Epitaxy

Type of physical science: Epitaxy, Chemistry

Field of study: Chemical processes

Epitaxy involves the growth of one crystal on the surface of a dissimilar crystal through molecular interactions between crystals. This molecular interaction can be successfully harnessed into the creation of optoelectronic devices and integrated circuits with exacting optical and electrical quality and specificity.

Overview

Epitaxy involves the growth of one crystal on the surface of another crystal. The surface of one crystal provides, through its lattice structure, the preferred positions for the deposition of a second crystal. The two crystals do not necessarily have to have the same chemical composition. The orientation of the crystal being deposited follows the lattice network of the first crystal or substrate.

This process does not involve a chemical reaction between the two materials; rather, molecular forces between the atoms composing each crystal form the interaction between the different crystals. In this context, it is beneficial to describe how atoms and molecules bond and interact before discussing how crystals interact with other crystals in an epitaxial mechanism.

The molecules within a crystalline lattice interact on one another through one or more molecular forces, including hydrogen bonding, electrostatic interactions, and van der Waals forces. All crystalline lattices are unique in that their molecular structures are different from every other molecular structure. Every crystalline structure has a molecular unit that repeats throughout the entire crystal lattice. Just as a single phrase of music in a symphonic fugue will repeat over and over again, so will a "unit cell" repeat itself over and over again in a crystal. In other words, crystals are long-range repeats of a basic structural unit, and all crystal structures are uniquely different.

The symmetry evidenced in a crystal structure is essentially of three types: translational, rotational, and reflectional (or mirrored). Permutations and combinations of these three types of symmetry operations total to 230 possible symmetry groups. Every crystal structure has a symmetry representative of one of the continual three-dimensional symmetric repetition of some basic structural unit, the unit cell, which can be defined by a set of axial lengths and interaxial angles termed "lattice constants." Axial lengths and interaxial angles are measured through X-ray crystallography as the distance and angle between atoms composing a single unit cell.

As a pure crystal grows, molecules become arranged in the proper crystallographic orientation, form the necessary chemical contacts, and become an integral part of the crystalline lattice. Epitaxial growth involves the oriented overgrowth of one crystalline lattice onto another. Whereas, in the growth of a pure crystal, all unit cells have the exact same distance and angles between atoms (that is, the axial lengths and interaxial angles), epitaxial growth involves unit cells with slightly different axial lengths and interaxial angles. The two crystalline lattices are in molecular contact with each other, and the atomic arrays at the epitaxial contact zone between the two crystals must be both chemically and structurally compatible. The mechanical and electrical properties of polycrystalline materials (those composed of more than one crystal having undergone epitaxial growth) depend dramatically upon these boundaries between the two different crystals. The crystalline structure and orientation of the substrate determine how the epitaxial layer is deposited on the substrate.

Successful epitaxy is based on matching lattice distances between two crystals. Crystal interaction in epitaxial growth can be simply thought of as a hand fitting into a glove. If the spacing between the fingers (representing atoms) is exactly the same as the spacing between the fingers on the glove (also representing atoms), then there is a perfect match, and crystalline growth can occur. This situation is said to have "0 percent misfit." Few crystals share the same lattice distance. Aluminum arsenide and gallium arsenide have the same crystal structure and the same lattice parameters to within 0.1 percent misfit, and they therefore grow excellent crystals on each other.

Lattice distances, however, do not have to be near-perfect in order for crystalline growth to occur. Misfits between two crystalline lattice distances can be 5 percent, 10 percent, and in some cases even 15 percent without inhibiting crystalline growth. Though it may not be a perfect fit, a small hand can fit into a larger glove; similarly, two crystals not having an exact fit can still support epitaxial growth.

When the crystal and substrate are of the same material, the process is called "homoepitaxy." Silicon layers of different impurity content, for example, are grown on silicon substrates in the manufacture of computer chips.

When the crystal and substrate are different materials, the process is called "heteroepitaxy." Silicon is often used as a substrate on which epitaxial growth is to occur. Silicon substrates are readily available and anatomically smooth, making them ideal for the growth of other crystals such as gallium arsenide, germanium, cadmium tellurite, and lead tellurite in the production of semiconductors. Any flat substrate, however, can be used for epitaxy, such as rock salt (NaCl) and magnesium oxide.

Several kinds of solid-state electronic devices, such as high-speed transistors and integrated circuits, are fabricated by the production of an epitaxial layer of one semiconductor on a wafer of another that has a different type of chemical composition and, hence, a different electrical conductivity.

Several methods to achieve epitaxial growth have been developed. Molecular-beam epitaxy is an ultra-high-vacuum technique in which beams of atoms or moleculars of the constituent elements of the crystal provided impinge upon a heated substrate crystal. The amount of gas molecules can be controlled to grow one layer at a time or multiple layers at a time. This method is slow, since molecular beams have low densities of atoms, but it allows highly reproducible growth and very thin epitaxial layers, as thin as 0.5 nanometer.

