Electronic Materials Production

Summary

While the term “electronic materials” commonly refers to the silicon-based materials from which computer chips and integrated circuits are constructed, it technically includes all materials upon which the function of electronic devices depends. This includes the plain glass and plastics used to house the devices to the exotic alloys and compounds that make it possible for them to function. Production of many of these materials requires not only rigorous methods and specific techniques but also high-precision analytical methods to ensure the structure and quality of the devices.

Definition and Basic Principles

Electronic materials are those materials used in the construction of electronic devices. The major electronic material today is the silicon wafer, from which computer chips and integrated circuits (ICs) are made. Silicon is one of a class of elements known as semiconductors. These materials do not conduct electrical currents appreciably unless acted upon or “biased” by an external voltage. Another such element is germanium.

The construction of silicon chips requires materials of high purity and consistent internal structure. This, in turn, requires precisely controlled methods in the production of both the materials and the structures for which they are used. Large crystals of ultrapure silicon are grown from molten silicon under strictly controlled environmental conditions. Thin wafers are sliced from the crystals and then polished to achieve the desired thickness and mirror-smooth surface necessary for their purpose. Each wafer is then subjected to a series of up to five hundred, and sometimes more, separate operations by which extremely thin layers of different materials are added in precise patterns to form millions of transistor structures. Modern CPU (central processing unit) chips have between 10⁷ and 10⁹ separate transistors per square centimeter etched on their surfaces in this way.

One of the materials added by the thin-layer deposition process is silicon, to fill in spaces between other materials in the structures. These layers must be added epitaxially in a way that maintains the base crystal structure of the silicon wafer.

Other materials used in electronic devices are also formed under strictly controlled environmental conditions. Computers could not function without some of these materials, especially indium tin oxide (ITO) for what are called transparent contacts and indium nitride for light-emitting diodes in a full spectrum range of colors.

Background and History

The production of modern electronic materials began with the invention of the semiconductor bridge transistor in 1947. This invention, in turn, was made possible by the development of quantum theory and the vacuum tube technology with which electronic devices functioned until that time.

The invention of the transistor began the development of electronic devices based on the semiconducting character of the element silicon. Under the influence of an applied voltage, silicon can be induced to conduct an electrical current. This feature allows silicon-based transistors to function somewhat like an on-off switch according to the nature of the applied voltage.

In 1960, the construction of the functional laser by American physicist and Nobel laureate Arthur Schawlow began the next phase in the development of semiconductor electronics, as the assembly of transistors on silicon substrates was still a tedious endeavor that greatly limited the size of transistor structures that could be constructed. As lasers became more powerful and more easily controlled, they were applied to the task of surface etching, an advance that has produced ever smaller transistor structures. This development has required ever more refined methods of producing silicon crystals from which thin wafers can be cut for the production of silicon semiconductor chips, the primary effort of electronic materials production (though by no means the most important).

How It Works

Melting and Crystallization. Chemists have long known how to grow large crystals of specific materials from melts. In this process, a material is heated past its melting point to become liquid. Then, as the molten material is allowed to cool slowly under controlled conditions, the material will solidify in a crystalline form with a highly regular atomic distribution.

Modern-day molten silicon is produced from a material called polysilicon, which has been stacked in a closed oven. Specific quantities of doping materials such as arsenic, phosphorus, boron, and antimony are added to the mixture according to the conducting properties desired for the silicon chips that will be produced. The polysilicon melt is rotated in one direction (clockwise). Then, a seed crystal of silicon, rotating in the opposite direction (counterclockwise), is introduced. The melt is carefully cooled to a specific temperature as the seed crystal structure is drawn out of the molten mass at a rate that determines the diameter of the resulting crystal.

To maintain the integrity of the single crystal that results, the shape is allowed to taper off into the form of a cone, and the crystal is then allowed to cool completely before further processing. The care with which this procedure is carried out produces a single crystal of the silicon alloy as a uniform cylinder, whose ends vary in diameter first as the desired extraction rate is achieved and then due to the formation of the terminal cone shape.

Wafers. In the next stage of production, the nonuniform ends of the crystal are removed using an inner-diameter saw. The remaining cylinder of crystal is called an ingot, and is then examined by X-ray to determine the consistency and integrity of the crystal structure. The ingot then will normally be cut into smaller sections for processing and quality control.

