Q: Is there anything one can do, short of putting a slip joint between a slab and the masonry bearing wall that it rests on, to prevent breaking the bond to the mortar joint between them?

We built a 12x15-foot vault inside a building for fire protection of documents. The top slab has now curled up about 1/4 inch at the corners, leaving a gap. We expect to tuckpoint the joint, but it will not look as neat as originally. Slab curling doesn't appear to be a problem with slabs on grade.

Is this because it is not noticeable or because the subgrade controls the rate of curing better?

Is there such a thing as curling in footings?

I would appreciate any information or references on curling and concrete shrinkage.

A.: We haven't heard of any problem with curling of footings, but curling is fairly common in slabs on grade. It is not always noticeable, but it can be troublesome in slabs where furniture or high shelving rests partly on a curled corner or where truck traffic crosses a joint between curled slabs.The main cause is greater shrinkage on the top surface than on the bottom. Anything that can be done to keep the moisture content of the slab more nearly uniform from top to bottom will help minimize curling. In Section 11.11 of ACI 302.1R-80, "Guide for Concrete Floor and Slab Construction," there is a fairly complete list of measures that can be taken to minimize curling, some of which would apply to your problem. This useful publication of 46 pages is available from the American Concrete Institute, Box 19150, Detroit, Michigan 48219. The price is $25.25 ($18.50 to members of the Institute).

Reader Response:

The deflection is probably caused by the slab acting as a plate. For example, take an 8 1/2x11-inch sheet of cardboard backing from a writing tablet and cut it to 8 1/2 inches square (don't use corrugated paper, which is stiffer in one direction than the other). Set four books of equal size on edge to support the cardboard. Using your hand, load the cardboard downward as uniformly as possible. You should observe an upward deflection at the corners.

Plate action can cause upward curling of slab supported by masonry walls. Corners at A and B represent such curling, while C and D represent corners where curling has been restrained by anchoring (see text).

Concrete slabs constructed on masonry walls will behave the same way. In Figure 1, A and B represent two such corners, while C and D represent corners that have been anchored as discussed later. Normally all four corners would curl upward and undergo the cracking indicated in Figure 2. Park and Gamble (Reference 1) say that if slab corners are "not positively held down . . . [they] will lift off the support." Ferguson (Reference 2) recommends that "slabs should be anchored down (to a substantial mass of masonry . . .) and have reinforcing steel provided for the restraining moment thus developed." If the slab corner is attached incorrectly, however, it will cause horizontal cracks to develop in the wall.There are two possible solutions. The first is to anchor the slab corners into the masonry to a depth sufficient to prevent the slab from lifting the top of the wall and cracking it. Corner D of Figure 1 represents such anchorage. Corner C represents a slab that caused a horizontal crack at point E near the top of the wall because it was not anchored deeply enough. The other solution is to wait for the slab corners to deflect upward. Then either tuckpoint the joint or, if the slab loading can vary, provide a movable joint between the wall and the slab. Either solution can generate problems, so be careful. No matter which solution is used, special reinforcement is required in the corners of the concrete slab to prevent the kind of slab cracking shown in Figure 2 (below).

Figure 2:

Corner crack pattern observed in simply supported slabs


  • Park, R., and Gamble, W. L., Reinforced Concrete Slabs, John Wiley & Sons, New York, New York, 1980, pages 30-31.
  • Ferguson, R M., Reinforced Concrete Fundamentals, Fourth edition, John Wiley & Sons, New York, New York, 1979, page 35
  • Bruce Suprenant Department of Civil Engineering University of South Florida