Coordination modes of 3-hydroxypicolinic acid: Synthesis and structural characterization of polymeric mercury(II) complexes

Article abstract of DOI:10.1016/j.poly.2006.09.094

Novel mercury(II) compounds of 3-hydroxypicolinic acid (HpicOH; IUPAC name: 3-hydroxy-2-pyridinecarboxylic acid) were synthesized and characterized. HgCl(picOH) (1) and HgBr₂(HpicOH) (2) were obtained as reaction products from the reaction of the corresponding mercury(II) halide with HpicOH, irrespective of the molar ratio of the reactants. From the reaction of HpicOH and mercury(II) acetate, Hg(picOH)₂ (3) was obtained, while mercury(II) nitrate monohydrate gave the 1/1 solvate with water Hg(picOH)₂ · H₂O (3a). Infrared, ¹H and ¹³C NMR spectroscopic data were analyzed for complexes 1, 2 and 3. X-ray crystal structure analysis of 1 and 2 revealed their polymeric nature and different coordination modes of HpicOH. In 1 the deprotonated picolinic acid is N,O-chelating and bridging, while in 2 HpicOH is a O-monodentate weakly bound ligand. Compound 1 consists of HgCl(picOH) moieties with two linear covalent bonds, Hg-N 2.143(4) and Hg-Cl 2.298(1) A?, and four additional Hg?O contacts (2.460(3)-2.904(3) A?) in which both oxygen atoms from the carboxylic group are bridging and involved in coordination to three neighboring mercury atoms, thus forming infinite layers. The coordination of mercury is 2 + 4. 2 consists of {HgBr₂(HpicOH)} moieties, which are linked into chains by means of mercury to bromine secondary long range interactions. The coordination sphere of mercury can be described as irregular 2 + 3 formed by two covalently bonded bromine atoms (Hg-Br 2.277(1) and 2.366(1) A?), two bridging bromine atoms (Hg?Br 3.309(1) and 3.247(1) A?) and by the HpicOH ligand attached to mercury in the zwitterionic form via the carboxylic oxygen atom (Hg?O 2.602(7) A?).

Full text of DOI:10.1016/j.poly.2006.09.094

Polyhedron 26 (2007) 1045–1052
www.elsevier.com/locate/poly
Coordination modes of 3-hydroxypicolinic acid: Synthesis
and structural characterization of polymeric mercury(II) complexes
a,
a,
*
b
*
ˇ
´
´
´
´
Zora Popovic , Dubravka Matkovic-Calogovic , Jasminka Popovic ,
a
c
c
´
´
´
´
Ivan Vickovic , Marijana Vinkovic , Drazˇen Vikic-Topic
a
Laboratory of General and Inorganic Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102 A, HR-10000 Zagreb, Croatia
b
-
Ruder Bosˇkovic´ Institute, Division of Materials Physics, Bijenicˇka cesta 54, HR-10000 Zagreb, Croatia
c
Ruder Bosˇkovic´ Institute, NMR Center, Bijenicˇka cesta 54, HR-10000 Zagreb, Croatia
-
Received 21 June 2006; accepted 27 September 2006
Available online 10 October 2006
Abstract
Novel mercury(II) compounds of 3-hydroxypicolinic acid (HpicOH; IUPAC name: 3-hydroxy-2-pyridinecarboxylic acid) were syn-
thesized and characterized. HgCl(picOH) (1) and HgBr2(HpicOH) (2) were obtained as reaction products from the reaction of the cor-
responding mercury(II) halide with HpicOH, irrespective of the molar ratio of the reactants. From the reaction of HpicOH and
mercury(II) acetate, Hg(picOH)2 (3) was obtained, while mercury(II) nitrate monohydrate gave the 1/1 solvate with water Hg(picO-
1
H)2 Æ H2O (3a). Infrared, H and 13C NMR spectroscopic data were analyzed for complexes 1, 2 and 3. X-ray crystal structure analysis
of 1 and 2 revealed their polymeric nature and different coordination modes of HpicOH. In 1 the deprotonated picolinic acid is N,O-
chelating and bridging, while in 2 HpicOH is a O-monodentate weakly bound ligand. Compound 1 consists of HgCl(picOH) moieties
˚
˚
with two linear covalent bonds, Hg–N 2.143(4) and Hg–Cl 2.298(1) A, and four additional Hgꢀ ꢀ ꢀO contacts (2.460(3)–2.904(3) A) in
which both oxygen atoms from the carboxylic group are bridging and involved in coordination to three neighboring mercury atoms,
thus forming infinite layers. The coordination of mercury is 2 + 4. 2 consists of {HgBr2(HpicOH)} moieties, which are linked into chains
by means of mercury to bromine secondary long range interactions. The coordination sphere of mercury can be described as irregular
˚
2 + 3 formed by two covalently bonded bromine atoms (Hg–Br 2.277(1) and 2.366(1) A), two bridging bromine atoms (Hgꢀ ꢀ ꢀBr 3.309(1)
˚
and 3.247(1) A) and by the HpicOH ligand attached to mercury in the zwitterionic form via the carboxylic oxygen atom (Hgꢀ ꢀ ꢀO
˚
2.602(7) A).