To produce the rough wafers that will become the substrates for chips, the ingot pieces are mounted on a solid base and fed into a large wire saw. The wire saw uses a single long moving wire to form a thick network of cutting edges. A continuous stream of slurry containing an extremely fine abrasive provides the cutting capability of the wire saw, allowing the production of many rough wafers at one time. The rough wafers are then thoroughly cleaned to remove any residue from the cutting stage.

Another procedure rounds and smooths the edges of each wafer, enhancing its structural strength and resistance to chipping. Each wafer is also laser-etched with identifying data. They then go on to a flat lapping procedure that removes most of the machining marks left by the wire saw, and then to a chemical etching process that eliminates the marking that the lapping process has left. Both the lapping process and the chemical etching stage are used to reduce the thickness of the wafers.

Polishing. Following lapping and rigorous cleaning, the wafers move into an automated chemical-mechanical polishing process that gives each wafer an extremely smooth mirror-like and flat surface. They are then again subjected to a series of rigorous chemical cleaning baths, and are then either packaged for sale to end users or moved directly into the epitaxial enhancement process.

Epitaxial Enhancement. Epitaxial enhancement is used to deposit a layer of ultrapure silicon on the surface of the wafer. This provides a layer with different properties from those of the underlying wafer material, an essential feature for the proper functioning of the MOS (metal-oxide-semiconductor) transistors that are used in modern chips. In this process, polished wafers are placed into a programmable oven and spun in an atmosphere of trichlorosilane gas. Decomposition of the trichlorosilane deposits silicon atoms on the surface of the wafers. While this produces an identifiable layer of silicon with different properties, it also maintains the crystal structure of the silicon in the wafer. The epitaxial layer contains no imperfections that may exist in the wafer, and that could lead to failure of the chips in use.

From this point on, the wafers are submitted to hundreds more individual processes. These processes build up the transistor structures that form the functional chips of a variety of integrated circuit devices and components that operate on the principles of digital logic.

Applications and Products

Microelectronics. The largest single use of silicon chips is in the microelectronics industry. Every digital device functions through the intermediacy of a silicon chip of some kind. This is as true of the control pad on a household washing machine as it is of the most sophisticated and complex CPU in an ultramodern computer. Silicon chips are also used in smartphones and drones. They are commonly used in the automobile industry, particularly as the industry is pressured to make efficient electric vehicles.

Digital devices are controlled through the operation of digital logic circuits constructed of transistors built onto the surface of a silicon chip. The chips can be exceedingly small. In the case of integrated circuit chips, commonly called ICs, only a few transistors may be required to achieve the desired function.

The simplest of these ICs is called an inverter, and a standard inverter IC provides six separate inverter circuits in a dual inline package (DIP) that looks like a small rectangular block of black plastic about one centimeter wide, two centimeters long, and 0.5 centimeters thick, with fourteen legs, seven on each side. The actual silicon chip contained within the body of the plastic block is approximately 5 millimeters square and no more than 0.5 millimeters thick. Thousands of such chips are cut from a single silicon wafer that has been processed specifically for that application.

Inverters require only a single input lead and a single output lead, and so facilitate six functionalities on the DIP described. However, other devices typically use two input leads to supply one output lead. In those devices, the same DIP structure provides only four functionalities. The transistor structures are correspondingly more complex, but the actual chip size is about the same. Package sizes increase according to the complexity of the actual chip and the number of leads that it requires for its function and application to physical considerations such as dissipation of the heat that the device will generate in operation.

In the case of a modern laptop or desktop computer, the CPU chip package may have two hundred leads on a square package that is approximately four centimeters on a side and less than 0.5 centimeters in thickness. The actual chip inside the package is a very thin sheet of silicon about one square centimeter in size, but covered with several million transistor structures that have been built up through photo-etching and chemical vapor deposition methods, as described above. Examination of any service listing of silicon chip ICs produced by any particular manufacturer will quickly reveal that a vast number of different ICs and functionalities are available.

Solar Technology. There are several other uses for silicon wafer technology, and new uses are yet to be realized. Large quantities of electronic-grade silicon wafers are used in the production of functional solar cells, an area of application that is experiencing high growth, as nonrenewable energy resources become more and more expensive. Utilizing the photoelectron effect described by Albert Einstein in 1905, solar cells convert light energy into an electrical current. Three types are made, utilizing both thick (> 300 micrometers [μm]) and thin (a few μm) layers of silicon. Thick-layer solar cells are constructed from single-crystal silicon and from large-grain polycrystalline silicon, while thin-layer solar cells are constructed by using vapor deposition to deposit a layer of silicon onto a glass substrate.

