Palladium(II) and platinum(II) complexes of ((2-pyridyl)pyrazol-1-ylmethyl)benzoic acids: Synthesis, Solid state characterisation and biological cytotoxicity

Article abstract of DOI:10.1016/j.ica.2016.02.038

The new ligands 3-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L2) and 5-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzene 1,3-dicarboxylic acid (L3) are reported and the synthesis and characterisation of [PdCl₂(L)] and [PtCl₂(L)] complexes of these and the previously reported 4-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L1) are described. In the solid state, the square planar complexes assemble via hydrogen bonding interactions involving COOH and M-Cl groups as well as by various π-stacking interactions involving the aromatic rings on the ligands and, notably, the chelate rings. Hirshfeld surface analysis has been used to gain insight into the assembly of the molecules. Preliminary studies of the biological cytotoxicity of the [PtCl₂(L)] complexes against A549 and MDA-MB-231 cancer cell lines are reported.

Full text of DOI:10.1016/j.ica.2016.02.038

Accepted Manuscript
Palladium(II) and Platinum(II) Complexes of ((2-pyridyl)pyrazol-1-ylmeth-
yl)benzoic Acids: Synthesis, Solid State Characterisation and Biological Cyto-
toxicity
Aidan P. McKay, Warrick K.C. Lo, Dan Preston, Gregory I. Giles, James D.
Crowley, Jonathan E. Barnsley, Keith C. Gordon, David A. McMorran
PII:
S0020-1693(16)30054-8
http://dx.doi.org/10.1016/j.ica.2016.02.038
ICA 16908
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
11 December 2015
15 February 2016
18 February 2016
Please cite this article as: A.P. McKay, W.K.C. Lo, D. Preston, G.I. Giles, J.D. Crowley, J.E. Barnsley, K.C. Gordon,
D.A. McMorran, Palladium(II) and Platinum(II) Complexes of ((2-pyridyl)pyrazol-1-ylmethyl)benzoic Acids:
Synthesis, Solid State Characterisation and Biological Cytotoxicity, Inorganica Chimica Acta (2016), doi: http://
dx.doi.org/10.1016/j.ica.2016.02.038
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Palladium(II) and Platinum(II) Complexes of
((2-pyridyl)pyrazol-1-ylmethyl)benzoic Acids: Synthesis, Solid State
Characterisation and Biological Cytotoxicity.
Aidan P McKay1, Warrick K C Lo1, Dan Preston1, Gregory I Giles2, James D Crowley1, Jonathan E
Barnsley1, Keith C Gordon1 and David A McMorran1*
1 Department of Chemistry, University of Otago, PO Box 56, Dunedin, New Zealand
2
Department of Pharmacology and Toxicology, University of Otago, PO Box 56, Dunedin, New
Zealand
davidm@chemistry.otago.ac.nz
Abstract
The new ligands 3-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L2) and 5-(3-(2-pyridyl)pyrazol-
1-ylmethyl)benzene 1,3-dicarboxylic acid (L3) are reported and the synthesis and characterisation of
[PdCl2(L)] and [PtCl2(L)] complexes of these and the previously reported 4-(3-(2-pyridyl)pyrazol-1-
ylmethyl)benzoic acid (L1) are described. In the solid state, the square planar complexes assemble
via hydrogen bonding interactions involving COOH and M-Cl groups as well as by various π-
stacking interactions involving the aromatic rings on the ligands and, notably, the chelate rings.
Hirshfeld surface analysis has been used to gain insight into the assembly of the molecules.
Preliminary studies of the biological cytotoxicity of the [PtCl2(L)] complexes against A549 and
MDA-MB-231 cancer cell lines are reported.
1. Introduction
The development of an understanding of the myriad factors that drive how molecules assemble
together into the solid state continues to be an important area of study in chemistry [1-6]. Key to
such an understanding is a cataloging of the different possible interactions for a given set of
molecular functionalities, or synthons, and a determination of the predictability of assembly motifs
for such functionalities. Central to such a determination is the nature of the intermolecular force that
drives the synthon-synthon interaction. Of the various intermolecular forces available, hydrogen
1

bonding has emerged as an important player, due to the strength and directionality of the interaction,
and also the ease with which hydrogen bond donors and/or acceptors can be synthetically
incorporated into molecules of interest [7]. The carboxylic acid (COOH) group, for example, has
been shown to form R2 (8) homodimers [8] in half of all structurally characterised compounds
2
containing only carboxylic acid functionalities [9]. Interactions involving π systems have also
received increasing attention [10]. While these tend to be weaker than hydrogen bonds, they also
have a degree of directionality and, given the widespread use of aromatic rings in the frameworks of
organic and inorganic molecules, π-stacking interactions play a significant role in the solid state
assembly of many reported structures.
As the field of crystal engineering matures, systems that include more than one type of hydrogen
bonding motif have begun to be explored, with the aim of determining a hierarchy of functionalities
in terms of their hydrogen bond donor or acceptor abilities. For example, Vishweshwar et al. have
shown that, in structural studies on pyridine and pyrazine monocarboxylic acids, the OH⋅⋅⋅N
hydrogen bonding drives molecular assembly, rather than R2 (8) dimer formation [11] and Du et al.
2
have reported similar results in their studies on cocrystallisation of dipyridyl species with
dicarboxylic acids [12]. Very recently, Duggirala et al. have shown that charge assisted hydrogen
bonds between phenol groups and chloride anions prevail over COOH⋅⋅⋅Cl hydrogen bonds or R2 (8)
2
dimer formation in cocrystallisation experiments [13]. Such studies allow for the development of a
quantitative ranking of, for example, hydrogen bond donors, where structural data from the
Cambridge Structural Database (CSD) is used in combination with appropriate calculations [14].
We have recently reported the synthesis of 4-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L1),
which combines two nitrogen donors in a potential metal-chelating motif and a distal carboxylic acid
group separated by a flexible methylene hinge [15]. The X-ray structure of L1 showed that the
molecules assembled via OH⋅⋅⋅N hydrogen bonds (between the COOH group of one molecule and
the pyridine nitrogen of the next) into helical chains (consistent with the results of Vishweshwar et
al. and Du et al.) – the helicity arising from the prochirality that the methylene hinge imparts to the
ligand. Upon reaction of silver(I) salts with L1, [Ag(L1)2]+ complexes were obtained, in which the
silver ion adopted a distorted tetrahedral geometry and the L1 ligands wrapped around the metal
such that chiral metallosynthons were generated. Hydrogen bonding interactions, either directly
between the COOH groups of adjacent molecules or between COOH and solvent or counterion
species, assemble these into either one-dimensional helical or meso-helical chains.
2

