Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How to etch (100) p-tye Si

Tags:
  1. Aug 27, 2016 #1
    Im trying to etch 525um (100) p-type silicon, and i intend to use HF+HNO3+CH3COOH for high etching rate since i need to etch 525um of Si. does anyone know what ratio should i use to obtain high etching rate?

    I am trying to look for a mask that is able to withstand the attack of the etching solution, would the teflon tape withstand the cp4 attack? if not, can anyone with experieces let me know what can withstand etching solution's attack?

    thank
     
  2. jcsd
  3. Aug 28, 2016 #2

    phyzguy

    User Avatar
    Science Advisor

    Below are a couple of references to get you started. The problem with HF/HNO3/CH3COOH is finding a mask that will with stand this very corrosive mixture long enough to etch through the silicon. I suggest you look instead at anisotropic etchants instead. A simple KOH/IPA mixture will etch through the silicon and is easily masked with SiO2. Plus it is much safer and easier to work with.


    https://www.seas.upenn.edu/~nanosop/documents/IsotropicSiliconEtch.pdf

    https://www.ee.washington.edu/research/microtech/cam/PROCESSES/PDF FILES/WetEtching.pdf

    https://en.wikipedia.org/wiki/Etchi...t_etching_.28Orientation_dependent_etching.29
     
  4. Aug 28, 2016 #3
    Thanks for reply, if I use anisotropic KOH, it will etch very slowly once it hits 111 plane, then I won't be able to etch through the 525 um thickness, so I should use corrosive HF/HNO3/Acetic acid method.

    I'm wondering if it is possible to use Teflon as a mask to withstand its attack? Such as the Teflon etching cell like this http://sailorgroup.ucsd.edu/research/porous_Si_intro.html
     
  5. Aug 28, 2016 #4

    phyzguy

    User Avatar
    Science Advisor

    What does the pattern that you are trying to etch look like? How small of a pattern are you trying to etch all the way through the silicon? Bear in mind that with an isotropic etch it will etch sideways at least as far as down. So the minimum pattern that you can etch will be 2* 525 microns. You could achieve the same result by upsizing the pattern by 525 microns per side and using an anisotropic etch. As long as the pattern is larger than 2 * 525 microns, the anisotropic etch will etch all the way through.

    I don't know. Maybe. How will you deposit and etch the teflon?
     
  6. Aug 28, 2016 #5
    my die is about 9x9 mm and i am trying to etch a square of 7x7mm on the center of the die, so it is quite large.
    I have a question about anistropic etching using such as KOH, the etching rate would dramatically slow once it hit the 111 plane of silicon, resulitng a praymaid like etched structure, which is not what i want. Therefore, i am thinking it would be better to use isotropic etching, which it won't have such problem.
     
  7. Aug 28, 2016 #6

    phyzguy

    User Avatar
    Science Advisor

    With anisotropic etching you will get an inverted pyramid with well defined sloped sides. With isotropic etching you will get rounded sides and the degree of rounding will depend on exactly how long you etch - it will be very hard to control. Do you want the opening 7mm wide at the top or at the bottom?

    But it is your choice. If you use isotropic etching with teflon as a mask, how will you deposit and etch the teflon?
     
  8. Aug 28, 2016 #7
    i would like to have the opening 7x7mm at the bottom. I have bonded copper pillar beanath the silicon die; i need to etch 7x7mm Si on the top to all the way down to the bottom in order to see the bonded structure.

    If i use teflon as a mask, I am planning just ot use teflon tape, which is can tape around the die tightly and tape it in place on the teflon glass container, and then pour in the etching solution to it, and it will etch the exposed region on the silicon die. That is how i am planning to do, but i haven't had any Si etching experience before.
     
  9. Aug 28, 2016 #8

    phyzguy

    User Avatar
    Science Advisor

    (1) I predict that the HF/Nitric/Acetic will eat away the adhesive on the teflon tape and you will end up etching away all of your silicon.

    (2) What do you expect to happen to your copper pillar when the HF/Nitric/Acetic reaches it?
     
  10. Aug 28, 2016 #9
    1) really?, then i might look for anther way in which customize a teflon cell/holder so i can mount my silicon die onto the teflon holder with exposed 7x7mm square on the top where i can pour in etching solution, just like the etching cell sailor's lab is using http://sailorgroup.ucsd.edu/research/Standard_cell_top_tapped_all.pdf
    do you think it will work?

    2) actually, i have Ni coating on the outside of copper pillar, and as long as i could keep track of as it apporach to the end of 525 um Si, it won't etch that much of my structure.
     
  11. Aug 29, 2016 #10

    phyzguy

    User Avatar
    Science Advisor

    Well, I'm pessimistic, but maybe I'm wrong. Go ahead and try your scheme and see if it works.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: How to etch (100) p-tye Si
  1. Micro Etching (Replies: 15)

  2. Selective Etching (Replies: 4)

Loading...