Analyst, October 1997, Vol. 122 (1161–1165)
1161
Sorption–Catalytic Determination of Manganese Directly
on a Paper-based Chelating Sorbent
M. K. Beklemishev, T. A. Stoyan and I. F. Dolmanova*
Analytical Chemistry Division, Department of Chemistry, Moscow State University, Moscow
119899 GSP, Russia. E-mail: [email protected]
A hybrid technique based on a catalytic reaction carried
out on the surface of a paper-based sorbent is proposed.
It is shown that MnII exhibits its catalytic action in the
oxidation of 3,3A,5,5A-tetramethylbenzidine with periodate
in aqueous solution as well as on filter-paper with or
without attached diethylenetriaminetetraacetate
(DETATA) groups. Optimum conditions differ for the
reaction in solution and on filter-paper. For equal catalyst
concentrations, higher initial reaction rates are attainable
on the filter-papers. Preconcentration of manganese on
the DETATA sorbent combined with the subsequent
catalytic reaction improves the selectivity, reduces the
limit of determination down to 5 3 1026 mg l21 (as
compared with 6 3 1025 mg l21 in solution) and expands
the linear range by an order of magnitude (to 5 3
1026–2.5 3 1023 mg l21). The precision of manganese
determination on the sorbent is high (the RSDs are @5%
for !6 3 1024 mg Mn). Samples of tap and river water
were analyzed by use of the proposed sorption–catalytic
technique. A rapid procedure for the determination of
manganese with visual detection was also developed.
Keywords: Catalytic kinetic method of analysis; chelating
sorbent; diethylenetriaminetetraacetate; preconcentration of
manganese
Catalytic procedures are appreciated by analytical chemists for
their sensitivity and simplicity in realization. Catalytic indicator
reactions can be applied to the determination of a large number
of compounds including the catalyst, inhibitors, activators, and
compounds which convert the catalyst into an active state
(oxidizers, ligands1), and catalytically inactive metal ions
through the use of the competitive complexation principle.
Catalytic methods become especially powerful when combined
with analyte separation/preconcentration, which allows the
selectivity to be increased and the detection limits to be lowered.
One of the approaches involves adsorption of metal ions on
chelating sorbents with subsequent desorption and catalytic
determination.2 However, this type of procedure, while being
convenient in flow systems,2 is tedious in other cases owing to
the need for desorption of the analyte. Here, we propose to
eliminate the desorption stage by conducting the catalytic
indicator reaction directly on the sorbent used for preconcentration purposes. Little information has been published to date on
the foregoing principle.3 This novel area requires studies of the
peculiarities of catalytic indicator reactions as they occur on the
surfaces of sorbents, including kinetics, concentration effects,
and the metrological characteristics of the procedures.
The main purpose of this study was the practical realization
of the aforementioned sorption–catalytic principle. For the
investigation, the preconcentration of manganese on diethylenetriaminetetraacetate (DETATA) paper-based chelating sorbent was chosen as it combines effectiveness of metal sorption
and simplicity of operation.2,4 As an indicator reaction, the
oxidation of 3,3A,5,5A-tetramethylbenzidine (TMB) by potassium periodate (KIO4) catalyzed by MnII was selected.
Various amines other than TMB were previously studied in this
reaction;5–7 however, TMB is more stable in air, less carcinogenic and its oxidation provides higher absorbing products. In
the present work, the catalytic effect of MnII on the reaction
between TMB and KIO4 in solution was studied. The aim was
to employ this system to ascertain the analytical possibilities of
the sorption–catalytic approach for the determination of
manganese with both instrumental and visual detection. A
potential benefit of the sorption–catalytic approach may be
realized in ‘rapid-test’ techniques, such as spot-tests, field tests
and similar procedures,8 in which a visible signal appears on the
surface of paper strips or other supports. The development of
rapid tests for manganese may be a subject for further work
originating from this study.
Experimental
Reagents, Solutions and Apparatus
TMB ‘for analysis’ was obtained from Riedel-de Häen,
Hannover, Germany; boric, hydrochloric and sulfuric acids
(‘special purity’) from Reakhim (Moscow, Russia) were used.
