ASTM D5580 for elements in finished gasoline by gas chromatography
ASTM D5580 Standard Test Method for Determination of Benzene, Toluene, Ethylbenzene, p/m-Xylene, o-Xylene, C9 and Heavier Aromatics, and Total Aromatics in Finished Gasoline by Gas Chromatography
8. Preparation of Columns
8.1 TCEP Column Packing:
8.1.1 Use any satisfactory method, that will produce a column capable of retaining aromatics from nonaromatic components of the same boiling point range in a gasoline sample. The following procedure has been used successfully.

8.1.2 Completely dissolve 10 g of TCEP in 100 mL of methylene chloride. Next add 40 g of 80/100 mesh Chromosorb P(AW) to the TCEP solution. Quickly transfer this mixture to a drying dish, in a fume hood, without scraping any of the residual packing from the sides of the container. Constantly, but gently, stir the packing until all of the solvent has evaporated. This column packing can be used immediately to prepare the TCEP column.

8.2 Micro-packed TCEP Column:
8.2.1 Wash a straight 560-mm (22-in.) length of 1.6-mm (1/16-in.) outside diameter, 0.76-mm (0.030-in.) inside diameter (stainless steel tubing with methanol and dry with compressed nitrogen.

8.2.2 Insert 6 to 12 strands of silvered wire, a small mesh screen or stainless steel frit inside one end of the tube. Slowly add 0.14 to 0.15 g of packing material to the column and gently vibrate to settle the packing inside the column. Insert silvered wire, mesh screen, or frit to the other end of the tube to prevent the packing material from falling. When strands of wire are used to retain the packing material inside the column, leave 6.0 mm (0.25 in.) of space at the top of the column.

8.3 WCOT Methyl Silicone Column - It is suggested that this column be purchased directly from a suitable capillary column manufacturer (see 6.3.2.1).

9. Sampling
9.1 Every effort should be made to ensure that the sample is representative of the fuel source from which it is taken. Follow the recommendations of Practice D4057, or its equivalent, when obtaining samples from bulk storage or pipelines.

9.2 Appropriate steps should be taken to minimize the loss of light hydrocarbons from the gasoline sample to be analyzed. Upon receipt in the laboratory, chill the sample in its original container from 0 to 5°C (32 to 40°F) before and after sub-sampling is performed.

9.3 If necessary, transfer the chilled sample to a vaportight container and store at 0 to 5°C (32 to 40°F) until needed for analysis.

10. Preparation of Apparatus and Establishment of Conditions
10.1 Assembly - Connect the TCEP and WCOT column to the valve system (Fig. 1) using low-volume connectors and narrow bore tubing. It is important to minimize the volume of the chromatographic system that comes in contact with the sample, otherwise peak broadening will occur.

10.2 Initial Operating Conditions - Adjust the operating conditions to those listed in Table 1, but do not turn on the detector circuits. Check the system for leaks before proceeding further.

10.2.1 If different polar and nonpolar columns are used, or WCOT capillary columns of smaller inner diameter or different film thickness, or both, are used, it may be necessary to use different optimum flows and temperatures.

10.2.2 Conditions listed in Table 1 are applicable to the columns described in 6.3. If a WCOT column of a different film thickness is used, the conditions chosen for the analysis must sufficiently separate toluene from the internal standard (first analysis) and ethylbenzene from the xylenes (second analysis).

10.3 Flow Rate (Carrier Gas) Adjustments:
10.3.1 Attach a flow measuring device to the precolumn vent (or Detector B) with the valve in the RESET or forward flow position and adjust the pressure of the capillary injection port (Fig. 1) to give 10.0-mL/min flow (17 to 20 psi). Soap bubble flow meters are suitable. This represents the flow through the polar precolumn.

10.3.2 Attach a flow measuring device to the split injector vent and adjust the flow from the split vent using the flow controller to provide a flow of 100 mL/min. Recheck the column vent flow set in 10.3.1 and adjust, if necessary. The split ratio should be approximately 11:1.

10.3.3 Switch the valve to BACKFLUSH position and adjust the variable restrictor to give the same precolumn vent flow set in 10.3.1. This is necessary to minimize flow changes when the valve is switched.

10.3.4 Switch the valve to the RESET position and adjust the auxiliary flow controller to give a flow of 10 mL/min at the Detector A (FID) exit.

