Admiralty (CDA No. 443), aluminum brass, and 

 nickel brass, were attacked by parting corrosion in 

 degrees varying from slight to severe. The zinc con- 

 tent varied from 10 to 42% with the severity of 

 parting corrosion generally increasing with the zinc 

 content. Although the Arsenical Admiralty contained 

 about 30% zinc, the addition of about 0.03% arsenic 

 rendered it immune to parting corrosion. Because of 

 the 2% aluminum in aluminum brass and the 8% 

 nickel in the nickel brass, they were also immune to 

 parting corrosion even though they contained 20 and 

 40% zinc, respectively. 



3.2.3. Effect of Depth 



The effect of depth on the corrosion of the 

 brasses after 1 year of exposure in seawater is shown 

 in Figure 9. Although there was a slight tendency for 

 the brasses to corrode more slowly at depth than at 

 the surface, this slight decrease is not significant. 



3.2.4. Effect of Concentration of Oxygen 



The effect of the concentration of oxygen in sea- 

 water on the corrosion of the brasses after 1 year of 

 exposure is shown in Figure 10. The corrosion of the 

 brasses increased linearly, but slightly, with increasing 

 oxygen concentration. 



3.2.5. Stress Corrosion 



Two brasses, CDA No. 280 and 443, were 

 exposed in seawater while stressed at values equi- 

 valent to 50 and 75% of their respective yield 

 strengths for the periods of time and at the depths 

 given in Table 16. Neither alloy was susceptible to 

 stress corrosion at either depth, 2,500 or 6,000 feet, 

 after 400 days of exposure. 



3.2.6. Mechanical Properties 



The effects of corrosion on the mechanical pro- 

 perties of three brasses are given in Table 17. The 

 mechanical properties of Arsenical Admiralty were 

 not impaired, while those of Muntz Metal and nickel- 

 manganese bronze were impaired. The degree of 

 impairment increased with the time of exposure at 

 both depths, 2,500 and 6,000 feet. The degree of 



impairment in both alloys roughly paralleled the 

 severity of the parting corrosion. 



3.2.7. Corrosion Products 



The corrosion products which formed on cast 

 nickel-manganese bronze during 403 days of exposure 

 at a depth of 6,000 feet were analyzed by X-ray dif- 

 fraction, spectrographic, infra-red spectrophotometer, 

 and quantitative analyses methods. The corrosion 

 products were composed of cupric chloride 

 (C uCl 2 '2 H 2 O ) ; copper hydroxychloride 

 (Cu 2 (OH) 3 Cl); copper as metal, 35.98%; minor 

 amounts of aluminum, iron, silicon, and sodium; 

 chloride ions as Cl, 0.91%; sulfate ions as S0 4 , 

 11.53%; small quantities of an organic compound or 

 compounds present due to decomposed algae and 

 vegetative materials. 



3.3. BRONZES 



The chemical compositions of the bronzes are 

 given in Table 18, their corrosion rates and types of 

 corrosion in Table 19, their resistance to stress cor- 

 rosion in Table 20, and the changes in their mechan- 

 ical properties due to corrosion in Table 21. 



3.3.1. Duration of Exposure 



The effects of the duration of exposure on the 

 corrosion of the bronzes in seawater at depths, at the 

 surface, and in the bottom sediments are shown in 

 Figure 12. 



Since the corrosion rates of all the bronzes, 

 except those of alloys CDA No. 65 3 and 65 5 (silicon 

 bronzes), were so comparable, the average values for 

 any one time, depth, or environment were used to 

 prepare the curves in Figure 12. Because the cor- 

 rosion rates of the silicon bronzes were so much 

 greater than those of the other bronzes, they were 

 not averaged with the others. They are shown in 

 Figure 12 as a separate curve that includes the rates 

 for both depths as well as those for the surface. The 

 average corrosion of the bronzes in seawater and in 

 the bottom sediments was essentially constant with 

 increasing duration of exposure. Also, it was essen- 

 tially the same at the 2,500-foot depth as at the 



45 



