Waste Management 33 (2013) 1825–1834

Contents lists available at SciVerse ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Production, quality and quality assurance of Refuse Derived Fuels (RDFs)

R. Sarc ⇑, K.E. Lorber
Department of Environmental and Energy Process Engineering, Montanuniversitaet Leoben, Franz-Josef-Straße 18, 8700 Leoben, Austria
Chair of Waste Treatment Technologies and Landfilling, Montanuniversitaet Leoben, Franz-Josef-Straße 18, 8700 Leoben, Austria

a r t i c l e i n f o a b s t r a c t

Article history: This contribution describes characterization, classification, production, application and quality assurance
Received 28 February 2013 of Refuse Derived Fuels (RDFs) that are increasingly used in a wide range of co-incineration plants. It is
Accepted 2 May 2013 shown in this paper, that the fuel-parameter, i.e. net calorific value [MJ/kgOS], particle size d90 or d95
Available online 5 June 2013
[mm], impurities [w%], chlorine content [w%], sulfur content [w%], fluorine content [w%], ash content
[w%], moisture [w%] and heavy metals content [mg/kgDM], can be preferentially used for the classification
Keywords: of different types of RDF applied for co-incineration and substitution of fossil-fuel in different industial
Refuse derived fuels
sectors. Describing the external production of RDF by processing and confectioning of wastes as well
Waste processing
Fluidized bed incinerator plant
as internal processing of waste at the incineration plant, a case study is reported on the application of
Quality assurance RDF made out of different household waste fractions in a 120,000 t/yr Waste to Energy (WtE) circulating
fluidized bed (CFB) incinerator. For that purpose, delivered wastes, as well as incinerator feedstock mate-
rial (i.e. after internal waste processing) are extensively investigated. Starting with elaboration of sam-
pling plan in accordance with the relevant guidelines and standards, waste from different suppliers
was sampled. Moreover, manual sorting analyses and chemical analyses were carried out. Finally, results
of investigations are presented and discussed in the paper.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction a relevant extent for the purpose of energy generation and which
satisfies the quality criteria laid down in this directive. . .’’.
A nearly unlimited broad range of solid, liquid and gaseous In fluidized bed incineration plants, different types of wastes
waste materials from household, commerce, forestry, agriculture or waste fuels (municipal and industrial solid waste, sewage
and industry, which have a certain calorific value, may be applied sludge, waste from paper production plants, etc.) are used. One
as ‘‘waste fuel’’ or ‘‘Refuse Derived Fuel’’ (RDF) in Waste to Energy of the main preconditions for waste utilization in such types of
(WtE) or co-incineration plants after having undergone different incinerators is reduction of particle size (d95 < 100 mm) in
levels of prior processing. Generally speaking, sewage sludge, mechanical treatment plants. Different types of mechanical waste
waste wood, high calorific fractions from mechanical–physical sorting plants are operated in Austria, where various SRF-quali-
(MPT) and/or mechanical–biological treatment (MBT) plants, calo- ties are produced and delivered to the incineration plant. Addi-
rific fractions of household and commercial waste, shredder light- tionally, some fluidized bed incineration plants have their own
weight fractions (e.g., from used old vehicles, electrical and mechanical treatment facilities, which ensure that the particle
electronic equipment (WEEE)), scrap tyres, food byproducts (fats, size (d95) of used incinerator feedstock material is lower than
animal meal, etc.) waste oil, used solvents and viscose plant off- ca. 100 mm.
gas, etc. may be considered as ‘‘waste fuel’’ or ‘‘Refuse Derived
Fuel’’ (RDF). In the narrow sense of definition, solid fuels which
are prepared from sorted or mixed solid wastes (municipal waste 1.1. Quality criteria for waste fuels
fractions, commercial wastes, production wastes, lightweight frac-
tions from MBT/MPT-plants, etc.) are described as ‘‘Solid Recovered The European Union Directive 2000/76/EC on the incineration
Fuel’’ (SRF). of waste (EC, 2000) limits emissions to air only, but when waste
In Austria, the definition for ‘‘waste fuels’’ or ‘‘Refuse Derived fuels are burnt in co-incineration plants, there is no limit of the
Fuels’’ (RDFs) is given in the ‘‘Waste Incineration Directive pollutant levels in the fuels, products or residues themselves so
(WID)’’ (BMLFUW, 2010) as: ‘‘. . .waste that is used entirely or to far. That is different in Austria, where energy related quality crite-
ria for waste fuels burnt in different types of co-incineration plants
are laid down in the WID (BMLFUW, 2010), see Table 1. There are
⇑ Corresponding author. Tel.: +43 3842 4025105; fax: +43 3842 4025102. also other countries (e.g., Germany and Switzerland) which have
E-mail address: [email protected] (R. Sarc). defined strict quality criteria (not always in mg/MJDM but mg/

0956-053X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.wasman.2013.05.004
1826 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834

Table 1
Limit values [mg/MJDM] for heavy metal concentrations in waste fuels, related to heating value and depending on type of co-incineration plant, laid down in the Waste
Incineration Directive (BMLFUW, 2010).

