ziqquratu
Hazard to Others
Posts: 385
Registered: 15-11-2002
Member Is Offline
Mood: No Mood
|
|
Using oil in a mercury diffusion high vac pump
Hi everyone,
I recently scored an old all-glass mercury diffusion pump for our lab, which is in great condition, but is currently mercury-free. I work in a
professional lab, so I can in principle get plenty, however I imagine our OHS people may frown upon such a purchase and would proceed to make my life
miserable.
Does anyone have experience using oil in a pump designed for mercury as the working fluid? I could imagine that there might be differences which may
mean oil wouldn't work - but I can equally believe that it might be just fine! Either way, I'm not quite willing to jump into experimentation
(especially since we'd have to buy oil, too).
On that note, can anyone recommend good oil for one of these pumps?
Thanks!
|
|
|
DJF90
International Hazard
Posts: 2266
Registered: 15-12-2007
Location: At the bench
Member Is Offline
Mood: No Mood
|
|
I believe one of the glass blowing manuals available online (perhaps in the sciencemadness library) details the manufacture of these diffusion pumps
(theres a portable vacuum assembly on the front cover IIRC?) and its the final project in the book. As I recall I don't know if there are any
differences between those designed for oil or for mercury, but recall you need a bypass so that you can evacuate the diffusion pump before turning it
on (prevents oxidation of the working fluid). I also recall mention of diamyl phthlalate as a diffusion pump oil. I'll have a breif look on my
computer and see if I can find the cited text.
EDIT: Find attached the relevent pages.
[Edited on 3-8-2012 by DJF90]
Attachment: Diffusion pump pages.pdf (596kB)
This file has been downloaded 568 times
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by ziqquratu  | | Does anyone have experience using oil in a pump designed for mercury as the working fluid? I could imagine that there might be differences which may
mean oil wouldn't work [...] |
The main difference will be operating temperature. I had to look up a bit of
physical data as a reminder, but as I recall all the non-mercury working fluids boil at a higher temperature than mercury. They're within the working
range of glass, though. You might have to upgrade the heating element to get adequate pumping rate. References: A NIST report on the vapor pressure of mercury. Article by a manufacturer on their product, with comparative data.
If you want a shakedown run to see if your pump works at all, you don't need the expensive diffusion pump oil for a test. You can use any reasonable
pure, high-boiling hydrocarbon. You don't want to use it for production, because you'll get significant backstreaming contamination. You'll also want
to change your backing pump oil to a hydrocarbon oil for the duration of the test, since you'll get significant outgassing from an oil not produced
specifically for high vacuum. But you can run an operational test without huge expense.
I might also recommend putting such a test oil under backing pump vacuum with stirring but without heat to get rid the bulk of low boiling fractions;
you don't want to risk contaminating your vacuum gauge. Similarly, ramp up the heat on the diffusion pump slowly when first operating with test oil.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by DJF90 | | I believe one of the glass blowing manuals available online (perhaps in the sciencemadness library) details the manufacture of these diffusion pumps
[...] attached the relevent pages. |
What book is this from?
There's a more recent design for glass diffusion pumps from Mike Wheeler. Plans are available from the ASGS, though I believe you have to join to get them.
(Although if you've got the gear and skill to build one, you probably want to join anyway.) Do look at the link; it's complete glassblowing porn.
(More pictures from the University of Utah glass shop.) It's an interesting story, too. It was first conceived of as a way of demonstrating how a
diffusion pump works. It ended up pumping much faster than other designs.
|
|
|
ziqquratu
Hazard to Others
Posts: 385
Registered: 15-11-2002
Member Is Offline
Mood: No Mood
|
|
I had considered the temperature might be a factor - although I admit I expected I'd have to lower it (having never looked at boiling points of some
of those oils before!). Other than that, though, it sounds like it's worth a shot - particularly since I'm fairly sure I can get my hands on a usable
test fluid.
Thanks for the help guys. I doubt I'll get a chance to play for a couple of weeks, but I'll report back when I do.
|
|
|
DJF90
International Hazard
Posts: 2266
Registered: 15-12-2007
Location: At the bench
Member Is Offline
Mood: No Mood
|
|
@Watson That excerpt is from the book "Techniques of Glass Manipulation in Scientific Research" by J. D. Heldman. I found it a good read. Here's a d/l
link: http://filecloud.io/ibwou06r
Those pictures are awesome too. I'd love to have the infrastructure in place to be able to try my hand at glassblowing. Would be an excellent ability
to have as an amatuer chemist.
|
|
|
l-PEI
Harmless
Posts: 2
Registered: 27-1-2012
Member Is Offline
Mood: No Mood
|
|
Sorry for the thread necromancy.
I recently tried to use oil in a mercury pump. It didn't work. The oil is too viscous, and forms droplets in the jet forming nozzles, blocking them.
As for recommending a pump, remember that any line has very limited conductance. A 4" 3-stage Wheeler might be capable of 375 l/s, but once connected
to a schlenk type line this is immediately throttled to nearer 0.5 l/s or lower by the low conductance of the stopcocks. Simple upjet Hickman's are
fine in this application. If you get one the important thing to observe is where the ring breaks - there should be a distinct "ring" where the
condensation of the working fluid starts. It should be somewhere around twice the height of the diameter of the chimney.
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
