In this blog post, we’ll explore the principles behind how MOSFET transistors enabled the miniaturization of electronic devices and what that means.
These days, no matter where you go, it’s easy to spot people using smartphones. People browse the internet or watch videos while walking down the street, and they can make payments with their smartphones even without a credit card or cash. Compared to the era when computers were first invented, this represents an astonishing level of progress. The “ENIAC,” the first electronic general-purpose computer, was a massive machine measuring about 30 meters in length and weighing approximately 30 metric tons, occupying a vast amount of space. However, compared to today’s smartphones, its performance was actually far inferior. So how did computers, which were once as massive as dinosaurs, evolve into the small, fast smartphones we have today?
The key lies in the advancement of transistor technology. Early electronic devices primarily used BJTs (bipolar junction transistors). While BJTs were excellent components for amplifying current or controlling its flow, they had limitations when it came to increasing integration density to miniaturize electronic devices. Later, with the advent of the MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor), it became possible to integrate semiconductors at a much higher density, enabling the miniaturization and high performance of electronic devices. Let’s now examine the structure and operating principles of the MOSFET and explore how it made the miniaturization of electronic devices possible.
A MOSFET consists of three terminals: the gate, source, and drain. Beneath the gate is a very thin oxide layer that prevents electrons from passing through directly. In a typical n-channel MOSFET, the source and drain are made of n-type semiconductor material, while the substrate surrounding them is made of p-type semiconductor material.
So, what exactly are n-type and p-type semiconductors? To understand this, we must first understand the bonding structure of silicon, the basic material of semiconductors. Atoms consist of a positively charged nucleus at the center and electrons orbiting around it. Silicon has four valence electrons, which it shares with neighboring silicon atoms to form stable bonds.
In other words, a single silicon atom bonds with four neighboring silicon atoms by sharing electrons with each of them. However, in a material composed solely of pure silicon, electric current does not flow easily. For current to flow, electrons that can move freely are needed, since covalently bonded electrons cannot move freely. Therefore, silicon undergoes a process called doping, in which very small amounts of trivalent or pentavalent elements are added. Adding a 5-valent element creates excess electrons, resulting in an n-type semiconductor that conducts electricity, while adding a 3-valent element creates “holes”—vacancies where electrons are missing—resulting in a p-type semiconductor that conducts electricity.
Now let’s examine how a MOSFET allows current to flow. A MOSFET is structured such that when a voltage is applied to the gate, current flows between the source and drain. This is easy to understand if you think of it as a water faucet.
The gate voltage acts like the handle of a water faucet. If the gate voltage is lower than a certain threshold voltage, no current flows between the source and drain. However, when the gate voltage exceeds the threshold voltage, electrons begin to accumulate beneath the oxide layer; once enough electrons have gathered, a channel—the conductive path connecting the source and drain—is formed. It’s similar to a tunnel being created between two islands. As electrons move through this channel, current flows. The higher the gate voltage, the better the channel forms, allowing more current to flow; similarly, the wider the MOSFET, the wider the pathway through which current can pass, enabling it to carry a larger current.
So how did MOSFETs make the miniaturization of electronic devices possible? The key lies in the structural differences between BJTs and MOSFETs. If we compare these two devices to cars, a BJT is like a high-performance car that is large and expensive to manufacture, while a MOSFET is like a car that, although its individual performance is somewhat lower, is small and easy to manufacture. In other words, while the BJT has advantages in terms of individual device performance, the MOSFET is very small, has low power consumption, and is much more suitable for manufacturing integrated circuits. Thanks to these characteristics, it has become possible to integrate more than billions of MOSFETs onto a single semiconductor chip, which has ultimately enabled much higher computational performance. Furthermore, as MOSFETs become smaller, more transistors can be packed into the same area, allowing electronic devices to become smaller and evolve more rapidly.
Looking at the numbers, the number of MOSFETs integrated onto a single semiconductor chip has increased exponentially over the past few decades. The latest smartphone processors incorporate over 100 billion transistors, and computational performance and data processing capabilities have also improved dramatically. Compared to the ENIAC, which performed about 5,000 operations per second, today’s smartphone processors can perform over a billion operations per second using multiple cores. Thus, the improvement in transistor density can be considered one of the most important driving forces behind the advancement of modern information technology.
People are now moving beyond smart devices toward a hyper-connected society that integrates artificial intelligence and the Internet of Things (IoT). With semiconductor chips embedded in various devices, information is accessible anytime and anywhere, and services such as telemedicine are gradually expanding. However, as the size of MOSFETs shrinks to the nanometer scale, they are facing challenges such as scaling limits, increased leakage current, power consumption, and heat generation. To overcome these technical limitations, new transistor technologies such as the Gate-All-Around (GAA) structure are being adopted, and research into new semiconductor materials and process technologies is actively underway. It will be exciting to see what further innovations semiconductor technology—which transformed computers from slow, cumbersome machines into smartphones that fit in the palm of your hand—will bring in the future.
