This method covers the following groups of components:
Chlororganic pesticides:
Polychlorinated biphenyls,
These components may be determined in air samples, described in chapter, 3.13. Filter and PUF plugs are extracted separately with a hexane/diethylether 9:1 mixture in a soxhlet extractor. The extracts are concentrated and then cleaned by using adsorption chromatography (silica). After the concentration to the appropriate volume and addition of the recovery standard, the components are separated and quantified by using gas chromatography combined with mass spectrometry (MS).
Vial, 0.9 ml, Chromacol 0.9-CTV or similar with septum cap (teflon-coated silicon seal), Brown cat. no.151261 or similar quality
Vial, 1.5 ml Brown cat. no.150900 with screw top and teflon seal, Brown cat. no.150931 or similar quality
Vial, 2 ml with a capillary opening (diameter 1.5 mm and length 20 mm) and screw cap (teflon seal Schott GL14), (home made)
Vial, 8 ml with screw cap (teflon seal), Supelco cat. No. 2-3295
Vial with 100 ml insert, Chromacol 0.3- FIV
samples, both untreated and extracts after chemical treatment
standards
HP 5890-II gas chromatograph with split/splitless injector and HP 7673 autosampler
HP 5989 mass spectrometer (MS Engine)
HP G1034C MS Chemstation integration system
HP 5890-II gas chromatograph with split/splitless injector and HP 7673 autosampler
VG GC/MS interface with lock mass substance inlet system
VG AutoSpec three sector high resolution mass spectrometer with EI-ion source
EC 3100 data system with OPUS MS- software system
60 m and 30 m length* 0.25 mm
i.d., 0.10 mm
film thickness, 95% dimethyl 5% diphenyl polysiloxane, immobilized, e.g. Rtx-5
from Restek Corporation,
Bellefonte, USA
25 m length* 0.20 mm i.d. * 0.11 mm film thickness, 5% phenylmethylpolysiloxane HP-5 (Ultra), Hewlett Packard Company, Amsterdam, Holland
50 m length *0.22 mm i.d. *0.15 mm 8% phenylpolycarbonsiloxane e.g. HT-8 SGE, Australia
30 m length *0.25 mm i.d. *0.10 mm, 14% cyanopropylphenyl 86% methylpolysiloxane, immobilized, e.g. Rtx-1701 from Restek Corporation, Bellefonte, USA
30m length *0.25 mm i.d. *0.10 mm, 90% biscyanopropyl 10% phenylcyanopropyl, e.g. Rtx-2301 from Restek Corporation, Bellefonte, USA
All solvents, except diethyl ether can be used without cleaning.
Inorganic chemicals, adsorbents and various accessories
See chapter 3.16.3.1
See chapter 3.16.3.2
See chapter 3.16.3.3
See chapter 3.16.3.4
All solvents must be of "pesticide grade" or equivalent quality. The solvent must give chromatograms free from interfering peaks (S/N < 3) in the elution range from a-HCH to OCN. Performing a complete method blank test may also check this.
The diethyl ether must be cleaned because it contains an inhibitor to prevent formation of peroxides: 250 ml diethyl ether is filtered using a chromatography column (diam. 20 mm) packed with 20 cm basic aluminiumoxide. Diethyl ether without inhibitor may, with time, form peroxides, which represents an explosion hazard, especially by pre-concentrating samples using a rotary evaporator. Therefore not more than an amount sufficient for one months consumption is cleaned. The solvent is stored in the dark at a temperature <5°C.
Sodium sulphate is put in a porcelain dish and heated to 600°C for 8 hours in an oven. Let cool to room temperature in a desiccator. Store in a stoppered glass bottle. The bottle must be labelled with the expiry date of the sodium sulphate. Maximum storage time is one month. After the expiry date the sodium sulphate is discarded.
Ca. 400 g silica is put in a porcelain dish and heated at 600°C for at least 8 hours in an oven. After heating the silica is left to cool in a desiccator and shall not be used before it has reached room temperature. The silica is stored in a glass bottle with a glass stopper. The bottle must be labelled with the expiry date of the silica. Maximum storage time is one month
Soxhlet thimbles are cleaned by soxhlet extraction using n-hexane for 8 hours. The thimbles are dried in a fume hood at room temperature overnight and wrapped individually in aluminium foil.
