Coordination ability of free or silica immobilized Schiff bases towards Hg(II), Cd(II) and Pb(II) ions

Article abstract of DOI:10.1016/j.ica.2013.10.005

A series of ligands, consisting of Schiff bases and their reduced amine derivatives, was prepared by [2+1] condensation of 3-methoxy (or 3-ethoxy)-2-hydroxybenzaldehyde or 2-hydroxybenzaldehyde with 3,4-diaminobenzoic acid or 1,2-diaminobenzene, and by su

Full text of DOI:10.1016/j.ica.2013.10.005

Inorganica Chimica Acta 410 (2014) 29–38
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Coordination ability of free or silica immobilized Schiff bases towards
Hg(II), Cd(II) and Pb(II) ions
Andrea Magro a, Laura Crociani a, Cristina Prinzivalli a, Pietro Alessandro Vigato a
Pier Luigi Zanonato b, Sergio Tamburini a,
⇑
a Institute for Energetics and Interphases, CNR, Padova 35127, Italy
b Department of Chemical Sciences, University of Padova, Via Marzolo 1, Padova 35131, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 June 2013
Received in revised form 11 September 2013
Accepted 6 October 2013
Available online 14 October 2013
A series of ligands, consisting of Schiff bases and their reduced amine derivatives, was prepared by [2+1]
condensation of 3-methoxy (or 3-ethoxy)-2-hydroxybenzaldehyde or 2-hydroxybenzaldehyde with
3,4-diaminobenzoic acid or 1,2-diaminobenzene, and by subsequent reduction with hydrogen or NaBH4.
The complexation behavior of these ligands towards Cd(II), Pb(II) and Hg(II) salts was tested and the
resulting complexes characterized by elemental analysis, IR and NMR spectroscopy, ESI-MS mass spec-
trometry and SEM–ESEM microscopy. Ligand-bearing silicas were prepared by the reaction of a carbox-
ylic group at the periphery of the coordinating moiety of the Schiff bases or amine derivatives with
functionalized silicas bearing NH2 or Cl groups. Ligand-bearing silicas were characterized by elemental
analysis, IR spectroscopy and SEM–ESEM microscopy. Finally, the capability of these materials to seques-
trate the above mentioned metal ions from water solutions was studied. Preliminary data on the selective
recognition of mercury ions from cadmium and lead ions are reported.
Keywords:
Schiff bases
Functionalized silica
Grafting reactions
Mercury recognition
Metal ion separation
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
surfaces, electrochemical probes, sensors, solid phase extractants,
etc. [10–15].
The design of Schiff bases, their reduced derivatives and the
related complexes represents a relevant and qualified goal in sev-
eral fundamental and applied aspects of science [1–3]. Remarkable
efforts are currently aimed at the preparation, characterization and
application of new Schiff bases and related complexes. Further-
more, a renewed interest persists in a deeper evaluation of possi-
ble, diversified applications of the already acquired systems.
The polydentate nature makes Schiff bases excellent hosts
for specific metal ions and a large variety of compounds was
synthesized and successfully tested [4,5]. The presence of specific
functionalities at the periphery of the coordinating moiety of these
ligands favors the formation of quite appealing systems, capable
of selective recognitions [6], specific separation and transport
processes [7], activation and catalysis [8] or surface modification
when grafted on appropriate solid platforms [9].
Silica gel can be easily functionalized after activation by reflux-
ing in 10% HCl in order to increase the surface Si–OH groups
[16–18]. The resulting material is prone to react with compounds
of the type (OR)3Si–(CH2)3–X to form the modified silica gel
(Fig. 1) which can be further derivatized according to the proper-
ties of the X group.
A Schiff base bearing a carboxylic group at the periphery of the
coordinating moiety, can be grafted on such modified silica by
esterification (X = Cl, 3-chloropropyltriethoxysilane modified sil-
ica) or amidation (X = NH2, 3-aminopropyltriethoxysilane modified
silica). These covalently bound groups are resistant to removal
from the surface by organic solvents or water. Recent research fo-
cused on the preparation of functionalized amorphous silicas since
these systems were expected to exhibit a specific binding ability
toward highly toxic heavy metals ions such as Hg(II), Pb(II) and
Cd(II) as a consequence of a soft Lewis acid–soft Lewis base
interaction.
The contamination of water sources by heavy metal pollutants,
like cadmium, lead or mercury, is an important environmental
concern and their removal from contaminated water is required
before discharging it into the environment. Several materials such
as activated carbons [19–23], biomasses [24,25], resins [26,27],
clays [28] and silica [29] have been studied for the adsorption of
heavy metal ions from water. Moreover, modification of
Among the various materials considered as potential solid sup-
ports (metals, oxides, organic polymers), silica gel was widely
studied because of its relative inertness at pH < 10, high mechani-
cal properties and ability to be organically functionalized via its
surface-located silanol groups. Quite promising hybrid systems
have been developed to create new devices, catalysts, modified
⇑
Corresponding author. Fax: +39 0498295951.
E-mail address: sergio.tamburini@cnr.it (S. Tamburini).
0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2013.10.005

30
A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
(0.761 g, 5.00 mmol) in methanol (15 mL), gave a red mixture
X
which was refluxed for 1 h. The resulting red precipitate was fil-
tered, washed with methanol and dried in vacuo.
(CH2)3
O
Yield: 0.570 g (27%). Anal. Calc. for C23H20N2O6: C, 65.71; H,
4.79; N, 6.66. Found: C, 65.82; H, 4.71; N, 6.77%. IR (KBr pellet,
Si
O
cmꢀ1): 1687 ( C@N). 1H NMR ([D6]dmso, ppm):
m C@O); 1624 (m
O
12.98 (s, 1H, C(14)COOH), 12.82 (s, 1H, C(2)OH), 12.61 (s, 1H,
C(20)OH), 9.02 (s, 1H, C(7)H@N), 8.97 (s, 1H, C(70)H@N), 7.96 (s,
1H, C(12)H), 7.94 (d, 3J = 8.1 Hz, 1H, C(10)H), 7.53 (d, 3J = 8.1 Hz,
1H, C(9)H), 7.35–7.09 (m, 2H, C(6)H, C(60)H), 6.96–6.86 (m, 4H,
C(40)H, C(4)H, C(5)H, C(50)H), 3.82 (s, 3H, C(3)OCH3), 3.81 (s, 3H,
C(30)OCH3). ESI/MS (m/z): 421 ([H3L2+H]+, 100%), 443 ([H3L2+Na]+,
20%).
Fig. 1. Structure of the surface of functionalized silica.
adsorbents with organic functional groups has been shown to im-
prove the capacity and selectivity of the adsorbents for binding de-
sired metal ions and has recently drawn much attention [30–48].
The present work deals with the preparation of some Schiff
bases and their reduced derivatives, their ability to coordinate
Hg(II), Pb(II) or Cd(II) and their linkage on modified silica. The abil-
ity of these materials to secure the above mentioned metal ions
from water solution was investigated. The role of peripheral func-
tionalization on the formation and stability of metal complexes is
also discussed.
H3L3: The addition of a methanol solution (25 mL) of 2-hydro-
xy-3-ethoxybenzaldehyde (1.662 g, 10.00 mmol) to a stirred sus-
pension of 3,4-diaminobenzoic acid (0.761 g, 5.00 mmol) in
methanol (15 mL) at room temperature, afforded a brown solution
which was refluxed for 2 h, then evaporated to dryness. The result-
ing oil was suspended in chloroform (25 mL) and the obtained pre-
cipitate was filtered, washed with chloroform and dried in vacuo.
