Chromatography Forum

Hi all,

I've been working with a USP Pharmacopoeia method for the determination of impurities but I've run into a few problems with the inlet pressure shutting down.

The method states a linear velocity of 25cm/s (at 50°C) with the carrier gas being Helium which is what we're using. The column details are, 30m, 0.32mm and a film thickness of 0.5μm.

The inlet conditions were setup as follows;

Temperature: 150°C
Column Flow: 1.4ml/s
Mode: Splitless
Septum Purge: Off
Purge Flow: Off
Purge Time: Off

With these conditions the inlet pressure is preset at ~7psi but the pressure would shutdown as the inlet pressure was unable to rise to this preset value. After some research on this forum and understanding the theory of GC inlet systems I have set the following conditions;

Temperature: 150°C
Column Flow: 1.4ml/s
Mode: Splitless
Septum Purge: 3ml/min
Purge Flow: 30ml/min
Purge Time: 0.5

The pressure is easily maintained with these conditions but I'm wondering if this is the correct approach to take? On a side note after performing a SST injection it was noted that the early eluting peaks had a symmetry factor of <1.5 but a peak just before the solvent peak showed shouldering and a poor symmetry factor of ~3.5. I'm not sure if my Purge Flow / Purge Time is a factor in this?

Thanks.

Those first conditions were the problem with the Purge Flow and Purge Time set to Off. With it set to Off there was on flow present to bring the pressure up to the setpoint.

There is flow through the port until just before injection, when the purge valve closes and flow is diverted into the split vent line, then after the splitless injection portion is finished it returns flow through the split vent to sweep out high boiling contaminates while the run proceeds. With it set to off on flow the pressure will fall to near zero.

The peak shapes can be effected by the purge timing, initial column conditions and holds, injection speed and many other things. You may have to experiment with those to sharpen the peaks as needed.

Did the USP method specify if they were using a split/splitless injection port or a direct inject injection port? If transferring the method from a direct inject to split/splitless port you will have to do some tweaking of the parameters to match the performance.

The USP method made no mention of a split ratio, the specific monograph was Benzyl Alcohol. We were using a split/splitless injection port that also had a head-space element attached.

However we ran it on an alternative Agilent GC with a higher purge flow of 60ml/min and a slightly shorter purge time which produced excellent chromatography. All our peaks used for quantification had a symmetry of 0.98 to 1.03 so this issue can be considered closed now.

Thanks for the additional clarification!

The USP method made no mention of a split ratio, the specific monograph was Benzyl Alcohol.

I found many USP Monograph methods to be poorly written and/or poorly-developed. Sad.

Of course our QA Director - who lacked hands-on knowledge and experience - felt USP was god. I had to fight with him about the recovery limit for our finished cGMP products to get him to OK 95-105% recovery as in FDA-ORA regulations.

I E-mailed USP "owners" of monographs a few times, like did "standardize frequently" under "Volumetric Solutions" mean every month, every day, or every 10 minutes? And I asked about how was content of pure alcohol tested, response was from specific gravity or refractive index, and not in the monograph.

Of course my own supervisor did not agree with allowable modifications detailed in <621> and in FDA-ORA, even though they were printed in black ink using the same 26 alphabet letters we learned in kindergarten. I never did get information as to if one could combine two of such changes (like decrease column from 4.6mm to 3.0mm and increase column temperature 5 degrees).

The USP method made no mention of a split ratio, the specific monograph was Benzyl Alcohol.

I found many USP Monograph methods to be poorly written and/or poorly-developed. Sad.

Of course our QA Director - who lacked hands-on knowledge and experience - felt USP was god. I had to fight with him about the recovery limit for our finished cGMP products to get him to OK 95-105% recovery as in FDA-ORA regulations.

I E-mailed USP "owners" of monographs a few times, like did "standardize frequently" under "Volumetric Solutions" mean every month, every day, or every 10 minutes? And I asked about how was content of pure alcohol tested, response was from specific gravity or refractive index, and not in the monograph.

Of course my own supervisor did not agree with allowable modifications detailed in <621> and in FDA-ORA, even though they were printed in black ink using the same 26 alphabet letters we learned in kindergarten. I never did get information as to if one could combine two of such changes (like decrease column from 4.6mm to 3.0mm and increase column temperature 5 degrees).

My first real encounter(and it wasn't actually a true encounter) was in a job interview for a QC chemist(not really my field, but they called me in for the interview when I hadn't applied for that specific job) and I was asked if I followed USP methods in any of my work. I was pretty well shot down and said "you can't do that" when I told them that in academia I never know what someone was going to bring me and that I developed a lot of my own methods or modified existing published ones. They were also less than impressed when I said that I do follow EPA methods both because I occasionally have environmental work and also because a lot of instrument comparison/testing seems to happen on EPA methods. They kept repeating in seeming disbelief that I'd never used a USP method.

Looking some up after that, I had much the same reaction. EPA methods can at times be annoyingly too detailed with things like DFTPP tunes. USP methods-like you said-are often a nightmare to try and figure out what instrument parameters you're supposed to use.

