EFFECT OF HEAT FLUX ON THE CORROSION OF ALUMINUM BY WATER. PART IV. TESTS RELATIVE TO THE ADVANCED TEST REACTOR AND CORRELATION WITH PREVIOUS RESULTS

Abstract

The corrosion of 8081 and X-8001 aluminum alloys and the resultant formation of an adherent corrosion product on the corroding surface were investigated under conditions (except radiation) comparable to those that will exist on the surface of fuel-element cladding during operation of the Advanced Test Reactor. Since previous experiments indicated that corrosion penetration of the aluminum clad was unlikely during a reactor cycle provided the water chemistry is properly controlled, most of the studies in this investigation were concerned with the effect of variables on the rate of formation of corrosion- product films. These films have low thermal conductivity and can be a major factor in producing high fuel-element temperatures which lead to fuel-plate instabilities. The experimental procedures were the same as used in a similar study for the High Flux Isotope Reactor, but the ranges of variables investigated were greater than in previous studies. The 6061 and X-8001 alloys corroded to the same extent under the same test conditions until the corrosion product (boehmite) that formed on the surface bccame thick enough to spall spontaneously from the surface, usually about 2 mils thick. Spallation from the surface of the 6061 alloy was always accompanled by localized attack of the underlying metal, whereas only uniform attack was observed with the X-8001 alloy under all conditions. Previously it was shown that the 1100 aluminum alloy behaved like the 6061 alloy in all respects. All three alloys developed corrosion-product coatings at the same rate when tested under the same conditions. The thermal conductivity of the corrosion product was determined to be 1.3 Btu hr/sup -1/ ft/ sup -1/ ( deg F)/sup -1/. The pH of the water was an important variable in determining the rate of corrosion-product buildup on aluminum. Under the same conditions the rate of oxide formation was 2.7 times greater when the pH of the water was 5.7 to 7.0 than when the pH was 5.0 (with nitric acid); pH was not a significant variable within the former range. Changes in pH during a test resulted in abrupt changes in the rate of oxide formation and presumably also in corrosion. The data indicated that up to the point of film spallation about 70% of the aluminum that was oxidized remained on the surface as boehmite regardless of test conditions; the rest was lost to the water. The rate of oxide formation on the specimens decreased with exposure time and increased as the surface temperature increased. The controlling temperature was that at the aluminum oxide--water interface (surface temperature). Considering only the data acquired with the water at a coolant pH of 5.0 and at heat fluxes between 1 and 2 x 10/sup 6/ Btu hr/sup -1/ ft/sup -2/, the thickness of oxide built up on the corroding surface was described reasonably well by the empirical equation X = 443 theta / sup 0.778/ exp (-4600/K), where X is corrosion-product thickness in mils, theta is time in hours, and K is surface temperature in degrees Kelvin. With pH values between 5.7 and 7.0, the same relation applied except that the first coefficient in the above equation was 1200 instead of 443. Within the range of 1 to 2 x 10/ sup 6/ Btu hr/sup -1/ ft/sup -2/, heat flux was unimportant except in the manner in which it influenced surface temperatures. Below 1 x 10/sup 6/ Btu hr/sup -1/ ft/sup -2/, heat flux per se was a significant variable; considerably lower oxide thicknesses were observed than calculated by the above correlation. The results of the experimental program indicate that the 6061 aluminum alloy, which has mechanical strength superior to either 1100 or X-8001, will have adequate corrosion resistance for use as cladding for both the Advanced Test Reactor and the High Flux Isotope Reactor fuel plates, provided the pH of the water is maintained at 5.0 with nitric acid. However, the increase in fuel-plate temperatures resulting from the buildup of the corrosion-product layer must be evaluated carefully to determine its effect of the mechanical and irradiation behavior of individual iuel plates and on the overall fuel-element stability. (auth)