SelectiveCoprecipitationofChromium（III）inWater withScandium HydroxidePriortoGraphiteFurnace AtomicAbsorptionSpectrometricDetermination

(1)

Introducti on

There i s substanti al i nterest i n speci ati on of chromi um i n w ater.

Chromi um （III ） and chromi um （VI ） are foundi naqueousmedi a.Thechromi um （III ） speci es are chi efl y present as hydroxo compl exes.Thechromi um （VI ）speci esare domi natedbychromateandarehi ghl ytoxi c.

Speci ati on of chromi um （III ） and chromi um （VI ） on the basi s of copreci pi tati onhasbeenwi del yi nvesti gated because of i ts si mpl i ci ty. Sel ecti ve copreci pi tati on of chromi um （III ） was studi edusi ng i ron hydroxi de,

¹⁵

al umi ni um hydroxi de,

⁶

gal l i um hydroxi de,

⁷

gal l i um phosphate,

⁸

hafni um hydroxi de

⁹

andyttri um

hydroxi de.

¹⁰

Forsel ecti vecopreci pi tati onof chromi um （VI ） ,reagentsempl oyedwerel ead sul fate

^11,12

and zi nc-di ethyl di thi carbamate.

¹³

In some cases, si nce the presence of a copreci pi tantaffects the determi nati on of chromi um,i ti s necessary to remove the copreci pi tant

⁵

ortoaddachemi calmodi fi er

⁷

pri ortothedetermi nati on.

Al thoughrareearthhydroxi dessuchas l anthanum hydroxi de,

^14,15

terbi um hydrox- i de

¹⁶

anderbi um hydroxi de

¹⁷

wereal soappl i ed to the copreci pi tati on of chromi um, speci ati on of chromi um （III ） and chromi um （VI ） hasnotbeenachi evedexcept for yttri um hydroxi de. We have been studyi ngthesystemati csofcopreci pi tati on

Sel ecti veCopreci pi tati onofChromi um （III ） i nWater wi thScandi um Hydroxi dePri ortoGraphi teFurnace

Atomi cAbsorpti onSpectrometri cDetermi nati on

TomoharuM INAMI * ,Yoshi kiS OHRIN * andJoi chiU EDA **

原著論文

* Insti tuteforChemi calResearch,KyotoUni versi ty,Uj i611-0011,Japan

** Facul tyofEducati on,KanazawaUni versi ty,Kakumamachi ,Kanazawa920-1192,Japan Accepted8February2006

Abstract

Chromi um （III ）wasseparatedfrom chromi um （VI ）bythecopreci pi tati onwi thscandi um

hydroxi deanddetermi nedbygraphi tefurnaceatomi cabsorpti onspectrometry （GF-AAS ） .In

thepresenceof0. 28moldm

^－3

sodi um sul fateand0. 28 ×10

^－3

moldm

^－3

di sodi um hydrogen

phosphate,chromi um （III ）upto10・ gdm

^－3

wasquanti tati vel ycopreci pi tatedwi thscandi um

hydroxi deatpH 9. 8,whi l e50 ・ gdm

^－3

ofchromi um （VI ） washardl ycopreci pi tated.Underthi s

condi ti on,chromi um （III ） was separated from chromi um （VI ） when the concentrati on of

chromi um （VI ） wasl essthanfi veti mesthatofchromi um （III ） .Thesum ofchromi um （III ） and

chromi um （VI ）wasdetermi ned i n a si mi l arway afterthereducti on ofchromi um （VI ）to

chromi um （III ） wi th2. 5gdm

^－3

ofhydroxyl ammoni um chl ori deat0. 12moldm

^－3

ofhydrochl ori c

aci d.Thi scopreci pi tati on method w asappl i ed to thedetermi nati on ofchromi um （III ）and

chromi um （VI ） i nawatersampl ecol l ectedfrom anurbanri ver.

