Author's School

School of Engineering & Applied Science

Author's Department/Program

Mechanical Engineering and Materials Science


English (en)

Date of Award

Spring 4-25-2013

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Shankar M.L. Sastry


As large countries such as China begin to industrialize and concerns about global warming continue to grow, there is an increasing need for more environmentally friendly building materials. One promising material known as a geopolymer can be used as a portland cement replacement and in this capacity emits around 67% less carbon dioxide. In addition to potentially reducing carbon emissions, geopolymers can be synthesized with many industrial waste products such as fly ash.

Although the benefits of geopolymers are substantial, there are a few difficulties with designing geopolymer mixes which have hindered widespread commercialization of the material. One such difficulty is the high variability of the materials used for their synthesis. In addition to this, interrelationships between mix design variables and how these interrelationships impact the set behavior and compressive strength are not well understood. A third complicating factor with designing geopolymer mixes is that the role of calcium in these systems is not well understood.

In order to overcome these barriers, this study developed predictive optimization models through the use of genetic programming with experimentally collected set times and compressive strengths of several geopolymer paste mixes. The developed set behavior models were shown to predict the correct set behavior from the mix design over 85% of the time. The strength optimization model was shown to be capable of predicting compressive strengths of geopolymer pastes from their mix design to within about 1 ksi of their actual strength. In addition to this the optimization models give valuable insight into the key factors influencing strength development as well as the key factors responsible for flash set and long set behaviors in geopolymer pastes.

A method for designing geopolymer paste mixes was developed from the generated optimization models. This design method provides an invaluable tool for use in future geopolymer research as well as commercial development of geopolymer products.


This work is not available online per the author’s request. For access information, please contact or visit

Permanent URL: