B.P. Wood

Displacement current and multiple pulse effects in plasma source ion implantation

In plasma source ion implantation (PSII), a target to be implanted is immersed in a weakly ionized plasma and pulsed to a high negative voltage. Plasma ions are accelerated toward the target and implanted in its surface. In this article, two factors in the analysis of these discharges are examined for the first time: (1) displacement current across the expanding sheath results in increased implant current and decreased implanted ion energy, with respect to existing models; and (2) ion depletion around the target due to high pulse repetition rates results in decreased implant current and dose. These effects are studied with analytic models and particle‐in‐cell simulations. Simulation results are compared to previously published PSII models.

Advances in PSII techniques for surface modification

Abstract Recent activities in plasma source ion implantation (PSII) technology include scale-up demonstrations for industry and development of variations on the original PSII concept for surface modification. This paper presents an overview of the continued growth of PSII research facilities world-wide and the industrial demonstrations within the USA. In order to expand the applicability of PSII, Los Alamos is actively researching a PSII-related technique called plasma immersion ion processing (PIIP). In one case, a pulsed-biased target can be combined with cathodic arc sources to perform ion implantation and coating deposition with metal plasmas. Erbium plasmas have been combined with oxygen to deposit erbia (Er 2 O 3 ) coatings that are useful for containment of molten metals. In a second case, hydrocarbon, inorganic and organometallic gases are utilized to create a graded interface between the substrate and the coating that is subsequently deposited by using pulsed-bias techniques. PIIP represents a significant advance since it allows coating deposition with all the strengths of the original PSII approach. Diamond-like carbon (DLC) and boron carbide are two such coatings that will be highlighted here for tribological applications.

Particle-in-cell and Monte Carlo simulation of the hydrogen plasma immersion ion implantation process

Hydrogen plasma immersion ion implantation into a 200-mm-diam silicon wafer placed on top of a cylindrical stage has been numerically simulated by the particle-in-cell (PIC) and transport-and-mixing-from-ion-irradiation (TAMIX) methods. The PIC simulation is conducted based on the plasma comprising three hydrogen species H+, H2+, and H3+ in a ratio determined by secondary ion mass spectrometry. The local sputtering losses and retained doses are calculated by the Monte Carlo code TAMIX. The combined effect of the three species results in a maximum retained dose variation of 11.6% along the radial direction of the wafer, although the implanted dose variation derived by PIC is higher at 21.5%. Our results suggest that the retained dose variations due to off-normal incident ions can partially compensate for variations in incident dose dictated by plasma sheath conditions. The depth profile becomes shallower toward the edge of the wafer. Our results indicate that it is about 34% shallower at the edge, but with...

Initial operation of a large‐scale plasma source ion implantation experiment

In plasma source ion implantation (PSII), a workpiece to be implanted is immersed in a weakly ionized plasma and pulsed to a high negative voltage. Plasma ions are accelerated toward the workpiece and implanted in its surface. A large‐scale PSII experiment has recently been assembled at Los Alamos, in which stainless steel and aluminum workpieces with surface areas over 4 m2 have been implanted in a 1.5 m diam, 4.6 m length cylindrical vacuum chamber. Initial implants have been performed at 50 kV with 20 μs pulses of 53 A peak current, repeated at 500 Hz, although the pulse modulator will eventually supply 120 kV pulses of 60 A peak current at 2 kHz. A 1000 W, 13.56 MHz capacitively coupled source produces nitrogen plasma densities in the 1015 m−3 range at neutral pressures as low as 0.02 mTorr. A variety of antenna configurations have been tried, with and without axial magnetic fields of up to 60 G. Measurements of sheath expansion, modulator voltage and current, and plasma density fill‐in following a pu...

Recent advances in plasma source ion implantation at Los Alamos National Laboratory

Abstract Plasma source ion implantation (PSII) is an environmentally benign, potentially cost-effective alternative to conventional lineof-sight, accelerator-based implantation and wet-chemical plating processes. PSII offers the potential of producing a high dose of ions in a relatively simple, fast and cost-effective manner, allowing the simultaneous implantation of large surface areas (many square meters), complex shapes and multiple components. The dynamics of the transient plasma sheath present during PSII have been modeled in both 1 1/2-D and 2 1/2-D (one or two spatial dimensions, plus time), and recent results from these efforts are compared with measurements of the uniformity of the implanted ion dose in complex configurations. Ammonia gas (NH 3 ) has been used as a nitrogen source for PSII processing of electroplated hard chromium. A retained dose of 2.2 × 10 17 N atoms cm −2 has been demonstrated to increase the surface hardness of the electroplated Cr by 24%, and decrease the wear rate by a factor of four, without any evidence of increased hydrogen concentration in the bulk material. By adjusting the repetition rate of the applied voltage pulses, and therefore the power input to the target, controlled, elevated temperature implantations have been performed, resulting in enhanced diffusion of the implanted species with a thicker modified surface layer. Experimental work has been performed utilizing cathodic arcs as sources of metallic ions for implantation, and preliminary results of this work are given. The area of ion-beam-assisted deposition (IBAD) has been explored utilizing PSII, with large surface area diamond-like carbon (DLC) layers being generated which can exhibit hardnesses in excess of 20 GPa.

Magnetic insulation of secondary electrons in plasma source ion implantation

The uncontrolled loss of accelerated secondary electrons in plasma source ion implantation (PSII) can significantly reduce system efficiency and poses a potential x‐ray hazard. This loss might be reduced by a magnetic field applied near the workpiece. The concept of magnetically insulated PSII is proposed, in which secondary electrons are trapped to form a virtual cathode layer near the workpiece surface where the local electric field is substantially reduced. Subsequent electrons that are emitted can then be reabsorbed by the workpiece. Estimates of anomalous electron transport from microinstabilities are made. Insight into the process is gained with multidimensional particle‐in‐cell simulations.

APPLICATION OF PARTICLE-IN-CELL SIMULATION TO PLASMA SOURCE ION IMPLANTATION

The powerful numerical technique of particle‐in‐cell simulation has been used to study sheath formation and dynamics of plasma source ion implantation (PSII). Two‐dimensional cylindrical calculations permit us to study the process when sheath dimensions are large compared to feature scales of the implanted object, where conformality is not assured. Plasma chambers as large as the PSII system at Los Alamos have been modeled. Densities of 1014–15 m−3 are initialized in the numerical configuration. Voltages of −50 to −100 kV have been modeled, with rise times on this pulse of 0.1–1.0 μs. Calculations of the ambient electron/N+2 ion plasma are presented. The implantation flux and dose on a variety of different shapes, including long cylinders and bores in flat plates, have been investigated. Finally, magnetic fields have been added to the calculations. The effect of this field on the process is measured with and without the effect of secondary electron emission from the surface.

Plasma source ion implantation of metal ions : synchronization of cathodic-arc plasma production and target bias pulses

Abstract An erbium cathodic-arc has been installed on a plasma source ion implantation (PSII) experiment to allow the implantation of erbium metal and the growth of adherent erbia (erbium oxide) films on a variety of substrates. The operation of the PSII puiser and the cathodic-arc are synchronized to achieve pure implantation, rather than the hybrid implantation/deposition being investigated in other laboratories. The relative phase of the 20 μs PSII and the cathodic-arc pulses can be adjusted to tailor the energy distribution of the implanted ions and suppress the initial high-current drain on the pulse modulator. We present experimental data on this effect and make a comparison with the results from particle-in-cell simulations.

Stochastic electron heating in a capacitive RF discharge with non-Maxwellian and time-varying distributions

In capacitively coupled radio frequency discharges, the electrons gain and lose energy by reflection from oscillating, high voltage sheaths. When time-averaged, this results in stochastic heating, which at low pressure is responsible for most of the electron heating in these discharges. Previous derivations of stochastic heating rates have generally assumed that the electron distribution is a time-invariant, single-temperature Maxwellian, and that the sheath motion is slow compared to the average electron velocity, so that electrons gain or lose a small amount of energy in each sheath reflection. Here we solve for the stochastic heating rates in the opposite limit of fast sheath motion and consider the applicability of the slow and fast sheath equations in the intermediate region. We also consider the effect of a two-temperature Maxwellian distribution on particle balance and the effect of a time-varying temperature on the heating rates and densities. >

Characterization and performance of diamond-like carbon films synthesized by plasma- and ion-beam-based techniques

Diamond-like carbon (DLC) films have been deposited on dissimilar substrates using three different deposition processes. Two well-studied deposition methods, employing cathodic arcs and r.f. plasma self-bias, have been used. These two processes differ in that cathodic arc processes use gas pressures less than 1 mTorr (0.13 Pa) and deposit atomic ions with energies less than 100 eV, while r.f. plasma self-bias processes use gas pressures greater than 1 mTorr (0.13 Pa) and deposit molecular ions with energies greater than 100 eV. In addition, DLC films have been deposited using a new plasma-based, pulsed-bias process. The pulsed-bias process uses gas pressures greater than 1 mTorr (0.13 Pa) and deposits molecular ions with energies greater than 1000 eV. Both the self-bias and the pulsed-bias processes utilized hydrocarbon gases as the carbon source. Cathodic arc processes generally rely on the arc between two graphite electrodes as the carbon source. Deposited films from all three processes have been characterized using ion backscattering techniques, elastic recoil spectrometry, transmission electron microscopy (TEM) and selected-area diffraction. Films deposited using the cathodic arc process are virtually hydrogen free while the self-bias and pulsed-bias films contain up to 40% H. TEM results indicated that the films are homogeneous and amorphous. The hardness, elastic modulus, coefficient of friction and wear rate of the films are also reported.