The electrical conductivities of glasses in the system Li
2O-K
2O-Fe
2O
3-P
2O
5 were observed in the temperature range 20° to 160°C. Whereas alkali ions, Li
2O and K
2O, give rise to ionic conduction, Fe
2O
3 to electronic conduction. In the series (40-
x) Li
2O-
xK
2O-60P
2O
5 the conductivity was found to have a minimum and the activation energy to have a maximum at the composition corresponding to
x=20mol%. In the series (20-
x) Li
2O-
xK
2O-20Fe
2O
3-60P
2O
5 the curves of logσ versus the content of K
2O were shown to have the minimum at about
x=5mol%, however, the activation energy curves decreased with increasing K
2O content without having any maximum. This can be considered from the fact that a part of Li
2O combines with Fe
2O
3 in some ways, i.e. by way of producing a phase LiFeO
2 or LiFe
5O
8. Li
+ utilized for the ionic conduction may actually decrease.
In complicated composition of glass, the simple additivity law does not hold. The experimental values of logσ
100°C in the 20Li
2O-20Fe
2O
3-60P
2O
5 glass was about one order of magnitude lower than those calculated from the formula of additive property,
σ=
K1⋅
y⋅
e-ΔHa/RT+
K2⋅(1-
y)
2⋅
c(1-
c)⋅
e-ΔHe/RTwhere
K1,
K2 are constants,
ΔHa the activation energy for the glass 40R
2O-60P
2O
5,
ΔHe the activation energy for the glass 40Fe
2O
3-60P
2O
5,
y the concentration of R
2O, (1-
y) the concentration of Fe
2O
3,
c the ratio of Fe
2+/Fe
2++Fe
3+,
R gas constant, and
T absolute temperature.
On the other hand the experimental values of the conductivity in
xK
2O-(40-
x)Fe
2O
3-60P
2O
5 glasses were found to be approximately equal to the calculated values from the above formula of additivity law. In this experiment, therefore, the additivity law holds only for the glass containing K
2O as alkali component. Accordingly the formula of the conductivity in the glasses containing both Li
2O and Fe
2O
3 is considered to be supplemented by a correction term owing to the producing of the phase LiFeO
2 and LiFe
5O
8 which hinder both ionic and electronic conduction; that is, we have a conduction formula,
σ=
K1⋅
y⋅
e-ΔHa/RT+
K2⋅(1-
y)
2⋅
c(1-
c)⋅
e-ΔHe/RT+σ
c.
σ
c is a correction term.
View full abstract