Proposed Physics Experiments

for Structure-based Laser-driven Acceleration in a Vacuum



Submitted to

The Steering Committee of the Brookhaven National Laboratory

For Access to the Accelerator Test Facility


Submitted by

Yen-Chieh Huang, Principal Investigator


Assistant Professor of Electrical Engineering

National Tsinghua University

Hsinchu, Taiwan 30043




May, 2000




        The goal of this proposal is to experimentally confirm electron acceleration from a laser-driven accelerator structure in a vacuum. The project takes advantage of the 75 MeV electron beam and the CO2 pulse laser at ATF. Due to the high electron energy and the long laser wavelength, the accelerator structure can be as large as 10 cm, consisting of 5 accelerator cells. The predicted electron energy gain is about 1 MeV. The experimental result of this project is to answer the question of the possibility of vacuum laser-driven particle acceleration in a solid structure. The accelerator structure will be fabricated and optically tested at National Tsinghua University, Taiwan. The laser-driven acceleration experiment is to be carried out at the ATF facility, Brookhaven National Laboratory, USA. The experimenters include Y.C. Huang, the principal investigator, Y.H. Chen, the postdoctoral research associate of Huang’s group, and two graduate students from Huang’s group. The amount of beam time needed is approximately three weeks near the end of the fiscal year. The funding is primarily from National Science Council, Taiwan, with the possibility of joint supports from other agents in the US.



National Tsinghua University, Taiwan


1)  Structure fabrication of a CO2 laser-driven accelerator structure.

The structure consists of five accelerator cells with electron transmitting holes along the acceleration path. The whole accelerator structure is installed in a vacuum chamber that may fit into the ATF beam line.

2) Optical test of the accelerator structure

The test includes laser coupling to the accelerator structure, excitation of the acceleration laser mode, and the phase control of individual accelerator cells.



ATF, Brookhaven National Laboratory, USA.


1)  Laser acceleration chamber installation.

2)  Cold test of the accelerator structure by using the ATF CO2 laser.

3)  Electron beam energy gain/loss measurements


1.  Introduction

Laser driven particle acceleration finds potential applications in two areas: 1) high-gradient acceleration, leading to table-top accelerators or linear colliders; 2) ultra-short electron bunch generation, leading to coherent x-ray generation. Current laser driven particle acceleration schemes can be divided into two categories: one with a medium and one without a medium in the electron acceleration path. The one with a medium includes the plasma-based laser acceleration [1] and the inverse Cherenkov acceleration [2] ; the one without a medium includes inverse free-electron (FEL) acceleration [3] , and structure-loaded vacuum linear acceleration [4] . The acceleration schemes adopting media along the particle acceleration path often comprise with material properties, including scattering and stability. The inverse FEL acceleration suffers from excess radiation loss when scaled to high energies. A structure-based laser accelerator, although having the material stability, has a lower acceleration gradient compared to other laser-driven accelerators due to structure damage. Nonetheless the laser acceleration technology has been making significant progresses, and problems are being solved one by one. Among all the laser acceleration schemes, the structure-loaded vacuum laser-driven acceleration still requires a proof-of-principle experiment to let the technology grow. The goal of this proposal is to overcome this barrier by verifying electron energy gain in a laser-driven accelerator structure in a vacuum.

Since the laser wavelength is on the order of a micron, the accelerator structure size scaled from the RF accelerators will be nearly impossible to fabricate or operate. Consequently novel designs for laser-driven accelerator structures are necessary. The possibility of having an accelerator structure size exceeding thousand times the laser wavelength does exist. However the overall size of the accelerator structure is still roughly scaled with the driving wavelength. Another factor that influences a laser accelerator structure size in a test experiment is electron slippage due to a small relativistic factor g.

    For ease of the experiment, the design criterion of the accelerator structure is to have a large structure size and a large electron energy gain. The acceleration gradient, on the other hand, is not taken into the consideration. Currently a dielectric-based vacuum laser acceleration is being carried out at Stanford Hansen Experimental Physics Laboratory. Compared to the Stanford project, this proposal takes advantage of the 10 times longer wavelength and two times higher electron energy at the ATF. As a result, the overall size of the accelerator cell is on the order of centimeter and can be handled by human hands. The proposed project also adopts a multiple-cell design that predicts an overall electron energy gain exceeding 1 MeV.

2. The Proposed Experiment

A.  Theoretical Background

Several structure-based vacuum laser-driven acceleration schemes have been proposed in the past. For this particular project, we employ the TEM10 laser-mode field for electron acceleration based upon the following considerations:

1). As a proof-of-principle experiment, the experimental results can be interpreted directly from the mode concepts in a conventional RF accelerator.

2)  The accelerator structure mimics a laser resonator or a lens array that can be assembled by commercial optical components.   

3)  The excitation of the TEM10 laser mode is a standard, well-known mode-matching technique.

4)  The characterization and analysis of the TEM10 laser mode is well developed in laser technologies.