We were interested in extending these studies to square planar metal-based systems, and also in
modifying the basic structure of the ligand. To this end, we herein report the new ligands 3-(3-(2-
pyridyl)pyrazol-1-ylmethyl)benzoic acid (L2) and 5-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzene 1,3-
dicarboxylic acid (L3), where the position, and number, of the COOH group(s) has been varied
compared to L1 (Scheme 1). Initially we attempted to prepare [Pd(L)2]2+ complexes, expecting that
the palladium(II) ions would prefer a square planar geometry. However, these attempts were
unsuccessful, due, we believe, to the steric clash between the methylene groups on the two ligands
(in a trans arrangement) or the methylene group and a pyridine CH group (in a cis arrangement).
This result is consistent with the literature: while a small number of palladium(II) and/or platinum(II)
complexes containing pyridylpyrazole-type ligands have been reported [16-21], the only reports of
[M(L)2]2+ systems are those where the N2 nitrogen of the pyrazole is unsubstituted [22-29]. We
therefore moved our focus to square planar complexes of the type [PdCl2(L)] and [PtCl2(L)]. These
should not suffer the same steric issues as the [M(L)2]2+ complexes and also, being neutral rather
than cationic, should modify the ways in which molecules will interact with each other in the solid
state compared with the silver(I) complexes, where the role of the counterion was important. While
intermolecular hydrogen bonding between COOH groups might be expected, as has been observed in
related palladium systems [30-32], the potential M-Cl hydrogen bond acceptor might also be
expected to play a role [5, 7, 33].
Synthesis and characterisation of the six [MCl2(L)] (M = Pd(II), Pt(II)) complexes is reported. The
X-ray crystal structures of four of these, along with that of L2, were obtained and show a variety of
packing arrangements, facilitated by a variety of intermolecular hydrogen bonding interactions
(OH⋅⋅⋅N, OH⋅⋅⋅Cl, OH⋅⋅⋅O) as well as by π-π stacking interactions involving the chelate rings, a
recently recognised [34-40] but probably not uncommon type of supramolecular motif. Indeed, it has
been shown that, in certain cases, such interactions can be as strong as hydrogen bonds [38].
Hirshfeld surface analysis provides further insight into the nature of the intermolecular interactions
in the complexes. Finally, given their similarity to the well-known chemotherapeutic agent cisplatin,
the cytotoxicity of the complexes was explored.
3

Scheme 1. The 2-pyridylpyrazole-based ligands discussed in this work.
2. Experimental
2.1. General methods
The ligand 4-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L1) [15] and the ligand precursors 3-
bromomethylbenzoic acid [41], 1,3-dimethyl-5-(bromomethyl)benzene-1,3-dicarboxylate [42] and 2-
(1H-pyrazol-3-yl)pyridine [43] were prepared by published procedures. cis-[PdCl2(CH3CN)2] was
prepared by heating PdCl2 in acetonitrile and cis-[PtCl2(dmso)2] was prepared by adding dmso to an
aqueous solution of K2PtCl4. All other chemicals were purchased commercially and used as received.
1
1H NMR and H-1H COSY spectra were recorded on a 400 MHz Varian spectrometer at 298 K,
referenced to the residual solvent signal. Microanalyses were performed at the Campbell
Microanalytical Laboratory at the University of Otago. Mass spectra were collected on a Bruker
micrOTOF-Q spectrometer. Infrared spectra were recorded on a Bruker ALPHA FT-IR spectrometer
with an ALPHA P ATR measurement module.
2.2. Synthesis of 3-(3-(2-pyridyl)pyrazol-1-ylmethyl)benzoic acid (L2).
2-(1H-Pyrazol-3-yl)pyridine (0.339 g, 2.33 mmol) and 3-bromomethylbenzoic acid (0.500 g, 2.33
mmol) were added to a solution of 40% aqueous NaOH (3.5 mL), benzene (10 mL), and Bu4NOH (4
drops), and the resulting solution was refluxed at 80 °C overnight. After cooling to room
temperature, the colourless organic layer was separated from the yellow aqueous layer. The aqueous
layer was then washed with ethyl acetate (2 × 10 mL) before being acidified to pH 3 using aqueous
HCl (6 M), at which point a yellow precipitate formed. This precipitate was extracted into ethyl
acetate (3 × 60 mL) that was then washed with water (2 × 50 mL) and brine (50 mL). The organic
layer was then dried over magnesium sulfate before having the solvent removed under reduced
o
pressure. Recrystallization from ethyl acetate/petroleum ether (40-60 C) gave L2 as a pale brown
4

solid (0.224 g, 34%). Anal. calcd for C16H13N3O2.0.25H2O C, 67.71; H, 4.80; N, 14.80. Found: C,
67.94; H, 4.67; N, 14.72. 1H NMR (400 MHz, dmf-d7): δ (ppm) = 5.61 (2H, s, Hg), 6.95 (1H, d, J =
2.3 Hz, He), 7.30 (1H, ddd, J = 7.4, 4.8, 1.2 Hz, Hb), 7.54 (1H, t, J = 7.7 Hz, Hh), 7.64 (1H, dt, J =
7.7, 1.5 Hz, Hk), 7.83 (1H, td, J = 7.7, 1.8 Hz, Hc), 8.01 – 7.95 (3H, m, Hd, Hj and Hi), 8.05 (1H, d, J
= 2.3 Hz, Hf), 8.60 (1H, dt, J = 4.9, 1.3 Hz, Ha). HRESI-MS (dmf/MeOH): m/z calcd for
-
C16H12N3O2 : 278.0935 [L2-H]-: found: 278.0930. Selected IR ν / cm-1: 2415,1690, 1598, 1566,
max
1492, 1294, 768.
2.3. 5-(3-(2-Pyridyl)pyrazol-1-ylmethyl)benzene 1,3-dicarboxylic acid (L3)
2-(1H-pyrazol-3-yl)pyridine (0.500 g, 3.45 mmol) and 1,3-dimethyl-5-(bromomethyl)benzene-1,3-
dicarboxylate (0.990 g, 3.45 mmol) were added to a solution of 40 % aqueous NaOH (5 mL),
benzene (15 mL), and Bu4NOH (5 drops), and the resulting solution was refluxed at 80 °C overnight.
After cooling to room temperature, the colourless organic layer was separated from the yellow
aqueous layer. The aqueous layer was then washed with ethyl acetate (2 × 10 mL) before being
acidified to pH 3 using aqueous HCl (6 M), at which point a white precipitate formed. The solvents
were then removed in vacuo and the residue dissolved in ca. 1 mL of dmf. Addition of ethyl acetate
gave a pale yellow solid, which was filtered off, washed with ethyl acetate and diethyl ether and
dried (0.354 g, 32%). Anal. calcd for C17H13N3O4.2H2O: C, 56.83; H, 4.77; N, 11.69. Found: C,
56.78; H, 3.82; N, 11.55. 1H NMR (400 MHz, dmf-d7): δ (ppm) = 5.71 (2H, s, Hg), 6.96 (1H, d, J =
2.3 Hz, He), 7.30 (1H, ddd, J = 7.5, 4.9, 1.2 Hz, Hb), 7.83 (1H, td, J = 7.7, 1.8 Hz, Hc), 7.97 (1H, dt, J
= 7.7, 1.8 Hz, Hd), 8.12 (1H, d, J = 2.3 Hz, Hf), 8.21 (2H, m, Hh), 8.63-8.56 (2H, m, Ha, Hi). HRESI-
-
MS (dmf/MeOH): m/z calcd for C17H12N3O4 : 322.0833 [L3-H]-; found: 322.0826 (100%). Selected
IR νmax / cm-1: 2415, 1703, 1599, 1463, 1278, 766.
2.4. Synthesis of [PdCl2(L1)] (1)
cis-[PdCl2(CH3CN)2] (68 mg, 0.262 mmol) was added to a solution of L1 (76 mg, 0.272 mmol) in
methanol (25 mL). This solution was stirred and heated at 60 °C for 30 min during which time an
orange precipitate formed in the yellow solution. The solution was then cooled to room temperature
and the resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo
to give an orange solid, 1 (97 mg, 79%). Anal. calcd for C16H13N3O2Cl2Pd: C, 42.09; H, 2.87; N,
1
9.20. Found: C, 41.54; H, 2.95; N, 9.13. H NMR (400 MHz, dmf-d7) δ (ppm) = 6.41 (2H, s, Hg),
7.50 (1H, d, J = 2.9 Hz, He), 7.58 (2H, d, J = 8.1 Hz, Hh), 7.73 (1H, m, Hb), 8.04 (2H, d, J = 8.0 Hz,
Hi), 8.32 – 8.34 (2H, m, Hc, Hd), 8.44 (1H, d, J = 2.9 Hz, Hf), 9.15 (1H, d, J = 5.9, Ha). HRESI-MS
5

