Guidelines for sampling, preparation and analysis

of alternative fuels in the Cimpor Group

CIMPOR TEC

CIMPOR TEC | Gonalo Almeida | May 2012

Table of Contents
1 - Introduction ................................................................................................................................................. 3
2 Quality Control ........................................................................................................................................... 4
2.1 Quality Control Concept for alternative fuels ...................................................................................... 4
2.2 Inspection Plans and Equipment necessary ....................................................................................... 5
3 Sampling .................................................................................................................................................... 6
3.1 Sampling Plan ..................................................................................................................................... 9
3.1.1 Definition of the overall objectives ................................................................................................ 9
3.1.2 Definition of a lot and determining lot size ................................................................................... 9
3.1.3 Determination of the sampling procedure .................................................................................. 10
3.1.4 Determination of the number of increments ............................................................................... 13
3.1.5 Determination of minimum sample size ..................................................................................... 13
3.1.6 Determination of the minimum increment size ........................................................................... 16
3.1.7 Determination of the effective increment and sample sizes ....................................................... 19
3.1.8 Selection of distribution of increments over a lot........................................................................ 19
3.1.9 Determination of the distribution of the increments when sampling from a material flow .......... 20
3.1.10 Determination of the distribution of the increments when sampling from a vehicle(s) ............. 21
3.1.11 Implementation of sampling from a static lot ............................................................................ 22
3.2 Implementation of the sampling plan ................................................................................................ 22
4 Preparation .............................................................................................................................................. 24
4.1 Sample preparation procedure ......................................................................................................... 25
4.2 Methods for sample division ............................................................................................................. 27
4.2.1 - Riffling ......................................................................................................................................... 27
4.2.2 Strip mixing ................................................................................................................................. 28
4.2.3 Long pile ..................................................................................................................................... 28
4.2.4 Manual increment division .......................................................................................................... 29
4.2.5 Rotary divider ............................................................................................................................. 29
4.2.6 Coning and quartering ................................................................................................................ 30
4.3 Methods for reducing laboratory samples to sub-samples and general analysis samples .............. 30
4.4 Methods for the preparation of the test sample from the laboratory sample .................................... 33
4.4.1 Sequence of treatment techniques ............................................................................................ 35
4.4.2 - Guideline for choosing sample treatment techniques ................................................................. 36
4.4.3 Characteristics of the laboratory sample for chemical analysis ................................................. 42
5 Analysis ................................................................................................................................................... 43
5.1 Net / Gross Calorific Value ................................................................................................................ 43
5.2 Determination of carbon (C), Hydrogen (H) and Nitrogen (N) .......................................................... 44
5.3 Determination of total Sulphur (S)..................................................................................................... 45
5.4 Determination of Humidity - Moisture................................................................................................ 46
5.4.1 Determination of total moisture .................................................................................................. 46
5.4.2 Determination of moisture in general analysis samples ............................................................. 48
5.5 Determination of water content in liquid alternative fuels ................................................................. 50
5.6 Determination of Volatile matter........................................................................................................ 51

Guidelines for sampling, preparation and analysis of alternative fuels Page 1

CIMPOR TEC | May 2012
 5.7 Determination of Ash content ............................................................................................................ 52
5.8 Determination of major elements by XRF ......................................................................................... 53
5.9 Determination of trace elements by XRF .......................................................................................... 54
5.10 Determination of mercury (Hg) by direct analyser .......................................................................... 54
5.11 Determination of Chloride (Cl) ........................................................................................................ 55
5.12 Determination of Fluoride (F) .......................................................................................................... 55
5.13 Determination of biomass content .................................................................................................. 56
6 - Bibliography .............................................................................................................................................. 58

Guidelines for sampling, preparation and analysis of alternative fuels Page 2

CIMPOR TEC | May 2012
1 - Introduction
This document has the purpose to cover all the aspects involved in the sampling, preparation and analysis
of alternative fuels in the CIMPOR Group.
The use of alternative fuels in its different shapes and states has a growing importance in the cement
industry and his use in the CIMPOR Group is a present and future trend that the plants must be ready to
deal with it.
The cement kiln operating conditions mainly working temperature, residence time and oxidation
conditions maximise the retention of potential pollutants (such as heavy metals) without reducing the
quality of the main final product, cement.
The burning of alternative fuels in the clinker kilns implies a thorough evaluation of the production process
and a certain number of adjustments are necessary accordingly to the physical and chemical
characteristics of the fuels.
Taking into account this factors its needed to adjust the process control and to implement an adequate
delivery control for the reception of the alternative fuels, not forgetting to impose all the necessary
specifications to the suppliers of this materials.
There are several types of alternative fuels that can be in solid or liquid state, and the most common
problems in analysing them is their heterogeneity, which makes the sampling and preparation steps crucial
to obtain a representative analysis.
This factor can be less difficult if the supplier provides a homogenous material with a reduced particle size,
but this is not always available, so its important to have the correct tools to perform these tasks and obtain
a representative sample for analysis.
In the performance of the tests, besides the fact that there are no standards of these materials, it is critical
to know how to overcome the problems created by the physical state of the several alternative fuels and to
adapt the test methods in an accurate way.
The main types of alternative fuels used and tested at the CIMPOR plants are:
RDF Residue Derived Fuels: extremely heterogeneous material that may include all types of solid
urban waste, tyres and other solid fuels.
Tyres: can be delivered previously crushed or not, the presence of metal alloys causes difficulties
to the preparation and analysis.
Animal meal: the presence of fat complicates the preparation.
Residues of end-of life vehicles: heterogeneity and presence of metal alloys causes problems in
preparation.
Used fuels, solvent and hydrocarbon wastes: the contamination with water and heterogeneity of the
samples are the main problems in the preparation and analysis.
Other alternative fuels not so commonly analysed: grapes and olive waste, biomass, rice shell,
carpet wastes, electric and electronic equipments waste, etc.
Recently there has been a new classification in the European standards with the following specifications:
SRF Solid Recovered Fuels: solid fuel prepared from non-hazardous waste to be utilised for energy
recovery in incineration or coincineration plants.
Note: "prepared" here means processed, homogenised and up-graded to a quality that can be traded
amongst producers and users.
This classification includes all the solid alternative fuels that are treated like urban wastes, tyres, animal
meal that are not derived from natural sources like the solid biomass. In practice the SRF are a new
designation for RDF but for now, both classifications are still used.

Guidelines for sampling, preparation and analysis of alternative fuels Page 3

CIMPOR TEC | May 2012
2 Quality Control
The use of alternative fuels in a cement production process implies a change in the quality control scheme
in such a way that the quality of the final product and the stability of the process are not severally affected.
For this purpose its important to know the physical state of the alternative fuels to be used, which can be
delivered in the following states.
Table 1 Physical states of alternative fuels

Alternative Fuels physical state Characteristics Examples

Solid lump fuel incl. whole tires (all 3- Heavy material; cannot be Whole tires, coarse shredded
dimens. solids >50mm and 2- carried by kiln gases; burns at tires and plastics
dimens. foils >200mm) kiln inlet

Solids <50mm (3-dimens.) incl. foils Can be carried by kiln gases Tyre chips, shredded plastics
<200mm (2-dimens.) (suited for precalciner) and textiles

Solids < 5mm (3-dimens.) incl. foils Can be carried easily by kiln Plastics, impregnated saw
<50mm (2-dimens.) gases (suited for main firing) dust, animal meal

Sludge Pumpable with piston pump Petroleum sludge

Liquids Can be atomized with Waste oil, solvents

compressed air (solid particles
in liquid <2-4mm)

For solids as listed above there are only 3 categories. One category for the main burner, one category for
the calciner and one category for kiln inlet.
In the categories for the main burner and calciner a distinction has to be made between light, 2-dimensional
material such as foils which can be easier carried by the kiln gases and heavy, 3-dimensional material.
In case of mixtures of 2-dimensional and 3-dimensional material (e.g. shredded mixes of plastic foils, hard
plastics, textiles, wood and cardboard) often compromises in terms of particle size are made to reduce the
cost of shedding (e.g. <25mm for the mix which is burned at the main burner).
If such mixtures contain significant amounts of 3-dimensional material the impacts on production rate,
availability etc. will be higher because the oversize 3- dimensional pieces injected at the main firing will not
burn-out in the flame, they will fall on to the clinker bed and burn out on the material. Finally it is an
economical decision of weighing the preparation (shedding) cost against the cost of the additional impacts.

2.1 Quality Control Concept for Alternative Fuels

Avoidance and minimization of many negative impacts of alternative fuels is possible and should take
priority over compensation efforts, and so an appropriate quality control concept must therefore be
developed. Two specific sets of control procedures can be distinguished: the control of the alternative fuel
itself and increased/modified process control due to the use of alternative fuel.
Acceptance Qualification: The acceptance of an alternative fuel source has to go through a formal
process. A prerequisite for acceptance of an alternative fuel from suppliers/customers is the relay of all
possible information about the material.
Communications should include sufficient information to evaluate the alternative fuel for potential
contaminant compounds or elements (e.g. the process producing the alternative fuel), define closely
typical physical properties, potential for heterogeneity as well as the possibility of OH&S concerns.
A 'typical' alternative fuel sample from the supplier should be provided and analysed (independently from
the supplier) to confirm the specified properties.

Guidelines for sampling, preparation and analysis of alternative fuels Page 4

CIMPOR TEC | May 2012
Delivery Control: Strict, site-specific, acceptance and rejection criteria and procedures should be
established.
Quality control is material dependent with non-critical alternative fuels (e.g. low grade coal, plastics and
sewage sludge) best managed through delivery agreements with reliable (and authorized) suppliers,
regularly scheduled sampling of incoming material and periodic analysis of composite samples. Typical
analyses would include water content, calorific value, ash content and composition, granulometry, flash
point and specific elemental compositions (e.g. P2O5).
More critical alternative fuels (with potential relevance to the environment, OH&S, permit conditions, and/or
plant specifications) requires more frequent sampling (normally per delivery) and extended analysis,
including Cl, PCBs, heavy metals or other parameters deemed necessary for a 'fingerprint' analysis.
Quality control procedures for alternative fuels are described in more detail throughout this document;
these analyses may require the appropriate investment in the necessary laboratory equipment for which a
guidance table is also presented further.
Modified Process Control when using alternative fuels
With the use of alternative fuels, it is normally necessary to adapt the quality control of the processed
material stream (i.e. raw meal, kiln feed, hot meal, and clinker). General recommendations for material
control are given in the table below.

Table 2 General recommendations for Process Control changes when using alternative fuels

Material Frequency Parameter

Raw meal real time to 1 hourly Chemical composition, fineness

Kiln feed every 2 hours Chemical composition

Hot meal every 2-4 hours Cl, SO3, LOI (calcination degree)

every 2 hours Chemical composition,

Clinker mineralogy (Rietveld if available)

hourly Free lime

Table 3 Recommended Equipments for Process Control

Material Equipment Parameter

Hot meal Automatic sampler and analyser Cl, SO3

Clinker Automatic sampler and analyser Free lime

2.2 Inspection Plans and Equipment necessary

The general concept of quality control for alternative fuels was established in the previous section, and with
that information its possible to define the inspection plans for daily use.
In the files bellow mentioned with links some examples are presented for supplier specification, and for
delivery and control plans, separated by type of alternative fuel, mainly considering their physical state,
parameters to control, some classified as mandatory and others as recommended and frequency of
sampling and testing, either at the laboratory of the plant or in an outside laboratory for less frequent or
more specific tests.
Quality control for Solid Alternative Fuels: AF_Quality Control_Solids.xlsx

Guidelines for sampling, preparation and analysis of alternative fuels Page 5

CIMPOR TEC | May 2012
 Quality control for Hazardous Liquid Alternative Fuels: AF_Quality Control_Liquids.xlsx

In the table in the next page file its mentioned the equipment necessary to perform these tests including a
resume of the equipment available at the Central Laboratory of the Cimpor Group, the mandatory
equipments for Plant laboratories, and a list of recommended suppliers.

Table 4 Equipment necessary for preparation and testing of alternative fuels at Plant Laboratory

Purpose Description Mandatory at Equipment at LCGC Recommended

Plants laboratories Suppliers

Preparation of Solids Cutting Mill X Fritsch Pulverisette 25 Fritsch; Retsch

Preparation of Solids Ring Mill X Herzog HSM-100H Herzog; Retsch; Fritsch

Drying and Humidity Oven X Memmert ULM 700 Memmert, etc.

Water content Karl Fischer X Crison Titromatic KF Crison; Metrohm; etc.

1S
(only liquids)

Calorific Value Calorimeter X Parr 6400 Parr; LECO

Humidity, Ash and Thermo- LECO TGA 701 LECO

Volatiles (automatic) Gravimetric
Analyser

Ash and Volatiles Furnace X Nabertherm N7 Nabertherm; etc.

Elemental analysis CHN and S LECO Truspec CHN S LECO; Parr

(CHNS) analyser

Major elements and X-Ray X Panalytical Axios Panalytical; Bruker;

Cl Fluorescence Cement ARL

Trace elements X-Ray Axios Pro-Trace Panalytical

Fluorescence Program

Mercury analyser Mercury AMA-254 AMA (LECO)

analyser

Potentiometry Automatic Metrohm Titrino 751 Metrohm; etc.

(Fluoride) Titrator GDP

3 Sampling
The testing of alternative fuels enables informed decisions about their subsequent handling and use. In
order to carry out a test on an alternative fuel a representative and significant sample of the material is
required.
Before any sampling operation is devised it is important that the objectives for sampling are clearly
identified and subsequently well executed to ensure that the expectations of any involved parties are
recognized and satisfied.

Guidelines for sampling, preparation and analysis of alternative fuels Page 6

CIMPOR TEC | May 2012
The following procedure is based in the EN 15442 standard, which specifies methods for taking samples of
solid recovered fuels from production plants, from deliveries or from stock. This standard its not applicable
to samples that are formed by liquid or sludge, but it includes dewatered sludge.
The identification of objectives helps to define the level of testing required, e.g. thorough examination or
routine testing and in addition desired reliability of testing / assessment and frequency of testing. The
sampling objectives, along with the sequence of operations required to fulfil them are detailed in an overall
sampling plan. After a sampling plan has been prepared the sampling of solid recovered fuels (SRF's) itself
can be implemented.

