Synthesis and characterization of a 3-hydroxy-4-pyridinone chelator functionalized with a polyethylene glycol (PEG) chain aimed at sequential injection determination of iron in natural waters

Article abstract of DOI:10.1016/j.poly.2015.09.015

The synthesis of a highly water soluble 3-hydroxy-4-pyridinone ligand, functionalized with a hydrophilic ethylene glycol chain (PEG-HPO), was successfully achieved and is reported together with that of its iron(III) complex. The improved hydrophilicity of both the PEGylated 3,4-HPO ligand and its iron(III) complex were fully investigated in an analytical application and, the new chelator is proposed for the spectrophotometric sequential injection determination of iron in waters. The new ligand provided better sensitivity and a lower LOD for iron determination than that obtained for N-alkyl-3-hydroxy-4-pyridinone ligands. The developed sequential injection method presents a dynamic working range of 0.10-1.00 mg Fe/L with a LOD of 48 μg/L. Due to the use of a sequential injection system, the overall effluent production was <2 mL corresponding to the consumption of 44 μg of PEG-HPO, 0.71 mg of NaHCO₃, 0.92 mg of HNO₃ and 500 μL of sample. Two reference samples were assessed for accuracy studies and a relative deviation <5% was obtained; eight waters samples were analyzed and the results compared with the reference procedure, and no statistical difference was observed for the two sets of results.

Full text of DOI:10.1016/j.poly.2015.09.015

Polyhedron 101 (2015) 171–178
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Synthesis and characterization of a 3-hydroxy-4-pyridinone chelator
functionalized with a polyethylene glycol (PEG) chain aimed at
sequential injection determination of iron in natural waters
Raquel B.R. Mesquita , Tânia Moniz b,1, Joana L.A. Miranda , Vânia Gomes b,1, André M.N. Silva b,
a,1
a,1
b
a,
⇑
c,
⇑
J.E. Rodriguez-Borges , António O.S.S. Rangel , Maria Rangel
a
CBQF – Centro de Biotecnologia e Química Fina – Laboratório Associado, Escola Superior de Biotecnologia, Universidade Católica Portuguesa/Porto,
Rua Arquiteto Lobão Vital, Apartado 2511, 4202-401 Porto, Portugal
REQUIMTE-UCIBIO, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, 4169-007 Porto, Portugal
b
c
REQUIMTE-UCIBIO, Instituto de Ciências Biomédicas de Abel Salazar, Universidade do Porto, 4050-313 Porto, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a highly water soluble 3-hydroxy-4-pyridinone ligand, functionalized with a hydrophilic
ethylene glycol chain (PEG-HPO), was successfully achieved and is reported together with that of its iron
Received 9 August 2015
Accepted 6 September 2015
Available online 12 September 2015
(III) complex. The improved hydrophilicity of both the PEGylated 3,4-HPO ligand and its iron(III) complex
were fully investigated in an analytical application and, the new chelator is proposed for the spectropho-
tometric sequential injection determination of iron in waters. The new ligand provided better sensitivity
and a lower LOD for iron determination than that obtained for N-alkyl-3-hydroxy-4-pyridinone ligands.
The developed sequential injection method presents a dynamic working range of 0.10–1.00 mg Fe/L with
Keywords:
3
-Hydroxy-4-pyridinone
Iron
PEGylation
Sequential injection
Water solubility
a LOD of 48
l
g/L. Due to the use of a sequential injection system, the overall effluent production was
2 mL corresponding to the consumption of 44 g of PEG-HPO, 0.71 mg of NaHCO , 0.92 mg of HNO
L of sample. Two reference samples were assessed for accuracy studies and a relative deviation
<
l
3
3
and 500
l
<
5% was obtained; eight waters samples were analyzed and the results compared with the reference pro-
cedure, and no statistical difference was observed for the two sets of results.
Ó 2015 Elsevier Ltd. All rights reserved.
1
. Introduction
lipophilic balance (HLB) without significantly changing its chelat-
ing properties. The variation in HLB can be achieved by introducing
appropriate substituents on the endocyclic nitrogen atom of the
pyridinone ring. Therefore, a more hydrophilic ligand is expected
to provide a more sensitive method with lower detection limit
than a previously used 3,4-HPO ligand [2].
Herein, we reported the synthesis of a 3,4-HPO ligand with
improved water solubility that was achieved by introducing an
ethylene glycol chain in the endocyclic nitrogen atom of the pyridi-
none framework. The aim was to increase the sensitivity and
decrease the detection limit for the spectrophotometric determina-
tion of iron by sequential injection analysis, which was limited by
the low concentration of the ligand although a saturated solution
was used [2].
Diverse chromogenic reagents, such as thiocyanate,
,10-phenantroline, bathophenanthroline, 2,2-bipyridyl, erio-
1
chrome cyanine R, cetyltrimethylammonium, are well known for
their analytical usage in the determination of iron [1] but also for
their high toxicity.
In a recent work, we tested low toxicity 3-hydroxy-4-pyridi-
none (3,4-HPO) chelators as chromogenic reagents and success-
fully applied them for determination of iron in bathing waters
[
2]. However, its efficiency was limited by the ligand water solubil-
ity as the sensitivity for the determination of iron increased with
the increase of ligand concentration in the chromogenic solution
[
2]. In this scenario, it would be rather advantageous to obtain a
more water soluble 3,4-HPO ligand. This option is feasible as the
structure of 3,4-HPO ligands allows tailoring of their hydrophilic/
Our group has long been interested in the synthesis and solu-
tion properties of 3,4-HPO ligands and their complexes with M
(
II) and M(III) metal ions for several applications [2–5]. To improve
the water solubility of the ligands we considered the use of oligo
ethylene glycol)s (OEGs) substituents that have been described
to provide high water solubility [6]. Herein we report the synthesis
⇑
Corresponding authors. Tel.: +351 220402593 (M. Rangel).
E-mail addresses: mrangel@icbas.up.pt, mcrangel@fc.up.pt (M. Rangel).
Authors contributed the same.
(
1
http://dx.doi.org/10.1016/j.poly.2015.09.015
277-5387/Ó 2015 Elsevier Ltd. All rights reserved.
0

1
72
R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
NMR and Mass spectrometry analyses were performed at Labo-
ratório de Análise Estrutural, Centro de Materiais da Universidade
do Porto (CEMUP) (Portugal). Elemental analyses were performed
at the analytical services of University of Santiago (Spain).
