RT shift (and replacing columns on 3000 microgc

[emorris1000]

Howdy. First post. Word of warning I'm no GC expert. I've gotten stuck maintaining a couple of GCs on some test reactors I work on, and am learning but, like I said, no expert.

Anyways. I've been having problems for the last year or so of RT drift on one of my GCs, an Agilent/Inficon 3000 MicroGC. I've dealt with it by updating the method as I go, but this GC does online monitoring and the more I push back my RTs the worse my sampling resolution gets. Right now its taking almost 4.5 minutes between each run. The limiting factor is hexene which comes in ~4 minutes.

I have 2 columns, a molsieve backflush with a PLOTU precolumn and an Al column

Some info.

1) Columns are ~10 years old

2) Previous owner never installed driers on carrier gas (sample gas is dry). It exclusively uses UHP carrier gas, but I don't buy that that is enough. When I have an acceptable level of downtime I will be installing a couple of driers.

3) Previous owner installed 'home-made' in line filters which was an emptied check valve with a 3 um T-Type filter element held in place with a spring. (Surprisingly, previous owner no longer works here.)

4) To try and push RT times down I increased column temp/pressure. They were within the "safe" bounds given by Agilent, but right at the edge.

My guess is that RT shifts are due to column degradation. What's surprising to me is that the RT shifts have accelerated dramitically since I took over operation of the GC. I've seen almost a 100% shift in the last year. As far as I can tell the shifts were much slower in the years before. Only thing I can guess is that increasing the column temp/pressure rapidly increased the column degredation.

Also about 7 months ago we sent the equipment in for repairs to the OBC, while there the inlet ass'y was replaced, and EPC valve and pressure sensor was replaced for the molsieve/plotu column, and the detector and flow ass'y die was replaced for the Al column. Could any of these faults have hidden RT shifts?

What other possible causes can you think of for the RT shifts?

---------------

Second question has to do with replacing GC columns on a 3000 MicroGC. The column modules themselves are prohibitively expensive, roughly 5-7k each. At that price I am more likely to upgrade this GC to the Varian/Agilent MicroGC I have on my other reactor. According to Inficon, however, I should be able to replace the column itself without replacing the whole module, which should be substantially cheaper.

Anyone have any experience with this? How tough is it? I'm pretty mechanically oriented and have little difficulty working with 1/16" tubing and the like (I can tell the difference between Bucci and Swagelok 1/16" ferrels.)

-------------

Final question. May be better suited for the Software subforum. The Agilent/Inficon MicroGC uses Cerity, which I hear quite a bit of whinging about. I can make it work for me, and to be perfectly honest, it is FAR superior to the Star software we are using on my Varian/Agilent MicroGC.

We're planning on replacing the Star software (its almost older than me) and I am trying to decide whether to go with EzChrom or Chemstation (or maybe Chromeleon.) What I want out of it, more than anything else, is the ability to work with different manufacturers GCs. It looks like EzChrom is the way to go with that in mind, but I'm still not sure exactly what Chemstation will support (keep hearing different things).

-----------------

Anyways, Thanks for reading my book. I hope to publish it one day.

[Henri_sra]

Hello emorris,

maybe my colleagues can help you. My company is a specialised on MicroGC especially on 3000 MicroGC.
We also have a software designed only for MicroGC for Online analysis.

If you want i can give you our coordinate.

Regards
Henri - SRA

Howdy. First post. Word of warning I'm no GC expert. I've gotten stuck maintaining a couple of GCs on some test reactors I work on, and am learning but, like I said, no expert.

Anyways. I've been having problems for the last year or so of RT drift on one of my GCs, an Agilent/Inficon 3000 MicroGC. I've dealt with it by updating the method as I go, but this GC does online monitoring and the more I push back my RTs the worse my sampling resolution gets. Right now its taking almost 4.5 minutes between each run. The limiting factor is hexene which comes in ~4 minutes.

