- Email: [email protected]
Determination of silver ions in solution with a glass electrode
JOURNAL OF ELECTROANALYTICAL
DETERMINATION
OF
SILVER
GLASS
IONS
CHEMISTRY
IN
35
SOLUTION
WITH
A
ELECTRODE*
ALLAN
(Received
L. BUDD
April
26th.
1962)
INTRODUCTION Since the discovery of the hydrogen ion function at glass membranes by CREMER~, has been a subject HABER AND KLEMENSIEWICZ~ and BORELIUS 3, the glass electrode of curiosity_ Its application in the measurement of pH is well known One of the problems encountered with the glass electrode in pH measurements is the sodium erroI-_ HOROWITZ~ and SCHILLER~ studied the response of glass electrodes to other ions, the relationship besuch as sodinm and potassium_ LENGYEL AND BLuM~ studied tween composition of the glass and response and concluded that the oxides of aluminum and boron enhance the sodium function of glass electrodes. EISENMAN, RUDIN AT;ID CASBY~ estended the work of LENGYEL AND BLUM and produced sodium aluminum silicate glasses with high alumina content and very satisfactory sodium function_ Their IS M per-cent alumina glass was about 250 times more sensitive to sodium than potassium_ LEO~ARD~ reported on two glass electrodes; one a sodium aluminum silicate and the other a lithium aluminum silicate_ The former was found to be equally responsive to sodium and potassium, while the latter showed a IOOO to I selectivity for sodium over potassiumNone of the above workers was able to find a glass which was specific for any metal ion over hydrogen. As a result, it was found necessary to raise the pH of a solution if the cation
function
of the glass
electrode
was
to be observed.
EXPERIMENTAL Electrodes used 111 this study were commercially available glass electrodes (Beckman Electrode No_ 39275 of lithium aluminum.silicate composition and No- 3gI37 of sodium al uminum silicate composition) _ A pH meter with a full scale sensitivity of 1400,700, 200, or IOO mV (Beckman No. -76000) was used for all cation measurements_ All pH measurements were made with a type E-z glass-electrode to minimize cation error. Analytical reagent grade chemicals were used in all cases. RESULTS The l
electrodes
were
tested
for their
response
A Paper Presented at the 13th Pittsburgh~$Ionference
Spectroscopy,
March,
to a large
number
on -Analytical
of cations Chemise
and
and
the
~Xp&ied
rg62J_ Eleclroanal.
Chem..
-=j (1963)
:
35-39
36
A_ I__BUDD
results aregiveninTableI_Inallcaseswhereresponsewasobserved,itwas
mately
55 mV
multivalent
per decade
change
in concentratiori.
No
approxi-
response
was
observed
for
ions. TABLE1 RESPONSE
OF
THE
TWO
Ion
GLASS
No-
ElECTRODE No.
39137
H+ Naf *+
+
IA+ GL’f
i
MgZ* Ba"+ Cuf
-
TO
VARIOUS
IONS
39’76
+ + -
T
-
-
Tl+ *g+
+
NH_,+
+
+
+Responds - Does not respond The new andinteresting periments were performed
observationisthe to determine how
response to silver ions and further exsensitive these electrodes are for silver.
Electrode ~0.3g137 showedno great selectivity for any oftheions excepthydrogen_ The silver ion was favored over other monovalent metal cations. The electrode was about rsootimesmore sensitive to hydrogenion and 25 times more sensitive to silverthanto otherions- The order of sensitivity for various ions was H+ s Agf > K+, NHA+
>
Na+
>
Li+.
pK=2 pNa vaned
0 mV
_
-!a
-100 -150
1
2
3
4
5
,(ion)
Fig. I- Response
of electrode 39278
relativeto (Na+).