All other reagents were of analytical-reagent grade. Humic and
fulvic acids were obtained from Moscow river water by
adsorption on poly(styrene–divinylbenzene)sorbent (XAD)
with subsequent desalination. All aqueous solutions were
prepared using distilled water. Ethanolic solutions of TMB
(0.025 mol l21) were prepared every 10 d; less concentrated
solutions were obtained by dilution with ethanol when necessary. A stock solution of periodate in water (1 g l21 KIO4) was
diluted to an appropriate concentration (in most cases 0.1 g l21)
every 4–5 d. A stock solution of anhydrous MnSO4 (1 g l21 Mn)
was standardized by titration.9 Solutions with lower contents of
the metal were prepared every 3 d (10 and 1 mg l21) or daily
(0.01 mg l21) by dilution of the stock solution. All the
manganese solutions were acidified with H2SO4 to pH 1.8–1.9;
it was found that such acidification was necessary in order to
obtain reproducible results. Hydrochloric acid was used to
provide pH values ranging from 2 to 4; it was purified by
isothermal distillation of the concentrated acid (reagent grade).
Borate buffer of pH 6.8 was prepared by dropwise addition of
(a)
Me
Me
H2N
NH2
Me
Me
(b)
O
N
N
N
OH O
OH O
OH O
OH
Scheme 1 (a) TMB; (b) DETATA groups attached to filter paper.
1162
Analyst, October 1997, Vol. 122
0.2 mol l21 KOH to a solution of 3.3 g of boric acid in 1000 ml
of water until the desired pH value was reached. Buffers for the
pH range 4–6 (0.1 mol l21 in sodium) were prepared from acetic
acid and sodium acetate, and buffers for pH values of 7–11 (0.1
mol l21 in borate) were prepared from boric acid, Na2B4O7·10H2O and KOH, by using standard procedures.10 The
other buffer solutions with pH > 4 were purified by pumping
through two consecutive DETATA filters placed in a special
polyethylene holder (‘Biospectr’, St. Petersburg, Russia) at a
rate of 8–10 ml min21, which removed transition metal
impurities.
The chelating sorbent was synthesized as described previously by chemical attachment of aminocarboxylic (diethylenetriaminetetraacetate, DETATA) groups to cellulose filterpaper.11 The filter-paper was 2.5 cm in diameter.
Water was distilled in a commercial apparatus (D3-4-2M,
Ioshkar-Ola, Russia). A KFK-3 spectrophotometer (ZOMZ,
Zagorsk, Russia) was used for measurements of absorbance. To
spray the filter-papers with periodate solution, a hand-operated
sprinkler for thin-layer chromatograms was used.
Reaction in Solution
Preliminary studies of various mixing orders of the components
of the indicator reaction showed that maximum absorbance was
achieved when MnII and borate buffer (pH 6.8) were mixed first,
followed by addition of TMB and finally of KIO4. Based on
this, the following procedure was used. In a 15 ml glass test-tube
were placed the reagents in the following sequence (the
amounts are given for the final procedure for manganese
determination): 3.9 ml of the buffer (pH 6.8), 0.5 ml of the
solution, containing MnII at a pH of 1.8–1.9, 0.1 ml of a 2.5 3
1024 mol l21 ethanolic solution of TMB and 0.5 ml of 4.3 3
1022 mol l21 periodate. The addition of periodate was taken as
the time of the start of the reaction. After agitation, the reaction
mixture was transferred into a 0.5 cm cell and the absorbance of
the oxidation product was measured at 650 nm against water.
The absorbance value at 3 min (A3) was taken as the analytical
signal. The experiments were performed at ambient temperature, which was kept at 24 ± 1 °C.
dependence of the analytical signal on the amount of KIO4 on
the filter-paper, as shown below.
The absorbance of a filter-paper was measured in its wet state
against the same filter-paper without added periodate (i.e., not
coloured with the reaction products) using the following
procedure. On addition of TMB solution, but before drying, the
filter-paper was placed between two glass plates fixed together
with a clip and then placed in the cell compartment of the
spectrophotometer perpendicular to the light beam. The absorbance of this filter-paper was taken as zero. The filter-paper
was then dried, sprayed with KIO4 as described above, and the
wet specimen was again placed between the glass plates for the
measurements. The absorbance at 2 min (A2) was taken as the
analytical signal.
Calculation of Metrological Characteristics
The limit of determination was defined as the concentration of
manganese for which the RSD did not exceed 33%; this
concentration was taken as the lower limit of the linear range.