Sampling procedures are provided for a range of process streams and common storage conditions. The
sampling technique adopted depends on a combination of different characteristics of the material and
circumstances encountered at the sampling location. The determining factors are:

the type of solid recovered fuel;

the situation at the sampling location / the way in which the material occurs (e.g. in a stockpile, on a
conveyor belt, in a lorry);
the (expected) degree of heterogeneity (e.g. monostreams, mixed fuels, blended fuels).
The next shows the links between the essential elements of a testing program.

Guidelines for sampling, preparation and analysis of alternative fuels Page 7

CIMPOR TEC | May 2012
Figure 1 Links between the essential elements of a testing program.

Guidelines for sampling, preparation and analysis of alternative fuels Page 8

CIMPOR TEC | May 2012
3.1 Sampling Plan
The sampling plan must comply with the defining principle of a good sampling, which is that every particle
in the lot or sub-lot to be represented by the sample should have an equal probability of being included in
the sample. When this principle cannot be applied in practice, the sampler shall note the limitations in the
sampling plan.
In the following items all the necessary elements for the development of a sampling plan are described in
detail.

3.1.1 Definition of the overall objectives

In the definition of the overall objectives of a sampling plan all parties involved must agree in the
requirements and objectives, if necessary two or more sampling plans shall be made to correspond to all
the testing samples that might be needed.
The sampling plan shall specify its primary objectives and meet all of his the requirements of objectives. If it
is not possible to meet all requirements in one single document, two or more sampling plans shall be made
in order to achieve adequate sampling plans for all objectives.

3.1.2 Definition of a lot and determining lot size

A lot for this purpose is a defined quantity of alternative fuel for which the quality is to be determined.
The definition of a lot and the determination of the lot size: shall be defined on the basis of the way the
material is used (upon delivery, upon acceptance, upon storage, etc.).
The lot size relates to the quantity of material delivered on the basis of specifications agreed by contract as
a unit.
The maximum weight of a lot or sub-lot, for sampling purposes, shall be no more than 1500 tons. If the
contracted weights are bigger it shall be split in two or more separate lots.
Several lot definitions are possible accordingly to the sampling point:
Lot from sampling from a material flow: period of production or in which a certain amount of solid recovered
fuel is transported through the material flow.
Lot in transport by a vehicle: it shall compromise the contents of the entire series of 1 or more vehicles
(lorry or truck) used.
Lot in transport by a ship: it shall compromise the contents of the entire series of 1 or more ships used, but
remember if the ship contains more than 1500 tons; it shall be divided in as many sub-lots as necessary.
Sampling from a static lot: if the material has been stored with permission from the supplier, it relates to the
quantity and specifications agreed on beforehand in a contract within a demarcated area.

Guidelines for sampling, preparation and analysis of alternative fuels Page 9

CIMPOR TEC | May 2012
3.1.3 Determination of the sampling procedure
The sampling method that shall be representative, therefore, the material shall preferably be sampled from
a moving transport medium.
There are several sampling methods that are next showed in order of preference, because the
representativeness decreases. A less representative method can only be used if a better one is not
possible.

(a) Mechanically from a drop flow.

For sampling from a drop flow, either mechanically or manually, a collection tray shall be used with the
following requirements:
- the direction of movement shall be perpendicular to the direction of movement during the sampling:
- the size of the collection tray shall be large enough to enable all the material in the drop flow to end up
in the collection tray;
- the size of the collection tray shall be such that no more than 75 % of its capacity is taken up by one
increment;
- the particles that fall on the edge of the collection tray moving through the material flow shall have an
equal chance of ending up or not ending up in the tray;
- the material of the collection tray shall not have any impact on the material sampled.

It is important that the collection tray shall be moved through the entire flow at a constant velocity when
taking the sample of a drop flow. All locations from the cross-section of the drop flow shall be sampled,
using the collection tray, during the same time period.
In practice, the most convenient way of doing this is for the collection tray to be moved through the drop
flow at a right angle to the direction of the transport system. Figure 2 is a schematic drawing of the taking of
an increment.
It is also necessary to choose a constant velocity that enables a sufficiently large increment to be taken, at
the times determined in the sampling plan in a stratified random way.

Figure 2 Schematic representation of sampling from drop flow.

Guidelines for sampling, preparation and analysis of alternative fuels Page 10

CIMPOR TEC | May 2012
(b) Mechanically from a moving conveyor.
For mechanical sampling from a moving conveyor, the equipment used must satisfy the following
requirements:
- the dimensions shall be large enough for the required increment;
- the equipment shall be sufficiently robust;
- use of sampling equipment shall not lead to chemical or physical changes in the (sampled) material;
- the sampling equipment shall sample all material that is directly in front of the sampler at the sampling
time. As the transport velocity via the conveyor increases and the velocity of the sampling equipment
decreases, the equipment will have to be placed at more of an angle to the conveyor (see Figure 3);
- the cutting action shall be executed at a constant velocity;
- the sampling equipment shall sample all material;
- the particles on the edge of the cutting face shall have an equal chance of being included or not being
included in the sample;
- the material shall be sampled across the full breadth of the conveyor using the sampling equipment.

The equipment should be set up in such a way that the increments can be taken directly, preferably at the
times drawn in a stratified random way. If just manual operation is present, the mechanical sampling shall
be put into action at the times drawn.

Figure 3 Schematic representation of sampling from the conveyor

Guidelines for sampling, preparation and analysis of alternative fuels Page 11

CIMPOR TEC | May 2012
(c) Manually from a stationary conveyor.
For manual sampling from a stationary conveyor it must be checked if the use of a sampling frame is
possible, because this is a requisite for this type of sampling.

The sampling frame is placed on the stationary

transport flow with a distance between the two
side plates of at least three times the nominal top
size
The dimensions of the sampling frame shall be
large enough for the required increment, and
shall cover the full breadth of the conveyor
(length of sampling frame).
The plates shall follow the form/cavity of the
conveyor, for example straight for a flat conveyor,
arched for a concave conveyor arch and angular
for a V-shaped conveyor; Figure 4 Sampling frame

The sampling frame shall be sufficiently robust; its use shall not lead to chemical or physical changes in the
(sampled) material.
The cutting action shall be perpendicular to the direction of transport through the positioning of the
sampling frame.
The particles on the edge of the cutting face of the sampling frame shall have an equal chance of being
included or not being included in the sample;
In the sampling of granular alternative fuels all material between the two plates in the location of the
increment shall be included.
If fluff-type alternative fuels are being sampled and the material is being transported from left to right, any
material that lies partly on the left of the left plate (after the increment has been taken) is counted as part of
the increment. Any material that lies partly on the right of the right plate when the increment is being taken
is not counted as part of the increment.
For the sampling stop the conveyor, preferably at the times drawn in a stratified random way, then place
the sampling frame with a sawing movement perpendicular to the direction of movement of the conveyor,
and take the increments.

(d) Manually from a drop flow

The sampling procedure is similar to the described for the mechanical sampling as mentioned above in
section (a).

(e) Manually from a vehicle

Divide the lot or store into as many strata of equal dimensions as the number of increments to be taken.
Inside each stratum, take an increment the size of the effective increment size in a (preferably) random
way.

(f) From a temporary store location.

Follow the same procedure described for sampling from a vehicle.

Guidelines for sampling, preparation and analysis of alternative fuels Page 12

CIMPOR TEC | May 2012
3.1.4 Determination of the number of increments
An increment is a portion of fuel extracted in a single operation of the sampling device.
The minimum number of increments shall be 24, for example, for a lot size of 120 tons on the average
every 5 tons an increment has to be taken.
Its possible to take more increments, like for instance if more sample material is required, or if its easier to
stratify the lot in a different number of strata.

3.1.5 Determination of minimum sample size

The minimum sample size or dimension of the sample required during sampling and sample preparation
from the point of view of preserving its representativeness.
The minimum sample size is equal to the effective increment size multiplied by the number of increments,
and is linked directly to the nominal top size.
The minimum sample size shall be large enough in order to have enough particles from the lot being
sampled in order to obtain a representative sample for the total lot.
The following factors shall be determined in order to determine the minimum sample size:
1) the nominal top size of a particle, d95.
Aperture size of the sieve used for determining the particle size distribution of the sample through which at
least 95% by mass of the material passes.

2) the shape factor f;

For materials which are more or less granular or have a smaller nominal top size smaller than 50 mm is not
useful to determine the shape factor. The shape factor can be taken as 1,0 for all more or less granular
materials.
For these materials the shape factor will very close to 1,0 and the practical saving in terms of sample size
reduction will not outweigh the effort of determining the shape factor. If the shape factor is not determined a
value 1,0 shall be used.

To determine the shape factor the following equation shall be used:

Where:
3 3
f is the shape factor, in mm /mm ;
V95 is the maximum volume of a fluff particle (a mass fraction of 95 % of the particles are smaller than V95),
in mm3 (where V = l b h);
d95,l is the maximum length of a fluff particle (a mass fraction of 95 % of the particles are smaller than d95l),
in mm.

The shape factor f is not constant, but depends on the type of fluff-type material.

3) the particle density p

Determination of the bulk density should be done using the standard for the material we are testing, for
example in Solid Recovered Fuels it shall be determined as specified in CEN/TS 15401.

Guidelines for sampling, preparation and analysis of alternative fuels Page 13

CIMPOR TEC | May 2012
4) the distribution factor g;
The distribution factor g, which corrects for the distribution in the particle size, depends on the ratio
between the nominal top size, d95, and the minimum particle size, d05.
Which value for g shall be used shall be deducted from Table 1.
Table 4 Distribution factor
Ratio d95/d05 g
d95/d05 > 4 0,25
2 d95/d05 4 0,50
1 d95/d05 < 2 0,75
d95/d05 = 1 1,00

Note: d05 is the minimum particle size (a mass fraction of 5 % of the particles is smaller than d05).
In the case of fluff-type solid recovered fuels, there is generally a large distribution in the particle size. In
many cases this distribution results in a g of 0,25. Almost all granular materials, on the other hand, have a
similar particle size. Consequently, g for pellets is generally 1,00.

5) the factor p (= the fraction of the particles with a specific characteristic, such as contaminant);
The factor p refers to the fraction of the particles with a specific characteristic (such as contaminant). If no
value is known or determined for the material, a fixed value of 0,10 shall be maintained for this factor.

6) the coefficient of variation cv.

The coefficient of variation cv of 0,1 shall be accepted when taking the aforementioned assumptions to
determine a correct minimum sample size.

Calculation of the minimum sample size

The parameters above mentioned shall be used in the Equation to calculate the minimum sample size.
( )
( )
Where:
mm is the mass of the minimum sample size, in kilograms as received;
d95 is the nominal top size of a particle, in mm;
Note If the recovered fuels are of the fluff type, the value for d95l can be used here. For granular materials,
both the d95 of the particles in the pellets and the d95 of the pellets themselves can be used.
3 3
f is the shape factor, in mm /mm ;
3
is the average particle density of the particles in the solid recovered fuel, in kg/m as received;
g is the correction factor for distribution in the particle size;
p is the fraction of the particles with a specific characteristic (such as a specific contaminant), in kg/kg, and
is equal to 0,1;
cv is the coefficient of variation, here set to 0,1.
Some examples of the way in which the minimum sample size can be calculated for both fluff-type and
granular solid recovered fuels are given the next page.

Guidelines for sampling, preparation and analysis of alternative fuels Page 14

CIMPOR TEC | May 2012
Example for granular solid recovered fuel:
A granular solid recovered fuel consisting of pellets has a particle density of 1 000 kg/m3 and a bulk density
of 300 kg/m3. Approximately a mass fraction of 95 % of the pellets has a diameter of 20 mm; however, the
d95 of the particles in the pellets amounts to 10 mm.
A value of 0,1 can be used for the coefficient of variation cv and the factor p. The distribution in the grain
size is small, so a value of 1,0 can be assumed for g. As pellets are more or less granular, a value of 1,0
can be used for the shape factor.
This leads to a minimum sample size with a weight of: 1,0 /(6 109) 103 1,0 1 000 (1-0,1)/(0,12
0,1) = 0,471 kg and a volume of 0,471/300 = 0,001 571 m3 = 1,57 l.

Example for fluff-type solid recovered fuel:

The density (particle density) amounts to 1 000 kg/m3 and the bulk density amounts to 80 kg/m3, V95
3
amounts to 95 000 mm , d95 amounts to 190 mm, and d05 amounts to 50 mm.
The shape factor f then amounts to 95 000/1903 = 1/72,2. The ratio d95l/d05l amounts to approximately 3,8;
accordingly a value of 0,50 applies to g. A value of 0,1 is maintained for the factor p and the coefficient of
variation cv.
Based on the above values, this leads to a minimum sample size with a weight of:
/(6 109) (190)3 1/72,2 1 000 0,50 (1-0,1)/(0,12 0,1) = 22,384 kg, and a volume of 22,4/80
1 000 = 280 l for the fluff-type solid recovered fuel concerned.

Quick determination of minimum sample size

A quick determination of the minimum sample size is also available in tables in the EN 15442 standard:
Table for fluff-type solid recovered fuels: shows the minimum sample size for a large distribution in the
3
particle size. For the particle density a value of 1 000 kg/m has been assumed, a value of 0,05 has been
used for the shape factor; the distribution factor of the particle size g has been set at 0,25. A value of 0,1
has been assumed for the factor p, as for the coefficient of variation cv.
Table for granular solid recovered fuels: shows the minimum sample size for samples with a small
3
distribution in the grain size. For the particle density a value of 1 000 kg/m has been assumed, a value of
0,05 has been used for the shape factor; the distribution factor of the particle size g has been set at 0,25. A
value of 0,1 has been assumed for the factor p, as for the coefficient of variation cv.
Table 5 Table for determination of minimum sample size in fluff-type solid recovered fuels.

Guidelines for sampling, preparation and analysis of alternative fuels Page 15

CIMPOR TEC | May 2012
3.1.6 Determination of the minimum increment size
The minimum increment size is minimum dimension or size of the increment that is taken from a lot, from
the point of view of preserving its representativeness.
The product of the minimum increment size and the number of increments to be taken should never be
smaller than the minimum sample size.
The minimum increment size depends directly of the type of sampling used and there are different
procedures for the following:
- mechanical sampling from a drop flow;
- manual sampling from a drop flow;
- sampling from a conveyor, in this case no distinction is made between mechanical sampling from a
moving conveyor and manual sampling from a stationary conveyor;
- sampling from static lots or vehicles (lorry, truck or ship).
The size of an increment shall be large enough so that all particles have a chance to be part of the
increment. Besides this for increments of material flows and conveyors the particles over the whole breadth
of the material flow or conveyor shall have an equal chance of ending up in the increment.