to
(Ag*)
Fig_ 2. Response
of electrode 39278 relative to (K+)-
to (Na*)
Electrode No_ jg278 had some ve+ interesting selectivity properties. In Fig_ I, potentials aregivenfor~mixt&es of~silverandsodiumions asoneis variedrelativeto the:other. It-can be se&rthat_even a Iooo-fold excess of sodium over si1ver.hasver-y little;effectmon.the-reqknse of-the electrode to silver. Similar data for mixtures of stidium &id potassmm
iGnsSare$Gen-in Fig, z and reveal that potassium has no measur-m
DETERMINATION
OF
SILVER
IONS
37
ableeffectonthesodIummeasurement.Likewise,potassiumhasnoeffectonthesilver measurement (Fig. 3)_ Apparently,-thelithimn aluminumsilicate electrodeis moresensitiveto silverions, thanto hydrogenions. It canbe seenin Fig. 4thatthe hydrogenion seems to have_ muchless effect onthesilverresponsethan‘silverhas onthehy&rogenresponse_ The indicated change in response, for a given silver concentration with chang&g pH. isshowninFig.5.WhenthepHis oneunit_belowpAgthereisapparentlynohydrogen ion error_
250
7
mV 200
200
mV 150
0
1
-
2
3
4
5-
1
3
2
4
p(ion)
p(ion) Fig. 3_ Response of electrode 39278 relative to (K+).
2
Fig_ 4. Response of electrode 39278 relative to (H+)).
to (Ag+)
3
4
5
6
to
(Ag+)
7
PH Fig_5_Responseofelectrode39278to
(Ag+)withchangingpH.
DISCUSSION
The mechanism of the glass electrode process is complex and, as yet, not fully understood_ It may be assumed that the following factors influence the response for agivenion: (I) the size of the ion relative to the size of the "electron rich holes" in the glass membrane_ Lewis acidity oftheion p~olarizability of the ion. energy of election transfer (proportionaltothe E~.oftheion-metalequilibriu~)_ (4) relative population of the ions, i_e..concentration. (5)
tz
A_ I-
38
BUDD
Thedegreetowhichtheelectroderespondsto oneioninthepresenceof anotherisprobablybaseduponabilityoftheionto competefortheaciivesitesontheelectrode surfaceInthecaseofpHglasselectrodes,thesizeofthe "holes" in thenglass membrane is suchthatonlyhydrogenionscanfit_ As aresult,theseelectrodesrespondto otherions only at high pH values_Thisresponse for sodium, often calledsodium error,isthen dueprimarilytosodiumions occupy-ingactivesites. A ros-role-foldexcess ofsodium ions isrequired, however, to compete with the hydrogen ion. Ifthe active sites,or the "holes", are madelarger,then otherions canmore easily occupy the sites.This appears to be the case with electrodes with increased cation functions If the above considerations are true, it is logical that an electrode which responds to sodium and/or potassium should also respond to silver and, in fact, probably be moreresponsivetosilverth~theothertwo.Conductancedatagshowthattheslzeof the silverionisinthe same range as sodium and potassium_ The available sites for covalentbondingmake_silver abetter Lewis acid than sodium or potassium; silveris morepolarizablethaneithersodiumorpotassiumandhasan EO of-o_7ggrns.N.H.E., whereas sodium andpotassium arehigh onthepositiveside l'J_Allthesefactorstendto favorsiZveroversodium~dpotassium_Themostimportantfactoristhenatureofthe glasssurface. Ionicproperties such as size,Lewisacidity,andpolarizability,willhave greaterorlessereffect, depending onthe natureofthe glass The datapresentedhere cert~ysupportthetheoreticalinferencethatsilverionwillbefavoredoversodium and potassium_ CONCLUSIOFiS
Themostimportantfacttoariseoutofthisstudyistheobservationthataglasselectrode canrespondin aNemstianmannerto changesin silverconcentrationinthepresence of all otherions.Such anelectrodehas manypracticalanalytical applicationsinthe areaofargentometric determinations. Theglass electrodeisnotpoisonedby cyanide, sulfide or anyothersubstance except fluoride and concentrated alkali_Figure 6is a typical titration curve using the 39278 lithium aluminum silicateglass electrode as the sensor. RIoreworkmustbedonebeforethemechaIllsmoftheresponseoftheglasselectrode can be fully explained. The present work suggests that an adsorption and electron transfertype of equilibrium controls the electrode response-
50 -25 -
I
I 0 Fig_
IO6_ Typi&al
20
~36
ml-
:40~-
Agy
~50
solution
tiix+~oti~curve: J_ EZedroanaZ_
Chzem..
5 (1963)
35+3cj
DETE_RMINATION
OF
SILVER
IONSy
39
SUMMARY
Two glass electrodes were found $0 respond to changes in silver ion‘concentration. One of the electrodes iS somewhat more sensiti%reto silver than to hydrogen ions, andis highly selective for the silver ion over other cations.Ar_eGonse-of 55 mV per decadeis~observed atzs". Theoreticalimplications andapplications are discussed. REFtiRENCEs 1 M. CREMER,~. 2. F. HABERXND 3 G. BORELIUS. a-K_ s H_
Biol.. 47 (1906) 562_ Z_~EJIENSIE\~CZ.Z.P~~~~R_C~~~-, Ann.
Pkysik_, 2. Pkysik.. Am=. Physik..
45 [4] (1914) 15 (197-3) 38g_ 74 [4] (191-4)
67 (Igog)
385-
gzg_
HOROWITZ, SCHILLER; 105. 6 B_ LENGYEL~AND E_ BLUM, Trans_ Fuvaday Sot., 30 (1934) 461~ 7 G- EISE,~BI~, D_ 0. RUDIN AxD J_ U_ CASE&,SC~~TZC~. 126 (1957) 831~ 8 J_ E. LEONARD, Paper presented a& the Instrument Society of America. 5th I~zstrunzental Methods Term, Beckman Reprint R-6146. of Analysis Symposizmnz, May 1959. Housto~r. 9 H. HARNED A&G B. B. OWEN. Physicab Chemistry ofEZer;troZytic SoZzctims. 2nd ed.. Reinhold, NewYork, N-Y_. Ig5o,p_53S_ 10 XV. L-x-rInuzR. Ike Oxidation States of the EZenzemts and their Poten:iaZs in Aqrreotrs SoZutioHs. 2nd ed., Prentice-Hall, Englewood Cliffs,N. J-, 1952, pp. 34-343. J.
EZectvoamaZ.
Chem.,
5 (1963)
35-39
DETERMINATION
OF
SILVER
GLASS
IONS
CHEMISTRY
IN
35
SOLUTION
WITH
A
ELECTRODE*
ALLAN
(Received
L. BUDD
April
26th.
1962)
INTRODUCTION Since the discovery of the hydrogen ion function at glass membranes by CREMER~, has been a subject HABER AND KLEMENSIEWICZ~ and BORELIUS 3, the glass electrode of curiosity_ Its application in the measurement of pH is well known One of the problems encountered with the glass electrode in pH measurements is the sodium erroI-_ HOROWITZ~ and SCHILLER~ studied the response of glass electrodes to other ions, the relationship besuch as sodinm and potassium_ LENGYEL AND BLuM~ studied tween composition of the glass and response and concluded that the oxides of aluminum and boron enhance the sodium function of glass electrodes. EISENMAN, RUDIN AT;ID CASBY~ estended the work of LENGYEL AND BLUM and produced sodium aluminum silicate glasses with high alumina content and very satisfactory sodium function_ Their IS M per-cent alumina glass was about 250 times more sensitive to sodium than potassium_ LEO~ARD~ reported on two glass electrodes; one a sodium aluminum silicate and the other a lithium aluminum silicate_ The former was found to be equally responsive to sodium and potassium, while the latter showed a IOOO to I selectivity for sodium over potassiumNone of the above workers was able to find a glass which was specific for any metal ion over hydrogen. As a result, it was found necessary to raise the pH of a solution if the cation
function
of the glass
electrode
was
to be observed.
EXPERIMENTAL Electrodes used 111 this study were commercially available glass electrodes (Beckman Electrode No_ 39275 of lithium aluminum.silicate composition and No- 3gI37 of sodium al uminum silicate composition) _ A pH meter with a full scale sensitivity of 1400,700, 200, or IOO mV (Beckman No. -76000) was used for all cation measurements_ All pH measurements were made with a type E-z glass-electrode to minimize cation error. Analytical reagent grade chemicals were used in all cases. RESULTS The l
electrodes
were
tested
for their
response
A Paper Presented at the 13th Pittsburgh~$Ionference
Spectroscopy,
March,
to a large
number
on -Analytical
of cations Chemise
and
and
the
~Xp&ied
rg62J_ Eleclroanal.