(m/z) (dmf/MeOH): calcd for [PdCl2(C16H12N3O2)]-: 455.9348, found: 455.9336. Selected IR ν
/
max
cm-1: 2528, 1689, 1606, 1570, 1434, 1290, 766.
2.5. Synthesis of [PtCl2(L1)] (2)
cis-[PtCl2(dmso)2] (114 mg, 0.270 mmol) was added to a solution of L1 (77 mg, 0.274 mmol) in
methanol (20 mL). This solution was stirred and heated at 60 °C for 30 min during which time a pale
yellow precipitate formed in the solution. After 30 min the reaction was cooled to 40 °C and was
stirred overnight at this temperature. The solution was then cooled to room temperature and the
resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo to give
a pale yellow solid, 2 (107 mg, 73%). Anal. calcd for C16H13N3O2Cl2Pt: C, 35.24; H, 2.40; N, 7.71.
1
Found: C, 34.94; H, 2.50; N, 7.45. H NMR (400 MHz, dmf-d7) δ (ppm) = 6.52 (2H, s, Hg), 7.54
(1H, d, J = 3.0 Hz, He), 7.57 (2H, d, J = 8.1 Hz, Hh), 7.75 (1H, dt, J = 1.9, 7.6 Hz, Hb), 8.04 (2H, d, J
= 8.0 Hz, Hi), 8.35 (1H, d, J = 7.8 Hz, Hd), 8.37 (1H, q, J = 7.9 Hz, Hc), 8.50 (1H, d, J = 3.0 Hz, Hf),
9.54 (1H, d, J = 6.0 Hz, Ha). HRESI-MS (m/z) (dmf/MeOH): calcd for [PtCl2(C16H12N3O2)]-:
543.9958, found: 543.9950. Selected IR νmax / cm-1: 2530, 1689, 1609, 1571, 1429, 1291, 766.
2.6. Synthesis of [PdCl2(L2)] (3)
cis-[PdCl2(CH3CN)2] (38 mg, 0.145 mmol) was added to a solution of L2 (40 mg, 0.143 mmol) in
methanol (20 mL). This solution was stirred and heated at 60 °C for 30 min during which time an
orange precipitate formed in the yellow solution. The solution was then cooled to room temperature
and the resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo
to give a an orange solid, 3 (39 mg, 59%). Anal. calcd for C16H13N3O2Cl2Pd: C, 42.09; H, 2.87; N,
1
9.20. Found: C, 41.90; H, 2.75; N, 9.14. H NMR (400 MHz, dmf-d7) δ (ppm) = 6.41 (2H, s, Hg),
7.50 (1H, d, J = 2.9 Hz, He), 7.58 (1H, t, J = 7.7 Hz, Hi), 7.72 (1H, m, Hb), 7.79 (1H, d, J = 7.6 Hz,
Hh, 8.00 (1H, d, J = 7.7 Hz, Hj), 8.10 (1H, s, Hk), 8.31 – 8.33 (2H, m, Hc,Hd), 8.48 (1H, d, J = 2.9 Hz,
Hf), 9.15 (1H, d, J = 5.7 Hz, Ha). HRESI-MS (m/z) (dmf/MeOH): calcd for [PdCl2(C16H12N3O2)]-:
455.9348, found: 455.9340. Selected IR νmax / cm-1: 2530, 1714, 1605, 1440, 1291, 770.
2.7. Synthesis of [PtCl2(L2)] (4)
cis-[PtCl2(dmso)2] (62 mg, 0.147 mmol) was added to a solution of L2 (40 mg, 0.145 mmol) in
methanol (20 mL). This solution was stirred and heated at 60 °C for 30 min during which time a pale
yellow precipitate formed in the solution. After 30 min the reaction was cooled to 40 °C and was
stirred overnight at this temperature. The solution was then cooled to room temperature and the
6

resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo to give
a yellow crystalline solid, 4 (32 mg, 41%). Anal. calcd for C16H13N3O2Cl2Pt: C, 35.24; H, 2.40; N,
1
7.71. Found: C, 35.45; H, 2.39; N, 7.61. H NMR (400 MHz, dmf-d7) δ (ppm) = 6.52 (2H, s, Hg),
7.53 (1H, d, J = 2.9 Hz, He), 7.57 (1H, t, J = 7.7 Hz ,Hi), 7.75 (1H, ddd, J = 7.6, 6.0, 1.9 Hz, Hb),
7.79 (1H, d, J = 7.8 Hz, Hh), 7.99 (1H, d, J = 7.8 Hz, Hj), 8.09 (1H, s, Hk), 8.41 – 8.31 (2H, m, Hc,
Hd), 8.53 (1H, d, J = 3.0 Hz, Hf), 9.53 (1H, d, J = 5.9 Hz, Ha). HRESI-MS (m/z) (dmf/MeOH): calcd
for [PtCl2(C16H12N3O2)]-: 543.9958, found: 543.9981. Selected IR ν
/ cm-1: 2530, 1715, 1610,
max
1442, 1230, 766.
2.8. Synthesis of [PdCl2(L3)] (5)
cis-[PdCl2(CH3CN)2] (40 mg, 0.155 mmol) was added to a solution of L3 (50 mg, 0.155 mmol) in
methanol (25 mL). This solution was stirred and heated at 60 °C for 30 min during which time an
orange precipitate formed in the yellow solution. The solution was then cooled to room temperature
and the resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo
to give an orange solid, 5 (59 mg, 76%). Anal. calcd for C17H13N3O4Cl2Pd.H2O : C, 39.37; H, 2.92;
N, 8.10. Found: C, 39.15; H, 2.84; N, 8.06. 1H NMR (400 MHz, dmf-d7) δ (ppm) = 6.48 (2H, s, Hg),
7.53 (1H, d, J = 2.8 Hz, He), 7.76-7.69 (1H, m, Hb), 8.30 (2H, d, J = 1.6 Hz, Hh), 8.37 – 8.32 (2H, m,
Hc, Hd), 8.56 (1H, d, J = 2.9 Hz, Hf), 8.60 (1H, t, J = 1.6 Hz, Hi), 9.15 (1H, dt, J = 5.9, 1.1 Hz, Ha).
HRESI-MS (dmf/MeOH): m/z calcd for C17H10N3O4Pd-: 425.9719 [5-H]-, found: 425.9730. Selected
IR νmax / cm-1: 2530, 1710, 1605, 1440, 1275, 761.
2.9. Synthesis of [PtCl2(L3)] (6)
cis-[PtCl2(dmso)2] (65 mg, 0.155 mmol) was added to a solution of L3 (50 mg, 0.155 mmol) in
methanol (25 mL). This solution was stirred and heated at 60 °C for 30 min during which time a pale
yellow precipitate formed in the solution. After 30 min the reaction was cooled to 40 °C and was
stirred overnight at this temperature. The solution was then cooled to room temperature and the
resulting precipitate was filtered, washed with methanol and diethyl ether, and dried in vacuo to give
a bright orange solid, 6 (58 mg, 64%). Anal. calcd for C17H13N3O4Cl2Pt: C, 34.65; H, 2.22; N, 7.13.
Found: C,35.23; H, 2.20; N, 7.03. 1H NMR (400 MHz, dmf-d7) δ (ppm) = 6.59 (2H, s, Hg), 7.57 (1H,
d, J = 3.0 Hz, He), 7.75 (1H, ddd, J = 6.8, 5.9, 2.3 Hz, Hb), 8.29 (2H, d, J = 1.6 Hz, Hh), 8.43 – 8.33
(2H, m, Hc, Hd), 8.66-8.56 (2H, m, Hf, Hi), 9.57-9.50 (1H, m, Ha). HRESI-MS (dmf/MeOH): m/z
-
calcd for C17H12N3O4PtCl2 : 587.9843 [6-H]-; found: 587.9872 (100%). Selected IR ν
/ cm-1:
max
2530, 1687, 1627, 1448, 1257, 758.
7

