Spectrophotometric studies on the simultaneous determination of cadmium and mercury with 4-(2-pyridylazo)-resorcinol.

Article abstract of DOI:10.1016/S1386-1425(01)00590-X

A new method for direct spectrophotometric determination of cadmium with 4-(2-pyridylazo)-resorcinol is reported. Absorption maximum, molar absorptivity and Sandell's sensitivity of the 1:1 (M:L) complex are 510 nm, 2.5 x 10(5) l mol(-1) cm(-1) and 3.55 ng cm(-2), respectively. A linear calibration graph is obtained up to 4.49 microg ml(-1). The zero-crossing measurement technique is found suitable for the direct measurement of the first-derivative value at the specified wavelengths. Cadmium(II) (0.42-9.2 microg ml(-1)) and mercury(II) (0.35-7.4 microg ml(-1)) in different ratios have been determined simultaneously. A critical evaluation of the proposed method is performed by statistical analysis of the experimental data. The developed method was applied to the simultaneous spectrophotometric determination of Cd and Hg in some synthetic mixtures and was found to give satisfactory results.

Full text of DOI:10.1016/S1386-1425(01)00590-X

Spectrochimica Acta Part A 58 (2002) 1401–1410
www.elsevier.com/locate/saa
Spectrophotometric studies on the simultaneous
determination of cadmium and mercury with
4-(2-pyridylazo)-resorcinol
Elham Y. Hashem *
Chemistry Department, Faculty of Science, Assiut Uni6ersity, Assiut, Egypt
Received 30 May 2001; accepted 10 July 2001
Abstract
A new method for direct spectrophotometric determination of cadmium with 4-(2-pyridylazo)-resorcinol is
reported. Absorption maximum, molar absorptivity and Sandell’s sensitivity of the 1:1 (M:L) complex are 510 nm,
2.5×105 l mol−1 cm−1 and 3.55 ng cm−2, respectively. A linear calibration graph is obtained up to 4.49 mg ml−1
.
The zero-crossing measurement technique is found suitable for the direct measurement of the first-derivative value at
the specified wavelengths. Cadmium(II) (0.42–9.2 mg ml−1) and mercury(II) (0.35–7.4 mg ml−1) in different ratios
have been determined simultaneously. A critical evaluation of the proposed method is performed by statistical
analysis of the experimental data. The developed method was applied to the simultaneous spectrophotometric
determination of Cd and Hg in some synthetic mixtures and was found to give satisfactory results. © 2002 Elsevier
Science B.V. All rights reserved.
Keywords: Cadmium(II); Mercury(II); 4-(2-pyridylazo)-resorcinol; Derivative spectrophotometry
made regarding the spectrophotometric determi-
nation of cadmium and several metal ions with
1. Introduction
PAR, solution studies on the complexation equil-
Previous studies on the acid–base properties of
bria of PAR with Cd(II) have not been fully
4-(2-pyridylazo)-resorcinol (PAR) indicated that
understood. Hence, it would be suitable to study
in water-organic solvent mixtures, the reagent
the complexation equilibria of PAR with Cd(II).
PAR exists in four chromophoric species [1] viz
Derivative spectrophotometry has been shown
H3L+, H2L, HL− and L2− exhibiting the absorp-
to be more versatile than classical spectrophoto-
tion maxima at 420, 392, 414 and 502 nm, respec-
metry for solving analytical problems. It leads,
tively. Although many investigations have been
not only to an increase in selectivity, but also in
many cases, to an increase in sensitivity [2–4].
The scale of this increase depends on the shape of
the normal absorption spectra of the analyte and
the interfering substances, as well as on instru-
* Corresponding author. Fax: +20-88-342708.
E-mail
address:
elham hashem@yahoo.com
(E.Y.
–
Hashem).
1386-1425/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1386-1425(01)00590-X