Yield: 0.415 g (18%). Anal. Calc. for C25H24N2O6: C, 66.95; H,
5.39; N, 6.25. Found: C, 66.96; H, 5.75; N, 6.25%. IR (KBr pellet,
cmꢀ1): 1686 ( C@N). 1H NMR ([D6]dmso, ppm):
m C@O); 1626 (m
2. Experimental
13.26 (s, 1H, C(14)COOH), 12.85 (s, 1H, C(2)OH), 12.64 (s, 1H,
C(20)OH), 9.02 (s, 1H, C(7)H@N), 8.93 (s, 1H, C(70)H@N), 8.12 (s,
1H, C(12)H), 7.89 (d, 3J = 8.2 Hz, 1H, C(10)H), 7.70 (d, 3J = 8.2 Hz,
1H, C(9)H), 7.73–7.57 (m, 2H, C(6)H, C(60)H), 6.97–6.87 (m, 4H,
2.1. Materials and measurements
3
C(40)H, C(4)H, C(5)H, C(50)H), 4.08 (q, J = 6.9 Hz, 4H, C(3,30)OCH2-
The amine and formyl precursors 3,4-diaminobenzoic acid
(DBA), salicylaldehyde (Sal), 2-hydroxy-3-methoxybenzaldehyde
(Metiob), 2-hydroxy-3-ethoxybenzaldehyde (Etiob), 1,2-diamino-
benzene, the reagents for silica functionalization 3-aminopropyl-
triethoxysilane (NH2Ps), 3-chloropropyltriethoxysilane (ClPs),
N-(3-Dimethylaminopropyl)-N0-ethylcarbodiimide hydrochloride
(EDC), N-methylmorpholine (NMM) and metal salts mercury chlo-
ride, lead nitrate, cadmium chloride and zinc chloride were Sigma–
Aldrich products and used without further purification.
Elemental analyses were carried out by using a Fison 1108 ana-
lyzer. IR spectra were recorded as KBr pellets with a Mattson FTIR
spectrometer and as powders with a Nicolet Magna-560 equipped
with a Continuum Spectra-Tech microscope. NMR spectra (1H, 13C)
were recorded with a Bruker AMX300 spectrometer equipped with
direct and inverse broad-band multinuclear probes. Mass spectro-
metric measurements were performed on an ESI-MS spectrometer
(ThermoFisher LCQ-Fleet) with methanolic solutions of the sam-
ples (10ꢀ5 to 10ꢀ6 M). SEM–EDS measurements were performed
on a Fei-Esem FEI Quanta 200 FEG instrument, equipped with a
field emission gun. Mercury concentrations were determined by
ICP-AES using a Thermo Jarrell Ash Iris Advantage spectrometer.
3
CH3), 1.35 (t, J = 6.9 Hz, 3H, C(3,30)OCH2CH3). ESI/MS (m/z): 499
([H3L3+H]+, 100%).
H2L4: A methanol solution (10 mL) of 1,2-diaminobenzene
(0.541 g, 5.00 mmol) and salicylaldehyde (1.4 mL, 13.10 mmol),
stirred at room temperature for 1 h, separated a yellow solid which
was filtered, washed with methanol and diethyl ether and dried in
vacuo.
Yield: 1.230 g (78%). Anal. Calc. for C20H16N2O2: C, 75.93; H,
5.10; N, 8.86. Found: C, 73.46; H, 4.97; N, 8.21%. m.p. 138 °C. IR
(KBr pellet, cmꢀ1): 1614 (
m
C@N). 1H NMR ([D6]dmso, ppm):
12.92 (s, 2H, C(2)OH, C(20)OH), 8.91 (s, 2H, C(7)H, C(70)H), 7.64 (d,
3J = 7.9 Hz, 2H, C(9)H, C(12)H), 7.48–7.39 (m, 6H, C(4)H, C(40)H,
C(5)H, C(50)H, C(10)H, C(110)H), 7.00–6.95 (m, 4H, C(3)H, C(30)H,
C(6)H, C(60)H). ESI/MS (m/z) 317 ([H2L4+H]+, 100%), 655
([(H2L5)2+Na]+, 15%).
H3L5: To a methanol suspension (15 mL) of H3L1 (1.009 g,
2.77 mmol) in an autoclave vessel, palladium catalyst (Pd on car-
bon, 10 wt%, Aldrich) was added and the reduction was run at a
H2 pressure of 50 bar and at 90 °C for 24 h with continuous stirring.
The resulting reaction mixture was filtered to remove the catalyst
and the solution was reduced to a small volume (ꢁ5 mL). The pre-
cipitate formed by slow addition of diethyl ether was filtered,
washed with diethyl ether and dried in vacuo.
2.2. Synthesis of the ligands
H3L1: The orange Schiff base H3L1 was prepared according to lit-
erature by [2+1] condensation of salicylaldehyde with 3,4-diami-
nobenzoic acid [49].
Yield: 0.580 g (57%). Anal. Calc. for C21H20N2O4: C, 69.22; H,
5.53; N, 7.69. Found: C, 66.66; H, 4.97; N, 7.74%. 1H NMR
([D6]dmso, ppm): 7.15 (s, 1H, C(12)H), 7.09 (d, 3J = 7.9 Hz, 1H,
3
Anal. Calc. for C21H16N2O4: C, 69.99; H, 4.48; N, 7.77; O, 17.76.
C(10)H), 7.03 (d, J = 7.5 Hz, 2H, C(6)H, C(6)H0), 6.97 (t, 3J = 7.7 Hz,
3
Found: C, 69.37; H, 4.12; N, 7.65%. IR (KBr pellet, cmꢀ1): 1684 (
m
2H, C(4)H, C(4)H0), 6.76 (d, J = 7.9 Hz, 2H, C(3)H, C(3)H0), 6.67 (t,
C@O); 1616 (
m
C@N). 1H NMR ([D6]dmso, ppm): 13.09 (s, 1H,
3J = 7.3 Hz, 2H, C(5)H, C(5)H0), 6.50 (d, 3J = 7.9 Hz, 1H, C(9)H), 2.11
(s, 4H, C(7)H). ESI/MS (m/z) 365 ([H3L5+H]+, 100%).
C(14)COOH), 12.79 (s, 1H, C(2)OH), 12.58 (s, 1H, C(20)OH), 9.04 (s,
1H, C(7)H@N), 8.97 (s, 1H, C(70)H@N), 7.98 (s, 1H, C(12)H), 7.95
H2L6: To a suspension of H2L4 (0.650 g, 2.05 mmol) in methanol
(20 mL), cooled in an ice bath, NaBH4 (0.400 g, 10.51 mmol) was
slowly added under stirring in about 1 h. The resulting pale yellow
solution was stirred until gas evolution ceased. The addition of
water (20 mL) under vigorous stirring separated a white solid
which was filtered, washed with water and dried under vacuum.
Yield: 0.380 g (58%). Anal. Calc. for C20H20N2O2: C, 74.98; H,
6.29; N, 8.74. Found: C, 73.64; H, 6.37; N, 8.54%. m.p. 98 °C. IR
3
(d, J = 8.6 Hz, 1H, C(10)H), 7.74–7.70 (m, 2H, C(3)H, C(30)H), 7.53
(d, 3J = 8.2 Hz, 1H, C(9)H), 7.47–7.40 (m, 2H, C(6)H, C(60)H),
7.02–6.95 (m, 4H, C(40)H, C(4)H, C(5)H, C(50)H). ESI/MS (m/z): 361
([H3L1+H]+, 100%).
H3L2: The addition of a methanol solution (25 mL) of 2-hydro-
xy-3-methoxybenzaldehyde (2.282 g, 15.00 mmol) at room
temperature to a stirred suspension of 3,4-diaminobenzoic acid

A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
31
(KBr pellet, cmꢀ1): 3291 (s,
v
N–H), 1237 (s,
v
C–O). 1H NMR ([D6-
v
C@N), 1388 (s,
v
NO3). 1H NMR ([D6]dmso, ppm): 12.89 (s, 1H,
]dmso, ppm): 9.48 (s, 2H, C(2)OH, C(2)OH), 7.17 (d, 3J = 7.5 Hz, 2H,
OH), 8.43 (s, 2H, C(7)H@N, C(70)H@N), 7.88–7.85 (m, 3H, C(9)H,
C(10)H, C(12)H), 7.64–7.61 (, m, 2H C(6)H, C(60)H), 6.91–6.68 (m,
2H, C(4)H, C(40)H), 6.56–6.50 (m, 2H, C(50)H, C(5)H), 4.17 (q, 4H,
3
C(6)H, C(60)H), 7.01 (t, J = 7.6 Hz, 2H, C(4)H, C(40)H), 6.78 (d,
3J = 7.9 Hz, 2H, C(10)H, C(11)H), 6.70 (t, 3J = 7.4 Hz, 2H, C(5)H,
C(50)H), 6.47–6.38 (m, 4H, C(3)H, C(30)H, C(9)H, C(12)H), 5.06–
3J = 6.7 Hz, C(3,30)OCH2CH3), 1.48 (6H, t, J = 6.7 Hz, C(3,30)OCH2-
CH3). EDS: Pb 100%.