Microelectronic and Mechanical Systems. Silicon chips are also used to construct microelectronic and mechanical systems (MEMS). Exceedingly tiny mechanical devices such as gears and single-pixel mirrors can be constructed using the technology developed to produce silicon chips. Devices produced in this way are by nature highly sensitive and dependable in their operation, and so the majority of MEMS development is for the production of specialized sensors, such as the accelerometers used to initiate the deployment of airbag restraint systems in automobiles. A variety of other products are also available using MEMS technology, including biosensors, the micronozzles of inkjet printer cartridges, microfluidic test devices, microlenses and arrays of microlenses, and microscopic versions of tunable capacitors and resonators.

Other Applications. Other uses of silicon chip technology include mirrors for X-ray beams, mirrors and prisms for application in infrared spectroscopy—as silicon is entirely transparent to infrared radiation—and the material called porous silicon, which is made electrochemically from single-crystal silicon and has presented an exceptionally varied field of opportunity in materials science.

Many other materials fall into the category of electronic materials. Some, such as copper, gold, and other pure elements, are produced normally and then subjected to zone refining and vapor deposition techniques to achieve high purity and thin layers in the construction of electronic devices. Many exotic elements and metallic alloys, as well as specialized plastics, have been developed for use in electronic devices. Organic compounds known as liquid crystals, requiring no extraordinary synthetic measures, are normally semisolid materials that have properties of both a liquid and a solid. They are extensively used as the visual medium of thin liquid crystal display (LCD) screens, such as in wristwatches, clocks, calculators, laptop and tablet computers, almost all desktop monitors, and flat-screen televisions.

Another example is the group of compounds comprising indium nitride, gallium nitride, and aluminum nitride. These produce light-emitting diodes (LEDs) that provide light across the full visible spectrum. Growing these LEDs on the same chip now offers a technology that could completely replace existing CRT (cathode ray tube) and LCD technologies for visual displays.

Social Context and Future Prospects

Moore's law has successfully predicted the progression of transistor density that can be inscribed onto a silicon chip. There is a finite limit to that density, however, and the existing technology is very near or at that limit. Electronic materials research continues to improve methods and products to push the Moore limit.

New technologies must be developed to make the use of transistor logic as effective and as economical as possible. To that end, there exists a great deal of research into the application of new materials. Foremost is adding substances in chips, including graphene, black phosphorus, transition metal dichalcogenides, and boron nitride nanosheets. Collectively, these substances are known as two-dimensional (2-D) materials because they are flat sheets that are only an atom or two thick. Using these materials, new products will be developed in the next decade, including infrared night-vision mode in smartphones and creating stretchable electronic nanocomposites.

Bibliography

Dumé, Isabelle. "New Semiconductor Cools Computer Chips." Physics World, 5 Aug. 2021, physicsworld.com/a/new-semiconductor-cools-computer-chips. Accessed 12 Mar. 2022.

Glavin, Nicholas R., et al. "Emerging Applications of Elemental 2D Materials." Advanced Materials, vol. 32, no. 7, 2020. Wiley Online Library, doi.org/10.1002/adma.201902743.

"Graphene Transistor Could Mean Computers That Are 1,000 Times Faster." Science Daily, 13 June 2017, www.sciencedaily.com/releases/2017/06/170613145144.htm. Accessed 31 Aug. 2018.

Khan, Aftab A. P., et al. Advances in Electronic Materials for Clean Energy Conversion and Storage Applications. Woodhead Publishing, 2023.

Mims, Christopher. "Graphene and Beyond: The Wonder Materials That Could Replace Silicon in Future Technology." The Wall Street Journal, 27 Mar. 2021, www.wsj.com/articles/graphene-and-beyond-the-wonder-materials-that-could-replace-silicon-in-future-tech-11616817603. Accessed 12 Mar. 2022.

Morrison, Geoffrey. "How Quantum Dots Could Challenge OLED for Best TV Picture." CNET, 22 Jan. 2018, www.cnet.com/news/how-quantum-dots-could-challenge-oled-for-best-tv-picture. Accessed 31 Aug. 2018.

Rezende, Sergio M. Introduction to Electronic Materials and Devices. Springer, 2022.

Zhang, Q., et al. “A Two-Wafer Approach for Integration of Optical MEMS and Photonics on Silicon Substrate.” IEEE Photonics Technology Letters 22 (2010): 269-271.

Zheng, Y., et al. “Graphene Field Effect Transistors with Ferroelectric Gating.” Physical Review Letters 105 (2010).