Transcript Report on recent dielectric wakefield experiments: tunable and

Recent Euclid Wakefield
Experiments @ AWA
C. Jing, S. Antipov, A. Kanareykin, P. Schoessow, Euclid Techlabs, LLC
M. Conde, W. Gai, W. Liu, J. Power, Z. Yusof, HEP, ANL
HG Workshop, Feb. 2011
I. Experiment on Transformer Ratio
Enhancement Using a Ramped
Bunch Train
2
Wakefield Transformer Ratio
Collinear Wakefield Acceleration
W+max
q
Q
W-min
Transformer ratio R =
Max energy gain of the witness bunch
Max energy loss of the drive bunch
Transformer ratio limited: R≤2 @ a longitudinally symmetric drive bunch, but it
can be enhanced greater than 2 using asymmetric bunch.
3
To Enhance the TR
Scheme I---Single
Triangular Bunch
c
 (z)
W+
z W
Reference: Bane et. al., IEEE Trans. Nucl. Sci. NS-32, 3524 (1985)
+
W
Scheme II---Ramped
Bunch Train
(z)
z Wd
d
d
Reference: Schutt et. al., Nor Ambred, Armenia, (1989)
RBT: d=(1+1/2)λ, acceleration for the second bunch, Q1=3Q0,
W+=(3-1)W0+=2W0+, W0- =(3-2)W0- =W0-, R=2R0
Rn= nR0~2*n for the large number of bunches
4
The previous experiment by joint effort
from Euclid Techlabs and AWA (2006)*
Measured
Enhancement
factor of
R2/R1=1.31
Inferred R2=2.3
5
* Funded by DoE SBIR Phase
II
The latest experiment by joint effort
from Euclid Techlabs and AWA (2010)*
What’s the same comparing to the previous experiment?


Same Ramped Bunch Train Technique.
Same DLA Structure
What’s new comparing to the previous experiment?


Laser stacking technique to elongate the bunch
length of AWA beam.
Improved data taking conditions (upgraded LLRF, remotely
controlled delay line for the witness bunch, independent controlled shutter for
each bunch, etc.)
*Results will appear in PRSTAB soon.
6
FWHM~8 ps
14 mm
a-BBO
-20
-10
0
time (ps)
10
7 mm
a-BBO
20
2 crystal set
Intensity (arb. unit.)
Intensity (arb. unit.)
Laser stacking for RBT
Experiment
FWHM~24 ps
-20
-10
0
10
20
time (ps)
Streak camera
measurement
FWHM~
24 ps
Bunchlength=2.7mm
from Parmela
simulation
7
Measurement

Direct measurement of the wakefield transformer ratio for a
single bunch:
Normalized selfwake
41.1KeV / nC
W -d  
 2.5KeV / nC
79.7 KeV / nC

W

d
Normalized wake behind

 8.7 KeV / nC
(mean)
( STD)
(mean)
( STD)
R1=1.94
W - d 1 d 2
1
W - d1

Tune spacing and charge ratio to achieve:

Direct measurement of the wakefield transformer ratio
Enhancement after the 2nd bunch:
Wake behind the 1st bunch W  d 1  82.4 KeV
 15.1KeV
Wake behind the 2nd bunch W  d 1 d 2  
144.6 KeV
 21KeV
(mean)
( STD)
(mean)
( STD)
R2=3.4
R2/R1=1.75
8
II. Experiment on the 1st Tunable
DLA Structure*
* Funded by DoE SBIR Phase II
9
Tunable DLA Structure
By introducing an extra nonlinear ferroelectric
layer which has dielectric constant sensitive to
temperature and DC, the frequency of a DLA
structure can be tuned on the fly by controlling
the temperature or DC voltage.
Geometric and accelerating parameters
value
Radius: b0, b1, b2
4.79mm, 6.99mm, 7.49 mm,
Effective Length
101.6mm
Dielectric constant: dielectric, ferroelectric
6.8, 310 (at room Temp.)
Loss tangent: dielectric, ferroelectric
2*10-4, 2*10-3
Freq. of two dominant wakefield modes
7.8GHz, 14.1GHz (at room Temp.)
Q of two dominant wakefield modes
385, 1250
Peak wakefield by 50nC drive bunch (z=2.3mm)
16MeV/m
10
Bench Test
Ferroelectric layer
Ceramic layer
11
Wakefield Experiment
The experiment demonstrated that by varying
the temperature of the structure over a 50C
temperature range, the energy of a witness
bunch at a fixed delay with respect to the drive
beam could be changed by an amount
corresponding more than half of the nominal
structure wavelength.
26.2cm
12
Bench Test---DC Voltage
Although introduction of a high DC voltage to a tunable DLA structure in the vacuum
environment appears to be challenging at this moment, this approach is still attractive
because of its extremely short response time compared to the temperature control.
One can conclude the best solution for the future tunable DLA structures would be a
combination of ”coarse” but slow temperature tuning by 100s of MHz and rapid fine
tuning with high voltage dc biasing applied.
13
Summary
•Wakefield transformer ratio of 3.4 has been achieved
in the recent experiment at AWA facility with help of
the elongated bunch length.
•A novel low loss BSTM ferroelectric material has been
used in dielectric based accelerators as a method of
frequency tuning. Wakefield acceleration experiment
show an excellent tuning capability through control of
either temperature.
14