Electrochemical Study of Metallic Oxides in Fused lithium Chloride-Potassium Chloride Eutectic

The oxygen electrode does not behave reversibly in LiC1-KC1 at temperatures of 400~176 and at oxide ion concentrat ions of 0.1-0.2M. The following electrode systems behave revers ibly as electrodes of the second kind: Cu/ Cu~O,O=; Pt/PtO,O=; Pd/PdO,O=; Bi/BiOC1,O=. The Ni/NiO,O= and Bi/Bi~O~,O: electrodes did not show reversible behavior. Because of the relat ively high solubilities of the metal oxides, the applicabil i ty of the metal oxide electrodes to the measurement of oxide ion act ivi ty is l imited to solutions containing oxide ion in concentrations comparable with, or greater than, that contr ibuted by dissolution of the oxides. This l imit is 10-sM or higher, depending on the system. The relat ively high solubil i ty of heavy metal oxides in this solvent is a t t r ibuted to the great complexing tendency of chloride ion. It is probable that these oxides would be much less soluble in ni t ra te and sulfate melts and thus their ut i l i ty as acidity electrodes could be extended to lower oxide concentrations. A m o l t e n a lka l i me t a l chlor ide so lvent m a y be said to be "buffered" w i th respect to the add i t ion of Lewis acids such as A1C1, because of the presence of a large excess of the w e a k l y basic chlor ide ion. On the o ther hand , such a so lven t can be r e n d e r e d basic b y the add i t ion of subs tances such as oxide ion which are more s t rong ly basic t h a n chlor ide ion. The purpose of the p re sen t i nves t iga t ion was to d e t e r m i n e the feas ib i l i ty of us ing the o x y g e n o x i d e ion e lect rode or m e t a l m e t a l ox ide -ox ide ion elect rodes for the e lec t rochemica l m e a s u r e m e n t of oxide ion ac t iv i ty in m o l t e n l i t h i u m ch lo r ide -po ta s s ium chloride eutect ic at 450 ~ A secondary objec t ive was to d e t e r m i n e the solubi l i t ies of me t a l oxides f rom emf m e a s u r e m e n t s . There is cons iderab le ev idence for the r eve r s i b i l i ty of the o x y g e n o x i d e ion electrode at t e m p e r a tu res of the order of 1000~ in va r ious mel t s cont a in ing oxyanions . The ea r ly w o r k of B a u r and E h r e n b e r g (1) on borates , silicates, etc., and of T readwe l l (2) on me ta l oxides, us ing qua r t z or por cela in as an e lec t ro ly te and s i lver as the ind ica tor electrode, had ind ica ted tha t the o x y g e n o x i d e ion was exh ib i t i ng r eve r s ib l e behavior . Cs~ki and Dietzel (3) r epor ted on emf m e a s u r e me n t s in cells i nvo l v i ng p l a t i n u m ind ica tor e lect rodes in bora te melts , b u t the i r i n t e r p r e t a t i o n of the resul ts has been cri t icized b y Flood, Fo r l and , and Motzfeld (4) . A n t i p i n (5) used a g raph i te electrode, which had b e e n anodized to m a k e it b e h a v e as an oxygen electrode, as a r e fe rence electrode in cryol i te at 1050~ The inves t iga t ions of Lux (6) l a rge ly in su l fa te and phospha te sys tems at 950~ were the first which c lear ly showed the feas ib i l i ty of oxide ion ac t iv i ty m e a s u r e m e n t s us ing a p l a t i n u m elect rode wi th oxygen gas. S imi la r m e a s u r e m e n t s were m a d e by Flood and co -worke r s (4, 7, 8) who d e m o n s t ra ted tha t the dependence of the po ten t i a l on oxide ion ac t iv i ty obeyed the Nerns t equat ion . Work at lower t e m p e r a t u r e s has been m u c h more l imi ted in scope. Rose, Davis, a nd E l l i n g h a m (9) made m e a s u r e m e n t s i nvo l v i ng oxygen electrodes in m o l t e n NaOH con ta in ing smal l concen t ra t ions of t in or lead at 4000-700 ~ and f ound tha t the cells b e haved revers ib ly . Hi l l and co -worke r s (10) meas u r ed emfs of cells of the type M / M O / C a O in Li~SO4-K~SOJPt,O~ [A] 706 JOURNAL OF THE ELECTROCHEMICAL SOCIETY A u g u s t 1 9 6 0 at 550~176 and concluded tha t for iron, copper, and (at t e m p e r a t u r e s be low 658~ nickel , the cell emf agreed wi th tha t ca lcu la ted f rom the free e n ergy of fo rma t ion of the me ta l oxide. Since both of the electrodes used by Hill, e t al . theore t i ca l ly r e sponded equa l l y to oxide ion act ivi ty , the cell emf data did no t show w h e t h e r the potent ia l s of the two types of oxide ion electrodes i n d i v i d u a l l y va r ied p roper ly w i t h oxide ion act ivi ty . Selis and c o w o r k ers (11) in i n t e r p r e t i n g the ga lvan ic behav io r of a Mg-Ni cell in mois t l i t h ium ch lo r ide -po ta s s ium chloride eutectic, cons idered tha t the p o t e n t i a l d e t e r m i n i n g sys tems were Mg/MgO,O = and e i ther N i / NiO,O= (at t e m p e r a t u r e s of 445~176 or H~, N i / OH-,O = (at t e m p e r a t u r e s of 390~176 Once again, the i n d i v i d u a l e lect rode behav io r aga ins t a re fe rence e lect rode tha t was i n v a r i a n t w i th oxide ion ac t iv i ty was not de te rmined . In the p re sen t inves t iga t ion , a p l a t i n u m ( I I ) p l a t i n u m electrode (12) was used as a re fe rence for the m e a s u r e m e n t of the po ten t ia l s of the o x y g e n o x i d e ion e lect rode and of va r ious m e t a l m e t a l oxideoxide ion electrodes as a func t ion of the concen t r a t ion of oxide ion. P rev ious a t t empts to measu re revers ib le oxygen potent ia ls , us ing a p l a t i n u m foil or g raph i t e rod as an ind ica tor electrode, had been unsuccess fu l (13). In the p re sen t s tudy, a porous g raph i te e lect rode was used, aga in wi th unsuccess fu l resul ts , as wi l l be seen below. For an e lect rode of the second kind, composed of a me t a l M, its spa r ing ly soluble oxide MO~j2, and free oxide ion, the Nerns t equa t ion is g iven by E = E~ . . . . ~R T / n F In a~,+ [ 1 ] = E~ + R T / n F In (K , ,~ /ao : ) "j~ [2] : E~ M + R T / 2 F In Ka,, R T / 2 F In ao = [3] where E~ ~ is the s t a n d a r d electrode po ten t i a l 1 for the me ta l i o n m e t a l electrode, and Ko~, = a~ ~+~ln ao =, is the ac t iv i ty p roduc t of the spa r ing ly soluble oxide. The first two t e rms on the r igh t h a n d side of Eq. [3] can be combined to give the s t anda rd po ten t i a l of the electrode of the second kind, or E : E~ . . . . R T / 2 F In ao: [4] which is the Nerns t equa t ion for the electrode r e -