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Mercury(II) compounds; 3-Hydroxy-2-pyridinecarboxylic acid; IR and NMR spectra; X-ray diffraction
1. Introduction
(i) O-monodentate via the deprotonated carboxylic group,
as we found in this work in the structure of HgBr2(Hpi-
2ꢁ
3-Hydroxypicolinic acid (HpicOH) is widely used as a
DNA matrix system in matrix-assisted laser desorption/ion-
ization time-of-flight mass spectrometry (MALDI-
TOFMS) [1]. Because of its physiological activity [2]
and rich coordination modes, many metal complexes have
been investigated. As a ligand, 3-hydroxypicolinic acid,
HpicOH, shows the following coordination modes:
cOH); (ii) non-symmetrical picO2 bridging through the
oxygen of the deprotonated carboxylic and hydroxyl groups
and pyridine nitrogen atom, as found in the binuclear com-
plex {l-picO-[Ru(bipy)2]2}2+ [3]; (iii) N,O-chelating
through the pyridine nitrogen and the carboxylate group,
forming a five-member ring; and (iv) O,O-chelating through
the carboxylate group and deprotonated hydroxyl group,
forming a six-membered ring [4] (Scheme 1). The N,O-che-
lation is the most common mode for picOHꢁ and was
reported for rhenium [5,6], vanadium [7], palladium and
platinum [5], molybdenum and tungsten [8], and zinc,
*
Corresponding authors. Tel.: +385 1 4606340; fax: +385 1 4606341.
´
E-mail addresses: zpopovic@chem.pmf.hr (Z. Popovic), dubravka
ˇ
´
´
@chem.pmf.hr (D. Matkovic-Calogovic).
0277-5387/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2006.09.094

1046
Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
The one- and two-dimensional homo- and heteronuclear
M
M
N
N
N
N
1H and 13C NMR spectra were recorded with a Bruker AV
O
O
O
O
M
1
C
O
C
600 spectrometer, operating at 600.133 MHz for the H
C
O
C
O
nucleus and 150.917 MHz for the 13C nucleus. Samples
were measured from DMSO-d6 solutions at 22 ꢁC (295 K)
in 5 mm NMR tubes. Chemical shifts, in ppm, are referred
O
O
OH
O
OH
M
M
i
ii
iii
iv
1
to TMS as internal standard. FID resolution in H NMR
and 13C NMR spectra was 0.29 and 0.54 Hz per point,
respectively. The following measurement techniques were
used: standard 1H, 13C gated proton decoupling, APT,
COSY, HMQC and HMBC. The 2D NMR spectra were
measured in the pulsed field gradient mode (z-gradient).
X-ray and NMR data are given in Tables 1 and 2,
respectively.
Scheme 1.
manganese and copper [9,10]. In [CdX(picOH)(HpicOH)-
(H2O)]2 (X = Clꢁ, Brꢁ) we established two modes of Hpi-
cOH binding: N,O-chelating and O-monodenate with the
pyridine nitrogen atom protonated [11].
Our former investigations pointed out that the mer-
cury(II) ion is capable of forming various binding coordi-
nation modes with N,O-donor atoms when provided by
suitable ligands [12–16]. Our present interest on mer-
cury(II) coordination complexes has been directed towards
the understanding of the competition between halide ions
and ligands containing N,O-donors for coordination sites
of the mercury(II) ion. So far, we found that the tendency
of mercury to achieve effective coordination by spatial
arrangement of donor atoms of different types of acids,
leads to various, mostly highly irregular, coordination
polyhedra of mercury. Mono- and dicarboxylic acids con-
taining a pyridine ring, or their hydroxy derivatives, are
good ligand systems with N,O-donor atoms at appropriate
sites for modelling various mercury coordination polyhe-
dra in discrete or polymeric complexes of the formulas
HgXL, HgX2L or HgX(L)(LH) (LH and L stand for the
protonated and deprotonated form of the ligand, respec-
tively, X = Clꢁ, Brꢁ, Iꢁ, SCNꢁ).
Our wider interest includes investigation of toxicity
effects of various metal complexes of the 12th group of ele-
ments on microbial communities. For such a study well
characterized complexes are needed. This paper describes
new mercury(II) complexes with 3-hydroxypicolinic acid.
They were synthesized, identified by chemical analysis,
characterized by the appropriate spectroscopic methods,
both in solution and in the solid state, and two crystal struc-
tures were determined by the X-ray diffraction method.
Table 1
General and crystal data, summary of intensity data collection and
structure refinement of compounds 1 and 2
1
2
Formula
Mr
C6H4ClHgNO3
374.14
293(2)
C6H5Br2HgNO3
499.52
296(2)
Temperature (K)
Wavelength (A)
Color
Crystal system
Space group
˚
0.71073
0.71073
colorless
monoclinic
P21/c
colorless
monoclinic
P21/c
Unit cell parameters
˚
a (A)
10.4838(19)
7.9647(12)
9.8583(16)
106.206(18)
10.0030(13)
4.1601(5)
23.720(3)
97.378(10)
˚
b (A)
˚
c (A)
b (ꢁ)
3
˚
V (A )
790.5(2)
4
3.144
672
978.9(2)
4
3.389
888
Z
Dcalc (g cmꢁ3
F(000)
)
Crystal size (mm3)
h Range for data
collection (ꢁ)
0.173 · 0.111 · 0.022 0.412 · 0.111 · 0.070
4.8, 30.0
27,119
2295
4.66, 26.98
12,503
2114
Number of measured
reflections
Number of independent
reflections
Number of observed
reflections, I P 2r(I)
Number of parameters
Absorption correction
2092
2006
110
integration
19.769
123
integration
23.859
2. Experimental
l (mmꢁ1
)
2.1. Materials and physical measurements
Range of transmission
factors
0.027, 0.427
0.050, 0.191
Ra, wRb [I P 2r(I)]
R, wR (all data)
0.0266, 0.0308
0.0621, 0.0639
0.0344, 0.9218
1.124
0.0506, 0.0525
0.1320, 0.1347
0.0768, 5.3103
1.134
All reagents were supplied by Aldrich Chemical Co. and
were used as received without further purification. The
g1, g2 in wc
Goodness-of-fit on F2, Sd
Minimum and maximum
CHN-microanalyses were measured by the Central Analyt-
-
´
ical Service of the Ruder Bosˇkovic Institute, Zagreb. The
ꢁ1.68, 2.61
ꢁ3.185, 1.831
e
˚
electron density (e A)
mercury content in the complexes was determined by com-
plexometric titration with sodium diethyldithiocarbamate
of the solution obtained after decomposition of the speci-
men in aqua regia [17].
