Laser spectroscopy of the $4s4p\hspace{0.5em}{}^3{P}_2\hspace{0.5em}–\hspace{0.5em}4s3d\hspace{0.5em}{}^1{D}_2$ transition on magnetically trapped calcium atoms

Laser excitation of the $4s4p\hspace{0.5em}{}^3{P}_2\hspace{0.5em}–\hspace{0.5em}4s3d\hspace{0.5em}{}^1{D}_2$ transition in atomic calcium has been observed and the wavelength determined to 1530.5298(6) nm. The metastable $4s4p\hspace{0.5em}{}^3{P}_2$ atoms were magnetically trapped in the quadrupole magnetic field of a magneto-optical trap. This state represents the only “loss” channel for the calcium atoms when laser cooled on the $4s^2\hspace{0.5em}{}^1{S}_0\hspace{0.5em}–\hspace{0.5em}4s4p\hspace{0.5em}{}^1{P}_1$ transition. A rate equation model shows that an order of magnitude more atoms are trapped in this state compared with those taking part in the main cooling cycle. Excitation of the ${}^3{P}_2$ atoms back up to the $4s3d\hspace{0.5em}{}^1{D}_2$ state provides a means of accessing these atoms. Efficient repumping is achieved if the 1530-nm laser is used in conjunction with a 672-nm laser driving the $4s3d\hspace{0.5em}{}^1{D}_2\hspace{0.5em}–\hspace{0.5em}4s5p\hspace{0.5em}{}^1{P}_1$ transition. In the present experiment, we detected about $4.5\times {10}^4$ trapped ${}^3{P}_2$ atoms, a relatively low atom density, and measured a lifetime of approximately 1 s, which is limited by background collisions.

Figure 1

Low energy levels of the calcium atom relevant for the present work. Vacuum wavelengths are indicated and the solid arrow is the 423-nm ${}^1{S}_0{–}^1{P}_1$ transition, 672-nm ${}^1{D}_2{–}^1{P}_1$ transition (long dashed double arrow), and 1530-nm ${}^3{P}_2{–}^1{D}_2$ transition (dash dotted double arrow), respectively. The brown dashed arrows show the decay paths of the atoms from the $4s4p\hspace{0.5em}{}^1{P}_1$ state of the 423-nm MOT and $4s5p\hspace{0.5em}{}^1{P}_1$ state. The 430-nm transition (dashed double dotted double arrow) used in an alternative repumping scheme [12] is also shown with the decay path (brown dashed arrow) via the ${}^3{P}_1$ state to the ground state. The numbers in italic to the left of the levels indicate the labeling used in the rate equation analysis presented in Sec. 4.

Figure 2

Setup for infrared spectroscopy of the metastable cold calcium atoms. The 423-nm magneto-optical trap (MOT) (blue solid arrows) of calcium is loaded from the Zeeman slowed and deflected atomic beam (red dots) from an oven. The 672-nm repump laser (red dashed arrow) along with the 1530-nm laser (magenta dotted arrow) used to increase the number of atoms in the 423-nm MOT is also shown.

Figure 3

Resonance of the $4s4p\hspace{0.5em}{}^3{P}_2\hspace{0.5em}–\hspace{0.5em}4s3d\hspace{0.5em}{}^1{D}_2$ transition at 1530 nm observed as a decrease in the 423-nm MOT fluorescence (blue plus symbol). The 423-nm MOT is operated continuously, while the 1530 nm is switched on only for 100 ms. The solid magenta curve and red dashed curve are obtained from the rate equation model detailed in Sec. 4.

Figure 4

Pulse sequence of laser beams used for spectroscopy and for lifetime measurement of the magnetically trapped calcium atoms in the ${}^3{P}_2$ state. The 423-nm MOT is loaded with the molasses laser beams switched on for 500 ms. The MOT beams are then gated off for a hold time varied from 100 to 1000 ms. The 1530-nm laser beam is switched on for 100 ms at the end of this hold time, enabling detection of the remaining metastable ${}^3{P}_2$ atoms trapped in the magnetic quadrupole field of the MOT. The 672-nm laser remains either on or off throughout this sequence depending on the desired measurement.

Figure 5

(a) Detection of the metastable ${}^3{P}_2$ state atoms as a 423-nm MOT fluorescence with the 1530-nm laser pulse. Details of pulse sequence are described in the text. (b) Increase in the 423-nm MOT fluorescence with a frequency stabilized 672-nm laser on resonance with the 1530-nm laser pulse. In both methods, the 1530-nm laser pulse is on for only 100 ms. The solid (magenta) curve is obtained from the rate equation model described in Sec. 4.

Figure 6

Effects on the steady-state populations of the variables in the rate equations describing the system with two repumpers. Curve a shows the variation of the optimal observed 423-nm fluorescence signal with ${\beta }_1$, which is a geometrical factor accounting for the fact that only a fraction of the magnetically trapped ${}^3{P}_2$ atoms interact with the 1530-nm laser at any given time. The signal is normalized to the value in the absence of any repumper. The number of atoms trapped in the ${}^3{P}_2$ state is shown in curve b as a function of the 672-nm excitation rate. The population is normalized to that of levels 1 and 2 in the absence of any repumper. Curves c (black), d (green), e (blue), and f (magenta) show the trapped ${}^3{P}_2$ state population (again normalized to that of levels 1 and 2 in the absence of any repumper) as a function of 1530-nm laser excitation and for the ${\beta }_1$ values $0.1\%$, $1\%$, $10\%$, and $100\%$, respectively.

Figure 7

Number of atoms detected as 423-nm fluorescence in the 423-nm MOT are plotted as a function of different hold times of the calcium atoms in the $4s4p\hspace{0.5em}{}^3{P}_2$ state to measure the lifetime. The solid line shown is the exponential fit to the experimental data. The measured lifetime is 928(78) ms.