Improved Solid phase Spectrophotometry for the Microdetermination of Chromium(VI) in Natural Water
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
1445
2009 © The Japan Society for Analytical Chemistry
improved solid-phase spectrophotometry for the
microdetermination of Chromium(Vi) in Natural Water
sulistyo saPUTRO,*1 Kazuhisa YOsHimURa,*2† shiro maTsUOKa,*3 Kô TaKEHaRa,*2 and Narsito*4
*1 Department of Mathematics and Natural Science Education, Faculty of Teacher Training and Education
Sciences, Sebelas Maret University, Surakarta-57126, Indonesia
*2 Department of Chemistry, Faculty of Sciences, Kyushu University, Hakozaki, Higashi,
Fukuoka 812–8581, Japan
*3 Department of Environmental Science, Faculty of Science, Niigata University, Ikarashi,
Niigata 950–2181, Japan
*4 Department of Chemistry, Faculty of Mathematics and Natural Sciences, Gadjah Mada University,
Yogyakarta-55281, Indonesia
A simple and sensitive solid-phase spectrophotometry procedure was improved for the microdetermination of Cr(VI). A
0.06 cm3 portion of a cation exchanger, Muromac AG 50W-X2, was used to concentrate the target Cr(VI) in a 20 cm3
water sample, and resin beads were introduced in a flow cell of 1.5 mm diameter and having a 10 mm light path length
for measurements using a UV-visible spectrophotometer. Three lenses were used for focusing the incident light beam and
for recovering light scattered by the solid phase in the cell. The sensitivity achieved was higher by a factor of 277
compared with that of the solution method, and the detection limit was 0.014 µg dm–3. The recovery on spiked real water
samples by the standard addition method was 96 – 101%. Favorable working and performance characteristics made it
possible to directly determine sub-µg dm–3 amounts of Cr(VI) in natural water samples.
(Received October 8, 2009; accepted November 7, 2009; Published December 10, 2009)
introduction
Chromium is a metal that occurs in oxidation states, ranging
from +2 to +6. However, as far as environmental protection is
concerned, only the two most common oxidation states, namely
Cr(III) and Cr(VI), need to be considered. Cr(III) and Cr(VI)
are drastically different in physicochemical properties as well as
chemical and biochemical reactivities. Cr(III) is well-known as
an essential trace element for humans, required for the
maintenance of normal glucose, cholesterol and fatty acid
metabolism. On the other hand, water-soluble Cr(VI) is highly
toxic to humans and animals,1 and other studies have indicated
that it is an extremely toxic carcinogen.2
Several types of wastewater, such as those discharged in the
course of dye, pigment and leather production, and by the
process of plating and electroplating and by mining may contain
undesirable amounts of Cr(VI).3 In principle, because the health
effects are determined largely by the oxidation states, different
guideline values for Cr(III) and Cr(VI) should be derived.
However, current analytical methods and the variable speciation
of chromium in water favor a guideline value for a total
chromium of 50 µg dm–3.4–6 If simple, sensitive and rapid
analytical techniques selective for trace Cr(VI) are applicable to
various environmental samples, the situation can be improved.
Although atomic spectrometries, such as flame atomic
†
To whom correspondence should be addressed.
E-mail: [email protected]
absorption spectrometry (FAAS), inductively coupled
plasma–atomic emission spectrometry (ICP-AES) and
inductively coupled plasma–mass spectrometry (ICP-MS) are
sensitive and selective for Cr, only the total amounts of Cr can
be determined despite the fairly expensive equipment.
Solid-phase spectrophotometry (SPS) is based on the direct
spectrophotometric measurement of a solid phase that has
sorbed a sample component. This method made it possible to
determine trace components in natural and other water samples
without preconcentration, because a sensitivity enhancement
was easily accomplished by increasing the sample volume.7,8
By employing diphenylcarbazide (DPC) as a coloring agent,
SPS selective for Cr(VI) has also been developed.9
In the case where the distribution ratio of the sample species
is high enough for complete adsorption, the absorbance of the
target colored analyte species in the solid phase (ARC) can be
expressed as Eq. (1) if the analyte, whose concentration (C0) in
a V cm3 sample solution is concentrated into v cm3 of the solid
phase,
ARC = εRCC0lRCV/v,
(1)
where εRC is the molar absorptivity of the colored species and
lRC the light path length in the solid phase. Alternative ways to
enhance the sensitivity of SPS by using a smaller volume of the
sample solution is to apply a larger lRC and/or smaller v systems.
As described in a later section, the use of a black flow-through
cell in which a small volume of solid particles is packed as an
adsorbent for the target chemical species is very effective for
1446
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Fig. 2 Cell compartment of a spectrophotometer for SPS. A, Light
detector; B, reference beam; C, perforate metal plate of attenuance 0.4
or 1.0 (Hitachi); D, sample beam; E, concave lens (focal distance
–200 mm, Sugitoh); F, cylindrical convex lens (focal distance 25.4 mm,
Sugitoh); G, black flow cell; H, focusing lens (Jasco); I, XY stage; J,
disposal syringe.
Fig. 1 Black flow cell for SPS. A, Inlet of the black cell connected
to a 10-cm3 PE syringe (SS-10SZ, Terumo) using PTFE and silicone
tubes; B, outlet of the black cell with inserted PTFE tube; C, light path
which is filled with ion-exchange resin (1.5 mm i.d. 10 mm long); D,
PP filter tip; E, incident light beam; F, three-way stopcock.
higher sensitivity.
However, different from conventional
solution spectrophotometry, SPS basically suffers from weak
light intensities due to light scattered by the solid phase.10 In
addition, attenuation of the incident light intensity by a black
flow cell also has to be considered if the area of the incident
light beam is wider than that of the light-path portion of the
black flow cell. For this reason, it is necessary to make some
improvement in the equipment or the light measurement system
to reduce the background light attenuance by the solid phase
and the cell, and this makes it difficult to apply this SPS method
widely to various trace analyses.
The aim of this study was to develop and evaluate a simple
and sensitive solid-phase spectrophotometric batch method
using a commercially available, simple spectrophotometer,
which would be applicable to the determination of sub-µg dm–3
amounts of Cr(VI) in natural water samples or drinking water.
The most interesting features of the improved method are
simplicity in operation, high sensitivity and fast application
without any previous sample treatment including many laborious
steps. This paper reports on the main characteristics of this
improved method and the optimum experimental conditions, as
well as the analytical application of the improved method to
natural water samples.
Experimental
Reagents
All reagents used were of analytical grade. Highly purified
water prepared with a Milli-Q SP system (Millipore, Milford,
MA) was used throughout. A standard Cr(VI) solution
(1000 mg dm–3) for atomic absorption spectrometry (Kishida,
Osaka, Japan) was used. A sulfuric acid solution (about
0.5 mol dm–3) was prepared by diluting 6.8 cm3 of concentrated
sulfuric acid with water up to 250 cm3. A coloring reagent
solution was prepared by dissolving 0.25 g of DPC
(diphenylcarbazide, Wako, Osaka, Japan) and diluting to
100 cm3 with acetone. A calcium solution (2000 mg dm–3) was
prepared by dissolving 0.74 g of CaCl2·2H2O (Kishida) in
100 cm3 of water. A sodium chloride solution (1 mol dm–3) was
prepared by dissolving 5.84 g of NaCl (Wako) and diluting to
100 cm3 with water. A standard solution of NaOH (0.1 mol dm–3,
Wako) was used for acid-base titration of the ion exchanger.
