GB/T 11140-2008 Determination of sulfur content in petroleum products—Wavelength dispersive X-ray fluorescence spectrometry

AI-translated method card — human verification pending. This is not an official translation of the standard. Unclear source values remain unverified; check the original standard and product certificate before use.

Incomplete source card: Sections 11, 12, 13, 14 are absent from the available source. Only available content is translated; source completion is pending.

Card type: Test method

1. Method number / full title / English title / replacement relationship

• Standard number: GB/T 11140-2008
• Chinese title: Determination of sulfur content in petroleum products—Wavelength dispersive X-ray fluorescence spectrometry
• English title: Standard test method for sulfur in petroleum products by wavelength dispersive X-ray fluorescence spectrometry
• Replacement relationship: Replaces GB/T 11140-1989 "Determination of sulfur content in petroleum products (X-ray spectrometry)"
• Adoption relationship: Modified adoption of ASTM D2622:2007 "Method for determination of sulfur content in petroleum products by wavelength dispersive X-ray fluorescence spectrometry"
• Publication/implementation: Published on 2008-08-25, implemented on 2009-02-01
• Secretariat: Subcommittee on Petroleum Fuels and Lubricants of the National Technical Committee for Standardization of Petroleum Products and Lubricants
• Drafting organization: Shanghai Gaoqiao Branch of China Petroleum & Chemical Corporation
• Main drafters: Cao Yichun, Sang Guocui

2. Scope of application and non-applicability

• 1.1 applies to the determination of total sulfur content in petroleum and petroleum products that are single-phase and liquid at room temperature, that become liquid upon appropriate heating, or that are soluble in hydrocarbon solvents. Applicable objects include: diesel fuel, jet fuel, kerosene, other distillate oils, naphtha, residual oil, lubricating base oil, hydraulic oil, crude oil, motor gasoline, alcohol-containing oils, and biodiesel.
• 1.2 uses ASTM D6259 statistical process to calculate the pooled determination limit of this standard as 3 mg/kg.
• 1.2.1 For a specific laboratory instrument, its determination limit and precision depend on the instrument's X-ray tube power (low or high power), sample type, and the laboratory's specific operation of this method.
• 1.3 Samples with sulfur content exceeding 4.6% (mass fraction) may be diluted to bring their sulfur content within the applicable range; the determination results of diluted samples will have higher error than the determination results of undiluted samples mentioned in Chapter 14.
• 1.4 Volatile samples (such as high-vapor-pressure gasoline or light hydrocarbons) may not conform to the precision of this standard due to selective loss of light components during analysis.
• 1.5 Basic assumption: The standard sample and the sample to be tested have very matching matrix materials, or the matrix materials are different as described in 12.2. Differences in the carbon-to-hydrogen mass ratio between the standard sample and the sample to be tested, or the presence of other interfering elements or substances (see Table 1), will cause mismatch of matrix materials.
• 1.6 adopts SI International System of Units.
• 1.7 only describes certain special hazards related to the test process, and does not mention all safety issues.
• 5.4 points out: SH/T 0631 is often used as a recommended test method for determining lubricating oils and lubricating oil additives with sulfur content greater than 100 mg/kg; this standard is not suitable because it cannot determine the elements present in lubricating oils and their additives that cause matrix material mismatch.

3. Method principle

Place the sample in the X-ray beam, and measure the intensity of the sulfur Kα spectral line at the wavelength of 0.5373 nm. Subtract the background intensity measured at the recommended wavelength of 0.5190 nm (for a rhodium target X-ray tube, it is 0.5437 nm) from the highest intensity, and use the net count rate as the result to compare with a pre-established calibration curve (see Chapter 12), thereby obtaining the sulfur content expressed as mass fraction or milligrams per kilogram (mg/kg).

