Talanta 80 (2010) 1129–1132

Contents lists available at ScienceDirect

Talanta
journal homepage: www.elsevier.com/locate/talanta

Preparation of a new solid state fluoride ion selective electrode and application
Güler Somer ∗ , Şükrü Kalaycı, İbrahim Başak
Gazi University, Department of Chemistry, 06500 Ankara, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: A new solid state fluoride ion selective electrode composed of 70% Ag2 S, 10% Cu2 S and 20% CaF2 has been
Received 29 June 2009 developed. An analytically useful potential change occurred, from 1 × 10−6 to 1 × 10−1 M fluoride ion. The
Received in revised form 21 August 2009 slope of the linear portion (1 × 10−1 –1 × 10−5 M) was about 26 ± 2 mV/10-fold concentration changes in
Accepted 27 August 2009
fluoride. It was found that pH change between 1 and 8 had no effect on the potential of the electrode.
Available online 2 September 2009
There was no interference of most common cations such as K+, Na+ , Ca2+ and Mg2+ and anions such as
Cl− , NO3 − , SO4 2− and PO4 3− . The lifetime of the electrode was more than 2 years, when used at least 4–5
Keywords:
times a day, and the response time was about 60 s.
Fluoride electrode
Determination of fluoride
The measurements were made at constant ionic strength (0.1 M NaNO3 ) and at room temperature.
Solid state This electrode has been used for the determination of fluoride ion in Ankara city tap water and in bottled
Fluoride in water spring water using standard addition method. The validation of the electrode has been made with a
commercial fluoride ion selective electrode (Orion) and high consistency was obtained.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction
The first investigations relating to the ion selective electrodes
were done at the beginning of 1900. Increasing number of investi-
gations about ion selective electrodes are being continuing as they
have certain advantages such as they have good selectivity and
give specific and sensitive results in short time. These electrodes
are being used in clinical analysis and in routine control analysis
of some ions and some biological species, because of their ease in
use and selectivity characters [1,2]. They can be prepared rapidly,
they can simply be produced with a very low cost and above all
they allow us to analyze the solution without pre-treatments. Silver
salts of spare solubility are used mostly for their preparation. It was
found that solid state electrodes prepared from silver halogenides
were sensitive to halogenides, however these electrodes could also
be used for the determination of some cations such as mercury
[3–5] indirectly. Pungor and Hollos [6] could determine iodide ion
in the presence of chloride ion using an electrode prepared from
silver iodide and paraffin. It was found that the electrodes made
from two sulfide salts such as Ag2 S–PbS and Ag2 S–CdS were sen-
sitive to lead and cadmium ions respectively [7]. A solid state ion
selective electrode was prepared using Ag2 Se and Cu2 S which was
selective to selenite [8].
Some of the ion selective electrodes are in solid state such as
lead and cadmium [7], selenite [8], and hydroxide electrodes [9],
some of them are prepared using PVC or silicon rubber membrane
∗ Corresponding author. Tel.: +90 312 202 1150; fax: +90 312 2122279.
E-mail address: gsomer@gazi.edu.tr (G. Somer).
such as halide and sulfide [10], selenite [11], ferric ion [12], iodide
[5] and glucose electrodes [1].
Small amounts of fluorine are vital for the human body, although
larger quantities are toxic.
For two reasons the production of fluoride ion selective elec-
trode is necessary.
(1) Fluoride is important in commerce and technology.
(2) Determining fluoride by previous techniques is difficult.
Some feeling for importance of a rapid and reliable method for
monitoring fluoride can be gained by considering a list of mate-
rials in which fluoride occurs: (a) consumer products (toothpaste),
(b) biological materials (present in tooth enamel, and bone), (c)
pharmaceuticals (e.g. vitamins), (d) public water supplies (fluo-
ridation), (e) water and air (pollution by discharge of industrial
wastes), (f) electroplating and pickling baths (continuous monitor-
ing for fluoride level regulation) and (g) manufactured materials
(e.g. fertilizers, pesticides, glass, etc.).
For fluoride ion detection, LaF3 membrane electrode [13] has
been developed which is probably the most widely applied fluo-
ride ion selective electrode for practical measurements. Thorium,
calcium and the rare earth fluorides were also used for study as
being suitably insoluble [14]. However, detailed investigation of
precipitation procedures was necessary since normal precipitates
of these fluorides tend to be gelatinous. In a recent study [15] a flu-
oride selective electrode with LaF3 membrane and stainless steel
internal contact has been prepared.
From the literature survey it maybe seen that there are very few
electrodes for the determination of fluoride. The aim of this work

0039-9140/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.talanta.2009.08.037
1130 G. Somer et al. / Talanta 80 (2010) 1129–1132

was to fulfill this gap and to prepare an electrode selective for flu-
oride ion which is easy to prepare, easy to handle and has long
lifetime. The present work describes the preparation and applica-
tion of a new potentiometric fluoride selective electrode, which
fulfills the above given futures.