1402
E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
mental parameters and the measurement technique
(e.g. peak-to-trough or zero-crossing), chosen by
the analyst in a given analytical procedure [5–7].
The zero-crossing method, developed by O’Haver
[8], has been applied successfully for the simulta-
neous determination of metal ions in their binary
mixtures through the formation of complexes with
some organic ligands [9,10]. This method is based
on the measurements of the absolute value of the
derivative spectrum of the mixture at an abscissa
value (wavelength) where the intensity of one
component of the mixture goes to zero. At this
wavelength, the intensity is directly proportional to
the other component.
Therefore, a rapid, sensitive and inexpensive
spectrophotometric determination of Cd using
PAR as a chromogenic reagent is reported. The
present work has aimed at establishing the equi-
libria that exist in solution and at determining the
basic characteristics of the complexes formed.
Simultaneous determination of Cd and mercury in
mixtures by first-derivative spectrophotometry, us-
ing the zero-crossing technique, based on the
coloured complexes that both Cd and Hg form
with PAR, in 50% (v/v) ethanol–water medium,
has also been developed. The proposed method has
been applied successfully to the simultaneous deter-
mination of Cd and Hg in some synthetic mixtures.
A stock solution of 2.5×10−4 mol l−1 PAR
was prepared by dissolving an accurately weighed
amount of the purified reagent in ethanol.
A 2.5×10−4 mol l−1 stock solution of cadmiu-
m(II) chloride was prepared by dissolving the
required amount of the Analar product in deion-
ized water. A 2.5×10−4 mol l−1 stock solution
of mercury(II) nitrate was prepared by dissolving
the required amount of the Analar product in
deionized water. Cd and Hg solutions were stan-
dardized complexometrically with EDTA [12]. All
solutions included sodium perchlorate (1 mol l−1
)
prepared from standard solutions of sodium hy-
droxide and perchloric acid. The ionic strength of
solutions was maintained at a constant value of
I=0.1 mol l−1 (NaClO4). All measurements were
made in 50% (v/v) ethanol–water medium at
25 °C.
2.1. Standard procedure
Into an aliquot containing 112.4 mg/25 ml of
Cd(II), 5 ml of 1×10−4 mol l−1 solution of PAR
and 2.5 ml of 1 mol l−1 NaClO4 were added. After
mixing with the appropriate amount of pure etha-
nol, the pH of the solution was adjusted to 5.5 with
dilute HClO4. The resulting solution was filled up
to 25 ml with deionized water and the absorbance
was measured at 510 nm against a reagent blank
similarly prepared and of the same pH as the test
solution but containing no Cd(II).
1.1. Experimental instruments
Absorption measurements were made on a Shi-
madzu UV-2101 recording spectrophotometer us-
ing 10-mm matched quartz cells and slit width 2
nm. The pH of solutions were measured using a
Radiometer pH meter (M210) equipped with a
Radiometer combined glass electrode pH values in
water–ethanol medium and corrected as described
elsewhere [11]. The spectra of Cd–PAR system was
measured against a reagent blank, similarly pre-
pared but containing no metal ions.
2.2. Simultaneous determination of Cd(II) and
Hg(II)
An aliquot of a sample solution, containing up
to 9.2 mg ml−1 of Cd, and up to 7.4 mg ml−1 of
Hg, was transferred into a 25 ml calibrated flask,
then, 10 ml PAR (1×10−4 mol l−1) was added,
the pH was adjusted to 5.5 and diluted to the full
volume while keeping a final 50% (v/v) ethanol
content. The derivative spectrum was recorded in
the range from 400–600 nm against a reagent
blank. The absolute value of the first-derivative
was measured at 510 nm (zero-crossing point of Cd
complex) and 522 nm (zero-crossing point of Hg
complex) for the determination of Cd and Hg,
respectively.
2. Chemicals
All chemicals were of analytical-reagent grade
and deionized water (or pure ethanol) was used for
the preparation of solutions.

