Author Topic: Making a Memristor (a tutorial...sort of)  (Read 14275 times)

0 Members and 1 Guest are viewing this topic.

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Making a Memristor (a tutorial...sort of)
« on: November 08, 2013, 05:53:15 am »
Howdy!

I have been working on making some memristors so i can do some experiments on them.  I think i have finally managed to make a memristor.  If i understand properly, memristors are characterized by their pinched hysteresis loops, their ability to "remember" their resistance.


A memristor I-V curve

I thought that it would be a good idea to get some others opinions on weather or not i have actually made a memristor.

I will start with a quick explanation of my methods for creating a memristor.
This:

Is my memristor.
Anyone familiar with firearms will notice that it is built inside a 0.22 cal. casing that is now very blackened. don't mind all the other stuff on my bread board, it isn't relevant.
The black stuff is Copper Sulfide(i am terrible in chemistry so if someone out there knows what this really is please tell), this is the stuff that makes all the magic happen.  Before you blacken the cartridge you first have to tin the bottom where the firing pin hits the bullet because it is difficult to solder a wire onto the black stuff. 
Back to the Copper Sulfide.  It is created when i filled the empty cartridge with sulfur powder and heated it up until all the sulfur is burnt away, do this with safely the vapors from burning sulfur are horrible to breathe.
Next I strip about 1" of insulation off a piece of 22 AWG wire and wrap the uninsulated part around a small screwdriver 2 times to make a coil that just fits into the casing.  While keeping gentle pressure on the wire taking care to only let the wire touch the bottom of the casing i filled the remaining space with epoxy.  After the epoxy has dried your memristor is ready to go.




I have several I-V curves from my memristor:


It turns out making a memristor has been fairly hit and miss untill i perfected my method.  This is a typical trace of my early memristors.  Not quite the perfect figure 8, and it has two points where it crosses itself.  Alot of the time I would get a flat line (open circuit) or a vertical line (short circuit).

Ocasionally i would end up with some strange capacitor-like trace.


After lots of work, it all payed off and I got this trace

If I'm not mistaken that is the trace of an ideal memristor.

Out of curiosity i put a 100Hz square wave through my memristor and was rewarded with one of the most interesting traces I have ever seen.


So that's my memrisor.  Here is the whole gallery of images. http://imgur.com/a/ER89Y
For anyone looking to recreate this experiment, you can talk to your pharmasist and ask them to order you "Precipitated sulfur".  It will be less than $15CAD for half a kilo of it(enough for 1,000,000 memristors).
I look forwards to your thoughts on my project.
also First Post! yay.

I would like to credit this web-page with the basic idea of how to make this type of memristor.
http://makezine.com/2011/11/03/how-to-homemade-memristor/
« Last Edit: November 08, 2013, 08:03:41 am by Maximus »
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Making a Memristor (a tutorial...sort of)
« Reply #1 on: November 08, 2013, 08:28:53 am »
Interesting experiment, thanks for sharing !  :-+

Offline Whales

  • Super Contributor
  • ***
  • Posts: 1899
  • Country: au
    • Halestrom
Re: Making a Memristor (a tutorial...sort of)
« Reply #2 on: November 08, 2013, 10:14:48 am »
Indeed!   :-+  Plecos for sharing -- I'm going to have to try making my own now.  To the crusty sulphurous deposits!

Offline clifford

  • Regular Contributor
  • *
  • Posts: 64
  • Country: at
    • www.clifford.at
Re: Making a Memristor (a tutorial...sort of)
« Reply #3 on: November 08, 2013, 12:11:27 pm »
Wow! Really cool!

I've googled a bit but haven't found much about this type of memristor implementation. Can anyone explain how it is working? I'd assume it is some kind of electro-chemical process?
 

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #4 on: November 08, 2013, 03:13:05 pm »
I've googled a bit but haven't found much about this type of memristor implementation. Can anyone explain how it is working? I'd assume it is some kind of electro-chemical process?

