- Email: [email protected]
A soft material for chromium speciation in water samples using a chemiluminescence automatic system
Accepted Manuscript A soft material for chromium speciation in water samples using a chemiluminescence automatic system Cintia J. Fernandez, Claudia E. Domini, Marcos Grünhut, Adriana G. Lista PII:
S0045-6535(17)32156-2
DOI:
10.1016/j.chemosphere.2017.12.178
Reference:
CHEM 20559
To appear in:
ECSN
Please cite this article as: Fernandez, C.J., Domini, C.E., Grünhut, M., Lista, A.G., A soft material for chromium speciation in water samples using a chemiluminescence automatic system, Chemosphere (2018), doi: 10.1016/j.chemosphere.2017.12.178. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights A sorbent based on MWCNTs and ionic liquid was used for the speciation of chromium
RI PT
The speciation and detection of chromium was performed by using an automatic system Cr(III) and Cr(VI) were detected by chemiluminescence reactions
AC C
EP
TE D
M AN U
SC
Cr(VI) was determined by the oxidation of luminol with the Cr2O7= desorbed
1
ACCEPTED MANUSCRIPT 1
A soft material for chromium speciation in water samples using a
2
chemiluminescence automatic system
3
Cintia J. Fernandez, Claudia E. Domini, Marcos Grünhut, Adriana G. Lista*
5
INQUISUR, Departamento de Química, Universidad Nacional del Sur
6
(UNS)-CONICET, Av. Alem 1253, B8000CPB, Bahía Blanca, Argentina
RI PT
4
7
Running title: A soft material for chromium speciation in an automatic system
SC
8
M AN U
9
Correspondence:
11
Prof. Adriana G. Lista
12
Departamento de Química, Universidad Nacional del Sur
13
Av. Alem 1253, B8000CPB, Bahía Blanca, Argentina
14
Fax: +54 0291 4595160
15
E-mail address: [email protected]
TE D
10
EP
16
Abbreviations: BmimCl: 1-Butyl-3-methylimidazolium-chloride; C: channel;
18
CNT: carbon nanotubes; EA: electronic actuator; FBA: Flow-batch analysis; IL:
19
Ionic liquids; MDC: mixing detection chamber; MFS: multicommutated flow
20
system; MWCNT: Multiwalled carbon nanotubes; V: solenoid valve; PP:
21
peristaltic pump.
AC C
17
22 23
Keywords: Automatic system, Chromium speciation, Ionic liquids, Multiwalled
24
carbon nanotubes, Chemiluminescence detection
25 1
2
ACCEPTED MANUSCRIPT Abstract
27
A soft material formed by multiwall carbon nanotubes and 1-butyl-3-methyl
28
imidazolium chloride was used as sorbent material to perform the chromium
29
speciation in natural waters. This soft material was not yet used for the
30
speciation of metals as chromium. Thus, a multicommutated flow system
31
containing a minicolumn packed with the soft material was designed. The
32
procedure was based on the capacity of the sorbent to retain Cr(VI) as Cr2O7=
33
and allow to pass Cr(III) through the column. Then, a fully automated flow-batch
34
analysis
35
chemiluminescence detection. Thus, Cr(III) was determined as catalyst of the
36
luminol and hydrogen peroxide reaction and Cr(VI) as oxidant of luminol
37
reaction. This represents a new approach because the oxidation of luminol
38
using Cr2O7= has not been reported in literature. The variables of the two
39
systems were optimized. The limits of detection were 1.4 µg L-1 for Cr(VI) and
40
4.0 µg L-1 for Cr(III). The precision of the method was 3.8% and 7.0% for Cr(VI)
41
and Cr(III), respectively. The present method was applied to real water samples
42
with recoveries between 95% and 107%. Besides, these results were in
43
accordance with those obtained using inductive coupled plasma-optical
44
emission spectrometry technique.
46
developed
to
SC
was
quantify
both
species
using
EP
TE D
M AN U
system
AC C
45
RI PT
26
47 48 49 50 2
3
ACCEPTED MANUSCRIPT 1. Introduction
52
Chromium speciation is always important because of the bioavailability and
53
toxicity of some of its chemical forms. Even though there are many oxidation
54
states of chromium, the most commonly found are Cr(III) and Cr(VI), especially
55
as environment pollutants. While Cr(III) is considered an essential trace element
56
for preservation of glucose, lipids and protein metabolism (Shah et al., 2012),
57
Cr(VI) is irritant and toxic to human body tissue due to its water solubility and its
58
oxidizing potential. It has been demonstrated that inorganic species of Cr(VI)
59
are carcinogenic (Saçmaci et al., 2012). For these reasons, the determination of
60
the different species of chromium presents greater interest than the
61
determination of total chromium. There are several methods to carry out the
62
speciation of chromium in environmental samples applying different techniques.
63
In most of them, the direct determination of Cr(VI) is proposed and Cr(III) is
64
determined after its oxidation to Cr(VI) (Tuzen., Soylak., 2007). Ren et al.
65
present the determination of Cr(III) by electrothermal atomic spectrometry with
66
the reduction of Cr(VI) to Cr(III) (Ren et al., 2007). On the other hand,
67
separative techniques as ion chromatography, capillary electrophoresis and LC-
68
ICP-MS also have been used (Séby et al., 2003, Lin et al., 2016) but all these
69
methods need a pretreatment step to remove the analytes.
70
environmental pollutants such as heavy metal and organic compounds have
71
been extracted using solid phase extraction with different sorbent materials.
72
These materials, like activated carbon (Feist et al., 2014), resins (Tokalioğlu et
73
al., 2017) carbon nanofibers (Mubarak et al., 2017), and carbon nanotubes
74
(CNTs) (Tarigh et al., 2013; Thines et al., 2017; Mubarak et al., 2015) have an
75
outstanding role in the enrichment efficiency of the analytes. It is well known the
AC C
EP
TE D
M AN U
SC
RI PT
51
Usually,
3
4
ACCEPTED MANUSCRIPT excellent properties of carbon nanotubes as sorbent material but one of the
77
major disadvantages is their tendency to form aggregates. For this reason,
78
different alternatives were proposed to disperse the nanotubes using different
79
solvents such as surfactants, organic solvents and ionic liquids (ILs) (Lu et al.,
80
2012). In recent years, ILs have been combined with sorbent materials to
81
improve the efficiency and selectivity of the extraction procedure (Tokalıoğlu et
82
al., 2016). The role of ILs is to enhance the dispersion of CNTs, creating in this
83
way a new kind of sorbent material called soft material. This material was used
84
to extract and preconcentrate organic analytes (Polo-Luque et al., 2013). Until
85
now, to the best of our knowledge, there are not reports in which this soft
86
material was applied to evaluate chromium speciation. Only a few works report
87
this speciation in a previous step to the determination of analytes. One of these
88
uses nanostructured α-alumina (Monasterio et al., 2009) and another uses
89
MWCNTs for the adsorption of Cr(VI)-ammonium pyrrolidine dithiocarbamate
90
chelate (Tuzen., Soylak., 2007). On the other hand, up to now, the use of
91
chemiluminescence technique for determining both species of chromium has
92
not been reported. Seitz et al. described the determination of Cr(III) by
93
chimiluminescence
94
reaction
95
chemiluminescence technique are high selectivity and specificity. Moreover, it
96
presents a high compatibility and easy coupling to automatic systems. The
97
automatic systems based on multicommutation and flow-batch analysis are
98
interesting alternatives to make feasible a method or to transform batch
99
processes into automated processes. Both MFS and FBA systems may perform
100
different steps in the analytical procedure, such as separation and/or
AC C
EP
TE D
M AN U
SC
RI PT
76
(Seitz
et
analysis using the analyte as catalyst of the luminol al.,
1972).
Some
of
the
advantages
of
using
4
5
ACCEPTED MANUSCRIPT preconcentration of analytes and determination that may include the coupling to
102
conventional analytical instruments. Conformation and advantages of FBA
103
systems have been described in literature (Diniz et al., 2012). Particularly, the
104
complete programmability of the process (including, control of sampling, flow
105
rates, filling of mixing chamber, detection and evacuation) and the possibility to
106
perform the measurements of the analytical signal directly inside the mixing
107
chamber can be highlighted. Both are important facts for the automation of
108
methods where chemiluminescence detection is used (Grünhut et al., 2011).
109
In this work, a new analytical method to determine Cr(VI) and Cr(III) in water
110
samples was developed. For this purpose, a MFS to perform the speciation of
111
chromium, which included a minicolumn packed with a soft material, was
112
designed. The soft material, composed of MWCNTs and IL, improved the
113
absorption of Cr(VI) as Cr2O7= in comparison to MCNTs, due to the synergic
114
effect between them. Additionally, this soft material allowed the separation of
115
both chromium species. Then, a FBA system with chemiluminescence detection
116
was developed to carry out the quantitation of Cr(VI), as oxidant of luminol, and
117
Cr(III) as catalyst in the reaction of the luminol and hydrogen peroxide, both in a
118
basic medium. According to our knowledge, this is the first time that the
119
chemiluminescence technique is used to determine both analytes, particularly
120
using the same automatic system.
122
SC
M AN U
TE D
EP
AC C
121
RI PT
101
2. Materials and Methods
123 124
2.1 Reagents and Solutions
5
6
ACCEPTED MANUSCRIPT All solutions were prepared by using analytical-grade reagents and ultra-pure
126
water (18MΩ cm-1) from a Milli-Q System (Millipore, USA). A 0.2 mol L-1
127
potassium chloride (Merck) solution was mixed with a 0.2 mol L-1 sodium
128
hydroxide (Merck) solution in order to obtain a buffer solution of pH 12.00. A
129
0.01 mol L-1 luminol stock solution was prepared by dissolving 0.1770 g of
130
luminol (Sigma-Aldrich) with 100 mL of the buffer solution. This solution was
131
stored in a dark bottle at 4°C and it was stable fo r one week. The luminol
132
working solution (5x10-3 mol L-1) was daily prepared by appropriate dilution of
133
stock solution in buffer solution. Hydrogen peroxide solution was prepared by
134
diluting of 100 v/v H2O2 solution to obtain a 0.01 mol L-1 work solution. A 5x10-3
135
mol L-1 EDTA solution was prepared by diluting of 0.1 mol L-1 stock solution
136
(Merck Titrisol). Stock solutions of Cr(III) and Cr(VI), both of 100 mg L-1, were
137
prepared by dissolving Cr(NO3)3.9H2O (Mallinckrodt) and K2Cr2O7 (Sigma-
138
Aldrich) in water, respectively. These solutions were acidified with nitric acid.