2.10. Cytotoxicity studies
Reagents. 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) and cell culture
reagents were purchased from Life Technologies (Auckland, NZ). All other chemicals were obtained
from Sigma-Aldrich (Auckland, NZ). Cell Culture: All cell lines were purchased from the American
Type Culture Collection (ATCC, Manassas, VA, USA) and maintained in Dulbecco’s Modified
Eagles Medium (DMEM) enriched with 2% fetal bovine serum (FBS) and 1% antibiotic-antimycotic
solution. All cells were cultured at 37 °C in a humidified atmosphere with 5% CO2 levels.
Cytotoxicity evaluation: Cell viability was assessed using the MTT assay.[44] Cell lines tested
against were A549 (lung cancer) and MDA-MB 231 (breast cancer). Two 96 well plates were seeded
with 5,000 cells per well, and the cells were left to adhere for 24 hours prior to treatment. The cells
were then exposed to complexes solubilised in DMSO for 24 hours. To control for the effects of
DMSO, all cell culture medium had a constant DMSO concentration of 0.5% (v/v). Following
compound administration, cells were washed with phosphate buffer solution (PBS) and MTT (0.4
mg mL-1 in DMEM) was added. After MTT incubation (3 hours), the medium was aspirated and the
residual crystals were dissolved in DMSO. Cell number was then calculated at λ = 550 nm. [45]
2.11. X-ray crystallography
X-ray data for L2, 3.dmf, 4, 5.iPrOH and 6.dmf were collected at 100 K on an Agilent Technologies
Supernova system using Cu Kα radiation with exposures over 1.0°. X-ray data were treated using
CrysAlisPro [46] software. X-ray crystal structures were solved using SIR-97 [47] and weighted full-
matrix refinement on F2 was carried out using SHELXL-97 [48] running within the WinGX [49]
package. In each case all non-hydrogen atoms were refined anisotropically and hydrogen atoms were
placed in calculated positions and refined using a riding model. Crystallographic and structure
refinement data are provided in Table 1.
Colourless crystals of L2 were grown by vapour diffusion of diisopropyl ether into an acetone
solution. The structure was solved in the orthorhombic space group Pna21 and refined to an R1 value
of 3.9%. The asymmetric unit contains one molecule of L2. Yellow crystals of 3.dmf were grown by
vapour diffusion of diethyl ether into a dmf solution of 3. The structure was solved in the monoclinic
space group P21/c and refined to an R1 value of 4.2%. One molecule of 3 and one solvent dmf
molecule are present in the asymmetric unit. Yellow crystals of 4 were obtained from the hot
methanol reaction mixture. The structure was solved in the triclinic space group P-1 and refined to
8

an R1 value of 2.3%. One molecule of 4 is present in the asymmetric unit. Yellow crystals of
5.iPrOH were grown by layering isopropyl alcohol (iPrOH) onto a tetrahydrofuran solution of the
complex. The structure was solved in the monoclinic space group P21/c and refined to an R1 value of
5.5%. One molecule of 5 and one solvent iPrOH molecule are present in the asymmetric unit. Pale
yellow crystals of 6.dmf were grown by vapour diffusion of diethyl ether into a dmf solution of 6.
The structure was solved in the triclinic space group P-1 and refined to an R1 value of 8.7%. Two 6
molecules and two dmf solvent molecules are present in the asymmetric unit. The ellipsoids of atoms
Cl3, C6, C13, C14, C15, C26, C28 and C31 became either elongated or flattened when refined
anisotropically. These atoms were modelled with the ISOR command. The carboxylate groups
containing C16, O1 and O2, C17, O3 and O4, and C34, O7 and O8 were disordered. A combination
of ISOR and DFIX commands was used to model these carboxylate groups. The crystal lattice
contained a small amount of diffuse electron density that could not be appropriately modelled. The
SQUEEZE routine within PLATON was employed to resolve this problem, resulting in a void
electron count of 84 that was assigned to two disordered diethyl ether solvent molecules (84 in total).
CCDC numbers: 1441894 (L2), 1441895 (3.dmf), 1441896 (4), 1441897 (5.iPrOH) and
1441893 (6.dmf).
9

Table 1. Crystallographic data and structure refinement data
L2
3.dmf
4
5.iPrOH
6.dmf
chemical formula C16H13N3O2
C19H20Cl2N4O3Pd C16H13Cl2N3O2Pt C20H21Cl2N3O5Pd C20H20Cl2N4O5Pt
formula weight
temperature (K)
wavelength (Å)
crystal system
space group
a (Å)
279.29
529.69
545.28
560.70
662.39
100.0(1)
1.54184
Orthorhombic
Pna21
12.8393(9)
21.6933(13)
4.7745(4)
90
100.0(1)
1.54184
Monoclinic
P21/c
10.0259(13)
29.2584(3)
7.0185(8)
90
99.110(2)
90
2032.8(4)
4
100.0(1)
1.54184
Triclinic
P-1
7.4273(3)
9.9086(3)
11.3726(4)
86.884(3)
86.278(3)
78.323(3)
817.18(5)
2
100.0(1)
1.54184
Monoclinic
P21/c
17.8499(4)
16.1431(4)
7.7594(2)
90
101.104(2)
90
2194.03
4
100.0(1)
1.54184
Triclinic
P-1
7.6293(2)
17.9665(8)
19.1008(9)
68.956(4)
80.542(3)
84.648(3)
2408.67(18)
4
b (Å)
c (Å)
α (o)
β (o)
90
90
γ (o)
volume (Å3)
Z
1329.83(17)
4
Dcalcd (Mg m−3)
µ (mm−1)
Tmin, Tmax
θ range (o)
reflections
collected
data/parameters
Rint
1.395
0.774
0.968, 0.995
4.001-74.025
5695
1.731
10.036
0.529, 0.843
3.021-76.668
15664
2.216
19.201
0.335, 0.520
3.899-76.738
9827
1.697
9.395
0.692, 0.905
3.724-74.715
23218
1.827
13.269
0.214, 0.699
4.226-74.709
36493
2059/191
0.0488
4237/265
0.0543
3407/218
0.0329
4461/284
0.0704
9663/583
0.0917
0.0390 (0.0913) 0.0417 (0.1141) 0.0228 (0.0575) 0.0547 (0.1799) 0.0866 (0.2342)
R1(wR2) [I > 2σ]
R1(wR2) (all data) 0.0500 (0.0965) 0.0437 (0.1195) 0.0237 (0.0582) 0.0653 (0.1864) 0.1112 (0.2628)
3. Results and discussion
The ligands L1-L3 were prepared in moderate yields by a phase transfer alkylation reaction, as
described previously [15, 51, 52] (Scheme 2). For L2, 3-bromomethyl benzoic acid, prepared in turn
from 3-toluic acid [41], was refluxed with 2-pyridyl pyrazole in a mixture of benzene and 40%
NaOH(aq), with a few drops of tetrabutylammonium hydroxide added to act as a phase transfer
catalyst. To prepare L3 5-methylbenzene-1,3-dicarboxylic acid was converted to the dimethyl ester
and then brominated following literature procedures [42]. The resulting 1,3-dimethyl-5-
(bromomethyl)benzene-1,3-dicarboxylate was then submitted to the phase transfer alkylation
reaction with 2-pyridyl pyrazole directly, as the highly basic reaction media not only enabled the
coupling reaction, but also the de-esterification to the desired dicarboxylic acid L3.
10

