Blackbody radiation shift of theCs133hyperfine transition frequency

Abstract

We report the theoretical evaluations of the static scalar polarizability of the 133 Cs ground state and of the black body radiation shift induced on the transition frequency between the two hyperfine levels with m F = 0. This shift is of fundamental importance in the evaluation of the accuracy of the primary frequency standards based on atomic fountains and employed in the realization of the SI second in the International Atomic Time (TAI) scale at the level of 1 × 10 −15 .Our computed value for the polarizability is α 0 = (6.600±0.016)×10 −39 Cm 2 /V in agreement at the level of 1×10 −3 with recent theoretical and experimental values. As regards the black body radiation shift we find for the relative hyperfine transition frequency β = (−1.49 ± 0.07) × 10 −14 at T = 300 K in agreement with frequency measurements reported by our group and by Bauch and Schröder [Phys. Rev. Lett. 78, 622, (1997)]. This value is lower by 2 × 10 −15 than that

Frequency Shift of the Cesium Clock Transition due to Blackbody Radiation

“…There is some disagreement on the value of this shift. Early measurements [2,4,7] and ab initio calculations [5,8] support a value which is close to −2.2 × 10 −10 Hz/(V/m) 2 while more recent measurements [10,11] and semiempirical calculations [3,9,12] claim that actual number might be about 10% smaller.In the present work we have performed fully ab initio calculations of the radiation frequency shift and have identified the source of the disagreement between different theoretical results as the contribution of the continuum spectrum states into summation over the complete set of intermediate states. The continuum spectrum was included in all the ab initio calculations and missed in the semiempirical considerations.…”

“…There is some disagreement on the value of this shift. Early measurements [2,4,7] and ab initio calculations [5,8] support a value which is close to −2.2 × 10 −10 Hz/(V/m) 2 while more recent measurements [10,11] and semiempirical calculations [3,9,12] claim that actual number might be about 10% smaller.…”

“…In contrast, analysis performed in [3,9,12] is limited to discrete spectrum. Adding −0.34 × 10 −1 Hz/(V/m) 2 (which is total tail contribution from all three terms (6), (7) and (8) found in our calculation) to the results of Feitchner et al [3] and Micalizio et al [9] brings them to excellent agreement with ab initio calculations.…”

“…It is not surprising therefore that a semiempirical consideration, involving only discrete spectrum states, gives very good results for the atomic polarizabilities (see, e.g. [9]). However, let us stress once more that the calculation of polarizabilities checks only term (8) and tells us very little about the accuracy of other two terms, (6) and (7).…”

“…Therefore, if good accuracy is achieved for polarizabilities, even better accuracy should be expected for the term (8) (see also Ref. [9]). However, the behavior of the other two terms, (6) and (7), is very different and calculation of polarizabilities tells us little about accuracy for these terms.…”