4. Reagents and materials (purity / preparation)

• 7.1 Reagent purity: All chemical reagents used in the test are of analytical grade. If reagents of other purities can ensure that the accuracy of the determination results is not reduced, they may also be used.
• 7.2 Di-n-butyl sulfide: superior grade pure. The molecular formula is C₈H₁₈S (the original text source-text extraction is CsHgS, requires confirmation against the source standard (source-text extraction)). When calculating the accurate concentration of the calibration sample, the certified sulfur content and its purity should be used (see 9.1). Warning: Di-n-butyl sulfide is flammable and toxic. Its prepared solution has a shelf life of only a few months. Note: Since impurities may also be combined sulfur, it is necessary to know not only the purity of di-n-butyl sulfide but also the sulfur concentration of di-n-butyl sulfide. Its sulfur content can be determined by comparative analysis with reference materials from NIST or other major standardization bodies, and dilution with sulfur-free white oil.
• 7.3 Drift correction check substance (optional): A suitable drift check substance should be a substance whose measured value is stable when repeatedly placed in the X-ray. Stable liquids such as polysulfides, glass, or metal-type samples are suitable drift check substances. Liquids, powders, and solid materials that decompose cannot be used. Suitable sulfur-containing substance samples include renewable liquid petroleum samples, metal alloys, or fused glass disks. The count rate of the check substance over a certain period should be less than the relative calculation error of 1%. During calibration (see 9.4) and analysis (see 10.1), the count rate of the check substance should be measured to calculate the drift correction factor (see 11.1).
• 7.3.1 Although manual calculation is possible, drift correction is usually performed by software. For a highly stable X-ray spectrometer, the magnitude of a single drift correction factor may not differ significantly.
• 7.4 Polysulfide oil: generally a nonyl polysulfide diluted with a hydrocarbon and having a known sulfur content. Warning: May cause allergic skin reactions. Note: Polysulfide oil is generally a high-molecular-weight oil containing 50% (mass fraction) sulfur content. When completely dissolved in white oil, it still maintains its original characteristics, such as low viscosity, low volatility, and long-lasting effectiveness. Its sulfur content can be determined by comparative analysis with reference materials from NIST or other major standardization bodies, and dilution with sulfur-free white oil.
• 7.5 Mineral oil: white oil with a sulfur content lower than 2 mg/kg, analytical grade, or other suitable matrix material with a sulfur content lower than 2 mg/kg. When the predicted sample sulfur content is lower than 200 mg/kg, the sulfur content of the matrix material should also be included in the calculation of the calibration standard concentration (see 9.1). When the sulfur content of the solvent or sample is unknown, its sulfur content should be determined. Use the purest grade of substance for calibration. In addition, determining its carbon-to-hydrogen mass ratio is also very important (see Chapter 12 and Figure 1).
• 7.6 X-ray transparent film: Any transparent film resistant to sample corrosion, without a cup (requires confirmation against the source standard (source-text extraction)), sufficient to be penetrated by X-rays. Films include polyester film, polypropylene film, polycarbonate film, and polyimide film. However, samples with high aromatic content can dissolve polyester film and polycarbonate film.
• 7.7 Helium: minimum purity of 99.9% (volume fraction).
• 7.8 Counting gas: used for instruments with a flow-through proportional counter. Its purity should be consistent with the specifications in the instrument manual.
• 7.9 Sample cell: consistent with the geometric dimension requirements of the sample and the fluorescence spectrometer. For samples with sulfur content lower than 50 mg/kg, disposable sample cells should be used.
• 7.10 Calibration control sample: liquid petroleum or petroleum product with known sulfur content (including polysulfide oil, di-n-butyl sulfide, thiophene, etc.) and not used to prepare the calibration curve. Generally used to determine the accuracy and precision of the initial calibration curve (see 9.5).
• 7.11 Quality control sample: stable petroleum product or solid sample that can represent the characteristics of the sample being tested, used to check whether the system is in a satisfactory working state (see Chapter 13). Note 1: It is recommended to check system control by using quality control samples and control charts. Each laboratory must have a quality control procedure. Note 2: If these typical samples are relatively stable, suitable quality control samples can often be prepared by mixing their retained portions. As check substances, solid materials are commonly used. The shelf life of quality control samples must be very long.

