Correlation of infrared spectra of zinc(II) carboxylates with their structures

Article abstract of DOI:10.1016/j.saa.2006.02.050

The correlation of the infrared spectra of zinc(II) carboxylates with their structures was investigated in the paper. The complexes with different modes of the carboxylate binding, from chelating, through bridging (syn-syn, syn-anti, monatomic), ionic to

Full text of DOI:10.1016/j.saa.2006.02.050

Spectrochimica Acta Part A 66 (2007) 262–272
Correlation of infrared spectra of zinc(II) carboxylates with their structures
∗
ˇ´
´
¨
´
V. Zelenak , Z. Vargova, K. Gyoryova
ˇ
Department of Inorganic Chemistry, Faculty of Science, P.J. Safa´rik University, Moyzesova 11, 041 54 Kosˇice, Slovak Republic
Received 30 December 2005; received in revised form 22 February 2006; accepted 22 February 2006
Abstract
The correlation of the infrared spectra of zinc(II) carboxylates with their structures was investigated in the paper. The complexes with different
modes of the carboxylate binding, from chelating, through bridging (syn–syn, syn–anti, monatomic), ionic to monodentate were used for the
study, namely [Zn(C6H5CHCHCOO)2(H2O)2] (I) with chelating carboxylate group (C6H5CHCHCOO = cinnamate), [Zn2(C6H5COO)4(pap)2]
(II) with syn–syn bridging carboxylate (C6H5COO = benzoate; pap = papaverine), [Zn(C6H5CHCHCOO)2(mpcm)]n (III) with syn–anti car-
boxylate bridge (mpcm = methyl-3-pyridylcarbamate), [Zn(C5H4NCOO)2(H2O)4] (IV) with ionic carboxylate group (C5H4NCOO = nicotinate),
[Zn(C6H5COO)2(pcb)2]n (V) with monodentate carboxylate coordination (pcb = 3-pyridylcarbinol) and [Zn3(C6H5COO)6(nia)2] (VI) with syn–syn
and monatomic carboxylate bridges (nia = nicotinamide). First, the mode of the carboxylate binding was assigned from the infrared spectra using
the magnitude of the separation between the carboxylate stretches, Δexp = νas(COO−) − νs(COO−). Then the values Δexp were compared with those
calculated from structural data of the carboxylate anion (Δcalc). The conclusions about the carboxylate binding which resulted from the Δ values,
were confronted with the crystal structure of the complexes. The limitations and recommendations were formulated to assign the mode of the
carboxylate binding from the infrared spectra. The dependence of the Δexp values on the magnitudes of Zn–O–C angles in bidentate carboxylate
coordination was observed.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Zinc; Carboxylate; IR; Stretching vibration; Structure
1. Introduction
These criteria are applied to assign the type of the carboxylate
coordination in inorganic complexes [9–16] and biomolecules
as well [13,17,18]. It was shown that frequencies of the asym-
metric carboxylate stretches of alkali-metal halogenobenzoates
and the values of Δ correlate with the ionic potential of
metals and halogens as substituents [19,20]. Generally, the
following order is proposed for divalent metal carboxylates
[9–12]:
Metal carboxylates are subject of extensive research for
few decades. They are used to study the phenomenon of the
antiferromagnetic interaction [1], as model compounds of met-
aloenzymes [2] or they show antibacterial properties [3–6]. The
carboxylate anions are nowadays studied as important build-
ing blocks of “metal–organic frameworks” (MOFs), the porous
compoundschallengingasthehydrogenstoragesystems(energy
source) [7,8].
Δ (chelating) < Δ (bridging) < Δ (ionic) < Δ (monodentate),
The carboxylate group has versatile coordination behaviour.
It can be ionic, monodentate, bidentate chelating or bridging.
One method to determine the mode of the carboxylate binding
is the infrared spectroscopy (IR). The frequency of the asymmet-
ric carboxylate vibration in the IR spectra, νas(COO−), and the
magnitude of the separation between the carboxylate stretches,
Δ = νas(COO−) − νs(COO−), are often used as spectroscopic
criterions to determine the mode of the carboxylate binding.
where Δ ionic is approximately 160–170 cm−1 for acetates
[9,12]. In the monodentate coordination the redistribution of the
electron density takes place and the shift of νas(COO−) to higher
wavenumbers is observed in comparison with the ionic group,
increasing the value of Δ. On contrary the chelating coordina-
tion shifts the position of the asymmetric carboxylate stretch
to lower wavenumbers in comparison with the ionic group and
lowers the value of Δ. In the bridging coordination, when one
divalentmetalcationisboundtooneoftheoxygensoftheCOO−
group and another divalent metal cation to the other oxygen,
the band νas(COO−) is located at the same position as that of
∗
Corresponding author. Tel.: +421 55 6228114.
E-mail address: zelenak@upjs.sk (V. Zelenˇa´k).