A Muromac 50W-X2 cation exchanger (100 – 200 mesh,
Muromachi, Tokyo, Japan) was used.
Apparatus
Absorbance measurements were made with a double-beam
UV-visible spectrophotometer (Model V-630, Jasco, Tokyo,
Japan). A flow cell Model FLM 220B-B-1.5 (Fig. 1) was
supplied from GL Science (Tokyo, Japan). It was black-sided,
10 mm in length and 1.5 mm in diameter. At the outlet of the
cell, a polytetrafluoroethylene (PTFE) tube (1 mm i.d.) was
connected to a silicone tube. Inside the end of the tube, a
polypropylene (PP) filter tip was placed so as to block the
ion-exchanger beads in order to pack them in the light-path
portion. At the sample light beam, a cylindrical convex lens
(focal distance 25.4 mm, Sugitoh, Tokyo, Japan) and a concave
lens (focal distance –200 mm, Sugitoh) were placed between
the incident light window and the cell for focusing the light
beam at the light-beam entrance of the cell, and the lens (Jasco)
at the detector window for recovering the scattered light. A
diagram of the cell holder for the determination of Cr(VI) is
shown in Fig. 2.
The ion exchanger was measured with an ion-exchanger
aliquotting device. A PTFE tube (1.0 mm i.d. and 7 cm long)
was fitted on one side with a PP resin filter tip and connected to
a 10 cm3 disposable syringe.11
Collection of natural water samples
Natural water samples (from No. 1 downstream to No. 7
upstream) were collected from a surface stream of the Ochozu
Experimental Watershed in Fukuoka Experimental Forest,
Kyushu University, in a mountainous watershed of Japanese
cypress located about 15 km east of Fukuoka City in western
Japan. The predominant forest soil is yellow-brown, and the
underlying bedrock consists of serpentinite containing chromite
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
1447
Table 1 Improvement of the light measurements by SPS using
lenses
Without lenses
With the lensesa;
without the Jasco
lensb
With the lenses; with
the Jasco lens
A540nmc
A700nmd
∆Ae
Over 4
2.434
2.382
2.312
1.931
1.936
1.924
Over 4
2.238
2.175
2.101
1.732
1.738
1.726
—
0.196
0.207
0.211
0.199
0.198
0.198
a. For focusing the incident light beam on the flow cell.
b. For recovering the scaterred light from the cell.
c. Absorbance at the wavelength characteristic of colored species.
d. Absorbance at the wavelength which the sample species does not
absorb.
e. The difference between the two of absorbances.
The light measurements were done against air. Cell: 1.5 mmf, 10 mm
long. Ion exchanger: Muromac AG 50W-X2, 100 – 200 mesh.
and chlorite schist.12 Water samples No. 8 and No. 9 were
collected from a surface stream near Ochozu Experimental
Watershed with a different geological condition. Water samples
No. 10 and No. 11 were collected from Karst springs on
Hirao-dai plateau, Kitakyushu, Fukuoka. Water samples were
also collected from the Central Java Province, Indonesia. Tlatar
(Tl) and Ngabean (Ng) are natural groundwater sources that
flow continuously, and are located at about 7 and 10 km,
respectively, from Boyolali, Central Java (near the Merapi
volcano). Tawangmangu (Tw) is a natural waterfall at 1100 m
in altitude, covered with forest, and located 30 km from
Surakarta, Central Java. The underlying bedrock consists of
volcanic rocks with lava and andesite breccias. The natural
water was filtered through a 0.20-µm membrane filter at the site
and stored without being acidified in a polyethylene (PE) bottle
for Cr(VI) determination. For total chromium determination,
the sample was stored in a PTFE bottle, and 1 cm3 of highly
purified nitric acid (Kishida, Japan) was added to a 100-cm3
sample in order to avoid the adsorption of Cr(III) on the
container wall. The pH and temperature of the water sample
were measured with a pH meter (HM-14P, TOA DKK) at the
sampling point, and the dissolved organic carbon (DOC) content
was measured with a TOC apparatus (TOC-VE, Shimadzu).
Determination of Cr(VI) by means of SPS
To a 20 cm3 water sample containing 1.0 cm3 of a 2000 mg
dm–3 Ca2+ solution, 1.0 cm3 of a H2SO4 solution and 0.5 cm3 of
a coloring agent solution, 0.06 cm3 of the ion exchanger was
added using an aliquotting device, and the mixture was stirred
for 20 min at 20°
C. After allowing the ion exchanger to settle,
the supernatant solution was removed, and about 1 cm3 of the
mixture was transferred into a disposable PE syringe (SS-10SZ,
Terumo, Tokyo, Japan) connected to a flow cell, as shown in
Fig. 1. The absorbances were directly measured at 540 nm
(absorption maximum wavelength) and 700 nm (non-absorption
wavelength), and the difference between the two absorbances
was used for Cr(VI) analyses.
After the absorbance
measurement, the ion-exchanger beads were removed from the
cell for the next measurements.
Analytical method for total chromium
The total chromium concentration of the natural water samples
was determined with an ICP-MS Agilent Model 7500cx
(Yokogawa, Japan).
Fig. 3 Attenuation spectra of a solid background against a 0.4
attenuation disk. A, B: 1.5 mmf cell; C, D: 3 mmf cell. A, C:
Muromac 50W-X2; B, D: SP-Sephadex.
Results and Discussion
Improvement of the spectrophotometer for SPS
As shown in Eq. (1), an effective way to enhance the sensitivity
is to employ a smaller v and/or longer cell systems. On the
other hand, these systems may involve a difficulty in packing
ion exchanger beads in a cell and/or exceed the limit of
detection by the light detector used. Therefore, we made some
improvement in the spectrophotometer, as shown in Fig. 2. The
cell was placed at the nearest position to the light detector of a
silicon photocell, and the sample light beam was adjusted to
focus at the flow cell using a cylindrical lens (focusing length of
25.4 mm, Sugitoh, Japan) and a convergence lens (focusing
length of –200 mm, Sugitoh). The optimum position of the cell
was adjusted using an XY stage (TASB-402, Sigma Koki,
Tokyo) attached the cell in order to obtain the strongest light
intensities. A commercially available lens (Jasco, Japan) for
recovering the scattered light from the cell was also installed at
the detector window of the spectrophotometer. The achievement
of the light measurement improvement is shown in Table 1 as
the decrease in the background attenuances of the ion exchanger
packed in the flow cell. Without using lenses, the background
attenuances were over 4, and by using a cylindrical convex lens
and a concave lens at the sample light beam, the values were
reduced to around 2. Background attenuance values of around
1.7 were achieved when a commercially available lens (Jasco)
was also used. Both the focusing lens and the recovery lens are
very effective for reducing the background attenuances of the
flow cell and the ion-exchanger beads. The cell was connected
to a 10-cm3 disposal syringe (SS-10SZ, Terumo, Japan) with a
PE three-way stopcock for introducing colored ion exchanger
beads in the cell, which could make the operation in the solid
phase absorbance measurements simple, easy and reproducible.
