Professional Documents
Culture Documents
Uncertainty in Chemistry Vol. 3 Examples
Uncertainty in Chemistry Vol. 3 Examples
Uncertainty in Chemistry Vol. 3 Examples
Traceability,
Validation and
Measurement
Uncertainty in
Chemistry: Vol. 3
Practical Examples
Traceability, Validation and Measurement
Uncertainty in Chemistry: Vol. 3
Nineta Hrastelj Ricardo Bettencourt da Silva
•
Editors
Traceability, Validation
and Measurement
Uncertainty in Chemistry:
Vol. 3
Practical Examples
123
Editors
Nineta Hrastelj Ricardo Bettencourt da Silva
European Association for Chemical Faculdade de Ciências
and Molecular Sciences (EuCheMS) Centro de Química Estrutural
Brussels, Belgium Universidade de Lisboa
Lisbon, Portugal
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword
Chemical and bio-analytical measurements are omnipresent and often very impor-
tant in our society. Just think of the quality of the food we eat, the air we breathe, the
role of these measurements in health care, in trade and in research. In all these cases,
people strive to get reliable data. There is an international standard for assuring the
quality of measurement data, namely EN ISO/IEC-17025. It contains particular
management as well as technical requirements. These technical requirements are
linked to the science behind the measurements, meaning that metrological issues
such as traceability, uncertainty and validation are at the heart of this.
So as to provide Life-Long Learning in this area, the TrainMiC® programme
[www.trainmic.org] was conceived in 2001 by the Institute for Reference Materials
and Measurements of the European Commission Joint Research Centre. First, it
addressed the need arising with those countries wanting to enter the EU at that
moment. Rather than approaching this training in an anecdotal way and organizing
ad hoc events, a programme was set up—called TrainMiC®—to create harmonised
training material as well as to disseminate knowledge in the various countries via a
network of authorised TrainMiC® trainers. Afterwards, the TrainMiC® programme
spread to the rest of the EU and Europe’s largest Life-Long Learning programme in
this area was created. Up to now, 20 national TrainMiC® teams have been set up
and more than 9000 people have been trained all across Europe at the end of 2014.
Trainers quickly realised the importance of having suitable examples for their
training events. They realised soon that creating examples adapted to the various
audiences is quite a labour-intensive activity. For this reason, sharing such exam-
ples proved to be an attractive proposition. Today, examples are reviewed and then
published mainly in an e-collaboration environment which is only available for the
authorised trainers. It was decided to publish part of the examples also in the format
of a series of books, with a first volume that appeared in 2010, a second one in 2011
and this now being the third volume.
v
vi Foreword
In the meantime, there is a lot of experience in using these examples for training
purposes. First and foremost these examples should be seen as a vehicle for
problem-based learning. Their intention is not to be a definitive guidance document,
explaining a unique way to approach a given problem. Typically, they are used in a
classroom context, where small groups of trainees will try to provide answers to the
questions raised. It is recommended that they have group discussions, because
discussion creates learning. Later on, the different groups present the outcome in a
session in plenum. Also from this, there is learning. The answers provided in the
book, are most likely one of the possibilities. As we all know, a lot depends on
interpretation.
vii
viii Preface
ix
Editors and Contributors
xi
xii Editors and Contributors
Contributors
1.1 Introduction
Phosphorus occurs in natural waters almost solely as phosphates. These are clas-
sified as free phosphate (orthophosphate), condensed phosphates (pyro-, meta-, and
other polyphosphates) and organically bound phosphates.
Various classes of phosphorus compounds can be quantified by molecular
spectrophotometry after different sample pre-treatment (Fig. 1).
The free phosphate and some condensed phosphates are quantified by molecular
spectrophotometry performed on an untreated aliquot (direct spectrophotometry).
This phosphorus compounds fraction is known as “reactive phosphorus” in specific
experimental conditions. Most of the condensed phosphates and some organically
bound phosphate are also analysed by spectrophotometry after the hydrolysis of the
sample aliquot with sulphuric acid at boiling water temperature. This phosphorus
compounds fraction excluding reactive phosphorus is known as acid-hydrolyzable
phosphorus in conditions prescribed by the measurement procedure. Finally, the
total phosphorus can be quantified in the samples after more drastic digestion
conditions (Fig. 1).
Therefore, the presented analytical procedure does not allow the discrimination
of the phosphorus compounds as their chemical classes.
Water sample
Reactive Total
(A+B)*
phosphorus (A) phosphorus (C)
Acid- Organic
hydrolyzable phosphorus (D)
phosphorus (B)
[D=C-(A+B)]
[B=(A+B)-A]
(i) Preliminary sample treatment: To 100 mL of sample, add one drop of phe-
nolphthalein indicator. Discharge pink colour with a strong acid. If more than
0.25 mL (5 drops) is required, take a smaller sample aliquot, discharge the
pink colour with acid and dilute to 100 mL with distilled water.
(ii) Colour development: Add, with thorough mixing after each addition, 4.0 mL
molybdate reagent (0.02 mol L−1 ammonium molybdate; 5.2 mol L−1 sul-
furic acid) and 0.5 mL stannous chloride reagent (0.11 mol L−1 stannous
chloride in glycerol). Hold samples, calibrators, and reagents within 2 °C of
one another and in the temperature range between 20 ºC and 30 °C.
After 10 min, but before 12 min, using the same specific interval for all
determinations, measure colour photometrically at 690 nm and compare with
a calibration curve, using a purified water blank.
(iii) Preparation of calibrators:
Calibrators were prepared from KH2PO4 salt using procedure described in
Fig. 2.
mðlg P) MðP2 O5 Þ
cðmg L1 of P2 O5 Þ ¼
VðmL) 2MðPÞ
c(mg L−1) Reactive phosphorus (RP) mass concentration in the sample expressed as
P2O5;
m(µg P) RP mass interpolated in the calibration curve expressed as P;
V(mL) Sample volumetric aliquot;
M(P2O5) Molar mass of phosphorus pentoxide;
M(P) Molar mass of phosphorous.
The measurement aims at assessing the compliance of surface water, intended for
the abstraction of drinking water, with limits settled by Directive 75/440/EEC. This
monitoring must be performed using a measurement procedure meeting require-
ments of Directive 79/869/EEC.
The validation of the measurement procedure involved the following studies and
the evaluation of the linearity of the spectrophotometer response (not included in
this document): (Tables 1, 2, 3, 4 and 5).
Table 2 Results from the analysis of a 2 µg P calibrator (µg P units) obtained under within-lab
reproducibility conditions (i.e. intermediate precision conditions)
1.8 2.1 1.9 2.3 1.9
1.9 2.0 2.1 2.1 2.0
1.7 2.2 2.2 – –
Mean value: 2.015 µg P
Standard deviation: 0.1725 µg P
Variance: 0.02974 (µg2 P)
6 R. Bettencourt. da Silva et al.
4.1 Equations
rffiffiffiffiffiffiffiffiffiffiffi
sy=x 1
uInt ¼ þ1
b n
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uP
u ðyi ^yi Þ2
t i
sy=x ¼
n2
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s ffi
tol 2 V DT aH2 O 2
uð V Þ ¼ pffiffiffi þ ðsr Þ þ 2
pffiffiffi
3 3
2 ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
Þ
uð R sobs uðcCRM Þ 2 sR 2 uðcCRM Þ 2
pffiffiffiffi þ pffiffiffiffi þ
¼
R cobs N cCRM
¼
R N cCRM
Calibrators (µg P in
Signal (a.u.) Repression parameters
100 mL volume)
0 0.002 Intercept (a.u.) 0.00206
2 0.023 Slope (a.u. µg-1) 0.0102
3 0.032 sy/x (a.u.) 0.00110
5 0.052
7.5 0.080
10 0.103 Linear correlation parameter
R(Pearson) 0.99966
2. Does the analytical procedure fulfil the requirement(s) for the intended
use?
