Enhanced Metal Recovery in Fluidized Bed Electrodes with a Fin-Type Current Feeder

(1)

2344

J. Electrochem. Soc., Vol. 138, No. 8, August 1991 9

The Electrochemical Society,

Inc.

7 0 , I 6 0 5O v 30 20 10 t i i i i - 9 0 0 - 8 0 0 - 7 0 0 - 6 0 0 - 5 0 0 - 4 0 0 - 3 0 0 Potential E ( m V )

Fig. 5. Warburg coefficient-potential curve for preoxidized Ni in fused Na2S04 at 1200 K.

irreversible process, % a n d Ro/s can be expressed as fol- lowing R T % - [ 2 ]

n2F2~ 2~o~

Ro/s

--R T n2FaC~ok~h~ exp (-~<~)

where C~o a n d Do are the b u l k concentration a n d the diffusion coefficient of $20~- in the melt, respectively, (~ is the cathodic transfer coefficient, k~h is the rate constant of the cathodic reaction, a n d ~b = (nF/RT)(E - Eo). R, T, n, a n d F have their usual meanings. Since the salt melt was not ex- posed to the SO2-O2 gas m i x t u r e u n t i l the e x p e r i m e n t started, the b u l k concentration of $20~- (dissolved SO3) was low at first. As the i m p e d a n c e m e a s u r e m e n t s pro- ceeded with increasing polarization, the dissolution of SO3 c o n t i n u e d a n d the b u l k concentration of $20~- gradually increased. Thus, instead of a constant value, a decreasing % was observed. The c h a n g i n g b u l k concentration of $20~-, coupled with the overpotential term exp (-~(h), also

caused Ro/s to decrease as the cathodic polarization increased, u n t i l the reaction became diffusion limited.

Conclusions

Double layer capacitance in a fused Na2SO4 film on a preoxidized Ni electrode was measured by electrochemical i m p e d a n c e spectroscopy. A m a x i m u m capacitance was observed at a potential of -0.6 V (vs. Ag/Ag2SO , reference electrode), which, based o n the polarization resistance, diffusion impedance, a n d p o t e n t i o d y n a m i c polarization measurements, resulted from the additional c o n t r i b u t i o n of a faradaic reaction. The decrease in polarization resistance a n d diffusion i m p e d a n c e with increasing polarization potential was attributed to the overpotential effect and the t i m e - d e p e n d e n t dissolution of SO~. Application of the ac i m p e d a n c e t e c h n i q u e is ideal in this case, because the sys- t e m has b e e n thoroughly studied (9) a n d an equivalent cir- cuit u n a m b i g u o u s l y decided.

Acknowledgments

This research was sponsored b y the National Science F o u n d a t i o n u n d e r G r a n t No. DMR 8620311. Critical reading of this m a n u s c r i p t by R. A. Rapp is gratefully ac- knowledged.

Manuscript s u b m i t t e d Dec. 10, 1990; revised m a n u s c r i p t received March 11, 1991.

The Ohio State University assisted in meeting the publi- cation costs o f this article.

R E F E R E N C E S

1. I. L. Cooper, J. A. Harrison, a n d D. R. Sandbach, Elec-

trochim. Acta,

23,

527 (1978).

2. I. L. Cooper a n d J. A. Harrison, ibid., 29, 1147 a n d 1165 (1984).

3. W. Wilhelmsen a n d T. Hurlen, ibid., 32, 95 (1987). 4. C. Clerc a n d D. Landolt, ibid., 33, 859 (1988).

5. M. Weidenauer a n d K. G. Weft, Ber. Bunsenges. Phys.

Chem., 92, 1368 (1988).

6. A. D. Graves a n d D. I n m a n , J. Electroanal. Chem., 25, 357 (1970).

7. K. R. Painter, P. Ballone, M. P. Tosi, P. J. Grout, a n d N. H. March, Surf. Sci., 133, 89 (1963).

8. H. Shih a n d H. W. Pickering, This Journal, 134, 1943 (1987).

9. Y. M. Wu a n d R. A. Rapp, This Journal, Accepted for publication.

10. M. Sluyters-Rehbach a n d J. H. Sluyters, in "Electro- analytical Chemistry," Vol. 4, A . J . Bard, Editor, Marcel Dekker, New York (1970).

Enhanced Metal Recovery in Fluidized Bed Electrodes with a

Fin-Type Current Feeder

**Shi-Chern Yen* and Ching-Yih Yao**

Department o f Chemical Engineering, National T a i w a n University, Taipei, Taiwan, China

ABSTRACT

A rectangular side-by-side fluidized bed electrode for metal recovery from dilute copper solutions was investigated. Because the contacts b e t w e e n fluidized particles and current feeders are particularly important, fin-type c u r r e n t feeders were used to e n h a n c e the deposition rate. Polarization curves with a fin-type c u r r e n t feeder were measured at various Rey- nolds n u m b e r s . It is shown in the experiments that fin-type current feeders provide higher deposition rates for the same electrode potential t h a n planar current feeders in fluidized bed electrodes.

Fluidized bed electrodes have been proposed since 1966 (1, 2). The conductive particles as electrodes are fluidized i n the cell by flows of electrolytes, a n d electrochemical reactions proceed on electrode surfaces by applying electrical voltage to the cell. The fluidized bed electrode could result in high productivity, even in the condition of low current densities d u e to its high specific surface area. Be- cause of the particle m o t i o n in fluidized bed electrodes, ag-

* Electrochemical Society Active Member.

glomeration due to metal deposition is avoided, u n l i k e in packed bed electrodes (3, 4), a n d this also allows continu- ous removal of metals by i n t r o d u c i n g small particles at the top of the bed, with the withdrawal of grown particles at the bottom. Several potential applications have been studied (5-8), b u t fluidized bed electrodes are especially i m p o r t a n t in metal recovery where the m a i n application can be expected (9-12).

Many f u n d a m e n t a l investigations have been carried out a n d have concluded that the electrochemical behavior of

(2)

J. Electrochem. Soc., Vol. 138, No. 8, August 1991

9 The Electrochemical Society,

Inc.

2 3 4 5 fluidized bed electrodes is affected by bed expansion, bed

height or bed thickness, current density, and the reactant concentration in the electrolyte (10, 11, 13, 14), and a m o n g them, bed e x p a n s i o n is a particularly i m p o r t a n t factor. It has b e e n found that in the direction normal to the current feeder, the active regions are near the current feeder as well as the m e m b r a n e , and t h e r e exists an anodic region in w h i c h the potential of c o n d u c t i v e particles is higher than that of the electrolyte phase, and the region will be enlarged by increasing bed e x p a n s i o n s (13). This p h e n o m e - n o n arises from discontinuity of the particle phase, in w h i c h it cannot be regarded as equipotential any more.

As a matter of fact, effective solid conductivity is the key point of e l e c t r o c h e m i c a l behavior in fluidized bed electrodes. Because particulate electrodes contact with each other instantaneously, the order of m a g n i t u d e of the effective c o n d u c t i v i t y is the same as that of the electrolyte phase (15, 16). Theoretical estimation of effective conducti- vities is based on w h a t m e c h a n i s m of charge transfer is as- sumed. There are a few m e c h a n i s m s proposed, and two of the m o s t i m p o r t a n t are described briefly as follows:

1. Collision m e c h a n i s m (17, 18).--The electrical double layers of particles are charged while contacted with current feeders or other particles w h i c h carry different charges, and are t h e n m o v e d to other positions for being discharged by either sharing with other particles or by e l e c t r o c h e m i c a l reactions.

2. C o n d u c t i v e m e c h a n i s m (15, 19).--Aggregated particles in the fluidized bed contact with others to form chains w h i c h m a y t o u c h either the current feeder or other chains, and charges will pass t h r o u g h the chains by electronic conduction. Because the chain exists for a short time pe- riod and t h e n is destroyed and reformed as another particle chain, potential fluctuation has been observed (20). The aggregated particle chains can be classified by w h e t h e r or not t h e y t o u c h current feeders. The chains touching current feeders are called " m o n o p o l a r , " and those not touching current feeders are called "bipolar." Either reduction or oxidation reactions can occur on the surface of a monopolar particle, but both of t h e m occur on the opposite side surfaces of bipolar particles at the same time, respectively. L e e et al. (21) h a v e m e a s u r e d the bipolarity in fluidized b e d electrodes. S a b a c k y and Evans (15), and H u h and E v a n s (22) h a v e c o n c l u d e d t h a t t h i s c o n d u c t i v e mecha- n i s m is the m o s t probable in their e x p e r i m e n t a l investigations.

In our investigations the m e c h a n i s m of charge transfer is c o n s i d e r e d as a c o n d u c t i v e m e c h a n i s m , and we tried to r e m o v e the limitation of bed e x p a n s i o n in the applications of fluidized bed electrodes by the appropriate design of the current feeder, for e x a m p l e using a fin-type current feeder. Macroscopic p e r f o r m a n c e s of the fluidized bed electrodes with a fin-type current feeder have been carried out by m e a s u r i n g their polarization curves. T h e cell p e r f o r m a n c e of cupric ion r e m o v a l from dilute solutions in fluidized bed electrodes was also investigated by circulating the electrolytes t h r o u g h the fluidized cell. Concentration of cupric ion in the reservoir was m e a s u r e d with the deposition time. It was found in our m e a s u r e m e n t s that the fluidized electrolytic cell with a fin-type current feeder can offer higher current densities and give broader allowance of bed expansion.

Experimental

E x p e r i m e n t a l a p p a r a t u s . ~ T h e rectangular side-by-side

fluidized bed cell is depicted in Fig. 1. The cell is separated into u p p e r and lower parts. The lower part is the electrolyte entrance and flow d e v e l o p i n g region, which consists of glass beads of 1.5 m m diam. The u p p e r part is the region of the electrolytic cell and is separated into two sections by an anion e x c h a n g e m e m b r a n e (Asahi Glass Company). The left side is the cathodic part, containing a current feeder with copper particles ( G r e d m a n n Company), sieved into the size range of 425-600 ~m as cathodes. The average d i a m e t e r of c o p p e r particles is 512 ~m and the unfluidized bed porosity is 0.688. The right side is the anodic section w h i c h is packed with platinized titanium screens, where o x y g e n evolution occurs. The length of both u p p e r and lower parts is 22.5 cm, with 3.8 c m in side width, and 2 c m

9 10

I

5 ~2

I

1: c a t h o d i c z o n e 2: a n o d i c z o n e 3: d i s t r i b u t o r 4: a n i o n e x c h a n g e m e m b r a n e 5,6: c u r r e n t f e e d e r s 7: c a t h o l y t e i n l e t 8: a n o ] y t e ~alet 9,10: e l e c t r o l y t e o u t l e t

Fig. 1. The fluidized bed cell.

in depth. Anolytes and catholytes were p u m p e d into the cell at the l o w e r parts respectively from b o t h reservoirs, and the anolytes contained sodium sulfate 0.8M. The catholytes consisted of cupric sulfate in the range of several mM, as well as supporting electrolytes of sodium sulfate 0.8M, and buffers to control pH in a range b e t w e e n 3 and 4. The solution c o n d u c t i v i t y is 0.065 (ll cm) -1. All c h e m i c a l s m a d e b y M e r c k C o m p a n y are analytical grades.

A fin-type current feeder m a d e of titanium is shown as Fig. 2. T h r e e different current feeders were used in the experiments. Two of t h e m were fin types, and the length L of fins was 0.7 and 1.2 cm respectively; they were respectively called short-fin and long-fin current feeders for con- venience. T h e other one was a planar current feeder. Each of the fin-type current feeders had five fins with a thickness of 0.9 mm. Because the fluidized bed length in our experiments n e v e r e x c e e d e d 5.5 cm, about half the surface area of the current feeder was m a s k e d to m i n i m i z e the area e x p o s e d directly to the electrolytes. A copper wire, func- tioning as the reference electrode, was placed above particle beds near the m e m b r a n e . A potentiostat/galvanostat, HA320 from H o k u t o Penko, Limited, was used to supply electric p o w e r and to control potentials b e t w e e n w o r k i n g and reference electrodes.

E x p e r i m e n t a l procedures.--Steady-state polarization

curves of the fiuidized bed electrodes with t h r e e types of current feeders w e r e m e a s u r e d at various Reynolds num- bers (Re = vdp/~), w h i c h is based on the particle's diameter dp, fluid superficial velocity v, and kinematic viscosity v. The electrolytes were circulated from a reservoir of 40 liters, in w h i c h the variation of the concentration can be ne-

/ 3.8cm -~ - ' ~ .',1, " \ ~ " " ' ~ 0 . 8 4 c m 10cm I O.09cm

1

e l e c t r o l y t e f l o w

(3)

2 3 4 6

J. Electrochem. Sac., Vol. 138, No. 8, August 1991 9

The Electrochemical Society, Inc. glected. The circulating operation was carried out at a con-

trolled potential, and concentration of cupric ion in the reservoir was d e t e r m i n e d by a s p e c t r o p h o t o m e t e r (Miltin- roy Company, Spectronic 601). Wavelength was set on 810 nm. Copper particles w e r e treated with acetone and di- luted nitric acid before being used, and the catholytes in the reservoir were purged by high-purity nitrogen before and during the experiments.

Results and Discussion

P o l a r i z a t i o n curves.--Polarization curves of fluidized bed electrodes with the planar current feeder measured at various Reynolds n u m b e r s (Re = vdp/v) are shown in Fig. 3. The current densities calculated in all figures are based on the e x p o s e d surface area of the planar current feeder, 12.16 cm ~ (3.2 • 3.8 cm), w h i c h is covered by copper particles before being fluidized (the initial height of bed: 3.2 cm; width: 3.8 cm). The concentration of cupric ion used in the e x p e r i m e n t s was 0.02M. The parameter e, appearing in Fig. 3, represents the porosity of the fluidized bed and the unfluidized bed porosity was m e a s u r e d at 0.668. As s h o w n in Fig. 3, the current densities with the planar current feeder increase with the applied potentials. There is no limiting-current plateau observed, i.e., the cur- rent flows in the fluidized bed electrodes at the range of applied voltages are not controlled by mass transfer. The same result was also obtained by F l e i s c h m a n n and Kelsall (13), who e x p l a i n e d that n o n u n i f o r m potential distribution arose from the low effective conductivity of fluidized bed electrodes. Thus, polarization curves were almost linear in this situation. A r e m a r k a b l e result can be seen in Fig. 3 in that the current densities of polarization curves do not rise as increasing Reynolds n u m b e r s (Re = v d J , ) , which can be related to bed expansions (i.e., the ratio of increased bed length by fiuidization to the initial bed length). The same result is also observed in Fig. 4, w h i c h shows polarization curves of fluidized b e d electrodes with the short-fin current feeder. B u t as shown in Fig. 5, the current densities of polarization curves with the long-fin current feeder are raised higher as increasing Reynolds n u m b e r s or bed expansions. It has to be pointed out here that it can further be operated at bed expansions higher than 48%. It is found that the long-fin current feeder offers the best performance.

Effects o f bed e x p a n s i o n . - - I t can be found in Fig. 3-5 that the long-fin current feeder gives higher current densities at the same applied potential, and the short-fin current feeder does not offer obvious e n h a n c e m e n t in current densities at the same condition. Figure 6 shows current densities vs. bed expansions for these t h r e e current feeders at an applied voltage of -0.8 V, and there is a significant difference

15 0 0 10 ga @ 5 rj 0 0.0 Re b e d o eymbol ( l - e ) e x p a n s i o n ( ~ ) / o 2 3 9 , 9 " r 4 0 21 / 6 8 4 2 [] 49 2 7 0 / / t I t I t I 0.5 1.0 1.5 P o t e n t i a l , ( - v o l t s )

Fig. 4. Polarization curves with the short-fin current feeder.

of deposition rates b e t w e e n the long-fin current feeder and the other two. As s h o w n in Fig. 6, the current densities decrease as bed e x p a n s i o n s increase by using the planar and short-fin current feeders, but the current density by the short-fin current feeder is higher than that by the planar current feeder, and asymptotic current densities seem to be reached at higher bed expansions. These asymptotic current densities s e e m to be c o n t r i b u t e d mostly from the particles near the surface of the current feeders, because the ratio of surface area of c u r r e n t feeders, short-fin to planar, is 2.47, and the ratio of a s y m p t o t i c current densities short-fin to planar, is 1.5. F u r t h e r m o r e , our e x p e r i m e n t a l results (23) indicated that the limiting current densities of a fluidized cell w i t h inert particles, w h i c h using the planar and short-fin current feeders, were in the range of 70-100A/m 2 and 100-200A/m 2, respectively; they were close to t h e a s y m p t o t i c current densities. The current densities with the long-fin current feeder are raised as increasing b e d expansions, and t h e y are m u c h higher than those of the other two types.

The contribution of metal deposition on the current feeder itself could also be investigated. In our experimental results the c o n t r i b u t i o n of the long-fin current feeder itself to the total current w a s m e a s u r e d b y the anodic dissolution of the current feeder individually after the m e t a l deposition w a s c o m p l e t e d . It w a s f o u n d that less than 5 %

o o 8'] I=1 I.,i rj

oO'

b e d e x p a n s i o n (~) o 11 4 ~ " x 19 9 / ,~ 30 ~e J ~

+

6 6 61

/w/

D 78 74

/ f l

, 3 9 3 0 p / A 0 5 2 4 8 / / ~

i '

t I = I f I 0.5 1.0 I. P o t e n t i a l , ( - v o l t s ) Fig. 3. Polarization curves with the planar current feeder.

b e d

so

72 ~

J

~ 40 19 ~" O 51 26 / o a 24 4 9 ~ o o 18 O / j , "~-

20

9 o 10 N r ~ R~; I I r I I I 0.6 1.0 1.5 P o t e n t i a l , ( - v o l t s ) Fig. 5. Polarization curves with the long-fin current feeder.

(4)

J. Electrochem. Soc.,

V o l . 1 3 8 , N o . 8, A u g u s t 1991 9 The Electrochemical Society, Inc. 2347 20

g-

o 15 0 ffl / x / A p p l i e d voltage : - 0 . 8 volts ~,mbol C u r r e n t f e e d e r l o n g - f i n s h o r t - f i n ~ . ~ ~ p l a n a r I w I I I ~ I I 20 40 60 80 B e d E x p a n s i o n , p e r c e n t 1 0

Fig. 6. Effects of bed expansion on current density for various current feeders. 1.0J 0.8 ~ ~ . 0.8 O u 0.4 - - - 0 . 5 v o l t ~ ' - 1 . 0 v o l t ~ x , C u r r e n t f e e d e r h% O.Z o p_lanar __ " ~ v s h o r t - f i n l o n g - f i n b e d e x p a n s i o n : 5 0

0.00

t 1 J I , I flO 100 150 O p e r a t i n g t i m e , ( m i n u t e s ) Fig. 8. Effects of applied voltage on the performance of metal recov- ery (initial concentration: 350 ppm).

of the metal deposition was deposited at the surface of the long-fin c u r r e n t feeder, a n d it can be concluded that the c u r r e n t e n h a n c e d obviously b y the fin-type c u r r e n t feeder was mostly c o n t r i b u t e d b y the metal deposition on the fluidized particles.

Because bed e x p a n s i o n s can greatly affect the effective conductivity of the particle phase, which will be decreased by rising bed expansions, it can be regarded that while it behaves i n this way the charge-transfer rate in the solid phase dominates c u r r e n t flows i n fluidized bed electrodes. The other behavior is that c u r r e n t densities are increased as Reynolds n u m b e r s increase, a n d both the behavior of fluidized bed electrodes with the long-fin current feeder a n d packed bed electrodes belong to this case. This reveals that mass transfer perhaps dominates c u r r e n t flows in the fluidized bed cells with the long-fin current feeder.

Copper recovery experiments.--Since the polarization

curves reveal that a suitable fin-type current feeder will greatly enlarge the c u r r e n t densities, it is appropriate to verify the feeder i n its performance in metal recovery. The performance of a cell removing copper ions from the dilute solution with three types of current feeders is shown in Fig. 7. The potential was controlled at -1.0 V, and the bed expansions were operated at 25, 50, a n d 75% respectively. The v o l u m e of the reservoir was 10 liters, a n d 132 g of cop-

1.0 0.8 0.6 0 r ( J 0.4 0.2 0.0 0 ~ x4*~ expansion ~ 50~ 7f~ x x ,,xx plauex o $ 9 short--fin ~ a 9 , I I I t I , I , 50 100 150 200 250 O p e r a t i n g t i m e , ( m i n u t e s )

Fig. 7. Comparisons between the three current feeders in perform- ance of metal recovery (initial concentration: 350 ppm; applied poten- tial: - 1 V).

per particles were used as cathodes. The initial concentration of the ~upric ion was 350 ppm. Figure 7 shows that the fin-type current feeder actually enhances the performance of fluidized bed electrodes in metal recovery. After 100 m i n of operation with 75% bed expansion, the concentration of cupric ion in the reservoir was reduced to 10% of the initial concentration, 350 ppm, b y u s i n g the long-fin current feeder. The concentration of cupric ion was reduced to 88 a n d 76% Of the initial concentration while using the planar and short-fin current feeders, respectively. Figure 7 indicates that the e x p e r i m e n t a l curves of the same type of current feeders are very close to each other, a n d they can be classified into three groups by various c u r r e n t feeders. The concentration curves by using the long-fin c u r r e n t feeder are lowest in Fig. 7, a n d it means that the long-fin c u r r e n t feeder gives a better performance in metal recovery.

The effect of applied voltage on the performance of fluidized bed electrodes in metal recovery is depicted in Fig. 8. The concentration decrease with time b y the long- fin c u r r e n t feeder at -0.5 V is less t h a n that at -1.0 V, b u t more t h a n that by the short-fin c u r r e n t feeder at -1.0 a n d -0.5 V. Furthermore, the performance b y the short-fin current feeder at -0.5 V is better t h a n that b y the planar current feeder at -1.0 V. Because the applied voltage is low- ered by using the fin-type current feeder to achieve the same level of performance in metal recovery, the fin-type current feeders can also achieve some saving of electric energy.

Discussion.--The fin length of the long-fin current feeder

is three-fifths of the thickness of the cathodic reaction part in the cell; longer fins seem to favor current flows in fluidized bed electrodes, which have b e e n verified by our experimental results. This m a y be explained b y considering the charge transfer process as the conductive mechanism, i n which contacts b e t w e e n particles a n d c u r r e n t feeders are very important. As has been described, aggregated particles contact with current feeders to form m o n o p o l a r particles where metal electrodeposition occurs. As the possibility of contacts b e t w e e n particles a n d current feeders is raised, the ratio of monopolar particles in the fluidized bed will be increased, a n d the c u r r e n t will be enhanced. The fin-type current feeder would increase the possibility of contacts b e t w e e n fluidized particles a n d the current feeder; the deposition rate could be e n h a n c e d in this way. N o n u n i f o r m fluidization was observed at high bed expansions in our experiments, while the long-fin current feeder was used. Partial particles circulated in the bed, especially at the top of the bed. The circulation of the particles m a y have e n h a n c e d the contact frequency b e t w e e n the particles and the c u r r e n t feeder.

