications. The H-bonding
patterns of neutral carboxylic acid groups are often consid-
ered as the most predictable ones. In most cases molecules
with the COOH group aggregate into dimers, described by the
2
2
2
graph set R (8), while much less frequent are chains, so
called catemers, described by the graph set C(4). Also other
aggregates of the carboxylic acid are possible, like tetra- or
hexamers. Our survey of monocarboxylic acids in the Cam-
3
bridge Structural Database (CSD – version November 2013)
2
revealed 3449 R (8) dimers, 408 C(4) catemers, one 3-mem-
bered cyclamer, seventeen 4-membered cyclamers and two
-membered cyclamers. However, equally important finite
2
4
5
,6
6
OH⋯O bonded catemeric intervals of carboxyl–carboxylate
groups have been studied much less and no rules describing
their aggregation have been reported. There are three groups
of compounds containing the carboxyl and carboxylate
Faculty of Chemistry, Adam Mickiewicz University, Umultowska 89b, 61-614
Poznań, Poland. E-mail: hpc.amu.edu.pl, katran@amu.edu.pl
†
Electronic supplementary information (ESI) available: Detailed experiment and
structure description, atomic coordinates and equivalent isotropic displacement
parameters, selected bond lengths, bond angles and torsion angles, hydrogen
bonds and shortest contact distance (Tables S1–S8), the reciprocal space
projections for the diffraction patterns, comparison of the aggregates and crystal
structures by their superposition, the differential scanning calorimetry diagram,
and the X-ray powder diffraction diagrams (Fig. S1–S6). For ESI and crystallo-
graphic data in CIF and other electronic format, see DOI: 10.1039/c5ce00054h
Scheme
complexes
1
Examples
of
carboxyl–(carboxylate
stoichiometries; (a)
anion) : cation
sodium 3,5-
of different
8
9
dinitrobenzoate; (b) potassium hydrogen bisIJ4-nitrobenzoate); (c)
ij3,5-dinitrobenzoate–IJ3,5-dinitrobenzoic acid) ] potassium IJbenzo-15-
crown-5); and (d) di[tetramethylammonium difumarate].
2
1
0
11
This journal is © The Royal Society of Chemistry 2015
CrystEngComm, 2015, 17, 4143–4149 | 4143

Paper
CrystEngComm
organic acids, if one carboxyl group is deprotonated, the most
common aggregation type is a dimer, but H-bonded aggre-
gates built of three, four or more carboxyl–carboxylate units
are also possible. Such aggregates require only one cation to
balance the charge equal to −1 and it is independent of the
8
–11
number of acid units in the aggregate (Scheme 1).
Our
survey of the CSD aimed at identifying the interval aggregates
of carboxylic and carboxylate groups with cationic counter-ions
yielded 1815 dimers, 206 trimers and 32 tetramers.
Betaines and amino acids are zwitterions with a positively
charged nitrogen atom and negatively charged carboxylate
7
group. The best known and most common are their base–
acid hydrogen-bonded aggregates of 1 : 1 stoichiometry. The
base molecule is protonated and the carboxyl group is hydro-
gen bonded to an anion (COOH⋯anion). There are 318 such
structures of betaine salts deposited with the CSD. The next
most frequently represented are the aggregates of stoichiome-
try 2 : 1, where the protonated betaine interacts by
COOH⋯OOC hydrogen bond with the zwitterion. Such a
homoconjugated cationic dimer is stabilized also by electro-
static forces between itself and the counter-anion. There are
3
2 such salts deposited with the CSD. There are no com-
Scheme
2
Examples of trimethylammonium derivatives: (a)
16
plexes of stoichiometry 3 : 2 and no betaines forming O–
H⋯O bonded complexes of different stoichiometries with the
same acid.
