Factors Affecting QE and Dark Current in Alkali Cathodes. John Smedley Brookhaven National Laboratory

Size: px
Start display at page:

Download "Factors Affecting QE and Dark Current in Alkali Cathodes. John Smedley Brookhaven National Laboratory"

Transcription

1 Factors Affecting QE and Dark Current in Alkali Cathodes John Smedley Brookhaven National Laboratory

2 Outline Desirable Photocathode Properties Low light detection Accelerator cathodes Factors Affecting Performance Practical Experience with K2CsSb Monte Carlo modeling Cathode studies

3 Photoinjector

4 What makes a good photocathode? Photoinjector High QE at a convenient λ Low dark current Dominated by field emission Spatially Uniform Long lifetime in challenging vacuum environment Photodetector High QE in range of interest Low dark current Dominated by thermal emission Spatially Uniform Large area Low response to stray Chemical poisoning light Ion bombardment Low intrinsic energy spread Reproducible (thermal emittance) Long lifetime in sealed Typical pulse length of 10 system 50 ps Cheap, easily manufactured Peak current density can be >10kA/cm2

5 Three Step Model Semiconductors 1) Excitation of e hν Empty States Reflection, Transmission, Interference Energy distribution of excited e Eg Filled States Energy No States Φ Ea Lase r Medium Vacuum 2) Transit to the Surface e lattice scattering mfp ~100 angstroms many events possible e e scattering (if hν>2eg) Spicer s Magic Window Random Walk Monte Carlo Response Time (sub ps) 3) Escape surface Overcome Electron Affinity

6 Factors Affecting QE Reflection Choice of polarization and angle of incidence Light traps (microstructures) Nonproductive absorption Semiconductor cathodes (especially NEA materials) Narrow valence band Work function reduction (Schottky effect, dipole layers) Electron scattering (electron mfp) Stay within the magic window, φ<eγ<2egap Minimize photon absorption length (surface plasmons) Good crystals minimize defect and impurity scattering Substrate material, cathode thickness, sequential vs co deposition, substrate temperature, cooling time, oxide layer formation Deposition parameters Vacuum environment Ion back bombardment, electron stimulated desorption, chemical poisoning Operating environment Thermal stability, space charge

7 Factors Affecting Dark Current Field emission Electric field at cathode Surface morphology (field enhancement) Work function Thermal emission Temperature Work Function I = A T2 exp[ eφ/(kt)] Ion bombardment Vacuum Work function Low work function reduces the threshold photon energy and improves QE, especially near threshold But, it increases dark current Hamamatsu Tech Note => Optimal work function depends on application

8 K2CsSb (Alkali Antimonides) Work function eV, Eg= ev Good QE (4% 532 nm, 355nm) Deposited in <10 10 Torr vacuum Typically sequential (Sb >K >Cs) Cs deposition used to optimize QE Oxidation to create Cs O dipole Co deposition increases performance in tubes Cathode stable in deposition system (after initial cooldown) D. H. Dowell et al., Appl. Phys. Lett., 63, 2035 (1993) C. Ghosh and B.P. Varma, J. Appl. Phys., 49, 4549 (1978) A.R.H.F. Ettema and R.A. de Groot, Phys. Rev. B 66,

9 Laser Propagation and Interference Laser energy in media 0.8 Calculate the amplitude of the Poynting vector in each media 0.6 Not exponential decay nm Vacuum K2CsSb Copper 200nm

10 Monte Carlo Modeling Thickness 543 nm Ref 0.09 trans 0.08 Total QE 0.5 QE w/o R&T Thickness (nm) QE Transmission/Reflection 0.7

11 Monte Carlo Modeling

12 Deposition System Sb K Cs Sequential deposition with retractable sources (prevents cross contamination) Cathode mounted on rotatable linear motion arm Typical vacuum 0.02 ntorr (0.1 ntorr during Sb deposition)

13 Substrate & Recipe Copper Substrate Recipe Following D. Dowell (NIM A ) Polished Solid Copper Stainless Steel Shield Stainless Section ~30 nm Copper Sputtered on Glass 100 Å Sb 200 Å K Cs to optimize QE Substrate Temperature 150 C 140 C 135 C Cool to room temperature as quickly as possible (~15 min)

14 140 Current Cs Deposition Temp min to cool to 100C Lose 15% of QE Photocurrent A μ ) ( Time (min) Substrate Temperature (C) 120

15 Spectral Response

16 Temperature Dependence On a ftershield HC test) (8 days On a ftershield HC test) (8 days On a T 7 fter 7C) = Shield HC test (8 days QE hv (ev)

17 Position Scan (532 nm) 8 7 Initial Scan 24 hrs 20 days SS Cath 6 After Oxygenation 5 Transmission Mode SS Shield 4 Cu 3Cath Photocurrent A μ ) ( Position (mm) 510 Cu transmission ~20% Transmission Photocurrent A μ ) ( Window

18 Copper vs Stainless nm ma On Fork (24 hrs) On Cathode (24 hrs) On Shield (48 hrs) On Shield (7 days) QE High current Wavelength (nm) 690

19 Summary Alkali Antimonide cathodes have good QE in the visible and near UV Narrow valance band from Sb 5p level Band gap depends on which alkali metals used Work function depends on surface termination (and metals used) May be room for improvement by growing better crystals Optimal work function depends on wavelength range of interest For thin cathodes, it may be possible to enhance the QE by tailoring the thickness to improve absorption near emission surface Practical aspects, such a choice of substrate material, surface finish of substrate, and cooling rate after deposition can have a dramatic effect on the QE Thanks for your attention!

20 Additional Slides

21 Photoinjector Basics Why use a Photoinjector? Electron beam properties determined by laser Timing and repetition rate Spatial Profile Bunch length and temporal profile (Sub ps bunches are possible) High peak current density 105 A/cm2 Low emittance/temperature <0.2 µm rad Cathode/Injector Properties Quantum Efficiency (QE) QE= e γ emitted =hν incident I P Lifetime: time (or charge) required for QE to drop to 1/e of initial Response Time: time required for excited electrons to Q bunch escape I = p τ bunch Peak Current: G. Suberlucq, EPAC04, 64 JACoW.org

22 SUPERFISH simulation SW, TM010 f = MHz Q0 = 7.07x109 Pd = 5.1W Bmax/Emax = 2.2 mt/(mv/m) Emax/Ecathode = SUPERFISH simulation Electric Field Photoinjector SW, TM010 f = MHz Q0 = 7.07x109 Pd = 5.1W Bmax/Emax = 2.2 mt/(mv/m) Emax/Ecathode = Time

23 Thin Cathode QE % QE in reflection mode Position in mm

24 QE Decay, Small Spot 80 µm FWHM spot on cathode (532 nm) 1kV bias 2kV bias 3kV bias QE 1.3 ma/mm2 average current density (ERL goal) Hours

25 Linearity and Space Charge 80 µm FWHM spot on cathode