-6, 2007

-2007, BARC, January 44-

REFERENCE MATERIALS IN ANALYTICAL

CHEMISTRY

K.L. Ramakumar
Radioanalytical Chemistry Section
Bhabha Atomic Research Centre
Mumbai 400 085

Klram@barc.gov.in
RACNT-
RACNT

January 6, 2007 KLR, RACS, BARC 1

 -6, 2007

Reference Materials in Analytical Chemistry

-2007, BARC, January 44-

Presentation theme
Indispensability of accurate analytical measurements
Role of reference materials in the field of chemical
metrology..
metrology
Definitions, Characteristic features, certification and
use, care and storage of reference materials
Efforts for preparation and characterisation of
reference materials for use in nuclear technology are
RACNT-
RACNT

also enumerated
enumerated..

January 6, 2007 KLR, RACS, BARC 2

 -6, 2007
Statement of Purpose

Any analytical measurement should be reliable and

-2007, BARC, January 44-

invariant
in time
in space
method independent
analyst independent

Realized by standardisation, harmonisation and

assuring the quality of measurements.
measurements.

Particularly true in nuclear technology where

RACNT-
RACNT

rigorous chemical quality control and assurance

measurements are order of the day.
day.
January 6, 2007 KLR, RACS, BARC 3
 Statement of Purpose
-6, 2007

The goal of any analytical measurement is to get

accurate, reliable and consistent data
data..
-2007, BARC, January 44-

Prerequisites for achieving accurate results

results::
Correct sampling
Correct weighing of the sample and standards
Use of well maintained and calibrated equipment
Qualified operators
Validated methods and procedures
Use of accurate standards or certified reference
materials..
materials
Reference materials provide the benefit of comparability—
comparability—
RACNT-
RACNT

between results obtained at different times, in different

places, by different people, and using contrasting
methods..
methods
January 6, 2007 KLR, RACS, BARC 4
 When is an analytical result the result acceptable?
-6, 2007

The result is acceptable when

it has been obtained by an established method
-2007, BARC, January 44-

all the uncertainties (systematic and random)

have been taken into consideration in the
uncertainty value being quoted
the method used has been validated by
analysing a known secondary (working)
reference material
the method used has been calibrated with a
RACNT-

known certified reference material whose

RACNT

value is traceable to international metrological

unit
January 6, 2007 KLR, RACS, BARC 5
 -6, 2007

Primary standard:
standard: Material designated or widely
acknowledged as having the highest metrological
-2007, BARC, January 44-

qualities and whose value is accepted without

reference to other standards of the same quantity
quantity..
e.g. uranium metal or plutonium metal
Secondary standard
standard:: Material whose value is
assigned by comparison with a primary standard of
the same quantity
quantity..
e.g. uranium dioxide pellets or plutonium oxide
oxide..
Their reference value is assigned by comparison
RACNT-
RACNT

with the respective metal standard.

standard.

January 6, 2007 KLR, RACS, BARC 6

 -6, 2007

Working standard
standard:: Used routinely to calibrate or
check material measures, measuring instruments or
-2007, BARC, January 44-

reference materials
materials..
e.g. Mixed solutions of uranium and plutonium
prepared in-
in-house
Certified reference material (CRM)(CRM):: A reference
material, accompanied by a certificate, one or more of
whose properties are certified by a procedure which
establishes traceability to an international unit of
metrology..The certified value is accompanied by an
metrology
RACNT-

uncertainty at a stated level of confidence.

confidence.
RACNT

January 6, 2007 KLR, RACS, BARC 7

 Requirements for (certified) reference material
-6, 2007

•The compounds and concentration should be as similar

as possible to the unknown sample
sample..
-2007, BARC, January 44-

•The material should be 'matrix matched'.

