Journal of Colloid and Interface Science 240, 97104 (2001)
doi:10.1006/jcis.2001.7649, available online at http://www.idealibrary.com on
Quaternary Ammonium Salts Immobilized on Silica Gel: Exchange
Properties and Application as Potentiometric Sensor for Perchlorate Ions
Elvio A. de Campos, Antonio A. da Silva Alfaya, Rosilene T. Ferrari, and Creusa Maieru M. Costa1
Departamento de Qumica, Universidade Estadual de Maringa, Av. Colombo 5790, 87020-900, Maringa-PR, Brazil
Received October 23, 2000; accepted April 21, 2001
Ammonium chlorides immobilized on silica gel, SA+ /Cl
and SE+ /Cl , were obtained from silica previously modified
with 3-aminopropyltriethoxysilane and N-[3-(trimethoxysilyl)-
propyl]ethylenediamine, respectively. Both materials showed
potential use as an anion exchanger: they are thermically stable
(up to 413 K), achieve equilibrium rapidly in the presence of suit-
able exchanger ions, and are easily recovered. The exchange ca-
pacities observed for SA+ /Cl and SE+ /Cl are 0.70 and 1.19 mmol
Cl g1 , respectively. Through the exchange isotherms and compet-
itive Cl X exchange (X = F , Br , I , N
3 , NO3 , SCN , ClO4 )
it was observed that: (i) SE+ /Cl exchanges the counterion Cl more
easily than SA+ /Cl ; (ii) SA+ /Cl presents higher selectivity than
SE+ /Cl ; and (iii) SA+ /Cl presents high affinity for ClO
4 . Because
of these two latter properties presented by the SA+ /Cl , the deriva-
tive SA+ /ClO 4 was used as a potentiometric sensor for this anion,
prepared from the supported material on a mixture of graphite pow-
der with epoxy resin. The electrode showed a nernstian behavior
and a limit of response of 0.13 mmol L1 . Potentiometric selectiv-
pot
ity coefficients, K A,B , were obtained for some interfering anions,
and the following interference order was observed: F > SCN >
NO
3 > Br > Cl > CH3 COO . The electrode showed fast and
stable responses and was useful for approximately 200 measures.
C 2001 Academic Press
Key Words: quaternary ammonium salts; silica gel; anion ex-
change; potentiometric electrode; perchlorate ion.
INTRODUCTION
Silica gels have been chemically modified with suitable
orgafunctional groups, with the purpose of application in the
adsorption of metal ions from solutions (14), in preconcen-
tration processes and further analysis of metal ions at trace
levels in aqueous and nonaqueous solutions (5, 6), and as
materials to pack columns for use in high-pressure liquid
chromatography (7).
The application of these materials in chromatography includes
the adsorption, preconcentration, and ion-exchange techniques
(811); in the latter, charged groups are introduced onto the
silica surface by chemisorption. Thus, cation or anion exchang-
1 To whom correspondence should be addressed. Fax: 55 44 263 57 84.
E-mail: cmmcosta@uem.br.
97
ers are obtained depending on the immobilized functional group
charge. The application of these materials in ion exchange is de-
pendent on their exchange capacity, stability, and selectivity.
The exchange capacity is governed by the concentration of sur-
face hydroxyl groups before the modification, by extension in
which the reactions occur between the silanol groups and the
modifier, and by the nature of the kind of modifier (12). An ion
exchanger anchored onto silica gel is stable in solutions over
the pH range 08 and shows much more rigidity than the or-
ganic matrices and swelling negligible (13, 14). The selectivity
is usually expressed through the separation factor (derivative of
the isotherms) and through the selectivity coefficient (retention
time), although some authors prefer to use the thermodynamics
equilibrium constants (K Y ) for the exchange process (15, 16).
The anion exchangers immobilized onto silica gel containing
primary, secondary, or terciary aminic groups show a behavior
that is weakly acidic, while the quaternary salts are strongly
acidic. However, these properties change according to hy-
drophilicities of the modifier functional groups (8). In this work
we report the synthesis and the anion-exchange properties of
two systems of different hydrophilicities containing ammonium
chloride derivatives of 3-aminopropyltriethoxysilane and N -
[3-(trimethoxysil)propyl]ethylenediamine supported onto silica
gel, SA+ /Cl and SE+ /Cl , respectively. The high affinity to
perchlorate ions presented by the SA+ /Cl through the equi-
librium studies of anion exchange led us to its application
as an ion-selective electrode for the perchlorate anion. As is
known, the determination of the perchlorate ion in the presence
of other anions and also in complexes, propellants, and explo-
sives analysis is very difficult and tedious (1719). Few analyt-
ical methods are available and most of them are indirect; e.g.,
the perchlorate anion is determined after it has been reduced
to chloride ion (19). Therefore, these methods are susceptible
to halides interference (19, 20). Thus, studies have been pro-
posed on the basis of the construction of perchlorate selective
electrodes.
Some perchlorate-selective liquid membrane electrodes have
been developed (2128), and more recently, electrodes with solid
membranes applied on epoxy supports have also been used (29
31). This second type of electrode is more advantageous than
liquid membrane electrodes due to its easy construction and high
chemical stability.
0021-9797/01 $35.00