5. Instruments and equipment

• 6.1 Wavelength dispersive X-ray fluorescence spectrometer, with a matched wavelength detector range of 0.52 nm~0.55 nm (most suitable at 0.537 nm). It has the most suitable sensitivity for sulfur, and the instrument should be equipped with the following devices:
  • 6.1.1 Optical path, specified by the instrument supplier; helium is more suitable, followed by air and nitrogen.
  • 6.1.2 Pulse height monitor, or other energy discrimination device.
  • 6.1.3 Detector, used to detect X-rays with wavelengths in the range of 0.52 nm~0.55 nm.
  • 6.1.4 Analyzing crystal, suitable for dispersing sulfur Kα and background X radiation within the detector angle range. Pentaerythritol or germanium is usually used. Other crystals may also be used according to the instrument supplier's recommendations.
  • 6.1.5 X-ray tube, capable of exciting sulfur Kα radiation. Although other anode tubes may be used, anode tubes with rhodium, chromium, and scandium are most commonly used. Note: Exposure to large amounts of high-energy radiation such as that generated by X-ray spectrometers is very harmful to the human body. Analysts should take certain protective measures, including not only primary X-rays but also secondary and scattered X-rays. Therefore, X-ray spectrometers should be used correctly in accordance with operating procedures.
• 6.2 Analytical balance, accurate to 0.1 mg, with a maximum range of 100 g.

6. Recommended operating conditions

• 10.1 Instrument setup: Before using any wavelength dispersive X-ray fluorescence spectrometer, the instrument must be set up according to the manual. Refer to the manual and perform a quality control check on the spectrometer.
• 10.2 Goniometer installation: Ensure that the position of the goniometer is accurate. Before calibrating the goniometer, for each element and its applied background, the angle is set by the pulse height monitor. The angle should be checked first, then the pulse height monitor; if the setting of the pulse height monitor is changed, the angle must be checked again. For the determination of samples with high sulfur content, the sulfur Kβ line may be selected, whose sensitivity is lower than the 10% of the Kα line.
• 10.3 Explanation of instrument interference factors: including crystal fluorescence, overlap of tube lines, and spectral contamination of certain elements in instrument materials. Lead is a very important influencing factor in sulfur determination. Many interference factors can be avoided by selecting window settings, such as the installation of a pulse height monitor; for elemental interference, different analytical lines can be used, and high-resolution collimators can be used to minimize tube line overlap and crystal selection.
• 10.4 When using coefficient F' in equation (8) (Chapter 11), periodically analyze blank samples to determine coefficient F'. Use sulfur-free samples such as matrix materials to determine the appropriate sulfur peak intensity and background angle.
• 10.5 Using a technique consistent with good usage practices for the specific instrument, place the test specimen into an appropriate sample cell. Ensure that the sample cell is filled beyond the minimum depth; when the filled sample exceeds this depth, the change in count rate becomes insignificant. Usually, fill the sample cell to more than one third of its capacity. The sample cell should have a small vent hole unless the sample cell is sealed.
• 10.6 Place the test specimen in the X radiation beam so that the X radiation path length reaches equilibrium.
• 10.7 Determine the intensity of sulfur Kα radiation at the 0.5373 nm wavelength by measuring the count rate at the precise angular position for this wavelength. Note: It is recommended to take a sufficient number of counts to satisfy a coefficient of variation (% relative deviation) of 1.0% or below 1.0%. When, because of either sensitivity or concentration, sufficient counts cannot be collected to achieve a coefficient of variation of 1.0%, use a recognized technique that can obtain the maximum statistical precision within the time specified for each analysis.
• 10.9 Measure the background count rate at the previously selected and fixed angular position, close to the sulfur Kα peak intensity. Note: The suitability of all background positions depends on the X X-ray tube anode used. It is recommended that when chromium or scandium is used, the wavelength is 0.5190 nm; while the most suitable wavelength for a rhodium X-ray tube is 0.5437 nm (the source text source-text extraction is "germanium X-ray tube", requires confirmation against the source standard (source-text extraction)). Table 3 lists the 2θ, peak intensity and background, and angles for various crystals.
• Table 3 2θ angles for various crystals:
  • Pentaerythritol (002), 2d/nm 0.8742, sulfur Kα (0.5373 nm) 75.85°, background 0.5190 nm 72.84°, background 0.5437 nm 76.92°
  • Germanium (111), 2d/nm 0.6532, sulfur Kα (0.5373 nm) 110.68°, background 0.5190 nm 105.23°, background 0.5437 nm 112.68°