E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
1403
3. Result and discussion
stant (i) of (Cd–PAR) complex is related to the
equilibrium constant (Keq) by the expression i=
−2
KeqKal
.
3.1. Complexation equilibria of cd(II) with PAR
The calculated values of the apparent equi-
librium constant and stability constant at pH 5.5
are given in Table 1.
The complexation equilibria of Cd(II) with
PAR were studied in solutions containing 50%
(v/v) ethanol–water medium within the pH range
3.0–10.0. The solution spectra were recorded in
equimolar solutions and in solutions containing
an excess of one component. The absorption spec-
trum of the Cd complex shows an absorbance
maximum at 510 nm, where the reagent PAR does
not absorb.
3.3. Stoichiometry of the complex
The composition of the complex was further
verified by the method of continuous variation
[16] in solution having Co=CM+CL=2.5×
10−5 mol l−1 at pH 5.5, the plot of absorbance at
510 nm versus mole fraction of Cd shows a maxi-
mum at 0.5, suggesting the formation of 1:1
(M:L) complex.
The optimum concentration of PAR for maxi-
mum colour development was determined by car-
rying out a set of experiments using a 7.86 mg
ml−1 Cd2+ concentration. The reagent concen-
tration was varied over a wide range from 1×
10−5 M to about 5×10−3 M. The maximum
absorbance was attained using 2×10−4 M PAR.
The complex formation starts at pH 3.0 and
increases up to pH 6.0 as evidenced by increase in
absorbance. Maximum colour development is at-
tained at pH 5.0–6.0. At higher pH values, a
decrease of absorbance is observed which is pre-
sumably due to the formation of hydroxo-com-
plexes. The absorbance versus pH graphs for the
Cd–PAR system were interpreted using the rela-
tionships derived by Voznica et al. [13], and by
the generalized approach described previously
[14,15]. By considering the acid–base equilibria of
PAR in 50% (v/v) ethanol, one can assume that
the molecular form of the reagent [H2L] is the
prevalent species in the pH range (3.0–5.5). At
this pH range, the graphical analysis of the ab-
sorbance versus pH graphs for solutions with
different component ratios of Cd(II) complex
(Fig. 1), indicate that the best fit of analysis is
obtained with the libration of one proton and
formation of [CdHL]+ complex according to
equilibrium (1),(Fig. 2).
3.4. Calibration graph and reproducibility
Under the optimum conditions cited above, a
linear calibration graph for Cd–PAR system was
obtained up to a concentration 4.49 mg ml−1 Cd
with a molar absorptivity of 2.5×105 l mol−1
cm−1 at 510 nm. A Ringbom plot showed that
the optimum working range for the determination
of Cd was 1.01–4.04 mg ml−1. The Sandell’s
sensitivity [17] of the reaction of Cd2+ was found
to be 3.55×10−2 ng cm−2
.
3.5. Effect of di6erse ions
The effect of diverse ions on the determination
of Cd2+ with PAR was studied by adding a
known quantity of the desired ion to a solution
containing 101.2 mg/25 ml of Cd(II) which was
determined as described in the given procedure.
The tolerance criterion for a given ion was taken
as deviation of the absorbance values by more
than 92% from the value expected for Cd2+
alone. The determination of Cd(II) as [CdHL]+
was possible in the presence of Li+, Na+, K+,
NH+4 , Ag+, Mg+2, Ba2+, Sr2+, Mn2+, Pb2+
Cu2+, Ni2+, Co2+, Ca2+, Y3+, Al3+, La3+
,
,
Cd2+ +H2L X [CdHL]++H+……Keq.
(1)
Bi3+, Sb2+, As3+, and Mo6+, 6.20 mg (ꢀ60-
fold), and I−, Br−, NO3−, SO42−, ClO4−, ace-
tate, borate 10.3 mg (ꢀ100-fold). Interferences
caused by Zn(II) could be masked with CN− ions.
Under the present experimental conditions, it was
3.2. Equilibrium and stability constant
The equilibrium constant Keq was determined
by considering equilibrium (1). The stability con-

1404
E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
observed that Cd(II) could not be determined in
the presence of EDTA, HPO4
ingly, the above reagents could not be used as
masking agents.
zero-order spectrophotometry is frequently
difficult. The procedure for the simultaneous anal-
ysis, which involves the measuring of the total
absorbance of the mixture at the two wavelengths
of maximum absorption followed by simultaneous
equations, was investigated. However, results of
poor accuracy and reproducibility are obtained.
This disadvantage has been overcome by use of
derivative spectrophotometry.
and F−, accord-
2−
3.6. Simultaneous determination of cadmium and
mercury in mixtures
3.6.1. Absorption spectra
The absorption spectra of Cd–PAR and Hg–
PAR complexes are shown in (Fig. 3). The total
absorption spectrum of a mixture of Cd and Hg
complexes is given. Due to the fact that absorp-
tion spectra of Cd and Hg complexes overlap, the
determination of Cd and Hg in their mixtures by
3.6.2. Deri6ati6e spectra
Appropriate parameters were selected to record
the first-derivative spectra using the zero-crossing
method to select suitable wavelengths for obtain-
ing linear calibration graphs. The reproducibility
Fig. 1. Absorbance versus pH graphs for Cd(II)–PAR systems of various concentrations of components. (a) CL=CM=2.5×10−5
M; at umax 510 nm. (b) CL=2.5×10−5M, CM=5×10−5M, at umax 510 nm. (c) CL=25×10−5 M, CM=1.0×10−5 M, umax 510
nm.

E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
1405
Fig. 2. Absorption spectrum of Cd(II)–PAR complex at different pH values. CL=2.5×10−5 M; CM=1.0×10−5M; pH (1) 3.7,
(2) 4.9, (3) 5.2, (4) 5.8, (5) 6.0, (6) 6.9, (7) 7.7, (8) 8.8, (9) 9.1.
h2 in the first-derivative spectra of mixtures (Fig.
5(a), (b)) corresponding to values taken at 510 nm
(zero-crossing point of Cd complex) and at 522 nm
(zero-crossing point of Hg complex) are propor-
tional to Cd and Hg concentrations, respectively.
The calculated regression equations are
of zero-crossing wavelengths of derivative spectra
was checked by recording the first-derivative spec-
tra of Cd–PAR and Hg–PAR systems (pH 5.5) at
different concentrations of the analytes (Fig. 4).
The zero-crossing wavelengths of Cd and Hg in the
first-derivative spectra are obtained at 510 and 522
nm, respectively, (Fig. 4(a), (b)), whereas the first-
derivative spectra of two series containing increas-
ing amounts of Cd or Hg ions are shown in (Fig.
5(a), (b)). Cd or Hg can be determined using the
absolute value of the total derivative spectrum at
a wavelength corresponding to the zero-crossing
point of the other component. The heights h1 and
D1=1.78×10−4
=1.70×10−4
C
C
Cd+0.15×10−3, D1
Hg+0.65×10−3
where D1 is the value of the first-derivative signal
(vertical distance from peak to trough) and CCd
,
CHg are the concentrations g ml−1 for Cd and Hg,
respectively.

1406
E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
Table 1
Mean values of equilibrium (log Keq), stability constants (log i) and molar absorptivities of Cd(II)–PAR complexes
Equilibriuma
Constant
Log constant
Molar absorptivity (l. mol−1 cm−1
)
[CdHL]+[H+]/[Cd][H2L]
Keq
(−0.290.01)b
(−0.190.02)c
(−0.690.01)d
2.5×105
[CdHL]+/[Cd][HL−
]
i
(9.2490.01)e
a Charges are omitted. Values are taken as average for various component concentration, I=0.1 M (NaClO4), 20 °C, 50% (v/v)
ethanol.
b From the absorbance versus pH graphs for solutions with equimolar concentrations.
c From the absorbance versus pH graphs for solutions with excess metal.
d From the absorbance versus pH graphs for solutions with excess ligand.
e log i=log Keq+pKa1.
Fig. 3. Absorption spectrum of (1) the Cd(II)–PAR complex; (2) the Hg(II)–PAR complex; (3) a mixture of Cd(II)– and
Hg(II)–PAR complexes, pH 5.5, 50% (v/v) ethanol. [Cd(II)=7.86 mg ml−1 (curve 1); Hg(II)=4.49 mg ml−1 (curve 2); Cd(II)=7.86
mg ml−1, Hg(II)=4.49 mg ml−1 (curve 3)]. PAR=2×10−4 mol l−1
.

E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
1407
Fig. 4. First-derivative spectra of Cd(II)–PAR (a) and Hg(II)–PAR (b) complexes. [Cd(II)=(1) 1.12; (2) 3.37; (3) 5.62; (4) 6.74; (5)
8.99; (6) 10.11; (7) 11.24; mg ml−1]; [Hg(II)=(1‘) 0.12; (2‘) 3.27; (3‘) 5.62; (4‘) 6.74; (5‘) 8.99; (6‘) 10.11; (7‘) 11.24 mg ml−1].
PAR=2×10−4 mol l−1, 50% (v/v) ethanol, pH 5.5.