Scheme 2. Synthesis of ligands and complexes. (i) C6H6, 40% NaOH(aq), Bu4NOH, reflux,
overnight: (ii) [PdCl2(CH3CN)2], MeOH, 30 °C, 30 min: (iii) cis-[PtCl2(dmso)2], MeOH, 60 °C, 30
min, 40°C, overnight.
The compositions of the ligands were confirmed by elemental analysis. High resolution electrospray
ionisation (HRESI) mass spectra displayed peaks corresponding to [L-H]- ions in each case (Figures
1
S3 and S4). H NMR spectra were also consistent with the desired compounds and were similar to
that of L1 [15]. The complexes in turn were prepared in good yield by simply stirring the appropriate
ligand with one equivalent of either [PdCl2(CH3CN)2] or [PtCl2(dmso)2] in hot methanol (Scheme 2).
The complexes are poorly soluble in most solvents, and dmf and dmso were found to be most useful
for characterisation.
The [MCl2(L)] compositions of the complexes were confirmed by elemental analysis. HRESI mass
spectra of dmf solutions of the palladium complexes 1, 3 and 5 contained multiple peaks, including
11

ones corresponding to free ligand (Figures S5, S7, S9). Peaks corresponding to [PdCl2(L-H)]- was
observed for 1, 3 and 5, although for 5 this peak was very small. By contrast, the spectra of the Pt
complexes comprised the [PtCl2(L-H)]- peak almost exclusively (Figures S6, S8, S10). These results
suggest that, in dmf solution under the conditions of the HRESI-MS experiment, the Pt(II)
complexes are much more stable than the Pd(II) ones.
1H NMR spectra of the complexes were recorded in dmf-d7 solution. In each case significant
coordination induced downfield shifts were observed for the signals for the protons on the pyridine
and, to a lesser extent, the protons on the pyrazole rings (Figure 1 and S1, S2). While the signals due
to the protons on the benzoic acid rings show almost no shift at all, consistent with the COOH group
not being involved in the metal ion coordination, a large downfield shift is observed for the
methylene protons (Hg). This is mostly likely due to their proximity to the chloride ions in the square
planar complex. The fact that the signal for the methylene protons appears as a singlet also shows
that, despite the prochiral nature of the ligand, in solution there is free rotation about the methylene
group.
Figure 1. Stacked 1H NMR plot of L1 (a), 1 (b) and 2 (c) in dmf-d7 solution. * = dmf.
3.1. Structural determinations
To explore both how the methylene hinge influences the nature of the complexes, and how the
various supramolecular synthons act to assemble the molecules into the solid, X-ray crystallography

was employed. While the solution studies show that there is free rotation about the methylene hinge
in the complexes, in the solid state this will not be the case and chiral complexes should be formed.
The relative orientations of the two ring systems in the ligands can be defined by two torsion angles
T1 and T2, as described in Scheme 3. For any value of T1 ≠ 0o endo and exo faces of the
pyridylpyrazole ring system can be defined, where the endo face is the one towards which the
benzoic acid ring is oriented.
Scheme 3. Conformational flexibility in the ligands. (a) T1 is defined by the torsion N2-N3-C9-C10,
(b) T2 is defined by the torsion N3-C9-C10-C11.
While L3 could not be obtained in a suitable crystalline form, L2 crystallises in the orthorhombic
space group Pna21, with the asymmetric unit containing one molecule of L2 (Figure 2). The pyridine
and pyrazole rings, which are almost coplanar (the angle between the pyridine ring and pyrazole ring
is 2.2o), adopt the expected transoid disposition, thereby minimising repulsive interactions between
the nitrogen lone pairs. The torsion angles T1 and T2 are 92.7o and -0.1o, respectively with the COOH
group oriented towards the pyridylpyrazole ring system.
Figure 2. X-ray structure of L2. Ellipsoids are drawn at the 50% probability level.

Consideration of the extended structure shows that OH⋅⋅⋅Npy hydrogen bonding interactions join
adjacent molecules of L2. The O⋅⋅⋅N distance is 2.65 Å, the O-H⋅⋅⋅N angle is 171.3o (Table 3) and the
angle between the pyridine and benzoic acid rings is 25.1o. Thus, rather than forming carboxylate
dimers, molecules of L2 assemble enantiospecifically into one-dimensional helical chains, which
propagate along the crystallographic c axis with a pitch of 4.77 Å (corresponding to two molecules)
(Figure 3). The chains are stabilised an offset π(py)⋅⋅⋅π(pz) stacking interaction (centroid⋅⋅⋅centroid
distance is 3.606 Å, centroid displacement angle [53] is 37.7o) between one molecule and a second
one two further along the chain. Similar chains of opposite helicity run parallel and interact with
each other through two weak CH⋅⋅⋅O hydrogen bonds between methylene and aromatic hydrogens
and carboxylic acid C=O groups (O⋅⋅⋅C = 3.156, 3.390 Å, O⋅⋅⋅H-C = 121.53, 140.63o, respectively).
Figure 3. Extended structures of L2 (left) and L1 (right), showing helical chains. Hydrogen atoms
not involved in hydrogen bonding are omitted for clarity.
The observed structure of L2 is similar to that found for L1 [15] in that they both adopt 1-
dimensional helical chains in the solid state, with adjacent molecules assembling via O-H⋅⋅⋅N
hydrogen bonds. However, the more distant placement of the COOH group in L1 leads to a much
less compact helix, with the pitch being 16.0 Å.
Frustratingly, despite numerous attempts, crystals suitable for X-ray analysis of 1 and 2 could not be
obtained. However, suitable crystals of the remaining four complexes could be obtained. 3
14

crystallises in the monoclinic space group P21/c with the asymmetric unit containing one molecule of
3 and one dmf solvent molecule (Figure 4). Selected bond lengths and angles are listed in Table 2
and are unremarkable. The Pd(II) cation is coordinated by the N donors of the pyridine and pyrazole
rings along with the two chloride ions and adopts a square planar geometry with a τ value of 0.06.
4
(τ = 1 for a perfect tetrahedral geometry and 0 for a perfect square planar geometry [54]). The
4
torsion angles T1 and T2 are 175.0o and 114.7o, respectively.
Figure 4. X-ray structure of 3.dmf. Ellipsoids are drawn at the 50% probability level.
The solvent dmf molecule is hydrogen bonded to the COOH group of the ligand (O1⋅⋅⋅O3 2.577 Å,
O1-H1A⋅⋅⋅O3 164.8o). The angle between the planes of the benzoic acid ring and the dmf molecule is
4.07o. However, this is the only strong hydrogen bonding interaction observed in the structure –
rather the molecules assemble primarily due to π-stacking interactions. The pyridylpyrazole ring
systems of molecules of alternating chirality stack along the crystallographic c axis in an exo-endo
fashion, with the angle between adjacent ring system planes being 7.39o (Figure 5). The
centroid⋅⋅⋅centroid distance between the pyridine rings is 3.739 Å (with a centroid displacement
angle of 19.9o), values typical of a relatively strong π(py)⋅⋅⋅π(py) interaction [53]. However, the

distance between the centroids of the two chelate rings is smaller at 3.532 Å (with a centroid
displacement angle is 4.3o), suggesting that, in concert with the π(py)⋅⋅⋅π(py) interaction, there is also
a strong π(chelate)⋅⋅⋅π(chelate) interaction. Such interactions have recently been shown to be quite
widespread in square-planar metal complexes and CSD searches have shown that there are three
common motifs, related to the relative orientations of the two chelate rings (Scheme 4). The parallel
motif has the torsion angle (τ) joining the chelate ring centroids via the metal ions, being ca. 0o,
antiparallel has τ = ca. 180o and the cross motif has τ = ca. 90o [38, 55]. In complexes of the type
discussed here, the antiparallel motif is most common, while the parallel motif is least common. The
value for τ for 3 is 80.9o, so it adopts a cross motif.
Scheme 4. Commonly observed structural motifs for π(chelate)⋅⋅⋅π(chelate) interactions.
Figure 5. (a) View of the π-stacking of the pyridylpyrazole ring systems in 3.dmf (dmf molecules
omitted) and (b) view down the c axis, showing cross motif interactions between adjacent molecules.
These interactions are supplemented by two intermolecular CH⋅⋅⋅Cl hydrogen bonds: one to a CH on
the benzoic acid ring (C⋅⋅⋅O = 3.656 Å, CH⋅⋅⋅O 147.35o) and one to a CH of the methylene bridge
(C⋅⋅⋅O = 3.555 Å, CH⋅⋅⋅O 143.61o).