Microwave-assisted reactions were carried out in a CEM Discov-
ery Labmate circular single-mode cavity instrument (300 W max
magnetron power output) from CEM Corporation.
2
2
.1.2. Synthetic procedures
.1.2.1. Synthesis of the azide-PEGylated chain: 1-azide-2-(2-(2-
Fig. 1. Formula of 3-hydroxy-4-pyridinone functionalized with
a hydrophilic
ethylene glycol chain (PEG-HPO).
ethoxyethoxy)ethoxy)ethane (b). Tri(ethylene glycol)monomethyl
ether (a) (9.80 mL, 56.1 mmol) was dissolved in anhydrous DCM
(
40 mL) and the solution was placed in a schlenk tube, which
was then closed under argon atmosphere. To the solution was
added anhydrous Et N (11.7 mL, 84.2 mmol), under magnetic
of an amine-terminated OEGs with a methyl group in the end of
the chain and of the corresponding 3-hydroxy-4-pyridinone ligand,
PEG-HPO (Fig. 1) obtained through reaction of the amine with 2-
methyl-3-hydroxy-4-pyrone (maltol).
The conjugation of these amino-PEGylated chains with 3,4-HPO
chelating units improves the hydrophilicity of both the ligand and
their metal ion complexes thus allowing their wider use in biolog-
ical and analytical applications. The iron(III) complex of the new
ligand was also synthesized and characterized in order to validate
the detection method.
The efficiency of this new compound in solution as a novel chro-
mogenic reagent for determination of iron was tested in a sequen-
tial injection system. The complexation reaction of the synthesized
ligand and iron(III) was studied in-line and applied to two certified
waters samples. The results obtained in terms of sensitivity, selec-
tivity and limit of detection, were critically compared with those
previously described using the less hydrophilic 3,4-HPO ligand
3
stirring and cooled to 0 °C. To this mixture, a solution of mesyl
chloride (5.21 mL, 67.3 mmol), previously prepared by dissolution
in 40 mL of anhydrous DCM, was added dropwise, during 90 min.
The reaction mixture was kept with stirring, 1 h at 0 °C and the fol-
low 15 h at room temperature. The reaction mixture was then
washed with aqueous solutions of HCl (3%) (50 mL) and saturated
solution of NaCl (50 mL). The product was purified by liquid/liquid
extraction with DCM (3 ꢁ 50 mL). The organic phase was dried
with Na
includes the reaction of the last product obtained in anhydrous
DMF (40 mL), in a schlenk tube, with NaN (9.12 g, 140 mmol), at
5 °C during 15 h. The reaction mixture was then washed with
2 4
SO and concentrated by evaporation. The next step
3
6
water (10 ꢁ 30 mL) and with saturated solution of NaCl (40 mL).
The product was then extracted from the aqueous phase with
AcOEt (3 ꢁ 50 mL) and the organic phase was dried with Na
2 4
SO
[
2]. The choice of sequential injection (SI) analysis, as flow analysis
and concentrated by evaporation to afford 10.2 g of product (b)
technique, was based not only on its extensive use as an efficient
tool for water analysis [7] but also to attain an appropriate com-
parison with the previously developed method. In the end, a SI
method for the determination of iron is proposed, based on a newly
synthesized hydrophilic 3,4-HPO ligand as an alternative reagent
for iron determination.
as a light brown oil (90% yield) [6] (Fig. 5A).
1
RMN H (400.15 MHz, CDCl
3
, ppm): d 3.68–3.62 (m, 8H,
O–), 3.59–3.55 (m, 2H, –OCH CH ), 3.51 (q,
J = 7.0 Hz, 2H, –OCH CH ), 3.37 (t, J = 5.1 Hz, 2H, –OCH CH ),
.19 (t, J = 7.0 Hz, 3H, –CH CH ).
–
(OCH
2
CH
2
)
2
2
2 3
N
2
3
2
2 3
N
1
2
3
2
.1.2.2. Synthesis of the amine-PEGylated chain: 2-(2-(2-ethox-
2
. Materials and methods
yethoxy)ethoxy)ethylamine (1). A solution of product (b) (2.02 g,
.94 mmol) in methanol (40 mL) was placed into a hydrogenation
vessel. The air was removed with N , a catalytic amount of 1% Pd/C
w/w) was added and the mixture was stirred at room tempera-
9
2
2
.1. Synthesis of the 3,4-HPO ligand
2
(
.1.1. Materials and physical measurements for the ligands synthesis
2
ture, with H at 50 PSi for 24 h. The reaction mixture was filtered,
washed with methanol and the solvent evaporated in vacuum to
give the light brown oil product. The resulting residue was dried
Chemicals were obtained from Sigma–Aldrich (grade puriss,
p.a.) or Fluka (p.a.) and were used as received unless otherwise
specified.
under vacuum to give 1.68 g of 1 (97% yield) (Fig. 5A).
NMR spectra were recorded on a Bruker Avance III 400 spec-
1
RMN H (400.15 MHz, CDCl
3
, ppm): d 3.65–3.43 (m, 12H,
–), 2.81 (t, J = 5.2 Hz, 2H, –CH NH ), 1.50
), 1.16 (t, J = 7.0 Hz, 3H, –CH ).
1
trometer, operating at 400.15 MHz for H and 100.62 MHz for
–
CH
2
(OCH
2
CH
2
)
2
OCH
2
2
2
1
3
C atoms, equipped with pulse gradient units, capable of producing
magnetic field pulsed gradients in the z-direction of 50.0 G/cm.
(
s, 2H, –CH
2
NH
2
3
1
1
Two-dimensional H/ H correlation spectra (COSY), gradient
selected H/ C heteronuclear single quantum coherence (HSQC)
and H/ C heteronuclear multiple bond coherence (HMBC) spectra
were acquired using the standard Bruker software.
High resolution electrospray ionization mass spectra (ESI-MS)
were obtained in a Thermo Scientific LTQ-Orbitrap XL mass spec-
trometer, externally calibrated with a standard kit provided by
the manufacturer. The spectrometer was operated in the positive
ionization mode setting the capillary voltage to +3.0 kV, sheath
gas flow to 6 and the temperature of the ion transfer capillary to
1
13
2.1.2.3. Synthesis of the PEGylated 3,4-HPO. The PEGylated 3,4-HPO
was obtained using the typical procedure utilized to prepare 3,4-
HPOs in which a 3-hydroxy-4-pyrone reacts with a primary amine
1
13
[
8,9] (Fig. 5B).
