Progress on Light Scattering From Degenerate Fermions Seth

Progress on Light Scattering From Degenerate Fermions Seth A. M. Aubin University of Toronto / Thywissen Group May 20, 2006 DAMOP 2006 Work supported by NSERC, CFI, OIT, PRO and Research Corporation.

Outline Ø Motivation Ø Apparatus Ø Light Scattering: Simple approach Ø Light Scattering: next generation

Light Scattering with Fermions Objective: Modify the lifetime/linewidth of an excited state with quantum statistics. Motivation: Ø Trapping environment reduces the number of recoil states lifetime increases. Ø Analogous phenomena observed in cavity QED systems. Ø Similar phenomena frequently observed in condensed matter systems. See for example, A. Högele et al. , Appl. Phys. Lett. 86, 221905 2005).

Rb + K

Signatures of Degeneracy 87 Rb Bose-Einstein Condensate: 104 - 105 atoms 0 Observation of Pauli Pressure EF Fermi-Dirac Statistics Boltzmann Statistics 200 400 Radial distance ( m) EK, release/EF Optical Density Fermion (40 K) momentum distribution 0. 1 TF with 4 104 40 K atoms S. Aubin et al. , Nature Physics (2006). k. TRb/EF

Signatures of Degeneracy 87 Rb Bose-Einstein Condensate: 104 - 105 atoms Fermion (40 K) momentum distribution Observation of Pauli Pressure Fermi-Dirac Statistics Boltzmann Statistics 0 EK, release/EF Fit Residuals EF 200 400 Radial distance ( m) 0. 1 TF with 4 104 40 K atoms S. Aubin et al. , Nature Physics (2006). k. TRb/EF

Light Scattering with Fermions: Simple Approach Degenerate Fermions: Pauli Blocking of light scattering Probe Laser Ø Fermi sea reduces number of states an excited atom can recoil into. DFG Ø Atomic lifetime increases, linewidth decreases. B. De. Marco and D. Jin, Phys. Rev. A 58, R 4267 (1998). Th. Busch et al. , Europhys. Lett. 44, 755 (1998). Erecoil = 0. 4 K EFermi = 1. 1 K k. F

Further difficulty with Fermions We want this process kx More likely process Fermi Sea kx krec oil kx Fermi Sea krec oil Almost no Pauli blocking. kx

Solution ? IDEA: different states can have different Fermi energies/momentum (i. e. different populations), but still be in thermal equilibrium. Excite mf = 7/2 atoms. kx Look for Pauli blocking of decay into mf = 9/2. Fermi Sea DFG, mf=7/2 krecoil kx Non-DFG, mf=9/2

How well does it work ? Suppression factor: M, suppresion factor EF, 2 = 4 Erecoil T=0 EF, 2 = 6 Erecoil EF, 2 = 8 Erecoil EF, 1 EF, 2 Theory for a spherical harmonic trap, based on: B. De. Marco and D. Jin, Phys. Rev. A 58, R 4267 (1998). Th. Busch et al. , Europhys. Lett. 44, 755 (1998).

Implementation 11/2 F = 11/2 9/2 7/2 Procedure: 5/2 Ø State preparation: prepare DFG in mf=7/2, and non-DFG in mf=9/2. Ø Apply weak excitation pulse (atom scatters less than 1 photon). Non. DFG 9/2 7/2 5/2 F = 9/2 Ø Measure population ratios Ø Look for a change in ratio as T is decreased.

Potential Difficulties Ø Rescattering of scattered light. far off resonance probe Ø Unwanted transitions to unsuppressed levels. dipole trap + large Zeeman splittings Ø Heating due to probe. short pulse

Dipole Trap Currently installing a 1064 nm dipole trap: Aligned with Z-wire trap. It works! ~100% loading efficiency with 87 Rb. Loading into the optical trap: 105 87 Rb atoms at ~ 1 µK

Summary EF Ø Degenerate Bose-Fermi mixture on a chip. Ø New scheme for light scattering with fermions Fermi Sea Ø Dipole trap installed. krecoil

Thywissen Group S. Aubin D. Mc. Kay B. Cieslak S. Myrskog M. H. T. Extavour A. Stummer T. Schumm Colors: Staff/Faculty Postdoc Grad Student Undergraduate L. J. Le. Blanc J. H. Thywissen

Atom Chip for Bose-Fermi mixtures Advantages: Ø Short experimental cycle (5 -40 s). Ø Single UHV chamber. Ø Complex multi-trap geometries. Ø On-chip RF and B-field sources. Trap Potential: Z-wire trap Chip by J. Esteve, Orsay.

Simple Version 11/2 F = 11/2 9/2 7/2 Procedure: Ø State preparation: prepare DFG in mf=9/2, and nothing in mf=7/2. 5/2 Ø Apply weak excitation pulse to intrap atoms. (atom scatters less than 1 photon) Ø Use Stern-Gerlach to image the states separately. DFG empty 9/2 7/2 5/2 F = 9/2 Ø Measure population ratios. Ø Look for a change in ratio as T is decreased.

Cross-Section plot

Implementation #2 11/2 9/2 F = 11/2 7/2 5/2 9/2 7/2 F = 9/2 5/2 Procedure: Non. DFG 9/2 7/2 5/2 F = 9/2 Ø State preparation: prepare DFG in mf=9/2, and non-DFG in mf=7/2. Ø Apply 2 -photon excitation pulse (1 RF + 1 optical). Ø Look for a decrease in scattering rate as T is decreased.

Rb-K cross-section (nm 2) Sympathetical Cooling