JOURNAL
OF
THE
ELECTROCHEMICAL
AND
ELECTRI:3RHE:MIBAL
SOBIE:TY
SCIENCE
TECHNOLOGY
DECEMBER
,,
1979
The Electrochemical Behavior of Alkali
and Alkaline Earth Metals in Nonaqueous Battery Systems-The Solid Electrolyte Interphase Model
E. Paled*
Ir~t~tutr of Chemistry, Te~-Aviv University, Te~ Aviv, Israe~
ABSTRACT
It is suggested that in practical nonaqueous b a t t e r y systems the alkali
a n d alkaline earth metals are always covered by a surface layer which is i n s t a n t l y formed b y the reaction of the metal with the electrolyte. This
layer, which acts as an interphase b e t w e e n the metal and the solution, has
the properties of a solid electrolyte. The corrosion rate of the metal, the
m e c h a n i s m of the deposition-dissolution process, the kinetic parameters, the
q u a l i t y of the m e t a l deposit, and the half-cell potential depend on the character of the solid electrolyte interphase (SEI).
The electrochemical behavior of the alkali metals i n
nonaqueous systems was extensively studied i n the
last 15 years, m a i n l y i n connection with high energy
b a t t e r y systems. This work is summarized i n Ref.
(1-4). The electrode kinetics of the alkali metals i n
these systems were treated and analyzed following the
B u t l e r - V o l m e r equation. It was generally assumed (14) that the r a t e - d e t e r m i n i n g step (rds) is the transfer
of electrons from the metal to the ions i n the solution.
Some e x p e r i m e n t a l results indicate that this assumption
is oversimplified and the real situation is m u c h more
complex. For example, the anodic transfer coefficient
(aa) m e a s u r e d for the reaction.
M + -{- e ( m ) ~ M ~ (M ----Li, Na)
[1]
was i n m a n y cases 0.45-0.13 (1-5). I n one case, ~a,
(for reaction [1]) was found to be insensitive to the
concentration of Li + ions in the solution (7) while in
a n o t h e r case (6) it was e x t r e m e l y sensitive to the conc e n t r a t i o n of Na + in the solution. T h e most complex
behavior of these systems is d e m o n s t r a t e d in Table I.
This table contains six similar nonaqueous s y s t e m s .
However, w h e n these systems are electrolyzed, the first
three yie! d a l u m i n u m as expected from t h e r m o d y n a m i c
considerations, while the other three yield l i t h i u m or
m a g n e s i u m in a p p a r e n t contradiction to t h e r m o d y namics. Each of the first three systems is used as a
b a t h for the electroplating of a l u m i n u m . The surfaces
of the electrodes i m m e r s e d i n these systems are in
direct contact with the solution either during electrolysis or u n d e r open-circuit conditions. The last three
systems are examples of nonaqueous b a t t e r y (NAB)
systems. I n these systems the electrodeposited m e t a l
is the more reactive one.
Recently, it was concluded b y Dey (13), a n d R a u h
and B r u m m e r (14) that i n some NAB systems the lithi u m m e t a l is covered by a Li + conducting film. A more
detailed model for such a film has been described for
9 Electrochemical Society Active Member.
Key words, battery, electrolyte, corrosion.
the m a g n e s i u m electrode i n thionyl chloride solutions
(12).
I n this paper it will be emphasized that the electrode
kinetics and the deposition-dissolution mechanism of
the alkali and alkaline earth metals i n NAB systems
are entirely different from those characteristic of other
nonaqueous or aqueous electrochemical systems. A
model will be proposed aimed at explaining the complex and u n i q u e behavior of NAB systems.
The SEI modeL--It is suggested that in practical NAB
systems the alkali a n d alkaline earth m e t a l s are always
covered by a surface l a y e r at least 15-25.k thick (12).
