Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Shape comparison between 0.4 υm to 2.0 υm and 20 υm to 60 υm cement particles



L Holzer, R Flatt, S.T. Erodgan, Jeffrey W. Bullard, Edward Garboczi


Portland cement powder has a wide particle size distribution, from approximately 0.1 υm to 60 υm. This wide powder size distribution arises via the grinding of much larger clinker particles in fact, several percent of the world s energy expenditures are consumed by this grinding. Many other powders are manufactured in the same way. An important question for cement, or indeed any other similarly ground particle, is: in what manner, if any, does particle shape depend on particle size? When one is modeling cement particles and their transformation via hydration into a three dimensional microstructure, one needs to know if any shape dependence on size needs to be addressed as one builds models. The shape of cement particles can affect the properties of cement paste (cement plus water), especially at early ages, where reaction can be controlled by nucleation on cement particle surfaces and where setting is controlled by percolation of a solid network that bridges cement particles. This paper describes how we have investigated this question for portland cement. For the same cement, we have used X-ray microcomputed tomography to examine the 3-D shape of particles in the 20 υm 60 υm size range, and a focused ion beam technique to examine the 3-D shape of cement particles found in the 0.4 υm 2.0 υm size range. Both size ranges were defined by sieving. Analyzing each set of data with a spherical harmonic technique, we are able to compute many shape-related parameters. By comparing various kinds of particle shape data for each size class, we are able to conclude that, within experimental uncertainty, the smaller size class particles, 0.4 υm 2.0 υm, tend to be somewhat more prolate than the 20 υm 60 υm size class. The practical effect of this shape difference on early-age hydration and setting were approximately assessed using the Virtual Cement and Concrete Testing Laboratory to simulate the set point of cement powders, all with the same chemistry, and having a variety of shapes: spheres, rectangular boxes with the same aspect ratios are the real particles, and the real particle shapes from the two size classes. We found that the two sets of real particle shapes have, within numerical uncertainty, a degree of hydration at set that is different from the spherical particles but similar to each other, indicating that the small but real shape difference between the real cement particles probably does not have a large effect on cement hydration. The box particles were found to give set points nearly identical to the real particles.
Powder Technology


Cement, Particle Size, Grinding, Shape, X-ray Microcomputed Tomography, Focused Ion Beam


Holzer, L. , Flatt, R. , Erodgan, S. , Bullard, J. and Garboczi, E. (2010), Shape comparison between 0.4 υm to 2.0 υm and 20 υm to 60 υm cement particles, Powder Technology, [online], (Accessed June 14, 2024)


If you have any questions about this publication or are having problems accessing it, please contact

Created June 14, 2010, Updated October 12, 2021