- Email: [email protected]
Commissioning of the electron injector for the AWAKE experiment
Journal Pre-proof Commissioning of the electron injector for the AWAKE experiment S.-Y Kim, S. Doebert, O. Apsimon, R. Apsimon, G. Burt, M. Dayyani, S. Gessner, I. Gorgisyan, E. Granados, S. Mazzoni, J.T. Moody, M. Turner, B. Williamson, M. Chung
PII: DOI: Reference:
S0168-9002(19)31488-3 https://doi.org/10.1016/j.nima.2019.163194 NIMA 163194
To appear in:
Nuclear Inst. and Methods in Physics Research, A
Received date : 1 November 2019 Revised date : 27 November 2019 Accepted date : 28 November 2019 Please cite this article as: S.-Y. Kim, S. Doebert, O. Apsimon et al., Commissioning of the electron injector for the AWAKE experiment, Nuclear Inst. and Methods in Physics Research, A (2019), doi: https://doi.org/10.1016/j.nima.2019.163194. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Journal Pre-proof Revised version of manuscript Click here to view linked References
of
Commissioning of the electron injector for the AWAKE experiment
pro
S.-Y Kima , S. Doebertb , O. Apsimonc , R. Apsimond , G. Burtd,e , M. Dayyanif , S. Gessnerb , I. Gorgisyanb , E. Granadosb , S. Mazzonib , J. T. Moodyg , M. Turnerb , B. Williamsonc,e , M. Chunga a Department
Abstract
re-
of Physics, Ulsan National Institute of Science and Technology, Ulsan 44919, Republic of Korea b CERN, Geneva 1211, Switzerland c Manchester University, M13 9PL Manchester, United Kingdom d Lancaster University, LA1 4YW Lancaster, United Kingdom e Cockcroft Institute, WA4 4AD Warrington, United Kingdom f Institute for Research in Fundamental Sciences, 19395-5531, Tehran, Iran g Max Planck Institute for Physics, 80805 Munich, Germany
urn al P
The advanced wakefield experiment (AWAKE) at CERN is the first proton beam-driven plasma wakefield acceleration experiment. The main goal of AWAKE RUN 1 was to demonstrate seeded self-modulation (SSM) of the proton beam and electron witness beam acceleration in the plasma wakefield. For the AWAKE experiment, a 10-meter-long Rubidium-vapor cell together with a high-power laser for ionization was used to generate the plasma. The plasma wakefield is driven by a 400 GeV/c proton beam extracted from the super proton synchrotron (SPS), which undergoes a seeded self-modulation process in the plasma. The electron witness beam used to probe the wakefields is generated from an S-band RF photo-cathode gun and then accelerated by a booster structure up to energies between 16 and 20 MeV. The first run of the AWAKE experiment revealed that the maximum energy gain after the plasma cell is 2 GeV, and the
Jo
SSM mechanism of the proton beam was verified. In this paper, we will present the details of the AWAKE electron injector. A comparison of the measured electron beam parameters, such as beam size, energy, and normalized emittance, Email addresses: [email protected] (S. Doebert), [email protected] (M. Chung)
Preprint submitted to Elsevier
November 27, 2019
Journal Pre-proof
with the simulation results was performed.
1. Introduction to AWAKE
pro
1
of
Keywords: Plasma, Wakefield, AWAKE, Electron injector
The advanced wakefield experiment (AWAKE) at CERN studies electron
3
beam acceleration in proton beam-driven plasma wakefields [1, 2]. The proton
4
beam driver has a momentum of 400 GeV/c, a bunch length of 6-12 cm (RMS),
5
and a transverse beam size of 0.2 mm (RMS). The plasma is generated in a
6
10-meter-long heated Rubidium-vapor source that can reach plasma densities
7
of the order of 1 × 1014 ∼ 1 × 1015 /cm3 . The ionization laser has an energy
8
of 450 mJ and a pulse length of 120 fs (FWHM), allowing full ionization of a
9
plasma channel with a radius of 1 mm. The laser pulse and the proton beam
10
co-propagate through the vapor source to create the plasma, and seed the self-
11
modulation within the long proton bunch. The maximum accelerating gradient
urn al P
re-
2
14
of the plasma wakefield [3] Emax = me wpe c/e is determined by the plasma p electron density ne , where me is the electron mass, wpe = ne e2 /me 0 is the
15
is the electron charge. In the case of AWAKE, the nominal plasma density is
16
7 × 1014 /cm3 , and the maximum accelerating gradient is therefore expected to
17
be on the order of GV/m.
12
13
plasma frequency, 0 is the vacuum permittivity, c is the speed of light, and e
18
An important condition particular to AWAKE for the generation of a high
19
plasma wakefields is the seeded self-modulation (SSM) of the proton beam [4,
20
21
wpe /c, and σz is the RMS bunch length of the drive beam [6]. Therefore, the
Jo
22
5]. In the linear plasma wakefield theory, the optimal condition to drive a √ plasma wakefields is kp σz ∼ = 2, where kp is the plasma wavenumber defined by
23
initial proton beam extracted from the super proton synchrotron (SPS) does
24
not satisfy the condition needed to generate high-gradient wakefields owing to
25
its long bunch length. However, once the long proton bunch has propagated
26
for a distance in the plasma, it splits into micro-bunches that have a length
27
corresponding to the plasma wavelength and meet the condition for driving the
2
Journal Pre-proof
high-gradient plasma wakefield [7]. The self-modulation of the proton beam is
29
reliably seeded by the front of the ionization laser, which is placed in the center
30
of the proton beam. The self-modulation of the proton beam density starts from
31
this point, enabling phase stable wakefields.
of
28
In the first run of the AWAKE experiment to demonstrate electron beam ac-
33
celeration through the plasma wakefields, an electron injector that consists of an
34
S-band RF photocathode gun and a traveling wave booster structure were used
35
to produce electron beams with an energy of 16-20 MeV [8]. The electron beam
36
was then passed through an electron transfer line, and injected into the plasma
37
cell co-propagating with the proton beam and the ionization laser [9, 10]. An
38
energy gain of the electron beam in the 10 m plasma source of up to 2 GeV [11]
39
has been demonstrated experimentally. In addition, the SSM phenomenon has
40
been successfully observed and studied by direct measurement of the modulated
41
proton bunch using streak cameras, and by indirect measurement of the proton
42
beam divergence due to the transverse wakefield [12, 13]. A schematic of the
43
AWAKE experiment can be seen in Fig. 1 and main parameters of the proton,
44
electron beam and ionization laser can be seen in Table 1.
urn al P
re-
pro
32
Details of the electron beam parameters required for electron beam acceler-
46
ation in the plasma wakefield are summarized in Refs. [14, 15]. Since the main
47
goal of the AWAKE RUN 1 experiment was to determine whether the electron
48
beam is accelerated in the proton beam-driven plasma wakefield, the focus was
49
on the matching of the electron beam to the plasma channel containing the
50
wakefields. The biggsest constrain was the transverse RMS beam size at the
51
entrance of the plasma cell and therefore the emittance of the electron beam. In
52
order to control the electron beam size and the emittance along the transfer line,
53
and to satisfy the parameters required at the plasma entrance, understanding of
54
the electron injector was essential. Therefore, the main parameters of the elec-
55
tron beam along the electron injector have been measured and compared with
56
the simulations to determine the operational conditions. In this paper, we will
57
present the details of the electron injector. A description of the electron injector
58
and commissioning results such as transverse beam distributions on screens and
Jo
45
3
of
Journal Pre-proof
Table 1: Main parameters of the proton, electron beam and ionization laser for the AWAKE
pro
experiment.
Parameters
Value
Proton beam (extracted from SPS)
Bunch length σz Bunch radius σr
400 GeV
re-
Energy
12 cm
0.2 mm 3.5 mm mrad
Energy spread σδ
0.45 %
urn al P
Normalized emittance n
Electron beam
Initial set value
(at the plasma entrance)
Value used for the experiment
Energy
16 MeV
16 - 20 MeV
Charge
200 pC
140 - 650 pC
Bunch length σz
4 ps
1 - 4 ps
Bunch radius σr
0.25 mm
0.25 mm
Normalized emittance n
2 mm mrad
2 mm mrad
Energy spread σδ
0.5 %
0.5 %
Ionization laser (Fiber Ti:Sa)
450 mJ
Pulse length σz
100 - 120 fs
Focused spot size σr
1 mm
Jo
Energy
4
re-
pro
of
Journal Pre-proof
Figure 1: Schematic of the AWAKE experiment at CERN. The upper right corner shows an
urn al P
example of a beam spot as observed in the spectrometer. The inserts in the lower left corner illustrate the injection conditions and the process of seeded self-modulation.
59
beam emittance measurements using pepper-pot and quadrupole scan methods
60
will be shown in Sections 2 and 3. A careful comparison of the measurements
61
with the simulation results in order to characterize the operating condition of
62
the electron injector will be presented in Section 4. Finally, the results of the
63
commissioning will be concluded in Section 5.
