Making a laser diode

What do you guys think it would take to make a laser diode, i know about semiconductor lasers, i looked around on the net and found 1 inch crystals of gallium arsenide, i think it would be a fun project but is it plausible to make one without having a high tech lab?

I know the active region in most diodes is no bigger than a few grains of sand, but would it work if scaled up proportionally?
I guess the hardest part would be achieving stable inverse population, without a chip, you could manage only pulsed laser non Continuous-Wave.


i cant find OC and HR mirrors anywhere tough...

Not possible short of a multimillion dollar investment. The construction of laser diodes does require very expensive equipment capable of depositing molecule-thin layers of material to form the resonant "cavity".

The OC and HR mirrors of a laser diode are the cleaved ends of the die itself. The transition point between the substrate and the air forms a reflective surface (kind of like what you see if you are underwater looking at the surface. The reflectivity is caused be the transition between water and air.).

ah i see, so the mirrors are the effect of simulated emission exchanging mediums.

Oh well, i guess ill have to find some other way.

Take a peek at some optical physics for the mathematics on refraction, even though it's not optical glass the math is the same. Pay particular attention to the NA information (numerical aperture). In laser systems from cavities to fiber coupling, NA reigns supreme.

And as far as making a diode itself, it only works because it's small. It won't work unless it's nano-scale. So you're looking at, as already said, millions of dollars worth of equipment.

The mirrors are fairly easy though, IF several things are planned for in advance. If you plan your laser such that the cleavage planes of the material line up with the laser die, you can just cleave the facets, and you're set. Many laser diodes nowadays though have etched facets (required another tool) or, even more difficult, they have facet coatings to make better mirrors (requiring another tool). But if you already have all the tools to make a laser die, you likely already have those tools.

But at the least, you need substrates (you mentioned GaAs, but for other colors you would need InP, GaN, sapphire, etc.), you need a tool for epitaxial growth (MBE or MOCVD/MOVPE, over a $million), and you need all the processing tools: photolithography application and exposure tools, wet etching chemicals and equipment, likely dry etching tools, film deposition, and annealing tools of various kinds.

Yeah, it's really complicated, and there's no real way to "scale up", it works because it's small, at least because the epitaxial layers are thin.

pullbangdead, thanks for the correction, post edited ;D

Making diodes requires growing a film layer of material on a substrate...and its no amateur act.
Just to give you a frame of reference look up "Molecular Beam Epitaxy" with google set to images...for an even more impressive scenary, set the images to show extra large images

Yeah, your wallets not going to be able to outsource that kinda capital

die4thing on the site die4lasers, has an instruction set for building a laser diode.

It is located here: http://www.die4laser.com/

However, he used an already made die (the part that requires millions of equipment). So, if you can get your hands on a die and have the equipment/skills shown on the site, then you could in theory build your own laser diode.

Illuminum3415 said:

Making diodes requires growing a film layer of monocrystalline silicon on a piece of monocrystalline substrate...and its no amateur act.
Just to give you a frame of reference look up "Molecular Beam Epitaxy" with google set to images...for an even more impressive scenary, set the images to show extra large images

Yeah, your wallets not going to be able to outsource that kinda capital

Click to expand...

Except it's not silicon. Computer chips are silicon, but LEDs and laser diodes are not silicon. Ever.

Silicon has an indirect bandgap, so it never gets used for things that emit light. There is some research out there about putting III-V semiconductors on silicon substrates, but even then the parts doing the work are not silicon, they're the other thing being put on silicon.

The materials used for laser diodes are III-V compound semiconductors, generally much more complicated than silicon in most every way.

And MBE is one of the ways to do epitaxial growth, most laser diodes mass produced, though, are made with MOCVD/MOVPE. But you're right that if you're looking for images, you should look at MBE, they're much more impressive to look at. MOCVD tools are like a big box, you can't see anything going on inside it, very boring to look at.

Illuminum3415 said:

Yeah, your wallets not going to be able to outsource that kinda capital

Click to expand...

You don't know me!   ;D

Yeah i get it, ill figure out something else...

If you want to build a laser completely from scratch, build a gas laser. It can be done with some pyrex, a few mirrors, a good vac pump and a lot of troubleshooting. It's the way to go if you want to build something from the ground up without spending millions. You can also build a DPSS laser, but then you're still not building the diodes. Or you could build a flashlamp pumped yag or something, but even then it's more like you're assembling pre-built components.

Check out sam's FAQ and build youeself a HeNe laser, fun stuff there!!!

It's no diode, but you could make a TEA laser from scratch.
Maybe, if you have the inclination, you could take the design and see how far you could miniturize it.

http://www.sparkbangbuzz.com/tealaser/tealaser7.htm

Reaver789 said:

[quote author=Illuminum3415 link=1233828326/0#5 date=1233883103]

Yeah, your wallets not going to be able to outsource that kinda capital

Click to expand...

You don't know me! ;D
[/quote]

no I don't, but I didn't say your account couldn't, did I?