Optimization of SPS for microdetermination of Cr(VI)
Type of ion exchanger. Polystyrene-type ion exchangers were
appropriate for the determination of Cr(VI) using the present
SPS. In the case of cross-linked dextran ion exchangers, it was
difficult to measure the light absorption due to the high
background absorbance. Therefore, a Muromac 50W-X2 cation
exchanger (100 – 200 mesh) was used. Figure 3 shows the
differences in background attenuation spectra of the two types
of ion exchangers. It was clear that by using Muromac 50W-X2,
the background absorbances were lower than those of
SP-Sephadex C-25 cation exchanger. On the other hand,
Muromac 50W-X2 also had a limitation, especially if
absorbances at wavelengths lower than 460 nm had to be
1448
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Table 2 Effect of adding an excess of Ca2+ in the analysis of
natural water samples
Cr(VI)/µg dm–3
Fig. 4 Effect of stirring time on color development at 20°
C. Sample,
1 µg dm–3 Cr(VI), 20 cm3; ion exchanger, Muromac 50W-X2,
0.06 cm3; ∆A = A540nm – A700nm.
measured. In this case, SP-Sephadex C-25 could still be used.
Diameter of the black flow cell. Figure 3 also shows that a
3-mm diameter cell provides a lower attenuance background
than that of a 1.5-mm diameter cell. When the attenuance
exceeds 2, it is recommended that the 3-mm diameter cell be
used, although the sensitivity is decreased. To balance the light
intensities between the sample beam and the reference beam, a
light attenuation disk of 0.4 or 1.0 absorbance unit, Hitachi
(Tokyo, Japan) was placed in the reference beam.
Aliquotting of cation exchanger. For precise aliquotting of a
small amount of ion-exchanger beads, a device made of a PTFE
tube was used.11 In order to use a very small amount of resin
with high reproducibility, the device was made with a PTFE
tube 1.0 mm i.d. and 7 cm long. The ion exchanger was
collected into the tube with a disposal syringe. One side end
connected to the syringe had an inserted polypropylene filter tip.
The wet volume of the ion-exchanger used in the device was
determined by titration. The H+ ion of 1.32 cm3 of the resin
packed in a column was exchanged with Na+ by passing a
1 mol dm–3 NaCl solution through the column, and the H+ in the
effluent solution was titrated with a standard NaOH solution.
An aliquotted ion exchanger was also titrated with the same
standard solution, and the volume of wet ion exchanger beads
could be estimated. The reproducibility of the aliquotting of the
ion exchanger beads was determined by this method to be
0.0600 ± 0.0008 cm3, with a relative standard deviation (RSD)
of 1.4%, which included the titration error.
Time dependence of color development. The effect of the
stirring time on the adsorption of the purple species on the
cation exchanger is shown in Fig. 4. The color development of
the resin was influenced by the stirring time. In this experiment,
the stirring time was fixed at 20 min. The absorbance of the
adsorbed purple color species was nearly constant in the
C.
temperature range 10 – 30°
Effect of co-existing ions. As has already been demonstrated in
SPS with DPC as the coloring agent,9,13 metals such as V(V),
Cr(III), Mn(II), Co(II), Ni(II), Zn(II), Mo(VI), Cd(II), Sn(IV),
Hg(II) and Pb(II) do not interfere when present up to 1000-times
the concentration of Cr(VI). In the case of Cu(II), its presence
at 10-times the concentration of Cr(VI) is tolerable.
The presence of divalent cations, such as Ca2+ in real water
samples, caused higher background attenuance (A700nm, i.e., the
attenuance at 700 nm) than that of a standard solution in the
absence of divalent cations, as shown in Table 2. A background
attenuance change due to shrinkage of the cation exchanger in
the cell is often observed13,14 when polyvalent cations are
adsorbed. The resin particles shrink when the counter ions are
exchanged with divalent cations, which causes an increase in the
A540nma
A700nmb
Without adding 100 mg dm–3 Ca2+
1.373
1.525
0
1.329
1.605
0.5
1.330
1.729
1.0
1.335
2.0
1.954
1.630
1.828
Tw-1 sample
1.616
1.801
1.617
1.794
With adding 100 mg dm–3 Ca2+
1.828
2.023
0
1.781
2.051
0.25
1.818
0.5
2.177
1.822
2.343
1.0
1.889
Tw-1 sample
2.102
1.881
2.093
1.904
2.119
∆Ac
Cr(VI)/µg dm–3
0.152
0.276
0.399
0.619
0.198
0.185
0.177
0.20
0.14
0.11
0.195
0.270
0.359
0.521
0.213
0.212
0.215
0.06
0.06
0.07
a. Absorbance at the wavelength characteristic of colored species.
b. Absorbance at the wavelength which the sample species does not
absorb.
c. The difference between the two of absorbances.
Sample: Tw-1, Tawangmangu waterfall, upstream of Tw-2,
Karanganyar, Central Java, Indonesia. The light measurements were
done against a 0.4 attenuation disk.
Fig. 5 Effect of coexisting Ca2+ and Mg2+ on solid-phase absorbance
measurements. ▲, Blank; ■, Cr(VI) 1 µg dm–3; ∆A = A540nm – A700nm.
amount of resin in the cell, and also in the effective light path.
In this paper, the effect of the concentration of divalent ions
common in natural water, Ca2+ and Mg2+, was studied (Fig. 5).
The existence of Ca2+ showed a larger effect in producing a
higher background attenuance compared to that of Mg2+. This
phenomenon was due to the higher selectivity of Ca2+ than that
of Mg2+. It is effective to add Ca2+ in constant excess to suppress
any change in the amounts of the ion exchanger in the light path
(Table 2).
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Calibration and sensitivity
For calibration, the absorbance difference (∆A) between the
absorbances at 540 and 700 nm was practically employed:
∆A = ARC + ∆A (for the blank). The calibration curve was
reasonably linear for the 20 cm3 water sample used, and was
expressed as follows:
∆A = 0.194C + 0.198
(sample volume, 20 cm3; R2 = 0.998), (2)
where C is the Cr(VI) concentration in µg dm–3. In the case of
the corresponding solution method, a 1 mg dm–3 Cr(VI) solution
showed an absorbance of 0.704, and therefore the sensitivity
achieved was 277-times higher for a sample volume of 20 cm3
than that of the solution method. The respective theoretical
value of the sensitivity enhancement estimated by the ratio V/v
was 333-times that of the solution method, which was in fairly
good agreement with the obtained results.
Precision and detection limit
The precision was measured with samples of Tw from
Tawangmangu, Karanganyar, Central Java province, Indonesia
and No. 10 from Mizutori-no-ana spring, Hiraodai plateau,
Fukuoka, Japan. Using the standard addition method, the
concentrations of Cr(VI) were 0.04 ± 0.006 (n = 5) for the
sample Tw-2 and 0.34 ± 0.03 (n = 5) for sample No. 10, with a
Table 3 Standard addition method in Cr(VI) determination for
natural water by SPS
Sample
Cr(VI) added/
µg dm–3
Cr(VI) found/
µg dm–3
Recovery, %
0
0.20
0.40
0.60
0
0.30
0.60
0.90
0.04 ± 0.006 (n = 5)
0.23
0.45
0.64
0.34 ± 0.03 (n = 5)
0.61
0.95
1.24
—
98
101
100
—
96
101
100
Tw-2a
No. 10b
a. Tawangmangu waterfall, downstream of Tw-1, Karanganyar, Central
Java, Indonesia.
b. Mizutori-no-ana spring, Hiraodai Karst plateau, Fukuoka.