3
3
5.02 (m, 2H, NH), 4.20 (d, J = 4.9 Hz, 4H, C(7)H, C(70)H). ESI/MS
(m/z) 321 ([H2L6+H]+, 100%), 663 ([(H2L6)2+Na]+, 30%).
2.3. Template synthesis of [Cd3(Ln)2] (n = 1, 2)
2.5. Direct synthesis of [Hg3(L4)2(Cl)2]
General method: A methanol solution (30 mL) of 3,4-diamino-
benzoic acid (0.152 g, 1.00 mmol) and hydrated cadmium chloride
(1.00 mmol) was added drop wise at room temperature over
10 min to a methanol solution (20 mL) of the appropriate aldehyde
(2.00 mmol) containing NaOH (0.160 g, 4.00 mmol). The resulting
colored suspension was refluxed for 6 h, then allowed to stand
overnight. The precipitate was collected by filtration, washed three
times with methanol (10 mL) and dried in vacuo.
Anhydrous HgCl2 (0.033 g, 0.12 mmol) and NaOH (0.010 g,
0.24 mmol) were added to a methanol solution (10 mL) of H2L4
(0.038 g, 0.12 mmol). The reaction mixture was stirred at room
temperature for 2 h and the resulting orange precipitate was fil-
tered, washed three times with methanol and vacuum dried.
Anal. Calc. for [Hg3(C20H16N2O2)2(Cl)2]ꢂ3H2O: C, 36.29; H, 2.89;
N, 4.23. Found: C, 35.23; H, 2.94; N, 4.25%. IR (KBr pellet, cmꢀ1):
v
C@N). 1H NMR ([D6]dmso, ppm): 8.72 (s, 2H,
Yield of [Cd3(L1)2]ꢂ(NaCl): 0.737 g (19%), red-orange solid. Anal.
Calc. for [Cd3(C21H13N2O4)2]ꢂ4(H2O)ꢂ(NaCl): C, 42.67; H, 2.90; N,
4.74. Found: C, 42.91; H, 2.87; N, 4.67%. IR (KBr pellet, cmꢀ1):
1606 (s,
C(7,70)H@N), 7.63–7.60 (m, 2H, C(10)H, C(11)H), 7.41–7.35 (m,
4H, C(5)H, C(50)H, C(9)H, C(12)H), 7.22 (t, 3J = 7.4 Hz, 2H, C(4)H,
3
C(40)H), 6.69 (d, J = 8.4 Hz, 2H, C(6)H, C(60)H), 6.52 (dd, 2H,
1609 (s,
v
C@N). 1H NMR ([D6]dmso, ppm): 8.65 (s, 1H,
3J = 7.4 Hz, C(3)H, C(30)H). EDS: Hg:Cl = 3:2.
C(7)H@N), 8.64 (s, 1H, C(70)H@N), 8.07 (s, 1H, C(12)H), 7.90 (d,
3J = 7.7 Hz, 1H, C(9)H), 7.60 (1H, d, 3J = 7.7 Hz,C(10)H), 7.39–7.30
(2H, m, C(3)H, C(30)H), 7.17–7.11 (2H, m, C(6)H, C(60)H), 6.60–
6.57(2H, m, C(40)H, C(4)H), 6.41–6.35 (2H, m, C(5)H, C(50)H). EDS:
Cd:Na:Cl = 3:1:1. ESI/MS (m/z) 453 [(L1)Cd-OH+H]+.
2.6. Template synthesis of [Zn3(HL3)2(Cl)2]
Yield of [Cd3(L2)2]ꢂ(NaOH)1.5: 0.780 g (32%), orange solid. Anal.
Calc. for [Cd3(C23H17N2O6)2]ꢂ(NaOH)1.5: C, 44.85; H, 2.90; N, 4.55.
Found: C, 45.60; H, 2.97; N, 4.33%. IR (KBr pellet, cmꢀ1): 1611 (s,
A
mixture of 2-hydroxy-3-ethoxybenzaldehyde (1.272 g,
7.66 mmol), 3,4-diaminobenzoic acid (0.582 g, 3.83 mmol) in THF
(20 mL) and a 0.5 M solution of [ZnCl2(THF)2] in tetrahydrofuran
(15.3 mL, 7.66 mmol) was refluxed for 1 h. Removal of tetrahydro-
furan from the resulting dark green solution and addition of meth-
anol gave [Zn3(HL3)2(Cl)2] as a nearly insoluble yellow solid
(Scheme 2).
v
C@N), 1537 (s). 1H NMR ([D6]dmso, ppm): 8.64 (s, 1H,
C(7)H@N), 8.63 (s, 1H, C(70)H@N), 8.00 (s, 1H, C(12)H), 7.81 (d,
3J = 5.8 Hz, 1H, C(9)H), 7.52 (d, 3J = 5.9 Hz, 1H, C(10)H), 6.98–6.92
(m, 2H, C(6)H, C(60)H), 6.78–6.74 (m, 2H, C(40)H, C(4)H), 6.33–
6.28 (m, 2H, C(5)H, C(50)H), 3.61 (s, 6H, C(3)OCH3, C(30)OCH3).
EDS: Cd:Na = 2:1. ESI/MS (m/z) 1173 [Cd3(L2)2]ꢀ.
Yield: 2.400 g (95%), yellow solid. Anal. Calc. for [Zn3(C25H22N2-
O6)2(Cl)2]ꢂ9H2O: C, 44.31; H, 4.46; N, 4.13. Found: C, 44.77; H, 4.46;
N, 3.93%. IR (KBr pellet, cmꢀ1): 1617 (s, C@N). 1H NMR ([D6]dmso,
v
2.4. Direct synthesis of [Pb3(Ln)2] (n = 1, 2, 3)
ppm): 13.14 (s, 1H, C(14)COOH), 9.10 (s, 1H, C(7)H@N), 9.10 (s, 1H,
C(70)H@N), 8.38 (s, 1H, C(12)H), 8.01 (d, 3J = 8.5 Hz, 1H, C(9)H), 7.92
(d, 3J = 8.5 Hz, 1H, C(10)H), 7.16–7.06 (m, 2H, C(6)H, C(60)H), 6.90–
6.88 (m, 2H, C(40)H, C(4)H), 6.45–6.39 (m, 2H, C(5)H, C(50)H), 4.10
(q, 3J = 6.8 Hz, 4H, C(3,30)OCH2CH3), 1.38 (t, 3J = 6.8 Hz, 6H,
C(3,30)OCH2CH3). EDS: Zn:Cl = 3:2. ESI/MS (m/z): 1123
[Zn3(HL3)2(Cl)]+.
General method: A solution of the desired ligand (1.00 mmol) in
methanol (10 mL) was added drop wise at room temperature over
10 min to a methanol solution (10 mL) of Pb(NO3)2 (0.331 g,
1.00 mmol) and NaOH (0.080 g, 2.0 mmol). The resulting suspen-
sion was refluxed for 6 h, and then allowed to stand overnight.
The solid was collected by filtration, washed three times with
methanol (10 mL) and dried in vacuo.
Yield of [Pb3(L1)2]ꢂ(NaOH): 0.670 g (15%), orange solid. Anal.
Calc. for [Pb3 (C21H13N2O4)2]ꢂ7(H2O)ꢂ(NaOH): C, 33.94; H, 2.78; N,
3.77. Found: C, 33.54; H, 2.06; N, 3.61%. IR (KBr pellet, cmꢀ1):
2.7. Transmetalation synthesis of [Zn2Cd(HL3)2(Cl)2]
1613 (s,
v
C@N). 1H NMR ([D6]dmso, ppm): 8.46 (s, 1H,
CdCl2 (0.011 g, 0.06 mmol) in methanol (5 mL) was added to a
solution of [Zn3(HL3)2(Cl)2] (0.058 g, 0.05 mmol) in THF (15 mL).
