736 
The Philippine Journal of Science 
1921 
not the case, due to the considerable difference in size between 
the oxygen and chlorine kernels. Consequently CL0 4 (I) and 
Cl 2 0 5 
(III) are not known. Some 
reactions 
which we might 
expect to yield CL0 5 give instead CIO., and a perchlorate, while 
others give 0 L , and CL by an exothermic rearrangement. 
0 
0 
0 
8 
8 
8 
Cl — 0 — Cl cc 0 
Cl 
— Cl 
8 
8 
8 
0 
0 
0 
I 
II 
0 
0 0 
0 
8 
8 8 
8 
Cl 
0 — Cl 0 B Cl 
— 0 
Cl CC 0 
8 
8 8 
8 
0 
0 0 
0 
III 
IV 
It seems that only quadricovalent chlorine 
(as in CIA, for- 
mula IV) has sufficient binding strength to unite with dico valent 
oxygen. This is the writer’s interpretation of the fact that 
chlorine shows a tendency “to share all 4 pairs of its electrons 
if it has to share any.” Chlorine heptoxide is therefore a rel- 
atively stable compound, but it is strongly endothermic as com- 
pared to Cl 2 and 0 2 , because the chlorine shell electrons are all 
displaced from their normal positions. This quasi-stability ot 
chlorine heptoxide illustrates the fact that chemical problems 
deal much more with metastable than with stable equilibrium. 
Especially in organic chemistry it is well known that the relative 
potential energy of two sets of possible reaction products has 
often very little bearing on the question of which will be formed. 
This applies in a somewhat more limited way also to inorganic 
chemistry, as will be illustrated in the further discussion of the 
chlorine oxides. 
CL0 4 (formula I) probably exists to some extent in liquid 
chlorine dioxide. It would seem that formula II represents a 
much stabler compound, which may be present in certain mix- 
tures of chlorine oxides of disputed composition, but would not be 
formed readily by the association of C10 2 or by the decomposi- 
tion of chlorates. The reason for this is the distinct polarity of 
C10 2 and C10 3 "", caused principally by the fact that the oxygen 
kernel has a positive charge of six, and the chlorine kernel seven. 
The unions in C10 3 are, therefore, what the writer has called 
