Sunday, September 8, 2013

Quartz crystal thickness monitor for vacuum deposition

I describe how using an oscillating quartz crystal can be used to monitor the thickness of deposited coating in a vacuum chamber.

11 comments:

  1. Hi Ben!

    Great post, as always. I especially appreciated your discussion of the things that don't work as well as your successes. For us spectators it is easy to forget that science and invention requires a LOT of trial and error and perseverance. One of my favorite quotes on the subject is from Edison "Invention is 1% inspiration and 99% perspiration".
    My other favorite, I'm not sure of the origins: "In science, never say oops. Only, 'Ah...interesting'". Good luck, I'm sure you'll figure it out soon!

    ReplyDelete
  2. I wrote out a nice big long post and then I lost it in the log on rigmarole. They might check your ability to post before presenting you with a dialog to enter a new comment, he said shaking his head disapprovingly.

    I ran you down on my pedigree– what a bigshot I am. Here's what I said: try charging the substrate with electricity – see if you can coax the vapor to deposit like the morning dew. |I am Addison Phillips on YouTube.

    ReplyDelete
  3. A little acceleration can't hurt. Also, which you probably know since you're using the better method already, some of the evaporated molecules are ionized and with a little bias voltage can result in a measurable current representing the material flow (deposition rate). This may require UHV.
    Thanks for sharing, btw, interesting projects well executed..

    ReplyDelete
  4. Are all 6MHz xtals sufficiently similar in geometry that deposition mass (to change the frequency) corresponds directly to a certain change in thickness? Or do you somehow calibrate for the geometry parameters of the canned xtals?

    ReplyDelete
  5. beambot, it's a good question. I would likely need to calibrate the TTL oscillator crystals, since their outer diameter is different than the typical 6MHz crystals used for thickness monitoring. Luckily, someone gave me a Inficon sensor head, so I'll be using the correct crystals when I get everything mounted.

    ReplyDelete
  6. This comment has been removed by the author.

    ReplyDelete
  7. After some more research, Ive got some questions. Why did you attach the capacitors to the TTL? how is that all connected to the coaxial cable? Why not just purchase a TTL oscillator that matches the frequency your monitor uses? My monitor says it sends 8 volts dc and needs to get back 2 volts peak to peak with a frequency of either 5 or 6 MHz. I was reading through the manual and it looks like its possible to recalibrate it, but I know almost nothing about electronics (I studied biotech in college). Could you point me in the right direction here?

    ReplyDelete
  8. if the difference in diameter is known couldnt you just adjust the rate for the difference in mass that would be deposited due to the different surface area?

    ReplyDelete
  9. asldsvrslhf, The TTL oscillator has three pins (gnd, Vcc, and the oscillator output). However, the thickness monitor only has a single coaxial connection (gnd and signal) going to the sensor. The monitor provides a DC voltage supply on the signal connector, and the oscillator must couple the 6MHz signal back onto the signal same signal line. In the video, I used the capacitor to AC-couple the oscillator output to the signal line, and connected the line directly to the Vcc pin for supplying DC power. I used another capacitor between Vcc and gnd to smooth out the voltage supply on the Vcc pin for proper oscillator operation. There are better ways to do this, since the original system had a dedicated oscillator in-line with the sensor. The thickness and diameter of the crystal will probably matter. You might be able to calibrate it by depositing MgF onto glass, and watching for anti-reflection coatings being generated every 1/4 wavelength. Search for a book titled Vacuum Deposition of Thin Films by L. Holland. Good luck

    ReplyDelete
  10. ok my monitor has the oscilator box, say I strip open a bnc line, where would I connect the things?

    ReplyDelete
  11. ive got oscillator boxes with 4 pins and 2 pins... I wish I didnt major in biotech XD

    ReplyDelete