139
Standard solutions of Cr(III) and Cr(VI) were freshly prepared by dilution of the
140
stock solutions. MWCNTs with external diameters of 40-60 nm and purity >95%
141
(Bayer) and a 0.5 mol L-1 BmimCl (Merck) solution were used to perform the
142
speciation of chromium.
143
AC C
EP
TE D
M AN U
SC
RI PT
125
144
2.2 Instrumentation and Software
145
An
146
photomultiplier operated at 900 V was used. The emission of photons was
147
registered at 425 nm. A Gilson® Minipuls 3 and an Ismatec® peristaltic pumps
148
and Tygon® tubes were used to pump the fluids. The rest of the tubing was
149
made of Teflon® (0.5 mm i.d.). NResearch® three-way solenoid valves were
Aminco Bowman®
Series
2
luminescence
spectrometer
with
the
6
7
ACCEPTED MANUSCRIPT used to handle all solutions into the systems. The MDC was constructed with
151
Teflon® and was designed in our laboratory in order to prepare solutions and
152
perform the detection in the same place. A lab-made stirrer system was
153
designed to improve the mixing inside the MDC. An EA connected to a
154
Pentium® 4 microcomputer was used to control all the components of the
155
systems. The software used for controlling both, MFS and FBA systems, was
156
developed in Labview® 5.1 visual programming language. A model 710 A
157
Orion® pHmeter with an Orion-Ross® model 81-02 electrode was used to
158
perform the pH measurements.
159
The composition of the soft material was examined by X-Ray Energy Dispersive
160
Microanalysis (EDS) (X-Max50, Oxford). The physical structure and morphology
161
of the soft material, before and after the sorption of Cr(VI), were analyzed by the
162
Scanning Electron Microscope (SEM) (Leo, 40 XVP, England). In both cases,
163
the sample preparation was carried out metallizing it with gold in a Sputter
164
Coater (Pelco 91000).
SC
M AN U
TE D
EP
165
RI PT
150
2.3 Chromium Speciation Procedure
167
A soft material based on the combination of BmimCl and MWCNTs was
168
prepared to perform the speciation of chromium. A MFS that included a
169
minicolumn packed with this material was developed (Fig. 1). The soft material
170
was prepared weighing 0.1 g of MWCNTs and adding 0.5 mL of BmimCl. This
171
mixture was homogenized in a mortar and it was kept at 4º C for 24 h. Then, a
172
Tygon® tube (length: 30 mm; inner diameter: 4mm) was used to pack this
173
material. The MFS consisted of two three-way solenoid valves (V1 and V2) and
174
only one channel (C1) where the minicolumn was inserted. As it can be seen in
AC C
166
7
8
ACCEPTED MANUSCRIPT Fig. 1, different steps were performed in the MFS. When V1 and V2 valves were
176
OFF, the soft material was conditioned passing 1 min of water and 2.5 min of
177
0.05 mol L-1 of BmimCl solution through it. Then, 5.0 mL of standard/sample
178
solution was loaded and a vial 1 was placed at the end of the channel. At this
179
time, the Cr(III) contained in the standard/sample passed through the column
180
and it was collected in the vial 1, while the Cr(VI) was retained on the soft
181
material. After this, the vial 1 was replaced by a waste vial and the column was
182
washed with 1min of water and then 1 min of air was passed through it to
183
remove the water. All these steps were performed at 0.85 mL min-1. Finally, the
184
elution step was performed at 1 mL min-1 by turning ON V1 and V2 valves and in
185
reversal flow mode. In this way, 300 µL of BmimCl alkaline solution with NaCl
186
passed through the column. When the volume of eluent was all inside the
187
column, the flow was stopped for 10 min. Then, the flow was restored to collect
188
the amount of Cr(VI) in vial 2, at the end of the channel.
SC
M AN U
TE D
189
RI PT
175
2.4 FBA system with chemiluminescence detection
191
A schematic diagram of the proposed FBA system is showed in Fig 2a. The
192
system consisted of five channels, through which luminol (C2), sample (C3),
193
hydrogen peroxide (C4), water (C5) and waste (C6) flowed. The direction of the
194
flow in the C2 to C5 channels was controlled by three-way solenoid valves. Then,
195
when the valves (V3 to V6) were OFF, the solutions were recycled to the
196
respective flasks. When the valves were ON, the respective solutions were
197
pumped to the MDC. As it can be seen in Fig. 2b, the lab-made MDC has been
198
designed with four inlets for incoming solutions and one output for emptying. It
199
was equipped with a quartz window, which was located at the bottom part of the
AC C
EP
190
8
9
ACCEPTED MANUSCRIPT chamber in order to minimize the final volume of the reaction mixture (1.50 mL).
201
In this way, the MDC was used to perform both the chemical reaction and the
202
chemiluminescence detection. In addition, a little magnetic stirring system was
203
constructed using a cooler fan motor obtained from an Intel® microprocessor
204
and placed below the MDC (Fig. 2b). The MDC coupled to the stirring system
205
was placed in the spectrofluorimeter by means of a lab-made cell holder (Fig.
206
2c). On the other hand, the fifth channel (C6) was used for emptying the MDC
207
by means of an auxiliary peristaltic pump (PPA) Thereby, when the
208
chemiluminescence reaction finished, the solution in the MDC was pumped to
209
the waste.
M AN U
SC
RI PT
200
210
2.5 FBA procedure
212
As an initial step, the flow rate for each channel was studied in order to
213
establish the time of activation of each solenoid valve. These times
214
corresponded to the precise volumes of solution that were introduced into the
215
MDC. For this purpose, each channel was filled with bidestilated water and the
216
water was collected and weighed in an analytical balance by activating the
217
peristaltic pump for 30 s. The corresponding volume was calculated by means
218
of the density data taking into account the experimental conditions (25 ºC and 1
219
atm). Finally, the flow rates obtained for each channel were 2.30, 2.60, 2.65,
220
2.30 and 2.28 mL min−1 for luminol (C2), sample (C3), hydrogen peroxide (C4),
221
water (C5) and waste (C6), respectively. Before starting the procedure, the
222
valves were set in a way that all solutions in their respective channels were
223
recycled toward their flasks (OFF position). Then, V3, V4, V5 and V6 valves were
224
switched ON for 10 s and the solutions were pumped toward the MDC in order
AC C
EP
TE D
211
9
10
ACCEPTED MANUSCRIPT to fill the channels between the valves and the MDC. Immediately, the PPA was
226
switched ON and the excess of the solutions contained in the MDC
227
wasaspirated to the waste for 20 s. This operation consumed a total time of 30
228
s. Then, the system was ready to perform the preparation of the calibration for
229
Cr(III) and Cr(VI) and the sample analysis. In the case of Cr(III) determination,
230
the analysis of standard solutions and the samples was performed by
231
sequentially switching ON valves V3, V4 and V5 during previously defined
232
intervals of time (26, 7 and 7 s, respectively). Thus, 1.00 mL of luminol, 0.30 mL
233
of Cr (III) and 0.20 mL of hydrogen peroxide were pumped toward the MDC. For
234
Cr (VI) determination, the analysis of standard solutions and the samples was
235
performed by sequentially switching ON valves V3 and V4 during previously
236
defined intervals of time (31 and 7 s, respectively). In this way, 1.20 mL of
237
luminol and 0.30 mL of Cr (VI) were pumped toward the MDC. In both cases,
238
the corresponding chemiluminescence signal at 425 nm was immediately
239
measured as peak height. When the register of the signal was finished, the
240
solution was aspirated and thrown into the waste by switching ON the PPA for
241
45 s. The magnetic stirring was ON during all the analysis. This procedure was
242
repeated for each Cr (III) and Cr (VI) standard solution and each sample. On
243
the other hand, the system was always cleaned between measurements. The
244
MDC cleaning procedure was carried out by switching ON V6 valve and the
245
magnetic stirring for 40 s. The total emptying of the MDC was assured by
246
switching ON the PPA for 45 s.
AC C
EP
TE D
M AN U
SC
RI PT
225
247 248 249
3. Results and discussion 10
11
ACCEPTED MANUSCRIPT 3.1 Characterization of the soft material before and after adsorption of
251
Cr(VI)
252
The results obtained by EDS (Fig. 3) showed that the soft material was free
253
from impurities. The five areas marked in figure 3a showed the same results, as
254
it was expected. In the analysis the only elements found were the
255
corresponding ones to the composition of the soft material: carbon, oxygen,
256
nitrogen and chloride as it can be seen in Fig. 3b. (corresponding to the
257
analysis of area 3 in figure 3a). On the other hand, Fig. 4 shows the physical
258
structure and morphology of the soft material before and after adsorption of
259
Cr(VI) obtained by SEM. Smooth MWCNTs were observed forming an
260
intertwined pattern, probably due to its combination with the ionic liquid (Fig. 4
261
a). Besides, the morphology of the soft material shows that the MWCNTs have
262
uniform diameters between 35 and 45 nm, which demonstrate that the soft
263
material is homogeneous. In general, the tubes became wider, especially at the
264
ends, after the adsorption process (Fig. 4 b), showing slightly bigger diameters
265
between 47 and 52 nm. Also, the surface of the tubes was not as smooth
266
(Mubarak et al., 2016) and looked more compact, probably due to the
267
incorporation of chromium into the material.
SC
M AN U
TE D
EP
AC C
268
RI PT
250
269
3.2 Optimization of speciation variables
270
In this work, a soft material was prepared to perform the speciation of
271
chromium. A combination of BmimCl and MWCNTs was used instead of
272
MWCNTs, because the soft material presented more efficiency to adsorb Cr(VI)
273
as Cr2O7= , while Cr(III) passed through the column. This fact is based on the
274
synergic effect between the IL and the CNTs. It has been demonstrated that in 11
12
ACCEPTED MANUSCRIPT a soft material, CNTs are surrounded by ILs, taking into account that the atoms
276
of the alkyl groups are nearer to CNTs surface than the nitrogen and carbon
277
atoms of imidazolium rings (Wang et al, 2008). A physisorption phenomenon is
278
produced, where electrostatic interactions and Van del Waals forces take place
279
on the MWCNTs surface and charged ion clusters are surrounding this surface
280
(Martinis et al., 2017). In this way, the anion of IL (Cl-) is interchanged with
281
Cr2O7= and the analyte is retained on the soft material, while Cr(III) passes
282
through the column. This mechanism is showed in Fig. 5. In order to validate
283
this selectivity of the soft material, the typical spectrophotometric reaction for
284
Cr(VI) with 1,5-diphenylcarbazide was used (Bose, 1954). Cr(VI) was
285
determined after its elution and Cr(III) was determined after oxidation with
286
hydrogen peroxide (Andersen, 1998). The obtained results demonstrated that
287
Cr(III) was not retained on the soft material.
