Ion sources for high and low energy extremes of ion implantation

Ady Hershcovitch

doi:10.1109/plasma.2008.4590620

Abstract

Various types of ions, but mostly B, P, Sb, and As, are implanted, over a wide range of energies into some of the materials used in the construction of semiconductors. These energies range from as low as approximately 100 eV for shallow implantations, to as high as multi-MeV for deep implantation into the substrate. State of the art ion sources meet industry needs for the energy range of about 10 keV to about 300 keV. But, at the two extremes (100's of eV and multi-MeV) of the energy range, source improvement is needed, due to space charge limitations at the low energy range and due to inefficiency in acceleration at the higher energy range. A joint research and development among HCEI, Tomsk, Russia, ITEP, Moscow Russia and BNL collaboration is focusing on meeting industry needs for steady state, intense ion sources has been in progress. Originally, the collaboration started to develop pulsed metal vapor ion sources, to enhanced charge states, with an external electron beam in two ion sources provisionally dubbed E-MEVVA. Lead and Bismuth, which previously achieved doubly charged ions, were ionized to ion charge states of Pb+7 & Bi+8 with ion currents exceeding 200 mA. The natural next step was to adapt these charge enhancement characteristics to ion sources that generate steady state multi-charged B, P, As, and Sb ions. These technical enhancements can be adapted to DC ion implanters in order to improve upon present day high- energy ion implanters that use rf accelerators. This endeavor has resulted in record steady state output currents of higher charge state Antimony and Phosphorous ions: P 2+ (8.6 pmA), P 3+ (1.9 pmA), and P 4+ (0.12 pmA) and 16.2, 7.6, 3.3, and 2.2 pmA of Sb 3+ Sb 4+ , Sb 5+ , and Sb 6+ respectively. However, the semiconductor industry has greater needs in the area of low energy (100's of eV) ion implantation, where space charge problems associated with lower energy ionbeams limit implanter ion currents, thus leading to low production rates. To tackle the space charge problem, two approaches were followed: using molecular ions and ion beam deceleration with space charge compensation. To date, 1 emA of positive Decaborane ions were extracted at 10 keV and a somewhat smaller current of negative Decaborane ions were also extracted. Some simulations of a novel gasless/plasmaless ion beam deceleration method were also performed. Finally, a spin-off result of a Bernas-Calutron ion source, from which over 70% of the extracted ion beam consists of singly charged Boron, was achieved (compared to the 20% of current implanters). This paper is a synopsis of an extensive ion source R&D program designed to address industry needs.

Full Text