Impregnation occurred for 3 hours at 15 psig overpressure. Excess 

 monomer was drained and hot water (85 to 95 C) was introduced into the 

 chamber to initiate polymerization of the monomer. After 4 hours the 

 aggregate was removed to an oven for heating overnight at 110 C to 

 assure complete polymerization. 



Table 3 shows that after the first impregnation the percentage 

 weight gain of polymer loading in the aggregate could be increased by a 

 second impregnation. A second impregnation was conducted, which brought 

 the polymer loading values closer to that calculated as the maximum. 

 The data showed that a certain portion of the void volume (about 7.5%, 

 8.6%, and 12.7% by volume for the coarse sand, 5/16-inch and 3/8-inch 

 aggregate, respectively) remained empty after the second impregnation. 

 Figure 4a shows a scanning electron microscope photograph at 15 times 

 magnification of a PFA aggregate particle. Polymer in many of the voids 

 is separated from the wall of the void as if shrinkage occurred during 

 the polymerization process. For comparison. Figure 4b shows a regular 

 lightweight aggregate particle. 



Concrete 



Table 4 gives the mix designs for the concrete. The basis for the 

 designs of Mix no. 1 through 3 was manufacturers' technical literature." 

 Mix no. 4 was a modification of Mix no. 3 in which a greater proportion 

 of large aggregate was used. The aggregate sizes were blended to meet 

 ASTM specifications C-33 for grading of concrete aggregates. 



The aggregate proportions in Table 4 are for regular lightweight 

 particles in a dry or "as received" condition from the manufacturer. 

 The manufacturer packages oven-dry material in paper sacks, but moisture 

 is picked up by the aggregate during storage. The aggregate weights 

 used during batching were from the slightly moisture-laden aggregates. 

 Without having the oven-dry weights, the quantity of PFA to use in each 

 batch could not be calculated using weighing methods. Therefore, a 

 volume batching method was used. 



Slump was used to control the quantity of water added to each batch 

 of concrete. The significantly different water-to-cement ratios between 

 PFA and regular lightweight concrete resulted from using the totally 

 saturated condition of the regular aggregate and the nonsaturated condi- 

 tion of the PFA. The quantity of water added to the mixes was the 

 amount used in calculating the water-to-cement ratio. 



All specimens were fog-cured for 28 days prior to testing for 

 strength or before placement in other environmental conditions for unit 

 weight measurements. 



EXPERIMENTAL PROCEDURES 



The compressive strength tests were conducted on 4 by 8-inch 

 cylinders in accordance with ASTM C-39, and splitting tensile strength 



''Lightweight Processing Co. Technical reference manual for rocklite con- 

 crete. Glendale, Calif., 1966. 