The group of compounds 2-, 3- and 4-IJtrimethylammonium)-
benzoates (benzbetaines), their esters, hydrates and salts with
4-trimethylammonium benzoate; (b) 4-trimethylammonium benzoic
15
acid chloride;
(c) bisIJ4-trimethylammonium-benzoate)hydrogen
and (d) ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
trimethylammonium-benzoate)]diiodide (this work).
18
iodide;
12–20
halogen anions,
is particularly important due to their bio-
15
chemical significance as well as applications in the pharmaceu-
tical and the nutrition industries. Of the trimethylammonium
benzoates, the adducts with halides in the ratios 1 : 1 (ref. 12,
4-trimethylammonium-benzoic acid chloride, 4-trimethylamm-
onium-benzoate hydrate, methyl-4-trimethylammonium-benzoic
acid iodide,
iodide hydrate and methyl ester of 3-trimethylammonium
benzoic acid iodide. Tadros and Sakla showed the con-
nection between the quaternization of dimethylamino-
16
1
7
bisIJ4-trimethylammonium-benzoate) hydro-
1
8
15, 20) and 2: 1 (ref. 18) have been obtained (Scheme 2). Their
1
9
20
structures were studied by diffraction, spectroscopy and theoret-
ical methods. Recently, we have studied the structures of
Fig. 1 The symmetry-independent unit in (a) polymorphs 1-I and 1-II of ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-trimethylammonium benzo-
ate)]diiodide and (b) 3-trimethylammonium-benzoic acid iodide (2). The hydrogen bonds have been marked by cyan dashed lines. Thermal ellip-
soids have been drawn at the 50% probability level.
4144 | CrystEngComm, 2015, 17, 4143–4149
This journal is © The Royal Society of Chemistry 2015

CrystEngComm
Paper
benzoic acid in an excess of methyl iodide in methanol and
the temperature and time of the reaction, as well as the posi-
tion of the dimethylamino group at the benzoic acid ring.
To the best of our knowledge no bisIJ3-trimethylamm-
onium-benzoate) hydrohalide 2 : 1 salts have been obtained
until now. Also, no salts in the zwitterion : anion 3 : 2 ratio, or
any of higher stoichiometries described by the general for-
mula n : (n − 1), were found in the CSD. Thus the structure of
was added and stirred at 25 °C for 35 h. The crude product
was washed with acetonitrile. Yellow crystals were obtained,
m.p. 211–213 °C. Analysis for C10H14NO I, m.wt. 309.13; calc.
2
%C, 39.11; %H, 4.59; %N, 4.56; found: %C, 39.24; %H, 4.52;
%N, 4.31.
We tried to obtain complexes of the 2 : 1 stoichiometry by
increasing and reducing the acid concentration; however,
all subsequent analyses of the precipitate showed that either
1 : 1 and 3 : 2 complexes were obtained or, in the excess of HI,
esterification took place and the compound was converted
to the methyl ester of 3-trimethylammonium benzoic acid
ijdi-IJ3-trimethylammonium-benzoic
acid)–IJ3-trimethylamm-
onium-benzoate)]diiodide (1) presented in this work is the
first example of the 3 : 2 aggregate. In the reiterated attempts
to change the ratio of this complex we found that 1 exists in
two polymorphic forms, 1-I and 1-II (Fig. 1). They have been
studied by single-crystal X-ray diffraction, thermal DSC analy-
sis, FTIR spectroscopy and DFT calculations. The main objects
of our concern were the O–H⋯O catemeric interval aggregates
of carbonyl and carboxylate groups in the zwitterionic type
compounds. The 1 : 1 complex 3-trimethylammonium-benzoic
acid iodide has the expected O–H⋯I bonded structure.
1
9
iodide.
