Halogen substituted quinolylsalicylaldimines: Four halogens three structural types

Article abstract of DOI:10.1016/j.molstruc.2012.11.028

A series of halogen substituted 5-X-N-(8-quinolyl)salicylaldimines (Hqsal^X, X = F 1, Cl 2, Br 3 and I 4) have been prepared, characterized and the crystal structures of all four are reported. The compounds form stacks, in most cases held togeth

Full text of DOI:10.1016/j.molstruc.2012.11.028

Journal of Molecular Structure 1036 (2013) 439–446
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Halogen substituted quinolylsalicylaldimines: Four halogens three structural types
a
b
Jitnapa Sirirak a, Wasinee Phonsri a, David J. Harding a, , Phimphaka Harding , Pimonrat Phommon ,
⇑
Wannapa Chaoprasa b, Rebecca M. Hendry c, Thomas M. Roseveare c, Harry Adams c
a Molecular Technology Research Unit Cell, Walailak University, Thasala, Nakhon Si Thammarat 80161, Thailand
b Department of Chemistry, Faculty of Science, Naresuan University, Phitsanulok 65000, Thailand
c Department of Chemistry, University of Sheffield, Sheffield S3 7HF, United Kingdom
h i g h l i g h t s
"
"
"
"
"
The structures of four halogen substituted quinolylsalicylaldimines are reported.
The structures exhibit hydrogen bonds,
The chloro derivative features an unexpected ‘inverted’ conformation.
The conformation determines the strength of the hydrogen bond.
DFT calculations indicate conformational preferences are halogen dependent.
p
–
p
and CAHꢀ ꢀ ꢀN/O interactions.
a r t i c l e i n f o
a b s t r a c t
A series of halogen substituted 5-X-N-(8-quinolyl)salicylaldimines (HqsalX, X = F 1, Cl 2, Br 3 and I 4) have
been prepared, characterized and the crystal structures of all four are reported. The compounds form
Article history:
Received 18 October 2012
Received in revised form 12 November 2012
Accepted 12 November 2012
Available online 23 November 2012
stacks, in most cases held together either by
an intramolecular OAHꢀ ꢀ ꢀN hydrogen bond with 2 also displaying an intermolecular OAHꢀ ꢀ ꢀO hydrogen
interactions serve to link neighbouring HqsalX mol-
p–p or lone pair(N)–p interactions. All compounds exhibit
bonding square. Additional CAHꢀ ꢀ ꢀN/O and CAHꢀ ꢀ ꢀ
p
ecules with 3 and 4 forming narcissistic dimers. While the halogen has a profound effect on the structure
it is not involved in either hydrogen or halogen bonding in any of the structures. DFT calculations suggest
that the conformational preference is dependent on the halogen.
Keywords:
Salicylaldimines
p–p Stacking
Hydrogen bonding
Ó 2012 Elsevier B.V. All rights reserved.
DFT calculations
1. Introduction
aforementioned interactions is halogen bonding [12–15]. Halogen
bonds are weak supramolecular non-covalent interactions and
The construction and design of molecules which crystallize in a
predictable manner leading to expected structures remains the
ultimate goal of crystal engineering [1,2]. One of the reasons why
it is so difficult to predict a crystal structure based purely on the
molecular structure relates to the interplay between the various
interactions that in turn determine which structure is the most sta-
ble [3–10]. It follows that understanding how different interactions
combine to form a complete structure and which are the structure
determining interactions is vital if some degree of predictability is
to be achieved [11].
are often similar to hydrogen bonds where the hydrogen is re-
placed by a halogen which acts as a halogen bond donor and are
attributed to electrostatic forces. The halocarbons can effectively
be the electropositive site that can interact with an electron donat-
ing atom/group to produce higher dimensional structures. How-
ever, they may also function as halogen bond acceptors forming
CAXꢀ ꢀ ꢀN/O interactions as well as halogen–halogen interactions
[16–19]. This flexibility is associated with the polarizability of
the halogens which have a positive
r-hole along the CAX axis
and belt of negative charge perpendicular to this CAX bond [20].
The more polarizable the halogen the stronger the interaction will
be, consequently halogen bonding typically increases on moving
from F to I.
Halogen bond strength can possibly be tuned by modifying the
substituents on the carbon skeleton [21]. Our attention was drawn
to the halogen substituted 5-X-N-(8-quinolyl)salicylaldimines
(HqsalX, X = F, Cl, Br and I) [22,23] as these contain two different
To do this requires utilization and understanding of a wide
range of supramolecular interactions including classical hydrogen
bonds [3,4], weak hydrogen bonds (e.g. CAHꢀ ꢀ ꢀN/O interactions)
[5,6] and
p–p
interactions [8,9]. A more recent addition to the
⇑
Corresponding author. Tel./fax: +66 75 672004.