matched'. Unfortunately
full matrix matching frequently is an unrealistic
requirement..
requirement
•The reference material should be homogeneous.
homogeneous.
•If there is a risk for segregation during transport or
storage, the material must be re
re--homogenized before use
use..
•The certified properties of the reference material and the
matrix should be stable.
stable. The material should be checked
RACNT-

for stability as part of the verification process

process..
RACNT

•The uncertainty of the value should be estimated for

certified reference materials
materials..
January 6, 2007 KLR, RACS, BARC 8
 -6, 2007
Test area

1 2 3
….n
Sampling pattern
-2007, BARC, January 44-

Sampling site Sampling point

Sampling unit

Increment/Primary sample

Composite/aggregate sample

Sub-sample

Laboratory sample
RACNT-

Drying, sieving, reducing, milling

RACNT

Test sample

Analytical Operations

January 6, 2007 KLR, RACS, BARC 9

 RACNT
RACNT--2007, BARC, January 44-
-6, 2007

January 6, 2007
KLR, RACS, BARC
10
 -6, 2007
-2007, BARC, January 44-

u2CRM = u2bb + u2its + u2sts + u2char

RACNT-
RACNT

A. Van der Veen, Advances in Chemical

metrology, November 18, 2003

January 6, 2007 KLR, RACS, BARC 11

 -6, 2007
General Challenges in the Preparation and Certification
Reference materials (RMs) serve as the basis for millions of
measurements performed each year
year..
-2007, BARC, January 44-

The “reliability” of these materials, the associated reference

value and the stated uncertainty are of key importance
importance..
Critical requirement in NUMAC
Total SNM inventory depends on the analytical result and
accompanying uncertainty obtained on an aliquot
aliquot..
A high degree of confidence needs to be put in the
measurements and in the reference materials
materials..
Preparation and certification of reference materials is a very
RACNT-
RACNT

demanding task. It requires a thorough understanding of the

processes involved.

January 6, 2007 KLR, RACS, BARC 12

 -6, 2007

Preparation of Reference Materials

-2007, BARC, January 44-

1. Definitive method
The method must be based on first principles, have
very high precision and essentially zero systematic
error.. An example is the use of isotope dilution
error
mass spectrometry for the characteristics of trace
level elements in natural matrix elements.
elements. The
certification is done in a single laboratory
RACNT-
RACNT

January 6, 2007 KLR, RACS, BARC 13

 -6, 2007
Methods identified as having the potential to be
primary
-2007, BARC, January 44-

- isotope dilution with mass spectrometry (IDMS)

- coulometry
- gravimetry a) gas mixtures and b) gravimetric analysis
- titrimetry
- determination of freezing-
freezing-point depression

Measurements of amount of substance, to be considered

primary, must be made using a method which is specific for
a defined substance and for which the values of all
RACNT-

parameters, or corrections which depend on other species or

RACNT

the matrix, are known or can be calculated with appropriate

uncertainty..
uncertainty

January 6, 2007 KLR, RACS, BARC 14

 -6, 2007

Preparation of Reference Materials

2. Independent measurement method
-2007, BARC, January 44-

Two or more reliable independent methods are

used.. The method must be proven to give
used
accurate results
results.. The certification is done in a
single laboratory
laboratory..
3. Interlaboratory consensus method
A number of laboratories analyze in replicate one
or more units of the material being characterized
characterized..
The participating laboratories may choose their
own method or all laboratories may use the same
RACNT-
RACNT

method.. The consensus value is usually taken as

method
the mean
mean..

January 6, 2007 KLR, RACS, BARC 15

 Total measurement uncertainty
-6, 2007

Measurement errors
-2007, BARC, January 44-

Systematic errors Random errors

Sampling
Corrected uncorrected error
Systematic Systematic
errors errors

Result Measurement uncertainty

RACNT-
RACNT

u Total measurement
X uncertainty

January 6, 2007 KLR, RACS, BARC 16

 -6, 2007
Selection Criteria of CRMs
A reference material is selected by comparing the
contribution of the reference material (Ir) to the total
-2007, BARC, January 44-

analytical uncertainty (IT) and by evaluating this as

satisfactory, acceptable or incompatible.
incompatible.