7. Sampling and Test Specimen Preparation

• 8.1 Unless otherwise specified, sampling shall be carried out in accordance with GB/T 4756 (the source text source-text extraction is "GB/T478", requires confirmation against the source standard (source-text extraction)).
• 8.2 If reusable sample cells are used, the sample cell should be cleaned and dried before each use; disposable sample cells cannot be reused. A new X radiation-transparent film shall be used for each sample determination, avoiding contact with the interior of the sample cell, the window film portion, or exposure to the instrument window of X radiation. When analyzing low-sulfur content samples, oil stains from fingerprints can affect the determination results, and wrinkling of the film can also affect the sulfur X radiation transmittance (the source text source-text extraction is "transmission event degree", requires confirmation against the source standard (source-text extraction)). To ensure the reliability of test results, the film should be kept taut and clean; if the type and thickness of the window film change, the analyst should recalibrate as required. After the sample cell is filled, there should be a small vent hole unless the sample cell is of the fully sealed type.
• 8.3 Select an appropriate method for sampling, maintain the integrity of the sample under storage conditions, and preserve the sample and mixed sample. Cool straight-run oil (requires confirmation against the source standard (source-text extraction)) or other similar volatile samples, keep the sample... (the source text source-text extraction is incomplete) before determination, it should be brought to room temperature. Then leave it in the ambient environment for a period of time so that the sample is suitable for analysis.
• 8.4 The impurities or thickness (the source text source-text extraction is "ju mo", requires confirmation against the source standard (source-text extraction)) of each roll/batch of polyester film vary, and all can affect the determination results of low-sulfur content samples. Therefore, before starting to use a new roll or new batch of film, the calibration situation should be checked.
• 10.5 Sample filling requirements: fill the sample cell to more than one third of its capacity, and there should be a small vent hole unless the sample cell is sealed.
• 10.11 When the count rate is greater than the count rate of the highest point of the calibration curve, dilute the sample with matrix material until the sulfur count rate is within the limit range of the calibration curve, and repeat the determination steps described in 10.5~10.10.
• 10.13 When the concentration of interfering substances contained in the test specimen is greater than the concentrations listed in Table 1, the specimen should be diluted with matrix material to the concentrations listed in the table.
• 10.13.1 When the difference between the mass absorption coefficient multiplied by mass fraction of the test specimen and the mass absorption coefficient multiplied by mass fraction of the calibration standard is not greater than 4%~5% of the latter, the collected data (see Note to 5.1) show good X radiation results. Absorption interference is additive and can only be minimized by dilution, not completely eliminated. Table 1 can serve as reference concentrations for interfering elements. Note 1: The influence of matrix material interference can also be corrected empirically or theoretically. Note 2: Add ethanol (or methanol) to a mixture of hydrocarbons and di-n-butyl sulfide until the total mass coefficient increases by 5%; at this point the concentrations of ethanol and methanol can be obtained by calculation. When using this standard to determine the sulfur content of alcohol-containing gasoline (or M85 and M100), the allowable concentration of ethanol (or methanol) is given in Table 1.
• 10.13.2 Mix the mixture thoroughly (the mixing method mainly depends on the type of matrix material) to ensure uniformity and transfer it to the measuring instrument.
• 10.13.3 Determine the sulfur content of the mixture by the method steps described in 10.5~10.9, and calculate the sulfur content of the initial standard sample according to Chapter 11.

8. Analytical Procedure

• 10.1 Instrument setup: set up the instrument according to the instructions, and perform a quality control check on the spectrometer.
• 10.2 Goniometer installation and calibration: ensure that the goniometer position is accurate; first check the angle, then check the pulse height monitor; if the pulse height monitor setting is changed, recheck the angle.
• 10.3 Explanation of instrument interference factors (crystal fluorescence, line overlap, spectral contamination, etc.); lead is an important influencing factor; avoid interference through window settings, pulse height monitor installation, different analytical lines, high-resolution collimators, crystal selection, etc.
• 10.4 Periodically analyze blank samples and determine the coefficient F'.
• 10.5 Place the test specimen into an appropriate sample cell, filling beyond the minimum depth, usually to more than one third of capacity; the sample cell should have a small vent hole unless sealed.
• 10.6 Place the test specimen in the X radiation beam so that the X radiation path length reaches equilibrium.
• 10.7 Determine the intensity of sulfur Kα radiation at the 0.5373 nm wavelength.
• 10.8 Calculate the coefficient of variation CV according to formula (6): CV = [100(Np + Nb)^(1/2)] / (Np - Nb) (the source text source-text extraction is CV = [100 (N® + N.) E] / (NP N.), requires confirmation against the source standard (source-text extraction)). Where: CV— coefficient of variation, %; Np— the number of counts collected for the sulfur spectral line at wavelength 0.5373 nm; Nb— the number of counts collected at the background wavelength during the same time spent collecting Np counts.
• 10.9 Measure the background count rate at the previously selected and fixed angular position, close to the sulfur Kα peak intensity.
• 10.10 As shown in Chapter 11, determine the corrected count rate and calculate the sulfur content of the test specimen.
• 10.11 When the count rate is greater than the count rate of the highest point of the calibration curve, dilute the sample with matrix material until the sulfur count rate is within the limit range of the calibration curve, and repeat the determination steps described in 10.5~10.10.
• 10.12 For test specimens with total sulfur content ≤100 mg/kg, duplicate determinations are required. Each time, a new test specimen should be used for the determination, and analysis should be carried out according to 10.5~10.10. The error between the two analyses should be equal to or lower than the repeatability value specified in 14.1.1. If the error is large, carefully analyze the various possible causes of contamination during test specimen preparation and reanalyze. Duplicate determination is also to ensure the precision of low-sulfur samples.
• 10.13 When the concentration of interfering substances contained in the test specimen is greater than the concentrations listed in Table 1, the specimen should be diluted with matrix material to the concentrations listed in the table.
• 10.13.1 Explanation of absorption interference and dilution requirements (see Section 7).
• 10.13.2 Mix the mixture thoroughly to ensure uniformity and transfer it to the measuring instrument.
• 10.13.3 Determine the sulfur content of the mixture by the method steps described in 10.5~10.9, and calculate the sulfur content of the initial standard sample according to Chapter 11.

9. Qualitative and Quantitative Determination and Calculation Formulas

• 9.1 Preparation of calibration curve: dilute authenticated di-n-butyl sulfide (see 7.2) with sulfur-free white oil or other appropriate matrix material to obtain a series of calibration samples of different concentrations, so as to prepare the calibration curve. The content of unknown samples must be within the determination range of this curve. The relevant standard sulfur content ranges are listed in Table 2. When calculating samples with sulfur content lower than 0.02% (mass fraction) (200 mg/kg), taking into account the sulfur content in the matrix material, calculate using formula (1):
  • S = [(m_DBS × S_DBS) + (m_wo × S_wo)] / (m_DBS + m_wo) … (1)
  • Where: S— sulfur content in the calibration sample, expressed as % (mass fraction); m_DBS— actual mass of di-n-butyl sulfide, in grams (g); S_DBS— sulfur content in di-n-butyl sulfide, generally 21.91% (mass fraction); m_wo— mass of white oil, in grams (g); S_wo— sulfur content of white oil, expressed as % (mass fraction).
  • Note: If required, calibration samples outside the concentration range listed in Table 2 may be prepared and analyzed.
• Table 2 Sulfur calibration curve concentration range:
  • 0 (mg/kg)~1 000 (mg/kg): 0.0, 0.100, 0.250, 0.500, 1.000, 2.0, 3.0, 4.0, 5.0, 10*, 100*, 250, 500, 750, 1000 (* matrix material; repeatedly analyze these calibration samples and use their average value for the calibration curve)
  • 0.10% (mass fraction)~1.00% (mass fraction): 0.100, 0.250, 0.500, 1.000
  • 1.0% (mass fraction)~5.0% (mass fraction): 1.0, 2.0, 3.0, 4.0, 5.0
• 9.1.1 Calibration samples can be prepared by mixing certified reference materials of the same matrix substance; the sulfur content and its uncertainty of these mixtures must have been certified by a certification body.
• 9.1.2 Calibration samples can also be prepared by diluting polysulfide oil with white oil. The calibration curve prepared with polysulfide oil should be corrected with certified reference materials. After the polysulfide oil calibration curve has been established, the calibration samples should be placed in amber glass bottles, stored at room temperature, and protected from direct sunlight. The concentration range of polysulfide oil calibration samples can vary from a few parts per million to a few percent. Mix each newly prepared polysulfide oil calibration sample thoroughly to make it homogeneous. These sulfur mixtures have relatively high molecular weights and low vapor pressures, which suppress membrane diffusion of the X radiation through the transparent film. Therefore, an autosampler can be used during the measurement. Note: If the accurate sulfur content of a commercially available sample is known and its concentration is approximately within the concentration range listed in Table 2, it may also be used.
• 9.2 According to the procedures specified in Chapter 10 and Chapter 11, carefully measure the net sulfur radiation intensity emitted by each calibration sample to establish the calibration curve data.
• 9.2.1 Calibration samples with a total sulfur content equal to or less than 100 mg/kg should be measured twice; either the two measured values or their average may be used to construct the calibration curve. All samples within the measurement range (see 10.12) should be analyzed twice, and the measurement results are given in 12.1.1.
• 9.3 Use the software provided by the equipment supplier to establish a calibration model. This calibration model may take any of the following forms (the exact formula established with reference to the software provided by the instrument supplier):
  • Linear calibration: Cs = a + B·I … (2)
  • Calibration for matrix substance effects: … (3) (the original formula source-text extraction is incomplete)
  • Optional calibration for matrix substance effects: Cs = a + (1 + Σα_j·C_j)·B·I … (4) (the original source-text extraction is incomplete, requires confirmation against the source standard (source-text extraction))
  • Quadratic term calibration: Cs = a + B·I + c·I² … (5) (the original source-text extraction is incomplete, requires confirmation against the source standard (source-text extraction))
  • Where: Cs— sulfur content, % (mass fraction), depending on the values of the calibration variables a, B, c; I— net intensity of the measured sulfur radiation; a— intercept of the calibration curve; B— slope of the calibration curve; c— correction coefficient of the quadratic term; α_j— influence factor of the interfering element (j) on the determination of sulfur (i); when a is cancelled by the matrix substance, the interfering element may be sulfur, and the empirical correction value may be used directly; C_j— concentration of the interfering element (j).
  • Note: The coefficients α_j in the formula can be determined empirically and theoretically. Instrument vendors can often provide software for calculating the theoretical α_j.
• 9.3.1 When constructing the calibration curve, if necessary, calibration curves for multiple ranges may be selected according to the range of sulfur content of the samples routinely measured. For example (Table 2): the ranges of sulfur content (mass fraction) include: 0%~0.1%, 0.1%~1.0%, 1.0%~5.0%. Note: The calibration curve is linear with the sulfur content in the range of 0.10% (mass fraction). When plotting the curve for this range, the analyst should choose a linear model. When plotting the curve for higher sulfur content, other calibration models (matrix substance effect calibration or quadratic term calibration) may be chosen.
