September i6, 1915] 



NATURE 



67 



I 



variably found to diminish, if the conditions are such 

 that this is possible. If we have, therefore, a system 

 in which, by any change of distribution of the con- 

 stituents, free energy can be decreased, such a change 

 of distribution will take place. This is one form of 

 the well-known " Principle of Carnot and Clausius." 



Now, practically any substance dissolved in water 

 lowers the surface tension present at the interface 

 between the liquid and another solid or liquid phase 

 with which it is in contact. Moreover, up to a certain 

 limit, the magnitude of this effect is in proportion 

 to the concentration of the solute. Therefore, as was 

 first pointed out by Willard Gibbs, concentration of 

 a solute at an interface has the effect of reducing free 

 energy and will therefore occur. This is adsorption. 

 As an example, we may take the deposition of a dye- 

 stuff on the surface of charcoal, from which it can be 

 removed again, unaltered, by appropriate means, such 

 as extraction with alcohol. Charcoal plus dye may, 

 if any satisfaction is derived from the statement, be 

 called a compound. But, since its chemical composi- 

 tion depends on the concentration of the solution in 

 which it was formed, it is much more accurate to 

 qualify the statement by calling it an "adsorption- 

 compound." Moreover, the suggestion that the union 

 is a chemical one tends to deprive the conception of 

 chemical combination of its characteristic quality, 

 namely, the change of properties. Dye-stuff and 

 charcoal are chemically unchanged by adsorption. 



The origin of adsorption from surface tension is 

 easily able to explain why it is less as the temperature 

 rises', as we find experimentally. As we have just 

 seen, surface tension diminishes with increase of 

 temperature. 



Let us next consider what will happen if the liquid 

 phase contains in solution a substance which lowers 

 surface tension and is also capable of entering into 

 chemical reaction with the material of which the other, 

 solid, phase consists. For example, a solution of 

 caproic acid in contact with particles of aluminium 

 hydroxide. On the surface of the solid, the concentra- 

 tion of the acid will be increased by adsorption, and, 

 in consequence, the rate of the reaction with it will 

 be raised, according to the law of mass action. 

 Further, suppose that the liquid phase contains two 

 substances which react slowly with each other, but 

 not with the solid phase. They will be brought into 

 intimate contact with each other on the surface of the 

 solid phase, their concentration raised and the rate of 

 their interaction increased. One of the reagents may 

 clearly be the solvent itself. But in all these cases the 

 rate of the reaction cannot be expressed by a simple 

 application of the law of mass action, since the active 

 masses are not functions of the molecular concentra- 

 tions, but of the surface of the phase boundaries. The 

 application of these considerations to the problem of 

 the action of enzymes and of heterogeneous catalysis 

 in general will be apparent. That the action of 

 enzymes is exerted by their surfaces is shown, apart 

 from the fact that they are in colloidal solution, by 

 the results of experiments made in liquids in which 

 the enzymes themselves are insoluble in the usual 

 sense, so that they can be filtered off by ordinary filter 

 paper and the filtrate found to be free from enzyme. 

 Notwithstanding this insolubility, enzymes are still 

 active in these liquids. The statement has been found, 

 up to the present, to apply to lipase, emulsin, and 

 urease, probably to trypsin, and the only difficulty in 

 extending it to' all enzymes is that of finding a sub- 

 strate soluble in some liquid in which the enzyme itself 

 is not. That adsorption is a controlling factor in the 

 velocity of enzyme action has been advocated by myself 

 for some years, but it is not to be understood as imply- 

 ing that the whole action of enzymes is an " adsorption 



NO. 2394, VOL. 96] 



phenomenon," whatever may be the meaning of this 

 statement. The rate at which the chemical reaction 

 proceeds is controlled by the mass of the reagents 

 concentrated on the surface of the enzyme phase at 

 any given moment, but the temp>erature coefficient 

 will, of course, be that of a chemical reaction. 



The thought naturally suggests itself, may not the 

 adsorption of the reacting substances on the surface 

 of the enzyme suffice in itself to bring about the equi- 

 librium at a greater rate, so that the assumption of 

 a secondary chemical combination of a chemical nature 

 between enzyme and substrate may be superfluous? I 

 should hesitate somewhat to propose this hypothesis 

 for serious consideration were it not that it was given 

 by Faraday as the explanation of one of the most 

 familiar cases of heterogeneous catalysis, namely, the 

 union of oxygen and hydrogen gases by means of the 

 surfaces of platinum and other substances. The in- 

 sight shown by Faraday into the nature of the pheno- 

 mena with which he was concerned is well known and 

 has often caused astonishment. Now, this case of 

 oxygen and hydrogen gases is clearly one of those 

 called "catalytic" by Berzelius. The fact that the 

 agent responsible for the effect did not itself suffer 

 change was clear to Faraday. I would also, in paren- 

 thesis, direct attention to the fact that he correctly 

 recognised the gold solutions which he prepared as 

 suspensions of metallic particles — that is, as what we 

 now call colloidal solutions. Although the systematic 

 investigation of colloids, and the name itself, were due 

 to Graham, some of the credit of the discovery should 

 be given to the man who first saw what was their 

 nature. Adsorption, again, was accurately described 

 by Faraday, but without giving it a name. 



I confess that there are, at present, difficulties in 

 the way of accepting concentration by adsorption as a 

 complete explanation of the catalytic activities of 

 enzymes. It is not obvious, for example, why the 

 same enzyme should not be able to hydrolyse both 

 maltose and saccharose, as it is usually expressed. 

 Another difficulty is that it is necessary to assume that 

 the relative concentration of the components of the 

 chemical system must be the same on the surface of 

 the enzyme as it is in the body of the solution ; in 

 other words, the adsorption of each must be the same 

 function of its concentration. Unless this were so, 

 the equilibrium position on the enzyme surfaces, and 

 therefore in the body of the solution, would be a 

 different one under the action of an enzyme from that 

 arrived at spontaneously or brought about by a homo- 

 geneous catalyst such as acid. This consideration was 

 brought to my notice by Prof. Hopkins, and re- 

 quires experimental investigation. We know, indeed, 

 that in some cases there is such a difference in the 

 position of the equilibrium position, for which various 

 explanations have been suggested. But it would be 

 a matter of some interest to know whether this differ- 

 ence has any relation to different degrees of adsorption 

 of the components of the system. 



At the same time, adsorption is under the control of 

 so many factors, surface tension, electrical charge, and 

 so on, that the possibilities seem innumerable. There 

 are, moreover, two considerations to which I may be 

 allowed to direct your attention. Hardy has pointed 

 out that it is probable that the increased rate of 

 reaction at the interface between phases may be due, 

 not merely to increased concentration as such, but that 

 in the act of concentration itself molecular forces may 

 be brought into play which result in a rise in chemical 

 potential of the reacting substances. In the second 

 place, Barger has shown that the adsorption of iodine 

 by certain organic compounds is clearly related to the 

 chemical composition of the surfaces of these sub- 

 stances, but that this relationship does not result in 



