Detection of hydroxyl radicals

Discussions about GC and other "gas phase" separation techniques.

15 posts Page 1 of 1
Hello everybody! I am a physicist and I don't know so much about gas chromatography. I am working with plasma in interaction with water in order to generate hydroxyl radicals.
I am using GC-MS and I would like to know which column I could use. Firstly I want to separate the air gases (Oxygen, Nitrogen) from the Argon and the water. Then I want to see this OH· radicals or even ozone and H2O2. The current column does not separate anything, everything come in one peak.
Thanks for your help and best wishes!
The reactive nature of the hydroxyl radical is going to cause you many problems when it comes to getting it through the GC column. I don't see that you'll have success with the direct approach you're suggesting.

If it were me, I'd find a way to trap the hydroxyl radicals. Make them react with something else that will give you something you can determine by GC.

Years ago we evaluated a technology that was supposedly generating free hydroxyl radicals, blowing them around the room, and killing microorganisms on the surfaces in the room - by merely circulating the air in the room. We didn't find it to be incredibly efficacious in its intent BUT as part of our testing, I created a gas standard of toluene in nitrogen. We fed it my gas standard and then collected a gas sample coming out of it. Sure enough, when you were standing in the outlet stream of the device, you could smell the "barn yard" coming out. Hydroxyl radicals react with toluene to make cresols (meta and para isomers, also very stable). So, the device was generating free hydroxyl radicals but we didn't find much evidence for disinfection of the room. As I recall, the yield was not great but the residence time in the device wasn't long.

O2 and N2 are not difficult but you need a specialized column (molecular sieve) and I'd recommend a pulsed-discharge helium ionization detector to see them. A thermal conductivity detector will also work for this as long as the required limit of detection is not too small. Mass spec is tough because you almost always have as little bit of O2 and N2 showing up in your tune. It's difficult to get an absolutely perfect seal.

I did find a reference for determination of ozone by GC:

https://pubs.acs.org/doi/abs/10.1021/ac ... ode=ancham

H2O2 presents the same problems as hydroxyl radicals. I'm not optimistic about that one either.

My $0.02, is that something like what I describe here will likely be your best bet. To do this right, I fear that it's not an easy undertaking.
Thanks for your answer!
So as far as I understood hydroxyl radicals are too reactive so they react before they pass throuh the column? Or is it another reason?
The main problem is that we only have an MS-detector attached to the GC. I can try with OES or any other technique, but using GC implies using MS and vice-versa.
That is what I'm saying.

One of the basic premises of GC is that you assume all you're doing is vaporizing the sample and the analytes therein. When chemistry occurs during the vaporization/transfer-to-and-through-the-column process, all bets are off. I've never seen a reference that indicated you could get hydroxide radicals by GCMS. I've been in this game for 27 years.

For instance, if you have a 5-phase (DB-5, Rtx-5, etc.) in your GC system, there will be 5% of the stationary phase that is benzene rings. They will undergo the same chemistry that I described above for toluene in the disinfection gizmo - thus trapping your hydroxide radicals and now incorporating them into your stationary phase. They'll be in there forever. In every scenario I can picture, there will be some sort of chemical reaction between the hydroxide radicals and the stationary phase.

Most people who are looking for free radicals in the condensed phase use a technique called electron paramagnetic resonance (EPR or ESR, electron spin resonance) spectrometry. EPR won't do much for you on your other analytes. You're going to need a number of different analytical techniques to do a good job getting all of that analytes you've listed.
Thanks again for your answers, they are very useful.
I think we are going to try to make the OH radicals react with toluene to see what happens. We have the following columns:

-Agilent 123-1065 DB-1. Length: 60 m, I.D. 0.32 mm, Film: 5 micrometres.
-Agilent 19091Z-716 HP-1. Length: 60 m, I.D. 0.32 mm, Film: 5 micrometers.
-ValcoBond VP-Molesieve 5·A. Length: 30 m, I.D. 0.32 mm, Film: 30 micrometers.
-Restek Rtx-5MS (Crossbond 5% diphenyl - 95% dimethyl polysiloxane). Length: 15 m, I.D. 0.25 mm,Film: 0.25 micrometer.

Which one do you think could be the best for this purpose? I am sorry if my questions are too general, but I do not know so much about CG.
Best regards.
The molecular sieve is only good for the fixed gases. Nothing else will come through it. You have to go really hot with your oven temperature to get CO2 to come through it. I wouldn't put a molecular sieve into my mass spec anyway.

I punched some numbers into the Agilent flow calculator with that 15 m Rtx-5 phase. It is too short to run in a GCMS system (outlet of the column is under vacuum). The vacuum pulls too hard on the inlet so you can't really set a head pressure on the column and achieve a somewhat optimum flow rate. Your 1-phases are pretty thick. This makes your retention times longer and the bleed into your MS will be greater. But, one of those will probably be your best option if you must work with what you have in inventory.

