Flow injection kinetic spectrofluorimetric determination of trace amounts of osmium

Article abstract of DOI:10.1016/j.saa.2004.09.003

A flow injection (FI) kinetic spectrofluorimetric method is described for the determination of osmium(IV) and the possible mechanism of catalytic reaction is discussed. The method is based on the fluorescence enhancing reaction of o-vanillin furfuralhydrazone (OVFH) with potassium bromate, which is catalyzed by Os(IV) in water medium at pH 6.10 and 45 °C. OVFH is newly synthesized and its ionization, IR and elemental analysis are established. Under these experimental conditions, the oxidized product of OVFH has excitation and emission maxima at 337 and 490 nm, respectively. The linear range of this method is 0-600 ng ml^-1 with the R.S.D. of 1.2%. The detection limit is 1.0 ng ml^-1 of Os(IV). A high analysis rate of 24 samples h^-1 is obtained by the FI method. The proposed method is applied successfully to determine Os(IV) in synthetic mixture and mineral samples, and the results are well consistent with the standard values.

Full text of DOI:10.1016/j.saa.2004.09.003

Spectrochimica Acta Part A 61 (2005) 2239–2244
Flow injection kinetic spectrofluorimetric determination
of trace amounts of osmium
Bo Tang∗, Hui Zhang, Yan Wang
College of Chemistry, Chemical Engineering and Materials Science, Shandong Normal University, Jinan 250014, PRChina
Received 20 April 2004; accepted 1 September 2004
Abstract
A flow injection (FI) kinetic spectrofluorimetric method is described for the determination of osmium(IV) and the possible mechanism
of catalytic reaction is discussed. The method is based on the fluorescence enhancing reaction of o-vanillin furfuralhydrazone (OVFH) with
potassium bromate, which is catalyzed by Os(IV) in water medium at pH 6.10 and 45 ◦C. OVFH is newly synthesized and its ionization,
IR and elemental analysis are established. Under these experimental conditions, the oxidized product of OVFH has excitation and emission
maxima at 337 and 490 nm, respectively. The linear range of this method is 0–600 ng ml−1 with the R.S.D. of 1.2%. The detection limit is
1.0 ng ml−1 of Os(IV). A high analysis rate of 24 samples h−1 is obtained by the FI method. The proposed method is applied successfully to
determine Os(IV) in synthetic mixture and mineral samples, and the results are well consistent with the standard values.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Osmium(IV) determination; Flow injection; Catalytic kinetics; Spectrofluorimetric analysis; o-Vanillin furfuralhydrazone (OVFH)
1. Introduction
trace amounts of osmium has been successfully determined
by the catalytic kinetic spectrofluorimetry [10].
As a very valuable and precious metal element, osmium is
widely used. Recently, osmium has been used in analysis field
[1–5]. Osmium usually coexists with other platinum group
metals in ores, rocks, soils, minerals and alloys. In order to
obtain accurate values of trace osmium in these samples, it
is necessary to find highly sensitive and selective method for
determining this element.
In recent years, many new methods for osmium analysis
have been proposed such as radiochemical neutron activation
analysis (RNAA), inductively coupled plasma mass spec-
trometry (ICPMS), inductively coupled plasma atomic emis-
sionspectrometry(ICPAES) [6], electroanalytical method [7]
and catalytic spectrophotometry [8,9]. The catalytic kinetic
spectrofluorimetric method has many advantages, for exam-
ple, it has high sensitivity and good selectivity and exper-
imental operation and equipment are simple. In our work,
The flow injection (FI) method is a very useful tech-
nique for the rapid determination of many samples, because it
can accomplish high sampling automatism with good preci-
sion. Moreover, FI technique can make reproducible timing
analysis, eliminate sample pretreatment and minimize sol-
vent and reagent consumption [11,12]. The determination
of osmium with some new methods as follows: trace anal-
ysis of osmium using flow injection kinetic catalytic spec-
trophotometric methods were reported, such as ␥-acid–O2
system [13], bromphenol red–H2O2 system [14] and m-
acetylchlorophosphonazo–osmium system [15]. The flow in-
jection catalytic kinetic fluorescence method combines the
advantages of the flow injection method with those of cat-
alytic kinetic fluorescence method, so it is a kind of more
perfect determination method. But so far, the flow injection
catalytic kinetic fluorescence analysis for osmium has not
been reported.
In this work, a new fluorescent reagent OVFH is synthe-
sized and a novel flow injection spectrofluorimetry is de-
veloped to determine osmium. As osmium is determined
∗
Corresponding author. Tel.: +86 531 6180010; fax: +86 531 6180017.
