Sciencemadness Discussion Board

Keeping crap out of my vacuum pump?

jgourlay - 15-6-2010 at 11:56

I've lately taken a little-kid like entertainment in doing various reactions under vacuum. The problem is, lots of stuff starts to evaporate under low pressure (like water), that I would really rather not have in my vacuum pump. They're expensive, and things like nitric acid probably don't mix all that beautifully with the vacuum pump oil.

So...how do the pro's do this?

Lambda-Eyde - 15-6-2010 at 12:16

With a dry ice or liquid nitrogen trap.

bahamuth - 15-6-2010 at 16:09

If you have a good rotary vane pump, no liq. nitrogen or dry ice trap is going to keep water out of your pump, especially if your runs take a long time.

Rather get yourself a couple of KOH traps or one KOH trap and a wash bottle with sulfuric acid, with KOH trap closest to pump. A very dense filter of glass wool should protect your pump against any KOH dust entering your pump.

Plus these are cheap chemicals and last relativly long compared to both dry ice or nitrogen..

not_important - 15-6-2010 at 20:06



Residual Water mg H2O/L Dried Air (numbers after a ';' are from alternative references)


KOH     0.3; 0,01-0,9

Molecular Sieve Activated Type 4A    0.001

Calcium Sulfate    0.005; 0,005-0,07

Silica Gel 6-16 Mesh    0.03

Magnesium Sulfate    1.0

Cupric Sulfate    1.4

Calcium Chloride (<20 Mesh) &nbsp;&nbsp; 0.14-0.25

LN2 trap &nbsp;&nbsp;&nbsp; < 10<sup>-6</sup>



altech6983 - 16-6-2010 at 10:09

My design may or may not be all that great but I will post and see what y'all think.

I am using four cylinders of PVC pipe connected top to bottom on all four with the bottom of the first going to the apparatus and the top of the last going to the vacuum pump. The first cylinder contains water. The second a Molecular Sieve Type 4A. Third and fourth contain activated charcoal with dense glass wool on top.

I figure the water will dissolve any water soluble gases, at least some of them. The Sieve will trap the water. The last two filters of activate charcoal will trap any remaining gasses and the back streaming (not that is really going to matter at this small of vacuum) of oil from the pump. It is pretty cheap to build and should protect my 1500 dollar vacuum pump. This should work for you should the forum members approve.

What do y'all guys think of the design?

jgourlay - 17-6-2010 at 06:19

Altech, looking forward to it!!

densest - 17-6-2010 at 06:42

Activated charcoal is a good filter material to remove chemically active substances but it is not good for removing gross amounts of particulate matter, oil mists, and other relatively inert matter. A film of oil on charcoal will block all the pores making it totally inert. Charcoal is often used to absorb acid fumes as well.

Zeolites (molecular sieves) are destroyed by acid. If you anticipate a lot of acid fumes, some soda lime or other alkaline prills before the sieves will protect them.

On a $1500 pump, I'd advise putting at least two more traps before the ones you mentioned:

a large empty jar to lower air velocity enough that large particles fall out
then
glass wool to remove smaller particulates & mists

These would not be affected by acid, etc. and they would reduce the load on the other filters, prolonging their life.

An empty jar and/or more glass wool next to the pump would be good to catch most of the backflow and keep the charcoal active. That would also catch stray charcoal. Charcoal is reasonably effective removing acid fumes as well as organics.

If you use the water trap, the molecular sieves have to be removed and dried frequently. A pass through silica gel might be useful to get most of the water out first. It would probably be enough to protect the pump by itself. Also, it's hard to scrub gases through liquid thoroughly - the gas has to be in tiny bubbles to get enough surface area. I'd skip the water trap.

So... I'd recommend:

dirty gas -> empty trap -> glass wool -> soda lime or equivalent -> charcoal -> silica gel -> glass wool -> pump

Yes, this is pretty complex... is it worth $1500? Your call.

not_important - 17-6-2010 at 07:58

Face it, if you're throwing crap like that at an expensive mechanical vacuum pump, you're using it wrong. Get/build a recirculating aspirator pump to use for filtering, solvent removal, low vacuum operations like distilling HNO3, exploding small animals, sucking the blood out of infants, and so on. Save the high vac pump for true vacuum operations, use the aspirator to pretreat anything that might really hurt the mech pump. An awful lot of vacuum work was done with water aspirators - 'water pumps', old chemistry texts use the phrase 'at the pump' and don't talk about traps because that's what they used.

Putting a container of water in line next to whatever you are evacuating means effectively you'll never do better than the vapour pressure of water in the apparatus, and it will be seeing that vapour; you might as well stick to an aspirator and save the hassle and cost. And the molsieves get the full steam of damp vapour, so they'll fairly quickly saturate in that design.

Using an alkaline filter isn't a bad idea, just remember that it may limit your ultimate vacuum level by out-gassing water back into the apparatus. Alkaline washed alumina is an alternative, it gives less dust than simple sodalime, doesn't form aqueous films like KOH, and will absorb some water too as well as serving as a dust filter.

Mechanical particle traps are good, but I think metal gauze like that used in oil traps for pumps is more effective than glass wool. Consider it at least for the pump-side filter, where it also has to deal with oil backstreaming, and you should have one on the pump's vent as well. Rather than loose glass wool, mats or similar woven forms are better as they are less likely to have channels where particles can sneak through.

For molecular sieves you want both 4A to trap water and 10X or 13X for organic vapours, A 5A stage wouldn't hurt either, it picks up typical solvent alcohols and smaller decomposition products from organics. Silica gel would be best as a front end to molecular sieves if you're going to be throwing a lot of water at the pump, it's much easier to regenerate, but silica gel lets through about 20-30 times the water of 3A or 4A sieves.

Activated carbon grabs the larger molecules like xylenes and bigger, use it after the water traps - water can displace the organics - and dry it before filling the trap with it as it can pick up considerable amounts of water in sitting around.

Remember that all of these offer resistance to the flow of gases, and thus make the pump work harder and may limit the vacuum you can reach. Try to make cross sections as wide as possible, rather than having long narrow tubes. Keep the connecting tubes a cm or so in diameter.

Large glass jars are a bad idea with pumps that pull any sort of vacuum, you're asking for implosions. Metal or heavy walled plastic is the way to go, thermoplastics may soften if the incoming gases can be warm. PVC and other plastics may outgas plasticisers, adding to the filter load and possibly getting into the pump oil, commercial filter products use plastics that don't have those types of additives.

altech6983 - 17-6-2010 at 11:46

Thanks for the feedback, I haven't actually used the filter or built it yet I I was just planning it. I didn't know that Zeolites were destroyed by acids, good to know. I originally bought the pump on eBay for $300 and I was going to use it for a particle accelerator. I ran into money problems with college and all so that has been put on hold.

I'm not looking for super vacuums, something like an aspirator or a little better, mainly at this point for white nitric acid distillation and hydrogen peroxide concentration. I always thought he water was kind of a bad idea.

I had it on hand so I figured I would use it but since it is really that complicated to trap it so that it stays safe I guess it will invest in a recirculating aspirator system. Deals like that don't come along all the time.

Thanks for the help not_important and densest.

Panache - 18-6-2010 at 03:25

you may already know this but i used a ice cream maker as a cold trap for the past few years (until it died/was murder just recently). I found it an excellent tradeoff cf/ solid CO2/N2 which are time consuming and expensive and solid traps which clog and are tedious to clean and replace.
It was an italian machine for domestic use, it's cavity easily fitted a laboratory reagant bottle (1L) which i used as the trap and had methylated spirits as the coolant. I generally left it on for weeks at a time and it never missed a beat sitting stably at -35-40. I had rigged extra insulation around the trap area as it would ice over otherwise in a few days. I'm now needing to replace it but am looking into small commercial machines typically sold to restaurants for boutique ice-cream, the cavity is around 2.5L and it sits at -50.
The drawback is it costs $1800 were-as the domestic ones were around 400, i got mine second hand for $200.
I assume the warranty would be voided for the use i intend so i'm waiting for one to come up second hand.


As an aside i thought teflon pumps negated the need for cold traps/acid etc, just particulate filter is needed. Is this not so?

It is an extremely conienient way to go, iii left it on constantly and the pump was always protected, initially i used a bleed within whatever setup i was dealing with however I found out this often would over-run the little icre-cream maker, so i switched the bleed to position between the trap and the pump and it was never a problem again.
That said i think i would perhaps be more pedantic i i made a swish pump.

altech6983 - 18-6-2010 at 21:43

Actually that reminds me I have seen some people on overclocking forums build a auto-cascading compressor system that was able to reach -104F. I might have to look into that when I do decide to build my particle accelerator as I will need a good cold trap. As for now, I ordered an aspirator, and my dad has a usable water pump that I can use (doesn't care if it gets eaten up, though that will probably take a long time) to build a recirculator.

I not really sure about Teflon pump negating that, I would think that water vapor and other things could build up in the pump (The link below has some pumps that have a auto purge cycle that "blast" those out) raising the ultimate pressure you can achieve but then again your probably not using a Teflon pump to reach the millibar levels.

I've been surfing eBay for a good Teflon pump price but they are too expensive, just have to wait for a good deal. A Google search found this: Pumps, love to have one of the lower ones on the page.

Also, sorry jgourlay for hijacking your thread.

jgourlay - 21-6-2010 at 07:41

Thanks all. Will repost when I have some results.

Contrabasso - 26-6-2010 at 23:58

Many reactions are better done under reduced pressure but not even nearly vacuum.