These stacks are directional, with each of the benzoic acid groups pointing in the same way (Figure
5b). This generates a V-shaped assembly, with the space between the ‘arms’ occupied by a second
stack, held in place by a CH⋅⋅⋅Cl hydrogen bond between one of the chloride ligands and pyridyl ring
CH (C⋅⋅⋅Cl = 3.495 Å, CH⋅⋅⋅Cl 123.27o) and a CH⋅⋅⋅O hydrogen bond between the C=O of a benzoic
acid ring and a pyrazole ring CH (C⋅⋅⋅O = 3.428 Å, CH⋅⋅⋅O 158.96o) (Figure 6a). This nesting of
stacks generates directional sheets along the crystallographic a axis. These sheets of nested stacks
associate via pairs of R2 (7) cyclic hydrogen bonding motifs, involving two CH hydrogens on one
3
benzoic acid ring in one sheet and the benzoic acid-dmf OH⋅⋅⋅O hydrogen bond in the next, to
generate the overall structure (Figure 6b).
Figure 6. Hydrogen bonding interactions in 3.dmf that connect the stacks of molecules into the
three-dimensional structure. Hydrogen atoms not involved in hydrogen bonding are omitted for
clarity.

Table 2. Selected bond lengths (Å), angles (o) and torsions (o)
3.dmf 5.iPrOH
6.dmf a
L2
4
M-Cl1
2.2818(7) 2.2908(8) 2.2793(15) 2.287(3), 2.273(6)
2.2843(8) 2.2946(8) 2.2781(14) 2.295(3), 2.281(4)
M-Cl2
M-N1
2.047(3)
2.049(3)
88.16(2)
2.040(3)
2.047(3)
86.88(3)
2.047(5)
2.069(5)
87.03(5)
2.033(10), 2.045(13)
2.056(10), 2.023(11)
86.67(12), 87.3(2)
79.2(4), 79.2(5)
M-N2
Cl1-M-Cl2
N1-M-N2
Cl1-M-N2
Cl2-M-N1
T1 (N2-N3-C9-C10)
79.51(10) 79.45(11) 79.8(2)
172.58(8) 173.11(8) 172.45(14) 173.0(3), 172.3(4)
178.54(7) 176.32(8) 179.37(14) 178.0(3), 178.8(4)
92.7 175.0
114.7
73.9
45.6
92.4
73.3
-94.6, -76.8
3.2, -29.6
T2 (N3-C9-C10-C11) -0.1
a Metrics for the two molecules in the asymmetric unit
4 crystallises in the triclinic space group P-1 with the asymmetric unit containing a single molecule
of 4 (Figure 7). The platinum(II) ion is coordinated by the N donors of the pyridine and pyrazole
rings along with the two chloride ions (Table 2). The coordination sphere around the Pt is square
planar, with a τ value of 0.06. The torsion angles T1 and T2 are 73.9o and 45.6o, respectively.
4
Figure 7. X-ray structure of 4. Ellipsoids are drawn at the 50% probability level.

Molecules of 4 of alternating chirality assemble via a pair of endo-endo π(chelate)⋅⋅⋅π(py)
interactions between the chelate ring of one molecule and the pyridyl ring of the next with the
intercentroid distance being 3.828 Å and the ring-centroid displacement 22.47o (Figure 8). The motif
could be described as offset antiparallel, with the τ value being 180o (by symmetry) – the proximity
of the benzoic acid rings prevents a truly antiparallel arrangement. Additionally, each molecule also
interacts with a second molecule of the same chirality via concerted exo-exo π(chelate)⋅⋅⋅π(chelate)
(intercentroid distance 3.485 Å, ring centroid displacement 4.49o) and π(pz)⋅⋅⋅π(py) (intercentroid
distance 3.669 Å, ring centroid displacement 20.60o) stacking interactions. These combine to
generate stacks of molecules along the crystallographic a axis. The π-stacking interactions are
complemented by a pair of CH⋅⋅⋅π interactions between CH protons on the pyridine ring on one
molecule and the benzoic acid ring on the other (H⋅⋅⋅centroid 2.656 Å, CH⋅⋅⋅centroid 158.89o) (Figure
8a).
Adjacent stacks of molecules assemble via bifurcated R2 (4) OH⋅⋅⋅Cl hydrogen bonds (O1⋅⋅⋅Cl1 3.24
1
Å, O1⋅⋅⋅Cl2 3.36 Å, O1-H1⋅⋅⋅Cl1 157.3o, O1-H1⋅⋅⋅Cl2 128.9o) and bifurcated R1 (7) CH⋅⋅⋅O hydrogen
2
bond between the C=O of the COOH group and hydrogens on the pyridine and pyrazole rings
(C4⋅⋅⋅O2 3.25 Å, C7⋅⋅⋅O2 3.26 Å, C4-H4A⋅⋅⋅O2 169.1o, C7-H7A⋅⋅⋅O2 153.2o) in concert with R2 (10)
2
CH⋅⋅⋅O hydrogen bonds between the OH of the COOH group and a hydrogen on the benzoic acid
ring (C13⋅⋅⋅O1 3.479 Å, C13-H13⋅⋅⋅O1 166.43o) (Figure 8b). Orthogonal π(acid)⋅⋅⋅π(acid)
interactions (intercentroid distance 3.711 Å, ring centroid displacement 21.80o) generate the three-
dimensional structure.
19

Figure 8. (a) π-stacking interactions and (b) hydrogen bonding interactions that assemble molecules
of 4 in the solid state. Hydrogen atoms not involved in hydrogen bonding are omitted for clarity.
5 crystallises in the monoclinic space group P21/c with the asymmetric unit containing one molecule
of 5 and one isopropanol solvent molecule (Figure 9). The palladium(II) ion is coordinated by the N
donors of the pyridine and pyrazole rings along with the two chloride ions and adopts square planar
geometry (Table 2), with a τ value of 0.08. The torsion angles T1 and T2 are 92.4o and 73.3o,
4
respectively.

Figure 9. X-ray structure of 5.iPrOH . Ellipsoids are drawn at the 50% probability level. Isopropanol
solvent molecule omitted for clarity.
Molecules of 5 of the same chirality assemble via R2 (8) hydrogen bond dimers, generating zig-zag
2
chains along the crystallographic b axis (O1⋅⋅⋅O4 1.81 Å, O3⋅⋅⋅O2 1.78 Å, O1-H1A⋅⋅⋅O4 175.9o, O3-
H3A⋅⋅⋅O2 170.4o) (Figure 10). The angle between adjacent benzoic acid rings is 8.6o. The chains are
directional, with the PdCl2 units all pointing the same way but lying on alternating sides of the plane
generated by the benzoic acid rings. One of the chloride ions forms a hydrogen bond to the OH of
the isopropanol solvate molecule (Cl1⋅⋅⋅O95 3.297 Å, Cl1⋅⋅⋅H95-O95 179.35o) and also a weaker
hydrogen bond to a pyrazole CH on the neighbouring chain (Cl1⋅⋅⋅C7 3.550 Å, Cl1⋅⋅⋅H7-C7 140.77o).
These Cl⋅⋅⋅CH hydrogen bonds assemble the zig-zag chains into sheets which lie parallel to the ac
diagonal. The sheets in turn stack along the crystallographic a axis via two exo-endo
π(chelate)⋅⋅⋅π(py) interactions (centroid⋅⋅⋅centroid distances are 3.438 and 3.442 Å, ring-centroid
displacements are 4.13o and 4.12o) in an unusual offset pseudo-cross motif (τ = 121.5o).