E-mail address: tangb@sdnu.edu.cn (B. Tang).
1386-1425/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2004.09.003

2240
B. Tang et al. / Spectrochimica Acta Part A 61 (2005) 2239–2244
in a positive catalyst to the fluorescence enhancing sys-
tem of OVFH–KBrO3, the linear range of this method is
0–600 ng ml−1 with a detection limit of 1.0 ng ml−1. No co-
existing ion interferes with the determination of the same
level of Os(IV). The method has been successfully applied
to determine osmium in synthetic mixtures and mineral sam-
ples.
2.3.1. Synthetic route of o-vanillin furfuralhydrazone
(OVFH)
2.70 ml of methyl-furoate (about 0.025 mol) and 3.00 ml
of 85% hydrationhydrazine (about 0.025 mol) were put into
a 50 ml flask, and then 25.0 ml of 95% ethanol was added.
The mixture was refluxed for 5 h in a water bath at 80–90 ◦C,
then it was distilled by decompression to eliminate ethanol.
After the red-brown and ropy liquid was cooled, white and
wax solid was obtained with a yield of 75%. The melting
point of the solid was 75–76 ◦C.
2. Experiment
1.26 g (0.010 mol) of formacylhydrazine was added into
40.0 ml of 95% ethanol, and the mixture was heated until
it was dissolved completely, 1.52 g (0.010 mol) of o-vanillin
was added into 20.0 ml of 95% ethanol, then it was heated to
be dissolved. The solution of o-vanillin was added into the
formacylhydrazine solution, then the mixture was refluxed
for 2 h in a water bath at 90–100 ◦C, cooled and pumped out.
The precipitation was washed with 10.0 ml of 95% ethanol
for three times. 2.10 g of yellowish powder were obtained
with a yield of 81%. The melting point of the powder was
147–148 ◦C.
2.1. Apparatus
All fluorescent measurements were carried out on a
CARY-Eclipse (Varian, Australia) spectrofluorimeter, with
a xenon lamp and an 18 ␮l flow quartz cell. A FIA-3100
flow injection analysis device (Wantuo Experiment Device
Works, Beijing, China) was used. The UV spectrum was
recorded on the UV-265 recording spectrophotometer (Shi-
madzu, Kyoto, Japan). All pH measurements were made with
a pHS-3c digital pH-meter (Shanghai Leici Device Works,
Shanghai, China) with a combined glass–calomel electrode.
Perkin–Elmer 983 infrared spectrophotometer (KBr discs,
Norwalk, CT, USA)andPerkin–Elmer2400elementaryanal-
ysis apparatus (Norwalk, CT, USA) were used. Constant
temperature container CS50 (Chongqing Experiment Device
Works, Chongqing, China) was also used.
2.3.2. Structure and ionization constants of OVFH
Elemental analysis gave a composition of: C 59.92%, H
4.61% and N 10.81%, which was in good agreement with the
theoretical composition of OVFH: C 59.97%, H 4.65%, and
N 10.77%.
The infrared spectrum of OVFH (KBr discs cm−1) was ob-
tained and the bands were assigned as follows: υO H (3300),
υC= (1660), υN and υC (1555), υC= (1608), υC
2.2. Reagents
O
H
N
N
N
(1250), υAr O (1238). The appearance of υC N showed that
the target compound of Scheme 1 synthetic reaction was
formed.
All chemicals were of analytical reagents grade or the
highest purity available. De-ionized water was used through-
out. A stock solution of osmium(IV) (100 ng ml−1) was pre-
pared by dissolving 0.02308 g of spectrographically pure
(NH4)2OsCl6 in20 mlof2 mol l−1 HClanddilutingto100 ml
with water. OVFH solution (5.00 × 10−4 mol l−1) was pre-
pared by dissolving 0.0130 g of the reagent in 100 ml of
absolute ethanol. The working solutions were prepared by
stepwise dilution as required. Potassium bromate solution
(0.040 mol l−1) was prepared by dissolving 1.6700 g of the
reagent in 250 ml of de-ionized water. A HAc–NaAc buffer
(0.16 mol l−1, pH 6.10) was used.
Determination of ionization constants of OVFH:OVFH
was insoluble in water but solved in ethanol more easily.