Of course, I also follow a fair few that have been journal published, and those can be hit or miss. Even in some good high-impact journals(JACS, Angewante, etc), it almost seems like the reviewers sometimes rubber-stamp chromatography methods without stopping to ask if they make sense. I seem to have better luck with good methods in lower impact but more focused journals like ACA. When you branch out of chemistry journals, it can be even more of a nightmare.

As a bit of an example, someone handed me a publication one time and asked if I could replicate the method/results in it. As it so happened, it was published on a 5890/5971 that I actually service and know fairly well...and have a copy of that lab's "universal" method on my 5890/5971 and 7820/5975. The published method said a "5% methylsiloxane column" which is fine and what I run as a standard column, but didn't give dimensions. I know it's a 20mx.20mm, which is a bit different from I run, but not a big deal as I run a larger diameter longer column that actually gives both similar retention times and a similar resolution. That was all good and well, but the paper claimed a 5mL/min flow rate. The pressure gauge on the front of a 5890 pegs at 30psi, and you need ~50psi to push helium through that column at 250ºC. Aside from that, I don't think the vacuum system on a 5971 is up to that kind of flow. That's not to mention that the peaks would probably look terrible moving that fast through that column. Fortunately, I know they keep their MS pressure set to 13psi, which is ~1mL/min depending on temperature, and I can replicate it decently well by using either my 5890 or 7820 set to constant flow at 1mL/min.

When scientists were reviewing my first publication for a journal, they asked a couple of questions, for clarity. Stuff that I did every time might not be so obvious to a reader. For example, when I wrote to filter through fast filter paper, reviewer asked if by gravity or with using vacuum, so I updated the wording.

The USP method made no mention of a split ratio, the specific monograph was Benzyl Alcohol.

I found many USP Monograph methods to be poorly written and/or poorly-developed. Sad.

Of course our QA Director - who lacked hands-on knowledge and experience - felt USP was god. I had to fight with him about the recovery limit for our finished cGMP products to get him to OK 95-105% recovery as in FDA-ORA regulations.

I E-mailed USP "owners" of monographs a few times, like did "standardize frequently" under "Volumetric Solutions" mean every month, every day, or every 10 minutes? And I asked about how was content of pure alcohol tested, response was from specific gravity or refractive index, and not in the monograph.

Of course my own supervisor did not agree with allowable modifications detailed in <621> and in FDA-ORA, even though they were printed in black ink using the same 26 alphabet letters we learned in kindergarten. I never did get information as to if one could combine two of such changes (like decrease column from 4.6mm to 3.0mm and increase column temperature 5 degrees).

My first real encounter(and it wasn't actually a true encounter) was in a job interview for a QC chemist(not really my field, but they called me in for the interview when I hadn't applied for that specific job) and I was asked if I followed USP methods in any of my work. I was pretty well shot down and said "you can't do that" when I told them that in academia I never know what someone was going to bring me and that I developed a lot of my own methods or modified existing published ones. They were also less than impressed when I said that I do follow EPA methods both because I occasionally have environmental work and also because a lot of instrument comparison/testing seems to happen on EPA methods. They kept repeating in seeming disbelief that I'd never used a USP method.

Looking some up after that, I had much the same reaction. EPA methods can at times be annoyingly too detailed with things like DFTPP tunes. USP methods-like you said-are often a nightmare to try and figure out what instrument parameters you're supposed to use.

Of course, I also follow a fair few that have been journal published, and those can be hit or miss. Even in some good high-impact journals(JACS, Angewante, etc), it almost seems like the reviewers sometimes rubber-stamp chromatography methods without stopping to ask if they make sense. I seem to have better luck with good methods in lower impact but more focused journals like ACA. When you branch out of chemistry journals, it can be even more of a nightmare.

As a bit of an example, someone handed me a publication one time and asked if I could replicate the method/results in it. As it so happened, it was published on a 5890/5971 that I actually service and know fairly well...and have a copy of that lab's "universal" method on my 5890/5971 and 7820/5975. The published method said a "5% methylsiloxane column" which is fine and what I run as a standard column, but didn't give dimensions. I know it's a 20mx.20mm, which is a bit different from I run, but not a big deal as I run a larger diameter longer column that actually gives both similar retention times and a similar resolution. That was all good and well, but the paper claimed a 5mL/min flow rate. The pressure gauge on the front of a 5890 pegs at 30psi, and you need ~50psi to push helium through that column at 250ºC. Aside from that, I don't think the vacuum system on a 5971 is up to that kind of flow. That's not to mention that the peaks would probably look terrible moving that fast through that column. Fortunately, I know they keep their MS pressure set to 13psi, which is ~1mL/min depending on temperature, and I can replicate it decently well by using either my 5890 or 7820 set to constant flow at 1mL/min.

When I first started running volatiles on a 5970 and 5971, we used a 105m 0.53mm column flowing at 7ml/min. Of course we were also using a jet separator so the flow into the MS was less than 1ml/min. If you leave out that fact it is really scary to think what might happen with the MS, but some older method assume that is what everyone uses with purge and trap.