Copyright C 2001 by Academic Press
All rights of reproduction in any form reserved.
98 DE CAMPOS ET AL.
Thus, besides the exchange properties of SA+ /Cl and
SE+ /Cl , we report in this work the construction and the study
of the performance of an ion-selective electrode to analyze the
perchlorate anion by using the derivative SA+ /ClO 4 supported
onto a mixture of graphite powder with epoxy resin, without an
inner reference solution.
EXPERIMENTAL
Reagents
All of the reagents used analytical grade and were supplied by
Aldrich. The water used was Milli-Qplus quality with resistivity
18 M cm.
Apparatus
Potentials were measured with a Digimed pH meter, model
DMP-3, by using a combined silver electrode. Infrared spectra
were obtained with a Bomen FT-IR 100-MB spectrophotometer.
The TG analysis was made using a Shimadzu Model TGA 50
Thermobalance, at a heating rate of 293 K min1 and a temper-
ature range of 2981273 K.
Preparation of Functionalized Silica Gels
Silica gel with a specific area of 500 m2 g1 , degassed at 423 K
under high vaccum (103 atm) for 8 h, was used to obtain both
functionalized materials. For one of them, 5.0 mL (0.0213 mol)
of 3-amimopropyltriethoxysilane was dissolved in 100 mL of
dry toluene and mixed with 15.0 g of activated silica gel. For
the other, 5.0 mL (0.0227 mol) of N -[3-(trimethoxysilyl)propyl]
ethylenediamine was used. The mixtures were refluxed under ni-
trogen atmosphere for 24 h and then filtered and washed with
ethanol in Soxhlet extractor for 24 h. The resulting functional-
ized silica gels, Si(CH2 )3 NH2 (SA) and Si(CH2 )3 NH(CH2 )2
NH2 (SE), were filtered and dried under vaccum at 353 K for 8 h.
Preparation of the Quaternary Ammonium Salts Derivatives
of SA and SE
Twelve grams of each of the functionalized silica gels was
suspended in 50 mL of N ,N -dimethylformamide, and 4.0 mL
(0.0643 mol) of methyl iodide was added gradually. The mix-
tures were stirred for 2 h in the dark, at room tempera-
ture. The solid products were filtered and washed with water,
then with a 2% sodium bicarbonate solution, and again with
water. The materials were dried under vaccum at 353 K for
6 h. The methylation reactions were repeated two times for
SA and four times for SE, to ensure a good yield of the de-
sired quaternary ammonium products, Si(CH2 )3 N(CH3 )3 I
and Si(CH2 )3 NH(CH2 )2 N(CH3 )3 I, respectively. Finally, about
0.1 g of the iodide ammonium-functionalized silica gels was dis-
persed in 50 mL of 10 wt% sodium chloride (perchlorate) solu-
tion. The mixtures were slowly stirred for 5 h at a temperature of
298.0 0.5 K. The solids were filtered, washed repeatedly with
water, ethanol, and diethyl ether, and then dried in air. The result-
ing derivative chlorides, Si(CH2 )3 N+ (CH3 )3 /Cl (SA+ /Cl )
and Si(CH2 )3 NH(CH2 )3 N+ (CH3 )3 /Cl (SE+ /Cl ), were used
in the anion-exchange experiments, while the corresponding
derivative perchlorate of the SA+ was applied in the ion-selective
electrode experiments.
Characterization
The nitrogen content of the aminated silica gels, SA and SE,
was determined by the Kjeldahl method (32).
The chloride ion present on SA+ /Cl and SE+ /Cl was deter-
mined by potentiometric titration. About 0.2 g of the materials
was shaken with 50 mL of 0.1 mol L1 HNO3 solution for 1 h.
The solids were filtered and washed repeatedly with water until
the chloride anion had disappeared. The amount of Cl present
was determined by potentiometric titration with 0.01 mol L1
AgNO3 standard solution. This value corresponds to the limit
amount of adsorbed Cl on the supports.
The infrared spectra of the unmodified silica gel and the func-
tionalized silica gels were performed by using the pure material
pressed disk technique (1).
To obtain information about the thermal stability of the studied
materials, the analysis of TG was made by using the dynamic
heating method (33).
Anion-Exchange Isotherms
The anion-exchange isotherms in aqueous medium were
obtained for SA+ /Cl and SE+ /Cl by the batch technique at
298.0 0.1 K. Adequate weights of the materials containing
0.1 mmol of the immobilized quaternary ammonium salts were
mechanically stirred for 2 h with 1.68.0 mmol L1 NaX so-
lutions (X = F , Br , I , N
3 , NO3 , SCN , and ClO4 ). The
amounts of exchanged Cl present in the supernatant solutions
were measured potentiometrically (0.01 mol L1 AgNO3 ).
Competitive Exchange
The competitive Cl X exchanges were carried out by us-
ing mixtures containing 0.14 g of the functionalized materi-
als, SA+ /Cl and SE+ /Cl , fixed amounts (2.0 mmol) of NaCl
and NaX, and 5.0 mL of water. The mixtures were stirred for
12 h at 298.0 0.1 K, filtered, washed exhaustively with water,
followed by ethanol and diethyl ether, and finally dried under
vacuum. The amount of adsorbed Cl ion on the supports was
determined by treating the solids with 0.1 mol L1 HNO3 so-
lution and quantified by potentiometric titration (0.01 mol L1
AgNO3 ). These values, subtracted from the limit amounts of
adsorbed chloride on the surfaces, give the amounts of X ions
on the supports.
Construction of the Selective Electrode for ClO
4
Using SA+ /ClO 4
The electrode was prepared by intimately mixing 1.0 g of
graphite powder with 1.2 g of epoxy resin (28 wt% of the harder
and 72 wt% of resin) (29, 30). The mixture was inserted at the
 QUATERNARY AMMONIUM SALTS IMMOBILIZED ON SILICA GEL 99

