150 / 220
Dual Lab Validation of a Method for the Determination of Fructans in Infant
Formula & Adult Nutritionals /
25 Jul 2016
40
Table 4
Quadratic calibration models for the HPAEC‐PAD quantification of
glucose and fructose
Constituent
Concentration range
(mg/l) injected analyte
HPAEC‐PAD
instrument
Calibration model
Correlation
coefficient
24 March 2016
Glucose
0.054 – 21.78
ICS5000_E
y = 0.052 x
2
+ 6.539 x +0.006
R
2
= 0.9999
Fructose
0.887 – 177.4
ICS5000_E
y = 0.409 x
2
+ 17,534 + 0.070
R
2
= 0.9999
10 May 2016
Glucose
0.054 – 21.78
ICS5000_E
y = 0.058 x
2
+ 5.680 x +0.005
R
2
= 0.9999
Fructose
0.887 – 177.4
ICS5000_E
y = 0.1199 x
2
+ 12.574 + 0.037
R
2
= 0.9999
21 June 2016
Glucose
0.051 – 20.49
ICS3000_D
y = 0.044 x
2
+ 5.754 x +0.003
R
2
= 1.0000
Fructose
0.894 – 178.7
ICS3000_D
y = 0.811 x
2
+ 18.129 + 0.251
R
2
= 0.9998
In which:
Y = calculated sugar (glucose or fructose) content in the injected 20 μl sample
X = peak area of the sugar (glucose or fructose) in the chromatogram
The relative residuals were calculated of the predicted concentration and the actual concentration
was, with exception of the lowest standard sample, always less than 5% relative. For the lowest
calibration standards in these extended quadratic calibration models the lack of fit at most about 15%
relative.
In annex B the raw data of the calibration models and the calculated lack of fit of the individual
calibration standards are given.
c.
LOD and LOQ
The no fructan containing sample 7 (Infant formula powder partially hydrolyzed milk based (lot
number 410057652Z)) was selected for the determination of the LOD and LOQ by standard addition
experiments. The measurements have been carried out at two different days, being February 28
th
,
2016 and May 23
rd
, 2016. The standard addition has been done in the reconstituted sample. The Orafti
P95 fructan was applied for the spiking. For the standard addition 100 μl of an fresh prepared aqueous
Orafti P95 solution with a fructan content of 942,6 mg/50ml (Febr. 28) respectively 941.4 mg/50ml
(May 23) was added to 4.00 ml of the reconstituted sample. With this addition the fructan
concentration in the reconstituted sample was adjusted to 0.046 % which is slightly higher than the
LOQ target of 0.03 %.
The results are summarized in table 4.
All the experimental results concerning the determination of the LOQ are presented in Annex C
VALIDATION REPORT
FOR ERP USE ONLY
DO NOT DISTRIBUTE