64
2. Design and construction of the injector An electron injector providing adequate beam quality to demonstrate the
66
first proton driven plasma wakefield acceleration experiment was designed and
Jo
65
67
constructed by AWAKE collaborators. The original target parameters for the
68
injector were energy of 16 MeV, a bunch charge of 200 pC, an energy spread
69
below 1%, a bunch length of 4 ps, and an emittance of 2 mm mrad. To achieve
70
these parameters, an RF photo-injector was chosen using a 3 GHz, 2.5-cell
71
RF-gun and a one-meter-long booster structure. The RF-gun including a load-
5
Journal Pre-proof
lock system was available at CERN [16], while the booster structure was newly
73
developed for this purpose [17]. The load-lock system enables the use of Cs2 Te
74
cathodes fabricated at CERN with a quantum efficiency of Qe ∼ 10−2 . The laser
75
beam was derived from the main Ti-Sa laser used to ionize and seed the proton
76
beam self-modulation. Consequently, an adequate timing synchronization could
77
be achieved. A small fraction of the initial laser power was frequency tripled to
78
a wavelength of 262 nm and sent with an off-axis mirror on the cathode. The
79
UV-beamline allows users to vary the spot size and energy on the cathode as well
80
as the bunch length as the IR-laser beam has to be compressed before the UV
81
conversion [18]. The RF-gun accelerates the electron bunches to an energy of
82
5.5 MeV, and then the booster can add a maximum energy of 16 MeV. A single
83
klystron is used to power the gun and the booster structure. A high power wave-
84
guide attenuator and phase-shifter allow for individual phasing and powering of
85
the two structures. A particular challenge for the design and construction was
86
the severe space constraints in the AWAKE experiment. The complete injector
87
has a total length of only 5 m before the beam transport towards the plasma
88
cell starts. The injector was equipped with a number of beam diagnostics as
89
described in more detail in the next section.
90
3. Beam Commissioning results
Jo
urn al P
re-
pro
of
72
Figure 2: Schematic electron injector layout emphasizing the location of magnets and diagnostics mentioned throughout the paper.
91
Figure 2 shows a schematic view of the electron injector indicating, in par-
92
ticular, the locations of focusing and beam diagnostic elements used in this 6
Journal Pre-proof
section. The electron injector consists of an S-band RF photocathode gun and
94
booster structure. The main solenoid and the corresponding bucking coil are
95
placed around the RF gun to control the beam focusing and emittance compen-
96
sation. Corrector magnets are used to correct the beam trajectory, which can
97
be measured with several strip-line BPM’s [19]. A pepper-pot beam diagnostic
98
instrument is used to measure the beam emittance out of the RF-gun, and the
99
screen behind the retractable pepper-pot also allows imaging of the transverse
100
distribution of the beam. A quadrupole triplet behind the booster structure
101
is used to match the beam for further beam transport and for the emittance
102
measurements using the quadrupole scan method at higher energies. A YAG
103
screen (BTV430042) is used to measure the transverse beam distribution. The
104
electron beam charge at the end of the injector is measured by a Faraday-cup
105
with high precision. The first dipole magnet of the following transport line to-
106
gether with a second screen also serves as a spectrometer to measure the beam
107
energy and energy spread. During the commissioning, as many beam parame-
108
ters as possible were measured so that they can be used as input parameters for
109
simulations.
urn al P
re-
pro
of
93
First, the initial transverse distribution of laser pulse was measured using
111
the so-called virtual cathode camera. This camera images a fraction of the
112
laser beam in similar imaging conditions as the laser pulse sent to the cathode.
113
The laser energy can be varied with optical density (OD) filters to adjust the
114
final beam charge. The beam charge during the commissioning reported here
115
was measured as 140 ± 10 pC. The initial transverse distribution of the laser is
116
shown in Fig. 3. Figure 3 (a) shows the case without the OD filter, which is used
117
to enhance the UV imaging. The RMS transverse beam size can be determined
118
as σx = 0.33 mm in the horizontal plane, and σy = 0.34 mm in the vertical
119
plane. In the experiment, however, an OD filter was used to reduce the beam
120
charge, which resulted unfortunately in a poorer quality image of the virtual
121
cathode as can be seen in Fig. 3 (b). The transverse size of the laser obtained
122
by analyzing this image appears slightly reduced owing to the cut-off of the laser
123
energy, and we found that σx and σy are 0.27 mm. The smaller values were used
Jo
110
7
Journal Pre-proof
for the particle tracking simulations later on. The bunch length of the electron
125
beam was not measured during the experiment. Previous measurements of the
126
UV laser beam suggested a laser pulse length of σz = 2.2 ps (FWHM 5.2 ps). Experimental data Fit curve
1.5
0.5 200
0 -0.5
1
300
100
20
Experimental data Fit curve
15
0.5
0
-0.5
10 5
-1
-1
-1.5
0
-1.5 -1 -0.5 0 0.5 1
x (mm)
0
-1.5 -1 -0.5 0 0.5 1
re-
-1.5
(b)
pro
y (mm)
1
(a)
y (mm)
1.5
of
124
x (mm)
Figure 3: Initial transverse distribution of the laser at the cathode without an OD filter (a)
urn al P
and with OD filter (b).
127
The energy of the electron beam out of the RF gun was calculated by mea-
128
suring the position of the beam at the pepper-pot screen while scanning the
129
first corrector magnet, as no actual spectrometer is available after the gun.
130
The electron beam momentum out of the RF-gun was determined to be 5.8
131
MeV/c, which corresponds to a nominal energy of 5.3 MeV. However, as this
132
measurement has a much lower precision compared to that of a real spectrom-
133
eter magnet, we estimated the error of the electron beam energy calculation
134
to be about 10%. The energy of the electron beam after the traveling wave
135
structure was measured using the first dipole magnet in a proper spectrometer
136
setup. The momentum of the electron beam after acceleration was determined
137
as 18.5 MeV/c with an error of 0.3%.
To optimize the operating conditions of the electron injector during com-
139
missioning, scans of the main solenoid around the RF-gun responsible for emit-
140
tance compensation have been performed. Beam sizes and emittances have been
141
measured before and after the booster structure as a function of the solenoid
142
current. First, the transverse beam size of the electron beam was measured
143
on the pepper-pot screen. Figure 4 (a) depicts the beam size and normalized
Jo
138
8
Journal Pre-proof
emittance as a function of the solenoid current. The minimum beam sizes σx
145
and σy at the pepper-pot screen are 0.47 mm and 0.38 mm for a solenoid cur-
146
rent of 185 A. We found during the commissioning that it was very difficult to
147
obtain reliable and reproducible emittance measurements with our pepper-pot
148
setup. The cause of this observation is still under investigation. At this point,
149
we tend not to rely on these measurements. We will come back to that issue in
150
the simulation section.
1
5
0.5
170
180
190
200
0
urn al P
0
2.5 2
Solenoid current (A)
7
(b)
6 5
1.5
4
1
3 2
0.5
0 165
170
175
1 180
185
190
0 195
Normalized emittance (mm mrad)
10
1.5
pro
2
15
(a)
re-
rms beam size (mm)
2.5
Normalized emittance (mm mrad) rms beam size (mm)
of
144
Solenoid current (A)
Figure 4: Measured transverse beam size and normalized emittance of the electron beam at the pepper-pot (a) and BTV 430042 screen (b).
In addition to the measurement at the pepper pot screen, the beam size
152
and normalized emittance after the traveling wave structure have been mea-
153
sured. This time a well-established quadrupole scanning method was used as
154
the beam is no longer space charge dominated. The beam size was measured at
155
the BTV430042 screen, while the normalized emittance was determined from
156
a quadrupole scan by varying the magnetic field of the center quadrupole of
157
the triplet (MQAWA430034). Figure 4 (b) shows the beam size at the screen
158
(BTV430042) and the normalized emittance as a function of the solenoid cur-
Jo
151
159
rent. The minimum beam size was found at a solenoid current of 175 A: σx is
160
0.29 mm and σy is 0.30 mm. The minimum normalized emittance, on the other
161
hand, was found at a solenoid current of 185 A, which is 1.0 mm mrad in both
162
the horizontal and the vertical plane. Relative error of the emittance in Fig. 4
163
(b) was determined as follows: When using the quadrupole scan method, beam 9
Journal Pre-proof
size was measured ten times at each quadrupole magnet strength. Then, the
165
polynomial curve fitting was performed to find ten sets of emittance values. The
166
relative error of the emittance in the range of the solenoid current was about
167
10%.
In the next section, we will show a comparative analysis of experimental
pro
168
of
164
169
measurements with the results obtained from the simulations.
170
4. Comparison with simulations
To compare measured beam parameters with the simulation results, particle
172
tracking was performed mainly using the ASTRA [20] code. The cathode mate-
173
rial used in AWAKE is Cs2 Te, and the characteristic parameters of the electron
174
beam emitted from the cathode such as the thermal emittance and momentum
175
distribution can be defined in terms of the Fermi-Dirac distribution [21, 22].
176
In ASTRA, the momentum distribution, and thermal emittance of the initial
177
beam at the cathode are defined as follows [23]:
urn al P
re-
171
σ px = σ py =
σE =
r
Ephoton − φef f , 3me c2
Ephoton − φef f √ , 3 2
x,y = σx,y
r
Ephoton − φef f , 3me c2
(1)
(2)
(3)
where σpx,y is the RMS value of the transverse momentum, Ephoton is the photon
179
energy of the laser, φef f is the effective work function of the cathode material,
180
σE is the energy spread, and x,y is the thermal emittance.
