Overwiev of photocathodes for high brigthness beamsphotocathodes2011.eurofel.eu/e103312/e110146/photo...Overwiev of photocathodes for high brigthness beams L. Cultrera Cornell Laboratory

Overwiev of photocathodes for high brigthness beamsfor high brigthness beams

L. Cultrera

Cornell Laboratory for Accelerator-based Sciences and Education

Cathodes for electron guns

Photocathodes Field Emitter Thermionic

Metals Semiconductors Spindt Si nanoCNT

CeB6

Pure Coated PEA NEA

Cu

Mg

Pb

Y

Cu-CsBr

Cu-MgF2

Cu-Cs

W-Cs

K2CsSb

Cs2Te

Cs:GaAs

Cs:GaN

Cs:GaAsP NOT EXAUSTIVE LIST!

BaO

SEY Diamond Window

Alloys

Brigthness of an electron beam

ynxn

InB

,,

2εε=

• Brightness is essentially limited by:– thermal emittance

• Intrinsic property of each photcathode

– achievable current• Photocathode => QE and laser damage, response time

ynxn ,,

( ) ( )( )( )

( )( )effFeff

eff

ee

opt E

RQE

φφφν

νλνλ

νν+

−

+

−≈

−

81

12

h

φ−

QE and thermal emittance

D.H. Dowell and J.F. Schmerge, Phys. Rev. ST Accel. Beams, 12, 074201 (2009)

( )23mc

v effxth

φσνε

−=

h

Generally speaking to get higher QE from metallic photocathode it should be accepted that this will give higher thermal emittance.

Copper

• Is this a really simple case?

1 cm x 1 cm scale

Cu(111): φwf = 4.94 eV

Cu(100): φwf = 4.59 eV100111

110

M. Greaves, PPP Workshop, BNL October 12-14 (2010)

Micro XRD analysis and profilometry carried out over the surface of a LCLS Cu photocatode revealed that the surface is far to be considered “ideal”.

Roughness of the surfaceKrasilnikov (FEL 2006) developed a simple mathematical model taking accountcat hode surface roughness aiming to predict the TE increase due to local surface orientation

Electric field lines near the surface bumps may deviate electrons from ideal trajectories focusing or defocusing the beam and increasing the effective TE by local Schottky lowering of the vacuum level

0

2

2

4

3

πεβφ

φφϖσε

Ee

cm

Schottky

e

Schottkylaserth

=

+−≈

h

Metal photocathodes properties

Dowell et al., NIM-A, 622, (2010) 685-697

Metallic photocathode: lifetimeDespite their lower claimedcontamination sensitivity even inUHV (10-9 mbar range) low workfunction metals as Mg, Y but alsothe most inert Cu may suffer fromthe contamination due to chemicalspecies present in residual gases(H2, CO, CO2, H2O).

Background Pressures

w/o RF ~ 5x10-10 mbar

with RF ~ 2x10-9 mbar

Cu

Y Mg

Surface cleaning to restore QE• Laser cleaning (rastering focused laser beam)

– Performed on Cu cathode improves QE values and uniformity cut at expenses of surface roughness

+6° -6°

• Similar treatment performed on Mg photocathode does not seems to affect surface roughness: – difference in contaminants: Carbon for Cu and Oxygen for Mg

generates different kind of chemical bonds.– Removal of different chemical species may require different

approach…

H-ion bombardment and Ozone

• H-ion gun sputtering of Cu surface has been demonstrated to be effective in removing the Carbon contaminants from the surface (practical implementation for in-situ may be challenging)

• Similarly the exposure of internal Cu surface of the gun/cathode assembly to ozone flux has been demonstrated to be effective on removing Carbon contamination

Dowell et al, PRSTAB, 9, 063502 (2006)Penco , IPAC 2010

Magnesium

• Many years of intense R&D due to the high QE of Mg illuminated with UV

• Incorporation of the cathode onto Cu backplate by– friction welding (BNL)– thin film deposition (INFN)

• High QE (>10-3) demonstrated • High QE (>10-3) demonstrated either in – low DC field (INFN)– RF gun (BNL)

• Laser cleaning necessary to remove surface MgO

• Low thermal emittance (0.5 mm mrad /mm) measured in RF gun (BNL)– Surface photoemission

H.J. Quian et al., App Phys Lett, 97, 253504 (2010)L. Cultrera et al., Phys. Rev. ST Accel. Beams, 12, 043502 (2009)

• Wide band gap thin film coatings are givinginteresting and in some way unexpected results

Improving lifetime

Accurate choice of the coating material to be transparent to the laser wavelength. The thickness may be designed to create an

antireflecting coating at the cathode surface.

Metals with CsBr coating

• Transmission mode @ 257 nm

• Photoemission arises from intraband states

• Electrons injected through the metal insulator junction sustain the emission

• Reflection mode @ 257 nm

• 50 times higher QE

• No strong degradation of QE due to air exposre

Z. Liu et al., Appl. Phys. Lett., 89, 111114 (2006)J. Maldonado et al., Phys. Rev. ST Accel. Beams, 11, 060702 (2008)

Cu with MgF2 coatingMgF2 CB bending

CBPotential with ext E field

Bare metal

P. Musumeci, PPP Workshop

• Short response time

• Initially designed to be AR coating at 266 nm but then used for multiphoton emission @800 nm, some aspect deserve to be investigated:

• Higher Shottky effect due to the electric field inside MgF2?

• Can we expect lower dark current due to higher width of the barrier?

VB

Yttrium • Bare metals require UV photons (e.g. 266 nm, 3rd harmonic Ti:Sa)

to be operated;

• Yttrium demonstrated linear photoemission in visible range (400nm, 2nd harmonic Ti:Sa) after laser cleaning of the surface

QE ~ 3.4x10-5

1 nC should require about 100 µµµµJ laser pulses @ 400 nm

Time (s)

QE

Pbg=10-9 mbar

E = 1.7 MV/m

λ = 406 nm

Metal-based alloys (Mg–Ba, Al–Li, and Cu–BaO)

V.G. Tkachenko, A.I. Kondrashev, I.N. Maksimchuk, Appl Phys B, 98, 839 (2010)

All alloying additives irrespective of their nature reduce the work function values for simple metals pushing the emission threshold from UV to VIS wavelength.

Semiconducting photocathodes properties

Dowell et al., NIM-A, 622, (2010) 685-697

Cesium Telluride• Routinely used in photoinjectors (Pitz, FLASH, CLIC..)

• Simple synthesis recipe (10-15 nm Te exposed to Cs vapors)

• QE suffers of large decrease during the initial stages of run but the value stays high enough to ensure months of operation

• UHV Load-lock system allow to test large number of cathodes

V. Miltchev, FEL 2005P. Michelato, EPAC 2008

Alkali AntimonideLaboratory Los Alamos CEA Wuppertal BOEING

Project HIBAF/APEX ELSA SRF injector APLE

Year 1993 1990 1989 1993

Cathode CsK2Sb CsK2Sb Cs3Sb CsK2Sb

• In past years several efforts have been made to obtain a reliable photoinjector source using alkali antimonides based photocathodes.

• Boeing 433 MHz NC gun operating with CsK2Sb (typ. QE 8% at 543 nm) photocathode set the 35 mA average current (still the world record for such injector) but the lifetime of the cathode was limited to few hours due to the poor vacuum conditions.

