The following is an edited version of the David Turnbull Lectureship address, given by recipient Didier R. de Fontaine at the 1995 MRS Fall Meeting, De Fontaine received the lectureship “in recognition of fundamental contributions and insights in the fields of order/disorder phenomena in materials and computational techniques for phase diagrams.” De Fontaine is a professor in the Department of Materials Science and Mineral Engineering at the University of California—Berkeley and holds a joint appointment with the Lawrence Berkeley National Laboratory.

A plasma, commonly referred to as the “fourth state of matter,” is an ensemble of randomly moving charged particles with a sufficient particle density to remain, on average, electrically neutral. While their scientific study dates from the 19th century, plasmas are ubiquitous, comprising more than 99% of the known material universe. The term “plasma” was first coined in the 1920s by Irving Langmuir at the General Electric Company after the vague resemblance of a filamented glow discharge to a biological plasma.

Plasmas are studied for many reasons. Physicists analyze the collective dynamics of ions and electron ensembles, utilizing principals of classical electromagnetics, and fluid and statistical mechanics, to better understand astrophysical, solar, and ionospheric phenomenon, and in applied problems such as thermonuclear fusion. Electrical engineers use plasmas to develop efficient lighting, and high-power electrical switchgear, and for magneto-hydrodynamic (MHD) power conversion. Aerospace engineers apply plasmas for attitude adjustment and electric propulsion of satellites. Chemists, chemical engineers, and materials scientists routinely use plasmas in reactive ion etching and sputter deposition. These methods are commonplace in microelec tronics since they allow synthesis of complex material structures with submicron feature sizes. A substantial portion of the multi-billion-dollar market for tooling used to manufacture semiconductors employs some form of plasma process. When compared with traditional wet-chemistry techniques, these dry processes result in minimal waste generation. Plasmas are also useful in bulk processing—for example as thermal sprays for melting materials.

While the quest for controlled thermonuclear fusion dominated much of plasma research in the 1960s and 1970s, in the last 20 years it has been the application of plasmas to materials processing that has provided new challenges for many plasma practitioners. It is not surprising that the guest editors and several of the authors for this issue of MRS Bulletin come from a fusion plasma-physics background.

Chemically reactive plasma discharges are widely used to process materials. A plasma is a primarily electrically neutral collection of free charged particles moving in random directions. The simplest plasma consists of electrons and one kind of positive ions. This article deals primarily with plasma discharges, which are plasmas having the following features:

(1) They are driven electrically.

(2) Charged-particle collisions with neutral-gas molecules are important.

(3) There are boundaries at which surface losses are important.

(4) Ionization of neutrals sustains the plasma in the steady state.

One simple discharge consists of a voltage source that drives current through a low-pressure gas between two conducting plates or electrodes. The gas “breaks down” to form a plasma, usually weakly ionized—that is, the plasma density is only a small fraction of the neutral-gas density.

The plasmas used in materials processing present an enormous range of charged-particle densities n and of temperatures Te, Ti, and T for electrons, ions, and processing gas, respectively. High-pressure (atmospheric) discharges are in near-thermal equilibrium (Te ~ Ti ~ T ~ 0.1–2 eV). Plasma temperatures are usually given in equivalent electron-volt units: One eV is equivalent to 11600 K through the Boltzmann constant. As discussed in the article by Boulos and Pfender in this issue of MRS Bulletin, these thermal discharges have high densities n ~ 1014-1019 particles/cm3 and are mainly used as heat sources. Low-pressure (1 mTorr–10 Torr) discharges are not in thermal equilibrium (Te ~ 2–5 eV ≫ Ti ~ T) and have low densities n ~ 109–1012 particles/cm3. As discussed in several of the following articles, these discharges are used as miniature chemical factories in which feedstock gases are broken into positive ions and chemically reactive etchants, deposition precursors, etc., which then flow to and physically or chemically react at the surface of a substrate.

The ever-shrinking dimensions of microelectronic devices has mandated the use of plasma processing in integrated circuit (IC) factories worldwide. Today the plasma-processing industry has grown to over $3 billion in revenues per year, well in excess of predictions made only a few years ago. Plasma etching and deposition systems are also found throughout flat-panel-display (FPD) factories despite the much larger dimensions of thin-film transistors (TFTs) that are used to switch picture elements (pixels) on and off. Besides the use of plasma in etching and depositing thin films, other processes include the following: removal of photoresist remnants after development (descumming), stripping developed photoresist after pattern transfer (ashing), and passivating defects in polycrystalline material. Why are plasma processes so prevalent?

