PAPERS ON PHYSICS 295 



THERMAL CONDUCTIVITY OF CONCRETE 



Prof. A. P. Carman and R. A. Nelson 

 University of Illinois 



The cylinder method was used. A long cylinder had a 

 circular hole along the axis in which there was an electric 

 heating coil. By preliminary tests it was shown that the 

 flow of heat in the middle of the cylinder was radial. The 

 heat generated in the coil in this middle part could be cal- 

 culated directly from the electric current and e. m. f. per 

 unit length of the coil. The temperature gradient was 

 measured by thermocouples which were placed in holes 

 parallel to the axis at different distances from the axis. 

 The thermal conductivity was then calculated from the for- 

 mula 



K= Q /R2\ 



2-1-L (T, = T2 ) ^^ ^R.'' 



In this formula, Q is the quantity of heat generated in 

 unit length at the middle, t^ and tg are temperatures at 

 radial distances r, and r; when a steady flow is reached. 



1 2 



Over fifty cylinders of various standard concrete mix- 

 tures have been tested at temperatures ranging from 50 C 

 to 300 C. The cylinders were made of various standard 

 concrete mixtures, the ratios of the mixture to the ag- 

 gregate being 1 :2, 1 :3, 1 :4, 1 :5, 1 :7, 1 ;9, and different pro- 

 portions of the mixing water were also used. The thermal 

 conductivity in c. g. s. physical units for "neat" cement 

 was found to be .00147 and that for mixtures of different 

 aggregates was about .00344 to .00384. The thermal con- 

 ductivity of "neat" cement is thus about 1/2 of that for 

 any concrete mixture. The thermal conductivity of con- 

 crete mixtures did not vary much with the richness in ce- 

 ment of the mixtures. 



The above values are for temperatures ranging from 

 lOO^C to 200^C, but the effect of temperatures on thermal 

 conductivity below 300''C was not marked. All the above 

 concretes were thoroughly dried and had an age of from 

 28 days to 120 days. 



