SH/T 0803-2007 Determination of polychlorinated biphenyl contaminants in insulating oil — Capillary gas chromatography

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.

Card type: Test method

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

• Standard number: SH/T 0803—2007
• Chinese name: Determination of polychlorinated biphenyl contaminants in insulating oil — Capillary gas chromatography
• English name (source-text extraction): Determination of polychlorinated biphenyls (PCBs) in insulating liquids — Method of determination by capillary column gas chromatography (the original English name source-text extraction is severely noisy; this treatment is only for readability and requires confirmation against the source standard (source-text extraction))
• Adoption relationship: modified adoption of IEC 61619:1997 (MOD)
• Replacement relationship: not provided in the available source
• Publication/implementation: published by 2007-08-01, implemented by 2008-01-01
• Publishing unit: National Development and Reform Commission of the People's Republic of China
• Custodian: Sinopec Research Institute of Petroleum Processing
• Drafting unit: Karamay Lubricating Oil Research Institute of PetroChina Lubricant Company
• Main drafters: Wang Jinhui, Hou Yudong, Li Yungang, Wang Peng, Zhang Bo, Liu Ying

2. Scope of application and non-applicability

• 1.1 specifies a method for determining the content of polychlorinated biphenyls in non-halogenated insulating oils, using a high-resolution capillary gas chromatograph equipped with an electron capture detector (ECD).
• 1.2 is applicable to the determination of the total content of polychlorinated biphenyls (PCB) in insulating oils, and is especially suitable for detailed analysis of PCB congeners. When detailed congener analysis is not required, other methods may also be used, such as IEC 60997.
• 1.3 is applicable to the determination of PCB in unused, regenerated (including dechlorinated, chemically/physically treated) and used insulating oils.
• The application of 1.4 may involve certain hazardous materials, operations and equipment, but it does not purport to address all safety problems associated with this. Users are responsible for establishing appropriate safety and protection measures before use and for clarifying the restricted scope of application.
• Not applicable: no explicit exclusion clause is provided in the available source.

3. Method principle

• The congeners of 4.1 PCB are determined by temperature-programmed gas chromatography, and the chromatograph uses a high-performance capillary column to separate PCB into individual congeners or similar congener groups.
• The presence of 4.2 mineral oil reduces the sensitivity of the electron capture detector. To minimize this effect, the sample is diluted 100 times.
• The 4.3 sample preparation (purification) process is used to remove most impurities that may affect the determination.
• In 4.4, a reference substance is introduced when calculating the experimental relative retention time (ERRTs), and the ERRTs data files of peaks are compared to identify individual congeners or unresolved congener groups. For quantitative determination, an internal standard substance is introduced.
• In 4.5, the relative response factors (RRF) obtained from the data files (10.1 entries), after correction by the experimental relative response factors (ERRFs) of the reference substance, are used to identify peaks and quantitatively determine individual (or groups of) congeners, and the results are summed to obtain the total PCB content.

