Mat Foundation Pour Exceeding the 70 °C Peak Hydration Limit
A 30-storey tower rests on a 2,400 m³ mat foundation, 2.2 m thick, cast in one continuous pour in July with an ambient temperature of 33 °C. The approved mix contains 380 kg/m³ of binder (Type I/II cement with 15 % slag replacement) at w/b = 0.42, and concrete arrives on site at 30 °C. The specification limits peak hydration temperature to 70 °C, surface-to-core differential to 20 °C, and cooling rate to 3 °C/h. Insulation blankets were fixed to the side forms only; the top surface received curing compound and nothing else. At 40 hours the thermocouple array records a core temperature of 79 °C and a top-surface temperature of 53 °C, and the differential is still widening at about 0.8 °C/h. Two of the eight sensors sit within 150 mm of the top surface and read 49 °C. The second mat, identical in size, is scheduled in three weeks and the same mix and method are proposed.
1. Assess the thermal state at 40 hours against the specification and state which risk governs the engineering response.
Read the worked answer
Three checks, each with a different consequence:
- Peak temperature. Core 79 °C against a 70 °C limit - exceeded by 9 °C. The limit exists to control delayed ettringite formation, which is a long-term, irreversible durability problem: once the heat cycle has damaged the paste, no amount of curing or later protection recovers it. This is the governing risk for the asset's life.
- Differential. 79 − 53 = 26 °C against a 20 °C limit, and still widening at 0.8 °C/h, so the peak differential has not yet occurred and may reach roughly 30 °C if nothing changes. Surface-and-core separation drives surface tensile stress and early-age cracking, which is the governing immediate risk because the pour is still live.
- Cooling rate. Not yet meaningful while the section is still heating, but it becomes the controlling parameter once the core turns over; the 3 °C/h limit is what prevents a thermal shock crack from the outside in.
The near-surface sensors at 49 °C are the most useful for cracking assessment, because the steepest gradient sits in the outer 150 mm.
Conclusion: the placement is non-compliant and the differential is actively worsening, so action is required now; the peak-temperature exceedance is already locked in and must be handled as a durability issue through the thermal control plan and later condition monitoring.
2. What immediate actions would you direct on the pour in progress, and what would you deliberately not do?
Read the worked answer
The core is already at 79 °C, so the peak cannot be reduced - the only controllable variables are the gradient and the subsequent cooling path.
Actions now:
- Insulate the top surface. Wet burlap plus the same blanket specification used on the sides, immediately. Raising the surface from 53 °C toward the core closes the differential; adding insulation does not raise the peak further in any meaningful way because the heat is being generated in the interior and the top is a heat-loss boundary.
- Do not remove side forms early. Forms are part of the insulation. Keep them at least 7 days and strip only when the core-to-ambient gradient permits.
- Control cooling actively. Once the core turns over, target ≤ 3 °C/h; if ambient drops at night, add blankets rather than letting convective cooling accelerate.
- Maintain curing moisture under the insulation so the surface does not dry and shrink while it is hot.
- Increase monitoring frequency to hourly at the two near-surface sensors and add a sensor at 75 mm depth to define the real gradient.
What not to do: do not water-cool or spray the hot surface to bring it down quickly - that manufactures exactly the differential the specification is trying to prevent, and could crack the mat within hours. Do not accept the placement as compliant or waive the limits on the basis of the 40-hour reading alone.
3. The second mat is due in three weeks with the same mix and method. Set out the changes you would require and the basis for each.
Read the worked answer
The pour schedule is not the problem; the heat generated per cubic metre and the boundary condition are.
Reduce the heat source. Cut total binder and change its composition: a 50–60 % slag or 25–30 % Class F fly ash mix at the same w/b will cut adiabatic temperature rise substantially and slow the release, which lowers the peak. Reducing binder below about 320 kg/m³ is only acceptable if the 28-day strength and durability requirements are re-verified by trial mix - do not trade a thermal crack for a strength shortfall.
Reduce the placing temperature. Target ≤ 27 °C by chilling mix water, cooling aggregates, and pouring at night. Lower starting temperature shifts the entire curve down roughly one-for-one, which is the cheapest single lever available.
Change the boundary condition. Insulate the top from the start, not just the sides, and specify a written thermal control plan with trigger temperatures and pre-agreed responses, so the site is not improvising at 02:00.
Add active control if the predictions still exceed 70 °C. Embedded cooling pipes in the core, or a two-lift pour with a designed construction joint, are the standard fallbacks; a trial pour or a heat-flow model of the section should be used to confirm that the revised mix plus chilled concrete actually lands below 70 °C before the mat is cast.