1408
E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
Fig. 5. First-derivative (a) spectra of a mixture of Hg(II)–PAR complex and an increasing concentration of Cd(II)–PAR complex.
Hg(II)=4.49 mg ml−1; Cd(II)=(1) 1.12; (2) 2.25; (3) 3.37; (4) 5.62; (5) 6.74; (6) 7.86; (7) 10.11. (b) Spectra of a mixture of
Cd(II)–PAR complex and an increasing concentration of Hg(II)–PAR complex. [Cd(II)=7.86 mg ml−1; Hg(II)=(1‘) 1.12, (2‘)
2.25; (3‘) 3.37; (4‘) 5.62; (5‘) 6.74; (6‘) 7.86; (7‘) 10.11.

E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
1409
Table 2
Statistical analysis of the determination of Cd and Hg in mixtures with PAR using first-derivative spectrophotometry
Metal ion
umax (nm)
Regression equationb
Correlation Coefficient (r)
Cd(II)
Hg(II)
522a
510a
D=1.78×10−4
D=1.70×10−4
C
C
Cd+0.15×10−3
Hg+0.65×10−3
0.9921
0.9953
a Recommended procedures.
b D, derivative values; C, concentration mg ml−1
.
Table 3
Comparison of various spectrophotometric reagents for Cd(II)
Reagent
umax (nm)
m (l mol−1 cm−1
)
Ref.
Xylenol Orange
1-(2-pyridylazo)2-naphthol
Dithizone
2-[5-chloro-(2-pyridylazo]5-methylaminophenol
Benzoyl(thiobenzol)methan
5-dimethylaminophenol
4-(2-pyridylazo)-resorcinol
575
555
520
550
406
555
510
2.75×104
4.90×104
8.8×104
1.2×105
3.0×104
1.4×105
2.5×105
[19]
[20]
[21]
[22]
[23]
[23]
This work
N +N
'
3.7. Calibration graphs and statistical analysis of
the results
l
b
Dxmin=xl−xb\tSb
NlNb
The calibration graphs, prepared by plotting the
first-derivative value (h) versus Cd or Hg concen-
tration, gave straight lines passing through the
origin, confirming the mutual independence of the
derivative signals of the two complexes. The cali-
bration graphs obtained by the recommended
methods were linear over a range 0.42–9.20 mg
ml−1 of Cd in the presence of 0.35–7.40 mg ml−1
of Hg, and vice versa. A critical evaluation of the
proposed methods as obtained by statistical analy-
sis of the experimental results are given in Table 2.
In order to test the accuracy and precision of the
methods, five successive measurements were car-
ried out with standard solutions containing 4.5 mg
ml−1 of Cd(II) and 3.6 g ml−1 of Hg(II). The
relative standard deviations for Cd(II) was 1.5%
and for Hg(II) 1.1%, respectively. The detection
limits at the 95% confidence level of the proposed
method for the mean of four analysis (Nl) were
calculated. The minimum detectable amount
Dxmin, is given by [18].
where the subscript b refers to the blank determi-
nation. The statistical parameter being t=2.36 for
seven degrees of freedom and 95% confidence.
The calculated detection limits are 21 ng ml−1
for Cd and 34 ng ml−1 for Cd and Hg,
respectively.
The effect of presence of diverse ions in the
simultaneous determination of Cd and Hg using
PAR was studied under optimum conditions. Ab-
sorbance measurements in the first-derivative
modes proved useful in overcoming interferences
in a number of cases where the use of a masking
agents could not help. When a precipitate ap-
peared, it was removed by centrifugation before
measuring the absorbance of the supernatant liq-
uid. The tolerance of the method to foreign ions
was investigated with solutions containing 0.23 mg
of Cd and 0.16 mg of Hg per 25 ml and various
amounts of foreign ions Table 3.
The simultaneous determination of Cd2+ and
Hg+2 was possible in the presence of Li+, Na+,