• 9.4 When using a drift correction monitor, measure the intensity of the drift correction monitor during calibration. The measured value is consistent with the coefficient A of formula (7) in 11.1.
• 9.5 Immediately after completing calibration, measure the sulfur content of the calibration sample (see 7.10). The difference between the two measured values should be within the repeatability range of this standard (see 14.1.1); otherwise, the stability of the instrument and the repeatability of sample preparation are not credible, and calibration should be performed again. When evaluating the calibration curve, the degree of matrix substance mismatch between the sample to be tested and the calibration sample should also be considered. As described in Chapter 13, if this method is within statistical control, a statistical quality control chart for these substances may also be established.
• 11.1 When using the drift correction monitor described in 7.3, calculate the correction factor for daily instrument sensitivity changes according to formula (7): F = A/B … (7). Where: F— correction factor; A— count rate of the drift correction monitor measured during calibration (see 9.4); B— count rate of the drift correction monitor measured during analysis (see 10.1).
• 11.2 Determine the corrected net count rate (intensity) according to formula (8): R = [(Np/S1) - (Nb·F'/S2)]·F … (8). Where: R— corrected net count rate; Np—0.5373 total counts collected at wavelength nm; Nb— total counts collected at the background position selected from clause 10.8; S1 and S2— time required to collect Np, Nb counts, s; F'—count rate/s of a sulfur-free sample at wavelength 0.5373 nm under the selected background; F— optional factor (see Note). Note: For some instruments, this coefficient in formula (8) is not necessary or required. In this case, F is uniformly specified. It is recommended that users plot the F coefficient chart based on the stability of instrument use and standard quality control principles, and establish relevant criteria.
• 11.2.1 The coefficient F' in formula (8) is optional. Usually, a multichannel spectrometer should be used to measure peak intensity and background intensity. Note: Plotting the F' coefficient chart, even if not used in formula (8), can alert the user to changes in instrument operation caused by contamination of system components such as crystals, high-resolution collimators, and fixed windows.
• 11.3 Insert the corrected net count rate of formula (8) into the calibration model of Chapter 9 to calculate the sulfur content of the test sample. In many cases, the instrument vendor will provide calculation software.
• 11.4 Calculate the sulfur content of the diluted test sample according to formula (9): S = S' × [(ms + mo)/ms] … (9). Where: S— sulfur content of the test sample, % (mass fraction); S'—sulfur content of the test sample after dilution, % (mass fraction); ms— mass of the test sample, g; mo— mass of the diluent, g.
• 11.4.1 When input of the required mass is required, the instrument vendor may provide software to perform this calculation.
• 11.5 When analyzing fuels with high ethanol or high methanol content (see Table 1) using a calibration curve prepared with white oil as the matrix substance, correct the measurement result obtained for 11.3 according to formula (10) (see Note 2 in 10.13.1): S_corr = M/F … (10). Where: S_corr— corrected sulfur content of the test sample, % (mass fraction); M— measured sulfur content of the test sample, % (mass fraction); F— correction factor, for example: for M85, F=0.59; for M100, F=0.55.
• 11.5.1 As described in 5.2, if the matrix substances of the calibration sample and the test sample are consistent, correction by formula (10) may be omitted.
• 11.6 For information on regression calculation, refer to the content in ASTM D7343.
• 12.2 When the matrix substance used for the calibration curve is white oil, correction can be performed using formula (11): S_C/H = [1.195 - 0.0164(C/H)_whiteoil] / [1.195 - 0.0164(C/H)] × S_whiteoil … (11). Where: S_C/H— sulfur concentration of the test sample after correction for matrix substances with different carbon-to-hydrogen mass ratios; (C/H)_whiteoil— carbon-to-hydrogen mass ratio of white oil, obtained from Figure 1, with a value of 5.7; (C/H)— carbon-to-hydrogen mass ratio of the sample; S_whiteoil— sulfur concentration of white oil obtained from the calibration curve.
• 12.2.1 For unknown samples, the value of the carbon-to-hydrogen mass ratio can be input; for different carbon-to-hydrogen mass ratios, S_whiteoil requires correction. If different solvents are used to dilute the calibration samples, the carbon-to-hydrogen mass ratio of the solvent can be replaced by (C/H)_solvent. Differences in the carbon-to-hydrogen mass ratio between the calibration sample and the test sample will lead to errors in the measurement results (see Table 6). If the error is too large, formula (11) should be used for correction.

