Anyone have video of memory silicon visably changing state?

[jnz]

So, probably most of you guys know what flash memory looks like under a microscope, how you can see 0 and 1 states. Does anyone have a video of change state happening? I realized I have no idea what it would look like and now I kinda want to see it.

EDITED TO REMOVE "FLASH"

[westfw]

what flash memory looks like under a microscope, how you can see 0 and 1 states.

Really?  I had no idea.  Do you have a link to a site with relevant photomicrographs?
Aren't flash cells among the smallest features made with semiconductor materials?  I thought that by the time you were able to resolve individual cells, you were usually looking at electron microscope pictures, or at least deep UV illuminated photos that had been artifically colored...

[jnz]

I just saw some color photos of 0-1 gates and the guy posted a link to software to OCR them back into data.

I'll try and find the link.

[Yansi]

I don't think so you can see anything, unless it is a mask rom memory. 

[sasa]

Modern flash memory devices use floating gate transistors, which can hold charge very long time: 1 for no charge, 0 - charge. In that case, no visible change under the microscope.

I just saw some color photos of 0-1 gates and the guy posted a link to software to OCR them back into data.

ROM is another story. I believe you refer on Piter Monta's page about physically reading memory from memory cans in HP-35:
http://www.pmonta.com/calculators/hp-35/

[jnz]

Modern flash memory devices use floating gate transistors, which can hold charge very long time: 1 for no charge, 0 - charge. In that case, no visible change under the microscope.

I just saw some color photos of 0-1 gates and the guy posted a link to software to OCR them back into data.

ROM is another story. I believe you refer on Piter Monta's page about physically reading memory from memory cans in HP-35:
http://www.pmonta.com/calculators/hp-35/

It definitely wasn't that. It was some English guy boiling acid to decap.

Ok, my bad, it must have been ROM and not flash. OK. I'll see if I can edit. So, I can definitely expect no video of ROM changing state, maybe writing EEPROM though?

[hamster_nz]

Maybe this will help... I don't have access to the actual paper, but the keywords might lead to something.

http://iopscience.iop.org/article/10.1088/0957-4484/17/7/S14/meta

"Visualization of charges stored in the floating gate of flash memory by scanning nonlinear dielectric microscopy"

[helius]

You can see TTL gates changing state with a SEM—a scanning electron microscope—if they are connected to bond wires. The bond wires are large enough that they can be visualized by the amount of deflection they cause to the electron beam. The gates themselves are not easily visualized this way, but a more sensitive instrument (an AFM or atomic force microscope) can sense very small electric fields on a chip. Of course with a SEM you have the problem of disrupting the chip itself with the electron beam.

[T3sl4co1l]

This might be the closest thing out there:



This is reading out the memory electronically, so it's not visible in the sense of a microscope watching the die.

You can't photograph that which has no image: Flash, EPROM, DRAM and SRAM shouldn't have any visible change whatsoever, and are off the list.  Possibly, FeRAM is visible, because of the special material used in its construction; the same may be true for any memory made with ferromagnetic or phase-change materials.

You also can't use an electron microscope, for the problems highlighted above: electron bombardment easily flips states, corrupting the data.  A low energy (< 10 eV?) electron microscope (I'm not sure which types can do this?) might sense the electric field at the surface, which would be an excellent way to tell.  Or likewise, an atomic force microscope with a charged probe.

Tim

[chicken]

With the SHARP memory LCDs, the traces visibly change from light to dark with charge. Though, that may be a side effect of the chip-on-glass process.

https://youtu.be/zmsshlkiYkU

[jnz]

Those are helpful. Thanks everyone, that was more or less what I was looking for.

I wish I had the giant rom image (visable by a microscope) I was looking for from a PIC, but that's fine.

[HwAoRrDk]

It definitely wasn't that. It was some English guy boiling acid to decap.

Were you thinking of Ken Shirriff? His website is at www.righto.com.

I think I recall reading something he wrote about recovering data from a ROM by visually examining the die under a microscope (or rather, from many stitched together hi-magnification photos of the entire die).

[sasa]

I wish I had the giant rom image (visable by a microscope) I was looking for from a PIC, but that's fine.

If that is interesting for you, on Piter Monta's page you have quite large complete die image one of the ROM for HP-35:
http://www.pmonta.com/calculators/hp-35/chips/index.html

[medical-nerd]

Hiya

Old tech I know, but I've always found bubble memory fascinating - here's a very short video showing the magnetic bubbles moving through the chip (they are 'serial' devices).



Cheers

[chicken]

This one may also be of interest: Photonic Emissions Analysis
https://youtu.be/Gt6VyuLZBww?t=1824

[NiHaoMike]

What about OTPROMs? They work using microscopic fuses so it should be possible to visibly see the writing process.

[Neomys Sapiens]

For anyone versed in German: Optisches Testen von ICs (optical testing of integrated circuits)
see attached file
also covers capacitive coupled measurement and stimulation.
At least the references should be usable for anyone else.
If someone is interested, I could post it in several zip parts (5M total)

[Gyro]

What about OTPROMs? They work using microscopic fuses so it should be possible to visibly see the writing process.

Aren't most OTPROMs just EPROMs in an opaque epoxy package? I don't think there are many parts (mayble old PLDs or Bipolar ROMs) that actually zap anything.