288
As mentioned before, the separation process was carried out in the minicolumn
289
inserted in a channel of the MFS, packed with the soft material. So, it was
290
necessary to optimize all variables of the MFS. Table 1 shows the range of all
291
these variables and their optimum values. On the other hand, the direction of
292
elution and type of eluent were also optimized. For these purposes, three
293
solutions were tested: IL, alkaline IL and alkaline IL with NaCl. The pH of
294
alkaline IL solution was varied between 8.00 and 12.00 and the best recoveries
295
were obtained at pH 11.00. At this pH, the predominant specie of Cr(VI) is
296
CrO4=. So it was necessary to add a drop of concentrated nitric acid to the
297
eluate, in order to obtain Cr2O7= and then get it into the FBA system. Then, to
298
enhance the recovery values, 0.01 g mL-1 of NaCl was added to test the salting
299
out effect and effectively, the recoveries were better. Taking into account the
AC C
EP
TE D
M AN U
SC
RI PT
275
12
13
ACCEPTED MANUSCRIPT contact time of the eluent with the analyte, it was supposed that the best results
301
would be obtained with low flow rates for the elution. However, the best
302
chemiluminescence signals were obtained when higher flow rates were used
303
and a stop flow step was included. The results were improved when the elution
304
was performed in a reverse mode regarding the sample load direction. This
305
elution mode avoids the tight packing of the sorbent and decreases the
306
dispersion of the concentrate during the elution. Thus, the selected optimum
307
values are showed in Table 1.
308
Moreover, the soft material needed a conditioning step. For this purpose, a 0.05
309
mol L-1 BmimCl solution, with and without NaOH, was passed through the
310
column during different times (1 min to 3 min). The IL solution without NaOH
311
(pH = 5.00) flowing for 2.5 min gave the best results in terms of recovery values
312
for Cr(VI). Besides, two different lengths of the column were tested (15 and 30
313
mm) and the best results, based on the best chemiluminescence signals, were
314
obtained with the longer column.
SC
M AN U
TE D
EP
315
RI PT
300
3.3 Optimization of the chemiluminescence reactions and the FBA system
317
The luminol in presence of an oxidant and in alkaline medium produces a
318
chemiluminescence signal at 425 nm. So, using Cr2O7= as oxidant it was
319
possible to determine Cr(VI). According to the literature, this is the first time that
320
Cr(VI) is determined by chemiluminescence, using the analyte as oxidant for the
321
luminol reaction. Usually, Cr(VI) is reduced to Cr(III) and is determined as
322
catalyst of the luminol reaction. On the other hand, taking into account that the
323
chemiluminescence reaction of luminol can be catalyzed by different metals, the
324
determination of Cr(III) was performed using hydrogen peroxide as oxidant and
AC C
316
13
14
ACCEPTED MANUSCRIPT Cr(III) as catalyst. The concentration of all chemical reagents involved in both
326
reactions and the reaction media should be optimized. The luminol
327
concentration was studied between 1 x 10-4 mol L-1 and 1 x 10-2 mol L-1 and the
328
optimum value was 5 x 10-3 mol L-1 for both analytes. The studied pH range was
329
10.00-13.00. Again, the optimum value (12.00) was the same for the two
330
analytes. These facts represented an advantage in the determination of both
331
analytes. The other optimized variables were only for the determination of Cr(III)
332
and they were: H2O2 concentration (studied range: 1 x 10-4 mol L-1 and 3 x 10-2
333
mol L-1, optimum value: 1 x 10-2 mol L-1), EDTA concentration (studied range:1 x
334
10-3 mol L-1 and 1 x 10-2 mol L-1, optimum value: 5 x 10-3 mol L-1). Also, the
335
variables related to the FBA system were optimized in order to obtain the best
336
signal. FBA variables such as order and volumes of the reagents added into the
337
MDC and stirring time were optimized based on the higher chemiluminescence
338
signal and the repeatability of the measurements obtained for both Cr(III) and
339
Cr(VI) determinations. As regards to the order of the reagents added into the
340
MDC, three different combinations were tested: luminol and oxidant, oxidant
341
and luminol, and luminol-oxidant simultaneously. The optimum order was
342
luminol and oxidant. On the other hand, the total volume of reaction was 1.50
343
mL, which included 1.00 mL of luminol, 0.20 mL of sample and 0.30 mL of
344
hydrogen peroxide for Cr(III) and 1.20 mL of luminol and 0.30 mL of sample for
345
Cr(VI).
346
corresponding oxidant (H2O2 for Cr (III) and Cr2O4= for Cr (VI)) was added so
347
fast as possible to the MDC (See section 2.5). Finally, the MDC coupled to the
348
stirring system was designed in order to obtain a rapid and efficient mixture of
349
the solutions. As the speed of stirring presented a significant effect in the
AC C
EP
TE D
M AN U
SC
RI PT
325
Optimal
chemiluminescence
signals
were
obtained
when
the
14
15
ACCEPTED MANUSCRIPT analytical signal, it was necessary to maintain a high and stable speed of
351
stirring (500 rpm) during the chemiluminescence reaction.
352
3.4 Analytical performance
353
The different analytical parameters were calculated using the optimized
354
proposed methods. The calibration curve for Cr(VI) was y = (59.396 ± 569) x +
355
(1968 ± 30), R2 = 0.9998, and for Cr(III) was y = (75.250 ± 3088) x – (1254 ±
356
162), R2 = 0.9960. In both cases, y: chemiluminescence signal and x: [µg L-1].
357
The limit of detection (LOD) and limit of quantitation (LOQ) for Cr(VI) were 1.4
358
µg L-1 and 38 µg L-1, respectively. For Cr(III) the LOD was 4 µg L-1 and the LOQ
359
was 12 µg L-1 (Miller and Miller; 2010). The precision of the method,
360
represented by repetitively (% RSD) was 7.0 and 3.8 for Cr(III) and Cr(VI)
361
respectively, calculated from eight independent measurements of a standard
362
solution containing 50 µg L-1 for both analytes.
363
TE D
M AN U
SC
RI PT
350
3.4.1 Selectivity
365
The effect of common cations and anions present in water samples in the
366
determination of 0.08 mg L-1 of Cr(VI) and 0.07 mg L-1 of Cr(III) was examined
367
applying the whole method. Different cations present in water samples could
368
catalyze the luminol reaction instead of Cr(III), so a study of these interferences
369
was performed. The first step in this study was to compare the concentration of
370
a standard solution of Cr(III) (0.07 mg L-1) with the concentration corresponding
371
to the same standard solution containing the interference. 6.00 mg L-1 of Cu
372
(II), 0.06 mg L-1 of Fe(III), 6.00 mg L-1 of Zn(II), 14.00 mg L-1 of Ba(II), 10.00 mg
373
L-1 of Ca(II) and 10.00 mg L-1 of Mg(II) were tested. Only Fe(III) presented
374
interference. This fact was solved adding a 5 x 10-3 mol L-1 EDTA solution to the
AC C
EP
364
15
16
ACCEPTED MANUSCRIPT samples. In this case, it was important to check the formation of Cr(III)-EDTA
376
complex, because under this chemical form, there was not chemiluminescence
377
signal. The Cr(III)-EDTA complex is kinetically slow to form, and the other
378
cations form EDTA complex rapidly. Hamm (Hamm., 1953) studied the Cr (III)-
379
EDTA system and it was showed that the rate of complex formation was
380
independent of EDTA concentration and at a higher concentration of Cr(III) the
381
rate was proportional to this concentration. Besides, the rate of the formation of
382
the complex increased with increasing pH. Thus, a study taking into account
383
these considerations was performed. First of all, it is important to notice that in
384
this case, the Cr(III) concentration levels were low, and the samples were
385
acidified with a drop of 1 mol L-1 HNO3. Hence, the EDTA concentration was
386
tested between 1 x 10-2 and 1 x 10-3 mol L-1, working with the Cr(III) standard
387
solution. The best chemiluminescence signal was obtained with 5 x 10-3 mol L-1.
388
Then, the same standard solution containing different concentrations of Fe(III)
389
was prepared and the respective signals were measured. In this way, 6 mg L-1
390
of Fe(III) did not interfere. It is important to highlight that EDTA should be added
391
just prior to analysis, to avoid the possible formation of Cr(III)-EDTA complex
392
and to obtain a good reproducibility. Conversely, the anions present in the water
393
samples could be adsorbed on the soft material together with Cr(VI), so it was
394
necessary
395
concentrations of anions were added to a standard solution of Cr(VI) (0.08 mg
396
L-1). Thus, it was possible to demonstrate that the proposed method supported
397
4 mg L-1 of Cl- and NO3-, 3 mg L-1 of PO3= and 2.5 mg L-1 of SO4=. Due to the
398
analyzed samples usually contain higher concentration levels of SO4= than 2.5
AC C
EP
TE D
M AN U
SC
RI PT
375
to
test
these
possible
interferences.
Therefore,
different
16
17
ACCEPTED MANUSCRIPT 399
mg L-1, this interference was solved by adding BaCl2 to the samples and filtering
400
them with a 0.45 µm nylon filter.
401
3.5 Analysis of real samples
403
The proposed method was applied to three different water samples. These real
404
samples showed that both chromium species were not present in them, so they
405
were spiked with the respective standard solutions. A recovery study was
406
carried out and the same spiked samples were analyzed by inductively coupled
407
plasma optical emission spectrometry, in order to validate the method. Table 2
408
shows the obtained results and the respective recoveries which agreed with
409
those recommended for water samples. On the other hand, the obtained results
410
applying inductively coupled plasma optical emission spectrometry were in
411
accordance with those obtained with the proposed method.
SC
M AN U
TE D
412
RI PT
402
4. Concluding remarks
414
A new automatic FBA method with chemiluminescence detection was
415
developed to determine Cr(VI) and Cr(III). The speciation of both analytes was
416
performed by using a MFS which contained a minicolumn packed with a soft
417
material. One of the novelties of this work is the use of a soft material
418
composed by MWCNTs and IL that allows the separation of both chromium
419
species. Cr(VI) was retained on the soft material and Cr(III) passed through the
420
minicolumn. Besides, it is the first time that the chemiluminescence technique
421
was used to determine both species, Cr(VI) as oxidant of luminol reaction and
422
Cr(III) as catalyst of the luminol and hydrogen peroxide reaction using the same
AC C
EP
413
17
18
ACCEPTED MANUSCRIPT concentration of luminol and pH value. This was an additional advantage when
424
FBA systems were used.
