Longitudinal compression oC a tailored-vciocity, intcmc neutralized ion heam hns been demonstmtcd. The compression tnlm plnec in a 1-2 m driit seetion Blled with plam~n to provide spacedierge neutduntioo. An induction cell produces a head-to-tnil velocity ramp that longitudinally compresses theneutrdized beam, enhancingthe b-peak current hyafndor of 50 audprodudag a pulse duration OC about 3 us. T h i s m m e m e n t hns been confirmed indopmdsntly with two darercot diagnostic eystems. The simultaneous l m v e r a e nnd longihdinal compression oI an ion heam is reguired t o achieve the high intensities necrssnry to create high energy density matter and fusion conditions. A recent driver study for inertial fusion, for Longitudinal compression of space-chargedominated beams has been studied extensively intheory and simulations [ll-E]. The compression is initiated by imposing a linear head-to-tail velocity tilt to a driitiig beam. Longitudinal space-&urge forces limit the beam compression ratio, the ratio oI the initial to i i nmal current, to about ten in most applications. iments on NDCX. To provjde the head-ta-tail velocity tilt, aninduction module withvariablewltagewaveronn is placed immedintely downstream of the last quadrupole mnpet. This is IoUmved hy a neutralized drii section which consists of a one -meter-long plasma column produced hy an AI cathodic cm: [ZO]. A diagnostic hmc is located at the downstream end of the plwarna column Thebeampmducedfromthesourcehasa5 paflat-top. The inductiontit voltage 'c~rves' out a -300 115 segment ofthe flat-top which compresseslongitudinallyas it driits through the plasma column. The final compressed beam is m e w e d in the dormetrenm diagnostic box.The induction cell consists oi 14 independently-driven magnetic cores in a preastnizad gas @Fa) region that is separated Erom the vnouum by a conventional high voltage insulator. The rvnveforms applied t o the 14 coria inductively add at the acceleration gap. Each core is driveu by a thyratronawitched modulator. Because the modulntor for each core can be designed to produce different waveiorms and can be triggered independently, a variety or wavdorms CM bs produced nt the acceleration gap using the 14 discrete building bloclw.The plasma column is formed hy two pulsed a u m i n u m cathodic arc sources loceted at the d m s t r e a m ond. Each source is equipped with a 45O open-arcutechm

A multibeamlet approach to a high current ion injector, whereby a large number of beamlets are accelerated and then merged to form a single beam, offers a number of potential advantages over a monolithic single beam injector. These advantages include a smaller transverse footprint, more control over the shaping and aiming of the beam, and more flexibility in the choice of ion sources. A potential drawback, however, is a larger emittance. In this paper, we seek to understand the merging of the beamlets and how it determines the emittance. When the constraints imposed by beam propagation physics and practical engineering issues are included, the design is reduced to a few free parameters. We describe the physics design of a multibeamlet injector and produce a design for an example set of parameters. Extensive use of 2D and 3D particle simulations was made in understanding the injector. Design tolerances and sensitivities are discussed in general and in relation to the example.

We have demonstrated experimental techniques to provide active neutralization for space-charge dominated beams as well as to prevent uncontrolled ion beam neutralization by stray electrons. Neutralization is provided by a localized plasma injected from a cathode arc source. Unwanted secondary electrons produced at the wall by halo particle impact are suppressed using a radial mesh liner that is positively biased inside a beam drift tube. We present measurements of current transmission, beam spot size as a function of axial position, beam energy and plasma source conditions. Detailed comparisons with theory are also presented.

Intense beams of heavy ions are well suited for heating matter to regimes of emerging interest. A new facility, NDCX-II, will enable studies of warm dense matter at ϳ1 eV and near-solid density, and of heavy-ion inertial fusion target physics relevant to electric power production. For these applications the beam must deposit its energy rapidly, before the target can expand significantly. To form such pulses, ion beams are temporally compressed in neutralizing plasma; current amplification factors of ϳ50-100 are routinely obtained on the Neutralized Drift Compression Experiment ͑NDCX͒ at the Lawrence Berkeley National Laboratory. In the NDCX-II physics design, an initial non-neutralized compression renders the pulse short enough that existing high-voltage pulsed power can be employed. This compression is first halted and then reversed by the beam's longitudinal space-charge field. Downstream induction cells provide acceleration and impose the head-to-tail velocity gradient that leads to the final neutralized compression onto the target. This paper describes the discrete-particle simulation models ͑one-, two-, and three-dimensional͒ employed and the space-charge-dominated beam dynamics being realized.