- Email: [email protected]
A PVC-coated carbon rod ion-selective electrode for thallium and its application to the analysis of rocks and minerals
SHORT ~O~~~AT~~~S A ~YC-CRATED CARBON ROD ION-S~LE~~~ ELECTRODE FOR THALLIUM AND ITS APPLICATION TO THE ANALYSIS OF ROCKS AND MINERALS
(Received 22 July 1987. Revised 28 January 1988. Accepted 5 February 1988)
Smmn~ary-A new Tl(III) ion-selective electrode prepared by coating a graphite rod with a PVC membrane has been made and investigated. The optimum membrane ~~~~tiou is 3~~lrh~a~ne E-‘I’lw 5 mg, ~~~~~~~ 0.1 zrx&dketyf ph~alate 8.4 n& PVC O.17 g. The &&rode exhibits I+.znstian response to n(m) ovee tile concentration range 6.5 X I&%2.5 x IO-%. The detection limit is 3.5 x IO-?&f and the slope of the electrode response is 53 rt 1 mV per decade. The electrode shows high stabiiity and selectivity. The electrode is easy to make and store, and inexpensive. Thallium, which shows crystallochemical affinities with the ~~he~i~~~ ~~~o~a~~ e-lements ~tasai~~ and ~~~~~~~ is an &rnerXt of ~I~~~ interest. 1% has been proposed as a geochernicai indicator in problems such as petrogenesis, rock-sea-water interaction, and evaluation of the degree of sulphur saturation in a silicate melt.‘*2 The abundance of thallium in geological material is very low so its dete~na~o~ requires highly sensitive analytical methode, such as fluorimetry, neutron activation, atomic-absorption spectrometry (AAS) and anodic stripping voltammetry (ASV). Fluorimetry has severe limitations due to interference, neutron activation requires expensive facilities and a long time for ~~a~a~o~ and coo&g of samples,arrdAAS and ASV often require enrichment methods. We have already developed a PVC membrane electrode based on the Butylrhodamine B-AuCl; ion-pair as electraactive agent,3 and now propose Eu~lrh~~i~e B-TlC& as an eleetroactive agent for use in a PVC coating on a carbon rod efeetrcrde. This electrode, which needs nu internal reference solution and internal reference electrode, is easier and cheaper to make than conventional PVC membranes or all solid-state PVC membranes.4 The electrode has been used successfully for the dete~nation of Tl(II1) in ~olo~~~ reference samples. EXPERIMENTAL
&tgeFm Analytical-reagent grade chemicals were used. Tl(III) stock solution,5 1 x 10-2M, dituted as required, w&b MM sodium chloride and 0.5&f hydr~lo~ acid. Carbon rod, ~t~~o~y pure, 8 mm d&meter. Preparation of efectmde
Weigh 0.2 g of Butylrhodamine B and dissolve it in 5 ml of demineralized water, then add 1 ml of concentrated hydrochloric acid, 0.7 g of sodium chloride and IO ml of
lO%f Tl(II1) stock solution dropwise. Centrifuge and wash the red precipitate with two 54 portions of 20% hydro&lo&z acid. Dry the prod& at So”. The ODeium i%kei?lbmne6xxllwition Ras been four& to be 5 mg of Butylrhodarnine BlTlq, 0.1 ml of &Iorobenzene. 0,4 ml of dioctvl ohthalate and 0. I7 Kof PVC. Mix these reagents in the ~160s stated, add 10 ml Gf tetrahydrofuran, stir, and let stand for 10 hr. Coat the carbon rod by inunersing it in the aoIution, removing it and allowing it to dry” Repeat the coating%taking care to get a ~~-d~s~but~ meznbrane. Leave the rueznbraoe in the air for about 12 hr. A thin membrane has poor stability and a short lifetime, but the response is slow if the membrane is too thick. A faur-layer membrane is a suitable compromise. Figure 1 shows the structure of the coated carbon rod PVC membrane electrode. Before use, the electrode should be activated For 2 hr by min&alized water until the potential &a&e to an SCE is stable. The cell is carbon rod 11 PVC sensorlunknown soln. 11 SCE
Tl(II1) hydrolyses and precipitates at pH >5, A reasonable pH range for use is O-5. The membrane resistance measured by the parallel resistance method is 0.1 MB The electrode response is immediate if the ~u~tmtiou of Tl@f) is >10-% The response time is about 30 set in ~~-~-~~-~~ Tt(rrI). The
potential reading taken every 30 min aver a period of 3 hr had a drift of only rfr1 mV. Potential readings taken five times in lo-‘M and 10e6M Tl(II1) alternately, showed no memory effect. Table 1 shows the ~p~du~bi~ty of the electrode response over a per&d of 4 days. Figure 2 shows a typical Tl(HF) calibration graph. The response is Nernstian (53 :‘s 1 mV/decade at lSo) over the concentration range 6.3 x lo-‘2.5 x IO-3M. The detection limit is 3.5 x IO-‘M.
