ISO 9516-1:2003

INTERNATIONAL ISO
STANDARD 9516-1
First edition
2003-04-01
Iron ores — Determination of various
elements by X-ray fluorescence
spectrometry —
Part 1:
Comprehensive procedure
Minerais de fer — Dosage de divers éléments par spectrométrie de
fluorescence de rayons X —
Partie 1: Procédure détaillée
Reference number
©
ISO 2003
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.

©  ISO 2003
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2003 — All rights reserved

Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Normative references . 2
3 Principle . 2
4 Reagents and materials. 2
5 Apparatus. 6
6 Sampling and samples . 7
7 Procedure. 7
8 Calculation of results. 17
9 General treatment of results . 20
10 Test report. 24
Annex A (normative) Preparation of flux A. 25
Annex B (normative) Preparation of flux B or flux C . 27
Annex C (normative) Preparation of synthetic calibration standard . 28
Annex D (normative) Standard deviation of specimen preparation. 30
Annex E (normative) Spectrometer precision tests. 35
Annex F (normative) Determination of the dead time and maximum count rate of the equipment. 39
Annex G (informative) Air cooling block for fused discs . 46
Annex H (informative) Computer program for calculation of results. 47
Annex I (informative) Sample of data for use with calculation program . 60
Annex J (normative) Flowchart for acceptance of results . 65

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 9516-1 was prepared by Technical Committee ISO/TC 102, Iron ore and direct reduced iron,
Subcommittee SC 2, Chemical analysis.
This first edition, together with ISO 9516-2, cancels and replaces ISO 9516:1992 by the augmentation of the
range of elements under analysis and the diversification into two procedures.
ISO 9516 consists of the following parts, under the general title Iron ores — Determination of various elements
by X-ray fluorescence spectrometry:
 Part 1: Comprehensive procedure
 Part 2: Simplified procedure
iv © ISO 2003 — All rights reserved

Introduction
In this part of ISO 9516, Table 1 indicates that some determinations may be used for referee purposes and
others for routine analysis only.
A simplified procedure for routine use with all determination will be published in ISO 9516-2.

INTERNATIONAL STANDARD ISO 9516-1:2003(E)

Iron ores — Determination of various elements by X-ray
fluorescence spectrometry —
Part 1:
Comprehensive procedure
WARNING — This part of ISO 9516 may involve hazardous materials, operations and equipment. This
part of ISO 9516 does not purport to address all of the safety problems associated with its use. It is
the responsibility of the user of this part of ISO 9516 to establish appropriate health and safety
practices and determine the applicability of regulatory limitations prior to use.
1 Scope
This part of ISO 9516 sets out a wavelength dispersive X-ray fluorescence procedure for the determination of
iron, silicon, calcium, manganese, aluminium, titanium, magnesium, phosphorus, sulfur, potassium, tin,
vanadium, chromium, cobalt, nickel, copper, zinc, arsenic, lead and barium in iron ores. The method has been
designed to cope with iron ores having high ignition losses.
The method is applicable to iron ores regardless of mineralogical type. The concentration range covered for