The limit of detection (cmin) was calculated as 3a/b; for the
logarithmic plot y = a + bx, where y is absorbance and
x = log(cMn), there are some complications. The following
procedure was used: (1) shift of the graph y = a + bx to the
origin (x = y = 0), i.e., transformation of the graph to the form
Y = A + bX where A = 0 and b is the previous value. This
required a recalculation: X = x + (a/b) [for example,
X = log(cMn) + 4.82 for the reaction on DETATA filters]; (2)
calculation of the regression parameters for the new graph
Y = A + bX, including sA; (3) calculation of log(cmin) as
3sA/b.
Results and Discussion
Reaction Products
Reaction on Filter-paper
At least two different coloured products are observed in the
course of the studied reaction catalyzed by MnII in solution. A
bluish green product (absorption maxima 370 and 650 nm) is
formed under conditions where there is a deficiency of the
oxidant (less than 1 3 1024 mol l21 KIO4 for a TMB
concentration of 2.5 3 1024 mol l21, Fig. 1). According to the
literature,12 this species may be a dimeric dication (lmax = 380
In preliminary experiments the optimum mixing order of the
reactants was sought with the use of visual detection. Coloration
of the filter-paper was most intense when TMB and the buffered
manganese solution were mixed first and periodate was added at
the last step. The role of the humidity of the filter-papers was
also studied (the filter-papers were stored for up to 3 d in a
desiccator with controlled 75–95% humidity); no effect of
humidity on the reaction rate was observed.
The final procedure for the reaction on filter-paper (with or
without attached DETATA groups) is given below. MnII was
first applied onto a filter-paper either (a) in the form of a 0.02 ml
aliquot of a solution (0.4 mg l21 Mn) added to the centre of the
filter-paper by an Eppendorf pipette, or (b) by pumping a
buffered manganese solution of pH 6.8 (5 3 1025–0.05 mg l21
MnII, 0.08 mol l21 in borate) through the filter-paper by use of
a peristaltic pump at a rate of 8 ml min21. The filter-paper was
then dried with a gentle flow of pressurized air until no traces of
moisture could be detected, after which TMB was pipetted onto
the filter-paper (0.02 ml of a 4 3 1024 mol l21 solution)
followed by an identical drying procedure. The oxidant was
added by sprinkling the filter-paper with a 4.3 3 1024 mol l21
solution of KIO4, which was taken to be the beginning of the
reaction. The coloured reaction zone was about 20 mm in
diameter. Sprinkling rather than pipetting of periodate was
necessary in order to obtain uniform coloration of the filterpapers; the sprinkling technique was applicable owing to a weak
Fig. 1 Absorbance of the indicator reaction products as a function of KIO4
concentration. Curves 1 and 2, reaction in solution; curves 3 and 4, reaction
on filter-paper with attached DETATA groups (KIO4 applied by sprinkling
the solution onto the filter-paper). For the reaction in solution (curve 1): 2.5
3 1024 mol l21 TMB, 1 3 1023 mg l21 MnII; the measurements in solution
(A3) were made at 650 nm for cKIO4 < 5 3 1025 mol l21 (bluish green
product) or at 460 nm for cKIO4 > 5 3 1025 mol l21 (orange product). For
the reaction on DETATA filter-paper (curve 3): 8 3 1028 mol TMB, 8 3
1023 mg l21 MnII; all the measurements of the filter-paper absorbances (A2)
were made at 650 nm (bluish green product). For curves 2 and 4, the
conditions are the same as for curves 1 and 3, respectively, but with no
MnII.
Analyst, October 1997, Vol. 122
1163
and 650 nm), probably in a mixture with a meriquinone (lmax
= 655 nm). An increase in periodate concentration over 1 3
1023 mol l21 results in an orange product (lmax = 465 nm),
which is probably the result of a more extensive oxidation and
may be ascribed a quinonediimine structure12 (lmax = 465 nm).
Intermediate concentrations of the oxidant provide a brown
mixture with absorbance maxima at 380, 650 and 465 nm. As
can be observed from Fig. 1, the maximum difference in the
rates of MnII-catalyzed and non-catalytic reactions in solution
corresponds to the formation of the bluish green product,
whereas for higher periodate concentrations (and, consequently,
for orange product formation) the difference decreases. For the
determination of manganese, the bluish green product (370, 650
nm) was used.
Kinetic Curves
Kinetic curves for the TMB–MnII–periodate reaction are
depicted in Fig. 2. In order to compare the data for the
homogeneous and heterogeneous variants of the reaction, the
catalyst concentrations should be presented in units which are
common to both the solution and the filter-paper. For instance,
the mass of manganese per square of the cross-section of the
spectrophotometer light beam (mg cm22) would be a common
value for both the sorbent and a cell with the solution. If the
kinetic curves for the same concentration of manganese (in
mg cm22, Fig. 2) are compared, it can be seen that the slope of
the ascending portion of the curve is markedly higher for the
reaction on filter-paper (with or without DETATA groups) than
for that in solution. The optimum conditions for the reaction in
solution and on the filter-papers differ, so it can only be noted
that the highest attainable initial rate on the filter-papers is
higher than that in solution.
The absorbance of the filter-paper specimen (which corresponds to the formation of the bluish green product) increases
rapidly for the first 0.5 min (counting from the start of the
reaction), after which a slow decrease in absorption follows. In
solution this decrease leads to the formation of a scarcelycoloured final product which requires a few hours. On the dry
filter-paper the bluish green product is more stable (the
coloration is not diminished during 24 h). The slow increase in
the absorption of filter-paper specimens at 6–10 min is likely to
be caused by processes in the cellulose filter-paper itself, e.g.,
swelling (wet filters without the reagents also exhibit an
Fig. 2 Kinetic curves for the KIO4–MnII–TMB reaction at pH 6.8 in
solution (1, 4), on filter-paper with attached DETATA groups (2, 5) and on
filter-paper (3, 6). Curve 1, 2.5 3 1024 mol l21 TMB, 2.6 3 1025 mol l21
KIO4, 2.6 ng cm22 MnII; 2, 8 3 1028 mol TMB, 4.3 3 1024 mol l21 KIO4,
2.6 ng cm22 MnII. For curve 3, Mn solution (20 ml) was applied by pipetting.
For curve 2, Mn solution (20 ml) was pumped through the DETATA filterpaper. The solution of KIO4 was applied onto the filter-paper (curves 2, 3,
5, 6) by sprinkling. For curves 4, 5 and 6, the conditions are the same as for
curves 1, 2 and 3, respectively, but without MnII. The measurements were
made at 650 nm.
Fig. 3 Absorbance of the reaction products as a function of pH in solution
(3, 4) (A3) and on filter-paper with attached DETATA groups (1, 2) (A2).
Curve 1, 2.5 3 1024 mol l21 TMB, 2.6 3 1025 mol l21 KIO4, 1 ng ml21
MnII; 3, 8 3 1028 mol TMB, 4.3 3 1024 mol l21 KIO4, 8 ng ml21 MnII. For
curves 2 and 4, the conditions are the same as for curves 1 and 3,
respectively, but without MnII. The measurements were made at 650 nm. To
study the effect of pH, manganese solution (0.04 mg l21) of the appropriate
pH value was pumped through the DETATA filter-paper and the reaction
was carried out as described under Reaction on Filter-paper.
Fig. 4 Absorbance of the reaction products as a function of amount of
TMB in solution (2, 4) (A3) and on filter-paper with attached DETATA
groups (1, 3) (A2). Curve 1, 4.3 3 1024 mol l21 KIO4, 8 ng ml21 MnII; 2,
2.6 3 1025 mol l21 KIO4, 1 ng ml21 MnII. For curves 3 and 4, the conditions
are the same as for curves 1 and 2, respectively, but without MnII. The
measurements were made at 650 nm and at pH 6.8.
Fig. 5 Relative standard deviations of absorbance of the reaction products
as a function of manganese concentration for the reaction carried out in
solution (1) and on filter-paper with attached DETATA groups (2). Curve 1,
MnII amount denotes cMn/mg cm22 in solution (reaction conditions: 2.5 3
1024 mol l21 TMB, 2.6 3 1025 mol l21 KIO4); curve 2, MnII amounts were
calculated as cMn 3 0.02, where cMn is the MnII concentration in the solution
that was pumped through the DETATA filter-paper and 0.02 l is the solution
volume (conditions for the reaction on the filter-paper: 8 3 1028 mol TMB,
4.3 3 1024 mol l21 KIO4). The measurements were made at 650 nm and at
pH 6.5.
