u6 PHYSICAL CHEMISTRY 



at 25 °, when this ion is monovalent and is ten times as concen- 

 trated on one side of the membrane than on the other. 



It is characteristic of most of the separate phase membranes 

 that have so far been studied that they are very poorly permeable 

 to water. Their ionizing power is often very small, and except- 

 ing the metallic membranes, they are poor electric conductors. 

 Hence the measurement of the emf with a galvanometer, capil- 

 lary electrometer, or even a potentiometer, is sometimes im- 

 possible. A quadrant electrometer of Lord Kelvin's or Dole- 

 zalek's type is usually used. 



For the general case of a membrane permeable to both ions 

 of the solute, Nernst's ideal formula in millivolts at 25 ° is, 



U — V U' V' Cjl 



emf = 59 ( ) log — 



u+v u'+v' c 2 



where u and v are the speeds of cation and anion in water and 

 u' and v' in the membrane. Though the emf so calculated would 

 be correct if the two aqueous solutions were connected so as to 

 form a closed circuit, the emf could not be measured under such 

 conditions. If calomel electrodes are used, the solutions con- 

 necting this concentration chain to the electrodes would have an 

 effect on the total emf as measured. 



Examples of membranes permeable to only one of the two ions 

 of an electrolyte are given by Cremer (1906) and Haber and 

 Klemensiewicz (1909). Membranes of ice, benzol, toluol, 

 metaxylol, nitrophenol, and thin, water soaked glass, acted as. 

 though they were permeable only to H ions. When the H ion 

 concentration on one side of the membrane was ten times as 

 great as on the other side, the average emf with soft glass was 

 52 mv and with hard glass 59 mv, whereas that calculated from 

 Nernst's formula is 58 mv at 20 . The emf with the other mem- 

 branes was almost as great. Haber and Klemensiewicz used 

 almost pure solutions of acids and bases (without buffers) and, 

 in their tables, did not correct for dissociation. Furthermore, 

 they made no accurate measurements near the neutral point. 

 Since the reaction of all living cells or fluids bathing them is 

 near the neutral point, it seemed worth while to make such de- 

 terminations, in order to ascertain whether such membranes 