each of the component elements is given in Table 1. The determination of total iron cannot be used for referee
purposes.
Table 1 — Range of application of the method
Component Concentration range for Concentration range for
element referee purposes analysis
% %
Fe 38 to 72
Si 0,2 to 6,5 0,2 to 6,5
Ca 0,019 to 12,7 0,019 to 12,7
Mn 0,02 to 0,82 0,02 to 0,82
Al 0,1 to 3,5 0,1 to 3,5
Ti 0,016 to 4,7 0,016 to 4,7
Mg 0,2 to 2,0 0,2 to 2,0
P 0,006 to 0,6 0,006 to 0,6
S 0,04 to 0,6 0,007 to 0,6
K 0,008 to 0,45 0,012 to 0,45
Sn 0,006 to 0,015
V 0,001 7 to 0,3 0,001 7 to 0,3
Cr 0,006 to 0,024
Co 0,006 to 0,018
Ni 0,011 to 0,013
Cu 0,012 to 0,061
Zn 0,006 9 to 0,166 0,005 to 0,166
As 0,008 to 0,06
Pb 0,018 to 0,32 0,018 to 0,32
Ba 0,036 to 0,4
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
ISO 3082:1998, Iron ores — Sampling and sample preparation procedures
ISO 7764:1985, Iron ores — Preparation of predried test samples for chemical analysis
3 Principle
The glass discs for X-ray fluorescence measurement are prepared by incorporating the test portion of the iron
ore sample, via fusion, into a borate glass disc using a casting procedure. By using a fused glass disc, particle
size effects are eliminated. Sodium nitrate is added to the flux to ensure complete oxidation of all components,
particularly iron and sulfur. Any of three methods for glass disc preparation may be used: two use lithium
borate as flux; the other uses sodium borate.
X-ray fluorescence measurements are based on the “line only” principle. It is not necessary to measure
backgrounds on each glass disc, as background equivalent concentrations (BEC) are determined on several
blank glass discs at the line position using concentration-based line-overlap corrections. If desired,
backgrounds can be measured to obtain net line intensities. The method is applicable to data from
simultaneous and sequential X-ray fluorescence spectrometers.
The method relies on measuring all components of the sample, other than volatiles. If some components are
not measured, then errors will result in the measured components (see 7.2.2).
Calibration is carried out using pure chemicals. Results are obtained after matrix corrections for inter-element
effects.
4 Reagents and materials
During analysis, only reagents of recognized high purity shall be used.
NOTE 1 Where reagents have been ignited, they should be covered during cooling in the desiccator and weighed as
soon as possible.
NOTE 2 Reagents 4.2, 4.5, 4.7, 4.8, 4.9, 4.11, 4.13, 4.15, 4.16, 4.18 and 4.20 are used only for the preparation of the
synthetic calibration standard, and are not required if the synthetic calibration standard is available commercially.
4.1 Silicon dioxide, (SiO ), nominally 99,999 % SiO
2 2
The silicon dioxide shall contain less than 3 µg/g of each of the other elements listed in Table 1. It shall be
heated to 1 000 °C in a platinum crucible for a minimum of 2 h and cooled in a desiccator.
4.2 Aluminium oxide, (Al O ), analytical reagent grade, α form
2 3
If the α form is used, it shall be heated to 1 000 °C in a platinum crucible for a minimum of 2 h. If the
aluminium oxide is not the α form, it shall be converted to the α form by heating to 1 250 °C in a platinum
crucible for a minimum of 2 h. It shall be cooled in a desiccator and weighed as soon as it is cool.
2 © ISO 2003 — All rights reserved