Intended use of the To check compliance of water with limits set by Directive 75/440/EEC
results for water intended for the abstraction of drinking water. The
measurement performance requirements are laid down in Directive 79/
869/EEC
Parameters to be Value requested by the Value obtained during validation
evaluate customer
LOD LOD 0.02 mg L−1 LOD = (3sy/x/b)/V = 0.0032 mg L−1 of P;
of P2O5 0.0032 2.29 = 0.0073 mg L−1 of P2O5
(where sy/x, b and V are the residual standard
deviation and the slope of a least squares
calibration, and sample aliquot volume
respectively; 2.29 = M(P2O5)/(2 M(P))
LOQ Not specified(a)
(continued)
10 R. Bettencourt. da Silva et al.
(continued)
Repeatability 2 standard Instrumental signal repeatability:
deviation 0.04 mg 2 (sy/x/y) (m/V) 2.29 =
L−1 of P2O(b)
5 ¼ 2 (0.00110/0.103) (10/100) 2.29 =
¼ 0.0049 mg L−1 of P2O5; (where y is the
signal estimated for the largest mass of the
calibration interval, m, 10 µg of P which
corresponds to 0.23 mg L−1 of P2O5)
or
Measurement repeatability estimated from the
standard deviation of the difference, sd, of
duplicate sample results:
2 (0.00134 mg L−1 of P/sqrt(2)) 2.29 =
¼ 0.0043 mg L−1 of P2O5
(Table 5 of yellow pages)
Within-lab 2 standard 2 0.001725 2.29 = 0.0079 mg L−1 of
reproducibility deviation 0.04 mg P2O5
L−1 of P2O(b)
5 (estimated for a mass concentration of
0.046 mg L−1 of P2O(c) 5 ; Yellow pages:
Table 2)
(a)
Since LOQ = LOD/0.3, acceptable LOD would result in an acceptable LOQ
(b)
According to Directive 79/869/EEC “the range within which 95% of the results of measurements
made on a single sample, using the same method, are located”, defined as precision, shouldn’t be
larger than 10% of the limit of P2O5 in the water set in Directive 75/440/EEC: 0.4 mg L−1
(10% 0.4 mg L−1 = 0.04 mg L−1 of P2O5). Assuming the normal distribution of results, the half
range of results specified in the Directive is estimated by two times the standard deviation of
measurements precision. The Directive does not specify precision conditions. Therefore, this limit
is used for estimated repeatability and within-lab reproducibility (i.e. intermediate precision)
standard deviations
(c)
Ideally this information should be estimated for 0.4 mg L−1 of P2O5
(…)
examined and the average experimental value obtained”, defined as accuracy, shouldn’t be larger
than 20% of the limit of P2O5 in the water set in Directive 75/440/EEC: 0.4 mg L−1
(20% 0.4 mg L−1 = 0.08 mg L−1 of P2O5). In this Directive accuracy term is used for the
absolute mean error
Measurement of Total Reactive Phosphorus in Natural Water … 11
(e)
Assuming random and systematic effects affecting measurement results are adequately quantified
by the intermediate precision and mean analyte recovery respectively, the target (maximum
admissible) uncertainty is 0.10 mg L−1 of P2O5. The following equations are used to combine the
target values for precision and trueness to estimate the target expanded uncertainty, Utg [2]
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
0:042 0:082
U ¼ ku ¼ 2u ¼ 2
tg tg tg
2
p ffiffi
þ 3 ¼ 0:10 mg L1 of P2 O5
where utg and k are the target standard uncertainty and coverage factor respectively. The first term
within the square root is the square of the intermediate precision standard deviation and the second
term is the square of the bias standard uncertainty assuming mean error has a rectangular
distribution. This equation assumes that the reference value of recovery tests is affected by a
negligible uncertainty
(continued)
Participation in PT schemes The z-score should be estimated with a reference
standard deviation (“>”) 0.05 mg L−1 of P2O5 (half of
the target expanded uncertainty). Satisfactory z-score are
within the interval [−2, 2]
Measurement uncertainty Target expanded uncertainty “>” 0.10 mg L−1 of P2O5
Other-state
c(mg L−1) Reactive phosphorus (RP) mass concentration in the sample expressed as
P2O5;
m(µg P) RP mass interpolated in the calibration curve expressed as P;
V(mL) Sample volumetric aliquot;
M(P2O5) Molar mass of phosphorus pentoxide;
M(P) Molar mass of phosphorus.
3. Identify (all possible) sources of uncertainty
RP mass interpolated in the calibration curve, m(µg P) (statistical interpolation and
calibrator mass uncertainty components)
sample volumetric aliquot, V(mL)
and CRM certified value uncertainty components)
Recovery (mean analyte recovery, R,
Dilution of sample volumetric aliquot by acid addition, fdil (negligible component for most
samples)
Molar mass of phosphorus pentoxide (g mol−1) (negligible component)
Molar mass of phosphorus (g mol−1) (negligible component)
Measurement of Total Reactive Phosphorus in Natural Water … 13
(continued)
Input Standard Unit Remark
quantity uncertainty
fstd 0.0166 – The standard uncertainty associated with fstd, u(fstd), is
estimated from the RP mass relative standard uncertainty
of the calibrator with lowest quantity except the blank,
uStd2/2 (i.e. u(fstd) = fstd uStd2/2) [3]
M(P2O5) negligible g mol−1
M(P) negligible g mol−1
(a)
The estimated mass of RP in the diluted sample is affected by both the statistical interpolation of
sample signal in the calibration curve and by calibrator mass uncertainty. The ratio of the masses
of RP of any pair of calibrators should have a negligible uncertainty given the instrumental signal
precision to allow the reliable estimation of the interpolation uncertainty from least squares
regression model [2]
Calculations:
fstd:
V:
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
tol 2 V 4 aH2 O 2 0:08 2 100 4 0:00021 2
uðV Þ ¼ pffiffiffi þ ðsr Þ þ 2
pffiffiffi ¼ pffiffiffi þ ð0:014Þ þ 2
pffiffiffi
3 3 3 3
¼ 0:0684
tol Tolerance associated with the pipette nominal volume (0.08 mL);
sr repeatability of pipette manipulation (0.014 mL);
aH20 volume expansion coefficient for water (2.1 10−4 °C−1).
m:
rffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffi
sy=x 1 0:00110 1
uInt ¼ þ1 ¼ þ 1 ¼ 0:117lg P
b n 0:0102 6
7. Calculate the combined standard uncertainty (uc) of the result and specify
units
Using: Mathematical solution; Spreadsheet Approach; Commercial
Software
Calculations:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 2 uðV Þ2 uðR Þ 2
uðfStd Þ 2
Int
uðcÞ ¼ c þ þ þ
m V R fStd
s ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2 ffi
0:117 0:0684 0:0135 2 0:0166 2
uðcÞ ¼ 0:14590 þ þ þ
6:39 100 1 1
uðcÞ ¼ 0:0041 mg L1 of P2 O5
8. Calculate the expanded uncertainty (Uc) & specify the coverage factor k
and the units
Calculations:
Example: Percentage contribution, PInt, of the statistical interpolation uncertainty to
the combined standard uncertainty:
u 2
Int
¼ m
PInt
uInt 2 uðfStd Þ 2 uðV Þ 2 uR 2
þ þ þ
m fStd V R
ð0:0183Þ2
PInt ¼
ð0:0183Þ þ ð0:0166Þ2 þ ð0:000684Þ2 þ ð0:0135Þ2
2
PInt ¼ 0:422½42:2%
References
1. APHA, AWWA, WEF, Standard Methods for the Examination of Water and Wastewater, 22th.
Ed., USA, 2012.
2. R. J. N. B. Silva, Water 5 (2013) 1279–1302.
3. R. J. N. B. Silva, M. F. G. F. C. Camões, Anal. Lett. 43 (2010) 1257–1266.
chemistry while collaborating in teaching at national and foreign universities. Ricardo’s research
includes the development of approaches for the detailed evaluation of the uncertainty associated
with complex measurements and the assessment of the sources of lack of comparability of mea-
surements in some analytical fields. Ricardo has been a member of the IPAC Accreditation of
Chemical Laboratories Working Group since 2006, the Eurachem/CITAC Measurement
Uncertainty and Traceability Working Group since 2010, the Portuguese TrainMiC® team since
2008 and the TrainMiC®.Editorial Board since 2010.
as an assessor of the Portuguese Accreditation Body (IPAC), of the Israel Accreditation Body
(ISRAC) and of the Turkish Accreditation Body (TURKAK). Since 2007 Florbela Dias has been
in Portuguese TrainMiC team and since 2011 has been process manager and GLP inspector in the
scope of GLP principles of OECD.
Andreja Drolc
For nitrate we have here below text extracts of an ISO method and also the
requirements in the directive for nitrate.
International Standard EN ISO 10304-1 Water quality – Determination of dis-
solved anions by liquid chromatography of ions – Determination of bromide,
chloride fluoride, nitrate, nitrite, phosphate and sulphate.
1.1 Scope
1.1.1 General
The method is specified for the determination of nitrate in water (drinking water,
ground water, surface water, waste water, leachates and marine water) in the fol-
lowing range: <0.1 mg L−1 nitrate. The lower limit of application depends on the
matrix and the interferences encountered.
In certain cases, the range of application may be changed by variations in the
working conditions (e.g. sample volume, dilution, separating columns, pre-
concentration techniques, sensitivity ranges of detectors, etc.).
A. Drolc (&)
National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
e-mail: andreja.drolc@ki.si
1.1.2 Interferences
Some organic acids (malonic acid, maleic acid, malic acid) may interfere in the
determination of inorganic anions if they are present in high concentration. Sulfite
may cause positive bias for sulphate due to autooxidation. In this case the sample
may be adjusted to pH 10 and formaldehyde solution is added in order to stabilize
sulfite, if necessary.
In the presence of metals possibly interfering with the anions of interest, should
be checked and eliminated with the aid of special cation exchangers.