1449
The relative standard
recovery of 96 – 101% (Table 3).
deviations for five measurements were 15 and 8.8%, respectively,
due to their low concentrations. However, the Cr(VI) from 0.61
to 4.72 µg dm–3 in water samples from the Ochozu Experimental
Watershed was determined with an RSD of less than 5%
(Table 4). It is clear that the recovery for each sample solution
was acceptable, although these samples contain cations such as
magnesium and calcium ions at 20 mg dm–3 levels.
In order to determine the detection limit, a blank signal was
repeatedly measured. When the detection limit is defined as the
concentration that gives an absorbance corresponding to 3σ for
the standard deviation of fluctuation of the blank, the value was
0.014 µg dm–3 (n = 5) for 20 cm3 samples. The concentration
level of Cr(VI) below 1 µg dm–3 could be effectively determined
by the developed method.
Applicability of this method
The combination of SPS for Cr(VI) and ICP-MS for the total
Cr made it possible to carry out speciation of the dissolved Cr in
water. Except for the Ng sample, the predominant species of
dissolved Cr was Cr(VI). For water samples from the Ochozu
Experimental Watershed, serpentine containing chromite is
distributed in the studied area and the samples contained slightly
higher Cr(VI) concentrations than those of other areas of granite
and limestone.15 We could not find any relationships between
the [Cr(VI)]/[Cr(III)] ratios and the pH values or the
concentrations of DOC. A further study is necessary to clarify
the source of Cr(III) for the Ng sample.
As shown in Table 5, the sensitivity of the present method is
the highest among those of previous batch methods if the sample
volume is the same. There is a limitation of the batch method,
especially in packing all of the ion exchanger into the light-path
portion in the black cell; however, a 0.06-cm3 ion exchanger, a
three-times volume of the light-path portion was enough to get
high sensitivity and reproducibility.
The sensitivity is
comparable between the present method and the flow method of
SPS,13,14 but the present batch method is easy to set up and
simple in operation without any pumps or other accessories for
the flow system. The flow method may be convenient for the
routine analysis of many samples, but the present method can be
as an alternative for laboratory use.
The improvement of this method using another type of
double-beam spectrophotometer was done with a Shimadzu UV
1601 PC spectrophotometer. A focusing system using cylindrical
convex lenses (50 and 60 mm in focus) and convex lenses (25.4
Table 4 Analytical data for Cr(VI) and Cr(III) in natural water samples
Sampling site
Date
pH
Water temp./°
C
DOC/mg dm–3
Cr(VI)/µg dm–3
Cr(III)/µg dm–3
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 10
No. 11
Tl
Ng
Tw
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/18/08
Oct/18/08
Oct/18/08
Aug/30/08
Aug/30/08
Aug/26/08
8.04
8.15
8.24
8.06
8.22
7.69
7.86
7.84
n.d.
n.d.
n.d.
6.38
6.35
7.81
19.6
19.5
19.4
18.6
18.8
18.1
18.6
18.3
n.d.
n.d.
n.d.
25.5
28.0
21.2
1.58
1.82
1.67
1.43
1.52
0.77
1.12
1.16
1.87
0.67
0.55
0.82
0.78
0.78
0.61 ± 0.02a
1.25 ± 0.03
1.98 ± 0.06
2.04 ± 0.02
3.60 ± 0.17
4.72 ± 0.18
3.19 ± 0.04
0.91 ± 0.01
0.20 ± 0.005
0.34 ± 0.03
2.51 ± 0.08
0.08 ± 0.002
0.09 ± 0.009
0.07 ± 0.005
0.03b
0.04
0.50
0.53
0.71
1.19
0.58
0.05
0.04
0.09
0.35
0.00
0.48
0.00
a. n = 3, 1σ. b. The difference between the values of total Cr and Cr(VI). n.d. = not measured.
1450
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Table 5 Comparison of sensitivity
Sample volume/cm3
Solution method
Batch method of SPS
200
1000
1000
5
20
Flow method of SPS
4.4
7.9
Solid volume/cm3
0.5
0.5
0.5
0.06
0.06
Cell length/mm
Aa
Detection limit/µg dm–3
Reference
10
0.000704
1
1
10
10
10
0.016
0.037
0.310
0.087
0.194
0.072
0.014
7
10
10
Present study
Present study
10
10
0.083
0.244
0.055
0.009
13
14
a. Absorbance of a 1 µg dm–3 Cr(VI) solution.
and 40 mm in focus) also gave lower background absorbances,
and made it possible to analyze Cr(VI) in natural water samples.
No. 19310011 for K. Y. (2007 – 2009), from the Ministry of
Education, Science, Sports and Culture, Japan.
Conclusions
References
The favorable operation and performance characteristics of the
improved procedure of the SPS method made it possible to
determine Cr(VI) at sub-µg dm–3 to µg dm–3 levels in natural
water within 20 min. As shown in Table 5, the use of a narrow
black-sided cell could reduce the amount of required
ion-exchange material, and a higher sensitivity could be obtained
by using a smaller amount of sample solution. The procedure is
simple and easy in operation. An improvement in any
less-expensive spectrophotometers can be achieved along with
the basic idea shown in this paper, which will extend the SPS to
other wider demands for trace analyses.
1. J. Kotás and Z. Stasicka, Environ. Pollut., 2000, 107, 263.
2. A. M. Zayed and N. Terry, Plant Soil, 2003, 249, 139.
3. P. A. Kumar, M. Ray, and S. Chakraborty, J. Hazard.
Mater., 2007, 143, 24.
4. World Health Organization, “Chromium in Drinking-water”,
2006, World Health Organization (WHO), Geneva.
5. Guidelines for Canadian Drinking Water Quality, in Health
Canada on behalf of the Federal-Provincial-Territorial
Committee on Drinking Water, 2007.
6. EU’s Drinking Water Standards, Council Directive
98/83/EC on the Quality of Water Intended for Human
Consumption, November 3, 1998.
7. K. Yoshimura and H. Waki, Talanta, 1976, 23, 449.
8. K. Yoshimura and H. Waki, Talanta, 1985, 32, 345.
9. K. Yoshimura and S. Ohashi, Talanta, 1978, 25, 103.
10. K. Yoshimura and H. Waki, Talanta, 1987, 34, 239.
11. U. Hase and K. Yoshimura, Anal. Sci., 1993, 9, 111.
12. J. Ide, O. Nagafuchi, M. Chiwa, A. Kume, K. Otsuki, and
S. Ogawa, J. Forensic. Res., 2007, 12, 45.
13. K. Yoshimura, Analyst, 1988, 113, 473.
14. S. Matsuoka, Y. Tennichi, K. Takehara, and K. Yoshimura,
Analyst, 1999, 124, 787.