2.1.2.3.1. Synthesis of the protected PEGylated 3,4-HPO (3). A mix-
ture of amine 1 (2.1459 g, 0.01832 mol) dissolved in dried ethanol
(6 mL) was placed in a 10 mL reaction vial, which was then closed
under argon atmosphere and placed in the cavity of a CEM micro-
wave reactor. The reaction vial was irradiated (1 min ramp to
160 °C and 120 min hold at 160 °C, using 100 W maximum power).
The reaction solvent was evaporated and the crude oil resultant
was dissolved in 50 mL of water. The product was separated from
the starting material 2 by liquid/liquid extraction with 3 ꢁ 30 mL
of diethyl ether. The organic phase was rejected and the aqueous
phase was concentrated and then purified by chromatography,
2
75 °C. Spectra were recorded for m/z values between 250 and
2
000 in the Fourier Transform (FT) mode with resolution (FWHM)
set at 60000. Samples were prepared in water, diluted in a
0% (v/v) water:methanol mixture immediately before analysis
and directly infused into the electrospray ion source utilizing the
5
ꢀ1
syringe pump in the mass spectrometer at 10 ll min .

R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
173
England) of 1000 mg/L. Working standards in the dynamic range
.1–2.0 mg/L were weekly prepared from dilution of the stock
solution in 0.03 M of nitric acid.
0
Fig. 2. Formulae and numbering of compound 3 and chelator PEG-HPO.
2.2.2. Sequential injection manifold and procedure
The sequential injection manifold used for the study of the com-
plexation reaction and for the determination of iron(III) is depicted
in Fig. 3.
using a mixture of chloroform/methanol (9:1), affording 1.7520 g
of conjugate 3 (49% yield) (Figs. 2 and 5B).
Solutions were propelled by a Gilson Minipuls 3 peristaltic
pump, equipped with PVC pumping tube connected to the central
channel of a ten port selection valve (Valco VICI Cheminert C25-
3180EUHB). All tubing connecting the different components was
made of PTFE (Omnifit), with 0.8 mm i.d. An Ocean Optics USB
4000 charged coupled device detector (CCD) equipped with a pair
1
4
H NMR (400.15 MHz, MeOD-d , ppm): d 1.55 (t, J = 7.1 Hz, 3H,
0
0
–
CH
2
CH
3
); 2.22 (s, 3H, 2-CH
3
); 3.46–3.61 (m, 10H, H3 –H7 ); 3.68
0
0
(
t, J = 4.6 Hz, 2H, 2 -CH
2
); 4.13 (t, J = 4.6 Hz, 2H, 1 -CH
2
); 5.07 (s,
2
–
H, –CH
2 6 5
C H ); 6.48 (d, J = 7.4 Hz, 1H, H5); 7.27–7.42 (m, 5H,
1
3
CH C H ); 7.69 (d, J = 7.4 Hz, 1H, H6). C NMR (100.62 MHz,
2 6 5
0
MeOD-d
4
, ppm): d 12.9 (2-CH
3
); 15.2 (–CH
2
CH
); 116.4 (C5); 128.9–
); 138.1 (C6); 141.9 (C2); 145.1
3
); 54.4 (C1 ); 70.5
of 400
halogen light source and a Hellma 178.710-OS flow-cell with
10 mm light path and 80 L inner volume, was used as detection
lm fiber optic cable and a Mikropack DH-2000 deuterium
0
0
0
(
C2 ); 71.0–71.3 (C3 –C7 ); 74.1 (–CH
C
2 6
H
5
1
29.1 (–C
6
H
5
); 129.8 (Cq, –C
6
H
5
l
(
C3); 174.1 (C4).
system. Data acquisition was performed through the Ocean Optics
2.1.2.3.2. Synthesis of the PEGylated 3,4-HPO (PEG-HPO). A solution
– Spectrasuite software at 459 nm.
of ligand 3 (2.0117 g, 0.00536 mol) in methanol (40 mL) and HCl
0.01 mL) was placed into a hydrogenation vessel. The air was
removed with N2, a catalytic amount of 10% Pd/C (w/w) was added
and the mixture was stirred at room temperature, with H at 40 PSi
for 6 h. The reaction mixture was filtered, washed with methanol
and CH Cl and the solvent evaporated in vacuum to give the
A personal computer (HP Pavilion zt3000) equipped with a
National Instruments DAQcard – DI0 interface card, running a
homemade software, was used to control the selection valve posi-
tion and the peristaltic pump direction and speed.
The protocol sequence used for the iron determination in natu-
ral waters using the newly synthesized ligand is described in
Table 1.
(
2
2
2
brown oil product. The resulting residue was dried under vacuum
to give 1.6378 g of 4 (95% yield) (Figs. 2 and 5B).
The first step was the aspiration of 3,4-HPO ligand solution
(step A) to the holding coil, followed by the aspiration of the buffer
and the standard (steps B and C). Mixing was promoted by the flow
reversal while propelling the aspirated plugs towards the detector
(step D).
+
MS: calculated for C14
weight M ), found: MS: 286.16261. H NMR (400.15 MHz, MeOD-
H
24NO
5
: 286.16 (monoisotopic molecular
+
1
d
3
4
, ppm): d 1.04 (t, J = 7.1 Hz, 3H, –CH
2
CH
3
); 2.54 (s, 3H, 2-CH
3
);
0
0
0
.37–3.56 (m, 10H, H3 –H7 ); 3.79 (t, J = 4.8 Hz, 2H, 2 -CH
2
); 4.48
0
(
2
t, J = 4.8 Hz, 2H, 1 -CH ); 7.03 (d, J = 7.4 Hz, 1H, H5); 8.04 (d,
1
3
J = 7.4 Hz, 1H, H6). C NMR (100.62 MHz, MeOD-d
4
, ppm): d 13.2
2.3. Accuracy assessment – sample collection
0
0
0
(
2-CH
3
); 15.4 (–CH
2
CH
3
); 57.0 (C1 ); 67.5 (C7 ); 70.1 (C2 ); 70.5–
0
0
7
1
1.5 (C3 –C6 ); 111.4 (C5); 140.0 (C6); 143.2 (C2); 144.6 (C3);
60.3 (C4).
