29 / 258
1584
H
aselberger
&
J
acobs
:
J
ournal of
AOAC I
nternational
V
ol
. 99, N
o
. 6, 2016
glucoheptose for each level of working standard. Fit the
peak area ratios (responses) and corresponding standard
concentrations, in µg/mL, to a quadratic model. Do not force
intercept to zero. Identify glucoheptose as the internal standard.
(d)
Sample weight factors
(SWF).—
Calculated as follows:
(
1
) R
TF products and commodities.
SWF = SW1 (g) × SW2 (g)
× 10000
(2)
Powder products.
1
2
10 000
( )
( )
( )
( )
( )
=
+
×
×
×
SWF
PW g
PW g WW g
SW g SW g
where SW
1
= weight of material delivered to a volumetric
flask (RTF, or reconstituted powder) and diluted as per sample
table guidelines. For example, an RTF product of 0.21–1.0%
fructan would have an SW
1
of 7 g (±10%) into a 200 mL
volumetric flask. SW
2
= weight of solution from volumetric
flask per sample table guidelines (volumetric dilution of SW
1
).
Value should be 0.2 g ± 10%. PW = weight of powder product
weighed to be reconstituted (5 g ± 10%). WW = weight of water
used to reconstitute powder products (40 g ± 10%).
(e)
Dilution factor (DF).—
Calculated as follows:
(1) Low-level fructan (RTF 0.03–0.2%, powder 0.27–1.0%).
DF = 50 g × 2 mL
(2) Mid-level fructan (RTF 0.21–1.0%, powder 1.1–10%).
DF = 200 g × 2 mL
(3) High-level fructan (RTF 1.1–5.0%, powder 11–45%).
DF = 500 g × 2 mL
(4) Commodity DF (46–100%).
DF = 1000 g × 2 mL
(f)
Commodity factors (CF).—
As determined, based on
analysis in Part I. The factors chosen for the three groups listed
in section
G
, in Part I, were done in such a way so as to cover
the spectrum of fructan commodities currently in use. Any
factor should then impart a bias of no more than ~5% for the
grouping that it covers.
If the average DP of the fructan is known, this factor can be
explicitly calculated according to:
CF
DP
DP
DP
Avg
Avg
Avg
(
1
) *(0.9 0.1 )
=
−
+
(g)
Sample total fructan (g/100 g).—
Interpolate the
concentration of fructose in the injected sample (µg/mL) using
the corresponding fructose response parameters. Calculate the
total fructan concentration using the proper weights and factors:
Total Fructan g
g
Interpolated Fructose x DF x CF
SWF
100
=
where DF = dilution factor; CF = commodity factor; SWF =
sample weight factor.
Validation Protocol
Linearity
A total of 43 standard curves (two bracketed sets of
standards for each curve) were collected during the course of
Figure 2016.06E. Control powder, milk based, chromatogram (Part II).
)58&7$1 >PRGLILHGE\+$6(/3$@
&/& B5
('B
Q&
PLQ
)UXFWRVH
*OXFRKHSWRVH
Table 1. Recovery from spiked in-house GOS containing placebo
Nominal
0.3% spike
Nominal
3% Spike
Sample type
No. of
replicates
a
Spiking
material
b
Avg. % rec.
RSD, % Avg. % rec.
RSD, %
Infant formula powder, low lactose milk based with GOS
9
scFOS
100
5.43
96.2
2.85
Infant formula powder, low lactose milk based with GOS
9
Oligofructose
99.0
6.09
96.0
2.35
Infant formula powder, low lactose milk based with GOS
9
Oligofructose/inulin 98.4
2.85
97.2
2.04
a
Triplicates on each of 3 days.
b
In-house material. scFOS = Ingredion Nutraflora FOS powder; oligofructose = Beneo Orafti P95 powder; oligofructose/inulin = Beneo Orafti Synergy 1.
29