ANALYTICAL MEASUREMENT

UNCERTAINTY ON PESTICIDE
RESIDUE ANALYSIS

EURACHEM GUIDE ON MU 3rd EDITION

Modified QuEChERS
Measurement Uncertainty
 Every step may contribute to the over-all uncertainty
 Identify uncertainty components
 Construct a fishbone diagram
 Quantify uncertainty components
 Calculate combined and expanded uncertainty
Definitions of Terms
 Uncertainty- relates to the general concept of doubt
 Uncertainty components- factors that contribute to the CSU
 Standard uncertainty- expressed as standard deviation
 Combined standard uncertainty- combination of all SU from
uncertainty components
 Expanded standard uncertainty- an interval that gives higher
confidence level of the measurand value
 Measurand- a quantity intended to be measured
 Error- difference between an individual result from the true value
Modified QuEChERS
15.0 g homogenized
NOTES:
sample
-1.0 kg homogenized sample
15.0 mL ACN
Extraction -Spiked sample
7.2 g MgSO4
LT- 0.2 mL 0.75 ppm
1.7 g NaCl HT- 0.2 mL of 7.5 ppm

6.0 mL aliquot
Clean-up 1 C18 pre-packed SPE
20 mL acetonitrile

Reduce to 2 mL
Figure 1. Procedure flow of multi-pesticide
Clean-up 2 GCB/PSA
residue analysis using Modified QuEChERS
12 mL 60:40 ACN: DCM

Final Volume 3.0 mL acetone

Fishbone Diagram
Equivalent
Linear calibration curve mass/Aliquot msample
volume calibration
sensitivity
replicates
temperature calibration linearity
mtare
linearity repeatability sensitivity
linearity
Regression coefficients calibration mgross

canalyte
temperaturerepeatability calibration

temperature temperature
Dilution volume
mgross repeatability
repeatability
linearity
manalyte calibration
calibration
linearity purity
Final volume mtare
***Spiked std (recovery VACN
only)

Figure 2. Fishbone diagram showing the identified

uncertainty components
Measurand
( 𝑅 − 𝐵 0)
𝑐 ( 𝑎𝑛𝑎𝑙𝑦𝑡𝑒 ) = 𝑥¿¿
𝐵1
Where,

Canalyte concentration of analyte in mg/Kg or ppm

R response of the instrument due to analyte
B0 y-intercept of the linear calibration curve
B1 slope of the line
VACN volume of acetonitrile added to sample
msample mass of sample
maliquot equivalent mass of sample aliquot
Vfinal volume final volume acetone
Measurand
( 𝑅 − 𝐵 0)
𝑐 ( 𝑎𝑛𝑎𝑙𝑦𝑡𝑒 ) = 𝑥¿¿
𝐵1
Where,
0 .5
Canalyte concentration of analyte in mg/Kg or ppm
R =
R response of the instrument due to analyte TO
AC
B0 y-intercept of the linear calibration curve
F
B1 slope of the line
O N
T I
VACN volume of acetonitrile added to sample
LU
msample mass of sample DI
maliquot equivalent mass of sample aliquot
Vfinal volume final volume acetone
Uncertainty Components
 Mass of homogenized sample
 Volume of Acetonitrile added
 Equivalent mass aliquot
 Final volume
 Recovery spike
 Linear calibration curve

 Note: does not allow possibility of spurious errors

 Formula for Combined Standard Uncertainty

√( )( )( )( )( )
𝟐 𝟐 𝟐 𝟐 𝟐
𝒖 ( 𝒎𝒔𝒑𝒍 ) 𝒖 (𝑽𝑨𝑪𝑵 ) 𝒖 ( 𝒎𝒂𝒍𝒊𝒒 ) 𝒖 ( 𝑽 𝒇𝒊𝒏𝒂𝒍 ) 𝒖 ( 𝒄𝒂𝒍 )
𝒖 ( 𝒄 )=𝒄(𝒂𝒏𝒂𝒍𝒚𝒕𝒆)∗ + + + + +¿¿ ¿
𝒎(𝒔𝒂𝒎𝒑𝒍𝒆) 𝑽 ( 𝑨𝑪𝑵 ) 𝒎(𝒂𝒍𝒒) 𝑽 ( 𝒇𝒊𝒏𝒂𝒍) 𝒄(𝒄𝒂𝒍)