Limited parameter Cement kiln Powerstation Other plants

Median 80th Percentile Median 80th Percentile Median 80th Percentile
610% 615%
(mg/MJDM) (mg/MJDM) (mg/MJDM)
Co-incineration sector
As 2 3 2 3 2 3 1 1.5
Pb 20 36 23 41 15 27 15 27
Cd 0.23 0.46 0.27 0.54 0.17 0.34 0.17 0.34
Cr 25 37 31 46 19 28 19 28
Co 1.5 2.7 1.4 2.5 0.9 1.6 0.9 1.6
Ni 10 18 11 19 7 12 7 12
Hg 0.075 0.15 0.075 0.15 0.075 0.15 0.075 0.15
Sb 7 10 7 10 7 10 7 10

The following formula shall be used for the conversion: Pollutant content [mg/MJ] = Pollutant content [mg/kgDM]/Net calorific value [MJ/kgDM].
The following formula applies to liquid waste fuels: Pollutant content [mg/MJ] = Pollutant content [mg/kg]/Net calorific value [MJ/kg].

kgDM) for SRF used in co-incineration plants (Flamme and Geiping, Additionally, technical aspects have to be considered, when
2012). using SRF in different mono- or co-incineration plants. Technical
properties may be divided into a few groups, e.g., chemical (i.e.
separation between combustible/non-combustible substances, vol-
1.2. Characterization of waste fuels
atile matter content, etc.), mechanical (density of combustible/
non-combustible substances, bulk properties, etc.), calorific (i.e.
There are different possibilities and systems for quality classifi-
heating value, adiabatic combustion temperature, etc.) and reac-
cation and/or characterization of waste fuels (Lorber et al., 2012;
tion technical properties (i.e. ignition temperature, combustion
Flamme and Geiping, 2012; CEN, 2011).
behavior, slag formation, corrosion potential, etc.) (Beckmann
Apart from legal requirements (see Table 1), additional fuel
et al., 2012).
specifications (i.e. chemical and physical parameters) are usually
As shown in Fig. 1, fuel properties like heating value (i.e. net cal-
laid down in the contract between the RDF supplier and user,
orific value [MJ/kgOS]) and particle size i.e. d90 or d95 in [mm]
which may contain.
strongly influence the application field of industrial utilization of
Grain or particle size (d95) or (d90), net calorific value (MJ/kgOS),
waste fuels. Individual application fields are based on average val-
chlorine content (w%DM), sulfur content (w%DM), fluorine content
ues of investigations for different waste recovery plants (Pomber-
(w%DM), ash content (w%DM), water content (w%OS), biogenic car-
ger, 2007).
bon content (%), as well as bulk density (kg/m3) and restrictions
for heavy metals content (mg/kgDM) like: As, Sb, Pb, Cd, Cr, Co,
Cu, Zn, Ni, Hg, Tl, V, Sn and Mn (Lorber and Sarc, 2012). 2. Recovery of fuels from wastes
Regardless of its application in different sectors and types of
mono-incineration or co-incineration systems, RDF must fulfill For higher qualities of RDF and/or SRF, a multi-stage separation
general quality requirements in order to be safely and efficiently process is necessary for manufacturing, including the unit opera-
utilized, like: tions of classifying and sorting of waste material fractions, as well
as the separation of ferrous and non-ferrous metals and also un-
 well defined calorific value, wanted heavyweight inert materials (e.g. stones, glass, ceramics,
 low chlorine content, etc.) followed by confectioning of the fuel according to specifica-
 quality controlled composition (few impurities), tions given (Lorber and Sarc, 2012; Lorber et al., 2012).
 defined grain size, As far as net calorific value (LHV) and grain size (d95) are con-
 defined bulk density and cerned, quality controlled waste fuels (i.e. RDF and SRF) may com-
 availability of sufficient quantities with required specifications. prise a broad range between 11–25 MJ/kgOS and 5–300 mm
(Kunter and Wellacher, 2010; Lorber et al., 2011, 2012).
As shown in Fig. 2, separate treatment of different waste mate-
rial streams is necessary for processing of different RDF qualities
and specifications. In this connection, the discrimination between
2-dimensional (2D) and 3-dimensional (3D) grains is also advisable
(extensively discussed in Lorber and Sarc, 2012), because cubic
grains show different combustion behavior compared to plane par-
ticles, i.e. foils and matters alike. In any case, a multi-stage process
with repeated unit operation steps (i.e. several times crushing,
sieving and separating) is necessary for obtaining the required
quality and uniformity of waste fuels (RDF or SRF), due to the
inherent complexity of the input material ‘‘waste’’. The final qual-
ity of RDF (or SRF) will ultimately depend on the composition of
the input (feeding) materials, as well as on the extent and the
intensity of the applied recovery process.
Fig. 1. Fuel properties of SRF and application area (d90: 90th Percentile of grain size As already mentioned before (see Section 1.2), one of the
in mm). (Pomberger, 2007; Lorber et al., 2012). requirements for RDF is the availability of sufficient quantities that
 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834 1827

Fig. 2. Simplified processing scheme for different RDF specifications. (Lorber et al., 2011; Pomberger, 2007; Kunter and Wellacher, 2010).

plant for approximately 2 years (from 2011 to 2013). The CFB

waste fuel (RDF) incinerator plant has a fuel capacity of
120,000 t/yr and a heating capacity of 40 MW. According to the fire
performance graph of the plant, the heating value range is between
6 MJ/kgOS and 19 MJ/kgOS, corresponding to a throughput (i.e.
waste fuel consumption) between 21 and 5 t/h respectively. The
design point is 10 MJ/kgOS and 14 t/h. The steam parameters are
40 bar and 400 °C. (BMLFUW, 2002; UBA, 2007) In order to attain
the ‘‘energy recovery’’ status and not the ‘‘disposal status’’, the
R1 formula had to be considered for the present plant (EU, 2008).
R1 criteria (use principally as a fuel or other means to generate en-
ergy, (EU, 2008)) for energy efficiency determines, that municipal
solid waste incineration plants must have an efficiency equal to
or above:

