Oddbean

there’s an interesting (more radical) suggestion in this metrologia paper to make the second more like the other new SI units “fixing the numerical value of one more fundamental constant, in addition to c, h and e. From the fundamental standpoint, a good choice for this constant is the electron mass m_e, in which case the system of units is set by the relations: m_e = M kg, where M is the defining value, completed by the other defining relations for c, h, e, k_B, N_A and K_cd.”

later on they say “Currently, the value of m_e has an uncer-tainty of 3.0 parts in 10^10, while the uncertainty in the Rydberg constant is 1.9 part in 10^12. These uncertainties are several orders of magnitude larger than the present realizations of the unit of time of the current SI system (few parts in 10^16) and even further away from the capabilities of optical frequency standards (10^−18 or better). Consequently, Option 3 is not practical in the current state of science and technology.”

there’s a table of candidate optical standards, ranging from 0.4 to 1.1 THz, using Hg or Al or Yb or Sr or Ca, in trapped ion optical clocks or neutral atom optical lattice clocks it says 4 labs have Yb lattices and 3 have Sr lattices in production contributing to TAI then one lab using oldskool rubidium, i’m pretty sure that’s the USNO’s rubidium fountains, which are unusual in (a) using rubidium instead of caesium and (b) operating continuously rather than for calibrating other clocks

seriously high-quality jargon “In addition to continued advances in cavity performance mentioned earlier, there are efforts in parallel to develop novel measurement protocols that mitigate the limitations caused by reference cavity noise, such as” get this “zero-dead time interrogation, correlation spectroscopy, and dynamic decoupling of laser phase noise in compound atomic clocks.”