extremity of a glass tube with an internal diameter of 1.0 cm.

A layer (1 mm thick) of SA+ /ClO 4 was physically fixed on
the external surface of the epoxy support. The internal surface
of the electrode was covered with a layer of colloidal silver. A
coaxial cable was connected to this layer.

Potentiometric Measurements
With the purpose of evaluating the sensitivity of the elec-
trode SA+ /ClO 4 for perchlorate ions, the analytical curve was
obtained by the separated solution method (34, 35), by using
20.0 mL of 101 105 mol L1 NaClO4 solutions. The response
of the electrode was studied by the successive additions of
101 and 102 mol L1 NaClO4 standard solutions in 20.0 mL of
water. These additions were made by using a Gilmont S-1200A
microburet (0.002 mL accuracy). The potentiometric selectiv-
pot
ity coefficients, K A,B , were evaluated by varying the concentra-
tion of the primary ion A (ClO 4 ) at a fixed concentration of the
interfering ion B (F , Cl , Br , NO
3 , SCN , and CH3 COO ).

To carry out the experiment, two solutions were prepared:

solution A, 9.0 102 mol L1 NaClO4 solution, containing
interfering ions at 1.0 103 mol L1 ; and solution B, 1.0
103 mol L1 interfering ion solution. The electrochemical cell FIG. 1. Reactions for obtaining the quaternary ammonium salts immobi-
was filled with 20.0 mL of solution B, and aliquots of solution lized on silica gel, SA+ /Cl and SE+ /Cl .
A were added.
The influence of pH on the electrode response was examined
over the pH range 2.28.0, by using a 1.0 103 mol L1 between 413 and 1013 K for SE+ /Cl . Therefore, the materials
NaClO4 solution and citric acid/Na2 HPO4 buffer. are thermically stable up to 413 K.

