CHINESE JOURNAL OF CHEMISTRY 202, 20, 153-159 153

Sol-Gel Derived Carbon Ceramic Electrode Bulk-Modified with Tris ( 1 10-phenanthroline-5 6-dione ) iron ( 11 ) Hexafluorophos- phate and Its Use as an Amperometric Iodate Sensor

WANG, Peng(3a) ZHU, Guo-Yi'(&#%) Laboratory of E,?ectmwdytical Chemhtry, Changchun Institute of Applied Chhtv, Chinese Academy of S c i e m , Chungchwr , Jilin 130022, China

A surfacerenewable tris( 1,l&pbenaathroline-5, &done) irOa (n ) hexaauorophosphate (Fern) modified carboll ceramic electrode was mmtmckd by dispersing FeH) and graphite powder in mthyI~thoxysihe (MTMOS) based gek. 'Ihe F e P D d e d eiechode presented pH-dependent vdtarnnet- ric behavior, and its peak cunw~ts were dimsiorr contmued in0.1 d L N ~ S O , + H ~ S O , ~ I J ~ ~ O I I ( p H = 0 . 4 ) . In the preseace of iodate, dear eterhpeatalytic reductjon waves were observed and thus the &emidly modified electrode was used afaa ampemmetric sensor for iodate in common salt. The linear raage, sensitivity, detection limit and respolrse time of the iodate semor were 5 x 106-1 x 1O'md/L, 7.448 ~ A - W d, 1 . 2 x l Q 6 W L a n d 5 s . rospeetivdy. Adistiactad- vantage of this smor is ib good repddbility of surface-re-

d b Y s i m p l e - * pdishing.

KeymKds phemthroline-5, bdione) iron( II 1, iodate

sol-gel, carbon ceramic electrode, tris ( I , 10-

Introduction

'Ihe sol-gel pmess provides a versatile method for the fabrication of inorganic and inorganic-organic hybrid materials via the hydrolysis and condensation of suitable metal alkoxides.13 'Ihese materials possess notable ad- vantages over other inorganic-organic materials for the encapsulation of various dopants and the development of practical sensom and catalytic supports, which include ease of preparation, high pomsity, photochemical and electrochemical stability, and enhanced mechanical and chemical durability. Many studies on silica-modified

electrodes have been reported in the last decade. Recent advances in various fields and applications of sol-gel in electrochemistry were described in several excellent re- view artic1es.49 ~n 1994, ~ e v et d. '' first introduced a new class of sol-gel carbon c e h c e l e c d e s ( CCE) , which comprised a dispersion of graphite powder in an organically modified or unmodified silicate porous ma- trix. The hydrophobicity of these electmdes can be con- mlled easily using an organically modified sol-gel pre- cursor. An interesting feature of this kind of chemically modified electrodes is that the active section of the elec- trodes is not clogged upon repeated polishing due to the brittleness of the sol-gel silicate backbone, and thus the electrodes can be renewed by a simple mechanical polish after every use or contamination. Many efforta have been devoted to preparing chemically modified CCEs and us- ing them as senson for metal ions, glucose, and other important chemical and biological substances. "-I7

'Ihe action of 1 ,lO-phenanthrOline-5, 6-dione [ PD , Fig. 1 ( A) ] and its complexes with metal cations as oxi- dizing agents for organic amines and inorganic reducing agents and for biological reductants such as NADH has attracted considerable interest, and several experimental studies have appeared. "he electrochemical behavior of a variety of metal complexes of the ligands has also been described. In 1985, Goas and Abruna3' synthe- sized tris( 1 ,IO-phenanthroline5,6-dione)iron (II) hex- atluorophosphate [ FePD, Fig. 1 ( B) ] and studied its electrochemical and electrocatalytic properties for the

* E-mail: zhuguoyi@ m. ciac. jl. cn

Received March 26. 2001; revised June 25, 2001; accepted September 18, 2001. Roject supported by the Ministry of Science and Technology of China (No. 200lBA210AcW)

154 Carbon ceramic electrode WANG & ZHU

first time. Recently, Tobalina et ad. 32 reported the elec- trocatalytic oxidation of NADH at FePD-modified carbon paste electrodes. Our interest in FePD stemmed from its strong spontaneous adsorption on d t e and its possible interaction with silanol p u p , which means that FePD doped in CCE matrix may not leak out in aqueous 401~- ticn. In the present paper, FePD and graphite powder were first dispersed in methyltrimethoqdane (MTMOS) based gels to fabricate surface-renewable FePD-modified CCEs. The chemically modified electrode showed a high electrocatalytic activity toward iodate reduction, and thus was used as an amperometric sensor.