Measurements
The single-crystal X-ray diffraction data of 3-trimethylamm-
onium-benzoic acid iodide 1 : 1 complex
2 and ijdi-IJ3-
trimethylammonium-benzoic acid)–IJ3-trimethylammonium-
benzoate)]diiodide, polymorphs 1-I and 1-II, were measured
using an Oxford Diffraction Excalibur EOS-CCD diffractome-
ter. The structures were solved by direct methods and refined
Experimental
Syntheses
2
1
using SHELXL. H atoms were located according to the ideal
geometry (hydroxyl 0.82 Å, arene 0.93 Å, and methyl 0.96 Å).
Anisotropic thermal parameters were refined for non-H-
atoms, and isotropic Uiso parameters for H atoms were
linked to Ueq of their carriers. The crystal data, details of data
collections and structure refinements are given in Tables 1
and S1–S2† and the final atomic coordinates are in Tables S3
and S4 (ESI†). The crystal data have been deposited with the
Cambridge Crystallographic Database Centre as supplemen-
tary publications CCDC 1041713 (polymorph 1-I), 1041714
(1-II) and 1056077 (2).
Preparation of ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
trimethylammonium-benzoate)]diiodide 1 (complex 3 : 2)
A solution of 3-dimethylamino-benzoic acid (2.56 g; 0.0155 M)
3
in methanol (30 cm ) and methyl iodide (1.47 g, 0.103 M)
was stirred at 25 °C for 28 h. The crude product, 2.93 g, was
filtered off, washed with a small amount of acetonitrile.
ijDi-IJ3-trimethylammonium-benzoic acid)–IJ3-trimethylammonium-
benzoate)]diiodide in the form of colourless crystals was sepa-
rated. After recrystallization from an ethanol–acetonitrile
(
10 : 1) mixture, m.p. 191–192 °C, it was identified as 1, with
The FTIR spectra of 1 and 2 were measured using the KBr
disks on a Bruker IFS 66v/S instrument, with a resolution of
a considerable content of polymorph 1-I. The filtrate was
evaporated to half of the volume and the remaining crystals,
mainly of 1-I, were filtered off, 0.05 g. From the filtrate evapo-
rated to dryness, 1.01 g of the crystals of polymorph 1-II
could be isolated and recrystallized from a mixture of aceto-
nitrile–methanol (3 : 1), m.p. 188–189 °C, and then several of
the picked crystals were identified as polymorph 1-II. This
suggests that polymorph 1-II is more soluble in methanol
than 1-I. The powder diffraction patterns of the products
always contained a mixture of polymorphs 1-I and 1-II,
and also of another compound that was later identified as
the 1 : 1 complex (see below). The ratio of 1-I to 1-II is 3 : 1.
−
1
2 cm . The spectra were recorded by accumulation of 64
scans. The NMR spectra of 1 were recorded using a Varian
VNMRS-400 spectrometer operating at 402.6435 and 101.2440
1
13
MHz for H and C, respectively. The spectra were measured
in DMSO-d relative to an internal standard of TMS. The ther-
6
mal analysis of 1-I was performed with a DSC THASS XP-10
Table 1 Selected crystal data of complex 3 : 2, polymorphs 1-I and 1-II,
and complex 1:1 (2) of 3-trimethylammonium-benzoate with hydroiodic
acid (cf. Tables S1–S2 in the ESI)
1
NMR (DMSO-d , ppm); H: 3.68, CH ; 7.80, H(5); 8.13, H(6);
8
1-I
1-II
2
6
3
1
3
.30, H(4); 8.41, H(2); C: 56.38, CH ; 121.11, C(2); 125.10,
3
Formula weight
C
H I N O
30 41 2 3 6
C
H I N
30 41 2 3
O
6
10 2
C H14INO
C(6), 130.63, C(4); 130.67, C(5); 132.59, C(1); 147.40, C(3); 166.05,
C(7). Analysis for C H N O I , m.wt. 793.48; calc. %C, 45.41;
Crystal system
Space group
Unit cell dimensions
Monoclinic
P2 /c
Orthorhombic Monoclinic
1
Pccn
1
P2 /n
30 41 3 6 2
%H, 5.20; %N, 5.30; found: %C, 45.23; %H, 4.95; %N, 5.18. All
a (Å)
b (Å)
c (Å)
β
18.7965IJ10)
13.9047(8)
12.8843(6)
96.010IJ4)°
3348.9(3)
4
37.3738IJ17)
13.9167IJ11)
12.8674(7)
90°
6692.6(7)
8
8.1929(2)
11.3278(3)
13.1285(3)
105.735IJ3)°
1172.76(5)
4
attempts to obtain one of the polymorphs from recrystalliza-
tion resulted in concomitant polymorphs 1-I and 1-II.