 0.60 for installation in operation and permitted in accordance

with applicable Community legislation before January 1, 2009
Fig. 3. Relationship between recovery rate (quantity) and net calorific value and
(quality) of waste fuels. (Pomberger, 2008).
 0.65 for installation permitted after December 31, 2008 using
will match with specifications. In practice, this may lead to the the defined formula (EU, 2008).
controversial challenge between high recovery rate and high qual-
ity of the product, as shown in Fig. 3, where for municipal waste Applying this formula to the present waste to energy plant, the
(MW) and commercial waste (CM), the recovery rate of the wanted minimum energy efficiency point is 24 MW.
fuel is plotted against the required quality (i.e. net calorific value).
The solid graph represents the practical relationship (i.e. results of 3.1. Waste fuel processing for RDF-incinerator feedstock
a SRF recovery plant), whereas the dotted line reflects the theory
(i.e. results of manual sorting analysis). It becomes clear that one The input materials: Household Waste (EU waste code
cannot expect a high recovery rate and a high quality at the same 20_03_01-mixed municipal waste), Commercial Wastes (EU waste
time. For premium quality of high calorific SRF (i.e. LHV P 22 MJ/ code 20_03_01-mixed municipal waste), Bulky Waste (EU waste
kgOS) used as primary burner fuel (PBF) in cement kilns, the recov- code 20_03_07) and Construction Site Waste (EU waste code
ery rate is about 20% for commercial waste input and only 7% for 17_09_04-mixed construction and demolition wastes other than
municipal waste input. Also, it does not make sense to mix these those mentioned in 17_09_01, 17_09_02 and 17_09_03) (EPA,
two different input materials before starting the recovery process 2002) are undergoing external (i.e. by the different suppliers) and
in the SRF processing plant, as the separating conditions become internal (i.e. at the plant) treatment to produce RDF-incinerator
more complex, resulting in lower quality and/or lower recovery feedstock material.
rates of the required product.
Also the results on collection and treatment of different waste
3.1.1. External waste processing
types (i.e. household waste and commercial waste), presented in
As shown in Fig. 4, a multi-stage mechanical separation process
the study ‘‘KRIGEZ’’ (Amt der Steiermärkischen Landesregierung,
is used by the suppliers for the external processing of waste fuel
2008), confirm these findings. Separate collection of dried (i.e.
(i.e. ‘‘Premium Quality’’, ‘‘Medium Quality’’ and ‘‘EBS-Low’’
higher heating value) commercial waste and its treatment in spe-
Fraction).
cialized processing plants is recommended.
The outputs of the Star Screen separation step are assigned the
EU-waste catalogue code 19_12_12. The ‘‘Premium Quality’’ frac-
3. Case study: Refuse Derived Fuels (RDFs) for circulating tion is used in a circulating fluidized bed incinerator and/or in a ce-
fluidized bed incinerator plant ment plant as so-called ‘‘primary burner fuel’’ (PBF). The ‘‘Medium
Quality’’ fraction is used in the circulating fluidized bed incinerator
A comprehensive study was carried out on the characterization plant of this case study. The ‘‘EBS Low’’ fraction is used as HotDisc-
of waste fuels burnt in a circulating fluidized bed (CFB) incinerator Chamber feedstock of a cement plant (cf. Fig. 1).
1828 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834

Fig. 4. External mechanical sorting plant (MSP) for waste fuel processing in the reported case study.

Summarized, it means that a modern and advanced mechanical of Fe-metals), at least one eddy-current separator (rejecting of
sorting plant (MSP) for RDF consists of at least two or even three NON-Fe-metals, mostly only for fine fraction) and, depending on
shredding steps, at least two magnetic separation steps (rejection customer requirements, at least two sieving steps. Different types
 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834 1829

of multi-stage mechanical sorting plants operating in Austria are

depicted in detail in UBA (2006 and 2008).

3.1.2. Internal waste processing

The pretreated waste fuels from the suppliers are delivered to
the fluidized bed incinerator plant by walking floor trucks (ca.
21–25 t weight and 90 m3 volume) which empty their load into
the coarse material bunker, see Fig. 5. As shown in Section 3.1.1,
different types of MSP are used for external waste fuel preparation.
To increase the incinerator feedstock quality (homogeneous parti-
cle size (d95), water and ash content, cf. also Fig. 2: Requirements
and specifications) and to reduce its fuel-technical differences, an
additional internal waste processing plant is required. By using
an in-plant multi-stage shredding process, the portion of 3D-grains
is automatically reduced, which results in improved fuel-technical
characteristics. The 3D-material usually contains more impurities
(inert materials like stones). Hence, the 3D waste stream has to un-
dergo a more complex preparation process. To remove the metals
efficiently from SRF, it is important to apply magnetic separation
after each size reduction (shredder) step (Lorber et al., 2012).
After sieving, separation of Fe-metals (2) and shredding, the
material is ready for combustion and it is stored in the incinerator
feedstock bunker.

3.2. Waste fuel characteristics

Fig. 5. Internal in-plant processing for RDF-incinerator feedstock material burnt in
For investigation of the different waste fuel streams, represen- circulating fluidized bed incinerator plant.
tative sampling was performed from the unloading walking floor
trucks (i.e. pretreated coarse input bunker material) in accordance
with Section 4, as well as from the material stream after internal as shown in Figs. 6 and 7, with a field sample (composed of 10
RDF processing at the incineration plant (i.e. RDF-incinerator feed- increments) amounting to between 90 and 100 kg for each
stock material ready for combustion). The sampling procedure was material stream.