15. T. Tsuruhara, K. Takehara, K. Yoshimura, S. Matsuoka, S.
Saputro, and J. Aizawa, J. Ion Exchange, 2007, 18, 524.
acknowledgements
The authors would like to thank Prof. Kyoichi Otsuki from the
Department of Forest and Forest Product Sciences, Faculty of
Agriculture, Kyushu University, for guidance on site sampling
at the Ochozu Experimental Watershed, and Dr. Yoshika
Tennichi from the Kyushu Environmental Evaluation Association
of Fukuoka for ICP-MS measurements of total chromium. This
work was partially supported by the JSPS Ronpaku Program
(DGHE-10715) for S. S. (2007 and 2008) and by a Grant-in-Aid
for Scientific Research (C), No. 18550067 for S. M. (2006 –
2007), and by a Grant-in-Aid for Scientific Research (B),
1445
2009 © The Japan Society for Analytical Chemistry
improved solid-phase spectrophotometry for the
microdetermination of Chromium(Vi) in Natural Water
sulistyo saPUTRO,*1 Kazuhisa YOsHimURa,*2† shiro maTsUOKa,*3 Kô TaKEHaRa,*2 and Narsito*4
*1 Department of Mathematics and Natural Science Education, Faculty of Teacher Training and Education
Sciences, Sebelas Maret University, Surakarta-57126, Indonesia
*2 Department of Chemistry, Faculty of Sciences, Kyushu University, Hakozaki, Higashi,
Fukuoka 812–8581, Japan
*3 Department of Environmental Science, Faculty of Science, Niigata University, Ikarashi,
Niigata 950–2181, Japan
*4 Department of Chemistry, Faculty of Mathematics and Natural Sciences, Gadjah Mada University,
Yogyakarta-55281, Indonesia
A simple and sensitive solid-phase spectrophotometry procedure was improved for the microdetermination of Cr(VI). A
0.06 cm3 portion of a cation exchanger, Muromac AG 50W-X2, was used to concentrate the target Cr(VI) in a 20 cm3
water sample, and resin beads were introduced in a flow cell of 1.5 mm diameter and having a 10 mm light path length
for measurements using a UV-visible spectrophotometer. Three lenses were used for focusing the incident light beam and
for recovering light scattered by the solid phase in the cell. The sensitivity achieved was higher by a factor of 277
compared with that of the solution method, and the detection limit was 0.014 µg dm–3. The recovery on spiked real water
samples by the standard addition method was 96 – 101%. Favorable working and performance characteristics made it
possible to directly determine sub-µg dm–3 amounts of Cr(VI) in natural water samples.
(Received October 8, 2009; accepted November 7, 2009; Published December 10, 2009)
introduction
Chromium is a metal that occurs in oxidation states, ranging
from +2 to +6. However, as far as environmental protection is
concerned, only the two most common oxidation states, namely
Cr(III) and Cr(VI), need to be considered. Cr(III) and Cr(VI)
are drastically different in physicochemical properties as well as
chemical and biochemical reactivities. Cr(III) is well-known as
an essential trace element for humans, required for the
maintenance of normal glucose, cholesterol and fatty acid
metabolism. On the other hand, water-soluble Cr(VI) is highly
toxic to humans and animals,1 and other studies have indicated
that it is an extremely toxic carcinogen.2
Several types of wastewater, such as those discharged in the
course of dye, pigment and leather production, and by the
process of plating and electroplating and by mining may contain
undesirable amounts of Cr(VI).3 In principle, because the health
effects are determined largely by the oxidation states, different
guideline values for Cr(III) and Cr(VI) should be derived.
However, current analytical methods and the variable speciation
of chromium in water favor a guideline value for a total
chromium of 50 µg dm–3.4–6 If simple, sensitive and rapid
analytical techniques selective for trace Cr(VI) are applicable to
various environmental samples, the situation can be improved.
Although atomic spectrometries, such as flame atomic
†
To whom correspondence should be addressed.
E-mail: [email protected]
absorption spectrometry (FAAS), inductively coupled
plasma–atomic emission spectrometry (ICP-AES) and
inductively coupled plasma–mass spectrometry (ICP-MS) are
sensitive and selective for Cr, only the total amounts of Cr can
be determined despite the fairly expensive equipment.
Solid-phase spectrophotometry (SPS) is based on the direct
spectrophotometric measurement of a solid phase that has
sorbed a sample component. This method made it possible to
determine trace components in natural and other water samples
without preconcentration, because a sensitivity enhancement
was easily accomplished by increasing the sample volume.7,8
By employing diphenylcarbazide (DPC) as a coloring agent,
SPS selective for Cr(VI) has also been developed.9
In the case where the distribution ratio of the sample species
is high enough for complete adsorption, the absorbance of the
target colored analyte species in the solid phase (ARC) can be
expressed as Eq. (1) if the analyte, whose concentration (C0) in
a V cm3 sample solution is concentrated into v cm3 of the solid
phase,
ARC = εRCC0lRCV/v,
(1)
where εRC is the molar absorptivity of the colored species and
lRC the light path length in the solid phase. Alternative ways to
enhance the sensitivity of SPS by using a smaller volume of the
sample solution is to apply a larger lRC and/or smaller v systems.
As described in a later section, the use of a black flow-through
cell in which a small volume of solid particles is packed as an
adsorbent for the target chemical species is very effective for
1446
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Fig. 2 Cell compartment of a spectrophotometer for SPS. A, Light
detector; B, reference beam; C, perforate metal plate of attenuance 0.4
or 1.0 (Hitachi); D, sample beam; E, concave lens (focal distance
–200 mm, Sugitoh); F, cylindrical convex lens (focal distance 25.4 mm,
Sugitoh); G, black flow cell; H, focusing lens (Jasco); I, XY stage; J,
disposal syringe.
Fig. 1 Black flow cell for SPS. A, Inlet of the black cell connected
to a 10-cm3 PE syringe (SS-10SZ, Terumo) using PTFE and silicone
tubes; B, outlet of the black cell with inserted PTFE tube; C, light path
which is filled with ion-exchange resin (1.5 mm i.d. 10 mm long); D,
PP filter tip; E, incident light beam; F, three-way stopcock.
higher sensitivity.
However, different from conventional
solution spectrophotometry, SPS basically suffers from weak
light intensities due to light scattered by the solid phase.10 In
addition, attenuation of the incident light intensity by a black
flow cell also has to be considered if the area of the incident
light beam is wider than that of the light-path portion of the
black flow cell. For this reason, it is necessary to make some
improvement in the equipment or the light measurement system
to reduce the background light attenuance by the solid phase
and the cell, and this makes it difficult to apply this SPS method
widely to various trace analyses.
The aim of this study was to develop and evaluate a simple
and sensitive solid-phase spectrophotometric batch method
using a commercially available, simple spectrophotometer,
which would be applicable to the determination of sub-µg dm–3
amounts of Cr(VI) in natural water samples or drinking water.
The most interesting features of the improved method are
simplicity in operation, high sensitivity and fast application
without any previous sample treatment including many laborious
steps. This paper reports on the main characteristics of this
improved method and the optimum experimental conditions, as
well as the analytical application of the improved method to
natural water samples.
Experimental
Reagents
All reagents used were of analytical grade. Highly purified
water prepared with a Milli-Q SP system (Millipore, Milford,
MA) was used throughout. A standard Cr(VI) solution
(1000 mg dm–3) for atomic absorption spectrometry (Kishida,
Osaka, Japan) was used. A sulfuric acid solution (about
0.5 mol dm–3) was prepared by diluting 6.8 cm3 of concentrated
sulfuric acid with water up to 250 cm3. A coloring reagent
solution was prepared by dissolving 0.25 g of DPC
(diphenylcarbazide, Wako, Osaka, Japan) and diluting to
100 cm3 with acetone. A calcium solution (2000 mg dm–3) was
prepared by dissolving 0.74 g of CaCl2·2H2O (Kishida) in
100 cm3 of water. A sodium chloride solution (1 mol dm–3) was
prepared by dissolving 5.84 g of NaCl (Wako) and diluting to
100 cm3 with water. A standard solution of NaOH (0.1 mol dm–3,
Wako) was used for acid-base titration of the ion exchanger.