1386-1425/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2006.02.050

V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
263
the ionic group [17]. A number of works use this criterion for
the assignment of the carboxylate binding, especially when no
structural data are available. However the range 160–170 cm−1
was derived for acetates and in general the comparison of the
Δ value of the respective complex with the Δ value of the par-
ticular sodium salt should be used for the assignment following
the guidelines: (i) bidentate chelating coordination occurs when
Δ(COO−)studied complex ꢀ Δ(COO−)sodium salt; (ii) the biden-
tate bridging carboxylate exists when Δ(COO−) studied com-
plex ≤ Δ(COO−)sodium salt; (iii) monodentate coordination is
characterised by Δ(COO−)studied complex ꢁ Δ(COO−)sodium salt
[9–11,21,22].
The relation between the values of Δ and the structure of the
carboxylate anion was proposed and derived by Nara et al. from
the molecular orbital calculations of the vibrational frequencies
of carboxylate compounds [23]:
Δ = 1818.1δr + 16.47(θOCO − 120) + 66.8
(1)
˚
where δr is difference between the two CO bond lengths (A) and
θOCO is the OCO angle (◦). This relation was applied to study of
zinc(II) acetate dihydrate and anhydrous zinc(II) acetate [15].
It was found a good agreement between the calculated and the
experimental Δ values for anhydrous zinc(II) acetate, but not
for the zinc acetate dihydrate. Indeed, these two compounds
have different modes of the carboxylate binding (bridging or
chelating one, respectively). Therefore the aim of our study is
to investigate the applicability and coherence of the aforemen-
tioned criteria for the set of zinc(II) carboxylates with various
Fig. 1. (a) Infrared spectrum of the complex [Zn(C6H5CHCHCOO)2(H2O)2]
(I) and (b) the structure of the complex [Zn(C6H5CHCHCOO)2(H2O)2] (I) [24].
modes of the carboxylate binding, from chelating, through
bridging, ionic to monodentate. Aromatic zinc(II) carboxylates
with known molecular and crystal structures were used for
this purpose, namely [Zn(C6H5CHCHCOO)2(H2O)2] (I) with
chelating carboxylate group (C6H5CHCHCOO = cinnamate)
[24], [Zn2(C6H5COO)4(pap)2] (II) with syn–syn bridging
Scheme 1.

264
V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
Table 1
Assignments of the principal IR bands for the compounds I–VI
Assignment
Compound
I
II
III
IV
V
VI
ν(O H)
ν(N H)
3418br, m
–
–
–
–
3249br, s
–
3436m
–
–
3247w
3128w
3376s
3200s
ν(C H)ar
3059w
3028w
3064w
3062w
3072w
–
3057w
3066w
3027w
ν(C H)aliph
–
2992w
2934w
2834w
2949w
2932w
2866w
–
ν(C O)
–
–
–
–
–
1734s
1641s
–
–
–
–
1670s
δ(H
O
H)
1612s
–
–
–
ν(C C)cinn
1641s
ν(C N)ar, ν(C C)ar
1578s
1642s
1620s
1591s
1578s
1589s
1618s
1607s
1600s
νas(COO−)
1523s, br
1572s
1561s
1523s
1570s
1552s
1630s
1568s
1541s
ν(C C)ar
1497s
1452s
1512s
1487m
1494m
1496w
1452s
1482w
1437s
1492w
δ(CH3) + δ(CCH)
1464w
1450w
1426m
1451s
1441s
1448m
νs(COO−)
1411s
1404s
1418s
1376s
1391s
1392s
1401s
1343s
ν(C N)
δ(CH2) + δ(CH3)
ν(C O)
–
–
–
–
1273m
–
1336w
1295w
1257m
–
–
–
–
1351s
1313m
1308m
–
–
δ(CCH) + δ(CH3, CH2)
1292w
1254m
1203w
1072w
1030w
1214w
1162m
1058w
1027m
1195w
1134w
1079m
1059m
1027w
1329w
1247w
1199w
1158w
1119w
1054m
1033w
1249m
1197m
1177m
1113m
1059s
1204m
1176w
1133m
1097w
1071w
1062m
1024m
1033w
ν(C
O
C)
–
1240m
–
1239m
–
–
–
–
–
–
γ(CCH)cinn
978s
999m
980m
γ(CCH)ar + ρ(CH3, CH2)
876m
773s
685s
992w
935w
856m
786w
879m
810w
774s
844m
765m
647w
993w
951m
926w
824s
794m
703s
942w
843m
802w
789w
695s
687m
680s
650w
618w
639w
δ(COO−)
γ(COO−)
715m
592m
718s
679m
717m
592w
700m
600w
718s
676s
723s
–
–
567w
Abbreviations: s, strong; m, medium; w, weak; br, broad; ar, aromatic; aliph, aliphatic; cinn, cinnamate.

