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

Best way to go for precision aluminum parts?

  1. Feb 7, 2014 #1
    I might need to go to a metal shop/fabricator/what have you soon to have work done on a conceptually simple part that has some elements with measurements on the microscopic scale. Think cylinders with a diameter of .2 microns. Lots of them. I would want it done in aluminum. What would be the best way to go here? Can 3d printing accomplish such a feat yet? Should I be looking at CNC milling (and if so, what kind of tooling should I be looking for here)? I know work on this scale can be done - heck, they do 22nm silicon transistors, so surely they could do .2 micron cylinders in aluminum - but I'm not exactly sure by whom or how. Any clarification here would be of great help. Thanks!
     
  2. jcsd
  3. Feb 7, 2014 #2
    There's no way any normal, or even high end, fabricator could machine for you aluminum cylinders with a diameter of 0.2 mircons.

    You are looking at an incredibly an expensive and highly niche market, if it even exists.
     
  4. Feb 7, 2014 #3
    Hmm. Okay, thanks! I might be barking up the wrong tree then. I know aluminum is hard to work with lasers . . .
     
  5. Feb 7, 2014 #4

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    A laser or an EDM would still have a tough time with that. You are venturing into the realm of nanotechnology here.
     
  6. Feb 7, 2014 #5

    berkeman

    User Avatar

    Staff: Mentor

    How long do the cylinders need to be? How precise do the dimensions need to be?

    You may be able to start with a 0.2 micron "wire" and chop it up...

    Or you may be able to etch them, if they don't need to be very long.
     
  7. Feb 7, 2014 #6
    I'm not sure how you could get a .2 micron wire of aluminum though...that's like 700 atoms thick.

    Edit:
    These guys deal with small scale aluminum products. Maybe contact them and see what's up. I have never used them and have no idea of their capabilities or quality. http://www.americanelements.com/almw.html

    Edit: Also, contact high-end microphone manufacturers as they sometimes use micron-thick aluminum ribbons, so they may be able to help out with their supplier.
     
    Last edited: Feb 7, 2014
  8. Feb 7, 2014 #7

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    In any event, conventional machining technology would not be able to fabricate such tiny parts. I'm not even sure they could be seen without a powerful magnifying glass or microscope.

    This article discusses surface finishes, especially for metal parts:

    http://en.wikipedia.org/wiki/Surface_finish

    A feature size of 0.2 microns is a fine surface polish. If you want to manufacture parts to this size,
    you don't go to a machine shop, you go to a lab.
     
    Last edited: Feb 7, 2014
  9. Feb 7, 2014 #8
    Have you also considered gold and silver? There's probably a bit more research in those areas as well, as those materials tend to oxidize less, which will be critical at this size.
     
  10. Feb 19, 2014 #9
    Yes, I would imagine so. I know silicon immersion lithography has achieved amazing precision
    with 193nm UV lasers, but that wouldn't be so good for aluminum.

    I had originally planned for .5 mm length cylinders. Interesting thought about wires. I am now looking at 250 micrometer diameter cylinders of 3 mm length (yes, I know, different length/diameter ratio) instead.

    Thanks for the link. I'll talk to those guys and see if they have time for my inquiries.

    Also thanks for the link. I had made some assumptions based on the prevalence of nanoscopic
    chip manufacture, but that's silicon, not aluminum, and good luck getting access to someone's
    $2 billion fab for a small production run of tiny little cylinders.

    Not previously, but now that you mention it, gold might not be a bad idea. Ruthenium could
    also be interesting. Silver would be cheaper.
     
  11. Feb 19, 2014 #10

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    The thing about silicon chip production is you are not making tiny pieces of silicon X nm wide, you are using a photomask to etch features of this size onto a much larger silicon wafer. I don't know what your ultimate purpose is in fabricating such tiny components, but a sneeze or a breeze could disperse them however they are made.
     
  12. Feb 19, 2014 #11

    cjl

    User Avatar

    One thing to keep in mind is that 0.25mm diameter aluminum cylinders will be incredibly fragile, especially with the L/D ratio of >10 that you are looking at. What is your application - why do you need such small cylinders of aluminum in the first place?
     
  13. Feb 24, 2014 #12
    Yes, and actually, that manufacturing technique would work nicely provided the etching could reach a depth equal to the intended length of the aforementioned cylinders. Well, maybe. And there's still the issue of using any kind of laser on aluminum.

    They would be quite fragile. I am looking at a special-purpose heat sink of unibody construction with a pin-type configuration, hence the need for cylinders. Or squared pins if rounded surfaces prove to be difficult at this scale. American Elements might be able to help me with it, but that remains to be seen.
     
  14. Feb 24, 2014 #13

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Silicon chip production does not use lasers, AFAIK. The photomask is a special chemical which is coated on the surface of a blank silicon wafer. When the photomask is exposed to light, the chemical undergoes a reaction in the areas exposed to light, like taking a photograph on film, leaving the rest of the chemical unaffected. The exposed chemical can be removed using a special solvent, leaving an exposed portion of the original silicon wafer, which can be then doped with the prescribed amount of impurities to produce a semiconductor. Eventually, the remainder of the photomask is removed from the silicon.

    For your heatsink application, perhaps a similar process could be devised, what is called chemical milling.

    http://en.wikipedia.org/wiki/Chemical_milling

    Trying to build up a heatsink by assembling a bunch of near-microscopic components would be an uneconomical nightmare.
     
  15. Mar 5, 2014 #14
    Yes, that's generally what I was thinking of when I mentioned lasers. AMD/Globalfoundries and Intel (at the very least) both use immersion lithography, which is a form of the process which you describe. I believe the typical form of light used for immersion lithography is a 193 nm ultraviolet laser:

    http://en.wikipedia.org/wiki/Immersion_lithography

    I am not sure if the material to be etched has any bearing on the effectiveness of the light exposure part of the process. If it does, aluminum could be problematic. If not, hey, all the better.
     
  16. Mar 5, 2014 #15

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    If the goal, then, is to make a heat sink, I am not sure something that small will be very helpful. Basically, the fins would help cool your components through convection with the surrounding air. The problem is, with such small surfaces, you will be so low in the boundary layer that they will practically be at zero velocity anyway, so it's not clear that these would be suited to convective cooling in the first place. That leaves only conduction with the air, and those cylinders aren't likely to noticeably increase heat transfer of that type.

    Just a few things to think about, anyway. Maybe you've already gone through that thought progression.
     
  17. Mar 5, 2014 #16

    Integral

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Lasers are used in wafer processing. HP inkjet pens all are processed using lasers to cut the ink feed features. They use a 30W green beam focused to a 30micon spot. The laser has allowed them to make the ink feed slots much longer, enabling the page wide array pens which can spit out an amazing amount of ink.
     
  18. Mar 5, 2014 #17

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    The wavelength of the light used is dependent on the size of the smallest feature which must be etched onto the silicon wafer. A wavelength of 193 nm is huge compared to the features on some chips. By using Jedi mind tricks and fancy optics, along with electron beams when no one is looking, Intel Core processors use 32 nm feature sizes, and the industry is now moving to 22 nm feature sizes.
     
  19. Mar 7, 2014 #18

    Baluncore

    User Avatar
    Science Advisor

    Al is highly reactive so the surface will oxidise instantly it is exposed to air.

    For a heat sink you need tapered protrusions to conduct heat out to the end of the fibre.

    Why machine material to a cylinder when you might grow single crystal whiskers?
    Whiskers are very much stronger than machined structures in the same material.

    Parallel whiskers could be grown to have a tapered profile by controlling the growth medium parameters and potential gradient.
     
  20. Mar 11, 2014 #19
    I was working on a hydrogen sulfide detector once and I went to my local university's EE department and they were able to do what I assume was like a CVD-type process to apply gold to a surface for me. I'm sure the gold was on the order of 10s-100s of microns thick and very pure. I don't know if something similar would work for your application though.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Best way to go for precision aluminum parts?
Loading...