Pressed KBr pellets were used for IR measurements
within the range 4000–400 cmꢁ1 with a Perkin–Elmer FT-
IR spectrometer 1600 Series.
P
P
a
b
c
R = jjFoj ꢁ jFcj/ jFoj.
h
i
1=2
P
P
2
2
wR ¼
ðF 2o ꢁ F 2c Þ = wðF 2oÞ
.
w ¼ 1=½r2ðF 2oÞꢂ þ ½g1P þ g2Pꢂ where P ¼ ðF o2 þ 2F 2c Þ=3.
P
2
1=2
d
e
S ¼ ½wðF 2o ꢁ F 2c Þ =ðNobs ꢁ NparamÞꢂ
.
The maximum electron density in the last difference Fourier map is
near Hg.

Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
1047
2.2. HgCl(picOH) (1)
off, washed with cold ethanol and dried. Yield: 0.37 g
(90.5%). Anal. Calc. for C6H5Br2HgNO3 (mol. wt.
499.52): C, 14.43; H, 1.01; N, 2.80; Hg, 40.16. Found: C,
14.13; H, 1.07; N, 3.15; Hg, 40.07%. IR max. (cmꢁ1):
3585m, sharp, 3526m, sharp, 3065m-w, 2856w, br, 1655s,
1613s, 1534s, 1470s, 1412s, 1377m, 1335s, 1283s, 1252m,
1214m, 1152m, 1120m-w, 1048m-w, 874s, 816vs, 802s,
682s, 667vs, 576w, 549w, 446m.
To an aqueous solution of mercury(II) chloride (0.30 g;
1.1 mmol in 15 ml H2O) an aqueous solution of 3-hydrox-
ypicolinic acid was added dropwise (0.16 g; 1.15 mmol in
15 ml H2O). The reaction mixture was allowed to stand
for a few days at room temperature to crystallize. The crys-
talline product was filtered off, washed with cold water and
dried in a desiccator. Yield: 0.31 g (75.6%). Anal. Calc. for
C6H4ClHgNO3 (mol. wt. 374.14): C, 19.26; H, 1.08; N,
3.74; Hg, 53.61. Found: C, 19.38; H, 1.18; N, 3.67; Hg,
53.81%. IR max. (cmꢁ1): 3087w, 1621vs, 1601s, 1570s,
1527m, 1476s, 1456vs, 1375vs, 1312vs, 1249s, 1229s,
1149m, 1064w, 893m, 872m-w, 829s, 805m-s, 692s, 578m.
When the same reaction is performed in ethanol 1 is also
obtained, however the yield is better from aqueous
solution.
The same reaction performed in an aqueous solution
also results in 2, but with traces of unreacted HgBr2 which
is less soluble in water than ethanol.
2.4. Hg(picOH)2 (3) and Hg(picOH)2 Æ H2O (3a)
To an aqueous solution of mercury(II) acetate (0.3 g;
0.94 mmol in 15 ml H2O with a few drops of 2 M acetic
acid) an aqueous solution of 3-hydroxypicolinic acid was
added (0.3 g; 2.15 mmol in 15 ml H2O). The yellowish
product was filtered off, washed with water and dried.
Yield (3): 0.40 g (88.9%). From the reaction of 3-hydrox-
ypicolinic acid (0.24 g; 1.72 mmol in 15 ml H2O) and mer-
cury(II) nitrate monohydrate (0.3 g; 0.86 mmol in 15 ml
H2O with a few drops of 20% nitric acid) the 1/1 solvate
of Hg(picOH)2 with water was obtained. Yield (3a): 0.38
g (87.8%). Anal. Calc. for C12H8HgN2O6 (mol. wt.
2.3. HgBr2(HpicOH) (2)
To an ethanolic solution of mercury(II) bromide (0.3 g;
0.83 mmol in 15 ml of EtOH) an ethanolic solution of
3-hydroxypicolinic acid was added dropwise (0.12 g;
0.86 mmol in 25 ml EtOH). The reaction mixture was left
to stand for a few days. The crystalline product was filtered
Table 2
1H and 13C NMR chemical shifts (d/ppm)a and mercury induced chemical shifts (Dd/ppm),b H–H coupling constants (nJHH/Hz)c and C–H coupling
constants (nJCH/Hz)c of 3-hydroxypicolinic acid (HpicOH) and its mercury compounds
Molecule
Atom
HpicOH
HgCl(picOH)
1
HgBr2(HpicOH)
2
Hg(picOH)2
3
H-2,3
d
Dd
D
14.81 (bs)
15.42 (bs)
0.61
7.65
14.34 (bs)
ꢁ0.47
15.31 (bs)
0.50
7.78
8.46 (d)
0.02
7.88
8.05; 5.10 (dd)
0.18
8.29 (bs)
H-4
7.76
8.55 (d)
7.74
3JHH
Dd
D
8.44 (d)
ꢁ0.11
8.43 (d)
ꢁ0.02
H-5
H-6
7.70
8.61; 5.07 (dd)
7.74
7.76
3JHH
Dd
D
8.19; 5.02 (dd)
0.04
8.26
4.20 (d)
0.21
131.21 (s)
1.79
8.55; 4.93 (dd)
0.06
8.13
4.80 (d)
0.08
129.55 (s)
0.13
8.05
4.94 (d)
3JHH
Dd
d
0.24
131.11 (s)
1.69
160.63 (s)
0.46
C-2
C-3
C-4
129.42 (s)
160.17 (s)
Dd
d
Dd
d
JCH
Dd
D
JCH
Dd
D
160.62 (s)
0.45
159.87 (s)
ꢁ0.30
131.92
168.06 (d)
129.98
169.90 (d)
ꢁ1.94
131.73
168.19 (d)
ꢁ0.19
130.04
167.34 (d)
ꢁ1.88
C-5
C-6
129.63
170.48 (d)
130.01
167.61 (d)
0.38
129.83
171.11 (d)
0.20
130.41
170.25 (d)
0.78
133.56
189.93 (d)
139.94
190.64 (d)
6.38
168.03 (s)
2.46
135.16
190.94 (d)
1.60
165.93 (s)
0.36
140.02
189.94 (d)
6.46
168.09 (s)
2.52
JCH
Dd
d
COOH
165.93 (s)
Dd
a
Recorded in DMSO-d6 solutions. Referred to TMS.
Mercury induced chemical shifts (Dd/ppm) are obtained by subtracting the chemical shifts of the parent compound (HpicOH) from the chemical shifts
b
of the Hg-compounds. Sign (+) denotes deshielding effect, while (ꢁ) denotes shielding effect of mercury.
c
n Denotes the number of bonds between nuclei in spin–spin coupling; (s) singlet, (d) doublet.

1048
Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
˚
476.80): C, 30.23; H, 1.69; N, 5.88; Hg, 42.07. Found: C,
30.37; H, 1.75; N, 5.92; Hg, 42.18%. IR max. (cmꢁ1):
3435m, br, 3100m, br, 2935w, 1647vs, 1597s, 1572m,
1470s, 1455s, 1400s, 1380s, 1314s, 1248m, 1230s, 1147m,
1066w, 991w, 939w, 917w-m, 890s, 821s, 691s, 579m-w,
561w. Anal. Calc. for C12H10HgN2O7 (mol. wt. 494.82):
C, 29.13; H, 2.04; N, 5.66; Hg, 40.54. Found for: C,
29.25; H, 1.93; N, 5.62; Hg, 40.24%. IR max. (cmꢁ1):
3436m, br, 3030w, 2924w, 1636s, 1590m-s, 1470m, 1455s,
1381s, 1312s, 1229m, 1144m-w, 1040w, 1000w, 980vw,
918w, 886m, 819m, 800m, 687m, 670w, 578w, 542w, 448w.
and were refined freely [the bond distance is 0.8(1) A].
The other hydrogen atoms which belong to the pyridine
carbon atoms, as well as the hydroxyl hydrogen atom, were
2
˚
generated at ideal geometrical positions [Csp –H 0.93 A,
˚
Uiso(H) = 1.2 Ueq(C); O–H 0.82 A; Uiso(H) = 1.5 Ueq(O)],
and were refined applying the riding model. The molecular
graphics were created by means of the PLATON98 program
[21].
3. Results and discussion
3.1. Preparation
2.5. X-ray crystallographic structure determination of 1 and
2
Compound 1 HgCl(picOH) was obtained from the reac-
tion of 3-hydroxypicolinic acid (HpicOH) with mercury(II)
chloride in an aqueous solution. Similarly, reaction with
HgBr2 in ethanolic solution gave HgBr2(HpicOH) (2). In
both cases the reaction products were not influenced by
the molar ratio of reactants (1:1 or 2:1). Both mercury(II)
acetate and nitrate monohydrate react with HpicOH in
aqueous media giving typical metal carboxylates i.e.
Hg(picOH)2 (3) and Hg(picOH)2 Æ H2O (3a), respectively.
All the isolated compounds are soluble in solvents with
pronounced coordination properties such as dimethyl sulf-
oxide (DMSO), N,N-dimethyl formamide (DMF) or pyri-
dine. Crystals suitable for X-ray diffraction studies were
obtained from the mother liquor for compounds 1 and 2,
but all our attempts to prepare crystals of compounds 3
and 3a failed.
The general and crystal data, and summary of intensity
data collection and structure refinement for compounds 1
and 2 are given in Table 1.
The data collection for both structures was carried out
on an Oxford Diffraction Xcalibur four-circle kappa geom-
etry single-crystal diffractometer with Sapphire 3 CCD
detector, by applying the CrysAlis Software system [18].