Cross sensitivity (lack of resolution) may occur in the case of large differences in
concentration between the anions determined
1.2 Principle
1.3 Reagents
1.3.1 Eluents
Different eluents are used, their choice depending on the type of separating column
and detectors (follow column manufacturer’s instructions).
1.3.2 Water
Deionised water with electrical conductivity of <0.1 lS/cm and shall be free of
particulate matter with diameter >0.45 lm.
Measurement of Concentration of Nitrate in Drinking Water 23
1.4 Apparatus
where
tR1 is the retention time, in seconds, of the 1st peak
tR2 is the retention time, in seconds, of the 2nd peak
W1 is the peak width, in seconds, of the 1st peak
W2 is the peak width, in seconds, of the 2nd peak.
Use vessels made of PTFE or polyethylene for sampling. After arrival of the sample
in the laboratory, filter it through a membrane filter (of pore size 0.45 lm). Stabilize
the sample by cooling it to between 4 and 6 °C or by deep freezing until the
analysis is performed.
24 A. Drolc
1.7 Procedure
1.8 Calculation
When the analytical system is first evaluated, and at intervals afterwards, establish a
calibration function for the measurement according to ISO 8466-1 (Water Quality –
Calibration and evaluation of analytical methods and estimation of performance
characteristics).
Estimate the mass concentration, in milligrams per litre, of the nitrate in the
solution using the peak areas or peak heights as follows:
A B0
C¼
B1
where A, B1 and B0 are measured area of the sample chromatographic peak, slope of
the linear least square calibration curve and calculated blank.
Slope of the linear least square calibration curve B1 and calculated blank B0 are
calculated from equations:
P
n
ðAi AÞ
ðCi CÞ
i¼1
B1 ¼ P
n
2
ðCi CÞ
i¼1
B1 C
B0 ¼ A
X
n
¼1
A Ai
n i¼1
X
n
¼1
C Ci
n i¼1
Measurement of Concentration of Nitrate in Drinking Water 25
where Ci and Ai are concentration of reference solution on ith level (C1, …, Ci, …, Cn)
and areas of chromatographic peaks of ith reference solution (A1, …, Ai, …, An),
respectively.
Take into account all the dilution steps.
Report the results to a minimum of three significant figures.
1
98/83/EC, of 3 November 1998, on the quality of water intended for human consumption.
26 A. Drolc
5.41
5.31
5.37
5.41
B0 14,967
B1 112,837
Abl 5104
5216
5072
Cobserved 4.56 mg L−1
Cmatrix 2.32 mg L−1
Cspiked 2.22 mg L−1
Equations
P
n
xi
i¼1
x ¼
n
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
ðxi xÞ2
s¼
n1
ALOD b0
LOD ¼
b1
cobserved cmatrix
R¼
cspiked
2
Here you may also want to include the relevant certificates.
Measurement of Concentration of Nitrate in Drinking Water 27
pre-analytical phase
Collection of
water sample
from a tap
Preparation and
storage of sample
analytical phase
Separation:
liquid
chromatography
Preparation of
calibration
Conductometric Conductometric
detection detection
Result
28 A. Drolc
Equations
A B0
C¼
B1
P
n
ðAi AÞ
ðCi CÞ
i¼1
B1 ¼ P
n
2
ðCi CÞ
i¼1
X
n
¼1
A B1 C
Ai B0 ¼ A
n i¼1
X
n
¼1
C Ci
n i¼1
A B0 cobserved cmatrix
CNO3 ¼ R¼
B1 R cspiked
Measurement of Concentration of Nitrate in Drinking Water 29
Model equation
The concentration of nitrate, C, is calculated by:
A B0
C¼
B1
where A, B1 and B0 are measured area of the sample chromatographic peak, slope of
the linear least square calibration curve and calculated blank. Slope of the linear
least square calibration curve B1 and calculated blank B0 are calculated from
equations:
P
n
ðAi AÞ
ðCi CÞ
i¼1
B1 ¼
P
n
2
ðCi CÞ
i¼1
B1 C
B0 ¼ A
X
n
¼1
A Ai
n i¼1
X
n
¼1
C Ci
n i¼1
where Ci and Ai are concentration of reference solution on ith level (C1, …, Ci, …, Cn)
and areas of chromatographic peaks of ith reference solution (A1, …, Ai, …, An),
respectively.
Measurement of Concentration of Nitrate in Drinking Water 31
4. List the reference standards needed and state the information regarding
traceability of the reference value
For the analyte
Selectivity, interferences
6. Measuring range
Linearity
Upper limit
LOD
LOQ
7. Spread – Precision
Repeatability
Reproducibility (within Lab)
Reproducibility (between Lab)
8. Robustness
Variation of parameters
9. Quality Control
Control charts
Participation in PT schemes
34 A. Drolc
12. Does the analytical procedure fulfil the requirement(s) for the intended
use?
Parameter Value requested Value obtained The requirement
by the customer during validation is fulfilled
(the same as stated Yes/No
in question 3)
LOD 5 mg L−1 0.5 mg L−1
LOQ – –
Repeatability – –
−1
Within-lab 5 mg L 0.06 mg L−1
reproducibility
Trueness 5 mg L−1 0 (R = 1)
Measurement – –
uncertainty
Other – –
Model equation:
A B0
CNO3 ¼
B1 R
where A, B1, B0 and R are measured area of the sample chromatographic peak,
slope of the linear least square calibration curve, calculated blank and method
recovery, respectively. Slope of the linear least square calibration curve B1 and
calculated blank B0 are calculated from eqns:
P
n
ðAi AÞ
ðCi CÞ
B1 ¼ i¼1 B1 C
B0 ¼ A
P
n
2
ðCi CÞ
i¼1
X
n X
n
¼1
A Ai ¼1
C Ci
n i¼1
n i¼1
cobserved cmatrix
R¼
cspiked
where Ci and Ai are concentration of reference solution on ith level (C1, …, Ci, …, Cn)
and areas of chromatographic peaks of ith reference solution (A1, …, Ai, …, An),
respectively.
3. Identify (all possible) sources of uncertainty
Uncertainty of concentration of reference solutions
Uncertainty of measurements of peak area
(continued)
Measurement of Concentration of Nitrate in Drinking Water 37
(continued)
Method bias
Matrix effect
Other:
Other:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Þ
uðB1 C uðB1 Þ 2 uðCÞ 2 831 2 0:016 2
¼ þ ¼ þ
B1 C B1 C 112837 3:4770
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 5:4237 105 þ 2:1175 105 ¼ 7:5412 105
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 112837 3:4770
uðB1 CÞ 7:5412 105 ¼ 3407
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðB0 Þ ¼ uðAÞ 2 þ uðB1 CÞ 2 ¼ 1462 þ 34072 ¼ 3410
A B0
CNO3 ¼ CNO3 ¼ 5:4 mg L1
B1 R
38 A. Drolc
7. Calculate the combined standard uncertainty (uc) of the result & specify
units
Using: Mathematical solution; Spreadsheet Approach; Commercial
Software
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðcÞ uðB1 Þ 2 uðRÞ 2 uðA B0 Þ 2
¼ þ þ
c B1 R A B0
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2 2
831 0:01 5425:54
¼ þ þ
112837 1 594350 þ 14967
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
¼ 23:35235 105
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðA B0 Þ ¼ uðAÞ2 þ uðB0 Þ2 ¼ 42202 þ 34102 ¼ 5425:54
8. Calculate expanded uncertainty (Uc) & specify the coverage factor k and
the units
Uc ¼ 0:2 mg L1 ðk ¼ 2Þ
9. Analyse the uncertainty contribution & specify the main three input
quantities contributing the most to Uc
1 Method recovery
2 Calibration curve
3 Measurement of the peak area of the sample
A1= 51437.0 51513 51437 51437 51437 51437 51437 51437 51437
A2= 98583.0 98583 98787 98583 98583 98583 98583 98583 98583
A3= 200037.0 200037 200037 200188 200037 200037 200037 200037 200037
A4= 305822.0 305822 305822 305822 306134 305822 305822 305822 305822
A5= 418793.0 418793 418793 418793 418793 419108 418793 418793 418793
A6= 531318.0 531318 531318 531318 531318 531318 532030 531318 531318
A7= 645226.0 645226 645226 645226 645226 645226 645226 645942 645226
A8= 767727.0 767727 767727 767727 767727 767727 767727 767727 768022
A-average= 377367.875 377377.375 377393.375 377386.75 377406.875 377407.25 377456.875 377457.375 377404.75
9.5 25.5 18.875 39 39.375 89 89.5 36.875
u(A-average)= 146.4894941 90.25 650.25 356.265625 1521 1550.390625 7921 8010.25 1359.765625
Measurement of Concentration of Nitrate in Drinking Water
39
40
C1= 0.488 0.4899 0.488 0.488 0.488 0.488 0.488 0.488 0.488
C2= 0.976 0.976 0.9795 0.976 0.976 0.976 0.976 0.976 0.976
C3= 1.95 1.952 1.952 1.959 1.952 1.952 1.952 1.952 1.952
C4= 2.93 2.928 2.928 2.928 2.939 2.928 2.928 2.928 2.928
C5= 3.904 3.904 3.904 3.904 3.904 3.918 3.904 3.904 3.904
C6= 4.88 4.88 4.88 4.88 4.88 4.88 4.899 4.88 4.88
C7= 5.856 5.856 5.856 5.856 5.856 5.856 5.856 5.877 5.856
C8= 6.832 6.832 6.832 6.832 6.832 6.832 6.832 6.832 6.857
C-average= 3.4770 3.4772375 3.4774375 3.477875 3.478375 3.47875 3.479375 3.479625 3.480125
0.0002375 0.0004375 0.000875 0.001375 0.00175 0.002375 0.002625 0.003125
u(C-average)= 0.005316337 5.64062E-08 1.91406E-07 7.65625E-07 1.89062E-06 3.0625E-06 5.64062E-06 6.89062E-06 9.76562E-06
A. Drolc
STEP 3: Calculation of B1
C1= A1= C2= A2= C3= A3= C4= A4= C5=
Values 0.488 51437 0.976 98583 1.952 200037 2.928 305822 3.904
Uncertainties 0.044 76 0.043 204 0.044 151 0.044 312 0.044
C1= 0.488 0.532 0.488 0.488 0.488 0.488 0.488 0.488 0.488 0.488
A1= 51437 51437 51513 51437 51437 51437 51437 51437 51437 51437
C2= 0.976 0.976 0.976 1.019 0.976 0.976 0.976 0.976 0.976 0.976
A2= 98583 98583 98583 98583 98787 98583 98583 98583 98583 98583
C3= 1.952 1.952 1.952 1.952 1.952 1.996 1.952 1.952 1.952 1.952
A3= 200037 200037 200037 200037 200037 200037 200188 200037 200037 200037
C4= 2.928 2.928 2.928 2.928 2.928 2.928 2.928 2.972 2.928 2.928
A4= 305822 305822 305822 305822 305822 305822 305822 305822 306134 305822
C5= 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.948
A5= 418793 418793 418793 418793 418793 418793 418793 418793 418793 418793
C6= 4.88 4.88 4.88 4.88 4.88 4.88 4.88 4.88 4.88 4.88
A6= 531318 531318 531318 531318 531318 531318 531318 531318 531318 531318
C7= 5.856 5.856 5.856 5.856 5.856 5.856 5.856 5.856 5.856 5.856
Measurement of Concentration of Nitrate in Drinking Water
A7= 645226 645226 645226 645226 645226 645226 645226 645226 645226 645226
C8= 6.832 6.832 6.832 6.832 6.832 6.832 6.832 6.832 6.832 6.832
A8= 767727 767727 767727 767727 767727 767727 767727 767727 767727 767727
C-average= 3.477 3.477 3.477 3.477 3.477 3.477 3.477 3.477 3.477 3.477
A-average= 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88
B1= 112837.2713 113250.1549 112831.1121 1.132E+05 1.128E+05 1.130E+05 1.128E+05 1.129E+05 1.128E+05 1.128E+05
412.8836783 -6.159118274 329.2488931 -13.83320764 193.7876004 -6.243467101 56.59986455 -4.644147184 -71.38606829
u(B1)= 831.0006894 170472.9318 37.93473791 108404.8336 191.3576335 37553.63408 38.98088144 3203.544667 21.56810306 5095.970746
% u(B1)= 0.736459399
41
42
C1= 0.488 0.488 0.488 0.488 0.488 0.488 0.488 0.488 0.488 0.488
A1= 51437 51437 51437 51437 51437 51437 51437 51437 51437 51437
C2= 0.976 0.976 0.976 0.976 0.976 0.976 0.976 0.976 0.976 0.976
A2= 98583 98583 98583 98583 98583 98583 98583 98583 98583 98583
C3= 1.952 1.952 1.952 1.952 1.952 1.952 1.952 1.952 1.952 1.952
A3= 200037 200037 200037 200037 200037 200037 200037 200037 200037 200037
C4= 2.928 2.928 2.928 2.928 2.928 2.928 2.928 2.928 2.928 2.928
A4= 305822 305822 305822 305822 305822 305822 305822 305822 305822 305822
C5= 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.904 3.904
A5= 418793 419108 418793 418793 418793 418793 418793 418793 418793 418793
C6= 4.88 4.88 4.924 4.88 4.88 4.88 4.88 4.88 4.88 4.88
A6= 531318 531318 531318 532030 531318 531318 531318 531318 531318 531318
C7= 5.856 5.856 5.856 5.856 5.901 5.856 5.856 5.856 5.856 5.856
A7= 645226 645226 645226 645226 645226 645942 645226 645226 645226 645226
C8= 6.832 6.832 6.832 6.832 6.832 6.832 6.878 6.832 6.832 6.832
A8= 767727 767727 767727 767727 767727 767727 767727 768022 767727 767727
C-average= 3.477 3.477 3.477 3.477 3.477 3.477 3.477 3.477 3.4926 3.477
A-average= 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377367.88 377513.88
B1= 112837.2713 1.128E+05 1.126E+05 1.129E+05 1.125E+05 1.129E+05 1.124E+05 1.129E+05 1.128E+05 1.128E+05
3.646846346 -199.3082064 27.08424298 -332.4772175 46.18346366 -460.0447772 26.83450429 -5.955889584 0
u(B1)= 831.0006894 13.29948827 39723.76115 733.5562177 110541.1001 2132.912316 211641.1971 720.0906207 35.47262073 0
% u(B1)= 0.736459399
A. Drolc
Measurement of Concentration of Nitrate in Drinking Water 43
STEP 4: Calculation of B0
A-average Caverage B1
Values 377367.8 3.477 112837.270
Uncertainties 146.48 0.005316337 831
B0
B1
Fdil
R
44 A. Drolc
1.1 Principle
_
I2 + N ⋅SO2 + 2 N + H2O → 2 N H+ I + N ⋅SO3
ð1Þ
_
N ⋅SO3 + ROH → N H + ROSO3 ð2Þ
The reaction is very fast and with strict stoichiometry. Solution of iodine, SO2
and pyridine dissolved the alcohol ROH is the titrant solution. In the classical KF
titrant the alcohol ROH is methanol. In modern commercial titrants ROH is often
methoxyethanol and pyridine is often replaced by imidazole (both because of
potential toxicity). Titration is carried out in ROH or in a mixture of ROH and some
other solvent (if samples are not soluble in ROH).
The end point of the titration is indicated by a small amount of unreacted iodine
in solution. End-point is usually determined voltammetrically: alternating current of
constant strength is applied to a double Pt electrode. Potential difference between
the Pt wires is monitored. Even small quantities of iodine lead to a dramatic drop of
the potential difference.
1.2 Scope
1.3 Procedure
1.4 Interferences
Many compounds (strong oxidizing and reducing agents, aldehydes, etc.) can
interfere with KF reaction. However, no such compounds are present in edible oils.
Thus within the scope of application there are no interferences.
Measurement of Moisture Content (Water Content) in Edible Oil … 47
The oil sample is shaken to ensure its homogeneity and a suitable amount is aspired
into an air-tight plastic syringe.
1.7 Calculation
Vt Tt
cwater ¼ 1000 ð3Þ
ms
1.8 Results
The result is presented with number of decimal digits corresponding to the obtained
uncertainty.
48 I. Leito and L. Jalukse
In this case the customer is a producer of refined rapeseed oil. The measurement is
needed to test if the moisture content in the oil under question is below or above the
maximum permissible limit, which is 500 mg/kg. This permissible limit has been
set by the quality standard of the producer.
3.1 Equations
The data are obtained on different days with different samples. From these data the
pooled repeatability standard deviation sr can be calculated. It can be seen that there
is no correlation between the water content and standard deviation. Thus sr can be
assumed to be reasonably constant over the whole concentration range (Table 1).
The equation to use for the ISO GUM modelling approach is similar to the equation
used for calculating the result. There is only one difference: the repeatability
components of Vt and ms are not included in the uncertainties of these quantities but
are taken into account separately by the repeatability factor fr. The reason is the
following: the amount of injected sample is measured gravimetrically and each time
the sample amount is slightly different.
The equation on which the uncertainty calculation is based is the following:
Vt Tt
cwater ¼ fr 1000 ð5Þ
ms
Table 2 presents the data relevant for the particular measurement for which the
uncertainty is calculated.
50 I. Leito and L. Jalukse
The notations are the same is used in the Nordtest handbook for uncertainty
evaluation.1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uc ¼ uðRw Þ2 þ uðbiasÞ2 ð8Þ
u(Rw) is the uncertainty component that takes into account precision, within lab
reproducibility.
u(bias) is the uncertainty component that takes into account the trueness, i.e. the
(possible) lab bias. The trueness estimate in turn is composed of two components:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðbiasÞ ¼ RMS2bias þ uðCref Þ2 ð9Þ
1
Handbook for calculation of measurement uncertainty in environmental laboratories (2017)
Nordtest TR 537, 4th ed. Espoo, Finland (available from the Internet: http://www.nordtest.info/).
