The Czochralski method for growing single-crystal semiconductors | Hacker News Reader

The Czochralski method for growing single-crystal semiconductors

spectrum.ieee.org

76 pointssohkamyung14 comments

If you think this is neat, you'll probably also like the "grain selector" technique they use to cast single crystal turbine blades. This is useful for the blades operating in the hottest parts of the turbine, and helps avoid creep deformation.

To do this, they add a little spiral to the end of the wax turbine blade pattern before making the investment casting.

The mold is then filled with molten metal, and the entire casting is moved very slowly through a temperature controlled furnace with a hot and a slightly-less-hot zone. The metal solidifies at the interface between those two zones, and as the solidified "front" passes through the spiral, one crystal orientation ends up winning out, and then the remainder of the turbine blade solidifies into a crystal with that orientation.

There's not a lot of public information about this process (lots of trade secrets), but it's super cool.

https://www.americanscientist.org/article/each-blade-a-singl...

https://www.youtube.com/watch?v=zkUN9ch07dM

Apparently they use a somewhat similar dual-furnace technique to grow single crystal boules of III-V semiconductors like GaAs:

https://en.wikipedia.org/wiki/Bridgman%E2%80%93Stockbarger_t...

That was a fascinating article indeed--thanks.

I wonder if you could do the same thing with water ice for clear ice cubes.

During my school years every child in Poland knew that name because growing a salt crystal using this method was part of the 6th grade physics curriculum.

I failed because I procrastinated and apparently you can't grow such a crystal overnight.

Wow!!! What a great lab exercise!

Tell me, please: how were the students supposed to reach the melting point of the salt? It seems... too high for me to try this at home.

I think it refers to growing a salt crystal out of a super-saturated saltwater at room temperature.

Pictures of fully grown silicon crystal ingots always impresses me:

- https://ars.els-cdn.com/content/image/3-s2.0-B97801280358180...

That narrow section at the top is where the crystal started growing from, and holds the entire ingot during the process, while it's being pulled up from the molten silicon and crystallized as it cools down.

The first method for refining high-purity silicon was discovered by Carl Marcus Olson.

https://en.wikipedia.org/wiki/Carl_Marcus_Olson

His undergraduate degree was from Augustana College. My evening walks take me through the grounds quite often.

http://smithsonianchips.si.edu/olson/article.htm

p.s. The quartz vessels also have their own story... https://www.wired.com/story/book-excerpt-science-of-ultra-pu...

I think the author is using 'silica' and 'silicon' interchangeably in the fifth and sixth paragraphs, ie:

> boron or phosphorus, could be added to the molten silica in precise amounts to change the silica’s carrier concentration. Depending on what dopants he added, the silica turned into p-type or n-type silicon

In almost every case the author probably means silicon since this is a method used for growing semiconductor crystals, not quartz crystals.

Jan Czochralski had also interesting life. Polish, born and educated in Germany, moved to Poland after independence in 1925. During WW2 he had been allowed by Germans to run his laboratory (and had been helping Resistance at the same time). For that readon there was no place for him under Communism. He run a small cosmetics lab in his homerown until his death in 1953.

[] https://en.m.wikipedia.org/wiki/Jan_Czochralski

Something that I've never found a satisfactory answer to when I've read about this method before has been the source for seed crystals. Where did the 'original' seed crystal come from in this process? I suppose once a new crystal with proper structure fabricated, you could use the newly created crystal as seed for further creations. Usually this was glanced over in my textbooks on silicon fab while in college.

The article states the process was discovered

> while investigating the crystallization rates of metal, Czochralski dipped his pen into molten tin instead of an inkwell. That caused a tin filament to form on the pen’s tip

And later, describes the process for silicon

> Once the silicon melted, he placed a small piece of polycrystalline material—a seed crystal

It seems then, that the seed crystal is not anything special. A typical piece of metal (such as a the pen tip) is made of up multiple single-crystal "grains" with non-crystalline "grain boundaries" between them; this is a polycrystalline material.

The question then is, why does only a single crystal form, rather than multiple crystal filaments oriented at different directions according to whichever grain contacted the starting point of the filament...

> The question then is, why does only a single crystal form, rather than multiple crystal filaments oriented at different directions according to whichever grain contacted the starting point of the filament...

The silicon wants to be in the lowest energy state and it does so by forming a face centered cubic crystal (diamond structure). The formation of crystals depends on the cooling rate. If matter cools to fast, it can't form a crystal structure.

If the start of crystallization is sufficiently difficult, growth begins with only one random grain and when the other grains start growing, a significant time later, they find themselves shut off by previous growth.

Reply on news.ycombinator.com