4. Reagents and materials (purity / preparation)

• 6.1 General: all reagents and materials, including those used in the purification process, shall not contain PCB or substances that cause a response in ECD.
• 6.1.1 Solvent: hexane, heptane, cyclohexane or isooctane (2,2,4- trimethylpentane), chromatographic grade, free of PCB and with low content of substances that cause a response in ECD.
• 6.1.2 Hexachlorobenzene: purity greater than 99%, used to test detector sensitivity.
• 6.1.3 Insulating oil: verified to contain no PCB or other interfering substances.
• 6.1.4 Congener 30 solution (C30): content in solvent 10 mg/L. Purchase the solution or prepare it from pure substances (purity at least 99%).
• 6.1.5 Congener 209 (DCB), decachlorobiphenyl solution: content in solvent 10 mg/L. Purchase the solution or prepare it from pure substances (purity at least 99%).
• 6.1.6 Calibration solution containing selected PCB: a calibration mixture with a defined content in solvent, including at least the following PCB congeners, each at a content of 10 μg/L: 18, 28, 31, 44, 52, 101, 118, 138, 149, 153, 170, 180, 194 and 209.
  • Note: 18, 28, 31, 44, 52, 101, 118, 138, 149, 153, 170, 180, 194 and 209 are the numbers assigned by IUPAC to PCB congeners.
• 6.2 Commercially available PCB reference materials
  • 6.2.1 Aroclor 1242, Aroclor 1254 and Aroclor 1260 solvent solutions: required concentration 50 mg/L or higher, usually 1000 mg/L.
    • Note: Aroclor is a registered trademark of the American company Monsanto; similar products include Clophen from the German company Bayer, Phenoclor and Pyralene from the French company Prodelec, and Kane-chlor from the Japanese company Kanegafuchi, etc. This standard uses Aroclor.
  • 6.2.2 Aroclor 1242, Aroclor 1254 and Aroclor 1260 oil solutions: solutions of Aroclor 1242, Aroclor 1254 and Aroclor 1260 containing 50 mg/kg in unused insulating oil, either purchased standard solutions or prepared from pure substances.
• 6.3 Gases for gas chromatography
  • 6.3.1 Carrier gas: nitrogen or hydrogen, high-purity gas, purity at least 99.99%.
  • 6.3.2 Auxiliary gas: argon:methane (95%:5%); nitrogen with purity at least 99.99% may also be used.
• 6.4 Internal standard substances/reference solutions
  • Note: reference materials should be stored in a cool, dark place.
  • 6.4.1 Internal standard solution 2 (IS2): 2 mg/L C209 (DCB), 2 mg/L C30. Use a pipette to transfer 5 mL DCB solution and 5 mL C30 solution into a 25 mL volumetric flask, and dilute to the mark with solvent.
  • 6.4.2 Internal standard solution 0.5 (IS0.5): 0.5 mg/L C209 (DCB), 0.5 mg/L C30. The preparation procedure is the same as 6.4.1, using a 100 mL volumetric flask.
• 6.5 Test mixed solution: in a 25 mL volumetric flask, weigh in 0.625 g of 50 mg/kg of Aroclor 1260, 0.625 g of 50 mg/kg of Aroclor 1254 and 1.25 g of 50 mg/kg of Aroclor 1242 insulating oil solution, add 1 mL IS2 solution with an accurate dropper, and dilute to volume with solvent. The solution is treated according to 12.1.3 before use.
• 6.6 Calibration-congener mixture stock solution: in a 25 mL volumetric flask, weigh 2.5 g insulating oil to the nearest 0.001 g, add 1 mL calibration PCB congener mixture, and dilute to the mark with solvent.
• 6.7 Congener mixture calibration solution (for response factors): use the 500 μL solution in the purification process, and the final solution is used to determine the relative response factors. Prepare a fresh solution once a month.

5. Instruments and equipment

• 5.1 Gas chromatograph
  • 5.1.1 High-resolution gas chromatograph: has highly precise and reproducible oven temperature control; when a suitable column and operating conditions are selected, the test mixture can be resolved at least as in Appendix A (90 peaks), with relative retention times reproducible within ±0.0015.
  • 5.1.2 Gas inlet lines (carrier gas and auxiliary gas): water vapor and oxygen filters shall be installed.
  • 5.1.3 Carrier gas supply system: the use of nitrogen or hydrogen shall be sufficient to open a separation column of 50 m and provide sufficient column head pressure.
  • 5.1.4 Oven temperature control program: the temperature control range can be set to meet the required resolution.
• 5.2 Injector: either an “on-column (ON COLUMN)” or a “split/splitless (SPLIT/SPLITLESS)” injector may be selected.
• 5.3 Separation column: a fused silica capillary column coated with a cross-linked 5% phenylmethylsilicone stationary phase, or a column with a similar chemically bonded phase, may be used. Its dimensions are as follows:
  • Length: 50 m or 60 m
  • Internal diameter: 0.2 mm～0.35 mm
  • Film thickness: 0.1 μm～0.25 μm
• 5.4 Detector: high-temperature electron capture detector, which shall have sufficient sensitivity; when 1 pg hexachlorobenzene is injected into the chromatographic column, the signal-to-noise ratio shall be greater than 20. The detector shall be operated within its linear range.
• 5.5 Data processing system: any data processing system that can run appropriate software on the instrument and operate according to the following steps may be used:
  • Raw data → peak data to be integrated (area or peak height) → calculate each peak's ERRT → identify peaks within the window, compare with the "all probabilities" or "all possibilities" library file → calculate each peak's PCB content (using the internal standard reference material method) → sum each peak's PCB content to calculate the total PCB content.
• 5.6 glassware
  • 5.6.1 volumetric flasks: 100 mL, 50 mL, 25 mL, 10 mL, 5 mL.
  • 5.6.2 syringes and pipettes: 50 μL syringes; 1 mL～5 mL pipettes; 10 μL gas chromatography precision syringes.
  • 5.6.3 bulb pipettes: 1 mL, 2 mL manual.
• 5.7 extraction columns and accessories used for sample preparation
  • 5.7.1 purchased or self-packed extraction columns: 3 mL silica gel column absorbent mass 500 mg, particle size 40 μm; 3 mL benzenesulfonic acid column, absorbent mass 500 mg, particle size 40 μm.
  • 5.7.2 column connection adapter: used to connect two extraction columns.
  • 5.7.3 vacuum manifold column processor (optional).

6. Recommended operating conditions

• 9.1 General: the reference operating conditions given below are feasible, but for a specific gas chromatograph they still need to be optimized so that the chromatogram is similar to the results obtained with the diluted test mixture in Appendix A. With hydrogen as carrier gas, 30 min～40 min can achieve satisfactory separation; with helium as carrier gas, the time required for separation is longer, being 55 min～60 min.
• 9.2 Injector: set up the injector according to the manufacturer's instructions; typical settings in this analytical method are as follows:
  • Split/splitless injector:
    • Splitless mode: vaporization chamber temperature is 240℃～280℃;
    • Split mode: vaporization chamber temperature is 250℃～280℃, split ratio 5:1～50:1.
  • On-column injector: vaporization chamber temperature is 50℃～110℃, depending on the solvent used.
• 9.3 Oven temperature control (original table, source-text extraction heavy noise, the following is a readable reorganization, requires confirmation against the source standard (source-text extraction)):

• 9.4 Carrier gas flow rate: adjust the inlet pressure so that at 130℃, the flow rate of hydrogen in the chromatographic column is 1 mL/min, and for helium it is 270 kPa.
  • Note: hydrogen as carrier gas can effectively reduce column pressure and shorten analysis time.
• 9.5 Electron capture detector (ECD) settings
  • 9.5.1 Temperature: 300℃～350℃.
  • 9.5.2 Electrical control: set according to the manufacturer's recommendations so that the detector achieves optimum linearity.
  • 9.5.3 Auxiliary gas: flow rate 20 mL/min～50 mL/min, according to the manufacturer's recommendations.

7. Sampling and test sample preparation

• 8.1 Sampling
  • 8.1.1 Sampling should be carried out as described in GB 7597.
  • 8.1.2 To avoid sample cross-contamination, it is recommended that all auxiliary materials used (tubes, connectors, stopcocks, fittings), etc. should be disposable and free of interfering substances.
• 8.2 Sample preparation
  • 8.2.1 Except for replaceable dropper tips and extraction columns which are made of plastic, only glass and metal apparatus may be used for sample preparation and determination; all apparatus should be free of PCB and interfering substances.
  • 8.2.2 If an aqueous phase is present in the sample, it should be removed before analysis, for example by centrifugation. Emulsified water will appear turbid; this can be remedied by adding a certain proportion of sodium sulfate and shaking until a clear sample is obtained.
  • 8.2.3 The sample should be homogeneous; it may be shaken by hand for 3 min, and an ultrasonic bath may also be used.

8. Analytical procedure

• 12.1 Sample treatment (purification)
  • 12.1.1 Test portion: weigh 0.9 g～1.0 g of sample, accurate to 0.001 g, into a 10 mL volumetric flask, add with a dropper 1 mL of internal standard substance IS2 or IS0.5, make up to the mark with solvent, and mix by shaking. If the sample has a high water content, the solution will appear opaque; in this case add anhydrous sodium sulfate and shake until a clear solution is obtained; this solution is designated solution A.
    • Note: For samples with unknown PCB content (see 12.2), for samples expected to have PCB content lower than 20 mg/kg, it is best to use IS0.5.
  • 12.1.2 Extraction column preparation: connect the silica gel column to the sulfonic acid column with an adapter. Wash three times with 2 mL of solvent to purify the extraction column, taking care not to let the absorbent in the silica gel column dry out.
  • 12.1.3 Purification process
    • 12.1.3.1 Add 500 μL±5 μL of solution A to the top of the upper sulfonic acid column.
    • 12.1.3.2 Add 0.5 mL of solvent, and apply slight vacuum so that the sample is evenly distributed in the packing of the sulfonic acid column; maintain the flow-through process for at least 30 s.
    • 12.1.3.3 Elution should be carried out at the maximum flow rate 2 mL/min; after each batch of elution from the extraction column, the liquid level is just above the adsorbent (finally with nitrogen blow-down).
    • 12.1.3.4 Wash the connection part of the extraction column twice with 1 mL of solvent each time, collect the eluate in a 5 mL volumetric flask, remove the upper sulfonic acid column and adapter, and elute the silica gel column twice with 0.5 mL of solvent. Collect the eluate in the same volumetric flask, dilute to the mark with solvent, and mix well. This solution is used for chromatographic analysis and is designated solution B.
    • 12.1.3.5 For certain samples, other purification techniques may also be used if necessary.
  • 12.1.4 Recovery
    • 12.1.4.1 Dilute Aroclor solution with solvent to 5 mg/L; this solution is designated solution C.
    • 12.1.4.2 Use 500 μL of solution C for purification; the eluate is designated solution D.
    • 12.1.4.3 Dilute solution D with solvent to 5 mL, at a concentration of 0.5 mg/L; this solution is designated solution E.
    • 12.1.4.4 Carry out by adding 50 μL of internal standard substance solution to solution D and solution E.
    • 12.1.4.5 Inject solution D and E into the chromatograph to obtain chromatograms and area tables for each solution, and use the internal standard method (see 12.7.2.2) to calculate the total PCB content in each solution. Calculate the recovery R using formula (3):
      • R = PCB(D) / PCB(E) × 100 ...... (3)
      • Where: R—— recovery, %; PCB(D)—— total PCB content in solution D, mg; PCB(E)—— total PCB content in solution E, mg.
    • 12.1.4.6 The calculated recovery should be greater than 85%.
  • 12.1.5 Commercially available PCB reference materials: use 50 mg/kg of a commercially available oil solution of the PCB mixture. If a low-concentration standard solution is needed, it may be diluted with unused insulating oil that does not contain PCB. Commercially available PCB reference materials are treated in the same way as samples and also require the purification process.
• 12.2 Background check: each batch of new solvent and blank sample prepared from insulating oil free of PCB should be tested by chromatograph as described in clause 12.1 to ensure there are no interfering peaks. A blank test is required for each batch of samples and at least every 20 samples.
• 12.3 Determination
  • 12.3.1 Inject the test sample, the commercially available PCB reference material solution, and solution B into the chromatograph, and test the chromatograph under the preferred conditions.
  • 12.3.2 Before each batch of samples, the instrument must be calibrated once with the reference material solution; if the number of samples in the same batch exceeds 10, calibrate once every 10 samples. Select a reference material similar to the sample, most typically 10 mg/kg and 50 mg/kg of Aroclor 1260 reference material oil solution (this is the most common product) or other required reference materials. To ensure quality, the blank samples and test samples used in the test should be properly stored.
    • Note: The chromatograms of all commercially available materials used should be retained to facilitate future type identification.
  • 12.3.3 Integrate the chromatogram; the chromatograph integration system will produce a list of peak numbers, retention times, peak areas and/or peak heights.
• 12.4 Determination of the test relative retention time (ERRT)
  • 12.4.1 Test the test mixture prepared in 12.1, using the same operating conditions as for sample analysis, identify all peaks by comparison with the examples in Appendix A, and calculate the ERRT of each peak according to formula (4):
    • ERRT = (t_i - t_30) / (t_209 - t_30) ... (4)
    • Where: i—— the selected peak; t—— the retention time from injection; 30—— the homologue 30 (reference substance); 209—— the homologue 209 (reference substance and internal standard substance).
    • List the results as shown in Appendix B, and store the results in the data file.
    • Note: For different gas chromatography systems, ERRT should be determined and stored in the data file; if the chromatographic working conditions change, the system needs to be recalibrated.
  • 12.4.2 Reference peaks of the data processing system: Use the ERRT values of the homologues in Table 2 to assign values to the reference peaks in the data processing system.