Fig. 6. Schematic description of representative sampling for waste fuels in accordance with relevant standards (- - Collective Sample- - may be skipped). (Lorber et al., 2012).
1830 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834

Truck Delivery PROCEDURE IN FIELD PROCEDURE IN LAB

10 Increment
Increments m = ca. 10 kg
a ca. 10 kg 10 10 d95 < 120 mm

x 10

Combined
Sample a Combined Sample
1. 2. m = 10 x 10 = 100 kg Sorting Out of Extraneous
90 - 100 kg
(< 120 mm) (min m = 50 kg) Materials
d95 < 120 mm

Size
Reduction
Size Reduction 1 Drying
d95 : 120 mm => 30 mm 40 °C

1. 2.
Field Sample
Size Reduction 2
Field Sample
Coarse Cutting Mill
m = 100 kg
m = ca. 3 kg
d95 < 30 mm
Mass d95 : 30 mm => <10 mm
Reduction

Sorting Out of Extraneous

Materials Mass Reduction Mass Reduction
100 kg => 50 kg ca. 3 kg => 1.0 - 1.3 kg
PROCEDURE 25 kg => 12.5 kg => ca. 6 kg d95 < 10 mm
IN
LAB

Size
Reduction Size Reduction 3
Reference Sample Lab Sample
Fine Cutting Mill
m = ca. 3 kg m = ca. 3 kg
m = 1.0 - 1-3 kg
d95 < 30 mm d95 < 30 mm
(min m = 0.8 kg)
d95 : < 10 mm => 0.5 mm (0.25 mm)

1. 2.
Test Sample
Test Samples
Determination of
Examination Storage Parameters

Fig. 7. Concept (sampling procedure and sample preparation scheme) for waste materials delivered to incinerator plant by truck. (Reproduced from Lorber et al., 2012).

3.2.1. Characterization of supplier materials from external waste catalogue code 200301) are also reported in Table 3. It can be seen
processing that samples from the waste fuel suppliers (i.e. A1, A2 and B) con-
The samples taken from the walking floor trucks have been tain a significantly greater share of ‘‘Fine Fraction < 16 mm’’ (A1, A2
investigated and analyzed according to the relevant standards and B) and of ‘‘Plastics and Lightweight Fraction’’ (A2 and B) but
(ASI, 2006, 2009, 2011a, 2011b, 2011d) and some summarized re- less portions on ‘‘Organic/biogenic waste’’ and ‘‘Paper, cardboard
sults are presented in Tables 2 and 3. and cardboard packaging’’.
It becomes obvious that the treated waste materials delivered
from the suppliers A and B are showing a rather heterogeneous
composition (i.e. distribution of Cl, Ash, Cu). Also the waste fuel in- 3.2.2. Characterization of RDF-incinerator feedstock material after
put streams A1 (having greater portions of construction site internal waste processing
waste), A2 (having greater portions of household waste) and A3 Samples taken from the incinerator feedstock bunker (see
(having greater portions of commercial waste), which are originat- Fig. 4) have been investigated and analyzed according to the rele-
ing from the same mechanical sorting plant shown in Fig. 4 are vant standards (ASI, 2006, 2009, 2011a, 2011b, 2011c, 2011d) and
quite different, as far as their physical–chemical composition is some selected results are presented in Tables 4 and 5.
concerned (e.g., LHV). The data given in Table 4 reveals that by internal waste process-
The results from manual sorting analysis (field sample size ca. ing applied at the fluidized bed incinerator plant (see Fig. 5), a
90–100 kg each) shown in Table 3 confirms the outcome from homogenization of the different input material streams from sup-
the physical–chemical analyses of the supplier materials shown pliers takes place. Compared to the waste fuel supplier materials
in Table 2. The low net calorific values of sample A1 (and A3) can (i.e. A1, A2, A3 and B), the RDF-incinerator feedstock is more
be explained by the high portion of ‘‘Fine Fraction < 16 mm’’ and homogeneous for most of the measured parameters (e.g., LHV,
‘‘Inert Materials’’. For comparison, the results of the Federal Waste Cl). However, it is remarkable, that the mean concentration value
Management Plan 2011 (BMLFUW, 2011) on the Austrian residual for copper with 892 mg/kgDM is significantly higher compared to
waste from household and similar establishments (EU waste the 80th Percentile with 604 mg/kgDM. This reflects the extreme
 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834 1831

Table 2 Table 4
Selected results (mean values, n = 3, time period 2012) from physical–chemical Selected results (mean value, n = 48, time period 2011–2013) from physical–chemical
analysis of investigated supplier materials (A1, A2, A3 and B). (Note: OS: original analyses of investigated RDF-incinerator feedstock material. (Note: OS: original
substance, DM: Dry Matter, Inerts: incombustible materials like glass, stones and substance, DM: Dry Matter, Inerts: incombustible materials like glass, stones, and
metals, A1, A2 and A3: different truck deliveries from the same supplier A). metals).