425
The new method was applied to real water samples and the obtained results
426
presented good recoveries and were in agreement with those obtained with
427
inductively coupled plasma optical emission spectrometry method.
RI PT
423
428
Acknowledgment
430
The authors acknowledge financial support of Universidad Nacional del Sur. C.
431
Fernández, C. Domini and M. Grünhut also acknowledge to CONICET (Consejo
432
Nacional de Investigaciones Científicas y Técnicas).
M AN U
SC
429
433
Conflict of interest
435
The authors declare that they have no conflict of interest
438 439 440 441 442 443
EP
437
AC C
436
TE D
434
444 445 446 447 18
19
ACCEPTED MANUSCRIPT References
449
Andersen, J.E.T. 1998. Introduction of hydrogen peroxide as an oxidant in flow
450
injection analysis: speciation of Cr(III) and Cr(VI). Anal. Chim. Acta. 361, 125-
451
131.
452
Bose, M., 1954.The reaction of chromate with diphenylcarbazide. Anal. Chim.
453
Acta. 10, 201-208.
454
Diniz, P.G.D., Almeida, L.F., Harding, D.P., Araújo, M.C.U. 2012. Flow-batch
455
analysis. Trends Anal. Chem. 35, 39-49.
456
Feist, B., Mikula, B., 2014. Preconcentration of heavy metals on activated
457
carbon and their determination in fruits by inductively coupled plasma optical
458
emission spectrometry. Food Chem.147, 302-306.
459
Grünhut, M., Martins, V.L., Centurión, M.E., Araújo, M.C.U., Fernández Band,
460
B.S., 2011. Flow-batch analyzer for the chemiluminescence determination of
461
catecholamines in pharmaceutical preparations. Anal. Lett. 44, 67-81.
462
Hamm,
463
ethylenediaminetetraacetic acid complex with chromium (III). J. Amer. Chem.
464
Soc. 75, 5670-5672.
465
Lin, Y.A., Jiang, S.J., Sahayam, A.C., Huang, Y.L., 2016. Speciation of
466
chromium in edible animal oils after microwave extraction and liquid
467
chromatography inductively coupled plasma mass spectrometry. Microchem.
468
J.128, 274-278.
469
Lu, F., Zhang, S., Zheng, L., 2012. Dispersion of multi-walled carbon nanotubes
470
(MWCNTs) by ionic liquid-based phosphonium surfactants in aqueous solution.
471
J. Mol. Liq.173, 42-46.
1953.
Complex
Ions
of
Chromium
VI.
The
AC C
EP
R.E.,
TE D
M AN U
SC
RI PT
448
19
20
ACCEPTED MANUSCRIPT 472
Martinis, E. M., Grijalba, A. C., Pérez, M. B., Llaver, M., Wuilloud, R. G. 2017.
473
Synergistic
474
hybrids. TrAC Trends in Analytical Chemistry. 37, 333-344.
475
Miller, J.C., Miller, J.N. Statistics and Chemometrics for Analytical Chemistry,
476
Sixth Edition, Pearson Education Limited, Harlow, 2010.
477
Monasterio, R.P., Lascalea, G.E., Martinez, L.D., Wuilloud, R. G., 2009.
478
Determination of Cr(VI) and Cr(III) species in parenteral solutions using a
479
nanostructured
480
absorption spectrometry. J. Trace Elem. Med Biol. 23, 157-166.
481
Mubarak, N. M., Faghihzadeh, A., Tan, K. W., Sahu, J. N., Abdullah, E. C.,
482
Jayakumar, N. S., 2017. Microwave Assisted Carbon Nanofibers for Removal of
483
Zinc and Copper from Waste Water. J. Nanosci. Nanotechnol. 17(3), 1847-
484
1856.
485
Mubarak, N. M., Sahu, J. N., Abdullah, E. C., Jayakumar, N. S., Ganesan, P.,
486
2015. Novel microwave-assisted multiwall carbon nanotubes enhancing Cu (II)
487
adsorption capacity in water. J. Taiwan Inst. Chem. Eng. 53, 140-152.
488
Mubarak, N.M., Sahu, J.N., Abdullah, E.C., Jayakumar, N.S., 2016. Rapid
489
adsorption of toxic Pb(II) ions from aqueous solution using multiwall carbon
490
nanotubes synthesized by microwave chemical vapor deposition technique.
491
J.Environ.Sci. 45, 143-145.
492
Polo-Luque, M.L., Simonet, B.M., Valcárcel, M., 2013. Ionic liquid combined
493
with carbon nanotubes: A soft material for the preconcentration of PAHs.
494
Talanta.104, 169-172.
495
Ren, Y., Fan, Z., Wang, J., 2007. Speciation analysis of chromium in natural
496
water samples by electrothermal atomic absorbance spectrometry after
497
separation
498
immobilized on silica gel. Microchim. Acta.158, 227-231.
preconcentration
packed-microcolumn
ionic
and
liquid-nanomaterial
electrothermal atomic
AC C
EP
TE D
M AN U
SC
material
with
RI PT
analytical
/
preconcentration
with
nanometer-sized
zirconium
oxide
20
21
ACCEPTED MANUSCRIPT 499
Saçmaci, Ş., Kartal, Ş., Yilmaz, Y., Saçmaci, M., Soykan, C., 2012. A new
500
chelating resin: synthesis, characterization and application for speciation of
501
chromium (III)/(VI) species. Chem. Eng. J. 181, 746-753.
502
Séby, F., Charles, S., Gagean, M., Garraud, H., Donard, O.F.X., 2003.
503
Chromium
504
chromatography to inductively coupled plasma-mass spectrometry-study of the
505
influence of interfering ions. J. Anal. At. Spectrom.18, 1386-1390.
506
Seitz, W.R., Suydam, W.W., Hercules, D.M.,1972. Determination of trace
507
amounts of chromium(lll) using chemiluminescence analysis. Anal. Chem. 44,
508
957-963.
509
Shah, P., Strezov, V., Nelson, P.F., 2012. Speciation of chromium in Australian
510
coals and combustion products. Fuel.102, 1-8.
511
Tarigh, G.D., Shemirani, F., 2013. Magnetic multi-wall carbon nanotube
512
nanocomposite as an adsorbent for preconcentration and determination of lead
513
(II) and manganese (II) in various matrices. Talanta. 115, 744-750.
514
Thines, R. K., Mubarak, N. M., Nizamuddin, S., Sahu, J. N., Abdullah, E. C.,
515
Ganesan, P., 2017. Application potential of carbon nanomaterials in water and
516
wastewater treatment: A review. J. Taiwan Inst. Chem. Eng. 72, 116–133.
517
Tokalioğlu, Ş., Papak, A., Kartal, Ş., 2017. Separation/preconcentration of trace
518
Pb (II) and Cd (II) with 2-mercaptobenzothiazole impregnated Amberlite XAD-
519
1180 resin and their determination by flame atomic absorption spectrometry.
520
Arabian J. Chem.10, 19-23.
521
Tokalıoğlu, Ş., Yavuz, E., Şahan, H., Çolak, S.G., Ocakoğlu, K., Kaçer, M.,
522
Patat, Ş., 2016. Ionic liquid coated carbon nanospheres as a new adsorbent for
523
fast solid phase extraction of trace copper and lead from sea water, wastewater,
524
street dust and spice samples. Talanta. 159, 222-230.
by
hyphenation
of
high-performance
liquid
AC C
EP
TE D
M AN U
SC
RI PT
speciation
21
22
ACCEPTED MANUSCRIPT 525
Tuzen, M., Soylak, M., 2007. Multiwalled carbon nanotubes for speciation of
526
chromium in environmental samples. J. Hazard. Mater.147, 219-225.
527
Wang, J., Chu, H., Li, Y., 2008. Why single-walled carbon nanotubes can be
528
dispersed in imidazolium-based ionic liquids. ACS Nano. 2, 2540-2546.
RI PT
529 530 531
SC
532 533
M AN U
534 535 536 537
541 542 543 544 545
EP
540
AC C
539
TE D
538
546 547 548 549 22
23
ACCEPTED MANUSCRIPT Figure captions
551
Figure 1. MFS for chromium speciation. A: conditioner; B: sample; C: water;
552
C1: channel 1; D: air; E: eluent; MWCNTs/IL: Multiwalled carbon nanotubes/ionic
553
liquid; V: solenoid valve; PP: peristaltic pump.
554
Figure 2. a) FBA system for Cr(VI) and Cr(III) determination. C2 to C6: channel
555
2 to 6; F: luminol; G: hydrogen peroxide; H: water; I: waste; MDC: mixing
556
detection chamber; MS: magnetic stirrer; V: solenoid valve; PP: peristaltic
557
pump; PPA: auxiliary peristaltic pump. b) MDC coupled to the stirring system. c)
558
MDC and the stirring system coupled to the detector.
559
Figure 3. X-Ray Energy Dispersive Microanalysis: (a) five different analyzed
560
areas, (b) analysis of area 3.
561
Figure 4. SEM images of soft material: (a) before chromium adsorption (20000x
562
and 50000x), (b) after analyte adsorption (20000x and 50000x).
563
Figure 5. The mechanism of chromium adsorption on the soft material.
