CHEMICAL INTERACTIONS AMONG DEFECTS IN Gg AND Si 537 



appear in every respect to provide a medium for chemical reactivity 

 similar to liquids, particularly water. Such pehnomena as acid-base reac- 

 tions, complex ion formation, and electrolyte phenomena such as Debye 

 Hiickel effects, ion pairing, etc., all seem to take place. 



Besides the experiments theoretical work has been done in an attempt 

 to define the limits of validity of the mass action principle, to furnish 

 more refined electrolyte theories, and most importantly, to provide firm 

 theoretical bases for entirely new phenomena such as ion pair relaxation 

 processes. 



The consequence is that the field of diamond lattice^ semiconductors 

 which has previously engaged the special interests of physicists threatens 

 to become important to chemists. Semiconductor crystals are of interest, 

 not only because of the specific chemical processes occurring in these 

 substances, but also because they serve as proving grounds for certain 

 ideas current among chemists, such as electrolyte theory. On the other 

 hand renewed interest is induced on the part of physicists because chem- 

 ical effects like ion pairing engender new physical effects. 



The purpose of this paper is to present the field of defect interaction 

 as it now stands, in a manner intelligible to both physicists and chem- 

 ists. However, this is not a review paper. Most of the experimental re- 

 sults, and particularly the theories which are fully derived in the text or 

 the appendices are entirely new. Some allusion will be made to published 

 work, particularly to descriptions of the results of some previous theories, 

 in order to round out the development. 



The governing theme of the article lies in the analogy between 

 semiconductors and aqueous solutions. This analogy is useful not so 

 j much for what it explains, but for the experiments which it suggests. 

 : More than once it has stimulated us to new investigations. 

 1 In our work we have made extensive use of lithium as an impurity. 

 This is so because lithium can be employed with special ease to demon- 

 strate most of the concepts we have in mind. This specialization should 

 not obscure the fact that other impurities although not well suited to 

 the performance of accurate measurements, will exhibit much of the 

 same behavior. 



II. ELECTRONS AND HOLES AS CHEMICAL ENTITIES 



Since electrons and holes'* are obvious occupants of semiconductors 

 I like germanium and silicon, and are intimately associated with the pres- 

 [ence of donor and acceptor impurities,^ it is fitting to inciuire into the 

 f roles they may play in chemical interactions between donors and ac- 



