Uncertainty Calculation in Chemical Metrology
Uploaded by
Rosa Lilia JordanUncertainty Calculation in Chemical Metrology
Uploaded by
Rosa Lilia JordanCENTRO DE METROLOGIA QUIMICA
CALCULATION OF
UNCERTAINTIES
FOR THE RESULTS
REPORTED IN [Link]-P2
ANALYTICAL MEASUREMENT
As, Cu, Fe, Mn, Ni, Pb,
(Zn)
Perkin-Elmer AAnalyst700
Atomic Absorption
Spectrophotometer
equipped with a
longitudinal graphite
furnace
Cd
Flame Atomic Absorption
Spectrophotometry
Perkin-Elmer AAnalyst100
and using high
sensitivity nebulizer
ANALYTICAL MEASUREMENT
NIST-SRMs were used for the preparation
of calibration standards.
The quantification method used was
bracketing
Four independent aliquots of the sample
were analyzed, in all cases.
NIST-SMR 1643d was used as control
(except for Cd)
ANALYTICAL MEASUREMENT
Identify Uncertainty Sources
THE UNCERTAINTY ESTIMATION
PROCESS
Calculate combined standard uncertainty
Specify associated model
Quantify Uncertainty Sources
Reporting Uncertainty
Convert to standard uncertainty
ASSOCIATED MODEL:
BRACKETING
C
L
C
H
A
L
A
H
C
S
A
S
L S
L S
L H
L H
C C
A A
C C
A A
Sample concentration is
calculated through the equation
( )
( )
( ) d C C C
A A
A A
C L L H
L H
L S
S -
|
|
.
|
\
|
+
=
IDENTYFING UNCERTAINTY
SOURCES
repeatability
calibration
temperature
C
sample
d
A
L
, A
H
, A
s
repeatability
repeatability
C
L
, C
H
calibration
temperature
F(dil)
U-SMR
QUANTIFYING UNCERTAINTY
SOURCES
repeatability
calibration
temperature
C
sample
d
A
L
, A
H
, A
s
repeatability
repeatability
C
L
, C
H
calibration
temperature
F(dil)
U-SMR
Repeatability of A
(S,L,H)
n
s
u = ) (A Std) (S,
The uncertainty due to random variation of the EAA signal
measurement of samples and bracketing standards was calculated
as:
QUANTIFYING UNCERTAINTY SOURCES
Fe Determination
Std Low Std High SIM
Average Signal 0,1704 0,2448 0,1955
Std desv 0,0002 0,0019 0,0030
n 3 3 4
ui 0,00011 0,00107 0,00150
repeatability
calibration
temperature
C
sample
d
A
L
, A
H
, A
s
repeatability
repeatability
C
L
, C
H
calibration
temperature
F(dil)
U-SMR
QUANTIFYING UNCERTAINTY SOURCES
C
sample
repeatability
C
L
, C
H
calibration
temperature
F(dil)
n (dil) F C C SMR H , L - =
|
|
.
|
\
|
+
|
.
|
\
|
=
n
2
2
F
u
C
u
C u
SMR
*
(dil)
(F(dil)) (SMR)
H) (L, ) (C H L,
Equation for concentration assessment
Equation for the uncertainty calculation
U-SMR
2 2
V
u
V
u
Fdil u
|
.
|
\
|
+
|
.
|
\
|
- =
fin
(Vfin)
in
(Vin)
(Fdil)
C
sample
repeatability
C
L
, C
H
calibration
temperature
F(dil)
|
.
|
\
|
SMR C
u(SMR)
3
U
u
(certif)
(SMR)
=
|
.
|
\
|
Fdil
u ) Fdil (
( ) ( ) ( )
2 2 2
u u u T rep cer
f Vi,
u + + =
U-SMR
2
2
3
1
Fdil
u
C
u
C u * |
.
|
\
|
+
|
.
|
\
|
=
(Fdil)
SMR
) (C
(SMR)
(L) L
3 2 1 SMR Fdil Fdil Fdil C CL - - - =
C
L
= 9.827 g/L u
C
= 0.055 g/L
PREPARATION OF STANDARDS
Values and uncertainties
Description Nominal value standard uncertainty Relative std (u
(x)
/x)
2
u
(x)
uncertainty u
(x)
/x
C
SMR
(mg/L)
109,83 0,19 0,00173 2,99E-06
Fdil
1 0,0090 3,63E-05 4,04E-03 1,63E-05
Fdil
2 0,0998 1,42E-04 1,42E-03 2,03E-06
Fdil
3 0,0998 3,19E-04 3,20E-03 1,02E-05
repeatability
calibration
temperature
C
sample
d
A
L
, A
H
, A
s
repeatability
repeatability
C
L
, C
H
calibration
temperature
F(dil)
QUANTIFYING UNCERTAINTY SOURCES
U-SMR
( )
( )
( ) d C C C
A A
A A
C L L H
L H
L S
S -
|
|
.
|
\
|
+
=
CALCULATING THE CONCENTRATION
Values and uncertainties in determination of Fe on SIM-P2 sample
Description Value x Standard uncertainty
u
(x)
A
S SIM-P2 sample Absorbance 0,1955 0,0015
A
L Low standard Absorbance 0,1704 0,0001
A
H High standard Absorbance 0,2448 0,0011
C
L
Low standard Concentration (
g/L)
9,827 0,055
C
H
High standard Concentration (
g/L)
13,780 0,073
d Sample diluted factor 8,349 0,044
CALCULATING THE COMBINED STANDARD
UNCERTAINTY
( )
( )
( ) d C C C
A A
A A
C L L H
L H
L S
S -
|
|
.
|
\
|
+
=
2 2
C
2
C
2 2 2
d
S
H
S
L
S
A
H
S
A
L
S
A
S
S u
d
C
u
C
C
u
C
C
u
A
C
u
A
C
u
A
Cs
C uc
2
H
2
L
2
H
2
L
2
S
2
) Cs (
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
- =
|
|
.
|
\
|
=
i
2
x
2
i
1
u
x
f
u(y)
) (
c
c
) x ...., x , x ( f y i 2 1 =
2 2 2
) (x
n
) (x
2
) (x
1
n
2
2
2
1
2
u
x
y
........ u
x
y
u
x
y
u(y)
|
.
|
\
|
c
c
+ +
|
.
|
\
|
c
c
+
|
.
|
\
|
c
c
=
KRAGTEN
(*)
Finally
2
) y(x u x y u
x
y
) (x ) (x
2
2
|
.
|
\
|
~ +
|
.
|
\
|
c
c
1 1 1 1
1
) (
) (x
) (x ) (x
u
y u x y
x
y
1
1 1 1
1
) ( +
~
c
c
2
2
2
) (x *
) (x
) (x ) (x
) (x u
u
y u x y
u
x
y
|
.
|
\
|
+
~
|
.
|
\
|
c
c
1
1
1 1 1
1
1
) (
(*) J. Kragten, Analyst, 119, 2161-2166 (1994)
CALCULATION OF COMBINED STANDARD
UNCERTAINTY
A
S
A
L
A
H
C
L
C
H d
Value 0,19552 0,17039 0,24483 9,82683 13,78000 8,34912
u 0,00150 0,00011 0,00107 0,05523 0,07262 0,04367
A
S 0,1955 0,1970 0,1955 0,1955 0,1955 0,1955 0,1955
= x
1
+ u(x1)
A
L 0,1704 0,1704 0,1705 0,1704 0,1704 0,1704 0,1704
A
H 0,2448 0,2448 0,2448 0,2459 0,2448 0,2448 0,2448
C
L 9,8268 9,8268 9,8268 9,8268 9,8821 9,8268 9,8268
C
H 13,780 13,780 13,780 13,780 13,780 13,853 13,780
d 8,3491 8,3491 8,3491 8,3491 8,3491 8,3491 8,3928
Conc.(mg/L) 93,19 93,853 93,156 93,029 93,493 93,392 93,675
y =equation y'
1
=y(x
1
+ u(x1)) idem idem idem idem idem
u(y,x
i
) 0,6659 -0,0318 -0,1586 0,3054 0,2047 0,4874
A1
= y - y'
1
idem idem idem idem idem
u
2
(y), u
2
(y,xi)
0,84236 0,44344 0,00101 0,02515 0,09329 0,04190 0,23757
uncertainty (u) 0,92
Expanded uncert. (U) 1,84
Relative U (%) 1,97
UNCERTAINTY CONTRIBUTIONS
Uncertainty contributions
0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00
Cs
A S
A L
A H
C L
C H
d
u
n
c
e
r
t
a
i
n
t
i
e
s
s
o
u
r
c
e
s
u(y,xi)
Finally, the concentration and uncertainty
values for Fe in the sample [Link]-P2 are:
Concentration C
S
= 93.19
g/L
Combined uncertainty u
com
= 0.92
g/L
Expanded uncertainty U = 1.84 g/L
U=u*k, (k=2) (1.97%)
RESULTS OF Intec IN [Link]-P2
SUMMARY OF RESULTS OBTAINED ON SAMPLE SIM-P2
Element Average concentration Expanded Uncertainty REU
( g/L) ( g/L) (%)
Ni 62,86 2,52 4,0
Fe 93,19 1,84 2,0
Cu 17,06 0,44 2,6
As 50,94 1,02 2,0
Pb 21,86 0,66 3,0
Mn 39,44 1,58 4,0
Cd 144 4 2,8
RESULTS OF [Link]-P2
RESULTS OF [Link]-P2
CENTRO DE METROLOGIA QUIMICA