SC
M AN U
TE D
EP AC C
564
RI PT
550
23
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 1. Studied ranges and optimum values for chromium speciation variables. Studied ranges
Optimum values
Load flow rate (mL min-1)
0.80 - 4.04
1.21
Elution flow rate (mL min-1)
3.03 - 8.08 5 - 15
Elution volume (µL)
50 - 500
Stop-flow time (min)
5 -15
7.88 5
300 10
AC C
EP
TE D
M AN U
SC
Sample volume (mL)
RI PT
Variables
ACCEPTED MANUSCRIPT
Table 2. Analysis of real samples
Cr(III) [µg L-1]
Drinking water
Cr(VI) [µg L-1]
Added
Found*
Recovery (%)
Added
40
42 ± 1
105
10
60
58 ± 3
97
50
40
38 ± 2
95
10
60
62 ± 2
103
40
43 ± 3
107
60
58 ± 1
AC C
EP
TE D
*The results are average of five replicates
97
100
49 ± 2
98
10.2 ± 3
102
50
52 ± 5
104
10
9.5 ± 2
95
60 ± 4
102
M AN U
Filtered tap water
Recovery (%)
10 ± 1
SC
Tap water
Found*
RI PT
Sample
50
S0045-6535(17)32156-2
DOI:
10.1016/j.chemosphere.2017.12.178
Reference:
CHEM 20559
To appear in:
ECSN
Please cite this article as: Fernandez, C.J., Domini, C.E., Grünhut, M., Lista, A.G., A soft material for chromium speciation in water samples using a chemiluminescence automatic system, Chemosphere (2018), doi: 10.1016/j.chemosphere.2017.12.178. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights A sorbent based on MWCNTs and ionic liquid was used for the speciation of chromium
RI PT
The speciation and detection of chromium was performed by using an automatic system Cr(III) and Cr(VI) were detected by chemiluminescence reactions
AC C
EP
TE D
M AN U
SC
Cr(VI) was determined by the oxidation of luminol with the Cr2O7= desorbed
1
ACCEPTED MANUSCRIPT 1
A soft material for chromium speciation in water samples using a
2
chemiluminescence automatic system
3
Cintia J. Fernandez, Claudia E. Domini, Marcos Grünhut, Adriana G. Lista*
5
INQUISUR, Departamento de Química, Universidad Nacional del Sur
6
(UNS)-CONICET, Av. Alem 1253, B8000CPB, Bahía Blanca, Argentina
RI PT
4
7
Running title: A soft material for chromium speciation in an automatic system
SC
8
M AN U
9
Correspondence:
11
Prof. Adriana G. Lista
12
Departamento de Química, Universidad Nacional del Sur
13
Av. Alem 1253, B8000CPB, Bahía Blanca, Argentina
14
Fax: +54 0291 4595160
15
E-mail address: [email protected]
TE D
10
EP
16
Abbreviations: BmimCl: 1-Butyl-3-methylimidazolium-chloride; C: channel;
18
CNT: carbon nanotubes; EA: electronic actuator; FBA: Flow-batch analysis; IL:
19
Ionic liquids; MDC: mixing detection chamber; MFS: multicommutated flow
20
system; MWCNT: Multiwalled carbon nanotubes; V: solenoid valve; PP:
21
peristaltic pump.
AC C
17
22 23
Keywords: Automatic system, Chromium speciation, Ionic liquids, Multiwalled
24
carbon nanotubes, Chemiluminescence detection
25 1
2
ACCEPTED MANUSCRIPT Abstract
27
A soft material formed by multiwall carbon nanotubes and 1-butyl-3-methyl
28
imidazolium chloride was used as sorbent material to perform the chromium
29
speciation in natural waters. This soft material was not yet used for the
30
speciation of metals as chromium. Thus, a multicommutated flow system
31
containing a minicolumn packed with the soft material was designed. The
32
procedure was based on the capacity of the sorbent to retain Cr(VI) as Cr2O7=
33
and allow to pass Cr(III) through the column. Then, a fully automated flow-batch
34
analysis
35
chemiluminescence detection. Thus, Cr(III) was determined as catalyst of the
36
luminol and hydrogen peroxide reaction and Cr(VI) as oxidant of luminol
37
reaction. This represents a new approach because the oxidation of luminol
38
using Cr2O7= has not been reported in literature. The variables of the two
39
systems were optimized. The limits of detection were 1.4 µg L-1 for Cr(VI) and
40
4.0 µg L-1 for Cr(III). The precision of the method was 3.8% and 7.0% for Cr(VI)
41
and Cr(III), respectively. The present method was applied to real water samples
42
with recoveries between 95% and 107%. Besides, these results were in
43
accordance with those obtained using inductive coupled plasma-optical
44
emission spectrometry technique.
46
developed
to
SC
was
quantify
both
species
using
EP
TE D
M AN U
system
AC C
45
RI PT
26
47 48 49 50 2
3
ACCEPTED MANUSCRIPT 1. Introduction
52
Chromium speciation is always important because of the bioavailability and
53
toxicity of some of its chemical forms. Even though there are many oxidation
54
states of chromium, the most commonly found are Cr(III) and Cr(VI), especially
55
as environment pollutants. While Cr(III) is considered an essential trace element
56
for preservation of glucose, lipids and protein metabolism (Shah et al., 2012),
57
Cr(VI) is irritant and toxic to human body tissue due to its water solubility and its
58
oxidizing potential. It has been demonstrated that inorganic species of Cr(VI)
59
are carcinogenic (Saçmaci et al., 2012). For these reasons, the determination of
60
the different species of chromium presents greater interest than the
61
determination of total chromium. There are several methods to carry out the
62
speciation of chromium in environmental samples applying different techniques.
63
In most of them, the direct determination of Cr(VI) is proposed and Cr(III) is
64
determined after its oxidation to Cr(VI) (Tuzen., Soylak., 2007). Ren et al.
65
present the determination of Cr(III) by electrothermal atomic spectrometry with
66
the reduction of Cr(VI) to Cr(III) (Ren et al., 2007). On the other hand,
67
separative techniques as ion chromatography, capillary electrophoresis and LC-
68
ICP-MS also have been used (Séby et al., 2003, Lin et al., 2016) but all these
69
methods need a pretreatment step to remove the analytes.
70
environmental pollutants such as heavy metal and organic compounds have
71
been extracted using solid phase extraction with different sorbent materials.
72
These materials, like activated carbon (Feist et al., 2014), resins (Tokalioğlu et
73
al., 2017) carbon nanofibers (Mubarak et al., 2017), and carbon nanotubes
74
(CNTs) (Tarigh et al., 2013; Thines et al., 2017; Mubarak et al., 2015) have an
75
outstanding role in the enrichment efficiency of the analytes. It is well known the
AC C
EP
TE D
M AN U
SC
RI PT
51
Usually,
3
4
ACCEPTED MANUSCRIPT excellent properties of carbon nanotubes as sorbent material but one of the
77
major disadvantages is their tendency to form aggregates. For this reason,
78
different alternatives were proposed to disperse the nanotubes using different
79
solvents such as surfactants, organic solvents and ionic liquids (ILs) (Lu et al.,
80
2012). In recent years, ILs have been combined with sorbent materials to
81
improve the efficiency and selectivity of the extraction procedure (Tokalıoğlu et
82
al., 2016). The role of ILs is to enhance the dispersion of CNTs, creating in this
83
way a new kind of sorbent material called soft material. This material was used
84
to extract and preconcentrate organic analytes (Polo-Luque et al., 2013). Until
85
now, to the best of our knowledge, there are not reports in which this soft
86
material was applied to evaluate chromium speciation. Only a few works report
87
this speciation in a previous step to the determination of analytes. One of these
88
uses nanostructured α-alumina (Monasterio et al., 2009) and another uses
89
MWCNTs for the adsorption of Cr(VI)-ammonium pyrrolidine dithiocarbamate
90
chelate (Tuzen., Soylak., 2007). On the other hand, up to now, the use of
91
chemiluminescence technique for determining both species of chromium has
92
not been reported. Seitz et al. described the determination of Cr(III) by
93
chimiluminescence
94
reaction
95
chemiluminescence technique are high selectivity and specificity. Moreover, it
96
presents a high compatibility and easy coupling to automatic systems. The
97
automatic systems based on multicommutation and flow-batch analysis are
98
interesting alternatives to make feasible a method or to transform batch
99
processes into automated processes. Both MFS and FBA systems may perform
100
different steps in the analytical procedure, such as separation and/or
AC C
EP
TE D
M AN U
SC
RI PT
76
(Seitz
et
analysis using the analyte as catalyst of the luminol al.,
1972).
Some
of
the
advantages
of
using
4
5
ACCEPTED MANUSCRIPT preconcentration of analytes and determination that may include the coupling to
102
conventional analytical instruments. Conformation and advantages of FBA
103
systems have been described in literature (Diniz et al., 2012). Particularly, the
104
complete programmability of the process (including, control of sampling, flow
105
rates, filling of mixing chamber, detection and evacuation) and the possibility to
106
perform the measurements of the analytical signal directly inside the mixing
107
chamber can be highlighted. Both are important facts for the automation of
108
methods where chemiluminescence detection is used (Grünhut et al., 2011).
109
In this work, a new analytical method to determine Cr(VI) and Cr(III) in water
110
samples was developed. For this purpose, a MFS to perform the speciation of
111
chromium, which included a minicolumn packed with a soft material, was
112
designed. The soft material, composed of MWCNTs and IL, improved the
113
absorption of Cr(VI) as Cr2O7= in comparison to MCNTs, due to the synergic
114
effect between them. Additionally, this soft material allowed the separation of
115
both chromium species. Then, a FBA system with chemiluminescence detection
116
was developed to carry out the quantitation of Cr(VI), as oxidant of luminol, and
117
Cr(III) as catalyst in the reaction of the luminol and hydrogen peroxide, both in a
118
basic medium. According to our knowledge, this is the first time that the
119
chemiluminescence technique is used to determine both analytes, particularly
120
using the same automatic system.
122
SC
M AN U
TE D
EP
AC C
121
RI PT
101
2. Materials and Methods
123 124
2.1 Reagents and Solutions
5
6
ACCEPTED MANUSCRIPT All solutions were prepared by using analytical-grade reagents and ultra-pure
126
water (18MΩ cm-1) from a Milli-Q System (Millipore, USA). A 0.2 mol L-1
127
potassium chloride (Merck) solution was mixed with a 0.2 mol L-1 sodium
128
hydroxide (Merck) solution in order to obtain a buffer solution of pH 12.00. A
129
0.01 mol L-1 luminol stock solution was prepared by dissolving 0.1770 g of
130
luminol (Sigma-Aldrich) with 100 mL of the buffer solution. This solution was
131
stored in a dark bottle at 4°C and it was stable fo r one week. The luminol
132
working solution (5x10-3 mol L-1) was daily prepared by appropriate dilution of
133
stock solution in buffer solution. Hydrogen peroxide solution was prepared by
134
diluting of 100 v/v H2O2 solution to obtain a 0.01 mol L-1 work solution. A 5x10-3
135
mol L-1 EDTA solution was prepared by diluting of 0.1 mol L-1 stock solution
136
(Merck Titrisol). Stock solutions of Cr(III) and Cr(VI), both of 100 mg L-1, were
137
prepared by dissolving Cr(NO3)3.9H2O (Mallinckrodt) and K2Cr2O7 (Sigma-
138
Aldrich) in water, respectively. These solutions were acidified with nitric acid.