If you have funding to buy another, I'd get an 60 m Rtx-5 (either 0.25 mm or 0.32 mm in diameter) that has 0.25-0.5 µm film thickness. I searched for cresol in Restek's database (they make the Rtx columns) and found that in all cases, the 5-phase was used. In my own experience, I have had good success getting cresols with my Rtx-5. I also have a system that has a wax phase (Restek = Stabilwax) in it. That's a great phase for phenols.

I don't have a 1-phase in any system in my lab or I'd try it for you. Do you have any idea what concentrations of the phenols (you may get the ortho and para isomers) will be formed? Once formed, do you have any idea how you're going to get them out of the water and into the GCMS system?
Hello. I will try to buy another column, but meanwhile I will try with the 1-phase. Do you know the difference between DB-1 and HP 1?
I don't know which concentrations of the phenols we are going to get, I will ask if the group has done something like this before.
We are using a plasma inside a chamber in which OH radicals are formed. This chamber has a pumping system which takes the gases inside to the GC (in form of gas obviously). So I was thinking to introduce Toluene inside the chamber and take samples with the pumping system. I could also use a gas syringe but I do not understand from where is coming the water you were talking about.

I am sorry if I am disturbing you but I am new in this topic. Thanks again for your help.
Ah. From the original post, I somehow got it in my head that you you were generating the free radicals in a condensed phase (water). Gas phase makes it easier to sample for the stable materials.

The number 1 means all polydimethylsiloxane stationary phase. DB = J&W Scientific. HP = Hewlett Packard. So, it's all about the vendor. Actually, Agilent was spun off from Hewlett Packard years ago and Agilent bought J&W Scientific. So, those columns all come from the same vendor anyway. Rtx-1 is the same phase made by Restek, ZB-1 is made by Phenomenex, etc. Mostly, the column-coating technology is the same for all so you likely won't see a dime's worth of difference between them. The 5 (DB-5, HP-5, Rtx-5, ZB-5) means there's 95% methyl and 5% phenyl in the stationary phase. This phase is slightly more polar so the cresols will chromatograph a little better.

http://www.restek.com/Chromatography-Co ... s-by-Phase

Cresols give a wonderful molecular ion (m/z = 108) in the mass spectrum. Once you collect your total-ion chromatogram for your sample injection, you can go back and sort by 108 to make most everything else vanish.

Can you offer a little more detailed background on the project? What is the goal of all of these GCMS measurements? What do you hope to learn about your system? Don't give away anything that's "super top secret" but if I know more about what you're after, I might be able to offer suggestions on how to calibrate your system for quantitative analysis.
I did a quick experiment in a vial. I took 2, 20 mL headspace vials. I sealed one with a rubber septum (aluminum crimp cap) and to the other I added 1 drop of what the bottle says is 50% H2O2 in water to the other and sealed it. The peroxide solution is very old so it may not be 50% any more. Then to each vial, with a gas-tight syringe, I added 5 mL of vapor from the toluene reagent jar. I then cooked both vials in a 100 °C dri-bath heater for 1 hour. I analyzed the vapor space in the vials (used a technique called solid-phase microextraction).

I was able to detect the cresols in the vial containing the peroxide but not in the other (no peroxide). However, the yield of cresols was not fantastic.

https://1drv.ms/b/s!AkH-uI0tnY5LfiOwlv_eIZPiy3Y

Maybe toluene isn't the best "trap" for the hydroxyl radicals. It does trap at least some of them but the yield apparently isn't great. I don't have a good idea for what to use instead.
Of course I can explain more in detail our experiment:

We create a plasma with a TIA (Torch Inyection Axial), which basically creates the plasma with a gas and microwaves in order to maintain the discharge. This plasma is created inside the chamber I told you before, from where there is a pumping system which go to a loop in the CG to take the sample.
The gas used to create the plasma is Argon. The line of Argon is divided in 2: one go directly to the TIA and the other pass through a bubling vessel with water before going to the TIA. The idea is to change the concentration of H2O in the plasma changing the flow of both lines. This water in contact with the plasma is supossed to create some OH radicals. The main goal is to measure this OH radicals since they can be used for many purposes (cleaning air or liquids). So I though to add to the regular line of Argon another bubbling vessel with Toluene and see what happens.

One question: in your Mass Spectrum for the 21.9 min, do you get only the element 108? What about the other masses (90. 79. 63...)? How do you get all the different component of this spectrum? Do you compare with NIST data base?

I will try to summarize the data and funding we have to make the next step.

Best wishes!
You probably don't want to pass straight toluene vapor in there. You'll overload your GC system. With GCMS, a little goes a very long way. Perhaps you could get a toluene standard in argon (more dilute than straight reagent) and use that to feed one of the lines?

What I displayed is the mass spectrum at the apex of the 21.9 minute peak. It's a very good match for a cresol isomer. You can't really say which one because isomers like this all have a very similar fragmenation pattern. For true identification, you need to inject the authentic material and match the retention time as well. It's not the NIST database but it is very similar.

