Scaggs - thanks for pointing this out (I have not been to a Silex AGM - I live on the west coast). I think the main efficiency issues are the duty cycle, power and frequency stability rather than whether it works or not.. Covering each of these issues:
Duty cycle - reading the literature (that's difficult since details are secret) but the best source I have found is the Snyder report (see
HERE and
HERE) The issue would appear to be that Silex use a pulsed CO2 laser (with a Raman shifter) to provide the laser source for the enrichment process. The problem is that pulses have to multiplexed using a rotating mirror to multiplex several laser beams into a single (more continuous) beam. That's however not the end of the story since if its a circular shaped beam the gas will transit the top and bottom of the circle more quickly that the middle. Also a normal laser beam has a "bell shaped" intensity profile (i.e stronger in the centre than the edges) so the beam needs to be "shaped" to be consistent across its radius - all this requires R&D. The point is that you want to irradiate all the UF6/carrier gas stream both consistently (in time) and also evenly across the gas flow. If there is too much radiation in the middle it will cause dissociation into UF5 (not wanted) and at the edges will not provide sufficient power to split the dimers into monomers - that's the way the 235UF6 monomers are separated from the 238UF6/G dimers and if condensation repression is used then the dissociation needs to be preserved into the condensation stage (i.e the 235UF6 monomers still need to be excited during condensation so they condense later in the flow). These are not easy development and I assume these are "efficiency" issues that SLX will have been working on to improve the economics.
Power - the power in the laser beam needs to be sufficient to cause dissociation of the 235UF6/G dimers but also remain as UF6 (not UF5). I understand a lower (and upper) threshold to the power is required (i.e beam shaping)
Frequency stability - this is vital since Syder tells us that the v3 resonant frequency for UF5/G is 15.916 μm but for 238UF6/G is only 15.931 μm at low temperature (in expanding gas) - this requires extreme frequency stability for the laser beam (hence the need for the Beam Control system - BCS).
In conclusion I think SLX has had a lot of issues to resolve to get efficient separation; this is likely what they have been working on since the commissioning in 2013 and hopefully have now got these to a commercial standard so as to give us an edge over competition. Most of this is laser work - these also need to be highly reliable (run of the mill high pressure CO2 are not that reliable - you may well need redundant CO2 lasers built into the system to ensure reliability.for a commercial system)
All the above assumes thay are still using CO2 lasers/raman shifters - the may now be using Quantum cascade lasers (QCL) - this would make things easier but the problem would then be getting the required minimum power for the beam with all the QCL's being phase synchronized (so as to avoid frequency beating)
My conclusion is that SLX should (hopefully) have overcome numerous technical challenges to be where I hope they are now - these would have been major development issues.
(20min delay)