139
Standard solutions of Cr(III) and Cr(VI) were freshly prepared by dilution of the
140
stock solutions. MWCNTs with external diameters of 40-60 nm and purity >95%
141
(Bayer) and a 0.5 mol L-1 BmimCl (Merck) solution were used to perform the
142
speciation of chromium.
143
AC C
EP
TE D
M AN U
SC
RI PT
125
144
2.2 Instrumentation and Software
145
An
146
photomultiplier operated at 900 V was used. The emission of photons was
147
registered at 425 nm. A Gilson® Minipuls 3 and an Ismatec® peristaltic pumps
148
and Tygon® tubes were used to pump the fluids. The rest of the tubing was
149
made of Teflon® (0.5 mm i.d.). NResearch® three-way solenoid valves were
Aminco Bowman®
Series
2
luminescence
spectrometer
with
the
6
7
ACCEPTED MANUSCRIPT used to handle all solutions into the systems. The MDC was constructed with
151
Teflon® and was designed in our laboratory in order to prepare solutions and
152
perform the detection in the same place. A lab-made stirrer system was
153
designed to improve the mixing inside the MDC. An EA connected to a
154
Pentium® 4 microcomputer was used to control all the components of the
155
systems. The software used for controlling both, MFS and FBA systems, was
156
developed in Labview® 5.1 visual programming language. A model 710 A
157
Orion® pHmeter with an Orion-Ross® model 81-02 electrode was used to
158
perform the pH measurements.
159
The composition of the soft material was examined by X-Ray Energy Dispersive
160
Microanalysis (EDS) (X-Max50, Oxford). The physical structure and morphology
161
of the soft material, before and after the sorption of Cr(VI), were analyzed by the
162
Scanning Electron Microscope (SEM) (Leo, 40 XVP, England). In both cases,
163
the sample preparation was carried out metallizing it with gold in a Sputter
164
Coater (Pelco 91000).
SC
M AN U
TE D
EP
165
RI PT
150
2.3 Chromium Speciation Procedure
167
A soft material based on the combination of BmimCl and MWCNTs was
168
prepared to perform the speciation of chromium. A MFS that included a
169
minicolumn packed with this material was developed (Fig. 1). The soft material
170
was prepared weighing 0.1 g of MWCNTs and adding 0.5 mL of BmimCl. This
171
mixture was homogenized in a mortar and it was kept at 4º C for 24 h. Then, a
172
Tygon® tube (length: 30 mm; inner diameter: 4mm) was used to pack this
173
material. The MFS consisted of two three-way solenoid valves (V1 and V2) and
174
only one channel (C1) where the minicolumn was inserted. As it can be seen in
AC C
166
7
8
ACCEPTED MANUSCRIPT Fig. 1, different steps were performed in the MFS. When V1 and V2 valves were
176
OFF, the soft material was conditioned passing 1 min of water and 2.5 min of
177
0.05 mol L-1 of BmimCl solution through it. Then, 5.0 mL of standard/sample
178
solution was loaded and a vial 1 was placed at the end of the channel. At this
179
time, the Cr(III) contained in the standard/sample passed through the column
180
and it was collected in the vial 1, while the Cr(VI) was retained on the soft
181
material. After this, the vial 1 was replaced by a waste vial and the column was
182
washed with 1min of water and then 1 min of air was passed through it to
183
remove the water. All these steps were performed at 0.85 mL min-1. Finally, the
184
elution step was performed at 1 mL min-1 by turning ON V1 and V2 valves and in
185
reversal flow mode. In this way, 300 µL of BmimCl alkaline solution with NaCl
186
passed through the column. When the volume of eluent was all inside the
187
column, the flow was stopped for 10 min. Then, the flow was restored to collect
188
the amount of Cr(VI) in vial 2, at the end of the channel.
SC
M AN U
TE D
189
RI PT
175
2.4 FBA system with chemiluminescence detection
191
A schematic diagram of the proposed FBA system is showed in Fig 2a. The
192
system consisted of five channels, through which luminol (C2), sample (C3),
193
hydrogen peroxide (C4), water (C5) and waste (C6) flowed. The direction of the
194
flow in the C2 to C5 channels was controlled by three-way solenoid valves. Then,
195
when the valves (V3 to V6) were OFF, the solutions were recycled to the
196
respective flasks. When the valves were ON, the respective solutions were
197
pumped to the MDC. As it can be seen in Fig. 2b, the lab-made MDC has been
198
designed with four inlets for incoming solutions and one output for emptying. It
199
was equipped with a quartz window, which was located at the bottom part of the
AC C
EP
190
8
9
ACCEPTED MANUSCRIPT chamber in order to minimize the final volume of the reaction mixture (1.50 mL).
201
In this way, the MDC was used to perform both the chemical reaction and the
202
chemiluminescence detection. In addition, a little magnetic stirring system was
203
constructed using a cooler fan motor obtained from an Intel® microprocessor
204
and placed below the MDC (Fig. 2b). The MDC coupled to the stirring system
205
was placed in the spectrofluorimeter by means of a lab-made cell holder (Fig.
206
2c). On the other hand, the fifth channel (C6) was used for emptying the MDC
207
by means of an auxiliary peristaltic pump (PPA) Thereby, when the
208
chemiluminescence reaction finished, the solution in the MDC was pumped to
209
the waste.
M AN U
SC
RI PT
200
210
2.5 FBA procedure
212
As an initial step, the flow rate for each channel was studied in order to
213
establish the time of activation of each solenoid valve. These times
214
corresponded to the precise volumes of solution that were introduced into the
215
MDC. For this purpose, each channel was filled with bidestilated water and the
216
water was collected and weighed in an analytical balance by activating the
217
peristaltic pump for 30 s. The corresponding volume was calculated by means
218
of the density data taking into account the experimental conditions (25 ºC and 1
219
atm). Finally, the flow rates obtained for each channel were 2.30, 2.60, 2.65,
220
2.30 and 2.28 mL min−1 for luminol (C2), sample (C3), hydrogen peroxide (C4),
221
water (C5) and waste (C6), respectively. Before starting the procedure, the
222
valves were set in a way that all solutions in their respective channels were
223
recycled toward their flasks (OFF position). Then, V3, V4, V5 and V6 valves were
224
switched ON for 10 s and the solutions were pumped toward the MDC in order
AC C
EP
TE D
211
9
10
ACCEPTED MANUSCRIPT to fill the channels between the valves and the MDC. Immediately, the PPA was
226
switched ON and the excess of the solutions contained in the MDC
227
wasaspirated to the waste for 20 s. This operation consumed a total time of 30
228
s. Then, the system was ready to perform the preparation of the calibration for
229
Cr(III) and Cr(VI) and the sample analysis. In the case of Cr(III) determination,
230
the analysis of standard solutions and the samples was performed by
231
sequentially switching ON valves V3, V4 and V5 during previously defined
232
intervals of time (26, 7 and 7 s, respectively). Thus, 1.00 mL of luminol, 0.30 mL
233
of Cr (III) and 0.20 mL of hydrogen peroxide were pumped toward the MDC. For
234
Cr (VI) determination, the analysis of standard solutions and the samples was
235
performed by sequentially switching ON valves V3 and V4 during previously
236
defined intervals of time (31 and 7 s, respectively). In this way, 1.20 mL of
237
luminol and 0.30 mL of Cr (VI) were pumped toward the MDC. In both cases,
238
the corresponding chemiluminescence signal at 425 nm was immediately
239
measured as peak height. When the register of the signal was finished, the
240
solution was aspirated and thrown into the waste by switching ON the PPA for
241
45 s. The magnetic stirring was ON during all the analysis. This procedure was
242
repeated for each Cr (III) and Cr (VI) standard solution and each sample. On
243
the other hand, the system was always cleaned between measurements. The
244
MDC cleaning procedure was carried out by switching ON V6 valve and the
245
magnetic stirring for 40 s. The total emptying of the MDC was assured by
246
switching ON the PPA for 45 s.
AC C
EP
TE D
M AN U
SC
RI PT
225
247 248 249
3. Results and discussion 10
11
ACCEPTED MANUSCRIPT 3.1 Characterization of the soft material before and after adsorption of
251
Cr(VI)
252
The results obtained by EDS (Fig. 3) showed that the soft material was free
253
from impurities. The five areas marked in figure 3a showed the same results, as
254
it was expected. In the analysis the only elements found were the
255
corresponding ones to the composition of the soft material: carbon, oxygen,
256
nitrogen and chloride as it can be seen in Fig. 3b. (corresponding to the
257
analysis of area 3 in figure 3a). On the other hand, Fig. 4 shows the physical
258
structure and morphology of the soft material before and after adsorption of
259
Cr(VI) obtained by SEM. Smooth MWCNTs were observed forming an
260
intertwined pattern, probably due to its combination with the ionic liquid (Fig. 4
261
a). Besides, the morphology of the soft material shows that the MWCNTs have
262
uniform diameters between 35 and 45 nm, which demonstrate that the soft
263
material is homogeneous. In general, the tubes became wider, especially at the
264
ends, after the adsorption process (Fig. 4 b), showing slightly bigger diameters
265
between 47 and 52 nm. Also, the surface of the tubes was not as smooth
266
(Mubarak et al., 2016) and looked more compact, probably due to the
267
incorporation of chromium into the material.
SC
M AN U
TE D
EP
AC C
268
RI PT
250
269
3.2 Optimization of speciation variables
270
In this work, a soft material was prepared to perform the speciation of
271
chromium. A combination of BmimCl and MWCNTs was used instead of
272
MWCNTs, because the soft material presented more efficiency to adsorb Cr(VI)
273
as Cr2O7= , while Cr(III) passed through the column. This fact is based on the
274
synergic effect between the IL and the CNTs. It has been demonstrated that in 11
12
ACCEPTED MANUSCRIPT a soft material, CNTs are surrounded by ILs, taking into account that the atoms
276
of the alkyl groups are nearer to CNTs surface than the nitrogen and carbon
277
atoms of imidazolium rings (Wang et al, 2008). A physisorption phenomenon is
278
produced, where electrostatic interactions and Van del Waals forces take place
279
on the MWCNTs surface and charged ion clusters are surrounding this surface
280
(Martinis et al., 2017). In this way, the anion of IL (Cl-) is interchanged with
281
Cr2O7= and the analyte is retained on the soft material, while Cr(III) passes
282
through the column. This mechanism is showed in Fig. 5. In order to validate
283
this selectivity of the soft material, the typical spectrophotometric reaction for
284
Cr(VI) with 1,5-diphenylcarbazide was used (Bose, 1954). Cr(VI) was
285
determined after its elution and Cr(III) was determined after oxidation with
286
hydrogen peroxide (Andersen, 1998). The obtained results demonstrated that
287
Cr(III) was not retained on the soft material.
