Cleaning procedure, capillary column

The gas chromatographic determination of isomers of dinitrotoluene in seawater has been described by Flashimoto and co-workers [296,297]. These authors describe the complete separation of six dinitrotoluene isomers using gas chromatography with support-coated open tubular glass capillary columns and electron-capture detection. The method was applied to the qualitative and quantitative analyses of trace levels of isomers in seawater and the results were found to be satisfactory, with no need for further clean-up procedures. 

The use of capillary columns rather than packed column GC has improved resolution and sensitivity and shortened the analysis time (Washall and Wampler 1988). However, more stringent sample clean-up procedures are required for capillary column GC (Oliver and Nicol 1982b). The development of methods using whole column cryotrapping (Pankow and Rosen 1988 Pankow et al. 1988) and cryogenic refocusing (Washall and Wampler 1988) provide even greater sensitivity and resolution for GC analysis. 

Each laboratory was given a free choice of extraction, clean-up and conditions for the final determination. The latter, gas chromatography with high-resolution mass spectrometry, included choice of injection, capillary columns, ion masses monitored etc. [34]. In addition to the CRM, the participants received a calibrant solution containing the PCDDs/PCDFs of interest for this project as a control for their calibration procedure. 

Sub-samples of 1 g were taken for the analysis. The PAHs were extracted under reflux with toluene/acetone (1 1 v/v, 150 mL) for 30 minutes. The efficiency of the extraction procedure was tested by adding an internal standard (indeno[l,2,3-c t/]fluoranthene) and was found to be 95 99%. An additional extraction with tetrahydrofurane did not enhance the PAH yield. The extract was evaporated to about 2 mL. A cydohexane/dimethylformamide (DMF)/water (100 mL cyclohexane, 90 mL DMF and 10 mL water) partition was performed. The cyclohexane phase was discarded and the DMF/water phase diluted with 80 mL water. This phase was re-extracted with cyclohexane. The cyclohexane phase was evaporated to a volume of 2 mL and cleaned-up over a silica column (5 g, 9.1% water). The PAHs were eluted with 80 mL cyclohexane, which was evaporated afterwards to a volume of 0.1-0.5 mL. The obtained concentrate was placed on a Sephadex LH 20 column (10 g) and eluted with propanol-2. The fraction of 0-46 mL was discarded. The fraction from 46-170 mL was evaporated to a small volume and analysed by high resolution gas chromatography (on column injection) using a fused silica capillary column of which the conditions are described elsewhere [26]. 

Each laboratory used its own procedures for the sample preparation, clean-up, method of injection, choice of carrier gas, capillary columns and other chromatographic conditions. Each participant received two calibrant solutions prepared independently and gravimetrically with PCDD and PCDF of verified purity and identify [20] in addition, the participants received six - C labelled isomers in solution to be used as internal standards for determining the response factors and the recovery efficiency of the methods, as well as for calibration purposes [19],... 

A detailed description of the methods used is given in the certification report [33], including the internal standards used, GC conditions, calibration techniques etc. Clean-up procedures were based on well proven techniques to remove the bulk of the mineral oil and other organic compounds which would interfere with the chromatographic determination, e.g. decomposition with concentrated H2SO4, adsorption chromatography, separation on silica gel, or separation with Florisil or potassium silicate. For the separation, each CB was identified and confirmed using at least two capillary columns coated with different stationary phases to compare the relative retention times or... 

The 5 to 6 L of DCM recovered from the extractor was evaporated to 25 mL in a mantel-heated flask equipped with a 12-ball Snyder condenser. Final concentration to 1 mL was made in a water-bath-heated smaller flask with a Kuderna-Danish condenser. The suspended sediments were soxhiet extracted overnight with an acetone/hexane mixture. After back extraction with water to remove the acetone, the extract was concentrated to 1 mL by the above procedure. The extract was cleaned up on a minicolumn packed with 2 cm Na2S04/4 cm 40% H2SO4 in silica gel/2 cm florisil. Final gas chromatographic analysis was performed using dual capillary columns (SE54 and OV17) with electron capture detectors. Further method details and recovery efficiencies of the procedures have been previously reported (Oliver and Nicol 1982). 

Korfmacher et al. [71] employed a short clean-up procedure followed by electron capture gas chromatography for the determination of octachlorodibenzo-p-dioxin in soils using a furzed silica capillary gas chromatographic column. The technique was suitable as a routine screening procedure for samples taken from contaminated sites. 

Sediment and soil samples are homogenized and extracted. Clean-up procedures are required prior to analysis by GC/ECD or GC/MS techniques (Lopez-Avila et al. 1992 Moseman et al. 1977 Saleh and Lee 1978 Tiernan et al. 1990). For sediment, soil, and sludge, recoveries were good (>85%) with sensitivity in the low ppb range (Moseman et al. 1977 Saleh and Lee 1978). Precision is good ( <6% RSD) (Saleh and Lee 1978). Analytical difficulties (unacceptable recovery not detectable using second capillary GC column) were reported (Lopez-Avila et al. 1992 Tiernan et al. 1990). 

Recently. Liao et al. (1988) developed a relatively simple and rapid procedure for purifying human and bovine adipose tissue extracts so that trace levels of complex mixture of target analytes (including PAHs) could be detected and quantified by capillary GC/MS. By employing an activated Florisil column. Liao and co-workers showed that lipid contaminants bind effectively (more than 99.75%) with Florisil. thereby producing a relatively clean sample extract. A detection limit at a low ng/g level and an average sample recovery of 85% were achieved (Gay et al. 1980 Liao et al. 1988 Modica et al. 1982). 

The method of analysis for the final determination of these OCPs was based on gas chromatography with electron capture detection. Most of the participating laboratories used capillary GC. Mass spectrometry was used as a complementary technique but was not used to obtain quantitative data for the certification exercise. Each laboratory used its own procedures for sample preparation, extraction, clean-up, method of injection, choice of carrier gas, columns and other chromatographic conditions. Critical parameters were optimised by the analyst prior to undertaking these determinations [26]. 

Samples of fresh fig or rehydrated dry fig (100 g) were enclosed in a wide mouth Erlenmyer flask (250 mL) and the flask covered with aluminum foil. Samples of vapor (10 mL) from above the figs was drawn up into a clean dry glass syringe and Injected directly into the capillary GLC column (cf. 3). This procedure was used for both the capillary GLC-mass spectrometry analyses and the quantitative analyses. Quantitative analyses were made by comparison of peak areas with those from vapor samples above known concentrations of the compounds in water solution. 

A tuft of quartz wool is placed in a glass beaker and is leached for several hours in warm 10 % HCl. One layer of leached and rinsed quartz wool, approximately 5 mm, is placed in the undermost position of the augmented part of the capillary (Fig. 12-26). Above this layer, 80 % of the remaining space is filled with dry silica gel (approximately 1.5 mL of Chromo-sorb W/AW-DMCS). The column is closed with another 5 mm plug of quartz wool. The quartz column is coimected to a vacuum pump. The column is rinsed slowly with approximately 300 mL of 3 % HO and then with ultra-pure water until the obtained water passed is of neutral ph. The column is cleaned by performing repeated separation procedures as described in the following section until the blank values are constant. 

---