Parameter Unit Austrian Supplier materials Parameter Unit Austrian Mean Standard Median 80th
Standard (ASI) Standard deviation Percentile
A1 A2 A3 B
(ASI) (%)
LHV MJ/kgOS 15400 7 15 6 12
LHV MJ/kgOS 15400 9.0 12.3 8.9 10.3
Cl g/kgDM 15408 3 25 9 8
Cl g/kgDM 15408 9.3 35.8 9.2 12.3
S g/kgDM 15408 5 5 7 4
S g/kgDM 15408 3.4 46.0 2.9 3.7
F g/kgDM 15408 0.2 0.2 0.2 0.2
F g/kgDM 15408 0.18 60.9 0.14 0.22
Ash w% 15403 55 20 50 30
Ash w% 15403 41 16.6 40 45.4
DM w% 15414-1 75 80 70 80
DM w% 15414-1 73 6.6 74 77.6
Inerts w% – 20 5 3 9
Inerts w% – 21 30.1 21 26.7
Pb mg/kgDM 15411 250 550 300 100
Pb mg/kgDM 15411 312 133 215 334
Cd mg/kgDM 15411 3 10 3 2
Cu mg/kgDM 15411 892 230 190 604
Cr mg/kgDM 15411 150 550 250 150
Zn mg/kgDM 15411 882 126 580 955
Cu mg/kgDM 15411 900 100 300 200
K mg/kgDM 15410 1986 34.9 1919 2738
Ni mg/kgDM 15411 250 50 20 60
Na mg/kgDM 15410 3011 21.8 2972 3681
Zn mg/kgDM 15411 1500 500 1500 500

Table 5
Table 3
Selected results (mean value, n = 11, total field sample size ca. 100 kg, time period
Selected results (single values, time period 2012) from manual sorting analysis of
2012) from manual sorting analysis of RDF-incinerator feedstock, compared to results
supplier materials A1, A2 and B compared to results from Federal Waste Management
from FWMP 2011 for residual waste (Federal Waste Management Plan) (BMLFUW,
Plan for residual waste (FWMP) (BMLFUW, 2011).
2011).
Fractions (w%) Supplier FWMP 2011
Fractions (w%) RDF feedstock (waste FWMP 2011 (waste
materials
code 191212) code 200311)
A1 A2 B
Fine Fraction < 20 mm 55 19.6
Fine Fraction < 16 mm 50 30 50 19.6 Organic/biogenic waste 6 20.5
Organic/biogenic waste 8 10 10 20.5 Paper, cardboard and 6 12.4
Paper, cardboard and cardboard packaging 6 9 5 12.4 cardboard packaging
Sanitary articles 0 0 0 8.2 Sanitary articles 1 8.2
Plastics and lightweight fraction 10 33 19 9.7 Plastics and lightweight 16 9.7
Composite materials 3 5 2 9.5 fraction
Textiles 1 9 5 5.8 Composite materials 3 9.5
Glass 3 0 0 4.3 Textiles 5 5.8
Inert materials 15 1 3 3.4 Glass 4 4.3
Metals 2 2 4 2.9 Inert materials 3.4
Hazardous household waste 0.1 0.2 0.4 1.2 Metals 1 2.9
Other (unidentified) 1.9 0.8 1.6 2.5 Hazardous household 3 1.2
Total 100 100 100 100 waste
Other (unidentified) 2.5
Total 100 100

heterogeneity of this element in the investigated RDF-incinerator

feedstock material.
The Federal Waste Management Plan 2011 contains actual pub- feeding material in cement kilns) in Austria, the consumer has to
lished data on the composition of residual waste from household pay the supplier (contractor), but for lower RDF qualities, the co-
and similar establishments in Austria. Residual waste (EU waste incineration plant receives money from the waste fuel supplier.
catalogue code 20_03_01) is defined as all the solid waste that is This makes Quality Assurance of RDF (or SRF) an important issue.
normally generated by households and similar establishments, ex- Apart from legal requirements (see Section 1.1), additional specifi-
cept for bulky waste and separately collected waste, such as cations (see Section 1.2) are usually laid down in the contract be-
recoverables (paper, glass, metal, plastics, etc.), biogenic waste tween RDF supplier and user. Ensuring the necessary legal
and hazardous household waste. compliance with the limit values and additional specifications gi-
Comparing the RDF-incinerator feedstock with the residual ven in the supply contract requires monitoring of fuel quality.
waste, it becomes evident that RDF-incinerator feedstock contains For this, two different approaches are common (Lorber et al.,
significantly higher proportions of ‘‘Fine Fraction < 20 mm’’, ‘‘Plas- 2012):
tics and lightweight fraction’’ but much less percentage of ‘‘Organ-
ic/biogenic waste fraction’’, ‘‘Sanitary articles’’, ‘‘Composite – Supplier control model: External quality control of RDF (or SRF)
materials’’ as well as ‘‘Metals’’. for legal compliance is done by the suppliers, who pass down
This is due to the external and internal waste fuel processing all the relevant information to the consumer. In this case, the
steps (i.e. multi-stage shredding, repeated Fe-metals separation, suppliers have to prepare a sampling plan and conduct all
etc.) reported in Section 3.1. required analytical measurements, and the consumer (i.e. the
co-incineration plant) has to take random samples of the
incoming SRF in order to check the quality of the material.
4. Quality assurance for RDF These validation checks have to be carried out at least once a
year. If contamination of the waste fuel is suspected in the
For premium quality, high calorific waste fuels (e.g. SRF with course of a visual check or if the waste fuel does not seem right,
net calorific value P 22 MJ/kgOS that is used as primary burner further samples have to be taken and examined. For the
1832 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834

validation check, the analytical examination (lot size) refers to Table 6

an amount of 150 t or – in case of waste streams >40,000 t/yr Calculation of field sample amount (consisting of 10 increments) for different particle
size (d95) of delivered waste fuel. Bulk density of sample material (kb) is assumed to
– to an average daily amount. be about 250 kg/m3 for walking floor truck.
– Consumer control model: The incoming SRF delivered by differ-
ent suppliers is examined by the co-incineration plant itself. Particle size d95 kb (kg/m3) Increment Field sample
(mm) amount (kg) amount (kg)
Therefore, a sampling plan as well as a sample preparation con-
cept have to be established according to the norms, standards 60 250 1.5 15
70 250 2.3 23
and guidelines given or recommended by the authorities (Flam- 80 250 3.5 35
me and Geiping, 2012; ASI, 2011e). The volume and intensity of 90 250 4.9 49
investigations to be performed for quality assurance depend 100 250 6.8 68
strongly on the amount and frequency of SRF-deliveries to the 110 250 9.0 90
120 250 11.7 117
plant. A sampling plan is to be set up according to CEN/TS
15442 (ASI, 2011e) for every type and origin of SRF separately.

