It’s a confusing time for GC/MS analysts

I don’t know about you, but the more information I hear about switching from helium to hydrogen as a carrier gas, the more confused I get. When the helium shortage started, there were articles stating that this was a minor hiccup in the processing/supply chain, and capacity & inventories would be restored by the fall. Well, winter is (almost) here and supplies are still tight.

Then came along articles stating that we all were using much more helium than was capable of being produced, and it should be rationed to those industries whose livelihood depended upon it. Many scientists began calling for the end of helium-filled balloons, and other non-essential uses.

Eventually, slight panic set-in among chromatographers, and laboratories started to consider switching from helium to alternative GC carrier gases. The gases which were considered were nitrogen, argon, and hydrogen.  Unfortunately most mass spec manufacturers will adamantly state that you do not want to use either nitrogen or argon as carrier gas because sensitivity will suffer (due to the large size of the nitrogen and argon molecules), so the only option is hydrogen.

At that time, most articles focused on the positives of hydrogen (faster analysis time, more column efficiency, etc…) with the only negatives I remember reading about were:

A. Safety (as we all know, hydrogen is combustible).

B. If you were using old copper tubing to plumb the carrier gas, it was recommended to switch to new, clean stainless steel tubing (supposedly hydrogen can make older copper tubing brittle). If you were already using stainless steel tubing, but with other gases, it was recommended to replaced the used tubing with new, clean stainless to eliminate the “scrubbing” effect of hydrogen.

OK, so when switching to hydrogen, make sure you do not have any leaks and safely vent the carrier gas into vents/exhausts – check. Don’t use old copper tubing – check. Don’t reuse stainless steel tubing which was used for helium or other gases – check.  All will be well.

Things actually seemed to be headed in the right direction, and having hydrogen as a substitute carrier gas seemed like a done-deal. For most GC detectors, it probably was a done deal, but then came along several issues involving mass spec. Customers reported that right after the switch from helium to hydrogen, their baseline was very noisy, and ghost peaks filled their chromatogram. Others reported loss of compound sensitivity, and some reactive pesticides were disappearing altogether. Still others reported that compound peaks were tailing, and column lifetime decreased. Older instruments seemed to have the majority issues, with several not even able to pump-down properly. But, when they switched back to helium, everything returned to normal. Now it started to look like this was not going to be an easy switch.

I am happy to see that instrument manufacturers are taking the helium shortage seriously, and have provided webinars and other useful information to help customers make the switch to hydrogen. Overall, I feel they have done an excellent job. If you haven’t watched any of their webinars, or read any of their published information, I encourage you to contact your instrument manufacturer to see what they have available.

However, several important questions continue to go unanswered. Maybe I’m just an impatient person, and maybe I’m just too detailed-oriented for my own good, but there are a lot of “Why’s” that I still need answered. Maybe, hopefully, that will change in 2013. Below are the most common questions I hear, but still don’t know the answer.

1. Why do they recommend against using methylene chloride as a solvent? I have heard that HCl vapors can be produced in the injection port, and if moisture is present, even trace levels of hydrochloric acid. If this is true, I would think that every instrument would be negatively affected. Why have I only heard about this using mass spec? I have also heard not to use carbon disulfide as a solvent because an end-product (if moisture is present) can be sulfuric acid. Can the same be true for other chlorinated solvents (like chloroform) and sulfur-containing solvents (like dimethyl sulfoxide and carbon disulfide)?

2. It has been stated that no gas traps should be installed on the hydrogen carrier gas line to further purify the gas – why? I have heard of a “scrubbing effect”, but what exactly is getting scrubbed out of these traps (even ones designed for use with hydrogen)?

3. I’ve heard that fragmentation patterns will be different from those published in the NIST library. I have also heard that fragmentation patterns will be the same as those published in the NIST library. If the patterns are different for a particular instrument, is there any way to correct this through mass spec tuning (or other methods), so they once again match the NIST library?

4. I have heard you should use metal columns in your GC (to prevent breakage). I have heard that you should not use metal columns (because of possible arcing issues within the source). Are there certain models of mass specs where arcing is a problem? If so, I would like to know which models.

5. I have heard that injection port liners need to be replaced every day. While this may already be common for certain analysis, I’m getting the impression this will need to be done for all analysis when one switches to hydrogen. If so, why? What is occurring in the injection port that damages the inertness of a liner so quickly?

6. What exactly is getting “scrubbed” from the gas lines, source, injection port, ect… when one switches to hydrogen? I have heard that the fragmentation pattern resembles hydrocarbons. So, is it really hydrocarbons getting scrubbed from the gas lines, or something else?

7. What is causing peak tailing? If there are no leaks, and typical linear velocities are faster with hydrogen, why are certain peaks tailing?

Believe it or not, the last thing I planned was to write yet another post on switching from helium to hydrogen as a carrier gas until I had answers to the questions listed above, but hearing the frustration from our customers has been a difficult pill for me to swallow, because as a member of tech service, it’s my job to have answers. Honestly, I’m not interested in hearing statements like “Here is what I think is happening”, or “I have heard that this is happening”. I am only interested in the cold, hard facts, and the data to back it up. I do realize that no one may have these answers yet, but if that’s the case, just tell us. I would rather have someone state in a webinar “Don’t do that, but we don’t yet know why”, than simply “Don’t do that”.

So in summary, I (we) do feel your frustration. Hopefully we will get some answers in 2013, and end this confusing time for GC/MS analysts. Thanks for reading.

This useful Application Note was published after I wrote this post.  I think many of you will find it as informative as I did.


