当前位置:首页 >> >>

MEASUREMENTS OF PLASMA WAKE-FIELDS IN THE BLOW-OUT REGIME


MEASUREMENTS OF PLASMA WAKE-FIELDS IN THE BLOW-OUT REGIME
N.Barov, M.Conde and J.B.Rosenzweig Dept. of Physics and Astronomy, UCLA, Los Angeles, CA 90024 P.Schoessow, G.Cox, W.Gai, R.Konecny, J.Power, J.Simpson Argonne National Laboratory, 9700 S.Cass Ave., Argonne IL 60439
Initial results from nonlinear plasma wake-field experiments at the Argonne Wake-field Accelerator (AWA) test facility are reported. This nonlinear "blow-out" regime is characterized by the complete ejection of the plasma electrons from the beam channel. The wake-fields in this case are of notably high quality for acceleration of electrons, as the acceleration is independent of transverse position, and the focusing is linear and independent of longitudinal position within the electron depleted region, allowing self-consistent guiding of the majority of the driving electron beam. Initial measurements of the energy gain in a witness beam indicate a positive shift in its energy distribution of at least 0.5 MeV. single-cell L-band rf cavity, using a copper photocathode to generate the electron pulse, followed by two 1 meter linac sections. The design of this facility is optimized for high charge (up to 100 nC per pulse), and short bunch length. In the future, an additional electron gun will be added to the AWA to produce a 5 MeV witness beam for probing the wake fields in dielectric structures and plasmas. For the initial measurements, we realized that it was feasible to generate a stably propagating witness beam in the same cavity as the drive beam by delaying a portion (approximately 20%) of the laser pulse. We found that delays as long as 70 picoseconds were possible with laser path differences of about 3 cm. Figure 1 shows a resultant streak camera trace. This data is the result of inserting a 1 mm thick fused silica Cerenkov radiator in the path of the beam (inclined 11 degrees to the normal) and directing the light to a Hamamatsu C1587 temporal disperser. Since the delay time is dependent on the amount of RF compression, and hence on the injection phase of the bunches into the linac cavity, the streak camera was necessary to calibrate this delay time for a specific set of conditions.
1

I. INTRODUCTION
Presently, there is a great interest in experimental evidence of acceleration using the nonlinear plasma wakefield acceleration (NPWFA) concept [1]. In addition to our work, a group at KEK [2] has also been measuring this effect. The NPWFA an electron drive beam propagates in a plasma whose initial electron density is less than the beam's density (an underdense plasma). When such a beam is short (the rms bunch length, σz , is less than 2 collisionless skin depths, kp-1 ), it is well-suited to coupling power into a plasma wave in which the plasma electrons' motion is mostly radial. Because of the underdense condition, all of the plasma electrons are pushed out of the drive beam's way, resulting in the extremely strong focusing associated with an ion channel, which guides the drive beam and also the accelerated (witness) beam. This plasma electron depleted region assures that the focusing is linear and the acceleration field independent of radius. This accelerator scheme is exciting for the high acceleration gradient predicted: a 1.5 GeV/m is possible for a plasma density, n0 = 1 ×1014 cm ?3 and a 100 nC beam. The characteristic wavelength of the acceleration field is somewhat longer than 3.3 mm due to relativistic effects, and is suitable for accelerating short electron bunches with collider applications in mind. The NPWFA concept, however, requires care when considering multiple acceleration sections to increase the final energy. Krall and Joyce [3] are currently studying the electron hose instability in this context. This work suggests that very long, ramped drive beams are not desirable.

0.8

Current (arb. units)

0.6

0.4

0.2

0 0.0

20.0

40.0

60.0

80.0

100.0 120.0 140.0

t (psec) Figure 1. A streak camera image of the current profile of drive and witness beams. The data has been rescaled to reflect the charge in each pulse.