4.3 Iron(III) oxide, (Fe O ), nominally 99,999 % Fe O
2 3 2 3
The iron(III) oxide shall contain less than 3 µg/g of each of the other elements listed in Table 1. It shall be
heated at 1 000 °C in a platinum crucible for a minimum of 1 h and cooled in a desiccator.
4.4 Titanium dioxide, (TiO )
Analytical grade titanium dioxide shall be heated at 1 000 °C in a platinum crucible for a minimum of 1 h and
cooled in a desiccator.
Phosphorus is a common impurity in TiO and a reagent low in phosphorus shall be selected. The selected
reagent shall be checked, as even nominally high-purity reagents can be significantly contaminated, e.g. a
supposed 99,99 % TiO grade reagent has been found to contain about 0,5 % P O .
2 2 5
4.5 Potassium dihydrogen orthophosphate, (KH PO )
2 4
Analytical grade potassium dihydrogen orthophosphate shall be dried at 105 °C for 1 h and cooled in a
desiccator.
4.6 Calcium carbonate, (CaCO )
Analytical grade calcium carbonate shall be dried at 105 °C for 1 h and cooled in a desiccator.
.
4.7 Calcium sulfate, (CaSO 2H O)
4 2
Analytical grade calcium sulfate dihydrate shall be dehydrated at 700 °C for 1 h and cooled in a desiccator.
4.8 Manganese oxide, (Mn O )
3 4
Manganese oxide shall be prepared by heating analytical grade manganese oxide (MnO , MnO or Mn O ) for
3 4
15 h at 1 000 °C in a platinum crucible and then cooling. The lumpy material shall be crushed to a fine powder,
heated for 1 h at 200 °C and cooled in a desiccator.
4.9 Magnesium oxide, (MgO)
Analytical grade magnesium oxide shall be dried in a platinum crucible by slowly heating from room
temperature to 1 000 °C. After 1 h at 1 000 °C, the crucible containing the magnesium oxide shall be placed in
a desiccator and weighed as soon as it is cool, as magnesium oxide readily absorbs carbon dioxide from the
atmosphere.
4.10 Sodium nitrate, (NaNO )
Analytical grade sodium nitrate shall be dried at 105 °C for 1 h and cooled in a desiccator.
4.11 Tin oxide, (SnO )
Analytical grade tin oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.12 Vanadium(V) oxide, (V O )
2 5
Analytical grade vanadium(V) oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.13 Chromium(III) oxide, (Cr O )
2 3
Analytical grade chromium(III) oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.14 Cobalt oxide, (Co O )
3 4
Analytical grade cobalt oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.15 Nickel oxide, (NiO)
Analytical grade nickel oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.16 Copper oxide, (CuO)
Analytical grade copper oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.17 Zinc oxide, (ZnO)
Analytical grade zinc oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
.
4.18 Di-sodium hydrogen arsenate, (Na HAsO 7H O)
2 4 2
The analytical grade reagent shall be weighed as received.
4.19 Lead oxide, (PbO)
Analytical grade lead oxide shall be heated at 400 °C for a minimum of 1 h and cooled in a desiccator.
4.20 Barium carbonate, (BaCO )
Analytical grade barium carbonate shall be heated at 105 °C for a minimum of 1 h and cooled in a desiccator.
4.21 Ammonium iodide, (NH I)
Laboratory reagent grade ammonium iodide need not be dried, but shall be stored in a desiccator.
4.22 Desiccant
The desiccant shall be freshly regenerated self-indicating silica gel.
4.23 Flux
4.23.1 General
Flux A, flux B or flux C, as described in 4.23.2, 4.23.3 and 4.23.4, may be used. The levels of contamination in
the flux shall be checked (see 9.1). Because levels of contamination may vary from batch to batch, the same
batch of flux shall be used for all discs (iron ore, blank and calibration) involved in the batch of determinations.
4.23.2 Flux A
Flux A shall be prepared by fusion of a mixture of anhydrous lithium tetraborate (Li B O ) and anhydrous
2 4 7
lithium metaborate (LiBO ) using the procedure specified in Annex A. Flux shall be dried at 500 °C for a
minimum of 4 h and stored in a desiccator.
4.23.3 Flux B
Flux B shall be prepared using sodium tetraborate using the procedure specified in Annex B. Flux shall be
dried at 500 °C for a minimum of 4 h and stored in a desiccator.
4 © ISO 2003 — All rights reserved

4.23.4 Flux C
Flux C shall be prepared using lithium tetraborate using the procedure specified in Annex B. Flux shall be
dried at 500 °C for a minimum of 4 h and stored in a desiccator.
NOTE If this flux is used, sulfur will not be reported.
4.24 Calibration standard
Two independent (i.e. prepared on different days) batches (labelled Day 1 and Day 2) of calibration standard
shall be prepared by the procedure specified in Annex C. The composition of the calibration standard, given in
Table 2, approximates that of an iron ore. The contents of some elements are higher than would be expected
in an iron ore, but this is advantageous for obtaining a reliable calibration.
Prior to weighing, a sufficient aliquot of the calibration standard shall be heated at 900 °C for 20 min and
cooled in a desiccator.
Table 2 — Composition of the calibration standard
Component Content Oxide content
element
% %
Fe 64,000 Fe O
44,764
2 3
Si 4,44 9,500 SiO
Ca 3,067 4,2913 CaO
Mn 1,441 2,000 Mn O
3 4
Al 2,65 5,000 Al O
2 3
Ti 1,500 TiO
0,899
Mg
3,016 5,000 MgO
P 1,16 2,660 P O
2 5
S 0,921 2,300 SO
K 1,46 1,758 9 K O
Sn 0,157 5 0,200 SnO
V 0,112 0 0,200 V O
2 5
Cr 0,200 Cr O
0,136 8
2 3
Co 0,200 Co O
0,146 8
3 4
Ni
0,157 2 0,200 NiO
Cu 0,159 8 0,200 CuO
Zn
0,160 7 0,200 ZnO
As 0,084 7 0,111 8 As O
2 3
Pb 0,185 7 0,200 PbO
Ba 0,179 1 0,200 BaO
Na 0,052 0 0,070 1 Na O
5 Apparatus
5.1 General
The sample may be fused with the flux in a crucible and then poured into a separate mould or, if an
appropriately shaped crucible is used, the fusion may be carried out and the glass allowed to cool in the same
crucible. Both methods will produce glass discs of the same quality.
A conventional electric furnace, high-frequency furnace, or a gas burner may be used for heating.
There are disc-making machines commercially available, and these may be used to fuse and cast the discs.
A platinum lid may be used to cover the crucible if fusing in a furnace, but not if fusing over a flame, as this
enhances sulfur loss.
Where a high-frequency furnace or a gas burner is used for heating, a check shall be made to determine if
sulfur is lost during disc preparation. A mixture that contains 90 % Fe O and 10 % CaSO shall be prepared
2 3 4
and used to prepare replicate discs using normal fusion times and times of twice and thrice normal. The
intensity of SKα from the discs should not vary by more than 2 % relative.
5.2 Analytical balance, capable of weighing to four decimal places.
5.3 Crucible and mould
5.3.1 General
The crucible and mould shall be made from a non-wetting platinum alloy.
NOTE 1 Either platinum/gold or platinum/gold/rhodium alloys are suitable.
If more than one crucible or more than one mould is used for casting, these crucibles or moulds shall all be
used in the specimen preparation test in Annex D.
NOTE 2 It is essential to use all of the crucibles or moulds, as casting vessels may become distorted with use, giving
the analytical surface a curvature that will result in error.
Sometimes, even undistorted crucibles or moulds give curvatures unique to the particular crucible or mould.
5.3.2 Crucible
Where the crucible is used for fusion only, it shall have sufficient capacity to hold the flux and sample required
for fusion. Where the crucible is used as a mould as well as for fusion, it shall have a flat bottom, to enable
production discs to fit the spectrometer.
5.3.3 Mould
Because the bottom of the disc is the analytical surface, the inside bottom surface of the mould shall be flat
and shall be polished regularly with approximately 3 µm diamond paste to ensure that the glass disc releases
easily from the mould. To prevent deformation through repeated heating and cooling, the base shall be
greater than 2 mm thick.
5.4 Electric furnace, capable of maintaining a temperature of at least 1 050 °C.
The furnace shall be capable of maintaining higher temperatures where it is to be used for converting Al O to
2 3
the α form (1 250 °C), or for preparing flux A (1 100 °C).
The furnace may be of a conventional type with heating elements, or may be a high-frequency furnace. The
furnace shall be cleaned regularly to prevent contamination of the samples.
6 © ISO 2003 — All rights reserved

5.5 Gas-oxygen burner
Where fusions are made over a gas-oxygen flame, provision shall be made for oxygen enhancement of the
flame to minimize sulfur loss and crucible contamination. The temperature of the melt shall be in the range
1 000 °C to 1 050 °C. The temperature shall be checked using an optical pyrometer while the crucible
contains several grams of flux. Alternatively, if an optical pyrometer is not available, about 3 g of potassium
sulfate (m.p. 1 069 °C) shall be added to the crucible and the flame adjusted so that it all just melts in the open
crucible. A gas burner may be used for heating the mould, and it shall be adjusted so that the mould is a bright
red heat (approximately 950 °C). A Meker burner shall not be used, as loss of sulfur and the uptake of iron
from the glass into the platinum ware may result.
5.6 Desiccator
5.7 Spatulas, non-magnetic, for weighing of the test portion and for mixing.
Vibrating spatulas are not acceptable, because they can lead to segregation of the sample.
5.8 X-ray fluorescence spectrometer, of any wavelength dispersive, vacuum (or helium) path type, X-ray
fluorescence spectrometer, provided that the instrument has been checked. Performance checks shall be
carried out in accordance with the precision tests set out in Annex E, accumulating at least 2 × 10 counts for
each measurement.
The dead time for FeKα is determined in the method described in Annex F, and this dead time may be used
for all elements when using a sequential instrument. However, where separate counting channels are used for
the different elements (simultaneous instruments), or where the detector is changed, the dead time of each
channel shall be determined independently. The procedure is given in Annex F.
5.9 Ultrasonic bath, optional. It may be used to aid cleaning of the platinum ware.
5.10 Cooling device
NOTE It is recommended that the mould and glass be cooled using an air jet. Commercial disc-making machines use
this method. A drawing of a suitable device is given in Annex G.
Whatever the method of cooling, it is vital that samples be treated identically, as the curvature of the analytical
surface of the disc depends on the rate of cooling.
6 Sampling and samples
Samples shall be taken and prepared in accordance with ISO 3082. The predried test samples shall be
prepared according to the procedure specified in ISO 7764. The calibration standards shall be heated to
900 °C for 20 min prior to weighing and then cooled in a desiccator.
7 Procedure
7.1 Preparation of discs
7.1.1 General
Independent duplicate sets (Day 1 and Day 2) of test samples, blanks and calibration samples shall be
prepared. The expression “independent” implies that the repetition of the procedure be carried out at a
different time or by a different operator.
The operator shall have demonstrated the ability to consistently make discs with high precision. This ability
shall be verified each month by carrying out the procedure given in Annex D.
In preparing discs, great care shall be taken to avoid contamination and, in particular, the crucible in which the
fusion is carried out shall be thoroughly cleaned prior to use (see 7.1.8).
7.1.2 Weighing
Table 3 shows the components used in making the glass discs. Provided that the proportions are kept
approximate to those given in Table 3, the masses can be varied to suit mould diameter and shape (see
Note 1).
Table 3 — Masses of specimen components
Mass
a g
Standard masses
Component
Disc diameter
g
32 mm 40 mm
Flux 6,80 4,10 to 4,61 6,40 to 7,20
NaNO 0,40 0,24 to 0,27 0,38 to 0,42
Sample 0,66 0,41 to 0,44 0,64 to 0,68
a
Values used to calculate alpha coefficients.
The specified masses may be weighed as “catch” weights, recording the mass weighed to the nearest 0,001 g
for the flux and sodium nitrate portions, and to the nearest 0,000 1 g for the test and calibration portions.
If desired, ammonium iodide (4.21) can be used as a releasing agent. If added at this stage, no more than
0,01 g shall be added. Alternatively, a smaller amount may be added prior to casting (see 7.1.5)
NOTE 1 If a disc diameter used differs from those given in Table 3, masses should be adjusted to be approximately
proportional to the area of the glass disc. If masses used are higher than recommended, crystallization and segregation
with consequent cracking are likely to occur as the glass cools.
NOTE 2 Bromides are used as releasing agents but, since BrLα interferes with AlKα, they are not used in this part of
ISO 9516.
Because the components are hygroscopic, they shall be weighed as soon as possible after reaching room
temperature following heating and without any undue delay between weighings. Weighings may be made
direct into the crucible to be used in the fusion, or into a clean glass vial. Because of static effects, glass vials
are preferable to plastic. If a vial is used, care shall be taken to ensure complete transfer of the contents into
the fusion crucible.
7.1.3 Mixing
Thoroughly mix the components in the crucible using a microspatula or similar implement, taking care that no
material is lost. Brush any fine material adhering to the mixing implement back into the crucible. Gently tap the
bottom of the crucible on the bench top to ensure that any material adhering to the crucible wall, above the
general level of the mixed components, is reincorporated into the bulk of the mix.
It is imperative that the crucible be tapped gently on the bench top, as too severe an impact will result in the
loss of some of the finer material and possible deformation of the crucible.
NOTE The mixing implement used should be free of sharp or pointed edges, in order to ensure that the interior of the
crucible is not damaged by scratching.
8 © ISO 2003 — All rights reserved

7.1.4 Fusion
For samples containing sulfur as sulfide, the fusion mixture is to be preoxidized by heating to 700 °C for
10 min prior to fusion. Place the crucible in the electric furnace (5.4) or on the gas-oxygen burner (5.5) at a
temperature of 1 000 °C to 1 050 °C and maintain this temperature for 10 min. At least once during this period,
after the sample is dissolved, briefly swirl the mixture. While swirling, incorporate into the melt any material
that may be adhering to the sides of the crucible.
If a furnace is used for heating, it may be necessary to remove the crucible from the furnace for the purpose of
swirling. When the furnace is opened, the temperature may drop. The specified temperature shall be regained
before the time period starts.
7.1.5 Casting
If ammonium iodide was not added as a release agent earlier, it may be added to the melt just prior to casting.
In this case, no more than 0,002 g shall be added. Casting is then carried out by one of the following methods.
a) Casting in the crucible
If the glass is to be cast in the crucible, remove the crucible from the furnace, place on a suitable cooling
device (5.10) and allow the glass to solidify.
b) Casting in a separate mould
If the glass is to be cast in a separate mould, the mould shall be pre-heated over a gas flame to red heat
(900 °C to 1 050 °C). While the mould is still hot, pour the melt into the mould from the crucible. Remove the
mould from the heat source and place it on the cooling device (5.10) and allow the glass to solidify.
NOTE Failure to ensure that the mould is scrupulously clean prior to casting will result in discs sticking to the mould
and possibly cracking.
7.1.6 Visual inspection
Prior to storage, discs shall be inspected visually, paying particular attention to the analytical surface. The
discs shall not contain undissolved material, and shall be whole and free from crystallization, cracks and
bubbles. Defective discs shall be re-fused in the crucible, or discarded and substitute discs prepared.
7.1.7 Disc storage
As soon as possible (while the glass is still warm), transfer the discs to a desiccator so that absorption of
moisture and the possibility of contamination are minimized. When not being measured, discs shall be stored
in a clean desiccator.
To avoid contamination of the analytical surface, the specimen shall be handled by its edges and the surface
shall not be touched by hand or treated in any way. Specifically, it shall not be washed with water or other
solvents, ground or polished.
NOTE If paper labels are used on the backs of discs, great care should be taken to ensure that the labels do not
contact the analytical surfaces of other discs. Paper labels are clay coated and readily cause contamination by silicon and
aluminium. For the same reason, paper envelopes should not be used to store the discs.
7.1.8 Cleaning of platinum ware
Although the crucible and mould are fabricated from an alloy that is not wetted by the glass, in order to ensure
absolute precision they shall be cleaned between each fusion. Immersion in hot hydrochloric, citric or acetic
acid (approximately 2M), for about 1 h is usually sufficient, but they should be inspected to ensure that all
residual glass has been removed.
A rapid method of cleaning is to put the crucible or mould into a beaker containing the acid. Place the beaker
in a small ultrasonic bath for about 1 min or until all residual glass is removed, then rinse the mould in distilled
water and dry before using.
An alternative method of cleaning is to fuse several grams of flux in the crucible, moving the melt around to
clean the entire inner surface. The molten flux is then poured from the crucible. If a droplet adheres to the
crucible, this can easily be flaked off when the crucible is cold.
7.1.9 Monitor discs
To compensate for drifts in X-ray tube output intensity, all X-ray measurements shall be made relative to a
monitor disc. Although different monitor discs could be used for each component, it is most convenient to use
a single disc, containing all components to be measured. The requirements of the monitor disc are that it be
stable, at least for the time necessary to complete all the measurements associated with a batch of analyses.
Also, the monitor shall contain sufficient amounts of each element in order to ensure that each analytical line
is much higher than the BEC. Suitable stable monitor discs made for the analysis of iron ore are commercially
available.
Although discs prepared in accordance with the normal procedure described in 7.1.1 to 7.1.11 are not stable
over prolonged periods, they are stable for a sufficient period to allow such a disc to be used as a monitor.
One of the calibration discs would, therefore, be a suitable monitor, and in such a case this particular disc
shall be clearly identified. The same monitor shall be used for Day 1 and Day 2 measurements.
7.1.10 Calibration discs
Calibration shall be carried out using discs prepared according to the masses and various proportions set out
in Table 4, where w is the standard mass of chemical compound (referred to as “sample” in Table 4) prepared
in accordance with Clause 4.
If handled carefully and stored under desiccation, the glass discs can be used for several weeks. The
analytical surface shall under no circumstances be touched by hand.
Table 4 — Synthetic standard set
Sample components
Weight of compound Weight of SiO
Disc identity Number Description
g (4.1)
g
Si A, B 2 100 % SiO N/A 1,00 w
Fe A, B 2 100 % Fe O 1,00 w of Fe O (4.3) N/A
2 3 2 3
30Fe/Si A, B 2 30 % Fe O :70 % SiO 0,30 w of Fe O (4.3) 0,70 w
2 3 2 2 3
66Fe/Si A, B 2 66 % Fe O :33 % SiO 0,66 w of Fe O (4.3) 0,33 w
2 3 2 2 3
Ca/Si 1 10 % CaO:90 % SiO 0,179 w of CaCO (4.6) 0,90 w
2 3
Ti/Si 1 10 % TiO :90 % SiO 0,10 w of TiO (4.4) 0,90 w
2 2 2
V/Si 1 10 % V O :90 % SiO 0,10 w of V O (4.12) 0,90 w
2 5 2 2 5
Cr/Si 1 10 % Cr O :90 % SiO 0,10 w of Cr O (4.13) 0,90 w
2 3 2 2 3
Mn/Si 1 10 % Mn O :90 % SiO 0,10 w of Mn O (4.8) 0,90 w
3 4 2 3 4
Co/Si 1 10 % Co O :90 % SiO 0,10 w of Co O (4.14) 0,90 w
3 4 2 3 4
Pb/Si 1 10 % PbO:90 % SiO 0,10 w of PbO (4.19) 0,90 w
Zn/Si 1 10 % ZnO:90 % SiO 0,10 w of ZnO (4.17) 0,90 w
Ba/Si 1 10 % BaO:90 % SiO 0,129 w of BaCO (4.20) 0,90 w
2 3
a
SynCal A, B 2 Synthetic calibration standard 1,00 w of calibration standard (4.24) N/A
a
Use Day 1 calibration standard for the first set, and Day 2 calibration standard for the second set.
10 © ISO 2003 — All rights reserved

7.1.11 Test discs
One disc from each test sample shall be prepared. At least one certified reference material, of the same type
as the ore used in the test discs, should be prepared. Prior to fusing test discs, crucibles should be thoroughly
clean, particularly if the same crucibles were used to prepare the calibration discs, some of which are high in
trace elements.
7.2 Measurements
7.2.1 General
The analytical lines to be used and suggested conditions of measurement are given in Table 5. Other
instrument parameters (collimators and detectors) shall be selected according to the particular element.
In the set of discs listed in Table 4, Fe, 30Fe/Si, 66Fe/Si and Si are used to determine alpha(Fe,Fe). Discs
Ti/Si, V/Si, Co/Si, Pb/Si, Zn/Si and Ba/Si are used to determine overlap factors. The overlap factors should be
constant, but may change with spectrometer alignment. The discs listed in Table 4 need not be included with
each set of analyses; the factors determined previously can be used. However, all discs shall be included at
least once each four weeks and, if alpha(Fe,Fe) or line overlap factors vary significantly, the cause shall be
determined and the problem remedied.
Table 5 — Suggested analytical lines, crystals and operating conditions
a b
Component Line Voltage, kV Crystal Specific line
element overlaps
(7.2.3) (7.2.4) (7.2.5) (7.2.7)

Fe Kα 80 or 40 LiF(200)or LiF(220) —
Kβ 80 or 40 LiF(200)or LiF(220) —
Si Kα 30 or 50 PE —
Ca Kα 30 or 50 LiF(200) or PE or Ge(111) —
Mn Kα 80 or 50 LiF(200) CrKβ
Al Kα 30 or 50 PE BaLα(3), CrKβ(4)
Ti Kα 80 or 40 LiF(200) BaLα
Mg Kα 30 or 50 TlAP or multi-layer —
P Kα 30 or 50 Ge(111) or PE —
S Kα 30 or 50 Ge(111) or PE CoKα(3), PbMα
K Kα 30 or 50 LiF(200) —
Sn Lα 30 or 50 LiF(200) CoKα(2)
V Kα 80 or 50 LiF(200) TiKβ, BaLβ
Cr Kα 80 or 50 LiF(200) VKβ
Co Kα 80 or 50 LiF(200) FeKβ
Ni Kα 80 or 50 LiF(200) CoKβ
Cu Kα 80 or 50 LiF(200) —
Zn Kα 80 or 50 LiF(200) —
As Kα 80 or 50 LiF(200) PbLα
Lβ1 80 or 50 LiF(200) —
Pb
Mα 30 or 50 PE —
Lα 80 or 40 LiF(200) TiKα
Ba
Lβ1 80 or 40 LiF(200) TiKβ, VKα
a
The first figure will normally give better performance, but performance will depend on tube used.
b
The first crystal listed is preferred.
7.2.2 Effect of errors or omissions
There are various circumstances where all twenty elements may not be determined. If a simultaneous
instrument is used, there may not be analytical channels for all elements. In plant control, there may be no
point in doing all the minor elements. Where the source material is constant and known, it is probably
unnecessary to do the minor elements on all samples.
When converting intensities to concentrations, the intensities are multiplied by the absorption coefficient of the
glass (the matrix factor) and this factor receives contributions from all components of the sample, so if one or
more components are not determined then all other components will be in error. The method then relies on
the measurement of all components of the sample.
Table 6 shows the error, as a relative percentage, for each analyte where there is a 1 % error in a component.
The error may be not estimating a component if it has a significant concentration. The calculations have been
made for a typical iron ore.
Table 6 can be used to estimate at what level a minor element may be omitted from the analysis without
exceeding a predetermined error.
Where small errors are involved in a component, the resulting errors to other components are proportional to
the initial error, and if more than one component is in error, or omitted, the errors are additive.
The errors shown in Table 6 have been calculated on the basis of matrix errors only. Overlap errors can give
rise to additional errors. In such cases, the error to a component is an absolute concentration, i.e. the error is
not proportional to the concentration of the analyte. Overlap errors are more important for minor elements. No
attempt has been made to quantify these errors as they are dependent on instrument parameters.
Iron is the element required with high precision, and if 0,1 % Fe O is regarded as the maximum error that can
2 3
be tolerated then, using Table 6, it can be seen that the omission of measuring 250 ppm BaO will give such an
error. From Table 6 it can be seen that an error of 1 % in BaO gives a relative error of 4 % in Fe O , so that if
2 3
the Fe O content of the ore is 90 % the resulting error due to 0,025 % BaO is calculated as follows:
2 3
0,025 × 4,3 × 90/100 = 0,097 %.
7.2.3 Analytical lines
Line-only positions are measured. It is not necessary to measure background intensities, but if desired they
can be measured and net intensities recorded.
7.2.4 XRF generator settings
The voltage, in kilovolts, is not critical and normally with a simultaneous instrument will be set in the range
40 kV to 60 kV. Where using a sequential instrument, it may be advantageous to use a low voltage (40 kV) for
the lighter elements and a higher voltage for the heavier elements. If tube operating conditions are changed
during analysis, this may result in slight instability in the spectrometer output. Since Fe content is required to
be determined with very high precision, Fe has to be measured in a separate run with constant tube
conditions if conditions vary for the other elements.
Some spectrometers are limited to less than 80 kV. In these cases, use the highest available voltage. The
tube current is governed by the maximum power that can be applied to the X-ray tube; consult the
manufacturer’s specification for this information.
When XRF generators are powered up, it is common for the instrument to drift for some time, typically 30 min
to 60 min. Therefore, prior to measurement, the generator should be powered up and left to stabilize.
All measurements shall be made under vacuum, using a detector (proportional or scintillation counter)
appropriate to the wavelength being measured, and using specimen rotation if available. A Cr, Cr/Au, Sc,
Sc/Mo, Sc/W or Rh target X-ray tube shall be used. It is recommended that pulse height selection be used,
12 © ISO 2003 — All rights reserved

particularly in the case where low concentrations are being determined. Where count rates are very high (e.g.
Fe Kα or Ca Kα), either wide pulse height settings or no upper level shall be used.
In special circumstances (e.g. determining Cr or Mn using a Cr type tube), primary beam filters may be used,
but they should not be used as a method for reducing the count rate, as they will alter matrix effects. The
exception to this will be the determination of Mn and Cr using a Cr target X-ray tube where a filter will be
required to achieve low backgrounds.

14 © ISO 2003 — All rights reserved
Table 6 — Error, in relative %, resulting from a 1 % error in a component
Affected component
Fe O SiO CaO Mn O Al O TiO MgO P O SO KO SnO V O Cr O Co O NiO CuO ZnO As O PbO BaO Cl
2 3 2 3 4 2 3 2 2 5 3 2 2 2 5 2 3 3 4 2 3
Plus 1 %
SiO 0,57 0,54 0,58 0,22 0,70 0,43 0,65 1,33 0,59 2,40 1,02 1,16 0,67 1,40 1,64 1,39 2,35 1,17 1,73 0,54
CaO 2,01 0,55 2,04 0,54 2,32 1,03 0,83 1,71 0,94 4,31 3,44 3,94 2,37 4,96 5,79 4,93 8,33 1,56 5,76 1,28
Mn O 0,90 0,48 0,56 0,48 0,52 0,94 0,69 1,38 0,59 2,48 0,79 1,61 3,02 4,24 4,97 4,24 7,21 1,24 1,42 1,17
3 4
Al O 0,50 0,40 0,49 0,51 0,62 0,38 0,59 1,19 0,52 2,17 0,90 1,02 0,58 1,24 1,44 1,22 2,07 1,06 1,54 0,48
2 3
TiO 1,81 0,52 0,61 1,83 0,53 1,04 0,77 1,52 0,66 2,74 1,42 3,53 2,14 4,47 5,23 4,46 7,56 1,36 2,34 1,31
MgO 0,45 0,37 0,44 0,47 0,38 0,56 0,55 1,10 0,49 1,98 0,82 0,92 0,53 1,11 1,30 1,10 1,87 0,98 1,40 0,42
P O 0,63 0,24 0,59 0,65 0,24 0,77 0,47 1,43 0,63 2,64 1,13 1,28 0,74 1,56 1,83 1,54 2,63 1,28 1,91 0,58
2 5
SO 0,71 0,27 0,67 0,73 0,27 0,86 0,52 0,40 0,70 2,92 1,26 1,43 0,84 1,76 2,06 1,74 2,97 0,77 2,14 0,65
KO 1,97 0,53 1,71 2,00 0,52 2,29 0,99 0,80 1,66 4,26 3,39 3,88 2,32 4,85 5,67 4,82 8,14 1,52 5,68 1,22
SnO 3,11 0,95 1,45 3,15 0,94 3,54 1,77 1,44 2,93 1,49 5,25 6,04 3,67 7,59 8,85 7,56 12,68 2,63 8,72 2,15
V O 1,86 0,55 0,64 1,88 0,56 0,93 1,08 0,80 1,57 0,70 2,84 1,71 2,20 4,60 5,39 4,59 7,80 1,43 2,44 1,37
2 5
Cr O 2,53 0,72 0,84 1,32 0,73 0,95 1,41 1,04 2,04 0,91 3,72 1,85 2,99 6,23 7,28 6,22 10,52 1,86 3,09 1,77
2 3
Co O 1,09 0,62 0,74 0,47 0,62 0,61 1,20 0,91 1,80 0,80 3,26 0,89 1,00 2,88 6,40 5,48 9,33 1,63 1,53 1,50
3 4
NiO 1,15 1,25 1,50 1,21 1,25 1,66 2,39 1,83 3,62 1,63 6,61 2,46 2,85 2,57 5,36 10,97 18,39 3,29 4,15 2,96
CuO 1,13 1,33 1,60 1,22 1,34 1,77 2,54 1,97 3,85 1,73 7,11 2,60 2,97 1,36 4,22 4,80 19,56 3,53 4,38 2,81
ZnO 1,21 1,49 1,83 1,32 1,49 1,99 2,81 2,19 4,36 1,97 8,00 2,90 3,29 1,42 2,00 5,45 21,63 3,95 4,88 2,70
As O 1,39 1,67 2,11 1,56 1,50 2,36 1,26 2,48 4,91 2,25 9,15 3,40 3,80 1,59 2,41 2,67 2,14 4,49 5,70 1,54
2 3
PbO 2,91 0,94 2,54 2,97 0,91 3,43 1,67 1,46 3,07 2,56 10,41 5,11 5,89 3,45 6,83 7,97 6,81 13,08 8,48 2,00
BaO 4,30 1,28 1,60 3,99 1,28 2,36 2,34 1,90 3,81 1,73 7,06 3,61 6,35 5,09 10,47 12,21 10,48 17,49 3,45 2,75
Cl 0,39 0,33 0,38 0,40 0,34 0,50 0,66 0,49 1,00 0,42 1,75 0,72 0,80 0,46 0,97 1,13 0,95 1,63 0,87 1,22

7.2.5 Crystals
The crystals listed in Table 5 are those preferred for the measurements, particularly for sequential-type
instruments. Other crystals could, however, be used if these are not available. In the case of PKα, the
Ge(111) crystal is recommended because it does not give second-order wavelengths. If, however, this crystal
is unavailable and a PE crystal is used, pulse height selection shall be used and the settings very carefully
selected, so that the possibility of interference from the second-order wavelength of CaKβ is minimized.
7.2.6 Counting times
After assembly of measurement conditions for all elements, and prior to analysing samples, the required
counting times for each element shall be determined as set out below.
The sensitivity, m, as c/s/%, for each element is given by
RS.MS − RB.MB
m =
CS
where
m is the sensitivity;
RS is the intensity, counts per second, from the calibration standard SynCal A; measured
...