This layer is formed i n s t a n t l y by the contact of the
metal with the solution. The thickness of this freshly
f o r m e d layer is d e t e r m i n e d by the electron t u n n e l i n g
range. This layer consists of some insoluble products
of the reaction of the m e t a l with the solution. It acts
as an interphase b e t w e e n the m e t a l a n d the solution
and has the properties of solid electrolyte, t h r o u g h
which electrons are not allowed to pass. Therefore, it is
called "Solid Electrolyte I n t e r p h a s e (SEI)."
The existence of this electronic i n s u l a t i n g layer on
the anodes i n NAB systems is the principle difference
b e t w e e n NAB and a q u e o u s - l i k e systems such as, E x amples 1-3 in Table I. It should be emphasized that
Table I. Electrolyis of nonaqueous systems
Electrodeposited
metal
Ref.
Toluene
A1
(8)
Diethyl ether
A1
(9)
THF
AI
(10)
Thionyl chloride
Propylene carbonate
Li
(11)
Li
Mg
(1)
(12)
No.
Electrolyte
1
2
A1Brz + L i B r
A1Clz + LiC1 +
LiA1H,
AICIs+ LiCI +
LiAIH,
4
5
AICIa+ L i C I
A1CI~ + LiC1
6
F e C h + MgCI~
Thionyl chloride
3
Solvent
2047
Downloaded 03 Feb 2010 to 18.53.5.219. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp
2048
J. Electrochem. Sac.: E L E C T R O C H E M I C A L S C I E N C E A N D T E C H N O L O G Y
in N A B systems (according to this suggestion) t h e r e
is n e v e r a direct a n d free contact b e t w e e n the m e t a l
a n d the solution. Thus, these b a t t e r i e s will be called
S E I batteries. The electrochemical b e h a v i o r of S E I
electrodes will be g o v e r n e d b y the p r o p e r t i e s of the
SEL The following p r o p e r t i e s of t h e SEI should be
considered: ~ts m o z p n o m g y (compact or porous, size of
the c r y s t a l s ) ; its thickness ( L ) ; the t y p e and the conc e n t r a t i o n of the l a t t i c e defects; the t r a n s f e r e n c e n u m bers of electrons (te), cationic defect ( t + ) , anionic
defect ( t _ ) ; the m o b i l i t y of these defects. The p o t e n tial difference ~M/sol contains t h r e e components (12)
A'~M/so 1 = A~bM/sE -~- A~bsE ~- A~bsE/sol
[2]
the total o v e r p o t e n t i a l contains s i m i l a r terms.
The simplest and most i m p o r t a n t case for NAB systems is a S E i electrode having t+ : 1 a n d te = 0,
i.e., the SEI is a p u r e cationic conductor. I n this case,
t h e electrode will be r e v e r s i b l e for its o w n cations
and thus will obey the Nernst law. The dissolutiondeposition process for SEI electrode having S c h o t t k y
lattice defects is s c h e m a t i c a l l y d e s c r i b e d in Fig. 1. I t
has t h r e e steps: (i) on cathodic p o l a r i z a t i o n a metallic
cation sheds its solvent molecules, crosses the s o l u t i o n /
SE i n t e r f a c e a n d enters inside a v a c a n c y in the SEI;
(ii) it m i g r a t e s t h r o u g h the b u l k of t h e S E I b y
S c h o t t k y vacancies mechanism; (iii) finally, it reaches
the metallic surface and accepts an e l e c t r o n f r o m the
m e t a l a n d becomes a m e m b e r of the metallic lattice.
On anodic p o l a r i z a t i o n t h e m e t a l l i c cation moves in
M;mlSs § o (M) Z M;j
I
+
+
+
M
+
the opposite direction. The d i s s o l u t i o n - d e p o s i t i o n p r o c ess in a real SEI electrode is m o r e c o m p l i c a t e d as some
of the following p h e n o m e n a can be involved: c r y s t a l lization; g r a i n b o u n d a r y effect; shorts t h r o u g h the S E I
d u r i n g p r o l o n g e d deposition; b r e a k d o w n and r e p a i r of
the SEI d u r i n g p r o l o n g e d dissolution; adatoms
f o r m a t i o n in the M / S E interface; and a d s o r b e d ions in
the S E / s o l u t i o n interface. Some of these p h e n o m e n a
will be discussed later.