• Note: Except for C30 and C209, the remaining reference peaks are the main components of commercially available mixed solutions, and these reference peaks have never been found in oil samples; their numbers can only be obtained from the PCB mixed solution.
• 12.5 Calculation of the corrected relative response factors (CRRF)
  • 12.5.1 Due to the influence of the injection mode, the operating conditions of the electron capture detector, and different instruments, the test relative response factors (ERRFs) of the homologues may change, and the results may not be completely identical to literature data.
  • 12.5.2 The chromatogram is divided into nine parts (see Appendix A), each part representing one homologue listed in Table 3. These homologues usually exist as main components in commercially available mixtures.
  • 12.5.3 Using the preferred chromatographic conditions listed in Chapter 9, inject an appropriate amount of the mixture of purified homologues and calibration solvent. Determine the relative response factors of the homologues listed in Table 3 according to formula (5) (converted to DCB):
    • ERRF_i = (A_i × M_s) / (A_s × M_i) ... (5)
    • Where: A_s—— the peak area/peak height of the internal standard substance (DCB); A_i—— the peak area/height of the homologue i; M_s—— the content of the internal standard substance (DCB) (μg/mL); M_i—— the content of the homologue i (μg/mL).
    • Note: Some software packages generate the reciprocal of ERRF.
    • Take at least the average of three determinations of ERRF.

• 12.5.4 Based on the determined ERRF or the theoretical value of RRF in Table 3, calculate the correction factor K_i for each homologue according to formula (6):
  • K_i = ERRF_i / RRF_i ... (6)
  • For example: For segment 7, the calculation of homologue 180 is shown in formula (7), where the determined ERRF=1.030, the theoretical RRF=1.137 (see Appendix C).
  • K_180 = 1.030 / 1.137 = 0.906 ... (7)
• 12.5.5 Based on the calculated correction factors, for the selected homologues, multiply all peaks in each segment of the chromatogram by the correction factor. For example: For segment 7, homologue 180, each RRF is multiplied by 0.906. The resulting table of corrected relative response factors (CRRF) used for possibility and probability will serve as one of the data files.
• 12.6 Inspection of the chromatogram: Check the chromatogram for interfering peaks, chromatographic faults, or other interfering factors.
  • Note: The mixture of "tetrachlorobenzyltoluene" may be mixed in and misidentified as PCB. This can be determined by checking whether there are special peaks in the PCB region of the spectrum.
• 12.7 Calculation of results
  • 12.7.1 Qualitative analysis: Compare the sample chromatogram with the reference substance chromatogram to identify commercially available materials such as Aroclor 1242, Aroclor 1254, and Aroclor 1260.
  • 12.7.2 Quantitative analysis
    • 12.7.2.1 Calculation: The computer program or software should be able to generate a list including peak numbers and the corresponding DCB homologue numbers and the PCB content calculated from the internal standard substance DCB, mg/L.
    • 12.7.2.2 PCB content of each peak: The mass of each homologue or co-eluting homologue is calculated by the internal standard method using the corrected relative response factors, see formula (8). For non-dechlorinated samples, use the corrected response factors obtained from the "all probabilities" RRF column in Appendix B; for dechlorinated samples, use the values obtained from "all possibilities".
      • m_i = (m_s × A_i × RRF_s) / (A_s × CRRF_i) ... (8)
      • Where: i—— represents the homologue or co-eluting homologue; m_i—— the mass corresponding to the test sample peak i, mg; m_s——12.1.1 the mass of the internal standard substance in the test sample (such as 2 μg or 0.0005 mg), mg; A_i—— the area/height of peak i; RRF_s—— the relative response factor of the internal standard substance (=1.00); A_s—— the area/height of the peak of the internal standard substance; CRRF_i—— the corrected relative response factor of peak i.
    • 12.7.2.3 The amounts of all individual peaks are added together to obtain the total mass of PCB in solution B. The total PCB content in the sample is calculated from the initial mass of the sample. These data can be input into the software to calculate the total PCB content of the sample according to formula (9), mg/kg.
      • w = (Σ m_i) / m_0 ... (9)
      • Where: m_0—— the initial mass of the sample, g.