The quality assurance of SRF is comprehensively described in

The detailed sampling concept applied for the investigation of
Lorber et al. (2012).
waste deliveries by truck to the waste fuel incinerator plant is de-
picted in Fig. 7.
4.1. Representative sampling of investigated waste fuels

5. Results and discussions

An important part of Quality Assurance is the representative
sampling of waste fuel, schematically depicted in Fig. 6. The classi-
One could expect that the resulting final quality of waste fuels
cal analysis for quality assurance of solid recovered fuels is divided
like RDF (or SRF) will depend on the origin of the input materials
into:
as well as on the volume and intensity of the separation process
applied in fuel manufacturing. Looking at Tables 2 and 3, the qual-
 sampling in accordance with the relevant guidelines and
ity difference between the three materials, A1, A2 and A3, that
standards,
comes from one supplier only, is quite surprising. Apparently, the
 sample treatment including its digestion depending on require-
type of waste input into the RDF processing plant is more impor-
ments for the analysis to follow,
tant than the fuel recovery process itself. A high calorific, premium
 analysis and measurement of element concentrations and
quality of RDF can be recovered much more easily, when sepa-
 evaluation of results and application of different assessment
rately collected fractions of ‘‘clean’’ input materials are used (see
methods. (Flamme and Geiping, 2012; Rotter et al., 2011)
also Section 2). From Table 4 it becomes clear that the distribution
of analysis parameter in the readymade RDF-incinerator feedstock
During each step, possible uncertainties must be kept as low as
can be extremely different, due to the inherent heterogeneity of
possible. According to the error propagation law, errors add up and
waste fuel. E.g., the parameter LHV and Na, are more homogenous-
lead to a total error, so they cannot be considered in isolation. Sta-
ly distributed compared to the heavy metals Pb, Cu and Zn, which
tistical parameter of variance (s2) characterizes the error, whereby
are showing a relative standard deviation (SD) > 100%. Actually,
the total error is calculated as follows (Flamme and Geiping, 2012):
when comparing Tables 2 with 4, it becomes evident that the con-
S2total ¼ S2sampling þ S2sample treatment þ S2analysis þ S2interpretation of data ð1Þ tent of heavy metals (Pb, Cu and Zn) is more or less the same in the
supplier’s material and in the final RDF-incinerator feedstock. That
The statistical methods are described in detail in ASI (2011e), Flam- is plausible, because as shown in Figs. 3 and 4, neither the external
me and Geiping (2012) and Lorber et al. (2012). The minimum nor the internal waste fuel processing plant contains an Eddy-Cur-
amount of sample required for one increment (mm) may be calcu- rent separation step for removing NON-Fe metals out of the mate-
lated by the empirical equation: rial stream treated for waste fuel recovery. On the other hand, the
two stage air classifier shown in Fig. 4, which is integrated within
3
mm ½kg ¼ 2:7  108  d95 ½mm  kb ½kg=m3  ð2Þ the external waste treatment plant, apparently works quite effec-
tively in separating ‘‘Inerts’’ out of the waste stream, which results
with: in a relatively low standard deviation (i.e. about 20%) for the heat-
d95: 95th Percentile of grain size; ing value. A crucial parameter for waste fuel is chlorine (Cl) as well
kb: bulk density of sample material. as the ratio Cl:S, because it strongly influences the corrosion
For the determination of density of delivered waste, the starting behavior during combustion of waste fuel. Unlike fossil fuels,
situation is the material weight of one delivery, i.e. a truckload. The waste fuels usually contain more chlorine (Cl) than sulfur (S) and
bulk density of waste material (Medium Quality, cf. Fig. 4) is ca. the Cl:S-ratio is >1, making corrosion a severe problem in RDF-uti-
140 kg/m3, but in the truck the material is compressed in order lization (Lorber et al., 2011; Spiegel et al., 2012).
to fully use the truck load capacity. The fact is, that the density As shown in Tables 2 and 4, with the exception of sample A2,
of a loaded truck has to be used for the calculation of increment the chlorine content in the investigated waste fuel is usually less
weight, because the sampling weight, as shown in the empirical than 1%. Chlorine <1% in waste fuel has become a standard require-
formula (2), is directly related to the bulk density. That means that ment for RDF and SRF (Lorber and Sarc, 2012).
the sampling weight has to be in connection with the uploaded As shown in Table 2, the supplier material A2 is showing a rel-
material mass and its volume. A delivery by a walking floor truck, atively high net calorific value (LHV) of about 15 MJ/kgOS, together
as mentioned in Section 3.1.2, is between 21 and 25 tons and its with an extreme Cl-content of 2.5%. Additionally, the proportion of
volume is approx. 90 m3. By the assumption of 22.5 tons per truck the ‘‘Fine Fraction < 16 mm’’ is relatively small. This clearly indi-
and a truck volume of 90 m3, the mean value of bulk density be- cates that PVC-packaging material (probably from commercial
comes 250 kg/m3. Using Eq. (2), the calculation of sample amounts waste input) was enriched during the external recovery process
for one increment and the resulting sample amounts of field sam- shown in Fig. 4. To get chlorine out of the waste stream and to ob-
ples (i.e. 10 increments) is shown in Table 6. tain premium quality grade waste fuels (like primary burner
 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834 1833