0 0

A: PD

0 II -

0

N 3 0 I 2+

B: FePD

Experimental

Sodium hexafluomphosphate and high purity p- phite powder ( average particle diameter, 1-2 pm) were obtained firm Aldrich. MTMOS (197%) was purchased from ACROS and used without further purifi- cation, Other reagents were all of analytical grade and used as received. Ultrapure water obtained from a Milli- pore Milli-Q water purification system was used through- out the experiments. All experimental solutions were thoroughly deoxygenated by bubbling pure argon through them for at least 15 min.

IR spectra were recorded on a Bio-Rad FE-7 in- frared spectrophotometer in KBr pellets. ' H NMR exper- iments were carried out at 29 t on a Varian U n i t y a spectrophotometer. Chemical shifts were expressed in 6 using TMS as intemal standad. Elemental analysis was performed on a GmbH VarioEL elemental analysis system

and a TJA POEMS inductively coupled plasma atomic e- mission spectrometer. Solution acidities were measured with a Model 620D pH meter ( Shangha~, China). A CHI 660 Electrochemical Wolkstation connected to an IBM-FC compatible computer was used for control of the electrochemical measurements and for datum collection. A conventional b e l e c t r o d e system was used. The working electrode was an unmodified or FePD-modified CCE . An Ag/AgCl ( saturated KCl) electrode was used as a reference electrode and a Pt gauze as a counter electrode. AU potentials were m e a s d and reported v e m s the Ag/AgCl electrode. Electrochemical experi- ments were pedomed at rwm temperature.

The following synthetic pmedure for PD was based on the report by Paw and Esienberg,a and it yielded the p d u c t almost quantitatively. A solution of H2SOd (40 mL) and HN4(20 d) frozen to - 10 "c was in ad- vance added to the mixture of 4.4 g of 1,lO-phenanthro- line monohydrate and 4.0 g of KBr at - 10 qC . 'Ihe mixture was slowly heated and then stirred at reflwt for 3 h . After cooled, the solution was poured over 500 g ice and neutralized carefully with 40% NaOH q e o u s solu- tion until neutral to pH 6 . 5 . The resulting mixture was filtrated to obtain a yellow solid. 'Ihe filtrate was extract-

sulfate and removing solvent to give a yellow solid. The total pure solid was obtained with the yield of 94%. 'H

ed with C H C l 3 followed by drying with d y h sodium

NMR ( c D c 1 3 ) : 9.13 (d , J = 4 . 0 & , 2H, 2-Hand9- H), 8.51 (d , J = 8 . 0 Hz, 2H, 4-H and 7-H) , 7.60 (9, J=4 .0 Hz, 2H, 3-H and 8-H); IR Y: 1688 (C = 0) cm-'; Anal. calcd for C12HaNz&: C 68.57, H 2.88, N 15.22; found C 68.52, H, 2.85, N 15.28.

FePD was prepared according to the literature3' with some modifications. 3 .2 equiv. of PD in ethanol was added to 1 equiv. of ferrous sulfate dissolved in water. The formation of the complex was apparent by the imme-

Vol. 20 No. 2 2002 Chinese Journal of Chemistry 155

diate formation of a dark-red solution. The mixture was allowed to stir at room temperature for 30 min and then a proper amount of hydrogen chloride was added. The complex was precipitated as a dark-red solid by the addi- tion of sodium hexafluorophosphate. ‘Ihe solid was col- lected, washed with water, dried in z~cuo at 50 “c and confimed to be [Fe(PD)3] (PF6)2 *2H20. ‘H NMR

of three PD ligands), 7.92 ( d , J = 4.0 Hz, 6H, 2-H and 9-H of three PD ligands) , 7.75 (q, J = 4.0 Hz, 6H, 3-H and 8-H of three PD ligands); IR Y: 1698 ( C = O ) , 840 (P-F) cm-I; Anal. calcd for &Ha- F12FeN6GPz: C 42.71, H 2.19, N 8.30, Fe 5.52, p 6.12; found C 42.65, H 2.23, N 8.32, Fe 5.47, P 6.18.

(DMSO) 6: 8.76 ( d , J = 8.0 Hz, 6H, 4-H and 7-H

Fabrication of d $ e d and FePD-mod$ed CCEs

Unmodified CCEs were prepared according to the reference method. lo The FePD-modified CCEs were con- structed by the following procedure. 0.75 mL of saturat- ed FePD methanol solution, 0.35 mL of MTMOS, and 0.03 mL of hydrochloric acid ( 1 1 moVL) were ultrason- ically mixed for 2 min, then 1.5 g of graphite powder was added and shaken on a vortex agitator for an addi- tional 3 min. The mixture was added to glass tubes with 3 mm inner diameter and 8 cm length, and the length of composite material in the tubes was conmlled to be about 0.8 cm. In addition, a little extra mixture had to be retained on the top of the electrodes, and the mixture in the tubes was slightly pressed on smooth plastic paper with a glass stick through the back. After drylng at room temperature for 48 h, the electrodes were polished with 600-grit emery paper to remove extra composite material and wiped gently with filter paper. Electric contact was made by silver paint through the back of the electrode.

Results and discussion

Fabrication of FePD-mdjkd CCEs

Additional water for the hydrolysis step was provid- ed by air humidity.” Base (NaOH or NhOH) or neu- tral ( N b F ) catalysts failed to produce rigid monoliths and yielded only silica-covered graphite powder. Large dceages of hydrogen chloride and MTMOS resulted in the rapid gelation before adding graphite powder and the bad

conductivip of the composite, respectively. The Si-C bonds remained unchanged during the hydrolysis and polymerization, and the methyl p u p remained exposed at the surface of porous silicate network. In the process of fabrication of the FePD-modified CCEs, a little extra mixture was needed to be retained on the top of the elec- trodes in order to obtain the whole and mimr-like sur- faces conveniently when they were first polished. In ad- dition, the composite material became h l e and thus it was difficult to obtain smooth electrode surfaces if the gelation temperature was higher than 60 “c .

Electrochenid behavior Of the FQ-modified CCE

In general, transition complexes of PD are chemi- cally more stable and electrochemically more reversible than the PD ligand itself, especially in alkaline media. Fig. 2 ( A ) shows cyclic voltammograms of the FePD- modified CCE in the 0 . 1 moVL N%S04 t HZSO, solu- tion (pH = 0 . 4 ) at various scan rates. In the potential range from - 0.15 V to + 0.65 V , a well-defined redox mponse for the quinone moiety of FePD doped in CCE matrix was observed. As could be ascertained, instead of two PD-based one-electron reductions for FePD in acetonitrile, in aqueous media the FePD-modified CCE exhibited a reduction wave, which was assigned to the 2ee/2H+ reduction of the quinone moiety to the corre- sponding catechol . n*32 previous studies3’ showed that three PD ligands of FePD were reduced at the same po- tential, so the total reaction of FePD should be a 6e-/ 6H + process. At the scan rate of 10 mV/s, the anodic and cathodic peak potential s e p t i o n ( AE, ) is 28 mV. Along with the increase of scan rates, anodic peaks shift positively while cathodic peaks shift negatively, and AE, increases gradually. As shown in Fig. 2( B ) , both anodic and cathodic peak currents are proportional to the sqwm roots of gcan rates in the acidic solution (pH = 0.4), suggesting that peak currents are difbion-con- trolled in the investigated scan rate range.

In the present work, square-wave voltammetry was adopted to measure the pH-dependent electrochemical behavior conveniently. Fig. 3 ( A ) presents square-

vol- for the FePD-modified CCE in aqueous solution with dif€erent acidities. It can be clearly seen that along with increasing pH, peak potential gradually shift to the more negative potential direction and the peak currents also decrease. Reduction of FePD doped in

156 Carbon ceramic electrode WANG & ZHU

40-

30

20

10

. 4 0: \

-10

-20

-30

- - -

- - -

W

-5

-10

-I5

-20

$ -25 4

-30

-35

-50

-40 I . I . I . L . 1 .

0.1 0.2 0.3 0.4 0.5 0.6 0.7 E N vs. Ag/AgCI

- -

- -

- - - .

401 I

4 0 2 4 6 8 10 12

y '"/mV'" s - IR

Flg. 2 ( A ) Typical cyclic voltammognrrns of the FePD-modi-

lutjon ( p H = 0 . 4 ) at different scan rates. Fmm inner curve to outer m e : 10, 20, 40, 60, 80 and 100 mV/s, respectively. (B) 'Ihe dependence d (a) an- odic and ( b ) cathodic peak cumnts on the square motsdscanraes.

fied CCE in 0.1 moVL N%SO, + &SO, aqueor~, 90-

the CCE matrix is accompanied by the evolution of pro- tons fnw solution to the electrode wetting section to maintain c b neutrality. Along with the increase of pH, dower diffusion rate of protons should be the reason for the current decrease, and the mom negative duction potentials can be elucidated by the Nemst equation. Plots of peak potential Venus pH for the F e P D d i e d CCE show good linearity in the pH range from 0.04 to 6.8, as shown in Fig. 3 ( B ) . Slopes in this pH range are -69 rnV/pH which is close to the theoretical value -60 mV/ pH expected for the &/6H+ redox process at morn tem-

perature. In addition, the peak cunents am not pmpor- tional to pmton concentrations, which may be caused by the effect of sol-gel-derived methylsilicate matrix.

A

45' ' -0;s . o i o . o , i s . o.;o ' 0,;s . o.io . ' EN vs. AgJAgCI

0.5 . B

0.4 -

ti 0.3 - z a f 0.2 - a & 0.1 - vI

0.0 -

- I 0 1 2 3 4 5 6 7 8

PH

a. 3 ( A ) Square-wave vohmmopm of the FePDd-

(a) N+W4 + &SO4 aqueoua solution (pH = 0.4); (b) glycine tndfersolution (pH=2 .0 ) ; ( c ) glyck butTersolution (pH=2 .8 ) ; (d) acetate buffer solu- tion ( p H = 4 . 2 ) ; ( e ) phosphate buffer solution (pH =5.3); (e) phosphatctndfcrsolution ( p H = 6 . 8 ) . Increment: 4 mV; fresuency: 10 Hz; can rate: 40 mV/s. (B) Relationship bamm peak potential a d pH for the FePD-modified CCE.

6ed CCE in aqueous solutions with various aciditia.

The determination of iodate in common salt is of particular importance since the content of iodate has a

straightforward infuence on human health. Trace quality

Vol. 20 No. 2 2002 Chinese Jd of Chemistry 157

0 -

< 2 -3- 4

-6

-9

of iodate was usually deteminated by spectrophotometric and fluorescent quenching techniques, but seldom by electrochemical methods. 35*M Although the ampemmetric response of the F e P D - d i e d CCE in pH 0.4 aqueous solution is enhanced compared with that in high pH solu- tions, the steady electrocadytic response of iodate in such a high acidic solution can not be achieved until pH > 2 .O. ?herefore, pH 2.2 glycine buffer solution was selected as media for the electrocatalytic reduction of io- date. As is known, the electrochemical reduction of io- date requires a large oveptential, and no obvious re- sponse was observed in the potential range from t 0.6 V to - 0.05 V at an unmodified CCE in pH 2.2 glycine buf€er solution [shown in Fig. 4( A) ] . In this work, the FePDmodified CCE showed an electrocatalytic activity toward iodate reduction. As shown in Fig. 4 ( B ) , adding iodate to the electrolyte cell caused a dramatic change in the cyclic vo l t amrnop with an increase in

-

-

1.5 I

-3.0 -0.15 0.M) 0.15 0.30 0.45 0.60

E N vs. AgIAgCI

3 I . ( - \ '

-12 1 . 1 . I . t .

0.00 0.15 0.30 0.45 0.60 EN vs. AgfAgCI

Fig. 4

cathodic current and a concomitant decrease in anodic currents, indicating that iodate was electrocatalytically reduced by the reducing product from FePD. At present, a codimed rnechnism of the electrocatalytic reduction of iodate at the FePD-modified CCE is unknown.