Figure 10. (a) Assembly of molecules of 5 into sheets via R2 (8) hydrogen bond dimers, showing the
2
OH⋅⋅⋅Cl hydrogen bonds to the isopropanol molecules; (b) Assembly of sheets via π(chelate)⋅⋅⋅π(py)
stacking. Hydrogen atoms not involved in hydrogen bonding are omitted for clarity.
6 crystallises in the triclinic space group P-1 with the asymmetric unit containing two molecules of
6, each of which is hydrogen bonded to a dmf molecule via one of its COOH groups (Figure 11). The
structures of the two molecules of 6 are similar. In each case, the platinum (II) ions are coordinated
by the N donors of the pyridine and pyrazole rings and two chloride ions resulting in square planar
geometries (Table 2) - the τ values for each of the complexes are 0.06. By contrast, the torsion
4
angles T1 and T2 for the two molecules are quite different, with the values for the molecule
containing Pt1 being -94.6o and 3.2o, respectively, whereas those for the molecule containing Pt2 are
-76.8o and -29.6o, respectively.
Each molecule participates in two stacking interactions: an exo-endo π(chelate)⋅⋅⋅π(acid) interaction
(intercentroid distance 3.453 Å, ring centroid displacement 0.90o) and an endo-endo π(py)⋅⋅⋅π(acid)

interaction (intercentroid distance 3.453 Å, ring centroid displacement 24.18o). These two π-stacking
interactions generate stacks of molecules propagating along the crystallographic a axis (Figure 12).
Figure 11. X-ray structure of 6.dmf. Ellipsoids are drawn at the 50% probability level.
The molecule containing Pt1 further hydrogen bonds to a dmf via a ca. planar R2 (7) motif with one
2
strong hydrogen bond (O1⋅⋅⋅O10 2.561(15)Å, O1-H1A⋅⋅⋅O10 179.8o) and one weaker hydrogen bond
(O2⋅⋅⋅C40 3.231(15) Å, O2⋅⋅⋅H40-C40 127.0o) with an angle between the COOH and NCHO planes
of 15.6o. The second COOH group forms a weak O⋅⋅⋅HC hydrogen bond to the second dmf molecule
(O4⋅⋅⋅C36 3.604 Å, O4⋅⋅⋅H36B-C36 156.32o). In contrast, the molecule containing Pt2 forms only a
single hydrogen bond to the second dmf (O5⋅⋅⋅O9 2.58(2) Å, O5-H5A⋅⋅⋅O9 179.8o) as the angle
between the COOH and NCHO planes is 86.4o.
The chloride ligands make a number of Cl⋅⋅⋅HC hydrogen bonding interactions. Cl1 hydrogen bonds
to a methylene CH (Cl⋅⋅⋅C 3.761 Å, Cl⋅⋅⋅H-C 149.94o) and a pyrazole CH (Cl⋅⋅⋅C 3.571 Å, Cl⋅⋅⋅H-C

149.08o) within the stack and to a pyridine CH (Cl⋅⋅⋅C 3.675 Å, Cl⋅⋅⋅H-C 176.55o) in an adjacent
stack. Cl2 hydrogen bonds to the CHO hydrogen of one of the dmf solvates (Cl⋅⋅⋅C 3.596 Å, Cl⋅⋅⋅H-C
172.64o) and to a pyridine CH (Cl⋅⋅⋅C 3.527 Å, Cl⋅⋅⋅H-C 129.31o) on an adjacent stack. Cl4 forms
two hydrogen bonds, to a pyridine CH (Cl⋅⋅⋅C 3.683 Å, Cl⋅⋅⋅H-C 145.87o) within the same stack and
to a pyridine CH (Cl⋅⋅⋅C 3.577 Å, Cl⋅⋅⋅H-C 158.07o) in an adjacent stack.
Figure 12. Stacking of molecules of 6 along the crystallographic a axis.
Table 3. Selected hydrogen bond data
D-H⋅⋅⋅A
d(H⋅⋅⋅A) / Å
1.83
d(D⋅⋅⋅A) / Å ∠(DHA) / o
O1-H1A⋅⋅⋅N1#1
O1-H1A⋅⋅⋅O3
O1-H1A⋅⋅⋅Cl1#2
O1-H1A⋅⋅⋅Cl2#1
2.647(3)
2.577(3)
3.237(3)
3.363(3)
2.589(6)
2.626(6)
2.561(15)
2.58(2)
171.2
164.8
157.3
128.9
170.4
175.9
179.8
179.8
L2
3.dmf
4
1.78
2.47
2.49
5.iPrOH O3-H3A⋅⋅⋅O2#3
1.78
O1-H1A-O4#4
1.81
6.dmf
O1-H1A⋅⋅⋅O10
O5-H5A⋅⋅⋅O9
1.71
1.73
Symmetry transformations used to generate equivalent atoms: #1 –x+1, -y+2, z-1/2; #2 x-1, y+1, z;
#3-x+1, y+1/2, -z+1/2; #4 –x+1, y-1/2, -z+1/2.
24

3.2. Hirshfeld Surface Analysis
Hirshfeld surfaces [57, 58] and their associated 2D-fingerprint plots were generated from the cif files
for L2 and complexes 3 – 6 using CrystalExplorer [59]. These provide a way to quantify the
different types of intermolecular contacts between molecules in the crystal and so compare the
importance of each type between the structures. Figure 13 shows the Hirshfeld surface and the 2D
fingerprint plot for L2.
de
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
All
di
(Å)
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Figure 13. Hirshfeld Surface plot (superimposed over the X-ray structure) (left) and fingerprint plot
(all interactions) (right) for L2.
The Hirshfeld surface displays short intermolecular contacts in red, contacts at about the van der
Waals distance in white and regions with no close contacts in blue. Figure 13 shows large red
patches at the pyridine nitrogen and the OH of the carboxylic acid, corresponding to the observed
OH⋅⋅⋅N hydrogen bond. The smaller red regions correspond to a weaker interaction between the
oxygen of the OH group and a CH hydrogen on the pyrazole ring of the neighbouring molecule. The
fingerprint region plots di vs. de, where these are the distances from the Hirshfeld surface to the
nearest nuclei inside and outside the surface, respectively, with the colours changing from blue to
green as the frequency with which a particular (di/de) increases. While the bulk of the contacts relate
to general van der Waals contacts (H⋅⋅⋅H contacts account for 42.7% of all contacts), the di/de values
associated with the green region in the centre of the plot correspond to those expected for π-stacking
interactions (7.7%), while the characteristic pair of spikes at the lower left correspond to the OH⋅⋅⋅N
hydrogen bonding (12.7%).
25