It is possible to electrodeposit a l k a l i or a l k a l i n e
e a r t h m e t a l s (12) from NAB systems on an i n e r t c a t h ode. However, in these cases t h e m e t a l should s t a r t to
deposit o n l y a f t e r the electrode surface is c o m p l e t e l y
covered b y a passivating l a y e r w h i c h acts as a S E I
t h r o u g h which the electrons cannot pass (12).
Kinetics of the SEI etectrode.--As it was m e n t i o n e d
the most i m p o r t a n t a n d simple case is t+ --- i ( t - , Se
= 0). In this case, the rds m a y be a n y one of the
t h r e e steps d e s c r i b e d above. F o r h i g h l y conductive SEI
at s m a l l thickness the charge t r a n s f e r r e a c t i o n at one
of the SEI interfaces m a y be the rds. However, w h e n
the SEI is thick enough the m i g r a t i o n of ions t h r o u g h
it m a y be the rds. Moreover, it is k n o w n t h a t t h e ionic
c o n d u c t i v i t y of p u r e solid crystals is v e r y low at r o o m
t e m p e r a t u r e (15), a fact w h i c h s t r e n g t h e n s the last assumption. I n this case it is possible to use Eq. [3] d e scribing the m i g r a t i o n of ions in a solid c r y s t a l u n d e r
e x t e r n a l field (16). It is assumed t h a t L is l a r g e r t h a n
the space charge lengths in t h e SEI, and t h e r e is no
change in the concentration or m o b i l i t y of t h e m o b i l e
l a t t i c e defects t h r o u g h the SEI
i : 4ZFan+~ e x p ( - - W / R T ) sinh (aZFE/RT)
[3]
a : h a l f - j u m p distance, v = v i b r a t i o n f r e q u e n c y of
t h e ion in the crystal, n+ = concentration of the
l a t t i c e cationic defect, Z : t h e valance of t h e mobile
ion, W = a b a r r i e r e n e r g y for j u m p i n g , E = electric
field. T h e a s s u m p t i o n t h a t t h e m i g r a t i o n of the cation
in t h e SEI is the rds means that the l a r g e s t fraction of
the electrode o v e r p o t e n t i a l (~1) will develop on the SEI,
i.e.
-+
--
--
(~*.m)
December I979
SE
,l = E L
-
+
[4]
O n the substitution of Eq. [4] in Eq. [3], one gets
i : 4ZFan+v exp (-- W/RT)sinh(aZFTI/RTL)
MIGRATION
+
-- +
IN SE
+/
A t n : 0, the f o r w a r d and b a c k w a r d ionic c u r r e n t
t h r o u g h the SEI a r e equal and the ionic e x c h a n g e c u r r e n t d e n s i t y (io') can be defined
(~bSIE)
-- + ~ - -
+
9.~
-- +
+
-- +
io' = 2ZFayn+ exp ( - - W / R T )
F o r high field conditions (aZF~l > RTL)
like equation is o b t a i n e d
.(sol)so
+
--
+
--
+
-- +
--
§
--
+
.'/~
--
4-
--
+
-- +
SE
(A4)(SE/so I)
I
= io' exp (aZFn/RTL)
[8]
a Tafel[9]
a n d the Tafel slope is
-- +
-
[7]
a n d Eq. [5] becomes
-- +
i : 2io'sinh (aZF~/RTL)
VM+ + M+sol(SOl)n ~. MSE +
[8]
w h e r e io' is given b y (7)
: X / _ : "-:,/
-- +
q.
