The 2 percent aqueous solution of uranium tetrabromide is 38 percent uranium   
   tetrabromide by mass. Uranium tetrabromide is over 5 times denser than water   
   and since acceleration is equivalent to gravity, it should separate out of   
   solution and collect at    
   the back of the propellant tanks. We are talking about enough uranium to   
   trigger a gigaton level explosion if it goes off.    
      
   Sure you can add enough mass to the system to fix most of the problems with   
   this idea, but then you will not have torch drive performance. Plus Zubrin   
   makes a pretty good whopper of an unproven assumption about controlling the   
   reaction in the chamber.    
   His assumption that he can limit the reaction to only happen where he wants by   
   controlling the propellant velocity rests on the incorrect assumption that   
   thermal neutrons only travel at their average velocity. When in fact there is   
   a velocity    
   distribution of fission neutrons and bunch of those neutrons will be traveling   
   faster than the average. This could cause the continuously detonating fission   
   reaction to move closer to the hull. You may not have enough time to even turn   
   it off before the    
   back of your spacecraft gets vaporized.    
      
   Simple mathematical models, which the NSWR is, do not mean that you have a   
   workable solution. The best that you can do is to argue that it is a promising   
   avenue of research that NASA should look into, but you shouldn't fool yourself   
   into thinking that it    
   is necessarily feasible. The fact that Zubrin can't convince anyone to fund   
   either the testing of this idea or just engineering design work should be a   
   giant red flag by itself.    
      
   Seriously, would you want to spend money that you were responsible for on a   
   wild idea whose failure mode was to vaporize a multi-billion dollar spacecraft   
   in the largest explosion in human history? Even if Zubrin is 95% right, a fuel   
   pump malfunction    
   could cause a multi-megaton or even gigaton explosion. And that is if you add   
   machinery to continuously mix the solution to keep the uranium tetrabromide   
   from collecting at the back of the propellant tanks.    
      
   For NASA, torch drive performance is not necessary. Torch drives don't get you   
   there much faster than continuous acceleration in a sub-torch engine.   
   Decreasing acceleration by a factor of 16, only increases maximum travel time   
   by a factor of 4 if both    
   engines accelerate continuously during the trip. Since the torch engine will   
   most likely cut off early and coast, the travel time increase will actually be   
   less than this.    
      
   On the whole this is a very risky idea with rather little payoff compared to   
   alternatives.    
      
   In fact it is doubtful that even a space military would find low end torch   
   drive performance useful.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)