246 M. E. WiNFIELD 



There are two ways in which to satisfy the widespread claim that during the 

 catalatic cycle the valency of the iron atom does not change. One is to assume 

 with Chance and Fergusson (1954) that peroxide co-ordinates to Fe"^ of 

 catalase without at any time extracting electrons from the metal atom. The 

 co-ordinated peroxide essentially retains its identity until a second peroxide 

 molecule approaches. The two then interact without participation of the iron 

 atom, except in so far as it renders the first peroxide molecule more electro- 

 negative and therefore able to withdraw two electrons from the second. 



An alternative explanation assumes that the first molecule of HgO,, very 

 rapidly after its co-ordination to Fe^^^, does in fact remove two electrons 

 from the iron porphyrin unit, but without appreciable change in the number 

 of electrons in the orbitals of the iron atom. In other words, although there is 

 an electron rearrangement about the metal atom, and although the conventions 

 of co-ordination chemistry require us to write the iron atom as having a 

 formal valency other than three, the removal of electrons from its vicinity is 

 compensated by an approximately equal donation of electrons by one or 

 more of the six ligands. 



It is the second of the above alternatives that we favour, for the following 

 reasons : 



(i) In general peroxide, except when attached at both ends to a metal ion, 

 is a weak ligand. We are drawing here on unpubhshed experiments on the 

 reactions of HoOo and Oo with K3Co(CN)5, as well as the work of Werner 

 (191 1) and others on peroxocomplexes in general. It is unlikely that peroxide 

 can co-ordinate at one end to Fe^" strongly enough to account for the apparent 

 dissociation constant of less than 10~'^ of the so-called catalase-HaOg complex 

 (Chance et al, 1952). 



(ii) Substitution of peroxide for the — OH or — OHg group in position 6 

 on the iron atom cannot be expected to weaken drastically the Soret band, 

 which we may expect to have an intensity within the range found with 

 HgO, OH~, Cl~ or acetate ion as ligand. The large fall in extinction coefficient 

 which accompanies the conversion of catalase to Cat. H2O2 I indicates im- 

 paired resonance in the porphyrin ring, and can scarcely be attributed to 

 co-ordination of peroxide. King and Winfield (1959a) have pointed out that 

 simultaneous attachment of the peroxide to the porphyrin as well as to the 

 metal is improbable. 



(iii) The affinity of Cat. H2O2 I for NH2OH, fluoride, etc., is higher than 

 that of catalase, and the interaction is relatively slow (see, for example. 

 Beers, 1955) — observations which are inexplicable in terms of displacement 

 of the peroxide ligand by fluoride. 



Structure of Catalase Peroxide I 



When a co-ordination complex is oxidized in a succession of 1 -electron 

 steps, we may in general expect that electrons are lost from the metal atom 



