ISO 3924 Petroleum products - Determination of boiling range distribution - Gas chromatography method
10 Procedure
10.1 Sample preparation
10.1.1 The amount of sample injected shall not overload the column stationary phase capacity nor exceed the detector linear range.

NOTE A sample with a narrow boiling range requires the injection of a smaller amount than that with a wider boiling range.

10.1.2 The column stationary phase capacity can be estimated from the chromatogram of the calibration mixture (5.4). Different volumes of the calibration mixture can be injected to find the maximum amount of a component that the stationary phase can tolerate without overloading (see 8.5, Note). Record the peak height for this amount of sample. The maximum sample signal intensity shall not exceed this peak height.

10.1.3 Samples with a viscosity low enough to be sampled with a syringe at ambient temperature shall be injected undiluted. Samples that are too viscous or waxy for sampling with a syringe may be diluted with carbon disulfide (5.6).

10.1.4 Typical sample injection volumes are shown in Tables 5 and 6.

10.2 Sample analysis
Using the analysis sequence protocol (see 9.1), inject a sample aliquot into the gas chromatograph. At the time of injection, start the chromatograph time cycle and the integrator/computer data acquisition.

11 Calculation
11.1 Correct the sample area slices for the non-sample detector response by subtracting each blank analysis area slice from each sample area slice at the equivalent slice time. Sum the corrected area slices to obtain the cumulative corrected areas for each time interval during the run.

11.2 At the point on the chromatogram where the baseline at the end of the run first becomes steady, record the total cumulative area counts. Move back along the chromatogram until the cumulative area equals 99.5 % of the total area. Mark this point as the final boiling point.

NOTE Locating the final boiling point can be the most difficult step in this method. Some samples have extremely long tail-end portions due to gradually decreasing amounts of heavy material. This fact, coupled with the natural tendency of the chromatographic baseline to rise at the end of the run due to septum and/or column bleed and/or elution of traces of heavy components from previous samples, can preclude the possibility of the chromatogram returning precisely to the original baseline established prior to the initial boiling point of the sample. Thus, the most satisfactory procedure is to inspect the chromatogram and the area counts at each interval near the end of the run to determine the point at which the rate of change of the chromatographic signal has reached a constant low value of no greater than 0.00001 % of the total area counts per second.

11.3 Observe the area counts at the start of the run until the point is reached where the cumulative area count is equal to 0.5 % of the total area. Mark this point as the initial boiling point of the sample. If carbon disulfide is used as the solvent, its response shall be ignored in the calculations.

11.4 Divide the cumulative area at each interval between the initial and final boiling points by the total area and multiply by 100 to give the percentage of the sample recovered at each time interval.

11.5 Tabulate the cumulative percentage recovered at each interval and the retention time at the end of the interval. Using linear interpolation where necessary, determine the retention time associated with each percentage between 1 % and 99 %.

11.6 For each percentage and its associated retention time, determine the corresponding boiling point temperature from the calibration table (see 9.3.2). Use linear interpolation between data points.