377 / 471
(
d
) The method can also determine free choline and free carnitine by bypassing the microwave digestion
and basic hydrolysis. Simply dilute the same sample size with water, add 125 µL MIX-IS, and dilute to 10
mL in a tube. Mix and then dilute this solution 1 to 25 mL with water in another tube. Mix, filter about 1
mL of sample slurry, and then dilute 10x with acetonitrile as above. No nitric acid or ammonia is used.
G. Instrument Operating Conditions
(
a
)
Mass spectrometry conditions.
--
See
Tables
2014.04D
and
E
.
(
b
)
UPLC conditions.
–
See
Tables
2014.04F
and
G
.
(
c
)
UPLC analysis
.—Column stability was improved by storing the column in water-acetonitrile (5 + 95,
v/v) without additives (recommended by column supplier). After verifying equilibration of the UPLC
system, inject the mid-level WS four times to verify system suitability. %RSD of the peak areas from
these injections should be <5%. Once system suitability has been established, inject WS 1-3, followed
by a reagent blank, control sample, and samples. Reinject WSs approximately every 4 h (e.g., enough
t ime for 16 samples with analysis cycle time of 15 min).
H. System Suitability
(
a
) The RSD of the four standard injections to prove equilibration prior to run must be <5%.
(
b
) Calibration curve residuals must be ≤4%. Samples should be bracketed by two sets of such valid
calibration curves.
(
c
) A suitable control sample is NIST SRM 1849a, reconstituted as a normal sample powder (each packet
contains about 10 g). A control sample must be run concurrently with every sample set and a
corresponding control chart set up. The control chart RSD of the means of choline and carnitine must be
<4.0%.
(
d
) The method is valid for analytical solution concentrations between 50% of WS1 and 10% above WS3.
I. Calculations
(
a
) For each of the three WSs, the software plots each relative response (analyte/IS) versus its
corresponding WS concentration to obtain two separate calibration curves for choline and carnitine (two
data points for each concentration, one from the beginning of the analysis and one from the end). It
applies a linear regression model to the data, and obtains an equation for the best-fit line.
(
b
) For each sample injected, the instrument measures the response (analyte/IS) for choline and
carnitine and uses the linear regression equation to calculate the resulting concentration in analytical
solution.
(
c
) The concentration in the analytical solution (ng/mL) is multiplied by a dilution factor to project the
results back to the original sample, on a µg/g basis:
C
x
= C
s
× DF (Equation 1)
Candidates for 2016 Method of the Year
315