IT / Ir > 10:
10: Satisfactory
10 > IT / Ir > 4: Acceptable
4 > IT / Ir > 2: Acceptable if no other solution
exists;; upward revision of IT
exists
undoubtedly necessary
IT / Ir < 2: Unsatisfactory situation
situation;; implies
reviewing the total uncertainty, the
RACNT-
RACNT

method applied, and the choice of

reference material
material..

January 6, 2007 KLR, RACS, BARC 17

 Use of Reference Materials
-6, 2007

Method Validation and Measurement Uncertainty

Bias estimation One of the most difficult elements
-2007, BARC, January 44-

Appropriate RMs provide information within the limits of

the uncertainty in certified value(s) and the uncertainty of
the method
Verification of the Correct Use of a Method
RMs can be used for training
For checking infrequently used methods
For trouble shooting when unexpected results are
obtained..
obtained
Calibration
Normally a pure substance RM is used for calibration
RACNT-

Other components such as sample digestion, separation

RACNT

and derivatisation are not covered

covered..
The uncertainty associated with RM purity will contribute
to the total uncertainty of the measurement
measurement..
January 6, 2007 KLR, RACS, BARC 18
 Traceability of the measurements
-6, 2007

Consider two
laboratories, A and B,
-2007, BARC, January 44-

carrying out
measurements on
samples of broadly the
same type. Each
calibrates their equipment
using a reference
standard with a known
nominal concentration (x1
(x1
and x2 respectively). They
calculate their respective
results y1 and y2 from a
RACNT-

calibration equation
RACNT

including the respective

values of x.

January 6, 2007 KLR, RACS, BARC 19

 -6, 2007

If, however, the two reference

standards are both calibrated
against some common
-2007, BARC, January 44-

reference, a comparison
becomes meaningful.
meaningful. Now,
both results are derived from
the same value (x0). Both will
now have the same units of
measurement, (the same scale
and units as x0), and direct
comparison of the values y1
and y2 is now not only
possible but also meaningful.
meaningful.
By analogy, of course, x0
could also be derived from a
RACNT-
RACNT

yet higher reference to allow

global comparisons
comparisons..

January 6, 2007 KLR, RACS, BARC 20

 RACNT-
RACNT -6, 2007
-2007, BARC, January 44-

Traceabiliy of working standards to secondary and primary

standards
January 6, 2007 KLR, RACS, BARC 21
 Use of calibration standards (calibrants)
-6, 2007

Primary
-2007, BARC, January 44-

Reference material producer calibrator

Manufacturer’s sells to Secondary
working
calibrator calibrator

Manufacturer’s sells to End user’s

product laboratory
calibrator primary calibrator
End
user’s
RACNT-

Working calibrator
RACNT

laboratory
sample
January 6, 2007 KLR, RACS, BARC 22
 -6, 2007

Use of Reference Materials

Control charts and reference materials:
-2007, BARC, January 44-

Demonstration of measurement procedure

performing within given limits.
Warning limits and Control limits

If the control value is outside the limits, no analytical

results are reported and remedial actions have to be
taken to identify the sources of error, and to remove
such errors.
RACNT-
RACNT

January 6, 2007 KLR, RACS, BARC 23

 -6, 2007

Control charting is a powerful and simple tool for the daily

quality control of routine analytical work. The basis is that the
laboratory runs control samples together with the routine
-2007, BARC, January 44-

samples in an analytical run. Material of control samples can

be standard solutions, real routine samples, blank samples, in-
house control materials and certified reference materials.

