10. Procedure
10.1 Measure the intensity of the drift correction monitor(s) used in 9.7. The value determined corresponds to B in Eq 5, Section 11. This measurement may not be required on high stability instrumentation. Determine the value of F8 in Eq 6, Section 11 at regular intervals by measuring the peak and background count rates on the solvent blank. This measurement may not be needed on some instruments.
10.2 Place the sample in an appropriate cell using techniques consistent with good practice for the particular instrument being used. Although sulfur radiation will penetrate only a small distance into the sample, scatter from the sample cup and the sample may vary to such an extent that a specific amount or a minimum amount of sample shall be used. Generally, filling the cup to one-half capacity is sufficient. Once this amount is established for each instrument, this volume of sample is used for each measurement.
10.3 Place the sample in the X-ray beam, and allow the X-ray optical atmosphere to come to equilibrium.
10.4 Determine the intensity of the sulfur Ka radiation at 5.373 Å by making counting rate measurements at the precise angular settings for this wavelength.
NOTE 5 - Take a sufficient number of counts to satisfy at least a 1.0 % expected coefficient of variation when practical. When sensitivity or concentration, or both, make it impractical to collect a sufficient number of counts to achieve a 1.0 % coefficient of variation, accepted techniques, which will allow the greatest statistical precision in the time allotted for each analysis, should be used. Calculate the coefficient of variation as follows:
where:
Ns = gross number of counts collected at sulfur line, and
Nb = gross number of counts collected at background wavelength (on the base material) in the same time interval taken to collect Ns counts.
10.5 Measure background count-rate at a previously selected, fixed, angular setting, adjacent to the sulfur Ka peak.
NOTE 6 - Suitability of any background setting will depend on the X-ray tube anode employed. A wavelength of 5.190 Å is recommended where chromium or scandium is used, whereas 5.437 Å has been found suitable for rhodium. 20 peak and background angles for various crystals are listed in Table 2.
10.6 Calculate the corrected net counting rate using Eq 6 in Section 11.
10.7 If, from the measurements made in accordance with 10.3 and 10.4, the counting rate is higher than that of the highest point on the calibration curve, dilute the sample with base material until the sulfur count rate is within the limits of the calibration curve and repeat the procedure described in 10.2-10.6.
10.8 Quality Control - For the purpose of establishing the statistical control status of the testing process since the last valid calibration, QC samples prepared from material(s) selected and stored as described in 7.7 are regularly tested as if they were production samples. Results are recorded and analyzed by control charts (see Note 7) or other statistically equivalent techniques to ascertain the statistical control status of the total testing process. An immediate investigation for root cause(s) is triggered by any out of control data. The outcome of this investigation may, but not necessarily, result in instrument recalibration. In absence of other explicit requirements, the frequency of QC samples testing is dependent on the criticality of the quality being measured and the demonstrated stability of the testing process. It is recommended that at least one type of QC sample that is regularly tested be representative of samples routinely analyzed.
NOTE 7 - The precise test method of control charting, chart interpretation, and corrective action is left to the individual laboratory since the topic is outside the scope of this test method.
11. Calculation
11.1 Calculate the correction factor for changes in daily instrument sensitivity as follows:
F = A/B
where:
A = counting rate of the calibration check standard as determined at time of calibration, and
B = counting rate the calibration check standard as determined at time of analysis.
NOTE 8 - Inclusion of F in Eq 6 may not be necessary or desirable with some highly stable instrumentation.
11.2 Determine the corrected net counting rate as follows:
R = [CK/S1) - CBF'/S2)]F
where:
CK = total counts collected at 5.373 Å,
CB = total counts collected at sulfur background,
S1 and S2 = seconds required to collect C counts,
R = corrected net counting rate, and
F' = (counts/s at 5.373 Å)/(counts/s at background wavelength) on a blank sample containing no sulfur.
NOTE 9 - The F8 factor is optional and is offered for use in the case that it has a significant value.
11.3 Apply the corrected net counting rate to the appropriate calibration curve and read and report the sulfur concentration in mg/kg.
11.4 Calculate the concentration of sulfur in a diluted sample as follows:
Smg/kg = Sb x [(Ws + Wo)/Ws)]
where:
Sb = mg/kg sulfur in diluted blend,
Ws = mass of original sample, g, and
Wo = mass of diluent, g.