Novel sensing device based on potentiometric measurement for Lithium detection

S. Merzouk*,**, N. Zine *, M. Taha Janan ***, M. Agouzoul**, J. Bausells****, F. Teixidor*****,

N. Jaffrezic-Renault*, A. Errachid*

* Université Claude Bernard Lyon 1, Institut des Sciences Analytiques, Département LSA, Equipe SIMS ,

5 rue de la Doua ,69100 ,Villeurbanne ,France, safae_merzouk@yahoo.fr, nadia.zine@univ-lyon1.fr

Nicole.Jaffrezic@univ-lyon1.fr, abdelhamid.errachid@adm.univ-lyon1.fr **

UM5A, Ecole Mohammadia d’ingénieurs, ERD3M, BP 765, Rabat, Maroc, m.agouzoul@gmail.com ***

UM5S, Ecole Normale Supérieure de l'Enseignement Technique, 10100,Rabat,Maroc,

m.tahajanan@um5s.net.ma ****

Centre Nacional de Microelectrònica (IMB-CSIC), Campus U.A.B., 08193 Bellaterra, Spain,

joan.bausells@imb-cnm.csic.es ****

Institut de Ciència de Materials de Barcelona (CSIC), Campus UAB, 08193 Bellaterra, Spain,

teixidor@icmab.es

ABSTRACT

The main objective of this work is the determination of

lithium by Solid contact ion-selective microelectrodes (SC-

uISE) based on silicon technology. A film of polypyrrole

doped with cobaltabis(dicarbollide) anion [3,3’-Co(1,2-

C2B9H11)2] was deposited on gold microelectrodes by

electrochemical polymerization. The PPy[Co(C2B9H11)2]

was subsequently coated with poly(vinyl chloride) matrix

membrane containing Lithium ionophore III (ETH 1810).

The developed potentiometric SC-µISE provides

excellent performance towards the determination of lithium

with a high sensitivity of 52 ± 2 mV/decade, a low

detection limit of 1x10-6

M and a wide linear range from

4.10-5

to 1.10-1

M covering the clinically interesting Li+

range. The SC-µISE exhibits a good selectivities for lithium

over other cathions: logKpot

Li, Mg= -4.35, logKpot

Li, K= -4.09,

logKpot

Li, NH4= -3.39, logKpot

Li, Ca= -2.6 and logKpot

Li, Na= -

2.5. Furthermore, the developed SC-µISE will be a

promising chemical sensing device for the measurement of

Lithium in pharmaceutical preparations of Lithium

treatments.

Keywords: Ion-selective Microelectrode µISE , conducting

polymer, Potentiometry, lithium ion

1 INTRODUCTION

Ion-selective electrodes (ISEs) are very attractive for ion

activity measurements in biological and environmental

systems. The therapy of manic depressive psychosis with

lithium salts requires periodic measurements of the Li

concentration in whole blood, plasma, urine or serum [1-3].

The analysis is difficult because of the low lithium

concentration compared to the high content of the natural

cations (especially Na+) [4]. Lithium has a narrow

therapeutic range (0.5-1.5 mmol/l), and too low of a dosage

leads to ineffectiveness and too high leads to severe toxicity

[5]. Accurate and rapid monitoring of the Li activity in

blood for the patients in lithium therapy is critically

important as the gap between its therapeutic and toxic

levels are very close. Lithium-selective electrode has been

developed for such purpose, and several commercial

analyzers are now routinely used in many clinical

laboratories [6-13]. For this interest, we report a novel

sensing device based on potentiometric measurement for

the detection of Lithium. Lithium Ionophore III was used as

ionophore for membrane preparation. The selective

membrane was deposited onto gold microelectrodes

containing a conducting polymer (polypyrrole doped with

cobaltabis(dicarbollide) ions ([3,3′-Co(1,2-C2B9H11)2]−) as

solid contact layer.