Preparation of 3-trimethylammonium-benzoic acid iodide
3
Volume (Å )
2
(complex 1 : 1)
Z
To a solution of 5 g (0.03 M) of 3-dimethylamino-benzoic
acid in 30 ml of methanol, 3.3 g (0.031 M) of methyl iodide
Calculated density
1.574
1.575
1.739
−
3
(g cm )
This journal is © The Royal Society of Chemistry 2015
CrystEngComm, 2015, 17, 4143–4149 | 4145

Paper
CrystEngComm
calorimeter. The sample with a mass of 6.3 mg was stored in
an Al O crucible in helium atmosphere and the rate of tem-
2
3
−
1
perature change was 10 °C min .
DFT calculations
The DFT calculations were performed with the GAUSSIAN 09
2
2
program package. The calculations employed the B3LYP
exchange–correlation functional, which combines the hybrid
2
3,24
exchange functional of Becke
with the gradient-
2
5
correlation functional of Lee et al. and the split-valence
polarized 6-31G(d,p) basis set. The 6-31G(d,p) basis set by
Glukhovtsev et al. for iodine has been acquired from the
Fig. 3 Superimposed OH⋯O bonded aggregates and anions of ijdi-IJ3-
2
6
trimethylammonium-benzoic
acid)–IJ3-trimethylammonium-
2
9
benzoate)]diiodide in polymorph 1-I (green) and polymorph 1-II
yellow). The superimposed aggregates viewed from other directions
(
2
7,28
EMSI Basis Set Library.
The geometries in polymorphs 1-I
are shown in the ESI Fig. S2 and S3.†
and 1-II and in complex 2 were used as starting models for
the structure optimizations. All calculated IR frequencies are
positive, consistent with the minimum energies of the opti-
mized structures.
The structure of 1 : 1 complex 2 is built of OH⋯I bonded
ionic pairs (Fig. 1); the O⋯I distance is 3.446(6) Å. The next
shortest interionic bonds are all longer than the sum of the
van der Waals radii, except CIJ5)HIJ5)⋯OIJ2′) (symmetry code
Discussion
Crystal structure
1
.5 − x, −0.5 + y, 1.5 − z), with a HIJ5)⋯OIJ2′) distance of 2.476 Å
The crystals of ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
trimethylammonium-benzoate)]diiodide, 1, are composed of
two 3-IJtrimethylammonium)-benzoic acid cations (A and C),
one 3-IJtrimethylammonium)-benzoate zwitterion (B) and two
iodide anions (Fig. 1). Cations A and C are joined to both oxy-
gen atoms of the carboxylate group of zwitterion B by two
non-equivalent hydrogen bonds, OIJ1)–HIJ1)⋯OIJ11) and
OIJ21)–HIJ21)⋯OIJ12) (Fig. 1). The crystals of polymorph 1-I are
monoclinic, with space group P2 /c, while those of 1-II are
1
orthorhombic, with space group Pccn (Table 1).
The crystals of polymorphs 1-I and 1-II can be distin-
guished by their morphology (Fig. 2). In their structures, the
positions and orientations of the complex aggregates are very
similar and the main difference is in the location of the
shorter than the sum of van der Waals radii according to
Bondi by 0.274 Å.