E-mail address: hdavid@wu.ac.th (D.J. Harding).
0022-2860/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2012.11.028

440
J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
Table 1
2.2. X-ray crystallography
Crystallographic data and structure refinement for 1–4.
Crystal data processing parameters for the structures of 1–4 are
1
2
3
4
given in Table 1. X-ray quality crystals of 1–4 were grown by slow
evaporation of a diisopropyl solution of HqsalX. Crystals were
mounted on a glass fibre using perfluoropolyether oil and cooled
rapidly to 150 or 100 K in a stream of cold nitrogen. In the case
of 1 the structure was determined at 293 K. All diffraction data
were collected on a Bruker APEXII area detector with graphite
Formula
Molecular
weight
C
16H11FN2O C16H11ClN2O C16H11BrN2O C16H11IN2O
266.27
282.72
327.18
374.17
(g molꢁ1
)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
Monoclinic
P21/c
10.0551(5)
9.2429(4)
14.1107(7)
90
104.421(2)
90
293(2)
Monoclinic
P21/c
Monoclinic
P21/n
Monoclinic
P21/c
13.4764(14)
3.7721(4)
25.991(2)
90
16.4841(3)
4.68910(10)
17.1739(3)
90
17.014(4)
4.6068(10)
17.418(4)
90
monochromated Mo Ka (k = 0.71073 Å) [24]. After data collection,
in each case an empirical absorption correction (SADABS) was ap-
plied [25], and the structures were then solved by direct methods
and refined on all F2 data using the SHELX suite of programs [26].
In all cases non-hydrogen atoms were refined with anisotropic
thermal parameters; hydrogen atoms were included in calculated
positions and refined with isotropic thermal parameters which
were ca. 1.2 times the carbon atoms. The exception were the hy-
droxyl hydrogens which were located by low-theta difference Fou-
rier and then refined with isotropic thermal parameters 1.5 times
those of the oxygen atoms to which they are attached. All pictures
were generated using the POV-Ray interface in X-SEED [27,28].
a
(°)
b (°)
(°)
104.802(7)
90
99.7460(10)
90
99.835(4)
90
c
T (K)
100(2)
100(2)
1308.31(4)
4
150(2)
1345.1(5)
4
Cell volume (Å3)
Z
1270.10(10) 1277.4(2)
4
0.099
4
Absorption
coefficient
0.295
3.138
2.377
(mmꢁ1
)
Reflections
collected
Independent
reflections,
Rint
Max. and min.
transmission
Restraints/
parameters
Final R indices
16663
6562
13316
11812
3413, 0.027 2267, 0.055
3033, 0.026
3083,
0.055
2.3. Synthesis of 5-fluoro-N-(8-quinolyl)salicylaldimine HqsalF
0.9882 and
0.9795
0/182
0.9740 and
0.8963
0.5726 and
0.4064
0.7800 and
0.5168
0/181
0/181
0/182
5-Fluorosalicylaldehyde (0.701 g, 5 mmol) was dissolved in
diisopropyl ether (20 ml) giving a pale yellow solution. 8-Amino-
quinoline (0.721 g, 5 mmol) was added and the solution stirred
for 4 h giving an orange solution. Slow evaporation of the solution
to ca. 5 ml gave orange crystals which were isolated by filtration,
0.0492,
0.1476
0.0391,
0.1416
0.0246,
0.0767
0.0447,
0.1194
[I > 2r(I)]: R1,
wR2
1.22 g (92%).
m mOH), 3036, 2970 (mCH), 1613
max(KBr)/cmꢁ1 3435 (
(m
C=N). 1H NMR (CDCl3, 295 K; d; ppm) 13.67 (OH), 8.99 (dd, 1Ha,
types of aromatic rings, a hydroxyl group as well as the halogens
and may be expected to also engage in hydrogen bonding, CAHꢀ ꢀ ꢀ
and interactions in addition to any halogen bonds. In examin-
JHH 1.8, 4.2), 8.91 (s, 1Hg), 8.20 (dd, 1Hc, JHH 1.8, 8.4), 7.75 (dd,
p
1Hd JHH 1.5, 8.1), 7.45–7.61 (m, 3Hb,e,f), 7.13 (m, 2Hh,j), 7.02 (dd,
p–p
1Hi, JHH 4.2, 8.7). kmax(CH2Cl2)/nm
(e, ) 360
dm3 molꢁ1cmꢁ1
ing this series, we hope to develop a better understanding of the
interplay between the various types of interactions and which
dominate or cooperate with each other in such structures. Thus,
in this work we report the crystal structures of all four compounds,
explore the different structural motifs present and determine what
affect the different halogens have on the extended structures in
this apparently simple series.