6 Responses to “It’s a confusing time for GC/MS analysts”

  1. Ward D'Autry says:

    Dear Alan,

    Interesting questions for sure. Also in our lab we consider the switch from helium to hydrogen and some of these questions are also keeping us busy. Yet another problem that I “heard of” is the possible hydrogenation of unsaturated hydrocarbons, which may lead to false MS identification. Also here, we are curious for a good answer from someone that has really investigated this.

    About your first question (which is not related to hydrogen as carrier gas I assume? This problem exists with He too). I can confirm that injecting both methylene chloride or chloroform results in the presence of HCl in the mass spectra of the solvent peaks (presence of 36/38 3:1 ratio). Even with cold PTV injections, suggesting that HCl is already present as impurity in the analytical grades of these solvents. And yes, this is not solely a mass spec issue but your whole system suffers. Recently we did a test by filtering the solvents through basic alumina to remove HCl. It works very well. Although we might consider another supplier too :)

    Best wishes for 2013.

    Ward D’Autry

  2. Alan Sensue says:

    Hi Ward.

    Thanks for reading my post, and for your reply. The reason I mentioned methylene chloride, even though one would assume production of HCl vapors would not be the result of using hydrogen as the carrier gas, was that I personally had not heard of avoiding this solvent until I viewed a few webinars from a couple of instrument manufacturers. They recommended using alternative, non-chlorinated solvents when switching to hydrogen as the carrier gas. Maybe it is because one of the impurities from hydrogen generators is moisture, and since the use of gas traps (filters) is not recommended, this moisture can react with the HCl vapors to form hydrochloric acid. But, I am just speculating.

    Happy New Year.


  3. Romas Cesas says:

    What’s with the this “don’t use copper lines” with hydrogen? I have used hydrogen as a carrier gas for my GC’s for a few decades now and never saw an issue with my copper tubing. Is this just one of those over reactions? And as for helium in my MS, when I asked my supplier about rationing I was told I am OK and not to worry. I only use a little less than 3 cylinders a year and would prefer not to deal with the hydrogen issues on the MS. I tried hydrogen many years ago and saw the extreme noise on the baseline and promptly went back to helium.
    Also just a comment about hydrogen usage. It appears the scare mongers are still busy making it sound like hydrogen explosions are imminent. If you use a little common sense and do a bit of calculation you will find that the dangers are over blown and easily prevented. First the explosive limits are very narrow making it hard to have the proper air/hydrogen mix for an explosion, number two hydrogen dissociates very rapidly making it very difficult to get the proper mix ratio and finally the gas flow is rather low making it extremely difficult to achieve an explosive mixture. All this was evaluated back in the 1980’s and it was determined then that hydrogen use as a carrier gas was safe.

  4. Alan Sensue says:

    Hi Romas – thank you for reading my post, and for your feedback. You may want to review a previous post GC Carrier Gases – Alternatives to Helium , and more specifically, page 4 of the document below for information about copper tubing and brittleness when exposed to hydrogen. Since I wrote this post, I have seen references to what is called “Hydrogen Embrittlement” which is a term to explain the reaction of hydrogen with various metals.

    I will agree with you that I think the safety concerns may be slightly exaggerated (I personally would be more concerned about a natural gas explosion than a hydrogen explosion), but I do think that safe use of hydrogen needs to be taken seriously.

    Best wishes in the new year.


  5. Dave Macdonald says:

    I was forced to use hydrogen on a 5890 II/5971A since our vendor cut us off. Agilent recommended the stainless steel tubing and baking the source. Actually got Dr Cooks from Purdue on the phone (only expected one of his grad students), he confirmed that based on additional or changed ion ratios that too much hydrogen was in the source. Reduced column i.d. (20M x0.18 x0.18) and reduced flow to 0.6mL/min. This solved the majority of my problems and with some tweeking I was able to run SVOCs with hydrogen for several months. The problems that persisted were significantly decreased response and peak tailing of nitro-aromatics and others; the clue to what is going on is in the tail. Specifically the ions that are reacting in the source. For example Nitrobenzene, the principal ions in order of abundance, m/z 77, 51, 123, 65, 93, first of all with hydrogen m/z 93 will be much higher, looking at the tail the 93 ion increases in abundance relative to the others. Remember I’ve been using a 5971 and 5890 series II with EPC. This same effect is observed with Nitrobenzene-d5 m/z 98. In addition to adverse effects in the injection port one, must contend with different ionization patterns occurring in the mass spec. For 4,6-Dinitro-2-methylphenol additional ions m/z 108 and 80 show up and are more prevalent in the tail as well. Obviously you won’t get the same MDLs that you used to with helium at least for the more problematic compounds. The question that concerns me is that all the EPA GC/MS methodologies were written for helium. How will some regulatory agencies react to these events, even though you can explain and demonstrate the validity of your data?

  6. Alan Sensue says:

    Hi Dave – thanks for the information, especially concerning the tailing peaks. You, like many of our customers, are very concerned they will not be able to meet the requirements of many published methods if they are unable to use helium as the carrier gas. We both know that the EPA and other agencies are aware of the helium shortage, but I haven’t seen any recent method modifications/revisions acknowledging this situation. Honestly, I am not sure how laboratories who are not able to obtain helium are coping with less than optimal chromatography, loss of sensitivity, and changes to compound spectra. I wish you the best that you are able to resolve the handful of issues which still remain after switching to hydrogen.

Leave a Reply

Restek Domestic Customer Service



Your Full Name

Your Email

Company Name


Spam Block (Please leave this blank)

all fields required

Thank you

Your message has been sent. We will be in touch shortly.

Message not sent

Sorry, your message could not be sent at this time. Please try again later, or contact Restek or your local Restek representative via phone.