The ultraviolet spectra of a water–ethanol solution of the
reagent in media of various pH showed different absorbance
maxima (Fig. 1). OVFH behaved as a bibasic substance
with de-protonation of the hydroxy groups. Determined by
a spectrophotometric method [16] (Fig. 2), the correspond-
ing ionization constants were pKa = 3.82, pKa = 8.10, (in
1
2
0.10 mol l−1 KCl, at 20 1 ◦C).
2.4. Procedure
2.3. Synthesis and properties of OVFH
The experimental flow injection diagram is given in Fig. 3.
The positive and reversed phase flow injection systems were
Synthesis route:
Scheme 1.

B. Tang et al. / Spectrochimica Acta Part A 61 (2005) 2239–2244
2241
The procedure was as follows: after peristaltic pump was
started up, some OVFH solution was injected into reaction
pipe, the carrier (buffer and osmium solutions) carried the
injection zone flow through the reaction coil, in which they
met the KBrO3 solution. So the mixture produced enhanc-
ing fluorescence substance by catalytic oxidation reaction.
Then flow injection system stopped for 90 s in 45 ◦C water
bath. After flowing through the cold water bath to be cooled
down to the room temperature, the solution flowed into flow-
ing cell, in which the fluorescence intensity was measured
at λex/em = 337/490 nm. Then the blank experiment was done
with only buffer solution as carrier.
Fig. 1. Ultraviolet absorption of OVFH reagent (2.00 × 10−5 mol l−1): (1)
pH 2.80; (2) pH 6.10; (3) pH 9.30 and (4) pH 12.00.
3. Results and discussion
3.1. Excitation and emission spectra
The fluorescence maxima of OVFH (λex/em = 344/494 nm)
showed a hypsochromic change (λ ex/em = 337/490 nm) when
potassium bromate was present in the solution. In the pres-
ence of trace amounts of Os(IV) the oxidation rate of
OVFH increased owing to the catalytic effect of this ion
(Fig. 4).
3.2. Influence of experimental variables
Fig. 2. Effect of pH on UV absorbance of OVFH reagent
(2.00 × 10−5 mol l−1) at 305.6 nm.
The pH of the medium had a great effect on the fluorescent
intensity of the oxidized reagent (Fig. 5). In order to obtain
the lower fluorescent intensity of the reagent blank and the
higher signal-to-noise (the following optimum experimental
variables were chosen, according to this principle also), a pH
of 6.10 was fixed with acetate buffer. As the buffer injected
(from 0.12 to 0.20 mol l−1) had little effect on the fluorescent
intensity, 0.16 mol l−1 acetate buffer was used in subsequent
experiment.
The effect of temperature on both catalyzed and uncat-
alyzed reaction was studied over the range from 15 to 70 ◦C.
The fluorescent intensity of both systems increased smoothly
in the range of 15–50 ◦C, increased greatly in the range of
studied respectively. Positive phase FI was sample solution
(osmium solution) being injected and OVFH solution as car-
rier, but reversed phase FI was reagent solution (OVFH so-
lution) being injected and sample solution as carrier. It was
found that fluorescence curve of positive phase FI was not
stable, changed with time and had poor peak shape, but that
of reversed phase FI was very stable and had good peak shape.
So we chose reversed phase FI method, in which OVFH solu-
tion was injected, buffer and osmium solution were as carrier.
Allparametersweresetupasfollowing:λex/em = 337/490 nm,
sampling volume 100 ␮l, reaction pipe length 200 cm, sam-
pling time 38 s, injection time 22 s, peristaltic pump rotate
speed of 30 rpm, the flow rate of OVFH was 1.38 ml min−1
the flow rates of KBrO3 and sample solutions were both
2.60 ml min−1
,
.
Fig. 3. Experimental flow injection diagram. R: reagent solution
(OVFH + ethanol–water (v/v = 3:7) solution, 1.50 × 10−4 mol l−1); S: buffer
and osmium solution (pH 6.10, 0.16 mol l−1 HAc–NaAc); C: KBrO3
(0.040 mol l−1); P: peristaltic pump; V: sampling valve; K: reaction coil;
T: constant temperature container (45 ◦C water bath); T’: cooling device
(cold water, room temperature 20 ◦C); D: spectrofluorimeter and W: waste.
Fig. 4. Excitation (A) and emission (B) spectra. (1): OVFH; (2):
OVFH + KBrO3; (3): OVFH + Os(IV) + KBrO3. [OVFH]: 1.50 ×
10−4 mol l−1; [Os(IV)]: 100 ng ml−1; [KBrO3]: 4.00 × 10−2 mol l−1; pH
6.10, 0.16 mol l−1 HAc–NaAc.