2 EXPERIMENTS AND RESULTS

2.1 Reagents

Pyrrole (Aldrich Chemicals) was distilled under vacuum

prior to its use. All the membrane components (analytical

reagent grade), namely, Lithium ionophore III (ETH 1810)

(62558), Potassium tetrakis(4-chlorophenyl)borate (60591),

2-Nitrophenyl octyl ether (73732),Poly(vinyl chloride) high

molecular weight (81392) were purchased from

Fluka.Tris(hydroxymethyl)aminomethane (Tris) (252859)

was purchased Sigma-Aldrich.Standard solutions and

buffers were prepared with di-ionised water.

2.2 Electropolymerization of PPy [3,3′-

Co(1,2-C2B9H11)2]

The fabrication process for the microelectrodes has been

performed at the Centro Nacional de Microelectronica

(CNM) of Barcelona (see Fig.1) [9]. The

electropolymerization of poly(pyrrole) doped with

NSTI-Nanotech 2013, www.nsti.org, ISBN 978-1-4822-0586-2 Vol. 3, 2013 77

cobaltabis(dicarbollide) anion [3,3′-Co(1,2-C2B9H11)2]-

on

the surface of gold microelectrode has been made by

chrono potentiometry method using a potentiostat

galvanostat (voltalab PGZ401), in a single compartment

cell with a standard three-electrode system (the working

electrode is a gold microelectrode with size 300 m2, the

auxiliary platinum and the reference electrode is saturated

calomel) to a constant current of 100 µA during 20s [9].

The solution for the electropolymerization was 0.1 M

Pyrrole and 0.035 M Cs[3,3-Co(1,2-C2B9H11)2]- in

acetonitrile (MeCN) 1 wt.% in water .

Figure 1: image and encapculation process of the gold

microelctrode manifactured

The results of electrodeposition of PPy is presented in

the (see Fig. 2), which shows the gold microelectrode

before and after electropolymerization and the

chronopotentiometric response (see Fig.3).

Figure 2: Image step by step of the gold microelctrode

doped PPy [3,3′-Co(1,2-C2B9H11)2] in 0.1 M Pyrrole and

0.035 M Cs[3,3-Co(1,2-C2B9H11)2] in acetonitrile (MeCN)

1 wt.% in water during 20s at a constant current of 100 µA.

Figure 3: Chrono potentiometry response

2.3 Deposing of lithium membrane

The membrane was obtained by mixing: Lithium

ionophore III “ETH 1810”; polyvinyl chloride (PVC); o-

nitrophenyloctyl ether (NPOE) ; Potassium tetrakis(4-

chlorophenyl)borate in the following proportions: 1,2%

(w/w); 32,4 % (w/w); 66% (w/w) ; 0,6% (w/w),

respectively, diluted in 3 ml of tetrahydrofuran (THF). This

solution was then cast directly onto the PPy[Co(C2B9H11)2]

film. After complete evaporation of the solvent, a

transparent layer with a thickness of 100-150 micrometers

was obtained. Finally, before each use for measurements,

the Lithium-selective microelectrode was conditioned in 10-

3 mol/l of LiCl solution for 30 minutes, and stored in air

when not in use.

2.4 Potentiometric measurements

All measurements were carried out at room temperature,

using a homemade data adquisition set-up, eight multi-

channels microelectrodes connected to a personal computer.

LabVIEW5.0 software was used for data manipulation. The

potential was measured by immersing the microelectrode

and the reference electrode from Orion model 90-02

Ag/AgCl double junction in a beaker containing 25 ml of

Tris-HCl [c =0.05 M, pH = 7.2]. The calibration curves

were obtained by applying the addition method in which

known amounts of the measured ion (LiCl) were added to

the aqueous solution from 10-8

M to 10-1

M concentrations.

The activities of metal ions in the aqueous solutions were

calculated according to the Debye-Hückel equation. The

performance of all the microelectrodes was characterised by

the slope of the calibration graph, the detection limit and

the selectivity coefficients.