(2)

wi th rareearth hydroxi des.In aprevi ous study,wefoundthatscandi um hydroxi dei s promi si ng forthecopreci pi tati on oftrace metal s pri or to graphi te furnace atomi c absorpti onspectrometry （GF-AAS ） ,

^18,19

si nce scandi um di dnoti nterferewi ththeGF-AAS determi nati onsofsomemetal s.Inaddi ti on, wefound i n prel i mi nary experi mentsthat the copreci pi tated percentage of chromi um （VI ） was l ower for heavy l anthanoi dandscandi um hydroxi des,whi l e chromi um （III ） was quanti tati vel y copreci pi tated.

Inthi spaper,wedescri bethecondi ti ons ofcopreci pi tati onwi thscandi um hydroxi de for the separati on ofchromi um （III ） and chromi um （VI ） and the determi nati on of chromi um （III ）andchromi um （VI ）i nwater sampl es.

Experi mental Apparatus

AShi madzuAA-660Gatomi cabsorpti on spectrometer wi th a GFA-4A graphi te furnaceatomi zerwasusedforthechromi um determi nati ons.Thegraphi tetubeusedwas a hi gh densi ty graphi te tube （not pyrol yzed ） . The opti mum operati ng condi ti onsforchromi um aresummari zedi n Tabl e1.Thebackgroundwascorrectedusi ng aD

²

l ampthroughouttheexperi ments.A Perki n El merOpti ma 2000DV i nducti vel y- coupl ed pl asma atomi c emi ssi on spectrometer was used for prel i mi nary experi ments.A Hi tachi -Hori ba ModelM-5 pHmeterwi thagl ass-el ectrodewasusedfor

pH measurements.

Reagents

Reagentsofguaranteedreagentgrade were used except for scandi um oxi de.

Dei oni zed water （El i x-5-Gradi ent-A10, Mi l l i pore ） wasusedforthepreparati onofal l sol uti ons.

Chromi um sol uti ons. Chromi um （III ） sol uti on（1 g dm

^－3

） was prepared by di ssol vi ng770mgofchromi um （III ） ni trate nonahydrate （Nacal aiTesqueInc. ）wi th a smal lamountofconcentratedni tri caci dand di l uti ng to 100 cm

³

wi th dei oni zed water.

Chromi um （VI ） sol uti on （1 g dm

^－3

） was preparedbydi ssol vi ng283mgofpotassi um di chromate （WakoPureChemi calIndustri es, Ltd. ） i n100cm

³

of0. 1moldm

^－3

ni tri caci d.

Phosphatesol uti on. Phosphatesol uti on

（0. 28moldm

^－3

）waspreparedbydi ssol vi ng 1. 246 g of di sodi um hydrogen phosphate di hydratei n25cm

³

ofdei oni zedwater.

Scandi um sol uti on.Scandi um sol uti on

Tabl e1 Operati ng condi ti onsofGF-AAS forchromi um

Sampl evol ume/mm

³

20 Anal yti calwavel ength/nm 357. 9

Lampcurrent/mA 10

Sl i twi dth/nm 0. 50 Argongasfl owrate

Innergas/dm

³

mi n

^－1

1. 5

Outergas/dm

³

mi n

^－1

1. 5

Dryi ng Ramp,150 ℃,50s

Ashi ng Step,600 ℃,30s

Atomi zi ng Step,2500 ℃,4s

^a

Cl eani ng Step,2800 ℃,2s

a.Innergaswasstopped.

(3)

（1. 5gdm

^－3

） waspreparedbydi ssol vi ng230 mgofscandi um oxi de （99. 9%,WakoPure Chemi cal Industri es, Ltd. ） i n 8 cm

³

of concentrated ni tri c aci d on heati ng and di l uti ngto100cm

³

wi thdei oni zedwater.

