Reed Electronics GroupSemiconductor International
Welcome, Guest. ( Register / Log in here for extra features.)
News | Archives | Buyers Guide | Events | Newsletters | Webcasts  
Free magazine subscription
Info Channels
Emerging Tech
Wafer Processing
Yield Management
Inspection, Measurement & Test
Clean Processing
Materials Science
Semiconductor Packaging
Business/Financial Info
Market Research

Current Issue - Semiconductor International
Current Issue
Free magazine subscription
Sign up today for special features, including free e-letters, access to archives, and much more.
Tools and Services
Buyers Guide – SemiSource
In-Stat Research Store
Learning Center
RSS Feeds  RSS Feeds
Special Reports
Talk Back
Inside SI
About us
Contact Us
Editorial Calendar
Submit an Article
Submit a Press Release
Free Subscriptions

Reed Electronics
 Group Websites
Electronic Business
Electronic News
Test & Measurement
Reed Electronics Group

Powered By

How to Make a Chip

1. First, a 99.9999% pure silicon crystal ingot is sliced into thin wafers.
2. Polishing the wafer removes surface scratches and impurities, leaving a near-perfect base for building chips.
3. Portions of the silicon are chemically altered to create the source and drain regions of the transistor, which control the flow defined by photolithography, where the wafer is coated with a light-sensitive material called photoresist. Next, light is shined through a patterned mask onto a chip-size section of the wafer -- a process similar to printing a photograph from a negative. A machine called a stepper repeats this process for each chip on the wafer.
4. The exposed areas of the photoresist harden. During the development process, non-hardened photoresist is washed away.
5. Atoms of dopant materials -- such as boron or arsenic -- are forced into an area by ion bombardment in a process called doping, aand are "activated" by a thermal annealing step. The resist material blocks the dopants from entering any areas where they are not intended to be. After ion implantation, the hardened resist is stripped off, and the process is repeated for other types of dopants implanted in different areas. In subsequent steps, a similar patterning process is used, but the resist acts as an etch mask.
6. Next, the gate of the transistor is formed by depositing and patterning a layer of silicon dioxide (which forms the gate oxide) and then a layer of polysilicon, which is then heavily doped. This polysilicon gate acts as a "faucet" to turn the flow of electrons between the source and drain on and off.
7. The rest of the fabricaction steps involve forming the "wires" that connect the gate, source and drain of the transistors to one another and to the outside world. Layers of silicon dioxide -- a dielectric, or insulator -- are deposited on the wafer using chemical vapor deposition (CVD). During CVD processing, gases that contain atoms of the material to be deposited react on the heated wafer surface, forming a thin film of solid material. Metals, primarily aluminum, are deposited by the physical vapor deposition (PVD). During PVD -- also called sputtering -- gas ions accelerate toward a "target" of the material to be deposited. The ions chip off atoms of the target material, which fall and accumulate on the wafer.
8. Steps 3,4 and 5 are repeated to build up patterned layers of silicon dioxide, metals and other materials to complete the circuit design. A layer of conducting metal (usually aluminum) is deposited (CVD or PVD), exposed (photolithography), and etched to form tiny metal interconnects. Complex chips require several metal layers, with vertical connections between them called vias.
9. The wafer is cut up, or diced, to form chips. The chips are put in packages, and a wirebonder electrically connects the chips to the appropriate package pins or leads.

 Our Sponsors

Free magazine subscription
About SI    |    Editorial Opportunities    |    Editorial Calendar    |    Advertising    |    Register/Log In    |    Site Feedback   |   Free Subscriptions