Waller, H. F.
Staff engineersymposium chairman, Materials Research and Development Inc., Raleigh, N. C.
Pages: 2 Published: Jan 1969
ASTM Test for Quantitative Extraction of Bitumen from Bituminous Paving Mixtures (D 2172-67) describes four procedures for determining the amount of bitumen in bituminous mixture. While these procedures have the advantage of a relative simplicity and of having been used for many years, they are extremely time consuming to perform, and the results obtained are often of questionable value, especially when single tests are made. With the high production capacity of modern day asphalt plants, it is essential that test methods be available that can provide an accurate measure of bitumen content within minutes if a meaningful degree of control is to be employed. It is not uncommon for a modern plant to produce 100 to 200 tons of mixture per hour. Therefore, one readily may visualize the possible difficulties that could arise if three or four hours were required to obtain a test result. The problem could be complicated even further if a check test or tests were necessary. By this time several hundred tons of mixture would be laid and compacted, thereby making appropriate adjustments in bitumen content impossible for the mixture already placed on the grade. Perhaps the most commonly used of the four procedures described in Method D 2172-67 is the Rotarex Extraction Procedure. This procedure involves separation of the bitumen-aggregate mixture by use of a solvent; the liquid is then removed by a centrifuge type device. This test presents a number of operational problems including a time factor for completion. Added to the matter of the time required for performance of the test (usually one to four hours) is the relatively high standard deviation associated with the procedure. Most technologists agree that under normal operating conditions, a standard deviation (σ) in the order of 0.25 percent will be about the lowest that can be expected. When one considers the fact that most bituminous paving specifications allow an operating tolerance on bitumen content of ± 0.3 to ± 0.5 percent from the target value, it can be seen that the degree of assurance related to only one or two test results is very low. This, then, suggests that several tests should be performed before adjustments are made in the bitumen or aggregate percentages. Recognizing the dilemna facing the materials and testing engineer, this symposium was develped to explore alternate means of rapid measurement of bitumen content that may have a greater level of reliability, that is, lower standard deviation. It was hoped that a method or methods would be disclosed that would strike a proper balance between: (1) accuracy of test results, (2) cost of equipment, (3) speed with which the test can be performed, and (4) technician training required. To be of optimum value, such a method must be readily adaptable to field use as well as use in a central control laboratory.
Paper ID: STP48355S