SUPERLATTICE PHOTOCATHODES: An Overview Tarun Desikan PPRC, Stanford University [email protected] OUTLINE Spin polarized electrons quick study Semiconductor polarized electron sources Method and results Superlattice characterization The need for strain Simulation of superlattice band structures Uses and requirements X-ray diffraction and photoluminescence Results USES OF POLARIZED ELECTRONS High-energy physics Surface analysis and imaging “Quantum” applications SPLEEM and SPSPM Computing, cryptography Single-electron devices Spintronics Spin-polarized currents Enhanced performance REQUIREMENTS IN HEP High polarization High charge 1013 electrons per train High speeds Increases effective luminosity of Collider >90% Picosecond micro-bunches Semiconductor photocathodes SEMICONDUCTOR SOURCE CB E k mj = +1/2 + Eg +3/2 HH SO LH -1/2 S1/2 - +1/2 -1/2 +1/2 -1/2 -3/2 P3/2 P1/2 Photo-excitation by polarized laser beam HH -> CB populates one spin state LH -> CB populates the other spin state Maximum polarization = 50% NEGATIVE ELECTRON AFFINITY SURFACE GaAs Polarized electrons Cs2O Tunneling current CB VB Circularly polarized laser photons Polarized e- tunnel through to NEA material and escape Atomically clean surface at UHV STRAINED PHOTOCATHODES CB E mj = +1/2 Eg k + +3/2 HH SO LH +1/2 +1/2 -1/2 S1/2 -1/2 -1/2 -3/2 P3/2 P1/2 HH and LH no longer degenerate at k=0 HH -> CB populates one spin state, LH -> CB does not occur Maximum polarization = 100% SAMPLE STRUCTURE 1000 A 25mm 25mm Active Region GaAs0.64P0.36 Buffer GaAs(1-x)Px Graded Layer GaAs Substrate SUPERLATTICE PHOTOCATHODES Critical thickness (~100 A) limits the size of strained active region Practical limit is ~1000 A Active region partially relaxes Multiple quantum wells Strained material sandwiched between unstrained layers Strained region thickness < critical thickness Band engineering SUPERLATTICE BAND CALCULATIONS k•p transfer matrix method 1 Chuang (UIUC), David Miller, Jim Harris (Stanford) 2 3 4 N+1 AN 2 A1 B T B 1 N 2 T D1 .P2 D2 .P3 D3 ...PN 1 DN 1 N+2 SUPERLATTICE BAND CALCULATIONS Must account for CB, HH, LH and SO CB decoupled HH, LH and SO interact Matrix solution to Schrödinger's equation 8x8 Hamiltonian Strain effects incorporated into Hamiltonian Boundary conditions Reach MATLAB noise floor SINGLE QUANTUM WELL SIMULATION MULTIPLE QUANTUM WELL SIMULATION SIMULATION RESULTS Must use consistent parameters Easy wrap-around scripts Spot trends Compare with experiments? E Lw Effective Band Gap HH–LH Splitting X-RAY DIFFRACTION High-resolution XRD to analyze crystal Study layer attributes Thickness Composition Strain Tilt Vendor specifications XRD THEORY Bragg’s Law: n* = 2*d*sin() All lattice planes d contribute to Bragg diffraction (004), (224), (113) planes commonly used Every layer contributes a Rocking Curve peak Repeating series of thin layers causes additional peaks ROCKING CURVES RECIPROCAL SPACE MAP cps 2 RECIPROCAL SPACE MAP cps 2 SAMPLE STRUCTURE 1000 A 25mm 25mm Strained GaAs GaAs0.64P0.36 Buffer GaAs(1-x)Px Graded Layer GaAs Substrate OTHER CHARACTERIZATION TOOLS Photoluminescence Band structure analysis Check simulation predictions SIMS Doping profile Destructive STRAINED SUPERLATTICE SVT-3682 1000 A 25mm 25mm Active Region GaAs0.64P0.36 Buffer GaAs(1-x)Px Graded Layer GaAsP 30 A Strained GaAs 30 A GaAsP Strained GaAs GaAsP Strained GaAs GaAs Substrate BAND STRUCTURE SIMULATION BAND STRUCTURE CB1 1.65 eV HH1 0.86 eV GaAsP GaAs GaAsP LH1 GaAs GaAsP Photoluminescence confirms the simulation prediction LAYER THICKNESSES (004) scan [above] as well as (224) SVT-3682 ANALYSIS Active Region GaAs0.64P0.36 Buffer GaAs(1-x)Px Graded Layer GaAs Substrate aGaAs 6 aGaAs P 0.64 0.3 6 aGaAs aGaAs Ideal Actual Well Width = Barrier Width = 32 A Phosphorus fraction in GaAsP = 0.36 Strained GaAs does not relax significantly P 0.64 0.3 SVT-3682 PERFORMANCE Quantum Efficiency Polarization SVT-3682 PERFORMANCE Peak polarization of ~86% High QE A record at SLAC > 0.2 % is great No charge limit A great photocathode! Repeatable? CONCLUSIONS High performance superlattice photocathodes fabricated using GaAs/GaAsP Further improvement by optimizing parameters Need to test validity of band structure simulations Extend simulation model to calculate polarization