Determination of Calcium Sulfate Dihydrate and Hemihydrate in Cement

Introduction

Cement consists primarily of calcium silicates, calcium aluminates, and calcium aluminoferrites. Besides these main constituents, various additives are used in small quantities.

One important additive is gypsum in the form of the dihydrate (CaSO4·2H20) or the hemihydrate (CaSO4 ·½ H20). The gypsum is ground with the clinker and serves to retard the setting and hardening of the cement. Without gypsum, hydrated calcium aluminate crystallizes within about 10 min; gypsum retards this process by several hours or even days depending on the gypsum content. In the quality control of cement, it is therefore important to know the exact content of the CaSO4 dihydrate or hemihydrate.

 

Experimental Details

Thermogravimetry is good analytical method for determining the gypsum content. The dihydrate and the hemihydrate lose their water of crystallization on heating. In the case of the dihydrate, the elimination takes place in two steps (via the hemihydrate), and with the hemihydrate in just one step. The experimental difficulty is that both reactions occur close together in the same temperature range and, depending on the conditions, overlap. This makes quantification of the two types of gypsum difficult or even impossible. To achieve a separation of the two types of gypsum, a sample must be heated in an almost completely sealed crucible. This shifts the dehydration of the hemihydrate to higher temperatures thereby improving the separation of the two processes. The gypsum content in cement is usually in the percent range. From stoichiometric considerations, dehydration of the dihydrate is expected to give a total mass loss of 20.9%, made up of steps of 15.7% (dihydrate → hemihydrate) and 5.2% (hemihydrate → anhydrous form, based on the mass of the dihydrate). The dehydration of the hemihydrate is expected to give a mass loss of 6.2%.

Assuming, for example, a hemihydrate content of 1% in cement, dehydration of the hemihydrate in 1 mg of cement gives rise to a mass loss of 0.61 µg. For this reason, the samples analyzed must be as large as possible. Since only aluminum crucibles can be sealed, the experiments were performed using 100-µL aluminum crucibles sealed with a lid with a 50-µm hole.

Samples of pure calcium sulfate dihydrate, calcium sulfate hemihydrate and a mixture of the two were first analyzed. This was followed by three cement samples with known dihydrate and hemihydrate contents. 

 

Results

Figure 1 displays the measurement results of the pure samples and their mixtures. The results show that both the dihydrate and the hemihydrate samples already contained small amounts of the anhydrous form. Based on stoichiometry, the purities of the two hydrates were 93.1% (hemihydrate) and 94.5% (dihydrate). With these purities, a mass loss of 12.8% is expected for the mixture. This value agrees well with the measured value of 12.7%.

TGA of Pure DH, HH and a Mixture 

Figure 2 shows the measurements performed on the three cement samples containing different contents of the dihydrate and hemihydrate. The two mass loss steps that occur during the dehydration can be clearly seen. In contrast to the TGA curves of the pure calcium sulfate hydrates, a continuous loss of mass is observed in both cement samples over the whole temperature range measured. This is overlaid by the two mass loss steps of the dehydration. For this reason, the mass loss steps of the dehydration were determined using the tangential step evaluation program. Alternatively, the peaks from the first derivative of the mass loss curves could also be evaluated. 

Conclusions

The contents of calcium sulfate dihydrate and calcium sulfate hemihydrate can be determined in cement with good accuracy by thermogravimetry. A quantitative determination is only possible if the samples are measured in an almost hermetically sealed crucible. This is most easily done using aluminum crucibles sealed with a lid with a 50-µm hole. If a pierced lid is not available, the same effect can be achieved by placing a grain of sand (about 20 µm in diameter) on the rim of the crucible before sealing it.

Determination of Calcium Sulfate Dihydrate and Hemihydrate in Cement | Thermal Analysis Application No. UC 264 | Application published in METTLER TOLEDO Thermal Analysis UserCom 26