9
$\begingroup$

I've become very curious about (electrochemical) atomic layer deposition (ALD) via surface limited redox replacement.

Since the technique allows for deposition of metals in atom thick layers it's possible to create films of nearly exact atomic thickness. However, I haven't found any reference that describes how one can remove the film from the (usually Au) substrate that it's deposited on.

Hypothetically speaking, if one used ALD to generate a film of $\ce{Cu}$ that is 5 atoms thick (arbitrary length and width), how would one go about removing the film in-tact? Is it even possible?

$\endgroup$
  • 1
    $\begingroup$ Very carefully with a pair of small tweezers. $\endgroup$ – Ali Caglayan Jun 11 '15 at 11:55
4
$\begingroup$

In most cases I've seen, ALD is performed on the intended substrate - that is, there's no need to separate it. In other cases, you use a sacrificial substrate that you can remove easily.

If you wanted to separate the film from the substrate, there are various techniques, but nothing comes to mind for separating Au from Cu. I find it incredibly unlikely in the specific case of a 5-atom thick layer of Cu, which would likely not survive typical techniques (vs. bending, mushing, forming a larger packing, etc.)

Some typical techniques for separating two films:

  • Adhesive tape (as mentioned in a comment) - works for many films, e.g., graphene, other layered materials, where you physically pull apart layers
  • Swelling (e.g., separating Au from mica) - including various intercalation methods used in graphenes and graphene oxides and other layered materials, so that the layers separate more easily
  • Dissolving, etching, or eroding the substrate (e.g., back-etching a Si wafer to expose an $\ce{SiO2}$ or thin metal film deposited on the wafer)

In my group, for example, we make "template-stripped gold" where Au is deposited on cleaved mica substrates. Then the thin gold layer is glued to glass substrates and the mica is separated from the Au by swelling / degredation, leaving a very smooth Au surface.

This technique can work with fairly thin layers of gold (~10-30 nm) but there's a large difference between Au and mica, making separation possible. Not true with Au/Cu.

$\endgroup$
  • $\begingroup$ I'll comment that I would love to see some technique capable of separating 5-atom thick metal films. $\endgroup$ – Geoff Hutchison Jun 12 '15 at 2:15
  • $\begingroup$ +1 for dissolving or etching the substrate. It would be hard to selectively etch gold underneath copper, but with a proper choice of substrate, I would say that is the way to go. $\endgroup$ – Burak Ulgut Jun 12 '15 at 10:35
  • $\begingroup$ @BurakUlgut I'm just coming up empty with ways to etch out the gold and leave the copper, especially in this scenario. I guess if you timed it extremely carefully, you'd remove all the gold and leave copper. $\endgroup$ – Geoff Hutchison Jun 12 '15 at 12:56
-1
$\begingroup$

I'd suggest electrostatic repulsion. Charge both layers with electrons and they'd separate easily. Then use a electron microscope to manipulate the layer.

$\endgroup$
  • 3
    $\begingroup$ Adhesive tape would work for some combinations (used for graphene, for example), but since gold and copper alloy so well, I'd assume the copper would have interpenetrated the gold surface layer, to some extent. $\endgroup$ – DrMoishe Pippik Jun 11 '15 at 21:01
  • 3
    $\begingroup$ I agree with the comment above (i.e., some amount of alloying in such a thin film) but even if Au/Cu was fully segregated, I don't see how charging the film will separate the layers. The film will be highly conductive, so I don't see how electrostatic repulsion will occur - added electrons will be highly delocalized. $\endgroup$ – Geoff Hutchison Jun 12 '15 at 1:24
  • $\begingroup$ This could have been written as a comment. $\endgroup$ – user15489 Jun 12 '15 at 1:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.