1164
Analyst, October 1997, Vol. 122
Table 1 Equations for the calibration graphs and linear ranges for the determination of MnII by oxidation of TMB with KIO4 in solution and on DETATA
filter-papers (preconcentration of MnII from 20 ml of solution for DETATA filter-papers)
0.455
sa
0.037
b
0.104
sb
0.024
r
0.987
cmin/mg l21
1.5 3 1025
Linear range/
3 mg l21
6 3 1025–2 3 1023
0.385
0.015
0.077
0.006
0.996
2.5 3 1025
5 3 1026–2.5 3 1023
a*
Reaction
In solution
On DETATA
filter-paper
*
For the equation y = a + bx, where x = log (cMn) and y = A3 (absorbance measured 3 min after the start of the reaction; blank absorbance was not
subtracted).
increase in absorption). If the latter is not taken into account, the
shape of the kinetic curves (Fig. 1) may be tentatively
explained13 by reversible sequential–parallel reactions such as
TMB " Product I (370, 650 nm) " Product II (colourless)
or a set of two reversible reactions
TMB " Product I (370, 650 nm)
TMB " Product II (colourless)
If one of these schemes is true, the steady-state portion of the
absorbance–time plot will correspond to an equilibrium of the
bluish green product with a colourless product.
Effect of pH and Reagent Concentrations
It was interesting to study whether the influence of reagent
concentrations (TMB, KIO4) and pH on the signal differs for the
reaction on filter-paper as opposed to in solution. As shown in
Figs. 3–5, considerable differences exist. As can be seen from
the pH curve (Fig. 3), the maximum amount of the reaction
products is formed at pH 3.1 and 6.8 both in solution and on the
DETATA filter-papers, but only on the sorbent is there a
catalytic effect of manganese at pH 3.1. The signal on the
DETATA filter-papers is higher at pH 3.1 than at pH 6.8 but the
precision is poorer, viz., RSDs of 8 and 2%, respectively, are
Table 2 Tolerance limits for foreign ions (cion : cMnII) in the determination
of MnII by use of the catalytic reaction of TMB with KIO4 in solution and
on filter-paper with chelating DETATA groups
Foreign ion
Solution (0.001
mg l21 Mn)
5
50
500
500
700
FeII
ZnII
Cl2
FeII *
K, Na, Ca, Mg, Al,
FeIII, CuII, Br2,
!1000
SO422, acetate
* In the presence of 0.2 mg l21 KF.
DETATA filter-paper
(1 ng Mn;
preconcentration
from 20 ml)
50
150
500
!1000
700
!1000
Table 3 Effect of FeII and fluoride on the absorbance of the products of the
TMB–MnII–KIO4 reaction in solution 3 min after the start of the reaction
(A3). [TMB] = 2.5 3 1024 mol l21; [KIO4] = 2.6 3 1025 mol l21; l = 650
nm; l = 0.5 cm
Concentration of
MnII/mg l21
KF/mg l21
0
0
0.2
FeII/mg l21
0
0.28
0.28
0.0002
0.056
0.053
0.058
0.002
0.204
0.086
0.206
0.02
0.223
0.097
0.220
obtained (for 8 3 1028 mol TMB, 4.3 3 1024 mol l21 KIO4 and
8 ng ml21 MnII pumped through the DETATA filter-papers).
One reason for the low reproducibility at pH 3.1 may be
incomplete sorption of MnII at this pH value. Subsequent
reactions, both in solution and on the filter-papers, were carried
out at pH 6.8.
A study of the effect of the periodate concentration on the
reaction on the filter-papers showed that only the bluish green
product is formed even at high concentrations of the oxidant. In
solution (Fig. 1), the orange product was found at KIO4 : TMB
ratios !1 : 1, whereas on the filter-papers it was never obtained.
This implies some sort of stabilization by the filter-paper of the
bluish green oxidation product; a possible explanation is the
reducing properties of the filter-paper. Another property of the
reaction on the filter-papers is a virtual absence of the effect of
KIO4 concentration on the difference in absorbances for
catalytic and non-catalytic reactions (Fig. 1). The signal is only
slightly affected by periodate concentration in the range
1024–1023 mol l21. This permits the oxidant to be applied onto
the filter-papers without strict control of the amount; sprinkling
of the filter-papers with KIO4 was used.