In etching, plasmas are used for high-fidelity transfer of the photolithographically defined pattern that defines the device or circuit. More generally, plasma provides the means to taper sidewalls. In Si processing, the sidewalls must be nearly vertical to obtain high density integration and faster performance. However in making FPDs, sidewalls are tapered to obtain uniform step coverage and reduce shorting. In deposition, plasmas are used to enable processing at low temperature. For both etching and deposition, only plasma processing provides an economically viable means for processing large area substrates: 300 mm for Si and more than 550 × 650 mm for FPDs. It is the ability to scale uniform reactant generation to larger areas that sets plasma apart from beam-based processes that might otherwise offer the desired materials modifications. The nonequilibrium characteristics of plasma further distinguish this processing method. Energetic electrons break apart reactant precursors while ions bombard the surface anisotropically.

As adaptable as polymeric materials are in their many applications to our daily lives, the need exists to tailor the polymer surfaces to provide even more flexibility in regard to their uses. Plasma treatments offer an unprecedented spectrum of possible surface modifications to enhance polymers, ranging from simple topographical changes to creation of surface chemistries and coatings that are radically different from the bulk polymer. Furthermore plasma treatments are environmentally friendly and economical in regard to their use of materials.

Plasma processing can be classified into at least four categories that often overlap. These are the following: (1) surface preparation by breakdown of surface oils and loose contaminates, (2) etching of new topographies, (3) surface activation by creation or grafting of new functional groups or chemically reactive, excited metastable species on the surface, and (4) deposition of monolithic, adherent surface coatings by polymerization of monomeric species on the surface. Key features of these processes will be briefly discussed, with a rudimentary introduction to the chemistries involved, as well as examples. Focus is placed on capacitively coupled radio-frequency (rf) plasmas (see Figure 1 in the article by Lieberman et al. in this issue of MRS Bulletin) since they are most commonly used in polymer treatment.

The marketplace is continually demanding better functionality in the surface and near-surface properties of metal components and tools. In the last 20 years, plasma and ion-assisted techniques have enabled new methods to substantially improve wear, corrosion, and fatigue resistance. In most cases, plasmas are used to enhance well-established processes such as thermochemical diffusion treatment and physical and chemical vapor deposition (PVD and CVD), but their operation in a plasma atmosphere has led to major improvements in their reliability and efficacy, as well as extending the range of materials and components that can be treated.

Plasma-immersion ion implantation (PIII) is an emerging technology for the surface engineering of semiconductors, metals, and dielectrics. It is inherently a batch-processable technique that lends itself to the implantation of large numbers of parts simultaneously. It thus offers the possibility of introducing ion implantation into manufacturing processes that have not traditionally been feasible using conventional implantation.

In PIII the part to be treated is placed in a vacuum chamber in which is generated a plasma containing the ions of the species to be implanted. The plasma based implantation system does not use the extraction and acceleration methods of conventional mass-analyzing implanters. Instead the sample is (usually) repetitively pulsed at high negative voltages (in the 2–300 kV range) to implant the surface with a flux of energetic plasma ions as shown in Figure 1. When the negative bias is applied to a conducting object immersed in a plasma, electrons are repelled from the surrounding region toward the walls of the vacuum chamber, which is usually held at ground potential. Almost all the applied voltage difference occurs across this region, which is generally known as a sheath or cathode fall region. Ions are accelerated across the sheath, producing an ion flux to the entire exposed surface of the work-piece. Because the plasma surrounds the sample and because the ions are accelerated normal to the sample surfaces, implantation occurs over all surfaces, thereby eliminating the need for elaborate target manipulation or masking systems commonly required for beam line implanters. Ions implanted in the work-piece must be replaced by an incoming flow of ions at the sheath boundary, or the sheath will continue to expand into the surrounding plasma.

Plasma densities are kept relatively low, usually between 108 and 1011 ions per cm3. Ions must be replenished near the workpiece by either diffusion or ionization since the workpiece (in effect) behaves like an ion pump. Gaseous discharges with thermionic, radio-frequency, or microwave ionization sources have been successfully used.

Surface-enhanced materials are obtained through PIII by producing chemical and microstructural changes that lead to altered electrical properties (e.g., semiconductor applications), and low-friction and superhard surfaces that are wear- and corrosion-resistant. When PIII is limited to gaseous implant species, these unique surface properties are obtained primarily through the formation of nitrides, oxides, and carbides. When applied to semiconductor applications PIII can be used to form amorphous and electrically doped layers. Plasma-immersion ion implantation can also be combined with plasma-deposition techniques to produce coatings such as diamondlike carbon (DLC) having enhanced properties. This latter variation of PIII can be operated in a high ionenergy regime so as to do ion mixing and to form highly adherent films, and in an ion-beam-assisted-deposition (IBAD)-like ion-energy regime to produce good film morphology and structure.

Over the past decade, researchers in Japan, Russia, and the United States have been investigating the application of intense-pulsed-ion-beam (IPIB) technology (which has roots in inertial confinement fusion programs) to the surface treatment and coating of materials. The short range (0.1–10 μm) and high-energy density (1–50 J/cm2) of these short-pulsed (t ≥ 1 μs) beams (with ion currents I = 5–50 kA, and energies E = 100–1,000 keV) make them ideal flash-heat sources to rapidly vaporize or melt the near-surface layer of targets similar to the more familiar pulsed laser deposition (PLD) or laser surface treatment. The vaporized material can form coatings on substrates, and surface melting followed by rapid cooling (109 K/s) can form amorphous layers, dissolve precipitates, and form nonequilibrium microstructures.

An advantage of this approach over laser processing is that these beams deliver 0.1–10 KJ per pulse to targets at expected overall electrical efficiencies (i.e., the ratio of extracted ion-beam energy to the total energy consumed in generating the beam) of 15–40% (compared to < 1% for the excimer lasers often used for similar applications). Consequently IPIB hardware can be compact and require relatively low capital investment. This opens the promise of environmentally conscious, low-cost, high-throughput manufacturing. Further, efficient beam transport to the target and excellent coupling of incident ion energy to targets are achieved, as opposed to lasers that may have limited coupling to reflective materials or produce reflecting plasmas at high incident fluence. The ion range is adjustable through selection of the ion species and kinetic energy, and the beam energy density can be tailored through control of the beam footprint at the target to melt (1–10 J/cm2) or to vaporize (10–50 J/cm2) the target surface. Beam pulse durations are short (≥ 1 μs) to minimize thermal conduction. Some disadvantages of IPIB processing over laser processing include the need to form and propagate the beams in vacuum, and the need for shielding of x-rays produced by relatively low-level electron current present in IPIB accelerators. Also these beams cannot be as tightly focused onto targets as lasers, making them unsuitable for applications requiring treatment on small spatial scales.

Thermal plasmas in which the plasma state approaches local thermodynamic equilibrium may be produced by high-intensity arcs (current I > 100 A) or by radio-frequency (rf) discharges at high-power densities and at pressures exceeding 10 kPa (0.1 atm). In such plasmas, electron and heavy-particle temperatures are identical, and particle densities are only a function of the temperature (chemical equilibrium).

Thermal-plasma technology has passed through a drastic transition stage from primarily space-related activities in the 1960s to a more and more materials-oriented focus in the 1980s and 1990s, and thermal-plasma processing of materials is now attracting increasing interest for industrial applications.

Thermal-plasma processing of materials however must be viewed in the context of much broader technology trends, some of which are visible today. There is no question that materials and materials processing will be one of the most important technical issues as we approach the turn of the century. This trend will not be restricted to the development of new materials but will also include the refining of materials, the conservation of materials (by hard facing, coating, etc.), and the development of new processing routes that are more energy-efficient, more productive, and less damaging to our environment. Thermal-plasma processing is already playing an important role in these developments. Its potential for developing new materials-related technologies is increasingly recognized, and many research laboratories all over the world are engaged in advancing the frontiers of our knowledge in this exciting field. As a result of these efforts, thermal-plasma processing has emerged as a powerful technology that is finding numerous industrial applications.