I have access to quite a large database of scientific journals (especially electrical engineering related ones) and whatnot with my school.  I will give that a look-through today and start hunting for a reason for why these memristors work.
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline Tinkerer

  • Frequent Contributor
  • **
  • Posts: 346
Re: Making a Memristor (a tutorial...sort of)
« Reply #5 on: November 09, 2013, 12:34:41 am »
Whoa, its only been in the last couple years that they have been able to even take the memristor from concept and math to reality. To be able to do this on your own is impressive.
In regards their ability, yes, they are supposed to "remember" all previous resistances. Hence their name.
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8240
Re: Making a Memristor (a tutorial...sort of)
« Reply #6 on: November 09, 2013, 02:27:41 am »
This is impressive, on the level of making your own transistors :-+
 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7799
  • Country: us
  • adieu
Re: Making a Memristor (a tutorial...sort of)
« Reply #7 on: November 09, 2013, 02:31:58 am »
Anyone who can make their own components gets serious points from me. I've been meaning to try to make a transistor at some point, myself, but it's hard to find the time.
No longer active here - try the IRC channel if you just can't be without me :)
 

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #8 on: November 09, 2013, 06:42:33 am »
UPDATE:
It appears that the type of memristor i described is doomed to an early death.  This manufacturing process creates memristors that slowly decay into regular old resistors as they are used.  To make a more durable memristor we would need to expose a piece of copper to H2S for half a day.  Unfortunately H2S is super poisonous and likes to explode.  So until I can figure out a safe way use H2S I am stuck with my temporary memristors.

Sources:
Memristors based on an organic monolayer of molecules and a thin film of solid electrolytes
Islam, M.S. ; Johns, C. ; Long Do ; Ohlberg, D.A.A. ; Shih-Yuan Wang ; Williams, R.Stanley
retrieved from IEEE Xplore
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Making a Memristor (a tutorial...sort of)
« Reply #9 on: November 09, 2013, 08:19:28 am »
Apparently you didn't pay attention to the absolute maximum ratings from the manufacturer of that poor memristor.  :-DD

Seriously, if you're going to make new one again, please, do a complete step bt step documentation like photos on each progress, imo, that will be really helpful and much less trial & error for other people to follow your steps.

Great thread, bookmarked & subscribed.  :-+

Offline Len

  • Frequent Contributor
  • **
  • Posts: 547
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #10 on: November 09, 2013, 09:48:06 pm »
Pls post datasheet kthx. ;D
DIY Eurorack Synth: https://lenp.net/synth/
 

Offline Tinkerer

  • Frequent Contributor
  • **
  • Posts: 346
Re: Making a Memristor (a tutorial...sort of)
« Reply #11 on: November 10, 2013, 01:27:46 am »
UPDATE:
It appears that the type of memristor i described is doomed to an early death.  This manufacturing process creates memristors that slowly decay into regular old resistors as they are used.  To make a more durable memristor we would need to expose a piece of copper to H2S for half a day.  Unfortunately H2S is super poisonous and likes to explode.  So until I can figure out a safe way use H2S I am stuck with my temporary memristors.

Sources:
Memristors based on an organic monolayer of molecules and a thin film of solid electrolytes
Islam, M.S. ; Johns, C. ; Long Do ; Ohlberg, D.A.A. ; Shih-Yuan Wang ; Williams, R.Stanley
retrieved from IEEE Xplore
Well, the safe way here is to use tha standard basic precautions you should take with chemicals and then use a very small amount and allow it to sit somewhere it wont be disturbed. There should be guidelines written on safe handling of various chemicals out there.
 

Tac Eht Xilef

  • Guest
Re: Making a Memristor (a tutorial...sort of)
« Reply #12 on: November 10, 2013, 08:02:12 am »
There should be guidelines written on safe handling of various chemicals out there.

In the case of H2S, the handling and storage guidelines are basically "don't - unless you have the facilities, are well trained, really really have to, and have a spotter who isn't going to run in and become casualty #2 if you collapse".

It really is nasty stuff - not the least because the threshold for detection (i.e. you smell something odd but can't recognise it) is in the sub-ppb range; you can become temporarily desensitised to it and fail to recognise the smell from a few minutes of exposure at sub-recognition levels (a few ppb); it can knock you out almost immediately at a few hundred ppm; and less than 1ppt is considered lethal.

About 2/3rds of people who are killed by H2S die because they figure "I can't smell it any more, it must be safe", and wander in where it's a bit too concentrated. Most of the other 1/3rd die when they rush in to help their mate from the first 2/3rd who passed out.

Oh, and it works in much the same way as cyanide.

