Complete ground-state transition rates for the helium $2{}^{3}P$ manifold

This work presents experimental measurements of the $2{{}^{3}P}_2\to 1{{}^{1}S}_0$ and $2{{}^{3}P}_0\to 1{{}^{1}S}_0$ transition rates for helium. It completes the determination of the decay rates to the ground state from the $2{}^{3}P$ manifold and also completes the $2{{}^{3}P}_2$ decay rates for the metastable rare-gas series. The rate of photon emission from the $2{{}^{3}P}_2$ and $2{{}^{3}P}_0$ states to the $1{{}^{1}S}_0$ ground state is measured relative to the emission rate from the $2{{}^{3}P}_1\to 1{{}^{1}S}_0$ transition, which we have determined previously for the first time. The value for the $2{{}^{3}P}_2\to 1{{}^{1}S}_0$ transition is $0.324\pm 0.016{\text{s}}^{-1}$. An upper bound of $\sim 0.01{\text{s}}^{-1}$ is placed on the emission rate for the $2{{}^{3}P}_0\to 1{{}^{1}S}_0$ transition, which is predicted to be strictly forbidden. Along with our previous measurement of the $2{{}^{3}S}_1$ lifetime, the $2{}^{3}P$ manifold decay rates are in excellent agreement with current quantum-electrodynamic theory.

Figure 1

(Color online) Energy level diagram for helium showing laser excitation transitions and xuv photon emission with associated inverse decay rates $\tau$.

Figure 2

(Color online) (a) Historical progress of theoretical predictions (references shown) for the $2{{}^{3}P}_2\to 1{{}^{1}S}_0$ decay rate along with our experimental determination and associated uncertainty. Estimated uncertainties for the theoretical results are smaller than the data points. (b) Ratio of the experimental to the most recent theoretical calculation for the $2{{}^{3}P}_2\to 1{{}^{1}S}_0$ decay rates for the heliumlike isoelectronic sequence. Adapted from [18]: He $Z=2$, present work.

Figure 3

(Color online) (a) Channeltron signal for the $2{{}^{3}P}_2$ xuv photon decay flux recorded while atoms are held in the compressed MOT (blue data points and dashed black fitted exponential curve), followed by the large xuv peak when the P1 beam is pulsed on (solid red line). Note the change in scale due to more than three orders of magnitude difference in transition rates. (b) Photodiode signal for the corresponding 1083 nm fluorescence (dashed blue for signal from MOT and solid red for P1 laser pulse). All data averaged over $16000$ experimental cycles.