~o' = ~ = ~
--
+
[5]
2.3 R T L
b= - aZF
M+ (sOgln
ml
Fill. 1. A schematic description of the deposition-dissolution procam (,ee text).
[10]
T h a t means t h a t one cannot e x p e c t to g e t a sole
Tafel slope w h e n m e a s u r i n g the kinetics of the a l k a l i
a n d the a l k a l i n e e a r t h m e t a l s in N A B systems. T h e
m i n i m u m v a l u e of b can r o u g h l y be e s t i m a t e d using
Eq. [11]. F o r z = 1, a = 3A, L = 20A one gets b 400
mV
60/.,
b (mY) -~ aZ'
[U]
Downloaded 03 Feb 2010 to 18.53.5.219. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp
VoL 126, No. 12
However, Tafel slopes of several volts can be m e a s u r e d
Table II. The kinetic parameters for the Li electrode in
nonaqueous system
for a thicker SEs electrode.
For low field conditions, Eq. [5] reduces to Ohm's
law
4a2ZZF2vn+ exp ( - - W I R T )
i
=
"--
RT
L
[12]
a~o)
b (3)
(mV)
aA(~>
0.25
Or
=
--"
--
p
[13]
L
w h e r e p is the resistivity of the SEI. The reaction resistance of a n electrode is defined as
[14]
thus
Rr =
pL
[16]
The relation b e t w e e n io' a n d p is given by
RT
io'
--
-
-
2paZF
[16]
where
RT
P=
2049
NONAQUEOUS BATTERY SYSTEMS
4a2Z~F2~n+ e x p ( - - W / R T )
[17]
The reaction resistance of SEI electrode increases
l i n e a r l y with the thickness of the SEI (Eq. [15]). Thus
no sole value of Rr can be m e a s u r e d i n these NAB
systems.
A n o t h e r interesting aspect is the w a y the concent r a t i o n of the cation in the solution affects the reaction
resistance. I n the case of r e g u l a r electrode kinetics regarding a n electron transfer reaction the concentration
effect is given by (18)
io(CT) = const 9 C~+a~
[18a]
Rr = const 9 C M + - ~
[18b]
or
where aa is the anodic transfer coefficient and io is the
exchange c u r r e n t density ior the emctron charge t r a n s fer (CT) reaction.
However, i n the case of SEI electrodes two extreme
cases can be identified: (i) L > > space charge length
and the rds is m i g r a t i o n t h r o u g h the SEi. I n this case
the concentration o~ the cation i n the solution is not
expected to aftect Rr a n d the calculated a p p a r e n t aa
should be close to zero; (d) L ~ space charge length.
I n this case the concentration of the ions i n the solution m a y affect the concentration o~ the de~ects i n the
SEI a n d thus m a y affect Rr, io', and p. It can be expected that the a p p a r e n t transfer coefficient as defined
i n Eq. [18b] will be decreased to zero as L increases.
The direction of this erlect depends on the type of the
lattice demcts a n d o n the pzc of the SEL
Literature e x a m p l e s . - - T h e d o u b l e - l a y e r capacitance
in nonaqueous solutions has values of 10-30 ~F crn -2
(19). A n y nonaqueous alkali or alkaline e a r t h metal
system which shows electrode interracial capacitance
s m a l l e r t h a n 5 ~F cm -'~ should be suspected for an
anode being coated by SEt. Scarr (20) m e a s u r e d 2.4-5.2
~F cm -~ i n the L i / P C -{- LiC104 system. Values smaller
t h a n 1 ~F cm -2 were m e a s u r e d for the Mg/SOC12,
Mg 2+, a n d the Li/SOC12, Li + systems (21). The
anodic transfer coefficient ( ~ ) for the l i t h i u m electrode i n n o n a q u e o u s solution was found to be s m a l l e r
than 0.5 which m e a n s t h a t the Tafel slope (b) is larger
than 120 inV. The results are s u m m a r i z e d i n Table IL
Scarr (20) f o u n d a smaller ~a value where the "double
layer" capacity was smaller: aa was 0.25 where Cm was
2.4 ~F cm -2, a a was 0.5 where Cdl was 5 ~]F cm-~. This
can be explained b y different thicknesses of the SEI
in these two experiments. If we assume that the dielectric coefficient of the SEI is 10, t h e n i n the first
250
0.2
0.13
450
0.5
0.25
0.25
0.17-0.52 (~)
System
Ref.
Li (Hg), DMSO,
LiCI
PC, LLA1Ch
PC, LLA1Ch
PC, LiC10,
PC, LiC10~
PC, LiAICI~
TC, LiA1CI~
(22)
(7)
(5)
(20)
(20)
(1}
(23)
(1) F rom Tafel equation.
(~) From Equation [18a].
(a) Formally calculated using b = 6 0 / a = (mV).
(4) D e pe na mg o n the thicF.ness of the SEI.
e x p e r i m e n t L was about 40A and i n the second one
only 20A.
I n other cases it was found that a l i n e a r r e l a t i o n
b e t w e e n the overpotential a n d the c u r r e n t density is
continuous beyond the expected value predicted b y
the V o l m e r - B u t l e r equation. The l i t h i u m electrode i n
SOC12 shows ohmic behavior up to 600 mV (23). The
Li electrode i n pc exhibits ohmic behavior u p to 200
mV (7). The value of the Tafel slope (b) should be
at least five times larger than the highest overpotential
value on the linear plot. Thus the expected Tafel slopes
for these electrodes should be 3 and 1V (at least),
respectively.
Complications associated w i t h SEI batteries (SEI
w i t h ~ -- ~ O).--So far the simplest SEI anodes h a v i n g
t+ : 1 were discussed. These anodes are mostly desirable from a practical point of view. SEI anodes
h a v i n g t+ < 1, t - ~ 0 (re = 0) are u n d e s i r a b l e for
b a t t e r y applications and m a y exhibit a much more
complex behavior. The c u r r e n t t h r o u g h such SEI is
carried both by cationic and anionic defects. As a result, on anodic polarization (discharge of the b a t t e r y ) ,
anions will be injected from the solution into the SEI
and will finally reach the M / S E interface. I n such
systems, especially where t - > t+, this process m a y
iead to the increase of the thickness of the SEI and to
increase in the polarization. In NAB having such SEI,
the anode will be severely polarized and e v e n blocked
after a short period of discharge time.
A n o t h e r disadvantage of such a system is the effect
of ~- on the emf of the battery and on the anodic halfcell potential. These electrodes are not expected to follow a simple Nernst equation, b u t a more complicated
one (19)
ACM/so, = t+
E~ ----ln
nF
[Mn+]sol
t+-~
(AG side reaction)
[19]
The o v e r - a l l side reaction m a y be
•XsoI m - -~- m/P/-'-> ( M m X n ) SEI -'~
nine
[20]
w h e r e X m- is a n a n i o n i n the solution which is capable
of m i g r a t i n g t h r o u g h the SEI.
It seems that this case is r e l e v a n t to m a n y NAB systems which use calcium or m a g n e s i u m as the anode
since in m a n y m a g n e s i u m or calcium compounds, the
value of t+ is smaller than 1 (15) (e.g., chlorides a n d
oxides). This m a y be the reason we have no calcium
or m a g n e s i u m nonaqueous batteries.
It was found that calcium and m a g n e s i u m SOCIz
cells have less capacity t h a n the l i t h i u m one (24-26).
I n addition, it was observed that the Ca anode in
the SOC12 cell became coated b y a white precipitate
d u r i n g discharge u n t i l it was completely blocked (25,
26).
Downloaded 03 Feb 2010 to 18.53.5.219. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp
2050
J. EZectrochem. Soe.: E L E C T R O C H E M I C A L S C I E N C E A N D T E C H N O L O G Y
A s i m i l a r situation m a y exist i n the Mg-MnO2 dry
cell (24). I n this case the passivating l a y e r o n the
surface of the anode (MgO or Mg(OH)2) m a y have
tMs+2 < 1. As a result the OCP of this cell is lower
t h a n expected. A similar p h e n o m e n o n was observed
d u r i n g cycling of Li in LiBF4-PC solutions (27). The
stripping overpotential rose with time a n d a b u i l d u p
of a passivating layer on the top of the l i t h i u m deposit
was observed by SEM. This type of behavior m a y be
a t t r i b u t e d to SEI having t - > 0. A n a t t e m p t to i m prove the electrochemical performance of Mg electrode
in SOC12 solution b y doping the SEI with higher
valance cations was m a d e b u t with no success yet (24).
Shorts during deposition.--The plating c u r r e n t efficiency of l i t h i u m i n organic systems reaches 100%
(2, 14, 28, 27). However, the stripping efficiency is
much lower t h a n 100% and declines with wet standing
and with the plating c u r r e n t density (27, 28). The
reason for this p h e n o m e n o n seems to be the formation
of shorts through the SEI d u r i n g plating as a result of
local excessive heating a n d the high electric field on
the SEI. The increase of the plating rate should aggravate the formation of shorts. The formation mechanism of these shorts m a y be the dissolution and t r a p ping of metallic atoms i n the cationic vacancies a n d
of electrons i n the anionic vacancies. This type of phen o m e n o n occurs d u r i n g the electrolysis of solid alkali
halides at elevated t e m p e r a t u r e s a n d it causes the i n crease of the electronic conductivity of the solid (29).
A m o r e detailed description of this p r o b l e m can be
seen in Fig. 2. As a result of local heating, some part of
the SEI becomes a n electronic conductor (Fig. 2a), and
therefore, l i t h i u m metal begins to deposit i n the b u l k
o f the SEI (Fig. 2b). This process continues and a ball
o f l i t h i u m totally covered b y SEI is formed (Fig. 2c).As
t h e c u r r e n t diminishes, this short zone i n the SEI dis-
Li
SEI
SOLUTION
Li+I
appears a n d later, w h e n the b a t t e r y is discharged, this
insulated l i t h i u m ball will not be dissolved (Fig. 2d).
This results in a decrease i n the discharge c u r r e n t
efficiency. I n order to alleviate this problem the ionic
resistivity of the SEI should be decreased by doping.
This way both the electric field, which will be developed on the SEI, and the electrical heating of t h e
SEI will be reduced.
The growth rate of the SEI.--A slow corrosion r a t e ,
and therefore, a long shelf life is expected for systems
i n which, at least, one of the corrosion products is
very insoluble and i m m e d i a t e l y precipitates on t h e
anode surface to form a dense and hole-free SEI which
is a very good insulator for electrons. However, even i n
this case, the thickness of the SEI will slowly increase
with time of storage. The rds for this process will always be the cathodic reaction, i.e., the reduction of
the solvent. The functional relation b e t w e e n the thickness of the SEI and the storage time is d e t e r m i n e d b y
the mechanism of the cathodic reaction. Two e x t r e m e
cases can be identified. (i) The anode surface a n d the
SEI are not completely homogeneous, i.e., the anode
contains anodic areas in which metallic ions are dissolved a n d cathodic areas i n which the electrons a r e
migrating t h r o u g h the SEI a n d reducing the solvent.
This case may be r e l e v a n t to practical systems having
impurities which are able to create cathodic zones.
(ii) I n a n u l t r a p u r e system, the anode a n d the SEI are
completely homogeneous a n d thus no well-defined
cathodic areas can exist. I n this case, the rds will be
the diffusion of electrons t h r o u g h the SEI to the solution side.
For the first case, the growth rate of the SEI can be
formulated as follows
V(corrosion) = Veq.cell - - 11r - - ~la - - i R "-- 0
(a)
If we assume that (i) the corrosion c u r r e n t follows
Ohm's law a n d (ii) the electronic resistivity of the SEI
[p(e)] is constant with time t h e n we get
Veq,eelt
~cor# = - -
p(e). L
Li§ [
|
[21]
Veq,eell = OCP (i ---- 0), me, 11a = cathodic and anodic
overpotential; R = resistance of the solution. I n practical NAB systems both ~]a a n d iR can be neglected.
Thus
1% = Veq.ee11
[22]
Li +
@
D e c e m b e r I979
[23]
It was assumed that all the corrosion products precipitate on the anode, to give a homogeneous film thus
Lifo I
(b)
dL
c/t
K~corr
=
[24]
where K is a constant.
F r o m Eq. [23] and [24] one gets
dL
=
dt
--
|
.+
LJsoI
(c)
.§
LIso I
(d)
p(e). L
[25]
After i n t e g r a t i o n of Eq. [25]
L =
|
KVeq,cell
(
Lo2 -t-
2KVeq'cell
) V,
p(e)
9t
[26]
where Lo = L at t ~ 0. This is a parabolic law of
growth of the SEI.
The second m e c h a n i s m involves the diffusion of electrons through the SEI as the rds for the corrosion of
the anode. The corrosion c u r r e n t density will follow
Eq. [27]
FDCo
Li+---~
Fig. 2. The effect of a partial short on the deposition-dissolutle,1
process (see text),
/tort - - - -
L
[27]
where D is the diffusion coefficient of the electrons i n
the SEI a n d Co is the concentration of the electrons i n
Downloaded 03 Feb 2010 to 18.53.5.219. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp
VoZ. I26, No. 12
N O N A Q U E O U S BATTERY SYSTEMS
the SEI near the M/SE interface. From Eq. [24] and
[27]
dL
KFDCo
-
-
dt
-
~
[28]
L
After integration a parabolic law of growth is obtained
L = (Lo2 -t- 2KFDCot)'/,
[29]
Although the two corrosion mechanisms both lead to
parabolic law of growth they should be considered as
first approximation only. A practical system may be
more complicated due to the following: p(e) or D may
change with L; the SEI may contain cracks, holes, and
grain boundaries at which the greatest part of the electronic leakage current may take place. The SEI may
be nonhomogeneous and contain crystals of different
sizes (21). Therefore deviations from the parabolic law
of growth are expected.
The thickness of the SEI of Li in LiA1C14-SOCle
system was measured as a function of time at room
temperature. The value of a in Eq. [30] was calculated for several Li electrodes (21)
L -- const 9 t~
[30]
Not all the Li electrodes exhibited the parabolic law
of growth as values between 0.2 and 0.5 were obtained.
Summary
A model which is designed to explain the complex
and unique electrochemical behavior of alkali and
alkaline earth metals NAB systems was presented in
this paper. This model is supported by many experimental results. It was indicated that the depositiondissolution mechanism of alkali and alkaline earth
metals in SEI-NAB systems is entirely different from
that in aqueous or aqueous-like systems. Therefore
the improvement or control of this process should be
done by different means. It is concluded that a proper
anodic SEI is the key for the operation of NAB. It
seems that the controlling of the properties of the
SEI, i.e., reducing t - and te, increasing t + as close as
possible to unity, and reducing p (by doping the SEI),
will improve the performance of the SEI nonaqueous
batteries.
Manuscript submitted Dec. 27, 1978, revised manuscript received April 9, 1979. This was Paper 4 presented at the Atlanta, Georgia, Meeting of the Society,
Oct. 9-14, 1977.
Any discussion of this paper will appear in a Discussion Section to be published in the June 1980
JOURNAL. All discussions for the June 1980 Discussion
Section should be submitted by Feb. 1, 1980.
Publication costs o] this article were assisted by
TeN Aviv University.
REFERENCES
1. J. Jorne and C. W. Tobias, This Journal, 121, 994
(1974).
2051
2. R. Jasinski, in "Advances in Electrochemistry and
Electrochemical Engineering," Vol. 8, C. W.
Tobias, Editor, Wiley-.Interscience, New York
(1971).
3. J. W. Butler, in "Advances in Electrochemistry and
Electrochemical Engineering," Vol. 7, P. Delahay, Editor, Wiley-Interscience, New York
(1970).
4. W. H. Tiedemann and D. N. Bennion, This Journal,
l~O, 1624 (1973).
5. J. C. Cessna, Corrosion (Houston), 27, 244 (1971).
6. S. G. Meibuhr, This Journal, 118, 709 (1971).
7. S. G. Meibuhr, ibid., 118, 1320 (1971).
8. E. Peled and E. Gileadi, Plating (East Orange,
N.J.), 68, 342 (1975).
9. A. Brenner, in "Advances in Electrochemistry and
Electrochemical Engineering," Vol. 5, C. W.
Tobias, Editor, Wiley-Interscience, New York
(1967).
10. M. Yoshio and N. Ishiba'shi, J. Appl. Electrochem.,
3, 321 (1973).
11. J. J. Auborn, K. W. French, S. I. Lieberman, V. K.
Shah, and A. Heller, This Journal, 120, 1613
(1973).
12. E. Peled and H. Straze, Extended Abstract No. 127
presented at the 27th ISE Meeting, Zurich, Switzerland, September (1976); This Journal, 124,
1030 (1977).
13. A. N. Dey, Paper 62 presented at The Electrochemical Society Meeting, Atlantic City, New Jersey,
October 4-8, 1970.
14. R. D. Rauh and S. B. Brummer, Electrochim. Acta,
22, 75 (1977).
15. L. Heyen, ibid., 15, 1251 (1970).
16. L. Young, in "Anodic Oxide Films," Academic
Press, New York (1961).
17. M. J. Digman, in "The Anodic Behavior of Metals
and Semiconductors Series," Vol. 1, J. W. Diggle,
Editor, Marcel Dekker Inc., New York (1972).
18. E. Gileadi, E. Kirowa-Eisner, and J. Penciner, in
"Interracial Electrochemistry," Addison-Wesley
Publishing Co., Inc., Reading, Mass. (1975).
19. P. Payne, in "Advances in Electrochemistry and
Electrochemical Engineering," Vol. 7, P. Delahay
and C. W. Tobias, Editors, Wiley-Interscience,
New York (1970).
20. R. F. Scarf, This Journal, 117, 295 (1970).
21. E. Peled and H. Yamin, in "Proceedings of the
28th Power Sources Symposium," p. 237, A t lantic City, New Jersey, June 1978.
22. D. R. Cogley and J. M. Butler, J. Phys. Chem., 72,
4568 (1968).
23. E. Peled and H. Yamin, Dr. J. Chem., 18, 131 (1979).
24. E. Peled and A. Meitav, Paper 554 presented at
The Electrochemical Society Meeting, Seattle,
Washington, May 21-26, 1978.
25. M. Domeniconi, et a~., Interim Report No. N0001476-C-0524, October 1977.
26. J. R. Driscoll, et al., ECOM Report No. 74-0030-12,
April 1977.
27. P. J. Rommer and J. R. V. Beck, Paper 1 presented
at The Electrochemical Society Meeting, Atlanta
Georgia, Oct. 9-14, 1977.
28. R. Selim and P. Bro, This Journal, 121, 1457 (1974).
29. J. H. Schulman and W. D. Compton, in "Color Centers in Solids," Pergamon Press, New York
(1962).
Downloaded 03 Feb 2010 to 18.53.5.219. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp