How do I convert the Peak Area (pA*s) measurements in a GC-FID lab report into mass/weight percent of components?

How do I convert the Peak Area (pA*s) measurements in a GC-FID lab report into mass/weight percent of components for analysed crude oil samples?
How do I characterise the n-paraffins and estimate the single carbon number (SCN) distribution from the compositional analysis report?
How do I calculate Acentric factor, critical pressure, critical temperature, SG, MW, boiling point and ideal liquid density of crude oil components from GC-FID and basic physical properties (boiling point, SG, MW and density of the whole crude oil).
I do not have access to PVT equipment so have to make do with limited experimental data for PVT modelling. Every assistance would be really helpful.
Parents
  • >How do I convert the Peak Area (pA*s) measurements in a GC-FID lab report into mass/weight percent of components for analysed crude oil samples?

    Basically, not much conversion is needed. For ASTM methods, sometimes analytical "response factors" are utilized, other methods seem to perform just fine by assuming all hydrocarbons have an equivalent response by weight.

    Normally individual peak areas are not determined in crude oil except in methods for certain specific components or contaminants such as pristane and phytane.

    Instead the crude oil chromatogram is interpreted for "Simulated Distillation" (SimDist), so it's integrated by "slices" not peaks. The area under the whole chromatographic curve is accumulated up to known points of retention time which correspond to the literature boiling points of the paraffin hydrocarbons in the reference mixture.

    Also since all of the crude oil will not fully elute from the GC column, corrections need to be made to account for the heavy fraction that can not be detected.

    For this you need to deeply study ASTM D7169, and use a follow-up software treatment beyond the usual peak integration & calibration of ordinary GC work.

    Carbon Disulfide is the solvent for most SimDist work which was selected since it has such a small FID response. This is fine for Diesel fuel and other distillate products that are not supposed to contain light gasoline-range hydrocarbons, but these "light ends" of the crude oil will usually be badly obscured by the CS2 unfortunately. So for crude oil you usually need to determine the light ends on a different GC configured for ASTM D7900, then mathematically merge the data from both GC's to arrive at a more accurate boiling range distribution.

    D7900 does use regular peak data but it usually only goes up to n-Nonane.

    For SimDist the raw data from the GC run is Exported as an AIA format CDF file which basically consists of slices without any further peak integration.

    Then a Simulated Distillation software package will be used to calibrate the CDF files according to the boiling points of the reference materials, and to then calculate the crude oil according to expected distillation temperatures.

    Even if you are not actually interested in distillation properties, D7169 will give you some good guidelines on how to properly collect crude oil data from a lab GC.

    >How do I characterise the n-paraffins and estimate the single carbon number (SCN) distribution from the compositional analysis report?
    >How do I calculate Acentric factor, critical pressure, critical temperature, SG, MW, boiling point and ideal liquid density of crude oil components from GC-FID and basic physical properties (boiling point, SG, MW and density of the whole crude oil).

    From this point you you will have a lot of choices but not too much is very standard. You can process the same CDF files that were collected for SimDist, according to carbon number instead of temperature. SG is a physical measurement, easy with Stock Tank Oil using hydrometers of densitometers, and could be correlated to the GC measurements to some degree.
    For MW I am just fine with freezing point depression measured using a Beckmann thermometer, to get the average MW of the whole crude oil.
    These can be combined with GC data and be very meaningful and useful, but it's basically research-oriented applications within a PVT lab.
    For critical pressure & temperature, or anything else on actual reservoir fluid I think you really are going to need a basic PVT lab at least for real bench measurements.

Reply
  • >How do I convert the Peak Area (pA*s) measurements in a GC-FID lab report into mass/weight percent of components for analysed crude oil samples?

    Basically, not much conversion is needed. For ASTM methods, sometimes analytical "response factors" are utilized, other methods seem to perform just fine by assuming all hydrocarbons have an equivalent response by weight.

    Normally individual peak areas are not determined in crude oil except in methods for certain specific components or contaminants such as pristane and phytane.

    Instead the crude oil chromatogram is interpreted for "Simulated Distillation" (SimDist), so it's integrated by "slices" not peaks. The area under the whole chromatographic curve is accumulated up to known points of retention time which correspond to the literature boiling points of the paraffin hydrocarbons in the reference mixture.

    Also since all of the crude oil will not fully elute from the GC column, corrections need to be made to account for the heavy fraction that can not be detected.

    For this you need to deeply study ASTM D7169, and use a follow-up software treatment beyond the usual peak integration & calibration of ordinary GC work.

    Carbon Disulfide is the solvent for most SimDist work which was selected since it has such a small FID response. This is fine for Diesel fuel and other distillate products that are not supposed to contain light gasoline-range hydrocarbons, but these "light ends" of the crude oil will usually be badly obscured by the CS2 unfortunately. So for crude oil you usually need to determine the light ends on a different GC configured for ASTM D7900, then mathematically merge the data from both GC's to arrive at a more accurate boiling range distribution.

    D7900 does use regular peak data but it usually only goes up to n-Nonane.

    For SimDist the raw data from the GC run is Exported as an AIA format CDF file which basically consists of slices without any further peak integration.

    Then a Simulated Distillation software package will be used to calibrate the CDF files according to the boiling points of the reference materials, and to then calculate the crude oil according to expected distillation temperatures.

    Even if you are not actually interested in distillation properties, D7169 will give you some good guidelines on how to properly collect crude oil data from a lab GC.

    >How do I characterise the n-paraffins and estimate the single carbon number (SCN) distribution from the compositional analysis report?
    >How do I calculate Acentric factor, critical pressure, critical temperature, SG, MW, boiling point and ideal liquid density of crude oil components from GC-FID and basic physical properties (boiling point, SG, MW and density of the whole crude oil).

    From this point you you will have a lot of choices but not too much is very standard. You can process the same CDF files that were collected for SimDist, according to carbon number instead of temperature. SG is a physical measurement, easy with Stock Tank Oil using hydrometers of densitometers, and could be correlated to the GC measurements to some degree.
    For MW I am just fine with freezing point depression measured using a Beckmann thermometer, to get the average MW of the whole crude oil.
    These can be combined with GC data and be very meaningful and useful, but it's basically research-oriented applications within a PVT lab.
    For critical pressure & temperature, or anything else on actual reservoir fluid I think you really are going to need a basic PVT lab at least for real bench measurements.

Children
No Data
Was this helpful?