10.4 Detector Setup - Depending on the particular type of instrumentation used, adjust the hydrogen, air, and makeup flows to the flame ionization detector and ignite the flame. If a thermal conductivity detector (Detector B) is being used to monitor the vent effluent in the valve RESET position, set the reference flow and turn on the detector circuit.

10.5 Valve Backflush and Reset Times:
10.5.1 The time to BACKFLUSH and RESET the valve will vary slightly for each column system and must be determined as described in 10.5.1.1, 10.5.1.2, and 10.5.1.3. The start time of the integrator or computer system and valve timer must be synchronized with the injection to accurately reproduce the backflush time. This procedure assumes that a thermal conductivity detector is installed on the precolumn vent line as Detector B (see 6.1.4.1). If a detector is not available, the appropriate valve BACKFLUSH times, T1 and T2, must be determined experimentally. If the BACKFLUSH times, T1 and T2, are not set correctly (switched too late), it is possible that part of the benzene and ethylbenzene peaks will be vented.

10.5.1.1 Adjust the valve to RESET (forward flow) and inject 1.0 µL of a blend containing approximately 5 % each of benzene, ethylbenzene, o-xylene, and 2-hexanone in isooctane. This mixture is used to set the valve timing, therefore, the exact concentration need not be known. Alternatively, the calibration mixture can be used for this test. Determine retention time in seconds at which benzene and ethylbenzene start to elute as measured by Detector B. Subtract 6 s from each of these and call these times to BACKFLUSH, T1 and T2, respectively. The correct time for T1 and T2 is just prior to the elution of benzene and ethylbenzene from the TCEP precolumn.

NOTE 1 - Fig. 2 is an example chromatogram illustrating the elution of a calibration mixture from the polar precolumn using the procedure described in 10.5.1.1. Times to BACKFLUSH, T1 and T2, are indicated on the chromatogram. The times to BACKFLUSH, T1 and T2, should be optimized for each chromatographic system.

10.5.1.2 Reinject the calibration blend and turn the valve to BACKFLUSH at time T1. When the internal standard peak (2-hexanone) returns to baseline switch valve back to RESET (forward flow) position. Call this time T3.

10.5.1.3 Reinject the calibration blend and BACKFLUSH at time T2. When the o-xylene peak returns to baseline, switch the valve back to RESET (forward flow). Call this time T4.

10.6 Polar Precolumn Selectivity Check:
10.6.1 The selectivity of the polar precolumn is critical to allow for accurate determination of the C9 and heavier aromatics without non-aromatic interferences. The selectivity must be verified so that for the second analysis, when the time to BACKFLUSH T2 is properly adjusted, all of the C12 and lighter nonaromatic hydrocarbons are vented from the polar precolumn while the heavier aromatics are retained. The following test can be used to verify the precolumn performance.

10.6.1.1 Prepare a blend containing approximately 1.5 % n-dodecane in 2,2,4-trimethylpentane (isooctane). n-Dodecane is used to represent the high boiling nonaromatic hydrocarbons in gasoline. Inject 1.0 µL of the mixture under the conditions specified in 10.2 to 10.5 and actuate the valve at time T2 (BACKFLUSH) and time T4 (RESET). Record the signals from both the flame ionization (Detector A) and thermal conductivity (Detector B) detectors. Verify that n-dodecane fully elutes from the polar precolumn before BACKFLUSH time T2. When monitoring the thermal conductivity detector (Detector B), the n-dodecane peak should return to baseline before BACKFLUSH time T2. If not, part of the n-dodecane peak will be backflushed to the non-polar WCOT column and be detected by the flame ionization detector after the valve RESET time T4. If a thermal conductivity detector is not available on the precolumn vent line, the chromatogram obtained by the flame ionization detector can be used to verify that all the n-dodecane is being vented. This chromatogram should not show any significant response from n-dodecane after the RESET time T4.

10.6.1.2 If all of the n-dodecane peak is not completely vented from the polar precolumn, as measured by the thermal conductivity or flame ionization detector, recheck instrument parameters and valve backflush times (10.5) or replace the polar precolumn. If the valve is contained in a separate isothermal heated zone, it may be necessary to use a higher temperature to prevent absorption of small amounts of n-dodecane on the rotor or transfer tubing surfaces.