A Muromac 50W-X2 cation exchanger (100 – 200 mesh,
Muromachi, Tokyo, Japan) was used.
Apparatus
Absorbance measurements were made with a double-beam
UV-visible spectrophotometer (Model V-630, Jasco, Tokyo,
Japan). A flow cell Model FLM 220B-B-1.5 (Fig. 1) was
supplied from GL Science (Tokyo, Japan). It was black-sided,
10 mm in length and 1.5 mm in diameter. At the outlet of the
cell, a polytetrafluoroethylene (PTFE) tube (1 mm i.d.) was
connected to a silicone tube. Inside the end of the tube, a
polypropylene (PP) filter tip was placed so as to block the
ion-exchanger beads in order to pack them in the light-path
portion. At the sample light beam, a cylindrical convex lens
(focal distance 25.4 mm, Sugitoh, Tokyo, Japan) and a concave
lens (focal distance –200 mm, Sugitoh) were placed between
the incident light window and the cell for focusing the light
beam at the light-beam entrance of the cell, and the lens (Jasco)
at the detector window for recovering the scattered light. A
diagram of the cell holder for the determination of Cr(VI) is
shown in Fig. 2.
The ion exchanger was measured with an ion-exchanger
aliquotting device. A PTFE tube (1.0 mm i.d. and 7 cm long)
was fitted on one side with a PP resin filter tip and connected to
a 10 cm3 disposable syringe.11
Collection of natural water samples
Natural water samples (from No. 1 downstream to No. 7
upstream) were collected from a surface stream of the Ochozu
Experimental Watershed in Fukuoka Experimental Forest,
Kyushu University, in a mountainous watershed of Japanese
cypress located about 15 km east of Fukuoka City in western
Japan. The predominant forest soil is yellow-brown, and the
underlying bedrock consists of serpentinite containing chromite
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
1447
Table 1 Improvement of the light measurements by SPS using
lenses
Without lenses
With the lensesa;
without the Jasco
lensb
With the lenses; with
the Jasco lens
A540nmc
A700nmd
∆Ae
Over 4
2.434
2.382
2.312
1.931
1.936
1.924
Over 4
2.238
2.175
2.101
1.732
1.738
1.726
—
0.196
0.207
0.211
0.199
0.198
0.198
a. For focusing the incident light beam on the flow cell.
b. For recovering the scaterred light from the cell.
c. Absorbance at the wavelength characteristic of colored species.
d. Absorbance at the wavelength which the sample species does not
absorb.
e. The difference between the two of absorbances.
The light measurements were done against air. Cell: 1.5 mmf, 10 mm
long. Ion exchanger: Muromac AG 50W-X2, 100 – 200 mesh.
and chlorite schist.12 Water samples No. 8 and No. 9 were
collected from a surface stream near Ochozu Experimental
Watershed with a different geological condition. Water samples
No. 10 and No. 11 were collected from Karst springs on
Hirao-dai plateau, Kitakyushu, Fukuoka. Water samples were
also collected from the Central Java Province, Indonesia. Tlatar
(Tl) and Ngabean (Ng) are natural groundwater sources that
flow continuously, and are located at about 7 and 10 km,
respectively, from Boyolali, Central Java (near the Merapi
volcano). Tawangmangu (Tw) is a natural waterfall at 1100 m
in altitude, covered with forest, and located 30 km from
Surakarta, Central Java. The underlying bedrock consists of
volcanic rocks with lava and andesite breccias. The natural
water was filtered through a 0.20-µm membrane filter at the site
and stored without being acidified in a polyethylene (PE) bottle
for Cr(VI) determination. For total chromium determination,
the sample was stored in a PTFE bottle, and 1 cm3 of highly
purified nitric acid (Kishida, Japan) was added to a 100-cm3
sample in order to avoid the adsorption of Cr(III) on the
container wall. The pH and temperature of the water sample
were measured with a pH meter (HM-14P, TOA DKK) at the
sampling point, and the dissolved organic carbon (DOC) content
was measured with a TOC apparatus (TOC-VE, Shimadzu).
Determination of Cr(VI) by means of SPS
To a 20 cm3 water sample containing 1.0 cm3 of a 2000 mg
dm–3 Ca2+ solution, 1.0 cm3 of a H2SO4 solution and 0.5 cm3 of
a coloring agent solution, 0.06 cm3 of the ion exchanger was
added using an aliquotting device, and the mixture was stirred
for 20 min at 20°
C. After allowing the ion exchanger to settle,
the supernatant solution was removed, and about 1 cm3 of the
mixture was transferred into a disposable PE syringe (SS-10SZ,
Terumo, Tokyo, Japan) connected to a flow cell, as shown in
Fig. 1. The absorbances were directly measured at 540 nm
(absorption maximum wavelength) and 700 nm (non-absorption
wavelength), and the difference between the two absorbances
was used for Cr(VI) analyses.
After the absorbance
measurement, the ion-exchanger beads were removed from the
cell for the next measurements.
Analytical method for total chromium
The total chromium concentration of the natural water samples
was determined with an ICP-MS Agilent Model 7500cx
(Yokogawa, Japan).
Fig. 3 Attenuation spectra of a solid background against a 0.4
attenuation disk. A, B: 1.5 mmf cell; C, D: 3 mmf cell. A, C:
Muromac 50W-X2; B, D: SP-Sephadex.
Results and Discussion
Improvement of the spectrophotometer for SPS
As shown in Eq. (1), an effective way to enhance the sensitivity
is to employ a smaller v and/or longer cell systems. On the
other hand, these systems may involve a difficulty in packing
ion exchanger beads in a cell and/or exceed the limit of
detection by the light detector used. Therefore, we made some
improvement in the spectrophotometer, as shown in Fig. 2. The
cell was placed at the nearest position to the light detector of a
silicon photocell, and the sample light beam was adjusted to
focus at the flow cell using a cylindrical lens (focusing length of
25.4 mm, Sugitoh, Japan) and a convergence lens (focusing
length of –200 mm, Sugitoh). The optimum position of the cell
was adjusted using an XY stage (TASB-402, Sigma Koki,
Tokyo) attached the cell in order to obtain the strongest light
intensities. A commercially available lens (Jasco, Japan) for
recovering the scattered light from the cell was also installed at
the detector window of the spectrophotometer. The achievement
of the light measurement improvement is shown in Table 1 as
the decrease in the background attenuances of the ion exchanger
packed in the flow cell. Without using lenses, the background
attenuances were over 4, and by using a cylindrical convex lens
and a concave lens at the sample light beam, the values were
reduced to around 2. Background attenuance values of around
1.7 were achieved when a commercially available lens (Jasco)
was also used. Both the focusing lens and the recovery lens are
very effective for reducing the background attenuances of the
flow cell and the ion-exchanger beads. The cell was connected
to a 10-cm3 disposal syringe (SS-10SZ, Terumo, Japan) with a
PE three-way stopcock for introducing colored ion exchanger
beads in the cell, which could make the operation in the solid
phase absorbance measurements simple, easy and reproducible.