Anion-Exchange Properties
RESULTS AND DISCUSSION
To determine the exchange ability as a function of the time,
Characteristics of the Materials SA+ /Cl and SE+ /Cl were shaken in a 8.0 103 mol L1
NaBr solution. The results indicated that the maximum sorp-
The immobilization of the quaternary ammonium salts on the tion of the Br ion could be attained within 30 min for both
silica gel surface is represented by the reactions in Fig. 1. exchangers (Fig. 2).
The results of the chemical analysis for SA, SE, SA+ /Cl ,
Exchange isotherms. The anion-exchange reaction involv-
and SE+ /Cl are shown in Table 1. These results indicate that
ing monovalent anion can be described by the equilibrium
nearly 100% of the aminic nitrogen in SA and 50% in SE
were converted to quaternary ammonium salts. This is proba-
An+ Yn + n MX An+ Xn + n MY, [1]
bly due to the strong steric hindrance on the secondary nitrogen
in SE.
where An+ represents the anion exchanger and Y the counte-
The absorption bands of the FTIR spectra of unmodified silica
rion. Because the exchanges occur in equivalent amounts, the
and grafted silicas and the respective assignments are listed in
competing anion (X ) concentration at the equilibrium can be
Table 2.
inferred from the amount of the exchanged counterion (Y ),
The main evidence from the infrared spectra is the strong
considering the beginning X concentration. The equilibrium
band observed at 1610 cm1 in SA and at 1620 cm1 in SE,
which can be identified as bending vibrations of NH2 groups.
These frequencies are absent in the spectra of SA+ /Cl and TABLE 1
SE+ /Cl , in which strong bands at 1500 cm1 were observed; The Nitrogen and Chloride Analyses of Modified Silica Gels
they were attributed to the asymmetrical bending vibrations Sample N (mmol g1 ) Cl (mmol g1 )
of CH3 groups. This confirms the methylation of the NH2
groups. SA 0.70 0.01
According to TG analysis (curves not shown) the functional- SE 2.49 0.02
ized silicas presented a sudden weight loss at 413 K and a total SA+ /Cl 0.70 0.01 0.70 0.01
SE+ /Cl 2.49 0.01 1.19 0.01
moderate weight loss between 413 and 873 K for SA+ /Cl and
100 DE CAMPOS ET AL.

TABLE 2
Infrared Absorption Frequencies (cm1 ) of the Unmodified
and Functionalized Silica Gels
Unmodified
silica SA SA+ /Cl SE SE+ /Cl

3400 (s,b) 3400 (s,b) 3400 (s,b) 3400 (s,b) 3400 (s,b) OH
2990 (m) 2990 (m) 2990 (m) 2990 (m) CH2
1645 (s) 1645 (s) 1645 (s) 1645 (s) 1645 (s) HOH
1610 (s) 1620 (s) NH2
1500 (s) 1500 (s) CH2 ; CH3
1485 (w) 1480 (s) CCH2C

Note. s, strong; m, medium; w, weak; b, broad.

described in Eq. [1] can be analyzed according to the treatment

(36)

N0 1 1
= + , [2]
Ne n n K [X ]

where [X ] is the concentration (in mol L1 ) of the competing

anion in aqueous solution after the exchange, Ne is the change
of the concentration of X in the exchanger phase (in mol g1 ),
N0 is the exchange capacity of the material (in mol g1 ), n is the
molar fraction of exchanged anion (Y = Cl , in this work) to
the infinite concentration of the competing anion, and K is the
equilibrium constant of Eq. [1] (binding constant). Multiplying
both terms in Eq. [2] by [X ] and dividing by N0 , we get
FIG. 3. Exchange isotherms for some anions (X ) on the exchangers
SA+ /Cl (a) and SE+ /Cl (b).
[X ] [X ] 1
= + . [3]
Ne n N0 n N0 K
changes and that the active sites of the exhanger are uniformly

The exchange isotherms for each anion X competing with spread on the surface.
the counterion Y of the exchanger can be represented by plot- Figure 3 shows the exchange isotherms for the X ions (F ,
ting [X ]/Ne as a function of [X ]. Br , I , ClO
+
4 , N3 , NO3 , and SCN ) on SA /Cl and SE /Cl
+

A linear correlation will be obtained, considering that each according to Eq. [3], and Table 3 reports the n and K values
exchanged ion does not have influence on the successive ex- determined from these isotherms.