V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
265
Table 2
Wavenumbers of the COO− stretches and the experimental Δexp values in the compounds I–VI and the respective sodium carboxylates
νas(COO) (cm−1
)
νs(COO) (cm−1
)
Δexp. (cm−1
)
Δsodium salt (cm−1
)
a
Compound
[Zn(C6H5CHCHCOO)2(H2O)2] (I)
[Zn2(C6H5COO)4(pap)2] (II)
1523
1572
1411
1404
112
168
143
139
[Zn(C6H5CHCHCOO)2(mpcm)]n (III)
1561
1523
1376
1418
185
105
143
[Zn(C5H4NCOO)2(H2O)4] (IV)
[Zn(C6H5COO)2(pcb)2]n (V)
1570
1552
1391
1392
179
160
164
138
[Zn3(C6H5COO)6(nia)2] (VI)
1630
1568
1541
1343
1401
287
167
140
139
a
The respective νas(COO−) and νs(COO−) values are 1551 and 1408 cm−1 for cinnamate (C6H5CHCHCOO−), 1553 and 1414 cm−1 for benzoate (C6H5COO−),
1568 and 1404 cm−1 for nicotinate (C5H4NCOO−).
carboxylate (C6H5COO = benzoate; pap = papaverine) [25a],
[Zn(C6H5CHCHCOO)2(mpcm)]n (III) with syn–anti car-
boxylate bridges (mpcm = methyl-3-pyridylcarbamate) [25b],
[Zn(C5H4NCOO)2(H2O)4] (IV) with ionic carboxylate group
(C5H4NCOO = nicotinato) [25c], [Zn(C6H5COO)2(pcb)2]n
(V) with monodentate carboxylate coordination (pcb = 3-
pyridylcarbinol) [25d] and [Zn3(C6H5COO)6(nia)2] (VI) with
syn–syn and monatomic carboxylate bridges (nia = nicoti-
namide) [25e]. The types of the carboxylate binding studied in
the paper are summarised in Scheme 1.
ture (1/1, v/v) as a reaction medium. The complex II,
[Zn2(C6H5COO)4(pap)2], was synthesised in absolute ethanol
using zinc chloride, sodium benzoate and papaverine in the
molar ratio 1:2:2. For the preparation of the compound
III, [Zn(C6H5CHCHCOO)2(mpcm)]n, water/ethanol mixture
(1/1, v/v) of zinc sulphate, sodium cinnamate and methyl-3-
pyridylcarbamateinthemolarratio1:2:2wasused. Complex IV,
[Zn(C5H4NCOO)2(H2O)4], was prepared from aqueous solu-
tion of nicotinic acid and zinc(II) acetate in the molar ratio 2:1.
Finally, aqueous solution of zinc sulphate, sodium benzoate and
3-pyridylcarbinol or nicotinamide in the molar ratios 1:2:2 or
1.5:3:1, respectively were used in the synthesis of the complexes
V and VI. Examination of the composition of the compounds by
C, H, N elemental analysis and complexometric determination
of the zinc content confirmed the composition of the complexes.
The details of the syntheses, elemental analyses and the crystal
structures of the complexes are described in Refs. [24,25].
2. Experimental
To prepare the complexes I–VI we have used various solven-
ts and zinc(II) salts. The compound I, [Zn(C6H5CHCHCOO)2
(H2O)2], was synthesised from zinc sulphate and sodium
cinnamate in the molar ratio 1:2 using water/ethanol mix-
Table 3
Bond lengths and angles of the COO− groups and the respective values of Δcalc for the compounds I–VI
˚
Compound
Bond length (A)
Angle (◦)
Δcalc (cm−1
)
Ref.
[24]
I [Zn(C6H5CHCHCOO)2(H2O)2]
C1–O1
C1–O2
1.275
1.246
O1–C1–O2
119.9
118
a
a
II [Zn2(C6H5COO)4(pap)2]
C–OAav
C–OBav
1.2575a
1.2535a
OA–C–OBav
125.45a
164(153 + 175)
[25a]
[25b]
a
III [Zn(C6H5CHCHCOO)2(mpcm)]n
C13–O6
C13–O7
C11–O3
C11–O2
1.280
1.245
1.267
1.260
O6–C13–O7
O2–C11–O3
124.6
120.6
206
90
IV [Zn(C5H4NCOO)2(H2O)4]
V [Zn(C6H5COO)2(pcb)2]n
C6–O2
C6–O3
1.260
1.256
O2–C6–O3
O1–C1–O2
125.2
160
[25c]
[25d]
C1–O2
C1–O1
C11–O11
C11–O12
1.268
1.251
1.270
1.257
125.4
124.1
187
158
O11–C11–O12
VI [Zn3(C6H5COO)6(nia)2]
C21–O21
C21–O22
C30–O31
C30–O32
1.275
1.247
1.306
1.228
O21–C21–O22
O31–C30–O32
125.5
121.1
208
227
[25e]
a
The average values of C1–O1A and C8–O2A distances (denoted as C–OAav), C1–O1B and C8–O2B distances (denoted as C–OBav) and O1A–C1–O1B and
O2A–C8–O2B angles (denoted as OA–C–OBav) (see text for the details and explanation).