(5)

2348

J. Electrochem. Soc., Vol. 138, No. 8, August 1991 9 The Electrochemical Society, Inc.

Although the ratio of surface area, long-fin to planar, is 3.53, it produces higher e n h a n c e m e n t ratio corresponding to the ratio of the surface area. It is not e n o u g h to explain the current e n h a n c e m e n t caused by the long-fin current feeders, if it is based on contact frequency, especially at higher bed expansions. It m a y be explained further that reaction rates which occurred on the particle's surface are d e p e n d e n t on positions in the beds. When fins are ex- t e n d e d toward the m e m b r a n e , particles near the surface of the c u r r e n t feeder have more opportunities to become monopolar, a n d their potentials are very close to the current feeder. Like flow-by packed bed electrodes, current densities increase in the direction toward the m e m b r a n e (4),

i.e.,

the particles near the m e m b r a n e are more electrochemically active. So the deposition rates will be e n h a n c e d more, even though the surface area of the long-fin current feeder has n o t increased correspondingly. Using the short-fin current feeder has also enlarged the current, b u t with less en- hancement. It m a y be explained that less monopolar particles are formed by short-fin current feeders, a n d are still far from the m e m b r a n e , due to shorter extended fins. The long-fin c u r r e n t feeder could produce more monopolar particles, which also have more electrochemical activity t h a n the short-fin or planar current feeder. Thus, the en- h a n c e m e n t of c u r r e n t densities by the long-fin current feeder is m u c h better t h a n b y the other two types of current feeders.

Conclusion

F r o m the polarization a n d circulation experiments, it is found that fin-type current feeders could e n h a n c e the metal deposition rates i n fluidized bed electrodes. The relationship b e t w e e n current density a n d bed e x p a n s i o n is also altered b y increasing the fin length. Current densities are decreased as bed expansions are raised w h e n using planar as well as short-fin current feeders. But on the con- trary, by using the long-fin current feeder, the current density is e n h a n c e d with increasing bed expansions, and the performance of metal recovery is improved. I n addition to the e n h a n c e d metal recovery, the fluidized bed electrodes with an appropriate fin-type c u r r e n t feeder can also allow a more flexible range of bed expansions. The e n h a n c e m e n t of current feeders with fins extended can be explained by the fact that it has n o t only increased the contact probabil- ity b e t w e e n particles a n d c u r r e n t feeders, b u t has also in-

creased n u m b e r s of more electrochemically active monopolar particles in the cell.

Manuscript s u b m i t t e d Nov. 1, 1991, revised m a n u s c r i p t received Feb. 18, 1991.

National Taiwan University assisted in meeting the pub-

lication costs of this article.

R E F E R E N C E S

1. M. F l e i s c h m a n n , F. Goodridge, a n d J. R. Backhurst,

Brit. Pat.,

1,194,181 (1970).

2. F. Coeuret, P. le Goff, a n d F. Vergnes, F r e n c h Pat., 1,500,269 (1967).

3. D. N. B e n n i o n a n d J. Newman,

J. Appl. Electrochem.,

2, 113 (1972).

4. M. A. Enriquez-Granados, D. Hutin, and A. Storck,

Electrochim. Acta,

27, 303 (1982).

5. T. Berent, I. Fells, a n d R. Mason,

Nature,

223, 1054 (1969).

6. J. R. Backhurst, F. Goodridge, R. E. Plimley, a n d M. F l e i s c h m a n n ,

ibid.,

221, 55 (1969).