Jo
178
181
Through the equations above and Ref. [22], the transverse momentum and
182
energy spread are determined only by the photon energy of the laser and the
183
work function of the cathode material, regardless of the spatial distribution type
184
of the laser. As the momentum distribution is determined by Ephoton and φef f
185
irrespective of the laser shape, the initial beam distribution measured at the
10
Journal Pre-proof
cathode (see previous section) was used as an input to the simulation. As can
187
be seen in Fig. 3 (b), the distribution is not fully symmetric. To perform the
188
simulations, we converted the measured image into a transverse input distribu-
189
tion suitable for simulations. For the longitudinal components, a distribution
190
generated by ASTRA was used assuming a Gaussian laser pulse shape. For the
191
simulations, as we do not have a real bunch length measurement, we studied
192
three values: 2.2 ps as indicated by previous UV laser measurements and shorter
193
beams with 1.0 ps and 1.5 ps bunch lengths for comparison.
pro
of
186
The RF fields of the RF gun were overlaid with the magnetic fields of the
195
solenoids. The maximum accelerating gradient was set to be 79.6 MV/m to
196
obtain the measured beam energy of 5.3 MeV out of the RF gun. At 1.6 m,
197
a traveling wave structure (booster structure) was placed to increase the beam
198
energy to match the measured value of 18.1 MeV; thus its accelerating gradient
199
was 17.8 MV/m. In the ASTRA simulation, a full 3D field map of the booster
200
structure obtained from CST was used, while the field map of the RF gun
201
was determined by the 1D axial electric field owing to a lack of 3D data. The
202
phase of the gun was checked to achieve the maximum energy gain, nominally
203
corresponding to the smallest emittance as well. Therefore, the beam launch
204
phase was approximately 30◦ off crest in the simulations. In the experiment, we
205
tried to minimize the emittance by varying the phase of the gun and solenoid
206
current around the nominal working point. The booster structure was set to
207
on-crest acceleration as optimized for maximum energy in the experiment.
urn al P
re-
194
Solenoid scans were simulated with bunch lengths of 1.0 ps, 1.5 ps, and
209
2.2 ps. The simulation results together with the measured data are shown in
210
Fig. 5, upper plots for the pepper-pot screen and the lower plots for the BTV
211
430042 screen. In this case, misalignment of the RF structure and solenoid
212
magnet were not considered. One can see that the slopes of the beam size scans
213
match better with shorter bunch length cases on both screens. This results
214
indicate that the bunch length might have been shorter than 2.2 ps as indicated
215
by previous UV measurements. At the pepper-pot screen, the minimum beam
216
sizes σx and σy for the case of 1.0 ps bunch length are 0.32 mm and 0.31
Jo
208
11
2 1.5 1 0.5 0 160
170
180
190
200
210
Solenoid current (A)
10
5
0 160
170
180
190
200
2 1.5 1 0.5
(c)
170
175
180
185
210
7 6 5
re-
Normalized emittance (mm mrad)
2.5
urn al P
rms beam size (mm)
(b)
Solenoid current (A)
3
0 165
15
of
rms beam size (mm)
(a)
pro
2.5
Normalized emittance (mm mrad)
Journal Pre-proof
190
195
Solenoid current (A)
4 3 2
(d)
1
0 165
170
175
180
185
190
195
Solenoid current (A)
Figure 5: Transverse beam size and normalized emittance obtained from the ASTRA simulation at the pepper-pot screen (a, b) and BTV430042 screen (c, d) without misalignement.
mm; furthermore, the minimum values of the normalized emittance are 0.80
218
mm mrad in the horizontal plane and 0.78 mm mrad in the vertical plane. In
219
addition, the minimum beam sizes σx and σy at the BTV430042 screen are
220
0.48 mm and 0.44 mm, and the minimum values of the normalized emittance
221
are very similar to the values at the pepper-pot screen. However, the shape
222
of the scan for the emittance measurements does not correspond well with the
223
simulations. The emittance determination using the pepper-pot data seems to
Jo
217
224
fail because the larger measured values do not agree with the simulations, while
225
the measurements further downstream of the beam line using the quadrupole
226
scan method seem to match well with the simulation results, at least for the
227
minimum values. Therefore, at this point we concluded that the pepper-pot
228
results are not very reliable. The reason for this is still under investigation, and
12
Journal Pre-proof
there are hints that the location and dimensions of our device are not optimal
230
[24].
of
229
During the experiments, the machine operation and in particular the beam
232
alignment were optimized for the minimum emittance values at 185 A, as these
233
beams have been used for the plasma wakefield acceleration experiments. The
234
alignment of the beam was not corrected during the scan. Therefore, we suspect
235
misalignment of the beam due to steering of the solenoid magnet is responsi-
236
ble for the particular shape of the emittance measurements. Particle tracking
237
simulations were used again to study possible offsets of the booster structure
238
and the solenoids, as we had some experimental hints from the beam trajectory
239
data that this might have been the case.
2
urn al P
rms beam size (mm)
2.5
Normalized emittance (mm mrad)
re-
pro
231
1.5 1
0.5
(a)
0 165
170
175
180
185
190
195
Solenoid current (A)
6 5 4 3 2 1
(b) 0 165
170
175
180
185
190
195
Solenoid current (A)
Figure 6: Comparison of the transverse beam size (a) and normalized emittance (b) at BTV430042 screen taking into account a misalignment of the booster structure.
240
Various offsets of the main solenoid and the booster structure have been
241
studied using ASTRA with 3D space charge calculations and a 3D field map
242
for the booster structure. We found that the measurements are best described assuming a relatively big offset of the booster structure only. Figure 6 shows
244
a comparison of the experiment data and the simulation results obtained for a
245
-4.5 mm offset at the entrance of the booster structure. In this simulation, a
246
1.5 ps bunch length was used. Both the emittance and the spot size show good
247
qualitative agreement for these misalignment assumptions.
Jo 243
13
urn al P
re-
pro
of
Journal Pre-proof
Figure 7: Horizontal phase space of the beam at the input coupler section. Top: on-axis simulation; bottom: off-axis simulation with -4.5 mm booster structure offset. Solenoid current: (a, c) 175 A, and (b, d) 185 A.
The emittance growth seen in the simulations is attributed to an RF kick
249
within the input coupler of the booster structure due to its not perfectly compen-
250
sated RF fields. Phase space distortion due to the asymmetric field is illustrated
251
in Fig. 7, where the top row shows the horizontal phase space at the input cou-
252
pler of the booster structure obtained from an on-axis simulation, while the
Jo
248
253
bottom row images are with the booster structure offset. One can see that the
254
momentum distribution of the on-axis case is concentrated in the center. In the
255
case with a large beam offset, simulation results reveal that the momentum dis-
256
tribution is distorted with respect to the core, which leads to emittance growth.
257
However, for well-focused beams at the entrance of the booster (solenoid cur14
Journal Pre-proof
rents of 180 and 185 A), the effect is much smaller, and the emittance growth
259
is not significant.
of
258
In a second beam commissioning campaign, the misalignment of the beam in
261
the booster structure and the question of the not well-known bunch length were
262
addressed. The original misalignment of the solenoids could not be solved, so
263
we corrected the beam at the entrance of the booster structure for each solenoid
264
setting using a pair of corrector magnets when taking data. In addition, as
265
the UV pulse shape and length were not clearly determined during the last
266
measurement, UV pulse measurements using a streak camera were performed
267
again. Unfortunately, it was not possible to obtain reliable data from direct
268
bunch length measurements using light from an OTR screen on the streak cam-
269
era. The intensity arriving at the streak camera was simply not high enough.
re-
pro
260
Amplitude (a.u.)
urn al P
1
(b)
0.8 0.6 0.4 0.2 0 -15
-10
-5
0
5
10
15
Time (ps)
Figure 8: Initial transverse beam distribution at the cathode (a) and longitudinal profile of the UV laser pulse (b). Intensity of the UV laser is normalized.
The transverse beam distribution and charge of the beam were slightly dif-
271
ferent from those in the first campaign. Therefore, the following inputs were
Jo
270
272
used for the simulations. Figure 8 shows the measured transverse beam dis-
273
tribution at the cathode and the measured longitudinal profile of the UV laser
274
pulse. The initial beam size is slightly changed compared to that in Fig. 3 (b).
