CLIN.CHEM.37/8, 1375-1378 (1991)
Analysis of Ion Behavior and Potentials in a Na
Containing Methyl Monensin
Ion-Selective Membrane
Masato Fujiwara
Ion-selective electrodes (ISEs) are widely used in clinical
examinations. The mechanisms of ISE5 have been studied to develop a disposable ISE for “dry chemistry”
analyses. The behavior of Na and Ag in an ionselective polymer liquid membrane (ISM) containing
methyl monensin (MMO), which is selective for Nat has
been clarified by x-ray microanalysis and measurement of
electrical potential. The results indicate that Na is not
transported across the membrane because of a stabilityof
the Na-MMO complex, whereas Ag (from the Ag/A9CI
electrode) is transported with C1 across the membrane
by diffusion. The difference in potential, which follows a
Nernst gradient, is generated by the formation of a double
layer of electrons at the interface between the ISM and
the electrolyte solution. These results are explainable by
the stability of the trap in the polymer liquid membrane.
MMO acts as a stable trap of Na and as a metastable
site of Ag + for movement.
AdditionalK.yphrases: electrolytes
analysis
.
electrochemistiy
.
Ag
x-ray micropolymer liquidmembrane
.
electrochemical
application
analyses.
Surprisingly,
x-ray microanalysis
demonstrates
that
cannot be transported
across the polymer liquid
membrane,
whereas Ag from the standard
Ag/AgCl
electrode layer diffuses into the membrane along with
Cl-. The polymer liquid membrane should behave differently from the liquid membrane (14), and this difference should affect the mechanisms
of the potential
generation and the ion transport in the ISE.
Na
The
analysis
of the
potential
of the
ion-selective
polymer liquid membrane (ISM) shows that potential
differences are generated at the interface between the
ISM and the electrolyte
solution. The difference
in
potential, which depends on the activity of Na, behaves
according to a Nernst gradient.
.
Use of small and disposable test equipment is one of
the common trends in clinical examinations.
This approach has many advantages in electrochemical
analysis of biological trace substances and instruments, because the use of the sensors allows the use of a small
sample size. In recent years, disposable ion-selective
electrodes
(ISEs) have been used widely in electrochemical analysis of ions in clinical tests (particularly
in
blood), food analysis,
and environmental
analysis (13)#{149}1
Further
to generate potential differences proportional to
the concentration of Na (8-13). Consequently, both the
selectivity
and the ion transport of Na are very important for determining
the activity of Na in clinical
brane
of the dispos-
able ISE has been extended by “dry chemistry” analysis
(e.g., DRI-Chem from Fuji Photo Film, Asaka, Saitama,
Japan, and Ektachem from Eastman KOdak, Rochester,
NY) (4-7). The small volumes of blood and standard
solution are in direct contact with the two different
ion-selective membranes. The amount of ion in the blood
is decidedby the difference in potential between the
blood and the standard solution. The disposable ISE is
not used again.
Na is an important component of blood. ISEs for Na
have been fabricated by using a polymer liquid membrane and methyl monensin (MMO), which has a high
selectivity
for Na. MMO carries Na in a liquid mem-
Materials and Methods
Electrodes
having two different types of polymer liquid membrane were prepared as shown in Figure 1 (13,
14). The Ag/AgCl electrode was prepared by depositing
under very low pressure
an 800-nm layer of Ag metal
onto a polyethylene terephthalate
(PET) film. The Ag
was oxidized to AgC1 by HC1 and K2Cr2O7 (conversion
ratio = 25.0%). In the ISE shown in Figure la, an NaCl
layer is present between the ISM and the AgCl layer.
This ISE was prepared by coating polyvinyl
alcohol
solution (per liter, 50 g of polyvinyl alcohol and 0.2 mol
of NaCl) onto the AgCl layer and drying at 80#{176}C
for 1 h
(thickness of NaCl layer = 6 pm). The subsequently
accumulating polymer solution consisted of 5.0 mL of
_-
100 ---.....3
ISM
Ag/AgCI NaCI
()
3-
I----
30gm
ISM
PET
Asaka Research Laboratories, Fuji Photo Film Co., Ltd., Senzui,
Asaka-shi, Saitama-ken, 351 Japan.