Optimization of SPS for microdetermination of Cr(VI)
Type of ion exchanger. Polystyrene-type ion exchangers were
appropriate for the determination of Cr(VI) using the present
SPS. In the case of cross-linked dextran ion exchangers, it was
difficult to measure the light absorption due to the high
background absorbance. Therefore, a Muromac 50W-X2 cation
exchanger (100 – 200 mesh) was used. Figure 3 shows the
differences in background attenuation spectra of the two types
of ion exchangers. It was clear that by using Muromac 50W-X2,
the background absorbances were lower than those of
SP-Sephadex C-25 cation exchanger. On the other hand,
Muromac 50W-X2 also had a limitation, especially if
absorbances at wavelengths lower than 460 nm had to be
1448
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Table 2 Effect of adding an excess of Ca2+ in the analysis of
natural water samples
Cr(VI)/µg dm–3
Fig. 4 Effect of stirring time on color development at 20°
C. Sample,
1 µg dm–3 Cr(VI), 20 cm3; ion exchanger, Muromac 50W-X2,
0.06 cm3; ∆A = A540nm – A700nm.
measured. In this case, SP-Sephadex C-25 could still be used.
Diameter of the black flow cell. Figure 3 also shows that a
3-mm diameter cell provides a lower attenuance background
than that of a 1.5-mm diameter cell. When the attenuance
exceeds 2, it is recommended that the 3-mm diameter cell be
used, although the sensitivity is decreased. To balance the light
intensities between the sample beam and the reference beam, a
light attenuation disk of 0.4 or 1.0 absorbance unit, Hitachi
(Tokyo, Japan) was placed in the reference beam.
Aliquotting of cation exchanger. For precise aliquotting of a
small amount of ion-exchanger beads, a device made of a PTFE
tube was used.11 In order to use a very small amount of resin
with high reproducibility, the device was made with a PTFE
tube 1.0 mm i.d. and 7 cm long. The ion exchanger was
collected into the tube with a disposal syringe. One side end
connected to the syringe had an inserted polypropylene filter tip.
The wet volume of the ion-exchanger used in the device was
determined by titration. The H+ ion of 1.32 cm3 of the resin
packed in a column was exchanged with Na+ by passing a
1 mol dm–3 NaCl solution through the column, and the H+ in the
effluent solution was titrated with a standard NaOH solution.
An aliquotted ion exchanger was also titrated with the same
standard solution, and the volume of wet ion exchanger beads
could be estimated. The reproducibility of the aliquotting of the
ion exchanger beads was determined by this method to be
0.0600 ± 0.0008 cm3, with a relative standard deviation (RSD)
of 1.4%, which included the titration error.
Time dependence of color development. The effect of the
stirring time on the adsorption of the purple species on the
cation exchanger is shown in Fig. 4. The color development of
the resin was influenced by the stirring time. In this experiment,
the stirring time was fixed at 20 min. The absorbance of the
adsorbed purple color species was nearly constant in the
C.
temperature range 10 – 30°
Effect of co-existing ions. As has already been demonstrated in
SPS with DPC as the coloring agent,9,13 metals such as V(V),
Cr(III), Mn(II), Co(II), Ni(II), Zn(II), Mo(VI), Cd(II), Sn(IV),
Hg(II) and Pb(II) do not interfere when present up to 1000-times
the concentration of Cr(VI). In the case of Cu(II), its presence
at 10-times the concentration of Cr(VI) is tolerable.
The presence of divalent cations, such as Ca2+ in real water
samples, caused higher background attenuance (A700nm, i.e., the
attenuance at 700 nm) than that of a standard solution in the
absence of divalent cations, as shown in Table 2. A background
attenuance change due to shrinkage of the cation exchanger in
the cell is often observed13,14 when polyvalent cations are
adsorbed. The resin particles shrink when the counter ions are
exchanged with divalent cations, which causes an increase in the
A540nma
A700nmb
Without adding 100 mg dm–3 Ca2+
1.373
1.525
0
1.329
1.605
0.5
1.330
1.729
1.0
1.335
2.0
1.954
1.630
1.828
Tw-1 sample
1.616
1.801
1.617
1.794
With adding 100 mg dm–3 Ca2+
1.828
2.023
0
1.781
2.051
0.25
1.818
0.5
2.177
1.822
2.343
1.0
1.889
Tw-1 sample
2.102
1.881
2.093
1.904
2.119
∆Ac
Cr(VI)/µg dm–3
0.152
0.276
0.399
0.619
0.198
0.185
0.177
0.20
0.14
0.11
0.195
0.270
0.359
0.521
0.213
0.212
0.215
0.06
0.06
0.07
a. Absorbance at the wavelength characteristic of colored species.
b. Absorbance at the wavelength which the sample species does not
absorb.
c. The difference between the two of absorbances.
Sample: Tw-1, Tawangmangu waterfall, upstream of Tw-2,
Karanganyar, Central Java, Indonesia. The light measurements were
done against a 0.4 attenuation disk.
Fig. 5 Effect of coexisting Ca2+ and Mg2+ on solid-phase absorbance
measurements. ▲, Blank; ■, Cr(VI) 1 µg dm–3; ∆A = A540nm – A700nm.
amount of resin in the cell, and also in the effective light path.
In this paper, the effect of the concentration of divalent ions
common in natural water, Ca2+ and Mg2+, was studied (Fig. 5).
The existence of Ca2+ showed a larger effect in producing a
higher background attenuance compared to that of Mg2+. This
phenomenon was due to the higher selectivity of Ca2+ than that
of Mg2+. It is effective to add Ca2+ in constant excess to suppress
any change in the amounts of the ion exchanger in the light path
(Table 2).
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Calibration and sensitivity
For calibration, the absorbance difference (∆A) between the
absorbances at 540 and 700 nm was practically employed:
∆A = ARC + ∆A (for the blank). The calibration curve was
reasonably linear for the 20 cm3 water sample used, and was
expressed as follows:
∆A = 0.194C + 0.198
(sample volume, 20 cm3; R2 = 0.998), (2)
where C is the Cr(VI) concentration in µg dm–3. In the case of
the corresponding solution method, a 1 mg dm–3 Cr(VI) solution
showed an absorbance of 0.704, and therefore the sensitivity
achieved was 277-times higher for a sample volume of 20 cm3
than that of the solution method. The respective theoretical
value of the sensitivity enhancement estimated by the ratio V/v
was 333-times that of the solution method, which was in fairly
good agreement with the obtained results.
Precision and detection limit
The precision was measured with samples of Tw from
Tawangmangu, Karanganyar, Central Java province, Indonesia
and No. 10 from Mizutori-no-ana spring, Hiraodai plateau,
Fukuoka, Japan. Using the standard addition method, the
concentrations of Cr(VI) were 0.04 ± 0.006 (n = 5) for the
sample Tw-2 and 0.34 ± 0.03 (n = 5) for sample No. 10, with a
Table 3 Standard addition method in Cr(VI) determination for
natural water by SPS
Sample
Cr(VI) added/
µg dm–3
Cr(VI) found/
µg dm–3
Recovery, %
0
0.20
0.40
0.60
0
0.30
0.60
0.90
0.04 ± 0.006 (n = 5)
0.23
0.45
0.64
0.34 ± 0.03 (n = 5)
0.61
0.95
1.24
—
98
101
100
—
96
101
100
Tw-2a
No. 10b
a. Tawangmangu waterfall, downstream of Tw-1, Karanganyar, Central
Java, Indonesia.
b. Mizutori-no-ana spring, Hiraodai Karst plateau, Fukuoka.
1449
The relative standard
recovery of 96 – 101% (Table 3).
deviations for five measurements were 15 and 8.8%, respectively,
due to their low concentrations. However, the Cr(VI) from 0.61
to 4.72 µg dm–3 in water samples from the Ochozu Experimental
Watershed was determined with an RSD of less than 5%
(Table 4). It is clear that the recovery for each sample solution
was acceptable, although these samples contain cations such as
magnesium and calcium ions at 20 mg dm–3 levels.