Fingerprint plots for compounds 3 – 6, summarising all the intermolecular interactions present in
these structures, are shown in Figure 14 (and plots for each of the individual interaction types are
presented in Figure S19-S23). In each case the characteristic features corresponding to certain types
of interaction are highlighted, and the percentage of each type of interaction tabulated (Table S3).
The plots are asymmetric for 3, 5 and 6 due to presence of the solvent molecules in the structures,
which were not included in the calculation of the Hirshfeld surface. In each case the interactions
between molecules in the X-ray structures described above are found in the fingerprint plots, but the
quantification of each type of interaction in the different structures allows for comparison between
them. For example, while, in all cases, dispersive H⋅⋅⋅H are the predominant interactions (ca. 30% of
all the interactions in each case), H⋅⋅⋅O and H⋅⋅⋅Cl hydrogen bonding interactions combined make up
ca. 35% of total. Of these, the H⋅⋅⋅O percentage is greater, and the H⋅⋅⋅Cl percentage is smaller, for
the palladium complexes 3 and 5, than for the platinum complexes 4 and 6. The fact that there is a
greater percentage of H⋅⋅⋅O interactions in 4 and 6 might be expected, as these are the complexes
with two COOH groups. The data further suggest that the Cl is a better hydrogen bond acceptor
when coordinated to the palladium, which is consistent with the electron density calculations (Table
S1), which show a larger electron density on the chlorines attached to palladium as compared to
platinum. By contrast, the proportion of the total interactions attributable to π-stacking interactions
(estimated by summing the contributions from the C⋅⋅⋅C, C⋅⋅⋅N, M⋅⋅⋅C and M⋅⋅⋅N interactions) range
from 7.4% to 14.0%, significantly less than the hydrogen bonding. This perhaps surprising result
may be due to the fact that, while the predominant motif for molecular assembly observed in the X-
ray crystal structures is π-stacking (and, in particular, stacking involving the chelate rings), the
interactions between these stacks is largely hydrogen bonding in nature and, across the crystal as a
whole, these outweigh the π-stacking. The data in Figure 14 also reveal the significance of H⋅⋅⋅C
interactions, with these accounting for between 12.0 and 15.3% of the interactions. These
presumably arise from CH⋅⋅⋅π interactions between molecules (which appear as characteristic
‘wings’ on the plots). It is also interesting to note the diffuse nature of the plots for 5 and especially 6
in the top right hand corner, indicating less efficient packing of the molecules in the cases where L3
is present. This presumably is a result of the utilisation of both COOH groups in the intermolecular
hydrogen bonding: calculations using Mercury [60] confirm that the void volume (probe radius = 0.5
Å) in the structure of 6 (13.5%) is more than twice that of the other structures.
26

Figure 14. Fingerprint plots showing the important intermolecular interactions in the crystal packing
of compounds 3 – 6, with the associated percentages of each type of interaction.
3.4. Biological Cytotoxicity
Since the discovery of the therapeutic effects of cisplatin, there has been vigorous exploration of
platinum(II) complexes as anticancer agents [61], with interest also in palladium(II) complexes.[62]
While most attention has focussed on complexes with N2Cl2 donor sets, where the nitrogen donors
are aliphatic or aromatic, and often part of a chelating ligand, systems with extra functionality to
potentially enhance their effectiveness have also been explored [63-65]. Of particular relevance, Sun
et al. have shown that Pd(II) and Pt(II) complexes of 2,2’-bipyridyl-5,5’-dicarboxylic acid and 2,2’-
bipyridyl-4,4’-dicarboxylic acid possess appreciable cytotoxicity in both the neutral [66] and
deprotonated [67] acid forms.
Preliminary studies indicated that the present palladium(II) complexes did not possess the requisite
stability to act as cytotoxic agents (NMR spectra of dmso-d6 solutions containing 5 equivalents
histidine showed decomposition of the complexes). However, the platinum(II) complexes were
27

found to be more stable and were tested for cytotoxic effect against A549 (lung cancer) cells and
cisplatin resistant [68] MDA-MB 231 (breast cancer) cells (against which cisplatin has an IC50 =
9.4 0.3 µM [69] and 41.2 3.9 µM, [70] respectively). Unfortunately, the three platinum(II)
complexes all exhibited low cytotoxicity against both cell lines (IC50 > 60 µM) (Figure 15),
suggesting that in their current form these compounds possess little potential as anticancer agents.
Figure 15. Cell viability ((a) A549, (b) MDA MB 231) plotted against varying concentrations of
platinum(II) compounds. 2 (ꢀ), 4 (ꢀ), 6 (ꢀ). Cells were treated with 0 to 60 µM of compounds for
24 hours, and cell viability was assessed. Cell viability is expressed as mean standard error of
mean, where n = 8.
4. Conclusions
Ligands L1 – L3 readily form square planar complexes of the type [PdCl2(L)] and [PtCl2(L)] that are
usefully soluble only in dmf and dmso. X-ray crystallography shows that while the complexes adopt
quite similar structures in the solid state (except for the rotation of the benzoic acid group with
respect to the pyridylpyrazole ring system – Figure 15) the ways in which the molecules assemble
within the crystal vary considerably. In general, molecules arrange into stacks, facilitated by π-
stacking interactions involving (depending on the compound) all the available aromatic rings, as well
as the chelate ring of the complex. The stacks are then assembled into the overall structure by means
of hydrogen bonding interactions involving the COOH groups (donor and acceptor), the M-Cl
acceptor as well as weaker hydrogen bonding involving CH donors. Hirshfeld surface analysis shows
that these hydrogen bonding interactions account for about 35% of the total interactions within the
structures.
28

Thus, while the initial aim of assembling the molecules primarily by means of the appended COOH
groups and/or the chloride ligands was not realised (with the exception of 5), this may be in part due
to the inclusion of solvent molecules such as dmf (necessitated by the compounds’ poor solubility),
which competed for the hydrogen bond donors. It may also be the case that the π-stacking
interactions, and in particular the π(chelate)-stacking interactions, in the current molecules are
sufficiently strong to compete with the hydrogen bonding in these cases [38]. Studies to better
understand the hierarchies of the intermolecular interactions in these systems are continuing.
Acknowledgments
DP and WKCL thank the University of Otago for PhD scholarships, Lisa Bucke is thanked for
preparation of some of the figures and the University of Otago is acknowledged for financial
support.
Supporting Information
1H NMR spectra, mass spectra, details of the DFT calculations and Hirshfeld surfaces and fingerprint
plots can be found in the Supporting Information.
References
[1] K. Biradha, R. Santra, Chem. Soc. Rev., 42 (2013) 950-967.
[2] G.R. Desiraju, J. Am. Chem. Soc., 135 (2013) 9952-9967.
[3] M.G. Goesten, F. Kapteijn, J. Gascon, CrystEngComm, 15 (2013) 9249-9257.
[4] M.W. Hosseini, in: Molecular tectonics. An approach to crystal engineering, CRC Press,
2012, pp. 231-254.
[5] P. Metrangolo, G. Resnati, Chem. Commun. (Cambridge, U. K.), 49 (2013) 1783-1785.
[6] A. Mukherjee, Cryst. Growth Des., 15 (2015) 3076-3085.
[7] Z.-T.W. Li, Li-Zhu, Hydrogen Bonded Supramolecular Structures, Springer, 2015.
[8] M.C. Etter, J.C. MacDonald, J. Bernstein, Acta Crystallogr B, 46 ( Pt 2) (1990) 256-262.
[9] O. Ivasenko, D.F. Perepichka, Chem. Soc. Rev., 40 (2011) 191-206.
[10] E.R.T. Tiekink, Zukerman-Schpector, J., The Importance of Pi-Interactions in Crystal
Engineering, Wiley, 2012.
[11] P. Vishweshwar, A. Nangia, V.M. Lynch, J. Org. Chem., 67 (2002) 556-565.
[12] M. Du, Z.-H. Zhang, X.-J. Zhao, Cryst. Growth Des., 5 (2005) 1199-1208.
[13] N.K. Duggirala, G.P.F. Wood, A. Fischer, L. Wojtas, M.L. Perry, M.J. Zaworotko, Cryst. Growth
Des., 15 (2015) 4341-4354.
[14] C.B. Aakeroey, K. Epa, S. Forbes, N. Schultheiss, J. Desper, Chem. - Eur. J., 19 (2013) 14998-
15003.
[15] D.A. McMorran, Inorg. Chem., 47 (2008) 592-601.
[16] S.-Y. Chang, J. Kavitha, S.-W. Li, C.-S. Hsu, Y. Chi, Y.-S. Yeh, P.-T. Chou, G.-H. Lee, A.J. Carty, Y.-
T. Tao, C.-H. Chien, Inorg. Chem., 45 (2006) 137-146.
29