Measurement of Moisture Content (Water Content) in Edible Oil … 51
biasi is the bias found on i-th bias determination (difference between lab result and
reference value‚ Table 3).
If interlaboratory comparisons are used then u(Cref) is found as follows:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P 2
uðCref Þi
uðCref Þ ¼ ð11Þ
n
In this case the pooled standard deviation sr (data from Table 1) can be used as an
estimate of u(Rw).
Interlaboratory comparison data (Table 2) are used for finding RMSbias and u
(Cref).
52 I. Leito and L. Jalukse
Model equation
Calculation of water content in sample is carried out according to the following
equation (mathematical model):
Vt Tt
cwater ¼ fr 1000 ð13Þ
ms
4. List the reference standards needed and give also the information
regarding traceability of the reference value
For the analyte
6. Measuring range
Linearity
Upper limit
LOD
LOQ
7. Spread – Precision
Repeatability
Reproducibility (within Lab)
Reproducibility (between Lab)
8. Robustness
Variation of parameters
The following parameters should be varied:
- Stirring speed;
- Titrant flow rate (i.e. rate of piston movement in the burette);
- Age of titrant.
56 I. Leito and L. Jalukse
9. Quality Control
Control charts
Participation in PT schemes
12. The method fulfils the requirement for the intended use:
Parameter Value requested by the customer Value obtained during validation
LOD
LOQ 100 mg kg−1 80 mg kg−1
−1
Repeatability 8 mg kg 5.5 mg kg−1
Within-lab reproducibility 16 mg kg−1 11 mg kg−1
Measurement uncertainty uc 20 mg kg−1 ISO GUM Modeling: 7 mg kg−1
Nordtest: 15 mg kg−1
Trueness
Other
Yes No
13. Calculation of other parameters
_
I2 + N ⋅SO2 + 2 N + H2O → 2 N H+ I + N ⋅SO3
ð15Þ
_
N ⋅SO3 + ROH → N H + ROSO3 ð16Þ
The reaction is very fast and with strict stoichiometry. Solution of iodine, SO2
and pyridine dissolved the alcohol ROH is the titrant solution. In the classical KF
titrant the alcohol ROH is methanol. In modern commercial titrants ROH is often
methoxyethanol and pyridine is often replaced by imidazole (both because of
potential toxicity). Titration is carried out in ROH or in a mixture of ROH and some
other solvent (if samples are not soluble in ROH).
58 I. Leito and L. Jalukse
The end point of the titration is indicated by a small amount of unreacted iodine
in solution. End-point is usually determined voltammetrically: alternating current of
constant strength is applied to a double Pt electrode. Potential difference between
the Pt wires is monitored. Even small quantities of iodine lead to a dramatic drop of
the potential difference.
Model equation:
See Eq. 13.
3. Identify (all possible) sources of uncertainty
Uncertainty of concentration of the titrant
Uncertainty of measurements of the titrant volume
Method bias (taken into account within the other input quantities)
Matrix effect
Other: Sample mass
Other: Overall repeatability
Other:
Repeatability
fr 1.000 0.015
V⋅T
cwater = t t ⋅ f r ⋅ 1000
ms
Finding combined uncertainty:
2 2 2 2
⎛ u(V ) ⎞ ⎛ u(T ) ⎞ ⎛ u(ms ) ⎞ ⎛ u( fr ) ⎞
uc (cw ater) = c w ater ⎜⎜ t ⎟⎟ + ⎜⎜ t ⎟⎟ + ⎜⎜ ⎟⎟ + ⎜⎜ ⎟⎟
Result:
⎝ Vt ⎠ ⎝ Tt ⎠ ⎝ ms ⎠ ⎝ f r ⎠
u' (m s ) u' (V t ) u' (T t ) u' (f r)
c water 368.2510 % 0.00002 0.0042 0.0100 0.0150 6.8332
Tidža Muhić-Šarac
According to the EU directive 95/2/EC and regulations from the Statute on Quality
of Meat Products (B&H), polyphosphates can be used in production of all kinds of
sausages (except those with long shelf life) and in production of canned meat and
other products which are thermally processed. The main condition for their usage is
that the quantity in the final product does not exceed w = 0.5% or w = 0.3%
(m m−1) in relation to the quantity of the product, expressed as P2O5 content.
Polyphosphates (condensed phosphates) can be obtained by heating of primary or
secondary orthophosphates, which dehydrate and get condensed at the same time.
Only cyclical polyphosphates are used in meat processing and most frequently in the
form of: disodium phosphate (Na2HPO4; E339), mono-sodium phosphate (NaH2PO4;
E339), sodium-meta-phosphates (Na3P3O9; E452), tetrasodium-pyrophosphate
(Na4P2O7; E450) and acid disodium pyrophosphate (Na2H2P2O7; E450).
They function in the direction of protein hydration and puffing up of muscle
tissue, and better homogeneity, to achieve a firm consistence and stability of the
meat products. They also help in improvement of taste, and mistakes in production
are reduced to a minimum. From the point of health the possibility of harmful
T. Muhić-Šarac (&)
Faculty of Science, Department of Chemistry, University of Sarajevo, Zmaja od Bosne 33-35,
71000 Sarajevo, Bosnia and Herzegovina
e-mail: tidzamuhic@gmail.com
3 Na þ þ 12 MoO2
4 þ H3 PO4 þ 21 H
þ
¼ ðNaÞ3 PO4 ðMoO3 Þ12 þ 12 H2 O
1. Weigh approximately 5.0 g ± 0.1 mg (m1) of the sample (meat product) and
transfer it on ceramic hot plate.
2. The digestion of organic issue (mineralization of samples) was done by calci-
nations with long lasting (3 h) and gradual heating in the furnace, starting from
room to t = 530 °C temperature.
3. The dissolution of ash, which consists of oxides P2O5, K2O, CaO, MgO, Fe2O3,
and SiO2, chlorides and sulfates was done with 6 mol L−1 HCl and quantita-
tively taken over into a volumetric flask 100 mL (V1). This is stock solution of
sample (meat product).
Model Equation
where:
C Concentration of PO−3 4 in working standard solution mg L
−1
1.3.3 Measurement
1. Transfer 1.0 mL (V2) from the stock solution samples (meat products) to a
200 mL (V3) volumetric flask, and add distilled water till the mark.
2. Transfer 10.0 mL (V4) from the above solution into the spectrophotometric
cuvette. Add the “Reagent pillows” for developing color. Allow this solution to
stay at least two minutes to color developing.
3. Parallel prepare the blank probe in the way that you add the reagent pillow to
10 mL blank probe, without meat product.
4. Measuring of the optical density of the obtained blue-colored complex which is
in a linear dependence on the concentration of the phosphate was done by a
UV-VIS spectrophotometer product of the HACH, USA. type DR/4000U
according to program: 3025, for concentration area within 0–2.5 mg L−1 PO−3 4
in a 2 cm cuvette and with k = 890 nm wavelength.
5. Read the content of phosphate in mg L−1.
6. Calculate the content of phosphorous-pentoxide given by the formula.
Measurement of Polyphosphates in Meat Products … 67
msample ðgÞ R
where:
0.75 Concentration factor M(P2O5)/2M(PO−3
4 )
where:
wP2O5 Mass fraction wP2O5 in %
C1 Concentration of PO−3 −1
4 (mg L ) in meat product
V1 Volumetric flask 100 mL (±0.10 mL)
V2 Pipette 1 mL (±0.01 mL)
V3 Volumetric flask 200 mL (±0.15 mL)
V4 Pipette 10 mL (±0.1 mL)
m1 Mass of sample (g) (±0.1 mg)
F Concentration factor M(P2O5)/2M(PO−34 ) = 0.75
100 Factor for conversion (mg ! %)
10001 Factor for conversion (mg L−1 ! mg mL−1)
10 Factor of calculation
10002 Factor for conversion (g ! mg)
R Recovery
Rrep Repeatability
Icheck Intermediate checks of spectrophotometer-in one point
1 0.008
2 0.007
3 0.008
4 0.007
5 0.008
6 0.007
7 0.008
8 0.007
Mean
Sb
Signal = b0 + b1 c
Ybl = Signal of the ‘blank’;
Sbl = stdev of the ‘blank’ in signal domain
procedure, same location, and replicate measurements on the same objects over a
short period of time, involving changes operators (Analyst)
The linearity of method was evaluate for concentration rang of phosphates from
0.30 to 2.50 mg L−1 PO−34 . The using standard working solution is KH2PO4,
concentration 0.0015 mg mL−1 PO−34 .
