Silicon doping

you do realize that device grade silicon needs to be 99.99999% pure. Then you normally dope it in a vacuum chamber. It's not something you can really do by hand.

I'm very skeptical about whether or not it can be done.

--

/ O O \\ \\_____/ FTB. For email, remove my socks.

In science it often happens that scientists say, 'You know that's a really good argument; my position is mistaken,' and then they actually change their minds and you never hear that old view from them again. They really do it. It doesn't happen as often as it should, because scientists are human and change is sometimes painful. But it happens every day. I cannot recall the last time something like that happened in politics or religion.

- Carl Sagan, 1987 CSICOP keynote address

I don't think you can. I'm struggling to remember how it all works. I think the initial doping is done by the people who make the ingot. Making an ingot is pretty involved. You would have a very high utility bill if you decided to make an ingot yourself. But maybe you could buy ingots.

So, let's say it is a P-type ingot. Now you have to create N-type pockets in order for it to be useful. I think there are two main ways to do this. (I'm trying to remember. It's all a bit hazy.) I believe the normal way is that they mask off areas which are to remain P, and leave the N areas unmasked. Then they heat the wafer and allow N-type dopant to diffuse in. Like I say, it is all kind of hazy.

There would then be additional rounds of masking and diffusing where you convert part of your new N-type well back into P-type material (that's your base). And then there are also metalization steps. Then you would have to somehow attach leads to the metalized areas. It just seems too painful, and it's not like would be able to make anything fancier than the stuff you can buy.

Also, I do remember that nasty chemicals are involved, and the process control details are critical. Time, temperature, concentration/pressure. Clean rooms.

--Mac

Well, you need a cleanroom of the class appropriate to the size of your structures. If you stick with 2N3055 size, that should be doable. Then a ultra high vacuum chamber with handlers. It is less a the problem of a thight box than of the gear inside. Since it shouldn't gaz out, the materials have to be carefully selected. Then you'd need a source of ultrapure silicon and a way to handle, saw, etch, [whatever] it. I guess the dopants could be added in the etching chamber that runs with Ar-Ions. Our local University, the biggest around here, aspiring to belong to the top league decided against Silicon, but for GaAs to keep the cost limited. I once heard a number, but forgot it.

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net

sci.skeptic?

Rub Arsenic on the surface with your hand ?:-)

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

"How can I perform doping of silicon manually? "

What dopant do you want to apply? What is the nature of the silicon substrate? In general, you need a way to apply a closely controlled concentration of the dopant onto the surface. Arsnic and phosphorus can be applied as a liquid. Boron is typically applied in a vacuum either through ion implantation or LPCVD. There are many texts that explain the theory.

You missed the smiley ??

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

You need either a particle accelerator or an extremely deadly gas (Arsine, Phosphine or Di-borine depending on you dopant type). Once the dopant is on the surface of the wafer you will need to activate the dopant with a high temp furnace (typically a drive-in is done at 950C-1100C). If you used the particle accelerator you will need to aneal the wafer to restore the smooth surface. If you only want to dope certain sections of the wafer then you will need a lithography system. You should be able to pick up used equipment for about $1million US. Or you can hire a fab to do the work for you.

Also, you will get a better response from sci.engr.semiconductors.

Dwayne

Wear gloves...arsenic is poison!!!

--

/ O O \\ \\_____/ FTB. For email, remove my socks.

In science it often happens that scientists say, 'You know that's a really good argument; my position is mistaken,' and then they actually change their minds and you never hear that old view from them again. They really do it. It doesn't happen as often as it should, because scientists are human and change is sometimes painful. But it happens every day. I cannot recall the last time something like that happened in politics or religion.

- Carl Sagan, 1987 CSICOP keynote address

You can't.

-- American democracy has sunk to a level where elections are considered no different from sports and the consequences go no further than the record books. -- The Iron Webmaster, 3433 nizkor

flying saucers a2

On Sat, 04 Jun 2005 09:47:36 GMT, Matt Giwer wroth:

I remember that back in the late 50's that Bell Labs was distributing science project kits to schools to promote education. Remember that this was back when the US was playing catch-up to the Russians.

One of the lits was put together to allow school kids to make silicon solar cells right in the classroom with ordinary stuff found there. The kit included silicon wafer slices, some chemicals, a 115 volt heating element similar to the ones used in small radiant room heaters, some asbestos sheets for insulation, some fine carbide sandpaper, and a list of instructions.

You built an oven from the heater and the asbestos sheets. The heater was a ceramic cylinder with nichrome wire coiled around the outside and an Edison screw base. The inside of the cylinder was open and you broke the silicon wafer into pieces small enough to fit inside.

The wafer pieces were dipped into a water slurry of the chemical, I forget exactly which chemical (probably something with phosphorous in it), and placed in the heater/oven to get red hot. The original wafer pieces were probably grown with an N or P dopant and the subsequent difusion created a complemental doping.

The wafers were allowed to cool and then the carbide was used to remove the surface on one side of the wafer to get back down the original silicon. I forget exactly how the wires were added to each side, probably a loose flat spiral of bare copper held in contact mechanically. When finished the kids had a working solar cell.