1) Mechanical sampling from a drop flow

For mechanical sampling from a drop flow the dimensions of the sampling equipment shall meet the
requirement that the breadth of the increment shall be equal to at least three times the d 95 of the material
for sampling, unless the nominal top size d95 is smaller than 3 mm.

The mass of the increment size is calculated using the following equation:

Where:
mi is the mass of the increment size, in kg;
d is the drop flow, in kg/s;
b is the breadth of the increment, in m;
vc is the velocity at which the collection tray moves through the drop flow, in m/s ( 0,6 m/s).

Providing that the breadth of the increment is equal to the minimum increment breadth (three times d95),
and the velocity of the material collection is equal to the maximum velocity (0,6 m/s), the mass of the
minimum increment size for material with d95 greater than 3 mm can be calculated as follows:

where
mm is the mass of the minimum increment size, in kg;
d is the drop flow, in kg/s;
d95 is the nominal top size, in mm.

Note: The above equation can only be used for a (virtually) constant conveyor load. Major discontinuities in
the mass flow or conveyor load can therefore be avoided as far as possible.

Guidelines for sampling, preparation and analysis of alternative fuels Page 16

CIMPOR TEC | May 2012
2) Manual sampling from a drop flow
The breadth of the increment and the collection tray shall be at least equal to the breadth of the drop flow.
If it is assumed that the time when only part of the drop flow ends up in the collection tray (while the tray is
moving in and out of the drop flow) is negligible when compared with the time when the material flow falls
entirely into the collection tray, the mass of the increment size is equal to:

Where:
mi is the mass of the increment size, in kg;
d is the drop flow, in kg/s;
tm is the sampling time, in s.

The sampling time shall be chosen in such a way that the required number of increments results in
sufficient material in order to meet the minimum sample size.
Assuming that the collection tray is moved through the drop flow at a constant velocity, the mass of the
increment size is equal to:

Where:
mi is the mass of the increment size, in kg;
d is the drop flow, in kg/s;
bs is the breadth of the drop flow in the direction in which the collection tray is moved, in metres;
vc is the velocity at which the collection tray moves through the drop flow, in m/s ( 0,6 m/s).

The above equation shall only be used for a (virtually) constant conveyor load. Major discontinuities in the
mass flow or conveyor load shall therefore be avoided as far as possible.

3) Sampling from a conveyor

For determining the minimum increment size when sampling from a conveyor no distinction is made
between manual sampling from a stationary conveyor and mechanical sampling from a moving conveyor.
As regards the determination of the dimensions of the sampling equipment, these shall be determined
using the following requirements:
- the breadth of the increment shall be equal to at least three times the d 95 of the material for sampling,
if this has a d95 greater than 3 mm;
- for materials that have a d95 smaller than 3 mm, the breadth of the increment shall be equal to 10 mm;
- the length of the increment shall be as large as the breadth of the material flow on the conveyor (and
no more than equal to the breadth of the conveyor);
- the height of the increment shall be equal to the height of the conveyor load at the point where the
increment is taken.

Guidelines for sampling, preparation and analysis of alternative fuels Page 17

CIMPOR TEC | May 2012
If the d95 is greater than 3 mm and the breadth of the increment is chosen to be equal to three times the
d95, the minimum increment size shall be determined by the following equation:

Where:
mm is the mass of the minimum increment size, in kg;
b is the breadth of the increment, in m;
G is the conveyor load, in kg/m;
d95 is the nominal top size (d95), in mm.

If the d95 is smaller than 3 mm, the minimum increment size shall be determined by the following equation:

Where:
mm is the mass of the minimum increment size, in kg;
b is the breadth of the increment, in metres (= 0,01 m);
G is the conveyor load, in kg/m.
The above equation can only be used for a (virtually) constant conveyor load. Major discontinuities in the
mass flow or conveyor load should therefore be avoided as far as possible.

4) Sampling from static lots or vehicles (lorry, truck or ship)

The minimum increment size shall be in terms of volume at least three times the nominal top size in all
dimensions of the particles with the nominal top size.
To determine the mass of the minimum increment size for solid recovered fuels with a d95 greater than 3
mm, the next equation shall be used:

Where:
mm is the mass of the minimum increment size, in kg;
d95 is the nominal top size (a mass fraction of 95 % of the particles are smaller than d95), in mm;
b is the bulk density of the flow, in kg/m3.

If the d95 of a material is smaller than 3 mm, the next equation shall be used:

Where:
mm is the mass of the minimum increment size, in kg;
b is the bulk density of the flow, in kg/m3.

Guidelines for sampling, preparation and analysis of alternative fuels Page 18

CIMPOR TEC | May 2012
3.1.7 Determination of the effective increment and sample sizes
The effective increment size shall meet the three requirements below:
the effective increment size shall be at least as large as the minimum increment size;
the effective increment size shall be large enough in order to obtain a sample of the minimum sample
size with the selected number of increments;
the effective increment size shall be large enough in order to obtain a sample sufficiently large enough
for the analysis in the laboratory.

The effective sample size shall meet the three requirements below:
the effective sample size shall be at least as large as the minimum sample size;
the effective sample size shall be large enough in order to equal at least the minimum increment size
multiplied by the number of increments;
the effective sample size shall be large enough for the analysis in the laboratory.

Only if the nominal top size is more than 40 mm the effective sample size may be reduced in order to
downscale the sample size to realistic proportions.
The reduction of the particle size and the sample size shall be done according to the standards in use, for
instance for SRFs the EN 15443, but the sampling report and the analyses report shall clearly mention this
deviation and state that therefore the testing results are less representative.

3.1.8 Selection of distribution of increments over a lot

The increments shall be taken scattered all over the lot. Each particle in the lot shall have an equal chance
of ending up in the sample.
The following sampling methods (arranged in decreasing order of preference) shall be used:
a) stratified random sampling;
b) stratified sampling.
Stratified means that a quantity of material (expressed as a mass or volume) or a time interval is divided
into a specific number of equal strata (sections).
The increments can be located over static lot by either stratified or random sampling is applied. The stars
show where an increment can be taken.

Figure 5 Stratified sampling in static lot Figure 6 Stratified random sampling in a static lot

Key: 1 Boundaries strata ; 2 ground surface static lot.

Guidelines for sampling, preparation and analysis of alternative fuels Page 19

CIMPOR TEC | May 2012
3.1.9 Determination of the distribution of the increments when sampling from a
material flow
If the sampling is planned from a material flow, the time when each increment is to be sampled shall be
determined. This requires the following approach:
a) the lot shall be either defined as a period of production or as the period in which a certain amount of
alternative fuel is transported through the material flow;
b) this period shall be divided by the number increments in order to obtain equal strata (sub periods) for
each increment;
c) the exact time within each stratum shall be determined either random or fixed in the middle of the period.

The next figures show how the increments can be located over the lot which is specified as a certain
production period if stratified or random sampling is applied.
The arrows show where an increment can be taken. T 0 is the starting time and tend is the end time of the
production period.

Figure 7 Stratified sampling in a production period

Figure 8 Stratified random sampling in a production period

Guidelines for sampling, preparation and analysis of alternative fuels Page 20

CIMPOR TEC | May 2012
3.1.10 Determination of the distribution of the increments when sampling from a
vehicle(s)
If the sampling is planned from a vehicle, the locations in the vehicle(s) shall be determined. This requires
the following approach:
a) the lot shall be defined as a number of vehicles with a minimum of one vehicle;
b) from each vehicle the same number of increments shall be taken;
c) the number of increments per vehicle multiplied by number of vehicle shall at least the minimum number
of increments;
d) if it is possible and sufficiently safe for the sampler to do so, this increment may be taken directly from
the vehicle. If this is not possible, the increment may be taken directly after the material has been unloaded
in accordance with the system for sampling from a material flow or from a static lot;
e) for each vehicle selected, take the increments from the top, middle and bottom of the material
alternately, i.e. increment 1 from the top of the material in the first vehicle selected, increment 2 from the
middle of the material in the second vehicle selected, increment 3 from the bottom of the material in the
third vehicle selected, and so on.

For example the following describes the complete sampling procedure:

A lot of 3 106 kg solid recovered fuels consisting of pellets, will be transported to a purchaser by a total of
120 lorries, with an average load of 25 000 kg per lorry. The maximum lot size is 1,5 106 kg. The quantity
above the maximum permitted lot size should be regarded, for sampling purposes, as a new lot or sub-lot.
The total lot of 3 106 kg is therefore split into two separate sub-lots for sampling of 1,5 106 kg each. For
each sub-lot of 1,5106 kg, the minimum number of increments that are taken is 24.
Sampling requires the following approach:
Step 1) The weight of the total lot is 3 106 kg. For sampling purposes, this is regarded as two sub-lots
weighing 1,5106 kg each. The first sub-lot will be transported by a series of 60 lorries;
Step 2) At least 24 increments should be taken for each sub-lot of 1,5 106 kg, but because 30 increments
is easier regarding the number of 60 lorries it is decided that 30 increments will be taken.
Therefore from the first sub-series and then from every consecutive sub-series of 60/30 = two lorries, a
lorry is selected for sampling in a (preferably) random way each time.
For example, from the first sub-series (lorry 1 lorry 2), lorry number 1 is sampled, from the second
subseries (lorry 3 lorry 4) lorry number 4 is sampled, from the third sub-series (lorry 5 lorry 6) lorry
number 5 is sampled, and so on;
Step 3) The increment should be taken randomly from the top of the material in lorry number 1, from the
middle of the material in lorry number 4, from the bottom of the material in lorry number 5, and so on.
The dimensions of each increment shall be at least equal to the already determined minimum or effective
increment size. Steps 1 to 3 shall be completed again for the sampling process involving the second sub-lot
of 1,5 106 kg.
Note 1 If for example 2 increments should be taken from each of the lorry, it would be very useful to take
one increment at the beginning of unloading of the lorry (short stop during loading) and the second
increment at the middle/end of unloading the lorry in accordance with the system for sampling from a
material flow or static lot.
Note 2 If the dimensions of a lot are such that the number of vehicles used to transport the lot is equal to or
less than the (minimum) number of required increments, at least two (or more when necessary) increments
should be taken per vehicle.

Guidelines for sampling, preparation and analysis of alternative fuels Page 21

CIMPOR TEC | May 2012
3.1.11 Implementation of sampling from a static lot
Implementation of sampling from a static lot or store requires the following approach:
a) determine the dimensions of the lot or store;
b) divide the lot or store into as many strata of equal dimensions as the number of increments to be taken;
c) for each stratum, determine in a (preferably) random way the location where the increments will be
taken;
d) for each location, take the increments from the top, middle and bottom of the material alternately.

For example: a lot of solid recovered fuel of 1 106 kg, which shall be sampled, has a volume of 3 000 m3.
The lot is 60 m long, 40 m wide and 1,25 m high. Twenty-four increments are taken from the lot.
The following approach is used for the sampling process:
2
1) the surface area of the lot is 2 400 m , and the number of increments to be taken is 24. The lot can
2
therefore be divided into 24 sections of 10 m by 10 m (100 m );
2
2) for each stratum of 100 m , the increment location is determined (the x and y values) in a random way
and one increment is taken;
3) for each location drawn, take the increments from the top, middle and bottom of the material alternately,
i.e. increment 1 from the top of the material within stratum 1, increment 2 from the middle of the material
within stratum 2, increment 3 from the bottom of the material within stratum 3, and so on.

3.2 Implementation of the sampling plan

The following things shall be performed by the sampler before the actual sampling:
a) control of all relevant data in the sampling plan;
b) control of the nominal top size. If the nominal top size is larger than stated in the sampling plan, the
project leader shall be notified and a new effective sample and increment size shall be determined and
mentioned at deviations;
c) control of the lot size. If the lot size is deviating from the sampling plan it shall be adjusted;
d) control whether the lot appears to be one lot and does not origin from two or more different lots. Make
sure the lot does not origin from two or more different lots. If the lot origin is from more than one lot, each
lot shall be sampled separately;
e) make a situation sketch or pictures from different angles;
f) if steps a) to e) require adjustments of the sampling plan these shall be reported in the sampling report
and sufficiently justified.

The following things shall be performed during sampling:

a) if sample is taken manually it is important that the increment is taken in one movement and that no
material is removed from the increment after it is taken. In case the size of an increment is too large a new
increment shall be taken;
b) the sampler shall not judge the quality of an increment nor discard it and shall not take a new one if he
thinks the increment is not representative. This will influence the representativeness of the sample. In every
case the upper surface of the material to be sampled shall always be removed to avoid material that has
dried and/or become contaminated;
c) if during the performance of the sampling anything is performed deviating from the sampling plan this
shall be mentioned in the sampling report in the section for deviations. The deviations shall be sufficiently
motivated.

Guidelines for sampling, preparation and analysis of alternative fuels Page 22

CIMPOR TEC | May 2012
The following things shall be done after sampling:
a) the sampling plan shall be finished by the sampler;
b) the sampling containers shall be closed and sufficiently labelled.

Some examples of other sampling equipment

Figure 9 Examples of sampling equipment

Guidelines for sampling, preparation and analysis of alternative fuels Page 23

CIMPOR TEC | May 2012
4 Preparation
The main purpose of sample preparation is that a sample is reduced to one or more test portions that are in
general smaller than the original sample, in a way that analytical tests can be performed.
The main principle for sample preparation is that the composition of the sample as taken on site shall not
be changed during each step of the sample preparation and that possible requirements of the analysis
methods to be performed are obeyed.
Each sub-sample shall be representative for the original sample. To reach this goal every particle in the
sample before sample preparation shall have an equal probability of being included in the sub-sample
retained after sample preparation.
Also the loss of moisture and other volatile components shall be minimised if these components are
analysed or influence the properties to be analysed.
The following procedure is based in the EN 15443 standard, that specifies the methods for the preparation
of the laboratory sample for solid recovered fuels, in other type of alternative fuels adjustments must be
made taking into account the physical state of the material, like for instance in liquids.
The laboratory sample is the sample sent to or received by the laboratory, for many tests further
preparation is necessary to obtain the test sample, these methods will also be covered further in the
document.
The sample preparation technique adopted depends on a combination of different characteristics of the
material and circumstances encountered at the sampling location.
The determining factors are:
the type of alternative fuel;
the physical behaviour of the specific alternative fuel;
the (expected) degree of heterogeneity (e.g. monostreams, mixed fuels, blended fuels).
Two basic methods are used during the sample preparation. These methods are:
Sample division;
Particle size reduction of the sample.

