Alkali-aggregate research in New Brunswick

New Brunswick is located within the northeast-southwest trending Appalachia n Region. The basement rocks consist largely of metamorphosed sedimentary t ypes with some granitic intrusions and the composition of the natural grave ls reflects the bedrock types. Research into alkali-aggregate reaction (AAR ) problems in the province was sponsored initially solely by CANMET and mor e recently in association with the Department of Transportation. The resear ch consisted of (i) petrographic studies of aggregates, (ii) petrographic s tudies and case histories of existing concrete structures, and (iii) labora tory expansion testing of concrete and mortar bar specimens. Alkali-aggrega te reaction has been found to be one of the factors responsible for prematu re concrete deterioration in New Brunswick. Visual signs of Alkali-aggregat e reaction require up to 10 years to appear and the reaction takes up to 30 years to fully develop. Principal reactive rock types are greywacke, argil lites, and fine-grained volcanic rocks. The reactive component is thought t o be fine-grained quartz less than 100 mu m in size. Laboratory expansion t est results on concrete specimens are sensitive to alkali levels. Concrete prism tests (CSA A23.2-14A) with 5.5-5.4 kg/m(3) Na2O equivalent have been used to predict aggregate performance. By comparison the water soluble alka li contents of field concretes constructed since about 1970 have been found to range from 3.5 to 5 kg/m(3) Na2O equivalent. Accelerated mortar bar exp ansion test results (CSA A23.2-25A) do not correlate well with concrete pri sm expansion tests and indicate a much higher percentage of deleterious agg regates. Use of supplementary cementing materials such as fly ash and silic a fume have been found to be effective in long duration laboratory tests to control aggregate reactions.