450 



FORMATION AND DISSOLUTION OF AIR BUBBLES 



Figure 1. Cavitating model propeller. The picture was made with a 1/30,000-sec flash. Note the heavy tip vortices, 

 considerable laminar cavitation near the blade tips, and the start of burbling cavitation of the blade face near the hub. 

 This is a right-hand propeller and the water is flowing from left to right. 



position differs markedly from that of atmospheric 

 air because the solubiUty of nitrogen is twice that of 

 oxygen. Accordingly, the cavitation gas consists of 

 M oxygen and % nitrogen. The quantity of gas 

 which collects each second in a cavity in moving 

 water is proportional to the surface area of the 

 cavity and to the partial pressure of air dissolved in 

 the surrounding water but is essentially independent 

 of temperature and hydrostatic pressure. The con- 

 stant of proportionality is roughly 4 X 10~' mole 

 per sq cm per second per atmosphere.^ 



When the cavity collapses, the gas which has dif- 

 fused into it will be compressed, and a bubble will be 

 formed with a radius such that the gas pressure in- 

 side equals the hydrostatic pressure outside. The 

 cavities formed by blade cavitation collapse so 

 quickly that any air bubbles formed must be very 

 small indeed. However, the cavities originating in the 



tip vortices last much longer, since the centrifugal 

 force in the whirling vortex remains high for some 

 time. Thus, presumably it is the tip vortex cavita- 

 tion that is primarily responsible for most of the air 

 appearing as bubbles in propeller wakes. It has been 

 observed that sea water at all depths contains dis- 

 solved oxygen and nitrogen in amounts roughly cor- 

 responding to saturation at the surface. For this 

 reason it is undersaturated with respect to a bubble 

 of air or cavitation gas anywhere below the surface, 

 and a bubble of either gas will gradually disappear 

 as the gas reenters the water. The rate of solution 

 agrees with the same simple theory of diffusion as the 

 rate of accumulation of gas in a cavity; indeed, the 

 facts regarding the latter process are largely inferred 

 from a study of the former. The number of moles of 

 air which escape each second from a bubble is ap- 

 proximately proportional to the surface area of the 