288
As mentioned before, the separation process was carried out in the minicolumn
289
inserted in a channel of the MFS, packed with the soft material. So, it was
290
necessary to optimize all variables of the MFS. Table 1 shows the range of all
291
these variables and their optimum values. On the other hand, the direction of
292
elution and type of eluent were also optimized. For these purposes, three
293
solutions were tested: IL, alkaline IL and alkaline IL with NaCl. The pH of
294
alkaline IL solution was varied between 8.00 and 12.00 and the best recoveries
295
were obtained at pH 11.00. At this pH, the predominant specie of Cr(VI) is
296
CrO4=. So it was necessary to add a drop of concentrated nitric acid to the
297
eluate, in order to obtain Cr2O7= and then get it into the FBA system. Then, to
298
enhance the recovery values, 0.01 g mL-1 of NaCl was added to test the salting
299
out effect and effectively, the recoveries were better. Taking into account the
AC C
EP
TE D
M AN U
SC
RI PT
275
12
13
ACCEPTED MANUSCRIPT contact time of the eluent with the analyte, it was supposed that the best results
301
would be obtained with low flow rates for the elution. However, the best
302
chemiluminescence signals were obtained when higher flow rates were used
303
and a stop flow step was included. The results were improved when the elution
304
was performed in a reverse mode regarding the sample load direction. This
305
elution mode avoids the tight packing of the sorbent and decreases the
306
dispersion of the concentrate during the elution. Thus, the selected optimum
307
values are showed in Table 1.
308
Moreover, the soft material needed a conditioning step. For this purpose, a 0.05
309
mol L-1 BmimCl solution, with and without NaOH, was passed through the
310
column during different times (1 min to 3 min). The IL solution without NaOH
311
(pH = 5.00) flowing for 2.5 min gave the best results in terms of recovery values
312
for Cr(VI). Besides, two different lengths of the column were tested (15 and 30
313
mm) and the best results, based on the best chemiluminescence signals, were
314
obtained with the longer column.
SC
M AN U
TE D
EP
315
RI PT
300
3.3 Optimization of the chemiluminescence reactions and the FBA system
317
The luminol in presence of an oxidant and in alkaline medium produces a
318
chemiluminescence signal at 425 nm. So, using Cr2O7= as oxidant it was
319
possible to determine Cr(VI). According to the literature, this is the first time that
320
Cr(VI) is determined by chemiluminescence, using the analyte as oxidant for the
321
luminol reaction. Usually, Cr(VI) is reduced to Cr(III) and is determined as
322
catalyst of the luminol reaction. On the other hand, taking into account that the
323
chemiluminescence reaction of luminol can be catalyzed by different metals, the
324
determination of Cr(III) was performed using hydrogen peroxide as oxidant and
AC C
316
13
14
ACCEPTED MANUSCRIPT Cr(III) as catalyst. The concentration of all chemical reagents involved in both
326
reactions and the reaction media should be optimized. The luminol
327
concentration was studied between 1 x 10-4 mol L-1 and 1 x 10-2 mol L-1 and the
328
optimum value was 5 x 10-3 mol L-1 for both analytes. The studied pH range was
329
10.00-13.00. Again, the optimum value (12.00) was the same for the two
330
analytes. These facts represented an advantage in the determination of both
331
analytes. The other optimized variables were only for the determination of Cr(III)
332
and they were: H2O2 concentration (studied range: 1 x 10-4 mol L-1 and 3 x 10-2
333
mol L-1, optimum value: 1 x 10-2 mol L-1), EDTA concentration (studied range:1 x
334
10-3 mol L-1 and 1 x 10-2 mol L-1, optimum value: 5 x 10-3 mol L-1). Also, the
335
variables related to the FBA system were optimized in order to obtain the best
336
signal. FBA variables such as order and volumes of the reagents added into the
337
MDC and stirring time were optimized based on the higher chemiluminescence
338
signal and the repeatability of the measurements obtained for both Cr(III) and
339
Cr(VI) determinations. As regards to the order of the reagents added into the
340
MDC, three different combinations were tested: luminol and oxidant, oxidant
341
and luminol, and luminol-oxidant simultaneously. The optimum order was
342
luminol and oxidant. On the other hand, the total volume of reaction was 1.50
343
mL, which included 1.00 mL of luminol, 0.20 mL of sample and 0.30 mL of
344
hydrogen peroxide for Cr(III) and 1.20 mL of luminol and 0.30 mL of sample for
345
Cr(VI).
346
corresponding oxidant (H2O2 for Cr (III) and Cr2O4= for Cr (VI)) was added so
347
fast as possible to the MDC (See section 2.5). Finally, the MDC coupled to the
348
stirring system was designed in order to obtain a rapid and efficient mixture of
349
the solutions. As the speed of stirring presented a significant effect in the
AC C
EP
TE D
M AN U
SC
RI PT
325
Optimal
chemiluminescence
signals
were
obtained
when
the
14
15
ACCEPTED MANUSCRIPT analytical signal, it was necessary to maintain a high and stable speed of
351
stirring (500 rpm) during the chemiluminescence reaction.
352
3.4 Analytical performance
353
The different analytical parameters were calculated using the optimized
354
proposed methods. The calibration curve for Cr(VI) was y = (59.396 ± 569) x +
355
(1968 ± 30), R2 = 0.9998, and for Cr(III) was y = (75.250 ± 3088) x – (1254 ±
356
162), R2 = 0.9960. In both cases, y: chemiluminescence signal and x: [µg L-1].
357
The limit of detection (LOD) and limit of quantitation (LOQ) for Cr(VI) were 1.4
358
µg L-1 and 38 µg L-1, respectively. For Cr(III) the LOD was 4 µg L-1 and the LOQ
359
was 12 µg L-1 (Miller and Miller; 2010). The precision of the method,
360
represented by repetitively (% RSD) was 7.0 and 3.8 for Cr(III) and Cr(VI)
361
respectively, calculated from eight independent measurements of a standard
362
solution containing 50 µg L-1 for both analytes.
363
TE D
M AN U
SC
RI PT
350
3.4.1 Selectivity
365
The effect of common cations and anions present in water samples in the
366
determination of 0.08 mg L-1 of Cr(VI) and 0.07 mg L-1 of Cr(III) was examined
367
applying the whole method. Different cations present in water samples could
368
catalyze the luminol reaction instead of Cr(III), so a study of these interferences
369
was performed. The first step in this study was to compare the concentration of
370
a standard solution of Cr(III) (0.07 mg L-1) with the concentration corresponding
371
to the same standard solution containing the interference. 6.00 mg L-1 of Cu
372
(II), 0.06 mg L-1 of Fe(III), 6.00 mg L-1 of Zn(II), 14.00 mg L-1 of Ba(II), 10.00 mg
373
L-1 of Ca(II) and 10.00 mg L-1 of Mg(II) were tested. Only Fe(III) presented
374
interference. This fact was solved adding a 5 x 10-3 mol L-1 EDTA solution to the
AC C
EP
364
15
16
ACCEPTED MANUSCRIPT samples. In this case, it was important to check the formation of Cr(III)-EDTA
376
complex, because under this chemical form, there was not chemiluminescence
377
signal. The Cr(III)-EDTA complex is kinetically slow to form, and the other
378
cations form EDTA complex rapidly. Hamm (Hamm., 1953) studied the Cr (III)-
379
EDTA system and it was showed that the rate of complex formation was
380
independent of EDTA concentration and at a higher concentration of Cr(III) the
381
rate was proportional to this concentration. Besides, the rate of the formation of
382
the complex increased with increasing pH. Thus, a study taking into account
383
these considerations was performed. First of all, it is important to notice that in
384
this case, the Cr(III) concentration levels were low, and the samples were
385
acidified with a drop of 1 mol L-1 HNO3. Hence, the EDTA concentration was
386
tested between 1 x 10-2 and 1 x 10-3 mol L-1, working with the Cr(III) standard
387
solution. The best chemiluminescence signal was obtained with 5 x 10-3 mol L-1.
388
Then, the same standard solution containing different concentrations of Fe(III)
389
was prepared and the respective signals were measured. In this way, 6 mg L-1
390
of Fe(III) did not interfere. It is important to highlight that EDTA should be added
391
just prior to analysis, to avoid the possible formation of Cr(III)-EDTA complex
392
and to obtain a good reproducibility. Conversely, the anions present in the water
393
samples could be adsorbed on the soft material together with Cr(VI), so it was
394
necessary
395
concentrations of anions were added to a standard solution of Cr(VI) (0.08 mg
396
L-1). Thus, it was possible to demonstrate that the proposed method supported
397
4 mg L-1 of Cl- and NO3-, 3 mg L-1 of PO3= and 2.5 mg L-1 of SO4=. Due to the
398
analyzed samples usually contain higher concentration levels of SO4= than 2.5
AC C
EP
TE D
M AN U
SC
RI PT
375
to
test
these
possible
interferences.
Therefore,
different
16
17
ACCEPTED MANUSCRIPT 399
mg L-1, this interference was solved by adding BaCl2 to the samples and filtering
400
them with a 0.45 µm nylon filter.
401
3.5 Analysis of real samples
403
The proposed method was applied to three different water samples. These real
404
samples showed that both chromium species were not present in them, so they
405
were spiked with the respective standard solutions. A recovery study was
406
carried out and the same spiked samples were analyzed by inductively coupled
407
plasma optical emission spectrometry, in order to validate the method. Table 2
408
shows the obtained results and the respective recoveries which agreed with
409
those recommended for water samples. On the other hand, the obtained results
410
applying inductively coupled plasma optical emission spectrometry were in
411
accordance with those obtained with the proposed method.
SC
M AN U
TE D
412
RI PT
402
4. Concluding remarks
414
A new automatic FBA method with chemiluminescence detection was
415
developed to determine Cr(VI) and Cr(III). The speciation of both analytes was
416
performed by using a MFS which contained a minicolumn packed with a soft
417
material. One of the novelties of this work is the use of a soft material
418
composed by MWCNTs and IL that allows the separation of both chromium
419
species. Cr(VI) was retained on the soft material and Cr(III) passed through the
420
minicolumn. Besides, it is the first time that the chemiluminescence technique
421
was used to determine both species, Cr(VI) as oxidant of luminol reaction and
422
Cr(III) as catalyst of the luminol and hydrogen peroxide reaction using the same
AC C
EP
413
17
18
ACCEPTED MANUSCRIPT concentration of luminol and pH value. This was an additional advantage when
424
FBA systems were used.