(48,000). m/z (ESI) 266 [M]+. Calcd. for (found%) C16H11FN2O: C,
72.16 (72.33); H, 4.16 (4.36); N, 10.53% (10.81).
2.4. Synthesis of 5-chloro-N-(8-quinolyl)salicylaldimine HqsalCl
HqsalCl was prepared in an analogous manner to HqsalF, orange/
red microcrystals, 78%.
CH), 1624 (
C=N). 1H NMR (CDCl3, 295 K; d; ppm) 13.96 (OH),
8.98 (s, 1Hg), 8.91 (s, 1Ha), 8.19 (dd, 1Hc, JHH 8.1), 7.74 (s, 1Hh),
m mOH), 3042, 2994
max(KBr)/cmꢁ1 3436 (
(m
m
2. Experimental
7.60 (m, 4Hb,d,e,f), 7.04 (s, 1Hi), 7.01 (s, 1Hj). kmax(CH2Cl2)/nm (e,
dm3 molꢁ1cmꢁ1) 359 (65,500). m/z (ESI) 282 [M]+. Calcd. for
(found%) C16H11ClN2O: C, 67.96 (68.22); H, 3.92 (3.87); N, 9.91%
(10.04).
2.1. General remarks
All reactions were conducted in air using reagent grade sol-
vents. All other chemicals were purchased from Sigma–Aldrich
Chemical Company and used as received. Infrared spectra (as KBr
discs) were recorded on a Perkin–Elmer Spectrum One infrared
spectrophotometer in the range 400–4000 cmꢁ1. Electronic spectra
were recorded in CH2Cl2 on a Shimadzu 1800 UV–Visible spec-
trometer. Elemental analyses were carried out on a Eurovector
EA3000 analyser. ESI–MS were carried out on a Bruker Daltonics
7.0T Apex 4 FTICR Mass Spectrometer.
2.5. Synthesis of 5-bromo-N-(8-quinolyl)salicylaldimine HqsalBr
HqsalBr was prepared in an analogous manner to HqsalF, red
microcrystals, 88%.
m mOH), 3049, 2968 (mCH),
max(KBr)/cmꢁ1 3369 (
1617 (
m
C=N). 1H NMR (CDCl3, 295 K; d; ppm) 14.01 (OH), 8.99 (d,
1Ha, JHH 4.2), 8.91 (s, 1Hg), 8.21 (d, 1Hc, JHH 8.4), 7.75 (d, 1Hd JHH
8.1), 7.45–7.59 (m, 5Hb,e,f,h,j), 6.97 (d, 1Hi, JHH 8.7). kmax(CH2Cl2)/
Scheme 1. Synthesis of HqsalX.

J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
441
nm (e
, dm3 molꢁ1cmꢁ1) 350 (57,300). m/z (ESI) 326 [M]+. Calcd. for
(found%) C16H11BrN2O: C, 58.72 (58.60); H, 3.39 (3.56); N, 8.56%
(8.93).
2.6. Synthesis of 5-iodo-N-(8-quinolyl)salicylaldimine HqsalI
HqsalI was prepared in an analogous manner to HqsalF, dark red
microcrystals, 85%.
1616 (
C=N). 1H NMR (CDCl3, 295 K; d; ppm) 14.05 (OH), 8.98 (s,
1Hg), 8.92 (s, 1Ha), 8.89 (s, 1Hc), 8.21 (s, 1Hh), 7.59 (m, 4Hb,d,e,f),
m mOH), 3052, 2966 (mCH),
max(KBr)/cmꢁ1 3436 (
m
7.28 (d, 1Hi, JHH 7.2), 6.88 (d, 1Hj). kmax(CH2Cl2)/nm (e
, dm3 molꢁ1
cmꢁ1) 338 (61,800). m/z (ESI) 374 [M]+. Calcd. for (found%) C16H11
-
IN2O: C, 51.34 (51.46); H, 2.96 (3.15); N, 7.49% (7.69).
Fig. 1. POVRAY drawing of 3 with ellipsoids drawn at 50% probability showing the
intramolecular hydrogen bond.
