SIST-TP CEN/TR 15716:2008

SLOVENSKI STANDARD
SIST-TP CEN/TR 15716:2008
01-september-2008
7UGQRDOWHUQDWLYQRJRULYR'RORþHYDQMHQDþLQD]JRUHYDQMD
Solid recovered fuels - Determination of combustion behaviour
Feste Sekundärbrennstoffe - Bestimmung des Verbrennungsverhaltens
Combustibles solides de récupération - Détermination du comportement de la
combustion
Ta slovenski standard je istoveten z: CEN/TR 15716:2008
ICS:
75.160.10 Trda goriva Solid fuels
SIST-TP CEN/TR 15716:2008 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------

TECHNICAL REPORT
CEN/TR 15716
RAPPORT TECHNIQUE
TECHNISCHER BERICHT
June 2008
ICS 75.160.10

English Version
Solid recovered fuels - Determination of combustion behaviour
Combustibles solides de récupération - Détermination du Feste Sekundärbrennstoffe - Bestimmung des
comportement de la combustion Verbrennungsverhaltens
This Technical Report was approved by CEN on 21 January 2008. It has been drawn up by the Technical Committee CEN/TC 343.
CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland,
France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal,
Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre: rue de Stassart, 36  B-1050 Brussels
© 2008 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TR 15716:2008: E
worldwide for CEN national Members.

---------------------- Page: 2 ----------------------

CEN/TR 15716:2008 (E)
Contents Page
Foreword.3
Introduction .4
1 Scope .7
2 Combustion of solid fuels.7
2.1 Basis of solid fuel combustion.7
2.2 Basics of some common combustion systems that utilises SRF .8
2.3 Determination of characteristic parameters .9
2.4 Use of classification numbers.10
2.5 Combustion prediction tool.10
3 Thermal gravimetric analysis .13
4 Standard fuel analysis.17
4.1 General.17
4.2 Proximate analysis: Moisture, volatiles, and ash content.17
4.3 Ultimate analysis: C, H, N, S, Halogens.17
4.4 Gross calorific value (GCV)/net calorific value (NCV).18
4.5 Particle size distribution .18
4.6 Ash content and ash melting behaviour .19
5 Advanced laboratory methods for fuel characterisation.19
5.1 General.19
5.2 Determination of fuel composition .21
5.3 Composition and calorific value of the volatile matter.22
5.4 Kinetic properties .25
5.5 Image analysis method for particle size distribution.30
5.6 Apparent densities of particles and intermediates .32
5.7 Aerodynamic lift velocity .33
5.8 Slagging and fouling behaviour.34
6 Operational behaviour in the combustion process.35
7 Summary.38
Bibliography .40

2

---------------------- Page: 3 ----------------------

CEN/TR 15716:2008 (E)
Foreword
This document (CEN/TR 15716:2008) has been prepared by Technical Committee CEN/TC 343 “Solid
recovered fuels”, the secretariat of which is held by SFS.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. CEN [and/or CENELEC] shall not be held responsible for identifying any or all such patent rights.
3

---------------------- Page: 4 ----------------------

CEN/TR 15716:2008 (E)
Introduction
Historically, SRF goes back to the oil crises approximately 30 years ago, when refused derived fuel (RDF)
was promoted as a substitute low cost fuel. Contrary to that situation, the producers of SRF took the initiative
for the implementation of a quality system to meet and guarantee specified fuel classification and specification
parameters. Quality systems to check their production now exist in several EU member states and efforts are
being made by CEN/TC 343 to develop European Standards for SRF [1].
The production and thermal utilisation (energy recovery) of Solid Recovered Fuels (SRF) from bio wastes,
residues, mixed- and mono waste streams have significant relevance as a key component of an integrated
waste management concept.
The implementation of SRF production in an integrated waste management concept demands a potential
market for these products. Known proven markets are found in the European energy sector and in other more
product-oriented sectors like cement or lime industry by substitution of fossil fuels. The capacities for co-
utilisation of these products, to include utilisation in minor thermal shares, are enormous, especially in the new
European member states as most of the energy production of these countries relies on fossil fuels.
A successful application of solid recovered fuel in power plants and industrial furnaces would require a
thorough understanding of the fuel properties which include the combustion behaviour, emission potential,
impact on facility etc. The determination of combustion behaviour which is the main focus of this document
seeks to outline possible methods and procedures that can be adopted to analyse any given solid recovered
fuel. An approach has therefore been outlined where the determination of combustion behaviour is
categorised into four groups which combine to give a holistic impression of the combustion progress of SRF in
both mono and co-firing systems (see Figure 1).

Figure 1 — Scheme to determine combustion behaviour of SRF
While there are standardised methods, such as from the American Society for Testing and Materials (ASTM)
and the German Institute for Standardization (DIN Deutsches Institut für Normung e. V.), for determining
combustion behaviour for primary fuels (e.g. coal), the process is not the same for SRF. At present, there are
no standardised methods for SRF. Most of the available methods are in-house, usually designed for particular
types of SRF, e.g. waste, or bio-residue fractions to suit a specific combustion system like grate firing,
fluidised bed, pulverised fuel system, and cement kiln. Figure 2 gives an overview about the broad variety of
4

---------------------- Page: 5 ----------------------

CEN/TR 15716:2008 (E)
SRF utilisation routes using an example of co-combustion in power plants and industrial furnaces.
Co-combustion also includes indirect co-firing systems such as gasification (Lahti, Zeltweg) and pyrolysis
(ConTherm). While the environmental aspect of the thermal utilisation of SRF is very important, this report
focuses only on the combustion aspect.

Figure 2 — SRF utilisation routes
Solid recovered fuel can be made of any combustible non-hazardous waste and processed to a quality that
allows to classify it in accordance with CEN/TS 15359 and which fulfils specifications as agreed with the
customer. Considering this, the main problem becomes obvious: How to define reliable methods to describe
the combustion behaviour of solid fuels such as SRF, valid for all possible types of input material and
combustion systems? A systematic approach adopted herein to determine combustion behaviour is outlined in
Figure 1. It is grouped into four categories:
 standard fuel analysis;
 laboratory-scale tests with advanced methods;
 semi-technical and pilot-scale combustion tests;
 full-scale test.
In general, such a four-step procedure is an effective way to successfully integrate a new fuel in an existing
power plant or an industrial furnace. In any case, full scale tests are the most reliable but very expensive with
several bottlenecks (e.g. retrofits, permits, time, etc.) and that is the reason for the need to develop and
standardise methods which are reliable, fast, and not expensive according to the various firing systems are
essential. Besides the evaluation of parameters concerning combustion behaviour, the steps before full scale
implementation also forms substantial basis to reliably evaluate other areas of major interest such as grinding
and fuel feeding; slagging, fouling and corrosion; and lastly emissions and residues. The systematic
evaluation of these additional topics requires area specific analyses, tests, and measurements.
5

---------------------- Page: 6 ----------------------

CEN/TR 15716:2008 (E)
Concerning combustion behaviour, the standard analysis of the SRF will determine the basic parameters
about the combustible and incombustible matter. The amount of energy, the contents of water, volatiles, fixed-
carbon, ash, and particle size will roughly dictate the type of the combustion system that is best suited. In
addition to the standard analysis, a selected combustion system might require an advanced parameter
analysis, if possible, with a close relation to case specific process parameters. Such a correlation will
substantially enhance the reliability of transfer studies. An example, in the case of a pulverised firing system,
is the maximum particle size required for a complete combustion in order to avoid fuel plummeting into the
bottom ash.
Currently, the activities towards the combustion behaviour of SRF rely largely on standard analysis and
laboratory-scale tests, which were originally developed with certain limitations and applicable to solid fuels
such as lignite and hard coal. A common problem of these methods is that parameters related to SRF during
combustion are not sufficiently covered. These methods make sure consistent quality of the SRF supply rather
than to predict combustion performance. Therefore, the development of the so-called advanced test methods
to fill the gap and amending existing test apparatus and measurement conditions is required.
The driving force to introduce SRF rests much on economic factors. In most cases, the end user will be either
the operator of a power plant or an industrial furnace. The primary focus will be an unrestricted and reliable
operation of the facility. One wants to assess the possible risks and dangers. In case of retrofits, the end user
needs to calculate the required cost on modifications and operation. It can be assumed that due to possible
operational risks such as corrosion, the plant operators will select the fuel with the most appropriate qualities.
Such requirements are needed tools to control the quality of the SRF and to deliver them according to
specification. As such, the knowledge of the combustion behaviour is an essential aspect for the
commercialisation of SRF. It will allow the optimisation of the process and the assessment of possible risks
and dangers prior to full-scale application.
Some methods and parameters will be introduced in the subsequent sections, but whatever methods are to be
used in the future should be orientated towards the following aspects:
 reproducibility;
 repeatability;
 reliability;
 time efforts (rapid test methods);
 cost effectiveness;
 possibilities for automatic testing.
The authors summarise and refer to past and current activities trying to describe combustion behaviour of
SRF. The idea is to identify a common and successful practice where various approaches converge.
6

---------------------- Page: 7 ----------------------

CEN/TR 15716:2008 (E)
1 Scope
This Technical Report gives a review on determination methods for exploring how different SRFs behave in
different combustion systems, e.g. with respect to time for ignition, time for gas phase burning and time for
char burn out, including information on technical aspects like slagging and fouling, corrosion as well as
required flue gas cleaning for meeting the emission limit values induced by the Waste Incineration
Directive (WID).
2 Combustion of solid fuels
2.1 Basis of solid fuel combustion
Combustion of fuels shall be considered both from theoretical and practical perspectives. The former can
define combustion as the rapid chemical reaction of oxygen with the combustible elements of a fuel. While the
later where the engineer is concerned with boiler design and performance might define combustion as the
chemical union of fuel combustibles and the oxygen of the air, controlled at a rate that produces useful heat
energy. The two definitions implicitly consider many key factors. For complete combustion within a furnace,
four basic criteria shall be satisfied:
1) adequate quantity of air (oxygen) supplied to the fuel;
2) oxygen and fuel thoroughly mixed (turbulence);
3) fuel-air mixture maintained at or above the ignition temperature;
4) furnace volume large enough to give the mixture time for complete combustion.
Quantities of combustible constituents within the fuel vary by types. Figure 3 shows the significant change in
the combustion air requirements for various fuels, resulting from changes in fuel composition. It illustrates the
minimum combustion air theoretically required to support complete combustion.

Key
Y Stochiometric air demand in nominal cubic meter dry air per kilogram fuel
Figure 3 —Stoichiometric air to fuel ratio for some SRFs
7

---------------------- Page: 8 ----------------------

CEN/TR 15716:2008 (E)
In an ideal situation, the combustion process would occur with the stoichiometric quantities of oxygen and a
combustible based on underlying chemical principles. However, since complete mixing of air and fuel within
the furnace is virtually impossible, excess air shall be supplied to the combustion process to ensure complete
combustion. The amount of excess air that should be provided varies with the fuel, boiler load, and type of
firing system, and it is in the range of 0,1 ≤ λ ≤ 0,6 or even more.
Solid fuel combustion consists of three relatively distinct but overlapping phases:
 heating phase (time to ignition);
 gas phase combustion (time of gas phase burning);
 char combustion (time for char burnout).
Firstly, the time to ignition involves particle heat up due to radiation and convection in the furnace driving off
moisture and volatiles adsorbed in the solid. Solid fuels, especially fresh biomass, can release combustible
volatiles below 100°C and ignition can occur as soon as the particle is not completely surrounded by water
vapour. The time to ignition is relatively short. For plastics it is different, they do not contain volatiles in the
traditional meaning. They are often transparent so they heat up slowly and then start melting. Film plastics
tend to shrink and form molten droplets. At about 400°C de-polymerization starts (pyrolysis) where gaseous
combustible compounds release. The time to ignition is long compared to regular fuels of the same particle
size. Secondly, the time of gas phase burning involves the volatiles released through desorption and pyrolysis
burn in a flame around the particle until a solid char is left. This phase is long for plastics compared to coal
because plastics (except PVC) do not form a char at all. The flaming particle can fly as a warm air balloon.
Thirdly, the time for char burnout is a gas/solid reaction which for coal is the longest step and it is strongly
dependent on particle size and porosity etc. For wood this is intermediate and for polyolefin plastics it is close
to zero. The tests indicate that, for particles of the same size (50 mg) and same temperatures and oxygen
contents, the time for complete burning is in the following order: plastics < dry wood < coal.
The combustion process of metals present in SRF especially aluminium is complicated and cannot be
completely avoided. Ignition of such particles is preceded by the disruption of the oxide film at a temperature
> 1500°C (calculated); it react intensively with atmospheric oxygen, which leads to a further sharp increase in
temperature of the particle surface zone (see [2]). These high temperature regimes required to start the
ignition are usually not found in conventional boilers, therefore the molten aluminium droplets coagulate and
form large pieces upon cooling.
2.2 Basics of some common combustion systems that utilises SRF
Pulverised fuel combustion system (PF): In PF combustion, the fuel is ground to a specified fineness, e.g.
coal to a maximum particle size of 250 µm to 300 µm, depending on the reactivity. They are pneumatically
transported to the burners and injected via particle-laden jets into the combustion chamber. For lower
reactivity fuels, the fineness of grind is increased to create a larger specific surface area so as to improve
conditions for ignition and combustion. The transport air that carries the fuel from the mill to the burners is a
small fraction of the total combustion air. It is kept at low temperature, limited to about 373 K for coal and for
SRF much lower, for reasons of safety against ignition and explosion in the mill and in the pulverized fuel
transport pipeline between the mill and the burners. The rest of the combustion air, which can be preheated to
higher temperatures, is injected separately and admixed with the already ignited particle-laden jet in the
combustion chamber. The combustion chamber is typically of parallelepiped shape; the cross-sectional area
of a 300-MW coal-fired boiler would be about 15 m × 15 m and its height 45 m to 50 m (see [3]).
Fluidised Bed system: A fluidised bed is composed of fuel (coal, coke, biomass, SRF, etc.) and bed material
(ash, sand and/or sorbent) contained within an atmospheric or pressurised vessel. The bed becomes fluidised
when air or other gas flows upwards at a velocity sufficient to expand the bed. At low fluidising
velocities (0,9 m/s to 3 m/s), relative high solid densities are maintained in the bed and only a small fraction of
the solids are entrained from the bed. A fluidised bed that is operated in this range is referred to as a bubbling
fluidising bed (BFB). As the fluidising velocity is increased, smaller particles are entrained in the gas stream
and transported out of the bed. The bed surface becomes more diffuse and solids densities are reduced in the
8

---------------------- Page: 9 ----------------------

CEN/TR 15716:2008 (E)
bed. A fluidised bed that is operated at velocities in the range of 3,9 m/s to 6,7 m/s is referred to as circulating
fluidised bed (CFB) (see [4]).
Fluidised bed combustion (FBC) units are touted as being „fuel flexible“, with the capacity of firing a wide
range of solid fuel with varying heating value, ash content, and moisture content. Also, slagging and fouling
tendencies are minimised in FBC units because of law combustion temperatures. The advantages of FBC in
comparison to conventional pulverised fuelled units can be summarised as follows:
 SO can be removed in the combustion process by adding limestone to the fluidised bed, eliminating
2
the need for an external desulphurisation process;
 fluidised bed boilers are inherently fuel flexible and, and with proper design provisions, can burn a
variety of fuels;
 the combustion in FBC units takes place at temperatures below the ash fusion temperatures of most
fuels, consequently, tendencies for slagging and fouling are reduced with FBC;
 because of the reduced combustion temperatures, NO emissions are inherently low.
x
Stoker firing system: In this firing system, solid fuel is spread and combusted on a grate system. The grate
usually used is a continuous-cleaning, travelling grate. The lighter portion of the solid fuel burns in suspension
above the grate and the heavier portion burns on the grate. Air and or water banks are used to cool the grate.
Sometimes cooling is done by water. An over-fire combustion air is used to cause mixing of gases and
combustion above the grate. This system is used predominantly for processed as well as unprocessed solid
waste combustion. Other technologies used to generate heat and power from SRFs are explained in details
elsewhere (see [5]).
2.3 Determination of characteristic parameters
Several researchers (see [6], [7], [8]) have shown that particle size and reactive surface have large influence
on the combustion process, most especially SRF and solid biomass qualities available on the market. The
distinctions between fuel properties and process conditions which contribute to the total combustion process
are outlined in Table 1. Fuel properties and process parameters (operational conditions) together define how a
particular SRF behave during combustion.
Table 1 — Fuel and process parameters influencing the combustion behaviour of SRF
Fuel properties Process parameters
Heat capacity and conductivity Temperature profile in the furnace
Gross calorific (GCV) and net calorific value (NCV) Heat transfer in the furnace
Composition and distribution of organic and inorganic Oxygen partial pressure along the furnace profile
matter
Volatile release and char reactivity as a function of Velocities, turbulence and mixing behaviour
the particle temperature
Ash composition and ash fusion behaviour Residence time
Fuel particle size etc. Combustion system etc.

Solid fuel properties can be distinguished into chemical, mechanical (physical), calorific, and kinetic (reaction)
properties. The chemical properties describe aspects like burnable substances, major and minor elements,
etc. The mechanical properties describe particle and bulk densities, particle size distribution, shape and form,
etc.; whereas the heating value, air demand, heat capacity and the calculated adiabatic flame temperature
describe the calorific properties. These parameters can be transferred to SRF without restriction. The
evaluation of kinetic properties is more difficult as they depend on chemical, mechanical, and calorific
9

---------------------- Page: 10 ----------------------

CEN/TR 15716:2008 (E)
properties. Contrary to the procedures used for coals, where detailed investigations towards the combustion
properties are performed as a function of volatile content, heating value or particle size distribution, the
approach even though not totally suitable for heterogeneous fuel can be adopted after a few modifications.
2.4 Use of classification numbers
Some parameters for the combustion behaviour of fuel are indicated by different classification numbers (see
[9], [10]). This can be used to compare different primary and secondary fuels. Classification numbers are also
published for other areas of interest such as slagging, fouling, corrosion and emission formation. A transfer of
these numbers, which are mainly applied to different coal qualities, to SRF will have more limitation. However,
it could be an option in the classification of SRF, although the interaction with coal in case of co-incinerator
cannot be described sufficiently with this approach. Nevertheless, the approval of such classification numbers
should include an independent validation procedure.
2.5 Combustion prediction tool
For the development and successful implementation of particle combustion models into combustion predicting
tools, such as computational fluid dynamic (CFD) calculations, the kinetics of the fuel slagging; specifically
volatile and char kinetic data are of great interest. The former is paramount if de-volatilisation is the rate
determining step of the SRF during combustion, and vice versa. Most SRF have high volatile content, and
during combustion their release dictates the process. Figures 4 and 5 illustrate the interactions in the
combustion model as captured in CFD modules for plastics biomass and coal [11].

Figure 4 — Plastic combustion model [11]

10

---------------------- Page: 11 ----------------------

CEN/TR 15716:2008 (E)

Figure 5 — Modelled reaction scheme for coal and biomass
The combustion of SRF like any other solid fuel can be simulated by way of adopting and modifying existing
modules that have worked for pulverised coal combustion. The three primary phenomena that contribute to
the predictions using CFD calculations are chemical reactions, flow behaviour, and heat transfer [12]. Figure 6
shows how such tools have been used to visualise burnout profiles of different SRF particle sizes in a
boiler [13]. It shows that the burnout for larger particle sizes (d = 5 mm) were about 0,5 kg/kg according to
50 o
the burnout colour code. Particle tracking has also been performed for coal, the biogenic and the plastic
fraction of SRF and it is illustrated in Figure 7.
11

---------------------- Page: 12 ----------------------

CEN/TR 15716:2008 (E)


a) Biogenic fuel, d == 1,3 mm b) Biogenic fuel, d == 2,5 mm c) Biogenic fuel, d == 5,0 mm
50== 50== 50==
Key
Y Burnout colour code
Figure 6 — Simulation of SRF burnout in a boiler (see [13])



a) Coal, d ==== 0,13 mm b) Biogenic fuel, d ==== 1,3 mm c) Plastic fraction (PP),
50 50
  d == 1,3 mm
50==
Key
Y1, Y2, Y3 Burnout colour codes
Figure 7 — Particle tracks showing the burnout of coal, biogenic and plastic fuel (see [11])
For each case, the particle tracks show representatives of the whole particle spectrum, reproducing the
decrease of combustible substances in the particle. The maximum values in the legends refer in case of coal
to the initial daf-combustible substance, in case of biogenic and plastic to the initial af-combustible substance
of the particle. In comparison to coal, several differences in the combustion behaviour of the SRF fractions
can be observed. Partly they are due to the different particle size distributions, partly to the different behaviour
of the fuels themselves. Whereas the fate of the coal and the biogenic particles is characterized by time-
consuming de-volatilisation and char burnout steps, the plastic particles undergo a relatively long melting
12

---------------------- Page: 13 ----------------------

CEN/TR 15716:2008 (E)
process followed by a rapid decomposition phase (heat of decomposition is negligibly small), and transform
relatively quickly into the gaseous phase.
3 Thermal gravimetric analysis
Different methods and apparatus such as thermo-gravimetric analysis (TGA), differential thermo-gravimetric
analysis (DTG), high temperature wire mash (HTWM), and laboratory-scale batch reactors (drop tube
furnaces etc.) are used to determine kinetic data (see [6], [14]). Most of the apparatus dealing with single
particles, offers the ability to investigate kinetic properties of major fuel particle fractions, such as plastic,
paper, wood, etc. However, a transfer or prediction of how the real fuel mixture behaves is more difficult. In
this context, the requirements for sampling and sample reduction needs to be emphasised as TGA and
comparable laboratory test methods usually operate with very low sample amounts (milligrams to grams) [15].
To gain reliable results, similar effort is required to prepare test portion just like in the case for standardised
ch
...