590 Table 1. Reproducibility of electrode response Response, mV 1 x 10-4M Tl 1 x 10-5M Tl
1 x 10-3M Tl
Day 1 2 3 4
-271 -271 -271 -271
-321 -324 -323 -323
1 x 10-6iM Tl
-375 -378 -373 -374
-428 -430 -427 -429
Table 2. Potentiometric selectivity coefficients (separate solution method) Foreign ion (j)
KPo’ _,I Tq
3.9 x 1o-4
Na+ K+ Mg2+ Zn*+ Bi”+ Ga’+ In’+ Al3+
U’I ; ---H---
Foreign ion (j)
3.4 x IO-4 : 5.2 x 1.2 x 6.7 x 4.7 x
1O-6 10-S 1O-5 1o-6
Table 3. Determination Fig. 1. Diagram of coated carbon rod PVC membrane electrode. 1, Electrical connection; 2, electrode cap; 3, carbon rod; 4, PVC insulator; 5, PVC sensor.
4
P E 4r
-
-300
-
-350
-
-400
7
-450
-
1 4.5 x 1.2 x 6.4 x 2.6 x 9.5 x
10-d 10-6 IO-* 1O-2 10-d
*Severe interference. tNo interference.
0.8 cm
-250
As3+ AU’+ NO, PO:ClO; IBr-
Km’ _ TICI, .I
Sample*
of thallium
Recommended value, ppm
Found, PPm
RSD, %
GBW07302
1.90
1.92
1.0
GBW07304 GBW07305
1.16 1.08
1.05 1.07
9.5 1.0
GBW07306
1.20
1.20
0.0
*Chinese reference standard (formerly Chinese streamsediment reference standard). ution to a 50-ml standard flask, and dilute to volume with OSM sodium chloride/O.SM hydrochloric acid. Analyse for Tl(II1). Table 3 shows typical results. CONCLUSIONS I 2
I 3
I 4
I 5
I
I*
6
7
PCnc,;
Fig. 2. Typical calibration graph for ll(II1).
The potentiometric selectivity coefficients of the electrode for some interferent ions, measured by the separate solution method, are listed in Table 2. The interference of other ions is negligible except for Au(II1). Analysis of geological samples
Weigh 0.2 g of sample into a loo-ml beaker, and add 20 ml of freshly prepared aqua regia. Heat on a steam-bath until a moist residue remains, add 5 drops of concentrated hydrofluoric and perchloric acids, then heat on a hot-plate to fumes of perchloric acid. Heat with hydrochloric acid to drive off nitric acid, repeating the treatment twice, cool, transfer the sol-
The carbon-coated PVC membrane electrode is a new type of PVC membrane electrode with good characteristics. The electrode is easy to prepare, convenient to use and preserve, inexpensive and reliable, and has a long lifetime. There is good adherence of the PVC membrane to the carbon rod, and the rod can be used repeatedly. The technique is extremely economical. Acknowledgement-We thank Mr Xu Peifan and Mr Chang Huijun for experimental work. REFERENCES 1.P. J. McGoldrick, R. R. Keays and B. B. Scott, Geochim. Cosmochim. Acta, 1979, 43, 1303. 2. J. Hertogen, M. J. Janssen and H. Palme, ibid., 1980,44, 2125. 3. Dan Dezhong, Xu Peifang and Liu Lun, Fenxi Huaxue, 1987, 8, 706. 4. J. L. F. C. Lima and A. A. S. C. Machado, Analyst, 1986, 111,799. 5. A. G. Fogg, A. A. Al-Sibaai and N. Burgess, Anal. L&t., 1975, 8, 129.
