Page 45 - Volume 3, Issue 2, July 2010

Anagnostopoulos

et al.

76

combined with an ECD, or a DB-5-MS col-

umn and a DB-17 MS column each of them

combined with a NPD, or a DB-5-MS column

combined with an ECD and a DB-17 MS col-

umn combined with a NPD, or a DB-5-MS

column combined with a NPD and a DB-17

MS column combined with an ECD. Howev-

er, the quantification of fensulfothion sul-

fone must be done with a NPD detector. The

experiments for recovery of the analytes

fensulfothion-oxon and terdufos sulfoxide

failed at the lowest fortification level and for

omethoate failed at both levels.

Limit of Quantification (LOQ)

The limit of quantification (LOQ) was es-

tablished as the lowest concentration tested

for which recovery and precision were sat-

isfactory (70–110% and <20% RSD, respec-

tively) in accordance with the criteria es-

tablished for analysis of pesticide residues

in foods (4). Therefore, in most cases the

method’s limit of quantification (LOQ) was

the lowest fortification level, i.e. 3 μg/kg.

Conclusions

In the present study, 25 target pesticides

and metabolites, selected according to the

European Union Directives 2006/141 and

2006/125, were studied in cereal-based baby

foods. The extraction procedure was based

on the QuEChERS method and the deter-

mination was performed with GC–ECD and

NPD. The validation results presented good

accuracy with recoveries of 79.2 – 124.6%,

precision with RSD of 3.4 - 28% and limits of

quantification meeting the EU legislation re-

quirements for the maximum residue lim-

its. Concerning the reporting of the results,

from the data of the EU proficiency tests, a

default expanded uncertainty figure of 50%

(corresponding to a 95% confidence level

and a coverage factor of 2), in general cov-

ers the inter-laboratory variability between

the laboratories for residue analysis (5).

The disadvantage of the method was

that fensulfothion-oxon and terdufos sul-

foxide failed to be determined at the 3 μg/

n

is the number of the data points in the cal-

ibration, (

y

i

- y

) is the residual for the i

th

point

and

b

is the slope of the regression line. The

contribution of the uncertainty of the cali-

bration curve to the total uncertainty of

the method was calculated at the middle

(C=15 μg/l) of the calibration curve for each

compound and was estimated to be 3.8 –

23.5%.

Trueness, precision

The trueness was estimated by calculat-

ing the attained recovery from fortification

experiments. For validating a method, mean

recoveries of 70–120% with a repeatability

RSD ≤ 20% are considered acceptable, while

in routine analysis, the acceptable recover-

ies are in the range of the mean recovery ±2

relative standard deviations (SDr) %, which

usually fall within the range 60-140% (4).

The mean recoveries were determined from

spiked flour samples, at two concentration

levels with five replicates for each level.

The recoveries were calculated using

matrix-matched calibration standards.

As

shown in Table 3, mean recoveries of the

samples fortified at the lowest fortification

level (3 μg/kg) were between 79.2 – 124.6%

with SDr less than 17.2% with some excep-

tions, as shown in Table 3 and at the second

fortification level (30 μg/kg) between 68.2

– 118.1% with SDr less than 23.9%. These re-

sults indicate generally good method accu-

racy and precision.

Dimethoate, fensulfothion sulfone, fen-

sulfothion-oxon-sulfone and terbufos-sul-

fone are sensitive to both detectors and the

qualification can be achieved by both ECD

and NPD at the lowest fortification level.

However, the mean recoveries are accepta-

ble only by one of the detectors. Therefore,

we can use both detectors for confirmation,

but for quantification of the analyte we can

use only the detector that provides good ac-

curacy. For example, fensulfothion sulfone

can be confirmed with a DB-5-MS column

combined with an ECD and a NPD detec-

tor, or a DB-17 MS column combined with an

ECD and a NPD detector, or a DB-5-MS col-

umn and a DB-17 MS column each of them