Cotton wool is soxhlet extracted with 500 ml 2500 ml (depending of amount of cotton) n-hexane or dichloromethane for 24 hours. The cotton is dried in a vacuum desiccator at 60°C.
Helium GC carrier gas cleaning:
Nitrogen used for final sample blow-down/pre-concentration is cleaned using a metal cartridge filled with active carbon. Exchange adsorbent when exchanging the gas bottle.
Chrompack filters are discarded after use.
Re-activation of molecular sieves: Fill molecular sieves in a metal cartridge and activate at 300°C (3 h) in an oven, flushing the cartridge with 20 ml/min. pre-filtered helium.
Active carbon is discarded after use.
Sample pre-treatment, e.g. weighing of filters and extraction, is discussed in chapter 3.17.5
Small amounts of interfering substances may be removed from the sample with one single method. Depending on the components to be measured different cleaning procedures can be selected; treatment of the extract with sulphuric acid for acid-stable components, or with alkaline hydrolysis for acid-labile substances. Both treatments are followed by adsorption chromatography. When both acid-stable and -labile substances are measured and their relative amounts unknown, the concentrate should be divided into two equal parts before further treatment. After cleaning, the sample should be concentrated once more followed by addition of a recovery standard (TCN or OCN) in order to determine the amount of internal standard before the sample is ready for analysis by GC/MS.
The concentrated sample (0.51 ml volume) is transferred to a 1015 ml centrifuge tube, 8-10 ml conc. sulphuric acid added, and placed in a rack until the next day. The hexane fraction is transferred to a new centrifuge tube and 1 ml MilliQ water added drop by drop. The water phase is removed and the hexane phase dried with ½ teaspoon Na2SO4.
To 1 ml concentrated sample is added a solution of 0.2 g KOH in 1 ml ethanol and 0.1 ml MilliQ water. The mix is heated on a water bath at 50°C for 30 min. Add 5 ml MilliQ water, shake the sample, separate by a centrifuge, concentrate the organic phase by evaporation, and add a recovery standard before the GS/MS analysis. If the hydrolysed sample is strongly coloured after evaporation, the sample may be cleaned using silica-chromatography (next Section).
Cotton is put at the bottom end of a glass column (20 cm x 1.5 cm) and the column is filled with 4 g silica activated at 600°C. A layer of 1g Na2SO4 is added on top. Use a vibrator when filling. The column is washed with 30 ml 10 % diethyl ether in hexane. The column should never be allowed to run dry! The sample (0.51 ml volume) is transferred to the column after washing, and the sample container rinsed with additionally 23 ml diethyl ether/hexane mixture which is transferred to the column. Elute the column with 30 ml of diethyl ether/hexane mixture and the sample is collected in a TurboVap glass with 20 ml nonane as a keeper. The sample volume is carefully reduced with the TurboVap until 0.5 ml, and transferred to a sample glass with conical insert. The TurboVap vessel is rinsed three times with 0.15 ml hexane, and the liquid volumes added to the sample. The sample volume is reduced to the desired volume (0.10.5 ml) in a slow nitrogen stream. Recovery standard is added and the sample vessel sealed with a screw cap or a crimp cap with auto-injector septum. This sample is now ready for a GS/MS analysis. If the sample is not analysed at once (same day) it must be kept dark in a freezer at -20°C. If the sample is stored for more than 1 month, this should be noted in the data report.
Flush a pasteur pipette with a piece of cotton wool at the bottom with hexane. Pass the sample through the pipette and collect the sample. Wash with a small amount hexane before volume reduction to 0.10.5 ml.
A standard mixture containing known concentrations of components of interest is used for identification/quantification. Standard components should preferably be provided as crystalline solids with purity better than 99 per cent. If they can be provided in solutions only, the solutions must be certified or calibrated against certified standards from an international standardisation bureau as NIST or BCR.
Concentrated standards containing only one, or a small number of components are prepared and checked by GC/MS in full scan mode before further use. If impurities are discovered, their concentrations should correspond to less than 3% of the main components area. The other standards; calibration standards, internal standards, and recovery standards, are all prepared as different diluted mixtures based on the concentrated standards.
Weighing the proper amounts of crystalline substance of the standards should be performed with great care. Use disposable gloves and a mask. Weighing ships and spatulas should be rinsed in toluene and hexane before use and (air) dried. The tare of the ship is set to zero and the standard component transferred to the ship by a thin spatula in an amount as close to the estimated one as possible. The spatula must be rinsed and dried between each weighing in order to avoid contamination. When all components have been weighed, the content of the ship is transferred to a volumetric flask with n-hexane or iso-octane. The flask is filled with n-hexane or iso-octane to the correct volume and placed in an ultrasonic bath until all solids are dissolved. The concentrated standard is transferred to a flask equipped with screw cap and a teflon seal.
The weighed amounts and standard concentrations, a standard number, and the weight of the standard flask must be recorded in the standards logbook. The concentrated standards should be kept in a refrigerator at 4°C. When preparing the more diluted solutions, the concentrated standards should be removed from the fridge two hours before use. The concentrated standard should be sonicated for 5 min. in order to dissolve any solid substance. This is of particular importance for heavily soluble components e.g. b-HCH. In order to maintain a high accuracy in the final concentration, the dilution should be less than 1:100 in all steps, i.e. at most 100 ml in a 10 ml graduated flask. When more diluted solutions are needed, secondary standards should be prepared. The flask containing the concentrated standard should be weighed before and after removing a volume for dilution, and the weights recorded in the logbook.
Diluted solutions are prepared by using volumetric flasks and pipettes or syringes. Disposable pipettes are preferred to syringes used for different standards because of a possibility for contamination. The entire dilution process must be checked by weighing.
The following components may be included in a set of standards:
Pesticides |
Abbreviation |
Polychlorinated biphenyls |
IUPAC no. |
Hexachlorobenzene |
HCB |
2,2',5-TriCB |
18 |
a-Hexachlorocyclohexane |
a-HCH |
2,4,4'-TriCB |
28 |
b-Hexachlorocyclohexane |
b-HCH |
2,4',5-TriCB |
31 |
|
|
2,3,4-TriCB |
33 |
|
|
3,4,4-TriCB |
37 |
g- Hexachlorocyclohexane |
g-HCH |
2,2',4,4'-TetCB |
47 |
Trifluralin |
Trifl |
2,2',5,5'-TetCB |
52 |
Chlordene |
CDen |
2,3,4,4'-TetCB |
60 |
Heptachlor |
HepC |
2,3',4,4'-TetCB |
66 |
Oxy-Chlordane |
oxy-CD |
2,4,4',5-TetCB |
74 |
Cis-Heptachlorepoxide |
cis-Hepex |
2,2',4,4',5-PenCB |
99 |
Trans-Chlordane |
tr-CD |
2,2',4,5,5'-PenCB |
101 |
Cis- Chlordane |
cis-CD |
2,3,3',4,4'-PenCB |
105 |
trans-Nonachlor |
tr-NO |
2,3,4,4',5-PenCB |
114 |
cis-Nonachlor |
cis-NO |
2,3',4,4',5-PenCB |
118 |
Pesticides |
Abbreviation |
Polychlorinated biphenyls |
IUPAC no. |
|
|
2,3,3,4,5-PenCB |
122 |
a-Endosulfan |
a-Endo |
2',3,4,4',5-PenCB |
123 |
Dieldrin |
Diel |
2,2',3,3',4,4'-HexCB |
128 |
Aldrin |
Ald |
2,2',3,4,4',5'-HexCB |
138 |
|
|
2,2,3,4,5,5-HexCBC |
141 |
Endrin |
End |
2,2',3,4',5',6-HexCB |
149 |
o,p-Dichlorodiphenyldichloroethane |
op-DDD |
2,2',4,4',5,5'-HexCB |
153 |
p,p'-Dichlorodiphenyldichloroethane |
pp-DDD |
2,3,3',4,4',5-HexCB |
156 |
o,p'-Dichlorodiphenyldichloroethylene |
op-DDE |
2,3,3',4,4',5'-HexCB |
157 |
p,p'- Dichlorodiphenyldichloroethylene |
pp-DDE |
2,3',4,4',5,5'-HexCB |
167 |
|
|
|
|
Pesticides |
Abbreviation |
Polychlorinated biphenyls |
IUPAC no. |
o,p'-Diklorodifenyltrikloroehtane |
op-DDT |
2,2',3,3',4,4',5-HepCB |
170 |
p,p'-Diklorodifenyltrikloroethane |
pp-DDT |
2,2',3,4,4',5,5'-HepCB |
180 |
|
|
2,2,3,4,4,5,6-HepCB |
183 |
|
|
2,2',3,4',5,5',6-HepCB |
187 |
|
|
2,3,3',4,4',5,5'-HepCB |
189 |
|
|
2,2,3,3,4,4,5,5-OctCB |
194 |
|
|
2,2,3,3,4,4,5,5,6-NonCB |
206 |
|
|
2,2,3,3,4,4,5,5,6,6-DecaCB |
209 |
|
|
|
|
Internal standard pesticides |
Abbreviation |
Internal standards PCB |
IUPAC no. |
13C-p,p- Dichlorodiphenyldichloroethylene |
13C-p,p-DDE |
13C-2,4,4'-Trichlorobiphenyl |
13C-PCB-28 |
13C-g-Heksachlorosyklohexane |
13C-2D-g-HCH |
13C-2,2',5,5'-Tetrachlorobiphenyl |
13C-PCB-52 |
13C-a-Heksachlorosyklohexane |
13C-a-HCH |
13C-2,2',4,5,5'-Pentachlorobiphenyl |
13C-PCB-101 |
|
|
13C-2,3',4,4',5-Pentachlorobiphenyl |
13C-PCB-118 |
|
|
13C-2,2',4,4',5,5'-Hexachlorobiphenyl |
13C-PCB-153 |
13C4-Aldrin |
|
|
|
13C4-Dieldrin |
|
|
|
13C4-Heptachlor |
|
|
|
13C-Hexachlorobenzene |
13C-HCB |
13C-2,2',3,4,4',5,5'-Heptachlorobiphenyl |
13C-PCB-180 |
Recovery standards |
|
|
|
1,2,3,4-Tetrachloronaphtalene |
TCN |
|
|
Octachloronaphtalene |
OCN |
|
|
A standard for GC/MS should have concentrations similar to the expected concentrations of the components to be measured
The internal standard may be a solution with pesticides and/or PCBs which contains labelled isotopes.
A solution containing tetrachloronaphtalene is used for this purpose. The recovery standard is added to the sample as the last step before quantification.
Amounts of standards added before extraction (ISTD) and after the sample preparation (RSTD) should be similar to the expected concentrations in the sample.
The purity of the standards is checked in the GC/MS full scan mode before acceptance. If impurities are discovered in the concentrated standard, the amounts as expressed by its area in the chromatogram must be less than 3 % of the main components area. When concentrated standards are stored dark in a refrigerator, their stability will be very good. In order to document the stability, the full scan mode check is repeated at intervals not longer than 3 years. Normally this will be carried out when preparing a working standard after 2 years. The working standards are stored dark in a refrigerator, but new standards should be prepared every two years at most. Working standards kept in sample flasks with capillary tubes are not checked for weight loss (1 mg in 6 months when closed). The stability of working standards stored as described above is considered to be 2 years.
A newly prepared series of standards must always be compared with the previously used ones before use. Only differences considered to be less than the reproducibility of the analytical method are accepted. The working standards should be compared with certified reference material; NIST SRM 1492 Chlorinated pesticides in hexane and BCR CRM 365 Polychlorinated biphenyls in iso-octane, at least once every year. Standards from laboratory comparisons may also be used. The standards should be stored in a refrigerator.
The cleaned samples are analysed by gas chromatography/mass spectrometry (GC/MS). Standard mixtures are used for identification and quantification.
The individual components are identified by their GC retention and their mass fragments.
The quantification of the components is made by using internal standard. A calibration is performed with a standard mixture containing known concentrations of the components to be measured and one or more components not contained in the sample (internal standards). The calibration is followed by injection of the sample containing known amounts of internal standards. Quantification is relative to the internal standard. In this way, the sample extract volume will not be included in the calculations, and it is not necessary to accurately determine the final sample volume after evaporation or the injection volume.
The GC-parameters given are approximate and must be fine-tuned for each column, since equal columns may separate the actual compounds slightly differently.
Capillary column:
Rtx-5, 60 m x 0.25 mm x 0.10 mm:
Carrier gas: He, 185 kPa (1.85 bar, 27
psi)
GC-temperature program:
1µl injected splitless (autoinjector or
"hot needle" injection) at 60°C, 2 min. at 60°C,
60190°C
with 20°/min., 190-230°C with 3°/min., 230280°C and 280°C for
15 min. isothermally.
Capillary column: Rtx-5 or
equivalent, 30 m x 0.25 mm x 0.10 mm:
Carrier gas: He, 75 kPa (0.75 bar, 11.5
psi)
GC-temperature program:
1 ml injected splitless
(autoinjector or "hot needle" injection) at 60°C, 2 min. at 60°C,
60150°C
with 20 °/min.,
150280°C
with 1 °/min.
and 280°C
for 10 min. isothermally.
Capillary column: Rtx-2330 or
equivalent, 30 m x 0.25 mm x 0.10 mm:
Carrier gas: He, 83 kPa (0.83 bar, 12
psi)
GC-temperature program:
1 ml injected splitless
(autoinjector or "hot needle" injection) at 60 or 100°C
(depending on solvent), 2 min. at 60 or 100°C (depending on solvent),
to 170°C
with 20 °/min.,
170230°C
with 3 °/min.,
230270°C
and 270°C
for 6.5 min. isothermally.
Capillary column:
HP Ultra-2, 25 m x 0.20 mm x 0.11 mm:
Carrier gas: He, 110 kPa (1.1 bar, 15
psi)
GC-temperature program:
1 ml injected splitless
(autoinjector or "hot needle" injection) at 60°C, 2 min. at 60°C,
60150°C
with 20 °/min., 150230°C with 4 °/min. and 230280 with 25 C/min and 275°C for
5 min. isothermally.
In addition the following parameters are used:
Split gas flow: 40 ± 10 ml/min
Septum purge flow: 3 ml/min
Injector temperature: 260°C
GC/MS-interface temperature: 260°C280°C
To save carrier gas, the split gas flow is reduced to <5 ml/min when the instrument is not used.
Autoinjector conditions
(approximate):
Solvent A: toluene
Solvent B: n-hexane
Sample wash: 0
Sample pumps: 5
Sample volume: 1 µl
Solvent A washes: 8
Solvent B washes: 8
Solvents A and B for syringe cleaning must be exchanged each day. The solvent vials are cleaned when necessary.
The injector septum is exchanged after ca. 50 injections or once a week. The cleanness of the glass liner is checked after ca. 100 injections or if the GC-separation is poor.
For quantification GC/MS with either EI or NCI ionisation is used. To check the stability of the GC/MS-system, a calibration standard is injected before and after each sample batch.
Operation of the GC/MS-system is described in the instrument manuals.
Calibration- and detection conditions for
EI (VG-AUTOSPEC GC/MS)
Gas chromatograph
GC/MS-interface: 260°C
Ion source:
Electron impact (EI) ion source
Ion source temperature: 260-300°C
Max. acceleration voltage: 8000 V
Electron energy: 30 eV-40 eV
Lock substance: Perfluorokerosene
(PFK)
By mounting the capillary column, 1-2 mm of the exit of the column (on the MS-side) should extend into the ion source.
Using mass fragment m/z = 330,97 from perfluorokerosene (PFK, boiling point range 70°240°C) the instrument is optimised manually for ion gain and mass resolution. At resolution 10,000 (defined as m/Dm = 10,000 at 5% valley) the signal/noise ratio for 500 fg of g-HCH should be S/N ³3.
The mass scale for each SIM-function (single ion monitoring) is calibrated automatically if possible. Optimisation of ion source and mass resolution and calibration of mass scale is controlled for each single PFK-mass in each SIM-function.
To reduce the risk of false identification further, two masses in each fragment cluster are detected (see table SIM-program for pesticides).
The SIM-program described is sufficient for a semi-quantitative analysis. If a higher accuracy is desired, a 13C-labelled internal standard must be added to each SIM-group in order to compensate for differences in sensitivity between the different SIM-functions.
Since the mass spectrometer has a large linear range, injection of one calibration standard before a series of samples is sufficient.
Calibration
The response factor, Rfi, for each compound, i, relative to the internal standard (ISTD) is determined from an analysis of
a calibration standard with known concentrations:
Rfi : |
Response factor of compound i |
AmountISTD: |
Amount of internal standard injected |
Amounti : |
Amount of compound i injected |
Areai : |
Peak area of compound i |
AreaISTD: |
Peak area of internal standard |
Quantification
Using the response factors, Rfi, determined during the
calibration, a known amount of internal standard and the peak areas detected
during the quantitative analysis, the amount of each compound i is calculated.
Amounti: |
Amount of compound i in the sample |
AmountISTD: |
Amount of internal standard added to the sample |
Areai : |
Area of compound i |
Rfi : |
Response factor of compound i |
AreaISTD: |
Area of internal standard |
Recovery of internal standard (added before sample clean-up) is computed relative to amount of recovery standard (RSTD) added before the quantification. Relative response factors based on the recovery standard (RRFg) is calculated for each ISTD-compound from the quantification standard analysis.
Amt.ISTD : Amount internal standard added before extraction
Amt.RSTD : Amount of recovery standard added before quantification
AreaISTD : Peak area of internal standard
AreaRSTD : Peak area of recovery standard
SIM-program for PCB-compounds
SIM-function |
Isomer group |
12C-Mass 1 |
12C-Mass 2 |
13C-Mass 1 |
13C-Mass 2 |
1 |
HCB |
283,8102 |
285,8072 |
293,8244 |
295,8214 |
2 |
TCN |
263,9067 |
265,9038 |
|
|
3 |
TeCB |
289,9224 |
291,9194 |
|
|
4 |
PeCB |
325,8804 |
327,8775 |
337,9207 |
339,9177 |
5 |
HxCB |
359,8415 |
361,8385 |
|
|
SIM-program for DDT-compounds
SIM-function |
Isomer group |
12C-Mass 1 |
12C-Mass 2 |
13C-Mass 1 |
13C-Mass 2 |
1 |
TCN |
263,907 |
265,904 |
|
|
All mass fragmentograms and area lists are printed after the analysis. Mass fragmentograms must be evaluated on the following properties:
Calibration- detection conditions for NCI (HP 5989 GC/MS)
The instrument parameters are optimised using perfluortributylamine (PFTBA) either with automatic or manual tuning. To reduce the risk of false identification further, two masses (M and M+2) in each fragment cluster are detected (see table for SIM-program).
Since the mass spectrometer has a large linear range, injection of one calibration standard before a series of samples is sufficient. The analysis is performed using the same procedures described for EI GC/MS.
All mass fragmentograms and area lists are printed after the analysis.
SIM-program for POP (for guidance)
SIM-function |
Isomer group |
Mass 1 |
Mass 2 |
1 |
HCH |
252.9 |
254.9 |
2 |
TCN |
263.9 |
265.9 |
3 |
o,p-DDE |
245.9 |
247.9 |
4 |
o,p-DDD |
245.9 |
247.9 |
5 |
p,p-DDT |
280.9 |
282.9 |
6 |
PCB
156 |
359.8 |
361.8 |
The GC/MS-instrument should be calibrated every day. The sensitivity of the mass spectrometer can, for instance, be controlled daily by determining the signal-to-noise ratio for a given amount of a chosen component (one such component could be PCB-101).
Every new working standard should be compared to the existing standard before it is taken into use. Deviations within the reproducibility of the procedure are acceptable. At least once a year, the working standards should be controlled against a reference standard from an intercomparison or which has been certified from an international reference laboratory.
The accuracy should be within the uncertainty of the procedure (± 20 %). Measures to assure constant standard concentration is described in chapter 4.19.8.2 under Quality assurance of standards.
The quantification standard should be injected at the beginning of the GC-run of every series of samples. A maximum of 10 samples should be analysed before a new injection of the quantification standard is carried out. If the sample series consists of less than 10 samples, the quantification standard should be injected after the last sample. A control standard should also be injected with every sample series.
At the moment there is no certified reference material available that can be utilised for determination of organic compounds in air samples. It is therefore necessary for the laboratory to establish a control sample. This sample should be large enough to correspond to about 40 real air samples. The sample is extracted in the usual way. The extract is homogenised and split into 40 separate samples that are stored in suitable flasks at -20°C. Each year, at least 4 of these control samples should be analysed. The results for at least one component from each component group (for instance g-HCH, tr-CD and PCB-153) should be plotted on a quality control chart (QCC, Vogelsang, 1991). This quality control chart gives a good overview of the long time stability of the measurement results.
An internal standard (ISTD) should be added at the beginning of the procedure and a recovery standard (e.g. octachloronaphtalene) should be added just before the quantification step. In addition to this, recovery tests with spiked (including C12-components) samples or solvents should be carried out for every 100 sample of a certain type. In this work, analysis of control samples and blank samples are also important parts. The recovery of the internal standard should be between 40% and 120%, while the spiked C12-components should have a recovery that corresponds to the uncertainty of the procedure (for instance ± 20 %) relative to the theoretically added amount of each component. For volatile components, for instance HCB or HCH, which are prone to losses during the volume reducing steps, the lower recovery limit for the corresponding C13-spiked components are 20%.
The aim of quality assurance (QA) is to ascertain that the established results have the necessary accuracy and traceability.
The methods used for determination of organic compounds in environmental samples at very low concentrations may include a number of possibilities for errors:
To eliminate as many as possible of these error sources, the following demands should be fulfilled:
Operating procedures
The operating procedures are the fundament for every quality
assurance measure taken. The procedures assure continuity and dependability in
every working step carried out in the laboratory. They are also an important
part of the education and training of new operators and of the continuous
training of all personnel involved. The operating procedures should include
descriptions and specifications of the following:
As soon as the laboratory receives a sample, the sample should be registered in the sample journal or sample logbook. The journal should include information about sample type, sampling site, sampling date, sample amount, and, if necessary, place of storage. Every sample should be given a unique sample number, for instance year/serial number (99/102).
When the sample is registered in the sample journal a sample handling form for the sample should be established. In addition to information about sample type, sample number and so on, details of the important steps in the sample handling should be written on the form, especially deviations from the procedures. Amounts of added standards and name and location of electronic data files should also be specified. All notes in the form should be signed (initials) and dated.
Every analysis instrument should have an instrument logbook. In this journal every sample run should be registered together with method used, temperature program or other vital instrument parameters. Instrument deviations, for instance poor separation or tailing, should be registered. In addition, instrument sensitivity and simple maintenance of the instrument, e.g. change of septum or cleaning of the glass liner, should be registered.
Every standard should be given a unique identification. The standards should be registered in the standard journal. Concentration, solvent, date of preparation and weight of the container should be specified. A container should be weighed at room temperature before and after removal of an amount of standard. Weight and date should be registered in the journal. It is not necessary to weigh a working standard in a glass container with capillary outlet.
The following criteria should be fulfilled in order to achieve a necessary degree of certainty in the identification and quantification of organic compounds:
Are the mass fragmentograms pure and undisturbed or are there extra or missing signals?
Is the gas chromatographic separation adequate?
Retention times: Deviations in the relative retention times in
relation to the isotopic marked quantification standards should be between +3
and 0 seconds.
The deviation relative to non-isotopic marked
quantification standards should be between +2 and 2 seconds.
The signal-to-noise ratio should be larger than 3:1.
The ratio between two measured isotope signals should be inside ±20% of the theoretical value (alternatively the standard value).
The report should include:
The following should be stored:
It is not necessary to store paper copies of chromatograms or fragmentograms
There are always a number of possible errors that may affect the quality of the results. It is not possible to eliminate all these errors because samples are different due to a number of factors and because every step in the procedure has inherent possibilities for errors. Validation of a method must therefore be a continuous process.
The following is a list of some of the more important possible errors with measures or control routines:
An important part of the quality control of the results is the comparison of the measured sample concentration with the blank values of the method (calculated on the basis of the sample amount). Before the preparation and clean-up of every new series of samples or new sample type is started, a blank sample should be run through the procedure. The result for this sample will represent the blank value of the method.
In the case of larger series of the same type of samples, it will be enough to run a blank sample (filter and PUFs) for every 30 real sample unless there are other considerations that make it necessary with more frequent blank samples. One such consideration is the analysis of a sample with unexpected high concentration (more than 100 times above the normal level). A blank field sample (representing the blank value of the whole process including sampling, transport and preparation and clean-up) should be run 2 to 3 times a year for each sampling site.
Criteria for
acceptance of blank values
The results of a blank sample is accepted if the blank
values for every component to be quantified is lower than the limit of
detection (signal-to-noise ratio larger than 3:1) or at lower than 1/10 of the
lowest expected concentration level. For a larger blank sample series (more
than 5 blank samples) the limit of quantification may be utilised. This level
of quantification is defined as the average of the blank value (for a
component) plus 3 standard deviations.
Intercomparisons are an important tool for validating the operating procedure. The laboratory should try to achieve participation in at least one intercomparison a year for each sample type (air, precipitation, sediment and biological samples).
Vogelsang, J. (1991) The quality control chart principle: Application to the routine analysis of pesticide residues in air. Fresenius J. Anal. Chem., 340, 384-388.