‘Nonstandard abbreviations: ISE, ion-selective electrode; ISM,
ion-selective membrane; MMO, methyl monensin; and PET, polyethylene
terephthalate.
Received September 13, 1990; accepted June 3, 1991.
30,/hm-..
/6
1.,
Ag/AgCI
(b)
Fig. 1. Cross section of two polymer liquid membrane electrodes
CLINICALCHEMISTRY,Vol.37, No.8, 1991 1375
methyl ethyl ketone containing 0.9 g of copolymer
(polyvinyl chloride and polyvinyl acetate, 9/1, by wt), 0.3
g of ion-selective molecule (MMO), and 1.8 g of organic
solvent (dioctyl sebacate). The resulting
layer (30 m
thick) was dried at 60#{176}C
for 1 h.
The ISE shown in Figure lb contains no NaC1 layer;
the polymer solution was coated directly onto the Ag!
AgC1 electrode layer.
Aqueous solutions of NaCl at various concentrations
were left in contact with the surface of the ISM for
various
specified
times, and then were removed by
suction intoa filter paper. The portion of the ISE that
had touched the NaCl solution was embedded in an
epoxy resin (Semendain Super; Semedain
Co., Ltd.,
Tokyo, Japan) for about 16 h, after which a cross-section
of the ISM was cut with a diamond knife. The surface of
the cross-section was coated by carbon evaporated under
very low pressure. The distributions of the ions in the
ISM were measured
with an x-ray microanalyzer
(wavelength-dispersion type, HSM 2B XMA; Hitachi, Tokyo,
Japan).
The difference in potential between the two silver
layers of the ISEs (Figure 1) was measured as follows.
Aqueous NaCl solutions were placed on the surface of
the two ISMs, then interconnected
by a salt bridge. The
differences
in potential between the two Ag layers, and
between the Ag layer and a standard calomel electrode
(SCE), were measured
2 mm after the NaCl solution
was in contact with the voltmeter (Type 901; Orion
Research Inc., Boston, MA) arranged in a series.
The changes of the amount of Na on the ISM after
contact for 1-60 miii were analyzed by atomic absorption spectrometry
(Zeeman Analyzer type 701; Hitachi,
Tokyo, Japan).
Results
Figure 2 shows the distributions
of Na, Ag, and C1
in the cross-section of the ISM after contact with the
aqueous NaCl solution.
When the NaCl layer was present (Figure 2a), the
distribution of neither Na nor Ag showed any change
over time. Thus, Na did not diffuse into the ISM. This
means that the Na’-MMO complex (15) formed at the
interface between the NaCl layer and the ISM layer did
not diffuse into the ISM. This also shows that Ag did
not diffuse into the ISM, because the solubility of Ag
was severely diminished by Cl- in the NaCl layer
covering the Ag/AgC1 electrode.
However, if the NaC1 layer was not present (Figure
2b, c), Ag and Cl- diffused into the ISM. The same
results were obtained by using distilled water or serum
instead of NaC1 aqueous solution. The amount of Na
(from the solution that remained) on the surface of the
ISM was constant, regardless of the length of contact
(1-60 min). This result showed that Na did not migrate
into the ISM, and moreover that the phenomena were
not dependent on ion transport of Nat, but on the
permeation of water.
The diffusion behavior of Ag in the absence of the
NaCl layer was as follows: Ag diffused into the ISM
along with Cl- (Figure 2b, c). The amount of Ag
measured by the emission flux of the x-ray microanalyzer increased with increasing MMO and was proportional to the square root of the time in contact with the
water (Figure 3). The results indicate that Ag is
transported by diffusion process. The diffusion constant,
dependent on the amount of MMO, was estimated to be
4.4 x 1O_9cm2/s (MMO concentration = 110 pmolIL).