II. OVERVIEW OF THE AWA FACILITY
The layout of the Argonne Wakefield Accelerator (AWA) can be found in Ref. 4. The electron source is a

III. THE PLASMA SOURCE AND DIAGNOSTICS BEAM LINE
The plasma source used for this experiment is a modified version of that used in previous linear-regime acceleration [5] and focusing [6] studies (see Fig. 2). It is a

DC hollow cathode arc discharge with a moderate confinement magnetic field (up to 1 kG; typically 200 G,) using Argon at up to 10 ?2 Torr pressures. The plasma electron density is variable between 8 × 1012 and 1 × 1014 cm3 , as measured with a Langmuir probe. The validity of the probe data is also supported by ?m-wave (140 GHz) interferometry measurements, which yield an integrated plasma density along a cross section of the plasma; the two techniques agree to within 15%. The length of the plasma column is 13 cm and n0 is within 25 % of the peak value over 11 cm of this length.

a

c f

b

d

e

Figure 2. Experimental setup including a) focusing solenoid, b) cathode assembly, c) plasma confinement solenoid, d) anode assembly, e) bend magnet, f) phosphor screen. In order to access the underdense regime, the beam must be brought to a tight focus - ultimately close to the betatron-matched condition for the ion channel. In order to easily control the location of this focal spot, we chose a solenoid magnet, placed about 40 cm before the start of the plasma. The beam is diagnosed with a phosphor screen before entering the solenoid. In the middle of the plasma chamber, we intercept the beam with a mirrored surface (with no plasma) and use the resulting optical transition radiation (OTR) to measure the solenoid's focal spot. After the plasma, the beam's energy distribution is measured with a spectrometer magnet. This spectrometer was designed in such a way that it has nearly point to point imaging in the bend plane from anywhere inside the plasma to the phosphor screen at the spectrometer's focal plane. The goal of this design is that the energy distribution be accurately measured regardless of the plasma caused focusing and guiding.

acceleration is also supported by computer simulation results (the simulation method is explained in Appendix A.) By the time of the second experimental run, we had developed the technique of witness beam generation, and there was more charge available from the linac. Also, because of improvements in the beam transport and emittance, the beam could be focused to σr =360 ?m. To investigate the possibility of particles accelerated by the plasma wave, as predicted by the Run I data and simulations, we recorded images of the high energy end of the spectrometer's phosphor screen. Fig. 3 shows the intensity profiles for all possible combinations of switching on of witness beam and plasma. Each point on this plot is the result of several camera images, averaging the set of energies corresponding to a fixed intensity. With this averaging, we attempt to improve on the energy jitter, probably due to laser injection phase fluctuations in the photoinjector, evident on the phosphor screen. The plasma density for this data was 2.2 × 1013 cm-3 , the drive and witness beams combined had Q = 13 - 17 nC, with 20% of this charge belonging to the witness beam. The initial σr was 450 ?m at athe entrance to the plasma. With no plasma, the Fig. 3 distributions are identical. This suggests that prior to entering the plasma, the witness beam's high energy tail is at or below that of the drive beam nd does not appear in the plots. Without a method to diagnose the witness beam alone, we can only conclude that the lower limit on the witness beam's gain in energy as a result of the beam-plasma interaction is 0.5 MeV. Note that at lowest intensity, the gain in energy of the tail is approximately twice as large again. Therefore the average acceleration field is at least as large as 4.1 MeV/m, and the peak field measurable is perhaps twice this value.
160 140

Video camera intensity

120 100 80 60 40 20 0 -1.5 -1.0 -0.5 0.0 0.5

wit. and pl. pl., no wit. wit., no pl. no wit., no pl.

IV. MEASUREMENT OF WAKE FIELDS
In the first experimental run (Run I), we did not have use of the witness beam. However, this was a good opportunity to verify that energy was being deposited into the plasma, by observing the beam's deceleration. The charge per bunch was 6-8 nC, the beam was focused by the solenoid to an rms radius σ r ≤ 500 ? m and n0 = 8 × 1012 . Assuming the bunch length to be 7.5 mm FWHM, the underdense factor, nb / n0 = 1. The result of these measurements is that the low energy half maximum point of the energy moved down by as much as 0.5 MeV, while the high energy half maximum remained within 20 keV of the no plasma case. It is encouraging that this data was shifted primarily toward the low energy side, indicating that the energy distribution was being accurately measured. The lack of the beam's self-

1.0

1.5

?E (MeV) Figure 3. Observation of the high energy tail at spectrometer. In the legend, wit. indicates witness beam and pl. indicates plasma present. The curvature near the top of the plot is due to saturation of the video camera.

We also changed the plasma density to below 1 × 1013 and above 4 × 1013 cm-3 and observed the effect on the energy distribution. From visual observation alone, the energy gain seemed weaker for these conditions. When trying to compare this data with simulation results, we found that within the range of Q, n0 and the length of the drive beam, it was possible to cause the acceleration of the drive beam's trailing edge. This, however, does not contradict the fact that the no witness beam high energy tail becomes depressed when the plasma is turned on the drive beam's energy is initially likely to be negatively correlated with distance from the leading edge, due to space charge and running off-crest in the linac, but the extent of this effect is unknown. Thus, direct comparisons with simulation are not yet feasible. The simulations also reveal that a σ r = 450 ?m beam's core focuses to 180 ?m in the plasma and then oscillates, staying below 330 ?m. This implies that in places along the propagation the beam is 4 times more dense than the plasma, satisfying the underdense criteria. The witness beam in the experiment was positioned so far back that it overlapped with the second peak in the acceleration field. This simulation resulted in the formation of wake fields of 20 MeV/m averaged over the propagation, assuming a 13 nC drive beam, a number not in disagreement with measurements. A major possible sources of disagreement between the experiment and the simulations is the nonaxisymmetry in the beam distribution due to cathode nonuniformity, space-charge and transverse wake-field effects. cm-3

APPENDIX A: COMPUTER SIMULATION METHODS
The computer simulations for this paper have used the code NOVO [9], which solves the Maxwell equations, assuming that this solution is unchanging in a frame moving with the beam. The plasma electrons are assumed to behave like a cold fluid and are modeled with the Maxwell-Boltzman equation. We have added a beam model to this code which tracks a set of super-particles, consistent with the NOVO picture [8]. This contradicts the assumption that the fields are static in the beam's frame, but this discrepancy is not very large, if we can assure that the beam changes very slowly compared to a plasma oscillation ( k p >> 1 β ). A similar argument is used with spatial variations in the plasma density. In the most extreme case, the beam crosses a plasma/vacuum boundary, with the method and have been incorporated into the code. causing a transient response, which is neglected in our treatment. More gradual variations in the plasma density, like the droop at the ends of the column, are entirely consistent.

REFERENCES
[1] J.B.Rosenzweig, T.Katsouleas, and J.J.Su, Phys. Rev. A 44, R6189 (1991). [2] A.Ogata et al., Proc. of the 6th Workshop on Advanced Accelerators, Lake Geneva, WI (1994). [3] J.Krall and G. Joyce, these proceedings. [4] P. Schoessow et al., these proceedings. [5] J.B. Rosenzweig, et al., Phys. Rev. Letters 61, 98 (1988). [6] J.B. Rosenzweig, et al,, Phys. Fluids B 2, 1376 (1990). [7] N.Barov and J.B. Rosenzweig, Proceedings of the1993 PAC , Washington DC. [8] N.Barov and J.B.Rosenzweig, Phys.Rev E 49 4407 (1994). [9] B.N.Breizman, T.Tajima, D.L.Fisher, and P.Z.Chebotaev (unpublished).

V. FUTURE WORK
After improving the longitudinal shaping and transverse focusing of the drive beam, as well as reducing some aperture scraping in our beam transport system, the greater peak density of the electron beam, and more importantly shorter bunch length, should drive much more powerful fields in the plasma - on the order of 200 MeV/m. Even higher gradienst should be possible with the addition of a compressor chicage which is currently under construction. We will accompany this effort with measurements of the beam spot prior to exiting the plasma. A time resolved image of this spot [7] will confirm that the beam is propagating in a well focused state through the plasma, and that phase-mix damping leads a larger equilibrium radius near the beam's leading edge [8]. We also plan to measure the UV and VUV radiation emitted by the plasma as a result of the beam plasma interaction. The plasma wave is characterized by plasma electrons which are mildly relativistic ( v c ≤.5 ). These energetic electrons will further ionize the Ar ions and excite atoms, causing emission of UV line radiation.