The dimensions of units A, B and C (see Fig. 1) in both
polymorphs 1-I and 1-II are consistent within standard devia-
tions (Tables S5–S6 in the ESI†).
The presence of the COOH groups, in polymorphs 1-I and
1-II, in cations A and C is confirmed by C–OH bonds longer
than 1.300(6) Å and by CO bonds shorter than 1.204(6) Å.
In zwitterion B, the CIJ17)–OIJ1) and CIJ17)–OIJ2) bond lengths
are 1.255(10) and 1.240(11) Å in polymorph 1-I, and 1.256(6)
and 1.243(6) Å in polymorph 1-II, respectively. The dimensions
of hydrogen bonds OIJ1)–HIJ1)⋯OIJ11) and OIJ21)–HIJ21)⋯OIJ12)
linking cations A and C with zwitterion B, are very similar, in
polymorph 1-I of 2.529(9) and 2.533(9) Å, and also similar
in polymorph 1-II of 2.493(5) and 2.549(5) Å, respectively
3
-trimethylammonium group of the zwitterion (Fig. 3).
(
Table S7†). The aggregates are surprisingly similar in their
shape, too (Fig. 3): the planes fitted to cations A and C are
inclined to the plane of zwitterion B by 40.70° and 44.21° in
1
-I, and 41.21° and 45.86° in 1-II, respectively. Likewise, the
lattices (Table 1) of polymorphs 1-I and 1-II are surprisingly
similar, including the a-parameter, which is almost exactly
two times longer in 1-II than in 1-I (Fig. 4a and b). The super-
position of both structures shown in Fig. S5† illustrates
the differences mainly associated with the orientation of the
nearly identical aggregates (Fig. 3), as can be seen in the posi-
tion of trimethylammonium groups between Fig. 4a and b.
The iodide anions are located close to the nitrogen atom
of zwitterion B in the structures of both polymorphs; however
the NIJ11)⋯IIJ2) distances of 4.435(7) in 1-I and 4.433IJ4)Å in
1
4
-II are somewhat longer than the NIJ11)⋯IIJ1) distances of
.374(8) and 4.375IJ5)Å, respectively (Table S8†). Additionally,
Fig.
2 Crystals of ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
the hydrogen atoms of the methyl groups interact with the
iodide anions as well as with the ring protons at the
para-position. These interatomic distances vary from 3.900 to
3.968 Å (Table S7 in the ESI†).
trimethylammonium-benzoate)]diiodide: (a) polymorph 1-I and (b)
polymorph 1-II. Apart from these pure forms of concomitant
polymorphs 1-I and 1-II, their twin-like agglomerates also were
formed, with the “twinning plane” (010) (see Fig. S1†).
4146 | CrystEngComm, 2015, 17, 4143–4149
This journal is © The Royal Society of Chemistry 2015

CrystEngComm
Paper
DFT calculations
Three models of the adducts of 3-trimethylammonium-
benzoic acid with iodide in ratios of 1 : 1 (2′), 2 : 1 (3) and 3 : 2
(
4) were optimized at the B3LYP/6-31GIJd,p) level of theory
and their structures are shown in Fig. 5. In the optimized
structure 2′, the electrostatic attraction between the positively
charged nitrogen atom and the iodide anion significantly
contributes to the aggregation energy, as indicated by a short
+
−
N ⋯I distance of 4.306 Å (Table S8 in the ESI†). This differ-
ence illustrates the competition between the OH⋯I hydrogen
bond prevailing in the crystal and the electrostatic forces
between the charged trimethylammonium group and the
iodide anion in the optimized isolated complex.
Model 3 represents a hypothetical hydrogen-bonded com-
plex of stoichiometry 2 : 1. The 3-trimethylammonium-
benzoic acid cation A and its zwitterion B are linked through
the asymmetric and short COOH⋯OOC hydrogen bond of
+
−
2
.503 Å (Table S7 in the ESI†). The N ⋯I distance to cation
A is slightly shorter than that to zwitterion B (Table S8 in the
ESI†). The calculated energy of 3 is −8108.3556063 a.u. and
its dipole moment is 10.64 D.