3. Results and discussion
3.1. Synthesis and basic characterization of Hqsal-X
The compounds 5-X-N-(8-quinolyl)salicylaldimines (HqsalX,
X = F 1, Cl 2, Br 3 and I 4) were prepared in a simple one step reac-
tion involving addition of the appropriate substituted salicylalde-
hyde to 8-aminoquinoline in diisopropyl ether (see Scheme 1).
Previous syntheses have always involved the use of EtOH but we
found the products to be more difficult to crystallize from this sol-
vent whereas orange microcrystalline products are formed from
diisopropyl ether. It should be noted that HqsalCl and HqsalBr have
previously been synthesized but no experimental details or spec-
troscopic studies were reported and are included here in the inter-
ests of completeness [22,23].
Fig. 2. Diagrammatic representation of the angles, h and /.
Table 2
Packing parameters for the HqsalX ligands.
h (°)a
/ (°)b
p–p
The compounds have been characterized by IR, 1H NMR and
UV–Vis spectroscopy with IR spectroscopy showing an imine
stretch between 1613 and 1624 cmꢁ1 while the hydroxyl group
is present at 3369–3436 cmꢁ1 confirming the formation of 1–4.
1H NMR spectra were recorded in CDCl3 and display a characteris-
tic peak at 8.91 or 8.92 ppm for the imino proton while the hydro-
xyl proton is found to be strongly deshielded indicative of strong
hydrogen bonding to the imino nitrogen atom. The aromatic pro-
tons are found in their expected positions and are similar to values
reported for Hqsal [29]. The UV–Vis spectra of the HqsalX ligands
were recorded in dichloromethane. The HqsalX ligands exhibit a
single intense absorption between 338–360 nm, which by compar-
HqsalF
HqsalCl
1
14.7
133.7
22.6
5.1
89.7
133.7
81.3
86.9
3.88
3.77
–
2
3
HqsalBr
HqsalI 4
–
a
h represents the angle between the quinolyl and salicylaldimine moiety.
/ represents the angle between neighbouring HqsalX molecules.
b
ison with Hqsal have been assigned to a
p ?
pꢂ transition [29].
3.2. Structural studies of HqsalX
Single crystals of all the compounds were grown by slow evap-
oration of a solution of HqsalX in diisopropyl ether. Most of the
compounds crystallize in the monoclinic space group P21/c while
HqsalBr crystallizes in P21/n as does the previously reported Hqsal
and HqsalOMe [29,30]. Interestingly, despite this none of the com-
pounds are isostructural though some contain the same structural
motifs.
Fig. 3. POVRAY drawing showing the hydrogen bonding square in 2 (ꢂsymmetry
code = 1ꢁx, ꢁy, 2 ꢁ z).
Fig. 4.
p–p Interactions showing the dimers in the structure of 1.

442
J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
Fig. 5. Rectangular ‘tubes’ in 1 viewed down the c axis.
This is surprising given that studies on the halogen substituted
compounds 4-halo-N-(2-fluorophenyl)benzamide reveal that the
chloro, bromo and iodo substituted compounds are all isostructur-
al [16]. Similarly, (4-halophenyl)-N-(2-cyanophenyl)methanimine
(X = fluoro, chloro, bromo) are also isostructural [18].
A feature of the structures is the presence of a strong hydrogen
bond between the phenolic OAH group and the imine nitrogen
with OAHꢀ ꢀ ꢀN distances varying from 1.76 to 1.84 Å (see Fig. 1).
A similar interaction is also observed in Hqsal although in this case
the distance is much shorter, 1.593 Å [29].
The structures also involve stacks of the HqsalX molecules and
in most cases form a herringbone packing pattern. However, it is
important to note that the interactions used to form these stacks
are not always the same. In quantifying these stacks we have cho-
sen to use two angles h and / which represent the angle between
the quinolyl and salicylaldimine moiety and the angle between
neighbouring HqsalX molecules, respectively (see Fig. 2). Thus, for
HqsalX (X = F, Br and I) the quinoline ring is oriented such that
the ligand is preorganized to coordinate a metal atom (see
Fig. 1). However, the degree of coplanarity between the two aro-
matic rings varies from 5.1° in HqsalI to 22.6° in HqsalBr (see
Table 2).
