Lighting – Light Emitting Diodes [Part 2 – Fabrication]

Light Emitting Diodes [LEDs] have been one of the top hits in the 20th century in turns of revolutionary inventions. And it has been only recently that their commercial feasibility has been tweaked enough for them to be popularized in many applications from standard lightbulbs to car headlamps. And there’s a really good (and quite complex) reason for this.

When you look up “how an LED works” either through the internet, through a book, or a basic science teacher; they will probably give you the following image.

Note: To make the description easier, I did not use the term “current” as it represents the opposite direction of the flow of electrons. Turns out that someone (*Cough*Franklin*Cough) thought that current was the flow of positively charged particles, and (unfortunately) that convention stuck. XKCD shares my frustration in the following comic:


Electrons enter into a higher energy orbital in a “doped semiconductor” and drops in energy as it transitions into a low energy orbital in another similar material before progressing out of the system. A practical LED will NEVER look like this for a few reasons.

1) Semiconductors are given such a name because they do resist the flow of electrons significantly more than a metal will. A practical design will decrease the amount of distance an electron travels in the doped materials. Light is produced at the N/P boundary, which limits an LED’s active thickness area to a couple microns or less (depending on the doping differential).

2) Micron-scale manufacturing is best done from a top-to-bottom style of manufacturing. Practical design will exhibit vertical layering.

Thus, a practical (yet still simplified) LED design will look more like the image shown below.



A brief description of the fabrication process:

  1. Obtain a crystalline semiconductor wafer. Wafer processing is fascinating, as it is the pinnacle of material purification. From my experience, nothing else has been perfected to reach 99.9999999%. Describing this process is worthy of its own blog. For LEDs, the most common material is Gallium Arsenide (GaAs).
  2. The wafer is chemically altered through doping. Doping involves placing your wafer in close proximity to a elemental disc, putting them into an oven, and cranking up the heat (like >1000 C)! The elements in the discs will “boil” off and penetrate into the nearby silicon wafer. Hotter temperatures = deeper doping depth.
  3. Finish with metal contacts. The most unique step involves creating the metal arc contact above the doped conductor through “Wire bonding.”
  4. Protect your LED with a thick slab of plastic coating. The plastic can either be molded to focus the light or manufactured with scattering particles to do just the opposite (uniformly spread the light out).

Many modern LEDs have many more layers than the one described above and include many other unique manufacturing steps, including chemical vapor deposition (cheap, rough material coating), molecular beam epitaxy (costly, fine material coating), dry etching (isolate individual LEDs), and metal sputtering (for conductive contacts for the circuit).

There are lots of neat videos on Youtube that are worth checking out. Particularly inspiring is how fast automatic assembly lines can process and test individual LEDs including packaging and wire bonding.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s