Laser Cooling of Solids
Presented at the joint Conference on Lasers & Electro-Optics (CLEO) and Quantum Electronics & Laser Science (QELS) meeting as postdeadline paper #QPD17 on May 25, 1995 in Baltimore, MD.
Authors: T.R. Gosnell, C.E. Mungan, M.I. Buchwald, B.C. Edwards, and R.I. Epstein.
We report the first successful demonstration of an optically cooled solid. The sample is an Yb3+ doped fluorozirconate glass pumped at 1008 nm. Steady-state net cooling of ~0.3 K below room temperature is obtained.
The premise that anti-Stokes fluorescence of a solid material can form the basis of a refrigeration cycle is an old idea first proposed in 1929 . Consider a system of isolated impurities, each with a simplified three-level energy structure, in order to understand the physical principles: label the ground state "1" and suppose that the two excited states, labeled "2" and "3", are split in energy by a few kT. Now further suppose that the energy gap between the excited states and the ground state is large enough that nonradiative relaxation across this energy gap is negligibly slow. On the other hand, nonradiative thermal equilibration between the two excited states occurs rapidly compared with radiative relaxation between these levels and the ground state. Thus, if a laser is tuned to the 1 → 2 transition, fluorescence from the 2 → 1 and 3 → 1 transitions will occur with a mean photon energy higher than that of the input photons. Cooling results as long as the fluorescence quantum efficiency is close to unity.
In the 1960s, optical materials such as GaAs  and Nd:YAG  were proposed as candidates for solid-state optical refrigerators. Although refrigeration-like phenomena were observed, incomplete understanding of the spectroscopy of these solid-state systems, plus the probable nonradiative quenching of excited states by impurities, yielded parasitic heating effects that dominated the expected cooling.
Tremendous advances in the purity of optoelectronic materials have allowed us to revisit the problem of optical refrigeration. By using a sample of ZrF4-BaF2-LaF3-AlF3-NaF (ZBLAN) glass doped with 1 wt % Yb3+, we have observed net steady-state anti-Stokes fluorescence cooling of ~0.3 K below ambient temperature when the sample was optically pumped with a cw Ti:sapphire laser at 1008 nm, a wavelength 13 nm longer than the mean wavelength of the 2F5/2 → 2F7/2 emission band of Yb3+.
The microscopic cooling properties of this ZBLAN sample were first investigated by using a collinear photothermal deflection technique: a counterpropagating HeNe laser beam was focused through the same sample volume that was exposed to a chopped infrared pump beam; the resulting thermally-induced angular deflections were synchronous with the pump beam's 0.5 Hz chopper and measured with a digitizing oscilloscope. The adjacent figure shows the amplitude of the observed deflection signals as a function of the pump wavelength, normalized by the absorption coefficient of the sample and by the infrared pump power. The dashed vertical line in the figure marks the transition from heating to cooling wavelengths as the pump wavelength is increased. Noteworthy in this experiment was an unmistakable 180 degree phase shift in the photothermal deflection signal between the long and short pump wavelengths.
A steady-state cooling experiment was performed by suspending a 4 mm2 by 7 mm long sample in a vacuum chamber and illuminating the sample along its long axis with 800 mW of pump power at 1008 nm. Blackbody emission from the sample was measured with an InSb focal-plane array in order to confirm true solid-state laser cooling.