2242
B. Tang et al. / Spectrochimica Acta Part A 61 (2005) 2239–2244
tion time were fixed at 38 and 22 s, respectively. Under these
conditions, the analysis rate was 24 samples h−1
.
Because sampling volume had a great effect on sensitiv-
ity and sampling frequency, with different sampling volume
(50, 100, 200 and 300 ␮l) was examined. When the sam-
pling volume increased the fluorescence intensity increased
but the sampling frequency decreased. Hence, 100 ␮l of sam-
pling volume was optimum. The effect of different peristaltic
pump rotate speed on ꢀF was also examined. The pumping
rotate speed 30 rpm was chosen. Under these conditions, the
flow rate of 1.50 × 10−4 mol l−1 OVFH was 1.38 ml min−1
,
the flow rates of 4.0 × 10−2 mol l−1 KBrO3 and sample so-
Fig. 5. Effect of pH on fluorescence intensity. (1): Uncatalyzed reaction; (2):
catalyzed reaction. [OVFH]: 1.50 × 10−4 mol l−1; [Os(IV)]: 100 ng ml−1
;
lution were both 2.60 ml min−1
.
[KBrO3]: 4.00 × 10−2 mol l−1
.
3.3. Determination of Os(IV)
3.3.1. Analytical characteristics
Under the optimum experimental conditions, there was
a linear relationship between the fluorescent intensity and
Os(IV) concentration in the range of 0–600 ng ml−1 with a
correlation coefficient (r) of 0.9994. The regression equation
was ꢀF = 0.3280[Os(IV)] + 6.400, [Os(IV)]: ng ml−1, 45 ◦C.
Thestandarddeviationofthefluorescencemeasurementswas
0.1123 obtained from a series of 11 blank solutions. The
detection limit was determined to be 1.0 ng ml−1, according
to IUPAC definition [17]. The relative standard deviation was
1.2% obtained from a series of 11 standards each containing
100 ng ml−1 of osmium.
Fig. 6. Effect of temperature on fluorescence intensity. (1) Uncatalyzed
reaction; (2): catalyzed reaction. [OVFH]: 1.50 × 10−4 mol l−1; [Os(IV)]:
100 ng ml−1
;
[KBrO3]: 4.00 × 10−2 mol l−1
;
pH = 6.10, 0.16 mol l−1
HAc–NaAc.
3.3.2. Effect of interferences
50–70 ◦C, so 45 ◦C was chosen as the optimum reaction tem-
perature (Fig. 6).
Over thirty cations and anions and some complexing
agents were studied individually to find their effects on the
determination of 100 ng ml−1 Os(IV) by the procedure. An
error of 5% in the intensity values was considered tolera-
ble. No interference was encountered from (tolerance ratio
in mass): K+, Na+, Mg2+, Cl−, Br−, SO42−, tartrate (1000);
Ba2+, Ca2+, Cr3+, I−, SCN− (500); Pb2+, Ni2+, Mn2+ (200);
Fe2+, Be2+, Cd2+, Co2+, Pd2+, Sr2+, Fe3+, As(III), As(V)
(100); Hg2+, Sn2+, Z−n2+, Sb3+, F−, EDTA (50); Ag+, Pt(IV),
Au3+,Cr(IV), MnO4 (20); Cu2+, Ru3+, Rh3+, Ir(IV) (10);
Al3+ (2). Al3+ interfered with determination of Os(IV), but
they were avoided by separating osmium through distillation
[18].
When the temperature was fixed at 45 ◦C, the fluorescent
intensity of the catalyzed reaction had linear relationship
with the stopped-flow time, whereas the uncatalyzed reac-
tion rate increased smoothly. So the catalyzed reaction could
be regarded as pseudo-zero reaction within the time range of
30–210 s (Fig. 7). Hence, the stopped-flow time was chosen
as 90 s at a 45 ◦C water bath. The sampling time and injec-
3.3.3. Mechanism of reaction
The oxidation reaction of OVFH catalyzed by Os(IV) had
Michaelis kinetic character: catalyzer and KBrO3 could form
1:1 complex [19], which was the center of the oxidation re-
action. The electrons transferred from OVFH to the lack of
electrons center of osmium, which made OVFH oxidized and
Os (IV) was oxidized to Os(IV) by KBrO3. Reaction equation
was as follows:
Fig. 7. Effect of stopped-flow time on fluorescence intensity. (1) Uncat-
alyzed reaction; (2): catalyzed reaction. [OVFH]: 1.50 × 10−4 mol l−1
;
2BrO3− + OsCl62− + 2H2O
[Os(IV)]: 100 ng ml−1
;
[KBrO3]: 4.00 × 10−2 mol l−1
;
pH 6.10,
= (OsO2(OH)2)2+ + 2BrO2− + 6Cl− + 2H+
0.16 mol l−1 HAc–NaAc.

