Chromatography Forum

My method says 30cm/sec linear velocity of nitrogen with HP-5, 30m*0.53mm, 5micron column.

Can anybody explain how much flow/pressure I can apply in this case.

thanks in advance,
Bhaskar

Not too sure if this is what you are asking, but 30 cm/s is about 2 mL/min.

And the handy formula for converting that is

F = ((pi*(d^2)*u)/4j)*60

where,

pi = 3.14
d = column diameter in cm
u = linear velocity of gas
j = gas compressibility factor
60 = needed to convert /sec to /min

I know for He, j = 0.1662; not too sure what it is for N2....

J can be experimentally determined using inlet pressure (pi) and outlet pressure (po) by the following equation:

j = 1.5 * (((pi/po)^2) - 1)/(((pi/po)^3) - 1)

If you go to the following site, you can download a flow calculator program that will give you the answers that you need.

http://www.chem.agilent.com/cag/main.html#flowcalc205

Gasman

Optimal flow rate, Fopt, of nitrogen in a capillary column can be found as

Fopt = (2.5 mL/min/mm)*ID

where ID is a column internal diameter. For 0.53 mm column, the formula yields:

Fopt = (2.5 mL/min/mm)*0.53 mm = 1.3 mL/min.

(This corresponds to average velocity of about 11 cm/s)

Avoid using average velocity as a control parameter. In many cases, optimal average velocity is difficult to predict.

lmb

Why go to all the trouble of trying to calculate flow from approximate equations ?. From the length of the column, calculate how long something moving at 30 cm/s would take to get from one end to the other. Inject any substance that the detector can see and that is not retained on the column stationary phase (e.g. methane for an FID) and measure the retention time. Increase or decrease the inlet pressure according to the time being too long or too short compared to the calculated value. The you are completely certain that you have the velocity you want on the instrument as you are using it.

Incidentally; 30 cm/s is the optimum flow for helium. For nitrogen it is around 15 cm/s.

There has been another thread on the validity of accepted values for optimum flows, it is more complicated than it looks !

Peter

Dear Peter,

Suggesting to “avoid using average velocity as a control parameterâ€

You recommended 30 cm/s for average velocity of helium. I am familiar with this and similar recommendations but I am not aware of their justifications. I would greatly appreciate any info about the sources. From what I know, the number is about OK, but only for wide bore columns and not in GC-MS (not when the column outlet is at vacuum).

Recommended velocity of 30 cm/s for helium and 15 cm/s for nitrogen in GC gives the optimum linear velocity of the carrier gas in which case the hight equivalent to a theoretical plate (HETP) has the lowest value. Which means also achieving the best column efficiency.

Best regards

Dear zokitano,

I understand that we are talking about the optimal conditions (expressed via optimal average velocity of a carrier gas or via its optimal flow rate) that minimize a column plate height and, therefore, maximize the column efficiency.

What I do question is the number 30 cm/s as optimal average velocity of helium in ANY COLUMN. As I mentioned in my previous reply to Peter, optimal average velocity of helium in 1m×0.1mm column in GC-MS is NOT 30 cm/s, but close to 180 cm/s – a much bigger number.

I understand that it might or might not be important for the columns that you use, and you might or might not be interested in paying attention to this, but the fact that optimal average velocity of helium might be significantly different from 30 cm/s remains, and it might be important for other readers.

I also wanted to mention (as I did in several previous postings) that, generally, it is much more difficult to figure out optimal average velocity of a particular carrier gas in a particular column than to figure out optimal flow rate for the same column and gas. That’s all.

Best regards,
lmb

Dear Imb,

Chromatographic separations using capillary columns are achieved under constant pressure, which is necessary to produce a given linear velocity or the carrier gas. The magnitude of the pressure drop across capillary is a function of the particular carrier gas and the length and inner diameter of the column.
On the other hand, smaller flowrates are necessary when using columns with small i.d (0.1mm, 0.18mm, 0.25 mm) to obtain best results in GC-MS.

Yes, the recommended velocity of 30 cm/s for helium is optimal average velocity of helium in any column.

The thiner the column, the smaller column flowrate at constant pressure or constant carrier gas velocity. I recommend to use constant linear velocity (of about 30 cm/s for helium) when you're using your thin 0.1mm column (at the same time you'll measure low column flowrates, adequate for your GC-MS analysis).
Let's put it in this way, if you have 0.53mm (i.d) 30m column and you're using constant flowrate of for example 5mL/min you'll measure linear carrier velocities of about 30 cm/s (for helium as carrier gas). If you change the column with 0.1mm (i.d) column, that flow (5mL/min) will give you higher carrier linear velocity (for helium), which isn't the best for obtaining the best separation and efficiency.

Best regards

I suspect LMB is being taught how to suck eggs, assuming LMB is Leon M Blumberg.