TABLE 3
na and K b Values for Some Anions (X ), Determined
from Equation [3]

Anion exchanger

SA+ /Cl SE+ /Cl

Competing anion
X K 103 (mol L1 ) n K 103 (mol L1 ) n

F 7.6 0.58 8.0 0.92

Br 0.70 0.46 0.61 0.90
I 1.1 0.47 2.0 0.95
N3 1.3 0.49 1.0 0.84
NO 3 2.1 0.49 2.7 0.91
SCN 2.8 0.56 3.8 0.81
ClO 4 5.0 0.75 4.2 1.0

an = molar fraction of exchanged anion (Cl ) to the infinite X .

FIG. 2. Cl Br exchange ability ([Br ] = 8.0 103 mol L1 ) at 298 K, bK = equilibrium constant of Eq. [1] (binding constant), obtained from the
as function of the time: (a) SA+ /Cl , (b) SE+ /Cl . exchange isotherms.
 QUATERNARY AMMONIUM SALTS IMMOBILIZED ON SILICA GEL 101

The K values are directly related to the interaction forces
between the active sites of the exchanger and the counterions.
The stronger the interaction of the ionic pair, the higher the K
value will be. From the data of Table 3, it is observed that a
strong interaction of the ionic pair does not necessarily indi-
cate that all of the ions are occupying the active sites of the
exchanger. For example, though the K value for the F ion
on SA+ /Cl is higher than that for the other ions studied, only
58% of the active sites of the exchanger are occupied by the
fluoride to the infinite concentration of this ion. For the ClO 4
ion on SE+ /Cl , it is observed that when the concentration of
ClO
4 in the solution tends to infinity, all Cl ions of the sup-
port are exchanged (n = 1), but the K value is smaller than that
for the F , which occupies at the saturation, about 92% of the
exchanger active sites (n = 0.92). This occurs because the se-
lectivity of an ion exchanger is not only governed by ionion in-
teractions but also depends on solventsolvent, solventsupport,
and solventions interactions (37, 38). Thus, the K constant is
better expressed through the thermodynamic equilibrium con-
stant, K Y (37), which depends on the concentrations and on the
activity coefficients of the anions involved in the equilibrium in
solution, as well as in the exchanger phase.
Therefore, the equilibrium showed in Eq. [1] can be repre-
sented by the equation
X f (A+ X )[Y ]Y
K Y = , [4]
Y f (A+ Y )[X ]X
for the composition K Y = (k)X = 0.5, when log k is a linear
function of X , which does not always happen.
The linear log k vs X plots for the exchangers SA+ /Cl and
SE+ /Cl are shown in Figs. 4 and 5, respectively. From these
linear relationships, obtained the K Y values (for X = 0.5),
which are in Table 4.
Affinity coefficients. The affinities of the systems SA+ /Cl
and SE+ /Cl for the various X anions compared to Cl were
determined from the competitive Cl X exchange reactions,
as described under Experimental. The affinity coefficients ( A)
obtained are shown in Table 4 and confirm the selectivity order
of the exchangers, observed through the thermodynamic equi-
librium constants (K Y = K Cl ).
From the results of K Y and A (Table 4) and the values of
K (Table 3), we can write the following: the order of the values
of the thermodynamic equilibrium constants (K Y ) is different
from the order observed for the binding constants (K ) and is the
same as the affinity coefficients ( A). These observations confirm
that the ion-exchange process does not depend only on the inter-
actions between the exchanger and the counterion. Therefore,
the selectivity order of an ion exchanger cannot be set up only
by comparing the relative values of K .
The data of Table 4 suggest that SA+ /Cl presents high affin-
ity for perchlorate ion, and SE+ /Cl for fluoride ion. The high
affinities for these ions are presumably due to the great hy-
drophilicity of the anion in the case of fluoride (8) and to the

where X and Y are the molar fractions of X and Y in

the exchanger phase, f (A+ X ) and f (A+ Y ) are the activity
coefficients in the exchanger phase, X and Y are the activ-
ity coefficients in the solution phase, [Y ] is the concentration
(in mol L1 ) of the exchanged anion, and K Y is the constant
equilibrium of Eq. [1]. Equation [4] can be rearranged as

f (A+ X )
K Y = k , [5]
f (A+ Y )

where

X [Y ]Y
k= . [6]
Y [X ]X

Applying the GibbsDuhem equation to Eq. [5] we get

Z X
ln f (A+ Y ) = X ln k ln kdX . [7]
0

The interactions between the exchanger and the ions in

the solid phase and solution are expressed by k. In this
manner,K Y , f (A+ X ), and f (A+ Y ) will be related to these
interactions . Although the evaluation of K Y requires that k be
measured for a large number of X values in the range 01, k
is sufficiently calculated for a few values of X (13). Assuming
X = Y = 0.5 and plotting log k vs Y , one can obtain K Y FIG. 4. Cl X exchanges at 298 K on SA+ /Cl .
102 DE CAMPOS ET AL.

FIG. 6. Relative selectivity of the exchangers SA+ /Cl and SE+ /Cl , ex-
pressed as log K Y (Y = Cl ) .

indicates a higher selectivity of the exchanger. Thus it can be

FIG. 5. Cl X exchanges at 298 K on SE+ /Cl .
high polarizability and low anion hydration enthalpy in case of
perchlorate (13, 39), in comparison with other anions.
The higher values of K Y for SE+ /Cl than for SA+ /Cl
suggest that it changes the counterion (Y ) more easily than
SA+ /Cl . This behavior is shown in Fig. 6, where the log K Y
values for the SA+ /Cl and SE+ /Cl exchangers are related.
High values of log K Y indicate high affinity of the support
to the anions. A great separation between the log K Y values
concluded that, although the SA+ /Cl presents shorter log K Y
values than SE+ /Cl , it is more selective than SE+ /Cl because
it presents great diferences between the log K Y values.
Both exchangers, SA+ /Cl and SE+ /Cl , presented high hy-
drophilicity compared to the other hydrophilic anion exchangers
described in the literature (8). Due to this property, SA+ /X and
SE+ /X can be easily recovered in the chloride form, by treat-
ing first with a bicarbonate solution (highly hydrophilic anion),
followed by a hydrochloric acid solution. In the case of SE+ /X
a 10% NaCl solution instead of HCl was used to avoid protona-
tion of the secundary nitrogen not quaternarized. The recovery
of SE+ /X in the chloride form is favored by the lability of the
counterions of the support due to high equilibrium constants for
the exchange process (Table 4). Figure 7 shows the results ob-
tained for SA+ /Cl and SE+ /Cl after successive regenerations.

TABLE 4
Selectivity Parameters (K Y , A)a of the Exchangers SA+ /Cl
and SE+ /Cl for Some Competing Anions (X )

Support

SA+ /Cl SE+ /Cl

X K Y (Y = Cl ) A K Y (Y = Cl ) A

F 1.71 0.45 3.81 21.5 2.5 3.47

Br 0.0111 0.0010 2.01 0.887 0.039 1.28
I 0.102 0.009 2.37 3.78 0.71 1.99
N3 0.178 0.012 2.42 1.19 0.09 1.45
NO 3 0.198 0.014 2.53 4.78 0.48 2.59
SCN 0.530 0.038 2.97 3.13 0.14 1.90
ClO 4 3.65 0.29 4.61 10.1 0.7 2.85

aK
Y = thermodynamic equilibrium constant.
A = affinity coefficient determined from the competitive Cl X exchange
reactions. FIG. 7. Recovery capacity of the exchangers SA+ /Cl (a) and SE+ /Cl (b).
 QUATERNARY AMMONIUM SALTS IMMOBILIZED ON SILICA GEL 103