Warning limit = CL ±2s Control limit = CL ± 3s

CL + 3s
CL + 2s
CL
CL - 2s
RACNT-
RACNT

CL - 3s

Periodicity
January 6, 2007 KLR, RACS, BARC 24
 -6, 2007

Milenko Markovic, Bruce O. Fowler and Ming S. Tung,

Preparation and Comprehensive Characterization of a Calcium
Hydroxyapatite Reference Material J. Res. Natl. Inst. Stand.
-2007, BARC, January 44-

Technol. 109
109,, 553
553--568 (2004)
Detailed physicochemical characterization of HA, prepared from
an aqueous solution, was carried out employing different
methods and techniques
techniques:: chemical and thermal analyses, x-ray
diffraction, infrared and Raman spectroscopies, scanning and
transmission microscopies, and Brunauer, Emmett, and Teller
(BET) surface method.. The contents of calcium (Ca2+),
surface--area method
phosphate (PO43–), hydroxide (OH–), hydrogenphosphate (HPO42–
), water (H2O), carbonate (CO32–), and trace constituents, the Ca/P
molar ratio, crystal size and morphology, surface area, unit-unit-cell
parameters, crystallinity, and solubility of this HA were
RACNT-
RACNT

determined.. This highly pure, homogeneous, and highly

determined
crystalline HA is certified as a National Institute of Standards and
Technology (NIST) standard reference material, SRM 2910 2910..
January 6, 2007 KLR, RACS, BARC 25
 Availability of Nuclear Reference materials
-6, 2007

Institute for Reference Materials and Measurements

(EC, IRMM), 2440 Geel, Belgium;
-2007, BARC, January 44-

http://irmm.jrc.cec.eu.int/
New Brunswick Laboratory, (NBL), 9800 S. Cass Ave.,
Argonne IL 60439, USA; http://www.nbl.doe.gov/
CETAMA, CEA/Valrho-
CEA/Valrho-Marcoule, DCC/DRRV, BP 171,
30207 Bagnols sur Ceze, France
V.G. Khlopin Radium Institute, 2-
2-nd Murinsky Av. 28,
194021 St. Petersburg, Russia
RACNT-

Ural Electrochemical Integrated Plant (UEIP), Materials

RACNT

are available through TENEX

http://www.tenex.ru/izotop_en.html#uran
January 6, 2007 KLR, RACS, BARC 26
The link ed image cannot be display ed. The file may hav e been mov ed, renamed, or deleted. Verify that the link points to the correct file and location.

Laboratory ware used for storage and

-6, 2007

sample treatment
Borosilicate Glass
Used extensively.
extensively. Resistant to most acids
-2007, BARC, January 44-

Should not be used with HF or boiling H3PO4

Alkaline solutions should not be heated or stored.
stored.
Borosilicate glass can contribute a variety of contaminants.
contaminants.
Should not be heated using a hot plate (500 °C)
C)..
do not use borosilicate glass for ashing in muffle furnace

Porcelain
Popular material used for ashing purposes.
purposes.
Can be heated up to 1370 K
Coated with a glaze which is about 70 % SiO2, with roughly equal
amounts of the oxides of Al and Ca, and lesser amounts of Na and K.
If alkalis are present, then the sample is typically treated with conc
conc..
RACNT-

H2SO4 prior to ashing.

ashing.
RACNT

Do not use porcelain:

porcelain: HF;
HF; boiling H3PO4; and the oxides, hydroxides,
or carbonates of the alkali or alkaline earth elements

January 6, 2007 KLR, RACS, BARC 27

The link ed image cannot be display ed. The file may hav e been mov ed, renamed, or deleted. Verify that the link points to the correct file and location.

Quartz
-6, 2007

•Opaque quartz has the highest trace element concentration and should
not be used for trace analysis.
•Transparent quartz comes in four different varieties
-2007, BARC, January 44-

Types I & II are made from naturally occurring quartz crystals or sands.
Type I is created by electric melting and type II by flame melting. Type II
has slightly less impurities than type I (some impurities are volatilized by
the flame).
Type III quartz is made synthetically by vapor phase hydrolysis of pure
silicon compounds such as SiCl4. This type of quartz is more pure than
-
the natural quartz, with the exception of Cl which is ~ 50 ppm.
Type IV quartz is synthetically made from SiCl4 using a process involving
electrical fusion of the oxidized staring material. It is as pure as type III,
-
with respect to trace metal content, and contains far more Cl .
Use the synthetic type III quartz whenever possible.
Quartz is typically 99.8+% SiO2. It is attacked by HF, boiling H3PO4, and the
RACNT-

alkali and alkaline earth oxides, hydroxides, and carbonates. It can be heated
RACNT

to 1100 °C . Its main advantage over that of porcelain is that major

contamination occurs from only Si -- however, this contamination can be
significant

January 6, 2007 KLR, RACS, BARC 28

The link ed image cannot be display ed. The file may hav e been mov ed, renamed, or deleted. Verify that the link points to the correct file and location.

