From: bergervo@iae.nl   
      
   On 11/19/2018 4:05 AM, richalivingston@gmail.com wrote:   
     ..   
   >>  ...  What exactly is it about the Cesium 133 atom that makes it so   
   >> convenient for keeping time?  I assume they considered other atoms, but   
   >> settled on Cesium as most convenient.  Is there a good simple reason for   
   >> this?   
   >   
   > First, I'd recommend reading the Wikipedia article, that would be   
   > a good start.   
      
   The article actually mentions that for accuracy, the Caesium   
   clock is not the best. Currently that's the Strontium clock   
   running at 430 THz.   
      
   > All atomic clocks use an atomic transition as their frequency   
   > standard.  Almost any transition could conceivably be used, but   
   > there are practical matters that make some more useful than others:   
   >   
   > -A narrow transition (one with a long emission time) are better   
   > because the frequency is more precise   
   > -A higher frequency is better as it allows a finer resolution of time   
   > -Too high a frequency (e.g. light wavelengths) are too difficult   
   > to count.  Also, the photon absorbtion to excite the atom may result   
   > in the atom moving which causes a doppler shift with again distorts   
   > the frequency observed.   
   > -An atom that can be worked with as a gas tends to be better as the   
   > frequency from an isolated atom is less likely to be disturbed by   
   > collisions or the close proximity of other atoms.   
   > -A heavier atom will be moving more slowly at any given temperature,   
   > so that again is preferred.   
   >   
   > Cesium 137 is a heavy atom with a narrow emission at 9.192631770   
   > GHz which is about the highest frequency that it is easy to count   
      
   Well, nowadays counters up to an order of magnitude faster can be   
   made with the leading semiconductor technologies (SiGe, GaAs, InP).   
      
   The first divide-by-two block is the most difficult part of such a   
   counter. So that leaves the question how they do this with the 430   
   THz Strontium standard! It must use some injection-locked laser at   
   a sub-harmonic of the 430 THz, I guess.. Followed by other, similar   
   dividers at longer IR wavelengths, until frequencies in the electronic   
   domain are finally reached.   
      
   But a quick search doesn't give any information on this particular   
   part of the Strontium clock (most links just describe the tremendous   
   accuracy of the oscillation, but not how the "escapement" is made,   
   to put it in mechanical terms..)   
      
   --   
   Jos   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)