behavior for the fog-cured concrete is 

 consistent with existing knowledge. How- 

 ever, the strength gain behavior for the 

 ocean-cured concrete is unusual and 

 needs discussion. 



When a mass of concrete the size of 

 a control cylinder is placed in the ocean 

 under high hydrostatic pressure, water 

 fills the larger size voids within a period 

 of several days. The smaller size voids 

 become filled over a much longer time 

 period. For example, 6 x 12-inch (152 x 

 305-mm) cylinders under a pressure head 

 of 550 feet (168 m) were still absorbing 

 water after S,^ days (Ref ^). Saturated 

 concrete, that is, concrete whose voids 

 are mostly filled with water, has been 

 found previously to have a strength 

 reduction of 10% compared to companion 

 unsaturated concrete. This strength 

 reduction was observed from two 

 different types of tests: (1) specimens 

 were exposed to a pressure head of 1,125 

 feet (343 m) for 7 days and then tested 

 under uniaxial compression in a labora- 

 tory environment (Ref 1, 5); and (2) 

 specimens were placed under a pressure 

 head of 20,000 feet (6,096 m) for 60 days 

 and then loaded axially while in the 

 hydrostatic environment (if the concrete 

 was totally saturated, which was the 

 assumed condition, then this test was a 

 uniaxial compression test) (Ref 4). The 

 cause of these strength reductions was 

 most likely from pore pressure buildup. 

 During uniaxial loading, the water pres- 

 sure in some of the pores rose slightly, 

 thereby placing an additional component 

 of tensile strain within the specimen that 

 reduced the tensile strength in the radial 

 direction and, consequently, reduced the 

 compressive strength in the axial direc- 

 tion. 



In Figure 10, when concrete speci- 

 mens were placed in the ocean, a 

 decrease in strength of about 10% 

 occurred. At 1.6 years, ocean-cured con- 

 crete showed a compressive strength 

 that was still less than the 28-day fog- 

 cured strength. Two alternative paths 



are shown in estimating the relationship 

 between the time the concrete was 

 placed in the ocean and 1.6 years. For 

 the lower path, which intersects the 

 datum point at 1.6 years, the initial 

 strength reduction when placed in the 

 ocean had to be greater than 10%. The 

 higher path assumes that the data at 

 time 0.25 years to be accurate and those 

 at 1.6 years to be a "low" result. The 

 data at 1.6 years are from one concrete 

 block, producing four cores with a 

 coefficient of variation of ^.0%. 



At 5.6 years, the cement was com- 

 pletely hydrated for both the fog-cured 

 and ocean-cured concrete as determined 

 from x-ray diffraction analysis of the 

 concrete. Hence, the rate of strength 

 gain of the concrete, or the slope of the 

 curve, should be zero. However, strength 

 changes could be occurring due to other 

 chemical composition changes. 



Also, at 5.6 years, the ocean-cured 

 concrete showed a compressive strength 

 15% less than that of the fog-cured 

 concrete. This strength reduction had to 

 be due to a different reason than satura- 

 tion effect because at this advanced age, 

 both the fog- and ocean-cured concretes 

 are known to be saturated (Ref 4). 

 Therefore, the cause for the lower 

 strength of the ocean-cured concrete as 

 compared to the fog-cured concrete can 

 be speculated as mainly due to the 

 presence of seawater. Magnesium ions in 

 seawater replace some of the calcium 

 ions in calcium silicate hydrate (tobor- 

 morite gel). This causes the formation of 

 magnesium silicate hydrate, which is 

 more brittle than calcium silicate 

 hydrate (Ref 6). Specimens of pure 

 cement paste are required in order to 

 determine the presence of magnesium 

 silicate hydrate with a scanning electron 

 microscope. Hence, these concrete 

 samples could not be tested for magnesi- 

 um silicate hydrate. Appendix D presents 

 an extension to the sphere program 

 which will provide data so that the 

 phenomenon of magnesium ions replacing 

 calcium ions can be studied. 