9. Qualitative and Quantitative and Calculation Formulas

• Qualitative: 12.7.1 Compare the sample chromatogram with the reference substance chromatogram to identify commercially available materials such as Aroclor 1242, Aroclor 1254, and Aroclor 1260. 12.4 Qualitatively identify individual homologues or unresolved homologue groups by comparing ERRT with the data file in Appendix B.
• Quantitative: 12.7.2 Using the internal standard method, calculate the mass of each homologue or co-eluting homologue using the corrected relative response factors (CRRF), see formula (8); the amounts of all individual peaks are summed to obtain the total mass of PCB in solution B, and then the total PCB content of the sample (mg/kg) is calculated from the initial mass of the sample according to formula (9).
• Recovery calculation: Formula (3) R = PCB(D) / PCB(E) × 100.
• Sensitivity coefficient calculation: Formula (1) S_i = R_i / B_i.
• Correction coefficient calculation: Formula (6) K_i = ERRF_i / RRF_i.
• ERRT Calculation: Formula (4) ERRT = (t_i - t_30) / (t_209 - t_30).
• ERRF Calculation: Formula (5) ERRF_i = (A_i × M_s) / (A_s × M_i).

10. Precision

• 15 Precision: Judge the reliability of the test results according to the following provisions (95% confidence level).
• 15.1 Repeatability r: The difference between two replicate test results obtained by the same operator, using the same instrument, with the same method on the same test sample shall not exceed the value specified in formula (10).
  • r = 2 + 0.1x ... (10)
  • Where: x—— the average of the two replicate test results.
• 15.2 Reproducibility R: For two independent test results obtained by different operators, using different instruments, and the same method on the same test sample, the difference shall not exceed the value specified in formula (11).
  • In formula (11), the original text source-text extraction is missing, requires confirmation against the source standard (source-text extraction).
  • Where: x—— the average of two independent test results.

11. Report

• 13 Test report: report the total content of PCB in the test sample, expressed as mg/kg, and the data used shall retain “all probabilities” (original text source-text extraction noise, requires confirmation against the source standard (source-text extraction)).

12. Normative references

• GB 7597 Sampling method for power oils (transformer oil, turbine oil)
• IEC 60997 Determination of polychlorinated biphenyls in insulating oil—Packed column gas chromatography
• IEC 61619: 1997 Determination of polychlorinated biphenyls in insulating oil—Capillary gas chromatography (modified adoption)

13. Source and verification

This page is an AI translation of the available Chinese method card. Human verification is pending. It is not an official translation of the standard. Check the source standard and product certificate before use.

14. Quick-reference Q&A