Laser-driven particle acceleration by using the TEM10 laser field has been analyzed by E.J. Bochove [5] et al.  Without considering electron slippage due to a finite g, the maximum interaction length is two Rayleigh ranges , from which the electron acccumulates the largest possible energy gain from single-stage acceleration. The corresponding maximum, single-stage energy gain is given by

 (MeV),                                                              (1)

where P is the pumping laser power and h is the wave impedance. The accelerator structure under consideration is therefore a confocal laser resonator, as illustrated in Fig. 1.








Figure 1. The confocal laser resonator as a single-accelerator cell.


        The energy-related phase slippage is governed by the plane-wave phase term, given by

,                                                                          (2)

where  is the wave number,  is the laser angular frequency, v is the electron velocity, and L is the acceleration length. If one arbitrarily limits the phase slip to 10% of the 180° phase reversal , the interaction length becomes , which clearly shows the advantage of having a high electron energe g and a long laser wavelength. Thus for the 75 MeV electron beam and the 10 mm CO2 laser wavelength at ATF, the single-stage accelerator cell length is approximately  cm. With this kind of accelerator size, the structure can be fabricated, adjusted by using conventional optical technologies.

        The single-stage energy gain is subjected to the structure damage. For the 150 psec CO2 laser pulse width at ATF, the material damage fluence is about F = 30 J/cm2. The laser radius at the mirror is , where  is the laser waist. The maximum power that the mirror may sustain at the  psec is therefore  MW. Subsitituting P into Eq. (1), one obtains the maximum single-stage energy gain 230 KeV. Note that, for this proposal, it is not intented to demonstrate high gradient acceleration.The proposed work is to achieve measurable acceleration gain at the sacrifice of the acceleration gradient. In other words, a maximum slippage distance permits a large accelerator size and the ease of a proof-of-principle experiment. With the success of this project, the acceleration gradient may approach the structure damage field ~ GV/m if a smaller accelerator cell is chosen.

  B. Laser-driven LINAC Design

        In a practical accelerator cell, there exist two electron transmitting holes. In our computer simulation [6] we found no degradation in the laser mode and the single-stage energy gain, when we opened the transmitting holes with their diameter = 50 mm. The electron energy loss due to diffraction radiation is estimated to be less than 10 keV per transit. We are currently investigating the effects of opening an even larger electron hole. The robust of the laser mode and the low optical loss are primarily due to the null field at the center of the TEM10 laser mode.

        Cascading accelerator cells, as illustrated in Fig. 2, may increase the total electron energy gain in the test experiment. The focal length of each lens f is equal to the half the radius of curvature of the mirror in Fig. 1, or  . With 5 accelerator stages at a total length of 11.25 cm, the maximum energy gain is 1.15 MeV. The high reflector in the downstream removes the laser energy and prevents it from taking away electron energies in the decelerating phase. The phase reset of individual cells can be accomplished by tuning temperatures or PZT positions associated with the lenses. To reduce the laser loss, all lenses can be anti-reflection coated. In order to avoid the cumbersome process of converting the pump laser into the TEM01 mode, the first lens can be coated with an additional p-phase-shift layer in its upper half. This way the mode conversion from TEM00 to the acceleration mode is simplified and can be accomplished with good coupling efficiency.  







Figure 2. A lens array as a laser-acceleration linac.


        Once the structure is fabricated and optically tested on Taiwan, the structure, along with the vacuum chamber, will be installed to the ATF beam line. The size of the vacuum chamber is around one foot at most. The vacuum chamber is self-contained, with all the control and tuning electronics built on Taiwan.

The electrons intercepted by the 50 micron aperture is more than 50%, given normalized emittance = 2 pi-mm-mrad and a reasonable beta function ~ 0.5 m. With the nC per bunch electrons, half of it can be detected with ease. The important apparatus in the downstream beam line that ought to be provided by ATF is an energy spectrometer capable of discerning the ± 1.3% energy spread resulting from the 5 acceleration stages. With this amount of energy spread, a typical energy spectrometer should be good enough for the experiment.

C.     Comparison Between This Proposal with the Stanford LEAP Project

Since May 1997, Stanford has been engaged in a dielectric-loaded laser electron acceleration project (LEAP) [7] . LEAP is aimed to demonstrate vacuum laser-driven electron acceleration in a dielectric accelerator structure. Table 1 shows the comparison between this proposal and LEAP.


Table 1. The comparison between Stanford LEAP and this proposal.


This proposal

Stanford LEAP

Electron beam energy

75 MeV

40 MeV

Laser wavelength

10 mm (CO2)

800 nm (Ti:sapphire laer)

accelerator cell length

2.25 cm

2.5 mm

Accelerator structure

A 5-cell linac

A single accelerator cell

Total energy gain

1.15 MeV

300 keV

Acceleration field

TEM10 mode

Crossed laser beam


    When LEAP was initially proposed, the system parameters are limited to those available at the Stanford Hansen Experimental Physics Laboratory. Also, the mission of LEAP is to demonstrate a novel dielectric accelerator structure driven by a state-of-the-art solid-state laser. As a result, the Stanford LEAP is highly challenging and significant. 

