A dc proton injector is being developed for a 6.7-MeV CW RFQ at Los Alamos. The RFQ input beam requirements are 75-keV energy, 110-mA dc proton current, and 0.20 'mm-mrad rms normalized emittance. The injector has now produced a 75-keV, 117-mA dc proton beam (130-mA total current) with the required emittance. The emittance has been measured after a 2.1-m-long two-solenoid beam transport system. The measured emittance can be explained in terms of the ion source emittance and beam transport through the focusing elements. Measured proton fractions are 90 - 92% of the beam current. The engineering of the accelerating column high-voltage design is being improved to increase the injector reliability. Injector design details and status will be presented.

High-current (100 mA), high-energy (1 GeV) linacs are being designed for accelerator-driven transmutation technology (ADTT) applications [1]. A CW radio frequency quadrupole (RFQ) has been designed to accept a 75-keV, 110-mA proton beam from a dc injector to produce a CW 100-mA, 6.7-MeV final beam [2]. The dc proton injector and the CW RFQ are being developed for the low-energy demonstration accelerator (LEDA) project at Los Alamos [3].

This dc proton injector development began at Los Alamos with a collaborative program with Chalk River Laboratories (CRL) [4]. The microwave-driven proton source [5] developed at CRL has now been extended to meet most of the LEDA beam requirements as shown in Table 1. Beam diagnostic measurements and interpretation are described in the following sections. Other injector details may be found in a recent review [6].

Figure 1 shows the injector configuration which has been used in these initial 75-keV beam tests. It is a prototype to check the conceptual design before building the final

Table 1.

Summary of the LEDA injector requirements and present status.

Parameter	Req.	Status
Energy (keV)	75	75
Proton current (mA)	110	117
Duty factor (%)	100	100
H2 Gas flow (T-l/s)	0.04 - 0.1	0.04 - 0.09
Proton fraction (%)	>70	91
Reliability (%)	98	To do
Lifetime (hr)	>168	To do
Beam noise (%)	±1	±1
LEBT exit emit. ('-mm-mrad)	0.20	0.20

LEDA low-energy beam transport (LEBT). The total LEBT length is 2.1 m from the ion source to the emittance

measuring unit (EMU) entrance slit. The magnetic focusing system consists of two joined solenoids. The injector is designed to operate in the dc mode at 10-kW beam power, and all of the diagnostics (excepting the EMU) are of the non-interceptive type. The diagnostics with their distance (z) from the ion source extractor are: (1) Bergoz dc beam current monitor (0.30 m), (2) four-grid energy analyzer (FGA) beam space-charge neutralization monitor (0.41 m), (3) x,y video profile CCD imaging (0.42 m), (4) ac current toroid for measuring beam current fluctuations (0.53 m), and (5) the EMU (2.1 m).

Figure 2 shows a set of measurements for the total beam current (Bergoz dc monitor) and beam fluctuations plotted versus the ion source axial magnetic field. Accuracy of the measured axial magnetic field is estimated to be 3 - 5%. Two

maxima in the total current data (left scale) are seen: the first at 0.0875 T which satisfies the electron-cyclotron resonance (ECR) condition at 2.45 GHz, and the second sharper beam current resonance at 0.0935 T. This type of resonant behavior is typical [5].

The ac toroid is a Supermalloy transformer with a T_r = 1(A/V) transfer ratio with a flat bandwidth response from 1 kHz to 10 MHz. The rms beam noise (i_rms) data reported in Fig. 2 (right scale) is obtained from the integrated power P_t and the relation i_rms(A) = (P_t*R)^1/2 T_r where R = spectrum analyzer input impedance = 50_. The power sum is done over the beam noise frequency, f, from 12.5 kHz to 1 MHz. Measured beam noise is maximum at low frequencies, and decreases to background for f > 1 MHz. The beam is generally tuned to minimum noise at each magnetic field setting by minimizing the reflected power from the ion source by adjustment of the three-stub tuner in the 2.45 GHz waveguide. Quiescent beams are obtained over the lower field broad resonance, whereas it is somewhat more difficult to maintain a quiet beam at the higher magnetic field.

The FGA [7] has been used to measure the degree of beam space-charge neutralization f_sc within the LEBT. An energy distribution of the radially-flowing beam-plasma ions measured with this diagnostic is shown in Fig. 3 for a i_b = 100-mA, 75-keV hydrogen ion beam [8]. The derivative of the FGA Faraday cup current vs. the discriminating grid voltage (grid 3) is shown. The base width of this distribution

is Æø = 4V, and this leads to f_sc(%) = (1 - Æø/Æø_u)*100 = 98% where Æø_u = 240 V, the radial potential drop across an unneutralized uniform beam. The H₂ LEBT gas density = n₀ = 1.5 x 10¹² (cm)^-3.

The measured phase-space distribution of a 130-mA, 75-keV hydrogen-ion beam is shown in Fig. 4. The proton fraction is 90%, thus the proton current is 117 mA. The focusing solenoids were both excited to 0.17 T, which gives a 7-cm diam. beam (10% contour) at the EMU. This focusing strength gives an average power loading of 0.25 kW/cm² at the EMU slits. The measurement is made by a two-slit technique using a dc emittance-measuring device [9]. The contaminant H₂⁺ beam is focused less and is visible in Fig. 4. A Gaussian extrapolation procedure [6] is used to extract the proton rms normalized beam emittance of 0.20 ('mm-mrad) which is the design RFQ input emittance.

