Sunday, May 25, 2008

Vacuum Control - updated!

It took 88 cardboard boxes and 1.5 km of packing tape to get my belongings packed up and on its way in a container destined for Denmark. Hopefully, I'll see it all at the other end in 2 months time. So with that stage of the move over and done I believe it's time for a post.

Rotary evaporators are great and we use them all the time but solvents have a bad tendency to bump and splash resulting in a mess. One way to partly control the mess is the use of a splash guard but it is still a pain. Chemists in industry generally don't have these bumping issues because they can afford a vacuum controller. These are great little gadgets where you punch the vacuum you would like to achieve in and hit go. On the more fancy systems even this is unnecessary as a clever little vapour pressure sensing device regulates the pressure ensuring the perfect distillation. However, these things are expensive and high maintenance so universities don't normally have them. Recently, a good friend that works at one of Australia's top institutions introduced me to a simple piece of glassware that essentially replaces the fancy vacuum controller at a very low cost (see picture left). The principle is very simple. The tap has two setting one allows passage through a wide glass tube and the other through a capillary tube. When you start the rotary evaporator you have the wide tube open and when the distillation starts you switch to the capillary tube. The capillary tube basically ensures that the current vacuum is maintained and stops it from going further down. Too easy! In addition the solvent recovery is dramatically improve saving the planet and importantly also your pump. All specifications for the gadget can be found in this paper:
Prevent the Loss of Volatile Solvents in Rotary Evaporators with a Simple Device, Daan van Leusen, Journal of Chemical Education, 1994, 71 (1), pp. 54-55. D!
One of the regular readers just emailed me a picture with a similar set-up to mine that does the same trick (See comments and picture right). The main difference is that this alternative set-up doesn't allow you to turn the vacuum off by turning the tap. Were I work we have a house vacuum system that requires a lot of tap turning so the set-up above is nice as this simplifies turning the vacuum off. However, if you are using a diaphragm pump this alternative is perfect and presumably also significantly cheaper to produce. D!


Taitauwai said...

Hi Daniel! Good luck in your move. Hope the see more goodies from you soon. Take care.

Kai said...


thanks for bringing this "trick" to a wider audience and for digging up a paper on it.
I have been using a similar setup for a while, it is pretty much an all-glass version of what the article describes. Might be a bit cheaper than the one you show due to the simpler valve.

I also find that having just one diameter of capillary does not work for all solvents. The one I have is too wide for ether and DCM, and could be wider for some high-boiling solvents. It's pretty ideal for e.g. EtOAc and cyclohexane, though.

Not sure if I can post pictures here, but I'll email you one of my gadget...


jay said...

Did you ever hear about needle valves? Cheap to buy, do exactly the same job at much lesser cost. I really don't see the point in this glass device, maybe I'm missing something...

Daniel Sejer said...

Yes, I think that your are missing the point. Full vaccum, controlled vacuum, system isolated by the turn of a tap. Each operation takes half a second. Very simple to use and easily beats messing around with any conventional tap. Also the capillary tube ensures the optimum destillation speed every time. The needle valves I normally use are expensive (compared to this piece of glass ware) so if you can supply some info on what product you are buying and its cost that would be great. D!

peach said...

If the pressure is rising that quickly, you're probably going too low and the trap isn't cold / effective enough.

If the lab purchaser pays Buchi for their V series pumps, then a controller, they're an idiot.

It's also questionable if you connect any dual stage rotary vane to a solvent boil off in the first place, let alone a diffusion (as are run on quite a few manifolds, sometimes hidden away where ever the main pump is). The pressure is way below what it needs to be for solvent work.

Fridge compressors provide a lower pressure than most diaphragms and can manage sub aspirator pressures - minus the water requirements. They're so quiet people think they've actually switched off when they're under vacuum.

I've put cyclohexane / acetone / e acetate / xylene / toluene / DCM / some liquid water and others through them.

They're designed to handle things like iso-butane. Very little (if any) rubber in them, no bearings, no real ignition hazards, no real exposed conductors.

They intrinsically apply the vacuum gently as it takes 30s - one or two minutes to bottom out for liters of glassware.

A typical edward / alcatel / pfiffifififer / varian / buchi pump costs $1 - 2.5k. Then add on the controllers, the traps, the cryogenics, the delivery of those, storage bins / tanks / special pumps / training, the special taps, all because the pump is massively over powered for the task. Pressure is too low (drives the boiling point way below 0C - nomograph), CFM's are ridiculously high for a system that should require 0CFM's once it's running properly under vacuum.

There are guys in China doing HVAC pumps that replicate the performance of those brand names for $100. The big brands are still using the look and name to charge the money. Trust me, I've taken them all apart (something not many people in the labs do), they're nothing special behind that high tech oil box cover. As much as they want you to think they are. Neither does the newer model do something super special. It's the same thing as last years in this years colours. And countless labs still buy into it, like fashion students.

I have three lab rotary vanes on my shelf (an edwards and two alcatels), and still use fridge pumps over them when I'm pulling off bulk solvent. I've even used them for high BP work. The pressure is low enough to get them down to the 100C-150C range.

Typical rotary vane; sub millibar (way too low), a few CFM (far too fast), $2k+?

Typical fridge pump pressure, 80 - 20mBar (ideal). 0.2CFM (ideal). Price, $0.

gareth batty said...

One of the major advantages of the rotary vane pumps is that the design readily lends itself to become a variable displacement pump, rather than a fixed displacement pump such as a spur-gear (X-X) or a gerotor (I-X) pump. The centerline distance from the rotor to the eccentric ring is used to determine the pump's displacement.

thanks a lot for sharing