Next: Acknowledgements Up: Main Previous: Qualitative and quantitative
A number of years ago, the progress of CEMHYD3D was such that bulk cement chemistry was insufficient to track the course of hydration accurately. The phase composition within thousands of particles was then measured by a combination of scanning electron microscopy and X-ray microprobe mapping and then statistically built into model 3-D particles [5]. These model particles were still only spherical, however, even though 2-D cross-sections of cement particles, seen in scanning electron microscopy, were clearly not circular. Now this shortcoming can be overcome. The actual 3-D shapes of cement particles have now become available, at least for one kind of cement. Future work is planned at X-ray microtomography sites to collect more cement image data at 1 µm per voxel or smaller resolutions.
It is interesting to see these shapes, "dredged up from the deep" so to speak, but it is even more interesting to see what effect shape has on cement and cement paste and eventually concrete properties. There is some evidence that different cements have different water demands − this may be due to shape and thus surface area differences. Any physical property differences, between the same cement ground in a ball mill versus a roller mill, should be due to the almost certain particle shape differences between cement powders made with these two processes. Of course, there may also be changes in the gypsum form due to temperature differences between the grinding techniques. The effect of different grinding aids and grinding time on particle shapes, and how the shapes are distributed with respect to size, are interesting questions, which, at the 5 µm particle size level or above, can now be explored. Increased use of cement particle shape data, as well as quantitative particle size distribution measurements [10], should greatly accelerate the development of the computational and experimental materials science of cement and concrete.