BS EN 14110 Fat and oil derivatives - Fatty Acid Methyl Esters (FAME) - Determination of methanol content
7 Procedure
Two alternative procedures, the first using internal calibration and the second using external calibration are described in clauses 7.1 and 7.2, respectively.
7.1 Procedure A - Using internal calibration
This procedure is generally preferred when only a small number of samples is analyzed and when automatic head space equipment is not available. For manual procedure, see note in clause (4.10).
7.1.1 Internal calibration
(5 +/- 0.01) g from each calibration solution is transferred into a head space vial (4.1) and 5 µl of 2-propanol (3.2) is added into the liquid phase using a syringe (4.3), the addition being made into the liquid phase. The vials are then immediately crimped and shaken vigorously to ensure mixing.
Every 10 min, introduce into thermostatically-controlled bath or oven (4.12) a vial of the calibration solution which shall be kept there for exactly 45 min.
Preheat the gas syringe (4.4) to 60 °C in an oven (4.13). Sample 500 µl of gaseous phase (headspace) above the solution to be analysed and carry out the chromatographic analysis.
The calibration factor F is calculated for each calibration solution according to equation (1 ), shall be expressed rounded to the nearest 0.01.
F = (Cm x Si)/(Ci x Sm)
where
Ci is the 2-propanol content in the calibration solution, expressed in % (m/m);
(If 5 µl of 2-propanol are introduced into 5.0 g calibration solution, then Ci = 0.0785 % (m/m)).
Cm is the methanol content in the calibration solution, expressed in %(m/m);
Si is the peak area of 2-propanol;
Sm is the peak area of methanol.
The calibration factor values obtained for the three reference solutions shall exhibit a coefficient of variation below 15 %. If this value is exceeded, the experimental set-up shall be inspected for errors and the calibration procedure shall be repeated starting from clause (6). The mean of these calibration factor values (in the region of 0.7) is used for the calculation described in clause 7.1.2.
7.1.2 Analysis and Calculation using internal calibration
The samples shall be prepared and analysed with exactly the same experimental conditions used in the calibration runs (7.1.1). The methanol content Cm of a sample, expressed in % (m/m), is calculated according to equation (2) and rounded to two decimal places:
Cm = (F x Sm x Ci)/Si
where
F is the calibration factor obtained according to clause 7.1.1;
Sm is the peak area of methanol;
Ci is 2-propanol content added to the sample, expressed in % (m/m);
(If 5 µl of 2-propanol are introduced into 5.0 g sample, then Ci = 0.0785 %(m/m)).
Si is the peak area of 2-propanol.
7.2 Procedure B - Using external calibration
This procedure is usually preferred when automatic head space equipment is used and a large number of sampled is analysed. It is not recommended to use external calibration when the analysis is executed manually, i.e. without automatic head space equipment.
7.2.1 External calibration
2 ml from each calibration solution is transferred into a head space vial. The vial shall be crimped immediately. The vials are then placed into the head space sampler and the analysis is started according to the manufacturer's instruction manual.
The calibration function is calculated using linear regression, using the methanol contents as dependent variable and the peak areas as independent variable. Using the resulting slope, d, and y-intercept, e, the regression function is then converted using equation (3):
Cm = a + b x Sm
where
Cm is the methanol content in % (m/m);
a is the coefficient obtained from (-e/d);
b is the coefficient obtained from (1/d);
Sm is the area of the methanol peak;
d is the slope of regression line;
e is y intercept of regression line.
NOTE If the correlation coefficient is less than 0.95, the procedure should be inspected for errors and the calibration procedure should be repeated starting with clause 6.
7.2.2 Analysis and calculation using external calibration
The samples shall be prepared and analysed with exactly the same experimental conditions used in the calibration runs (7.2.1). The methanol content Cm of a sample, expressed in % (m/m), is calculated according to equation (3) and rounded to the nearest 0.01 % (m/m).