B. Tang et al. / Spectrochimica Acta Part A 61 (2005) 2239–2244
2243
Table 1
Determination of osmium in synthetic mixtures
Sample
Composition of solution (ng ml−1
)
Added
(ng ml−1
Found (ng ml−1
)
Average
(ng ml−1
)
Average
recovery
(%)
R.S.D.
(%)
)
A
Os(IV) (100) + Pd2+ (1000) + Ag+
(400) + Au3+ (400) + Pt(IV) (400)
Os(IV) (100) + Rh3+ (200) + Ir(IV) (100)
Os(IV) (100) + Ir(IV) (100) + Ru3+
(100) + Al3+ (60)
100.0
100.6, 101.0, 100.1, 99.60, 99.80
100.2 0.6
100
0.60
B
C
100.0
100.0
98.50, 100.3, 99.40, 100.2, 99.90
99.80, 100.5, 99.90, 100.2, 100.9
99.66 0.7
100.3 0.5
99.7
100
0.70
0.50
Table 2
−
−
(OsO2(OH)2)2+ + 2BrO3 = OsO2(OH)2(BrO3
)
2
Determination of osmium in mineral sample (n = 5)
Sample
Standard value Found value
R.S.D. (%)
4(OVFH)Red + OsO2(OH)2(BrO3−)2 + 2H2O
(g/T)
(g/T)
−
→ 4(OVFH)Ox + (OsO2(OH)22+ + 2BrO2
Mineral sample of the
geological system of
China (No. 974)
0.05800
0.05620 0.00180 3.2
+ 4H+(OVFH)Ox : oxide of OVFH
3.3.4. Catalytic kinetic reaction mechanism
The linear relationship between the fluorescent intensity
oftheoxidationproductandtheconcentrationofthecatalyzer
(Os(IV)) had been testified. Moreover, two working curves
had been made at different temperature
The reaction mechanism probably consisted of oxidation
scission of the azomethine group and oxidation of the result-
ing compounds [20]. ␴ coordinate bond was formed between
full load ␲2p molecular orbital of –C N– double bond and
empty 5d26s6p3 hybrid orbital of Os(IV) in the way of lat-
eral group complexation. Meanwhile, ␲ bond was formed
between electron-packed 5d6p hybrid orbital of Os(IV) and
empty ␲*2p orbital of –C N– double bond. This kind of
bond formed between –C N– and Os(IV) was similar with
␮ bond described by Dewar–Chatt–Duncanson model, which
made partial electrons of z ␲ bond in azomethine group’s ex-
cited to ␲* antibond orbital (␲ ␲* transition). Thus, the ␲
bond in azomethine group was weakened. That was to say
␲ bond was activated by the complexation between azome-
thine group and Os(IV). Because the donicity of azomethine
group was greater than its acceptability, unsaturated carbon
atom partially became positive ions which facilitate the nu-
cleophilic attack. Then, the azomethine group was scissioned
by oxidation. The oxidation was as follows: Os(IV) was ox-
idized to Os(VIII) by potassium bromate, then Os(VIII) ox-
idized OVFH to a fluorescence product and itself reduced to
Os(IV).
One of the oxidation products was considered to be o-
vanillic acid or its acid group [21], whose (␲ ␲*) singlet state
energy was not higher than that of (n ␲*). When it existed
in high polar hydrogen bond solvent, the charge transfer be-
tween carbonyl group and benzene ring would be reinforced
by the intermolecular hydrogen bond. As a result, the lowest
excitation singlet state changed from (n ␲*) to (␲ ␲*) and
the fluorescence quantum yield increased greatly [22].
The reaction was regarded as a pseudo-zero order reaction
within 30–210 s. Under the conditions established above, the
reaction rate equation was:
ꢀF = 0.3280[Os(IV)] + 6.4000,
r = 0.9994([Os(IV)] : ng ml−1, 45 ◦C)
(2)
(3)
ꢀF = 0.2056[Os (IV)] + 3.1200,
r = 0.9994([Os(IV)] : ng ml−1, 35 ◦C)
Two different rate constants could be gained from Eqs.
(2) and (3): K45 C = 0.2187 min−1, K35 C = 0.1371 min−1. So
the Arrhenius equation was: lnK = −4122.1/T + 11.432; ap-
parent activation energy was Ea = 34.271 kJ mol−1, and fre-
quency factor was: A = 9.225 × 104 min−1mol l.
◦
◦
3.4. Applications
The developed method had been successfully applied to
the determination of osmium in a series of synthetic mixtures
(Table 1) and in mineral sample (Table 2).
It was necessary to separate the osmium from the interfer-
ing ions in synthetic mixture C [18].
The procedure of mineral sample analysis wasasreference
[10].
4. Conclusions
The method was found to be a perfect spectrofluorimetric
determination method of osmium(IV) with good selectivity,
highsensitivityandwidelinearrange. Distillationwasusedto
separate osmium from interfering elements in the samples.
The proposed method has been successfully applied to the
determination of osmium in synthetic mixtures and mineral
sample.
d[OVFH]
dt
dF
dt
R = −
=
= K[Os(IV)]
(1)
the integrating equation of (1) is: F = K[Os(IV)]t.

2244
B. Tang et al. / Spectrochimica Acta Part A 61 (2005) 2239–2244
[10] B. Tang, F. Han, G.Y. Zhang, F.Z. Liang, Anal. Lett. 35 (2002) 517.
Acknowledgements
[11] Z.L. Fang, Flow Injection Analysis, Science Press, Beijing, China,
1999, pp. 4 and 150.
[12] M. Du, F.Z. Liang, B. Tang, H.X. Shen, Chemical J. Chin. U. (Chi-
nese) 20 (1999) 369.
[13] Z.F. Zhou, J. Gu, S.G. Ye, Chem. J. Chin. U. (Chinese) 14 (1993)
781.
[14] X.G. Chen, H.P. Gong, Q. Zhang, Z. Hu, Microchem. J. 52 (1995)
364.
Financial supports from Important Project of Natu-
ral Science Foundation of China (20335030) and Shan-
dong Province Nature Science Main Foundation of China
(Z2003B01) are greatly acknowledged.
[15] H.P. Xu, H.T. Liu, H.W. Wang, L.J. Dong, X.G. Chen, Z.D. Hu,
Fresen. J. Anal. Chem. 368 (2000) 780.
References
[16] N.U. Persic-Janjic, A.A. Muk, V.D. Canic, Anal. Chem. 45 (1973)
798.
˜
[1] S. Eloy, M. Pablo, I.S. MarAa, R. Julio, Instrum. Sci. Technol. 30
(2002) 281.
[17] H.M. Irving, H. Freiser, T.S. West (Eds.), IUPAC Compendium of
Analytical Nomenclature, Definitive Rules., Pergamon Press, Oxford,
1978.
[18] S.X. Cai, C. Huang, Analysis of precious metal, in: Metallurgy, first
ed., Industry Press, Beijing, China, 1984.
[2] V.L. Alicia, B. Fernando, R. Julio, Anal. Chim. Acta 443 (2001) 17.
[3] M. Lanqun, Y. Katsunobu, Anal. Chim. Acta 415 (2000) 143.
[4] Z. Chengxiao, T. Haruyama, E. Kobatake, M. Aizawa, Anal. Chim.
Acta 408 (2000) 225.
[5] K. Rene´, H. Ludeˇk, F. Miroslav, P.P. Emil, Bioelectrochem. 55 (2002)
119.
[19] Z.B. Li, Q.E. Cao, J.L. Wang, Q.H. Xu, Chin. J. Anal. Chem. 28
(2000) 361.
[6] C.R.M. Rao, G.S. Reddi, Trends Anal. Chem. 19 (2000) 565.
[7] S.B. Khoo, R.D. Ye, Electroanal. (NY) 14 (2002) 420.
[8] L.M. Jamshid, H.S. Mohammad, A. Mohammad, Talanta 53 (2000)
61.
[9] R. Reinhard, N. Muriel, S. Boris, R. Oliver, J. Chem. Soc., Perkin
Trans. 2 2 (1999) 1615.
[20] K.H. Huang, H.L. Wan, Pricinple of Catalysis, first ed., Science
Press, Beijing, China, 1983, pp. 72–74 and 174.
[21] S. Rubio, A. Gomer-Hens, M. Valcarcel, Analyst 109 (1984) 717.
[22] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.Z. Zheng, Z.B. Wang, Method
of Fluorescence Analysis, second ed., Science Press, Beijing, China,
1990.