In this blog post, we’ll examine why silicon—among countless elements—was chosen as the primary material for semiconductors. We’ll also explore the process by which silicon acquired the structural properties suitable for semiconductors and compare it with other candidate materials, such as germanium and carbon, in terms of their quantitative, physical, and chemical properties.
With over 100 elements, why “Silicon Valley” of all places?
What comes to mind when you hear the name of the element silicon? Many people probably think of “Silicon Valley,” the high-tech industrial hub in the United States. As semiconductor companies flourished in and around the Santa Clara Valley in California, the region came to be known as “Silicon Valley.” In particular, because so many semiconductor companies established themselves there and silicon was widely used as the primary material for semiconductors, the name “silicon” came to symbolize the high-tech industry itself. But out of more than 100 elements, why did silicon, of all things, become the primary material for semiconductors? Couldn’t it have been “Oxygen Valley” or “Carbon Valley” instead? In this article, we will first examine which elements, based on their structural properties, are suitable candidates for semiconductor materials. Then, we will explore why silicon, among the elements possessing such characteristics, became the primary material by comparing its quantitative, physical, and chemical properties with those of other candidate materials.
Tetravalent Elements Suitable as Semiconductor Materials
To simplify the concept, a semiconductor is a material with electrical properties that lie between those of a conductor—which conducts electricity well—and an insulator—which conducts almost no electricity. Electric current is generated by the movement of charged particles—free electrons and holes—and the degree to which electricity flows is referred to as electrical conductivity. Semiconductors have the property that the degree of electrical conduction can be controlled by adding impurities or by altering conditions such as heat, light, or electric fields. Simply put, this means that while electricity does not flow well under normal conditions, the degree of electrical conduction can be altered when specific conditions are applied. Because of this characteristic, semiconductors play a crucial role in various electronic devices that require the control of electric current flow.
Among the many elements, tetravalent elements are particularly well-suited for exhibiting semiconductor properties. An atom can be viewed as a structure in which the atomic nucleus is surrounded by several electron shells; “tetravalent” means that the atom’s outermost electron shell contains four electrons. Atoms become relatively stable when the electrons in their outermost shell are in a stable configuration; for many elements, having eight electrons in the outermost shell is the stable state, which is known as the octet rule. Therefore, atoms of tetravalent elements, such as silicon, can share their outermost electrons with neighboring atoms to form four pairs of covalent bonds. In crystals formed in this way, since the electrons are bound to the bonds between individual atoms, there are not many electrons that can move freely. Therefore, crystals in their pure state do not conduct electricity well.
However, in real-world environments, the temperature is not 0 K (absolute zero), so thermal energy exists within the material. Electrons that absorb some of this thermal energy break free from the bonds between atoms and become able to move freely. Electrons that break free from these bonds and can move freely within the crystal are called free electrons. When an electron leaves its original position, it leaves behind an empty space; this is called a hole.
Although a hole itself does not move as an actual particle, it appears to move in the opposite direction as a result of a repeated process in which a neighboring electron fills the vacancy, only to be replaced by another electron. When free electrons and holes move within the crystal, an electric current flows. Therefore, while a crystal of a pure tetravalent element appears to conduct almost no electricity, a very small amount of current can flow as some electrons break free from their bonds when the temperature rises.
Adding small amounts of trivalent or pentavalent elements can significantly alter the electrical properties of a semiconductor. This intentional addition of small amounts of other elements to a semiconductor is called doping. For example, when arsenic (As), a pentavalent element, is added to silicon, one of arsenic’s outermost electrons is used to form a bond with a neighboring silicon atom, leaving one electron unaccounted for. Since this extra electron is not strongly bound to the bond, it can relatively easily become a free electron. Consequently, the number of free electrons capable of carrying current within the silicon crystal increases. Conversely, when boron (B), a trivalent element, is added, one electron is missing from the bond with the silicon atom, creating a vacancy. This vacancy is a hole, and as surrounding electrons fill this vacancy one by one, the hole appears to move, causing current to flow. In other words, adding a pentavalent element increases the number of free electrons, while adding a trivalent element increases the number of holes. This allows for the control of the type and number of particles that carry current within the semiconductor.Although electricity does not flow well in their pure states, the degree of electrical conductivity can be significantly controlled by adding small amounts of impurities; therefore, tetravalent elements are suitable as semiconductor materials. However, silicon is not the only tetravalent element. Looking at the periodic table, carbon, silicon, and germanium are representative tetravalent elements. There are also compound semiconductors formed by the combination of trivalent and pentavalent elements, such as gallium arsenide (GaAs), and those formed by the combination of divalent and hexavalent elements, such as zinc sulfide (ZnS). Although various types of semiconductor materials exist, silicon has been the most widely used in general semiconductor devices. So, why was silicon chosen among the many candidates?Silicon atoms share electrons with one another to form a crystalline structure with a specific pattern. Each silicon atom shares electrons with neighboring silicon atoms to form covalent bonds, and the repetition of these bonds results in the formation of a silicon crystal. In this state, since the electrons are bound within the atomic bonds, there are not many electrons that can move freely. Therefore, electricity does not flow easily. However, when thermal energy is applied to the silicon crystal, some electrons break free from their bonds and become able to move freely. The space left behind by the departing electron is a vacancy, known as a hole. Free electrons carry current by moving directly through the crystal, while holes appear to move as surrounding electrons sequentially fill the vacancies, thereby contributing to the flow of current.By doping silicon with small amounts of other elements, this movement of charge can be controlled more effectively. Adding arsenic (As), a pentavalent element, creates excess electrons that can participate in chemical bonds, thereby increasing the number of free electrons; adding boron (B), a trivalent element, creates vacancies where one electron is missing, thereby increasing the number of holes. Ultimately, depending on which elements are added to silicon and in what quantities, the number of free electrons and holes that carry current can be controlled, and these characteristics form the crucial foundation for the operation of semiconductor devices.
Silicon, the All-Rounder: Dethroning Germanium
In fact, germanium was used as a key material in early semiconductor devices. However, due to the various advantages of silicon, the primary material in the semiconductor industry gradually shifted from germanium to silicon. Let’s examine the reasons why silicon came to be used more widely than germanium from three perspectives. First, from a quantitative perspective, silicon had the advantage of being a very abundant element. Silicon is the second most abundant element in the Earth’s crust after oxygen, accounting for approximately 28% of the crust’s mass. Although it exists in nature primarily in the form of silicates or silicon dioxide—combined with oxygen—rather than as pure silicon, the ability to produce silicon on a large scale using this abundant raw material was a major industrial advantage. Furthermore, since even small amounts of impurities can significantly alter the electrical properties of semiconductors, semiconductor materials of very high purity are required to ensure predictable device operation. Silicon also had the advantage of allowing for the mass production and refinement of high-purity materials based on its abundant raw material supply. Therefore, in the semiconductor industry, which requires mass production, silicon was an attractive material in terms of cost-effectiveness and supply stability.
Silicon also possessed significant advantages in terms of its chemical properties. In particular, silicon reacts with oxygen at high temperatures to form a stable oxide layer called silicon dioxide (SiO₂). This oxide layer acts as an insulating film that does not conduct electricity well and also protects the silicon surface. Above all, the ability to form and control the interface between silicon and silicon dioxide relatively stably was crucial for semiconductor device manufacturing. In contrast, germanium was at a disadvantage in controlling the surface of devices because, unlike silicon, it struggled to form a stable oxide layer useful for semiconductor manufacturing. Thanks to this ability to form an oxide layer and the excellent properties of the silicon-silicon dioxide interface, silicon became a highly suitable material for fabricating integrated circuits.