For granular materials generally the principle of the third-power law is accepted and shall be respected at
each sample division step. The equation for this third power law is the following:

Where:
m is the mass retained after each sample division step in g;
d95 is the nominal top size in mm;
3
is a constant over the whole sample preparation procedure for a particular material in g/mm .
The value and unit of constant is fixed by the nominal particle size, d 95, and the sample size, m, of the
sample before sample preparation.
Example: a sample of 10 kg of SRF fluff has d95 of 50 mm. For the analysis a test portion of 5 g is required.
The third power law results in = 10 000 g divided by 50 mm to the third power. The value of is now 0,08
3
g/mm . Using this value in the equation for a reduced sample size results in a nominal top size for the
3
particles in the test portion of 3,97 mm (cube root of 5,0 g divided by 0,08 g/mm ).

Guidelines for sampling, preparation and analysis of alternative fuels Page 24

CIMPOR TEC | May 2012
Below in the table are shown the figures.
Table 6 Reduction figures for example of sample preparation

(g/mm )
3
m (g) d95 (mm)
10 000 0,08 50
5 0,08 3,97

The reduction factor of the nominal top size can be calculated by dividing the current nominal top size by
the proposed nominal top size after size reduction.
For SRF or other alternative fuels, however, many materials turn out to be far from granular. For example in
fluff the particles turn out to be predominantly flat. Therefore, for solid recovered fuels, a correction can
made for non-granular materials.
If moisture content of the material (as sampled) is to be determined, a separate moisture analysis sample is
taken (as there is a risk of reducing the moisture content by sample preparation operations).
Also if a sub-sample is required for the determination of mercury content, a separate mercury analysis
sample is taken (as there is a risk of reducing the mercury content by sample preparation operations).
For materials that have to be examined for moisture and mercury content, care shall be taken for any
significant heat build-up and risk of loss of moisture and mercury.

4.1 Sample preparation procedure

The sample preparation procedure should be performed according to the following steps:

Step 1: Collection of the relevant information of the material to be sampled

In the first step of sample preparation information shall be collected about the material to be sampled:
a) the minimum sample size out of the sampling plan;
b) the actual size of the sample, m 0;
c) the nominal top size of the sample;
d) the shape factor of the sample;
e) the requirements in terms of size reduction for the analysis that need to be performed;
f) the required amounts for each of the size fractions and their restrictions to the sample preparation
methods.
Sample preparation prepares a sample for a number of tests which will be performed on the sample. Some
of these tests require no particle size reduction or drying of the material in the sample.
Other tests require very tiny homogenized sub-samples with small particle sizes. A sample preparation plan
shall have to meet all these requirements.

Step 2: Making a sample preparation plan

Sample preparation is a combination of sample division and particle size reduction. Until what level a
sample of alternative fuel shall be prepared on site depends on available equipment on site, the
requirements of the laboratory and the preferences of the client of the sampling activities.
These two essential activities are specified below.
Sample division
The aim of sample division of a sub-sample is to reduce the mass remaining sub-sample or to make
several duplicate sub-samples out of one original sub-sample available.

Guidelines for sampling, preparation and analysis of alternative fuels Page 25

CIMPOR TEC | May 2012
 During the performance of sample division it is of eminent importance that the minimum sub-sample size
shall be retained in order to sustain the representatively of the sub-sample for the original combined
sample.
Particle size reduction of a sample
The aim of particle-size reduction is to reduce the nominal top size of the particles in order to reduce the
minimum sub-sample size without losing representatively.
During the performance of particle size reduction it is important that all materials are included. Leaving out
metals causes significant errors in the measured values of these and possibly accompanying metals.
In order to make a valid sample preparation plan it shall contain at least the information mentioned in the
table below, which specifies all activities that shall be performed during the whole process of sample
preparation.
The structure of the table can be adjusted to the properties of the alternative fuel and the equipment
available in the laboratory where it takes place. This means that e.g. steps can be skipped or added if the
nominal top size is already smaller than e.g. 30 mm or the coarse shredder results in a nominal top size
different from 30 mm, etc.
Table 7 Sample preparation plan

Description Method of Used Mass Mass after Nominal Nominal Shape Shape Mass to Purpose of
reduction technique before reduction top size top size factor factor be product of
and reduction before after before after withheld this
apparatus reductio reduction reduction reduction for reduction
n analysis step

Step Sample division of the Sample Long pile 183 kg 143 kg 300 mm 300 mm 0,05 0,05 40 kg Bulk density,
1 combined sample in a sub- division reduction durability of
sample for further sample pellets,
preparation and a sub- particle size
sample of untreated distribution,
material etc.

Step Particle size reduction in Particle size shredder 143 kg 143 kg 300 mm 30 mm 0,05 0,5
2 order to make further reduction to
sample division possible < 30mm

Step Sample division in order to Sample Long pile 143 kg 1,43 kg 30 mm 30 mm 0,5 0,5 Sub-
3 reduce the remaining mass division reduction samples for
or obtain sub-samples as moisture
general analyses sample content, etc.

Step Particle size reduction, the Particle size Cutting 1,43 kg 1,43 kg 30 mm 1,0 mm 0,5 1,0
4 remaining sub-sample in a reduction to mill
sub-sample for further > 1,0 mm
sample preparation sub-
samples as general
analyses sample

Step Sample division of the Sample Incremental 1,43 kg 1,0 g 1,0 mm 1,0 mm 1,0 1,0 1,0 g Sub-samples
5 remaining sample material division method for ash,
into the required general calorific
analysis sample(s) value,
chemical
analysis, etc.

Step Particle size reduction in Particle size

6 order to make further reduction to
sample division possible < 0,25 mm

Step Sample division of the Sample Sub-samples

7 remaining sample material division for analysis
into the required test where < 0,25
portions mm is
required

Note: Mass can only be withheld during a mass reduction sample division step and not during a size reduction step.

Guidelines for sampling, preparation and analysis of alternative fuels Page 26

CIMPOR TEC | May 2012
During every sample division step, every particle in the sample before sample division shall have an equal
probability of being included in the sub-sample retained after sample division.
For non-granular materials with a shape factor significantly smaller than 1,0 a correction shall be made for
the changing shape factor.

Step 3 Performing the sample preparation

The sample preparation shall be performed as specified in the plan, and during the actual sample
preparation the following procedures shall be performed at any times:
Homogenize the sub-sample material at every step thoroughly;
Make sure no material of the sub-sample gets lost;
If possible do not leave out any components. If its necessary to remove hard substances like metal
(steel) to avoid damage to the mill that shall be reported. The report shall include the weight
percentage and the type of substance removed from the sample.
At all times enough sample material shall be withheld in order to perform all analyses.

4.2 Methods for sample division

The following methods shall be used to produce a smaller but representative sample.
The remaining sample size shall not be less than the minimum sample size. The sampling implements and
sample division equipment used here shall have apertures with a minimum size of 3 times d 95.

4.2.1 - Riffling
This may be used for materials that can be passed through the riffle without bridging. It is not suitable for
fluff or other materials containing elongated particles, or for wet materials. Brittle materials should be
handled with care to avoid fine materials to be produced.
Put the whole of the combined sample into one of the containers of the riffle so that it is evenly distributed
throughout the container. Place the other two containers in position under the riffle.
Pour the contents of the first container down the centre line of the riffle. Pour the solid recovered fuel
sufficiently slowly that bridging does not occur.
Do not move the container from side to side, this would cause the end slots to receive less fuel. Discard the
fuel that falls into one of the containers. Repeat the riffling process until a sub-sample of the required size is
obtained.

Figure 10 Example of a riffle Box

Guidelines for sampling, preparation and analysis of alternative fuels Page 27

CIMPOR TEC | May 2012
4.2.2 Strip mixing
This may be used for all materials, and will be a convenient method when a combined sample is to be
divided into a small number of laboratory samples.
Place the whole combined sample on a clean, hard surface and homogenise it by mixing with a shovel.
Position a vertical plate at each end of the strip. Use a shovel to distribute the material along the length of
the strip as evenly as possible, working from end to end and from both sides.
The more passes from end to end, while building the strip, will improve the quality of the division. The
length to width ratio of the strip shall not be less than 10:1.
Obtain a laboratory sample by taking at least 20 increments from locations evenly spaced down the length
of the strip. Take each increment by inserting two plates vertically into the strip and removing all the
material from between the plates.
The two plates should be inserted the same distance apart each time so that each increment contains the
same quantity of material. The distance between the plates should be chosen so that the method yields a
laboratory sample of the required size.

Figure 11 Strip mixing

4.2.3 Long pile

This may be used for all materials, and will be a convenient method when a combined sample is to be
divided into several laboratory samples.
In this method, the whole of the combined sample is divided into a number of laboratory samples. Start by
forming a strip as in the strip mixing method. Choose the number of laboratory samples to be produced so
that they will have the required mass.
Arrange this many containers conveniently near the strip. Use a shovel of a size such that each laboratory
sample will receive at least 20 shovelfuls. (If the laboratory samples are each to be of mass m laboratory sample
kg, then the shovel should have a capacity of no more than m laboratory sample / 20 kg).

Take shovelfuls of material always from the same

end of the pile, being careful not to leave fine
particles behind, and place a shovelful in each
container in turn until all the material has been
used.

Figure 13 Example of shovel

Figure 12 Example of a Scoop

Guidelines for sampling, preparation and analysis of alternative fuels Page 28

CIMPOR TEC | May 2012
4.2.4 Manual increment division
This is suitable for pellets and other solid recovered fuels with a small particle size that can be worked with
a scoop.
Place the whole combined sample on a clean, hard surface and homogenise it by mixing with a scoop. Use
the scoop to spread the combined sample into a rectangle with a thickness of not more than three times the
nominal top size, and lightly mark the surface of the rectangle with the scoop to divide it into at least 20
parts.
Use the scoop and a bump plate to take an increment from each of the 20 parts, inserting the scoop to the
bottom of the heap each time, and combine the increments to form the required sub-sample.

Figure 14 Manual increment division

4.2.5 Rotary divider

A mechanical method to reduce the mass of a sample is using a rotary sample divider. The rotary sample
divider shall have a feeder device adjusted so that the divider rotates at least 20 times while the sample is
being divided.

Figure 15 Examples of rotary sample dividers

Guidelines for sampling, preparation and analysis of alternative fuels Page 29

CIMPOR TEC | May 2012
4.2.6 Coning and quartering
This may be used for materials such as pellets, chips and fluff that can be worked with a shovel. It is
suitable for producing sub-samples of these materials down to approximately 1 kg.
Place the whole combined sample on a clean, hard surface. Shovel the sample into a conical pile, placing
each shovelful on top of the preceding one in such a way that the sample runs down all sides of the cone
and is evenly distributed and different particle sizes become well mixed.
Repeat this process three times, forming a new conical pile each time. Flatten the third cone by inserting
the shovel repeatedly and vertically into the peak of the cone to form a flat heap that has a uniform
thickness and diameter and is no higher than the blade of the shovel.
Quarter the flat heap along two diagonals at right angles by inserting the shovel vertically into the heap.
Discard one pair of opposite quarters. Repeat the coning and quartering process until a sub-sample of the
required size is obtained.
Note: This method is only used in case the other methods are not applicable.

Figure 16 - Quartering

4.3 Methods for reducing laboratory samples to sub-samples and general

analysis samples
In the following pages the methods and equipment necessary for the reduction of the particle size of the
samples.

1) Initial sample division.

If the initial mass of the laboratory sample exceeds what is needed for the test sample the laboratory
sample may be divided using one of the methods already presented.

2) Initial mass determination.

Before the sample is subject to any handling or treatment it shall be weighted and recorded.

Guidelines for sampling, preparation and analysis of alternative fuels Page 30

CIMPOR TEC | May 2012
3) Pre-drying
Pre-drying of wet samples is carried out to minimise moisture loss in the subsequent sample-division
processes, to facilitate the sample preparation processes, and to minimise biological activity. If it is
necessary to dry a sample by heating, it shall be dried in an oven at a temperature not exceeding 40 C.
If the moisture content of the original sample is unimportant, e.g. if only particle size distribution is to be
determined or if a separate moisture analysis sample is sampled as well, the calculation of the moisture
loss can be omitted. Also in this case it is not necessary to obtain completely equilibrium with the
temperature and the humidity in the laboratory.
All samples (including those that have been dried by heating) shall be spread out in a tray no more than a
few particles deep, and left for at least 24 h in the laboratory, until constant weight is achieved.
For among others coarse materials 24 h retention time under laboratory conditions is not enough to reach a
constant weight. Monitor the moisture content by placing the sample or a sub sample on an electronic
balance during the retention time in the laboratory.
If the information about the moisture losses during the predrying is required, determine the moisture
content from an undried and a dried sub-sample and calculate the difference.
Calculate the moisture loss during pre-drying as a percentage of the initial mass of the sample:

4) Coarse cutting (particle size reduction to < 30 mm)

If the material contains particles retained on a 30,0 mm sieve, separate the sample into a coarse fraction
(retained on the 30,0 mm sieve) and a fine fraction (passing the 30,0 mm sieve) and process the coarse
fraction using a coarse cutting mill so that it passes the 30,0 mm sieve.
Materials in the sample e.g. hard alloy steel and stones shall (if necessary) be removed from the sample
before coarse cutting to protect the grinder from damage. Magnetic materials can be removed by using a
magnet. These materials shall be weighed and reported in the test report.
For some types of SRF pre-treatment by freezing may be necessary before grinding. Place the sample in a
freeze resistant bucket and treat the sample with carbon dioxide ice or liquid nitrogen. Remove the sample
immediately from the bucket to the inlet of the moving cutting mill (grinder). Depending on the material
other mills than the cutting mill can be used.
Re-combine the processed coarse fraction and the fine fraction and homogenise the sample. If required,
the above procedure may be used to achieve particle size reduction to a size below 30 mm by the use of a
suitable coarse cutting mill and sieve.

Figure 17 Cutting mill at LCGC

Guidelines for sampling, preparation and analysis of alternative fuels Page 31

CIMPOR TEC | May 2012
5) Sample division of <30 mm material
Use one of the manual methods previously described. The masses of retained subsamples shall respect
the values stated for the particle size reduction.
Determine the masses of the containers into which the sub-samples are to be put before starting the
sample division.
Seal the container immediately if a sub-sample is required for moisture content determination, or another
test for which it is important not to lose moisture.
Determine the mass of each sub-sample and record this mass. When a sub-sample is to be used as a test
portion, the sub-sample shall have a mass not less than the minimum test-portion mass specified in the
appropriate test method.

6) Particle size reduction of < 30 mm material to < 1mm

When a sub-sample of material is required with a nominal top size of about 1 mm, use a cutting mill to
reduce the whole of a sub-sample to this particle size.
Carry out the process in several steps, if necessary, using a finer sieve in the mill at each step, finishing
with an appropriate sieve.
If subpopulations with a different physical behaviour (like sand and/or metal particles) are expected then
care shall be taken especially for the segregation of particles, so the use of sample dividers is
recommended.
Grinding or cutting to a nominal top size of 1 mm can cause problems, because some materials are difficult
to grind and subsequently generate large amounts of heat when reduced in particle size. This heat affects
the composition of the sample and may cause a loss of e.g. moisture and mercury.
Therefore, it may be necessary to grind or cut the sample under cryogenic conditions or with a slow rotating
grinder. If the temperature in the sample increases to more than 70 C the particle size reduction shall be
performed under cryogenic conditions.
In case of cryogenic cutting or grinding, the mills or cutters shall be suitable for grinding or cutting at very
low temperatures. Cooling shall be carried out with either solid carbon dioxide (-79 C) or liquid nitrogen (-
196 C).
The sample shall be mixed or immersed in the cooling medium. In some cases it is necessary to cool the
mill or cutter as well with the cooling medium in order to keep the temperatures low enough.
Another strategy of avoiding an increasing temperature is to decrease to the throughput or rotation
frequency of the mill or the cutter. This may also be applied as long as the temperatures do not rise above
70 C.
If the material contains granular particles, there is a risk that they will rotate in the mill or get stuck in the
screen. Likewise if the material contains very elongated particles, some of these particles may come to rest
on the screen and not pass through.
Examine the mill after the milling is completed. If any such particles are found, grind them manually until
they pass the sieve and add this material to the sub-sample.
Spread out the sub-sample in a tray no more than a few mm deep, and leave it for at least 4 h in the
laboratory, until it reaches equilibrium with the temperature and humidity in the laboratory.
Material that has been subjected to the particle size reduction operations described in this clause shall not
be used for the determination of the total moisture content of the fuel.

Guidelines for sampling, preparation and analysis of alternative fuels Page 32

CIMPOR TEC | May 2012
7) Sample division of < 1 mm material
Homogenise the material in the sample container by stirring with a spatula, and use the spatula to take out
the quantity required.
Also in this step if subpopulations with a different physical behaviour (like sand and/or metal particles) are
expected then care shall be taken especially for the segregation of particles, so the use of sample dividers
is recommended.
The sub-sample size taken to further grinding shall be large enough to ensure representativeness.

8) Particle size reduction of < 1 mm material to < 0,25 mm

When a sub-sample is required with a nominal top size of 0,25 mm, use a cutting mill to reduce the whole
of the sub-sample to this size.
Feed the cutting mill with small portions of material from the general analysis sample and let each portion
pass the 0,25 mm screen to prevent excess heat generation.
Note: Depending on the material other mills than the cutting mill can be used.

9) Storage and labelling of sub-samples

Sub-samples shall be stored in tightly-closed containers. Each sub-sample shall be labelled with a unique
identification containing the identification of the sample from which it was obtained.

4.4 Methods for the preparation of the test sample from the laboratory sample
The laboratory sample is reduced in particle size and mass using different apparatus and procedures
depending on the type of sample and the type of analysis to which the sample will be submitted.
Care should be taken to avoid loss of material and contamination of the sample via the air, by dust, or by
the use of an apparatus. In an apparatus can occur 3 types of contamination: abrasion, cross-
contamination or chemical release, also any chemical reaction due to generated heat in source of error and
material alteration.
The procedure followed is the one described in the EN 15413 standard, specifically for solid recovered
fuels, which are the majority of the alternative fuels used in the cement industry.
Preparation of the test portion can be a complex process, because of a number of factors: sample type and
its physical state, amount of laboratory sample, type and number of determinations to be carried out, etc.
The prepared test portions shall satisfy the following requirements at the same time:
each test portion shall be a representative of the laboratory sample;
the amount and the physical state (e.g. particle size) of each test portion have to comply with the
requirements of the respective analytical technique;
for each test portion, no losses of and no contamination with respective analytes of interest should
occur.
The preparation of the test portions from the laboratory sample, which has been taken according to the
sampling plan, is related to the requested analytical determinations. This means that, if needed, contact
has to be established among all involved parties.
The preparation of test portions in the laboratory will frequently involve a sequence of operations such as
homogenisation, fraction separation, drying, reducing particle size and sub sampling.
A number of decisions on the specific order of these operations for a particular laboratory sample have to
be made. In some cases, the sequence of operations to be applied is rather straightforward, but in more
complicated cases (e.g. when several determinations with different requirements have to be performed) it
can be critical to choose the right sequence of such operations.

Guidelines for sampling, preparation and analysis of alternative fuels Page 33

CIMPOR TEC | May 2012
In order to define the operations to be applied to a laboratory sample to produce one or more
representative test portions, three main steps have to be considered:
a) Definition of analytical requirements
1) methods to be used;
2) amount of test portions necessary;
3) the quantity and properties of the test portions necessary for each analytical procedure;
4) preservation requirements (e.g. time frame, temperature, addition of reagents).

It is recommended to prepare at least five times the amounts needed as test portions for the test sample.

b) Definition of sequence of operations

Then, the sequence of operations shall be defined according to the flow sheet in the next page, based on
the properties of the laboratory sample and the requirements of the analytical procedures: each single
operation of this sequence has to be considered like an independent module; available modules are:
1) fraction separation;
2) drying;
3) particle size reduction;
4) homogenisation;
5) sub-sampling.

For practical reasons it is recommended to group the parameters in a way that test samples with similar
requirements can be prepared for several parameters. The same test sample may be used for different
parameters if it fulfils the necessary requirements.
Frequently, different determinations have to be performed on the laboratory samples. In those cases,
modules have to be combined and/or repeated to obtain sub-samples, finally resulting in different test
portions.

c) Choice of appropriate procedures

According to the requirements of the respective analytical techniques and the properties of the sample the
appropriate sample treatment technique has to be chosen within each module by following the instructions
presented in the next pages.

Guidelines for sampling, preparation and analysis of alternative fuels Page 34

CIMPOR TEC | May 2012
4.4.1 Sequence of treatment techniques
The next flow sheet describes the procedure to enable decisions on the specific order of treatment
operations for a particular laboratory sample in order to yield in representative test portions.
It shall be applied on the starting laboratory sample and repeated on all sample fractions or sub-samples
subsequently obtained during the preparation, in an iterative cycle until all analytical requirements are
fulfilled.

Figure 18 Flow sheet Sequence of operations

Guidelines for sampling, preparation and analysis of alternative fuels Page 35

CIMPOR TEC | May 2012
4.4.2 - Guideline for choosing sample treatment techniques
The preparation of test samples from a laboratory sample will frequently involve a sequence of operations
such as homogenisation, phase separation, drying, particle size reduction and sub-sampling. Specific
forms of these operations are described in the next pages.

1. Homogenisation
Before each operation that implies sub-sampling, a homogenisation step is required, in order to guarantee
that all sub-samples or sample fractions have the same properties and composition.
The homogenisation technique to be used is chosen depending on the properties of the sample. In many
cases before homogenisation, particle size reduction may be necessary.

Manual homogenisation

When to use it: When not to use it:

For samples that form layers because of the

presence of particles of different density;
Generally usable; For samples with particles of such a large size that
In cases when mechanical homogenisation could homogenisation by manual mixing cannot be
lead to loss of volatile compounds of interest reached;
(mercury). For samples that form layers because of the
presence of particles with large differences in
particle size.

Procedure: Mix the sample with an appropriate tool (e.g. shovel, pestle and mortar). If there is a risk of
losses of volatile substances the manual homogenisation has to be done very carefully.

Mechanical homogenisation

When to use it: When not to use it:

Generally usable especially in cases when manual For samples that form layers because of the
homogenisation is not suitable; presence of particles of different density; in this
case, if homogenisation is not possible, separate
In cases of sample of large particle size; and treat each layer as a different sample;
For solid samples containing particles of nearly the When the apparatus may heat the sample and loss
same density (e.g. for materials that do not form of volatiles can occur during this process; in this
"layers" after shaking). case, a manual homogenisation shall be performed.

Procedure: Operate according to the manufactures instructions.

Note: For samples of small particle size, it may be also possible to use a ball-mill without balls for
homogenisation.

Guidelines for sampling, preparation and analysis of alternative fuels Page 36

CIMPOR TEC | May 2012
2. Fraction separation
For heterogeneous samples, depending on their nature and on the determinations of interest, one or more
techniques of fraction separation can be applied to obtain two or more different sub-samples that are to be
analysed separately.
For samples consisting of different fractions, separation of some fractions may be necessary. For this
reason, the weight of each separated sub-sample shall be directly or indirectly measured after the
separation, in order to allow a final weighed combination of different fractions' analysis results.
The test report shall clearly state the technique(s) used for fraction separation, the weight and analytical
results related to all sub-samples obtained from fraction separation, as well as the weighted results.
In cases of visible heterogeneity of separable fractions, the separation of different fractions may be
necessary, especially if this can make subsequent particle size reduction, homogenisation and sub-
sampling easier.

Manual separation

When to use it: When not to use it:

When several fractions can be distinguished;

When contamination or losses of analytes of interest
When non-crushable fractions (e.g. copper wire) may occur.
are existing.

Procedure: Manually select macroscopic pieces of different nature and store them in separate containers,
either by hand (with protective gloves) or by using appropriate tools (e.g. tweezers, magnet).

Sieving

When to use it: When not to use it:

When separation of fractions of different particle

size is necessary; When contamination or losses of analytes of interest
For checking the particle size of the sample or the may occur.
particle size distribution.

Procedure: Sieve the sample by shaking either by hand or apparatus through sieves with appropriate
mesh size and material.

3. Drying
Depending on the nature of the sample and the specific requirements of the test portion, a drying step
might be needed during sample treatment for test portion preparation.
For the determination of water content a separate sub-sample may be necessary. Drying is very likely to
introduce analytical errors for volatile compounds, and should be avoided when not strictly required.
If a sub-sample or test portion for volatile compounds determination is to be dried, the actual drying
technique shall be selected in order to minimise losses of volatile compounds.
The test report shall clearly state the technique(s) used for drying, along with the weight of sub-sample(s)
before and after each drying step.
It is likely that a certain drying technique is not applicable for all requested determinations. In such cases,
different sub-samples shall be dried in different ways, choosing the appropriate sequence of techniques for
each one.

Guidelines for sampling, preparation and analysis of alternative fuels Page 37

CIMPOR TEC | May 2012
The drying time will depend on the technique chosen, the thickness of the layer of the sample, the nature of
the sample, moisture content of the sample and of the air and the rate of ventilation.
The grade of dryness that shall be reached with the drying step depends on the subsequent treatments to
be applied to the sample. Typically, it is not necessary to wait until constant weight: e.g. the sample shall
be just dried enough to make crushing, grinding etc. possible.

Air drying at room temperature

When to use it: When not to use it:

In every case where drying can be reached in When time is critical, and the properties of the
appropriate time without alterations of analytes of sample do not allow a good drying in a reasonable
interest; time at room temperature, and a higher temperature
In case of mercury determination. drying step can be safely applied.

Procedure: Spread the sample on the trays in a thin layer and allow it to get dry enough. Care shall be
taken in order to minimise possible contamination e.g. by dust. The use of a desiccator may accelerate the
drying process for small amounts of (sub-)samples.

Oven drying at 40C

When to use it: When not to use it:

When time is critical, and the properties of the When the properties of the sample do not allow for
sample do not allow for good drying in a good drying in a reasonable time at this
reasonable time at room temperature; temperature, and a higher temperature drying step
In case of mercury determination. can be safely applied.

Procedure: Spread the sample on the trays in a thin layer and allow it to get dry in the oven at 40C. Air
renewal may accelerate the drying process, provided that it does not cause loss of dust-like particles..

Note: Oven drying at other temperatures may be used if they fit with analytical procedures to be applied.

Oven drying at 105C

When to use it: When not to use it:

On sub-sample for which mercury is not In case of mercury determination;

determined. In cases where auto-ignition can be expected.

Procedure: Spread the sample on the trays in a thin layer and allow it to get dry in the oven at 105C. Air
renewal may accelerate the drying process, provided that it does not cause loss of dust-like particles.

Guidelines for sampling, preparation and analysis of alternative fuels Page 38

CIMPOR TEC | May 2012
4. Particle size reduction
In order to achieve a homogeneous and representative test portion, one or more particle-size reduction
steps might be needed. The choice of the technique to be used depends strongly on the nature of the
sample and on the particle size needed.
Typically, particle-size reduction is a multi-step operation that implies the use of a sequence of different
techniques; in some cases, it might be necessary to repeat a step until the sample reaches the requested
particle size.
Particle-size reduction is a critical step in sample preparation because of potential loss of mercury due to
heating, because of loss of dust-like material and because of contamination coming from the equipment
itself or from other samples. Care shall be taken in selecting the appropriate equipment and keeping it
clean.
The test report shall clearly state the technique(s) and operating conditions used for particle-size reduction.
Non-crushable fractions (e.g. copper wire) shall be separated, weighed and, if needed, analysed as
separate sub-samples.

Crushing/grinding

When to use it: When not to use it:

When representative sub-samples cannot be taken

because of large particle size;
When not applicable because of the nature of the
When the particle size of the sample is larger than sample (e.g. soft materials);
the allowed inlet particle size for the milling or
grinding equipment; When contamination or losses of compounds of
interest may occur by the equipment.
When the analytical requirements demand a
particle size in the mm order of magnitude.

Procedure: Break large pieces of the sample and/or crush the sample with appropriate apparatus
according to the manufacturers instructions to the desired particle size.

Freeze crushing

When to use it: When not to use it:

When the sample contains large amount of

plastics;
When representative sub-samples cannot be taken
because of large particle size;
When contamination or losses of compounds of
When the particle size of the sample is larger than interest may occur by the equipment.
the allowed inlet particle size for the milling or
grinding equipment;
When the analytical requirements demand a
particle size in the mm order of magnitude.

Procedure: Wrap the sample in a polyethylene container. Fill a Dewar vessel with sufficient liquid nitrogen
and immerse the wrapped sample in the liquid nitrogen. Allow the container to stand until the liquid
nitrogen no longer boils vigorously. Cool for approximately 10 min. After complete cooling, retrieve the
container from the liquid nitrogen and break large pieces of the sample with a hammer and/or crush the
sample with appropriate apparatus according to the manufacturers instructions to the desired particle
size. Alternatively commercial apparatus for freeze crushing are available.

Guidelines for sampling, preparation and analysis of alternative fuels Page 39

CIMPOR TEC | May 2012
Milling

When to use it: When not to use it:

When representative sub-samples cannot be taken When the initial particle size is too coarse: a
because of large particle size; crushing step is required in these cases;
When the requested particle size is less than 1 When not applicable, because of the nature of the
mm. sample.

Procedure: Mill the sample according to the mills manufacturers instructions until it reaches the desired
particle size; in order to prevent losses of mercury, care shall be taken to avoid excessive heating of the
sample during milling: a sequence of short and low-speed millings is to be preferred to a long and/or high-
speed treatment; it is necessary to let the equipment cool down between each milling operation and the
subsequent one. The use of a freeze-head mill can minimise the loss of mercury.
Note: If the sample has a plastic consistency, freezing it down to low temperatures (e.g. 20C to 30C)
can make it easier to mill.

Cutting

When to use it: When not to use it:

When representative sub-samples cannot be taken

because of large particle size;
When contamination with compounds of interest
When the material of the sample is not hard may occur by the equipment.
enough to allow crushing or milling, e.g. plastic,
textile.

Procedure: Cut the sample according to the cutting mills manufacturers instructions to the desired
particle size; in order to prevent losses of mercury care shall be taken to avoid excessive heating of the
sample during cutting; it is necessary to let the equipment cool down between each cutting operation and
the subsequent one. To minimise loss of mercury a cooled cutting mill should be used or manual cutting
by scissors may minimize the loss.

Freeze Cutting

When to use it: When not to use it:

When the sample has a plastic or fibred

consistency;
When representative sub-samples cannot be taken
When contamination from the equipment may occur.
because of large particle size;
When the analytical requirements demand a
particle size in the mm order of magnitude.

Procedure: If necessary, wrap the sample in a polyethylene container. Fill a Dewar vessel with sufficient
liquid nitrogen and immerse the (wrapped) sample in the liquid nitrogen. Allow the container to stand until
the liquid nitrogen no longer boils vigorously. Cool for approximately 10 min. After complete cooling,
retrieve the sample from the liquid nitrogen, cut it with a preferably cooled cutting mill according to the
manufacturers instructions to the desired particle size.

Guidelines for sampling, preparation and analysis of alternative fuels Page 40

CIMPOR TEC | May 2012
5. Sub-sampling
The laboratory sample usually has to be divided into different test portions for the analyses. Most analytical
techniques allow only small quantities of test portions to be analysed. Laboratory samples, conversely, can
be in some cases very large, so only a small, representative portion of them shall be taken for analysis.
The sub-sampling shall be performed in a way that the obtained sub-samples are as representative as
possible, taking into account the quantity of test portion to be analysed, the quantity of laboratory sample,
particle size and homogeneity.
If the sample is not homogeneous enough, or if particle size is too coarse, a representative portion cannot
be taken: one or more particle size reduction steps followed by homogenisation steps are needed.

Manual division of solid samples by quartering

When to use it: When not to use it:

In case of a large amount of laboratory sample;

When the sample is dry enough and there are no
When the sample contains clods that cannot be
clods in it;
divided.
If there is no evidence of heterogeneous
distribution of chemical constituents.

Procedure: Spread the sample on a flat surface coated by an inert sheet in a circular shape; mix it with a
shovel and make a cone. Divide the sample in four slices, e.g. with a sheet metal cross, discard two
opposite slices, and mix thoroughly the remaining part of the sample.
This sequence shall be repeated until the requested quantity of sub-sample is reached, provided that the
obtained sub-sample is still representative of the original sample.

Mechanical division of solid samples

When to use it: When not to use it:

When the sample is dry enough and there are no

clods in it;
When the particle size is small enough for the used When the sample contains clods that cannot be
apparatus; divided.
If there is no evidence of heterogeneous
distribution of chemical constituents.

Procedure: Mechanical sample splitters can achieve similar or better results compared to manual
division. Divide the sample by using the sample splitter, e.g. riffle box, tyler divider, according to the
manufacturers instructions.

Guidelines for sampling, preparation and analysis of alternative fuels Page 41

CIMPOR TEC | May 2012
4.4.3 Characteristics of the laboratory sample for chemical analysis
The samples of alternative fuels for chemical analysis must fulfil certain requirements, and as an example
the next table from the EN 15413 standard shows what is mandatory for the laboratory samples of Solid
Recovered Fuels in the mentioned European Standard.
A maximum amount of laboratory sample of 10 kg and maximum particle size of 1 cm is established on the
basis of number and type of parameters to be determined, sample representativity and practical reasons
for handling samples.
In the following table the requirements are summarised both for single or grouped chemical parameters.

Table 8 Requirements for the laboratory sample for the analysis of SRF

Guidelines for sampling, preparation and analysis of alternative fuels Page 42

CIMPOR TEC | May 2012
5 Analysis
There is a wide range of specifications used for the characterization of alternative fuels, in the following we
will describe the tests needed for this classification with special detail on the issues that must be taken into
consideration with alternative fuels.

5.1 Net / Gross Calorific Value

The test method is basically the same as mentioned in the international standards for coal and coke
samples, but there are some adjustments to be made accordingly to the physical and chemical properties
of the alternative fuel.
In LCGC we follow the ASTM D5865 standard for coals and cokes, using a Calorimeter Parr 6400 and
accordingly to the internal instruction (IOL 107), but there is also a European standard for this test in solid
recovered fuels (EN 15400) that uses the same principle.
A weighed portion of the analysis sample is burned in high-pressure oxygen in a bomb calorimeter under
specified conditions. The effective heat capacity of the calorimeter is determined in calibration experiments
by the combustion of certified benzoic acid under similar conditions, accounted for in the certificate.
The gross calorific value is calculated from the corrected temperature rise and the effective heat capacity of
the calorimeter, with allowances made for contributions from ignition energy, combustion of the fuse(s) and
for thermal effects from side reactions such as the formation of nitric acid.
Furthermore, a correction is applied to account for the difference in energy between the aqueous sulphuric
acid formed in the bomb reaction and gaseous sulphur dioxide, i.e. the required reaction product of sulphur
in the sample.
The net calorific value is obtained by calculation from the gross calorific value at constant volume
determined on the analysis sample and requires information about the moisture and hydrogen contents of
the analysis sample.
In testing of alternative fuel the majority of this sample may present some difficulty to burn due to low
Calorific Value. In such conditions the use of physical or chemical combustion aids is recommended, but
its necessary to know correctly the calorific value of them:
Physical combustion aids: combustion bags or combustion capsules.
Chemical combustion aids: benzoic acid or paraffin oil.
In LCGC its used Benzoic Acid with known Calorific Value (CV) of 6318 cal/g, which is mixed
proportionally with samples difficult to burn.
Another issue is the influence of sample weight, because materials with low CV can lead to wrong test
results if a too high weight is chosen and no combustion aid is used.
In liquid samples with high contamination of water the Benzoic Acid doesnt ignite because it absorbs
water. To solve this problem we need to add a water absorber to facilitate the burn, in LCGC we use LECO
reagent COM-AID which is basically Aluminium Oxide.

Figure 19 Physical combustion aids Figure 20 Calorimeter Parr 6400

Guidelines for sampling, preparation and analysis of alternative fuels Page 43

CIMPOR TEC | May 2012
5.2 Determination of carbon (C), Hydrogen (H) and Nitrogen (N)
The test method is basically the same used for other fuels and mentioned in the international standards.
At LCGC we follow the ASTM D 5373 standard for coals and cokes and the internal instruction IOL 643.
The instrument used is a LECO Truspec CHN, and there is a European Standard for SRF (EN 15407)
that indicates the same test method.
The method is based on the complete oxidation of the sample which converts all organic substances
into combustion products. The resulting combustion gases pass through a reduction furnace and are
swept into the chromatographic column by the carrier gas (helium) where they are separated and
detected quantitatively by appropriate instrumental gas analysis procedures.
The samples are held in a suitable container (tin or other crucible) and then dropped inside the quartz
tube furnace at about 1 000 C in an oxygen stream for complete oxidation in the presence of a catalyst
layer. Excess oxygen is removed by contact with copper, while nitrogen oxides are reduced to
elemental nitrogen.
The main problem in the analysis of alternative fuels is to ensure that the results fall inside the
calibration curves (lower C and H, higher N). For that it might be necessary to adjust the amount of
sample so that the analysis is correct.
In the analysis of samples with fast combustion, like plastic residues, biomass and waste fuels, its
necessary to add a product to delay of the ignition, in LCGC its used the LECO reagent COM-AID
which is basically Aluminium Oxide.

In this equipment the analysis of alternative

fuels causes many contaminations in the
several absorbers used, so its necessary to
replace them frequently and check for
accumulations inside the instruments.

Figure 21 LECO Truspec CHN

Guidelines for sampling, preparation and analysis of alternative fuels Page 44

CIMPOR TEC | May 2012
5.3 Determination of total Sulphur (S)
For the determination of total Sulphur (S) at LCGC its used the test method described at ASTM D 4239
and also in the internal instruction (IOL 642), which is a high-temperature combustion method with
Infrared Absorption Detection.
The sample is burned in a tube furnace at a minimum operating temperature of 1350C in a stream of
oxygen to oxidize the sulphur. Moisture and particulates are removed from the gas by traps filled with
anhydrous magnesium perchlorate.
The gas stream is passed through a cell in which sulphur dioxide is measured by an infrared (IR)
absorption detector. Sulfur dioxide absorbs IR energy at a precise wavelength within the IR spectrum.
All other IR energy is eliminated from reaching the detector by a precise wavelength filter. Thus, the
absorption of IR energy can be attributed only to sulphur dioxide whose concentration is proportional to
the change in energy at the detector.
The equipment at LCGC is a LECO Truspec S, and for a correct analysis a set of Certified Reference
Materials (CRM) is used to obtain a good calibration curve.
For laboratories without this type of equipment there is also some analytical methods, for instance at
LCGC we can also use the method described at ASTM D 3177 and in the internal instruction (IOL 622),
and there is also the method in the European Standard for SRF (EN 15408).
In both cases the sample is oxidized by combustion in a bomb containing oxygen under pressure,
similar to the one used for Calorific Value determination, and the sulphur is precipitated as BaSO 4 from
oxygen-bomb calorimeter washings, and the precipitate is filtered, ashed and weighed.
The testing of alternative fuels in the instrument used at LCGC, LECO Truspec S, most be done
carefully taking into consideration some important aspects.

The main issue is to ensure that the results fall into

the calibration curve and it might be necessary to
adjust the amount of sample used.
In samples difficult to burn add a combustion aid,
like the LECO reagent COM-CAT, 95% Tungsten
(VI) Oxide and 5% Potassium Dihydrogen-
phosphate.
In liquid samples add a water absorber, like the
LECO reagent COM-AID, Aluminium Oxide.
In this equipment the analysis of alternative fuels
causes many contaminations in the several
absorbers used, so its necessary to replace them
frequently and check for accumulations inside the
instrument.
Figure 22 LECO Truspec S

Guidelines for sampling, preparation and analysis of alternative fuels Page 45

CIMPOR TEC | May 2012
5.4 Determination of Humidity - Moisture
The determination of humidity or moisture in samples of alternative fuel is important to evaluate the
burning behaviour of the fuel in the kiln.
The several basis in which the results of fuel tests can be presented also provide keen information for
its end use, namely:
dry basis: calculation basis in which the fuel is free from moisture.
as received basis: calculation basis for material at delivery.
dry ash free basis: calculation basis in which the fuel is free from moisture and inorganic matter.
as determined basis: moisture of the sample in the condition that the tests were done.
As it was showed before, the preparation of alternative fuels usually include some drying steps
necessary to prepare the sample for analysis, it may be useful to remove just the amount of water that
could interfere with other processes involved (e.g. during crushing or milling).
For the end determination of the total quantity of water in the fuel (moisture) it might be necessary to
obtain a sub-sample or test portion for moisture determination before every drying step during the
preparation. Like it was mentioned in the preparation section, the temperature and conditions in which
the drying process is done should avoid the introduction of analytical errors due to loss of volatile
compounds.

5.4.1 Determination of total moisture

The reference method for the determination of total moisture, the as received basis, in solid recovered
fuels is presented in a Technical Specification CEN/TS 15414-1. Since the total moisture content of
solid recovered fuels is not an absolute value and therefore standardised conditions for its determination
are indispensable to enable comparative determinations.
The term moisture content when used with recovered materials can be misleading since solid recovered
materials, e.g. biomass, frequently contains varying amounts of volatile compounds (extractives) which
can evaporate when determining moisture content by oven drying.
The sample of recovered fuel is dried at a temperature of 105 C in air atmosphere until constant mass
is reached. The percentage of moisture is calculated from the loss in mass of the sample. The method
includes a procedure for the correction of buoyancy effects.
For some materials present in solid recovered fuels there is a risk of self-ignition when drying at 105 C,
because of that the LCGC considers a good practice to perform this test at 40 or 60C, retaining the
sample in the oven for approximately two days, but maintaining the same criteria for sample stability.

Sampling and sample preparation

1 - The sample shall be taken and prepared in accordance with the EN 15442 and EN 15443. It shall be
delivered into the laboratory in sealed water resistant and airtight containers or bags.
Note: Precautions should be carried out to ensure that the moisture content remains constant during
preparation of the sample. Coarse materials, for example, small wood and chunk wood, should be
prepared by using equipment appropriate for the fuel type, e.g. slow rotation grinder, shredder, to a
thickness of maximum 30 mm for the test material.
2 - The sample shall be weighed immediately after the sample preparation. The sample mass shall be
at least 300 g but preferably greater than 500 g.
Note: Solid recovered fuels are heterogeneous materials in many cases. Therefore, a sample size of
minimum 300 g is necessary to obtain representative test portions.
Note: For large particle size samples with a nominal top size of 100 mm, a sample mass of 1 kg to 2 kg
should be preferred.
3 - During the course of its preparation, the sample may be pre-dried; in this case the total moisture
content shall be calculated using the equation presented further in the calculations description.

Guidelines for sampling, preparation and analysis of alternative fuels Page 46

CIMPOR TEC | May 2012
Procedure
1 - Weigh an empty clean dish or tray to the nearest 0,1 g, transfer the sample from the container or bag
2
to the dish or tray and spread the sample evenly, allowing about 1 g of sample to 100 mm of surface
2
area of the dish or tray respectively of about 0,5 g of sample per 100 mm in the case that the bulk
density of the sample is less than 100 kg/m3.
Weigh an identical empty clean dish or tray (reference dish respectively tray) to the nearest 0,1 g. In
case of moisture left on the inner surfaces of the bag or container, this amount of moisture shall be
included in the calculation of the moisture content.
Dry the sample packing (container, bag etc.) in the drying oven and weigh the packing before and after
drying. If the packing material cannot resist a temperature of 105 C, it shall be allowed to dry at room
temperature by placing it open in the laboratory.
Note: A reference dish respectively reference tray is included in the procedure for a correction of
buoyancy. To avoid absorption of moisture from the atmosphere, the dish or tray with the dried sample
is reweighed when still hot.
The mass of a dish or tray when still hot is, due to buoyancy, less than the mass of the cold dish or tray.
The magnitude of the buoyancy effect depends on the size and the mass of the dish or tray.
2 - Weigh the dish or tray together with the sample. Place the loaded dish or tray together with the
reference tray in the drying oven controlled at (105 2) C. Heat the dish or tray until constant mass is
reached as specified bellow.
Note: For some materials present in solid recovered fuels there can be a risk of self-ignition when drying
at 105 C.
3 - Solid recovered fuels are hygroscopic and therefore the loaded dish or tray together with the
reference dish or tray shall be re-weighed to the nearest 0,1 g when still hot within 10 s to 15 s to avoid
absorption of moisture. Use heat-insulating material on the balance pan to protect it from direct contact
with the hot dish/tray.
Mass constancy is reached if the change of mass not exceeds 0,2 % of the total loss in mass during a
further period of heating at (105 2) C, over a duration of 60 min. The drying time required depends on
the particle size of the sample, the rate of atmosphere change in the drying oven, the thickness of the
sample layer etc.
Generally the drying time should not exceed 24 h to prevent unnecessary losses of volatile compounds.
The required drying time should be determined in pre-tests on similar fuel types with comparable
particle size.

Calculation
The moisture content, Mar, in the solid recovered fuel, as received, expressed as mass fraction in
percentage, shall be calculated using the following equation:
( ) ( )
( )
Where:
m1 is the mass of the empty dish or tray, in grams;
m2 is the mass of the loaded dish or tray before drying, in grams;
m3 is the mass of the loaded dish or tray after drying, in grams;
m4 is the mass of the reference dish or tray before drying (mass at room temperature), in grams;
m5 is the mass of the reference dish or tray after drying (mass when still hot), in grams;
m6 is the mass of moisture associated with the packing, in grams.

Guidelines for sampling, preparation and analysis of alternative fuels Page 47

CIMPOR TEC | May 2012
If the sample was pre-dried before this moisture determination, calculate the total moisture, MT,
expressed as a percentage by mass, using the following equation:

( )
Where:
Mp is the loss of moisture caused by pre-drying of the original sample, in percentage by mass;
Mr is the residual moisture determined in the pre-dried sample by this procedure, in percentage by
mass.

Moisture content on dry basis: The relation between moisture on dry basis, Ud, or wet basis, Mar,
expressed as mass fraction in percentage, shall be calculated using the following equations:

5.4.2 Determination of moisture in general analysis samples

For the determination of moisture in the general analysis samples, in the as determined basis, already
prepared and with a reduced particle size, the method is simpler according to the EN 15414-3 European
Standard.
The analysis sample of recovered fuel is dried at a temperature of 105 C under air atmosphere,
nitrogen atmosphere or vacuum conditions. The percentage of moisture is calculated from the loss in
mass of the test sample. If the sample material is susceptible to oxidation (at 105 C), drying in nitrogen
atmosphere or vacuum conditions is performed.
Automatic equipments may be used if the method is validated by parallel measurements, and if it fulfils
all the requirements regarding sample size, heating procedure, temperature, atmosphere and weighing
accuracy.

Procedure
A minimum of two determinations shall be carried out on the test sample.
Dry an empty weighing dish with its lid at (105 2) C until constant mass is reached and allow it to cool
to room temperature in the desiccator.
Weigh the weighing dish with its lid to the nearest 0,1 mg.
Add minimum 1 g of the analysis sample into the weighing dish in an even layer and weigh the weighing
dish with its lid plus sample to the nearest 0,1 mg.
Heat the uncovered weighing dish and its lid together with the sample at (105 2) C until constant
mass is reached. Constancy in mass is defined as a change not exceeding 1 mg in mass during a
further period of heating at (105 2) C over a period of 60 min.
If the sample material is susceptible to oxidation at the given temperature, dry in nitrogen atmosphere or
under vacuum conditions. The drying time required is usually between 2 h to 3 h.
Note: For some materials present in solid recovered fuels there can be a risk of self-ignition when drying
at 105 C.
Replace the lid while the weighing dish is still in the drying oven. Transfer the weighing dish and its
contents to the desiccator. Allow it to cool to room temperature. Weigh the weighing dish and its lid with
the sample to the nearest 0,1 mg.
Since small particle size recovered fuels are very hygroscopic, weigh rapidly once the sample is cooled,
because their moisture content will vary with change of humidity of the atmosphere and therefore, the
moisture of the analyses sample should always be determined simultaneously when portions are
weighed out for other analytical determinations, for example, calorific value, carbon, nitrogen.

Guidelines for sampling, preparation and analysis of alternative fuels Page 48

CIMPOR TEC | May 2012
Calculation
For each determination, the moisture content, Mad, in the analysis sample, as analysed, expressed as
mass fraction in percentage, shall be calculated using the following equation:
( )
( )

Where:
m1 is the mass of the empty weighing dish plus lid, in grams;
m2 is the mass of the weighing dish plus lid plus sample before drying, in grams;
m3 is the mass of the weighing dish plus lid plus sample after drying, in grams.
The test result for each individual determination shall be calculated on analysed basis to two decimal
places and for reporting purposes, the mean value of the individual test results shall be calculated and
rounded to the nearest 0,1 %.

Tests at LCGC
This determination at LCGC is done in an automatic equipment, a Thermo Gravimetric Analyser (TGA)
701 from LECO, which also allows to perform an automatic proximate analysis with the determination of
moisture, volatile matter and ash content.

in TGA its the 1st step of a test that also

includes the volatile matter and the ash
content.

Figure 23 TGA analyser at LCGC

Some issues must be mentioned in this analysis:

For samples of animal meal we consider this test also has the humidity as received, because in the
preparation at the oven at 80C it gains weight due to the fat present.
This gravimetric method cannot be used for liquid waste fuels, because at this temperature also the
volatiles are removed and give a false result.
Dont use this method for very volatile samples that burn easily like the ones mentioned in the
preparation.

Guidelines for sampling, preparation and analysis of alternative fuels Page 49

CIMPOR TEC | May 2012
5.5 Determination of water content in liquid alternative fuels
The determination of humidity in liquid alternative fuels is not possible, its necessary to use another
method to have a quantification of the amount of water that is present in these materials. Water content
will affect the heating value of fuels directly and can contribute to instability in the operation of the kiln.
The waste fuels used in the cement industry are often the result of the washing of oil containers and can
be very heterogeneous, composed by solvents, spent oils, paints or pigments, substances with lower
volatilization temperature than water and present in different matter states.
The LCGC its able to do this determination by two different methods based on ASTM standards:
ASTM D 5530-94 (2009): Standard Test
Method for Total Moisture in Hazardous
Waste Fuel by Karl Fischer Titrimetry.
The Water Determination Test (Karl Fischer
Method) is designed to determine water content
in substances, utilizing the quantitative reaction
of water with iodine and sulfur dioxide in the
presence of a lower alcohol such as methanol
and an organic base such as pyridine.
In volumetric Karl Fischer, iodine is added
mechanically to a solvent containing the sample
by the titrators burette during the titration.
Water is quantified on the basis of the volume
of Karl Fischer reagent consumed.

Figure 24 Karl-Fischer Titrator at LCGC

However the Karl Fischer method has not been very well succeeded in the analysis of hazardous waste
fuels at LCGC, mainly due to the heterogeneity and high level of water in the samples.

ASTM D 95-05 (2010): Standard Test Method

for Water in Petroleum Products and
Bituminous Materials by Distillation.
The material to be tested is heated under reflux
with a water-immiscible solvent, which co-distills
with the water in the sample.
Condensed solvent and water are continuously
separated in a trap, the water settling in the
graduated section of the trap and the solvent
returning to the still.
Depending on the material to be tested the
solvent-carrier liquid can be aromatic (Xylene,
Petroleum Naphtha, etc.), Petroleum distillate or
Volatile Spirits (Petroleum Spirit, Iso-octane, etc.).
Its necessary to determine the amount of water in
the solvent in a blank determination (ml), so it can
be removed to the volume of water in the trap at
the end of the test. Then divide by the total volume
or mass in the test sample and put it in percentage.
Figure 25 Distillator at LCGC
The results with the distillation method have been better, both in interlaboratory comparisons and in the
analysis of reference standard materials.

Guidelines for sampling, preparation and analysis of alternative fuels Page 50

CIMPOR TEC | May 2012
5.6 Determination of Volatile matter
The volatile matter is determined as the loss in mass less that due to moisture, when alternative fuel is
heated out of contact with air under standardised conditions.
The test is empirical and, in order to ensure reproducible results, it is essential that the rate of heating,
the final temperature and the overall duration of the test are carefully controlled. It is also essential to
exclude air from the alternative fuel during heating to prevent oxidation. The fit of the crucible lid is
therefore critical.
The moisture content of the sample is determined at the same time as the volatile matter so that the
appropriate correction can be made. Mineral matter associated with the sample can also lose mass
under the conditions of the test, the magnitude of the loss being dependent on both the nature and the
quantity of the minerals present.
The reference method for this test is done in a furnace, usually to the dried sample, and it corresponds
to the loss of mass to the sample subjected to 900C during 7 minutes in a covered quartz crucible.
The specifications for the furnace and the quartz crucible and lid are given in the European Standard
EN 15402, and it is also mentioned that automatic equipment may be used if the procedure is validated
with reference materials of an adequate type and fulfils all the requirements regarding sample size,
atmosphere, temperatures and weighing accuracy.
In LCGC this test is done in the TGA as the 2nd step of the proximate analysis mentioned in the
humidity determination.
For the TGA analysis a calibration curve is done, using standards analysed by the reference method in
the furnace. The purpose is to correct the effect of the heating rate that exists in the TGA, because the
test is not started immediately at 900C like in the furnace and the sample is subjected to heat before
that.
The reference procedure mentioned in the EN 15402 Standard is the following:
Fill either a stand with one empty crucible and lid and insert it into the furnace. Maintain the temperature
at (90010)C for about 7 min. Remove the crucible(s) from the furnace and allow to cool to room
temperature on a thick metal plate.
As soon as they are cool, weigh each empty crucible and lid and weigh into each crucible, to the
nearest 0,1 mg, (10,1)g of the general analysis sample. Replace the lid and tap each crucible on a
clean hard surface until the test portion forms a layer of even thickness on the bottom of the crucible.
Place the charged crucible(s) in a cold stand, transfer to the furnace, close the door and leave for
(4205)s. Remove and allow it to cool to room temperature. If cool, weigh the crucible(s) to the nearest
0,1 mg in the same manner as for the empty crucible(s).
The volatile matter, V, in the sample as analysed, expressed as mass fraction in percentage, is given by
the following equation:
( )
( )
Where:
m1 is the mass of the empty crucible and lid, in grams;
m2 is the mass of the crucible and lid and test portion before heating, in grams;
m3 is the mass of the crucible and lid and contents after heating, in grams;
M is the mass fraction of moisture in the general analysis sample as analysed, in percentage.
Report the result as the mean of duplicate determinations, rounded to the nearest 0,1 % mass fraction,
and the results shall be reported on dry basis.

Guidelines for sampling, preparation and analysis of alternative fuels Page 51

CIMPOR TEC | May 2012
5.7 Determination of Ash content
The ash content can be defined as the mass of inorganic residue remained after ignition under specified
conditions.
The sample is heated in air atmosphere up to a temperature of (550 10) C under rigidly controlled
conditions of time, sample mass and equipment specifications. The ash content is determined by
calculation from the mass of the residue remaining after heating.
In the reference method for testing in a furnace, the sample is introduced in a quartz crucible and
subjected to a heating rate until it reaches 550C and then till constant weight.
All the specifications for this test in solid recovered fuels are presented in the EN 15403 European
Standard, a minimum of two determinations shall be carried out on the general analysis sample, and the
procedure is the following.
Heat the empty dish in the furnace to (550 10) C for at least 60 min. Allow the dish to cool down in a
desiccator. After the dish is cooled, weigh it to the nearest 0,1 mg and record the mass.
Place about 1 g of the general analysis sample on the bottom of the dish and spread in an even layer
over the bottom surface. Weigh the dish plus the sample to the nearest 0,1 mg and record the mass. If
the general analysis sample is oven-dried, both the dish and the sample shall be dried at (105 10) C
as a precautionary measure and then weighed.
Place the loaded dish in the cold furnace. Heat the sample in the furnace according to the following
heating routine:
a) raise the furnace temperature evenly to (250 10) C over a period of 50 min (i.e. a rise of 5 K/min).
Maintain at this temperature level for 60 min to allow the volatiles to leave the sample before ignition;
b) continue to raise the furnace temperature evenly to (550 10) C over a period of 60 min (i.e. a rise
of 5 K/min) and keep this temperature level for at least 120 min.
Remove the dish with its content from the furnace. Allow the dish and its content to cool on a thick metal
plate for 5 min to 10 min and then transfer to a desiccator without desiccant and allow to cool to ambient
temperature. Weigh the ash and the dish to the nearest 0,1 mg as soon as ambient temperature is
reached and record the mass.
If there is any doubt of complete incineration (for instance presence of soot at visual inspection), then
add droplets of water or ammonium nitrate to the sample before it is reloaded into the cold furnace and
reheated to (550 10) C for a period of further 30 min until the change in mass is lower than 0,2 mg.
Automatic equipments may be used if the method is validated with reference samples and fulfills all the
requirements given regarding sample size, heating procedure, atmosphere, temperatures and weighing
accuracy.
The ash content on dry basis, Adb, of the general analysis sample, expressed as mass fraction in
percentage, shall be calculated by the following equation:
( )
( )
Where:
m1 is the mass of the empty dish, in grams;
m2 is the mass of the dish plus the general analysis sample, in grams;
m3 is the mass of the dish plus ash, in grams;
Mad is the mass fraction of moisture of the general analysis sample on wet basis, in percentage, and the
result shall be reported as the mean of duplicate determinations to the nearest 0,1 %.
In LCGC this test is done in the TGA as the 3rd step of the proximate analysis together with humidity
and volatiles, following the requirements for automatic equipments.
In TGA analysis its not necessary to use the previous heating rate because all the humidity and
volatiles were already removed in the previous steps.

Guidelines for sampling, preparation and analysis of alternative fuels Page 52

CIMPOR TEC | May 2012
5.8 Determination of major elements by XRF
For the characterization of alternative fuels the determination of major elements such as Al, Ca, Fe, Mg,
P, K, Si, Na and Ti, can be helpful to predict the melting behaviour and slagging of the ash.
There is a European Standard EN 15410 for solid recovered fuels where some test methods are
mentioned in which the test portion is digested using a suitable acid mixture, and then the digested
sample is analysed by a possible number of analytical techniques such as: Inductively Coupled Plasma
with optical or mass detection, Flame Atomic Spectroscopy, Graphite Furnace Atomic Absorption
Spectrometry and X-ray fluorescence spectrometry (XRF).
The analysis by XRF allows the fast and simultaneous determination of these elements, but the direct
analysis of alternative fuels is not possible by XRF due to the sample inhomogeneity and because
suitable certified reference materials for calibration are not available.
The analysis by XRF is done to the ashes obtained at 550C and it corresponds to the chemical
analysis of the ashes that remain after the burning of the combustible matter present in the sample.
Ash sample for XRF analysis are prepared as described in the method reported in EN 15403, starting
from a quantity of material to obtain the amount of ash sufficient for the analysis (ash content can be as
low as a very few percentage on dry basis).
The analysis of the ashes from the solid alternative fuels by XRF at LCGC is done by fused bead,
unless theres any problem like the impossibility of obtaining a fused bead because of cracks and other
fusion problems that might occur due to the chemical composition of these materials.
Whenever its not possible to obtain a fused bead, the samples are analysed by pressed powder and
the quantification is done by a semi-quantitative analysis program.
Also when theres a great amount of non-dosed elements, the semi-quantitative analysis program is
used for the determination of the remaining elements that can be present in this type of samples, like Ni,
V and Zn.
The XRF equipment at LCGC is the Axios Cement
from Panalytical.
This analysis is done in a wide range calibration
curve that we use for ashes and where the
proportion of sample/flux is 0,5/4,8g to improve the
fusion.
The calibration curves at LCGC are not in the
format of the elements, but in their oxides like is
usual in the cement industry, however the
conversion to the element by calculations is
possible using their atomic weights.
The analysis of liquid alternative fuels and ashes of
some materials, like tyres, where is not possible to
obtain a pressed powder and the analysis has to
be done to loose powder, the XRF test is done in a
specific sample holder for liquids and is performed
in a Helium atmosphere, with quantification by the
semi-quantitative analysis program. Figure 26 XRF analyser at LCGC
In XRF analysis the results can be expressed in mass percentage or milligrams per kilogram (ppm) on a
dry basis. For the results in the all original sample the following equation shall be used:

Where:
Cd is the concentration on dry basis in the original sample;
Cash is the concentration in the ash;
Aash is the ash content (%).

Guidelines for sampling, preparation and analysis of alternative fuels Page 53

CIMPOR TEC | May 2012
5.9 Determination of trace elements by XRF
The accurate determination of trace element content in alternative fuels is important for environmental
and technical reasons both in the production and combustion stage.
The principle is the same as for the analysis of major elements, and the European Standard EN 15411
for this determination mentions the same digestion methods and analytical techniques, with the addition
of direct analysis methods, like for instance the determination of mercury in the AMA analyser.
The trace elements mentioned in the standard are mainly the ones that have specifications for the use
of alternative fuels, namely: As, Ba, Be, Cd, Co, Cr, Cu, Hg, Mo, Mn, Ni, Pb, Sb, Se, Tl, V and Zn.
The direct analysis by XRF is not mentioned due to the fact that these elements exist in trace quantities
which are lower than the detection and quantification limits for the XRF analysis.
However at LCGC the trace elements are directly analysed by XRF in pressed powder samples, using
the Pro-Trace analytical program developed by Panalytical, in which the detection and quantification
limits reach the order of very low quantity in ppm.
The Pro-Trace analytical program uses a set of calibration standards and corrections for the overlaps
and background coefficients, and at LCGC was improved with the addition of several certified reference
materials (CRM) to the calibration curves of the elements. It is a bit costly but also the analytical
techniques mentioned in the standard are, and more time consuming than this XRF program.
The Pro-Trace can only be used for pressed powder samples so, when thats not possible, in liquids
and loose powder, the quantification is done by the semi-quantification program, where the detection
and quantification limits are higher and is not always possible to detect all the trace elements.

5.10 Determination of mercury (Hg) by direct analyser

For the determination of mercury the LCGC uses a direct analyser AMA-254, like the one mentioned in
the EN 15411 European Standard.
This direct analyser is a single purpose atomic absorption spectrophotometer for direct mercury
determination in solids and liquids without the need of sample chemical pre-treatment. It uses a mercury
vapour generation technique, with high sensitivity independently of the sample matrix.

Direct sample input (ca. 500 mg) of solid and

liquid samples
Concentration range from 0,05 ppb up to 100
ppm Hg
Reproducibility at 10 ppb Hg = < 1,5 %
Long-time stable calibration for 2 measuring
ranges
Figure 27 AMA-254 analyser at LCGC

Guidelines for sampling, preparation and analysis of alternative fuels Page 54

CIMPOR TEC | May 2012
5.11 Determination of Chloride (Cl)
The determination of chloride in alternative fuels is very important since several materials present in
these might have high values of chloride that will affect the production process or even the end-use of
the cement if they are retained in the clinker during the production process.
The EN 15408 European Standard for SRF states for this determination a method based on full
oxidation of the sample by combustion in a bomb containing oxygen (same as for Calorific Value), the
conversion of halogenated and sulphur compounds into fluoride, chloride, bromide and sulphate,
absorption in water or KOH solution with H 2O2 and analysis by ion chromatography or other suitable
techniques such as ICP-MS, volumetric analysis, etc.
In LCGC for many time was used the similar method mentioned in the ASTM 2361 standard, with
combustion in the bomb and chloride determination by back titration, however this is now used only for
liquids and samples were a pressed powder sample for XRF is not possible.
At this moment the LCGC determines the chloride present in solid alternative fuels by XRF in pressed
powder samples, a calibration curve was prepared using samples analysed by the reference method
and some Interlaboratory and CRM samples, and the results obtained were proved to be accurate.

5.12 Determination of Fluoride (F)

The determination of fluoride is also important due to possible effects on the production process and the
quality of the clinker obtained, even though it might have a positive role since it may help the
mineralisation of the clinker.
The test method for the determination of fluoride is also mentioned in the EN 15408 like it was
described in the previous section for the case of chlorides.
In LCGC it is done by a potentiometric technique with an Ion Selective Electrode specific for fluoride.
This method starts with a fusion attack similar to the one its done to raw materials of the cement
production, then there is a calibration curve with standards prepared from NaF and afterwards its
possible to measure the quantity of Fluoride Ion present in the sample.

In the picture it can be seen the automatic

titrator used at LCGC for the measurement of
the Fluoride Ion with the use of an Ion Selective
Electrode.

Figure 28 Automatic Titrator for Fluoride analysis

Guidelines for sampling, preparation and analysis of alternative fuels Page 55

CIMPOR TEC | May 2012
5.13 Determination of biomass content
The determination of biomass content is important for the evaluation of emissions during the utilization
of alternative fuels, including the CO2 taxes in European countries according to the determinations of
the international protocols for the reduction of greenhouse gases.
For this reason we have to differentiate between fossil and biogenic carbon:
Fossil: from the Latin fossus, having been dug up. Fossils are preserved remains or traces of
animals, plants, and other organisms from the remote past. The carbon was fixed in the material in the
geological past.
Biomass: The carbon, present in the material, was produced from living organisms and embedded a
short time ago.
In summary the CO2 from biomass does not lead to a higher atmospheric CO2 concentration, because
the Carbon of biomass was taken recently from the atmosphere.
The burnable carbon of solid alternative fuels derives from: fossil sources (mainly in form of plastic),
mixed sources like tires and packing material and biomass (e.g. wood, paper).
The methods for the determination of biomass content in solid recovered fuels are specified in the EN
15440 European Standard and are: the method of selective dissolution, the manual sorting method and
the method based on the 14C content.
The method of selective dissolution is based on the reaction of biomass material with a mixture of
sulphuric acid and hydrogen peroxide. The manual sorting method is based on the separation of
different fractions by visual inspection. The determination of the biomass content using the 14C method
is based on the well-established analytical procedures that are used for the determination of the age of
carbon containing objects. The fraction of biomass can be expressed: by weight; by energy content
(gross or net calorific value) or by carbon content.
In LCGC its possible to do the selective dissolution method, for which the principle is based on the
determination of: the inert mass (e.g. aluminium, iron, glass) by combusting defined as ash content and
the non-biomass by dissolving in sulphuric acid.
Depending on the chosen procedure, the determination of: the calorific value of the SRF, the residue
and the inert mass or the carbon content of the SRF, the residue and the inert mass calculating the
biomass content of the overall sample.
At LCGC we use the procedure based on the determination of the carbon content, and this method
implies the determination of three carbon contents, one ash content and one insoluble residue,
according to the following scheme:

Figure 29 Selective dissolution method for biomass by carbon content

Guidelines for sampling, preparation and analysis of alternative fuels Page 56

CIMPOR TEC | May 2012
The most correct method is by 14C, but its also expensive and not enough tests are required for LCGC
to implement that method, while the selective dissolution method can be easily implemented in
laboratories that already perform the tests included.
However its important to know the limitations of the selective dissolution method, because its made the
assumption that there is a complete biodegradation of biomass materials and no biodegradation of non-
biomass materials, but some biomass materials dissolve only partly and some non-biomass materials
dissolve.
Moreover, according to the result of several tests, there is no reliability for biomass concentrations
below 5% and above 95%, which makes this method not applicable for:
pure fractions of waste
products classified as CO2 neutral
charcoal, peat, solid fossil fuels (e.g. hard coal, lignite)
SRF with more than 10 % rubber (natural, synthetic)
SRF with more than 5 % biodegradable plastics of fossil origin
SRF with more than 5 % non-biodegradable plastics of biogenic origin
SRF with more than 5 % wool, viscose, fat, nylon.

Guidelines for sampling, preparation and analysis of alternative fuels Page 57

CIMPOR TEC | May 2012
6 - Bibliography

Alternative Fuels Quality Control: Sampling, preparation, analysis Gonalo Almeida - Cimpor
Quality Meeting 2010.
Solid recovered fuels - Sampling and Preparation Standards Gonalo Almeida - Cimpor Group
Quality & Process Meeting 2012.
Alternative fuels: sampling, preparation, analysis - Marion Engel - ECRA-Seminar S10-06 -
Beckum, 9/10 November, 2010.
Low Grade Fuel Study Revision 2007 Holcim
Guidelines for the Selection and Use of Fuels and Raw Materials in the Cement Manufacturing
Process - Cement Sustainability Initiative (CSI) DRAFT December 2005 Version 1.0
EN 15442:2011 Solid recovered fuels - Methods for sampling
EN 15443:2011 Solid recovered fuels - Methods for the preparation of the laboratory sample
EN 15413:2011 Solid recovered fuels - Methods for the preparation of the test sample from the
laboratory sample
EN 15357:2011 Solid recovered fuels - Terminology, definitions and descriptions
EN 15359:2011 Solid recovered fuels Specifications and classes
EN 15400:2011 Solid recovered fuels - Determination of calorific value
EN 15402:2011 Solid recovered fuels - Determination of -the content of volatile matter
EN 15403:2011 Solid recovered fuels - Determination of ash content
EN 15407:2011 Solid recovered fuels - Methods for the determination of carbon (C), hydrogen (H)
and nitrogen (N) content
EN 15408:2011 Solid recovered fuels - Methods for the determination of sulphur (S), chlorine (Cl),
fluorine (F) and bromine (Br) content
EN 15410:2011 Solid recovered fuels - Methods for the determination of the content of major
elements (Al, Ca, Fe, K, Mg, Na, P, Si, Ti)
EN 15411:2011 Solid recovered fuels - Methods for the determination of the content of trace
elements (As, Ba, Be, Cd, Co, Cr, Cu, Hg, Mo, Mn, Ni, Pb, Sb, Se, Tl, V and Zn)
CEN/TS 15401:2010 Solid recovered fuels - Determination of bulk density
CEN/TS 15414-1:2010 Solid recovered fuels - Determination of moisture content using the oven
dry method - Part 1: Determination of total moisture by a reference method
CEN/TS 15414-2:2010 Solid recovered fuels - Determination of moisture content using the oven
dry method - Part 2: Determination of total moisture content by a simplified method
EN 15414-3:2011 Solid recovered fuels - Determination of moisture content using the oven dry
method - Part 3: Moisture in general analysis sample
EN 15440:2011/AC:2011 Solid recovered fuels - Methods for the determination of biomass content

Guidelines for sampling, preparation and analysis of alternative fuels Page 58

CIMPOR TEC | May 2012