266
V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
The infrared spectra of the complexes were recorded with a
EQUINOX 55 (Bruker) FT-IR spectrophotometer as KBr discs
(complex/KBr mass ratio 1:100) in the range 4000–500 cm−1
For the measurements the crystals from the same batch were
used than for the single-crystal X-ray analysis. The elemental
analyses were performed with a Perkin-Elmer 2400 CHN Ele-
mental Analyzer.
Table 1. The assignments were made according to the literature
data [5,6,10,26–30]. The stretching vibration of the carboxylate
is known to have a wide variation in position due to hydrogen
bonding and therefore the assignment of the ν(COO−) modes
had to be made carefully. This was especially true for the com-
pounds II, III, V and VI containing pap, mpcm, pcb, nia ligands
in addition to the carboxylate. To be confident about our IR
assignments we have compared and evaluated the IR spectra of
the compounds II, III, V and VI against spectra of (i) respective
carboxylic acids and their sodium salts, (ii) simple zinc(II) car-
boxylates, and (iii) respective organic ligands (pap, mpcm, pcb,
nia). For example, the assignment of the band at 1620 cm−1 (see
.
3. Results and discussion
The wavenumbers, intensities and the assignments of the
principal IR bands for the compounds I–VI are summarised in
Fig. 2. (a) Infrared spectrum of the complex [Zn2(C6H5COO)4(pap)2] (II) (solid line) and papaverine hydrochloride (dashed line); (b) the structure of
[Zn2(C6H5COO)4(pap)2] [25a]; (c) the mutual orientation of the aromatic ring and the carboxylate group in two non-equivalent benzoato ligands of the com-
plex II.

V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
267
Table 1) either to papaverine ring vibration or νas(COO−) vibra-
tion was examined for the compound [Zn2(C6H5COO)4(pap)2]
(II). After comparison of the spectrum of the complex II with
those of papaverine ligand and the papaverine hydrochloride, the
bandwasattributedtotheringvibrationofpapaverine. Similarly,
the comparison of the spectrum of [Zn(C6H5COO)2(pcb)2]n (V)
with that of 3-pyridylcarbinol showed the bands at 1618 and
1607 cm−1 to belong to the vibration of pcb ligand.
The Δ values as determined from the IR spectra of the com-
pounds I–VI together with the Δ values of the respective sodium
salts are summarized in Table 2. The values calculated by the
substitution of the structural data to Eq. (1) are in Table 3. For
the compound I, [Zn(C6H5CHCHCOO)2(H2O)2], the asym-
metric and symmetric carboxylate stretches were observed
at 1523 and 1411 cm−1, respectively (see Fig. 1a). The low
separation of the stretches, Δexp = 112 cm−1, is an indica-
tion of chelating coordination of the carboxylate group since
Δexp (112 cm−1) ꢀ Δsodium cinnamate (143 cm−1). The result is
in accordance with the crystal structure of the zinc(II) cinnamate
dihydrate (see Fig. 1b) [24]. The substitution of the parameters
␦r and θOCO of the cinnamate anion to Eq. (1) gives the value
Δcalc = 118 cm−1 (see Table 2). It is therefore obvious, that for
chelating carboxylate low, nearly equal experimental and cal-
culated Δ values resulted and both criteria reflected the chelate
carboxylate coordination in the complex I adequately.
For the compound II, [Zn2(C6H5COO)4(pap)2], the
νas(COO−) and νs(COO−) stretches were observed at 1572 and
1404 cm−1, respectively (Fig. 2a). The experimental Δ value
(168 cm−1) > Δsodium benzoate (139 cm−1), which suggested, that
monodentate carboxylate coordination should be present in
the complex II, which is not true. The structural analysis
has shown that complex II forms the paddle–wheel cen-
trosymmetric dimmer with four syn–syn carboxylato bridges
(see Fig. 2b) [25a]. The carboxylato ligands in the dim-
mer are not equivalent and the pairs of bridges differ
˚
in C–O distances and O–C–O angles (C1–O1A = 1.257 A,
Fig. 3. (a) Infrared spectrum of the complex [Zn(C6H5CHCHCOO)2(mpcm)]n (III) (solid line) and methyl-3-pyridylcarbamate (dashed line) and (b) the structure
of the complex [Zn(C6H5CHCHCOO)2(mpcm)]n (III) [25b].

268
V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
νs(COO−) in the IR spectrum. This assumption was supported
by the calculation, when the average C O bond lengths and
O–C–O angles were substituted to Eq. (1) (see Table 3). The
calculation gave Δcalc = 164 cm−1, i.e. the value very close to
the experimental one (168 cm−1).
˚
˚
˚
C1–O1B = 1.255 A, C8–O2A = 1258 A, C8–O2B = 1.252 A,
O1A–C1–O1B = 125.0◦, O2A–C8–O2B = 125.9◦). Accordly,
two values Δcalc resulted from Eq. (1) (153 and 175 cm−1) in
contrast to the one experimental value observed in the spectrum.
The existence of two distinct vibration levels is also suggested
by the geometry of the non-equivalent carboxylato ligands. In
one benzoato ligand, the carboxylate group is almost coplanar
with the aromatic ring, whereas in the other one it is not (see
Fig. 2c). The conjugation effects between aromatic ring and car-
boxylate group are more favourable in the coplanar orientation,
and consequently the electronic charge density in both carboxy-
lato ligands should be different. Therefore two vibrational levels
for the non-equivalent carboxylate groups should be expected in
line with the two distinct values Δcalc. However, this fact does
not agree with the experiment, where only one asymmetric and
symmetric stretch was observed. We suppose that because the
non-equivalent benzoato ligands have the same symmetry and
they are close to each other, their vibration levels may be also
very close. Therefore an interaction may exist between these
groups originating to only one pair of the bands νas(COO−) and
Two asymmetric (1561 and 1523 cm−1) and two sym-
metric (1418 and 1376 cm−1) carboxylate stretches were
observed in the IR spectrum of the compound III,
[Zn(C6H5CHCHCOO)2(mpcm)]n (see Fig. 3a), indicating the
existence of two different modes of the carboxylate binding.
The frequency of asymmetric and symmetric vibration depends
on the electronic charge density of the C O bonds in the car-
boxylate. Therefore we can suppose that the higher are dif-
ferences in electronic density of C O bonds (and C O bond
lengths) the higher is frequency of the asymmetric vibration and
the lower is the frequency of the symmetric vibration. Hence
in the compound III the asymmetric vibration at 1561 cm−1
appertains to symmetric vibration at 1376 cm−1. Similarly the
stretches at 1523 and 1418 cm−1 are related. The respective
experimental values, Δexp, are 185 and 105 cm−1. The first
Fig. 4. (a) Infrared spectrum of the complex [Zn(C5H4NCOO)2(H2O)4] (IV) and (b) the structure of the complex [Zn(C5H4NCOO)2(H2O)4] (IV) [25c].

V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
269
value is an indication of the monodentate carboxylate coordina-
tion as Δexp (185 cm−1) ꢁ Δsodium cinnamate (143 cm−1), while
the second one of the bidentate chelating coordination Δexp
(105 cm−1) ꢀ Δsodium cinnamate (143 cm−1). The comparison of
these findings with the structure of the complex III (see Fig. 3b)
shows, that the assignment of the monodentate group was cor-
rect, but the chelating one was not. Instead of chelating coordi-
nation the syn–anti bridges exist in the complex III. The Δcalc
values for these two ligands are 206 and 90 cm−1, respectively.
For the monodentate coordination the Δexp value is inferior of
the calculated one, probably due to hydrogen bonding of the
uncoordinated carboxylate oxygen with the hydrogen atom of
NH group of mpcm. This hydrogen bonding withdraws the elec-
tron density of the C O bond towards the oxygen atom and
decreases the frequency of νas(COO−). On contrary the syn–anti
bridge is reflected by very low values of Δexp and Δcalc. The low
values of separation for syn–anti bridge were also observed in
anhydrous zinc(II) acetate [15]. Hence the low separation of
the carboxylate stretches indicates not only chelating, but also
bridging coordination of carboxylate.
Next, let us examine the infrared spectrum of the complex
[Zn(C5H4NCOO)2(H2O)4] (IV). The IR spectrum of this com-
pound is shown in Fig. 4a. The two distinct experimental and cal-
culated Δ values were observed (see Tables 2 and 3), from which
it was not possible to identify expressly the mode of the car-
boxylate binding. The experimental value (Δexp = 179 cm−1) is
a superior of that observed for the sodium nicotinate (164 cm−1
;
see Table 2) and it is shifted to the values typical for the mon-
odentate coordination. On the other hand, the calculated value
Δcalc = 160 cm−1 is lower than the experimental one and is typ-
ical for bridging coordination. The structural analysis of the
complex IV has shown that ionic group is present in the com-
pound (Fig. 4b) [25c]. Both, conjugation effect and hydrogen
bonding may explain the higher value of Δexp. The carboxy-
late anion is perfectly coplanar with the aromatic ring. The
carboxylate group has polar character and the electron cloud
Fig. 5. (a) Infrared spectrum of the complex [Zn(C6H5COO)2(pcb)2]n (V) (solid line) and 3-pyridylcarbinol (dashed line) and (b) the structure of
[Zn(C6H5COO)2(pcb)2]n [25d].

270
V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
forming C O bonds is withdrawn from the centre in the direc-
tion of the oxygen atoms. Moreover, both carboxylate oxygens
form hydrogen bonds with the water ligands of the neighbouring
[Zn(C5H4NCOO)2(H2O)4] units. As a consequence the conju-
gation between pyridine ring and carboxylate anion increases
the electron density in the C O bonds. A savoir, slightly larger
Δexp value was observed than for carboxylate group in sodium
nicotinate. The Δcalc value is an inferior of the experimental one.
The electron density is equally distributed in both C O bonds,
which results in very small difference in C O lengths (param-
eter δr in Eq. (1)). Thus the evidence of the ionic group in the
complex IV using spectroscopic criteria is ambiguous and not
straightforward.
The infrared spectrum of the complex V, [Zn(C6H5COO)2
(pcb)2]n, is shown in Fig. 5a. The asymmetric and the symmet-
ric carboxylate stretches are observed at 1552 and 1392 cm−1
,
respectively and Δexp = 160 cm−1. It has to be noted that in
our study dealing with the structural characterisation of com-
Fig. 6. (a) Infrared spectrum of the complex [Zn3(C6H5COO)6(nia)2] (VI) (solid line) and nicotinamide (dashed line); (b) the structure of [Zn3(C6H5COO)6(nia)2]
[25e]; (c) the mutual orientation of the aromatic ring and the carboxylate group in two non-equivalent benzoato ligands of the complex VI.

V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
271
pound V [25d] we supposed the band at 1618 cm−1 to be
due to asymmetric COO− vibration. However, the compari-
son of the spectrum of the complex V and that of free 3-
pyridylcarbinol ligand (see Fig. 5a) showed, that this vibra-
tion is due to ring vibration of 3-pyridylcarbinol. The value of
Δexp (160 cm−1) > Δsodium benzoate (139 cm−1), which indicates
the monodentate coordination in accordance with the crystal
structure (see Fig. 5b). However for monodentate coordination
usually higher Δ values are typical. The strong intramolecular
hydrogen bond between uncoordinated carboxylate oxygen and
hydrogen atom of the symmetrically related OH◦group of 3-
Scheme 2.
“symmetric” (compare Figs. 2c and 6c). Both these factors can
lead to the higher differences in the electronic charge density
in the carboxylato ligands of the complex VI. Consequently
the existence of two distinct vibrational levels results in two
νas(COO) bands in the infrared spectra.
˚
pyidylcarbinol (O H· · ·O = 1.69 A; O H O = 168 ) withdraws
the electron cloud of the C O bond closer to the oxygen atom.
The C O bond order slightly decreases which shifts the asym-
metric vibration to the lower wavenumber. The calculated value
For the two non-equivalent syn–syn bridges in the compound
VI, Δcalc was 158 and 208 cm−1 and for the monatomic bridge
Δcalc = 227 cm−1 (see Table 3). These results suggest that for
syn–syn carboxylate bridges, which are not ideally symmet-
ric, or for monatomic bridge, the Nara’s equation looses its
validity. An interesting dependence between the “symmetry”
of the bidentate carboxylate coordination and the values of Δexp
was found. We have defined the parameter Σ = |240 − (α + β)|,
where α and β are Zn–O–C angles as depicted in Scheme 2.
For ideally symmetric coordination α = β = 120◦ and Σ = 0. The
respective Zn–O–C angles (α and β angles) are 97◦ and 85◦ for
the compound I, 125◦ and 132◦ or 128◦ and 129◦ for the two
non-equivalent syn–syn benzoato ligands in compound II, 120◦
and 226◦ for the syn–anti bridge in the compound III and finally
114◦ and 133◦ or 120◦ and 139◦ for the two syn–syn bridges in
the compound VI.
The dependence of Σ on the magnitude of the separation
between carboxylate stretches is shown in Fig. 7. From the figure
it is obvious, that as the parameter Σ increases (the “deforma-
tion” of the bidentate coordination increases), the magnitude of
the separation Δexp decreases. It is to note, that for the non-
equivalent benzoate ligands in the compound II two equal Σ
values were obtained (2 × 17◦) which corresponds to one exper-
imental value found from the IR spectrum (see Section 3 for the
compound II).
Δcalc is 187 cm−1
.