Physical properties also played a decisive role in the selection of silicon. Since germanium has a smaller bandgap than silicon, the amount of charge generated by heat increases as the temperature rises, making it prone to higher leakage current. In contrast, because silicon has a wider bandgap than germanium, it can maintain stable electrical characteristics even at relatively high temperatures. Ultimately, silicon possessed a combination of advantages: quantitative advantages stemming from the ability to produce high-purity material based on abundant raw materials; chemical advantages, such as the formation of a stable oxide layer and ease of controlling surfaces and interfaces; and physical advantages, such as the ability to maintain relatively stable electrical properties even as temperatures rise. Coupled with advances in manufacturing technologies for refining high-purity silicon and growing single crystals, silicon established itself as the primary material in the semiconductor industry.
The Elimination of Carbon, the Final Candidate
This leaves carbon as the final candidate among the tetravalent elements. From a quantitative perspective, carbon is one of the primary elements that make up living organisms and is widely present on Earth. Carbon exists in various forms in nature, most notably graphite, which has a layered structure, and diamond, which has an extremely hard crystalline structure. In particular, graphene—a single layer of carbon atoms isolated from graphite—and carbon nanotubes have long attracted attention as materials for electronic devices due to their outstanding electrical and mechanical properties. However, because graphite is a highly conductive material, it is not suitable for use as a typical semiconductor material that precisely controls the flow of current, as silicon does. Although graphene also possesses excellent electrical properties, there are distinct technical challenges involved in creating devices that can easily turn the current on and off, as is the case with conventional silicon semiconductors. As for diamond, even setting aside its cost, the energy required to allow current to flow—that is, the bandgap energy—is very large, so under normal conditions, it exhibits properties closer to those of an insulator, with almost no current flowing.
Although its electrical properties can be altered through doping, significant difficulties in manufacturing and device fabrication prevent it from being used as a general-purpose electronic device material like silicon. Therefore, while carbon—in its various forms such as graphene, carbon nanotubes, and diamond—possesses exceptional properties, it has not established itself as a general-purpose semiconductor material capable of replacing silicon.
Silicon: Ascending to the Throne
As we have seen, among the various elements and compounds that can be used as semiconductor materials, silicon simultaneously met several conditions necessary for semiconductor manufacturing, including abundant raw materials, the potential for high-purity refinement, the ability to form stable oxide films, and excellent thermal properties. In particular, the stable interface between the silicon dioxide layer formed on the silicon surface and the silicon itself has served as a crucial foundation for controlling the surface of semiconductor devices and manufacturing integrated circuits. Furthermore, as silicon has been used in the semiconductor industry for a long time, manufacturing technologies for mass production—such as wafer production, crystal growth, and microfabrication—have advanced to a very high level. The combination of these various advantages has ensured that silicon remains the central material in the semiconductor industry to this day.Of course, silicon is not the only material used in semiconductors. Other elements, such as germanium, are still in use, and compound semiconductors like gallium arsenide (GaAs) also play important roles in fields requiring properties different from those of silicon. These materials are utilized in applications where specific electrical or optical properties superior to those of silicon are required. For example, germanium is used in high-speed, high-frequency devices and in silicon-germanium (SiGe) materials combined with silicon, while gallium arsenide is used in high-frequency devices and optoelectronic devices. In particular, since silicon has limitations in efficiently emitting or absorbing light, compound semiconductors play a crucial role in fields such as light-emitting devices and optical communications. As such, each semiconductor material has its own distinct advantages and disadvantages, and the most suitable material varies depending on the intended application. Nevertheless, silicon—which possesses multiple advantages simultaneously, including abundant raw materials, high-purity purification technology, a stable oxide layer, excellent thermal properties, and sophisticated manufacturing processes—remains the mainstream material in the semiconductor industry. Silicon became the representative semiconductor material that gives “Silicon Valley” its name not simply because it is abundant on Earth, but because its electrical properties are easily controllable, it is suitable for mass production, and it possesses the various physical and chemical properties necessary for semiconductor manufacturing.