Amperometric sensing and uue$erence

On the basis of the voltammetric results described above, it is possible to use the FePD-modified CCE as ampemmetric iodate sensor. According to the potential dependence of electrocatalytic reduction currents at steady-state conditions, the optimum electrode potential was selected at t 0.23 V versus Ag/AgCl for ampero- metric measurement in order to obtain constant and high sensitivity. A typical hydrodynamic ampemmetry [ Fig. 5 ( A ) ] was obtained by successively adding iodate to con- tinuously stirred pH 2. 2 glycine buffer solution. The electrode response time was less than 5 s . 'Ihe fast re- sponse is attributed to the thin wetting section controlled by methyl group and the short penetration depth of io- date. Fig. 5 ( B) shows the calibration graph for iodate at the modified electrode. 'Ihe electrode response was linear for iodate within the concentration range 5 x 106-1 .Ox lo2 moVL, and the sensitivity is 7.448 pA* Urn1 (conelation coefficient of 0.999). ?he detection limit was 1.2 x l@' moVL when signal-to-noise was 3.

Interference effects were investigated by testing the response of the d i e d electrode to Na' , MH' , Ca2', Ba2', As3' , Pb2' , Zn2', Cl-, F and be- cause these species exist in common salt samples. Our experiments showed that 1OOO-fold excess of Na' , S@ and Cl-, UX)-fold excess of M$' , Ca2+ and Ba" , and 50-fold excess of Pb2' , Zn2 + , As3+ and F did not in- terfere in the determination of iodate. Therefore, the F e P D - d i e d CCE is used as ampemmetric sensor for iodate in COmmDn salt. The results of six replicate analy- ses were a mean of 44.10 d g with a relative standard deviation (ED) of 0.8% cornpaTed to 39.95 pg/g ob- tained by a standard titration method." It can be seen that the agreement between the two meth& is satisfacto-

(.4) Cyclic vol- of the unmodified CCE in pH 2 . 2 glycine M e r solution containing ( a ) 0 or ( b ) 1 mmoVL iodate. ( B) Cyclic voltammogram, of the FePDmodified CCE in pH 2.2 glycine M e r 90-

ry.

S&;ldy & mpwtabddy of sdme renewal

?he hydrophobic MTMOS monomer results in the lution containing ( a ) 0 , ( b ) 0.5 or ( b ) 1 m V L iodate. Scan rate: 10 mV/s.

electrode wetting in aqueous solutions to be controlled.

Ciuimn ceramic electrode RANG & ZHI' 158

tl60 s

A \ L

8 rn

8 I2O 100 i" rn

c/mol/L

Flg. 5 ( A ) Ampemmetric response for the FePDmodified CCE to the successive increments of 5 p V L iodate in pH 2.2 glycine buffer solution. Applied potential:

+0.23 V; stirring rate: loo0 rpm. ( B ) calibration w e for iodate.

Hence, a bulk modified electrode can be polished using an emery paper and a fresh surface exposed whenever needed. 'Ihis is especially useful for the electmcatalytic study since the catalytic activity is known to decrease when the electrode is fouled. Indeed, ten successive polishmp of a F e P D - d i e d CCE resulted in a RSD of 5.2%. In addition, no obvious current decrease was found when the electrode was i m m e d in a pH 2.2 glycine M e r solution for 15 d. We think that the high stability of the F e P D - d i e d CCEs is related to the sta- bility of the silicate matrix, the limited wetting section controlled by methyl group, the possible interaction be- tween the quinone moiety of FePD and silanol p u p s , 3 ' and mainly the strong adsorption of FePD on graphite .n*n

Conclusions

F e P D d i e d CCEs have been fabricated by the sol-gel method and their pexfomce has been evaluated under steady-state and dynarmc conditions. The chemi- cally modified electrodes showed a high catalytic activity toward iodate reduction and thus used as an amperornet- ric iodate sensor. 'Ihe hydrophobic, organically modified backbone of the electrode rejects water and leaves only a

thin active layer at the outermost section exposed to the analyte, which makes surface renewal by mechanical polishing possible. In addition, the sensor has other ad- vantages, such as simple preparation, fast response, good chemical and mechanical stability.

