New materials for optical cooling

Well-characterized solid-state laser materials are evaluated for performance in optical refrigeration as well as radiation-balanced laser systems. New figures-of-merit are developed and applied to ytterbium-doped materials. Superior performance is predicted for high-cross-section tungstate materials. Photothermal deflection experiments on samples of Yb3+-doped KGd(WO4)2 confirm anti-Stokes fluorescence cooling. This is the first observation of optical cooling in a crystal. PACS: 42.55.Rz; 42.70.Hj The physical mechanism of radiation cooling by anti-Stokes fluorescence was originally proposed in 1929 [1]. It can be easily understood. Absorption of a photon can, on average, temporarily push an atom away from thermal equilibrium with its surroundings. If the atom then spontaneously emits after thermal equilibrium has been re-established, any frequency shift in the fluorescence compared to the absorbed radiation results in a net heating or cooling of the material. While simple in concept, to obtain radiative cooling in practice requires materials with near-unity fluorescence quantum efficiencies. Such anti-Stokes cooling was first reported in 1981 for CO2 gas [2], in 1995 and 1996 for organic dye solutions [3, 4], and in 1995 for ytterbium-doped ZBLANP glass [5]. Mungan and Gosnell have published a detailed review of anti-Stokes cooling [6]. Recent commercial availability of high-brightness laser diodes opens the possibility of practical devices that utilize optical cooling. Refinements of the experiments in ytterbiumdoped ZBLANP glass have recently demonstrated optical refrigeration from 301 K down to 236 K [7] and modeling suggests that temperatures as low as 77 K can be obtained with this material [8]. Further, it has been proposed that solidstate lasers could be constructed in which the cooling of anti-Stokes fluorescence would offset heat generated by stimulated emission [9]. This mode of laser operation is referred to as radiation-balanced (RB) lasing. Unlike conventional exothermic laser systems, RB lasers would exhibit little or no internal heat generation. In principle, this technique would allow RB lasers to be scaled up to much higher average powers than conventional solid-state laser systems. Previous solid-state optical cooling experiments have concentrated on ytterbium-doped glasses [10]. In this paper, we evaluate a range of well-characterized ytterbium-doped laser materials for application to optical refrigeration and radiation-balanced lasing. Crystalline hosts are emphasized instead of glasses because of their generally higher cross sections and well-defined micro-environments. Optical cooling experiments on a few materials confirm the potential of crystals for anti-Stokes optical cooling. 1 Spectral optimization for optical cooling Detailed descriptions of the process of optical cooling in solids have been presented elsewhere [6–8], so a brief review will suffice here. For ytterbium-doped materials, the process begins with optical pumping near 1.0 μm which excites Yb3+ ions from the F7/2 ground state to the F5/2 excited state. Both of these states are Stark split into several closely spaced energy levels. The picosecond timescale for phonon coupling of these levels to the lattice combined with the interstate millisecond timescale for spontaneous emission ensures Boltzmann distributions for both of the (2J +1)/2 multiplets prior to radiative decay. Spontaneous emission of the excited ions yields a broadband fluorescence with a mean wavelength of λF = ∫ IF(λ)λdλ ∫ IF(λ)dλ , (1) where IF(λ) is the fluorescence spectral intensity. In the absence of radiative trapping or quenching, a redshift of the pump from the mean wavelength will result in a net cooling of the material. Letting PF be the total fluorescence power and PP be the pump power, the efficiency per unit length of the optical cooling process can be defined as ηC(λP)≡ 1 PP ∂(PF − PP)