The yellow starting solution, turned into deep red, was stirred for
16 h at room temperature then was evaporated to dryness. The res-
idue was treated with methanol (20 mL) and the resulting red solid
was collected by filtration, washed twice with methanol and dried
in vacuo.
C(7)H@N), 8.45 (s, 1H, C(70)H@N), 8.30 (s, 1H, C(12)H), 7.88 (d,
3J = 7.9 Hz, 1H, C(9)H), 7.81 (d, 3J = 8.0 Hz, 1H, C(10)H), 7.41–7.37
(m, 2H, C(3)H, C(30)H), 7.33–7.26 (m, 2H, C(6)H, C(60)H), 6.56–
6.54 (m, 2H, C(40)H, C(4)H), 6.45–6.40 (m, 2H, C(5)H, C(50)H).
EDS: Pb:Na = 3:1. ESI/MS (m/z) 971 [(HL1)2Pb+2Na+H]+.
Yield of [Pb3(L2)2]: 0.815 g (17%), orange solid. Anal. Calc. for
[Pb3(C23H17N2O6)2]ꢂ6(H2O): C, 35.32; H, 2.96; N, 3.58. Found: C,
Yield: 0.045 g (21%), yellow solid. Anal. Calc. for [Zn2Cd(C25H22-
N2O6)2(Cl)2]ꢂ2H2O: C, 47.05; H, 3.63; N, 4.39. Found: C, 46.54; H,
35.16; H, 2.50; N, 3.49%. IR (KBr pellet, cmꢀ1): 1611 (s,
v C@N).
1H NMR ([D6]dmso, ppm): 8.57 (s, 2H, C(7)H@N, C(70)H@N), 8.01
(s, 1H, C(12)H), 7.95 (d, 3J = 7.3 Hz, 1H, C(9)H), 7.62 (d, 3J = 7.3 Hz,
1H, C(10)H), 7.46–7.43 (m, 2H, C(6)H, C(60)H), 7.08–6.96 (m, 2H,
C(40)H, C(4)H), 6.49–6.44 (m, 2H, C(5)H, C(50)H), 3.87 (s, 6H,
C(3)OCH3, C(30)OCH3). EDS: Pb 100%.
3.56; N, 4.52%. IR (KBr pellet, cmꢀ1): 1618 (s, C@N). 1H NMR ([D6-
v
]dmso, ppm): 13.14 (s, 1H, C(14)COOH), 9.10 (s, 1H, C(7)H@N), 9.08
(s, 1H, C(70)H@N), 8.36 (s, 1H, C(12)H), 8.01 (d, 3J = 8.5 Hz, 1H,
C(9)H), 7.92 (d, 3J = 8.5 Hz, 1H, C(10)H), 7.16–7.06 (m, 2H, C(6)H,
C(60)H), 6.90–6.88 (m, 2H, C(40)H, C(4)H), 6.46–6.40 (m, 2H,
Yield of [Pb3(HL3)(L3)](NO3): 0.984 g (21%), pale yellow solid.
Anal. Calc. for [Pb3C50H43N5O15]ꢂ(H2O): C, 35.48; H, 2.98; N, 4.97.
Found: C, 37.69; H, 2.75; N, 4.39%. IR (KBr pellet, cmꢀ1): 1619 (s,
C(5)H, C(50)H), 4.06 (q, J = 5.4 Hz, 4H, C(3,30)OCH2CH3), 1.40 (t,
3
3J = 6.4 Hz, 6H, C(3,30)OCH2CH3). EDS: Zn:Cl = 1:1, Cd/Cl = 1:3, Zn/
Cd = 2:1. ESI/MS (m/z): 1171 [Zn2Cd(HL3)2(Cl)]+.

32
A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
2.8. Synthesis of 3-chloropropyl (SiClPs) and 3-aminopropyl
(SiNH2Ps) functionalized silicas
washed with distilled water and dried in vacuo. The amount of me-
tal ions on the silica was analyzed by SEM–EDS. The 30 min contact
time was determined in a previous series of experiments to be
enough to ensure the equilibrium achievement.
Silica gel (45.000 g, Aldrich, type 60, 70–230 mesh) was acti-
vated by 2 h refluxing in water/conc. HCl 1:1, thoroughly washed
with water until neutral pH, then washed with methanol and final-
ly dried in vacuo. Subsequently, in a glove box, 10.000 g of acti-
vated silica was weighed and placed in a glass bottle which was
then filled with toluene and the selected silylating agent (3-chloro-
propyltriethoxysilane or 3-aminopropyltriethoxysilane, 2 mmol/g)
was added. The resulting suspension was heated close to the boil-
ing point and stirred for three days, then filtered and the resulting
functionalized silica was washed several times with ethanol and
methanol and dried in vacuo.
2.12. Mercury adsorption isotherm measurement
About 0.075 g of functionalized silica were weighed on an ana-
lytical balance and placed in a vial with 50 mL of a Hg2+ solution
buffered at pH 3 with a 0.5 M acetic acid/sodium acetate solution.
After 30 min stirring the suspension was filtered and mercury
concentration in the solution was measured by ICP-AES. Mercury
adsorbed by functionalized silica was determined by the difference
between the concentrations before and after contact with the
adsorbent.
SiClPs: Anal. C, 4.47; H, 1.14%; EDS: Si/Cl = 11.6. Elemental anal-
ysis gave a chloropropyl loading of 1.24 mmol/g (EDS data gave
1.29 mmol/g). In subsequent calculations we assumed a value of
1.25 mmol/g.
3. Results and discussion
SiNH2Ps: Anal. C, 6.30; H, 1.86; N, 1.71%. Amino group back
titration with HCl and NaOH 0.1 N gave a loading of 1.13 mmol/g.
3.1. The ligands
2.9. Preparation of Schiff-base bearing silicas SiNCOH2L1 and
The synthesized ligands (Fig. 2) share a similar structural core,
while changing the peripheral functionalization to explore its
influence in the chelating properties towards Cd(II), Pb(II) and
Hg(II).
SiNCOH2L2
To a suspension of H3L1 (7.531 g, 20.90 mmol) or H3L2 (0.765 g,
1.82 mmol), in dichloromethane (respectively 375 and 25 mL), EDC
(respectively 21.00 and 1.83 mmol) and NMM (42.00 and
3.84 mmol) were added. To the resulting clear orange solution,
SiNH2Ps (9.000 and 0.800 g, corresponding to 10.2 and 0.91 mmol
of free amino groups respectively) was slowly added and the sus-
pension stirred for 24 h at room temperature in an apparatus
closed with a calcium chloride drying tube. Then, a Kaiser test
[50] was performed to check the presence of free amino groups.
If positive, another equivalent of EDC was added and the suspen-
sion further stirred for 2 h. Finally the functionalized silica was
recovered by filtration, washed with dichloromethane, methanol
and water, then treated in a Soxhlet extractor with ethanol for
8 h and dried in vacuo.
The Schiff bases ligands (H3L1, H3L2 H3L3, H2L4) and the reduced
derivatives of H3L1 and H2L4 (H3L5, H2L6), were synthesized to
compare their coordination ability towards Cd(II), Pb(II) and Hg(II)
metal ions. Four ligands bear a peripheral carboxylic group, which
was successfully engaged in the reaction with 3-chloropropyl or
3-aminopropyl modified silicas to prepare the related ligand bear-
ing silicas. The materials derived from the reaction of H3L1 or H3L5
with 3-chloropropyl silica were selected as prototypes of these
chelating systems and hence were carefully characterized, with a
particular attention to their ability to sequestrate metal ions from
aqueous solutions.
The ligands H2L4 and H2L6, without the carboxylic functional
group, were synthesized to mimic the coordination behavior of
H3L1 and H3L5 when immobilized on silica substrate by the amido
or ester bond.
SiNCOH2L1: Anal. C, 12.33; H, 1.94; N, 2.46%. Free amino-groups
back titration gave a ligand loading of 0.62 mmol/g.
SiNCOH2L2: Anal. C, 11.80; H, 1.38; N, 2.23%.
