This section is from the book "Cyclopedia Of Architecture, Carpentry, And Building", by James C. et al. Also available from Amazon: Cyclopedia Of Architecture, Carpentry And Building.
Prof. Chas. L. Norton made several experiments with concrete bricks, 3 by 3 by 8-inch, in which steel rods, sheet metal, and expanded metal were imbedded. The specimens were enclosed in tin boxes with unprotected steel, and were exposed for three weeks. One portion was exposed to steam, air, and carbon dioxide; another to air and steam; another to air and carbon dioxide; and another was left in the testing room. In these tests, Portland cement was used. The bricks were made of neat cement of 1 part cement and 3 parts sand; of 1 part cement and 5 parts stone; and of 1 part cement and 7 parts cinders. After the steel had been imbedded in these blocks three weeks, they were opened and the steel examined and compared with specimens which had been unprotected in corresponding boxes in the open air. The unprotected specimens consisted of rather more rust than steel; the specimens imbedded in neat cement were found to be perfectly protected; the rest of the specimens showed more or less corrosion. Professor Norton's conclusions were as follows:
1. Neat Portland cement is a very effective preventive against rusting.
2. Concrete, to be effective in preventing rust, should be dense and without voids or cracks. It should be mixed wet when applied to steel.
4. Cinder concrete, if free from voids and well rammed when wet, is about as effective as stone concrete.
5. It is very important that the steel be clean when imbedded in concrete.