275
Beam sizes σx and σy are both 0.25 mm. The measured beam charge at the
276
Faraday cup was 150 pC. In the case of the UV measurement, it was confirmed
15
Journal Pre-proof
that the RMS UV pulse length was 1.5 ps. In addition, one can see that there is
278
a secondary pulse behind the main pulse, but it is negligible, as the amplitude
279
is very small compared to the main pulse. This after-pulse was not taken into
280
account in the simulations. Even though the simulation results with a 1.5 ps
281
bunch length did not perfectly match with the experimental data indicated in
282
Fig. 5, it was confirmed that the measured bunch length is shorter than that
283
assumed previously.
pro
of
277
Using the new measured inputs, the ASTRA simulations were performed
285
again. The booster structure gradient was slightly changed to obtain the exper-
286
imentally measured 18.8 MeV/c momentum value in the simulation. This time,
287
we obtained a good agreement between simulations and measurements, con-
288
firming the assumptions concluded from the analysis of the first measurement
289
campaign. The measured emittance values obtained from quadrupole scans and
290
the simulation results are compared in Fig. 9. The emittance value obtained
291
from the simulation and the experiment are 0.74 mm mrad and 0.70 mm mrad,
292
respectively. In addition, the overall emittance of the beam was reduced owing
293
to a better and more careful setup of the beam. Normalized emittance (mm mrad)
urn al P
re-
284
3.5 3
2.5 2
1.5 1
0.5
0 165
170
175
180
185
190
195
Jo
Solenoid current (A)
Figure 9: Horizontal normalized emittance measured at the BTV 430042 screen compared to simulations.
16
Journal Pre-proof
294
5. Conclusions The electron injector for the AWAKE experiment was successfully con-
296
structed and commissioned. It enabled successful demonstration of the first ever
297
proton beam-driven plasma wakefield experiments. Despite the time constraints
298
of beam delivery for acceleration experiments, systematic beam commissioning
299
measurements were successfully performed with the electron injector. These
300
measurements were compared with intensive beam dynamics modeling using
301
ASTRA. A maximum number of measured input parameters was used to re-
302
produce the experimental results. There was good overall agreement, giving
303
us confidence in our injector model. To improve the agreement even further,
304
two main parameters were identified; a potentially large misalignment in the
305
booster structure and a tendency for a shorter initial bunch length, as originally
306
anticipated. Recent UV pulse measurement confirmed that the actual measured
307
pulse length was shorter than that predicted by the simulations. Correcting the
308
alignment of the beam through the booster structure for each solenoid setting
309
improved the emittance and resulted in a good agreement compared to simu-
310
lations assuming an on-axis beam. The next step for AWAKE is to inject a
311
very short bunch of 200 fs with a matched beta-function and bunch charge to
312
load the wakefield into the plasma. The goal will be to demonstrate emittance
313
preservation and a low energy spread at the end of the plasma acceleration.
314
This challenging task requires a new injector, which is currently under design.
315
The good agreement between the simulations and measurements for the current
316
injector gives us confidence that our setup can be used in future work.
317
6. Acknowledgements
Jo
urn al P
re-
pro
of
295
318
This work was partly supported by the National Research Foundation of Ko-
319
rea (Grants No. NRF-2016R1A5A1013277, No. NRF-2019R1F1A1062377, and
320
NRF-2016-Fostering Core Leaders of the Future Basic Science Program/Global
321
Ph.D. Fellowship Program).
17
Journal Pre-proof
323
References [1] A. Caldwell, et al.,
Path to awake: Evolution of the concept,
of
322
Nu-
clear Instruments and Methods in Physics Research Section A: Accelera-
325
tors, Spectrometers, Detectors and Associated Equipment 829 (2016) 3–16.
326
doi:10.1016/j.nima.2015.12.050.
pro
324
327
[2] E. Gschwendtner, Awake, a particle-driven plasma wakefield acceleration
328
experiment, CERN Yellow Reports 1 (2016) 271. doi:10.5170/CERN-2016-
329
001.271.
331
332
[3] J. M. Dawson, Nonlinear electron oscillations in a cold plasma, Physical
re-
330
Review 113 (1959) 383. doi:10.1103/PhysRev.113.383. [4] P. Muggli, et al.,
Awake readiness for the study of the seeded self-
modulation of a 400 gev proton bunch, Nuclear Instruments and Methods
334
in Physics Research Section A: Accelerators, Spectrometers, Detectors and
335
336
337
338
339
340
341
342
343
Associated Equipment 60 (2017) 014046. doi:10.1088/1361-6587/aa941c.
[5] M. H¨ uther, P. Muggli,
Seeding of the self-modulation in a long pro-
ton bunch by charge cancellation with a short electron bunch, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 909 (2018) 67–70. doi:10.1016/j.nima.2018.02.084.
[6] W. Lu, et al.,
Limits of linear plasma wakefield theory for elec-
tron or positron beams,
Physics of plasmas 12 (2005) 063101.
doi:10.1063/1.1905587.
[7] K. Lotov, Physics of beam self-modulation in plasma wakefield accelerators,
Jo
344
urn al P
333
345
Physics of Plasmas 22 (2015) 103110. doi:10.1063/1.1905587.
346
[8] K. Pepitone, et al., The electron accelerators for the awake experiment at
347
cernbaseline and future developments, Nuclear Instruments and Methods
348
in Physics Research Section A: Accelerators, Spectrometers, Detectors and
349
Associated Equipment 909 (2018) 102–106. doi:10.1016/j.nima.2018.02.044. 18
Journal Pre-proof
[9] J. S. Schmidt, et al., Status of the proton and electron transfer lines for the
351
awake experiment at cern, Nuclear Instruments and Methods in Physics
352
Research Section A: Accelerators, Spectrometers, Detectors and Associated
353
Equipment 829 (2016) 58–62. doi:10.1016/j.nima.2016.01.026.
of
350
[10] M. Turner, et al., External electron injection for the awake experiment, in:
355
2018 IEEE Advanced Accelerator Concepts Workshop (AAC), 2018, pp.
356
1–4. doi:10.1109/AAC.2018.8659402.
358
[11] E. Adli, et al., Acceleration of electrons in the plasma wakefield of a proton bunch, Nature 561 (2018) 7723. doi:10.1038/s41586-018-0485-4.
re-
357
pro
354
[12] M. Turner, et al., Experimental observation of plasma wakefield growth
360
driven by the seeded self-modulation of a proton bunch, Physical review
361
letters 122 (2019) 054801. doi:10.1103/PhysRevLett.122.054801.
urn al P
359
362
[13] E. Adli, et al., Experimental observation of proton bunch modulation in
363
a plasma at varying plasma densities, Physical review letters 122 (2019)
364
365
366
054802. doi:10.1103/PhysRevLett.122.054802.
[14] A. Caldwell, et al., AWAKE design report: a proton-driven plasma wakefield acceleration experiment at CERN, Technical Report, 2013.
367
[15] P. Muggli, The awake proton-driven plasma wakefield experiment at cern,
368
in: 6th Int. Particle Accelerator Conf.(IPAC’15), Richmond, VA, USA,
369
370
May 3-8, 2015, 2015, pp. 2502–2505. doi:10.18429/JACoW-IPAC2015WEPWA007.
[16] O. Mete, et al., Production of long bunch trains with 4.5uc total charge
372
using a photo injector, Physical Review ST Accelerators and Beams 15
Jo
371
373
(2012) 022803. doi:10.1103/PhysRevSTAB.15.022803.
374
[17] O. Mete, et al., Modeling of an electron injector for the awake project, in:
375
6th Int. Particle Accelerator Conf.(IPAC’15), Richmond, VA, USA, May 3-
376
8, 2015, 2015, pp. 1762–1764. doi:10.18429/JACoW-IPAC2015-TUPJE059.
19
Journal Pre-proof
[18] V. Fedosseev, et al., Generation and delivery of an ultraviolet laser beam
378
for the rf-photoinjector of the awake electron beam, in: 10th Int. Partile
379
Accelerator Conf.(IPAC’19), Melbourne, Australia, 19-24 May 2019, 2019,
380
pp. 3709–3712. doi:10.18429/JACoW-IPAC2019-THPGW054.
of
377
[19] S. Liu, et al., Development of the awake stripline bpm electronics, in:
382
International conference on Technology and Instrumentation in Particle
383
Physics, 2017, pp. 237–242.
386
387
388
389
390
A space charge tracking algorithm,
http://www.desy.de/ mpyflo/.
[21] K. Floettmann, Note on the thermal emittance of electrons emitted by Cesium Telluride photo cathodes, Technical Report, 1997. [22] D. H. Dowell, J. F. Schmerge, Quantum efficiency and thermal emittance of metal photocathodes, Physical Review Special Topics-Accelerators and Beams 12 (2009) 074201. doi:10.1103/PhysRevSTAB.12.074201.
391
[23] K. Floettmann, Astra manual version 3.2 (2017).
392
[24] O. Apsimon, B. Williamson,
393
394
395
A numerical approach to designing a
versatile pepper-pot mask for emittance measurement,
Nuclear In-
struments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 943 (2019) 162485. doi:10.1016/j.nima.2019.162485.
Jo
396
1997. URL:
re-
385
[20] K. Floettmann,
urn al P
384
pro
381
20
*Author Contributions Section
Journal Pre-proof Seong-Yeol Kim: Formal analysis, Writing - Original draft preparation, Visualization. Steffen Doebert: Conceptualization, Writing - Original draft preparation, Writing Review & Editing, Supervision, Funding acquisition. Oznur Apsimon: Resources. Robert Apsimon: Resources. Graeme Burt: Resources. Spencer Gessner: Resources, Project administration. Ishkhan Gorgisyan: Resources. Eduardo Granados: Resources, Visualization.
pro
Stefano Mazzoni: Resources.
of
Mohsen Dayyani: Formal analysis.
Joshua Timothy Moody: Resources. Marlene Turner: Resources. Barnaby Williamson: Resources.
Jo
urn al P
re-
Moses Chung: Writing - Review & Editing, Supervision, Funding acquisition.
1
Journal Pre-proof *Declaration of Interest Statement
Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Jo
urn al P
re-
pro
of
☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
PII: DOI: Reference:
S0168-9002(19)31488-3 https://doi.org/10.1016/j.nima.2019.163194 NIMA 163194
To appear in:
Nuclear Inst. and Methods in Physics Research, A
Received date : 1 November 2019 Revised date : 27 November 2019 Accepted date : 28 November 2019 Please cite this article as: S.-Y. Kim, S. Doebert, O. Apsimon et al., Commissioning of the electron injector for the AWAKE experiment, Nuclear Inst. and Methods in Physics Research, A (2019), doi: https://doi.org/10.1016/j.nima.2019.163194. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Journal Pre-proof Revised version of manuscript Click here to view linked References
of
Commissioning of the electron injector for the AWAKE experiment
pro
S.-Y Kima , S. Doebertb , O. Apsimonc , R. Apsimond , G. Burtd,e , M. Dayyanif , S. Gessnerb , I. Gorgisyanb , E. Granadosb , S. Mazzonib , J. T. Moodyg , M. Turnerb , B. Williamsonc,e , M. Chunga a Department
Abstract
re-
of Physics, Ulsan National Institute of Science and Technology, Ulsan 44919, Republic of Korea b CERN, Geneva 1211, Switzerland c Manchester University, M13 9PL Manchester, United Kingdom d Lancaster University, LA1 4YW Lancaster, United Kingdom e Cockcroft Institute, WA4 4AD Warrington, United Kingdom f Institute for Research in Fundamental Sciences, 19395-5531, Tehran, Iran g Max Planck Institute for Physics, 80805 Munich, Germany
urn al P
The advanced wakefield experiment (AWAKE) at CERN is the first proton beam-driven plasma wakefield acceleration experiment. The main goal of AWAKE RUN 1 was to demonstrate seeded self-modulation (SSM) of the proton beam and electron witness beam acceleration in the plasma wakefield. For the AWAKE experiment, a 10-meter-long Rubidium-vapor cell together with a high-power laser for ionization was used to generate the plasma. The plasma wakefield is driven by a 400 GeV/c proton beam extracted from the super proton synchrotron (SPS), which undergoes a seeded self-modulation process in the plasma. The electron witness beam used to probe the wakefields is generated from an S-band RF photo-cathode gun and then accelerated by a booster structure up to energies between 16 and 20 MeV. The first run of the AWAKE experiment revealed that the maximum energy gain after the plasma cell is 2 GeV, and the
Jo
SSM mechanism of the proton beam was verified. In this paper, we will present the details of the AWAKE electron injector. A comparison of the measured electron beam parameters, such as beam size, energy, and normalized emittance, Email addresses: [email protected] (S. Doebert), [email protected] (M. Chung)
Preprint submitted to Elsevier
November 27, 2019
Journal Pre-proof
with the simulation results was performed.
1. Introduction to AWAKE
pro
1
of
Keywords: Plasma, Wakefield, AWAKE, Electron injector
The advanced wakefield experiment (AWAKE) at CERN studies electron
3
beam acceleration in proton beam-driven plasma wakefields [1, 2]. The proton
4
beam driver has a momentum of 400 GeV/c, a bunch length of 6-12 cm (RMS),
5
and a transverse beam size of 0.2 mm (RMS). The plasma is generated in a
6
10-meter-long heated Rubidium-vapor source that can reach plasma densities
7
of the order of 1 × 1014 ∼ 1 × 1015 /cm3 . The ionization laser has an energy
8
of 450 mJ and a pulse length of 120 fs (FWHM), allowing full ionization of a
9
plasma channel with a radius of 1 mm. The laser pulse and the proton beam
10
co-propagate through the vapor source to create the plasma, and seed the self-
11
modulation within the long proton bunch. The maximum accelerating gradient
urn al P
re-
2
14
of the plasma wakefield [3] Emax = me wpe c/e is determined by the plasma p electron density ne , where me is the electron mass, wpe = ne e2 /me 0 is the
15
is the electron charge. In the case of AWAKE, the nominal plasma density is
16
7 × 1014 /cm3 , and the maximum accelerating gradient is therefore expected to
17
be on the order of GV/m.
12
13
plasma frequency, 0 is the vacuum permittivity, c is the speed of light, and e
18
An important condition particular to AWAKE for the generation of a high
19
plasma wakefields is the seeded self-modulation (SSM) of the proton beam [4,
20
21
wpe /c, and σz is the RMS bunch length of the drive beam [6]. Therefore, the
Jo
22
5]. In the linear plasma wakefield theory, the optimal condition to drive a √ plasma wakefields is kp σz ∼ = 2, where kp is the plasma wavenumber defined by
23
initial proton beam extracted from the super proton synchrotron (SPS) does
24
not satisfy the condition needed to generate high-gradient wakefields owing to
25
its long bunch length. However, once the long proton bunch has propagated
26
for a distance in the plasma, it splits into micro-bunches that have a length
27
corresponding to the plasma wavelength and meet the condition for driving the
2
Journal Pre-proof
high-gradient plasma wakefield [7]. The self-modulation of the proton beam is
29
reliably seeded by the front of the ionization laser, which is placed in the center
30
of the proton beam. The self-modulation of the proton beam density starts from
31
this point, enabling phase stable wakefields.
of
28
In the first run of the AWAKE experiment to demonstrate electron beam ac-
33
celeration through the plasma wakefields, an electron injector that consists of an
34
S-band RF photocathode gun and a traveling wave booster structure were used
35
to produce electron beams with an energy of 16-20 MeV [8]. The electron beam
36
was then passed through an electron transfer line, and injected into the plasma
37
cell co-propagating with the proton beam and the ionization laser [9, 10]. An
38
energy gain of the electron beam in the 10 m plasma source of up to 2 GeV [11]
39
has been demonstrated experimentally. In addition, the SSM phenomenon has
40
been successfully observed and studied by direct measurement of the modulated
41
proton bunch using streak cameras, and by indirect measurement of the proton
42
beam divergence due to the transverse wakefield [12, 13]. A schematic of the
43
AWAKE experiment can be seen in Fig. 1 and main parameters of the proton,
44
electron beam and ionization laser can be seen in Table 1.
urn al P
re-
pro
32
Details of the electron beam parameters required for electron beam acceler-
46
ation in the plasma wakefield are summarized in Refs. [14, 15]. Since the main
47
goal of the AWAKE RUN 1 experiment was to determine whether the electron
48
beam is accelerated in the proton beam-driven plasma wakefield, the focus was
49
on the matching of the electron beam to the plasma channel containing the
50
wakefields. The biggsest constrain was the transverse RMS beam size at the
51
entrance of the plasma cell and therefore the emittance of the electron beam. In
52
order to control the electron beam size and the emittance along the transfer line,
53
and to satisfy the parameters required at the plasma entrance, understanding of
54
the electron injector was essential. Therefore, the main parameters of the elec-
55
tron beam along the electron injector have been measured and compared with
56
the simulations to determine the operational conditions. In this paper, we will
57
present the details of the electron injector. A description of the electron injector
58
and commissioning results such as transverse beam distributions on screens and
Jo
45
3
of
Journal Pre-proof
Table 1: Main parameters of the proton, electron beam and ionization laser for the AWAKE
pro
experiment.
Parameters
Value
Proton beam (extracted from SPS)
Bunch length σz Bunch radius σr
400 GeV
re-
Energy
12 cm
0.2 mm 3.5 mm mrad
Energy spread σδ
0.45 %
urn al P
Normalized emittance n
Electron beam
Initial set value
(at the plasma entrance)
Value used for the experiment
Energy
16 MeV
16 - 20 MeV
Charge
200 pC
140 - 650 pC
Bunch length σz
4 ps
1 - 4 ps
Bunch radius σr
0.25 mm
0.25 mm
Normalized emittance n
2 mm mrad
2 mm mrad
Energy spread σδ
0.5 %
0.5 %
Ionization laser (Fiber Ti:Sa)
450 mJ
Pulse length σz
100 - 120 fs
Focused spot size σr
1 mm
Jo
Energy
4
re-
pro
of
Journal Pre-proof
Figure 1: Schematic of the AWAKE experiment at CERN. The upper right corner shows an
urn al P
example of a beam spot as observed in the spectrometer. The inserts in the lower left corner illustrate the injection conditions and the process of seeded self-modulation.
59
beam emittance measurements using pepper-pot and quadrupole scan methods
60
will be shown in Sections 2 and 3. A careful comparison of the measurements
61
with the simulation results in order to characterize the operating condition of
62
the electron injector will be presented in Section 4. Finally, the results of the
63
commissioning will be concluded in Section 5.