In earlier posts in the archives, Leon points to several references. Searches on "Blumberg" in both the archives and current Fora will find several informative and helpful posts..

Please keep having fun,

Bruce Hamilton

This is one of those discussions where theory and practical reality do not always give the same answer.

The oft-quoted 30 cm/s optimum for helium (and the corresponding 50 cm/s for hydrogen and 15 cm/s for nitrogen) is based on the minimum of the classic van Deemter curve for plate height vs velocity, and is as applicable as the equation is - which is to say it is a useful approximation to actual practise. There are other ways of calculating various optima for different sets of conditions, but like the van Deemter curves they should be used as guides, not gospel. In particular, narrow bore columns will always run better at higher flows/velocities (and the limiting factor might then become the inlet plumbing !) and the vacuum of an MS increases the optimum flow/velocity by increasing the diffusivity and decreasing the viscosity of the gas phase. Temperature programming causes continual changes in all the parameters (except the diameter and length of the column).

Unless the gas flow is very far off the optimum the effect of flow on resolution is surprisingly small - but of course it has very noticable effects on retention, wghich in turn impacts on analysis time and sample throughput, which in most (?) labs are what matters. As a working approximation, maximum resolution in unit time cames at flow nearly double the van Deemter minumum plate height flows.

Where theory and practise do agree is that nitrogen takes way longer to get the separation done than helium, and that hydrogen is fater than either (but cannot be used with an MS).

I like to measure the elution time of an unretained peak because it gives a completely independent check of the "black box" flow and pressure readings on a GC's front panel. ALso the shape of the peak shows whether there are problems with column installation.

Peter

Dear zokitano,
Dear Peter,
Dear All,

I have a simple question. What are the sources for the numbers like 50 cm/s for hydrogen, 30 cm/s for helium, 15 cm/s for nitrogen, etc.?

I am aware of van Deemter equation, Golay equation, etc. – the formulae that are frequently mentioned as the basis for those numbers. I am not asking for the formulae.

I am also familiar with several sources where the numbers were recommended, but no justification was provided. Those are not the sources that I am looking for either.

What I am asking about are the sources where the numbers were actually obtained – theoretically, experimentally, whatever. What are those sources? If you are aware of any text that says something like this: “30 cm/s is optimal average velocity of helium in any column, and here is how I/we came up with the numberâ€

Dear lmb

I've done some research and I came to conclusion that the theory doesn't match practice
Thank you for enlighten me!!

Sincerely

Dear lmb

As you are well aware, there is no source that states "“30 cm/s is optimal average velocity of helium in any column, " etc, for the simple reason that such a statement is not true, and cannot be true because there is no single velocity that is optimum under all conditions of column dimensions and temperature. A simple practical difficulty is at what temperature should the linear velocity (or the volume flow rate) be measured if the separation is to be run with a temperature programme ?. If gas velocity was a critical parameter in GC we would never get the separations done because velocity is constantly changing as the analyte moves down the column and the temperature changes during a programme.

Ultimately, chromatography is a tool for separating compounds, and the faster it can be done the better. The linear velocity guidelines are practical starting points for optimisation that can be be easily and reliably measured, and independently verified for any given GC setup. At the guideline flows on the "ordinary" columns that most separations are run on you get a good chromatogram is a reasonable time. Nobody disputes that with short narrow columns the guidelines do not apply, or that to get the separation done as fast as possible you need to run the gas faster than the magic flows. Under the conditions that most gas chromatography is done, column dimensions, stationary phase, injection and inlet conditions, and temperature programme are all much more important than gas velocity.

What I would really be interested in seeing is a set of chromatograms that clearly show a major influence of flow rate on how well two peaks separate.

Peter

Dear lmb

As you are well aware, there is no source that states "“30 cm/s is optimal average velocity of helium in any column, " etc, for the simple reason that such a statement is not true, and cannot be true because there is no single velocity that is optimum under all conditions of column dimensions and temperature. A simple practical difficulty is at what temperature should the linear velocity (or the volume flow rate) be measured if the separation is to be run with a temperature programme ?. If gas velocity was a critical parameter in GC we would never get the separations done because velocity is constantly changing as the analyte moves down the column and the temperature changes during a programme.

Ultimately, chromatography is a tool for separating compounds, and the faster it can be done the better. The linear velocity guidelines are practical starting points for optimisation that can be be easily and reliably measured, and independently verified for any given GC setup. At the guideline flows on the "ordinary" columns that most separations are run on you get a good chromatogram is a reasonable time. Nobody disputes that with short narrow columns the guidelines do not apply, or that to get the separation done as fast as possible you need to run the gas faster than the magic flows. Under the conditions that most gas chromatography is done, column dimensions, stationary phase, injection and inlet conditions, and temperature programme are all much more important than gas velocity.

What I would really be interested in seeing is a set of chromatograms that clearly show a major influence of flow rate on how well two peaks separate.

Peter