Cathode CsK2Sb CsK2Sb Cs3Sb CsK2Sb

Cornell R&D on alkali antimonide

• CsK2Sb for ERL photoinjector prototype

• Transfer of photocathode in vacuum suit

• Using CsK2Sb photocathode we were able to deliver 20 mA average current for 8 hours without any noticeable decrease of QE (still lack of uniformity)

GaAs with cesiated surface•Negative affinity achieved by “yo-yo” techinque;

•Thermal emittance values measured are very

good even with IR or VIS radiation;

•This cathode type suffers from ion backstream

that limits the lifetime to few hours;

•Achieved high QE (~15%), low thermal emittance

(0.4 mm mrad for 1mm σ) @532 nm;

•Response time < 1 ps @532 nm

I. Bazarov et al., Phys. Rev. ST Accel. Beams, 11, 040702 (2008)

GaAs Roughness

Photoemission model including also phonon scattering reproduces quite well experimental data

S. Karkare and I. Bazarov, APL (2011)

AFM carried out after each photocathode preparation step reveal unexpected increase on surface roughness after heat cleaning process

This model indicate that the surface roughness is dominant factor on the worsening of thermal emittance

GaAs Lifetime

20

15

Bea

m C

urre

nt (

mA

)

0.15

Exit Laser P

ower (W

)

1.2

1.0

0.8

QE

(re

lativ

e)11/16/2010

15 min

10

5

0

Bea

m C

urre

nt (

mA

)

500040003000200010000Time (second)

0.10

0.05

0.00

Exit Laser P

ower (W

)

0.6

0.4

0.2

0.0

QE

(re

lativ

e)

1 hr

min

8 min

2.5 hr

Ion back bombardment

Copper (1 nC @ 10Hz => 10 nA)

GaAs (20 mA)

CsTe (72 uC @ 10Hz => 72 uA)

Copper (1 nC @ 10Hz => 10 nA)

Penco , IPAC 2010S. Lederer et al, FEL 2007

Diamond Window Amplifier

TransparentConductor

(Sapphire w/ITOor thin metal )

LaserPrimary

electronsSecondaryelectrons

DiamondThin Metal

Layer(10-30 nm)

HydrogenTermination

Primaryelectrons

Secondaryelectrons

NEA surface to generate cold electron beam Secondary electron to get high cuurent beam

X. Chang et al, PRL 105, 164801 (2010)

3 – 10 kV

Photocathode(K2CsSb)

Laser electrons electrons

Diamond(30 µm)

electrons electrons

Diamond Window Amplifier

Capsule mounting has been engineered;Gain higher than 200 have been measured;

Electron beam amplified (gain 40) transported in vacuum outside the diamond and observed on a scintillator screen

X. Chang et al, PRL 105, 164801 (2010)

Photo field assisted cathode

L. Hudansky et al., Nanotechonology, 19, 105201 (2008)R. Ganter et al., Phys. Rev. Letter, 100, 064801 (2008)

Thermionic optically switchedMAX-lab linac injectorhas been equipped with a Ti:Sa laser to operate the BaO thermionic cathode as photocathode.

When used as photocathode BaO temperature is lowered from 1100 to 700 °C

S. Thorin et al. ,Nucl. Instr. Meth. A, 606, 291 (2009)

from 1100 to 700 °C

Photocathode Engineering

Mg O

Ag

limit the emission to a single surface band and make kmax as small as possible

The transverse intrinsic emittance is a function of the number of MgO overlayers;

Minimum at nMgO =2,3 with εth =0.06mm mrad

Maximum sensitivity to nMgO can be achieved with thin Ag (<<8ML) and MgO onboth top and bottom surfaces (work function greatly reduces from ~4.6 eV to 2.92eV due to the MgO overlayers)

K. Németh et.al, PRL,104, 046801 (2010)

Conclusions

• Photoinjectors represent the state of the art of electron sources for XFEL and ERL

• Despite the increasing interest on this field experimental data are still scattered

• Collaboration efforts between material science and accelerator research groups is

• Collaboration efforts between material science and accelerator research groups is required to push forward photocathode performances

• Requirement of dedicated photocathode development labs allowing synthesis and test in real gun of larger number of samples

Collaboration Website Deployment

http://photocathodes.chess.cornell.edu