-6, 2007
Platinum
Expensive but a popular container material.
Heats up and cools down rapidly, making it excellent for % ash determinations
where the % ash is at low levels.
-2007, BARC, January 44-

Resistant to attack by most acids and reagents.

Avoid concentrated H3PO4 at high temperatures, HCl + HNO3 mixtures and
fusions using Li2CO3, Na2O2, or the alkali hydroxides.
Avoid heating at prolonged temperatures in excess of 1370K.
0 +2
Avoid using Pt for a sample containing high levels of Cu or Cu especially if
present in an organic matrix. A sample containing trace levels of Cu in an
organic matrix will not ruin the platinum, but it is likely to be lost to the
platinum during the ashing process.
Since platinum has alloying tendency, it is best to avoid its use with samples
containing elements other than those that have no tendency to form the metal
(i.e. - alkali, alkaline and rare earth elements). Platinum is known to contain
RACNT-

trace amounts of the other precious metals and should not be used for their
RACNT

preparation. Avoid samples containing Hg in any form. Hg metal is easily

formed and alloys very readily at room temperature with platinum. Also avoid
ashing samples containing P in any form, including the phosphates.

January 6, 2007 KLR, RACS, BARC 29

The link ed image cannot be display ed. The file may hav e been mov ed, renamed, or deleted. Verify that the link points to the correct file and location.

-6, 2007

Graphite
Inexpensive and relatively clean, but very messy to work with.
It is an inexpensive way to perform Li2CO3 fusions where the
crucible slowly oxidizes away over the course of 7-
7-10 fusions.
-2007, BARC, January 44-

It does not wet by some melts which can be poured out

quantitatively.
Losses due to the porosity of graphite should exclude its use
for ashing samples containing trace metals.
Graphite's main advantage to the trace analyst is being a
material that can withstand fusions that might destroy
platinum.

Plastics
Plastics are very important to the trace analyst. Whenever
possible, the analyst should attempt to use plastics for sample
RACNT-
RACNT

collection, storage, preparation, and measurement. Their

major disadvantage is the inability to be used for high
temperature operations, such as ashing or fusion.
January 6, 2007 KLR, RACS, BARC 30
 -6, 2007

Indigenous Efforts
-2007, BARC, January 44-

The necessity of reference materials for quantifying

analytical measurements for QC / QA and other
requirements on nuclear materials had been
recognized long ago in BARC
Limited availability of reference nuclear materials
Not all of them accessible to us
RACNT-
RACNT

January 6, 2007 KLR, RACS, BARC 31

 Indigenous Efforts
-6, 2007

Uranium
1950-60’s
1950- 60’s:: Pure uranium metal available from
-2007, BARC, January 44-

Analytical Chemistry Division was being used as

reference material
From late 1960
1960’s
’s:: High purity stoichiometric U3O8
was prepared and used as reference material
material..
1990’s:: Rubidium uranium sulphate (Rb2U(SO4)3) was
1990’s
prepared and characterised for use as reference
material for uranium in RLG
RACNT-
RACNT

G.M. Nair et al.

al., BARC
BARC--560(
560(1971
1971))
K.D. Singh et al.
al., J. Radioanal
Radioanal.. Nucl
Nucl.. Chem
Chem.., 240,
240, 183(
183(1999)
1999)
January 6, 2007 KLR, RACS, BARC 32
 Indigenous Efforts
-6, 2007