Vapor-phrase epitaxy, also known as "chemical vapor deposition," is another form of epitaxy that makes use of the vapor-deposition technique of molecular-beam epitaxy. Vapor-phrase epitaxy, however, is much faster than molecular-beam epitaxy, since the atoms are delivered in a flowing gas rather than in a molecular beam. Synthetic diamonds are grown by vapor-phase epitaxy.

Liquid-phase epitaxy is the technique most closely related to the methods used for bulk crystal growth. The substrate crystal is placed into a super-concentrated solution of the material that is to undergo epitaxial growth on the substrate crystal. Growth by liquid-phase epitaxy is done in an apparatus in which the substrate wafer is sequentially brought into contact with solutions that are at the desired composition and may be supersaturated or cooled to achieve growth. Typically, structures grown by liquid-phase epitaxy, such as gallium arsenide, gallium aluminum arsenide, and gallium phosphide--all of which are used in modern electronics and optic devices--have four to six layers ranging widely in composition.

Applications

From wristwatches and computers to the electronic scanners used in most stores, barely an aspect of modern life is not touched by a product containing a semiconductor chip or optic device created using epitaxial techniques. As previously described, many innovative methods exploit epitaxial molecular processes that enable micro-manipulation of various substrates for the production of semiconductors for use as solid-state electronic devices. Thus, microchip technology for computers, lasers, and optical cable production often utilizes devices created by epitaxial techniques.

Semiconductors have been the backbone of the electronics industry for many years. Semiconductors did not enter the industry in a major way, however, until several years after the vacuum tube had been well established. In terms of perspective, it is interesting to note that at least one semiconductor device predated the vacuum tube in the early days of radio communication. This was the then-familiar galena crystal and accompanying whisker used in early crystal-set radio receivers.

In the mid-twentieth century, the Massachusetts Institute of Technology's Radiation Laboratory became active in the investigation of crystal rectifiers and engaged the development of very pure semiconducting materials, notably silicon and germanium. In 1947, the invention of the transistor by William Shockley, John Bardeen, and Walter H. Brattain at Bell Labs represented a major breakthrough in semiconductor technology, and the three researchers jointly were awarded the Nobel Prize in Physics in 1956.

The transistor had a tremendous impact, constituting the birth of modern electronics. The transistor led to the development of almost innumerable semiconductor device configurations and, ultimately, to the phasing out of the vacuum tube except in special and limited applications.

Innovations to fabricate semiconductors with improved performance and smaller size has been a continuing process. With the advent of each new fabricating technique has come a new class of integrated circuits.

It is only in recent years that serious concern has been expressed as to the possible physical limits on size and performance of these semiconductors. It is interesting to note that only a modest diminution in the size of electronic-circuit components is required before the scale of individual molecules is reached; in fact, many existing circuit elements could already be accommodated within the area occupied by a virus.

Thus far, advancements in fabrication (chip-making processes) such as electron-beam and molecular-beam epitaxy have allayed these concerns. Although some interest in nonsilicon material has always been present, there has been a reawakening of interest in the use of gallium, indium, arsenic, phosphorus, and antimony.

Nevertheless, researchers have begun looking into venturing out of the relative paucity of structures achievable with inorganic compounds and are looking to the richness and variety of organic materials available. Some now forecast a major shift from crystal electronics to molecular electronics and even supermolecular electronics, in which signal transport and control will be effected by atomic-scaled assemblies.

Principal terms

CRYSTAL: A homogenous solid that exhibits a regularly repeating atomic arrangement and symmetry; almost all pure elements and compounds are capable of forming crystals

LATTICE: A regular periodic arrangement of points in three-dimensional space

SUBSTRATE: A support material on which an integrated circuit is constructed or to which it is attached

SEMICONDUCTOR: A crystalline material having intermediate values of electrical resistivity; semiconductors are the basic material of various electronic devices used in telecommunications, computer technology, control systems, and many other applications

SYMMETRICAL: Having a shape or configuration in which parts that lie on opposite sides of an actual or imaginary central line are identical in size, shape, and structure; having atoms or groups at equal intervals in a molecule

TRANSISTOR: A solid-state electronic device, composed of semiconductor material such as silicon and germanium, that controls current flow without the use of a vacuum

Bibliography

Hartman, P. Crystal Growth: An Introduction. New York: American Elsevier, 1973. As the title implies, this book provides a discussion of the basics of crystallography.

Mandel, Gretchen, and Neil Mandel. "Crystal-Crystal Interactions." In Kidney Stones: Medical and Surgical Management, edited by F. L. Coe, et al. Philadelphia: Lippincott-Raven, 1996. While intended for a medical audience, this chapter provides a concise, clear discussion on crystallography and epitaxy for the nonchemist. Provides exceptionally interesting insight into the role epitaxy may play in kidney-stone formation, a common ailment.

Mathews, J. W. Epitaxial Growth. New York: Academic Press, 1975. Although this volume is somewhat dated, it discusses physical properties that are immutable. Requires some background in physics and chemistry, but should be intelligible to the general reader.

Pasahow, E. Electronics Ready-Reference Manual. New York: McGraw-Hill, 1986. A survey of basic electronics. Useful background for understanding the applications of epitaxy. Plenty of diagrams, easy-to-understand language.