7. C. Oloman a n d A. P. Watkinson,

Can. J. Chem. Eng.,

53, 268 (1975).

8. F. Goodridge, C. J. H. King, a n d A. R. Wright,

Electro-

chim. Acta,

22, 1087 (1977).

9. F. Goodridge a n d C. J. Vance,

ibid.,

24, 1237 (1979). 10. F. Coeuret,

J. Appl. Etectrochem.,

10, 687 (1980). 11. S. G e r m a i n a n d F. Goodridge,

Electrochim. Acta,

21,

545 (1976).

12. R. L. Leroy,

ibid.,

23, 815 (1978).

13. D. H u t i n a n d F. Coeuret,

J. Appl. Electrochem.,

7, 463 (1977).

14. M. F l e i s c h m a n n a n d G. H. Kelsall,

ibid.,

14, 269 (1984). 15. B. J. Sabacky a n d J. W. Evans,

Metall. Trans.,

8B, 5

(1977).

16. F. Goodridge,

Electrochim. Acta,

22, 929 (1977). 17. M. F l e i s c h m a n n a n d J.W. Oldfield,

J. Electroanal.

Chem.,

29, 211 (1971).

18. M. F l e i s c h m a n n a n d J. W. Oldfield,

ibid.,

29, 231 (1971). 19. R. E. Plimley a n d A. R. Wright,

Chem. Eng. Sci.,

39, 395

(1984).

20. T. H u h a n d J. W. Evans,

This Journal,

134, 308 (1987). 21. J. K. Lee, L. W. Shemilt, a n d H. S. Chun, J.

Appl. Elec-

trochem,

19, 877 (1989).

22. T. H u h a n d J. W. Evans,

This Journal,

134, 317 (1987). 23. S. C. Yen a n d C. Y. Yao, U n p u b l i s h e d results.

24. K. Kusakabe, S. Morooka, a n d Y. Kato,

J. Chem. Eng.

Jpn.,

14, 208 (1981).

An Investigation into Membrane Potentials Generated Across

Mullite in Mixtures of Molten Alkali Metal Bromides

M. L. Orfield*

Department of Chemistry, University of Wisconsin-Stout, Menomonie, Wisconsin 54751

ABSTRACT

Mullite functions as an ionic c o n d u c t o r of alkali metal cations at high temperatures. The m e m b r a n e potentials of two mullite m e m b r a n e s , MV30 a n d MV33, were measured in mixtures of the alkali metal cations. These potentials were predicted by the equations used to predict m e m b r a n e potentials in aqueous systems. The potential selectivity coefficient of MV30 for Na/K, Li/K, a n d Na/Li were calculated a n d the potential selectivity coefficient of MV33 for Na/K was evaluted. The range of composition over which the behavior of each mullite m a y be predicted was discussed.

Glassy m e m b r a n e s , sensitive to a single ion, form the basis for m a n y separators a n d ion selective electrodes in use i n aqueous systems. The success of such electrodes de- pends u p o n a n u n d e r s t a n d i n g of the relationship b e t w e e n the composition of the solution a n d the potential generated across the m e m b r a n e . I n aqueous systems this relationship has b e e n thoroughly Examined and exploited to prepare glasses selective for a variety of ions.

* Electrochemical Society Active Member.

Since glassy m e m b r a n e s are k n o w n to c o n d u c t ions well in m o l t e n salt solutions, it seems likely that they m a y obey the same k i n d of mathematical relationships applicable in aqueous systems. A few investigators have explored this area. Doremus measured the potentials across silica a n d Pyrex glasses in m o l t e n nitrates (1, 2). Notz a n d K e e n a n also measured potentials across Pyrex in m o l t e n nitrates (3, 4). v a n R e e n a n

et al.

a n d Stern investigated potentials across silica i n m o l t e n chlorides to a limited extent (5-7). However potentials across these glasses could not be