3.2. Ligands formation and their reactions with cadmium(II),
lead(II)and mercury(II) salts
2.10. Preparation of Schiff-base bearing silicas SiOCOH2L1 and
SiOCOH2L4
The condensation in alcoholic solution of 3,4-diaminobenzoic
acid or 1,2-diaminobenzene with the appropriate aldehyde forms
the Schiff bases of Fig. 2 as yellow (H3L1), orange (H2L4) or red
(H3L2, H3L3) solids, stable in air and soluble in the most common
organic solvents. H3L5 and H2L6, obtained by the reduction of the
related Schiff bases H3L1and H2L4 via catalytic hydrogenation with
Pd/C and hydrogen or with sodium borohydride, were isolated as a
light brown and white solid, respectively, both stable in air and
soluble in methanol. The ESI mass spectra of all the synthesized
ligands show the parent peaks at the appropriate m/z value that
corroborate the structures proposed in Scheme 1.
The addition of sodium hydroxide (0.200 g, 5.00 mmol) to H3L1
(1.802 g, 5.00 mmol) or H3L5 (1.822 g, 5.00 mmol) in 70 mL of
isopropanol/dimethylformamide (90:10) formed an orange and
yellow solution respectively. By reaction with SiClPs (2.000 g,
corresponding to 2.50 mmol of chloropropyl groups) and reflux
of the resulting suspensions for 24 h afforded the functionalization
of the silica, which was recovered by filtration on a gooch crucible
and washed with DMF, water and methanol, then treated in a
Soxhlet extractor with ethanol for 8 h and finally dried in vacuo.
SiOCOH2L1: pale yellow powder. Anal. C, 5.25; H, 0.91; N, 0.06%;
EDS: Si:Cl = 12.5.
The synthesized Schiff bases bearing a carboxylic group (H3Ln,
n = 1,2,3) are potentially tetradentate and compartmental hexa-
dentate ligands; they react with lead(II) nitrate in alcoholic solu-
tion to give the complexes [Pb3(L1)2]ꢂ(NaOH), [Pb3(L2)2]
(Scheme 1) and [Pb3(HL3)(L3)]NO3.
SiOCOH2L4: light brown powder. Anal. C, 5.81; H, 0.93; N, 0.16%;
EDS: Si:Cl = 13.1.
2.11. Metal adsorption preliminary screening
On the other hand the related cadmium(II) complexes have
been obtained by the template procedure (Scheme 1).
An aqueous solution (5 mL) of Hg, Pb and Cd nitrate (each in a
0.01 M concentration) was treated in a vial with 0.050 g of the de-
sired ligand-bearing silica and 1 mL of a pH 5 or pH 3 acetic acid/
sodium acetate buffer solution (both 0.5 M), then stirred for
30 min. The solid silica was recovered by filtration, thoroughly
Under similar conditions HgCl2 and H2L4 afford [Hg3(L4)2(Cl)2]
(Scheme 1). The tetradentate nature of the inner N2O2 chambers
of the compartmental ligands favors the coordination of metal
ions, as lead(II) or cadmium(II), in this compartment. Although

A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
33
R
3
3
HO
HO
4
4
5
2
2
9
9
H
N
N
N
5
8
8
10
1
10
1
7
6
7
6
6'
6'
1'
1'
11
13
7'
5'
4'
11
13
7'
5'
4'
R'
14
R'
N
H
12
12
2'
2'
HO
HO
3'
3'
R
R'
COOH
COOH
COOH
H
R
H
OMe
OEt
H
R'
COOH
H
H3L1
H3L2
H3L3
H2L4
H3L5
H2L6
Fig. 2. The synthesized ligands.
CHO
2
NH2
NH2
HOOC
NH2
NH2
OH
R
MeOH
room temperature
(R=H)
CH2Cl2
room temperature
R
N
N
OH
OH
HOOC
N
N
OH
OH
HOOC
NH2
NH2
CdCl2, NaOH
MeOH
reflux
Pb(NO3)2, NaOH
R
MeOH
reflux
HgCl2, NaOH
R
R
MeOH
room temperature
O
O
O
N
N
O
O
O
N
N
M
C
C
M
O
M
(Na X)y
O
R
R
M
R
X
y
1
0
1
Cl
N
N
O
O
O
N
Hg
Hg
Hg
Pb
Pb
Cd
Cd
H
OH
OMe --
Cl
Cl
O
N
H
OMe OH 1.5
Scheme 1. Schematic representation of the complex formation.

34
A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
the external O2O2 chamber could accommodate a second metal
ion, on the basis of the experimental evidences discussed hereinaf-
ter, the most probable structure of the complexes formed in the
presence of NaOH, involves a coordination of the third central me-
tal ion to the carboxylate groups of two ligand moieties, while the
ethereal oxygens, when present, weakly coordinate a sodium ion.
These complexes, sparingly soluble in many organic solvents
except DMSO, were characterized by elemental analysis, IR, 1H
NMR and ESI-MS. Their NMR spectra, performed in d6-DMSO, show
the typical patterns associable to the formation of a M(II)-centered
salophen complex.
spectrum of the free Schiff bases also shifted to considerably lower
frequency (1585–1540 and 1465–1445 cmꢀ1) in all the metal com-
plexes, indicating carboxylic oxygen deprotonation and its involve-
ment in metal coordination, possibly as chelating ligands. This is
also confirmed by the 1H NMR, where the chemical shift of the car-
boxylic proton (about 13 ppm in the ligands spectrum) disappears.
New weak bands not detected in the spectrum of ligands but pres-
ent in the spectra of complexes at 505–562 and 414–436 cmꢀ1, are
assignable respectively to m(M–N) and m(M–O) vibrations, support-
ing the involvement of nitrogen and oxygen atoms in complexation
to the metal ions.
A comparison of the 1H NMR spectra of the complexes with
those of the parent Schiff bases shows that the OH phenolic signal,
occurring in the spectrum of the ligands in the range 12.5–
12.8 ppm, completely disappears in the spectrum of their metal
complexes, indicating that the OH proton is removed by coordina-
tion with the metal ions. In addition, the imino protons bands are
significantly shifted, due to coordination through the azomethine
nitrogen atom of the ligand. Furthermore, involvement of the ethe-
real oxygens in coordination is confirmed by the slight shift of the
OCH2- and OCH3 signals in the complexes compared to the corre-
sponding chemical shifts in the ligands.
The IR spectra of the Schiff bases and the corresponding com-
plexes provided further information about the metal–ligand bond-
ing. The weak broad band at 3250–3352 cmꢀ1, due to a hydrogen
bound –OH group in the free Schiff base, disappeared in the infra-
red spectra of the complexes, indicating deprotonation of the phe-
nolic oxygen and coordination in the complexes (Scheme 1).
The ESI mass spectra of the ligands and their metal complexes
are in good agreement with the structures suggested by elemental
analysis and spectroscopic studies. The ligands showed the molec-
ular ion peaks which correspond to the [L+H]+ or [L+Na]+ species;
for instance, in the mass spectrum of [Cd3(L2)2ꢂ(NaOH)1.5], the
occurrence of the parent peak at m/z = 1173 and its isotopic pat-
tern (Fig. 3) are consistent with the respective deprotonated com-
pound supporting the {M3(L)2} trinuclear nature of the complex.
3.3. Transmetalation of trinuclear Zn(II) Salophen complex
Due to the difficulty in synthesizing [Cd3(L3)2], a transmetala-
tion reaction was tested. To this purpose [Zn3(HL3)2(Cl)2] was
firstly prepared by a template synthesis method with ZnCl2, 3,4-
diaminobenzoic acid and 3-ethoxysalicylaldehyde (see Section 2).
Both ESI/MS spectrometry (showing a peak at m/z = 1123 corre-
sponding to [Zn3(HL3)2(Cl)]+) and EDS analyses (yielding a Zn:Cl
ratio of 3:2) agree with the proposed trinuclear formulation. The
Zn3 complex formation is corroborated also by the 1H NMR spectra
in d6-DMSO, where the signals of the imino protons (9.10 ppm) are
downfield shifted respect to ligand signals (9.02–8.93 ppm). On the
Furthermore, the strong
(1619–1609 cmꢀ1) lay at considerably lower frequencies compared
to that of the free Schiff bases
(C@N) (1636–1624 cmꢀ1), further
m(C@N) bands of these complexes
m
corroborating azomethine coordination to the metal ion. The
carboxylic stretching band appearing at about 1700 cmꢀ1 in the
other hand, the IR spectrum shows the presence of a m(C@O) band
Fig. 3. ESI mass spectrum (a) of [Cd3(L2)2ꢂ(NaOH)1.5] with the detailed pattern (b) and MS simulation (c) of the [Cd3(L2)2]ꢀ species.

