How Semiconductors are Made?

Semiconductors are the backbone of modern electronics, enabling the functionality of devices like smartphones, computers, and solar cells. The process of manufacturing semiconductors is intricate and requires precision at each stage. This article delves into the detailed steps involved in semiconductor manufacturing, highlighting the technical methods employed to produce these essential components.

1. Raw Material Extraction and Purification

Silicon Extraction The most commonly used material in semiconductor manufacturing is silicon, which is derived from quartzite sand. Silicon dioxide (SiO?) is reduced in an electric arc furnace at high temperatures (around 1900°C) to obtain metallurgical-grade silicon. The reaction is as follows:

SiO2+2C→Si+2CO

This silicon contains impurities and must be further refined. Chemical Purification: The Czochralski Process The Czochralski process is a key method for purifying silicon to electronic-grade quality (99.9999% purity). In this process, polycrystalline silicon is melted in a quartz crucible. A single crystal seed is dipped into the molten silicon and slowly withdrawn while rotating. As it is pulled upwards, the silicon crystallizes in alignment with the seed, forming a cylindrical ingot of monocrystalline silicon.

2. Wafer Production

Ingot Slicing The cylindrical silicon ingot is sliced into thin discs, known as wafers, using a precision diamond saw in a process called wire sawing. The wafers are typically between 150mm to 300mm in diameter and about 0.5mm thick. Wafer Polishing and Cleaning The sliced wafers undergo a series of polishing steps to achieve a mirror-like finish on the surface. This process, known as chemical-mechanical planarization (CMP), uses a combination of mechanical forces and chemical reactions to remove surface irregularities and achieve the required flatness and smoothness. After polishing, the wafers are cleaned using a series of chemical baths, including RCA cleaning, which involves using solutions of hydrogen peroxide, ammonium hydroxide, and hydrochloric acid to remove organic contaminants and metals.

3. Photolithography: Patterning the Circuit

Photolithography is the core process for defining the intricate patterns of a semiconductor circuit on a wafer. Photoresist Application A light-sensitive material called photoresist is applied to the wafer surface via spin coating. The wafer is spun at high speeds (1000-5000 RPM), creating a thin, uniform layer of photoresist. Mask Alignment and Exposure A photomask, which contains the circuit pattern, is aligned with the wafer. The photomask is a glass plate with opaque regions corresponding to the circuit design. The wafer is exposed to ultraviolet (UV) light, typically from a deep ultraviolet (DUV) laser, through the mask. The light alters the chemical structure of the exposed photoresist. There are two types of photoresists:

• Positive Photoresist: The exposed areas become soluble and are removed during development, leaving the pattern of the mask on the wafer.
• Negative Photoresist: The exposed areas become insoluble, and the unexposed regions are removed, leaving the pattern.

Development After exposure, the wafer is developed by immersing it in a chemical developer that removes the soluble portions of the photoresist, revealing the underlying silicon.

4. Etching: Material Removal

Etching is the process of removing layers from the wafer to create the desired circuit patterns. Wet Etching In wet etching, the wafer is submerged in a liquid chemical etchant that selectively dissolves the exposed material. For instance, hydrofluoric acid (HF) is commonly used to etch silicon dioxide. Dry Etching Dry etching, or plasma etching, uses reactive gases (like CF?, SF?) in a plasma state to etch the material. The wafer is placed in a chamber where the gases are ionized, creating highly reactive species that attack the exposed areas of the wafer.

• Reactive Ion Etching (RIE): This method combines chemical etching with physical sputtering, providing greater precision and anisotropy (vertical etching), which is critical for defining fine features in semiconductor devices.

5. Doping: Modifying Electrical Properties

Doping introduces impurities into the silicon to alter its electrical properties, creating regions of n-type (electron-rich) and p-type (hole-rich) semiconductors. Ion Implantation In ion implantation, dopant atoms (like boron for p-type or phosphorus for n-type) are ionized and accelerated in an electric field towards the wafer. The ions penetrate the silicon lattice, embedding themselves in the substrate. The depth and concentration of the dopants are controlled by adjusting the ion energy and dose. Annealing The ion implantation process can damage the silicon lattice. Annealing is a high-temperature process (typically between 600°C and 1000°C) that repairs this damage and activates the dopants by allowing them to occupy substitutional lattice sites.

Learn more: How Semiconductors are Made?