Sensitivity and Precision of Manganese Determination
The absorbance of the TMB–KIO4 reaction products was found
to be proportional to the logarithm of the manganese concentration for the reaction both in solution and on DETATA filterpapers. The metrological characteristics of the determination
procedure are given in Table 1. In solution, the limit of
determination (6 3 1025 mg l21) is close to that reported for the
most sensitive reactions for manganese: viz., oxidation with
periodate of N,N-diethylaniline [1 3 1025 (ref. 14) or 1 3 1024
mg l21 (ref. 15)], p-phenetidine [1 3 1024 mg l21 (ref. 16)] and
o-dianisidine [2 3 1024 mg l21 (ref. 15)].
Preconcentration of MnII on DETATA filter-papers with the
determination directly on the filter-paper makes it feasible not
only to decrease the detection limit but also to expand the linear
range for MnII from 1.5 orders (in solution) to over 2.5 orders of
magnitude (on DETATA filter-papers) (Table 1). As regards the
precision of the determination, it was thought that it would be
fairly low on the filter-papers because of both the additional
preconcentration operation and irregularities in the paper
structure (hence, irregular colouring of the filter-papers).
However, the RSD values for the reaction on the filter-papers
are close to those in solution (Fig. 5), i.e., the precision of the
determination remains fairly high.
Interferences
The criterion for interference was taken as a change of ±5% in
the absorbance for 0.001 (reaction in solution) or 5 3 1025
(reaction on DETATA filter-papers) mg l21 of manganese in
the analyzed aqueous solution. No interference results from the
presence of a 1000-fold molar ratio of various ions (Table 2) or
of 1 mg l21 humic and fulvic acids from river water. The
selectivity for manganese in the reaction on DETATA filterpapers is higher than that in solution, and both procedures are no
Analyst, October 1997, Vol. 122
1165
Table 4 Concentrations of manganese in water (mg l21) found by using the TMB–KIO4 reaction and reference techniques. The RSDs were obtained from
five parallel runs
Catalytic method
Reaction on
Atomic
DETATA
absorption
Sample
Reaction in solution *
filter-paper †
spectrometry
Spectrophotometry ‡
23
23
23
Tap water
(1.1 ± 0.3) 3 10
(0.7 ± 0.1) 3 10
(0.6 ± 0.1) 3 10
—
River water
(1.2 ± 0.1) 3 1021
(0.9 ± 0.2) 3 1021
(0.76 ± 0.03) 3 1021
(1.4 ± 0.1) 3 1021
* [TMB] = 2.5 3 1024 mol21; [KIO ] = 2.6 3 1025 mol l21; sample volume = 0.1–1 ml; [KF] = 0.2 mg l21; pH, 6.8; l = 650 nm, l = 0.5
4
cm. † The analyzed solution with added buffer (pH 6.8) and 0.2 mg l21 KF was pumped through the DETATA filter-paper and the reaction was carried out
as described under Reaction on Filter-paper. ‡ Determined with formaldoxime.17
less (sometimes more) selective than those using the reactions
of periodate oxidation of other amines.14–16 The only exception
for the TMB–KIO4 reaction is FeII, which significantly
interferes by decreasing the reaction rate (tolerance limit is 5 : 1
FeII : MnII for the reaction in solution): in p-phenetidine
oxidation,16 a 100-fold amount of FeII was tolerated. At the
same time, FeIII does not interfere in large amounts. The effect
of FeII can be removed by adding 0.2 mg l21 potassium fluoride
(Table 3). The mechanism of fluoride action is not clear, neither
is the mechanism of FeII interference itself. Fluoride is not able
to complex strongly with FeII ions; however, FeIII may be formed
in situ, while fluoride can change the redox potentials of the
pairs FeIII–FeII and MnIII–MnII simultaneously in such a manner
that iron may no longer participate in the reaction.
might be due to the lower selectivity of these techniques. When
manganese is preconcentrated on the DETATA sorbent, it is
separated from interfering species and the results obtained agree
with those obtained by another selective technique (atomic
absorption).
The authors thank Dr. G.I. Tsysin for providing the DETATA
filter-papers and for fruitful discussions, Dr. N.M. Sorokina for
AAS measurements, Dr. T.V. Polenova for the humic acid
preparation, and the Russian Foundation for Basic Research for
financial support (grant No. 96-03-08854).
References
1
Rapid Determination of Manganese With Visual Detection
One of the potential advantages of sorption–catalytic techniques
is the feasibility of rapid determinations of analytes directly on
the sorbents with no use of instrumental detection. The TMB–
MnII–KIO4 reaction was conducted on DETATA filter-papers
after preconcentration of manganese from 20 ml of aqueous
solution, using the same procedure as for quantitative measurements (see under Reaction on Filter-papers). Instead of
measuring the absorbance of the filter-paper after sprinkling it
with periodate, it was dried with a stream of air (which required
about 3 min) and the colour was observed visually. The colour
remains stable in air for not less than 6 h.
Various concentrations in the range from 1 3 1022 to 100 ng
of manganese in 20 ml of solution were studied. It was found
that confident discrimination of the colour intensities can be
made for manganese concentrations which differ by not less
than half an order of magnitude (i.e., 3 times). The determination is reliable for 0.1–10 ng of manganese (5 3 1026–5 3 1024
mg l21 for a pumped volume of 20 ml), which allows a colour
scale to be constructed for the semiquantitative determination of
manganese in this range. The whole procedure requires 6–7
min, starting with the pumping of the manganese solution
through the DETATA filter-paper.
Analysis of Tap and River Water
For analysis, an aliquot of the sample (1.0 ml of tap water or
0.10 ml of river water preserved by adding sulfuric acid to pH
1.85 immediately after sampling) with 0.2 ml of KF (20 mg l21)
added was diluted to 20 ml with borate buffer (pH 6.8). The
analyses were performed as described under Reaction in
solution and Reaction on Filter-paper. The results agreed with
those obtained by spectrophotometry17 and/or flame atomic
absorption spectrometry (Table 4). The high values obtained
with the catalytic method in solution and by spectrophotometry
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Dolmanova, I. F., and Peshkova, V. M., Vestn. Mosk. Gosud. Univ.,
Ser. 2: Khim., 1977, 18, 599.
Kolotyrkina, I. Ya., Shpigun, L. K., Zolotov, Yu. A., and Tsysin, G. I.,
Analyst, 1991, 116, 707.
Tikhonova, L. P., Bakay, E. A., Prokhorenko, E. P., Tarkovskaya,
I. A., and Svarkovskaya, I. P., presented at the 5th International
Symposium on Kinetics in Analytical Chemistry, September 25–28,
1995, Moscow, Russia; Abstracts of Papers, Nauka, Moscow, 1995,
L24.
Varshal, G. M., Velyukhanova, T. K., Pavlutskaya, V. I., Starshinova,
N. P., Formanovsky, A. A., Seregina, I. F., Shilnikov, A. M., Tsysin,
G. I., and Zolotov, Yu. A., Int. J. Environ. Anal. Chem., 1994, 57,
107.
Naylor, F. J., and Saunders, B. C., J. Chem. Soc., 1950, 3519.
Hester, R. E., and Williams, K. P. J., J. Chem. Soc., Faraday Trans.
II, 1981, 77, 541.
Makemoto, K., and Maysunaka, M., Bull. Chem. Soc. Jpn., 1968, 41,
764.
Zolotov, Yu. A., Zh. Anal. Khim., 1994, 49, 149.
Pr̂ibil, R., Analytical Application of EDTA and Related Compounds,
Mir, Moscow, 1975, p. 200 (in Russian).
Lurye, Yu., Handbook in Analytical Chemistry, Khimiya, Moscow,
1989 (in Russian).
Tsysin, G. I., Mikhura, I. V., Formanovsky, A. A., and Zolotov,
Yu. A., Mikrochim. Acta, 1991, III, 53.
Saunders, B. C., and Watson, G. M. R., Biochem. J., 1950, 46,
629.
Denisov, E. T., Kinetics of Homogeneous Chemical Reactions,
Vysshaya Shkola, Moscow, 1988, p. 57.
Nikolesis, D. P., Anal. Chem., 1978, 50, 205.
Dolmanova, I. F., and Yatsimirskaya, N. T., Zh. Anal. Khim., 1971,
26, 1540.
Gragorovich, F. G., Fresenius’ Z. Anal. Chem., 1974, 271, 5, 354.
Alimarin, I. P., Practical Recommendations on Physico-Chemical
Methods for Analysis, Moskovskii Gosudarstvennyi Universitet,
Moscow, 1987, p. 58 (in Russian).
Paper 7/02595E
Received April 16, 1997
Accepted June 30, 1997