The measured emittance may be partially understood by examining possible emittance growth mechanisms in the LEBT. Estimates can be made for (1) solenoid aberrations, (2) nonlinear space charge, (3) beam fluctuations, and (4) power supply regulation. The effects of (1) and (2) are estimated with the code SCHAR [10], which calculates beam trajectories through the measured solenoidal magnetic fields and takes into account the residual beam space charge. The influence of effects (3) and (4) on the measured beam emittance may be estimated by applying the mismatch factor concept to beam emittance growth in LEBTs [11].

Figure 5 shows the higher-order SCHAR code prediction for 999 macroparticles traced from the ion source extractor to the EMU superimposed on the measured 10% contour from Fig. 4. The SCHAR starting beam distribution (phase-space orientation, emittance) is deduced from drifting the measured

beam parameters at the 10% threshold backwards through the LEBT with the first-order transport code TRACE [12]. The measured magnetic fields and a residual beam space charge corresponding to a current of 3.5 mA are included in the simulation. Onset of a third-order aberration is observed in Fig. 5 in both the measurement and prediction. The SCHAR code predicts an 18% emittance growth for the beam transport through this 2.1 m long LEBT.

Soloshenko [13] has shown that dynamic decompensation of space-charge neutralized beams will occur when n₊₀v_bn₀s_e/2f << an₊₀ where n₊₀ is the beam density, v_b is the beam velocity, s_e = 2 x 10^-16 cm² is the electron production cross section, and a is the beam noise amplitude. The inequality is satisfied when the electron density produced in the beam by ionization is much less than the beam current density fluctuations. This effect would be important for this injector operating at a = 1% when f > 1 MHz, but beam noise from the microwave source has typically reached background at f = 1 MHz. An effective emittance growth of <10% from beam current and voltage fluctuations is estimated using the TRACE code and the beam ellipse mismatch concept [11].

A maximum 30% emittance growth estimate from known beam-transport effects has been made. The optimum ion source only emittance is estimated to be 0.13 ('mm-mrad) by extrapolation of the published emittance vs. emission aperture radius to the 4.2 mm value used in these measurements [5]. It may thus be possible to reduce the injector emittance performance below the measured 0.20 ('mm-mrad) by optimizing the ion extraction system. Injector work has shifted to increasing extraction voltage stability in order to meet the 98% reliability requirement.

[1] G. P. Lawrence, "Transmutation and Energy Production with High Power Accelerators", Proc. 1995 Particle Accelerator Conf., IEEE No. 95CH35843, (Dallas, Texas), 1 (1995).

[2] J. H. Billen, S. Nath, J. E. Stovall, H. Takeda, R. L. Wood, and L. M. Young, "A Versatile, High-Power Linac for Accelerator Driven Transmutation Technologies", Proc. 1995 Particle Accelerator Conf., IEEE No. 95CH35843, (Dallas, Texas), 1137 (1995).

[3] J. David Schneider, these proceedings.

[4] J. D. Schneider, J. Bolme, V. Brown, M. Milder, G. Neuschaefer, D. Rees, P. Schafstall, J. Sherman, R. R. Stevens, Jr., T. J. Zaugg, G. E. McMichael, J. Sheikh, and T. Taylor, Proc. 17th Int. Linac Conf., (Tsukuba, Japan), 149 (1994).

[5] T. Taylor and J. F. Mouris, "An Advanced High-Current Low -Emittance dc Microwave Proton Source", Nucl. Instrum. and Methods in Phys. Res. A336, 1 (1993).

[6] J. Sherman, G. Bolme, C. Geisik, D. Gilpatrick, L. Hansborough, D. Hodgkins, P. Lara, E. Meyer, J. Power, C. Rose, D. Sandoval, P. Schafstall, J. D. Schneider, M. Stettler, R. R. Stevens, Jr., M. Thuot, R. Wright, T. Zaugg, D. Spence, G. McMichael, and T. Taylor, "Microwave Proton Source Development for a High-Current Linac Injector", Rev. Sci. Instrum. 67(3), 1296 (1996).

[7] J. Sherman, E. Pitcher, R. Stevens, and P. Allison, "H^- Beam Neutralization Measurements in a Solenoidal Beam Transport System", AIP Conf. Proc. No. 287, (Upton, New York), 686 (1992).

[8] R. Ferdinand, private communication and to be published.

[9] T. Taylor, M. S. de Jong and W. L. Michel, "An Emittance Measuring System for High-Current High-Brightness Multi-Beamlet Multi-Species Heavy-Ion Beams", Proc. of the 1988 Linear Accelerator Conf., CEBAF-Report-89-001, 100 (1988).

[10] Mark J. Jakobson and Richard J. Hayden, Nucl. Instrum. and Methods A258, 536 (1987).

[11] Paul Allison, "Emittance Growth Caused by Current Variation in a Beam Transport Channel", IEEE Trans. Nucl. Sci., NS-32, 2556 (1985).

[12] K. R. Crandall and D. P. Rusthoi, "Documentation for TRACE: An Interactive Beam Transport Code", Los Alamos National Lab Internal Report, LA-10235-MS (Jan., 1985).

[13] I. A. Soloshenko, "Physics of Ion Beam Plasma and Problems of Intensive Ion Beam Transportation", Rev. Sci. Instrum. 67(4), 1646 (1996).

*Work supported by the U. S. Department of Energy.