MEMS chip

[Andy Resnick]

Hello World! (1st forum post )

MEMS_5-1_b.jpeg (467.89 KiB) Viewed 1693 times

This is a decapsulated MEMS (Micro Electro-Mechanical Systems) chip I (most likely) scavenged from a broken phone. This is the sensor that, among other things, controls when the display rotates along with the device. Some folks also use this technology in a classroom setting (https://iopscience.iop.org/article/10.1 ... 552/abdee2).

What you see is a very thin single-crystal piece of Silicon freely suspended and capacitively coupled to another piece of Silicon (those comb-like structures on the left and right), seen better at when zoomed in a bit:

Clipboard.jpg (315.97 KiB) Viewed 1693 times

This stitched macro image was acquired under epi-illumination conditions (100W halogen) using a 20mm epi-Luminar w/0.8x reduction lens mounted to an Ultraphot III/b, with my camera (Nikon D810) placed where the 4x5 film holder is located. Stitching was done with Hugin. No post-processing (other than stitching), the stitched image was created using jpgs output from the camera.

Since a lot of people here have extensive experience with sample handling and preservation, I'd greatly appreciate some ideas on how to properly mount these things.

[Sure Squintsalot]

Hey, that's pretty good! Pretty clean and dust free; should make post processing a breeze.

Did you de-encapsulate using the hot nitric acid method, or something else? It's looking like there's already a bit of corrosion/oxidation on the turqoise colored traces.

I have to say that I'm not a big fan of the uniformly homogenous epi-illumination. I have it on my set-up (originally a chip inspection configuration). I get why it's necessary for defect analysis, unfortunately it deadens the WOW factor. Have you tried some oblique lighting? You'd be astonished by the effect. At least, that's just one noob's opinion.

As for storage: depending on how humid it gets where you live, you might consider storing with a silica bead dessicant, though if you acid-dissolved the potting compound, most of the damage has already been done and exposed traces likely wouldn't further corrode noticeably in your lifetime.

BTW....Welcome to the forum!

[PeteM]

Well done - thanks.

I'm also curious to know how you decapsulated this, especially If you have found a relatively safe way. We'd like to try this with some of the older kids, but handling nitric acid makes that problematic. I've tried milling, without much success.

[blekenbleu]

Brute force splitting can work for these bonded chip pairs:
https://www.youtube.com/watch?v=9X4frIQo7x0&t=100s

Heat gun and flexing reportedly works for some other chips:
https://hackaday.com/2020/03/11/chip-de ... -easy-way/

[FredH]

Andy,

Nice image. Thanks for posting this.

Can you explain in more detail what we are seeing? I see several groups of things that might be flex pivots or springs. Which parts actually move? Where are the capacitive sensors? Are there amplifiers or other electronic processing elements in the image?

[PeteM]

. . .
Heat gun and flexing reportedly works for some other chips . . .

Thanks. I'll buy another batch and give that a try.

[Andy Resnick]

Hey, that's pretty good! Pretty clean and dust free; should make post processing a breeze.

Did you de-encapsulate using the hot nitric acid method, or something else? It's looking like there's already a bit of corrosion/oxidation on the turqoise colored traces.

I have to say that I'm not a big fan of the uniformly homogenous epi-illumination. I have it on my set-up (originally a chip inspection configuration). I get why it's necessary for defect analysis, unfortunately it deadens the WOW factor. Have you tried some oblique lighting? You'd be astonished by the effect. At least, that's just one noob's opinion.

As for storage: depending on how humid it gets where you live, you might consider storing with a silica bead dessicant, though if you acid-dissolved the potting compound, most of the damage has already been done and exposed traces likely wouldn't further corrode noticeably in your lifetime.

BTW....Welcome to the forum!

Thanks!

As for your questions- yes, my decapsulation method involves both hot sulfuric and hot nitric acids (separately). H2SO4 decomposes the epoxy and when that is gone, HNO3 will open up the package.

I can use a variety of epi-illumination methods as future posts will demonstrate . Oblique illumination is definitely pretty cool!

[Andy Resnick]

Well done - thanks.

I'm also curious to know how you decapsulated this, especially If you have found a relatively safe way. We'd like to try this with some of the older kids, but handling nitric acid makes that problematic. I've tried milling, without much success.

Decapsulation is definitely not safe for 'civilians'. Some of the older chips I have were packaged in TO-8 cans, those can easily be opened up with a Dremel tool w/o chemicals. Good luck!

[Andy Resnick]

Andy,

Nice image. Thanks for posting this.

Can you explain in more detail what we are seeing? I see several groups of things that might be flex pivots or springs. Which parts actually move? Where are the capacitive sensors? Are there amplifiers or other electronic processing elements in the image?

Thanks!

Honestly, I'm not able to provide much detail. AFAIK, most of that top surface is what moves. In the zoomed-in image, if you look at the two interdigitated sections on the right, one side is free to move and the other is fixed. As the spacing between the 'tines' changes the capacitance changes, and this is associated with voltage changes which the onboard circuitry detects and a secondary chip amplifies.

I have some earlier generations that are easier to understand because they are significantly simpler.