Linearity of method
1,6
1,4 y = 0,5727x
R2 = 0,999
1,2
Absorbance
0,8
0,6
0,4
0,2
0
0 0,5 1 1,5 2 2,5 3
3-
Concentration of PO4 (mg/L)
Measurement of Polyphosphates in Meat Products … 71
3.6 Recovery
The Recovery was evaluate for seven concentration level with additional method.
Standard working solution is added in meat samples (matrix).
cobserved cmatrix
R¼
cspike
See Scheme 1.
1
Here you may also want to include the relevant certificates.
72 T. Muhić-Šarac
VEIGH KH2PO4
(m1)
B -TYPE
DILUTE (V1)
(mass/1000mL) A/B -TYPE
DILUTE (V3)
(1mL/200mL) B/A/B -TYPE
INTERMEDIATE A -TYPE
CHECKS
where:
C Concentration of PO−3 4 in working standard solution mg L
−1
A -TYPE
SAMPLE
A -TYPE
HOMOGENISE
B - TYPE
WEIGH (m1)
TERMIC
DESTRUCTION A - TYPE
DILUTE (V1)
B/A/B-TYPE
(ash/100 mL)
DILUTE (V3)
(1mL/200mL) B/A/B -TYPE
A/B - TYPE TTYPE
MEASUREMENT
A - TYPE
msample ðgÞ R
where:
0.75 Concentration factor M(P2O5)/2M(PO−3
4 )
where:
wP2O5 Mass fraction wP2O5 in %
C1 Concentration of PO−3 −1
4 (mg L ) in meat product
V1 Volumetric flask 100 mL (±0.10 mL)
V2 Pipette 1 mL (±0.01 mL)
V3 Volumetric flask 200 mL (±0.15 mL)
V4 Pipette 10 mL (±0.1 mL)
m1 Mass of sample (g) (±0.1 mg)
F Concentration factor M(P2O5)/2M(PO−34 ) = 0.75
100 Factor for conversion (mg ! %)
10001 Factor for conversion (mg L−1 ! mg mL−1)
10 Factor of calculation
10002 Factor for conversion (g ! mg)
R Recovery
Rrep Repeatability
Icheck Intermediate checks of spectrophotometer-in one point
See Table 2.
76 T. Muhić-Šarac
Model equation
Intermediate checks of spectrophotometer-in one point
Phosphate concentration in working standard solution:
1
mðKH2 PO4 Þ P 10 1000 V2
cPO3 mg L ¼
4 V4 V3 V1 F
where:
C Concentration of PO−3 4 in working standard solution mg L
−1
where
wP2 O5 Mass fraction wP2O5 in % (m m−1)
C1 Concentration of PO−3 −1
4 (mg L ) in meat product
V1 Volumetric flask 100 mL (±0.10 mL)
78 T. Muhić-Šarac
4. List the reference standards needed and state the information regarding
traceability of the reference value
For the analyte
6. Measuring range
Linearity
Upper limit
LOD
LOQ
7. Spread – Precision
Repeatability
Reproducibility (within Lab)
Reproducibility (between Lab)
8. Robustness
Variation of parameters
9. Quality Control
Control charts
Participation in PT schemes
82 T. Muhić-Šarac
12. Does the analytical procedure fulfil the requirement(s) for the intended
use?
Parameter Value requested by Value obtained during The
the customer validation requirement is
(the same as stated fulfilled
in question 3) Yes/No
LOD Not defined 0.0183 mg L−1 PO−3
4 Yes
0.0055% (m m−1) P2O5
LOQ Not defined 0.0305 mg L−1 PO−3
4 Yes
0.0091% (m m−1) P2O5
Repeatability Not defined ±0.015% (m m−1) P2O5 Yes
Within-lab Not defined 0.0161% (m m−1) P2O5 Yes
reproducibility
Trueness Not defined
Measurement Not defined ±0.016% (m m−1) P2O5 Yes
uncertainty
Recovery Not defined 1.016 ± 0.0117 Yes
Yes No
Note:
This new analytical method is validated in one laboratory.
Requested by the customer is not defined, except the maximum limit in the content
of phosphate (% m m−1 P2O5) in meat products.
Evaluation of the obtained values for the elements of validation may be given to the
statement: “The analytical procedure is fit for the intended use.”
Low LOD, LOQ. Satisfactory: linearity of measurement in the measuring range of
interest, repeatability and within-lab. reproducibility, and the value of Recovery
test.
For Measurement Uncertainty and Traceability refer to the corresponding sheets
3. Dissolve the resulting ash with 6 mol L−1 HCl and quantitatively transfer it to
the volumetric flask of 100 mL (V1). This is stock solution of sample (meat
product).
4. Transfer 1.0 mL (V2) from the stock solution sample (meat products) to a
200 mL (V3) volumetric flask, and add distilled water till the mark.
5. Transfer 10.0 mL (V4) from the above solution into the spectrophotometric
cuvette. Add the “Reagent pillows” for developing color. Allow this solution to
stay at least two minutes to color developing.
6. Parallel prepare the blank probe in the way that you add the reagent pillow to
10 mL blank probe, without meat product.
7. Measure the concentration of phosphate (C1 in mg L−1) in the UV-VIS spec-
trophotometer by 3025 program and wavelength of k = 890 nm.
8. Calculate the content of phosphorous-pentoxide given by the formula.
Model equation:
(a) Mass fraction wP2 O5 in % (m m−1)
c1PO3 ðmg L1 Þ 0:75 2
WP2 O5 ð%Þ ¼ 4
msample ðgÞ R
where:
0.75 Concentration factor M(P2O5)/2M(PO−3
4 )
where:
wP2O5 Mass fraction wP2O5 in % (m m−1)
C1 Concentration of PO−3 −1
4 (mg L ) in meat product
V1 Volumetric flask 100 mL (±0.10 mL)
V2 Pipette 1 mL (±0.01 mL)
V3 Volumetric flask 200 mL (±0.15 mL)
V4 Pipette 10 mL (±0.1 mL)
m1 Mass of sample (g) (±0.1 mg)
F Concentration factor M(P2O5)/2M(PO−34 )= 0.75
100 Factor for conversion (mg ! %)
10001 Factor for conversion (mg L−1 ! mg mL−1)
10 Factor of calculation
86 T. Muhić-Šarac
Analytical procedure
Analytical procedure
Analytical procedure
Comprehensive equation for mass fraction wP2 O5 in %:
7. Calculate the combined standard uncertainty (uc) of the result & specify
units
Using: Mathematical solution; Spreadsheet Approach; Commercial
Software
Intermediate checks of spectrophotometer-in one point
Analytical procedure
8. Calculate expanded uncertainty (Uc) & specify the coverage factor k and
the units
9. Analyse the uncertainty contribution & specify the main three input
quantities contributing the most to Uc
1 Repeatability – contributing 47.52% to the expanded uncertainty
2 Recovery – contributing 27.37% to the expanded uncertainty
3 Uncertainty of intermediate checks of spectrophotometer – contributing 17.11% to the
expanded uncertainty
%u(xi)e2
V4
P
F
m( KH2 PO4 ) P 10 1000 V2
m1
cPO 3 (mg / L)
4
V4 V3 V1 F
V1
V3
T. Muhić-Šarac
V2
Analytical procedure
mg m1 5000 5000,085 5000 5000 5000 5000 5000 5000 5000 5000 5000
mg/L c1 1,22 1,22 1,222 1,22 1,22 1,22 1,22 1,22 1,22 1,22 1,22
mL V1 100 100 100 100,068 100 100 100 100 100 100 100
mL V2 1 1 1 1 1,004 1 1 1 1 1 1
mL V3 200 200 200 200 200 200,117 200 200 200 200 200
mL V4 10 10 10 10 10 10 10,043 10 10 10 10
F 0,75 0,75 0,75 0,75 0,75 0,75 0,75 0,75001 0,75 0,75 0,75
R 1,016 1,016 1,016 1,016 1,016 1,016 1,016 1,016 1,0277 1,016 1,016
% Rrep. 1 1 1 1 1 1 1 1 1 1,015 1
Icheck 1 1 1 1 1 1 1 1 1 1 1,009
0,360236 0,360236 0,360236 0,360236 0,360236 0,360236 0,360236 0,360236 0,360236 0,360236
%(m/m) P 2O 5 0,360236 0,360230 0,360827 0,360481 0,358801 0,360447 0,361785 0,360241 0,356135 0,36564 0,363478
Measurement of Polyphosphates in Meat Products …
u(y,xi) -6,1E-06 0,000591 0,000245 -0,00144 0,000211 0,001549 4,8E-06 -0,0041 0,005404 0,003242
u(y)2 6,14E-05 3,75E-11 3,49E-07 6E-08 2,06E-06 4,44E-08 2,4E-06 2,31E-11 1,68E-05 2,92E-05 1,05E-05
%u(xi)2 6,1E-05 0,567613 0,097663 3,352465 0,072281 3,905248 3,75E-05 27,37478 47,52195 17,1079
u(%P2O5) 0,007838 6,12E-06 5,91E-04 2,45E-04 1,44E-03 2,11E-04 1,55E-03 4,80E-06 4,10E-03 5,40E-03 3,24E-03
%u(xi) 2
m1 V1 V2 V4
Icheck c1 V3 F c1 V4 V3 V1 100 F Rrep. I check
R
WP2O5 (%)
10001 10 V2 m1 10002 R
Rrep.