These phenomena are similar to that of Ag behavior in
a K ISM prepared with valinomycin (14, 16).
The behaviors of Na, Ag, and Cl- in the Na ISMs
are shown in Figure 4. NaCl aqueous solutions of
various concentrations were placed in contact with the
surface of the ISM. The differences in potential between
the ISE in Figure la and a standard
electrode, and
between the ISE in Figure lb and a standard electrode,
are shown in Figure 5. The gradients of the two plots
were approximately nernstian. However, the difference
in potential between the two ISEs was about 560 mV at
the same concentration, corresponding to the difference
in potential expected from the solubility constant of
AgCI (17).
Discussion
These phenomena for the ISEs in Figure 1 described
above are illustrated
in Figure 4. Sodium ions are
0
0..
4-.
‘U
L.
0mm
0.3
C
0
.1-i
‘U
N
5mm
60mm
PET
NI;
ISM
4/A
(a)
PET
I ISM
Ag/Acm
/T
ISM
Ag/AgCI
(b)
(C)
Fig. 2. Characteristic x-ray flux of Ag (solid line, a and b), Cl (c),
andNa (dotted line, a)
lime at 0
mm Is before contact with aqueous
indicate the duration of contact
NaCI solution; 5 and 60 mm
1376 CLINICALCHEMISTRY,Vol.37, No.8, 1991
.0
0.2
0.1
0
(I)
10
(Time/s.
V2
Fig.3. Diffusionbehavior of Ag with time (in
The solubilizatlon ratio = 8olublllzedAg/totaI Ag
seconds)
v-Io
1.Coiipl.x
l.c-
Na
2. Trip
C,
ci
LTr.p
ci
Cl
CI
3.Trinstir
or
§cr
4. Attract
#{231}Na
5. Dfftieton
5. A&tm5
4. Esthic
Layir
5. Rsputulon
cr
Nil
§N
Na
ISM
MaCI
Solution
crast.
AgCI
SM
a
Fig.4. Mechanisms forNa, Ag, andCl in the ISMs from Fig. is
(a)and
Fig. lb(b)
mV
(a)
(b)
Ui
U
U,
-1
‘U
C
a,
400
NaCIconcentrat)on (mmol/L)
Fig. 5. Nemst plotsof potentials measured with theISEshownin Fig.
1 a (a) and lb (1)
DifferenceIn potential of 100,140, and 180 mmot/L aqueoussolutions of NaCl
measured after 2 mm:0, valuesof ISE (is); #{149},
values of ISE (ib)
transported by MMO in the liquid membrane (8-11).
However, Na was not transported
across the ISM,
neither from the NaCl layer nor from the aqueous NaC1
solution touching the ISM surface. These differences of
behavior are elucidated by the mobilities of MMO and
the depth of the MMO trap. The trap for Na appears to
be very deep, because it gives a clear x-ray diffraction
pattern
(15). Because the MMO trap is so deep, Na
cannot move to the adjacent MMO by thermal vibration
energy alone; also, Na is attracted by the counter ion,
Cl, which does not diffuse into the membrane. Consequently, Na in the MMO trap does not move into the
polymer liquid membrane (Figure 2a).
The diffusion behavior of Ag is in contrast to this.
The movement of Ag should be a passive
transport
(18), because the rate-determining
step of the Ag
transport
is a diffusion process. The MMO in the ISM
can act as a metastable
site for Ag. Silver ion in the
metastable
site of MMO should move to the adjacent
MMO site by thermal vibration energy, which causes
the Ag diffusion into the ISM. The Ag attracts the
counter ion, Cl- (Figure 2c). Diffusion of the counter ion
into the ISM promotes the diffusion of Ag, because C1
prevents the repulsion between positive charges of Ag.
As described
above, MMO does not move as an ion
carrier, as it does in a liquid membrane, but acts as a
metastable site for Ag diffusion here. The same behavior is observed by valinomycin, which acts as a metsstable site for Ag as reported previously (14).
The differences in potential in Figure 4a are generated as follows. Because the Na entrapped in the deep
MMO trap is not pulled into the electrolyte layer and
attracts the counter ion, Cl-, the separation of charges
(a double layer of electrons) is formed at the interface
between the ISM containing MMO and the electrolyte
layer. Three double layers of electrons are formed at the
interface between (a) the Ag/AgCl electrode and the
NaC1 layer, (b) the NaCl layer and the ISM layer, and
(c) the ISM layer and the solution being studied. As the
water of the solution being examined permeates the
electrolyte layer, the NaCl layer becomes saturated.
The potential of the Ag/AgCl electrode remains stable
and constant (560 mY) by contact with the saturated
NaCl solution. The MMO traps (about 10_b mollcm2) at
interface b are fully occupied by Na. Because the
concentration of Na from the solution examined is
about 100 mmol/L, the MMO traps at interface care also
occupied by Na. Consequently, the potential difference
generated at interface b is compensated by that of
interface c. The potential of ISE as shown in Figure la
is represented by the following expression (19):
(RT/F)(ln[K8]
+
ln[a])
(1)
where R = gas constant, T = absolute temperature,
F =
Faraday constant, K8 = solubility constant of AgCl, and
[a] = activity of Na in the solution (16).
Because the potential of the ISE as shown in Figure
la depends only on the concentration of Na of the
examined solution, this ISE should be stable both during use and in storage.
In Figure 4b, Ag is transported by the diffusion
process. Ag enters the ISM layer, where MMO can act
only as a moderately deep trap for Ag. Therefore, the
thermal vibration energy enables Ag to move to the
adjacent MMO trap and thus move across the membrane. Ag trapped in MMO can attract the counter ion,
Cl-, from the Ag/AgC1 electrode to maintain
electrical
neutrality. Because this movement can occur at all
points in the ISM, Ag diffuses into the ISM along with
Ct, but the interface between the electrode and the
ISM is electrically neutral. The separation of charges
(the electron double layer) is generated only at the
interface between the surface of the ISM and the examined solution. In storage, the potential of this ISE is
changed by the permeated
water, which solubilizes the
AgCl. Thus, the measured potential difference should
not be correct.
CLINICALCHEMISTRY,Vol.37, No.8, 1991 1377
The potential difference of the ISE shown in Figure lb
is generated at the interface between the surface of the
ISM and the examined solution, because there is no
separation of the charges in the ISM or at the interface
between AgCl and ISM. Because the amount of MMO at
the surface of the ISM is about 10_b mollcm2, MMO is
fully occupied by the Na from the solution examined
(about 100 mmol/L). As Na is entrapped in the MMO,
Na in the ISM is separated
from the Cl in the
solution. Consequently,
there is the separation of
charges (the double layer of electrons) of ions formed at
the interface between the ISM and the solution examined. The difference in potential is represented by the
following
(RTIF)(ln[a])
(2)
When the activity of Na
is the same, the difference in
the two kinds of ISEs is estimated
to
be (RTIFXIn[KJ)
(= 566 mY), which agrees with the
experimental
value of 555 mY (Figure 5). Expressions
1
potential
between
and 2 show that the potentials of the two ISEs give a
gradient when the activity of Na in the
examined solution is changed. However, the second kind
of ISE (Figure lb) cannot be stable, because Ag is
solubilized continuously from the AgIAgCl electrode
into the ISM; i.e., the potential is gradually changed by
Ag concentration on the electrode. Nonetheless, the
amount of Ag solubilized from this ISE can be regarded
as nearly constant for a short period (1-2 miii), during
which the experimental
results show the observed
nernstian gradient and concentration of NaCl. On the
contrary, the potentials measured by the first kind of
ISE (Figure la) correspond to the real values of the
concentration of NaC1, because the potential of the
Ag/AgCl electrode in contact with the saturated NaCl
solution is constant.
To summarize
these mechanisms: MMO does not act
as an ion carrier, but rather acts as a deep trap for Na
in the ISM. Na cannot diffuse into the ISM because of
the stability of the MMO trap and the electrostatic force
of Cl-. However, Ag can diffuse across the ISM by
moving among the MMO traps along with Cl-. The
differences of the ion behavior between Na and Ag
depend fundamentally
on the stability of the MMO
traps. The differences in potential are generated by the
double layer of electrons at the interface between the
ISM and electrolyte layers.
These results should be useful for fabricating a stable
nernstian
1378
CLINICALCHEMISTRY,Vol.37, No. 8, i99i
and inexpensive disposable ISE for Na and also for
understanding the mechanisms of the generation of
potential on a molecular scale.
I greatly thank H. Amano, W. Ueno, and Y. Akiyoshi for their
helpful discussions, and K. Yamada and M. Nirasawa forsupplyingtheexperimental samples.
References
1. Moody GJ, Thomas JDR. Progress in designing calcium ionselective electrodes. Ion-Selective Electrode Rev 1979;1:3-.29.
2. Mack Brown H, Owen JD. Micro ion-selective electrodes for
intracellular ions. Ion-Selective Electrode Rev 1979;1:145-.85.
3. Radic N. Solid state electrode sensitive to mercury ions. IonSelective Electrode Rev’ 1989;11:177-88.
4. Curme HG, Columbus EL, Dappen GM, et al. Multilayer film
elements for clinical analysis: general concepts. Cliii Chem
1978;24:1335-42.
5. CateJC N, Hedrick R, Taylor M, McGlothlun CD. Evaluation of
an engineering modelof the “EktaChem”analyzer for glucose and
urea assay. Clin Chezn 1980;26:266-70.
6. Warren K, Kubsaik NP, Brody BB, Sine HE, D’Souza JP. The
multilayered
film analyzer glucose in serum, plasma, cerebrospinal fluid, and urine; and urea nitrogen in serum and plasma. Clin
Chem 198026:133-7.
7. Wu T-W, Zumbulyadis N, Gross 5, Gohlke R. Human conjugated bilirubin-isolation,
biosynthesis,
and molecular
character-
izations by direct spectroscopic analyses [Abstract]. Cliii Chem
1979;25:1137.
8. Lehn JM, Sauvage ,JP, Dietrich B. Cryptate cation exchanging
rate. J Am Chem Soc 1970;92:2916-8.
9. Eisenman G, Szaro 5, Ciani 5, MacLaughlin S, Krasne S. Ion
binding and ion transport produced by neutral lipid-solublemolecules. Prog Surface Membrane Sci 1973;6:139-241.
10. Cussler EL. Membranes which pump. AIChE J 1971;
17:1300-3.
11. Cuasler EL, Fennell Evans D, Matesich
MA. Theoretical and
experimental basis for a specific countertransport system in membranes. Science 1971;172:377-9.
12. Perry M, Lobel E, Bloch R. Mechanism of a polymeric valunomycin-based potassium specific electrode. J Membrane
Sci
1976;223-35.
13. Battaglia CJ, Chang JC, Daniel DS. Ion-selective electrode.
US patent 4 214 968, 1980.
14. Fujiwara M. Direct observation of ion behaviour in a membrane containing valinomycin. J Electroanal Chem 1990296:259-62.
15. Duax WL, Smith GD, Strong PD. Complexation of metal ions
by monensin. Crystal and molecular structure of hydrated and
anhydrous crystal forms of sodium monensin. J Am Chem Soc
1980;102:6725.-9.
16. Fujiwara M. The diffusion behavior of Ag ions in a polymer
membrane containing valinomycin. Chem Lett 1991;451-4.
17. James TH. The theory of the photographic process. New York:
Macmillan Publishing, 1977:8 (Table 1-5).
18.Choy EM, Fennell Evans D, Cussler EL. A selective membrane for transporting sodium ion against its concentration
gradient. J Am Chem Soc 1974;96:7085-90.
19 Tamamushi
R. Denki-Kagsku,
Tokyo: Kagaku
Dobjin,
1967:122.