column flows

[mjh32586]

I am fairly new to GC and have a GC with dual ecd's. I am using one injection port with 2 columns connected using a y splitter that are in 2 separate detectors. I am using helium as my carrier gas and nitrogen as my makeup gas. I have nitrogen set to 30 mL/min. I am using a splitless injection. My columns are 30m X 0.32 ID and typical flow rates I have seen are 1.2mL/min. Since I have 2 columns installed, do I want to double to helium column flow to 2.4 mL/min? Can someone also explain to me what the sampling time signifies? My understanding is it determines when the split vent opens but what effect will that have on my chromatogram?

[Peter Apps]

For two columns you need double the flow, and your flow rate is low for a 0.32 mm column anyway, so you could increase your volume flow to 4 ml/min.

What injection set-up do you have ? "Sampling time" might be what is usually referred to as splitless time (or purge on time, or vent on time) with a standard split-splitless injector, but I have never heard it referred to as such.

Peter

[GasMan]

The flow that you enter depends on how the GC system works. I assume from the posting that the user is using a capillary inlet. I am not aware of any GC manufacturer that is actually controlling the flow in a capillary column using a flow sensor. All flows in a capillary column are controlled by pressure. You input into the GC the length and internal diameter of the column, the type of carrier gas and the column temperature, and the flow that you want. A calculation is then performed to calculate the pressure needed to give the required flow for your column. This means that all reported flows for a capillary column are calculated, not measured. As you have two identical columns in parallel, both would need the same head pressure to give the same flow. I would therefore say that the flow that you enter should not be double, but the same.

Gasman

[Peter Apps]

Gasman

You are entirely correct, to the extent that I am tempted to edit my post !!

mjh32586, my apologies for posting with my brain on cruise control.

Peter

[mjh32586]

Yes, I am using two capillary columns with a guard column in the injector and then a y-connector for the other two columns. The two columns I am using are 30m x 0.32mm ID x 0.32 um df and 30m x 0.32 ID x 0.25um df. Should I have a column flow (He as my carrier) of 1.2 mL/min ? I currently am using the nitrogen as the make-up gas and it is set to 30 mL/min. Also, what do I want for a total carrier flow? I was under the impression total carrier flow should be 10mL/min per column??
Right now I have everything set as follows:

He-carrier gas
Flow control mode=linear velocity (or should it be pressure??)
pressure= 98.0 kPa
total flow=22.7 mL/min
column flow=1.79 mL/min
linear velocity= 25.0 cm/sec
purge flow= 3mL/min (I believe this is septum purge)
split ratio=10.0 (I am using a splitless injection but this number is here so that the program calculates the total flow to be 22.7mL/min) sampling time for the splitless injection is currently 0.47 min

Nitrogen-make-up gas
total flow 30.0 mL/min

Do these settings seem in line with what they should be?

[Peter Apps]

Depending on which instrument you have (and I am guessing that it is an Agilent, can you confirm this ?) you can set the flow in different ways. There is a lot of discussion in the forum archives that makes very interesting reading about which is the best way.

If you set the linear flow the instrument calculates a volume flow, then calculates the pressure required on the basis of what column you have (which you will have entered somewhere), then puts that pressure on the inlet.

The septum purge flow needs to be 3 - 5 ml min for most applications. The split ratio is set according to what you are injecting and what you need to detect. For a splitless injection the split flow needs only to be fast enough (usually 10 - 20 ml/min) to flush the inlet when the split opens. It is adjusted by the instrument's adjusting the flow INTO to the inlet - and this is what appears as a readout of total flow; total flow = split flow (NB not split ratio) + column flow (in volume/time not linear flow) + septum purge.

You certainly do not need 10 ml/min of carrier flow through the columns that you have - for a 0.32 mm i.d. column the optimum is 1.7 - 3 ml/min depending whether you want speed or resolution (have a look in the archives for more on this). Nonetheless at a total flow of less than 10 ml/min the electronic flow controls do not work very well.

Your splitless time is probably OK, but I would round it up to 0.5 min - if that little bit of difference makes a difference to the results then the method is certainly not robust. Depending on a lot of things you could increase it to a minute, but your solvent peak will get wider and your analyte peaks will probably not get much bigger.

Hope I got it right this time !

Peter

[mjh32586]

I am actually using a Shimadzu GC-2010Plus. Nobody seems to know much about their instruments
So I will keep my column flow at 1.79 mL/min right now since that seems correct but I'm still confused about the total flow. You stated that total flow needs to be at least 10 mL/min. If I am using nitrogen as the make-up and it is set to 30 mL/min, then does that mean I am covering the necessary total flow of at least 10mL/min? Does that also mean that I could essentially set the split ratio to 0 since it's a splitless injection and then the total flow for helium will be 4.8mL/min? (3.0 mL/min septum purge + 1.8 mL/min column flow) or do I need the total helium flow to be 10 mL/min?

[Peter Apps]

The total flow that I am talking about, and which appears on the readout of your GC is the flow into the inlet. It has nothing at all to do with makeup gas, which goes only to the detector.

So; you need a total flow to the inlet (helium) of at least 10 ml/min for the EPC to operate properly. Note that with a total flow of 10 ml/min your split is less than 5:1.

If you set the split to 0 you will get a huge tail on the solvent peak.

All of this information, and a lot of other useful stuff is in the user's manual - I recommend that you read it.

Peter

[AICMM]

Peter,

Unless I understand the question wrong (not like that's ever happened before) I think you were correct in your first statement. mjh32586 has two columns in parallel. Therefore, he/she needs to set the pressure to twice what the data system would "suggest" in order to get the flow's where he needs on both columns. Here's my logic: the EPC is going to calculate a head pressure for a single 0.32 column and set that. In reality what mjh32586 has two columns to feed so the EPC will not correctly calculate.

Of course, the simplest answer is to pull out the flow meter and actually measure the flow through each of the columns/detectors and set the head pressure where it needs to be.

mjh32586, with the Agilent, you set the column flow you want and the system will take care of providing enough flow to the inlet to feed the septum purge. I don't know about the Shimadzu, sorry. Again, the best advice is to actually measure both column flows with a meter (without make-up, without air/H2, etc...) Finally, with a 0.32, 1.8 mL/min is probably a bit on the low side, you might try something more like 3 mL/min and take advantage of the relatively flat VanDeemter.

Best regards,

AICMM

[GasMan]

AICMM

The EPC does not know that there are two columns in parallel. If both columns are identical, then they both need the same head pressure to give the same flow, so you only need to enter the data for one column. If the inlet gas is controlled in a pressure mode, and it is in the Agilent systems during splitless mode, the system will supply the necessary flow. I do not know about the Shimadzu system, but I would assume that they work on a similar principle.

I would agree that the best way is to set up the system and measure the resulting flows.

Gasman