The complex bases for the cracks are resolved in this month's finale, which includes the key factor that changes speculation into reality.

A meeting ensued that was attended by the owner, ready-mix concrete supplier, design engineers, admixture supplier, Type K cement manufacturer, plaintiff's petrographic expert, and a number of attorneys representing some of the participants.

The ready-mix concrete supplier reported that he had a separate investigation done by a technologist/petrographer well-versed in concrete made using Type K cement. The following was concluded from that investigation: the concrete met the mixture design requirements; the cause of the problem was excessive expansion that will be followed by continuing expansion; there is a probability that the concrete will soon self-destruct as a consequence of the continuing expansion; and the plaintiff's expert was incorrect in his evaluation and interpretation of the cause of the problem.

The admixture supplier reported that the admixture contained triethanolamine, and the amount of admixture sold to the ready-mixed concrete supplier correlated to the amount required by the mixture design and the volume of concrete used for the floor slabs, and that the plaintiff's petrographer was incorrect in his evaluation.

The manufacturer of the Type K cement reported that he did not originally concur with the plaintiff's investigator's explanation. So he conducted a laboratory study to evaluate the admixture's effects on concrete by comparing concrete expansion without the admixture to the admixture at the dosage rate required by the specified mixture design, and at double and triple that dosage. The laboratory study used a restrained expansion test, then called the Rubin-Bar Test, which now is a standard mortar bar prism expansion test method outlined in ASTM C878, “Standard Test Method for Restrained Expansion of Shrinkage-Compensating Concrete.”

The Rubin-Bar Test demonstrates the effects of increasing admixture dosage on the magnitude of decreasing expansions.
The Rubin-Bar Test demonstrates the effects of increasing admixture dosage on the magnitude of decreasing expansions.

The mortar bar study demonstrated that at the dosage rate required by the mixture design, the expansion is an acceptable 85% of the non-admixed concrete; at a double dosage the expansion is an unacceptable 55% of the non-admixed concrete; and at a triple dosage the expansion is an unacceptable 10% of the non-admixed concrete. The laboratory results confirmed the plaintiff's expert's conclusion.

It is interesting that two different laboratory investigators, each very knowledgeable about concrete and the use of Type K cement, widely differed in their interpretations of what happened. Apparently, one did not have all of the data needed to provide a realistic picture of cause and effect, and committed a gross error. It also is interesting that the admixture supplier's sales records supported the total amount of admixture reported by the concrete supplier to have been used in the concrete project. The source of the excessive amount of triethanolamine in the troubled concrete remains unresolved.

The concrete industry benefited from this experience and now cautions are provided about establishing the relationship of admixtures to concrete expansion when using shrinkage-compensating cements with ACI 223, “Standard Practice for the Use of Shrinkage-Compensating Concrete.”