For accuracy assessment, results obtained with the proposed
sequential injection method were compared to the certified values
of two certified water samples: SPS-SW2 (surface water) from LGC
standards and SLRS-4 (river water) from NRC-CNR National
Research Council Canada. Furthermore, eight river water samples
from recreational areas (ESI Table S1) were collected in polyethy-
lene plastic bottles of 0.5 L capacity at about 30 cm depth. The
samples, acidified at collection to pH ꢃ 2 (with HCl) according to
the collection procedure [10], were introduced directly in the
developed system without filtration.
3 3
2.1.2.4. Iron(III) complex of PEGylated 3,4-HPO (Fe(PEG-HPO) ). FeL
complex was prepared by dissolving stoichiometric amounts of Fe
NO O (0.283 g, 0.7 mmol) and ligand PEG-HPO (0.60 g,
ꢂ9H
.0021 mol) in water, adjusting the pH to 8 with a diluted solution
(
3
)
3
2
0
of NaOH. The reaction mixture was kept with stirring, for 3 days at
room temperature. A precipitate was formed, collected by filtration
and washed with water and a red power was isolated (quantitative
yield).
+
MS: [H
NaHL]+ (NaC14
Fe(L)
]+ (FeC28
2
L]+ (C14
H
24NO
), m/z = 286.16499
(
D
m = 0.3 ppm);
m = ꢀ0.7 ppm);
+
[
[
(
(
H23NO
5
), m/z = 308.14664
10), m/z = 624.23363 (
15), m/z = 909.39250 (
15), m/z = 931.37393 (
(
D
3. Results and discussion
+
2
44 2
H N O
D
m = ꢀ0.6 ppm); [H(Fe
m = 1 ppm); [Na(Fe
m = 0.4 ppm).
+
+
L)
L)
3
)] (FeC42
H
67
N
3
O
D
3.1. Synthesis of the new iron chelator
+
+
3
)] (NaFeC42
H
66
N
3
O
D
Ligands of the 3-hydroxy-4-pyridinone type are usually synthe-
sized through the reaction of a 3-hydroxy-4-pyrone and a primary
amine as depicted in Fig. 4 [11].
2
2
.2. Analytical procedure
.2.1. Reagents and solutions
All solutions were prepared with analytical grade chemicals
The use of the above reaction (Fig. 4) allows the introduction of
1
different R substituents on the heterocyclic nitrogen atom by
and boiled deionised water (specific conductance less than
selection of the appropriate amine. The use of amines with differ-
ent R substituents allows the synthesis of 3-hydroxy-4-pyridi-
S cm 1).
ꢀ
1
0
.1
l
Ligand solutions were obtained by dissolution of approximately
mg of the synthesized ligand in 50 mL of water corresponding to
a concentration of 180 mg/L.
The buffer solutions of hydrogen carbonate 0.25 M were pre-
pared by dissolving 1.05 g of sodium hydrogen carbonate in
0 mL of water. The pH was set to 10.5 with a 0.5 M sodium
hydroxide solution. An iron(III) stock solution, 10 mg/L, was pre-
pared by dilution from the atomic absorption standard (Spectrosol,
nones with variable physicochemical properties, such as water
solubility. Moreover, and as shown in previous studies
[4,9,11,12], such alterations on the substituents of the nitrogen
atom of the heterocyclic ring of 3,4-HPOs do not significantly
change the chelating properties of the 3,4-HPO ligands. Aiming to
increase the water solubility of the 3,4-HPOs used in the previous
study and considering the water solubility of oligo(ethylene glycol)
chains we synthesized an amino-PEGylated chain with a methyl
9
5

1
74
R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
Fig. 3. Sequential injection manifold to study the chromogenic reaction between PEGylated 3,4-HPO ligand and iron; SV, 10 port selection valve; HC, holding coil with 300 cm
of length; PP, peristaltic pump; PEG-HPO, 180 mg/L ligand solution; Buffer, 0.25 M hydrogen carbonate solution; S
spectrophotometer at 459 nm; W, waste.
i 1
, iron(III) standards; R , reaction coil with 6 cm; k,
Table 1
(1) by hydrogenation. The products of both reactions were
obtained with good yield, 90% and 97%, respectively.
Protocol sequence for the developed SI methodology for iron determination using a
PEG-HPO ligand as a colorimetric reagent.
Step
Port
Time (s)
Flow rate
L/s)
Volume
(lL)
Description
3
.1.2. Optimization of the experimental procedures the synthesis of the
(l
protected PEGylated 3,4-HPO (3)
A
B
C
D
1
2
3–8
9
4
2
8
30
62
17
62
62
248
34
496
1860
Aspiration of ligand
Aspiration of buffer
Aspiration of standard
Propelling to detector
The synthesized amine functionalized with an OGE chain (1)
reacts with the benzyl protected 3-hydroxy-4-pyrone (2) leading
to the displacement of the oxygen atom of the ring and its further
replacement by the nitrogen of the anime group of the chain pro-
ducing a benzyl protected 3-hydroxy-4-pyrone (Fig. 5B). Com-
pound
2 was synthesized in our laboratory following the
procedures described in the literature [8].
The synthesis of 3 was performed by adding 1 and 2 in ethanol
under reflux (oil-bath) for 72 h yielding 3 in 27% yield (a, Table 2).
In order to improve the reaction outcome to synthesize 3, some
modifications were performed on the synthetic procedure.
Microwave-assisted organic chemistry was used instead of the
traditional oil bath in order to obtain the desired compound 3 in
a good yield, decreasing the reaction time and consequently more
efficient synthetic protocol. Therefore, in a monomode reactor,
under closed-vessel conditions the reaction was performed using
an excess of 1.5 eq of the amine 1 and 100 mg of 2, during 2 h.
Then, the reacting mixture was analyzed by NMR and the conver-
sion of 2 into 3 was calculated and 38% of the limiting reagent 2
was converted in 3 (b, Table 2). The same conditions were used,
except for acidification of the medium with HCl and the percentage
of conversion was determined and 24% occurred (c, Table 2). This
result shows that the reaction occurs with higher percentage of
Fig. 4. Scheme of the reaction of the synthesis of 3-hydroxy-4-pyridinones.
group in the end of the chain and the corresponding 3-hydroxy-4-
pyridinone ligand as shown in Fig. 5.
3
.1.1. Synthesis of amino-PEGylated chain
In order to produce de amino-PEGylated chain, the hydroxyl
group of tri(ethylene glycol)monoethyl ether (a) was substituted
by a mesyl group to react with NaN and to form the azide group
b). The azide was subsequently reduced to produce the amine
3
(
Fig. 5. Synthetic routes of: A, the PEGylated chain; and B, the functionalized 3,4-HPO.

R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
175
Table 2
the protons of the benzyl ring appear as a multiplet between
.27 and 7.42 ppm. The carbon associated to this ethyl group is
Experimental conditions of the synthesis of modified 3,4-HPO (3).
7
Entry
Method
Fold
excess
of (1)
Amount
of (2)
Time
(h)
% of conversion
in (3) (calculated
by NMR)
at 74.1 ppm and the 5 carbons of the benzyl ring appear between
128.9 and 129.0 ppm. The resonance signal at 129.8 ppm was
found to give long range coupling to the resonance signals of the
protons of the ethyl group and the CH protons of the benzyl ring
was attributed to the quaternary carbon of the protecting group.
The last quaternary carbons at 141.9, 145.1 and 174.1 ppm were
attributed C2, C3 and C4 due to their correlations in HMBC with the
*
a
b
c
d
e
Oil bath
MW
MW
MW
MW
1.5
1.5
1.5
1.5
2
500 mg
100 mg
100 mg (+HCl)
500 mg
500 mg
72
2
2
2
2
27
38
24
59
84
0
protons of the pyridinone ring and with C1 for C2 carbon.
*
Yield obtained by product isolation after purification.
After the deprotection (PEG-HPO), significant differences in the
1
13
H and C spectra were detected as for example the shift of the
protons of methyl substituent in the position 2 of the pyridinone
ring (2-CH ) that moved from 2.22 to 2.54 ppm and the respective
carbon signal, 2-CH3, was shifted from 12.85 to 13.21 ppm. The
protons of –CH groups near to the nitrogen of the pyridinone
are also shifted, namely 2 -CH
and 4.48 ppm as well as their respective carbons from 70.5 to
3
conversion of 2 into 3 in non acidified medium and we tested the
same excess of 1.5 eq of the amine 1 during 2 h but scaling up the
amount of reagents to 500 mg of 2 (d, Table 2) and the conversion
obtained was 59%. In the next condition tested (e, Table 2), we
increased the excess of the amine 1 to 2 eq and we maintained
2
0
0
2
2
and 1 -CH , that moved to 3.79
7
0.1 and from 54.4 to 57.0, respectively.
5
00 mg of 2, and 2 h of reaction leading us to a percentage of con-
The resonance signals attributed to the –CH protons H5 and H6
version of 84%. This modification allowed formation of a higher
amount of the ligand 3 and concomitantly we obtained our desired
product with lower reaction time. All the experimental conditions
performed for the synthesis of 3 are summarized in Table 2.
In order to obtain the final 3-hydroxy-4-pyridinone the benzyl-
protecting group of 3 was removed under a hydrogen atmosphere
in the presence of Pd/C (10%) and HCl, to obtain the dihydrochlo-
ride salt of ligand PEG-HPO.
The modification of 3,4-HPO ligands with amino-PEGylated
chains was successfully achieved in this work leading to the syn-
thesis of a new 3,4-HPO ligand with an enhanced solubility in
water. The new type of 3,4-HPO ligands will significantly enhance
the water solubility of the ligands and its metal ion complexes and
for that reason will allow to enlarge the field of application of this
class of ligands.
revealed the major shift from 6.48 to 7.03 ppm and from 7.69 to
.04 ppm, respectively. Their respective carbons are also dislocated
8
to low field region, namely from 116.4 to 138.0 ppm in the pro-
tected ligand to 111.4 and 140.0 ppm in the deprotected form,
respectively for C5 and C6 carbons.
The quaternary carbons were assigned based on their long-
range correlations. The resonance signal at 143.2 has shown HMBC
0
correlation with H6, 1 -CH
2
and 2-CH
3
protons and was attributed
to C2. The carbon at 144.6 shows HMBC correlation with H5, H6
and protons and 2-CH and was attributed to C3. The pick at
60.3 ppm was attributed as carbons C4 due to their HMBC corre-
3
1
lations with H5 and H6.
The observed difference in the chemical shifts of 1H and 13
C
nucleus in the spectra of the protected and deprotected com-
pounds is primarily due to the deprotection of the hydroxyl group
and the acidic pH of the deprotection reaction that lead to isolate
the final ligands in the enolic form [13].
3.1.3. NMR spectroscopy
The structures of the ligands 3, and PEG-HPO in solution were
1
13
established by NMR analysis ( H and C, 1D and 2D experiments,
including COSY, HSQC and HMBC spectra for unequivocal assign-
ment of the most characteristic proton and carbon chemical shifts).
The assignment of the resonance signals in 13C NMR spectra of the
protected and de-protected compounds was achieved by analysis
3.1.4. Synthesis and characterization of the iron(III) complex of Ligand
PEG-HPO
3
The complex Fe(L) was synthesized by mixing the iron salt, Fe
(NO O, and the ligand in water at pH = 8. Upon precipitation,
3
)
3
ꢂ9H
2
1
13
1
13
of H/ C HSQC and H/ C HMBC spectra, which provide one and
the complex was isolated and its composition was determined by
ESI-MS. The mass spectrum of the isolated complex shows the free
1
13
multiple bond H– C connectivity.
1
+
+
The H NMR spectra of ligand 3 revealed that the resonance sig-
ligand (m/z = 286.16499 [H
2
L] and m/z = 308.14664 [NaHL] ) and
0
+
nals of the methyl protons from the terminal group (8 -CH
3
) appear
two charged iron complexes, [Fe(L)
2
]
(m/z = 624.23363) and Fe
)] and m/z = 931.37393 [Na(Fe
)] ), as observed for the alkylpyridinone iron(III) complexes
0
+
at 1.15 ppm and the correspondent carbon (C8 ) with HSQC corre-
lation appear at 15.17 ppm. The singlet at 2.22 ppm was attributed
(L)
(L)
3
(m/z = 909.39250 [H(Fe(L)
3
+
3
to the methyl substituent in the pyridinone ring (2-CH
correlation with resonance signal at 12.85 ppm allow us to attri-
bute this pick to 2-CH carbon. The multiplet in the aliphatic region
of the spectra (3.46–3.61 ppm) was assigned to protons of –CH
3
) and HSQC
[11]. The high accuracy mass determination together with the
observed isotopic patterns allowed to unequivocally assigning
the chemical composition of the ions. As revealed by solution spe-
ciation studies, the presence of both iron complexes is to be
expected at acidic pH values (pH 6 6), which can be provided by
the electrospray plume acidification occurring in positive ioniza-
3
2
0
0
groups (3 -CH
2
–7 -CH
2
) of the ethoxyl tail and their respective
carbons appear between 71.04 and 71.33 ppm. Signals at 3.68
and 4.13 ppm was assigned as the protons of the –CH
2
groups
and
tion mode [14,15]. In addition, it should be noted that the Fe(L)
3
complex is a neutral species and its ionization requires protonation
or sodium adduct formation. However, these reactions are likely to
0
linked to the nitrogen atom of the pyridinone ring (2 -CH
2
0
1
5
2
-CH ) and their respective carbons appear at 70.5 and
0
4.4 ppm. For carbons C1 , HMBC correlations with carbon 2-CH
3
promote ligand dissociation, leading to the detection of the more
+
was found, confirming the assignment performed.
