Splitless mode-purge flow affecting column flow

Why is the purge flow affecting the column flow, e.g. we set purge flow to 0.5 mL/min, and that leads  to the column flow of 1.4 mL/min not being reached.

  • Liquid injection Splitless mode changes things in the inlet over time.   You want it to look like the example below.

    During the time from injection to 0.8 minutes, the split vent flow is off/diverted (depending on the instrument) and all flow goes through the liner onto the column.  At 0.8 minutes, that flow changes to 50 ml/min to sweep any sample or solvent residue out of the liner.  Then at 2 minutes, the Gas Saver mode reduces the 50 ml/min to 20 ml/min, the minimum total flow that the SSL/MMI inlet requires to operate properly.

    The 0.8 minutes depends on your column flow, inlet liner volume, inlet temperature, and solvent injected. The easy way to choose is to think about how long your column flow takes to sweep the inlet liner volume.  You say you are using 1.4 ml/min column flow (a bit high for the ion source, not bad, but high), and the liner volume of most Splitless liners is about 800uL, so 0.8/1.4 = 0.57 minutes.  I would pad that time by at least 0.1 minutes but not all that much more and use 0.7 minutes.

    Make sure and read the Operations Manuals.  You can also view a variety of documents, webinars, and videos on Agilent.com. 

  • Hi Paul, 

    I used this logic minset for splitless method development. For example, I have a 800 uL liner and a column flow at 1 mL/min. It takes 0.8 min and I have used 1.0 minutes as purge time. 

    Some of my peaks are ugly but some are perfect gaussian. 

    I have a 0.18 mm ID column so I don't mind to reduced liner volume to decrease inlet residant time of some tricky components. 

    Do you have double tapered splitless liner with smaller size ? I can't see it in the Agilent website? 

    Another observation here, 

    I checked the flow before the purge time and I don't know why my total flow (septum + column) is not good when I read the actuals at the inlet. 

    For example, when the split vent is closed during the injection, I can read 3.7 mL/min. Should it be 4.0 mL/min ? 

    3 mL/min for septum purge and 1 mL/min for the column. 

    my pressure decay passed the inlet and my tune has 2% Nitrogen. 

    Can you help me ? 

  • "Some of my peaks are ugly but some are perfect gaussian" -- the chromatographic peak shape with splitless injections partially depends on the initial oven temperature during the splitless time. The recommended temperature is 20° C or more lower than the boiling point of the solvent used. This may require changing to a higher boiling point solvent or using oven cryo. 

    Reducing liner volume may or may not decrease inlet residence time depending on the sample and solvent injected and the volume injected. Use the vapor volume calculator to see the percentage of the liner volume created by your sample, inlet pressure, and solvent.  You definitely do not want a liner that is barely large enough to contain the sample vapor.  

    Try pulsed splitless as that will reduce the expansion volume and get the sample onto the column in a narrower band, which often helps chromatographic peak shapes.  This example shows a 20m × 0.18 mmid column using helium at 45° C initial oven temperature at 0.75 ml/min so the initial pressure would be 15.391 psi.'

    I arbitrarily set the pulse to make 2.0 ml/min column flow which is 34.5 psi. Using the vapor volume calculator, the difference in vapor expansion between 15.391 psi and 34.5 psi is significant. 

    15.391 psi vs 34.5 psi vapor volume

    =================

    This is MassHunter Acquisition 10.2 SR3 and shows the Pulsed Column Flow, a nice touch.  The pulse time is set quite short so that the injection port pressure can decrease slowly over the time between the end of the pulse and the Purge Flow to Split Vent time of 0.75 min.  There can be issues if the pulse time and the purge time are too close together. If the inlet is at the 34.5 psi pulse pressure and that time is the same as the Purge time, the inlet pressure will go from 34.5 psi down to 15.391 psi nearly instantaneously, but the column pressure can't go down that fast so the flow at the front of the column will go backwards towards the lower pressure inlet for a while before stabilizing and going forward again.  As you can imagine this can affect chromatographic peak shape.