相关文章:
...OF PLASMA WAKE-FIELDS IN THE BLOW-OUT REGIME.pdf
MEASUREMENTS OF PLASMA WAKE-FIELDS IN THE BLOW-OUT REGIME_专业资料。Initial results from nonlinear plasma wake-field MEASUREMENTS OF PLASMA WAKE-FIELDS IN ...
...a High Charge Bunched Electron Beam in Plasma.pdf
plasma behavior in the blow-out regime have been...In recent measurements with compressed beam at ...wake-field acceleration, where the electromagnetic ...
Highlights of AAC 2000 Workshop.pdf
plasma (PWFA[2]) of high density n0 , in ...the linear theory of Cerenkov wake-fields. In...in the blow-out regime which also reached ...
Two-Dimensional Nonlinear Regime in the Plasma Wake....pdf
Two-Dimensional Nonlinear Regime in the Plasma Wakefield Accelerator_专业资料。The effect of nonlinearity on plasma wake wave excited by a relativistic ...
Transition Wake-Fields in Resistive Tube.pdf
Transition Wake-Fields in Resistive Tube_专业资料...in steady state regime have been well understood ...plasma frequency of the free electrons in the ...
1 Design of A High Charge (10- 100 nC) and Short Pu....pdf
plasma wakefields in the blow out regime, ...should not exceed 1/3 of the excited wake field...1 peak axial electric fields in the gun [4]....
PRL-2004-93-223001-Moshammer.pdf
T. Costello National Centre for Plasma Science ...In this regime the photon density is so high ...measurements of the ion yield arising from double...
The damage mechanism of fused silica surface induce....pdf
that in the picosecond regime for the interaction of the laser with plasma...Experimental system for measurements of laser induced plasmas and shock waves ...
SECONDARY EMISSION IN COLD-CATHODE MAGNETRON INJECT....pdf
in short-pulse regime of the gun and 3 ...Such a regime is impossible at plasma explosion ...fields corresponding the frequency of electron ...
Magnetic Field of Relativistic Nonlinear Plasma Wave.pdf
eld of plasma wave is zero. The magnetic ?eld...(wake waves) excited in cold plasma by ...In the nonlinear regime the behavior of plasma ...
Electrochemistry of thorium in LiClKCl eutectic m....pdf
of all chemicals were carried out in a glove ...The electrochemical measurements (cyclic voltammetry,...the initial regime of each transient is characterized...
Effect of gas mixing ratio on etch behavior_图文.pdf
xAlxOy thin ?lm in Cl2 / BCl3 plasma. Chang...ux-limited etch regime of ionassisted chemical ...measurements and quadrupole mass spectrometry QMS ....
In-situ stability study of nitrocarburized 304 stai....pdf
The plasma immersion ion implantation (PIII) has ...measurements were made in Bragg-Brentano geometry ...regime Temperature (8C) As-prepared (RT) 300 ...
Do Ordinary Nuclei Contain Exotic States of Matter.pdf
out the strength of two-nucleon correlations in ...regime at x > 1 would allow us to measure ...average 4 early universe LHC quark-gluon plasma ...
Mid-infrared extraordinary transmission through Ga-....pdf
on silicon with plasma frequencies in the near-infrared...in this wavelength regime potentially giving ...measurements of square 2D-hole arrays in ZnGaO ...
Melting temperature of screened Wigner crystal on h....pdf
the plasma parameter in the melting temperature de...Therefore, such electrons in this regime obey the...measurements of the melting temperature in the two...
New electromagnetic materials emphasise_图文.pdf
the essence a plasma of free charged particles. ...regime e for compared to solid metal, about the...Magnetic fields in the we introduce capacitance ...
Low intensity threshold values for keV X-ray production in ....pdf
measurements, and short lifetime states down to ...plasma frequency in the cluster matches the laser...recorded in the regime of high laser intensity, ...
Evolution of surface roughness of AlN and GaN induc....pdf
and a high-frequency regime exhibits scaling properties...In the case of plasma etching, the ?rst and ...All measurements reported here are on nominally ...
大学专业英语.pdf
24. S. C. Brown, Basic Data of Plasma ...are made for the choice of the optimal regime....The possible fields of application of the PLE ...