1410
E.Y. Hashem / Spectrochimica Acta Part A 58 (2002) 1401–1410
Table 4
Application of the proposed method to simultaneous determination of Cd and Hg in synthetic samples
Sample no.a
Sample composition (%)
Statistical term Cd(II) found
(%)
Hg(II) found
(%)
1
2
3
Zn: 23.12; Ca: 14.12, Mn: 21.18; Cd: 36.48; Hg: 5.10
Pb: 9.77; Mg: 32.07; Ca: 23.60; Cd: 28.40; Hg: 0.50
x
s
36.48
0.04
36.48–36.51
28.40
0.002
5.10
0.04
5.12–5.15
0.50
0.02
95% CL
x
s
95% CL
x
28.41–28.43
1.83
0.52–0.54
12.6
Co: 24.6; Ni: 12.8; Fe: 23.2; Mn: 13.2; Ca: 11.8; Cd: 1.80;
Hg: 12.5
s
0.02
0.02
95% CL
x
1.81–1.85
2.62
12.6–12.9
6.1
4
Ni: 32.3; Cu: 16.1; Fe: 8.3; Mn: 1.3; Al: 33.5; Cd: 2.61;
Hg: 5.9
s
0.02
0.02
95% CL
2.64–2.68
6.2–6.5
a Number of determination for each sample n=4; x=mean recovery (%), s=standard deviation (%), CL confidence limit (%).
K+, Cs+, NH4+, Ba2+, Mo6+, As3+, Sb3+, Sb3+
,
[5] T.C. O’Haver, G.L. Green, Anal. Chem. 48 (1976) 312.
[6] J. Medinilla, F. Ales, F. Garcia Sanchez, Talanta 33 (1986)
329.
[7] F.G. Sanchez, M.H. Lopez, J.C.H. Gomez, Talanta 34
(1987) 639.
[8] T.C. O’Haver, Anal. Chem. 51 (1979) 91A.
[9] A. Youssef El-Sayed, Fresenius J. Anal. Chem. 355 (1996)
29.
[10] A.YoussefEl-Sayed,M.M.H.Khalil,Talanta43(1996)583.
[11] G. Douheret, Bull. Soc. Chim. Fr. (1967) 1412.
[12] A.I. Vogel, ATextBookofQuantitativeInorganicAnalysis,
vol. 3, Longmans, London, 1973.
Bi3+, Li3+, and Al3+ (12 mg), and I−, Br−,
NO3−, SO42−, ClO4−, acetate, and B4O27− (15
mg). Under the present experimental conditions, it
was observed that Cd(II) and Hg(II) could not be
2−
determined in the presence of EDTA, HPO4 and
F−, accordingly the above reagents could not be
used as masking agents.
4. Application
[13] P. Voznica, J. Havel, L. Sommer, Collect. Czech. Chem.
Commun. 45 (1980) 54.
[14] L. Sommer, V. Kuban, J. Havel, Folia. Fac. Sc. Univ. Brno
11 (1) (1970) 33.
[15] L. Sommer, M. Langova, V. Kuban, Collect. Czech Chem.
Commun. 41 (1976) 1317.
[16] P. Job, Ann. Chem. 9 (1928) 113.
[17] E.P. Sandell, Colorimetric Determination of trace Metals,
vol. 3, Wiley, New York, 1959, p. 97.
[18] D.A. Skoog, D.M. West, F.J. Moller, Fundamentals of
analytical Chemistry, vol. 6, Saunders Coll. Publishing,
Florida, 1992.
To confirm the usefulness of the proposed
method, Cd and Hg have been determined simul-
taneously in four synthetic samples. The results
obtained (Table 4) are in good agreement with the
certified amounts. Simplicity, sensitivity, large
range of linearity besides accuracy and precision
are the main advantages of the proposed method.
[19] R.A. Pasechnova, A. Mokhov, Zh. Analit. Khim. 27 (1972)
2146.
References
[20] S. Shibata, Anal. Chim. Acta 25 (1961) 348.
[21] Z. Marzenko, Spectrophotometric Determination of Ele-
ments, Ellis Horwood, Chichestes, 1976, p. 177.
[22] M. Villareal, L. Porta, E. Marchevsky, R. Olsima, Talanta
33 (1986) 375.
[1] R.G.Anderson,G.Nickless,TheAnalystaReview92(1967)
1093.
[2] T.C. O’Haver, Anal. Proc. 19 (1982) 22.
[3] H. Ishii, K. Satoh, Z. Anal. Chem. 312 (1982) 114.
[4] P. Levillain, D. Fompeydie, Analysis 14 (1986) 1.
[23] S. Shibata, Anal. Chim. Acta 82 (1976) 169.