64
2. Design and construction of the injector An electron injector providing adequate beam quality to demonstrate the
66
first proton driven plasma wakefield acceleration experiment was designed and
Jo
65
67
constructed by AWAKE collaborators. The original target parameters for the
68
injector were energy of 16 MeV, a bunch charge of 200 pC, an energy spread
69
below 1%, a bunch length of 4 ps, and an emittance of 2 mm mrad. To achieve
70
these parameters, an RF photo-injector was chosen using a 3 GHz, 2.5-cell
71
RF-gun and a one-meter-long booster structure. The RF-gun including a load-
5
Journal Pre-proof
lock system was available at CERN [16], while the booster structure was newly
73
developed for this purpose [17]. The load-lock system enables the use of Cs2 Te
74
cathodes fabricated at CERN with a quantum efficiency of Qe ∼ 10−2 . The laser
75
beam was derived from the main Ti-Sa laser used to ionize and seed the proton
76
beam self-modulation. Consequently, an adequate timing synchronization could
77
be achieved. A small fraction of the initial laser power was frequency tripled to
78
a wavelength of 262 nm and sent with an off-axis mirror on the cathode. The
79
UV-beamline allows users to vary the spot size and energy on the cathode as well
80
as the bunch length as the IR-laser beam has to be compressed before the UV
81
conversion [18]. The RF-gun accelerates the electron bunches to an energy of
82
5.5 MeV, and then the booster can add a maximum energy of 16 MeV. A single
83
klystron is used to power the gun and the booster structure. A high power wave-
84
guide attenuator and phase-shifter allow for individual phasing and powering of
85
the two structures. A particular challenge for the design and construction was
86
the severe space constraints in the AWAKE experiment. The complete injector
87
has a total length of only 5 m before the beam transport towards the plasma
88
cell starts. The injector was equipped with a number of beam diagnostics as
89
described in more detail in the next section.
90
3. Beam Commissioning results
Jo
urn al P
re-
pro
of
72
Figure 2: Schematic electron injector layout emphasizing the location of magnets and diagnostics mentioned throughout the paper.
91
Figure 2 shows a schematic view of the electron injector indicating, in par-
92
ticular, the locations of focusing and beam diagnostic elements used in this 6
Journal Pre-proof
section. The electron injector consists of an S-band RF photocathode gun and
94
booster structure. The main solenoid and the corresponding bucking coil are
95
placed around the RF gun to control the beam focusing and emittance compen-
96
sation. Corrector magnets are used to correct the beam trajectory, which can
97
be measured with several strip-line BPM’s [19]. A pepper-pot beam diagnostic
98
instrument is used to measure the beam emittance out of the RF-gun, and the
99
screen behind the retractable pepper-pot also allows imaging of the transverse
100
distribution of the beam. A quadrupole triplet behind the booster structure
101
is used to match the beam for further beam transport and for the emittance
102
measurements using the quadrupole scan method at higher energies. A YAG
103
screen (BTV430042) is used to measure the transverse beam distribution. The
104
electron beam charge at the end of the injector is measured by a Faraday-cup
105
with high precision. The first dipole magnet of the following transport line to-
106
gether with a second screen also serves as a spectrometer to measure the beam
107
energy and energy spread. During the commissioning, as many beam parame-
108
ters as possible were measured so that they can be used as input parameters for
109
simulations.
urn al P
re-
pro
of
93
First, the initial transverse distribution of laser pulse was measured using
111
the so-called virtual cathode camera. This camera images a fraction of the
112
laser beam in similar imaging conditions as the laser pulse sent to the cathode.
113
The laser energy can be varied with optical density (OD) filters to adjust the
114
final beam charge. The beam charge during the commissioning reported here
115
was measured as 140 ± 10 pC. The initial transverse distribution of the laser is
116
shown in Fig. 3. Figure 3 (a) shows the case without the OD filter, which is used
117
to enhance the UV imaging. The RMS transverse beam size can be determined
118
as σx = 0.33 mm in the horizontal plane, and σy = 0.34 mm in the vertical
119
plane. In the experiment, however, an OD filter was used to reduce the beam
120
charge, which resulted unfortunately in a poorer quality image of the virtual
121
cathode as can be seen in Fig. 3 (b). The transverse size of the laser obtained
122
by analyzing this image appears slightly reduced owing to the cut-off of the laser
123
energy, and we found that σx and σy are 0.27 mm. The smaller values were used
Jo
110
7
Journal Pre-proof
for the particle tracking simulations later on. The bunch length of the electron
125
beam was not measured during the experiment. Previous measurements of the
126
UV laser beam suggested a laser pulse length of σz = 2.2 ps (FWHM 5.2 ps). Experimental data Fit curve
1.5
0.5 200
0 -0.5
1
300
100
20
Experimental data Fit curve
15
0.5
0
-0.5
10 5
-1
-1
-1.5
0
-1.5 -1 -0.5 0 0.5 1
x (mm)
0
-1.5 -1 -0.5 0 0.5 1
re-
-1.5
(b)
pro
y (mm)
1
(a)
y (mm)
1.5
of
124
x (mm)
Figure 3: Initial transverse distribution of the laser at the cathode without an OD filter (a)
urn al P
and with OD filter (b).
127
The energy of the electron beam out of the RF gun was calculated by mea-
128
suring the position of the beam at the pepper-pot screen while scanning the
129
first corrector magnet, as no actual spectrometer is available after the gun.
130
The electron beam momentum out of the RF-gun was determined to be 5.8
131
MeV/c, which corresponds to a nominal energy of 5.3 MeV. However, as this
132
measurement has a much lower precision compared to that of a real spectrom-
133
eter magnet, we estimated the error of the electron beam energy calculation
134
to be about 10%. The energy of the electron beam after the traveling wave
135
structure was measured using the first dipole magnet in a proper spectrometer
136
setup. The momentum of the electron beam after acceleration was determined
137
as 18.5 MeV/c with an error of 0.3%.
To optimize the operating conditions of the electron injector during com-
139
missioning, scans of the main solenoid around the RF-gun responsible for emit-
140
tance compensation have been performed. Beam sizes and emittances have been
141
measured before and after the booster structure as a function of the solenoid
142
current. First, the transverse beam size of the electron beam was measured
143
on the pepper-pot screen. Figure 4 (a) depicts the beam size and normalized
Jo
138
8
Journal Pre-proof
emittance as a function of the solenoid current. The minimum beam sizes σx
145
and σy at the pepper-pot screen are 0.47 mm and 0.38 mm for a solenoid cur-
146
rent of 185 A. We found during the commissioning that it was very difficult to
147
obtain reliable and reproducible emittance measurements with our pepper-pot
148
setup. The cause of this observation is still under investigation. At this point,
149
we tend not to rely on these measurements. We will come back to that issue in
150
the simulation section.
1
5
0.5
170
180
190
200
0
urn al P
0
2.5 2
Solenoid current (A)
7
(b)
6 5
1.5
4
1
3 2
0.5
0 165
170
175
1 180
185
190
0 195
Normalized emittance (mm mrad)
10
1.5
pro
2
15
(a)
re-
rms beam size (mm)
2.5
Normalized emittance (mm mrad) rms beam size (mm)
of
144
Solenoid current (A)
Figure 4: Measured transverse beam size and normalized emittance of the electron beam at the pepper-pot (a) and BTV 430042 screen (b).
In addition to the measurement at the pepper pot screen, the beam size
152
and normalized emittance after the traveling wave structure have been mea-
153
sured. This time a well-established quadrupole scanning method was used as
154
the beam is no longer space charge dominated. The beam size was measured at
155
the BTV430042 screen, while the normalized emittance was determined from
156
a quadrupole scan by varying the magnetic field of the center quadrupole of
157
the triplet (MQAWA430034). Figure 4 (b) shows the beam size at the screen
158
(BTV430042) and the normalized emittance as a function of the solenoid cur-
Jo
151
159
rent. The minimum beam size was found at a solenoid current of 175 A: σx is
160
0.29 mm and σy is 0.30 mm. The minimum normalized emittance, on the other
161
hand, was found at a solenoid current of 185 A, which is 1.0 mm mrad in both
162
the horizontal and the vertical plane. Relative error of the emittance in Fig. 4
163
(b) was determined as follows: When using the quadrupole scan method, beam 9
Journal Pre-proof
size was measured ten times at each quadrupole magnet strength. Then, the
165
polynomial curve fitting was performed to find ten sets of emittance values. The
166
relative error of the emittance in the range of the solenoid current was about
167
10%.
In the next section, we will show a comparative analysis of experimental
pro
168
of
164
169
measurements with the results obtained from the simulations.
170
4. Comparison with simulations
To compare measured beam parameters with the simulation results, particle
172
tracking was performed mainly using the ASTRA [20] code. The cathode mate-
173
rial used in AWAKE is Cs2 Te, and the characteristic parameters of the electron
174
beam emitted from the cathode such as the thermal emittance and momentum
175
distribution can be defined in terms of the Fermi-Dirac distribution [21, 22].
176
In ASTRA, the momentum distribution, and thermal emittance of the initial
177
beam at the cathode are defined as follows [23]:
urn al P
re-
171
σ px = σ py =
σE =
r
Ephoton − φef f , 3me c2
Ephoton − φef f √ , 3 2
x,y = σx,y
r
Ephoton − φef f , 3me c2
(1)
(2)
(3)
where σpx,y is the RMS value of the transverse momentum, Ephoton is the photon
179
energy of the laser, φef f is the effective work function of the cathode material,
180
σE is the energy spread, and x,y is the thermal emittance.
