Making Silicon and Beryllium
Melting Beryllium and Silicon Dopants
Follow the procedures below when outgassing or melting a new Si or Be charge. These procedures are the best practice. Read the notes from John English and other recent procedure notes listed
Melting and reloading instructions
Per conversation with John English (12/13/2010 EMK)
Outgas dopant cell without and with 1.5cc PBN crucible(s)
Load the material:
Silicon: 1 to 1/2 cc of high quality silicon. Either buy pellets or break up pieces of a prime grade UID silicon wafer (after giving it an HF dip to remove oxidation). We crushed up pieces of a wafer inside a beaker using the ball end of an ultra-clean SS Allen wrench
Beryllium: Load 1/2 cc of Be into crucible (1 cc if you can afford it). To break up the Be chunk, use a disposable glove bag (an atmos bag) to avoid the toxic dust particles spreading. Hold the charge with ultra-clean pliers or vice grip and saw using a jewelers saw. Side-cutters work just as well. Fume hood, dis glove bag and respirators are a MUST.
Fit 1-2 tantalum foil melting caps over the crucible (to hold in the majority of the material). From 0.002" tantalum foil stock, we used scissors to cut 0.7" diameter circles with two 0.2" tabs (on opposite sides) to go under the wire holding the crucible in. Be sure to give them the treatment before putting them on the cell.
Ramp the cell up to ~200 C below the melting point at 20 C/min. Silicon melts at 1414 C. Beryllium melts at 1287 C.
Ramp the cell to the melting point at 50 C/min. For Silicon, hold it there for 2-3 minutes. For Beryllium, minimize the amount of time at high temperature. Beryllium will outgas like crazy and diffuse through the crucible into the cell at elevated temperatures. Silicon melts at 1414 C (we went up to ~1465 C). Beryllium melts at 1287 C (we went up to ~1290 C).
Ramp down the cell at 50-100 C/min.
Pull the cell out in a glovebag and make sure the charge melted. If the charge has melted properly and wetted to the crucible, there is no need to dump out the excess. You can turn the cell upside down to make sure the charge doesn't fall out. If it didn't melt, stick it back in the chamber and take it 10-15 C hotter.
2019 December dual dopant Si reload
Notes from SDM
Followed the John English and Veeco procedures with the following differences/details:
Did not HF dip/clean the crucible before reloading. Probably should have done this but did not notice it in the John English procedure
Si needed to be crushed in to small (< 1/2 cm2) or dust-like pieces. The smaller the pieces, the easier they melted with and fused to the wall of the crucible. Larger pieces did not melt/fuse and just fell out of the crucible when the cell was tilted (note, the dual dopant was a shallow downward-looking cell at the time). If the pieces are still largely intact after a melt attempt, it's likely that the charge did not fuse to the wall and will fall out. When dumping the Si material into the crucible, used a glass beaker. If a large piece was about to fall into the crucible, it was dumped into the trash to make sure no large pieces entered the crucible
Si was crushed up in the fume hood. (NOTE: Beryllium should never be crushed in a fume hood, only in a glovebag with respirators!!!)
Used an outgassing cap that was designed for a larger cell and had a hole in the center of it. This seemed to work fine and allowed us to reach sufficient melting temperatures. The first time the cap was used, the bake pressure was very high, probably because the cap had not been used for a long time and was dirty. Rebakes that followed had much lower bake pressures as the cap was likely much cleaner
Specific notes about temperature and pressure from the bakes:
Attempt 1: no outgassing cap, large pieces of Si used, max temp of bake ~1414 C (max temp cell could reach before power limited), max pressure ~ 5E-7, time at max temp ~3 minutes. Charge was barely melted and fell out when cell inverted and lightly tapped.
Attempt 2: with outgassing cap with hole in center, smaller pieces of Si used. max temp of bake ~ 1450 C (not power limited), max pressure ~ 2E-5 (suspect dirty outgassing cap, took ~ 8 hrs at 1200 C for pressure to drop below 1E-7), time at max temp ~3 minutes. Charge was ~1/2 melted but several large pieces still clearly visible and fell out when cell was inverted
Attempt 3: with outgassing cap, smaller pieces <1/2 cm2 or smaller and less material used (filled up crucible ~1/2 way), max temp of bake ~ 1465 C, max pressure ~ 5E-6, time at max temp ~ 3 minutes. Charge was almost completely uniform in appearance (only one small piece of wafer discernible from the melt). Charge did not fall out of cell when inverted and lightly tapped
Veeco Silicon/Beryllium Fusing Procedure
CAUTION! Avoid overheating. Molten silicon and beryllium react with PBN. This can lead to crucible failure, resulting in source damage.
Outgas the source and crucible.
Load a small amount of source material into the crucible (~1 gram). A single piece should be used. It should have maximum surface area contact with the crucible for best results.
Place an outgassing cap (Veeco P/N 211293) over the end of the crucible. If an outgassing cap is not available, you may substitute a clean piece of tantalum cut to the size of the end of the source (see Figure). The cap should “seal” the end of the source to minimize material and radiative losses.
Wire the cap in place to keep it attached.
Insert the source into an upward looking port.
Pump the system down to < 1x10-7 Torr.
Introduce liquid cooling around the source.
Ramp the source to 200 C below the melting point (~1100 C for Be, ~1200 C for Si) at 20 C/min. Maintain a system pressure below < 1x10-7 Torr (slow the ramp down if the pressure increases).
Allow source to stabilize and system pressure to recover for 10-20 minutes.
Ramp the source to 25 C below the melting temperature (~1240 C for Be, ~1380 C for Si) at 50 C/min.
Wait for 1-2 minutes.
Cool the source down to room temperature at 50-100 C/min.
Warm up the cooling panels and vent the system.
Remove the source.
Invert the source with the outgassing cap still in place and gently shake to see if the charge has fused.
If the charge has not fused, repeat this procedure going 5-10 C hotter in step 10.
Once fused, the cap may be removed.
The crucible should be wired back into the source without the outgassing cap.
The source can then be mounted in the desired downward looking port.
Contact: If there are any questions regarding this procedure, please contact Rich Bresnahan at (651) 494-5567 or rbresnahan@veeco.com, or contact info@appliedepi.com.
Ta Melting Cap Design
SJM cap creation notes
The Ta melting caps are used to keep in both heat and material. They should lay directly on the opening of the PBN crucible, and have no hole in them (as compared to the outgassing caps). We decided to use circles with side tabs to go under the Ta wire that's holding the crucible in. The original outgassing caps are ~0.85" in diameter, so we made them ~0.7" in diameter, with two 0.15" by 0.15" tabs on opposite sides. We cut them out of 0.002" thick tantalum with sharp scissors and layered 2-3 pieces on the opening of the crucible (after inserting the material).
-Scott Maddox 12/13/2010
2019 cap notes
Bravo dual dopant Si was successfully melted using a high-temp outgassing cap. This cap was larger than the cap described by John English and Scott Maddox since it was for a larger cell. This cap also had a hole in the interior of the cap, so it was not completely closed, like the John English and Scott Maddox caps. This worked fine and allowed the Si cell to go from a max temp of 1414 C (no cap) up to at least 1460 C (with this larger cap, probably could have gone a little warmer but 1465 C was enough to melt/fuse the charge)
Stephen March, 2019-12-29
Melting Beryllium: AFB notes
Performed 200224 The method above works well. Images are forthcoming once image upload issues are resolved.
Breaking the Be to fit the crucible was particularly hard, it might be worthwhile to invest in a jeweler's saw as EMK recommends. Wear a respirator when handling Be. Set up the individual glove bag in the front fumehood, taping the back with caution tape with a N2 line inserted (unscrew airgun). The bag should include clean foil, two pairs of clean pliers, beryllium, a clean beaker to place the 1 cc of Be in. The only method that I found successful for breaking the Be apart was grasping the Be with two pliers and tearing it apart into small pieces. Be is hard, cutting with a clean razor or tweezers does not work.
Follow the melting procedure above. At ~1290 C you could see the individual pieces of Be melt after which we ramped the cell down.