Page 626 - Handbook of Battery Materials
P. 626
600 17 Liquid Nonaqueous Electrolytes
This simple definition is invalid if the electrolyte is no longer ideal and contains,
for example, ion associates. The transference number has to be defined much more
generally: ‘the transference number of a cation- or anion-constituent R is the net
number of faradays carried by that constituent in the direction of the cathode or
anode, respectively, across a reference plane fixed with respect to the solvent, when
1 F of electricity passes across the plane’ [436]. With this definition the transference
number t R can also be negative, for example, if an ion A + is carried in a net
−
transport as an ion-constituent AX 2 to the anode.
To determine transference numbers of liquid electrolytes, generally ‘classical’
methods are used. These classical methods, meaning moving boundary [436],
Hittorf’s method, and combined data from emf [437], involve extensive experiments.
In the following, the different methods (Hittorf, emf, potentiostatic polarization,
conductivity, galvanostatic polarization, pfg-NMR and impedance measurements)
are explained as well as their advantages and disadvantages. This group studied [528]
some LiDFOB and LiPF 6 based solutions with four different methods, including
three electrochemical methods and NMR-measurements. Whereas three electro-
chemical methods yield transference numbers decreasing with concentration in
accordance with electrostatic theories, valid for low to intermediate concentrations
of the electrolyte, NMR measurements show increasing transference numbers with
increasing concentration.
17.4.6.2 Hittorf Method
In Hittorf’s method a known quantity of electricity passes through the solution
for a defined period. After electrolysis the cell is split into different sections that
are separated and analyzed [438, 439]. To ensure that no mixing between the two
electrode compartments has occurred, more than one middle compartment has to
be incorporated into the cell, see Figure 17.17. The concentrations in the middle
compartments must not change.
Although having the advantage of measuring the transference number directly,
some distracting effects constrain the procedure and result in insoluble problems
for nonaqueous electrolytes. First of all, the determination of transference numbers
depends on the accuracy of the analysis, because it is defined by a small concentra-
tion change between the two electrode compartments E A and E C . Analyzing can be
done by gravimetric [440] or potentiometric [441] methods, by radioactive isotopes
[442], and, for lithium electrolytes most commonly used, spectrophotometric [443,
444] methods. For appropriate evaluation a concentration gradient with sufficiently
high magnitude should be available, which is only accomplished by a long-time
+ E A M A M M M C E C −
Figure 17.17 Diagram of electrolysis cell with two electrode
compartments and three middle compartments.