It might not be up there with polonium, things with lots of nitrogens in them, or dimethyl-justaboutanything, but it's certainly not something for even an experienced home chemist to deal with in any quantity bigger than found in a dodgy egg.
« Last Edit: November 10, 2013, 08:03:51 am by Tac Eht Xilef »
 

Offline Tinkerer

  • Frequent Contributor
  • **
  • Posts: 346
Re: Making a Memristor (a tutorial...sort of)
« Reply #13 on: November 10, 2013, 09:43:57 pm »
I had a longer reply on enclosures and gloves and what not, but really at 1ppt dangerous, I would just find a lab that can help.
 

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #14 on: November 11, 2013, 04:46:35 am »
Maybe I will send a few memristors to Dave for Mail Bag Monday. 
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #15 on: November 11, 2013, 06:43:21 am »
Aren't you concerned about sending spent/treated shells through the mail? Customs may not like it.
There shouldn't be anything wrong with it.  May set off some sensors that sniff for gunpowder?  Maybe i could get the post office to check the package at the office to give it a 'passed inspection' stamp or something.
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2378
  • Country: de
  • Sounds like noise
Re: Making a Memristor (a tutorial...sort of)
« Reply #16 on: November 11, 2013, 11:18:20 am »
Hi,

as far as I understood what you need is the oxid on the brass, that can be made with any other process too. Your arrangement is close to the printed Cu/CuO/(AgO)/Ag memristors made by several other groups.

What you have here is brass/brass-oxid/(AlO)/Al and brass is based on copper. You might get better results using copper instead?

branadic
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 

Offline MaximusTopic starter

  • Contributor
  • Posts: 21
  • Country: ca
Re: Making a Memristor (a tutorial...sort of)
« Reply #17 on: November 11, 2013, 11:04:15 pm »
as far as I understood what you need is the oxid on the brass, that can be made with any other process too. Your arrangement is close to the printed Cu/CuO/(AgO)/Ag memristors made by several other groups.

What you have here is brass/brass-oxid/(AlO)/Al and brass is based on copper. You might get better results using copper instead?

branadic

As for the oxide:
I have tried using copper oxide for my memristors in the past.  This was actually one of my first experiments I did while waiting for my sulfur to arrive.  I created a little point contact junction between a penny(1/100$ Canadian coin.  They are copper plated.) that i had oxidized by heating it up, and a piece of aluminum.
(sorry queen Elizabeth II but i need to melt your face off for science)

The plastic card insulates the penny from the aluminum i built the rig out of.  any part of the aluminum is one contact, the red wire is the second contact


http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCoQFjAA&url=http%3A%2F%2Fanswers.google.com%2Fanswers%2Fthreadview%2Fid%2F397640.html&ei=7-KBUra9OqXhyQGZwYCwAg&usg=AFQjCNHPD-Ow_r90QZDXkb2FAcKueii5ww&bvm=bv.56146854,d.aWc&cad=rja
this page has some useful and interesting information about the conductive properties of copper oxides

It did work but I found that the copper oxide liked to scratch off and it seemed to have the same problems with limited lifespan.  Obviously more research, and more detailed documentation is needed on my part.

As for the Brass:
You may be right on that one, for some reason i thought the casings were all copper.  It appears that either:
A) brass doesn't have a serous effect, as the memristors all work great.
B) the casing is, in fact copper

-----------------------------------------
On a different note I have briefly looked into simulating memcapacitors and meminductors using a differential amplifier and a memristor.
This is all very exciting and i think it may be time to make a separate webpage or blog or something of my own about all these memory components so i stop clogging up this forum.
I imagine that memcapacitors would have some interesting applications in some sort of programmable RC circuits.
lots of fun stuff to learn about, i hope everyone finds this as fascinating as i do, i do realize its more into the physics/chemistry branch of things rather than the engineering. 
« Last Edit: November 12, 2013, 08:23:02 am by Maximus »
My field of study has become so complex that I can no longer just Wikipedia the answer
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2378
  • Country: de
  • Sounds like noise
Re: Making a Memristor (a tutorial...sort of)
« Reply #18 on: November 12, 2013, 07:48:01 am »
Quote
The black stuff is Copper Sulfide...

Are you sure? I would have said it's copper oxid, copper sulfid is green (green rust), isn't it?

Quote
B) the casing is, in fact copper

Sorry, I confused that with the "How-To: Homemade Memristor".
Anyway I think your arrangement is similar to the one of Nyle Steiner, but instead of brass you used copper --> Cu / CuO / AlO / Al.
However, I expect a similar effect to what is published for the printed memristors. The different is, they can't print aluminium, so they used silver instead.

http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=06553043
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf