Fun with my Vacuum Pump- Operation Questions

Quote: Originally posted by Wizzard

2. Pump backflows oil when turned off, and when oil gets inside the vessel, the slight water contamination causes a film buildup of sludge, I'm guessing. Prevention, besides the obvious?

Oh, I've got a needle valve on the way I appreciate any and all help.

Make and model of the pump is ULVAC G - 20D.

Ah, see, thats the information I need I've also been considering sealing it off with indium, which I have a copious amount of.

You can get down to 80mbar with a fridge compressor. They make excellent scum pumps.

I've heard people muttering about them running out of oil and stuff but I've had one running for probably about a decade now and it still works. And even if you toast it, they're disposable. You can score them from the dump. I've managed to get my removal time down to about three minutes with a screwdriver and pair of pliers.

I also never had the problem of oil back flowing - not in any quantity that it would scum on the glass.

You might want to consider fitting a trap of some description to your pump. If you;re sublimating high melting point stuff, water might be enough to catch it. The less shite you end up with in your pump oil, the better, particularly if it's a halogen.

Be VERY careful when you start scaling up. It turns into an entirely different beast as the volumes increase. HEED MY WARNINGS! A common misconception is that the harder you pull the vacuum, the more explosive the implosion will be. That's true to some extent, but even your average scum pump will remove a good 90% of the atmosphere inside a flask. A decent rotary pump will get almost all of it. As for the turbomolecular, getters and ion pumps, stepping up to them will cause almost zero increase in the violence of an implosion, they're removing the last, sub 1% fraction. It's probably more like that last 0.01%. The danger is produced when the 99% is roughed out by the rotary vane.

The energy contained in the implosion is proportional to the volume that's been evacuated, so a big flask will give a much bigger bang at the same pressures. It's common in proper labs to see vessels taped up so, if they do pop, the glass isn't sprayed everywhere. They also tend to cover the glass with that webbing you see on scuba cylinders to stop them getting scratched; same reason, catch the glass. Stay away from anything other than round flasks as the pressure drops and volumes increase.

I am genuinely worried around large volumes under vacuum. The slightest temperature variation or knock and they'll go; you have to warm them up VERY, VERY gently, I'm talking hours or even days for the big flasks. A bunsen will almost certainly fracture the glass due to the temperature variation induced stress; thats why proper labs use massive, expensive heating jackets or baths. Which raises another important point, if you don't heat them smoothly and uniformly, you'll develop stress in the glass, which wouldn't normally be a problem but it is now that you've pulled almost an atmosphere's worth of work out of them. It can take a few minutes or tens of minutes to pull the vacuum. That's like spending a few minutes or tens of minutes charging a capacitor. The amount of energy stored in the system is gigantic compared to the supply that's pumping it. A tazer works off 9v batteries, but will produce a shock so strong it'll drop anyone to the floor. The same applies for pumping down flasks.

There's no need for extremely hard vacuums in chemistry. The boiling / sublimation point has usually dropped so much by the time it's down to 100mbar that you're good to go. Saying that, I do wish a had my monster rotary vane handy sometimes to get down to maybe 10mbar for delicate organics.

Extremely hard vacuums are used for physics, in things like molecular beam epitaxy, where they need to have a volume that is basically empty of everything but the deposition beam. That's miles and miles away from what us chemists are getting up to. In MBE, they're looking for a volume that only has a few hundred molecules bouncing around.

You don't need high flow rates. Big industrial pumps with high CFM are used for ripping gas out of polymers and things like epoxy resin in the composite industries to avoid voids in the product. In chemistry, you should theoretically need 0 CFM once the system is at pressure, as the condenser should be neutralizing any vapor pressure. With the big industrial pumps, they need to rip the gas out out of the polymer as fast as possible as the polymer usually needs mixing with a catalyst before it can be used; epoxy glue is a perfect example of a polymer needing a catalyst, it's even one of the uber resins used in composites. The mixing beats air into the polymer (the betting the mixing, the better the cure and, unfortunatenly, the more air gets mixed in). Once the catalyst is in, the polymer is starting to solidify, so they need the gas out of it as fast as it can be ripped out so they can get it out of the chamber and get on with laying up the fabrics before it cures, as the longer they spend waiting the less it'll permeate the fabric and the poorer the end result will be.

Of coarse, not everything works out so fine and dandy in the world of chemistry, you'll have leaks and the condenser won't work perfectly. With things like DCM, if you're not running sub zero water through it, DCM will end up spitting out the exhaust of the pump as it slams into the wall of atmospheric pressure lurking outside. So having some CFM will help keep the pressure down there. It also indicates how important having a good scrubber / condenser setup is. DCM in your pump oil is not good. It'll dilute it down, killing your ultimate pressure ratio and halogenating the oil, which'll rot the insides of your pump; unless it's one of those ridiculously expensive PTFE diaphragm pumps that only reach around 100mbar. Acids are the worst. Luckily, I've never had the opportunity to experience that.

If you're distilling / subliming higher melting point substrates and don't want the expensive of buying a gas wash bottle setup, you could probably get away with just sticking a condenser in the way. With such high melting points, the substrates will solidify on their way to the pump. A scrubber would still do better I expect. And besides, gas wash bottles are the coolest thing around. Fuck buying a fume hood, you can solve 99.9% of your fume problems by simply venting the fumes through a wash bottle scrubber. I'd take a guess that you can scrub more out of the fumes this way than you can with a fume hood. AND, you get the benefit of being able to wash / fume / vent workups; I love wash bottle heads, I have about three or four of them. Active carbon is good, but it only absorbs, it doesn't neutralize. Scrubbing neutralizes, when it's done well.

Putting a wash bottle scrubber between your pump and flask will not only prevent things going from the workup to the pump, but will also destroy any chance of backflow from the pump. In proper labs they often use dry ice / IPA traps to get the temperature down REALLY low to liquefy / solidify anything trying to move between the two, other than gas. A well thought out scrubber will do just as well for most applications. E.g. filling the flask with something that's a good solvent for the pump oil will strip it out of the back flow. If you have acid vapors trying to get to the pump, stick an alkaline scrubber in the way, like NaOH or KOH, that'll tear apart any acid trying to get through. Use some mind power in selecting the scrubbers and you'll get rid of all the problems. For an example of how powerful this concept is, nerve gas is horrendously toxic to human. It's effective at the same dosages that LSD is, micrograms. But it also decomposes on contact with a base, so simply scrubbing the fumes with NaOH or KOH will destroy the compound, rendering an unstoppable leviathan into a harmless byproduct. [A NOTE TO THE AUTHORITIES BROWSING; I HAVE ZERO INTEREST IN PRODUCING WEAPONS. IT'S YOUR OWN MORALITIES AND IDEAS ON THIS TOPIC THAT YOU NEED TO RECONSIDER. SLEEP TIGHT, MURDERS. YES, YOU ARE. YOU BRING ME CLOSE TO THROWING UP]

However, even with my scum pumps (fridge compressors), their pathetic flow rates managed to fight off the leaks. If you grease up the tapers, you'll often see bubbles where they're leaking.

You can also use PTFE tape, but you need to wind that a little more carefully or you'll make things worse. Yo need to wing it around the tapers so it doesn't overlap it's self, if it overlaps you'll create ridges that aren't uniform and allow the atmosphere to creep around them. PTFE is good because it doesn't melt as it warms up or wick out your joints when a solvent goes past. It's quite common to find grease in your product if you're going cheapo and using vaseline. Proper vacuum greases are much more stable. PTFE tape also seems to dissolve when exposed to DCM, like my Nitrile gloves.

As far as a grease that'll seal up to 750C, I can't help you out much there, I think you may just have to live with the leaks. But first, concentrate on getting EVERYTHING else in your setup sealed properly. 750C is a SERIOUS temperature to expect a grease to be happy with.

What I'd suggest is that you use a cold finger condenser to collect the sublimate and then either a condenser and/or a scrubber to stop anything getting back to your pump. Choose a scrubber that'll strip the sublimate out with high efficacy but also render it in a form that'll allow you to easily recover the substrate if it's expensive, like iodine is.

Incidentally, I do have one major complaint against fridge compressors. They tend to have an annoying trait when it comes to quiescent regulation; when they bottom out at their lowest pressure.

The pressure will drop to 80mbar for example, then the pump will go silent, as if it's switched it's self off. The pressure will gradually rise to ~120-140mbar over a minute or two as it backflows or the system leaks and then it'll start up again and drop to 80mbar.

There are two issues with it doing that. Firstly, you can't nail a vapor temperature at the still head.

Secondly, and far, far, far more tedious is that during the time the pump is in it's off state, the mother liquor can accumulate more thermal energy, as it's now at a higher pressure which will prevent boiling until the temperature goes up.

When the pump starts up again and the pressure shoots down, the liquor is essentially in a superheated state at the new pressure (it has a ton of energy it would have normally boiled off as vapor now stored up in the exponentially more dense liquid format); ever done that trick where you microwave some water for five minutes and then stick a wooden spoon in it? Well something similar to that happens when the pressure drops, and you universally get severe bumping. I can't count the number of times I've messed around with this, lacking my monster pump, and had a distillation be rendered entirely useless when the liquor rockets into the product.

Also, sometimes being able to get down closer to 10mbar is handy because azeotropes will break at the lower pressures. So, for instance, you can produce dry ethanol by distilling below something like 60mbar, which the fridge pumps I've seen can't manage.

My genuine vacuum pump was incredible. I got it for $150 or so, because the seller had a terrible picture and described it as untested.

On arriving, it was HUGE. It was perfectly functional and better yet, it didn't have a mark on it; may as well have been new. It was so heavy I could only just pick it up. Checking the specs, this was one serious piece of gear, the kind of pump that retails for $5k or more for roughing down to diffusion; it could get down a few microns. I was kind of scared to go near the intake, anything getting too close would suddenly crack onto the intake like there was an immeasurably powerful magnet there. I practically stole it for that price.

I had to sell it in the end, but I still miss it. It was one of those bits of gear that makes you drool with envy, a true demon! I'm sure it could kick the ass of any BOC / Edwards pump commonly found in a genuine lab. I think it was probably intended for huge chamber evacuations or degassing polymers, it was way over specified for distillation. I miss you baby!

Take care with vacuums and high heats. As I've already said, the first time one blows on you, you'll fill your pants. And the combination of vacuum and extremely high temperatures is something you need to be very careful with. I've blown a 10 / 15l flask at around 80mbar and a hundred C. That was enough to make me jump. Luckily for me, the flask was packed with solid material I was distilling, so it essentially contained the implosion. My liquor material was also very stable and soaked in water, so there was no chance of it igniting. If a big flask implodes and it has a smaller amount of material in it at 750C, it'll spray everywhere and probably set fire to things. I have zero idea how the things you're planning to work with will behave when they're released into the atmosphere at 750C, but I'm guessing it could end up looking like a fireworks display as they oxidize. Iodine would fill your room with bright purple, choking fumes that'd stain the entire place, I'm guessing. I've had iodine on the carpet and in the bath, it NEVER goes away.

The other guys are also right that the melting point of the glass doesn't define it's limits. Glass is a fluid, and creeps whenever it's above absolute zero. Heating it so close to it's melting point will encourage it to collapse under the vacuum.

[Edited on 1-5-2010 by peach]