Increasing concerns regarding litigation and terrorism provide a strong dual motivation to decrease high explosives usage in the construction industry. This paper provides parameter considerations and initial guidelines for the application of expansive fracture agents, typically used for concrete and soft rock removal. This approach may be especially appropriate near environmentally and historically sensitive sites. Thirty-three unreinforced blocks (approximately a cubic metre each) of varying strengths, composed of sand, cement, and fly ash, were tested under various temperature environments, with differing expansive agents, confinement levels and post-cracking treatments. Cracking characteristics such as crack initiation and crack expansion were analysed. Although the performance of expansive cement was dependent on a highly complex set of variable interactions, higher ambient temperatures, higher agent mixture temperatures and chemical configuration designed for colder temperatures decreased the time to first crack and hastened the extent of cracking. Conversely, higher strength material required more time to first crack, as well as an extended time to achieve a 25 . 4 mm wide crack. Manual interference with the normal material volume expansion slowed the cracking process but did not truncate it, while the manufacturer's recommendation to introduce water post-cracking actually reduced and slowed the extent of cracking.

In many tunneling and excavation projects, free-field vertical ground movements have been used to predict subsidence and empirical limits have been employed to evaluate risk. Validity of such approaches given the reality of two-dimensional ground movements and the influence of adjacent applied loads has been largely unknown. This paper employed analytical and largescale experimental efforts to quantify these issues, in the case of a reinforced concrete frame structure adjacent to an excavation. Nearly half of all soil and building movements occurred prior to installation of the first tie-back, even when conservative practices were applied. Freefield data generated a trough half the size of that recorded near the building frames. Empiricallybased relative gradient limits generally matched the extent and distribution of the damage. Application of various structural limits also generally reflected global experimental response but did not fully identify local damage distribution. Fully free-field data or failure to include accurate two-dimensional soil displacements under-predicted building response by as much as 50% for low-rise concrete frames without grade beams.

No abstract