Where,

Canalyte concentration of analyte in mg/Kg or ppm

VACN volume of acetonitrile added to sample
msample mass of sample
maliquot equivalent mass of sample aliquot
Vfinal volume final volume acetone
Ccal concentration of test recovery obtained from replicated standard cal. Curves
Mspike amount of analyte spiked in a sample
Uncertainty Component
A. MASS OF SAMPLE

 Fifteen grams (15 g) of sample is weighed on a top loading

balance with a linear uncertainty 0.01 g. Using k=1.96
based on 95% confidence level,


g
 The contribution accounted twice because of the one for
the tare and second for the gross weight.
Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)

 3 major contributors
 Calibration

-Glassware specification
 Repeatability

-Standard deviation from 10 trials of volume delivery

 Temperature effects

-temperature limit, volume and coefficient of expansion

Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)

 CALIBRATION
 Assuming triangular distribution, the calibration
uncertainty is
Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)

 REPEATABILITY
 determined by computing the SD of 10 volumes dispensed
from the measuring glassware. The standard deviation at
10 trials is
 u(Vrep) =0.0032mL.
Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)
 REPEATABILITY (EXAMPLE)
50-mL Volumetric flask
Mass of empty Mass of filled container Volume of water
Trial container (g) (g) Mass of water (g) (mL)
1 34.7634 84.4637 49.7003 49.89438917
2 34.7613 84.4831 49.7218 49.91597314
3 34.7664 84.4954 49.729 49.92320125
4 34.7653 84.4623 49.697 49.89107629
5 34.7571 84.4825 49.7254 49.91958719
6 34.7704 84.475 49.7046 49.89870597
7 34.7615 84.4995 49.738 49.9322364
8 34.7688 84.4697 49.7009 49.89499152
9 34.762 84.4998 49.7378 49.93203562
10 34.7666 84.4713 49.7047 49.89880636
STDEV 0.016199
Uncertainty 0.016199
density of water @ 28.5C is 0.99611 g/mL
Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)

 TEMPERATURE EFFECTS

mL
Uncertainty Component
B. VOLUME ACN (15 mL volumetric pipet)

 Combined uVACN

Combining the 3 sources, we obtain

Uncertainty Component
C. EQUIVALENT MASS ALIQUOT 1:1(6 mL volumetric pipet)

 CALIBRATION

 REPEATABILITY at 10 trials= 0.0026 mL

 TEMPERATURE= 0.014 mL

following 1:1 ratio

Uncertainty Component
D. FINAL VOLUME (3 mL acetone)

 CALIBRATION

 REPEATABILITY at 10 trials= 0.0050 mL

 TEMPERATURE= 0.017 mL
Uncertainty Component
E. RECOVERY SPIKE

 Addition of 200 uL of 0.75 and 7.5 ppm standard solution is required

to check the QC and recovery
 contributors to the uncertainty- mass and purity of the primary
standards, volume used for dilution and volume spiked (e.g.)

 Every step has uncertainty!!!

Uncertainty Component
E. RECOVERY SPIKE

 i. PURITY OF LINDANE
given on the certificate of analysis as 0.99±0.01. Applying
rectangular distribution, the SU is
Uncertainty Component
E. RECOVERY SPIKE

 ii. MASS OF LINDANE

The tolerance of the analytical balance is ±0.1 mg.

At k=1.96 at 95% confidence level, uncertainty is
0.051 mg.

Accounting for gross and tare,

= 0.072 mg
Uncertainty Component
E. RECOVERY SPIKE

 iii. VOLUMES/DILUTION

Volume Calibration Repeatability Temperature effects Total uncertainty

(V) u(Vcal) u(Vrep) u(Vtemp) u(V) mL
0.200 0.00049 0.00030 0.00066 0.00088

2.50 0.0061 0.0082 0.0082 0.013

3.00 0.0061 0.0050 0.017 0.019

10.00 0.010 0.0097 0.033 0.036

50.00 0.024 0.017 0.16 0.17

 Uncertainty Component
E. RECOVERY SPIKE

 iv. COMBINED SU

√( )( )( )( )( ) ( )( )
2 2 2 2 2 2 2
𝑢( 𝑃 ) 𝑢 ( 𝑚) 𝑢 ( 𝑉 2.5 ) 𝑢 (𝑉 3.0 ) 𝑢 ( 𝑉 0.2 ) 𝑢 ( 𝑉 50 ) 𝑢 ( 𝑉 10 )
( 𝑚 .𝑎 ) =𝑚𝑔𝑠𝑝𝑖𝑘𝑒 ∗ + + + + +2 +
𝑃 𝑚 𝑉 2.5 𝑉 3.0 𝑉 0.2 𝑉 50 𝑉 10