In contrast with the other structures in this series, in HqsalCl the
quinoline ring points away from the OAHꢀ ꢀ ꢀN hydrogen bond with
the intramolecular angle between the rings 133.7°. A similar obser-
vation has beenmade for Hqsal3-SiPh3,5-Me, although this is thought to
be due to steric effects [31]. For HqsalCl the rotation of the quinoline
ring allows the formation of a hydrogen bonding square between
neighbouring phenolic OAH groups creating a dimer (2.545 Å, see
Fig. 3). Interestingly, the related compound 4-chloro-2-(1-naphthy-
liminomethyl)-phenol where the quinoline ring is replaced by a
naphthyl group reveals a normal conformation, i.e. h < 90 in this case
56.9° [32]. Moreover, the bromo substituted compound 4-bromo-2-
(1-naphthylimino-methyl)phenol [33] has essentially an identical
Fig. 6.
a axis.
p–p Interactions showing the 1D pillar in the structure of 2 viewed down the

J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
443
Fig. 7. View of the wave like planes in 2, viewed down the c axis.
Fig. 8. View of the CAHꢀ ꢀ ꢀN and CAHꢀ ꢀ ꢀO interactions that form the narcissistic dimer in 4 (ꢂsymmetry code = 1 ꢁ x, 1 ꢁ y, ꢁz).
other members of the series with / 133.7° and identical to h as the
molecule sits at the inflection point of the herringbone packing
motif.
Table 3
Parameters for the interactions in the HqsalX ligands.
DAHꢀ ꢀ ꢀA
DAH
(Å)
Dꢀ ꢀ ꢀA
Hꢀ ꢀ ꢀA
\DAHꢀ ꢀ ꢀA
(°)
Symmetry
The structure of HqsalF is unique in this series with inversely
(Å)
(Å)
parallel dimers being formed through weak offset p–p interactions
HqsalF
1
O(1)AH(1)
ꢀ ꢀ ꢀN(1)
0.82
0.93
2.571(2) 1.843
3.358(2) 2.536
147
x, y, z
between the phenyl and quinoline rings. Similar dimers are also
present in the structure of Hqsal although these are held together
by interactions between the CAO and C@N groups [29]. Moreover,
C(9)AH(9)
ꢀ ꢀ ꢀO(1)
147
x, ½ ꢁ y,
½ + z
HqsalCl
2
O(1)AH(1)
ꢀ ꢀ ꢀN(1)
0.94
0.94
0.95
2.589(3) 1.760
2.926(3) 2.545
3.551(3) 2.694
146
105
150
x, y, z
here there are additional
p–p interactions between the phenyl
rings of neighbouring dimers resulting in a 1D stepped chain
(Fig. 4).
O(1)AH(1)
ꢀ ꢀ ꢀO(1)
1 ꢁ x, ꢁy,
2 ꢁ z
C(9)AH(9)
ꢀ ꢀ ꢀO(1)
1 ꢁ x,
1 ꢁ y,
2 ꢁ z
HqsalBr
3
O(1)AH(1)
ꢀ ꢀ ꢀN(1)
0.84
0.95
0.95
2.577(2) 1.832
3.520(3) 2.660
3.287(3) 2.706
147
151
120
x, y, z
C(15)AH(15)
ꢀ ꢀ ꢀN(2)
ꢁx, ꢁy,
1 ꢁ z
C(14)AH(14)
ꢀ ꢀ ꢀO(1)
ꢁx, ꢁy,
1 ꢁ z
HqsalI 4 O(1)AH(1)
ꢀ ꢀ ꢀN(1)
0.84
0.95
0.95
2.567(8) 1.823
3.624(8) 2.732
3.228(9) 2.593
147
157
124
x, y, z
C(15)AH(15)
ꢀ ꢀ ꢀN(2)
1 ꢁ x,
1 ꢁ y, ꢁz
1 ꢁ x,
C(14)AH(14)
ꢀ ꢀ ꢀO(1)
1 ꢁ y, ꢁz
structure to the chloro derivative in stark contrast to our own find-
ings with 2 and 3, indicating that the quinoline influences which
conformation is preferred.
The angle between the stacks is remarkably similar for 1, 3 and
4 varying from 81.3° to 89.7° and does not seem to reflect the size
of the halogen or the angle, h. Once again, HqsalCl differs from the
Fig. 9. View of the offset stacks in 4.

444
J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
A further stepped chain also composed of
p
–
p
stacked dimers is
are then connected to neighbouring columns through the narcissis-
tic interaction described in Fig. 8.
perpendicular to the first chain and connected to it via CAHꢀ ꢀ ꢀO
interactions between the phenolic oxygen atom and a CAH group
on the quinoline ring (see Fig. S1). Combined these interactions
serve to create an extended rectangular ‘tube’ with identical ‘tubes’
isolated from each other (Fig. 5).