91
92 T. Muhić-Šarac
The two contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 1000 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV1 Þtotal ¼ 0:162 þ 0:52 ¼ 0:507 mL
V2 – 1 mL pipette
The volume has three major influences; calibration, repeatability and temperature
effects.
ffiffi ¼ 0:004 mL
(i) Calibration u ¼ 0:01pmL
6
(ii) Repeatability u ¼ S ¼ 0.0013 mL
(iii) Temperature ð1 4 2:1 104 Þ ¼ 0:0008 mL
0:0008 mL
u¼ pffiffiffi ¼ 0:0005 mL
3
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 1 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV2 Þtotal ¼ 0:0042 þ 0:00132 þ 0:00052 ¼ 0.004 mL
ffiffi ¼ 0:06 mL
(i) Calibration u ¼ 0:15pmL
6
(ii) Repeatability u ¼ S ¼ 0.025 mL
(iii) Temperature ð200 4 2:1 104 Þ ¼ 0:168 mL
0:168 mL
u¼ pffiffiffi ¼ 0:097 mL
3
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 200 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV3 Þtotal ¼ 0:062 þ 0:0252 þ 0:0972 ¼ 0.117 mL
V4 – 10 mL pipette
The volume has three major influences; calibration, repeatability and temperature
effects.
ffiffi ¼ 0:04 mL
(i) Calibration u ¼ 0:1pmL
6
(ii) Repeatability u ¼ S ¼ 0.015 mL
(iii) Temperature ð10 4 2:1 104 Þ ¼ 0:0084 mL
0:0084 mL
u¼ pffiffiffi ¼ 0:005 mL
3
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 10 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV4 Þtotal ¼ 0:042 þ 0:0152 þ 0:0052 ¼ 0.043 mL
P – purity of KH2PO4
The purity of the KH2PO4 is given on the certifikate as 99.95 ± 0.05% m m−1.
Purity is therefore 0.9995 ± 0.0005. Because there is no additional information
about the uncertainty value, a rectangular distribution is assumed:
0:0005
u(P) ¼ pffiffiffi ¼ 0:0003
3
For each element, the standard uncertainty is found by the IUPAC quoted
uncertainty as forming the bounds of a rectangular distribution.
The separate element contributions to the molar mass, together with the uncer-
tainty contribution for each, are:
The separate element contributions to the molar mass, together with the uncer-
tainty contribution for each, are:
M PO3
4 ¼ 30:9738 þ 63:9976 ¼ 94:9714 g mol1 ;
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u M PO3
4 ¼ 0:000292 þ 0:000682 ¼ 0:0007 g mol1
F MPO3 4
MKH2 PO4
u(F)
¼ 0:000009
1:43
u(F) ¼ 1:43 0:000009 ¼ 0.000013
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 100 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV1 Þtotal ¼ 0:042 þ 0:022 þ 0:052 ¼ 0.068 mL
V2 – 1 mL pipette
The volume has three major influences; calibration, repeatability and temperature
effects
96 T. Muhić-Šarac
ffiffi ¼ 0:004 mL
(i) Calibration u ¼ 0:01pmL
6
(ii) Repeatability u ¼ S ¼ 0.0013 mL
(iii) Temperature ð1 4 2:1 104 Þ ¼ 0:0008 mL
0:0008 mL
u¼ pffiffiffi ¼ 0:0005 mL
3
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 1 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV2 Þtotal ¼ 0:0042 þ 0:00132 þ 0:00052 ¼ 0.004 mL
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Finally, total standard uncertainty of the 200 mL volumetric flask is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðV3 Þtotal ¼ 0:062 þ 0:0252 þ 0:0972 ¼ 0.117 mL
V4 – 10 mL pipette
The volume has three major influences; calibration, repeatability and temperature
effects.
ffiffi ¼ 0:04 mL
(i) Calibration u ¼ 0:1pmL
6
(ii) Repeatability u ¼ S ¼ 0.015 mL
(iii) Temperature ð10 4 2:1 104 Þ ¼ 0:0084 mL
0:0084 mL
u¼ pffiffiffi ¼ 0:005 mL
3
The three contributions are combined to give the standard uncertainty u(V) of
the V.
Measurement of Polyphosphates in Meat Products … 97
0:1 mg
u ¼ pffiffiffi ¼ 0:06 mg
3
For each element, the standard uncertainty is found by the IUPAC quoted
uncertainty as forming the bounds of a rectangular distribution.
The separate element contributions to the molar mass, together with the uncer-
tainty contribution for each, are:
MðPO3 1
4 Þ ¼ 30:9738 þ 63:9976 ¼ 94:9714 g mol ;
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u M PO3
4 ¼ 0:000292 þ 0:000682 ¼ 0.0007 g mol1
The separate element contributions to the molar mass, together with the uncer-
tainty contribution for each, are:
98 T. Muhić-Šarac
MðP2 O5 Þ
F¼ ¼ 0:75
2 M PO3 4
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 0 12
u
uð F Þ u u M PO4 3
uð M P 2 O 5 Þ 2
t
¼ 2x @ A þ
F MPO3 4
MP2 O5
u(F)
¼ 0:000012
0:75
u(F) ¼ 0:75 0:000012 ¼ 0:000009 g mol1 ¼ 0.00001 g mol1
1.1 Purpose
1.2 Principle
The method is comprised of two parts. In the first the solution of potassium
dichromate is standardized by using the certified reference material (CRM Ilmenite
or CRM Titanium Slag), and in the second part the mass fraction of TiO2 and Fe-tot
is determined in a sample of Ilmenite or Ti-Slag by titration. Determination of the
equivalence points is potentiometric using a gold indicator electrode and an Ag/
AgCl reference electrode.
The sample of Ilmenite or Ti-Slag is dried at 110 °C to constant mass. Dried
sample is fused with melted potassium pyrosulfate and the melt is dissolved in
hydrochloric acid. By addition of the excess of the solution of chromium(II)
chloride Ti4+ is reduced to Ti3+ and Fe3+ is reduced to Fe2+ in the inert atmosphere
of carbon dioxide.
With potassium dichromate is titrated the firstly exceeded chromium(II) chloride
(1st equivalence point), then Ti3+ (2nd equivalence point) and then Fe2+ (3rd
equivalence point).
1.3 Chemicals
1.4 Equipment
• Nakazon’s reductor
• Automatic titrator Mettler DL70ES connected to PC and printer (calibrated by
producer)
• Gold (Au) indicator electrode
• Reference Ag/AgCl electrode
• Glass titration flasks, 250 mL
• Burettes Mettler (calibrated by producer)
• Gas flow meter, 400 ± 20 mL/min
• Volumetric flask − class A quality
• Volumetric pipette − class A quality
• Fumehood
• Sample and certified reference material are dried at 110 ± 2 °C to the constant
mass (min 2 h).
• It is cooled in the exicator to the room temperature.
1.6 Procedure
1.6.1. Fusion – Dissolution
1.6.3. Determination
Training and checking competence of the personnel is done according to the cor-
responding internal procedure.
Quality is assured by:
1.10 Safety
The value of the relative expanded uncertainty is defined as the level of uncertainty
that is broadly accepted as appropriate by the end-users of the data for the sector of
application of the results. A suggested definition of this »fitness-for-purpose« is
that it should comprise the uncertainty that minimized the combined costs of
analysis and the financial penalties associated with incorrect decisions multiplied by
their probabilities of occurrence.
The accuracy of test procedure was checked and evaluated by measuring the
appropriate CRM:
• CRM SARM 59 – Ilmenite
• CRM SARM 57 – Titanium Slag (Table 1).
Table 1 Reproducibility of TiO2 and Fe-tot measurements on CRM (from control charts)
CRM SARM 59 Ilmenite CRM SARM 57 Titanium
Slag
TiO2 [%] Fe-tot [%] TiO2 [%] Fe-tot [%]
n* 30 30 30 30
Average 48.81 35.23 85.34 8.24
SD 0.069 0.060 0.068 0.053
RSD [%] 0.14 0.17 0.08 0.64
CV** 48.8 35.2 85.4 8.25
95% CL*** 48.6–49.1 34.9–35.6 85.1–85.7 8.0–8.4
Recovery [%] 100.0 100.1 99.9 99.9
*Number of measurements
**Certified Value
***95% Confidence Limits
Task for a Laboratory Measurement of Mass Fraction … 107
Measurements were tested with the Grubbs test and were found to have no
outliers.
Laboratory reproducibility of TiO2 and Fe-tot measurement does not differ
substantially from laboratory repeatability.