Fig. 8. Simplified waste processing scheme for production and utilization of high quality RDF/SRF.

substitute fuel in the cement industry), a NDIR-sensor process step Acknowledgements

has to be integrated into the multi-stage waste fuel recovery pro-
cess, like the one used at the THERMOTEAM plant in Retznei, Aus- The authors are very grateful to the plant engineers and the
tria (Lorber and Sarc, 2012), by which the decreasing of the management of the waste fuel incinerator plant of our reported
chlorine content, lower than 0.6%, should be reached (Pomberger case study for enabling us to sample suppliers’ materials and
and Curtis, 2012). RDF-incinerator feedstock. In the interest of confidentiality, we
As already mentioned before, for premium quality of high calo- are not mentioning names. Many thanks also to our co-workers
rific waste fuel (RDF or SRF), the consumer has to pay the supplier, of the Chairs laboratory, who supplied us with numerous analytical
but for lower qualities it is different. data. Finally, the corresponding author would also like to express
This simple fact is causing a conflict of interests between tech- his sincere gratitude to Professor Karl E. Lorber, Chair of Waste
nical and economical requirements of a waste fuel incineration Treatment Technologies and Landfilling, for excellent scientific
plant: It receives more money from the suppliers of waste fuel supervision and extensive and inspiring discussions.
when it accepts a lower quality (lower heating value, higher con-
tent of chlorine, heavy metals and impurities (‘‘Inerts’’), less homo-
geneity) of RDF. This apparently also may be true for the reported References
case study, where low calorific waste fuel (cf. Table 4, mean: 9 MJ/
kgOS) is burnt in a plant designed for a fuel range between 6 and Amt der Steiermärkischen Landesregierung, Fachabteilung 19D (FA 19D), 2008.
Kriterien zur Trennung von Siedlungsabfall aus Industrie und Gewerbe als
19 MJ/kgOS and with a design point of 10 MJ/kgOS. Voraussetzung zur Zuordnung zu Behandlungsverfahren (KRIGEZ) (Criteria for
separation of municipal waste from industry and trade as precondition for its
allocation to the treatment processes). Austria.
Austrian Standards Institute (ASI), 2006. ÖNORM CEN/TS 15414-1-Solid Recovered
6. Conclusions
Fuels – Determination of Moisture Content using the Oven Dry Method – Part 1:
Determination of Total Moisture by a Reference Method. Vienna, Austria.
RDF or SRF is waste fuel that is normally made of high calorific Austrian Standards Institute (ASI), 2009. ÖNORM EN 15403-Solid Recovered Fuels –
fractions of municipal waste, commercial or industrial waste, bulky Determination of Ash Content. Vienna, Austria.
Austrian Standards Institute (ASI), 2011a. ÖNORM EN 15400-Solid Recovered Fuels
waste and construction site waste, which satisfies the fuel quality – Determination of Calorific Value. Vienna, Austria.
criteria according to the Austrian ‘‘Waste Incineration Directive’’ Austrian Standards Institute (ASI), 2011b. ÖNORM EN 15408-Solid Recovered Fuels
(BMLFUW, 2010) and the ‘‘Guideline for Wastes’’ (BMLFUW, – Methods for the Determination of Sulphur (S), Chlorine (Cl), Fluorine (F) and
Bromine (Br) Content. Vienna, Austria.
2008). Due to the inherent heterogeneity of wastes and waste Austrian Standards Institute (ASI), 2011c. ÖNORM EN 15410-Solid Recovered Fuels
fuels, the deviation (cf. Section 4.1) of analytical results on the con- – Methods for the Determination of the Content of Major Elements (Al, Ca, Fe, K,
tent of constituents (e.g., heavy metals) can be quite large. This Mg, Na, P, Si, Ti). Vienna, Austria.
Austrian Standards Institute (ASI), 2011d. ÖNORM EN 15411-Solid Recovered Fuels
may be one of the reasons why the limit values laid down in the – Methods for the Determination of the Content of Trace Elements (As, Ba, Be,
Waste Incineration Directive (BMLFUW, 2010) are statistically ex- Cd, Co., Cr, Cu, Hg, Mo, Mn, Ni, Pb, Sb, Se, Tl, V and Zn). Vienna, Austria.
pressed as Median and 80th Percentile, conceding that 20% of the Austrian Standards Institute (ASI), 2011e. ÖNORM EN 15442-Solid Recovered Fuels
– Methods for Sampling. Vienna, Austria.
measured values may be ‘‘runaways’’ on the higher side. Utilization Beckmann, M., Pohl, M., Bernhardt, D., Gebauer, K., 2012. Criteria for solid recovered
of RDF depends on characteristic fuel parameters, such as calorific fuels as a substitute for fossil fuels – a review. Waste Management and Research
value (11 MJ/kgOS 6 LHV 6 25 MJ/kgOS), grain size 30 (4).
Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft
(5 < d95 < 300 mm), chlorine content, mercury content and heavy
(BMLFUW), 2002. Stand der Technik bei Abfallverbrennungsanlagen (State of
metals content (CEN, 2011; BMLFUW, 2010). Production of high the Art by Waste Incineration Plants). Vienna, Austria.
quality RDF requires a multiple-stage waste treatment and separa- Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft
tion process (external, internal or combination of both (cf. Fig. 8)) (BMLFUW), 2008. Richtlinie für Ersatzbrennstoffe (‘‘Guideline for Waste Fuels’’),
Vienna, Austria.
like the one installed at specialized SRF production plant THERMO- Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft
TEAM (Lorber and Sarc, 2012). Common unit operations applied in (BMLFUW), 2010. Verordnung über die Verbrennung von Abfällen.
RDF production are crushing and shredding, classifying, rejects Abfallverbrennungsverordnung (AVV) (‘‘Waste Incineration Directive’’).
Vienna, Austria.
separation and confectioning. Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft
Basic requirements for utilization of RDF in co-incineration (BMLFUW), 2011. Bundesabfallwirtschaftsplan (‘‘Federal Waste Management
plants are legal compliance, legal validity of operating license Plan 2011’’). Vienna, Austria.
CEN, 2011. BS EN 15359:2011-Solid Recovered Fuels – Specifications and Classes.
and guarantee of supply of sufficient quantities with the required Brussels, Belgium.
quality, as well as quality assurance. Quality assurance is based European Commission Directive (EC), 2000. Directive 2000/76/EC of the European
on ‘‘CEN/TC 343 – Solid recovered fuels’’ guidelines and ‘‘Austrian Parliament and of the council of 4th December 2000 on the incineration of
wastes. Brussels, Belgium.
Standards Institute’’ norms (ASI). Potential problems with RDF uti- Environmental Protection Agency (EPA), 2002. European Waste Catalogue and
lization are plant corrosion (due to Cl, S, Na, K, heavy metals and Hazardous Waste List. Ireland.
‘‘inerts’’ content), as well as fire risk in production and storage. European Union (EU), 2008. Directive 2008/98/EC of the European Parliament and
of the Council of 19. November 2008 on waste and repealing certain Directives).
(Lorber et al., 2011; Spiegel et al., 2012) Due to limited production
Brussels, Belgium.
and increased demand for quality controlled high calorific pre- Flamme, S., Geiping, J., 2012. Quality standards and requirements for solid
mium SRF, consumers (e.g., cement plants) are already paying for recovered fuels – a review. Waste Management and Research 30 (4).
this waste fuel. The lower the waste fuel quality, the lower is the Kunter, A., Wellacher, M., 2010. Neue Entwicklungen bei der
Ersatzbrennstoffaufbereitung (Refuse Derived Fuel Production News). In:
payment from consumer to the supplier. For low RDF/SRF qualities Lorber K.E. et al. (Eds.), Proceedings of the 10th DepoTech Conference,
supplier has to pay the consumer. Leoben, Austria.
1834 R. Sarc, K.E. Lorber / Waste Management 33 (2013) 1825–1834