Jo
178
181
Through the equations above and Ref. [22], the transverse momentum and
182
energy spread are determined only by the photon energy of the laser and the
183
work function of the cathode material, regardless of the spatial distribution type
184
of the laser. As the momentum distribution is determined by Ephoton and φef f
185
irrespective of the laser shape, the initial beam distribution measured at the
10
Journal Pre-proof
cathode (see previous section) was used as an input to the simulation. As can
187
be seen in Fig. 3 (b), the distribution is not fully symmetric. To perform the
188
simulations, we converted the measured image into a transverse input distribu-
189
tion suitable for simulations. For the longitudinal components, a distribution
190
generated by ASTRA was used assuming a Gaussian laser pulse shape. For the
191
simulations, as we do not have a real bunch length measurement, we studied
192
three values: 2.2 ps as indicated by previous UV laser measurements and shorter
193
beams with 1.0 ps and 1.5 ps bunch lengths for comparison.
pro
of
186
The RF fields of the RF gun were overlaid with the magnetic fields of the
195
solenoids. The maximum accelerating gradient was set to be 79.6 MV/m to
196
obtain the measured beam energy of 5.3 MeV out of the RF gun. At 1.6 m,
197
a traveling wave structure (booster structure) was placed to increase the beam
198
energy to match the measured value of 18.1 MeV; thus its accelerating gradient
199
was 17.8 MV/m. In the ASTRA simulation, a full 3D field map of the booster
200
structure obtained from CST was used, while the field map of the RF gun
201
was determined by the 1D axial electric field owing to a lack of 3D data. The
202
phase of the gun was checked to achieve the maximum energy gain, nominally
203
corresponding to the smallest emittance as well. Therefore, the beam launch
204
phase was approximately 30◦ off crest in the simulations. In the experiment, we
205
tried to minimize the emittance by varying the phase of the gun and solenoid
206
current around the nominal working point. The booster structure was set to
207
on-crest acceleration as optimized for maximum energy in the experiment.
urn al P
re-
194
Solenoid scans were simulated with bunch lengths of 1.0 ps, 1.5 ps, and
209
2.2 ps. The simulation results together with the measured data are shown in
210
Fig. 5, upper plots for the pepper-pot screen and the lower plots for the BTV
211
430042 screen. In this case, misalignment of the RF structure and solenoid
212
magnet were not considered. One can see that the slopes of the beam size scans
213
match better with shorter bunch length cases on both screens. This results
214
indicate that the bunch length might have been shorter than 2.2 ps as indicated
215
by previous UV measurements. At the pepper-pot screen, the minimum beam
216
sizes σx and σy for the case of 1.0 ps bunch length are 0.32 mm and 0.31
Jo
208
11
2 1.5 1 0.5 0 160
170
180
190
200
210
Solenoid current (A)
10
5
0 160
170
180
190
200
2 1.5 1 0.5
(c)
170
175
180
185
210
7 6 5
re-
Normalized emittance (mm mrad)
2.5
urn al P
rms beam size (mm)
(b)
Solenoid current (A)
3
0 165
15
of
rms beam size (mm)
(a)
pro
2.5
Normalized emittance (mm mrad)
Journal Pre-proof
190
195
Solenoid current (A)
4 3 2
(d)
1
0 165
170
175
180
185
190
195
Solenoid current (A)
Figure 5: Transverse beam size and normalized emittance obtained from the ASTRA simulation at the pepper-pot screen (a, b) and BTV430042 screen (c, d) without misalignement.
mm; furthermore, the minimum values of the normalized emittance are 0.80
218
mm mrad in the horizontal plane and 0.78 mm mrad in the vertical plane. In
219
addition, the minimum beam sizes σx and σy at the BTV430042 screen are
220
0.48 mm and 0.44 mm, and the minimum values of the normalized emittance
221
are very similar to the values at the pepper-pot screen. However, the shape
222
of the scan for the emittance measurements does not correspond well with the
223
simulations. The emittance determination using the pepper-pot data seems to
Jo
217
224
fail because the larger measured values do not agree with the simulations, while
225
the measurements further downstream of the beam line using the quadrupole
226
scan method seem to match well with the simulation results, at least for the
227
minimum values. Therefore, at this point we concluded that the pepper-pot
228
results are not very reliable. The reason for this is still under investigation, and
12
Journal Pre-proof
there are hints that the location and dimensions of our device are not optimal
230
[24].
of
229
During the experiments, the machine operation and in particular the beam
232
alignment were optimized for the minimum emittance values at 185 A, as these
233
beams have been used for the plasma wakefield acceleration experiments. The
234
alignment of the beam was not corrected during the scan. Therefore, we suspect
235
misalignment of the beam due to steering of the solenoid magnet is responsi-
236
ble for the particular shape of the emittance measurements. Particle tracking
237
simulations were used again to study possible offsets of the booster structure
238
and the solenoids, as we had some experimental hints from the beam trajectory
239
data that this might have been the case.
2
urn al P
rms beam size (mm)
2.5
Normalized emittance (mm mrad)
re-
pro
231
1.5 1
0.5
(a)
0 165
170
175
180
185
190
195
Solenoid current (A)
6 5 4 3 2 1
(b) 0 165
170
175
180
185
190
195
Solenoid current (A)
Figure 6: Comparison of the transverse beam size (a) and normalized emittance (b) at BTV430042 screen taking into account a misalignment of the booster structure.
240
Various offsets of the main solenoid and the booster structure have been
241
studied using ASTRA with 3D space charge calculations and a 3D field map
242
for the booster structure. We found that the measurements are best described assuming a relatively big offset of the booster structure only. Figure 6 shows
244
a comparison of the experiment data and the simulation results obtained for a
245
-4.5 mm offset at the entrance of the booster structure. In this simulation, a
246
1.5 ps bunch length was used. Both the emittance and the spot size show good
247
qualitative agreement for these misalignment assumptions.
Jo 243
13
urn al P
re-
pro
of
Journal Pre-proof
Figure 7: Horizontal phase space of the beam at the input coupler section. Top: on-axis simulation; bottom: off-axis simulation with -4.5 mm booster structure offset. Solenoid current: (a, c) 175 A, and (b, d) 185 A.
The emittance growth seen in the simulations is attributed to an RF kick
249
within the input coupler of the booster structure due to its not perfectly compen-
250
sated RF fields. Phase space distortion due to the asymmetric field is illustrated
251
in Fig. 7, where the top row shows the horizontal phase space at the input cou-
252
pler of the booster structure obtained from an on-axis simulation, while the
Jo
248
253
bottom row images are with the booster structure offset. One can see that the
254
momentum distribution of the on-axis case is concentrated in the center. In the
255
case with a large beam offset, simulation results reveal that the momentum dis-
256
tribution is distorted with respect to the core, which leads to emittance growth.
257
However, for well-focused beams at the entrance of the booster (solenoid cur14
Journal Pre-proof
rents of 180 and 185 A), the effect is much smaller, and the emittance growth
259
is not significant.
of
258
In a second beam commissioning campaign, the misalignment of the beam in
261
the booster structure and the question of the not well-known bunch length were
262
addressed. The original misalignment of the solenoids could not be solved, so
263
we corrected the beam at the entrance of the booster structure for each solenoid
264
setting using a pair of corrector magnets when taking data. In addition, as
265
the UV pulse shape and length were not clearly determined during the last
266
measurement, UV pulse measurements using a streak camera were performed
267
again. Unfortunately, it was not possible to obtain reliable data from direct
268
bunch length measurements using light from an OTR screen on the streak cam-
269
era. The intensity arriving at the streak camera was simply not high enough.
re-
pro
260
Amplitude (a.u.)
urn al P
1
(b)
0.8 0.6 0.4 0.2 0 -15
-10
-5
0
5
10
15
Time (ps)
Figure 8: Initial transverse beam distribution at the cathode (a) and longitudinal profile of the UV laser pulse (b). Intensity of the UV laser is normalized.
The transverse beam distribution and charge of the beam were slightly dif-
271
ferent from those in the first campaign. Therefore, the following inputs were
Jo
270
272
used for the simulations. Figure 8 shows the measured transverse beam dis-
273
tribution at the cathode and the measured longitudinal profile of the UV laser
274
pulse. The initial beam size is slightly changed compared to that in Fig. 3 (b).
275
Beam sizes σx and σy are both 0.25 mm. The measured beam charge at the
276
Faraday cup was 150 pC. In the case of the UV measurement, it was confirmed
15
Journal Pre-proof
that the RMS UV pulse length was 1.5 ps. In addition, one can see that there is
278
a secondary pulse behind the main pulse, but it is negligible, as the amplitude
279
is very small compared to the main pulse. This after-pulse was not taken into
280
account in the simulations. Even though the simulation results with a 1.5 ps
281
bunch length did not perfectly match with the experimental data indicated in
282
Fig. 5, it was confirmed that the measured bunch length is shorter than that
283
assumed previously.
pro
of
277
Using the new measured inputs, the ASTRA simulations were performed
285
again. The booster structure gradient was slightly changed to obtain the exper-
286
imentally measured 18.8 MeV/c momentum value in the simulation. This time,
287
we obtained a good agreement between simulations and measurements, con-
288
firming the assumptions concluded from the analysis of the first measurement
289
campaign. The measured emittance values obtained from quadrupole scans and
290
the simulation results are compared in Fig. 9. The emittance value obtained
291
from the simulation and the experiment are 0.74 mm mrad and 0.70 mm mrad,
292
respectively. In addition, the overall emittance of the beam was reduced owing
293
to a better and more careful setup of the beam. Normalized emittance (mm mrad)
urn al P
re-
284
3.5 3
2.5 2
1.5 1
0.5
0 165
170
175
180
185
190
195
Jo
Solenoid current (A)
Figure 9: Horizontal normalized emittance measured at the BTV 430042 screen compared to simulations.
16
Journal Pre-proof
294
5. Conclusions The electron injector for the AWAKE experiment was successfully con-
296
structed and commissioned. It enabled successful demonstration of the first ever
297
proton beam-driven plasma wakefield experiments. Despite the time constraints
298
of beam delivery for acceleration experiments, systematic beam commissioning
299
measurements were successfully performed with the electron injector. These
300
measurements were compared with intensive beam dynamics modeling using
301
ASTRA. A maximum number of measured input parameters was used to re-
302
produce the experimental results. There was good overall agreement, giving
303
us confidence in our injector model. To improve the agreement even further,
304
two main parameters were identified; a potentially large misalignment in the
305
booster structure and a tendency for a shorter initial bunch length, as originally
306
anticipated. Recent UV pulse measurement confirmed that the actual measured
307
pulse length was shorter than that predicted by the simulations. Correcting the
308
alignment of the beam through the booster structure for each solenoid setting
309
improved the emittance and resulted in a good agreement compared to simu-
310
lations assuming an on-axis beam. The next step for AWAKE is to inject a
311
very short bunch of 200 fs with a matched beta-function and bunch charge to
312
load the wakefield into the plasma. The goal will be to demonstrate emittance
313
preservation and a low energy spread at the end of the plasma acceleration.
314
This challenging task requires a new injector, which is currently under design.
315
The good agreement between the simulations and measurements for the current
316
injector gives us confidence that our setup can be used in future work.
317
6. Acknowledgements
Jo
urn al P
re-
pro
of
295
318
This work was partly supported by the National Research Foundation of Ko-
319
rea (Grants No. NRF-2016R1A5A1013277, No. NRF-2019R1F1A1062377, and
320
NRF-2016-Fostering Core Leaders of the Future Basic Science Program/Global
321
Ph.D. Fellowship Program).
17
Journal Pre-proof
323
References [1] A. Caldwell, et al.,
Path to awake: Evolution of the concept,
of
322
Nu-
clear Instruments and Methods in Physics Research Section A: Accelera-
325
tors, Spectrometers, Detectors and Associated Equipment 829 (2016) 3–16.
326
doi:10.1016/j.nima.2015.12.050.
pro
324
327
[2] E. Gschwendtner, Awake, a particle-driven plasma wakefield acceleration
328
experiment, CERN Yellow Reports 1 (2016) 271. doi:10.5170/CERN-2016-
329
001.271.
331
332
[3] J. M. Dawson, Nonlinear electron oscillations in a cold plasma, Physical
re-
330
Review 113 (1959) 383. doi:10.1103/PhysRev.113.383. [4] P. Muggli, et al.,
Awake readiness for the study of the seeded self-
modulation of a 400 gev proton bunch, Nuclear Instruments and Methods
334
in Physics Research Section A: Accelerators, Spectrometers, Detectors and
335
336
337
338
339
340
341
342
343
Associated Equipment 60 (2017) 014046. doi:10.1088/1361-6587/aa941c.
[5] M. H¨ uther, P. Muggli,
Seeding of the self-modulation in a long pro-
ton bunch by charge cancellation with a short electron bunch, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 909 (2018) 67–70. doi:10.1016/j.nima.2018.02.084.
[6] W. Lu, et al.,
Limits of linear plasma wakefield theory for elec-
tron or positron beams,
Physics of plasmas 12 (2005) 063101.
doi:10.1063/1.1905587.
[7] K. Lotov, Physics of beam self-modulation in plasma wakefield accelerators,
Jo
344
urn al P
333
345
Physics of Plasmas 22 (2015) 103110. doi:10.1063/1.1905587.
346
[8] K. Pepitone, et al., The electron accelerators for the awake experiment at
347
cernbaseline and future developments, Nuclear Instruments and Methods
348
in Physics Research Section A: Accelerators, Spectrometers, Detectors and
349
Associated Equipment 909 (2018) 102–106. doi:10.1016/j.nima.2018.02.044. 18
Journal Pre-proof
[9] J. S. Schmidt, et al., Status of the proton and electron transfer lines for the
351
awake experiment at cern, Nuclear Instruments and Methods in Physics
352
Research Section A: Accelerators, Spectrometers, Detectors and Associated
353
Equipment 829 (2016) 58–62. doi:10.1016/j.nima.2016.01.026.
of
350
[10] M. Turner, et al., External electron injection for the awake experiment, in:
355
2018 IEEE Advanced Accelerator Concepts Workshop (AAC), 2018, pp.
356
1–4. doi:10.1109/AAC.2018.8659402.
358
[11] E. Adli, et al., Acceleration of electrons in the plasma wakefield of a proton bunch, Nature 561 (2018) 7723. doi:10.1038/s41586-018-0485-4.
re-
357
pro
354
[12] M. Turner, et al., Experimental observation of plasma wakefield growth
360
driven by the seeded self-modulation of a proton bunch, Physical review
361
letters 122 (2019) 054801. doi:10.1103/PhysRevLett.122.054801.
urn al P
359
362
[13] E. Adli, et al., Experimental observation of proton bunch modulation in
363
a plasma at varying plasma densities, Physical review letters 122 (2019)
364
365
366
054802. doi:10.1103/PhysRevLett.122.054802.
[14] A. Caldwell, et al., AWAKE design report: a proton-driven plasma wakefield acceleration experiment at CERN, Technical Report, 2013.
367
[15] P. Muggli, The awake proton-driven plasma wakefield experiment at cern,
368
in: 6th Int. Particle Accelerator Conf.(IPAC’15), Richmond, VA, USA,
369
370
May 3-8, 2015, 2015, pp. 2502–2505. doi:10.18429/JACoW-IPAC2015WEPWA007.
[16] O. Mete, et al., Production of long bunch trains with 4.5uc total charge
372
using a photo injector, Physical Review ST Accelerators and Beams 15
Jo
371
373
(2012) 022803. doi:10.1103/PhysRevSTAB.15.022803.
374
[17] O. Mete, et al., Modeling of an electron injector for the awake project, in:
375
6th Int. Particle Accelerator Conf.(IPAC’15), Richmond, VA, USA, May 3-
376
8, 2015, 2015, pp. 1762–1764. doi:10.18429/JACoW-IPAC2015-TUPJE059.
19
Journal Pre-proof
[18] V. Fedosseev, et al., Generation and delivery of an ultraviolet laser beam
378
for the rf-photoinjector of the awake electron beam, in: 10th Int. Partile
379
Accelerator Conf.(IPAC’19), Melbourne, Australia, 19-24 May 2019, 2019,
380
pp. 3709–3712. doi:10.18429/JACoW-IPAC2019-THPGW054.
of
377
[19] S. Liu, et al., Development of the awake stripline bpm electronics, in:
382
International conference on Technology and Instrumentation in Particle
383
Physics, 2017, pp. 237–242.
386
387
388
389
390
A space charge tracking algorithm,
http://www.desy.de/ mpyflo/.
[21] K. Floettmann, Note on the thermal emittance of electrons emitted by Cesium Telluride photo cathodes, Technical Report, 1997. [22] D. H. Dowell, J. F. Schmerge, Quantum efficiency and thermal emittance of metal photocathodes, Physical Review Special Topics-Accelerators and Beams 12 (2009) 074201. doi:10.1103/PhysRevSTAB.12.074201.
391
[23] K. Floettmann, Astra manual version 3.2 (2017).
392
[24] O. Apsimon, B. Williamson,
393
394
395
A numerical approach to designing a
versatile pepper-pot mask for emittance measurement,
Nuclear In-
struments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 943 (2019) 162485. doi:10.1016/j.nima.2019.162485.
Jo
396
1997. URL:
re-
385
[20] K. Floettmann,
urn al P
384
pro
381
20
*Author Contributions Section
Journal Pre-proof Seong-Yeol Kim: Formal analysis, Writing - Original draft preparation, Visualization. Steffen Doebert: Conceptualization, Writing - Original draft preparation, Writing Review & Editing, Supervision, Funding acquisition. Oznur Apsimon: Resources. Robert Apsimon: Resources. Graeme Burt: Resources. Spencer Gessner: Resources, Project administration. Ishkhan Gorgisyan: Resources. Eduardo Granados: Resources, Visualization.
pro
Stefano Mazzoni: Resources.
of
Mohsen Dayyani: Formal analysis.
Joshua Timothy Moody: Resources. Marlene Turner: Resources. Barnaby Williamson: Resources.
Jo
urn al P
re-
Moses Chung: Writing - Review & Editing, Supervision, Funding acquisition.
1
Journal Pre-proof *Declaration of Interest Statement
Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Jo
urn al P
re-
pro
of
☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: