Dear GC users,
We are contemplating purchase of gas generators to replace cylinders used to supply three varian 3400 GC FIDís plus one HP5890/5971 GCMS. Currently we use capillary and megabore columns for analysis of FAMEs, alkanes, alcohols, and some other general work. We use helium as a carrier gas, but if a hydrogen generator was purchased for the flame gas, it would be beneficial financially (no helium costs + faster runtimes) to switch the carrier to hydrogen also.
I am worried about the chance of explosion from hydrogen leakage into (a) the lab, and (b) the ovens, if a column breakage/loose fitting.
The lab is quite large and constantly vented through fume hoods so is danger (a) pretty minimal? I believe if a major leak from the generator into the lab occurs the generator would shut off too. In the event of (b), I donít know if the generator would shut off at all or in time to prevent buildup in the oven though? Would the oven fans remove the oven contents preventing explosion in the event of (b), and are there different considerations for the different instruments we have? Is the main problem if such a leak occurred whilst the GC is off but carrier on?
I think in the event of a power failure to GCs/generator the hydrogen generator would not restart until reset by a user Ė as opposed to cylinder hydrogen, which would continue to flow?
Are there additional precautions that can be recommended to improve safety?
I am in Australia and the one supplier I know of is Domnick Hunter. Does anyone know if other makes (Whatman, Packard, Ö.) are readily available in oz? Anybody with good/bad stories after using these or other makes?
Thanks in advance to anyone who is able to help with any of these questions
Regards, Andrew Poole
By Anonymous on Thursday, September 26, 2002 - 06:42 am:
I don't believe the lab-scale electrolytic generators have enough flow capacity to allow a flammable mixture to build up before diffusion removes it, even if the oven fan is off.
Domnick Hunter, Whatman and Packard have similar performance. You will always find users with some problems while others have had no problems with any of these. Perkin Elmer support Whatman, which may not suit you. Domnick Hunter and Packard might be better if you do your own maintenance.
What may be critical for users is - what happens if the generator fails and you can't fix it? Are you going to have a spare or can you wheel in a reserve H2 cylinder for a few days. Many labs are now forbidden to use H2 cylinders whether piped from inside or outside, under any circumstances. This makes you very interested in the supplier's maintenance deal.
By Anonymous on Thursday, September 26, 2002 - 07:41 am:
Andrew could always plumb-in a helium cylinder if his hydrogen generator went out. But first contact Agilent, maybe even on the web. First read their literature on using H2 as carrier for GCMS (we decided to stay with helium here). But definitely a good idea for your FID units, for the FID as well as for carrier, split gas, etc.
By Benjamin on Thursday, September 26, 2002 - 10:32 am:
I am not an expert on these matters but I would like to let you know about a Restek technical note I read years ago (by K. Grob) in which a thorough discussion of the subject was presented. Unfortunately I can not locate my copy of it to give you more details.
One item I remember well, is that is not easy to achieve the dangerous H2 level, under normal operation and ventilation conditions, to produce an explosion. Even when tried on purpose. I also remeber an ingeneous safety precaution you can take. This is a "reservoir tank" that you can fill from a regular cylinder. This reservoir can be installed on line and it is required to be of the right size to contain only the amount of H2 needed for lets say, 12 hrs of operation. This way even if you have leaks, there will never be enough H2 available to reach dangerous levels.
Check with your suppliers, most likely they can offer you this option, please some other safety precautions such as oven alarms, etc.
By Andrew Poole on Friday, September 27, 2002 - 12:40 am:
Thankyou all for the responses,
all very helpful. Anon1, we are not (yet!) barred from having H2 cylinders, so in event of generator problem could run with bottled He &/or H2 till generator repair.
Benjamin (and others interested) thanks for that pointer to the Grob article(s). I managed to locate them on the Restek website, this article and others are very interesting - i've spent the afternoon reading them! (http://www.restekcorp.com/advntage/konigrob.htm).
Regarding the reservoir tank, i think this only a precaution if still using bottled H2, as the generators i have investigated only have minimal stored H2 at any point in time, instead supplying as required.
There are some other very useful safety pointers referred to in the Grob article, including fitting of H2 sensors to oven ventilator which, following a 1% H2 level, automatically shut down heating and so purges the oven with air (or even causing a purging of the carrier line with N2 !!)
Thanks again for your useful comments,
any other comments welcome too!
By H W Mueller on Friday, September 27, 2002 - 07:43 am:
There was a discussion on this in
under "carrier and makeup question".
By Harvey G. on Saturday, September 28, 2002 - 12:42 pm:
I switched from helium to hydrogen as carrier gas and used it for 2 GCs. One was exclusively for identification and concentration determinations of alcohols and low level concentrations of dioxane by headspace. The other GC used an autosampler and a variety of quick connect columns for determining carbon chain distributions and deriving mean molecular weights for samples of fatty acids as methyl esters and alkylbenes, etc. Switching from helium to hydrogen as carrier, allowed me to change the make up gas from helium to nitrogen. The end result was an increase in response by a full magnitude. I am told that this was due to better separation in the column with hydrogen and better signal with nitrogen as make up gas at the FID. I did run into a problem. When I made the proposal, I suggested the Matheson hydrogen generator. It permits only hydrogen to reach the chromatograph. My supervisor got a better price for a unit that had a large container of drierite to remove water from the generated hydrogen on its way to my column. I had to spend a fortune on a system of filters and indicating columns placed between the generator and the injector to protect the column from water, oxygen and other contaminants. Every once in a while the generator failed due to the water in the tank being detected as unacceptable. I was using, city water that had been filterd, glass distilled and then polished with an Elga unit until it was in the 18.2 megaohm and extremely low TOC range. They had these silly little DI teabags hanging in the distilled water feed tank to soak up any ions from air contamination. Each time the system failed it took out all of my in line filters. I finally disconnected it and used ultrapure hydrogen in tanks. It worked better than the generator, was more dependable, easier to identify when it is going to run out and easier to replace. If you can get a dependable one such as that designed by Matheson with a filter that only allows hydrogen molecules to pass through, get it. I am sure that you can contact Matheson through the interet for information. They are in Canada and I assume in the USA. Otherwise use cylinders of pure hydrogen. I would never go back to helium.
By Benjamin on Monday, September 30, 2002 - 06:39 am:
I am glad you located the Grob article. It seems like know you have a range of options to implement safety measures.
Good Luck again.
By Anonymous on Monday, October 7, 2002 - 07:57 am:
The Wathman Hydrogen generator work similarly to Matheson generator cited by Harvey G.
It deliver hydrogen through a Pd membrane and so purity and dryness are assured.
We have two Wathman hydrogen generators by several years in our lab; annual maintenance is easy and performed in house and, at present, no matter for failure.
Good luck for choice.
By Andrew Poole on Wednesday, October 9, 2002 - 11:09 pm:
Thanks again to all posters for your useful information
By Jack Klein on Monday, October 14, 2002 - 06:08 am:
We've used Whatman hydrogen generators with great success and tried others. We found that the dessicant units like the Dominick Hunter and Packard drastically shortened the life of our columns. I guess I echo Harvey's comments to a degree. We looked at Matheson too but decided against the risk, they have not been making hydrogen generators very long.
By russ_uk on Tuesday, January 7, 2003 - 05:54 am:
We have Chrompack Hydrogen Safety Systems installed in all our GC's in which we have hydrogen as a carrier gas. These instruments cost about 3000GBP per unit and monitor the hydrogen concentration in the GC oven, we have configured these units to cut the power and switch the carrier over to nitrogen if the measurement reaches 25% LEL (1%vol H2).
We're probably going a bit far, and these units do occasionally shutdown the GC's in error, but we do feel a lot safer with these installed.
By Mark De Roy on Sunday, June 13, 2004 - 11:32 pm:
Could Russ UK get in touch as I have a Chrompack unit to install to a GC with no Manual and am interested in the swithing of carrier gas to nitrogen to make safer.
Contact on email@example.com