Fig.
4 Structures of ijdi-IJ3-trimethylammonium-benzoic acid)–IJ3-
trimethylammonium-benzoate)]diiodide complex 3 : 2 projected down
the [y] and [z] axes in (a) polymorph 1-I and (b) polymorph 1-II, as well
as (c) the structure of 3-trimethylammonium-benzoic acid iodide (2)
complex 1 : 1. The hydrogen bonds have been shown as the cyan lines.
The structure of 1 is very similar in its interactions
between the benzoic acid rings to that in the structure of
potassium
IJbenzo-15-crown-5)
ij3,5-dinitrobenzoate–IJ3,5-
1
0
dinitrobenzoic acid)2]. The main difference between these
structures is that catemeric interval 1 has positive charges
−
located on the nitrogen atoms stabilized by two I anions,
while that of 3,5-dinitrobenzoic acids has a negative charge
at the carboxylate group of 3,5-dinitrobenzoate, stabilized by
+
the K cation.
The differential scanning calorimetry (DSC) diagram of 1
(Fig. S4, in the ESI†) contains only the exothermic peak at
1
90 °C. However, this peak clearly consists of two super-
Fig. 5 Computed structures of the 1 : 1 (2′), 2 : 1 (3) and 3 : 2 (4-I and
imposed peaks that can be associated with the different melt-
4
-II) complexes of 3-trimethylammonium-benzoic acid with
ing points of 1-I and 1-II.
hydroiodic acid.
This journal is © The Royal Society of Chemistry 2015
CrystEngComm, 2015, 17, 4143–4149 | 4147

Paper
CrystEngComm
benzoate)]diiodide (1-I and 1-II) are similar. The spectrum is
shown in Fig. 6a. There are two broad absorptions in the
−1
spectrum: at 2480 cm attributed to the νO–H vibration of the
bond engaged in the medium-strong O–H⋯O hydrogen bond
−
1
−1
and in the 1300–850 cm region centered at 1100 cm , which
30
is typical of the O–H⋯O hydrogen bond shorter than 2.5 Å.
The ν COO and νCO vibrations appear at 1683 cm .
−
1
as
Fig. 6b shows the spectrum of trimethylammonium-
benzoic acid iodide (2). The strong and narrow band at
−
1
1
716 cm is attributed to the νCOOH vibration.
Conclusions
In the series of finite zwitterion : protonated-zwitterion car-
boxylate–carboxyl aggregates, a 3-membered interval of ijdi-IJ3-
trimethylammonium-benzoic acid)–IJ3-trimethylammonium-
benzoate)]diiodide complex (1), of stoichiometry 3 : 2, has
been obtained for the first time. Complex 1 crystallizes as
two polymorphs, at least. Also a 1 : 1 complex stoichiometry
of 3-trimethylammonium-benzoate with hydroiodic acid was
obtained in reiterated crystallizations of varied acidity solu-
tions. The interval aggregates in the 3 : 2 complexes are stabi-
lized by two short hydrogen bonds between the carboxylate
and the carboxyl groups, as well as electrostatic forces
between the iodide anions and the quaternary nitrogen
atoms. The theoretically optimized structures of the 1 : 1 and
Fig.
6 Solid-state FTIR spectra of (a) ijdi-IJ3-trimethylammonium-
benzoic acid)–IJ3-trimethylammonium-benzoate)]diiodide
trimethylammonium-benzoic acid iodide 2.