Standard deviations of measurement for different samples do not differ sub-
stantially, that is why we can calculate spooled:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðn1 1Þ s21 þ ðn2 1Þ s22 þ
spooled ¼
ðn1 1Þ þ ðn2 1Þ þ
The comparability of the TiO2 content results in Ilmenite and Ti-Slag was
checked with the inter laboratory comparison, organized by the Quality Control
department of Cinkarna Celje. 5 laboratories from Europe participated in this
inter-laboratory comparison (“Measurements of various parameters in some TiO2
based materials, March 2011”).
With the inter laboratory comparison we have confirmed the comparability of the
TiO2 content results in Ilmenite and Ti-Slag (z = −0.32 and z = 0.06) (Tables 6
and 7).
Table 6 Results of measurements of TiO2 content in Titanium Slag – inter laboratory comparison
Assigned value Xa 76.61%
Uncertainty on the assigned value u(Xa) 0.068%
Standard deviation of the proficiency assessment rpa 0.77%
Participant’s Result Standard Number of Method Reported z− f−
Code (Sample – X [%] deviation measurements standard score score
ID) [%] uncertainty [%]
CC-02/10 76.51 0.160 9 EN ISO 591-1 0.123 −0.13 −0.69
CC-02/11 76.20 0.046 3 UV-VIS 0.09 −0.53 −3.61
CC-02/12 76.70 0.086 3 Titration 0.07 0.11 0.88
CC-02/13 76.71 0.155 3 XRF 0.2 0.13 0.49
CC-02/14* 76.66 0.066 8 Potentiometric 0.1 0.06 0.38
titration
*Our result
Task for a Laboratory Measurement of Mass Fraction … 109
Ti-Slag
The repeatability of the entire process combines all the repeatability of indi-
vidual sources of uncertainty. Because we conduct all the analysis in two parallel
determinations, we have divided the standard deviations of measurements spooled
pffiffiffi
with 2 to obtain the standard uncertainty of the double determination.
Determination of mass fraction of TiO2 in Ilmenite [%]
4.1 Equations
Ilmenite
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s
uðWTiO2 Þ uðms Þ 2 uðH8 Þ 2 uðV100 Þ 2 uðVEX1 þ VEQ2 Þ 2 uðV500 Þ 2 uðPTiO2 Þ 2
¼ þ þ þ þ þ
WTiO2 ms H8 V100 ðVEX1 þ VEQ2 Þ V500 WTiO2
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uðWFe-tot Þ uðms Þ 2 uðH9 Þ 2 uðV100 Þ 2 uðVEX2 þ VEQ3 Þ 2 uðV500 Þ 2 uðPFe-tot Þ 2
¼ þ þ þ þ þ
WFe-tot ms H9 V100 ðVEX2 þ VEQ3 Þ V500 WFe-tot
Ti-Slag
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
uðWTiO2 Þ uðms Þ 2 uðH11 Þ 2 uðV100 Þ 2 uðVEX1 þ VEQ2 Þ 2 uðV500 Þ 2 uðPTiO2 Þ 2
¼ þ þ þ þ þ
WTiO2 ms H11 V100 ðVEX1 þ VEQ2 Þ V500 WTiO2
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
uðWFe-tot Þ uðms Þ 2 uðH12 Þ 2 uðV100 Þ 2 uðVEX2 þ VEQ3 Þ 2 uðV500 Þ 2 uðPFe-tot Þ 2
¼ þ þ þ þ þ
WFe-tot ms H12 V100 ðVEX2 þ VEQ3 Þ V500 WFe-tot
112 J. Pustinek et al.
Analyte TiO2
Fe-tot
Measurand Mass fraction of TiO2 in Ilmenite
Mass fraction of TiO2 in Titanium Slag
Mass fraction of Fe-tot in Ilmenite
Mass fraction of Fe-tot in Titanium Slag
Units %
Model equation
Mass fraction of TiO2 and Fe-tot is calculated by:
Ilmenite
Ti-Slag
4. List the reference standards needed and state the information regarding
traceability of the reference value
For the analyte
(continued)
3 Quantity/Equipment/calibration: Burettes Mettler – calibrated by producer
4 Quantity/Equipment/calibration: Volumetric flask – class A quality
5 Quantity/Equipment/calibration: Volumetric pipette – class A quality
6. Measuring range
Linearity
Upper limit
LOD
LOQ
7. Spread – Precision
Repeatability
Reproducibility (within Lab)
Reproducibility (between Lab)
Task for a Laboratory Measurement of Mass Fraction … 117
8. Robustness
Variation of parameters
9. Quality Control
Control charts
Participation in PT schemes
12. Does the analytical procedure fulfil the requirement(s) for the intended
use?
Parameter Value requested by the customer (the Value obtained The
same as stated in question 3) during validation requirement is
fulfilled
Yes/No
LOD
LOQ
Repeatability
Within-lab
reproducibility
Trueness
Ti-Slag
(continued)
Input Standard Unit Remark
quantity uncertainty
V500 0.2086 mL Volume of dissolved sample
ms 0.00015 g Mass of sample
H12 0.0000344 g Standardization of K2Cr2O7 with CRM
mL−1 Titanium Slag
VEX2 þ VEQ3 0.0163 mL Volume of K2Cr2O7 used for titration
Ilmenite – Fe-tot
7. Calculate the combined standard uncertainty (uc) of the result & specify
units
Using: Mathematical solution; Spreadsheet Approach; Commercial
Software
Ilmenite
TiO2 Fe-tot
Unit Value u urel Value u urel
P % – 0.068 0.00127 – 0.064 0.00209
V100 mL 100 0.0489 0.00049 100 0.0489 0.00049
V500 mL 500 0.2086 0.00042 500 0.2086 0.00042
ms g 0.5101 0.00015 0.00029 0.5101 0.00015 0.00029
H g/mL 0.009585 0.0000298 0.00311 0.006835 0.0000360 0.00527
VEX þ VEQ mL 5.7155 0.00836 0.00146 4.5732 0.00836 0.00183
W % 53.70 0.200 0.00373 30.64 0.184 0.00600
Titanium slag
TiO2 Fe-tot
Unit Value u urel Value u urel
P % – 0.066 0.00084 – 0.030 0.00476
V100 mL 100 0.0489 0.00049 100 0.0489 0.00049
V500 mL 500 0.2086 0.00042 500 0.2086 0.00042
ms g 0.5050 0.00015 0.00030 0.5050 0.00015 0.00030
H g/mL 0.004862 0.0000105 0.00216 0.003480 0.0000344 0.00989
VEX þ VEQ mL 16.2883 0.0174 0.00107 1.8276 0.0163 0.00892
W % 78.41 0.208 0.00265 6.30 0.089 0.0142
8. Calculate expanded uncertainty (Uc) & specify the coverage factor k and
the units
k=2 ILMENITE TITANIUM SLAG
TiO2 Fe-tot TiO2 Fe-tot
U* % 0.40 0.37 0.42 0.18
*U is expanded uncertainty calculated using a coverage factor of 2, which gives a level of
confidence of approximately 95%
124 J. Pustinek et al.
9. Analyse the uncertainty contribution & specify the main three input
quantities contributing the most to Uc
1 Standardization of K2Cr2O7 with CRM Ilmenite or CRM Titanium Slag – H
2 Volume of K2Cr2O7 used for titration – VEQ
3 Repeatability – P
Titanium Slag:
Further Readings
1. ISO 591-1: 2000 Titanium dioxide pigments for paints – Part 1: Specifications and methods
of test
2. J. D. Norris; Analyst, November 1984, Vol. 109: Determination of titanium in titanium
dioxide pigments, paints and other materials by chromium(II) chloride reduction and
automatic potentiometric titration
3. ISO 21748: 2010 Guidance for the use of repeatability, reproducibility and trueness estimates
in measurement uncertainty estimation
4. ISO TR 13587: 2012 Three statistical approaches for the assessment and interpretation of
measurement uncertainty
5. ISO 7870-2: 2013 Control charts – Part 2: Shewhart control charts
6. ISO 2854: 1976 Statistical interpretation of data – Techniques of estimation and tests relating
to means and variances
7. EURACHEM/CITAC Guide CG4: Quantifying Uncertainty in Analytical Measurement,
Third Edition, 2012
8. Eurolab Technical Report No. 1/2002: Measurement Uncertainty in Testing
9. NORDTEST Report TR 537: Handbook for Calculation of Measurement Uncertainty in
Environmental Laboratories, Edition 3.1, 2012
10. R. Bettencourt da Silva, E. Bulska, B. Godlewska-Żyłkiewicz, M. Hedrich, N. Majcen, B.
Magnusson, S. Marinčić, I. Papadakis, M. Patriarca, E. Vassileva, P. Taylor; Analytical
measurement: measurement uncertainty and statistics, European Commission, Joint Research
Centre, Institute for Reference Materials and Measurements, © European Union, 2012
Task for a Laboratory Measurement of Mass Fraction … 129