A public-private partnership has been chosen to ignite the introduction of RPC into the United States construction market. This research and development project is being conducted under the Construction Productivity Advancement Research (CPAR) program of the US Army Corps of Engineers. The project was initiated in the fall of 1994 and it will run for three years. The program goal is to verify product integrity and gain industry acceptance and commercialization by developing and demonstrating the technical and economic viability of RPC for producing culvert/sewer pipes, pressure pipes and piles. T h e primary technology transfer has been completed, US component material source identification has been brought into action and material property verification has been initiated. Other US products development efforts have been initiated. These applications include : 0 0 spun cast concrete poles, impact resistant railroad ties and grade crossing planks.

In order to enable rational and safe design with high performance concrete recommendations for this material are necessary. In the Netherlands an extended research program has been carried out focusing on aspects like behaviour in compression at various loading rates, shear friction in cracks, in-plane loading of cracked reinforced elements, splitting effects in the anchorage zone of prestressing strands, joints between precast columns, and creep. Furthermore trial casts have been carried out in order to get more experience with HPC at the building site. A four storey office building was completely built in HPC. During construction the temperature of the hardening concrete was measured at many locations, in order to investigate the development of temperature stresses and to get indications of the cracking probability. More-over a section of a box girder bridge was cast as an exercise for the construction of a 160 m span bridge in 1996. Both the labora-tory experiments and the site trials raised the confidence in suc-cessful applications of high performance concrete.

Reinforced concrete columns are typically made with higher strength concrete than are the floor slabs that they support. In construction, the slab is usually cast continuous through the region of the slab-column joint. As a result, load in the column above the slab must pass through a layer of weaker slab concrete before reaching the column below the slab. The column-slab joint may be viewed as a “sandwich” column, with high strength concrete above and below a layer of lower strength concrete. In design, the effective column concrete strength is based on a special weighted average of the column and slab concrete strengths. Because of confinement, the slab concrete in the joint region is assumed to be capable of carrying stresses well in excess of its specified strength. This confinement is, in turn, affected by gravity loading of the slab. Existing design procedures are based on tests of slab-column joints in which no load was applied to the slabs. This paper presents the results of a series of tests on interior column-slab joints in which service level loads were applied to the slabs prior to loading the columns. The major conclusions of this study are: (1) tests of sandwich slab-column joints with unloaded slabs consistently overestimate the strength of the connection and (2) the AC1 3 18-89 provisions for interior column-slab joints are unconservative for high ratios of column to slab strength and/or high ratios of slab thickness to column size.

Following is a brief overview of the techniques employed in developing the first generation of High Performance Concrete (from 50 to 130 Mpa/l9 000 psi), then the second generation and, most recently, the generation of Reactive Powder Concrete (from 200 to 800 Mpa/120 000 psi), the authors highlight the originality of the French approach as it has evolved, within the construction industry, over the past ten years. The basic principles underlying this originality are focused on : - high performance rather than high strengh, since improvements in other mechanical, physical and chemical properties have become, for many structural applications, crucial in the choice of construction materials, - application beyond sophisticated structures exclusively to encompass more basic construction uses or even many small, pre-fabricated structural elements, - a global analysis (design, development, maintenance) and a << systems B approach to construction that serves to emphasize the economic considerations behind High Performance C o n c r e t e . The second, and most detailed, part of this paper provides specific examples of the French approach through a discussion of not only : bridges and tunnels sized for heavy loads, major building projects, industrial structural framework (nuclear plant, offshore platform, oil tanker, etc), but also : short and medium-span bridges, small-scale, prefabricated components used in construction and public works, foundation work and structural repairs, etc. The third part, based on ten years of experience acquired through rather varied applications from French industry, provides an outlook on future development prospects and suggests new domains, both within or outside the field of construction.

Moisture migration in non-uniformly heated concrete is a complex phenomenon. It depends upon many factors, both intrinsic to the concrete mix and its local environment. At temperatures above 100°C pore vapour pressures dominate the mass transfer behaviour and lead to creation of dry zones containing superheated steam and zones of excessive wetness and physical saturation where condensation has occurred. Spalling of concrete, in fire, is strongly related to the water content of concrete at the time of heating and its moisture flow properties. During heating, as the temperature rises, the free water, contained in the porous structure of concrete, will expand whilst sustaining an increasing saturated vapour pressure. The continuous expansion of water together with the moisture flow frequently leads to physical saturation of the pores. Further heating will then generate additional strains in the solid envelope surrounding the pores and can lead to cracking and hydraulic fracture of the solid skeleton. High strength concrete is particularly vulnerable to this behaviour because of its inherent, low porosity, low permeability to water flow and high percentage of initial pore saturation. This paper describes numerical/theoretical modelling procedures, for the prediction of temperature-dependent moisture flow in non-uniformly heated concrete. The flow is considered to be governed dominantly by the pore pressures. A mathematical description is also provided to help understand the spalling process caused by the hydraulic fracture of the solid skeleton during heating of the water in saturated pores.