Distillation of Nitric acid produces red fuming acid at ambient pressure distilling in the region of 80C. If you reduce the pressure then it will distil over in the region of 50 - 60C and give white acid without the decomposition products. Also it's cheap and easy to cool a condenser for operation around 50C. If you pull a reasonable vacuum then nitric acid will boil at ambient temperature and getting the product to condense will need chilled condenser coolant and a chilled receiver and plenty if acid vapour will go to the pump.

Fit an air bleed valve near the vac pump to moderate the vacuum level and provide some flow through the pump, that way some of the pollutants that leave the cold trap and filters are diluted before the hit the pump and the flow helps to flush the oil.

Picric-A - 27-6-2010 at 13:26

Is corrosion due to solvent vapours a big problem in diaphram pumps? I have just bought one second hand and i am wondering how easily it will be attacked if used unprotected on a rotovap.

altech6983 - 27-6-2010 at 15:38

It really depends on what the diaphram is made out of. If Teflon, probably not much.

I would check to see what diaphram you have and then only use it with solvents that won't hurt that diaphram material.

Of course not much of an expert on things so I would wait for another reply from someone higher up.

Fleaker - 27-6-2010 at 15:47

I run DCM, toluene, DMF, and TFA vapours through my KNF Neuberger when rotovapping. I also filter aqua regia solutions with it. It's got PTFE/Kalrez heads and it's held up fine.

I lost one KNF Neuberger after years of abuse--a piece of copper wire inadvertently got into the vacuum hose and caused it to seize. I highly recommend those Laboport pumps.

Picric-A - 28-6-2010 at 01:54

I have Charles Austen Pump Model B100 SEC. As it is second hand it did not come with a manual and as a result i have no idea what the diaphram is made of. If anyone has access to a manual for this it would be extremly usefull if they could scan it in.

Fleaker- I have heard diaphram pumps make great lab pumps as they are virtually maintenance free and faily quiet running. Have you ever tried replacing the diaphram of your pump? Is it an easy' DIY' job or is it a 'send to the manufacurer' job?

jgourlay - 28-6-2010 at 04:52

I'm looking at diaphragm pumps on ebay. On most of them, I'm seeing the inlet and outlet, but not the little 3 prong thing that plugs into the wall.:)

Are these mostly sold without a motor, or am I missing something?

densest - 28-6-2010 at 06:49

Many diaphragm pumps are powered by compressed air or hydraulic pressure (either water or oil). Those are usually large units intended for industrial applications. The ones shown on EBay without motors are that kind as far as I looked.

Smaller ones are sometimes designed for an external motor drive. I have several which fit the de-facto "Masterflex" peristaltic pump head mounting pattern. That's a rectangle of screws around a drive shaft which looks like a stubby flat screwdriver. The pump has a shallow slot across the face of its drive shaft.

The pumps listed on EBay are almost all designed for specific applications often mentioned in the listing. If you find a "Vacuubrand" pump for an affordable price, look very carefully, because -some- of them are teflon-body teflon-diaphragm vacuum pumps which tolerate acids... check the model number, condition, etc. very carefully because they are in great demand!


Picric-A - 28-6-2010 at 08:15

I would consider the diaphram pump i have much smaller than most other pumps and it doesnt need and external motor drive, just plug it in and go! I would highly reccomend the Charles Austen Pump Model B100 SEC as i so far havnt had a problem with it and works well for filtering/distilling however i havnt tried it with and corrosives as i have no clue what the diaphram is made of :(

watson.fawkes - 28-6-2010 at 11:41

Quote: Originally posted by densest  
The pumps listed on EBay are almost all designed for specific applications often mentioned in the listing. If you find a "Vacuubrand" pump for an affordable price, look very carefully, because -some- of them are teflon-body teflon-diaphragm vacuum pumps which tolerate acids... check the model number, condition, etc. very carefully because they are in great demand!
As I recall, the Aro brand pumps that Ingersoll-Rand sells have valve-and-seal sets that are completely replaceable, so that chemical compatibility can be upgraded in the field as needed. Of course the pump body itself has to be compatible or else this doesn't matter. Grainger carries this brand, so parts are readily available in the US. Replacement diaphragms are available in PTFE for many (most?) of these pumps.

Picric-A - 30-6-2010 at 05:06

I have just found out from the manufacturers, the diaphram of my pump is made out of EPDM rubber and the o rings are made out of Nitrile rubber.

Does anybody know how chemical resistant these two rubbers are and if i can safely use this pump for things like DCM, Ether, EtOH, Acetone ect... (generally any solvent!) cheers

altech6983 - 30-6-2010 at 07:17

From Wikipedia:
"EPDM exhibits satisfactory compatibility with fireproof hydraulic fluids, ketones, hot and cold water, and alkalis, and unsatisfactory compatibility with most oils, gasoline, kerosene, aromatic and aliphatic hydrocarbons, halogenated solvents, and concentrated acids."

This is a table from another site:
This Link To Table contains a list of chemicals and their compatibility to EPDM rubber.