Lorber, K.E., Sarc, R., Pomberger, R., 2011. Herstellung und Einsatz von Production and Recovery of Refuse Derived Fuels in Austria). In: Thomé-
Ersatzbrennstoffen (EBS) in Österreich (Production and Application of Refuse Kozmiensky, K.J. Beckmann, M. (Eds.), Energie aus Abfall – Band 9. Neuruppin:
Derived Fuels in Austria). In: Waste-to-Resources 2011, 4. Internationale TK Verlag Karl Thomé-Kozmiensky.
Tagung MBA und Sortieranlagen Kühle-Weidemeier (Ed.), Cuvillier Verlag Rotter, V.S., Lehmann, A., Marzi, T., Möhle, E., Schingnitz, D., Hoffmann, G., 2011.
Göttingen, Germany. New techniques for the characterization of refuse-derived fuels and solid
Lorber, K.E., Sarc, R., 2012. Waste to energy by preparation of quality controlled recovered fuels. Waste Management and Research 29 (2).
solid recovered fuels (SRF). In: Proceedings of 4th ICET Conference, Hefei, China, Spiegel, W., Magel, G. Müller, W., 2012. Einfluss von Calciumspezies auf Korrosion
2012. und Verschmutzung in MVA und EBS-Kraftwerken (Impact of Calcium Phases
Lorber, K.E., Sarc, R., Aldrian, A., 2012. Design and quality assurance for solid on Corrosion and Fouling in Waste-to-Energy Plants). In: Lorber, K.E. et al.
recovered fuel. Waste Management and Research 30 (4). (Eds.), Proceedings of the 11th DepoTech Conference, Leoben, Austria.
Pomberger, R., 2007. Ersatzbrennstoffe aus Siedlungsabfällen – Anforderungen, Umweltbundesamt (UBA), 2006. Ist-Stand der Mechanisch-Biologischen
Möglichkeiten, Betriebserfahrungen (Solid recovered fuels from municipal Abfallbehandlung (MBA) in Österreich – Zustandsbericht 2006 (Status quo of
waste – Requirements, Opportunities, Operating Experiences). mechanical biological treatment (MBT) in Austria – State Report 2006). Vienna,
Vortragsmanuskript zu ÖWAV/UBA – Tagung. Vienna, Austria. Austria.
Pomberger, R., 2008 Entwicklung von Ersatzbrennstoff für das HOTDISC-Verfahren Umweltbundesamt (UBA), 2007. Abfallverbrennung in Österreich – Statusbericht
und Analyse der abfallwirtschaftlichen Relevanz (Development of SRF for the 2006 (Waste Incineration in Austria – State Report 2006). Vienna, Austria.
HOTDISC-process and evaluation of it’s importance for waste management). Umweltbundesamt (UBA), 2008. Mechanische Abfallbehandlung (MA) von
PhD Thesis at IAE-Institute, Montanuniversitaet Leoben. gemischten Siedlungs- und Gewerbeabfällen in Österreich (Mechanical Waste
Pomberger, R., Curtis, A., 2012. Neue Entwicklungen bei der Produktion und Treatment (MT) of mixed Household and Commercial Waste in Austria).
Verwertung von Ersatzbrennstoffen in Österreich. (New developments in Austria, Vienna.