My brother was given the kit by his science teacher to put together on his own for "extra credit". I suspect the teacher just wasn't up to the task of using the kit the way it was intended. I got the kit and played around with it.

So, in short, you CAN manually dope silicon without Billions of dollars of equipment. I know because I've done it.

Jim

Thanks, Jim, for that post. I had started to respond similarly here, with my own twists and then read what you wrote -- with more detail than I remembered, actually.

I think the kit that Bell Labs put out was back in the early-to-mid

1960's. I recently spoke with the people who currently own the rights to this kit (recently being a few years ago) and they are/were still selling it. Unfortunately, it would be very difficult for me to find the phone number, today -- it was in 2001 when I last had this information at hand. But they were on the east coast.

to the OP:

Modern, pure silicon wafers are rather cheap, thanks to the number of them being processed. I haven't directly purchased any, but someone at HP's Deer Creek facility told me they were only a "few dollars each" (my memory says the figure was near $4) when I pointed to the boxes and boxes of them they had laying around. This is sliced, polished, and cleaned so that no particles larger than a micron remained, if memory serves. Perhaps you could contact a fab or someone who might know someone at one who might help you get fragments of broken ones. I use such broken pieces (which I treat rather poorly) as convenient and well studied reflectors/filters, sometimes.

However you proceed, if it is based on choices you are making and not on someone else's well-thought-out design, you should research it well. If you are considering doing this "manually," then you should study how these things were done when there were no fabs and they were just experiments in laboratories. And run anything you come up with by a knowledgeable chemist before you try it. You will want an informed opinion about the risks and mitigations and advise about what to consider trying. Hot materials can easily produce noxious gases.

As a side bar, I've built ovens from spare parts that could easily raise a wafer to passivating temperatures, such as near 1500 C for rapid oxide growth, that sat in my garage. Just for testing some optical ideas, not for making wafers, though. Got plenty of free wafers from fabs, no problem. Used 'plenty' of dry nitrogen gas to fill the chamber (1500 C and open air with 20% oxy isn't so good an idea) and keep it at slightly higher pressure than ambient so oxygen doesn't get inside. Cripes, do things become a mess if you forget to run the nitrogen! You can construct quite a nice little "oven" of your own with a water cooled quartz jacket and some tungsten lamps from your local hardware or home construction store, a power controller, and a nickel plated chamber (much, much cheaper than gold and works "pretty good".) Just a tap water flow is usually enough. You don't want the quartz to get too hot, as it grows more opaque where it counts when it does and absorbs even more energy and melts in the resulting positive feedback. With a few kW of lamp energy and not enough cooling, bad things happen.

In any case, don't plan on building much other than a solar cell to start. And keep things really simple and as safe as you can, as you learn.

Jon

On Sat, 04 Jun 2005 21:25:38 GMT, Jonathan Kirwan wroth:

Of course, there's always the possibility of building point contact transistors out of silicon in the comfort of your garage or kitchen. With a flouride etchant to thin down a silicon wafer in the right spots and some indium solder you could probably make a passable junction transistor.

Jim

indium

Or you could buy one for $0.10

--

/ O O \\ \\_____/ FTB. For email, remove my socks.

In science it often happens that scientists say, 'You know that's a really good argument; my position is mistaken,' and then they actually change their minds and you never hear that old view from them again. They really do it. It doesn't happen as often as it should, because scientists are human and change is sometimes painful. But it happens every day. I cannot recall the last time something like that happened in politics or religion.

- Carl Sagan, 1987 CSICOP keynote address

for

Then you have the hard part of the job in the delivered kit.

of

it.

The problem is with the word _manually_. Obviously anything that can be done by a machine can be done by hand. Obviously with enough hardware for doing it and testing it and enough experimentation one can learn to get the results desired. However anyone who has looked into the process would know the right answer and not ask the question.

The question appears in line with the kit you mention. Is there a simple way to do it? is more like the question sounds and the answer is no. And with the kit efficiency was likely so low that a high impedance voltmeter was needed and no way to measure the current short of lab equipment the efficiency would be so low compared to even the cheapest commercial production.

So the answer is also no if the intention is a "try and see" bright idea. The idea might be sound but the success so low as to be unmeasurable. So again the answer is no. And again anyone who knew the difficulties involved would not ask the question.

--
America was made safer by turning the entire Muslim
world against America. As any Republican.
	-- The Iron Webmaster, 3449
  nizkor http://www.giwersworld.org/nizkook/nizkook.phtml
  http://www.giwersworld.org

indium

But it is far more difficult than finding a sweet spot on a gallium crystal for a radio. You can go through a lot of wafers before making one with little enough leakage to measure the effect. It took Schokley and company a while to get an effect that could not be attributed to measurement error beyond getting the effect in the first place.

--
Zionist refusal to admit Palestinians are people whose
ancestors converted from Judaism to Islam does not
help to bring peace.
	-- The Iron Webmaster, 3445
  nizkor http://www.giwersworld.org/nizkook/nizkook.phtml
  commentary http://www.giwersworld.org/opinion/running.phtml a5

plus $4 shipping and handling

--
When there is knowledge there is no need for belief.
	-- The Iron Webmaster, 3431
  nizkor http://www.giwersworld.org/nizkook/nizkook.phtml
  http://www.giwersworld.org