Q.: In Problem Clinic in the February issue you made recommendations for concrete to be used for bins in which rock salt (sodium chloride) was to be stored. We are having major problems with cracking, spalling and surface weathering of concrete used in salt storage. In this service the salt is being stored in damp conditions and it contains magnesium chloride. We are very interested to note that you consider the corrosive action of magnesium chloride to be more serious than the action of sodium chloride.

It would help us with our research into the problem if you could expand your answer further, stating previous research and the reasons for your conclusions.

A.: The critical statement in the earlier Problem Clinic item is "Although there is some slight possibility of expansion cracking [from absorption of sodium chloride solution], the more serious possible effect is from magnesium chloride, often found in rock salt as an impurity. This may cause cracking, followed by corrosive attack on the reinforcing steel." This statement is based on an item in Table A3 of ACI 515, "Guide for the Protection of Concrete Against Chemical Attack by Means of Coatings and Other Corrosion Resistant Materials," published in the Journal of the American Concrete Institute, December 1966, pages 1305-1392. The information actually came from earlier sources.

Since you are handling damp salt the concrete may in effect contact a solution saturated with sodium chloride and containing magnesium chloride in some lower concentration. ACI 201.2R-77, in a table showing the effect of commonly used chemicals on concrete, lists magnesium chloride solutions as having a slow rate of attack on concrete at ambient temperature. F. M. Lea, in The Chemistry of Cement and Concrete (third edition, Chemical Publishing Company, New York, page 673) says: "Solutions of magnesium chloride of 2 percent concentration and upwards produce a gradual diminution in the strength of portland cement mortars," and "Very strong [about 30 percent] solutions destroy portland cement concrete."

More general comments are given by Hubert Woods, in Durability of Concrete Construction, (American Concrete Institute Monograph Number 4, 1968, page 127). He says "The chlorides and nitrates of ammonium, magnesium, aluminum and iron all attack concrete, with those of ammonium being the most harmful." This is a summary of information based on the F. W. Locker and H. Pisters article, "Assessment of Waters Aggressive to Concrete," Zement-Kalk-Gips (Wiesbaden), Volume 17, 1964, pages 129-136.

Data showing considerable expansive effects of magnesium nitrate and magnesium sulfate solutions were reported by William H. Kuenning in "Resistance of Portland Cement Mortar to Chemical Attack--A Progress Report," Portland Cement Association Research Bulletin 204 (reprinted from Highway Research Record Number 113 [1966], pages 43-87). On the basis of several literature references the author also said "Most magnesium salts are destructive to concrete. They leach out calcium hydroxide to form the soluble calcium salt and deposit magnesium hydroxide, which is far less soluble than calcium hydroxide. They are also said to react with the other cement hydration products to form such compounds as magnesium hydroxide, soluble calcium salts, silica and alumina; the latter two possibly react with the magnesium hydroxide to form hydrated magnesium silicates and magnesium aluminate."

Most of the references cited here contain additional bibliographic material that may be useful to you.