Finally we shall examine how the IR spectrum reflects the
existence of carboxylate bridges of different types in com-
plex VI, [Zn3(C6H5COO)6(nia)2], shown in Fig. 6a. Three
bands due to νas(COO−) vibrations were observed in the spec-
trum at 1630, 1568 and 1541 cm−1. The symmetric νs(COO−)
stretches were observed at 1401 and 1343 cm−1. The signifi-
cantly higher/lower (1630/1343 cm−1) frequency of carboxylate
stretch than those observed for the complexes I–V indicates
higher/lower bond order and longer/shorter bond lengths, i.e.
the fact that these vibrations originate from the carboxylate
in which one of the C O distances is rather shorter than the
other one. The experimental Δ value (287 cm−1, see Table 2)
is rather higher than that of Δsodium benzoate (139 cm−1) and
shows inequality of both C O distances and the monodentate
coordination. Indeed, the structural analysis confirms the mon-
odentate bridging coordination (monatomic bridge) with the
C O bond lengths characteristic of localised single and dou-
˚
ble bonds, 1.306 and 1.219 A. Similar high frequency of the
asymmetric vibration was observed in other isostructural car-
boxylate complexes containing monatomic bridge [29,30]. The
other two νas(COO−) vibrations were observed at 1568 and
1541 cm−1 with the Δexp values 167 and 140 cm−1, respectively.
First Δ value (167 cm−1) > Δsodium benzoate (139 cm−1) and sug-
gests monodentate carboxylate coordination in the complex VI.
The second Δ value (140 cm−1) is almost equal to that observed
for sodium benzoate and indicates that ionic or bridging group
exists in the complex VI. With respect to the trinuclear character
of the complex VI, the bridging coordination is more feasible.
The structural results have shown that both vibrations (1568
and 1541 cm−1) originate from two non-equivalent syn–syn car-
boxylate bridges in the complex VI (see Fig. 6b). The high Δ
value for one of them (167 cm−1) is probably due to the defor-
mation of the bridge. The existence of two distinct bands of
νas(COO−) vibrations for two non-equivalent benzoates is in
contrast with only one band observed in the complex II. The
non-equivalent carboxylate groups in the complex VI have dif-
ferent orientation with respect to the aromatic ring. One of the
COO− group is almost coplanar with the aromatic ring, while
the other one includes angle greater than in the complex II (see
Fig. 6c). Moreover, in the complex VI the coordination of the
oxygen atoms of the carboxylato ligand to zinc atoms is less
Fig. 7. The dependence of Σ and Δexp. The dashed line is a guide for eyes.

272
V. Zelenˇa´k et al. / Spectrochimica Acta Part A 66 (2007) 262–272
[3] M. Melnik, M. Koman, D. Hudecov
4. Conclusions
a´, J. Moncol’, B. Dudova´, T. Glowiak,
J. Mrozinski, C.E. Holloway, Inorg. Chim. Acta 308 (2000) 1.
[4] V. Zelenˇa´k, K. Gyo¨ryova´, D. Mlynarcˇ´ık, Met. Based Drugs 8 (2002) 269.
[5] P. Koczon´, J. Piekut, M. Borawska, R. Swisłocka, W. Lewandowski,
The values of the separation between the carboxylate
stretches (Δ = νas(COO−) − νs(COO−)) in the IR spectra are
used in the literature to assign the mode of the COO− bind-
ing. The criterion is often applied unambiguously, especially in
works when no structural data are available. The present study
has shown that the assignment of the carboxylate binding based
only on the spectroscopic criterion has to be done very carefully
because the criterion does not have general validity and can be
influenced by hydrogen bonds and electronic effects. The fol-
lowing conclusions resulted from the study:
´
Spectrochim. Acta Part A: Mol. Biomol. Struct. 61 (2005)
1917.
´
[6] P. Koczo
Struct. 651–653 (2003) 651.
n´, J. Piekut, M. Borawska, R. Swisłocka, W. Lewandowski, J. Mol.
[7] H. Li, M. Eddaoudi, M. O’Keeffe, O.M. Yaghi, Nature 402 (1999)
276.
[8] N.L. Rosi, J. Eckert, M. Eddaoudi, D.T. Vodak, J. Kim, M. O’Keeffe, O.M.
Yaghi, Science 300 (2003) 1127.
[9] G.B. Deacon, R.J. Phillip, Coord. Chem. Rev. 33 (1980) 227.
[10] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, John Wiley & Sons, New York, 1997.
[11] B.S. Manhas, A.K. Trikha, J. Indian Chem. Soc. 59 (1982) 315.
[12] D. Martini, M. Pellei, C. Pettinari, B.W. Skelton, A.H. White, Inorg. Chim.
Acta 333 (2002) 72.
[13] W. Lewandowski, M. Kalinowska, H. Lewandowska, Inorg. Chim. Acta
358 (2005) 2155.
[14] D. Czakis-Sulikowska, A. Czylkowska, J. Therm. Anal. Cal. 71 (2003) 395.
[15] T. Ishioka, Y. Shibata, M. Takahaski, I. Kanesaka, Y. Kitagawa, K.T. Naka-
mura, Spectrochim. Acta Part A: Mol. Biomol. Struct. 54 (1998) 1827.