The packing in the structure of the two compounds is com-
pleted through CAHꢀ ꢀ ꢀ
ing pattern (see Fig. 10a). Interestingly while HqsalI achieves this
through interaction (C13AH13ꢀ ꢀ ꢀ
single type of CAHꢀ ꢀ ꢀ
p interactions resulting in a fishbone pack-
a
p
The structure of HqsalCl is composed of 1D pillars formed by
C13 = 2.768 Å, see Fig. S3), 3 employs three separate CAHꢀ ꢀ ꢀ
p
p–
p
interactions between the phenyl and quinoline rings but in
interactions (C13AH13ꢀ ꢀ ꢀC3 2.874 Å, C3AH3ꢀ ꢀ ꢀC14 2.814 Å,
C10AH10ꢀ ꢀ ꢀC10 2.875 Å; Fig. 10b) leading to a greater degree of
interconnectivity in this structure.
contrast to Hqsal [29] and HqsalF the aromatic rings are located
directly on top of one another (Fig. 6). The 1D pillar is then con-
nected to another pillar via the hydrogen bonding square described
It is interesting to note that in all the structures despite the
presence of halogens in these systems none of the halogens are
found to be involved either in CAHꢀ ꢀ ꢀX or CAXꢀ ꢀ ꢀX bonding. The
reason for this probably relates to the presence of additional inter-
in Fig. 3. Finally, two types of CAHꢀ ꢀ ꢀ
p
interactions
(C5AH5ꢀ ꢀ ꢀC14 = 2.858 Å, C14AH14ꢀ ꢀ ꢀC5 2.791 Å, Fig. S2) serve to
link the pairs of pillars to one another. Similar pillars are found
in 4-chloro-2-(1-naphthyliminomethyl)phenol, although these
are linked via CAHꢀ ꢀ ꢀO interactions. A perpendicular view of the
pillars reveals wave like planes with the HqsalCl molecule posi-
tioned at the ‘peaks’ and ‘valleys’ of the wave (see Fig. 7).
The structures of HqsalBr and HqsalI while not isostructural do
share many of the same structural motifs and as such will be
discussed together. The most important structural motif is the
presence of narcissistic dimers formed from two sets of comple-
mentary CAHꢀ ꢀ ꢀN and CAHꢀ ꢀ ꢀO interactions between neighbouring
hydrogens on the quinoline ring and the quinoline nitrogen and
phenolic oxygen, respectively (Fig. 8). However, there is a subtle
difference between the two structures such that for HqsalBr the
CAHꢀ ꢀ ꢀN interactions are stronger than the CAHꢀ ꢀ ꢀO interactions
while in HqsalI the situation is reversed (see Table 3). This is due
to the greater degree of non-coplanarity in HqsalBr which places
H14 further away from the phenolic oxygen. However, in HqsalI
the coplanarity of the two aromatic rings means that a close con-
tact between the phenolic oxygen and H14 limits the proximity
of H15 and N2.
actions, in particular
p
–p, CAHꢀ ꢀ ꢀN and CAHꢀ ꢀ ꢀO interactions
which necessarily limit the formation of halogen bonds. Similar
findings are reported in the series 4-halo-N-(fluorophenyl)benz-
amide [16]. This is somewhat surprising particularly in the case
of iodine which is the most polarizable of the halogens and more
commonly involved in halogen bonding [13,18]. It is important
to note that although the halogens do not engage in halogen bond-
ing they are key in determining which structure is the most favour-
able. Thus, fluorine which is the smallest member in the series and
most electronegative gives rise to a structure with extensive
p–p
stacking with the HqsalF inversely parallel. In contrast, the large
steric bulk of iodine prohibits such interactions and strongly
offset stacks are observed with the HqsalI now stacked parallel.
In both compounds the molecules form offset stacks through
interactions between the C@N group and the quinoline ring
(C7ꢀ ꢀ ꢀC9 = 3.175 Å in 3, C7ꢀ ꢀ ꢀC9 = 3.302 Å in 4, see Fig. 9). The
stacks are offset, in part, due to the bulky bromo and iodo substit-
uents which prevent the large degree of overlap found in 1 and 2.
In the case of HqsalI, there are additional weak N(lone pair)–
p
interactions (Cg 1ꢀ ꢀ ꢀN1 = 3.424 Å, where Cg 1 @ C8AC12,C16).
A similar interaction is found in the structure of [Cu2Cl2(N0-[1-
(pyridin-2-yl)ethylidene]acetohydrazide)2] between one of the
coordinated nitrogen atoms and the pyridyl ring (Cgꢀ ꢀ ꢀN = 3.421 Å)
[34]. Once again the non-coplanar nature of the aromatic rings in
HqsalBr precludes the formation of this interaction. These columns
Fig. 11. Overlay of the computed (blue) and molecular structure determined by X-
ray diffraction (grey) of HqsalBr 3. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.).
Fig. 10. (a) Fishbone packing in 4 viewed down the a axis and (b) CAHꢀ ꢀ ꢀ
p interactions in 3 viewed down the b axis, only hydrogens involved in interactions are shown for
clarity.

J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
445
Fig. 12. HOMOs and LUMOs of HqsalBr (right) and HqsalCl (left).
Electronics are clearly important too as evidenced by the structure
of HqsalCl where an additional OAHꢀ ꢀ ꢀO hydrogen bond forms as a
result of the quinoline group being rotated away from the hydroxyl
group.
between the imine nitrogen and hydroxyl group in the case of
HqsalCl and consistent with the stronger hydrogen bonding in this
compound. In contrast, the LUMOs are essentially identical
throughout the series being composed principally of pꢂ orbitals
with slightly more electron density being found on the quinoline.
Moreover, there is no electron density on the halogen in the LUMO
explaining the comparative similarity of the different HqsalX LU-
MOs (see Fig. 12).
3.3. Computational studies
In order to better understand the conformational preferences
for the different halogens and in particular why HqsalCl should
have the shortest OAHꢀ ꢀ ꢀO hydrogen bond we undertook DFT cal-
culations. The structures were all optimized with the B3LYP calcu-
lation using a 6-31Gꢂ basis set. Comparison of the optimized
structures reveals that the atom–atom bond lengths are in agree-
ment within 0.03 Å and bond angles within 5° indicating that the
calculations accurately model the HqsalX compounds. Further
proof of this is evidenced by an overlay of the molecular structure
determined by X-ray diffraction of HqsalBr and its calculated struc-
ture (Fig. 11).
Interestingly, the optimized structures of 1, 3 and 4 show low h
values while 2 displays a large h value such that the quinoline ring
is rotated away from the OAHꢀ ꢀ ꢀN hydrogen bond exactly match-
ing that found in the crystal structure. The inverted conformation
found in 2 must therefore be a consequence of the electronic ef-
fects of the chloro substituent. Thus, the hydrogen bonding square
(see Fig. 3) is a result of the inverted conformation rather than the
driving force for it. Moreover, single point calculations for all
HqsalX ligands both in the inverted and normal conformation re-
veal that in every case the OAHꢀ ꢀ ꢀN hydrogen bond is shorter when
the ligand is in an inverted conformation indicating that the short-
er hydrogen bond in 2 is due to the inverted conformation.
To provide further insight into the effect of the X group on the
structures we also examined the frontier orbitals of the com-
pounds. The compounds are all broadly similar with the salicylaldi-
mine moiety dominating the HOMO. However, the degree of
participation of the halogen in the HOMO varies with F contribut-
ing the least and I the most. Interestingly, there is orbital overlap
4. Conclusions
In conclusion, we have successfully prepared a complete set of
halogen substituted quinolylsalicyaldimines. The packing within
the structures is dominated by
p–p, CAHꢀ ꢀ ꢀN and CAHꢀ ꢀ ꢀO interac-
tions which form stacks although the angle and type of stacks
formed and the individual interactions which hold these stacks to-
gether vary depending on the halogen. The results clearly demon-
strate that both the steric and electronic effects of the halogens
need to be considered when attempting to predict what structure
will be obtained, although computational studies are able to pre-
dict conformational preferences of the individual molecules. More-
over, it is evident that introduction of a halogen into a molecule
does not guarantee the presence of halogen bonding in the result-
ing structure and further emphasizes the challenge of trying to pre-
dict crystal structures based purely on the molecular structure.
Acknowledgements
We thank the National Science and Technology Development
Agency (Grant No.: P-10-11181) for funding this research. We also
thank the Development and Promotion of Science and Technology
Talents Project for a Ph.D. scholarship (to W.P.) and Walailak Uni-
versity for a Postdoctoral Research Fellowship to J.S. (Grant No.:
WU55701). The University of Bristol is thanked for elemental
analysis.