(Received 22 July 1987. Revised 28 January 1988. Accepted 5 February 1988)
Smmn~ary-A new Tl(III) ion-selective electrode prepared by coating a graphite rod with a PVC membrane has been made and investigated. The optimum membrane ~~~~tiou is 3~~lrh~a~ne E-‘I’lw 5 mg, ~~~~~~~ 0.1 zrx&dketyf ph~alate 8.4 n& PVC O.17 g. The &&rode exhibits I+.znstian response to n(m) ovee tile concentration range 6.5 X I&%2.5 x IO-%. The detection limit is 3.5 x IO-?&f and the slope of the electrode response is 53 rt 1 mV per decade. The electrode shows high stabiiity and selectivity. The electrode is easy to make and store, and inexpensive. Thallium, which shows crystallochemical affinities with the ~~he~i~~~ ~~~o~a~~ e-lements ~tasai~~ and ~~~~~~~ is an &rnerXt of ~I~~~ interest. 1% has been proposed as a geochernicai indicator in problems such as petrogenesis, rock-sea-water interaction, and evaluation of the degree of sulphur saturation in a silicate melt.‘*2 The abundance of thallium in geological material is very low so its dete~na~o~ requires highly sensitive analytical methode, such as fluorimetry, neutron activation, atomic-absorption spectrometry (AAS) and anodic stripping voltammetry (ASV). Fluorimetry has severe limitations due to interference, neutron activation requires expensive facilities and a long time for ~~a~a~o~ and coo&g of samples,arrdAAS and ASV often require enrichment methods. We have already developed a PVC membrane electrode based on the Butylrhodamine B-AuCl; ion-pair as electraactive agent,3 and now propose Eu~lrh~~i~e B-TlC& as an eleetroactive agent for use in a PVC coating on a carbon rod efeetrcrde. This electrode, which needs nu internal reference solution and internal reference electrode, is easier and cheaper to make than conventional PVC membranes or all solid-state PVC membranes.4 The electrode has been used successfully for the dete~nation of Tl(II1) in ~olo~~~ reference samples. EXPERIMENTAL
&tgeFm Analytical-reagent grade chemicals were used. Tl(III) stock solution,5 1 x 10-2M, dituted as required, w&b MM sodium chloride and 0.5&f hydr~lo~ acid. Carbon rod, ~t~~o~y pure, 8 mm d&meter. Preparation of efectmde
Weigh 0.2 g of Butylrhodamine B and dissolve it in 5 ml of demineralized water, then add 1 ml of concentrated hydrochloric acid, 0.7 g of sodium chloride and IO ml of
lO%f Tl(II1) stock solution dropwise. Centrifuge and wash the red precipitate with two 54 portions of 20% hydro&lo&z acid. Dry the prod& at So”. The ODeium i%kei?lbmne6xxllwition Ras been four& to be 5 mg of Butylrhodarnine BlTlq, 0.1 ml of &Iorobenzene. 0,4 ml of dioctvl ohthalate and 0. I7 Kof PVC. Mix these reagents in the ~160s stated, add 10 ml Gf tetrahydrofuran, stir, and let stand for 10 hr. Coat the carbon rod by inunersing it in the aoIution, removing it and allowing it to dry” Repeat the coating%taking care to get a ~~-d~s~but~ meznbrane. Leave the rueznbraoe in the air for about 12 hr. A thin membrane has poor stability and a short lifetime, but the response is slow if the membrane is too thick. A faur-layer membrane is a suitable compromise. Figure 1 shows the structure of the coated carbon rod PVC membrane electrode. Before use, the electrode should be activated For 2 hr by min&alized water until the potential &a&e to an SCE is stable. The cell is carbon rod 11 PVC sensorlunknown soln. 11 SCE
Tl(II1) hydrolyses and precipitates at pH >5, A reasonable pH range for use is O-5. The membrane resistance measured by the parallel resistance method is 0.1 MB The electrode response is immediate if the ~u~tmtiou of Tl@f) is >10-% The response time is about 30 set in ~~-~-~~-~~ Tt(rrI). The
potential reading taken every 30 min aver a period of 3 hr had a drift of only rfr1 mV. Potential readings taken five times in lo-‘M and 10e6M Tl(II1) alternately, showed no memory effect. Table 1 shows the ~p~du~bi~ty of the electrode response over a per&d of 4 days. Figure 2 shows a typical Tl(HF) calibration graph. The response is Nernstian (53 :‘s 1 mV/decade at lSo) over the concentration range 6.3 x lo-‘2.5 x IO-3M. The detection limit is 3.5 x IO-‘M.