10. Precision

• 14.1 The reliability of test results is judged according to the following provisions (95% confidence level), including 27 types of samples such as gasoline, distillate oil, biodiesel, residual oil, and crude oil, and the precision of this standard was determined. For all types of samples, the pooled measurement limit is limited to approximately 3 mg/kg. Regardless of whether high-power or low-power X X-ray fluorescence spectrometers are used, the precision for gasoline and diesel samples is listed in Appendix A. These statistical tables apply to samples whose interfering substance content is lower than the values listed in Table 1 (see 1.4).
• 14.1.1 Repeatability (r): The difference between two results obtained by the same operator in the same laboratory, using the same instrument, following the procedures specified in the method, on the same sample for repeated determinations shall not exceed the value calculated by formula (12); the calculated values are shown in Table 4. r = 0.04519(X + 20)^0.99288 … (12). Where: X— total sulfur content, in milligrams per kilogram (mg/kg).
• 14.1.2 Reproducibility (R): The difference between two independent results obtained by different operators in different laboratories, using different instruments, following the procedures specified in the method, on the same sample shall not exceed the value calculated by formula (13); the calculated values are shown in Table 4. R = 0.1469(X + 20)^0.9988 … (13). Where: X— total sulfur content, in milligrams per kilogram (mg/kg).
• 14.1.3 The repeatability and reproducibility values for gasoline and diesel samples from the interlaboratory studies mentioned above are listed in Appendix A. The precision values and charts for high-power (>1 kW) instruments are also described in Appendix A.
• Table 4 Precision of all samples (S/(mg/kg) | Repeatability r/(mg/kg) value of formula (12) | Reproducibility R/(mg/kg) value of formula (13)):
  • 1.0 | 0.9 | 2.5
  • 5.0 | 1.1 | 2.9
  • 10.0 | 1.6 | 3.5
  • 25.0 | 2.3 | 5.0
  • 50.0 | 4.00 | 7.6
  • 100 | 9.8 | 12.5
  • 500 | 49 | 91
  • 1000 | 91 | 164
  • (Original source-text extraction Table 4 data is noisy, some values require confirmation against the source standard (source-text extraction))
• 14.2 Bias: The interlaboratory study included 10 working standard samples of NIST. Table 5 gives the certified sulfur content values, interlaboratory circulation test (RR) values, measured carbon-to-hydrogen mass ratios, absolute deviations, and relative deviations. Table 6 compares the NIST values with the sulfur concentration corrected for carbon-to-hydrogen mass ratio. It is assumed that the white oil has a carbon-to-hydrogen mass ratio of 5.698 (C₂₈H₅₈, requires confirmation against the source standard (source-text extraction)).
• 14.2.1 Figure 1 shows the relative sulfur sensitivity as a function of the carbon-to-hydrogen mass ratio.