In order to determine the detection limit, a blank signal was
repeatedly measured. When the detection limit is defined as the
concentration that gives an absorbance corresponding to 3σ for
the standard deviation of fluctuation of the blank, the value was
0.014 µg dm–3 (n = 5) for 20 cm3 samples. The concentration
level of Cr(VI) below 1 µg dm–3 could be effectively determined
by the developed method.
Applicability of this method
The combination of SPS for Cr(VI) and ICP-MS for the total
Cr made it possible to carry out speciation of the dissolved Cr in
water. Except for the Ng sample, the predominant species of
dissolved Cr was Cr(VI). For water samples from the Ochozu
Experimental Watershed, serpentine containing chromite is
distributed in the studied area and the samples contained slightly
higher Cr(VI) concentrations than those of other areas of granite
and limestone.15 We could not find any relationships between
the [Cr(VI)]/[Cr(III)] ratios and the pH values or the
concentrations of DOC. A further study is necessary to clarify
the source of Cr(III) for the Ng sample.
As shown in Table 5, the sensitivity of the present method is
the highest among those of previous batch methods if the sample
volume is the same. There is a limitation of the batch method,
especially in packing all of the ion exchanger into the light-path
portion in the black cell; however, a 0.06-cm3 ion exchanger, a
three-times volume of the light-path portion was enough to get
high sensitivity and reproducibility.
The sensitivity is
comparable between the present method and the flow method of
SPS,13,14 but the present batch method is easy to set up and
simple in operation without any pumps or other accessories for
the flow system. The flow method may be convenient for the
routine analysis of many samples, but the present method can be
as an alternative for laboratory use.
The improvement of this method using another type of
double-beam spectrophotometer was done with a Shimadzu UV
1601 PC spectrophotometer. A focusing system using cylindrical
convex lenses (50 and 60 mm in focus) and convex lenses (25.4
Table 4 Analytical data for Cr(VI) and Cr(III) in natural water samples
Sampling site
Date
pH
Water temp./°
C
DOC/mg dm–3
Cr(VI)/µg dm–3
Cr(III)/µg dm–3
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 10
No. 11
Tl
Ng
Tw
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/10/08
Oct/18/08
Oct/18/08
Oct/18/08
Aug/30/08
Aug/30/08
Aug/26/08
8.04
8.15
8.24
8.06
8.22
7.69
7.86
7.84
n.d.
n.d.
n.d.
6.38
6.35
7.81
19.6
19.5
19.4
18.6
18.8
18.1
18.6
18.3
n.d.
n.d.
n.d.
25.5
28.0
21.2
1.58
1.82
1.67
1.43
1.52
0.77
1.12
1.16
1.87
0.67
0.55
0.82
0.78
0.78
0.61 ± 0.02a
1.25 ± 0.03
1.98 ± 0.06
2.04 ± 0.02
3.60 ± 0.17
4.72 ± 0.18
3.19 ± 0.04
0.91 ± 0.01
0.20 ± 0.005
0.34 ± 0.03
2.51 ± 0.08
0.08 ± 0.002
0.09 ± 0.009
0.07 ± 0.005
0.03b
0.04
0.50
0.53
0.71
1.19
0.58
0.05
0.04
0.09
0.35
0.00
0.48
0.00
a. n = 3, 1σ. b. The difference between the values of total Cr and Cr(VI). n.d. = not measured.
1450
ANALYTICAL SCIENCES DECEMBER 2009, VOL. 25
Table 5 Comparison of sensitivity
Sample volume/cm3
Solution method
Batch method of SPS
200
1000
1000
5
20
Flow method of SPS
4.4
7.9
Solid volume/cm3
0.5
0.5
0.5
0.06
0.06
Cell length/mm
Aa
Detection limit/µg dm–3
Reference
10
0.000704
1
1
10
10
10
0.016
0.037
0.310
0.087
0.194
0.072
0.014
7
10
10
Present study
Present study
10
10
0.083
0.244
0.055
0.009
13
14
a. Absorbance of a 1 µg dm–3 Cr(VI) solution.
and 40 mm in focus) also gave lower background absorbances,
and made it possible to analyze Cr(VI) in natural water samples.
No. 19310011 for K. Y. (2007 – 2009), from the Ministry of
Education, Science, Sports and Culture, Japan.
Conclusions
References
The favorable operation and performance characteristics of the
improved procedure of the SPS method made it possible to
determine Cr(VI) at sub-µg dm–3 to µg dm–3 levels in natural
water within 20 min. As shown in Table 5, the use of a narrow
black-sided cell could reduce the amount of required
ion-exchange material, and a higher sensitivity could be obtained
by using a smaller amount of sample solution. The procedure is
simple and easy in operation. An improvement in any
less-expensive spectrophotometers can be achieved along with
the basic idea shown in this paper, which will extend the SPS to
other wider demands for trace analyses.
1. J. Kotás and Z. Stasicka, Environ. Pollut., 2000, 107, 263.
2. A. M. Zayed and N. Terry, Plant Soil, 2003, 249, 139.
3. P. A. Kumar, M. Ray, and S. Chakraborty, J. Hazard.
Mater., 2007, 143, 24.
4. World Health Organization, “Chromium in Drinking-water”,
2006, World Health Organization (WHO), Geneva.
5. Guidelines for Canadian Drinking Water Quality, in Health
Canada on behalf of the Federal-Provincial-Territorial
Committee on Drinking Water, 2007.
6. EU’s Drinking Water Standards, Council Directive
98/83/EC on the Quality of Water Intended for Human
Consumption, November 3, 1998.
7. K. Yoshimura and H. Waki, Talanta, 1976, 23, 449.
8. K. Yoshimura and H. Waki, Talanta, 1985, 32, 345.
9. K. Yoshimura and S. Ohashi, Talanta, 1978, 25, 103.
10. K. Yoshimura and H. Waki, Talanta, 1987, 34, 239.
11. U. Hase and K. Yoshimura, Anal. Sci., 1993, 9, 111.
12. J. Ide, O. Nagafuchi, M. Chiwa, A. Kume, K. Otsuki, and
S. Ogawa, J. Forensic. Res., 2007, 12, 45.
13. K. Yoshimura, Analyst, 1988, 113, 473.
14. S. Matsuoka, Y. Tennichi, K. Takehara, and K. Yoshimura,
Analyst, 1999, 124, 787.
15. T. Tsuruhara, K. Takehara, K. Yoshimura, S. Matsuoka, S.
Saputro, and J. Aizawa, J. Ion Exchange, 2007, 18, 524.
acknowledgements
The authors would like to thank Prof. Kyoichi Otsuki from the
Department of Forest and Forest Product Sciences, Faculty of
Agriculture, Kyushu University, for guidance on site sampling
at the Ochozu Experimental Watershed, and Dr. Yoshika
Tennichi from the Kyushu Environmental Evaluation Association
of Fukuoka for ICP-MS measurements of total chromium. This
work was partially supported by the JSPS Ronpaku Program
(DGHE-10715) for S. S. (2007 and 2008) and by a Grant-in-Aid
for Scientific Research (C), No. 18550067 for S. M. (2006 –
2007), and by a Grant-in-Aid for Scientific Research (B),