Use of SA+ /ClO

4 as a Potentiometric Electrode

The SA+ /Cl , instead of the SE+ /Cl , exchanger was cho-
sen to prepare an ion-selective electrode for perchlorate due to
its high affinity and selectivity for this ion (SE+ /Cl showed
a higher affinity for the F ion). The derivative SA+ /ClO 4 was
used as a selective potentiometric sensor for this anion. The con-
struction of the electrode and the procedure of the potentiometric
measurements are described in the Experimental section.
Electrode response. Figure 8 shows the potentiometric re-
sponse of the electrode made with the material SA+ /ClO 4.
The response to the primary ion (ClO 4 ) is linear in the con-
centration range from 101 to 104 mol L1 and is described
by the equation E SA+ /ClO4 = 291.5 + 58.2 log [ClO 4 ] with
r = 0.999. The angular coefficient of 58.2 is very close to the
59 mV required for a nernstian behavior. The limit of the elec-
trode response is nearly 1.3 104 mol L1 in the absence of
interfering ions, with a response time of about 20 s. The elec-
trode gave reproducible potentials with a standard deviation of
0.1 mV and was used over a period of 2 months (approximately
200 determinations) with a decrease of 20 mV in the potential. FIG. 9. Effect of pH on the SA+ /ClO
4 electrode response ([ClO4 ] =
The effect of pH between 2.2 and 8.0 using a citric 1.0 103 mol L1 ; buffer = citric acid/Na2 HPO4 ).
acid/Na2 HPO4 buffer on the electrode response was inves-
tigated. Figure 9 shows that the hydrogen ions interfere
in the electrode response in the pH range between 2.2 and 5.5. performance in the presence of interfering ions. The degree of
The useful pH range for the electrode is between 5.5 and 8.0. selectivity of the electrode for the primary ion, A, with respect
The interference at pH below 5.5 presumably is due to the ad- to an interfering ion, B, is expressed by the potentiometric selec-
pot
sorption of H3 O+ by the matrix. Above pH 8.0 the electrode tivity coefficient,K A,B , calculated by applying the equation (40)
does not work well due to the partial solubilization of silica.
pot
Selectivity of the electrode SA+ /ClO K A,B = (aA ) Z b /(aB ) Z a , [8]
4 . The main interest in
the development of a new ion-selective electrode is to know its
where aA = 18/Z a , aB is the activity of the interferent ion, and
Z b and Z a are the charges of the interferent and primary ions,
respectively.
The fixed interference method was used to investigate the
response of the electrode SA+ /ClO 4 to the interfering anions
F , Cl , Br , NO3 , SCN
, and CH
3 COO . In Table 5 the

TABLE 5
pot
Potentiometric Selectivity Coefficients, K A,B , for the Perchlorate
Selective Electrode, SA+ /ClO4 , and Some Other Solid-State Per-
chlorate Electrodes
Potentiometric selectivity
pot
coefficients (K A,B )
Interfering
anions SA + /ClO
4 Lorencetti a Fernandesb