RecommendedProcedure Determi nati onofchromi um( III)

Anal i quotofsampl esol uti on （200cm

³

） waspl acedi naPyrexgl assbeaker,and8gof sodi um sul fate anhydrous （0. 056 molas sul fatei on ） ,0. 2cm

³

ofphosphatesol uti on

（0. 056 ×10

^－3

molasphosphatei on ）and2. 7 cm

³

ofscandi um sol uti on （4mgasscandi um i on ） wereadded.ThepH wasadj ustedto9. 8 wi th sodi um hydroxi de sol uti on to form scandi um hydroxi de.Fresh watersampl es werekeptforseveralmi nutesandseawater sampl es were kept for ten mi nutes.The preci pi tatewascol l ectedona3G4si ntered- gl assfi l ter,washedwi thasmal lamountof dei oni zedwater,anddi ssol vedwi th0. 8cm

³

of concentrated ni tri caci d.Thesol uti on was madeupto10cm

³

wi thdei oni zedwaterand theatomi cabsorbanceofchromi um i nthi s sol uti onwasmeasuredundertheoperati ng condi ti onsgi veni nTabl e1.Thecal i brati on curveforchromi um （III ）wasprepared by usi nganaqueoussol uti oncontai ni ng1. 2mol dm

^－3

ni tri caci dand0. 4gdm

^－3

ofscandi um.

Determi nati onofchromi um( VI)

Anotheral i quotofthesampl esol uti on

（200cm

³

） waspl acedi naPyrexgl assbeaker, 0. 5gofhydroxyl ammoni um chl ori deand2 cm

³

ofconcentratedhydrochl ori caci dwere

added,andi twasl efttostandforanhouri n order to reduce chromi um （VI ） to chromi um （III ） .Thecopreci pi tati onmethod for chromi um was the same as thatfor chromi um （III ） menti onedabove.Chromi um i nthefi nalsol uti onwasdetermi nedbyGF- AAS,gi vi ngtheconcentrati onforthesum of chromi um （III ） and chromi um （VI ） . The concentrati on for chromi um （VI ） was obtai nedbysubtracti ngchromi um （III ） from the sum of chromi um （III ） and chromi um （VI ） .

Resul tsandDi scussi on

Determi nati onofchromi um byGF-AAS In these experi ments,sol uti ons were prepared by mi xi ng the sol uti ons of scandi um, chromi um and ni tri c or hydrochl ori caci dwi thoutacopreci pi tati on procedure.

Thecondi ti onsforthedetermi nati onof chromi um byGF-AASwereexami nedi nthe presence of scandi um. The presence of scandi um （upto7. 4gdm

^－3

） di dnoti nterfere wi th thedetermi nati on ofchromi um.The backgroundabsorbanceat357. 9nm wasnot affected by scandi um. Therefore, the presenceofscandi um doesnotcausespectral i nterferenceforchromi um.Iti sthoughtthat scandi um doesnotcauseanyi nterferencei n theashi ngoratomi zi ngstepssuchasl ossof chromi um or changes i n the atomi zi ng effi ci ency.

Theeffectoftheconcentrati onofni tri c

and hydrochl ori c aci ds on the chromi um

absorbance was exami ned.Al though the

(4)

absorbancedecreasedwi th i ncreasesi n the aci dconcentrati on,i twasal mostconstanti n therangeof1-2moldm

^－3

forni tri caci dand 1. 5-2moldm

^－3

forhydrochl ori caci d.Thus, 1. 2moldm

^－3

ofni tri caci dwasusedi nthe subsequentexperi ments.

The absorbance and concentrati on of chromi um showed a l i near rel ati onshi p, passi ng through theori gi n,up to200・ g dm

^－3

.Therel ati vestandard devi ati on was 4. 1%forthefi verepl i catemeasurementsof 100・ gdm

^－3

chromi um.Thedetecti onl i mi t, whi ch was cal cul ated as three ti mes the standarddevi ati onforthebl ank （contai ni ng 0. 4gdm

^－3

ofscandi um and1. 2moldm

^－3

of ni tri caci d ） ,was0. 2 ・ gdm

^－3

.