2. Materials and methods

2.1. Apparatus and reagents

A JENWAY 3040 Ion Analyzer has been used for potential

measurements and 924036 double junction Ag/AgCl reference elec-
trode was used as the outer reference electrode. Chiltern H531
magnetic stirrer was used.
Reagent grade chemicals were used without further purifica-
tion. Stock solutions were prepared with doubly distilled water and
kept in dark. In order to keep the ionic strength constant, 0.1 M
NaNO3 was used. Measurements were made at room temperature
22 ± 2 ◦ C and under nitrogen atmosphere.
2.2. Preparation of solid state crystal electrodes
For this purpose Ag2 S, Cu2 S and CaF2 were mixed in various
ratios so that the total weight was changing between 0.1 and 0.3 g.
Cu2 S was prepared by the addition of Na2 S to CuCl solution.
To protect CuCl from air oxidation, this reaction was made under
nitrogen atmosphere. About 2 g of CuCl were dissolved in 100 mL
de-aerated water. By slowly addition of 0.1 M Na2 S and passing
nitrogen, Cu2 S was precipitated. It was left waiting for about 30 min
at 60 ◦ C. Then it was filtered and washed until no precipitate for-
mation was observed with silver nitrate. It was dried in an oven at
about 110 ◦ C and left waiting in a desiccator.
Ag2 S was prepared as given above for Cu2 S, except there was no
need to protect from air oxidation but Ag2 S had to be kept in dark.
For the preparation of CaF2 , 4 g of Ca(OH)2 was dissolved in
100 mL water with 1–2 drops of HCl and then 0.1 M NaF was added.
CaF2 precipitate was filtered and dried.
A certain amount of the salt or salt mixture of Ag2 S, Cu2 S and
CaF2 (10–15 mg) was taken and pellets were made by holding first
under a pressure of 1940 atm for 1 min, then pressure was increased
to 7760 atm by adding each time 1940 atm for 1 min. The pellets of
7 mm diameter were sealed with epoxy resin (0.7 g epoxy and 0.9 g
hardener). To obtain a good sealing the epoxy resin on the edge
of the tubing had to wait for about 10 min and then the pellet was
sealed. One day after the pellet was sealed a silver wire is connected.
For this purpose the inside of the glass tube was filled to about
1 cm with a mixture of 0.5 g graphite powder and epoxy resin. The
electrode prepared had to wait for about 2 days so that the resin can
get dry. The surface of the electrode is washed and then polished
with a soft paper. These electrodes should be kept in air and in a
dark place when they are not in use. The lifetime of these electrodes
was more than 2 years, when used at least 4–5 times a day.
Fig. 1. Calibration curves for the new fluoride selective electrode (70% Ag2 S, 10%
Cu2 S, 20% CaF2 ) and for the commercial Orion electrode.
To obtain the calibration graph the potential readings were plot-
ted against −log of fluoride ion concentration. For each composition
four electrodes were prepared and their responses were measured.
The Ag2 S percentage varied from 60% to 80% and CaF2 from 5% to
30%.
The average slope of the curve in 1 × 10−4 –1 × 0.1 M concentra-
tion range was 5.6 mV (R2 = 0.9795) for the electrode having 80%
Ag2 S, 10% Cu2 S and 10% CaF2 composition. The slope became very
small at fluoride concentrations lower than 10−4 M. For the elec-
trode having 75% Ag2 S, 25% Cu2 S and 5% CaF2 composition the
slope was about 12 mV for 1 × 10−4 –0.1 M concentration range. But
no response was observed at concentrations lower than 10−4 M.
The highest slope was obtained for the electrode having 70% Ag2 S,
10% Cu2 S and 20% CaF2 composition. The slope was 26 ± 2 mV
(R2 = 0.9923) in 1 × 10−5 –1 × 0.1 M fluoride concentration range.
The calibration curve for this electrode is shown in Fig. 1, together
with the calibration curve obtained for the commercial Orion elec-
trode. For the electrode having 60% Ag2 S, 20% Cu2 S and 20% CaF2
composition, the slope was about 12 mV for 1 × 10−3 –0.1 M con-
centration range. The electrode with 30% CaF2 could not be used
since it was easily breakable.
The calibration curves for electrodes in different compositions
are shown in Fig. 2. As can be seen the highest slope was obtained
with the electrode in 70% Ag2 S, 10% Cu2 S and 20% CaF2 composition.
3.2. The effect of pressure on the electrode sensitivity
The pressure applied on the electrode during preparation was
changed from 4850, 7760, 9700 atm. The electrode with highest

3. Results and discussion

3.1. Effect of electrode composition

Membrane composition has a great effect over the sensitivity,

selectivity and stability of the ion selective membrane electrodes.
In this study, electrodes in four different compositions were pre-
pared to identify its effect. The measurements were made in 20 mL
0.1 M NaNO3 solution, in order to keep the ionic strength of medium
constant. The response of each electrode was measured after the
additions of 0.2 mL NaF solutions so that fluoride ion concentrations
changed from 0.1 to 1 × 10−6 M. Fig. 2. Calibration curves for fluoride electrodes in different compositions.
 G. Somer et al. / Talanta 80 (2010) 1129–1132
Fig. 3. pH dependence of (70% Ag2 S, 10% Cu2 S, 20% CaF2 ) fluoride electrode.
slope was used for all pressures which had the composition of 70%
Ag2 S, 10% Cu2 S and 20% CaF2 . It was found that the highest slope
was obtained when 7760 atm was applied.
3.3. Effect of pH
Effect of pH on the potential readings at various fluoride concen-
trations, were investigated. For this purpose to 20 mL 0.1 M NaNO3
solutions 0.02 mL aliquots of 1 M HCl was added and the potentials
after each addition was measured. The same procedure was applied
for the NaOH additions. The electrode with the highest slope (70%
Ag2 S) was used during the measurements. The change in potential
with pH was only 1–2 mV for each pH unit change from pH 1 to 8. At
pH values larger than 8, the change in potential with pH was about
15 mV for each pH unit. As can be seen from Fig. 3, this electrode
can be used between pH 1–8 safely.
3.4. The response time of the electrode
The response time of the electrode depends to the concentration
change. The electrode with the highest slope was used throughout
the measurements. The potential values were measured for each
concentration change after 30, 60, 90, 120 and 150 s. When fluo-
ride concentration was changed from 1 × 10−5 to 1 × 10−4 , 1 × 10−4
to 1 × 10−3 , 1 × 10−3 to 1 × 10−2 , and 1 × 10−2 to 1 × 10−1 M, the
response of the electrode was about 60 s, which is a short period
and it is an advantage. For this reason, this electrode can easily be
used for routine analyses.
The responses of electrode obtained for various concentration
changes are shown in Fig. 4. All these measurements were made at
constant temperature.
Fig. 4. The dependence of response time on the change of concentration.
1131
Fig. 5. The response of electrode against some anions in the presence of 10−3 M
fluoride (70% Ag2 S, 10% Cu2 S, 20% CaF2 ).
3.5. Conditioning of the electrode and lifetime
In order to identify the conditioning medium, electrodes having
the same composition were kept in air (dark) and in distilled water
and in 1 × 10−2 M NaNO3 solution under same duration of time and
their responses to fluoride ion were measured. Their performances
were found out to be close to each other. In further studies the
electrodes were kept in air (in dark) while not in use. The lifetime
of this electrode was more than 2 years when it was used at least
4–5 times per day.
3.6. Interference effects of various cations and anions
It usually is not possible to obtain an ISE which is only spe-
cific for one ion. Thus, in order to achieve correct and sensitive
results in analysis it is important to know the effects of all other
ions in the medium. These effects are identified through selectiv-
ity coefficient. In this study mixed solutions method was used to
pot
determine selectivity coefficient, kA,B [16], as it usually corresponds
more closely to the situation in samples. Selectivity coefficient val-
ues were calculated via making use of the below given equation:
kA,B aB nA /nB = aA {antilog[(E1 − E2 )/S]} − aA
pot
Here, S = 2.303RT/nA F (the slope of fluoride electrode), aA is the
activity of the primary ion, aB the activity of the interfering ion,
E1 the potential measured when only A is present, E2 the potential
responsive to the primary ion in the presence of interfering ion,
kA,B the selectivity coefficient and nA , nB are the charges of A and
B. In this method, solutions were prepared with a constant activ-
ity of the main ion (fluoride) and varying activity of interfering ion.
For this purpose, interfering ion was added to 10−3 M fluoride solu-
tion (in 0.1 M NaNO3 ) so that its concentration was 10−2 –10−5 M
and after each addition, the change of potential was measured. The
potentials obtained at different concentrations of interfering ions
are plotted against concentrations as shown in Fig. 5 for anions and
in Fig. 6 for cations. For comparison the curve obtained for fluoride
electrode is also given in the same figure.
As can be seen nearly no interference was observed for chlo-
ride, nitrate, sulfate and phosphate ions. While no interference was
observed for sodium, potassium, calcium and magnesium, slight
interference was observed for silver ions.
For the determination of the selectivity coefficients, the aB nA /nB
values are plotted against the right hand side of the above given
equation and the slope will be equal to the selectivity coefficient of
1132 G. Somer et al. / Talanta 80 (2010) 1129–1132

Table 2
Determination of fluoride in spring and tap water (N = 4).

Newly established fluoride Orion fluoride

electrode (mg/L) electrode (mg/L)

Spring water 0.25 ± 0.01 0.29 ± 0.01

Tap water 2.87 ± 0.03 3.21 ± 0.02

tions. Same procedure was applied four times and the result found
was 0.25 ± 0.01 mg/L. The same water sample was analyzed with
the commercial Orion fluoride electrodes and the result found was
0.29 ± 0.01 mg/L. The fluoride content in bottled drinking water
was given (on tag) as 0.20 mg/L. As can be seen there is about 12%
difference in results between these two electrodes. The slope of
Orion electrode is smaller at low concentrations, but then suddenly
Fig. 6. The response of electrode against some cations, in the presence of 10−3 M increases at higher concentrations Fig. 1. This large change in slope
fluoride (70% Ag2 S, 10% Cu2 S, 20% CaF2 ).
may be the cause of this difference.
The same procedure was applied to tap water of Ankara city. The
Table 1 fluoride content found with the new electrode was 2.87 ± 0.03 mg/L
pot
Selectivity coefficients (kA,B ) for the fluoride electrode in mixed solutions (in the (N = 4), with Orion electrode it was 3.21 ± 0.02 mg/L. The results for
presence of 10−3 M fluoride) (B: interfering ion; A: fluoride ion). two different kinds of water are given in Table 2.
pot
Ions kA,B
4. Conclusions
Cl− 3 × 10−3
NO3 − 8 × 10−3
SO4 2− 6 × 10−4 This electrode enables the determination of fluoride ion in
PO4 3− 3 × 10−4 1 × 10−6 –1 × 10−1 M concentration range in the presence of most
K+ 3 × 10−3 common ions. The slope of the linear portion (1 × 10−1 –1 × 10−5 M)
Na+ 8 × 10−3
Ag+ 1 × 10−2
was 26 ± 2 mV per decade change in fluoride ion. The present flu-
Ca2+ 1 × 10−4 oride ion selective electrode displayed very good selectivity for
Cu2+ 2 × 10−4 fluoride ion with respect to sodium, potassium, calcium, magne-
Mg2+ 5 × 10−3 sium, copper, chloride, nitrate, sulfate and phosphate ions. It was
possible to determine fluoride ion quantity in spring and tap water.
The preparation of the electrode is simple, it displays very good per-
the interfering ion. The selectivity coefficients obtained for men-
formance in regard to reproducibility, sensitivity and long lifetime.
tioned anions and cations are given in Table 1. As can be seen
the present fluoride ion selective electrode displayed very good
selectivity for fluoride ion with respect to sodium, potassium, cal- References
cium and magnesium cations and to chloride, nitrate, sulfate and
[1] Ş. Kalaycı, G. Somer, G. Ekmekci, Talanta 65 (2005) 87–91.
phosphate anions. [2] M. Mascini, G.G. Guilbault, Biosensors 2 (1986) 147.
[3] R.F. Overman, Anal. Chem. 43 (1971) 616–617.
3.7. Determination of fluoride in spring and tap water [4] G. Somer, Anal. Chem. 53 (1981) 2143–2145.
[5] G. Somer, Ş. Kalaycı, G. Ekmekci, Sens. Actuators B 81 (2001) 122–127.
[6] E. Pungor, R. Hollos, Acta Chim. Acad. Sci. Hung. 27 (1961) 63.
The aim of the sample study was to test the accuracy of the [7] G.A. Rechnitz, Z.F. Lin, S.B. Zamochnick, Anal. Lett. 1 (1967) 29.
results achieved by the analyses done with the fluoride sensitive [8] G. Ekmekci, G. Somer, Talanta 49 (1999) 91–98.
[9] G. Ekmekci, Ş. Kalaycı, G. Somer, Sens. Actuators B 101 (2004) 260–264.
electrode prepared in our laboratory. A sample of spring water [10] E. Pungor, Anal. Chem. 39 (1967) 28A.
(bottled) was analyzed with the newly established electrode. For [11] G. Ekmekçi, G. Somer, Talanta 49 (1999) 83–89.
this purpose first the potential of 20 mL 0.1 M NaNO3 is measured [12] G. Ekmekci, D. Uzun, G. Somer, S. Kalaycı, J. Membr. Sci. 288 (2007) 36–40.
[13] M.S. Frant, J.W. Ross, Science 154 (1966) 1553.
using fluoride ion selective electrode. Then 0.2 mL spring water [14] A.M.U. Macdonald, K. Toth, Anal. Chim. Acta 41 (1968) 99–106.
sample was added and once more potential was measured. The [15] M. Bralic, N. Radic, S. Brinic, E. Generalic, Talanta 55 (2001) 581–586.
fluoride content of this water was determined by standard addi- [16] K. Srinivasan, G.A. Rechnitz, Anal. Chem. 41 (1969) 1203–1208.