A crystal-detector distance of 60 mm and x scan type
was used for both structures. Data reduction, including
absorption correction, was done by the CrysAlis RED
application of the CrysAlis Software system [18]. Coordi-
nates of the non-hydrogen atoms were determined by direct
methods using the SHELXS program [19]. The coordinates
and anisotropic thermal displacement parameters for all
non-hydrogen atoms were refined by the least-squares
methods based on F2 using the SHELXL97 program [20].
The coordinates of the hydrogen atom belonging to the
nitrogen atom in 2 were found by examination of the elec-
tron-density Fourier map at the final stages of refinement
3.2. Structural description of 1 and 2
In 1 the shortest bonds to Hg are from nitrogen and chlo-
˚
rine atoms (Hg–N 2.143(4) and Hg–Cl 2.298(1) A) giving a
Fig. 1. Drawing of HgCl(picOH) (1) with the atom labelling scheme. The thermal ellipsoids are at the 50% probability level. Dashed lines represent
Hgꢀ ꢀ ꢀO contacts. Symmetry operation: a = x, 0.5 ꢁ y, z ꢁ 0.5.

Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
1049
Table 3
Selected interatomic distances (A) and angles (ꢁ) for 1
Table 4
˚
˚
Selected interatomic distances (A) and angles (ꢁ) for 2
Interatomic distances
Bond angles
Bond distances
Bond angles
Hg–N
Hg–Cl
2.143(4)
N–Hg–Cl
172.68(10)
73.47(12)
109.95(8)
85.13(12)
101.40(8)
83.38(10)
Hg–Br1
Hg–Br2
Hgꢀ ꢀ ꢀO1
Hgꢀ ꢀ ꢀBr1i
Hgꢀ ꢀ ꢀBr2ii
Hgꢀ ꢀ ꢀBr2iii
N–C5
N–C1
O1–C6
O2–C6
O3–C2
2.2766(10)
2.3660(10)
2.602(7)
Br1–Hg–Br2
Br1–Hg–O1
Br2–Hg–O1
168.22(4)
87.99(18)
103.63(18)
2.2975(13)
2.460(3)
2.609(3)
2.888(3)
2.904(3)
1.345(5)
1.331(6)
1.248(5)
1.266(5)
1.339(6)
N–Hg–O1
Cl–Hg–O1
N–Hg–O2i
Cl–Hg–O2i
O1–Hg–O2i
Hgꢀ ꢀ ꢀO1
Hgꢀ ꢀ ꢀO2i
Hgꢀ ꢀ ꢀO2ii
Hgꢀ ꢀ ꢀO1iii
N–C1
3.3093(10)
3.2469(10)
3.8475(10)
1.324(12)
1.492(13)
1.405(12)
1.211(12)
1.481(13)
N–C5
O1–C6
O2–C6
O3–C2
Symmetry transformations used to generate equivalent atoms: (i) x,
0.5 ꢁ y, z ꢁ 0.5; (ii) 1 ꢁ x, 1 ꢁ y, 2 ꢁ z; (iii) 1 ꢁ x, 1/2 + y, 3/2 ꢁ z.
Symmetry transformations used to generate equivalent atoms: (i) x, y ꢁ 1,
z; (ii) x, 1 + y, z; (iii) ꢁx, 1/2 + y, ꢁ1/2 ꢁ z.
˚
˚
slightly distorted characteristic linear coordination, N–Hg–
Cl 172.7(1)ꢁ [22,23]. These bonds are slightly longer than the
sum of the covalent radii of the corresponding atoms
because of the additional Hgꢀ ꢀ ꢀO contacts. Both carboxyl-
ate oxygen atoms of the deprotonated picolinic acid are
involved in coordination to two neighboring mercury atoms
(Hgꢀ ꢀ ꢀO1 2.460(3) and Hgꢀ ꢀ ꢀO2 (x, 0.5 ꢁ y, z ꢁ 0.5)
1.65 A; rVDW(O) = 1.40–1.52 A) [22–25]. All contacts fulfill
the effective coordination sphere of mercury, which is there-
fore 2 + 4. This behavior is in agreement with Grdenic’s
´
classification of a characteristic and effective coordination
sphere of mercury in crystal structures [22,23]. Through
these contacts (Fig. 2), infinite two-dimensional layers are
formed. There is an intramolecular hydrogen bond connect-
ing the hydroxyl and carboxylic group, O3–Hꢀ ꢀ ꢀO2 of
˚
2.609(3) A) thus bridging them and forming infinite zig-
˚
zag chains along the b axis, Fig. 1. Relevant bond distances
and bond angles are shown in Tables 3 and 4. Besides the
two already mentioned, there are two more long Hgꢀ ꢀ ꢀO
2.542(5) A. The intermolecular hydrogen bonds are very
weak (Table 5). A similar coordination, yet with slightly
longer covalent Hg–N and Hg–Cl bonds than in 1, is found
in the structure of (pyda Æ H)2[Hg(pydc)Cl2 Æ 2H2O] (pyda =
2,6-pyridinediamine;
acid). In this structure the distances amount to 2.198(6)
˚
contacts (Hgꢀ ꢀ ꢀO2 (1 ꢁ x, 1 ꢁ y, 2 ꢁ z) 2.888(3) A and
˚
Hgꢀ ꢀ ꢀO1 (1 ꢁ x, 0.5 + y, 1.5 ꢁ z) 2.904(3) A), still shorter
pydc = 2,6-pyridinedicarboxylic
than the sum of van der Waals radii (rVDW(Hg) = 1.55–
Fig. 2. Packing of molecules of HgCl(picOH) (1) in the unit cell, including one polymeric layer. The deformed octahedral geometry around Hg is shown
by double lines.

1050
Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
Table 5
structural type of HgBr2 (Hg–Br of 2.444(7) and
˚
Hydrogen bond geometry (distances (A), angles (ꢁ)) in 1 and 2
˚
2.445(7) A, 2 · Hgꢀ ꢀ ꢀBr of 3.240(5) and 2 · Hgꢀ ꢀ ꢀBr
˚
D–Hꢀ ꢀ ꢀA
Dꢀ ꢀ ꢀA
D–Hꢀ ꢀ ꢀA
3.243(5) A) [27]. The two bridging mercury to bromine
contacts as well as the mercury to oxygen contact are
longer than the covalent bond, but shorter than the van
der Waals interactions (rVDW(Br) = 1.85–1.95 A) [22–25].
1
O3–Hꢀ ꢀ ꢀO2
C4–Hꢀ ꢀ ꢀO3i
C5–Hꢀ ꢀ ꢀClii
2.542(5)
3.443(7)
3.414(5)
143
165
131
˚
The coordination polyhedron is best described as very
deformed square-pyramidal with an O atom at the apex.
The Hg to Br distance at the sixth coordination site of
2
O3–Hꢀ ꢀ ꢀO1
N–Hꢀ ꢀ ꢀO2
N–Hꢀ ꢀ ꢀO2iii
C3–Hꢀ ꢀ ꢀBr2iv
2.334(11)
2.662(10)
2.737(12)
3.495(12)
138
116(13)
133(14)
140
˚
3.8475(11) A is too long to be considered as a contact.
Analysis of all structures deposited with the Cambridge
Crystallographic Data Centre [28] that contain a mercury
to bromine bond, showed that the short Hg–Br1 bond in
2 is similar to the shortest one found so far in j-(d8-
ET)4[HgBr2 Æ Hg2Br6] (d8-ET = wholly deuterated bis(eth-
Symmetry transformations used to generate equivalent atoms: (i) ꢁx, 1/
2 + y, 3/2 ꢁ z; (ii) 1 ꢁ x, 1 ꢁ y, 1 ꢁ z; (iii) ꢁ1 ꢁ x, ꢁy, ꢁz; (iv) ꢁx, 1 ꢁ y,
ꢁz.
˚
ylenediothio)tetrathiafulvalene) where Hg–Br is 2.28(2) A
˚
and 2.310(2) A, respectively [26]. The greater elongation is
and Br–Hg–Br is 169.7(7)ꢁ. There are two additional con-
˚
caused by four Hgꢀ ꢀ ꢀO contacts that are in the range
tacts of 3.41(2) A at one side of this linear bond, yet the
˚
˚
2.488(5)–2.659(5) A.
Hg–Br distance of 3.77(2) A at the other side points to
Structure of 2 consists of polymeric chains (Fig. 3).
an absence of any additional interactions to Hg [29].
A similar polymeric structural motif to that of 2 was
found in the structure of HgCl2(tolnapH), (tolnapH =
N-p-tolyl-2-oxo-1-napthylideneamine) [30]. In the structure
of the polymeric HgBr(nic) (nicꢁ = nicotinate anion, pyri-
dine-3-carboxylate) [15] the effective Hg coordination is a
3 + 2 irregular polyhedron with Hg–Br bond distances of
There are two covalent bonds, Hg–Br1 2.277(1) and
˚
Hg–Br2 2.366(1) A forming the linear characteristic coordi-
nation, Br1–Hg–Br2 168.22(4)ꢁ (Table 4). The effective
coordination sphere of mercury can be described as irregu-
lar 2 + 3, consisting of three weak additional contacts, two
with bromine atoms from adjacent molecules (Hg–Br1 (x,
˚
˚
1 + y, z) 3.309(1) and Hg–Br2 (x, ꢁ1 + y, z) 3.247(1) A),
2.6528(9) and 2.6468(9) A, which are longer than in 2
and one with the carboxylic oxygen of HpicOH in the zwit-
and slightly longer than the tetrahedral Hg–Br covalent
˚
˚
terionic form (Hg–O1 2.602(7) A). Interestingly, both cova-
bond of 2.62 A [22,23,25].
lent Hg–Br bonds are shorter than the sum of the covalent
radii for Hg(digonal) and Br atoms (1.30 and 1.14 A,
respectively) [22–25] and also shorter than in the brucite
As in 1, the O3 hydroxyl atom forms an intramolecular
˚
˚
hydrogen bond O3–Hꢀ ꢀ ꢀO1 of 2.334(11) A, yet there is
another one involving the protonated nitrogen (N–Hꢀ ꢀ ꢀO2
Fig. 3. Drawing of HgBr2(HpicOH) (2) with the atom labelling scheme. The thermal ellipsoids are at the 50% probability level. Dashed lines represent
Hgꢀ ꢀ ꢀBr and Hgꢀ ꢀ ꢀO contacts. Symmetry operations: a = x, y ꢁ 1, z; b = x, 1 + y, z.

Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
1051
˚
2.662(10) A). Two polymeric chains are mutually con-
nected by centrosymmetric pairs via intermolecular hydro-
H6
C6
H5
H4
˚
gen bonds N–Hꢀ ꢀ ꢀO2 (ꢁ1 ꢁ x, ꢁy, ꢁz) of 2.737(12) A
C5
C4
N
(Fig. 4, Table 5). In such a manner three-center hydrogen
bonds are formed and the N–H group is distributed
between two proton acceptors.
C2
OH
C3
C
O
OH
3.3. Spectral analysis
Scheme 2.
The infrared spectra of the complexes proved that 3-
hydroxypicolinic acid is bound to the mercury(II) ion in
the deprotonated form in all the isolated compounds (see
ma(COO) in the experimental data), however different
modes of ligand coordination in 1 and 2 were unambi-
gously confirmed by X-ray structure analysis. Unfortu-
nately, due to overlaping of maxima and vibrational
coupling, the observed shifts of absorption bands sensitive
to metal coordination relatively to those of the free ligands
could not be unambigously assigned.
tra of 1, 2 and 3 in DMSO solution revealed narrow signals
and a lack of long range couplings, which is characteristic
of the monomeric form of the molecules. This is not unex-
pected since in solution the polymers almost certainly
break up into monomeric units. On the other hand, the
NMR data showed that the ligand is bound as in the solid
state with a deprotonated N and O in 1 and 3, while a neu-
1
tral O in 2. The obtained H spectra of 1, 2 and 3 display
four groups of signals. The assignments of the protons were
1
1H and 13C NMR data of 3-hydroxypicolinic acid and
its mercury(II) compounds are given in Table 2 while
Scheme 2 shows the labelling of the atoms. The NMR spec-
supported by COSY spectra. In the H NMR spectra, the
greatest shifts (Dd) after complexation are observed for
OH-protons which are five or six bonds remote from mer-
cury. Interestingly, the OH-proton of compounds 1 and 3 is
found to be deshielded, in comparison to the one in 2 which
is a shielded proton. Such behavior arises from the different
mode of mercury coordination. In all 1H spectra the multi-
plicity of the signals is alike. The mercury binding does not
influence significantly the other protons in the molecules of
both compounds 1 and 2. A small shielding effect is
observed on the H-4 atom in complexes 1 and 2, however
deshielding effects are observed on H-5 and H-6 atoms in
all three complexes.
The assignments of all the carbon atoms are confirmed
by HMQC and HMBC measurements. In the 13C NMR
spectra, the greatest change upon complexation with mer-
cury was observed for the chemical shifts of the C-6 atom.
In 1 and 3, C-6 is more deshielded than in 2, since in com-
plexes 1 and 3 the mercury atom is bonded to the nitrogen
atom and the carboxyl group, while in 2 only to the car-
boxyl group. In all three compounds, carboxyl carbon
atom was found to be not so deshielded as the C-6 atom.
Shielding effects are observed on C-3 and C-4 atoms
(ꢁ0.30 and ꢁ0.19 ppm, respectively) in complex 2, and
on the C-4 atom in 1 and 3 (ꢁ1.94 and ꢁ1.88 ppm, respec-
tively). The carbon atom C-2 is deshielded in all three com-
plexes (0.13–1.79 ppm) as expected, since it is close to the
mercury binding site. The mercury binding induces elec-
tronic redistribution throughout the molecule which makes
all the carbon atoms sensitive to mercury binding even if
they are distant from mercury.
4. Conclusion
The competition reaction of halide (X = Clꢁ, Brꢁ) with
the 3-hydroxypyridine-2-carboxylate ligand (picOHꢁ)
toward the mercury(II) ion was investigated. It was estab-
lished that the molar ratio of the reactants did not influence
Fig. 4. Packing of HgBr2(HpicOH) (2) molecules into polymeric layers.
The deformed square pyramidal geometry around Hg is shown by double
lines. Hydrogen bonds are shown by dashed lines.

1052
Z. Popovic´ et al. / Polyhedron 26 (2007) 1045–1052
´
[8] S.M.O. Quintal, H.I.S. Nogueira, H.M. Carapuc¸a, V. Felix, M.G.B.
the type of reaction product, and polymeric {HgCl(pi-
cOH)}n or {HgBr2(HpicOH)}n was always obtained.
X-ray crystal structural analysis of these polymeric com-
pounds showed interesting 2 + 4 and 2 + 3 coordination
polyhedra around the mercury atom, and confirmed two
different coordination modes of HpicOH. The NMR spec-
tra of all the isolated mercury compounds in DMSO solu-
tions revealed that they exist in the monomeric form.
Drew, J. Chem. Soc., Dalton Trans. (2001) 3196.
[9] C. Sun, X. Zheng, L. Jin, J. Mol. Struct. 646 (2003) 201.
[10] V.B. Di Marco, A. Tapparo, A. Dolmella, G.G. Bombi, Inorg. Chim.
Acta 357 (2004) 135.
´
´
´
´
´
´
[11] Z. Popovic, G. Pavlovic, M. Vinkovic, D. Vikic-Topic, M. Rajic
´
Linaric, Polyhedron 25 (2006) 2353.
ˇ
ˇ
[12] J. Popovic´, D. Matkovic´-Calogovic´, G. Pavlovic´, Z. Soldin, Z.
