A combination of ground tire rubber and asphalt cement was evaluated in the laboratory and in two full-scale experimental pavements. Tire rubber was added to asphalt in quantities from 18 to 26 percent by weight including whole tire and tread rubber produced from both ambient and cryogenic grinding processes. Rubber gradation was varied to provide graded and one-sized distributions of rubber particles in the resulting blends of asphalt-rubber.

Asphalt-rubber placed in the experimental pavements and comparable blends prepared in the laboratory are evaluated by laboratory tests. Four principal tests are presented for evaluation of asphalt-rubber behavior including force-ductility, double-ball softening point, rotational viscosity, and size exclusion chromatography. Results of laboratory testing indicates properties of field prepared asphalt-rubber can be duplicated in laboratory prepared blends. Condition survey results from three full-scale test pavements provide information on the most effective combinations of asphalt and tire rubber for interlayer construction.

A rotational viscometer was developed which simultaneously blends the rubber and asphalt and monitors changes in consistency. The variation in consistency with increasing rubber content is presented as a possible means of monitoring rubber concentration during construction.

Chemical analysis of the asphalt and rubber blends by gel permeation chromotography indicates that some chemical modification occurs to the asphalt as a function of blending temperature and time.

A method is presented which describes an extraction process for asphalt and rubber. Extraction of rubber from asphalt-rubber blends after various digestion periods indicates that a loss of integrity of solid rubber occurs as a function of digestion time, temperature and rubber type and size. Periods up to 24 hours resulted in a loss of solid rubber by weight of the blend of up to 34 percent due to digestion with asphalt.