I have 2 columns, a molsieve backflush with a PLOTU precolumn and an Al column

Some info.

1) Columns are ~10 years old

2) Previous owner never installed driers on carrier gas (sample gas is dry). It exclusively uses UHP carrier gas, but I don't buy that that is enough. When I have an acceptable level of downtime I will be installing a couple of driers.

3) Previous owner installed 'home-made' in line filters which was an emptied check valve with a 3 um T-Type filter element held in place with a spring. (Surprisingly, previous owner no longer works here.)

4) To try and push RT times down I increased column temp/pressure. They were within the "safe" bounds given by Agilent, but right at the edge.

My guess is that RT shifts are due to column degradation. What's surprising to me is that the RT shifts have accelerated dramitically since I took over operation of the GC. I've seen almost a 100% shift in the last year. As far as I can tell the shifts were much slower in the years before. Only thing I can guess is that increasing the column temp/pressure rapidly increased the column degredation.

Also about 7 months ago we sent the equipment in for repairs to the OBC, while there the inlet ass'y was replaced, and EPC valve and pressure sensor was replaced for the molsieve/plotu column, and the detector and flow ass'y die was replaced for the Al column. Could any of these faults have hidden RT shifts?

What other possible causes can you think of for the RT shifts?

---------------

Second question has to do with replacing GC columns on a 3000 MicroGC. The column modules themselves are prohibitively expensive, roughly 5-7k each. At that price I am more likely to upgrade this GC to the Varian/Agilent MicroGC I have on my other reactor. According to Inficon, however, I should be able to replace the column itself without replacing the whole module, which should be substantially cheaper.

Anyone have any experience with this? How tough is it? I'm pretty mechanically oriented and have little difficulty working with 1/16" tubing and the like (I can tell the difference between Bucci and Swagelok 1/16" ferrels.)

-------------

Final question. May be better suited for the Software subforum. The Agilent/Inficon MicroGC uses Cerity, which I hear quite a bit of whinging about. I can make it work for me, and to be perfectly honest, it is FAR superior to the Star software we are using on my Varian/Agilent MicroGC.

We're planning on replacing the Star software (its almost older than me) and I am trying to decide whether to go with EzChrom or Chemstation (or maybe Chromeleon.) What I want out of it, more than anything else, is the ability to work with different manufacturers GCs. It looks like EzChrom is the way to go with that in mind, but I'm still not sure exactly what Chemstation will support (keep hearing different things).

-----------------

Anyways, Thanks for reading my book. I hope to publish it one day.

[chromatographer]

I didn't think the columns in modules could be replaced by the user. If you find that this is possible I'd be very interested in knowing how it's done. In my experience alumina columns shift retention times quite a bit. I don't think the UHP carrier gas would be the problem. You can reduce the time shifts by adding time at the highest temp at the end of the run to bake out. If you change columns their OV1 has very good RT stability but not as good separation.

Chemstation won't work with the 3000 micro's. Cerity was very difficult for me after working with Chemstation but I was able to finally get used to it. It is very good for continuous process monitoring. I don't know how EzChrom would be for that - never used it.

[emorris1000]

Just found the e-mail on it. According to them you can just purchase the columns (instead of the whole module) depending on whether or not it is the new "modbox" style. Still haven't followed up on that with them as determining this would require some dismantling of my GC which I haven't had the opportunity to do.

I'm already running the Alumina column at 160C, the safe operating limit is 180. As it is I am already running at the highest temperature I can really. I do try and burn it out at 180 every couple of weeks, but from what I have seen its not making any difference these days.

Just stinks that it takes almost 5 minutes to have my hexene elute.

And yeah, cerity is pretty good for automated sampling. The integration system leaves a bit to be desired. Recently been having problems getting it to properly integrate my hexene peaks.

Edit: Burned the system out last night (~12 hours or so) and my 1-Hexene RT went from ~3.5 to 3.7, I think its time to replace this thing. Not looking forward to justifying the cost