F 0.85
SCN 0.27 0.91
NO3 0.10 0.28 0.0046
Br 0.08 0.21
Cl 0.03 0.15 0.30
CH3 COO 0.02 0.0063

FIG. 8. Response of the electrode SA+ /ClO

4 obtained by the method of
a Silica gel modified with pyridinium perchlorate (Ref. 30).
successive additions. b Silica gel modified with 1,4-diazobicyclo(2.2.2)octane perchlorate (Ref. 29).
104 DE CAMPOS ET AL.
values of the potentiometric selectivity coefficients are listed
and compared with the values obtained for other similar per-
chlorate electrodes made with modified silica gels (29, 30).
As can be seen from the selectivity coefficient values, fluo-
ride is the major interfering ion and acetate is the less inter-
pot
fering ion in this case (Table 5). The decreasing order of K A,B
is F > SCN > NO3 > Br > Cl > CH3 COO .
CONCLUSIONS
From the values of the thermodynamic equilibrium constants
and affinity coefficients obtained from the anion-exchange stud-
ies, the following selectivity setup can be established for both
exchangers: SA+ /Cl , ClO
4 > F > SCN > NO3 > N3 >
I > Br ; and SE /Cl , F > ClO4 > NO3 > I > SCN >
+
N3 > Br .
The exchange studies showed also that, although SE+ /Cl
exchanges the counterion Cl more easily than SA+ /Cl , the
latter is more selective than SE+ /Cl .
From the potentiometric studies it can be concluded that the
ion-selective electrode SA+ /ClO 4 showed good performance
in the presence of various interfering ions such as F , SCN ,
NO
3 , Br , Cl , and CH3 COO . This property, associated with
the high chemical stability and durability of the material, indi-
cates that this kind could be used as a potentiometric sensor to
determine the perchlorate anion
ACKNOWLEDGMENT
A.A.S.A. and R.T.F. are indebted to CAPES for the fellowship.
REFERENCES
1. Gushikem, Y., and Moreira, J. C., J. Colloid Interface Sci. 107, 70 (1985).
2. Gushikem, Y., and Silva, M. A. A., J. Colloid Interface Sci. 107, 81 (1985).
3. Airoldi, C., and Alcantara, E. F. C., Colloids Surf. 39, 291 (1989).
4. Iamamoto, M. S., and Gushikem, Y., J. Colloid Interface Sci. 129, 162
(1989).
5. Aleixo, L. M., Souza, M. F. B., Godinho, O. E. S., Oliveira Neto, G., and
Gushikem, Y., Anal. Chim. Acta 271, 143 (1993).
6. Pesco, C., Campos, E. A., and Costa, C. M. M., Mikrochim. Acta. 127, 229
(1997).
7. Wasiak, W., Chromatographia 23, 427 (1987).
8. Tundo, P., Venturello, P., and Angeletti, E., J. Am. Chem. Soc. 104, 6547
(1982).
9. Airoldi, C., and Goncalves, A. S., Colloids Surf. 28, 199 (1987).
10. Iamamoto, M. S., and Gushikem, Y., J. Colloid Interface Sci. 134, 275
(1990)
11. Desechler, U., Kleinschmit, P., and Panster, P., Angew. Chem. 25, 236
(1986).
12. Unger, K. K., Porous SilicaIts Properties and Use as Support in Colunn
Liquid Chromatography. Elsevier, Amsterdam, 1979.
13. Gushikem, Y., and Moreira, W. C., Colloids Surf. 25, 155 (1987).
14. Iler, R. K., The Chemistry of Silica. Wiley, New York (1979).
15. Inczedy, J., J. Chromatogr. 102, 41 (1974).
16. Schulthess, P., Amman, D., Krautler, B., Caderas, C., Stepanek, R., and
Simon, W., Anal. Chem. 57, 1937 (1985).
17. Jain, K. A., Jahan, M., and Tyagi, V., Analyst 112, 1355 (1987).
18. Jain, K. A., Jahan, M., and Tyagi, V., Anal. Chim. Acta 231, 69 (1990).
19. Baczuk, R. J., and Bolleter, W. T., Anal. Chem. 39, 93 (1967).
20. U.S. Department of the Navy, MIL-A23946 (AS). U.S. Govt. Printing
Office, Washington, DC, 1966.
21. Coetzee, C. J., and Freiser, H., Anal. Chem. 40, 2071 (1968).
22. Coetzee, C. J., and Freiser, H., Anal. Chem. 41, 1128 (1969).
23. Janes, H. J., Carmack, G. P., and Freiser, H., Anal. Chem. 44, 856
(1972).
24. Back, S., Anal. Chem. 44, 1696 (1972).
25. Luca, C., Semenescu, G., and Nedea, C., Rev. Chim. 25, 1015 (1974).
26. Sharp, M., Anal. Chim. Acta 62, 385 (1972).
27. Fogg, A. G., Pathan, A. S., and Brums, D. T., Anal. Chim. Acta 73, 220
(1974).
28. Kataoka, M., and Kambara, T., J. Electroanal. Chem. 73, 279 (1976).
29. Fernandes, J. R. L., Kubota, L., Gushikem, Y., and Oliveira Neto, G., Anal.
Lett. 26, 2555 (1993).
30. Lorencetti, L. L., Gushikem, Y., Kubota, L. T., Oliveira Neto, G., and
Fernandes, J. R. Mikrochim. Acta 117, 239 (1995).
31. Perez-Olmos, R., Rios, A., Martn, M. P., Lapa, R. A. S., and Lima,
J. L. F. C., Analyst 124, 97 (1999).
32. Vogel, A. I., Quantitative Inorganic Analysis, 3rd ed. London, 1961,
p. 256.
33. Wendlandt, W. W., Thermal Analysis, 3rd ed. Wiley, New York,
1986.
34. Guibault, G. G., Durst, R. A., Frant, M. S., Freiser, H., Hansen, H. E., Light,
T. S., Pungor, E., Rechnitz, G., Rice, N. M., Robin, T. J., Simon, W., and
J. Thomas, D. R., Pure Appl. Chem. 48, 127 (1976).
35. Guibault, G. G., Durst, R. A., Frant, M. S., Freiser, H., Hansen, H. E.,
Light, T. S., Moody, G. J., Pungor, E., Rechnitz, G., Rice, N. M., Robin,
T. J., Ruzicka, J., Simon, W., and Thomas, J. D. R., IUPAC 70 (1978).
36. Klotz, I. M., Walker, M. F., and Pivan, R. B., J. Am. Chem. Soc. 68, 1486
(1946).
37. Chu, B., Whitney, D. C., and Diamond, R. M., J. Inorg. Nucl. Chem. 24,
1405 (1962).
38. Dyer, A., and Malik, S. A., J. Inorg. Nucl. Chem. 43, 2975 (1981).
39. Aveston, J., Everest, D. A., and Wells, R. A., J. Chem. Soc. 231 (1958).
40. Bailey, P. L., Analysis with Ion Selective Electrodes, 2nd ed., p. 49. Hyden
& Son, 1980.