446
J. Sirirak et al. / Journal of Molecular Structure 1036 (2013) 439–446
[15] E. Corradi, S.V. Meille, M.T. Messina, P. Metrangolo, G. Resnati, Angew. Chem.,
Appendix A. Supplementary material
Int. Ed. 39 (2000) 1782–1786.
[16] S.K. Nayak, M.K. Reddy, T.N. Guru Row, D. Chopra, Cryst. Growth Des. 11
(2011) 1578–1596.
[17] D. Chopra, Cryst. Growth Des. 12 (2011) 541–546.
[18] W.H. Ojala, K.M. Lystad, T.L. Deal, J.E. Engebretson, J.M. Spude, B. Balidemaj,
C.R. Ojala, Cryst. Growth Des. 9 (2009) 964–970.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molstruc.2012.
11.028.
[19] A.G. Dikundwar, T.N.G. Row, Cryst. Growth Des. 12 (2012) 1713–1716.
[20] T. Clark, M. Hennemann, J. Murray, P. Politzer, J. Mol. Model. 13 (2007) 291–
296.
References
[21] G. Valerio, G. Raos, S.V. Meille, P. Metrangolo, G. Resnati, J. Phys. Chem. A 104
(2000) 1617–1620.
[1] D. Braga, G.R. Desiraju, J.S. Miller, A.G. Orpen, S.L. Price, CrystEngComm 4
(2002) 500–509.
[22] A.M. Donia, H.A. El-Boraey, Transition Met. Chem. 18 (1993) 315–318.
[23] K. Katuki, K. Isagai, Bunseki Kagaku 22 (1973) 1470–1474.
[24] Bruker. Bruker AXS Inc., Madison, Wisconsin, USA, 2005.
[25] Bruker. Bruker AXS Inc., Madison, Wisconsin, USA, 2003.
[26] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112–122.
[27] L.J. Barbour, J. Supramol. Chem. 1 (2001) 189–191.
[28] J.L. Atwood, L.J. Barbour, Cryst. Growth Des. 3 (2003) 3–8.
[29] J. Wagler, D. Gerlach, G. Roewer, Inorg. Chim. Acta 360 (2007) 1935–1942.
[30] D.J. Harding, D. Sertphon, P. Harding, K.S. Murray, B. Moubaraki, J.D. Cashion,
H. Adams, Chem. Eur. J. (2012), http://dx.doi.org/10.1002/chem.201202053.
[31] E. Kirillov, T. Roisnel, A. Razavi, J.-F. Carpentier, Organometallics 28 (2009)
2401–2409.
[2] K. Biradha, C.-Y. Su, J.J. Vittal, Cryst. Growth Des. 11 (2011) 875–886.
[3] T. Steiner, Angew. Chem., Int. Ed. 41 (2002) 48–76.
[4] G.R. Desiraju, Angew. Chem., Int. Ed. 50 (2011) 52–59.
[5] G.R. Desiraju, Acc. Chem. Res. 35 (2002) 565–573.
[6] M. Khan, V. Enkelmann, G. Brunklaus, Cryst. Growth Des. 9 (2009) 2354–2362.
[7] T.J. Mooibroek, P. Gamez, J. Reedijk, CrystEngComm 10 (2008) 1501–1515.
[8] C. Janiak, Dalton Trans. (2003) 2781–2804.
[9] C.R. Martinez, B.L. Iverson, Chem. Sci. 3 (2012) 2191–2201.
[10] M. Nishio, Y. Umezawa, K. Honda, S. Tsuboyama, H. Suezawa, CrystEngComm
11 (2009) 1757–1788.
[11] S.L. Price, Acc. Chem. Res. 42 (2009) 117–126.
[12] G. Cavallo, P. Metrangolo, T. Pilati, G. Resnati, M. Sansotera, G. Terraneo, Chem.
Soc. Rev. 39 (2010) 3772–3783.
[13] P. Metrangolo, F. Meyer, T. Pilati, G. Resnati, G. Terraneo, Angew. Chem., Int. Ed.
47 (2008) 6114–6127.
[14] J.-A. van den Berg, K.R. Seddon, Cryst. Growth Des. 3 (2003) 643–661.
[32] Z.-X. Li, X.-L. Zhang, Acta Crystallogr. Sect. E 61 (2005) o875–o876.
[33] X.-S. Tai, J. Yin, Y.-M. Feng, Acta Crystallogr. Sect. E 63 (2007) o699–o700.
[34] K. Das, A. Datta, C. Sinha, J.-H. Huang, E. Garribba, C.-S. Hsiao, C.-L. Hsu,
ChemistryOpen 1 (2012) 80–89.