√( )( )( )( )( ) ( )( )
2 2 2 2 2 2 2
0.0058 0.072 0.013 0.019 0.00088 0.17 0.036
𝑢 ( 𝑚 .𝑎 ) =0.0015 ∗ + + + + +2 +
0.99 25.3 2.5 3.0 0.2 50 10

u000019 mg
 Uncertainty Component
E. RECOVERY SPIKE
 iv. COMBINED SU

√( )( )( )( )( ) ( )( )
2 2 2 2 2 2 2
𝑢( 𝑃 ) 𝑢 ( 𝑚) 𝑢 ( 𝑉 2.5 ) 𝑢 (𝑉 3.0 ) 𝑢 ( 𝑉 0.2 ) 𝑢 ( 𝑉 50 ) 𝑢 ( 𝑉 10 )
( 𝑚 .𝑎 ) =𝑚𝑔𝑠𝑝𝑖𝑘𝑒 ∗ + + + + +2 +
𝑃 𝑚 𝑉 2.5 𝑉 3.0 𝑉 0.2 𝑉 50 𝑉 10

√( )( )( )( )( ) ( )( )
2 2 2 2 2 2 2
0.0058 0.072 0.013 0.019 0.00088 0.17 0.036
𝑢 ( 𝑚 .𝑎 ) =0.0015 ∗ + + + + +2 +
0.99 25.3 2.5 3.0 0.2 50 10

This is just for LINDANE! Compute each

u000019 mg
analyte individually!!
Uncertainty Component
F. LINEAR CALIBRATION CURVE
 The amount of analyte is calculated using a manually
prepared calibration curve
 concentrations 0.005 ppm, 0.01 ppm, 0.02 ppm, 0.05 ppm,
0.1 ppm and 0.2 ppm
 Use linear least square fit
 uncertainties of abscissa less than ordinate
 random variation in the signals/areas produced by the
instrument.
Uncertainty Component
F. LINEAR CALIBRATION CURVE
Concentration Peak Area
0.005 ppm 2293
0.01 ppm 4841
0.02 ppm 9639
0.05 ppm 27353
0.10 ppm 54613
0.20 ppm 111485
Unknown/recovery 12252
Uncertainty Component
F. LINEAR CALIBRATION CURVE
Uncertainty Component
F. LINEAR CALIBRATION CURVE
 Calculating Combined Standard Uncertainty for Lindane

√( )( )( )( )( )
𝟐 𝟐 𝟐 𝟐 𝟐
𝒖 ( 𝒎𝒔𝒑𝒍 ) 𝒖 (𝑽𝑨𝑪𝑵 ) 𝒖 ( 𝒎𝒂𝒍𝒊𝒒 ) 𝒖 ( 𝑽 𝒇𝒊𝒏𝒂𝒍 ) 𝒖 ( 𝒄𝒂𝒍 )
𝒖 ( 𝒄 )=𝒄(𝒂𝒏𝒂𝒍𝒚𝒕𝒆)∗ + + + + +¿¿ ¿
𝒎(𝒔𝒂𝒎𝒑𝒍𝒆) 𝑽 ( 𝑨𝑪𝑵 ) 𝒎(𝒂𝒍𝒒) 𝑽 ( 𝒇𝒊𝒏𝒂𝒍) 𝒄(𝒄𝒂𝒍)
Reporting concentration w/ MU

0.35 ppm ± 0.018 ppm

NOTE: The total uncertainty of all uncertainty components are

assumed to be fixed. However the reported expanded SU is
dependent on the value of the final concentration.
Assumptions
 Uncertainties estimated not intended to for spurious errors/
blunder
 errors arise from human technicalities or instrument
malfunction
 all accessories and consumables used in the method such
as centrifuge, SPE cartridges, rotary evaporator and others,
work well
 do not contribute to any loss/gain of analytes
Thank you!