• 14.2.2 Based on the analysis of 10 working standard samples of NIST, especially after applying the correction for carbon-to-hydrogen mass ratio, for any working standard sample and its type, within the range of the determined reproducibility, there is no significant deviation between the certified values and the results obtained from the interlaboratory study.
• Appendix A (Informative Appendix) Explanation of Precision:
  • A.1 Precision for gasoline: The interlaboratory study used 5 gasoline samples, with total sulfur content approximately in the range of 5 mg/kg~70 mg/kg. Samples: Sample 1 NIST SRM 2298 high-octane gasoline; Sample 2 NIST SRM 2296 ethyl tert-butyl ether gasoline containing 13%; Sample 3 NIST SRM 2299 reformulated gasoline; Sample 4 gasoline containing 5% ethanol; Sample 11 regular unleaded gasoline.
    • A.1.1 Repeatability (r): r = 0.5006X^0.4377 … (A.1). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.1, and the data plot is shown in Figure A.1.
    • A.1.2 Reproducibility (R): R = 1.4533X^0.4877 … (A.2). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.1, and the data plot is shown in Figure A.1.
    • Table A.1 Precision values for gasoline (S/(mg/kg) | Repeatability value of formula (A.1) r/(mg/kg) | Reproducibility value of formula (A.2) R/(mg/kg)): 5.0 | 1.01 | 2.94; 10.0 | 1.37 | 3.98; 25.0 | 2.05 | 5.95; 70.0 | 3.21 | 9.33.
  • A.2 Precision for diesel: The interlaboratory study used 5 diesel samples, with total sulfur content approximately in the range of 11 mg/kg~5 500 mg/kg. Samples: Sample 5 NIST SRM 2723a; Sample 7 NIST SRM 2770; Sample 8 NIST SRM 2724b; Sample 15 diesel; Sample 22 diesel containing 5% biodiesel.
    • A.2.1 Repeatability (r): r = 0.1037X^0.3000 … (A.3). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.2, and the data plot is shown in Figure A.2.
    • A.2.2 Reproducibility (R): R = 0.3856X^0.3000 … (A.4). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.2, and the data plot is shown in Figure A.2.
    • Table A.2 Precision values for diesel (S/(mg/kg) | Repeatability value of formula (A.3) r/(mg/kg) | Reproducibility value of formula (A.4) R/(mg/kg)): 11 | 0.71 | 2.63; 25 | 1.36 | 5.06; 100 | 4.13 | 15.35; 500 | 15.0 | 55.6; 0.100% | 26.0 | 96.9; 0.550% | 102 | 379.
  • A.3 Precision for high-power instruments: The power of the X X-ray fluorescence source affects the precision of its measurement results; greater than 1000 W is high power, less than 1000 W is low power. In the interlaboratory study, 5 participants used low-power instruments, and 10 participants used high-power instruments.
    • A.3.1 Repeatability (r): r = 0.08681X^0.8383 … (A.5). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.3.
    • A.3.2 Reproducibility (R): R = 0.3086X^0.8383 … (A.6). Where: X— total sulfur content, mg/kg. The calculation results are shown in Table A.3.
    • Table A.3 Precision values for all samples under high-power instruments (S/(mg/kg) | Repeatability value of formula (A.5) r/(mg/kg) | Reproducibility value of formula (A.6) R/(mg/kg)): 1.0 | 0.09 | 0.31; 5.0 | 0.33 | 1.19; 10.0 | 0.60 | 2.13; 25.0 | 1.28 | 4.58; 50.0 | 2.31 | 8.20; 100.0 | 4.12 | 14.66; 500 | 15.9 | 56.5; 0.100% | 28.4 | 101.0; 0.500% | 109.5 | 389.