field enhancement coefficient determination methods: dark current and schottky enabled...
TRANSCRIPT
Field enhancement coefficient determination methods: dark current and Schottky enabled photo-emissions
Wei Gai
ANL
CERN RF Breakdown Meeting
May 6, 2010
2
SCHOTTKY-ENABLED PHOTOEMISSION 1
Study a new regime of electron beam generation that has potential impact on the future Linear Collider and Light sources;
High brightness electron beam – minimize emittance; Scheme: Employ Schottky effect in the RF photocathode gun and with
low energy photons; This technique also produces a reasonable estimate of the field
enhancement factor without employing the Fowler-Nordheim model.
“… it now appears that the intrinsic cathode emittance will be the dominant quantity that limits our achieving beams of higher brightness from an rf gun.”
ANL Theory Institute Workshop on Production of Bright Electron Beams Sept. 22-26, 2003
In this presentation, thermal emittance = intrinsic emitance
Yusof et al., Phys. Rev. Lett. 93, 114801 (2004)
3
Emittance from RF photocathode {KJ Kim, NIM A275, 1989}
xscRFscRFthermal J 22222
where RF and sc are the emittances from RF field in the gun and from space charge effect, respectively. Jx is correlation between RF and sc and normally is zero.
There are ways to do emittance compensation to handle RF and sc {B. Carlsten, NIM A285, 313 (1989)}, so in principle, the only limitation left is the thermal emittance thermal:
SCHOTTKY-ENABLED PHOTOEMISSION 2 – Emittance in RF Photoinjector
4
where h : photon energy : material’s bulk work function : a constant : field enhancement factor : RF phase E : Electric field magnitude
Our scheme is to use h< , and then employ the Schottky effect to lower the effective work function eff, where
At the cathode 0 xpx
Therefore,
However, for a cathode in an electric field E: )( EhEkin
cm
Emx
cm
px kin
rmsrms
rmsrmsthermal0
0
0,
2
xxthermal pxpxcm
22
0
1
For a typical cathode: hEkin
)( Eeff
Minimizing Ekin will minimize thermal.
-eff
SCHOTTKY-ENABLED PHOTOEMISSION 3 – Thermal Emittance
5
Metal
Image potentiale2/160z
Electrostatic potential-eEz
Effective potential
z0
eff
EF
0
3
4Ee
eff = -
SCHOTTKY-ENABLED PHOTOEMISSION 4 – The Schottky Effect
6
Beam dump
ICT
Solenoid
Solenoid
Cathode
5 eV 3.3 eV
Mirror
YAG
Camera
Light source: Frequency-doubled Ti:Sapphire laser 372 nm (3.3 eV), 1 – 4 mJ, 8ps.
Photocathode: Mg, = 3.6 eV.
Example of Schottky effect on the cathode: at E() = 60 MV/m, ~ 0.3 eV
SCHOTTKY-ENABLED PHOTOEMISSION 5 – Schematic of Beamline
7
SCHOTTKY-ENABLED PHOTOEMISSION 6 - RF Scans
ee
eE() = - Emax sin()
Photocathode
h
E(
)
Phase (deg)
RF Phase
Laser injection
RF frequency = 1.3 GHz (Period ~ 770 ps)
Laser pulse length = 6 – 8 ps
Metallic photocathode response time ~ fs
We can safely assume that all the photoelectrons emitted in each pulse see the same E-field strength
8
SCHOTTKY-ENABLED PHOTOEMISSION 7 - Charge Obtained From Typical RF Scans
0 20 40 60 80 100 120 140
Cha
rge
RF Phase (deg)
1nC, 100 MV/m
Theoretical RF Scan
Theoretical result from PARMELLA simulation of our RF photoinjector (H. Wang)
We see the “expected” flat-top profile
Full range of charge detected ~ 130 degrees
0 20 40 60 80 100 120 1400.0
0.1
0.2
0.3
0.4
Cha
rge
Q (
nC)
RF Phase (deg)
28 MV/m 17 12
Experimental RF Scan
X.J. Wang et al. Proc. 1998 LINAC
Our scans
9
0 20 40 60 80 100 1200.00
0.05
0.10
0.15
0.20
0=20O
0=30o
Cha
rge
(nC
)
RF Phase (deg)
Emax(MV/m)
92
77
52
28
17
14
0=50O
h = 3.3 eV; = 3.6 eV
Laser beam diameter = 2 cm (0.35 mJ/cm2)
A noticeable shift of the onset of photoelectron production with decreasing RF power.
An RF phase scan allows us to impose different electric field magnitude on the cathode at the instant that a laser pulse impinges on the surface, i.e. E() = Emax sin().
h = 5 eV, = 3.6 eV
No change in the phase range over all RF power.
New observation Typical photoinjector conditions
0 20 40 60 80 100 120 1400.0
0.1
0.2
0.3
0.4
Cha
rge
Q (
nC)
RF Phase (deg)
28 MV/m 17 12
SCHOTTKY-ENABLED PHOTOEMISSION 8 – Experimental Result 1: RF Phase Scans
10
0 (deg) Emax (MV/m) E0 = Emax sin(0) (MV/m)
20 28 9.2 6.8
30 17 8.5 7.3
50 14 11 5.8
0 20 40 60 80 100 1200.00
0.05
0.10
0.15
0.20
0=20O
0=30o
Cha
rge
(nC
)
RF Phase (deg)
Emax(MV/m)
92
77
52
28
17
14
0=50O
2)( EhQ
0
max
4
)sin()(
e
EE
At threshold, Q = 0. This allows us to make a reasonable estimate of the maximum .
h = 3.3 eV; = 3.6 eVThis is a new and viable technique to realistically determine the field enhancement factor of the cathode in a photoinjector
SCHOTTKY-ENABLED PHOTOEMISSION 9 – Determination of Field-Enhancement Factor
11
SCHOTTKY-ENABLED PHOTOEMISSION 10 - Experimental Result 2 : Intensity Dependence
Parameters:
h = 3.3 eV
= 3.6 eV
E field on cathode: 80 MV/m
Laser spot diameter: 2 cm
0.4 0.6 0.8 1.0 1.2 1.40.04
0.08
0.12
0.16
Cha
rge
(nC
)
Laser Energy Per Pulse (mJ)
As we increase the laser intensity, we detect more charge. We definitely are detecting photoelectrons and not dark currents!
12
SCHOTTKY-ENABLED PHOTOEMISSION 11- Experimental Results 3 : Detection Threshold?
0 20 40 60 80 100 120 140 160 180
0.2
0.4
0.6
0.8
1.00.0
0.2
0.4
0.6
0.8
1.0
Am
plitu
de (
arb.
uni
ts)
(deg)
Am
plitu
de (
arb.
uni
ts)
0 20 40 60 80 100 1200.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
Cha
rge
(nC
)
(deg)
Simulated Detection Threshold Using A Sine Function
Experimental Observation
Emax = 28 MV/m
The shift in the photoemisson threshold is not due to the detection threshold.
Two different scans with different amount of charge produced, but with the same RF amplitude, show the same phase angle for the photoemission threshold.
No cutoff
With cutoff
13
The Effective Work Function for Cu Under External Electric Field
0 20 40 60 80 1001 .6
1 .8
2 .0
2 .2
2 .4
2 .6
2 .8
3 .0
3 .2
3 .4
3 .6
3 .8
4 .0
4 .2
4 .4
4 .6
ef
f (e
V)
A p p lie d F ie ld (M V /m )
E ffe c tiv e W o rk F u n c tio n fo r C u (0 = 4 .6 e V )
=1
=7
=10
=15
=20
=50
377 nm
400 nm
14
Estimate of the Field-Enhancement Factor in an RF Photoinjector
15
Estimate of the Field-Enhancement Factor in an RF Photoinjector
Measured dark current as a function of forward rf power in the gun (Fig. 8)
To extract the values of (i) current I and (ii) E-field, I used the following:
(i) E-field
P = kE2.Since at P=3 MW, E=120 MV/m, we have k=2.08×10-4 MW/(MV/m)2. From this, E-field can be extracted from the data.
(ii) Current I
Assuming that the dark current pulse length is 2.5 s, then I = Q/t. The values of I can then be extracted from Q.
16
Estimate of the Field-Enhancement Factor in an RF Photoinjector
An estimate of the field-enhancement factor is made using the Fowler-Nordheim model.
8 .0x10 -9 1 .0x10 -8 1 .2x10 -8 1 .4x10 -8 1 .6x10 -8-23 .8
-23 .6
-23 .4
-23 .2
-23 .0
-22 .8
-22 .6
-22 .4
-22 .2
-22 .0
Lo
g(I
/E2.
5 ) (A
.V-2
.5. m
2.5 )
1 /E (m /V )
The non-linear data points (red) have been excluded in the fit.
Using Eq. 14 from Wang and Loew (SLAC-Pub-7684), for the case of an RF field, the slope of the graph can be expressed as:
5.195.2
10 1084.2
)1(
)(log
Ed
EId
For Cu, = 4.6 eV. We then obtain the value of from the linear fit to be ~100.
Question: Is this the same β as in the Schottky experiments?A: Don’t know, but underline physics should be the same.
β~100
What next?
17
We are planning to perform further experiments with Cu cathodes in both L and S-band guns.(possible setup are at ANL and Tsinghua University)