On Saturday, January 11, 2014 12:17:23 PM UTC-5, haitic...@gmail.com wrote:   
   > I am working on a medical research project with the following constraints:   
   arteriole model with dia. of ca. 1.5 mm, freedom to surround arteriole with   
   any cavity, led wavelengths. The aim is to measure hemoglobin concentration.   
   Scattering by blood    
   cells is NOT an issue in this experiment, just the absorption by the Hb, due   
   to the particular aim of the experiment.   
   >    
   >    
   >    
   > I have tentatively selected a 5 mm blue led and a generic Si pin diode   
   arranged in several geometries. The geometries have included aiming the led   
   strait through the arteriole model and out the other side onto the detector   
   (A), jacketing the arteriole    
   model with a 2 mm channel and letting light bounce down it for a short   
   distance to an offset detector,(B),  or (C) creating a reflective chamber say   
   7 mm dia around the tube and fed by a blue led, with watching detector on   
   perimeter.   
   >    
   > This would be painted flat white inside.   
   >    
   >    
   >    
   > The first question is whether a blue led is best. I picked the 460 nm led   
   because, obviously, anything absorbing in the blue looks red. (ie Hb)  But of   
   course a white or red emitting led is going to be brighter.    
   >    
   >    
   >    
   > But here is my question: The apparent red color of hemoglobin may be there   
   just as a differential in the extinction coefficients between red and blue. So   
   a white led may still absorb in the red enough to give a useful signal. I   
   should mention in this    
   model, just as in vivo, the only absorption is from the Hb - there are no   
   other strong absorbers present.    
   >    
   >    
   >    
   > At this point I like the reflective cavity the best. (C). I tried the   
   constrained channel (B), but the signal was not very strong. (It could be   
   boosted, so is still a contender.) (A) attracted because the source and sensor   
   are in line, but is proving    
   to be tricky to align without the light going around the red tube.    
   >    
   >    
   >    
   > So I am left with (C0, a proven detection scheme for constrained targets.    
   >    
   >    
   >    
   > As well, I had the idea that (B0 is just a narrow version of (C). The light   
   is bouncing down the tube, in equilibrium with absorption by the target.   
   Probably the advantage of a proper reflective cavity is the collection gives a   
   better photon budget.   
   >    
   >    
   >    
   > Any ideas welcome.   
      
   B0, C0 = (B), (C)   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)