1 and (b)
Model 4 corresponds to the aggregation in the crystals of
1, composed of two cations, a zwitterion and two iodide
anions (Fig. 5). Two structures, 4-I and 4-II, have been opti-
mized, from the starting models of aggregates in polymorphs
2
3
: 1 complexes are much less preferred energetically than the
: 2 complex; however the theoretically optimized 1 : 1 com-
1-I and 1-II, respectively. The optimized model 4-I reproduces
the aggregate well in crystal structure 1-I. The carboxylic
groups of two 3-trimethylammonium-benzoic acid cations, A
and C, interact with the carboxylate oxygen atoms of zwitter-
ion B. The OIJ1)–H⋯OIJ11) and OIJ21)–H⋯OIJ12) distances in
the theoretically optimized structure 4-I are 2.480 and 2.443 Å,
respectively (Table S7†). The iodide anions are located on
both sides of the zwitterion molecule; however they are closer
to the nitrogen atoms of cations IJN ⋯I distances are 4.357
and 4.323 Å, Table S8 in the ESI†) than to the nitrogen atom
of the zwitterion IJN ⋯I distances are 5.019 and 4.678 Å).
The calculated energy of 4-I is −15 622.6608034 a.u. and the
dipole moment is 3.51 D. In model 4-II two cations of
plex significantly differs in the interionic interactions from
the experimentally determined structure, where OH⋯I bonds
are formed. It shows that the energies calculated for isolated
ionic aggregates can be irrelevant for predicting the com-
plexes of different stoichiometries. Nonetheless, the synthe-
sized 1 : 1 (2) and 3 : 2 (1-I and 1-II) complexes are the only
complexes of different stoichiometries between a benzbetaine
and a mineral acid deposited with the Cambridge Structural
Database so far. It is still not clear how strongly this stoichi-
ometry for specific stoichiometries of betaine complexes
depends on the molecular structure of betaines themselves.
Thus the understanding of these preferences still requires
further investigation. The lattice dimensions of polymorphs
+
−
+
−
3-trimethylammonium-benzoic acid, A and B, are linked by
an asymmetric and medium strong OIJ1)–H⋯OIJ11) hydrogen
bond of 2.585 Å. The so-formed dimer is linked to the
1
-I and 1-II are remarkably similar, which result in some sim-
ilarities in their powder-diffraction patterns (Fig. S5†). Com-
plexes 1 : 1 and 3 : 2 are clearly distinguishable by their FTIR
spectra, due to significantly different interactions in these
structures.
3-trimethylammonium-benzoate zwitterion C by the OIJ12)–
H⋯OIJ21) hydrogen bond with a distance of 2.494 Å (Table S7
in the ESI†). The orientation of the carboxylic group is cis in
cation A and trans in cation B. The benzoic acid moieties
have N ⋯I distances varying from 4.270 to 4.793 Å (Table S8
in the ESI†). The calculated energy of 4-II is −15 622.6660630
a.u. and the dipole moment is 7.60 D. The energy difference
between 4-II and 4-I, ΔE = 0.0052596 a.u., indicates that struc-
+
−
Acknowledgements
DFT calculations were performed at the Poznań Super-
computing and Networking Centre and supported in part by
PL-Grid Infrastructure.
−
1
ture 4-II is more stable by 13.8 kJ mol .
FTIR spectra
Notes and references
The solid-state spectra of the polymorphs of ijdi-IJ3-
trimethylammonium-benzoic acid)–IJ3-trimethylammonium-
1 L. Leiserowitz, Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem., 1976, 32, 775–802.
4148 | CrystEngComm, 2015, 17, 4143–4149
This journal is © The Royal Society of Chemistry 2015

CrystEngComm
Paper
2
3
4
5
M. C. Etter, J. C. MacDonald and J. Bernstein, Acta
Crystallogr., Sect. B: Struct. Sci., 1990, 46, 256–262.
F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58,
19 M. Szafran, A. Katrusiak, I. Kowalczyk, A. Komasa and Z.
Dega-Szafran, J. Mol. Struct., 2012, 1017, 115–122.
20 W. Tadros and A. B. Sakla, J. Chem. Soc., 1954, 1116–1119.
21 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
3
80–388.
P. Prayzner, E. C. A. Ojadi, J. A. Golen and P. G. Willard,
Acta Crystallogr., Sect. C: Struct. Chem., 1996, 52, 665–667.
D. Mospoch, N. Domingo, R. Ruiz-Molina, K. Wurst, G.
Vaughan, J. Tejada, C. Rovira and J. Veciana, Angew. Chem.,
Int. Ed., 2004, 43, 1828–1832.
22 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L.
Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E.
Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K.
Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, GAUSSIAN
09, Revision B.01, Gaussian, Inc., Wallingford CT, 2010.
23 D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
6
D. Mospoch, N. Domingo, R. Ruiz-Molina, K. Wurst, J.
Tejada, C. Rovira and J. Veciana, J. Am. Chem. Soc.,
2
004, 126, 730–731.
7
8
9
X. Domingo, in Amphoteric Surfactants, ed. E. G. Lomax,
Marcel Dekker, Inc., New York, 1996.
H. P. Jones, A. L. Gillon and R. J. Davey, Acta Crystallogr.,
Sect. E: Struct. Rep. Online, 2005, 61, m1131–m1132.
H. N. Shrivastava and J. C. Speakman, J. Chem. Soc.,
1
961, 1151–1163.
1
1
1
1
1
1
1
1
1
0 W. Xu, K. Clinger, M. L. Hackert and N. S. Poonia,
J. Inclusion Phenom., 1985, 3, 163–172.
1 J. E. Barry, M. Finkelstein, S. D. Ross, G. D. Mateescu, A.
Valeriu and C. Svensson, J. Org. Chem., 1988, 53, 6058–6061.
2 J. T. Edsall and J. Wyman Jr, J. Am. Chem. Soc., 1935, 57,
964–1975.
3 M. Hojo, M. Utaka and Z. Yoshida, Tetrahedron, 1971, 27,
255–4262.
4 J. Picha, R. Cibulka, F. Liška, P. Pařik and O. Pytela, Collect.
Czech. Chem. Commun., 2004, 69, 2239–2252.
5 M. Szafran, A. Katrusiak, Z. Dega-Szafran and I. Kowalczyk,
J. Mol. Struct., 2011, 996, 75–81.
1
24 D. Becke, J. Chem. Phys., 1997, 107, 8554–8560.
25 C. Lee, W. Yang and G. R. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785–789.
26 W. J. Hehre, L. Random, P. V. R. Schleyer and J. A. Pople, Ab
Initio Molecular Orbital Theory, Wiley, New York, 1986.
27 Feller, J. Comput. Chem., 1996, 17, 1571–1586.
4
6 M. Szafran, A. Katrusiak, Z. Dega-Szafran and I. Kowalczyk,
J. Mol. Struct., 2011, 1005, 144–151.
7 M. Szafran, A. Katrusiak, Z. Dega-Szafran, I. Kowalczyk and
A. Komasa, J. Mol. Struct., 2011, 1006, 330–336.
8 A. Komasa, A. Katrusiak, M. Kaźmierczak, Z. Dega-Szafran and
M. Szafran, Spectrochim. Acta, Part A, 2015, 136, 1149–1156.
28 L. L. Schuchardt, B. T. Didler, T. Eisethagen, L. Sun, V.
Gurumoorthi, J. Chase, J. Li and T. L. Windus, J. Chem. Inf.
Model., 2007, 47, 1045–1052.
29 M. N. Glukhovtsev, A. Pross, M. P. McGrath and L. Radom,
J. Chem. Phys., 1995, 103, 1878–1885.
30 A. Novak, Struct. Bonding, 1974, 18, 177–214.
This journal is © The Royal Society of Chemistry 2015
CrystEngComm, 2015, 17, 4143–4149 | 4149

Copyright of CrystEngComm is the property of Royal Society of Chemistry and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.