[16] T. Ishioka, Y. Shibata, M. Takahaski, I. Kanesaka, Spectrochim. Acta Part
A: Mol. Biomol. Struct. 54 (1998) 1811.
[17] M. Nara, M. Tasumi, M. Tanokura, T. Hiraoki, M. Yazawa, A. Tsutsumi,
FEBS Lett. 349 (1994) 84.
[18] F. Yumoto, M. Nara, H. Kagi, W. Iwasaki, T. Ojima, K. Nishita, K. Nagata,
M. Tanokura, Eur. J. Biochem. 268 (2001) 6284.
[19] W. Lewandowski, L. Fuks, M. Kalinowska, P. Koczon´, Spectrochim. Acta
Part A: Mol. Biomol. Struct. 59 (2003) 3411.
[20] P. Koczon´, P. Mos´cibroda, K. Dobrosz-Teperek, W. Lewandowski, J. Mol.
Struct. 565–566 (2001) 7.
• The infrared spectra can finely reflect the different modes
of the carboxylate binding in the molecule. The spectra
allowed resolving monodentate coordination and syn–anti
bridge (compound III) or even two different syn–syn carboxy-
late bridges (compound VI).
• The values of Δexp below 120 cm−1 frequently indicate
chelating carboxylate group (compound I). However, the
syn–anti bridges can yield the separation around this value
as well (compound III).
• For the compounds with the ionic COO− group or syn–syn
bridging coordination the observed values Δexp were
170 10 cm−1 (compounds II and IV, VI). However the val-
ues Δexp in this range does not confirm the ionic group or
syn–syn bridge definitely. In monodentate coordination with
uncoordinated carboxylate oxygen involved in the hydrogen
bonding the Δ values can fall also into this range (compound
V). Therefore the assignment of the carboxylate binding in
this range has to be made very carefully.
[21] W. Lewandowski, M. Kalinowska, H. Lewandowska, J. Inorg. Biochem.
99 (2005) 1407.
[22] D. Czakis-Sulikowska, A. Czylkowska, A. Malinowska, J. Therm. Anal.
Cal. 67 (2002) 667.
• The high values of Δexp (above 180 cm−1) were typical for
the monodentate coordination (compounds III, VI) because
the other types of coordination do not give such high values
[23] M. Nara, H. Torii, M. Tasumi, J. Phys. Chem. 100 (1996) 19812.
[24] H. Hosomi, S. Ohba, Y. Ito, Acta Cryst. C56 (2000) e123.
ˇ
[25] (a) V. Zelenˇa´k, M. Sabo, W. Massa, J. Cerna´k, Acta Cryst. C60 (2004) m85;
(b) V. Zelenˇa´k, I. C´ısaˇrova´, P. Lewellyn. Inorg. Chem. Commun., submitted
for publication;
of Δexp
.
• It was shown, that for bidentate carboxylate coordination
the magnitude of the separation between the carboxylate
stretches, Δexp, depends on the symmetry of the coordina-
tion of the bidentate carboxylate to zinc ion.
(c) Z. Vargova´, V. Zelenˇa´k, I. C´ısaˇrova´, K. Gyo¨ryova´, Thermochim. Acta
423 (2004) 149;
(d) V. Zelenˇa´k, I. C´ısaˇrova´, M. Sabo, P. Llewellyn, K. Gyo¨ryova´, J. Coord.
Chem. 57 (2004) 87;
(e) V. Zelenˇa´k, M. Sabo, W. Massa, P. Llewellyn, Inorg. Chim. Acta 357
(2004) 2049.
Acknowledgement
[26] L.J. Bellamy, The Infrared Spectra of Complex Molecules, vol. 1, Chapman
& Hall, London, 1975.
[27] W. Lewandowski, B. Dasiewicz, P. Koczon´, J. Skierski, K. Dobrosz-
The work was supported by the Grant Agency of the Slo-
vak Ministry of Education (VEGA grant No. 1/2474/05). The
support is gratefully acknowledged.
´
Teperek, R. Swisłocka, L. Fuks, W. Priebe, A.P. Mazurek, J. Mol. Struct.
604 (2002) 189.
[28] A. Civadze, G. Cinciadze, N. Gongadze, J. Charitonov, Koord. Khim. 9
(1975) 1221.
References
[29] W. Clegg, D.R. Harbron, Ch.D. Homan, P.A. Hunt, I.R. Little, B.P.
Straughan, Inorg. Chim. Acta 186 (1991) 51.
[30] W. Clegg, P.A. Hunt, B.P. Straughan, M.A. Mendiola, J. Chem. Soc. Dalton
Trans. (1989) 1127.
[1] O. Kahn, Molecular Magnetism, VCH, New York, 1993.
[2] T. Tanase, J.W. Yun, S.J. Lippard, Inorg. Chem. 35 (1996) 3585.