N2 for GC-MS "Rest" Gas

[benhutcherson]

At the moment, I have 6 GCs run off of a single He cylinder. Four of them are Buck/SRI TCD-FID instruments that are used infrequently(although heavily when used), and are plumbed separately from the GC-MSs. The SRIs get powered down completely when not in use, and have their helium supply shut off completely.

The GC-MSs go through periods of heavy use, and then at times might not get used for a couple of weeks at a time, or maybe get 1 sample a day and then go back to idle. Both of them(one a 5890/5971, the other a 7820/5975) are set up with a "rest" method that flows .5mL/min He and switches to splitless mode when not being used. Still, though, they can go through a fair bit of He when not being used(I'm at about 4 months on a cylinder, when I'd like 6-8 months).

Not too long ago, I was reading about the 8890/5977, and one of the features mentioned was the ability to switch to a less expensive gas when resting.

That got me to thinking that it might be possible to do something like that in my set-up. In my mind, it should be possible to put a selector valve on the carrier gas line leading to the GC-MSs, and then just plumb He to one side and N2 to the other side. Since loading a new method(our instruments are used by several different research groups and for several different projects, so we have probably over 100 different methods on them) usually takes several minutes for the temperatures to equilibrate, that would also give time to flush all of the N2 out of the lines and switch over entirely to He.

I haven't checked on the prices of UHP/Zero Grade N2, but I know I pay about $10 for the cheap stuff, and I'm sure UHP or Zero has to be less expensive than the $200+ I'm paying for zero grade He.

The long term plan is to switch over to H2 for the GC-MSs, but I have too many PhDs in progress to get away with that now. So, in the interim, does anyone have any thoughts on the above plan?

[LALman]

I do this for my backup GC-MS instrument. My workhorse is a 6890/5973 and the backup is a 5890/5971. The gas fittings on the 5890 tend to leak easily. So, I let them leak N2 instead. I keep a dewar of N2 for purge gas, makeup gas, and long term storage gas. Just realize you need some time before the switch back to He is complete. I usually vent the line to clear the N2 out of the line when I want to switch back to He to run. I do it the night before and its ready to run the next morning.

[James_Ball]

I have a setup on my 7890/7000C that I use to research the switch to H2. I just put a three way valve (two in, one out) just before the connection to the instrument. The one out line goes to the instrument, then He is plumbed to one inlet and H2 to the other, just turn the knob to select the gas, and it only takes a few minutes to flush what is in the controller and lines to the inlet. If you really want to save money over time, get a N2 or H2 generator. N2 generation would be very inexpensive if you have compressed air already plumbed into the lab.

[benhutcherson]

Thanks guys. Sounds like this is viable and a good move.

I have a compressed air line in the room, but it would take a bit more work to reactivate it. There aren't currently any free taps down in the air room(the NMRs suck up a lot of them) although that could be added. If I used the 5890 FID/NPD more often, I'd do it, but it's used infrequently enough that bottled air is fine(the SRIs all have built-in air compressors, one of the really nice little features of these instruments). I'm also concerned about space, power, and noise for an N2 generator-the first of those especially as the room is already rather cramped. Still, though, having a ready source of N2 would also be handy since one of the SRIs occasionally gets switched over to N2 for H2 analysis.

So, it sounds like I just need to buy/scavenge some plumbing(probably buy to make sure I'm not introducing contamination) and get this done.

[James_Ball]

This is the valve I used.

https://www.restek.com/catalog/view/46983

[benhutcherson]

Alright, I called today and got both UHP N2 and a new regulator ordered. I'm sure I could dig up a CGA 580 regulator, but considering the intended use I figure I'm better off starting with a new clean one. BTW, at $16/cylinder, I suspect a generator would take a long time to pay for.

I got the order list from McMaster-Carr put together today, but I missed the person with the ability to actually order it so I'll get that done tomorrow and hopefully get everything done Thursday. Of course the valve was the big thing I needed, but I figured it was a good time to stock up on 1/8" Swagelok nuts and ferrules also, plus a few other odds and ends.

[Consumer Products Guy]

Sounds like a good plan, Ben.

[LALman]

Hmmm, I pay $77 for a 192 CF UHP Nitrogen not including hazmat and delivery. You must be at a university.

[benhutcherson]

Yes, I am at a university.

As best as I can tell, our gas supplier only charges a small mark-up for common UHP gases vs. a lower grade. I think lower purity N2 is in the $14/cylinder range.

[James_Ball]

I think we have paid nearly $90 per N2 tank before, that is why I mentioned the generator. We were getting payback of the generator in less than a year. At your cost the tanks make more sense.

[Steve Reimer]

James, what is your maintenance on the nitrogen generator?
We run a large one that feeds several (four) LC/MS systems and it runs ~$2500 per year.

For low pressure/low flow we use bleed off from our 2000L liquid Dewar.

[James_Ball]

James, what is your maintenance on the nitrogen generator?
We run a large one that feeds several (four) LC/MS systems and it runs ~$2500 per year.

For low pressure/low flow we use bleed off from our 2000L liquid Dewar.

We just have one LCMS right now and went to the generator that uses the in house compressed air, membrane technology.

For the GC and TOC instruments we were using the Parker Balston PSA unit that produces 1200ml/min from in house air. That unit went down on us as since our demand I think is above what it can deliver now, we have begun to consider the bulk tank for UHP N2.

I haven't looked at the maintenance costs but the number sounds similar to what you list. We are currently going through 3-4 cylinders of N2 per week so at 3 per week and $70 per cylinder that comes to $10,920 per year just for UHP N2 which would easily cover yearly maintenance just on the PSA unit. But at current usage we probably need two of them, I think the reason we had two valves fail so quickly was because we are drawing more than it can supply(someone in another department added an instrument that uses quite a bit without telling us).

[LALman]

I keep an emergency 192cf UPH Nitrogen cylinder on hand. I pay $143 for a 230L UHP dewar plus $57 for delivery charges. Its short and squat and barely fits through the door way. It lasts me a couple months and is roughly equivalent to 23 cylinders of compressed gas. More of it leaks than I use (typically 1.5%/day), but its far cheaper than a bank of cylinders. But they get me on tank rental $3.73/day ($115/month). So call it about $215/month for all the gas I can use.

I use it as 8260 purge gas, GC-FID makeup gas, and standby gas.

[benhutcherson]

The valve is installed and working fine.

An experiment this morning showed that things were in good shape ~1 hour after switching back over to He, so I'll put that as part of the instructions. Of course, switching to a typical running flow rate(generally 1mL/min for our .25mm columns) should speed that up a bit more.

[James_Ball]

The valve is installed and working fine.

An experiment this morning showed that things were in good shape ~1 hour after switching back over to He, so I'll put that as part of the instructions. Of course, switching to a typical running flow rate(generally 1mL/min for our .25mm columns) should speed that up a bit more.

Set it to He and column flow of 1ml/min and split ratio to 100:1 and it will flush out in just a few minutes. If you put it into manual tune and scan until the N2 drops to almost nothing you know the flush is complete.