The Li+-SC-µISE exhibited linear responses to the

activity of lithium ions over the range 10-4

-10-1

M with a

slope of 52 ± 0.5 mV per decade and a correlation

coefficient of 0.9993 (Fig. 4). The microelectrodes based on

NSTI-Nanotech 2013, www.nsti.org, ISBN 978-1-4822-0586-2 Vol. 3, 201378

this configuration exhibited better performance to those

measured for PVC based on other types of ionophores [14].

The detection limit was evaluated as the abscissa of the

crossing point of extrapolation of the two linear parts of the

calibration graph, and the value was 5 x 10-4

M. we note

that the dynamic response of microelectrode potential

increases after each addition of the solution of various

concentrations of LiCl and stabilized there after (see Fig.

5), We deduce the efficiency of our μISE Lithium ion

selective. However, in order to confirm this sensitivity,

microelectrodes with membrane without ionphore were

tested. They have shown a slope and the detection limit

values that are clearly inferior to those obtained for

membranes containing Lithium ionophore III (ETH 1810).

Figure 4: Calibration plots of the Lithium microelectrodes

Figure 5: Dynamic response of Lithium microelectrode for

step changes in the concentration of Li+.

The behavior of the microelectrodes was also tested in

the presence of other cations (Mg2+

, K+, NH4

+, Ca

2+, and

Na+). The selectivity coefficients were determined by using

the fixed interference method (FIM) [15].

All the potentiometric interference measurements were

made in a 25 ml solution of interfering ion at a

concentration of 10-3

mol/l; we follow successively the

additions of the LiCl solution ranging from 1.10-8

to 1.10-1

mol/l (see Fig 6). Table 1 shows the results of KLi,j values

obtained for the different interfering ions. The KLi,j was

obtained using the equation (1). Where aA is the activity of

the primary ion and aB the activity of interfering ion and ZA

and ZB are their respective charges.

ZBZA)(

,

B

A

BA

a

aK (1)

Figure 6: Potential responses of the microelectrode for

different interfering ions

Table1: Potentiometric selectivity coefficients of Li+µISE

The cations as Mg2+

, NH4+ and K

+ do not cause any

disturbance to the performance of the developed

microelectrode. However, only Na+ and Ca

2+ produce small

interferences. The selectivity coefficients of the proposed

microelectrode based on ionophore III (ETH 1810) are

comparable with the corresponding values previously

NSTI-Nanotech 2013, www.nsti.org, ISBN 978-1-4822-0586-2 Vol. 3, 2013 79

reported for PVC- membrane Lithium-selective electrodes

based on different neutral ionophores [16].

The influence of pH on the potentiometric response of

the microelectrode was examined by following the potential

variation over a pH between 1,5 and 11 at a fixed

concentration of Li+ ion (1.10

-3 M). The pH of the solution

was adjusted by addition small volumes of chloridric acid

(1mol/l and 0.1 mol/l) to decrease the pH; the result shows

that the potential remains constant between pH 6,5 and 10

and the same may be taken as the working pH range (see

Fig 7). In addition, the microelectrode Li+ µISE can be used

for measurements in physiological fluid which has a neutral

pH ≈ 7.2, so the measures will not be influenced by such a

pH.

Figure 7: Effect of pH on the response of Li+ µISE

3 CONCLUSION

The experiment results founds in this work using SC-

ISE with PPy[Co(C2B9H11)2] as a layer of conducting

polymer between the Au microelectrode and selective

membrane for the detection of lithium, have shown a best

performance with respect to linear response, good

sensitivity covering the clinically interesting Li+ range (0.5–

1.5 mmol/l). The response was compared to different

interference cations for lithium detection which confirm

that, the analysis is difficult due to the low concentration of

lithium compared to the high content of cations (especially

Na+). Selectivity coefficients confirmed that sodium is the

largest interfering element compared to other species

studied containing in blood. Furthermore, the developed

µISE will be a promising chemical sensing device for the

measurement of Lithium in pharmaceutical preparations of

Lithium treatments and even for a practical control of

lithium concentrations to ensure conformity and to prevent

toxicity.

ACKNOWLEDGEMENTS

This work was funded by the European Communities

Seventh Framework Programme (FP7/2007-2013) under

the grant agreement No. 248763 (SensorART-IP) and

NATO SfP project under the grant No. CBP.NUKR.SFPP

984173 .

REFERENCES [1]. V.K. Gupta, S. Chandra, S. Agarwal , H. Lang, Sensors

and Actuators B. 107, 762-767, 2005

[2]. Marcos Fernando de S. Teixeira, Orlando Fatibello-

Filho , Luiz C. Ferracin, Romeu C. Rocha-Filho ,

Nerilso Bocchi , Sensors and Actuators B .67 ,96–100,

2000

[3]. M. Cretin, P. Fabry, Analytica Chimica Acta .354, 291-

299, 1997

[4]. K. Wilcox and G.E. Pacey, Analytics Chimica Actu.

245,235-242, 1991

[5]. M.I. Albero, J.A. Ortuno, M.S. García , M. Cuartero,

M.C. Alcaraz, Sensors and Actuators B .145, 133–138,

2010

[6]. Roger L Bertholf, M. Geraldine Savory, Kathy II.

Winbome, Judy C. Hundley, George M. Plummer and

John Savory, Clin. Chem. 3417, 1500-1502, 1988

[7]. Li-Xian Sun, Tatsuhiro Okada, Jean-Paul Collin,

Hideki Sugihara Analytica Chimica Acta. 329, 57-64,

1996

[8]. You Ra Kanga, Kyoung Moon Lee, Hakhyun Nam,

Geun Sig Ch, Sung Ouk Jung and Jae Sang Kim ,

Analyst. 122, 1445–1450, 1997

[9] N. Zine, J. Bausells, F. Teixidor, C. Viñas, C. Masalles,

J. Samitier, A. Errachid, Materials Science and

Engineering C .26, 399-404 , 2006

[10] O.A. Biloivan, S.V. Verevka , S.V. Dzyadevych, N.

Jaffrezic-Renault, N. Zine, J. Bausells, J. Samitier,

A.Errachid, Materials Science and Engineering C.

26,574 – 577, 2006

[11] N. Zine, J. Bausells, F. Vocanson , R. Lamartine , Z.

Asfari , F. Teixidor, E. Crespo, I.A. Marques de

Oliveira, J. Samitier, A. Errachid, Electrochimica

Acta.51, 5075–5079, 2006

[12] I.A. Marques de Oliveira, M. Pla-Roca, Ll. Escriche, J.

Casabó, N. Zine, J. Bausells, F. Teixidor, E. Crespo, A.

Errachid, J. Samitier, Electrochimica Acta, 51, 5070-

5074, 2006

[13] N. Zine, J. Bausells, A. Ivorra, J. Aguiló, M. Zabala, F.

Teixidor, C. Masalles, C. Viñas, A. Errachid, Sensenrs

and Actuators B, Chem. 91,76-, 2003

[14] I.A. Marques de Oliveira, D. Risco, F. Vocanson, E.

Crespo, F. Teixidor, N. Zine, J. Bausells, J. Samitier,

A. Errachid, Sensors and Actuators B, 130, 295-299,

2008

[15] V. Rumenjak, S. Milardovic, I. Kruhak and B. S.

Grabaric, Clin. Chim. Acta 335, 75-81, 2003

[16] Yoshio Umezawa, Kayoko Umezawa and Hithoshi

Sato, Pure & Appl. Chern. 67, 507-518, 1995

[17] E. Metzger, R. Dohner, W. Simon, D. J. Vonderschmit

and K. Gaustschi, Anal. Chem, 59, 1600-1603, 1987

NSTI-Nanotech 2013, www.nsti.org, ISBN 978-1-4822-0586-2 Vol. 3, 201380