Consideration of corrosion in the presence of water has led to the postulation of damage by hydrogen produced in the reaction. For cases where the metal surface is covered with a protective layer, it is considered that some hydrogen ions diffuse through the layer and form hydrogen atoms beneath it. The consequence of the production of hydrogen in this location is to decrease corrosion resistance. Three types of harmful effects are illustrated.

Aluminum alloy 1100 corrodes uniformly in nearly pure water at rates which increase with temperature. The reaction is characterized by an initial period of relatively rapid corrosion, followed by a constant lower rate of corrosion. The slope of the linear part of the curve or the “corrosion rate” changes from about 0.02 to 2.5 mg/dm2-day (.01 to 1.3 mil/yr) from 50 to 200 C. Above 200 C intergranular attack occurs, with resultant more rapid penetration and deterioration of the metal. The corrosion rate can be reduced by the addition of an acid to the water. The solution pH of minimum corrosive attack on aluminum goes down with increasing temperature. It is of the order of 6.5 at 50 C and 3 at 300 C. Corrosion in slightly alkaline solutions (pH 8.5) is more rapid than in neutral water at all temperatures. At 100 C the difference shown in stagnant tests is very slight, but increases considerably if the temperature is increased to the order of 200 C. Presumably the effect of solution flow rate would be greater in the case of alkaline water than in neutral water. At 200 C the addition of dichromate ion increased the rate of corrosion. A small concentration of sodium silicate resulted in a higher corrosion rate at pH 9.4, while at pH 3.5 it reduced the rate significantly. Chloride ion (5 ppm) caused some pitting and increased the corrosion rate slightly. Electrical coupling to more cathodic materials, such as graphite, zirconium or stainless steel produced no undesirable effects in distilled water at 200 C. At 315 C, the corrosion life of samples was extended by such coupling. 6.4.2

OXIDE DISSOLUTION ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 130.184.234.34 Downloaded on 2015-03-16 to IP

CT-1943(a-19 43(CT-l943(l) CT-1943((3'-1943(CT-19 43(CT-l943(l) CT-1943(CT-1943(l) CT-1943(l) a-1943(1) (=T-1943(Cr-1943(CT-1943(CT-1943(') BR-549 CT-29 25(2) w-549 (6) CT-127 2(1) CT-127 2 (1) CT-1272(1) m-549 (6) m-549 (6) CT-27 55(2) a-26 59 (2) CT-l272(Cr-1272(CT-1272(CT-1272(CT-270 3 (2, CT-26 68 (2, CT-28 57(*) (3"-29 2 5(

In oxygen‐saturated distilled water at 70°, the rate and amount of corrosion during short exposure are influenced by experimental conditions. One noteworthy effect is that contamination of the water by the reaction increases the corrosion rate. Subsequent to the first several days, the amount of corrosion varies with the logarithm of the exposure time. This behavior holds for at least 180 days; it is believed to hold for as long as tests have provided reasonable data, the longest being about 650 days. These results are interpreted in terms of local film rupture and growth. A method of averaging the over‐all corrosion rate on the basis of cyclical local reactions is derived.