The mass spec detector can operate in 2 ways: scanning and single-ion monitoring. In scanning mode, the detector scans rapidly throughout the chromatographic run, collecting spectra all along the way. When you're searching for an unknown, this is how you operate because at the end, you can match the spectra with libraries to identify the material. If you are looking for a particular material (say it's 5 parts-per-trillion trichloroanisole in water), you can run in SIM mode. The noise goes to almost zero so the signal-to-noise ratio for your analyte becomes very large, allowing for detection of very small amounts. The problem with SIM is that you only see the ions you specify. Everything else disappears.

Calibration of this type of system for quantitative analysis could be very challenging. It may be able to tell you some conditions are better than others for generating the hydroxyl radicals but you likely won't be able to determine the yield of hydroxyl radicals for a set of conditions.

Good luck! It's an interesting project.
Hello Rb6banjo.

At the end we made some changes in our project. But it was thanks to your idea of using toluene in order to detect the product of the reaction toluene-OH. I wanted to thank you.

The project is as follow: I put some water (4ml) and toluene (1ml) in a small glass. Since the are immiscible, and the toluene has lower density it will remain on the top. Then I treat this mixture with an Argon-plasma coming out from a small tube in T-shape. This plasma is very good at degrading toluene. Afterwords, I treat the mixture with the plasma for few minutes (1-2). Then I take 1 microlitre of toluene after treatment and I inject it with liquid chromatography. The results were nice since I got some interesting peaks of these toluene-OH products: Benzaldehyde, o-cresol, p-cresol and Benzylalcohol.

Then I am making studies changing some variables: the dependence with the power input for the plasma, with the flux of Argon, with distance from the tube to the liquid and with the treatment time. I measure the areas of the peaks and the results are more or less good (good lineal fits).

The thing I wanted to asked you is about the repeatability of the measurements with these areas. I am making 3 repetitions of each measurement. Some times this repeatability is good (around 5-10% of standard deaviation), but some times is really bad (30-40% SD). I don't know the reason of this and if this is normal since as I told you I am pretty new in Chromatography. I read something about internal pattern.. do you something about this? Do you think is a good idea to take into account?

Thanks again!
You can try to add what is called an internal standard (IS). An IS is a material that you add to your sample that is not reactive and helps to normalize out variability in injection process.

For instance, you add and IS to your sample and inject. The ratio of Asample/Ais should be very constant, even if the peak areas for each material vary.

Asample = RF1*Csample
Ais = RF2*Cis

A's are the peak areas, C's are the concentrations, RF's are the response factors for each material in your detector. Divide one equation by the other and you get:

Asample/Ais = RRF*Csample/Cis

RRF (RF1/RF2) is the "relative response factor" of the analyte to the internal standard. A plot of the ratios Asample/Ais vs. Csample/Cis should give a line and the slope of that line is RRF.

Csample = Asample/Ais*(Cis/RRF)

This approach can help to minimize your variability. I don't know how you'll add an IS to your system because no matter what you add, the plasma is likely to modifiy it.

Maybe you can pull your aliquot and do a dilution in your syringe somehow? That could be very tricky and it might not be very precise.

Perhaps you can take an aliquot of the sample, shoot it into a solvent (methylene chloride perhaps) that contains a known concentration of the IS. You could use this approach if you're not having any trouble detecting the analytes and you can live with a little dilution.
Thanks for your quick reponse!

I will think about the options you told me, since I tried diluting the sample in a solvent and the peaks were not so good. But it is very interesting your point about IS so maybe I can try somehow, since, as you said, the plasma would change it composition.

Anyway, maybe I will try your idea with the solvent. Any suggestion of what compound could be this IS?

Best regards
It could be anything. The requirements are basically that the IS is chemically stable, chromatographs well on your column, and under your chromatographic conditions - elutes in a place where there's nothing else. Make sure that the concentration of the IS is at least somewhat similar to your analytes. If you expect that your analytes are 25 ppm, then make the IS somewhere in that range (not 2,500 ppm).

My "go-to" IS for analysis of beer is methyl heptanoate. Methyl esters are not common in beer and it happens to elute in a place where there's not a lot of other stuff.

What if you tried to make your IS in toluene. Pull up 1 µL of your sample, pull a small air slug to separate the 2, then aspirate another known volume of your IS solution. Inject it all in one shot. Let the GC inlet do the mixing for you. I've never tried something like this so I can't say that it'll work. It will certainly be easy to try.

Be cautious that you're not over filling the inlet. Agilent has a "vapor volume calculator" that will help you decide how much solvent you can inject given the geometry of your inlet liner:

https://www.agilent.com/en/support/gas- ... alculators

It's a freebie thing. Overfilling the GC inlet could also be the source of some of your precision problems.
15 posts Page 1 of 1

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: No registered users and 1 guest

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food & Beverage, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry