Waste Management 21 (2001) 753 765
www.elsevier.com/locate/wasman
Characteristics, treatment and utilization of
residues from municipal waste incineration
a, b c
H.A. van der Sloot *, D.S. Kosson , O. Hjelmar
a
ECN Soil & Waste Research, PO Box 1, 1755 ZG Petten, The Netherlands
b
Vanderbilt University, Box 1831 Station B, Nashville, TN 37235, USA
c
DHI, Agernallee 11, 2960 DK HÅ‚rsholm, Denmark
Received 28 February 2000; received in revised form 23 December 2000; accepted 2 January 2001
Abstract
Beneficial utilization of residues from municipal solid waste incineration is an important objective for integrated waste manage-
ment in many jurisdictions. When residues are to be used as an aggregate substitute in construction applications, the release of
constituents of concern to soils and water through leaching is an important environmental consideration. In this paper, residue
characteristics that control constituent leaching and testing approaches for evaluating leaching are discussed. Quality control and
potential improvement in case of beneficial application are addressed. # 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Leaching; Treatment; Waste incineration; Residues; Environmental impact; Ash
1. Introduction and long-term environmental acceptability of such uti-
lization scenarios. In addition to the environmental
From waste-to-energy conversion of municipal solid aspects, the new material will have to meet technical
waste, in which high standards of emission control are specifications similar to those of natural materials tra-
reached, solid residues remain, including bottom ash, fly ditionally used for the same purpose. In Europe, utili-
ash and air pollution control residues, that need to be zation of MSWI bottom ash is either practised (e.g. the
dealt with in an environmentally acceptable manner. Netherlands, Denmark, Germany, France) or increas-
Advanced air pollution control measures in incinerators ingly considered as a viable option (e.g. Belgium).
shift constituents of concern from air emissions to solid National legislation has been implemented to regulate
residues, which potentially may lead to soil and water utilization of MSWI bottom ash in the Netherlands [2]
pollution. The evaluation of the environmental quality and in France [3]. In the Netherlands, bottom ash is
of such residues is necessary before decisions can be taken placed in a special category, because bottom ash, as
on the utilization, treatment or disposal of the residues. currently produced, does not always meet the regulatory
The quality of the residues from waste-to-energy conver- requirements. It is anticipated that improvement in ash
sion is very diverse, as has been detailed in the Interna- quality will bring the material within the regulatory
tional Ash Working Group s (IAWG) book   Municipal specifications [2]. Quality control programmes and cer-
Solid Waste Incinerator (MSWI) Residues  [1]. tification of bottom ash are in progress to ensure pro-
Management practices for incinerator residues are duction of a marketable product [4].
very different in different jurisdictions. As a result of In this paper, recent developments in characterization
recent developments in waste management, considera- of environmental properties of MSWI residues are dis-
tion is given to recycling and reuse of residues in con- cussed. Leaching data and relevant information from
struction. This necessitates a judgement on the short field studies are interpreted to reach conclusions on how
to predict long-term environmental impact. Since
MSWI residues as produced often do not meet envir-
* Corresponding author. Tel.: [phone]; fax: +31-224-
onmental criteria, treatment options are discussed. One
563-163.
E-mail address: [e-mail] (H.A. van der Sloot).
of the options is to control the input through waste
0956-053X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0956-053X(01)00009-5
754 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
Fig. 1. Leaching behaviour of Cd from MSWI residues (bottom ash, fly ash [11]) in comparison with Refuse derived Fuel (RDF) ash [1] and vitrified
MSWI fly ash [1]). All leaching tese were carried out at L/S=10 1/kg. Data illustrate role of increasing Cl-concentration on Cd leachability from
MSWI and related residues.
acceptance criteria. For this latter aspect, the relation- acterization of MSWI residues are now being standar-
ships between input and ultimate leaching behaviour of dized (Working Group 6). In the framework of this
residues are needed. Working Group a methodology guideline (ENV 12920)
has been described for the assessment of long-term
leaching behaviour [7]. It advocates evaluation of utili-
2. Materials and methods zation and disposal options based on the specific man-
agement scenario. This aspect will be addressed in more
The work mainly deals with MSWI residues. Some detail later.
earlier published data [1] have been used as reference for Comparison of leaching behaviour of different MSWI
more recent results. The sampling of MSWI residues residues and treated residues is possible through the
was carried out in accordance with the recommenda- information generated by the pH dependence test. Fig. 1
tions of the IAWG [1]. The leaching test methods presents the leaching behaviour of Cd from different
applied are preliminary versions of the pH dependence MSWI residue streams. The different leaching curves of
leaching test [5] and the percolation test [6] now stan- Cd as a function of pH are largely related to the chlo-
dardized in CEN TC 292 Characterization of waste  ride content in the residue for pH greater than 7. The
Working Group 6. amount released at pH less than 5 usually reflects the
total content of Cd in the residue stream, as almost all
Cd present is leachable. Refuse Derived Fuel (RDF) ash
3. Environmental quality of residues and vitrified MSWI fly ash are included for comparison.
The danger of indiscriminate use of an extraction test,
The main leaching character of MSWI residues has which provides data at only one condition, is illustrated
been largely outlined in the IAWG book [1]. Since the through the results of a single extraction test at the own
time of publication, more detailed information has been pH of the material. Clearly, this very limited evaluation
generated following this same basis of characterization could result in erroneous classification of a material1. If
by using the pH dependence leaching test and percola- local equilibrium is assumed, the actual release can be
tion tests [5,6]. Also a stronger emphasis has been quantified from the pH dependence leach curves of Cd
placed on modelling leaching behaviour to identify the with varying Cl levels. Fig. 1 illustrates that, if the pH
solubility controlling phases and binding mechan- would decrease over the long-term as a result of external
isms. In the framework of the European Standardi- influences, the leachability of Cd from RDF ash and
zation Committee (CEN) Technical Committee (TC) MSWI fly ash may increase dramatically. A compen-
292 Characterization of Waste, the two leaching
procedures  pH dependence test and percolation up- 1
The   own pH  of a material is the final pH of the leaching solu-
flow test  identified as forming the basis for char- tion when the material is extracted with deionized water.
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 755
sating factor for this dramatic increase may occur if Cl due to the Cl concentration and total Cd content in the
is released under high pH conditions when Cd is not material (note that the plateau is at pH<7 for high Cl
leachable, and by the time the pH decreases no Cl is concentrations but pH<5 for low Cl concentrations).
left to mobilize the Cd. In this case, the dynamics of For Crm the difference in leaching behaviour of bottom
leaching can lead to a substantially lower release in ash and fly ash at pH>5 is attributed to the presence of
practice. Such processes illustrate the need to carry out Cr as chromate in flue gas cleaning wastes, whereas in
more detailed characterization tests such as percolation bottom ash Cr is present as Cr III due to its initial slightly
tests for some cases. Batch leaching tests will not always reducing properties. In case of Mo, the leaching beha-
clarify such interrelated mechanisms. However, no viour as a function of pH is typical for an oxyanion [8].
extent of characterization will result in a perfect predic- In fly ash, the Mo level is increased relative to bottom
tion of future environmental performance and a balance ash but the shape of the curve is similar, indicating no
must be maintained to obtain information necessary for significant difference in chemical speciation between the
a sufficiently, but not overly, conservative decision and two types of residues. The difference between different
excessive testing requirements should be avoided. MSWI residues is not very marked at pH 8 12 in spite of
The consistency of data from different sources both in large differences in total composition. At pH>7 the level
time and place is illustrated in Figs. 2 and 3. For the ele- available for leaching is reached (plateau) with a difference
ments Cd, Cr, Mo and Zn, the characteristic behaviour of of only a factor of 2 3 between fly ash and bottom ash.
MSWI bottom ash and MSWI fly ash is clear. In case of The pH dependence leaching test provides a very
Cd the leaching from fly ash is similar in shape but shifted good means of mutual comparison of MSWI residues as
Fig. 2. Characteristic leaching behaviour of Cd and Cr from MSWI residues, in particular MSWI, bottom ash and MSWI fly ash [1,11,28]. Leaching
experiments carried out at L/S=10 1/kg. The smooth lines reflect the generic leaching behaviour of bottom ash (two data clusters) and fly ash.
BA=bottom ash; DTL=detection limit.
756 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
well as a comparison of leaching behaviour within one From V and Mo leachability as a function of pH and
and the same type of MSWI residue. as a function of contact time, further conclusions can be
drawn with respect to mechanisms of release, which in
turn can be used to guide activities to improve ash
4. Parameter settings in pHdependence and column tests quality and in assessing short- and long-term environ-
mental impact. Mo is leached as molybdate (oxyanion)
In the studies carried out to provide a basis for para- and its mobile fraction is almost completely washed out
meter setting in these standards worked out by CEN TC within L/S=2 l/kg. Molybdenum is therefor an example
292 Working Group 6, MSWI bottom ash has been of availability controlled leaching. When the percola-
used as a reference material, which owing to its hetero- tion test data are plotted in the pH dependence graph
geneous nature cannot be considered a simple choice. (test carried out at L/S=10 l/kg) the cumulative leached
Aspects such as particle size, contact time and percola- amounts approach the pH stat curve (determined at L/
tion rate have been addressed. In Fig. 4 the aspects S=10 l/kg) at the relevant pH from L/S=2 l/kg onwards.
percolation rate and contact time are highlighted. This indicates consistency between the two fully indepen-
Results of column tests performed at different flow dent leaching tests. Obviously, when concentration data
velocities (particle size: 95% < 4 mm) and pH depen- from the column test at low L/S are plotted in compar-
dence tests at different contact times (particle size: 95% ison to the pH dependence curve, greater concentrations
< 2 mm) are shown in Figs. 4 and 5, respectively. These are observed from the column test. However, this is also
results indicate that the hydraulic retention time within relevant information, as it will indicate what concentra-
the column is not a critical parameter within the range tions may be expected in leachate in the short-term or
of retention times evaluated. even long-term, depending on the level of infiltration.
Fig. 3. Characteristic leaching behaviour of Mo and Zn from MSWI residues, in particular MSWI, bottom ash and MSWI fly ash [1,11,28]. Leaching
experiments carried out at L/S=10 1/kg. The smooth lines reflect the generic leaching behaviour of bottom ash (two data clusters) and fly ash.
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 757
The relation between L/S (in l/kg) and time is given (dotted line in the middle figure in Fig. 4) in the cumu-
by [9]: lative release versus L/S curve [1]. Residence times
(shortened to RT in the graphs) have been calculated as,
L=S ź I t=ð hÞ
respectively, 2.7, 15.6 and 40.5 h for the fast, normal
and slow flow rates in the percolation test. The perco-
with I is the net infiltration of precipitation in mm/year, lation rate is apparently not a very critical factor as the
t the time in years, the dry bulk density of the material percolation rates differ significantly between the three
in kg/m3 and h the height of the application in m. For experiments, whereas the differences between element
open applications with a height of 0.5 m and an infil- concentrations are not as large [10].
tration rate of 200 mm per year, an L/S of about 10 l/kg The pH may also be a factor, which should not be
is reached in about 100 years, whereas in a situation forgotten in this type of comparison. The pH ranges in
with top cover for a height of 5 m and a substantially the fast, normal and slow column are respectively
reduced infiltration, an L/S of 1 l/kg may only be 10.43 10.94, 9.88 10.88 and 8.75 10.05. Clearly, the pH
reached in more than 1000 years. tends to become lower as the percolation time increases.
In the case of V, solubility controls release, which is Whether this is caused by reactions in the material
reflected by the horizontal line in the plot of concentra- matrix itself or is caused by external influences (atmo-
tion in percolate versus L/S and the slope of about 1 sphere) is at present not clear.
Fig. 4. Leaching of V and Mo from MSWI bottom ash in a pH dependence test (left) and in a percolation test (middle: cumulative release and right:
eluate concentrations). Dotted line in middle graph points at solubility control [28].
758 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
The contact time has been varied in the pH depen- spheric CO2. In Fig. 7, a typical ANC curve is
dence test. In Fig. 5 the data are given for V and Mo, illustrated. As can be concluded from this graph, it
that indicates under specific pH conditions (Mo at pH 4 takes about 0.3 mol/kg of acid equivalent to reach pH
and 6) reactions are continuing after 48 h, which is now around 8. This can come from the uptake of CO2 from
identified as the standard condition for testing. The the atmosphere as well as from degradation of organic
bottom ash is still reactive at pH 4 and 6 as indicated by matter generating CO2.
the acid addition needed to maintain the respective pH Based on this information and conditions specific to
levels constant (Fig. 6). It is therefore not surprising to the application which can be site-specific, the relevant
see that Mo is still reacting. For Mo and V, it is likely pH boundaries for the application can be identified. In
that slow sorption reactions (iron oxide phases) are the case of MSWI bottom ash, the lower boundary is set
responsible for the observed decrease with time, because by the calcite buffering around 8. In Fig. 8 the pH
otherwise, one would expect an increase in Mo leach- domain relevant for application of granular MSWI
ability as a result of more acid consumption. bottom ash in road base or embankment is given. This
The more critical pH values (4 and 6), in terms of can be set against hypothetical regulatory criteria at a
reaching stable conditions, are relevant from a leaching specified L/S ratio as indicated with the dotted lines. For
behaviour and modelling point of view. These condi- Cr, no problems are expected to occur as the actual
tions are less important for the evaluation of environ- leaching level is well below the critical value. In case of Zn,
mental impact as due to carbonation (calcite buffer the leachability may become critical, when the pH drops
formation) in MSWI bottom ash, it is not likely that to below 7.5, which is, however, unlikely under natural
these pH values will be reached under field exposure conditions. This type of evaluation is helpful in deciding
conditions. which elements need to be addressed in quality control
and for which elements improvement or additional con-
trols may be needed. It helps to define the relevant range
5. Impact evaluation from leaching data of pH as a result of external stresses and internal chan-
ges (mineralization, organic matter degradation).
In the framework of the network harmonization of To create durable material improvements in ash
leaching/extraction tests interrelations between different quality it is essential to base measures on understanding
test methods have been addressed [11,12]. Through the of the controlling mechanisms by using characterization
acid neutralization capacity (ANC) and base neu- data for evaluation of treatment. Therefore, it is impor-
tralization capacity (BNC) information generated in the tant to relate changes to the input, changes to the con-
pH dependence test, the external stresses on a material version process and changes after treatment to the basic
in a given environmental setting can be addressed. characterization as described before. Firstly, to avoid
These include acidification resulting from, for instance, solving one problem and creating a new one, and sec-
degradation of organic matter, sulphide oxidation, buf- ondly, to ensure that a reduced leaching level is durable
fering capacity of natural waters, acid rain and atmo- and not counteracted by exposure to the environment.
Fig. 5. Leaching of V and Mo as function of contact time in the pH dependence test (pH static mode L/S =10 1/kg) [28].
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 759
6. Field studies ing of MSWI bottom ash leads primarily to a significant
pH change with its inherent changes in leachability.
Several field verification studies have been carried out Field verification studies [13] have revealed that
[13]. Meima [14], in particular, has provided information leachable Cu is fractionated between a labile dissolved
on cores taken from a 20-year-old MSWI bottom ash organic matter (DOC; likely low molecular weight
deposit. Results from field work (8 year old road base of organic acids) and a more stable DOC complex (more
MSWI bottom ash [13] and 20-year-old landfill [14]) humic matter like) in MSWI bottom ash, as roughly
indicate that MSWI bottom ash is neutralized under 40% of the leachable Cu is retained in the soil directly
unsaturated conditions. This observation leads to the underneath an application of MSWI bottom ash as a
question of which testing conditions are most relevant road base stabilization layer. The labile Cu complex is
for MSWI bottom ash: the high pH resulting from either degraded or destroyed though exchange with soil
crushing fresh ash or the longer term stable condition of organic matter. About 60% has been transported to the
carbonated and neutralized material (which can be groundwater as a water soluble organic complex. This
mimicked by pH=8 control in testing). Upon weathering type of Cu fractionation in MSWI bottom ash has also
new phases may be formed [15]. Such effects are included been identified in laboratory studies [16]. Johnson et al.
in the field weathered samples that were tested. Thus age- [17], Baranger et al. [18] have also modelled MSWI
bottom ash leaching, thus providing an increasingly
better understanding of controlling factors under field
conditions.
7. Quality control
Once sufficiently well defined data are obtained
through characterization, reduction of the testing effort
can be achieved by limiting the number of leaching steps
that need to be measured and by limiting the effort to
constituents that are really relevant for the material
being evaluated. Fig. 9 shows quality control data rela-
tive to characterization data providing clues for action.
For example, reductions in Mo may be possible through
control of the waste sources and Cu leaching may be
controlled through reduction in residual organic matter
content. In contrast, the behaviour of Pb and Zn leach-
Fig. 6. Acid consumption in pH dependence test on MSWI bottom
ing from different facilities appears to be controlled only
ash illustrating continuation of dissolution reations at pH 6 and 4. A
and B denote acid or base addition [28]. by the material s own pH.
Fig. 7. Acid Neutralization Capacity of MSWI bottom ash obtained from acid/base consumption in the pH dependence test [28].
760 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
This type of approach provides a basis for QA/QC scale. The process is more costly than other solutions
testing and eventually certification, when proven to be and also requires further treatment of salt concentrates
sufficiently stable. By presenting the QA/QC data in (salts cannot be vitrified). In many cases impact assess-
perspective to previous characterization, the power of ments carried out in accordance with the environmental
decision based on a single data point increases drasti- assessment principle specified in the Methodology
cally. This aspect is grossly overlooked in current prac- Guideline ENV 12920 [7] will show that MSWI residues
tice. In several cases, pH, a crucial parameter is not even cannot be utilised or even landfilled without prior
measured, which makes such regulatory leaching data treatment. Various treatment options exists, several of
useless. New developments in this area provide a sound which are already part of the common MSWI manage-
basis for quality control [19,20], in which no more test- ment practice. Nearly all the available treatment meth-
ing is done than needed. However, as soon as results ods are based on one or more of three treatment
appear to be out of specification, the level of testing principles: physical or chemical separation, stabiliza-
needs to be increased. tion/solidification and thermal treatment. These treat-
ment principles and a number of the corresponding
7.1. Treatment of MSWI residues treatment processes or unit operations are listed in
Table 1, which also provides an evaluation of the
MSWI bottom ash does not always comply with reg- potential applicability of the various methods. Before
ulatory criteria for utilization. This implies that treat- selecting a treatment process for a given residue and a
ment prior to utilization may be required. A commonly given scenario, it is important to set target properties,
applied method to stabilize material properties is a which are in harmony with the short- and long-term
minimum sample storage period of several weeks up to conditions expected for the scenario. When evaluating a
a few months to age the material. However, this treat- given treatment option it is also important to take all
ment may not be sufficient and additional treatment to new waste streams created by the process into account.
remove critical components, e.g. through washing of In most cases, a particular treatment process will consist
salts, Cu and Mo may prove necessary. of a combination of several of the unit operations
The main issue in relation to MSWI fly ash and Air shown in Table 1. For instance, it will be very difficult to
Pollution Control (APC) residues is the content of stabilize acid gas cleaning residues without prior or
soluble salts [1,27]. Several options have been evaluated simultaneous removal of the high content of readily
to treat these residues. Controlled containment of the soluble salts, e.g. by aqueous extraction (and sub-
stabilized salt containing residues is an option for treat- sequent treatment and/or discharge of the extract).
ment. Removal of the soluble salts makes the remaining In addition to the treatment of residues from tradi-
inorganic residue much more manageable. However, it tional mass burn incinerators, new incineration tech-
is unlikely to lead to a marketable product. Vitrification nologies have been developed in which e.g. thermal
for fly ash has been tested and is practised on a small treatment of the residues are integrated [21]. To judge
Fig. 8. Leaching behaviour of Cr and Zn from MSWI bottom ash as a function of pH as a means of evaluating long-term environmental impact and
the role of external stresses on release [12]. Dotted line represents a hypothetical limit value based on BMD Cat I limit values in the Dutch Building
Materials Decree [2]. The box reflects the relevent pH domain for a given application. All data obtained for L/S=10 1/kg.
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 761
Fig. 9. Compliance leaching test data for MSWI bottom ash from different Dutch MSW incinerators [19] in relation to characterization data (pH
stat-solid line) for MSWI bottom ash as QC for application in road constructions. Extraction conditions were LS=20, 24 h, leachant demineralized
water. Each symbol fill pattern represents a different facility. BMD Cat I and BMD CatII [2] are limit values from the Building Materials Decree
converted from mg/kg to microgram per litre at L/S=10.
762 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
the performance of treatment options, a compliance test input as well as the chemical composition of the differ-
(single extraction method) does not generally suffice. A ent streams [25]. An inherent limitation of the study was
characterization of the changes in leaching behaviour that the sorting of the waste was carried on integral
caused by the treatment can be visualized by using the collected waste. This resulted in a situation where the
pH dependence test, as it provides a direct measure for composition of the wet organic fraction was sig-
such changes in leachability. nificantly biased by other non-organic sub-streams in
integral waste, such as vacuum cleaner dust and other
fine particulate contributions not originally belonging to
8. Waste input measures and mass balance the organic fraction. The organic fraction analysed by
the sorting of integrally collected waste is therefore sig-
An important question posed by government and nificantly different from the organic fraction for sepa-
municipal solid waste incinerator (MSWI) operators is rately collected organic waste. By comparing the
how the quality of MSWI residues is influenced by the composition of the organic fraction as obtained from
composition of the waste input to the incinerator. Also sorting integral waste with composition data derived
operating conditions, which are not discussed here, have from analysis of pure organic matter the level of con-
been identified as important controlling factors [22,23]. tamination can be clearly identified. This is particularly
A question linked to the input characteristics is to what relevant for the elements Cr, Pb, Cu, Cd.
extent specific (often smaller) waste streams are respon- At present, the leaching behaviour of MSWI residues,
sible for the final quality of the residues. This forms in particular, MSWI bottom ash, forms a limitation for
another option to improve the quality of the residues. the beneficial use of such residues in construction
Besides the chemical composition of the residues, a applications. Based on knowledge gained on how the
main aspect for the evaluation of MSWI residue quality quality of the residues is affected by specific materials or
is the leaching behaviour. This has been shown to be constituents in the input stream, the input to the incin-
unrelated to the bulk chemical composition for most erator could be managed in a manner in which such
elements of interest. Several studies have been carried inputs are minimized or eliminated. Such contaminating
out to address this question [1,19,23 25]. Characteriza- waste streams may form a minor fraction of the total
tion of the waste input to the incinerator into separate throughput, but greatly affect the overall quality.
sub-streams for their chemical composition is an essen- Before such measures will lead to measurable effects it
tial element in answering such a question. Back in 1973, is important to relate the leaching behaviour of MSWI
the National Public Health Institute (RIVM) conducted residues to the waste input quality. Based on work car-
a separation of curb collected domestic waste in 11 dif- ried out in the framework of the IAWG and other stu-
ferent sub-streams to be able to quantify both the mag- dies [1,8,11], the lack of correlation between
nitude of the different contributions to the total waste composition and leaching has been unambiguously
Table 1
Overview of principles and methods for treatment of MSWI residues (after [20])a
Treatment principle Examples of processes Bottom Fly Acid gas cleaning residues
and unit operations ash ash (with or without fly ash)
Separation Washing and extraction, e.g. at various pH values a a/b a/b
Chemical precipitation a/b a/b
Crystallization/evaporation b
Ion exchange b
Density and particle size based separation a b b
Distillation b b
Elektrolysis c
Elektrokinetic separation c
Magnetic separation a
Eddy current separation a
Stabilization and/or solidification Addition of hydraulic binders a a/c a/c
Addition of pore-filling additives a/b a
Chemical stabilization a a b
Thermal treatment Sintering a a/c c
Melting/vitrification a/c a/c c
a
a, Part of existing and proven treatment technology; b, have shown promising results, may be expected to be included in future treatment sys-
tems; c, currently under investigation or have been investigated and not found technically and/or economically feasible.
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 763
established. The elements Na, K, Mo, Br and Cl form stream [25]. Based on these data, the ash composition
an exception, as these do show a direct relation between of, respectively, bottom ash and fly ash is quantified by
input and leachability due to absence of matrix interac- assigning the ash resulting from a specific sub-stream in
tion, and thus are leached within L/S=2 l/kg. A control the waste input entirely or partially to either bottom ash
measure at the input for these elements will lead to a or fly ash. The rationale behind this concept is that
direct and proportional improvement of quality in terms during combustion, materials with a very low ash con-
of leaching. However, these elements are also readily tent release elements almost completely to the gaseous
removed by washing after the residue is produced. In phase which will be carried over to the fly ash. This also
Fig. 10 the positive correlation between total composi- applies to elements considered to be non-volatile. Ele-
tion and leachability is illustrated for Mo. ments from sub-streams with a high ash content are lar-
For elements (e.g. metals) controlled by solubility, gely or completely assigned to bottom ash. A few
measures at the input side are meaningless. There may materials will be distributed between the two. The result-
be other reasons to separate metals from the waste stream, ing predicted compositions can be verified against the
but not from an environmental impact evaluation. Only average Dutch bottom ash and fly ash composition [19].
when leachability is controlled by availability, can mea- The advantage of such a model is that a decision can be
sures at the input side of the installation be useful. taken on the acceptability of new or increased inputs of
To be able to quantify the contribution of different certain waste streams beforehand. In addition, it points
waste streams to the residue composition, a spreadsheet at the gaps in the present information on sub-streams.
has been developed, which uses the composition of the In Table 2, a first comparison of predicted bottom ash
separate sub-streams identified in the input to the and fly ash composition with the average bottom ash and
incinerator, their moisture content, their loss on ignition fly ash composition in the Netherlands is made. With
and their fractional contribution to the total waste this type of comparison a main drawback is the limited
Table 2
Comparison of calculated composition for bottom ash (BA) and fly ash (FA) with average composition of Dutch MSWI bottom ash and fly ash (all
data in mg/kg)
Cd Cu Mo Pb Sb Zn As Cr Cra
Bottom ash
Calculated BA from input 7.5 1278 6.2 252 14 545 5.2 178 160
Average BA Netherlands 4.0 2200 11.0 1270 34 1976 6.6 117
Fly ash
Calculated FA from input 208 2505 82 2948 504 4451 36 1174 479
Average FA Netherlands 345 1175 69 8975 736 28 245 60 192 600
a
Cr based on adjusted values for the organic fraction.
Fig. 10. Relation between total Mo in Bottom ash and leached Mo from ash at L/S=10 1/kg (pH range 10.5 12; n=64 [19].
764 H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765
number of elements for which composition data are bottom ash leads primarily to a significant pH change
available. This applies in particular for major elements. with its inherent changes in leachability. The more cri-
The Cr content reported for MSWI fly ash is low, tical pH values (4 and 6) in terms of reaching stable
which may be related to the type of sample destruction conditions are relevant from a leaching behaviour and
(aqua regia). Data from the   Mammoet  project [26] modelling point of view. These conditions are less impor-
point at higher average Cr contents in fly ash (complete tant for the evaluation of environmental impact of MSWI
matrix dissolution with HF). In view of the raw nature of bottom ash. Due to carbonation (calcite buffer forma-
the data, the agreement between the prediction and the tion) in MSWI bottom ash, it is not likely that these pH
measured data is quite good for Cd, Mo, Cu, Sb and As. values will be reached under field exposure conditions.
The Zn and Pb levels in bottom and fly ash are clearly The percolation rate within the range now used does
underestimated, which may be attributed to an insuffi- not appaer to be very sensitive. The type of release
ciently described composition of electronic parts in the mechanism (solubility, wash out, other) can be identi-
feed stock. The assessment is particularly interesting to fied readily. Defining the relevant range of pH for a
assess the qualities of the fly ash and APC residues as a scenario based on an evaluation of material behaviour
result of variations in low ash materials in the feedstock. (mineralization, organic matter degradation) and exter-
nal stresses is helpful in deciding which elements need to
be addressed in quality control and for which elements
9. Conclusions improvement or additional controls may be needed.
To create durable material improvements in ash
9.1. General quality it is essential to base treatment on understanding
of the controlling mechanisms by using characterization
To guide measures in acceptance of waste, in the data for evaluation of treatment. Therefore, it is impor-
conversion process and in treatment of residues, a more tant to relate changes to the input to the process, chan-
thorough understanding of factors controlling the ges to the conversion process and changes after
leachability of MSWI residues is needed. Waste titration treatment to the basic characterization using at least
to beat a single extraction test is not an acceptable time and pH dependent leaching data. Firstly, to avoid
approach, as it will generally lead to undesirable envir- solving one problem and creating a new one, and sec-
onmental impact. Instead, the emphasis should be on an ondly, to ensure that a reduced leaching level is durable
assessment of the long-term behaviour of the materials and not counteracted by exposure to external influences.
in their service life and ultimate fate (  end of life  ). The By presenting quality control data in perspective to
tools and associated modelling to make such assess- previous characterization data the power of decision
ments have grown substantially in recent years and are based on a single data point increases drastically. The
now subject of standardization. Leaching behaviour of potential of this mode of data interpretation is grossly
MSWI residues has been shown to be very systematic overlooked in current practice. For elements (e.g. metals)
and has been modelled with a great deal of success in controlled by solubility, measures at the input side to
identifying solubility controlling phases. The under- reduce environmental impact are meaningless. Only when
standing of controlling factors helps greatly in focusing leachability of elements is correlated with total compo-
the quality improvement activities and in defining the sition or availability, can measures at the input side of
necessary quality control for certification purposes. Field the installation be useful. Based on the assumption of
verification assists in keeping a balance between labora- complete transfer of inorganic elements from low ash
tory practice and realistic field conditions. Release content materials to the flue gas predictions of MSWI
modelling has already shown potential for prediction pur- bottom ash composition and fly ash composition have
poses. As the understanding grows, such assessments been made. The agreement between the prediction and
may have consequences for the regulatory framework the measured composition data is quite good for Cu,
and level setting. In addition, the process of evaluation Cd, Mo, Sb and As given the level of inherent variability
of environmental properties of materials for recycling in MSWI bottom ash composition. The Zn and Pb levels
and treatment for utilization is not unique to MSWI in bottom and fly ash are clearly underestimated.
residues, but also applies to other industrial residue
streams.
9.2. Specific References
[1] IAWG (International Ash Working Group; A.J. Chandler, T.T.
The pH dependence leaching test provides a very
Eighmy, J. Hartlen, O. Hjelmar, D.S. Kosson, S.E. Sawell, H.A.
good means of mutual comparison of MSWI residues as
van der Sloot, J. Vehlow). 1997. Municipal Solid Waste Incin-
well as a comparison of leaching behaviour within one
erator Residues. Studies in Environmental Science 67, Elsevier
and the same type of MSWI residue. Ageing of MSWI Science, Amsterdam.
H.A. van der Sloot et al. / Waste Management 21 (2001) 753 765 765
[2] Building Materials Decree. Staatsblad van het Koninkrijk der viour and solubility controlling phases of heavy metals in MSWI
Nederlanden, 1995, 567. ash. Waste Management 1996;16:129 34.
´
[3] Ministere de l Environment. Circulaire Relative a la Valorisation [18] Baranger Ph, Azaroual M, Lanini S, Piantone P, Freyssinet Ph.
de Machefers d Incineration de Residues Urbans en Techniques Modelling the weathering of a bottom ash heap. Waste Stabili-
Ć
Routieres, DPPR/SEI/BPSEID/FC no 94-IV-1, Paris, 1994. zation & Environment Conference, Eds. J. Mehu, G. Keck and
[4] Van der Hoek, E., van der Sloot, H.A., Korte testmethoden voor A. Navarro, Lyon, 13 16 April 1999, 79 83.
de beoordeling van de uitloging uit bouwmaterialen en afval- [19] Born JP. Quantities and qualities of MSWI residues in the Neth-
stoffen. KEMA 1998 (in Dutch). erlands. In: Goumans JJJM, van der Sloot HA, Aalbers ThG,
[5] CEN TC 292 WG6. Characterization of waste. Leaching beha- editors. Environmental aspects of construction with waste mate-
viour tests. Influence of pH under steady state conditions (in rials. Amsterdam: Elsevier Science, 1994. p. 633 44.
preparation) 1998. [20] Kosson DS, van der Sloot HA. Integration of testing protocols
[6] CEN TC 292 WG6. Characterization of waste. Leaching behaviour for evaluation of contaminant release from monolithic and gran-
tests. Percolation simulation leaching test (in preparation) 1998. ular wastes. In: Goumans JJJM, Senden GJ, van der Sloot HA,
[7] Methodology Document PrENV 12920, CEN TC 292 WG6, 1996. editors. Waste materials in construction  putting theory into
[8] Van der Sloot HA, Developments in evaluating environmental practice. Studies in Environmental Science 71. Amsterdam:
impact from utilization of bulk inert wastes using laboratory Elsevier Science Publishers, 1997. p. 201 16.
leaching tests field verification. Waste Management 1996;16(1 [21] Vehlow J. Thermische Behandlungsverfahren fur Hausmull im
¨ ¨
3):65 81. Vergleich. Beitrag zur Schulungsreihe: Restabfallgehandlung in
[9] Hjelmar O. Leachate from land disposal of coal fly ash. Waste der Steiermark. 3 April 1998, Graz. Forschungszentrum Karls-
Management & Research 1990;8:429 49. ruhe, Institut fur Technische Chemie, Bereich Thermische
¨
[10] Dijkstra JJ, van der Sloot HA, Comans RNJ. Process identifica- Abfallbehandlung, Karlsruhe, 1998.
tion and model development of contaminant transport in MSWI [22] Vehlow, J. Mitverbrennen von Rest-Abfallen aus dem Siedlungs-
¨
Bottom ash. Waste Management  special issue WASCON 2000 und Sonderabfallbereich  eine Frage der Zulassung oder der
(in preparation). Verfahrenstechnik. Seminar 02 im Rahmen der UTECH BER-
[11] Harmonization of leaching/extraction tests, 1997. Studies in LIN  96. Fortbildungszentrum Gesundheits- und Umweltschutz
Environmental Science, Volume 70. Eds H.A. van der Sloot, L. Berlin e.V. (FGU Berlin), 26 Februar 1996, pp. 199 210.
Heasman, Ph Quevauviller, Elsevier Science, Amsterdam. [23] Sawell SE, Chandler AJ, Rigo HG, Hetherington SA, Fraser J.
[12] Technical work in support of the Network Harmonization of The waste analysis, sampling, testing and evaluation program:
Leaching/Exctraction Tests. EU project SMT4-CT96-2066, 2000. effect of lead and cadmium spiking of MSW on the characteristics
[13] Schreurs JPGM, van der Sloot HA, Hendriks Ch F. Verification of MSWI residues. Proceedings Municipal Waste Combustion.
of laboratory-field leaching behaviour of coal fly ash and MSWI Air & Waste Management Association Pittsburg, Pennsylvania
bottom ash as a roadbase material . Proceedings WASCON 1997 1993;VIP 32:288 302.
Conference   Putting Theory Into Practice  , 4 6 June, 1997 [24] Steketee J. Onderzoek naar de relatie tussen samenstelling van
Houthem, The Netherlands. afvalcomponenten en de uitloging van AVI residuen. Tauw
[14] Meima J. Phd Thesis: Leaching properties of MSWI bottom ash. Milieu. R3586685.D03/jjs. 1998.
RU Utrecht (1997) Chapter 7. [25] Van de Beek AIM, Cornelissen AAJ, Aalbers ThG. Fysisch en
[15] Zevenbergen C, Bradley JP, Wood T, Brown RS, van Reeuwijk Chemisch Onderzoek aan Huishoudelijk Agval. Publ. RIVM,
LP, Schuiling RD. Weathering as a process to control the release Rap. 738505005, 1988.
of toxic constituents from MSW bottom ash. In: Geology and [26] Versluijs CW, Anthonissen IH, Valentijn EA. Mammoet  85: Inte-
Confinement of Toxic Waste, Proceedings of the Int. Symp. grale evaluatie van deelonderzoeken. RIVM report 738504008.
Geoconfine  93, Montpellier, France, 591 595, 1993. Bilthoven 1990.
[16] Meima JA, van Zomeren A, Comans RNJ. Complexation of Cu [27] Hjelmar O. Incineration: residues. In: Christensen TH, editor.
with dissolved organic carbon in municipal solid waste incin- Waste technology. Copenhagen: Teknisk Forlag, 1998. p. 221 61.
erator bottom ash leachates. ES&T 1999;33:1424 9. [28] CEN TC 292  Characterization of waste  Working Group 6-
[17] Johnson CA, Kersten M, Ziegler F, Moor HC. Leaching beha- Working documents. 1998.