Theopti mum condi ti onsforcopreci pi tati on ofchromi um( III)

As reported i n a previ ous paper,

¹⁸

scandi um hydroxi de quanti tati vel y preci pi tatesabovepH 8.Thepreci pi tatei s readi l y fi l terabl e and di ssol ves easi l y i n di l utedmi neralaci ds.

In these experi ments, sul fate and phosphate i ons were not added to the aqueoussol uti oncontai ni ngchromi um （III ） orchromi um （VI ） beforeei therchromi um wascopreci pi tatedwi thscandi um hydroxi de.

TheeffectofpHonthecopreci pi tati onof chromi um （III ） or chromi um （VI ） was exami ned wi th 200 cm

³

of the sol uti on contai ni ng4mgofscandi um and1 μgofthe anal yte.Theresul tsareshown i n Fi g.1.

Chromi um （III ） was quanti tati vel y copreci pi tatedwi thscandi um hydroxi dei na

pH range of 8. 3-11. 3.The copreci pi tated percentageofchromi um （VI ） was11%atpH 8anddecreasedathi gherpH val ues.Onthe otherhand,i ti spossi bl eforwatersampl es that preci pi tates such as magnesi um hydroxi deareformedathi gherpH val ues.

Thus,wedetermi nedtheopti mum pH tobe 9. 8 for sel ecti ve copreci pi tati on of chromi um （III ） .

Theeffectofthescandi um amount （1- 80mgdm

^－3

） oncopreci pi tati onof5 ・ gdm

^－3

chromi um （III ）wasexami ned.Morethan1 mg dm

^－3

ofscandi um was necessary for quanti tati ve copreci pi tati on of chromi um （III ） . Thus, 20 mg dm

^－3

of scandi um was adopted as the opti mum condi ti on.

Interferenceofdi versei ons

The effectof28 di verse i ons on the copreci pi tati on and determi nati on of chromi um （III ） wasexami nedusi ngasampl e sol uti on（50 cm

³

） contai ni ng 1 ・ g of chromi um （III ） and4mgofscandi um.Tabl e 2 shows that l arge amounts of sodi um,

0 50 100

8 9 10 11

pH

% coprecipitated

Fi g.1 EffectofpH on copreci pi tati on of 1・ g/200cm

³

ofchromi um （III ）（●）

andchromi um （VI ）（○）wi th4mgof

scandi um.

(5)

potassi um,magnesi um andcal ci um di dnot i nterfere wi th the copreci pi tati on.Itwas found thattheother i onstested di d not i nterferewhenthei on （mg ） /scandi um （mg ）

rati owasl essthan0. 25.

Tabl e2 Tol erancerati oofdi versei onstoscandi um forthecopreci pi tati onanddetermi nati onof 1 ・ gofCr （III ）

Tol erancerati o

^a

Ion ［mg ］ /Sc ［mg ］ Ion

600 Na

^＋b

260 K

^＋b

250 Ca

²^＋c

50 Mg

²^＋c

0. 25 Li

^＋

,

^b

Sr

²^＋

,

^c

Ba

²^＋

,

^b

Al

³^＋

,

^c

La

³^＋

,

^b

Th

⁴^＋

,

^c

Zr

⁴^＋

,

^b

MoO

⁴²^－

,

^d

WO

⁴²^－

,

^e

Mn

²^＋

,

^c

Fe

³^＋

,

^c

Co

²^＋

,

^c

Ni

²^＋

,

^c

PtCl

⁶²^－

,

^f

Cu

²^＋

,

^c

AuCl

⁴^－

,

^f

Zn

²^＋

,

^c

Cd

²^＋

,

^c

Ga

³^＋

,

^b

In

³^＋

,

^c

Sn

⁴^＋

,

^b

Pb

²^＋

,

^c

Sb

³^＋

,

^b

Bi

³^＋c

Copreci pi tati onwascarri edoutwi th4mgofscandi um i n50cm

³

ofthesampl esol uti onatpH 9. 8.

a. Theerrorsi nthedetermi nati onofchromi um werewi thi n±5%,whentherati oofthedi verse i onswasl essthantheval ue.

b. Counteri onwasachl ori dei on.

c. Counteri onwasani tratei on.

d. Counteri onwasanammoni um i on.

e. Counteri onwasasodi um i on.

f. Counteri onwasahydrogeni on.

Tabl e3 Copreci pi tati onofchromi um （VI ） i nthepresenceofani ons

Ani on Cr （VI ）

Speci es Added

moldm

^－3

Added

・ gdm

^－3

Copreci pi tated

%

None 0 50 20. 6 ±4. 8

Cl

^－

0. 028 50 18. 8 ±1. 4

NO

3^－

0. 028 50 16. 3 ±2. 2

SO

4²^－

0. 028 50 4. 3 ±0. 6

SO

4²^－

0. 28 50 3. 2 ±0. 1

CO

3²^－

0. 70 ×10

^－3

50 6. 0 ±0. 5

PO

4³^－

0. 28 ×10

^－3

50 3. 7 ±0. 0

SO

4²^－

＋PO

4³^－

0. 28 ＋0. 28 ×10

^－3

50 0. 9 ±0. 1 SO

4²^－

＋PO

4³^－

0. 28 ＋0. 28 ×10

^－3

5 1. 6 ±1. 2 Theresul tsaretheaverageandstandarddevi ati onforthreerepl i cateexperi ments.ICP-AESwas usedforthedetermi nati on.

Copreci pi tati onwascarri edoutwi th4mgofscandi um i n200cm

³

ofthesampl esol uti onatpH 9. 8.

Theani onswereaddedassodi um sal ts.

(6)

Separati on of chromi um( III) and chromi um( VI)

Inordertofi ndthebestcondi ti onfor separati ng chromi um （III ） from chromi um （VI ） , the copreci pi tati on of chromi um （VI ） wi th scandi um hydroxi de wasi nvesti gatedi nthepresenceofavari ety of ani ons. The ani ons such as sul fate, phosphate,carbonate,ni trateand chl ori de wereaddedassodi um sal ts.Theresul tsare shown i n Tabl e3.Twenty onepercentof chromi um （VI ） was copreci pi tated i n the absence of ani ons. The copreci pi tated percentagewassuppressedi nthepresenceof ani ons.Thesuppressi oneffecti ncreasedi n theorderofCl

^－

＜ NO

³^－

＜ SO

⁴²^－

＜ PO

⁴³^－

・ CO

³²^－

.In theexperi ments,phosphateand carbonatei onswereexami nedat0. 28 ×10

^－3

and 0. 70 × 10

^－3

mol dm

^－3

, respecti vel y, becausetherecoveryofscandi um decreased whentheconcentrati onsoftheseani onswere hi gher than these val ues. Addi ng both sul fateand phosphatei onswasthemost effecti vei nsuppressi ngthecopreci pi tati onof chromi um （VI ） .

Inthepresenceofbothsul fate （0. 28mol dm

^－3

）andphosphatei ons （0. 28 ×10

^－3

mol dm

^－3

） , copreci pi tati on of 5 ・g dm

^－3

chromi um （III ） wi th scandi um hydroxi de was tested. The average recovery and standarddevi ati onwere96. 3 ±5. 0% （n ＝ 3 ） . Thus,thi s condi ti on was adopted as the recommendedprocedure.

The separati on of 5 ・g dm

^－3

chromi um （III ） from chromi um （VI ） was tri edfol l owi ngtherecommendedprocedure.

Theresul tsareshowni nTabl e4.Fourrati os ofchromi um （VI ） /chromi um （III ） ,namel y,0, 1,5 and 10,were tested.The observed concentrati onsofchromi um di dnotshow a si gni fi cantdi fferencewhentherati owasl ess than 5.Thus,i t can be concl uded that chromi um （VI ） does not show si gni fi cant i nterference for the determi nati on of chromi um （III ） when the chromi um （VI ） /chromi um （III ）rati oi sl ess than5.

Tabl e4 Separati onofchromi um （III ） from chromi um （VI ） fol l owi ngtherecommendedprocedure

Added Found

Cr （III ）

・ gdm

^－3

Cr （VI ）

・ gdm

^－3

・ gdm

^－3

Percentageof

addedCr （III ）

5 0 4. 81 ±0. 25 96. 3 ±5. 0

5 5 4. 83 ±0. 13 96. 7 ±2. 5

5 25 5. 05 ±0. 05 101 ±1

5 50 5. 35 ±0. 32 107 ±6

Theresul tsaretheaverageandstandarddevi ati onforthreerepl i cateexperi ments.ICP-AESwas usedforthedetermi nati on.

Copreci pi tati onwascarri edoutwi th4mgofscandi um i n200cm

³

ofthesampl esol uti onatpH 9. 8.

(7)

Recoveri es of spi ked chromi um( III) and chromi um( VI)from watersampl es

Therecoveri esofspi kedchromi um （III ） from ri ver and sea water sampl es were exami ned.Theri verand sea waterswere col l ected from the Uj i Ri ver, Kyoto Prefecture and Nanao Bay, Ishi kawa Prefecture,respecti vel y.Thesesampl eswere fi l tered through an Advantecquanti tati ve

fi l terpaper （No.5C,di ameter300mm） and aci di fi edwi th0. 12moldm

^－3

ofhydrochl ori c aci d.Thi s was done to di ssol ve col l oi dal chromi um adsorbed on manganeseand/or i ron oxi de. The copreci pi tati on of chromi um （III ） wascarri edoutj ustafterthe addi ti onofhydrochl ori caci d.Asshowni n Tabl e 5, spi ked chromi um （III ） was quanti tati vel yrecovered.

Tabl e5 Recoveri esofchromi um from spi kedwatersampl es

Sampl e Cr （III ） Cr （VI ）

Sampl i ng

date Found

・ gdm

^－3

Added

・ gdm

^－3

Recovery

% Found

・ gdm

^－3

Added

・ gdm

^－3

Recovery

% Ri verwater

Jul .19 0. 15 0. 37

5 105 ±2 5 103 ±9

10 101 ±4 10 101 ±3

Seawater

Aug.6 0. 20 0. 09

5 105 ±8 5 102 ±1

10 102 ±8

Jul .24 － 0. 91*

10 97. 2 ±1. 8 Theobtai nedrecoveri esaretheaverageofthreerepl i cateexperi ments.

Chromi um wasdetermi nedbyacal i brati oncurvemethod.

Theri verandseawaterswerecol l ectedfrom theUj iRi ver,KyotoPrefecture,andNanaoBay,Ishi kawa Prefecture,respecti vel y.

*Thesum ofchromi um （III ） andchromi um （VI ） .

Tabl e6 Determi nati onofchromi um i ntheKanameRi verwater Cr （III ）

・ gdm

^－3

Cr （III ）＋Cr （VI ）

・ gdm

^－3

Cr （VI ）

・ gdm

^－3

Cal i brati oncurvemethod 26. 0 ±1. 8 29. 4 ±0. 2 3. 4 ±2. 2 Standardaddi ti onmethod 23. 2 ±1. 9 29. 0 ±0. 8 5. 8 ±2. 5 Theresul tsforCr （III ）andCr （III ）＋Cr （VI ）aretheaverageandstandarddevi ati onforthree repl i catesampl es.

Theconcentrati onforchromi um （VI ） wasobtai nedbysubtracti ngchromi um （III ） from thesum

ofchromi um （III ） andchromi um （VI ） .

(8)

Theprocedureforthedetermi nati onof chromi um （VI ）wasal soi nvesti gatedusi ng the reducti on of chromi um （VI ） to chromi um （III ）pri ortothecopreci pi tati on.

Thereducti oncondi ti onwasdesi gnedonthe basi softhepaperofKagayaetal .

⁸

Tothe sampl e sol uti on （200 cm

³

） contai ni ng hydrochl ori caci d （0. 12moldm

^－3

） ,0. 5gof hydroxyl ammoni um chl ori deasareduci ng agentwasaddedandi twasl efttostandfor anhour.Chromi um （VI ）addedtotheri ver and sea water sampl es was reduced to chromi um （III ） and quanti tati vel y copreci pi tated wi th scandi um hydroxi de

（Tabl e5 ） .

Bl ankrunswerecarri edoutusi ng200 cm

³

ofdei oni zedwaterasasampl e.Thebl ank val uewas0. 22 ±0. 03・ gdm

^－3

forchromi um

（n ＝ 3 ） .

Appl i cati on to the determi nati on of chromi um i nri verwater

Therecommendedprocedurewasappl i ed tothedetermi nati onofchromi um i nri ver water.Theri verwatersampl ewascol l ected from theKanameRi ver,Ishi kawaPrefecture on17August2005.TheKanameRi veri sa tri butaryfl owi ngi nanorthwestresi denti al areaofKanazawaCi ty,wherethereareafew factori esi ncl udi ngametalpl ati ngfactory.

Thepretreatmentoftheri verwatersampl e wassi mi l artothatmenti onedabove.The fi nalsol uti onoftherecommendedprocedure wasdi l utedtenti meswi th1. 2moldm

^－3

of ni tri c aci d, si nce the concentrati on of chromi um wasvery hi gh.Theresul tsare

shown i n Tabl e6.Thedetermi nati on was carri edoutby acal i brati on curvemethod andastandardaddi ti onmethod.Forboth chromi um （III ） and chromi um （III ）＋ chromi um （VI ） ,theconcentrati onsobtai ned wi th the two methods di d not show a stati sti cal l y si gni fi cant di fference on the basi sofat-test （p ＝ 0. 05 ） .

References

１．T.Fuj i naga,T.Kuwamoto,S.Murai ,S.

Ki hara and E. Nakayama, Ni ppon KagakuZasshi , 1971 ,92, 339.

２．R. E.CranstonandJ. W.Murray, Anal . Chi m.Acta, 1978 ,99, 275.

３．T.Shi mi zu,R.Ki moto,Y.Shi j oandK.

Sakai , BunsekiKagaku, 1981 ,30, 66.

４．F.Sugi moto,Y.MaedaandT.Azumi , Ni pponKai suiGakkai shi , 1988 ,42, 22.

５．E.Nakayama,T.Kuwamoto,H.Tokoro and T.Fuj i naga, Anal .Chi m.Acta, 1981 ,131, 247.

６．T.Shi gematsu,S.Gohda,H.Yamazaki and Y.Ni shi kawa,Bul l .Inst.Chem.

Res. ,KyotoUni v. , 1977 ,55, 429.

７．A.Boughri et,L.DeranandM.Wartel , J.Anal .At.Spectrom. , 1994 ,9, 1135.

８．S.KagayaandJ.Ueda, Bul l .Chem.Soc.

Jpn. , 1995 ,68, 2843.

９．J.Ueda,H.SatohandS.Kagaya, Anal . Sci . , 1997 ,13, 613.

10 ．M. C.MehraandA. S.Gui mond, Asi anJ.

Chem. , 1999 ,11, 409.

11 ．T. D. Marti n and J. K. Ri l ey, At.

Spectrosc. , 1982 ,3, 174.

12 ．G.Vos, Freseni usJ.Anal .Chem. , 1985 ,

(9)