Asphalt concrete pavement is subjected to several damaging actions from traffic loads, water (from precipitation and/or groundwater sources), and temperature. The durability of the asphalt-aggregate mixture, its ability to withstand these damaging actions for long periods, is a very important engineering property. While the durability of the asphalt-aggregates mixture depends on several factors such as the mixture's properties, construction methods, traffic loads and environmental conditions, they have to be evaluated to predict their field performance. Based on mixture evaluations, the mixtures that fail the test would have to be modified by additives or by changing the materials. The first objective of this thesis was to evaluate asphalt-aggregate mixtures for water damage using the Environmental Conditioning System (ECS), and rank the asphalt and aggregate types based on water sensitivity. The second objective was to relate the ECS ranking of the asphalt and aggregate types to Oregon State University (OSU) and University of Nottingham, UK (SWK/UN) wheel tracking test results, and to Net Adsorption Test (NAT) results. The third objective was to evaluate open-graded mixtures and rubber modified mixtures for water sensitivity using the ECS. The ECS test results indicate that performance ranking of mixtures by asphalt type or aggregate type alone cannot be made for the ECS test results due to the significant interaction between asphalt and aggregate. Water sensitivity in the ECS is significant for combinations of asphalt and aggregate. The ECS test results have shown that ECS performance ranking after one cycle is not statistically significant and does not correlate with ranking after three cycles. The results show that the ECS test program has similar aggregate rankings to those of the NAT and SWK/UN test program, while good agreement exists between SWK/UN wheel tracking results and the NAT test program results. However, poor agreement exists between the OSU wheel tracking results and those of the other two tests. Poor or very little agreement exists among the wheel tracking test results, ECS, and NAT test results in terms of asphalt type rankings. When considering the comparisons of materials ranking by different test procedures, one must keep in mind that the mechanisms leading to varying "performance" are not the same. The testing reported herein was aimed at measuring water sensitivity, but all the tests do not do so directly. The NAT procedure addresses only the potential for stripping (adhesion) and is not capable of evaluating cohesion loss. The other tests (ECS, OSU and SWK/UN wheel tracking) included all the mechanisms simultaneously, and these provided a gross effect without clearly separating the cause of failure in each case. Open-graded mixtures used by Oregon Department of Transportation (ODOT) performed well in the ECS in terms of water sensitivity. In the ECS evaluation, six mixtures passed the criteria of 75 % established for Indirect Retained Strength (IRS) test by ODOT, and one mixture was marginal. However, only one mixture passed the IRS evaluation, and another mixture was marginal. This confirms that the IRS test is a very severe test and is not suitable for water sensitivity evaluation of open-graded mixtures. Finally, the IRS test evaluation would suggest that these mixtures would fail prematurely after construction, but all of these mixtures have been used in projects which have been in service for more than three years with no visible signs of distress, or failures. / Graduation date: 1994

The Environmental Conditioning System (ECS) was designed to evaluate the water sensitivity of asphalt concrete mixtures. The ECS subjects asphalt concrete specimens to a series of conditioning cycles including water flow, elevated and/or lowered temperature, and repeated axial loading. The purpose of this research was to: (1) evaluate the ECS test apparatus and procedure, and (2) determine whether the ECS can identify asphalt concrete mixtures that will perform well, or poorly, in the field with regard to water sensitivity. Twelve primary field test sections were identified. For each section, specimens were prepared in the laboratory using the original mix design (or the mix design identified by extraction), and the original aggregates, asphalts, and admixtures. Specimens were tested using two procedures: the ECS and the Oregon State University (OSU) wheel tracker. Field cores were used to evaluate in-situ mixture performance. Nine additional mixtures that have historically experienced water damage were tested in a limited secondary test program. Analyses were performed to determine the mixture properties that were significant in the prediction of mixture performance in the ECS. Mixture type was consistently the most significant predictor of ECS modulus ratio (change in mixture stiffness), degree of visual stripping, and binder migration, which were the performance indicators for water sensitivity evaluated in the ECS. Additional analysis indicated the existence of correlations among the ECS response variables. Significant correlations were found between the coefficient of water permeability and the degree of visual stripping; and between specimen deformation and the degree of visual stripping and binder migration. Mixture performance was compared between the ECS and the OSU wheel tracker and the field. Results indicate that the ECS test procedure can distinguish the relative performance of mixtures, with regard to water sensitivity, and mixture performance in the ECS correlates well with performance in the OSU wheel tracker. No correlation was found between mixture performance in the ECS and mixture performance in the field for the primary test sections. However, the primary field sections are relatively young, and water damage is expected to manifest itself in the future in those pavements identified as water sensitive by the ECS. The ECS predicted failure in the secondary mixtures which were identified as having had poor performance with regard to water sensitivity. / Graduation date: 1994