Chem..
-=j (1963)
:
35-39
36
A_ I__BUDD
results aregiveninTableI_Inallcaseswhereresponsewasobserved,itwas
mately
55 mV
multivalent
per decade
change
in concentratiori.
No
approxi-
response
was
observed
for
ions. TABLE1 RESPONSE
OF
THE
TWO
Ion
GLASS
No-
ElECTRODE No.
39137
H+ Naf *+
+
IA+ GL’f
i
MgZ* Ba"+ Cuf
-
TO
VARIOUS
IONS
39’76
+ + -
T
-
-
Tl+ *g+
+
NH_,+
+
+
+Responds - Does not respond The new andinteresting periments were performed
observationisthe to determine how
response to silver ions and further exsensitive these electrodes are for silver.
Electrode ~0.3g137 showedno great selectivity for any oftheions excepthydrogen_ The silver ion was favored over other monovalent metal cations. The electrode was about rsootimesmore sensitive to hydrogenion and 25 times more sensitive to silverthanto otherions- The order of sensitivity for various ions was H+ s Agf > K+, NHA+
>
Na+
>
Li+.
pK=2 pNa vaned
0 mV
_
-!a
-100 -150
1
2
3
4
5
,(ion)
Fig. I- Response
of electrode 39278
relativeto (Na+).
to
(Ag*)
Fig_ 2. Response
of electrode 39278 relative to (K+)-
to (Na*)
Electrode No_ jg278 had some ve+ interesting selectivity properties. In Fig_ I, potentials aregivenfor~mixt&es of~silverandsodiumions asoneis variedrelativeto the:other. It-can be se&rthat_even a Iooo-fold excess of sodium over si1ver.hasver-y little;effectmon.the-reqknse of-the electrode to silver. Similar data for mixtures of stidium &id potassmm
iGnsSare$Gen-in Fig, z and reveal that potassium has no measur-m
DETERMINATION
OF
SILVER
IONS
37
ableeffectonthesodIummeasurement.Likewise,potassiumhasnoeffectonthesilver measurement (Fig. 3)_ Apparently,-thelithimn aluminumsilicate electrodeis moresensitiveto silverions, thanto hydrogenions. It canbe seenin Fig. 4thatthe hydrogenion seems to have_ muchless effect onthesilverresponsethan‘silverhas onthehy&rogenresponse_ The indicated change in response, for a given silver concentration with chang&g pH. isshowninFig.5.WhenthepHis oneunit_belowpAgthereisapparentlynohydrogen ion error_
250
7
mV 200
200
mV 150
0
1
-
2
3
4
5-
1
3
2
4
p(ion)
p(ion) Fig. 3_ Response of electrode 39278 relative to (K+).
2
Fig_ 4. Response of electrode 39278 relative to (H+)).
to (Ag+)
3
4
5
6
to
(Ag+)
7
PH Fig_5_Responseofelectrode39278to
(Ag+)withchangingpH.
DISCUSSION
The mechanism of the glass electrode process is complex and, as yet, not fully understood_ It may be assumed that the following factors influence the response for agivenion: (I) the size of the ion relative to the size of the "electron rich holes" in the glass membrane_ Lewis acidity oftheion p~olarizability of the ion. energy of election transfer (proportionaltothe E~.oftheion-metalequilibriu~)_ (4) relative population of the ions, i_e..concentration. (5)
tz
A_ I-
38
BUDD
Thedegreetowhichtheelectroderespondsto oneioninthepresenceof anotherisprobablybaseduponabilityoftheionto competefortheaciivesitesontheelectrode surfaceInthecaseofpHglasselectrodes,thesizeofthe "holes" in thenglass membrane is suchthatonlyhydrogenionscanfit_ As aresult,theseelectrodesrespondto otherions only at high pH values_Thisresponse for sodium, often calledsodium error,isthen dueprimarilytosodiumions occupy-ingactivesites. A ros-role-foldexcess ofsodium ions isrequired, however, to compete with the hydrogen ion. Ifthe active sites,or the "holes", are madelarger,then otherions canmore easily occupy the sites.This appears to be the case with electrodes with increased cation functions If the above considerations are true, it is logical that an electrode which responds to sodium and/or potassium should also respond to silver and, in fact, probably be moreresponsivetosilverth~theothertwo.Conductancedatagshowthattheslzeof the silverionisinthe same range as sodium and potassium_ The available sites for covalentbondingmake_silver abetter Lewis acid than sodium or potassium; silveris morepolarizablethaneithersodiumorpotassiumandhasan EO of-o_7ggrns.N.H.E., whereas sodium andpotassium arehigh onthepositiveside l'J_Allthesefactorstendto favorsiZveroversodium~dpotassium_Themostimportantfactoristhenatureofthe glasssurface. Ionicproperties such as size,Lewisacidity,andpolarizability,willhave greaterorlessereffect, depending onthe natureofthe glass The datapresentedhere cert~ysupportthetheoreticalinferencethatsilverionwillbefavoredoversodium and potassium_ CONCLUSIOFiS
Themostimportantfacttoariseoutofthisstudyistheobservationthataglasselectrode canrespondin aNemstianmannerto changesin silverconcentrationinthepresence of all otherions.Such anelectrodehas manypracticalanalytical applicationsinthe areaofargentometric determinations. Theglass electrodeisnotpoisonedby cyanide, sulfide or anyothersubstance except fluoride and concentrated alkali_Figure 6is a typical titration curve using the 39278 lithium aluminum silicateglass electrode as the sensor. RIoreworkmustbedonebeforethemechaIllsmoftheresponseoftheglasselectrode can be fully explained. The present work suggests that an adsorption and electron transfertype of equilibrium controls the electrode response-
50 -25 -
I
I 0 Fig_
IO6_ Typi&al
20
~36
ml-
:40~-
Agy
~50
solution
tiix+~oti~curve: J_ EZedroanaZ_
Chzem..
5 (1963)
35+3cj
DETE_RMINATION
OF
SILVER
IONSy
39
SUMMARY
Two glass electrodes were found $0 respond to changes in silver ion‘concentration. One of the electrodes iS somewhat more sensiti%reto silver than to hydrogen ions, andis highly selective for the silver ion over other cations.Ar_eGonse-of 55 mV per decadeis~observed atzs". Theoreticalimplications andapplications are discussed. REFtiRENCEs 1 M. CREMER,~. 2. F. HABERXND 3 G. BORELIUS. a-K_ s H_
Biol.. 47 (1906) 562_ Z_~EJIENSIE\~CZ.Z.P~~~~R_C~~~-, Ann.
Pkysik_, 2. Pkysik.. Am=. Physik..
45 [4] (1914) 15 (197-3) 38g_ 74 [4] (191-4)
67 (Igog)
385-
gzg_
HOROWITZ, SCHILLER; 105. 6 B_ LENGYEL~AND E_ BLUM, Trans_ Fuvaday Sot., 30 (1934) 461~ 7 G- EISE,~BI~, D_ 0. RUDIN AxD J_ U_ CASE&,SC~~TZC~. 126 (1957) 831~ 8 J_ E. LEONARD, Paper presented a& the Instrument Society of America. 5th I~zstrunzental Methods Term, Beckman Reprint R-6146. of Analysis Symposizmnz, May 1959. Housto~r. 9 H. HARNED A&G B. B. OWEN. Physicab Chemistry ofEZer;troZytic SoZzctims. 2nd ed.. Reinhold, NewYork, N-Y_. Ig5o,p_53S_ 10 XV. L-x-rInuzR. Ike Oxidation States of the EZenzemts and their Poten:iaZs in Aqrreotrs SoZutioHs. 2nd ed., Prentice-Hall, Englewood Cliffs,N. J-, 1952, pp. 34-343. J.
EZectvoamaZ.
Chem.,
5 (1963)
35-39