    Note that the Purge Flow to Split Vent is 50 ml/min. That high flow sweeps any possible sample vapor in the liner out quickly. If done right, this flow and timing cuts off a very long solvent tail while still getting the peaks on the column.   Use Gas Saver set to 20 ml/min at 2 minutes to reduce helium use. Don't go lower than 20 ml/min -- in fact many think 20 ml/min is insufficient to ensure that oxygen cannot get into the inlet subsystem.

    Upload some chromatograms before and after trying these parameters if you can.

  • Hi Paul, 

    I worked in the past with your colleague Ron Honnold with NDMA in drinkable water and he helped me with the pulsed splitless injection. I know this mode is more efficient than normal splitless. 

    But my quality manager was asked to fulfilled OMCL requirement about splitless instrument qualification testing. 

    I told him that pulsed splitless is also a modern mode of splitless and he accepted this morning to use this mode for protocol verification establisment.

    So I will try your advice in your response. 

    But for now, I will show you that I am not concern about the vapor volume used with 0.5 uL using Heptane as solvent. that's why I was asking if lower volume liner here could help. By using 400 uL liner I will still be around 20% only of vapor. 

    The oven is set at 70C so 20C below the boilling point of heptane (98C)

    When I tried Octanol, I had terrible shape on apolar HP-5MS column. 

    20 ppm might be too concentrated or maybe I need to get the inlet at lower temperature ? Do pulsed splitless will help this case? I will surely give it a try. 

    I also tried a agilent solution used to qualify FID (5188-5373) because it's a mixture of non polar compound in iso-octane. Iso-octane has the same boilling point than heptane. 

    My first peak has not a good shape but the 3 other peaks seems ok like if the earlier peak is more affecting by unknwon

    Finally, I also read that using Methanol (very polar solvent) on a apolar column like HP-5MS is not suitable for splitless because the solvent focusing is less effective since methanol will not stick correctly to the head column during the injection. 

    Are you agree with that ? 

    For example, Methanol can be good for polar column like Heavy wax agilent column. 

    Thank you for you help. 

  • Split/Splitless Injection Port or Multimode?  How far up is the column installed in the inlet? see: (5) Column Installation - Split/Splitless and Multimode inlets =and= SQ and TQ MS transferlines - Files - GC/MS - Agilent Community

    I'm not sure I'd call that peak shape "bad" ...maybe just not perfect?  It's still sharp and very big at 3M counts tall.  Exactly what column, flow rates, and temperatures are you using?  

    Most any solvent can work on most any column as long as the operator understands and accommodates the limitations and doesn't try to fix something that may not be broken.  With methanol's low boiling point of 64.7° C, the required initial oven temperature to do reasonable solvent focusing on the front of the column would probably need to be below 40° C which is difficult to achieve in many laboratories.  If the sample peaks are much higher boiling point it may not matter much. There are many variables to consider.

    The fast ALS injection speed means that droplets of sample exit the end of the syringe needle and fly until they hit something solid before evaporating. With your samples and only 0.5µL of a quite high boiling point solvent injected this is very likely.  Since the flow is going towards the column, the sample expansion probably stays quite near the bottom of the liner.   It might be interesting to try the fritted liner so that the sample hits the frit and vaporizes rather than hitting the bottom of the liner and possibly the end of the column first.  see: application-fritted-wool-liners-comparison-gc-ms-5994-2179en-agilent.pdf 

  • Hi Paul, I juste tried your pulsed conditions and it helps my shape. This is the Agilent 3 ppm solution of tridecane and tetradecane peaks. 

    I have S/SL inlet. My column is 0.5 mm long when I installed my self-tight nut at the inlet. 

    For now i'm injecting 0.5 uL, but I will go at 1 uL to have better % RSD on my replicates. 

    So I will try if this splitless can help more polar compound like FAME et octanol. 

    I will even try again my first test with caféine in Methanol with the oven at 40C. 

    Thank you again for your help. 

Was this helpful?