stable [Fe(L)
2
]
cation and the free ligand in its fully protonated
+
+
The resonance singlets at 6.48 and 7.69 ppm were assigned to
the aromatic CH protons (H5 and H6, respectively) of the pyridi-
none residue and show HSQC and HMBC correlation with carbons
at 116.4 and 138.1 ppm assigned as C5 and C6, respectively. The
signals related with the protecting group are the singlet at
([H
vides further evidence for the ability of L to bind ferric ions, form-
ing FeL and FeL complexes, as expected for a 3,4-HPO bidentate
ligand. The UV–Vis spectrum of an aqueous solution of FeL at
2
L] ) and sodiated ([NaHL] ) forms. Altogether, MS data pro-
2
3
3
pH P 7 exhibits a strong charge transfer band (kmax = 460 nm)
5
.07 ppm that corresponds to the protons of the ethyl group and
characteristic of this type of complexes [11].

1
76
R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
3
.2. Analytical procedure – sequential injection method
L iron(III) standard was measured for 1 min. According to the work
of Mesquita et al. [2], the absorbance was monitored at 460 nm,
corresponding to the maximum absorbance of the FeL complex.
The advantageous characteristics of sequential injection (SI)
3
analysis concept namely versatility, extent of automation and
low reagent consumption, made it an appropriate choice for
automation of the analytical application of 3,4-HPO ligands for iron
determination. In this context, it was important to confirm that the
complexation reaction was as fast as evidenced in the previous
work involving other 3,4-HPO ligands, as this would strongly influ-
ence the SI protocol to be designed. To do so, the absorbance of a
mixture, placed in a conventional cell, of 1 mL ligand solution
The color was observed almost immediately after mixing, indicat-
ing the complex formation. After the observed initial reaction,
there was no significant absorbance increase, 69%, during the mea-
sured time (1 min). As described before [2], the carbonate buffer
was needed to adjust the pH of the solution, obtained by mixing
sample/standard, ligand solution and buffer, to about seven. Con-
sidering that the standard solutions were acidified to pH ꢃ 2, the
buffer solution was used at a pH ꢃ 9.5. So, the SI protocol involved
the sequential aspiration of the following plugs: ligand solution,
buffer solution and sample/standard.
1
78.5 mg/L, 0.1 mL carbonate buffer 0.25 M and 0.5 mL of 10 mg/
3.2.1. Physical parameters
The buffer volume was set at 30 lL as it represents the minimal
reproducible value to be used in SI methods [16]. The ligand solu-
tion volume, of about 250 L, was also based upon previous studies
2]. However, considering the possibility of a significant increase in
l
[
the ligand concentration, over 10 times higher than the one used in
our previous work, as result of the improved solubility of the newly
synthesized ligand, the hypothesis of reducing the ligand volume
was evaluated. The previously used volume ensured a ligand
excess of 2.5 times above the needed stoichiometry 3:1 (ligand:
Table 4
Assessment of possible interfering ions in the determination of iron(III) according to
legislated values, UNFAO, United Nations Food and agriculture Organization [10].
µ
Possible
Legislation maximum Tested concentration
%
Fig. 6. Study of the influence of the volume of ligand solution (h) and sample (s)
on the slope (full lines) and intercept (dashed lines) of the calibration curves for iron
determination; the points in full black correspond to the chosen volumes of ligand
and sample.
interfering ion values UNFAO (mg/L) (mg/L)
interference
Al3
Ca
+
5a
15
5.00
15.0
0.10
20.0
0.01
5.00
ꢀ4.7
1.5
ꢀ2.8
ꢀ9.4
0.9
2+
b
2+
a
Co
Mg
0.1
2
+
+
b
5
2
b
Ni
Zn
Expected – 0.001
2+
a
2
ꢀ3
a
Irrigation waters.
Stream waters.
b
Table 5
Results obtained for the determination of iron using the developed SI method and
those obtained with the reference procedure, graphite furnace atomic absorption
spectrometry (GFAAS); RD, relative deviation.
Sample ID
GFAAS
SI
RD (%)
mg Fe/L
SD
mg Fe/L
SD
1
2
3
4
5
6
7
8
0.057
0.101
0.031
0.087
0.049
0.040
0.086
0.077
0.001
0.007
0.001
0.009
0.002
0.001
0.006
0.001
0.055
0.098
<LOD
0.091
<LOD
<LOD
0.091
0.074
0.004
0.000
ꢀ3.5
ꢀ3.0
–
4.6
–
0.009
Fig. 7. Study of the influence of the ligand concentration (s) and the buffer
concentration (h) in the calibration curves of the iron determination; full lines for
the slopes (sensitivity) and dashed lines for the intercept; the points in black
represent the slopes and intercept of the chosen concentrations of ligand and buffer.
–
0.009
0.008
5.8
ꢀ3.9
Table 3
Features of the developed methodologies for iron(III) determination in water samples using PEG-HPO ligand as a colorimetric reagent.
Typical calibration curvea A = mꢂ[Fe3+] + b
LOD (
lg/L)
Quantification rate (h
ꢀ1
)
Reagent/sample
consumption
Effluent production (
lL)
Dynamic range (mg/L)
0
.10–1.00
y = 0.0392 ± 0.0017 [Fe3+] + 0.008 ± 0.002
48
72
44.3 g 3,4-HPO
l
1860
2
R
= 0.996 ± 0.005
0.71 mg NaHCO
3
0
5
.92 mg HNO
00
3
l
L
a
n = 5.

R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
177
Table 6
Comparison of the developed SI method, using the newly synthesized ligand, PEG-HPO, with the previously described [2] less soluble form of the 3,4-HPO.
Ligand formula
Ligand concentration
mg/L)
Sensitivity (calibration
curve slope, L/mg)
LOD (
lg Fe/L)
Sample consumption/
Reference
(
determination (
lL)
1
78
S = 0.0392
48
500
This work
[2]
O
O
N
O
O
CH3
HO
CH
3
2
0
S = 0.0251
83
300
O
NH
HO
CH
3
iron) so the reduction of the volume for about 190
l
L (the tested
concentrations of the tested cations were obtained from proper
dilution of atomic absorption standards. Several standards, with
600 ppb of iron(III) and the tested concentration of interfering ions,
were prepared and analyzed with the developed SI method. The
absorbance values of the standard with and without interfering
ion were registered and the interference percentage calculated
(Table 4).
Overall, no significant interferences were observed, with almost
all the interference percentages being below 5%. Exception was
observed for magnesium, with an interference percentage close
to 10%. However, the tested concentration was 4 times the
expected concentration in water streams, aiming to test the possi-
ble application to sea water. So the <10% interference percentage
was a good indicator of that possibility.
valued) still ensured excess of ligand. Two calibration curves were
established and the slope and intercepts compared (Fig. 6).
The results showed that decreasing the volume resulted in a
decrease in sensitivity (slope) and increase in limit of detection
(
intercept) so the initial volume of about 250
wards, the sample volume was studied within the range 310–
58 L, and the sensitivity increased up to the volume of 496 L,
so that was the chosen value (Fig. 6).
lL was set. After-
5
l
l
3.2.2. Chemical parameters
Calibration curves with different ligand solution concentrations,
19, 178 and 238 mg/L, were compared, in terms of both the sen-
1
sitivity (slope) and limit of detection (intercept) (Fig. 7). The ligand
solution concentration of 178 mg/L was chosen as the results
showed that the sensitivity (slope) increase up to that the concen-
tration together with the intercept decrease (consequently of the
limit of detection), proving to be the best choice.
The requirement of buffering the complex formation estab-
lished in Mesquita et al. [2] work and the use of a carbonate buffer
was set. A study of hydrogen carbonate solution concentration was
carried out. From the tested concentrations, ranging from 0.10 to
3.3.2. Method validation
For accuracy assessment of the developed method, two certified
water samples were analyzed and the results compared to the cer-
tified value. A river water sample NRC-CNR SLRS-4, with an iron
content of 103 ± 5
ꢀ3% as the concentration calculated with the developed method
was 100 ± 5 g/L. The other certified water, a surface water SPS-
SW2 with an iron content of 100 ± 1 g/L, resulted in a relative
deviation of 4% as the calculated concentration with the developed
method was 104 ± 8 g/L.
lg/L and the relative deviation obtained was
l
0
.60 M, the maximum sensitivity was obtained with a concentra-
l
tion of 0.25 M, so that was the chosen concentration. The intercept
increased with the increase of the concentration but as the increase
in sensitivity was over 40%, 0.25 M was still the best choice.
l
For further accuracy assessment, several natural waters were
analyzed (Table 5) with the developed SI method and the results
obtained compared to those obtained with the reference proce-
dure, graphite furnace atomic absorption spectrometry (GFAAS).
A linear regression between the values obtained with the
developed SI method (CSI) and the reference procedure, graphite
furnace atomic absorption spectrometry (CGFAAS), was established
3.3. Features of the developed SI methodology
After the detailed studies for the determination of iron based on
the colored complex formed with the PEG-HPO bidentate ligand,
the characteristics of the developed method were summarized
(Table 3).
(ESI Fig. S1); the equation found was:
CSI = 1.046 (±0.441)
The limit of detection, LOD, was calculated according to IUPAC
CGFAAS – 0.0036 (±0.0366), where the values in parenthesis are
recommendations [17]: three times the standard deviation of ten
blank signals. A typical calibration curve corresponds to the mean
of five calibration curves obtained in consecutive days (with the
standard deviation values in brackets). The determination rate
was calculated based on the time spent per cycle; a complete ana-
lytical cycle took about 0.83 min. An analytical cycle is the sum of
the time needed for each step plus the time necessary for the port
selection in the selection valve. The presented consumption values,
for reagents and sample, and the effluent production were calcu-
lated per determination.
the 95% confidence limits. These figures show that the estimated
slope and intercept do not differ from the values 1 and 0, respec-
tively. Thus, there is no evidence for systematic differences
between the two set of results [18].
4
. Conclusions
The use of the new 3,4-HPO ligand, PEG-HPO, especially
designed to possess higher water solubility, proved to meet the
aim of improving the detection limit and sensitivity in the determi-
nation of iron (Table 6).
In fact, the sensitivity obtained with the developed SI method,
using the new ligand, showed an increase close to 60% when com-
pared to the previously described method [2] and a decrease of the
detection limit to about half (Table 6).
The repeatability was assessed by calculation of the
relative standard deviation (RSD) obtained by the mean of ten
consecutive injections of sample. The calculated RSD was 3.1%
(
0.225 ± 0.007 mg Fe/L).
3.3.1. Assessment of possible interferences
In that previous work [2], there was also the use of a micro
The possible interference of several bivalent and trivalent
sequential injection format (lSI-LOV), and a comparison within
cations was assessed, and the tested concentrations were
based in the available legislation limits from UNFAO [10]. The
the same conditions, without revisiting the determination
parameters, was made. The same sensitivity of the method was

1
78
R.B.R. Mesquita et al. / Polyhedron 101 (2015) 171–178
maintained proving the efficiency of the new ligand for iron deter-
mination in both flow formats.
[2] R.B.R. Mesquita, R. Suárez, V. Cerdà, M. Rangel, A.O.S.S. Rangel, Exploiting the
use of 3,4-HPO ligands as nontoxic reagents for the determination of iron in
natural waters with a sequential injection approach, Talanta 108 (2013) 38.
The choice of sequential injection as a flow technique proved to
be highly appropriated, enabling the detailed study of the com-
plexation reaction with a fast and automatic method. For a flow
analysis effective application, contributed the excellent character-
istic of the complex formation being almost immediate. Further-
more, the good results obtained with the certified water samples,
reinforced the successful application to the iron determination in
natural waters.
Finally, and from the point of view of synthesis of new com-
pounds, the newly synthesized ligand is not only important for this
particular work but also opens new perspectives for the synthesis
of other metal ion complexes with improved water solubility.
[
3] R. Suárez, R.B.R. Mesquita, M. Rangel, V. Cerdà, A.O.S.S. Rangel, Iron speciation
by microsequential injection solid phase spectrometry using 3-hydroxy-1(H)-
2-methyl-4-pyridinone as chromogenic reagent, Talanta 133 (2015) 15.
[
4] T. Moniz, M.J. Amorim, R. Ferreira, A. Nunes, A.M.G. Silva, C. Queirós, A. Leite, P.
Gameiro, B. Sarmento, F. Remião, Y. Yoshikawa, H. Sakurai, M. Rangel,
Investigation of the insulin-like properties of zinc(II) complexes of 3-
hydroxy-4-pyridinones: identification of a compound with glucose lowering
effect in STZ-induced type I diabetic animals, J. Inorg. Biochem. 105 (2011)
1
675.
[5] T. Moniz, C. Queirós, R. Ferreira, A. Leite, P. Gameiro, A.M.G. Silva, M. Rangel,
Design of water soluble 1,8-naphthalimide/3-hydroxy-4-pyridinone
conjugate: investigation of its spectroscopic properties at variable pH and in
a
3+
2+
2+
the presence of Fe , Cu and Zn , Dyes Pigm. 98 (2013) 201.
[6] L.N. Goswami, Z.H. Houston, S.J. Sarma, S.S. Jalisatgi, M.F. Hawthorne, Efficient
synthesis of diverse heterobifunctionalized clickable oligo(ethylene glycol)
linkers: potential applications in bioconjugation and targeted drug delivery,
Org. Biomol. Chem. 11 (2013) 1116.
Acknowledgments
[7] R.B.R. Mesquita, A.O.S.S. Rangel, A review on sequential injection methods for
water analysis, Anal. Chim. Acta 648 (2009) 7.
[
8] Z. Zhang, S.J. Rettig, C.C. Orvig, Lipophilic coordination compounds: aluminum,
gallium, and indium complexes of 1-aryl-3-hydroxy-2-methyl-4-pyridinones,
Inorg. Chem. 30 (1991) 509.
J.L.A. Miranda thanks to for the Grant PTDC/AAG-
MAA/3978/2012_BI_2. T. Moniz and R.B.R. Mesquita thank to Fun-
dação para a Ciência e a Tecnologia (FCT, Portugal) and Fundo
Social Europeu (FSE) through the program POPH – QREN for the
grants SFRH/BD/79874/2011 and SFRH/BPD/41859/2007, respec-
tively. This work was supported by European Union FEDER funds
through COMPETE and by National Funds through FCT, projects
PTDC/AAG-MAA/3978/2012, PEst-OE/EQB/LA0016/2013, PEst-OE/
EQB/LA006/2013. This work was supported by FCT, Portugal,
European Union, QREN, FEDER and COMPETE, projects NORTE-07-
[9] C. Queiros, M.J. Amorim, A. Leite, M. Ferreira, P. Gameiro, B. Castro, K. Biernacki,
A. Magalhães, J. Burgess, M. Rangel, Nickel(II) and cobalt(II) 3-hydroxy-4-
pyridinone complexes: synthesis, characterization and speciation studies in
aqueous solution, Eur. J. Inorg. Chem. 2011 (2011) 131.
[10] APHA-AWWA-WPCF, Standard Methods for the Examination of Water and
Wastewater, 20th ed., American Public Health Association, Washington DC,
1998.
[
11] J. Burgess, M. Rangel, Hydroxypyranones, hydroxypyridinones, and their
complexes, Adv. Inorg. Chem. 60 (2008) 167.
[
12] M. Rangel, A. Leite, A.M.N. Silva, T. Moniz, A. Nunes, M.J. Amorim, C. Queiros, L.
Cunha-Silva, P. Gameiro, J. Burgess, Distinctive EPR signals provide an
understanding of the affinity of bis-(3-hydroxy-4-pyridinonato) copper(II)
complexes for hydrophobic environments, Dalton Trans. 43 (2014) 9722.
13] A.M.G. Silva, A. Leite, M. Andrade, P. Gameiro, P. Brandão, V. Félix, B. Castro, M.
0
162-FEDER-000048, NORTE-07-0124-FEDER-000066/67. We thank
CeNTI, V.N. Famalicão, for making available a CEM Discover micro-
wave reactor. The Bruker Avance III 400 spectrometer is part of the
National NMR network and was purchased under the framework of
the National Programme for Scientific Re-equipment, contract
REDE/1517/RMN/2005, with funds from POCI 2010 (FEDER) and
[
Rangel,
Microwave-assisted
synthesis
of
3-hydroxy-4-pyridinone/
naphthalene conjugates. Structural characterization and selection of
fluorescent ion sensor, Tetrahedron 66 (2010) 8544.
a
[14] S. Zhou, B.S. Prebyl, K.D. Cook, Profiling pH changes in the electrospray plume,
Anal. Chem. 74 (2002) 4885.
15] M. Girod, X. Dagany, R. Antoine, P. Dugourd, Relation between charge state
distributions of peptide anions and pH changes in the electrospray plume. A
mass spectrometry and optical spectroscopy investigation, Int. J. Mass
Spectrom. 308 (2011) 41.
(
FCT). This work was also supported under CTQ2013-47461-R
[
project from Ministerio de Ciencia e Innovación (MINCINN, Spain).
Appendix A. Supplementary data
[
16] R.B.R. Mesquita, A.O.S.S. Rangel, A sequential injection system for the
spectrophotometric determination of calcium, magnesium and alkalinity in
water samples, Anal. Sci. 20 (2004) 1205.
17] Analytical chemical division, nomenclature, symbols, units and their usage in
spectrochemical analysis – II. data interpretation, Pure Appl. Chem. 45 (1976)
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.poly.2015.09.015.
[
99.
References
[18] J.C. Miller, J.N. Miller, Statistics for Analytical Chemistry, third ed., Ellis
Horwood, Chichester, UK, 1993.
[
1] Z.
Spectrophotometry in Inorganic Analysis, first ed., Elsevier, The Netherlands,
000.
Marczenko,
M.
Balcerzak,
Separation,
Preconcentration
and
2