A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
35
COOH
HOOC
N
O
N
O
O
O
O
N
N
O
N
O
Zn
Zn
CHO H2N
OH H2N
COOH
Cl
Cd
Cl
O
O
1. [ZnCl2(thf)2]
thf, 1h
2. MeOH
Cl
Zn
Cl
O
O
O
CdCl2
2
THF/MeOH
O
O
O
N
O
N
O
N
Zn
Zn
COOH
HOOC
Scheme 2. Synthesis and transmetalation of the Zn(II)–salphen complex.
at 1710 cmꢀ1, absent in the spectra of the other Schiff bases
cadmium(II) and lead(II) trinuclear complexes, while it occurs in
the spectrum of H3L3. This indicates that the carboxylic group is
protonated and not involved in metal coordination, suggesting
the proposed structure of Scheme 2. This is a quite different
structure, but it should be noted that the Zn3 complex was synthe-
sized without the use of a base, while in the synthesis of the
previously discussed cadmium(II) and lead(II) Schiff bases com-
plexes NaOH was employed as deprotoning agent. Attempts to
synthesize Cd(II) and Pb(II) complexes via a similar base-free
method were not satisfying; lead(II) showed no reaction and
cadmium(II) gave only a small yield of a mixture of not well char-
acterized complexes.
Then, [Zn3(HL3)2(Cl)2] was engaged in the reaction with CdCl2 at
room temperature in tetrahydrofuran (Scheme 2). Although tetra-
hydrofuran usually occupies the fifth apical coordination site in
salophen complexes, therefore stabilizing the Zn(II)–complex in
solution [51], the gradual color change of the reaction mixture
from yellow to red was attributed to Cd(II)–salphen formation. In
parallel ESI/MS spectrometry showed a peak at m/z 1171 with an
isotopic pattern consistent with a [Zn2Cd(HL3)2(Cl)]+ ion, thus con-
firming the formation of a Zn2Cd aggregate. The ratio between the
intensities of the peaks corresponding to [Zn2Cd(HL3)2(Cl)]+ (m/
z = 1171) and [Zn3(HL3)2(Cl)]+ (m/z = 1123) ions was nearly stable
after 16 h. The final product was recovered in 95% yield and
SEM/EDS confirms the transmetalation, being the ratio Zn:Cd very
close to 2:1. IR and NMR data of the transmetallated complex are
quite similar to the starting material suggesting the same general
structure of the Zn3 complex.
complex, as indicated by a drastic color change of the solution,
which however is not stable a long time resulting in a reduction
of mercury(II) ion and ligand degradation. The carboxylic group
must be strongly involved in this reactivity as H2L4 which does
not bear this group can form a more stable complex.
The reaction of H2L6 with HgCl2 was more carefully investi-
gated, as several evidences indicated
a complex formation,
although any attempt to isolate a Hg(II) amino complex was unsuc-
cessful. Mixing solutions of H2L6 (pale yellow) and HgCl2 (color-
less) in methanol, at once yielded a persistent pink solution. An
NMR spectrum of this solution showed a shift of the peaks belong-
ing to the aromatic protons of the 1,2-diaminobenzene moiety,
with an extent depending on the H2L6:Hg molar ratio (which
varied from 10:1 to 1:2). Furthermore, the ESI/MS spectrum of
the solution showed peaks corresponding to the [Hg(H2L6)2(Cl)]
(m/z = 876) and [Hg(H2L6)(Cl)] (m/z = 665) species, with isotopic
patterns clearly indicating the presence of a mercury ion. All these
data supported
a mononuclear complex formation, but the
isolation of a well defined compound was not possible due to
subsequent reactions yielding an orange solution and Hg2Cl2. The
exact nature of the products of L6 degradation could not be estab-
lished although mass spectra suggested a dimerization reaction
with the formation of a species with a 634 amu mass. Also 1H
NMR TOCSY-1D experiments showed the presence of three differ-
ent aromatic spin systems, each integrating for four protons, while
the methylenic protons band at d = 4.20 ppm disappears. While in
the absence of base the degradation reaction takes several hours to
complete at room temperature, indicating a relative stability of
mercury(II) complex in these conditions, it is very fast in the pres-
ence of triethylamine (2 equivalents) as deprotoning agent; in this
case the addition of mercury salt to an alcoholic solution of H2L6
causes the color change to orange and the precipitate formation
at once.
3.4. Reactions of the Schiff bases and related reduced derivatives with
mercury(II) salts
The peculiarity of the reactions of the prepared ligands with
mercury(II) salts needed a separate, more appropriate, discussion.
Only [Hg3(L4)2(Cl)2], derived from the reaction of H2L4 with HgCl2
in the presence of NaOH as deprotonating agent, could be obtained
in a solid form and fully characterized by the usual techniques.
Complex formation was confirmed by the IR imino stretching band
shift from 1614 to 1606 cmꢀ1 on going from the free ligand to the
complex, and by 1H NMR shift of the imino signal as observed for
Cd(II) and Pb(II) complexes.
In another experiment, a suspension of H2L6 in an aqueous
solution of HgCl2 was vigorously stirred and slowly gave a pink
solid very similar to that deriving from the above procedure.
Unfortunately conversion was never quantitative and the resulting
precipitate contained a mixture of the original ligand and the pink
solid. FT-IR and NMR spectra in deuterated methanol of this solid
do not differ significantly from the spectra of the free ligand
H2L6. However this solid is quite stable even when left in its
mother liquor for several weeks. This suggested the existence of
weak interactions, which did not cause significant NMR shift of
the ligand protons, between H2L6 and Hg(II) cations, possibly via
the phenolic oxygen atoms. These interactions were easily broken
when a coordinating solvent is present, making complex isolation
quite difficult.
On the contrary, either the reaction of H3L1 with mercury chlo-
ride or the template synthesis strategy, with or without a deprot-
onating base, always resulted in ligand degradation and
formation of a precipitate containing Hg2Cl2 and sometimes metal-
lic mercury. It can be hypothesized the formation of a mercury

36
A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
3.5. Ligands anchoring on silica gel and metal adsorption preliminary
screening
primarily as a test to verify if a substantial amount of one of the
metal ions of interest could be absorbed by the materials under
investigation. Among all the metal ions mercury(II) is the one pref-
erentially or exclusively adsorbed on all the ligand bearing silicas.
We can also notice that amide binding results in a remarkably
higher yield, since ester formation by direct SN2 of a carboxylate
group on an alkyl halogenide is a quite difficult reaction. In an
apparently according way, materials obtained from SiNH2Ps
adsorbed higher quantities of mercury than materials obtained
by SiClPs functionalization, but we have to observe that the ami-
no-modified silica used as precursor is itself capable of adsorbing
mercury, while SiClPs does not. This makes impossible for SiNH2Ps
derived materials to distinguish between mercury ions adsorbed
by the ligand moieties attached on the surface and mercury ions
adsorbed by remaining free amino groups. Moreover, functionali-
zation does not increase considerably the adsorption capacity. On
the other hand, Hg absorption by SiOCOH2L1 and SiOCOH2L5 is
totally due to Schiff base coordination. For this reasons we chose
the materials obtained from SiClPs for a more complete character-
ization of the adsorption isotherm.
The ability of the Schiff bases to form complexes with Cd(II),
Pb(II) and Hg(II) justified the successive study on novel function-
alized materials, obtained by anchoring selected ligands to silica,
as potential sequestrating agents of these metal ions. Two func-
tionalized silica gels, 3-aminopropyl-modified and 3-chloropro-
pyl-modified silica gel (in the following text we will refer to
these materials as SiNH2Ps and SiClPs respectively), were taken
into account as suitable precursors to anchor the synthesized li-
gands H3L1, H3L2 and H3L5. These precursors were obtained by
treating the activated silica gel with an appropriate silylating
agent of the type (EtO)3Si(CH2)nX (X = NH2, Cl) as reported in
the experimental section and widely described in the litera-
ture.[52–54] SiNH2Ps and SiClPs were obtained with a function-
alization loading of 1.13 and 1.25 mmol/g respectively, based on
elemental analysis results.