425
The new method was applied to real water samples and the obtained results
426
presented good recoveries and were in agreement with those obtained with
427
inductively coupled plasma optical emission spectrometry method.
RI PT
423
428
Acknowledgment
430
The authors acknowledge financial support of Universidad Nacional del Sur. C.
431
Fernández, C. Domini and M. Grünhut also acknowledge to CONICET (Consejo
432
Nacional de Investigaciones Científicas y Técnicas).
M AN U
SC
429
433
Conflict of interest
435
The authors declare that they have no conflict of interest
438 439 440 441 442 443
EP
437
AC C
436
TE D
434
444 445 446 447 18
19
ACCEPTED MANUSCRIPT References
449
Andersen, J.E.T. 1998. Introduction of hydrogen peroxide as an oxidant in flow
450
injection analysis: speciation of Cr(III) and Cr(VI). Anal. Chim. Acta. 361, 125-
451
131.
452
Bose, M., 1954.The reaction of chromate with diphenylcarbazide. Anal. Chim.
453
Acta. 10, 201-208.
454
Diniz, P.G.D., Almeida, L.F., Harding, D.P., Araújo, M.C.U. 2012. Flow-batch
455
analysis. Trends Anal. Chem. 35, 39-49.
456
Feist, B., Mikula, B., 2014. Preconcentration of heavy metals on activated
457
carbon and their determination in fruits by inductively coupled plasma optical
458
emission spectrometry. Food Chem.147, 302-306.
459
Grünhut, M., Martins, V.L., Centurión, M.E., Araújo, M.C.U., Fernández Band,
460
B.S., 2011. Flow-batch analyzer for the chemiluminescence determination of
461
catecholamines in pharmaceutical preparations. Anal. Lett. 44, 67-81.
462
Hamm,
463
ethylenediaminetetraacetic acid complex with chromium (III). J. Amer. Chem.
464
Soc. 75, 5670-5672.
465
Lin, Y.A., Jiang, S.J., Sahayam, A.C., Huang, Y.L., 2016. Speciation of
466
chromium in edible animal oils after microwave extraction and liquid
467
chromatography inductively coupled plasma mass spectrometry. Microchem.
468
J.128, 274-278.
469
Lu, F., Zhang, S., Zheng, L., 2012. Dispersion of multi-walled carbon nanotubes
470
(MWCNTs) by ionic liquid-based phosphonium surfactants in aqueous solution.
471
J. Mol. Liq.173, 42-46.
1953.
Complex
Ions
of
Chromium
VI.
The
AC C
EP
R.E.,
TE D
M AN U
SC
RI PT
448
19
20
ACCEPTED MANUSCRIPT 472
Martinis, E. M., Grijalba, A. C., Pérez, M. B., Llaver, M., Wuilloud, R. G. 2017.
473
Synergistic
474
hybrids. TrAC Trends in Analytical Chemistry. 37, 333-344.
475
Miller, J.C., Miller, J.N. Statistics and Chemometrics for Analytical Chemistry,
476
Sixth Edition, Pearson Education Limited, Harlow, 2010.
477
Monasterio, R.P., Lascalea, G.E., Martinez, L.D., Wuilloud, R. G., 2009.
478
Determination of Cr(VI) and Cr(III) species in parenteral solutions using a
479
nanostructured
480
absorption spectrometry. J. Trace Elem. Med Biol. 23, 157-166.
481
Mubarak, N. M., Faghihzadeh, A., Tan, K. W., Sahu, J. N., Abdullah, E. C.,
482
Jayakumar, N. S., 2017. Microwave Assisted Carbon Nanofibers for Removal of
483
Zinc and Copper from Waste Water. J. Nanosci. Nanotechnol. 17(3), 1847-
484
1856.
485
Mubarak, N. M., Sahu, J. N., Abdullah, E. C., Jayakumar, N. S., Ganesan, P.,
486
2015. Novel microwave-assisted multiwall carbon nanotubes enhancing Cu (II)
487
adsorption capacity in water. J. Taiwan Inst. Chem. Eng. 53, 140-152.
488
Mubarak, N.M., Sahu, J.N., Abdullah, E.C., Jayakumar, N.S., 2016. Rapid
489
adsorption of toxic Pb(II) ions from aqueous solution using multiwall carbon
490
nanotubes synthesized by microwave chemical vapor deposition technique.
491
J.Environ.Sci. 45, 143-145.
492
Polo-Luque, M.L., Simonet, B.M., Valcárcel, M., 2013. Ionic liquid combined
493
with carbon nanotubes: A soft material for the preconcentration of PAHs.
494
Talanta.104, 169-172.
495
Ren, Y., Fan, Z., Wang, J., 2007. Speciation analysis of chromium in natural
496
water samples by electrothermal atomic absorbance spectrometry after
497
separation
498
immobilized on silica gel. Microchim. Acta.158, 227-231.
preconcentration
packed-microcolumn
ionic
and
liquid-nanomaterial
electrothermal atomic
AC C
EP
TE D
M AN U
SC
material
with
RI PT
analytical
/
preconcentration
with
nanometer-sized
zirconium
oxide
20
21
ACCEPTED MANUSCRIPT 499
Saçmaci, Ş., Kartal, Ş., Yilmaz, Y., Saçmaci, M., Soykan, C., 2012. A new
500
chelating resin: synthesis, characterization and application for speciation of
501
chromium (III)/(VI) species. Chem. Eng. J. 181, 746-753.
502
Séby, F., Charles, S., Gagean, M., Garraud, H., Donard, O.F.X., 2003.
503
Chromium
504
chromatography to inductively coupled plasma-mass spectrometry-study of the
505
influence of interfering ions. J. Anal. At. Spectrom.18, 1386-1390.
506
Seitz, W.R., Suydam, W.W., Hercules, D.M.,1972. Determination of trace
507
amounts of chromium(lll) using chemiluminescence analysis. Anal. Chem. 44,
508
957-963.
509
Shah, P., Strezov, V., Nelson, P.F., 2012. Speciation of chromium in Australian
510
coals and combustion products. Fuel.102, 1-8.
511
Tarigh, G.D., Shemirani, F., 2013. Magnetic multi-wall carbon nanotube
512
nanocomposite as an adsorbent for preconcentration and determination of lead
513
(II) and manganese (II) in various matrices. Talanta. 115, 744-750.
514
Thines, R. K., Mubarak, N. M., Nizamuddin, S., Sahu, J. N., Abdullah, E. C.,
515
Ganesan, P., 2017. Application potential of carbon nanomaterials in water and
516
wastewater treatment: A review. J. Taiwan Inst. Chem. Eng. 72, 116–133.
517
Tokalioğlu, Ş., Papak, A., Kartal, Ş., 2017. Separation/preconcentration of trace
518
Pb (II) and Cd (II) with 2-mercaptobenzothiazole impregnated Amberlite XAD-
519
1180 resin and their determination by flame atomic absorption spectrometry.
520
Arabian J. Chem.10, 19-23.
521
Tokalıoğlu, Ş., Yavuz, E., Şahan, H., Çolak, S.G., Ocakoğlu, K., Kaçer, M.,
522
Patat, Ş., 2016. Ionic liquid coated carbon nanospheres as a new adsorbent for
523
fast solid phase extraction of trace copper and lead from sea water, wastewater,
524
street dust and spice samples. Talanta. 159, 222-230.
by
hyphenation
of
high-performance
liquid
AC C
EP
TE D
M AN U
SC
RI PT
speciation
21
22
ACCEPTED MANUSCRIPT 525
Tuzen, M., Soylak, M., 2007. Multiwalled carbon nanotubes for speciation of
526
chromium in environmental samples. J. Hazard. Mater.147, 219-225.
527
Wang, J., Chu, H., Li, Y., 2008. Why single-walled carbon nanotubes can be
528
dispersed in imidazolium-based ionic liquids. ACS Nano. 2, 2540-2546.
RI PT
529 530 531
SC
532 533
M AN U
534 535 536 537
541 542 543 544 545
EP
540
AC C
539
TE D
538
546 547 548 549 22
23
ACCEPTED MANUSCRIPT Figure captions
551
Figure 1. MFS for chromium speciation. A: conditioner; B: sample; C: water;
552
C1: channel 1; D: air; E: eluent; MWCNTs/IL: Multiwalled carbon nanotubes/ionic
553
liquid; V: solenoid valve; PP: peristaltic pump.
554
Figure 2. a) FBA system for Cr(VI) and Cr(III) determination. C2 to C6: channel
555
2 to 6; F: luminol; G: hydrogen peroxide; H: water; I: waste; MDC: mixing
556
detection chamber; MS: magnetic stirrer; V: solenoid valve; PP: peristaltic
557
pump; PPA: auxiliary peristaltic pump. b) MDC coupled to the stirring system. c)
558
MDC and the stirring system coupled to the detector.
559
Figure 3. X-Ray Energy Dispersive Microanalysis: (a) five different analyzed
560
areas, (b) analysis of area 3.
561
Figure 4. SEM images of soft material: (a) before chromium adsorption (20000x
562
and 50000x), (b) after analyte adsorption (20000x and 50000x).
563
Figure 5. The mechanism of chromium adsorption on the soft material.
SC
M AN U
TE D
EP AC C
564
RI PT
550
23
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 1. Studied ranges and optimum values for chromium speciation variables. Studied ranges
Optimum values
Load flow rate (mL min-1)
0.80 - 4.04
1.21
Elution flow rate (mL min-1)
3.03 - 8.08 5 - 15
Elution volume (µL)
50 - 500
Stop-flow time (min)
5 -15
7.88 5
300 10
AC C
EP
TE D
M AN U
SC
Sample volume (mL)
RI PT
Variables
ACCEPTED MANUSCRIPT
Table 2. Analysis of real samples
Cr(III) [µg L-1]
Drinking water
Cr(VI) [µg L-1]
Added
Found*
Recovery (%)
Added
40
42 ± 1
105
10
60
58 ± 3
97
50
40
38 ± 2
95
10
60
62 ± 2
103
40
43 ± 3
107
60
58 ± 1
AC C
EP
TE D
*The results are average of five replicates
97
100
49 ± 2
98
10.2 ± 3
102
50
52 ± 5
104
10
9.5 ± 2
95
60 ± 4
102
M AN U
Filtered tap water
Recovery (%)
10 ± 1
SC
Tap water
Found*
RI PT
Sample
50