Optical Spectroscopy of Potassium-Doped Argon Clusters. Experiments and Quantum-Chemistry Calculations

Abstract

Rare-gas clusters produced in a supersonic expansion into vacuum are doped with alkali-metal atoms using the pickup technique. We give here a detailed description of our experimental apparatus, which is suitable for electronic spectroscopy of any desired gas clusters doped with low-temperature melting metals. Potassium-doped Ar clusters (average size N approximately 2000) are investigated by beam-depletion (BD) spectroscopy as well as laser-induced fluorescence (LIF) spectroscopy. The observed BD spectra are strong and exhibit several broad peaks within the scan range of the Ti:sapphire excitation laser; only one weak band is observed instead in LIF spectra. This situation resembles the one previously observed for K-doped H2 clusters [Callegari et al., J. Phys. Chem. A 1998, 102, 4952] where it was argued that fluorescence is either quenched or strongly red-shifted by the cluster. We investigate BD spectra as a function of the potassium vapor pressure in the pickup cell, and we perform an extensive analysis of the pickup process, necessary to separate the overlapping spectral contributions of different species. We tentatively assign parts of the spectra to the monomer, singlet dimer, and doublet trimer, of potassium residing on the surface of the cluster. To support our trimer assignment, we perform complete active space self-consistent field (CASSCF) calculations to determine the lowest 12 electronic spin-doublet states of free K3. We find several candidate transitions in the region of interest. Although a definite assignment is not possible because the energy shifts due to the perturbation by the Ar cluster are not known, our theoretical and experimental data compare favorably with existing spectra, and with simulations, of the closely related Na3-molecule, upon suitable scaling of its energy-level structure.

The excited states of K3 cluster: The molecular symmetry adapted non-adiabatic coupling terms and diabatic Hamiltonian matrix

“…Ernst et al investigated potassium doped Ar clusters by beam depletion (BD) and laser induced fluorescence (LIF) spectroscopy, and performed CASSCF level calculation to assign experimental spectrum of doublet potassium trimer. Such an investigation [43] found the one to one correspondence between experimental data and theoretical prediction for the states, 1 2 A 0 2 , 4 2 A 0 1 and 3 2 B 0 2 . On the other hand, since the other low lying states (1 2 E 0 , 2 2 E 0 , 1 2 A 0 1 , 2 2 A 0 1 etc.)…”

“…These clusters are convenient for theoretical calculation [28][29][30][31][32][33][34][35][36][37][38] to interpret the experimentally measured data [39][40][41][42][43][44][45][46][47] due to one-electron-valence configuration and few number of vibrational degrees of freedom (DOFs). As for example, Yarkony et al confirmed [37] the presence of Conical Intersection (CI) at D 3h geometry in Li 3 cluster along with unusual CIs at C 2v points.…”

A Jahn–Teller analysis of K3 and Rb3 in the electronic states 12E′ and 12E″

“…In the case of potassium a minimum-sized full-electron active space CAS (9,8) including the 4s, three 4p, the 5s and the three 5p orbitals is sufficient to reproduce the experimental data. For rubidium we had to enlarge the active space to CAS (9,12) to obtain good agreement. At this level of theory about 77% of the experimental energy splitting is reproduced.…”

“…The alkali atoms reside on the surface of the rare-gas clusters and form clusters themselves, dissipating their binding energy by evaporation of rare-gas atoms. With this technique small oligomers of potassium on large argon and helium clusters were investigated by means of laser-induced fluorescence and beam depletion spectroscopy [12,13]. Unfortunately, the measured spectrum is always a mixture of contributions from atoms, dimers, trimers and higher oligomers.…”

“…[20]. In a recently published article [21] we presented the first ab initio calculations on the doublet states of the potassium trimer, and provided an assignment of three experimental peaks in the range between 10,000 and 14,000 cm À1 (3 2 A 1 , 1 2 A 2 , 4 2 A 1 ) [12]. A peak near 12,400 cm À1 remained unassigned.…”