References

1 2 3

4 5

6

7 8 9

10

11

12

13

14

15

16

17

18

19

Avnir ,D.Aoc. Chan. R a . 1 9 9 5 . 2 8 . 3 2 8 . Hench, L. L . ; West, J . K . Chem. Rev. 1990, 90, 33. Boury, B . ; Corriu. R. J . P.; !%at, V. L.; Delord, P.; Nobili, M . Angnu. Chem. , Int. Ed. 1999, 38, 3172. Wang, J. Anal. Chim. Aao 1999, 399.21. Lev, 0.; Wu, 2.; Bharathi, S.; Glezer, V . ; Modestov, A . ; Gun, J.; Rabinovich, L. ; Sampth, S. Chem. M e . 1997, 9,2354. Alber, K. S.; Cox, J . A . M h c h i m . Acta 1997, 127. 134. Walcarius, A. Electmud* 1998, 10, 1217. Collinmn, M. M . Mikochim. Ada 1998, 129, 149. C o h . M . M . G i r . R e u . A n a l . C h e m . 1 9 9 9 . 2 9 , 289. Tsionsky, M . ; Gun, J . ; Clezer V . ; Lev, 0. Anal. Chem. 19w. 66, 1747.

L i , J . ; C h i a , L . S . ; C o h , N . K . ; T a n , S . N . J . Ek- &. chm. 1999,460,234. Pankratov, I . ; Iev, 0. J . Ekunmal. Chem. 1995, 393, 35. Wang, J . ; Pamidi, P. V. A . ; Naacimento, V. B . ; Anpea, L. El.&mMl* 1997, 9 , 6 8 9 .

Ha,S .Y . ;Kim,S . J . Elumand. Chem.1999,468, 131. Wang, P . ; Yw, Y.; Wang, X.; Z ~ U , G. J . Ek- buand. C h . #)o, 493,130. Wang, P . ; Wang, X . ; Bi, L . ; Zhu, G . Analyst #)o,

125, 1291 * Wang. P . ; Wang. X . ; Jing, X . ; Zhu, G. A d . Chim. h. #)o, 424, 481. Eckert. T. S.; Bruice, T. C. J . Am. Chem. Soc. 1987, 105. 4431. Hilt, G . ; Steckhan. E. J . Chem. Soc., Chem. Corn-

Vol. 20 No. 2 2002 Chinese Journal of Chemistry 159

20

21

22

23

24

25

26

n 28

29

mwr. 1993, 1706. Eckert. T. S.; BNice, T. C . ; Gainor, J . A . ; Weinreb, S. A . Proc. Nad. Acad. sn'. U . S . A . 198L. 79,2553. Itoh, S.; Fukushima, H . ; Komatsu, M . ; Ohshiro, Y. Chem. La. 1992, 1583. G i R . D.;HiII, R. E . E. J . Chem. S O C . , LM&n T m . 1974, 1217. Pariente, F . ; hrenzo, E.; A b m , H. D. Anal. Chem. 19w, 66, 4337. Evans, D. H.; Gliffith, D. A . J . E M . Chem. 19QL, 134, 301. Evans, D. H.; Gliffith, D. A . J . Elecawnal. Chem. 1982, 136, 149. h i , Y . ; A n s o n , F . C . J. Am. Chem. SOC. 1995, 117, 9849. Shi, C . ; h n , F. C. Anal. Chem. 1998, 70, 1489. Hilt, G . ; Jarbawi, T . ; Heineman, W . R . ; Steckhan, E. Chem. Eur. J. 1997, 3, 79. Wu, Q.; Maskus, M . ; Pariente, F . ; Tobalina, F . ; Fer-

nanda, V. M . ; lorem, E . ; Abruna, H . D. Anal. h. 1996, 68,3688.

30 hi, Y . ; Shi, C . ; Anson, F. C. Inorg. Chem. 19%, 35, 3044.

31 Gas, C.; Abruna, H. D. Inorg. Chem. 1985, 24, 4263.

32 Tobalina, F.; Pariente, F . ; Hernandez, L . ; Abruna, H. D.; lorem, E. A d . Chim. Acta 1999, 395, 17.

33 Paw, W.; Wisenberg, R. Inorg. Gem. 1991, 36, 2287. 34 Cespedea, F . ; Martinez-Fabregas, E . ; Alegmt, S. Trend

Anal. 6 . 1 9 9 6 , l S , 296. 35 Song, W.; Chen, X . ; Jiang, Y . ; Lu, Y . ; Sun, C . ;

36 Koemimky, L . ; Bertotti, M . J . Ebctrwnal. Chan. 1999, 471, 37.

37 Analytical Chemistry Division of Central-South Institute of Mining and Metallurgy, Hadbook .f chanical M y s i r , Science Press, Beijing, IWlZ (in Chinese).

wang,x. Anal. chim. Acta1999, 3w,73.

(EO103262 LI, L. T. ; DONG, L. J . )