Homemade Memristor

About ten years ago while walking up this mountain road, I noticed a lot of small gravel and the smell of sulfur.

I also found some empty brass rifle cartridges in the gravel that had turned to a dark black color from corrosion.

After returning home, I proceeded to investigate the electrical properties of the black crud on the cartridges.
I layed a piece of aluminum on a cartridge and connected both of them to the curve tracer.

This typical pattern was often produced when ac voltage was applied across the aluminum and cartridge.

I recorded this in my notes as a curious memory phenomenon but didn't think much more about it until recently browsing the web and becoming aware of the memristor and the amount of attention it is getting.
Since the device shown here seems to fit very closely, all of the descriptions I've read about memristors, I decided that some might find this video interesting.

The exact middle of the screen is zero voltage and current. Notice how the curve always passes through zero. This is one of the requirements to qualify as a memristor.

Pieces of copper or lead that had been sitting in some powdered sulfur for a long time also acted as memristors.

Placing a small pile of powdered sulfur on a piece of copper sheet for just a few hours produced a black area that worked well as a memristor when in contact with a piece of aluminum.

The memristor is in series with pushbutton S1, an LED and enough voltage to light the LED. The LED shows the status of the memristor whenever S1 is pushed and closed. When the memristor is on it's resistance is low and the LED will light. When the memristor is off it's resistance is high and the LED will not light.

To turn the memristor on, S2 is added in series with a current limiting resistor and a 9 volt battery.

To turn the memristor off, S3 is added in series with a current limiting resistor and another 9 volt battery in the opposite polarity.

See more stuff at sparkbangbuzz.com

Category:

Science & Technology

Tags:

• memristor
• homemade
• semiconductor
• diode
• science
• fair
• k7ns
• nyle
• steiner

[infinatty]

fantastic discovery! I was able to reproduce the effect, but only measured changes from 300 to 700 Ohms. My black copper layer must still be too thin..

[mrmeval1]

That's pretty amazing. I looked to see what it might be and came up with copper sulfide or copper sulfate or a mixture alone with copper and copper oxide. There were diodes using these substances in the early days of radio. Their failure mode is high resistance. I believe this is the first time I've seen where the resistance can be changed.

[ThedudeVip3r1990]

Amazing!

[ScientificSolutions]

Thanks so much for sharing this discovery!! I replicated the response today and was amazed, stunned actually, at how high the frequency response is!! I originally thought you might have been observing some sort of simple micro thermal effect (and maybe we are) but the frequency at which the device can be reprogrammed is amazing. The curve (on my curve tracer) is so consistent with changing amplitudes and frequencies that it's difficult to imagine an alternative physical/thermal effect. Wow!!!

[richfiles]

This is amazing! Have you achieved the effect with any material other than aluminum in contact with the sulphur reacted surface? I imagine an etched copper clad board with specific pads covered in sulfur to create the reactive surface. If a solderable springy metal can be soldered to an adjacent pad, and the point allowed to make contact with the black pad... well, we may be talking about the possibility of full printed memristor circuits! Just gotta figure out that point contact.

[themanfromwem]

FASCINATING STUFF, THANK YOU FOR POSTING, hOWARD.

[c1ph37]

Never mind I recreated the effect. My problems were that the corrosion layer was too thin and my contact point was too large. With a nice thick layer on the copper I got a memristor topping out at around 1.2 meg and bottoming at about 20 ohms.

[c1ph37]

I haven't been able to recreate this effect. I have a very low resistance across the device (about 10 ohms). When first activated I can see a small reduction in resistance (from about 20 to 10 ohms - regardless of current direction) and can see no discernible change in resistance over time.

I'm using an unetched, uncoated copper backed glass fiber board (the type used for home board etching) and a sliver of kitchen foil.

I don't suppose anyone has any idea's where i could be going wrong?

[fermeister]

@MrRoorback Post your findings! :)

[MrRoorback]

Homemade memristor? OMG. I'm going to try this tomorrow :)