48 THE TRUE VALUE OF a OF VAN DER WAALS 1 



These numbers agree very well among themselves and they leave 

 little doubt that the true value for a is close to the mean value 

 of 27.71 X 10' 2 . The computation from the molecular weight and 

 valences, taking chlorine as trivalent so that the total number of 

 valences is 32, is almost exactly this number, namely 27.62 X H)' 2 - 

 A better agreement could not be wished. Van Laar's value is as 

 usual wrong, in this case being too high. He gives what cones- 

 ponds to 29.38 X 10 12 as the value of a. 



31. Methyl butyrate. (' b H^0 2 . 



(a), a = 2.03 RT C F C /1A)S — P r V. 2 . T, = 554.2 ; P,. = 20.055 

 mm.Hg.; ^ c = .3002; F c = 339.7... a == 26.86 X »0 12 



(/,). a = 0.5 P,.F 2 a = 26.07 X »0 12 



(c). a = p'M 2 /3 dp. ^ = 91.31. M = 1 02 . . a = 29.56 X 10' 2 



(a), a = C'BT C V C . C' = 1.824 a = 28.57 X I0 12 



(<?). a = 3 MN^T e CJd . d -= 1.1820. C= 2.22 



(Morgan and Schwartz.) a = 26.98 X N> 12 



The mean of these rather irregular values is 27.61 X 10 12 . The 

 value calculated from the molecular weight and the number of 

 valences, taking the number of valences as 34, is 26.96 X 10 12 , which 

 is almost exactly the value calculated from the surface tension, but 

 which is between 2 and 3 % lower than the mean. Considering the 

 variation in the values as calculated, the agreement must be consi- 

 dered as satisfactory. Van Laar gives the value of 0.245 fora 1 ' 2 

 which computed into the other units and for a gram mol makes 

 26.34 X l^ 12 m ^is case not very much below the real value. 



32. Methyl iso-butyraie. C 5 H 10 O 2 . 



(a), a = 2.03 RT c V c j 1 .03— P C F C 2 . T c = 540.55; P e = 25.740; 

 d c = . 3012; V c = 338.6 a = 26.07X10 12 



(b). a = 0.5 P r r; 2 a = 25.58 X 1« 12 



(c). a = {jl'M 2 /3 dp. fi! = 87.78 M = 102. . . a = 28.68 X 10 2 

 (a), a = C'BT C V C . C = 1.814 a = 27.62 X 10' 2 



(e). a = ;> ) J/A l: ''T l /'/d .d„= 1.1780; 6'= 2.25 

 between 100° and 132.5° (Ramsay and 



Aston) a = 26.76 X 10 12 



Again the latent heat formulae give a high value and do not agree 

 very well among themselves. The mean value of all is 26.94 X '0 12 . 

 This is almost exactly the value required by the computation from 

 the molecular weight and 34 valences of 26.96 X 10' 2 . Van Laar's 

 computed value is very close to this, being also 20.34. 