590 Table 1. Reproducibility of electrode response Response, mV 1 x 10-4M Tl 1 x 10-5M Tl
1 x 10-3M Tl
Day 1 2 3 4
-271 -271 -271 -271
-321 -324 -323 -323
1 x 10-6iM Tl
-375 -378 -373 -374
-428 -430 -427 -429
Table 2. Potentiometric selectivity coefficients (separate solution method) Foreign ion (j)
KPo’ _,I Tq
3.9 x 1o-4
Na+ K+ Mg2+ Zn*+ Bi”+ Ga’+ In’+ Al3+
U’I ; ---H---
Foreign ion (j)
3.4 x IO-4 : 5.2 x 1.2 x 6.7 x 4.7 x
1O-6 10-S 1O-5 1o-6
Table 3. Determination Fig. 1. Diagram of coated carbon rod PVC membrane electrode. 1, Electrical connection; 2, electrode cap; 3, carbon rod; 4, PVC insulator; 5, PVC sensor.
4
P E 4r
-
-300
-
-350
-
-400
7
-450
-
1 4.5 x 1.2 x 6.4 x 2.6 x 9.5 x
10-d 10-6 IO-* 1O-2 10-d
*Severe interference. tNo interference.
0.8 cm
-250
As3+ AU’+ NO, PO:ClO; IBr-
Km’ _ TICI, .I
Sample*
of thallium
Recommended value, ppm
Found, PPm
RSD, %
GBW07302
1.90
1.92
1.0
GBW07304 GBW07305
1.16 1.08
1.05 1.07
9.5 1.0
GBW07306
1.20
1.20
0.0
*Chinese reference standard (formerly Chinese streamsediment reference standard). ution to a 50-ml standard flask, and dilute to volume with OSM sodium chloride/O.SM hydrochloric acid. Analyse for Tl(II1). Table 3 shows typical results. CONCLUSIONS I 2
I 3
I 4
I 5
I
I*
6
7
PCnc,;
Fig. 2. Typical calibration graph for ll(II1).
The potentiometric selectivity coefficients of the electrode for some interferent ions, measured by the separate solution method, are listed in Table 2. The interference of other ions is negligible except for Au(II1). Analysis of geological samples
Weigh 0.2 g of sample into a loo-ml beaker, and add 20 ml of freshly prepared aqua regia. Heat on a steam-bath until a moist residue remains, add 5 drops of concentrated hydrofluoric and perchloric acids, then heat on a hot-plate to fumes of perchloric acid. Heat with hydrochloric acid to drive off nitric acid, repeating the treatment twice, cool, transfer the sol-
The carbon-coated PVC membrane electrode is a new type of PVC membrane electrode with good characteristics. The electrode is easy to prepare, convenient to use and preserve, inexpensive and reliable, and has a long lifetime. There is good adherence of the PVC membrane to the carbon rod, and the rod can be used repeatedly. The technique is extremely economical. Acknowledgement-We thank Mr Xu Peifan and Mr Chang Huijun for experimental work. REFERENCES 1.P. J. McGoldrick, R. R. Keays and B. B. Scott, Geochim. Cosmochim. Acta, 1979, 43, 1303. 2. J. Hertogen, M. J. Janssen and H. Palme, ibid., 1980,44, 2125. 3. Dan Dezhong, Xu Peifang and Liu Lun, Fenxi Huaxue, 1987, 8, 706. 4. J. L. F. C. Lima and A. A. S. C. Machado, Analyst, 1986, 111,799. 5. A. G. Fogg, A. A. Al-Sibaai and N. Burgess, Anal. L&t., 1975, 8, 129.