´
Popovic, in: Eleventh Slovenian–Croatian Crystallographic Meeting,
Bohinj, Slovenia, June 27–30, 2002, Book of Abstracts, 2002, 35.
ˇ
ˇ
´
´
´
[13] D. Matkovic-Calogovic, I. Picek, Z. Popovic, Z. Soldin, in: Tenth
Croatian–Slovenian Crystallographic Meeting, Lovran, Croatia, June
21–24, 2001, Book of Abstracts, 2001, p. 16.
Acknowledgements
ˇ
ˇ
´
´
´
´
[14] D. Matkovic-Calogovic, G. Pavlovic, Z. Popovic, Z. Soldin, in: 20th
European Crystallographic Meeting, ECM 20, Crystallography in
Natural Sciences and Technology, Krakow, Poland, August 25–31,
2001, Book of Abstracts, S6.M7.O4. pp. 116.
The authors would like to acknowledge the Ministry of
Science, Education and Sports of the Republic of Croatia
for financial support (Grants Nos. 0119633, 0119632,
0098067 and 0098059).
ˇ
ˇ
´
´
´
´
[15] Z. Popovic, G. Pavlovic, D. Matkovic-Calogovic, Z. Soldin, Acta
Crystallogr. C59 (2003) m165.
ˇ
´
´
[16] G. Pavlovic, Z. Soldin, Z. Popovic, in: Fourteenth Croatian–
Slovenian Crystallographic Meeting, Vrsar, Croatia, June 21–24,
2005, Book of Abstracts, 2005, p. 45.
Appendix A. Supplementary material
CCDC 602643 and 602642 contain the supplementary
crystallographic data for 1 and 2. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html, or from the Cambridge Crystallo-
graphic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223-336-033; or e-mail: depos-
it@ccdc.cam.ac.uk. Supplementary data associated with
this article can be found, in the online version, at
doi:10.1016/j.poly.2006.09.094.
[17] P. Gouverner, W. Hoedeman, Anal. Chim. Acta 30 (1964) 519.
[18] Oxford diffraction, 2004, Oxford Diffraction Ltd., Xcalibur CCD
System, CrysAlis Software System, Version 171.23.
[19] G.M. Sheldrick, Acta Crystallogr. Sect. A 46 (1990) 467.
[20] G.M. Sheldrick, SHELXL97, Program for the refinement of crystal
structures. University of Go¨ttingen, Germany, 1997.
[21] A.L. Spek, Acta Crystallogr. Sect. A 46 (1990) 34.
´
[22] D. Grdenic, Quart. Rev. 19 (1965) 303.
´
[23] D. Grdenic, Connections in the Crystal Structures of Mercury
Compounds, in: G. Dodson, J.P. Glusker, D. Sayre (Eds.), Structural
Studies of Molecules of Biological Interest, Clarendon Press, Oxford,
1981, p. 207.
References
ˇ
´
´
[24] D. Matkovic-Calogovic, Ph.D. Thesis (in Croatian; Abstract in
English) University of Zagreb, 1994.
[1] Y. Kim, G.B. Hurst, M.J. Doktycz, M.V. Buchanan, Anal. Chem. 73
(2001) 2617.
[25] L. Pauling, The Nature of the Chemical Bond, Third ed., Cornell
University Press, Ithaca, NY, 1960.
´
[2] E. Kiss, K. Petrohan, D. Sanna, E. Garribba, G. Micera, T. Kiss,
[26] A. Moghimi, A. Shokrollahi, M. Shamsipur, H. Aghabozorg, M.
Ranjbar, J. Mol. Struct. 701 (2004) 49.
Polyhedron 19 (2000) 55.
[3] V.L. Constantino, L.F.C. de Oliviera, P.S. Santos, H.E. Toma, Trans.
Met. Chem. 19 (1994) 103.
[27] V.I. Pakhomov, A.V. Goryunov, I.N. Ivanova-Korfini, A.A. Bogu-
slavskii, R.Sh. Lotfullin, Zh. Neorg. Khim. 35 (1990) 2476.
[28] (a) Cambridge Structural Database V5.27, Cambridge Crystallo-
graphic Data Centre, 12 Union Road, Cambridge, England, Novem-
ber, 2002;
´
[4] P.C.R. Soares-Santos, H.I.S. Nogueira, V. Felix, M.G.B. Drew, R.A.
Sa´ Ferreira, L.D. Carlos, T. Trindade, Chem. Mater. 15 (2003) 100.
[5] S.M.O. Quintal, H.I.S. Nogueira, V. Felix, M.G.B. Drew, New J.
Chem. 24 (2000) 511.
[6] S. Gatto, T.I.A. Gerber, G. Bandoli, J. Perils, J.G.H. du Preez, Inorg.
Chim. Acta 269 (1998) 235.
(b) F.H. Allen, Acta Crystallogr., Sect. B 58 (2002) 380.
[29] O.A. Dyachenko, V.V. Gritsenko, G.V. Shilov, R.N. Lyubovskaya,
R.B. Lyubovskii, Synth. Met. (1994) 193.
ˇ
[7] A. Shaver, J.B. Ng, D.A. Hall, B.S. Lum, B.I. Posner, Inorg. Chem.
32 (1993) 3109.
´
´
´
´
´
[30] Z. Popovic, V. Roje, G. Pavlovic, D. Matkovic-Calogovic, M. Rajic,
G. Giester, Inorg. Chim. Acta 322 (2001) 65.