    The physics experiment proposed in this project is more conservative while being historically important. With the success of this project, the accelerator physics community may answer the question of the possibility of vacuum laser acceleration in the last several decades. Moreover this project is proposed under the following practical considerations:

a.  The laser accelerator structure is large and can be handled by hands.

b.  The total energy gain exceed 1 MeV, allowing definitive energy measurements.

c.  The acceleration field is from a laser mode, permitting a direct comparison with the conventional RF accelerator theory. 

The data collected in this proposal at the 10 mm wavelength will be complimentary to those collected in the more advanced and yet difficult Stanford LEAP at the 800 nm laser wavelength.

3.      The Ability of Huang’s Group for Carrying out this Project

A.  Huang’s Group

The principle investigator, Y.C. Huang, has an extensive experience in laser, electron beam, and radiation. The following is a brief sketch of Y.C Huang.

a.  Under R.H. Pantell of Stanford, Huang designed, built, and characterized a compact far-infrared free-electron laser [8] .

b.  Under R.L. Byer, Huang initiated the Stanford LEAP project for laser-driven electron acceleration.

c.  Under M. Cornacchia, Huang worked part time at Stanford Synchrotron Radiation Laboratory as an infrared-laser-detection consultant Scientist.

Currently Huang’s group consists of 7 graduate students and one postdoctoral research associate, being supported by the National Science Council (NSC) and industrial affiliates on Taiwan. Huang is currently the principal investigator of the proposed NTHU Relativistic Photon-electron Dynamics Laboratory.

  B. Funding

        The expense of this project is fairly moderate, primarily for the accelerator fabrication on Taiwan and travel to ATF. The fabrication and optical test of the accelerator structure, and the manpower cost will be supported by an existing NSC-funded proposal.  In addition, Huang will submit a proposal to NSC-Taiwan as well as to DoE-USA for the travel budget. Huang’s group in the last two years was well funded at a level of US$300k per year. We do not expect undue difficulties in carrying the proposed work in this project.

4. Estimated ATF Support and Gantt Chart

        The ATF support can be divided into two parts: manpower and equipment. For manpower, the Huang’s group would appreciate some operator’s time during the experiment. If necessary, Huang’s group may have a graduate student trained at ATF prior to conducting the major experiment. Therefore the graduate student from Taiwan may be the assistant to the machine operator. Huang’s group will be responsible for all the other manpower necessary for the vacuum chamber installation and laser acceleration experiments, while welcoming any help and collaboration from local groups.

        The experiment is designed to fit into one of three high-energy beam lines at ATF. The chosen beam line should have an energy spectrometer in the down stream and easy access to the pulse CO2 laser. The required laser energy from the 150 psec CO2 laser is merely 20 mJ/pulse.

        For conducting the experiment at ATF, we would like to have three experiment sessions with one-week beam time in each session. Based upon the Stanford LEAP experience, we propose one-week intermittence between adjacent experiment sessions for any major modifications. To complete this project in a one-year period is not an easy task. If necessary, we will propose the continuation of the acceleration test in the following year based up the first-year performance.  Table 2 shows the Gantt charge of our tentative effort.


Table 2 The Gantt Chart of the Project Progress



work items




















Structure design













Structure fabrication













Optical Test













Vacuum chamber fabrication













Beam line installation and test













Laser acceleration test













If necessary, the acceleration test can be continued in the following year.





[1] See for example T. Tajima and J.M. Dawson, Phy. Rev. Lett. 43, 267 (1979); or C.E. Clayton, K.A. Marsh, A. Dyson, M. Everett, A. Lal, W.P. Leemans, R. Williams, C. Joshi, Phys. Rev. Lett. 70, 37, (1993).

[2] See for example, W.D. Kimura et al., “Laser Acceleration of Relativistic Electrons Using the Inverse Cherenkov effect,” Phys. Rev. Lett. Vol. 74, No. 4, (1995) pp. 546-549.

[3] See for example in P. Sprangle, “Laser driven electron acceleration in vacuum, gases, and plasmas,” Phys. Plasmas 3 (5) (1996) p. 2185.

[4] See for example Y.C. Huang et al., “A proposed high-gradient laser-driven electron accelerator using crossed cylindrical laser focusing,” Appl. Phys. Lett. 69 (15) (1996) pp. 2175-2177.

[5] E.J. Bochove, G.T. Moore, and M.O. Scully, “Acceleration of Particles by an Asymmetric Hermite-Gaussian Laser Beam,” Physical Review A,vol. 46, No. 10, 6640 (1992).

[6] Y.C. Huang, Proceedings Orion Workshop, Feb. 23-25, 2000, SLAC, Stanford, California.

[7] Y.C. Huang et al, “The Physics Experiment for a Laser-driven Electron Accelerator,” Nucl. Ins. Meth. A 407 (1998) 316-321.

[8] See for example Y.C. Huang et al., “Electron beam characterization for a compact far-infrared free-electron laser,” IEEE J. Quan. Electronics, Vol. 31, NO. 9 (1995) pp. 1637-1641.