Page 125 - Handbook of Thermal Analysis of Construction Materials

108 Chapter 3 - Formation and Hydration

Ramachandran and Beaudoin [51] followed the hydration of C AF
4
containing 5–30% gypsum, prepared at w/s ratios of 0.08, 0.13, 0.5, and 1.0
and exposed to a temperature of 25°C or 80°C. Techniques such as DTA,
XRD, and conduction calorimetry were used. Compounds were identified
by their characteristic thermal effects as follows. Gypsum indicated two
peaks in the range 150–200°C representing the stepwise removal of water.
At low concentrations, gypsum gave a single peak at about 150–160°C. Free
water gave an endothermal effect at 100°C. The hexagonal phase exhibited
an endothermal peak at about 160–175°C, the cubic form at 300–325°C.
Ettringite was identified by an endothermal peak effect at 110–125°C and
the low sulfoaluminate by an endothermal peak at about 200–210°C.
The relative amounts of unreacted gypsum and high sulfate alumi-
nate hydrate contained in samples cured for different times in a sample
hydrated at 25°C and a w/s ratio of 0.13 are given in Fig. 28. [53] Calculations
suggest that within 1 hour relatively more gypsum has reacted to form
ettringite in mixes containing larger amounts of gypsum. Conduction
calorimetric investigation also indicates that in the first 30 minutes a larger
amount heat of heat is developed in samples containing greater amounts of
gypsum. Although almost all gypsum has reacted at 7 hours in samples
containing 5–20% gypsum, substantial amounts are still present in the
mixture prepared with 30% gypsum. All samples show a general decrease
in the amount of ettringite after 3–7 hours of hydration and an increase in
the amount of low sulfoaluminate phase. It is generally believed that
ettringite begins to convert to the low sulfate form after all gypsum has been
consumed. This may be valid at low gypsum contents. At a 30% gypsum
level, although there is a decrease in the amount of ettringite and an increase
in the amount of low sulfate form, there is still a large percentage of
unreacted gypsum at 7 hrs. It appears that after this length of time the
reaction between ettringite and ferrite phase to form low sulfoaluminate
progresses at a faster rate than the reaction between gypsum and C AF to
4
form ettringite. Length measurements as a function of time in admixture
containing 30% gypsum show a higher expansion of 10.8% at 7 hrs
compared to only 2.9% for that containing 20% gypsum. A higher expan-
sion may result in a higher porosity and better availability of the C AF
4
surface for the reaction. It was also found that at very low w/s ratios and a
higher temperature the low monosulfate phase need not result from the
conversion reaction involving ettringite.
In Fig. 29, the thermal curves of C AF-CaSO •2H O mixes hy-
4 4 2
drated at a w/s ratio of 0.5 or 1.0 are given. Most samples hydrated for 2 days
at 25°C contain mainly the low sulfoaluminate. [51] The interconversions