510 
PACIFIC SCIENCE, VoL XX, October 1966 
Fig. 3. Exchange curves of silica gels. 
Figure 1, it is found that Ks/Ka for the acid 
functional groups of bentonite is about 10 2 or 
less. This implies that sodium ions are held on 
the montmorillonite clay with the same affinity 
as are the hydrogen ions. The amount of sodium 
ions adsorbed by halloysite and kaolinite is 
very small and partly pH-dependent. The in- 
flection points near pH 8 indicate that the 
acid-functional groups of the pH-dependent 
part have a Ks/Ka value of about 10 8 . The 
hydrogen ions associated with these functional 
groups cannot be replaced by sodium ions un- 
less the (Na+)/(H+) ratio of the solution 
becomes greater than 10 10 . This suggests cova- 
lent bond formation between the functional 
groups and the hydrogen ions; simple ionic 
interaction between the negative charges and 
the hydrogen ions seems to be inadequate to 
explain this phenomenon. A comparison of 
these curves with those of silica gels shown in 
Figure 3 indicates that these charges on kao- 
linite and halloysite most probably are due to 
the surface hydroxyls attached to silicon atoms. 
Aluminum Hydroxides 
Aluminum hydroxides A and B both failed 
to show cation exchange to an extent that could 
be considered of real importance. The amount 
of sodium ions adsorbed under the pH range 
studied were always less than 1 me. per 100 g. 
Aluminum hydroxide C had a maximum ex- 
change capacity of about 6 me. per 100 g, show- 
ing an inflection point near pH 8. The amount 
of sodium ions adsorbed at pH 4.5 was about 
1 me. per 100 g. 
Synthetic Aluminosilicate Gels 
That the tetrahedrally coordinated aluminum 
atoms produce exchange sites was suggested by 
Her (1955), and by De Kimpe, Gastuche and 
Brindley (1961). Such tetrahedrally coordi- 
nated aluminum atoms were postulated as the 
source of exchange capacity of soil allophanes 
by Iimura (1961). In this investigation, the 
aluminosilicate gels studied were prepared 
from alkaline solutions so as to arrange the 
aluminum atoms in the tetrahedrally coordi- 
nated positions, as much as possible. Their 
compositions, as well as the maximum cation 
exchange capacities, are listed in Table 1. Their 
exchange curves are shown in Figure 5. 
It is striking that the CEC/A1 ratios of the 
gels with silica-alumina ratios greater than 0.7 
are very close to 1 , indicating that each negative 
charge on the gel is associated with one alumi- 
num atom. If the CEC of the gel is entirely 
attributed to the tetrahedrally coordinated 
aluminum atoms in gel A 2 , it turns out that 
about 60% of the tetrahedral positions in the 
gel are occupied by the aluminum atoms. In 
gel A x the same calculation reveals that the 
tetrahedral positions occupied by the aluminum 
atoms would be as high as 62%, and that 
this amount may be the practical limit for the 
aluminum atoms to take the tetrahedral posi- 
tions in the aluminosilicate gels under the spe- 
cified conditions of preparation. Iler (1955), 
on the other hand, states that the maximum 
amount of aluminum atoms which can be tetra- 
hedrally coordinated in the aluminosilicate gels 
is 50%. 
The surface silanol groups of these synthetic 
gels apparently do not dissociate under the con- 
ditions of the experiment, and thus do not 
contribute to the CEC. The negative charges 
caused by these tetrahedrally coordinated alumi- 
num atoms are relatively strong in acid charac- 
ter (that is, readily dissociable), and their 
