agent. No one assumes that in this agent alkyl and halogen are
present in the state of free ions they are dissociated but only
in the sense that they are separately held by the metal. The
forces of residual affinity have been entirely disregarded by the
dissociationist school; and not being practised chemists, knowing
nothing of the organic side, they have left facts out of account.
The implications in Longfellow's lines, with reference to the sea
Only those who brave its dangers
Understand its mystery
will ever be true and generally applicable. It is necessary to give this warning to the coming generation of workers.
Granted that the interaction, in the case of the formation of water from hydrogen and oxygen, be determined by the presence of an electrolyte (an acid impurity), is this to be thought of as the catalyst? W hat is a catalyst?
The definition of a catalyst which is generally current is that it is an agent which merely accelerates a change in being: but this is based upon the gratuitous assumption that two pure sub- stances can interact. If this definition could be accepted, the acid impurity determining the rapid interaction of hydrogen and oxygen when a mixture is fired might be called the catalyst.
The issue is not quite so simple, however, since another class of activating agent has to be considered namely the solid, such as platinum and the enzymes of natural occurrence. Hydrogen and oxygen at once interact when brought into contact with platinum black or with a " clean " platinum plate. Again, it is customary to think of the platinum only as the determining agent; there can, however, be no doubt that, in this case also, the electrolyte must be present. It does not seem probable that platinum in itself would form a conducting circuit with hydrogen and oxygen: if the two gases were condensed at its surface even to the extent of being liquids, these would be non-conducting liquids.
The probability of this view is enhanced when the nature of the process is taken into account and it is realized that the pri- mary interaction is not even that of hydrogen and oxygen atoms but that, initially, the oxygen is converted into hydrogen perox- ide, acting as " depolarizer " in an electrolytic circuit, whilst the hydrogen is " hydroxylated," as shown in the following equation:
HHO HO HOH.. ..HO
HHO .................. HO HOH .................. HO
Hydrogen Electrolyte Oxygen Water Electrolyte Hydrogen Peroxide.
In the next change, the hydrogen peroxide acts as depolarizer, so that the reduction of the oxygen molecule is affected in two stages. According to this view, the electrolyte is the determining agent, the platinum exercising only an accelerating influence if the definition of a catalyst as an accelerator be retained, the platinum rather than the acid is to be regarded as the catalyst. Two other cases may be considered with advantage (i) that of a ferrous salt in promoting oxidation by means of hydrogen peroxide; (2) the hydrolytic action of enzymes. Hydrogen peroxide has little effect as an oxidizing agent and probably, if it could be used in pure solutions, it would be without action; oxidation at once sets in on the addition of a trace of ferrous salt. Familiar cases are the liberation of iodine from iodides (rendered evident by the presence of starch) and the oxidation of tartaric acid to dihydroxymaleic acid (rendered evident by the appear- ance of a violet colour on addition of excess of caustic soda). What is the function of the ferrous salt is it of such a kind that it is to be ranked as a catalyst? Its function would seem to be rather that of carrying the peroxide into action through the formation of a perhydrol which can act as an electrolyte, thus:
+
Ferrous sulphate.
HO
HO Hydrogen peroxide.
= Fe
/O.OH
+ OH 2
Ferrous sulphate perhydrol.
The case of the enzymes is more complex. These are all of natural origin and can only be judged by their actions. It is desirable to confine the term to hydrolytic agents.
Take the case of invertase, the enzyme present in ordinary yeast, which acts only on cane sugar and certain derivatives of
this sugar. Cane sugar is hydrolysed, more or less readily, by all acids, being converted into the two hexoses, glucose and fruc- tose:
CijHaOn +OH 2 . HX = C 6 H 12 O 6 + C 6 H 12 O 6 + HX.
It is similarly affected by invertase acting in a solution which is only faintly acid. Taking into account the amount of change effected by a small amount of enzyme compared with that effected by a relatively large amount of even a strong acid, it is clear that the enzyme is far more active than is any acid per se; it cer- tainly, therefore, can be regarded as an accelerator rather than as a determinant of change. The minute amount of acid which appears to be necessary may be regarded as active in the same way that a trace of acid is active in determining the interaction of hydrogen and oxygen at a platinum surface.
The essential differences between the two classes of agent, the acid and the enzyme, become obvious when the rates at which action proceeds are contrasted. In any interaction occurring in an aqueous solution, such as that in which cane sugar is hydro- lysed by an acid, so long as the solution be not too concentrated, the disappearance of water may be disregarded, owing to the relatively small extent to which it is withdrawn so that only a single changing substance need be considered. In such a case, the amount of change, during each successive interval of time, is proportional to the amount of unchanged substance present. If say 10% disappear during the first period, 10% of the remainder will disappear during each successive period, i.e. 10, 9, 8-1, 7-29, during periods i, 2, 3, 4, etc. The graph representing the rate of change is a logarithmic or exponential curve.
When cane sugar is hydrolysed by the enzyme invertase, the rate of change is of an entirely different order; within wide limits of concentration, well beyond the 50% limit, equal amounts are hydrolysed in each successive interval; the graph representative of the rate of change is, therefore, nearly a straight line. This be- haviour is characteristic of enzymes generally, though in many the rate is modified fairly soon by the reversal of the change or otherwise. The same behaviour is met with in solid catalysts, e.g. the reduction of the fatty oils and of unsaturated compounds such as ethylic cinnamate (C 6 H 5 .CH : CH.C0 2 Et) and anethol (CeH^CaHs.OCHs), by hydrogen in presence of finely divided nickel. Such results cannot well be explained except by the con- centration of the interacting materials at the solid surface; as enzymes behave like nickel, they too must be thought of as acting in a similar way and as merely suspended in the liquid in which they are brought into action. This explanation is rendered the more acceptable by the fact that enzymes will act even when suspended in alcohol, in which they certainly are insoluble. It thus appears desirable to confine " catalyst " to particulate agents acting at surfaces of concentration, and to apply the term " determinant " to agents, such as ferrous sul- phate or acids, acting under conditions of uniform distribution, in solution. The determinant may be said to be required in all cases, being the agent which constitutes the solvent an electro- lyte. It is here assumed that no liquid per se is an electrolyte, excluding fused salts as liquids.
The assertion of the Arrhenius school that water pure water is very slightly dissociated and that it can by itself determine some slight amount of change, is neither logical nor rational. Water is one of the most protean of compounds and has prop- erties which are altogether special. In considering the problems of particulate action, it is necessary that changes in water itself should be taken fully into account. Whilst it is admitted that water is a polymorphous substance and that we are not justified in assigning the simple formula H 2 to the molecule except it be in the state of dry steam, there is no agreement as to the consti- tution of the liquid or of ice. To avoid confusion, it is well to assign a special name to the simple molecule : hydrone, proposed by H. E. Armstrong, appears to be appropriate, most in accord- ance with its neutral character and justified by analogy, its or- ganic analogue being the ketone acetone, OC(CH 3 ) 2 . The passage of water through its three states is too commonly represented as a series of physical changes; actually there can be little doubt that a complex series of structural changes is involved. Accord-