Plutonium
-2007, BARC, January 44-

Anhydrous plutonium sulphate for use in

potentiometry
The material is stoichiometric with impurity content
less than 0.07
07%%.
In use for more than two decades
potassium plutonium sulphate (K4Pu(SO4)4) was
prepared and characterised for use as reference
material for plutonium in RLG
RACNT-
RACNT

G.M. Nair et al.

al., BARC/I
BARC/I--517
517((1979
1979))
K.D. Singh et al.
al., BARC
BARC--1496(
1496(1990
1990))
January 6, 2007 KLR, RACS, BARC 33
 Indigenous Efforts
-6, 2007

Trace Elements
Elements:: Non-
Non-Metallics
-2007, BARC, January 44-

Matrix matched reference materials are not available

F and ClCl:: NaF and BiOCl are homogeneously
blended with high purity U3O8
Carbon:: WC or stainless steel standards are used
Carbon
Generally employed for validation of the analytical
technique.. Hence a matrix matched reference
technique
material may not be strictly necessary
RACNT-
RACNT

G.M. Nair et al.

al., BARC/I
BARC/I--517
517((1979
1979))

January 6, 2007 KLR, RACS, BARC 34

 Indigenous Efforts
-6, 2007

Trace metallics
-2007, BARC, January 44-

1970’s:: First concerted effort for U3O8 with known boron

1970’s
content..
content

1980-90’s
1980- 90’s:: Sets of U3O8, ThO2 and graphite materials
containing known amounts of trace constituents
including rare earth elements were prepared for use in
spark source mass spectrometry
spectrometry..

K.L. Ramakumar et alal.., Fresenius’ Z. Anal.

Anal. Chem.
Chem., 318
318,, 12
12((1984)
1984)
RACNT-

K.L. Ramakumar, Ph.

Ph.D. Thesis, University of Bombay(1985
Bombay(1985))
RACNT

B.P. Datta, Spectrochimica Acta, B52 52,, 471(

471(1997)
1997)

January 6, 2007 KLR, RACS, BARC 35

 Indigenous Efforts
-6, 2007

2000’s: Interlaboratory experiments by NFC, and BARC to

characterize U3O8 and ThO2 with respect to trace elements.
-2007, BARC, January 44-

Prepared following a well-

well-defined analytical protocol.
Different analytical techniques employed in the
participating laboratories
Rigorous statistical treatment of the analytical results
Consensus values have been assigned to the individual
elemental concentrations.
DAE-Interlaboratory comparison experiment (ILCE) for trace metal
DAE-
assay of uranium-
uranium-Phase II: Experiments: development of certified
reference materials, compiled by radiochemistry division, BARC and
NFC, Hyderabad (2002)
RACNT-
RACNT

Preliminary report on the DAE interlaboratory comparison

experiments for determination of trace metals in thorium oxide,
Radiochemistry Division, BARC (2004)

January 6, 2007 KLR, RACS, BARC 36

 Conclusions
-6, 2007

Reference materials for any analytical measurement,

more so in nuclear technology whether it is for chemical
-2007, BARC, January 44-

quality control or for accountability purpose are

indispensable..
indispensable

Availability of reference materials enables any analyst

to come as close to absolute as analytically feasible
feasible..
At the same time these are not panacea to all the
analytical requirements and one has to be judicious
enough in their proper selection
selection..

It is necessary to prepare and characterise indigenous

RACNT-
RACNT

reference materials for use in nuclear technology, as

many of these are not accessible to the users in India
India..

January 6, 2007 KLR, RACS, BARC 37

 -6, 2007
Epilogue

1. Beware of contamination
-2007, BARC, January 44-

2. Understand matrix effects (*)

• Use standard addition where ever necessary(*)
• Calibrate the method (*)
• Determine precision and accuracy(*)
• Ensure proper sampling plan
• Know your Statistics
RACNT-
RACNT

Need for RMs.

Use appropriate reference materials
January 6, 2007 KLR, RACS, BARC 38