All the selected ligands bear a peripheral carboxylic group
which can be conveniently engaged in linkage reactions with the
modified silica by mean of an amide (SiNH2Ps) or ester (SiClPs)
bond formation. Detailed procedures on anchoring reactions are
reported in the experimental section and briefly summarized in
Scheme 3. After the anchoring reaction, all the materials were care-
fully washed by treatment in a Soxhlet apparatus with ethanol for
8 h to ensure total removal of the ligand not covalently bound to
the substrate surface.[55]
Subsequently, a preliminary metal adsorption test was per-
formed by using a water solution containing all the three metal
cations, i.e. Cd(II), Hg(II) and Pb(II), each in a 10ꢀ2 M concentra-
tion. After a 30 min contact with a buffered (pH 3 or 5) aliquot
of the water solution, the solid materials were analyzed with
SEM–EDS to check for the presence of adsorbed metal. The mod-
ified silica precursors were also tested. The test solution volume
was calculated in order to have an excess of each metal cation
with respect to the estimated amount of active sites on the silica
surface, so it should be pointed out that a ligand capable of bind-
ing more than one metal species, but with significantly different
strengths, is expected to coordinate only the metal with the high-
er bond strength.
3.6. Metal adsorption isotherms
Adsorption processes are frequently described by Langmuir (Eq.
(1)) or Freundlich (Eq. (2)) isotherms [56,57]
Kd ꢂ Ce
Nmax
Ne ¼
ð1Þ
Kd
Ce þ 1
Ne ¼ Kf ꢂ C1=n
ð2Þ
e
where Ce and Ne are the equilibrium concentrations of the adsorbed
specie in the liquid phase and on the adsorbent respectively. Nmax is
the maximum capacity of the adsorbent and Kd is the distribution
coefficient and both are constants for a given temperature. Kf and
n
also are constants of the Freundlich model for
a given
temperature.
Data for the adsorption isotherms of mercury onto SiNCOH2L1
and SiOCOH2L5 were measured as described in the experimental
section. The measurements were performed at pH 3 with a con-
stant buffer concentration and a variable concentration of mercury.
An acetate buffer with a concentration of about 0.05 M was
employed. A total of 7 points was measured for each material.
Experimental points were fitted both with a Langmuir and a Fre-
undlich type function and the results are reported in Table 2.
For both the materials, capacities resulting from Langmuir fits
correspond to a maximum stoichiometry of Hg3.5Ln (n = 1, 5),
The characterization of the synthesized materials and the re-
sults of the metal adsorption test at pH 3 are reported in Table 1.
The metal adsorption data at pH 5 were systematically lower and
are not reported. Ligand loadings were calculated from carbon
and nitrogen increases in the elemental analysis.
For sake of clarity, we point out that this test was not addressed
to give a measure of the metal loading on the substrate, but
R
O
HO
Si
O
X
O
Y
Y
O
Si
O
Z
H3Ln
exp
O
O
HO
n
1
exp
X
Z
R
H
Y
N
R
EDC/NMM
SiNCOH2L1
SiNCOH2L2
SiOCOH2L1
SiOCOH2L5
NH2
NH2
Cl
NH
NH
O
CHCl2, 24h r.t.
EDC/NMM
CHCl2, 24h r.t.
2
OMe
N
N
NaOH
isopropanol/DMF 90/10
24h reflux
1
5
H
H
NaOH
isopropanol/DMF 90/10
24h reflux
NH
Cl
O
Scheme 3. Synthetic routes exploited to anchor the ligands on SiNH2Ps and SiClPs.

A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
37
Table 1
Synthesized materials and data on preliminary metal adsorption test. Under the experimental conditions no cadmium or lead adsorption was observed.
Ligand loading (mmol/g) and % yielda
SEM–EDS metal adsorption results
% Si
% Hg (Hg/Si)
Ligand-functionalized silicas
SiNCOH2L1
0.25 (22%)
0.19 (17%)
0.026 (2.1%)
0.055 (4.4%)
95.4
94.5
98.5
96.8
4.6 (0.048)
5.5 (0.058)
1.5 (0.015)
3.2 (0.033)
SiNCOH2L2
SiOCOH2L1
SiOCOH2L5
Modified silica precursors
SiNH2Ps
SiClPs
1.13
1.25
94.4
100
5.6 (0.059)
0
a
% Yield is calculated as the percentage of NH2 or Cl sites which were functionalized with a ligand moiety.
Table 2
Experimental data and Langmuir and Freundlich fittings on adsorption isotherms of mercury onto SiNCOH2L1 and SiOCOH2L5 at pH 3. The ligand loadings, calculated from
nitrogen and carbon content as determined by elemental analysis results, are also reported.
Ligand loading (mmol/g)⁄
Langmuir fit
Freundlich fit
Kd (L/g)
Nmax (mmol/g)
R2
Kf (L/g)
n
R2
SiNCOH2L1
SiOCOH2L5
0.026
0.055
0.093
0.13
0.091
0.19
0.95
0.97
0.049
0.082
1.40
1.31
0.93
0.98
meaning that, for each ligand moiety anchored on the silica
substrate, up to 3.5 mercury ions can be absorbed. This is quite
different from the complex synthesized from the free ligand were
a stoichiometry Hg1.5L4 was the preferred situation.
Acknowledgments
The authors thanks Regione Veneto for financial support (pro-
jects no 436/1/1/1103/2010 and 436/1/2/1103/2010), Dr. Claudia
Battistella for her preliminary studies, Mrs. Anna Rosa Moresco
for elemental analyses and Mr. Valerio Corrado for technical
support.
On the other hand the two systems have marked differences.
Linking the ligands to the solid surface completely modifies their
ability to coordinate a metal ion not only because the carboxylic
group is no longer available but also because this limits the space
of possible molecular conformations. In addition, silica substrate
possesses hydroxylic groups which could cooperate in metal coor-
dination. It has to be taken into account also that complex synthe-
sis in solution was accomplished with a metal to ligand ratio of 1:1,
while the maximum loadings resulting from Langmuir fittings are
a theoretical asymptote corresponding to a large (infinite) excess of
metal.
References
[1] S. Brooker, Coord. Chem. Rev. 222 (2001) 33.
[2] V. McKee, J. Nelson, R.M. Town, Chem. Soc. Rev. 32 (2003) 309.
[3] (a) P.A. Vigato, S. Tamburini, L. Bertolo, Coord. Chem. Rev. 251 (2007) 1311;
(b) P.A. Vigato, S. Tamburini, Coord. Chem. Rev. 252 (2008) 1871;
(c) W. Huang, H.-B. Zhu, S.-H. Gou, Coord. Chem. Rev. 250 (2006) 414.
[4] P.A. Vigato, S. Tamburini, Coord. Chem. Rev. 248 (2004) 1717.
[5] A.L. Gavrilova, B. Bosnich, Chem. Rev. 104 (2004) 349.
[6] S. Wang, G. Men, Y. Wang, L. Zhao, Q. Hou, S. Jiang, IEEE Sens. J. 11 (2011) 137.
[7] S.G. Galbraith, L.F. Lindoy, P.A. Tasker, P.G. Plieger, Dalton Trans. (2006) 1134.
[8] (a) A. Kumar, G.S. Mishra, A. Kauman, Transition Met. Chem. 28 (2003) 913;
(b) S. Jana, B. Dutta, R. Bera, S. Koner, Langmuir 23 (2007) 2492;
(c) S. Bhunia, R. Sen, S. Koner, Inorg. Chim. Acta 363 (2010) 3993.
[9] P. Gomez-Romero, C. Sanchez (Eds.), Functional Hybrid Materials, Wiley–VCH,
Weinheim (D), 2004.
[10] A.G.S. Prado, A.H. Tosta, C. Airoldi, J. Colloid Interface Sci. 269 (2004) 259.
[11] L.N.H. Arakaki, J.G.P. Espinola, M.G. da Fonseca, S.F. de Oliveira, A.N. de Sousa,
T. Arakaki, C. Airoldi, J. Colloid Interface Sci. 273 (2004) 211.
[12] J.A.A. Sales, F.P. Faira, A.G.S. Prado, C. Airoldi, Polyhedron 23 (2004) 719.
[13] A.R. Cestari, E.F.S. Viera, E.C.N. Lopes, R.G. da Silva, J. Colloid Interface Sci. 272
(2004) 271.
4. Conclusion
The complexation behavior of the Schiff bases, derived from the
[2+1] condensation of 3-methoxy (or 3-ethoxy)-2-hydroxybenzal-
dehyde and 2-hydroxybenzaldehyde with 3,4-diaminobenzoic acid
or 1,2-diaminobenzene, and the related amino derivatives towards
Cd(II), Pd(II) and Hg(II) indicates the formation of different com-
plexes according to the employed experimental conditions. Mainly
a [M3(Ln)2] aggregation occurs.
The reaction of the carboxylic group at the periphery of the
coordinating moiety of the Schiff bases or the reduced derivatives
with functionalized silicas, bearing NH2 or Cl groups at the end of a
pendant arm, results in ligand-bearing silicas which show a selec-
tive recognition of mercury ions from cadmium and lead ions in
water, with a capacity of up to 0.19 mmol/g, corresponding to a
metal to ligand stoichiometry of Hg3.5L.
[14] N.L. Dias Filho, Y. Gushiken, Sep. Sci. Technol. 32 (1997) 2535.
[15] A. Corma, H. Garcia, Adv. Synth. Catal. 348 (2006) 1391.
[16] W.A. Aue, C.R. Hastings, J. Chromatogr. 42 (1969) 319.
[17] L.N.H. Arakaki, C. Airoldi, Polyhedron 19 (2000) 367.
[18] C. Airoldi, M.R.M.C. Santos, J. Mater. Chem. 4 (1994) 1479.
[19] J. Chenga, K.S. Subramanianb, C.L. Chakrabartia, R. Guoa, X. Maa, Y. Lua, W.F.
Pickeringc, J. Environ. Sci. Health A 28 (1993) 51.
[20] B.E. Reed, S. Arunachalam, B. Thomas, Environ. Prog. 13 (1994) 60.
[21] K. Ranganathan, Carbon 41 (2003) 1087.
The reported results show that it is not obvious to predict the
coordinating behavior of the supported ligands, as a different
coordination behavior was observed when the same ligands are
in solution or anchored to silica. Moreover, in the heterogeneous
extraction experiments with a metal ion to ligand ratio high as
5:1 with the designed ligand linked on the solid platform in a
low amount, different metal to ligand aggregation are favored.
We are currently experimenting different linkage reactions
capable of increasing the amount of grafted ligand.
[22] C. Namasivayam, K. Kadirvelu, Carbon 37 (1999) 79.
[23] E. Ekinci, T. Budinova, F. Yardim, N. Petrov, M. Razvigorova, V. Minkova, Fuel
Process. Technol. 77/78 (2002) 437.
[24] A. Dubey, S. Shiwani, Int. J. Environ. Sci. Technol. 9 (2012) 15.
[25] C. Fehrmann, P. Pohl, J. Appl. Phycol. 5 (1993) 555.
[26] P. Rama Devi, T. Gangaiah, G.R.K. Naidu, Anal. Chim. Acta 249 (1991) 533.
[27] R. Purohit, S. Devi, Talanta 38 (1991) 753.
[28] M.L. Schlegel, A. Manceau, D. Chateigner, L. Charlet, J. Colloid Interface Sci. 215
(1999).
´
´
´
´
[29] S.K. Milonjic, M.R. Boškovic, T.S. Ceranic, Sep. Sci. Technol. 27 (1992) 1643.
[30] I.L. Lagadic, M.K. Mitchell, B.D. Payne, Environ. Sci. Technol. 35 (2001) 984.

38
A. Magro et al. / Inorganica Chimica Acta 410 (2014) 29–38
[31] N.L. Dias Filho, Y. Gushikem, W.L. Polito, Anal. Chim. Acta 306 (1995) 167.
[32] L. Daza, S. Mendioroz, J.A. Pajares, Clays Clay Miner. 39 (1991) 14.
[33] L. Mercier, C. Detellier, Environ. Sci. Technol. 29 (1995) 1318.
[34] L. Mercier, T.J. Pinnavaia, Microporous Mesoporous Mater. 20 (1998) 101.
[35] B.L. Rivas, H.A. Maturana, S. Villegas, S. Pereira, Polym. Bull. 40 (1998) 721.
[36] Y.H. Park, J.M. Lim, C.R. Park, J. Appl. Polym. Sci. 63 (1997) 773.
[37] P. Metilda, K. Sanghamitra, J.M. Gladis, G.R.K. Naidu, T.P. Rao, Talanta 65 (2004)
192.
[48] M. Paul, N. Pal, P.R. Rajamohanan, B.S. Rana, A.K. Sinha, A. Bhaumik, Phys.
Chem. Chem. Phys. 12 (2010) 9389.
[49] S.S. Mandal, U. Varshney, S. Bhattacharya, Bioconjugate Chem. 8 (1997) 798.
[50] E. Kaiser, R.L. Colescott, C.D. Bossinger, P.I. Cook, Anal. Biochem. 34 (1970) 595.
[51] (a) A.J. Gallant, J.H. Chong, M.J. MacLachlan, Inorg. Chem. 45 (2006) 5248;
(b) J.K.-H. Hui, M.J. MacLachlan, Chem. Commun. 23 (2006) 2480;
(c) C.T.L. Ma, M.J. MacLachlan, Angew. Chem., Int. Ed. 44 (2005) 4178;
(d) A.J. Gallant, M.J. MacLachlan, Angew. Chem., Int. Ed. 42 (2003) 5307;
(e) S. Akine, T. Taniguchi, T. Nabeshima, J. Am. Chem. Soc. 128 (2006) 15765;
(f) S. Akine, S. Sunaga, T. Taniguchi, H. Miyazaki, T. Nabeshima, Inorg. Chem. 46
(2007) 2959.
[52] P. Di Bernardo, P.L. Zanonato, S. Tamburini, P.A. Vigato, Inorg. Chim. Acta 360
(2007) 1083.
[53] E.F.S. Vieira, J.A. Simoni, C. Airoldi, J. Mater. Chem. 7 (1997) 2249.
[54] E. Angeletti, C. Canepa, G. Martinetti, P. Venturello, J. Chem. Soc., Perkin Trans.
1 (1989) 105.
[38] K. Oshita, M. Oshima, Y. Gao, K.H. Lee, S. Motomizu, Anal. Chim. Acta 480
(2003) 239.
[39] P.K. Jal, S. Patel, B.K. Mishra, Talanta 62 (2004) 1005.
[40] A.M. Starvin, T.P. Rao, J. Hazard. Mater. 113 (2004) 75.
[41] R.R. Navarro, K. Sumi, M. Matsumura, Water Res. 33 (1999) 2037.
[42] R.R. Navarro, K. Sumi, N. Fuji, M. Matsumura, Water Res. 31 (1996) 2488.
[43] I.K. Tonle, E. Ngameni, D. Njopwouo, C. Carteret, A. Walcarius, Phys. Chem.
Chem. Phys. 5 (2003) 4951.
[44] D.M. Manohar, K. Anoop Krishnan, T.S. Anirudhan, Water Res. 36 (2002) 1609.
[45] M.E. Mahmoud, G.A. Gohar, Talanta 51 (2000) 77.
[46] E.M. Soliman, M.B. Saleh, S.A. Ahmed, Anal. Chim. Acta 523 (2004) 133.
[47] D. Chandra, S. Kumar Das, A. Bhaumik, Microporous Mesoporous Mater. 128
(2010) 34.
[55] T. Luts, W. Suprun, D. Hofmann, O. Klepel, H. Papp, J. Mol. Catal. 261 (2007) 16.
[56] M.D. LeVan, T. Vermeulen, J. Phys. Chem. 85 (1981) 3247.
[57] A.R. Cestari, E.F.S. Vieira, E.C.N. Lopes, R.G. da Silva, J. Colloid Interface Sci. 272
(2004) 271.