Full text of DOI:10.1080/10473289.2002.10470814

This article was downloaded by: [Universitat Politècnica de València]
On: 24 November 2014, At: 01:42
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of the Air & Waste Management Association
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uawm20
Stabilization of APC Residues from Waste Incineration
with Ferrous Sulfate on a Semi-Industrial Scale
Kasper Lundtorp a , Dorthe L. Jensen b & Thomas H. Christensen b
a Babcock & Wilcox Vølund Aps , Esbjerg , Denmark , USA
b Environment & Resources DTU , Technical University of Denmark , Lyngby , Denmark , USA
Published online: 27 Dec 2011.
To cite this article: Kasper Lundtorp , Dorthe L. Jensen & Thomas H. Christensen (2002) Stabilization of APC Residues from
Waste Incineration with Ferrous Sulfate on a Semi-Industrial Scale, Journal of the Air & Waste Management Association, 52:6,
722-731, DOI: 10.1080/10473289.2002.10470814
To link to this article: http://dx.doi.org/10.1080/10473289.2002.10470814
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Lundtorp, Jensen, and Christensen
TECHNICAL PAPER
ISSN 1047-3289 J. Air & Waste Manage. Assoc. 52:722-731
Copyright 2002 Air & Waste Management Association
Stabilization of APC Residues from Waste Incineration with
Ferrous Sulfate on a Semi-Industrial Scale
Kasper Lundtorp
Babcock & WilcoxVølund Aps, Esbjerg, Denmark
Dorthe L. Jensen and Thomas H. Christensen
Environment & Resources DTU, Technical University of Denmark, Lyngby, Denmark
ABSTRACT
flue gas cleaning systems at modern incinerators gener-
ate 10–50 kg of air pollution control (APC) residues per
ton of waste incinerated.3 Three gas cleaning systems are
commonly used: dry systems, semidry (SD) systems, and
wet scrubber systems. The first two systems inject lime or
other alkaline reactants into the gas to remove acid gases
and metals. The reaction products, excessive lime and fly
ash (FA), are removed as APC residues in bag filters or
electrostatic precipitators. The wet system usually removes
the FA before it reaches the wet scrubber. The wastewater
from the scrubber is then cleaned by traditional water
treatment methods, creating a sludge, which is disposed
of either separately or mixed with the FA.
A stabilization method for air pollution control (APC) resi-
dues from municipal solid waste incineration (MSWI) in-
volving mixing of the residue with water and FeSO4 has
been demonstrated on a semi-industrial scale on three
types of APC residues: a semidry (SD) APC residue, a fly
ash (FA), and an FA mixed with sludge (FAS) from a wet
flue gas cleaning system. The process was performed in
batches of 165–175 kg residue. It generates a wastewater
that is highly saline but has a low content of heavy met-
als such as Cd, Cr, and Pb. The stabilized and raw residues
have been subject to a range of leaching tests: the batch
leaching test, the pH-static leaching test, the availability
test, and the column test. These tests showed that the
stabilized residues have remarkably improved leaching
properties, especially with respect to Pb but also with re-
spect to Cd, Cu, and Zn. The release of Pb was reduced by
a factor of 250–36,000.
APC residues pose a threat to the surrounding envi-
ronment because of the high content of pollutants, espe-
cially trace metals such as Pb, Cd, and Zn, which easily
can be mobilized by infiltrating water. In particular, Pb
will be leached out because of the high pH in the leachate
from the alkaline residues.3,4 Furthermore, large amounts
of salts (i.e., Cl, K, and Na) and hydroxide will be leached
from the alkaline APC residues. Because of the high con-
centrations, this may also pose a threat to the environ-
ment. In addition, some of the anions may, at high
concentrations, enhance the mobility of the trace metals
because of complexation.
INTRODUCTION
In many parts of the world, incineration is a common
method of municipal solid waste disposal. In Europe, the
use of municipal solid waste incineration (MSWI) is ex-
pected to increase over the next 20 years, as landfilling
will become more and more restricted.1,2 The advanced
Several treatment methods have been developed to
reduce the leaching of trace metals from APC residues.
These include extraction, solidification, vitrification, and
chemical stabilization with phosphoric acid or CO2. A few
of these methods (e.g., cement stabilization, chemical sta-
bilization with CO2, and vitrification) have been used in
full scale,5,6 but they are not common because of limited
effect of the treatment, high costs, and the fact that the
problem is relatively new. Consequently, most of the APC
residues are landfilled either directly or in large bags. A
new directive bans the landfilling of hazardous waste with-
in the European Union without pretreatment.7 A suitable
pretreatment is, therefore, required to implement the new
IMPLICATIONS
The paper describes a stabilization method for reducing
the release of contaminants from FA and other APC resi-
dues from MSWI. The described stabilization method en-
ables a safe landfilling of the said residues with benefits
for the surrounding environment, caused by lower emis-
sions, and with a reduction in costs, because leachate
collection and treatment potentially can be avoided. The
state of the art for the stabilization method is described
and semi-industrial-scale use of the stabilization process
is demonstrated to assess the potential for full-scale
implementation.
722 Journal of the Air & Waste Management Association
Volume 52 June 2002

Lundtorp, Jensen, and Christensen
directive; however, technical objectives of the treatment
currently are not available.
A new method for chemical stabilization of APC resi-
dues, called Ferrox stabilization, has been developed and
documented in detail at the laboratory scale.8–10 The pro-
cess is based on the established facts that many trace
metals in natural environments, such as those in soils and
sediments, are bound to iron oxides11 and that engineered
iron oxides are able to bind substantial amounts of trace
metals.12 In the Ferrox process, the alkaline APC residue
is mixed with an FeSO4 solution. A precipitate of Fe(OH)2
is immediately formed on the residue. Subsequently, the
suspension is aerated, thereby oxidizing Fe(OH)2 to Fe(III)
oxides in the form of ferrihydrite.13 The trace metals are
bound to the Fe(III) oxides during the process, resulting
in low concentrations of trace metals in the wastewater.
Easily soluble elements (e.g., Cl, Na, and K) dissolve dur-
ing the process and are removed from the residue with
the wastewater. Laboratory investigations revealed that
the treated APC residues had significantly improved leach-
ing characteristics. For example, the leaching of Pb was
reduced by 3 orders of magnitude.
Figure 1. The Ferrox pilot-scale plant. The three major units are the
mixing tank, where FeSO4 solution, recycled water, and APC residue
are mixed; the process tank, where the oxidation of ferrous iron takes
place followed by a pH adjustment with either additional FeSO4 or
H2SO4; and the plate-and-frame filterpress, where the treated residue
is dewatered and washed. The process tank has a volume of ~1.5 m3.
separated from the treated residue and the residue was
washed with tap water. The mixing tank had a volume of
~1000 L and was equipped with a stirring aggregate (200
rpm). It was placed on three load cells for weighing of the
amount of saturated FeSO4 solution, water, recycled wash-
ing water, and residues that were added to the tank. Probes
for measuring density of the solution, pH, and specific
conductivity were mounted on the mixing tank. The satu-
rated FeSO4 solution was obtained from a tank contain-
ing water and FeSO4 in excess (solid FeSO4 was present on
the bottom of the tank).
To evaluate the Ferrox process at an industrial scale,
pilot plant studies were made in a 1:20 scale of a plant.
The pilot plant studies build on the results from the labo-
ratory-scale studies8–10 but provide much better insight
into scale-dependent issues such as process control, aera-
tion of the suspension, filtration, and washing of the
treated residues and provide much more reliable data
on wastewater composition, leaching characteristics of
treated APC residues, and mass balances. The primary
objective of the Ferrox process is to treat the APC resi-
dues so they can be landfilled in a safe way with a mini-
mum of emissions to the environment. This is related
both to the concentrations of metals in the leachate and
to the long time period during which leachate must be
collected and treated. A secondary objective may be that
the treated APC residue could be reused in construction
work depending on local legislation. This paper reports
on pilot-scale studies on Ferrox stabilization of three dif-
ferent APC residues: a residue from an SD system, an FA
from an electrostatic precipitator, and an FA mixed with
sludge (FAS) from a scrubber system.
The process tank had a volume of ~1500 L and was
mounted with a turbine (700 rpm) and an air inlet at the
bottom for dispersion of the air into the suspension. Probes
for pH, specific conductivity, and temperature were
mounted on the tank. The process tank had inlets for
addition of FeSO4 and H2SO4.
The filter press was a manually handled plate-and-frame
filter press with a filter volume of 169 L and a filter area of
12.5 m2. The wastewater outlet from the filter press was
equipped with pH and specific conductivity probes. The fil-
ter press had facilities for flushing the material with water.
In addition to the previously described components, there
were additional tanks for storing liquids, handling spills, and
storing dry residue. All data from pH, specific conductivity,
and temperature probes and from the load cells were logged
by a computer monitoring the process performance.
EXPERIMENTAL WORK
Pilot Plant
The pilot plant had a capacity of up to 200 kg APC resi-
dues per batch. The layout of the plant is shown in Figure 1.
The basic components were (1) the mixing tank where
FeSO4 solution, water, and residue were mixed; (2) the
process tank where the ferrous iron precipitate was oxi-
dized and the pH was adjusted; and finally (3) the plate-
and-frame filter press in which the wastewater was
Process Scheme
A saturated FeSO4 solution was pumped to the mixing
tank as a first step in the process. Accurate dosing of the
FeSO4 solution was obtained by online measurement of
the density and the weight of the solution. Water recycled
from the washing of the treated residue in the filter press
and untreated APC residue as a dry powder were added to
Volume 52 June 2002
Journal of the Air & Waste Management Association 723

Lundtorp, Jensen, and Christensen
the mixing tank. The initial pH of the FeSO4 solution was
acidic (pH 2–3), but it increased quickly after addition of
the alkaline APC residue. At a pH greater than 10, all Fe(II)
was precipitated as Fe(OH)2, and the suspension was trans-
ferred to the process tank, where the aeration took place.
The aeration (dispersion of air into the suspension
under heavy stirring) period was 20–50 min, depending
on the type of residue treated. Investigation of the oxida-
tion of Fe(II) during the aeration had shown that during
the first 20–50 min, 70% of the Fe(II) was oxidized, de-
pending on the type of residue treated. With additional
aeration for 24 hr, 83% of the Fe(II) was oxidized (data
not shown), which indicates that part of the Fe(II) either
is bound in Fe(II)-Fe(III) oxides and only slowly is trans-
formed or that some of the Fe(II) is not readily accessible
to oxygen. Because additional aeration time only resulted
in a minor increase in the amount of Fe(II) oxidized, the
aeration time was chosen to be 20–50 min, depending on
the type of residue treated.
The batch process was executed four times in a se-
quence for each type of residue to ensure that a pseudo-
steady state was reached with regard to recycling of the
water used for the washing process. The specific conduc-
tivity and Cl concentration of the wastewater discharged
revealed that a pseudo-steady state was reached. The spe-
cific conductivity and Cl concentration in the wastewater
from the first batch were lower than those from the fol-
lowing three batches (data not shown). All mass flows in
and out of the plant were recorded and sampled. All online
measurements of loads, pH, conductivity, temperature, and
Fe(II) concentration were recorded by a computer.
Materials
Three types of APC residue were studied. The chemical
composition of the residues specifically used in the
experiments is shown in the following sections.
•
An SD APC residue consisting of reaction products,
unreacted lime, and FA collected in a fabric filter
at I/S Amagerforbraending, a large MSWI (capac-
ity: 48 t/hr) in Copenhagen, Denmark;
After the suspension had been aerated for 20–50 min,
pH adjustment was initiated. Because the pH prior to the
pH adjustment is 11–12.5, FeSO4 or H2SO4 was used to
lower the pH. Table 1 reports on the consumption of
chemicals for the three residues treated, including the
FeSO4 added to the initial mixing. The pH was adjusted
to 10.4 for SD and 11.0 and 10.8 for FA and FAS, respec-
tively. The temperature increased in the tank because of
the chemical reactions, and the pH values reported were
at operating temperature (36 ºC for SD, 28 ºC for FA, and
25 ºC for FAS). After adjustment, the pH was maintained
at the desired level for 30–60 min, depending on the resi-
due treated before transfer to the filter press.
The suspension was dewatered in the plate-and-
frame filter press at 6 bars pressure. The filtrate, consist-
ing of process water, was discharged from the plant as
wastewater. Afterward, the treated residue in the press
was washed continuously with water (3.2 L/kg for treated
SD, 2.5 L/kg for treated FA, and 2.9 L/kg for treated FAS).
The water from this washing process was collected for
reuse in the mixing tank for the following batch. The
solids remaining in the filter press, called the Ferrox prod-
uct, had a dry matter content of 54–57%. The Ferrox
product was removed from the filter press and sampled
for subsequent characterization.
•
•
An FA from an electrostatic precipitator at I/S
Vestforbraending, a large MSWI (capacity: 50 t/hr)
in Copenhagen; and
An FA from an electrostatic precipitator mixed
with sludge from a wet scrubber gas cleaning
system at I/S Vestforbraending, a large MSWI
(capacity: 50 t/hr) in Copenhagen.
Analytical Work
The composition of the Ferrox products and the corre-
sponding untreated residues were determined by a total
digestion, first with a mixture of HNO3, HCl, and HF in a
microwave system. This was supplemented by fusion at
1000 ºC with lithiummetaborate for the elements Ba, Ca,
Cr, Fe, K, Mo, Na, Si, Sn, and Sr. Prior to digestion or fu-
sion, the samples were dried at 50 ºC (determination of
As, Cd, Co, Cu, Hg, Ni, Pb, S, and Zn) or 105 ºC (determi-
nation of Al, Ba, Be, Ca, Cr, Fe, K, La, Mg, Mn, Mo, Na,
Nb, P, Sc, Si, Sn, Sr, Ti, V, W, Y, and Zr). The composition
was related to the dry matter content determined at 105
ºC. The content of Cl– was analyzed by XRF (X-ray fluo-
rescence). The remaining elements were analyzed by ICP-
AES (inductively coupled plasma-optical emission
spectroscopy), ICP-QMS (inductively coupled plasma-mass
spectroscopy, Quadrupol), ICP-SMS (inductively coupled
plasma-mass spectroscopy, Sector), and AFS (atomic fluo-
rescence spectroscopy). The analysis was performed by
SGAB Analytica AB, Luleaa, Sweden, and is accredited by
Swedac (accepted in a range of countries in Europe and in
the United States) for all elements except Cl and S.
Table 1. The use of water and chemicals in the Ferrox treatment.
SD
FA
FAS
Water (m3/t untreated residue)
FeSO4 (kg Fe/t untreated residue)
H2SO4 (kg/t untreated residue)
3.9
49
9
2.7
20
0a
3.2
13
0a
The discharged wastewater from the treatment and
the eluates from the leaching tests (see later sections) were
aH2SO4 is not used for treatment of FA and FAS.
724 Journal of the Air & Waste Management Association
Volume 52 June 2002

Lundtorp, Jensen, and Christensen
measured for several parameters including pH, conductiv-
ity, Cl, Ca, Na, K, Cd, Cr, Cu, Hg, Ni, Pb, and Zn. The ele-
ments were analyzed with ICP-AES, ICP-SMS, AFS, and
atomic absorption spectrophotometry (flame and graph-
ite furnace technique). The leaching properties of the Ferrox
products and the corresponding untreated residues were
evaluated by the following leaching tests:
water, within the limits of the plant (maximum flow
2000 m3/hr), had no significant effect on removal of
salts (data not shown).
Table 1 shows a consumption of water between 2.7
and 3.9 m3/t of untreated residue. Treating SD required
the most water for washing out its higher salt content.
The water consumption was, in general, fairly low because
of reuse of the washing water in the mixing tank. This
was possible because the recycled washing water had a
specific conductivity of 20–25 mS/cm for SD and ~15 mS/
cm for FA and FAS, as compared with 70–140 mS/cm in
the wastewater discharged from the filter press (see Table 2).
The washing water, therefore, was not saturated with re-
spect to major ions. If the washing water had not been
recycled, the water consumption would have been be-
tween 5.6 and 7.0 m3/t untreated residue. Thus, recycling
of the washing water saved up to 50% of the water needed
for the process.
•
Batch leaching test: A two-stage compliance batch
leaching test performed at liquid-to-solid ratios
(L/S ratios) of 0–2 L/kg for 6 hr and 2–10 L/kg for
18 hr. The tested material determined the pH.
The test was performed according to ref 14, which
is almost identical to ref 15.
•
Availability test: A two-stage batch leaching test.
The first step was at L/S 100 L/kg and at fixed pH
7 (addition of HNO3 controlled by computer) for
3 hr. The second step was at 100 L/kg and pH 4
(addition of HNO3 controlled by computer) for
18 hr. Demineralized water was used as leachant,
and 8 g dry matter was used for each test. The
test was performed according to ref 16.
The amount of FeSO4 added was based on laboratory
experiences reported in refs 8 and 9. The content of hy-
droxides in FA and FAS was sufficient for precipitating an
adequate amount of ferrous iron, and a fixed pH was
maintained during the process by further addition of FeSO4
(this is included in the figures given in Table 1). SD had a
•
•
The pH-static leaching test: A one-step batch leach-
ing test with L/S ratio of 10 L/kg for 24 hr at fixed
pH (addition of HNO3 controlled by computer).
The test is performed for a range of pH values from
4 to 11 for each material tested. Demineralized
water was used as leachant and 40 g dry matter
was used for each test.
Table 2. Quantity and composition of the wastewater discharged from the process.
Residue
Treated
Unit
SD
FA
FAS
Column test: An up-flow column test. The test
was performed according to ref 17 except that
instead of using demineralized water acidified to
pH 4, pure demineralized water was used as
leachant. Furthermore, the collection of eluate
fractions was more frequent than proposed in ref
17, and the test was continued until an L/S ratio
of 3.4–6.2 L/kg was reached. The column had a
diameter of 5.0 cm and was filled with 27 cm of
the test material. The flow was between 0.06 and
0.09 L/kg/day.
Quantity
pH
m3/t
2.8
10.4
2.3
10.9
2.4
10.8
Cond.
Ca
mS/cm
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
140
91
72
19,000
52,000
0.050
7500
1.8
2000
28,000
<0.05
12,000
<0.9
1300
21,000
0.085
10,000
<1.4
Cl
Fe
K
Mg
Na
7400
280
10,000
1100
8400
1600
S
Al
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
µg/L
590
7.0
6100
<5
36,000
12
RESULTS AND DISCUSSION
As
Ba
Cd
Co
Cr
Water, Chemicals, and Time Consumption
3000
6.4
1400
2.6
420
5.7
Treatment of one batch (165–175 kg) took ~3 hr, 20 min
for SD and 2 hr, 15 min for FA and FAS. The aeration time
and the pH holding time determined the time needed for
the treatment. These were longer for SD because of the
excessive lime and the larger amount of iron added. If
mixing of one batch was performed (mixing tank) simul-
taneously with aeration/pH adjustment of another batch
(process tank), the overall time needed for the treatment
could be reduced to 2 hr, 15 min for SD and 1 hr, 15 min
for FA and FAS. A study of the washing process in the
filter press showed that the flow velocity of the washing
<0.25
68
<0.25
90
<0.25
4.0
Cu
Hg
Mn
Mo
Ni
7.3
<5
11
0.32
1.6
1.0
0.25
3.3
1.8
720
<2.5
69
1600
<2.5
28
830
<2.5
47
Pb
Zn
120
24
45
Volume 52 June 2002
Journal of the Air & Waste Management Association 725

Lundtorp, Jensen, and Christensen
Table 3. The content of selected elements in the examined APC residues. All figures
high and fairly variable content of excessive Ca(OH)2 from
the flue gas cleaning process. Therefore, it was decided to
add a fixed amount of FeSO4 and maintain a constant pH
by adding H2SO4. The amount of H2SO4 added varied with
the amount of excessive Ca(OH)2 in the SD APC residue.
A fixed pH was not maintained using FeSO4, as was done
in the case of FA and FAS, because the significant amounts
of FeSO4 needed would become a cost factor. H2SO4 was
readily available and cheap.
are based on dry residue.
Element
Unit
SD
FA
FAS
Ca
g/kg
g/kg
253
135
11
205
59
199
42
Cl
Fe
g/kg
17
18
K
g/kg
30
39
34
Na
g/kg
32
36
31
S
g/kg
34
30
38
Wastewater
Si
g/kg
65
104
151
1130
279
19
97
The process generates wastewater consisting of discharged
filtrate from the filter press. The composition and quan-
tity of the wastewater is shown in Table 2. The quantity
did not equal the water consumption shown in Table 1
primarily because part of the water was bound to the sol-
ids and left the plant with the treated residue as a wet
filter cake.
Table 2 shows that the wastewater had very high sa-
linity, suggesting that the easily soluble salts were effec-
tively removed. The content of Ca and Cl was highest in
the wastewater generated from treating SD because of the
high content of CaCl2 in this type of residue. Table 2 also
shows that the concentrations of all trace metals except
Al, Ba, and Mo were low in the wastewater. This proves
that the Ferrox process, despite washing the residues for
removal of salts, does not produce a wastewater with high
concentrations of trace metals. If a plant is located near a
marine water body, it might be possible to discharge the
wastewater directly to a marine environment after lower-
ing the pH. In other cases, the wastewater can be dis-
charged to the sewer system or evaporated to recycle the
salts for industrial purposes or use as a de-icing agent.
As
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
195
745
214
13
240
1160
275
19
Ba
Cd
Co
Cr
335
979
13
728
1640
1.8
617
1400
19
Cu
Hg
Mo
Ni
14
26
18
50
107
6620
661
448
20,100
83
Pb
5150
410
384
22,200
6740
585
479
19,300
Sn
Sr
Zn
Sum of
measured
elements
g/kg
637
615
574
elements except C, H, and O. Therefore, it has been nec-
essary to anticipate the mineral phases present based on
the assumption that all cation-forming elements were fully
oxidized. Furthermore, gypsum and ferrihydrite previously
had been identified in Ferrox-treated SD and FA.13 A cal-
culation of the total weight based on these minerals
showed that the assumptions made (e.g., full oxidation)
fit within 5–11% of the weight of untreated residues and
within 2–5% of the weight of the treated residues. The
greater difference for the untreated residues is probably
caused by the presence of elements such as Al, which is
not fully oxidized. Although the mineral phases are sug-
gestive only, Figure 2 clearly shows the major changes
caused by the Ferrox treatment. Ca was the most domi-
nant element in the untreated residue, but, although Cl
was washed out, Ca was not removed in any significant
amount because of its precipitation as gypsum (sulfate
was added as part of the FeSO4 solution).
Composition of Solids
The content of selected elements in the residues is given
in Table 3. The composition of the untreated residues
used in the experiments was within the range typical for
these kinds of residue as given by Chandler et al.3 As
expected, the content of Ca and Cl was higher in the SD
residue because of the injection in the flue gas of
Ca(OH)2, which reacts with HCl in the SD flue gas clean-
ing system. The content of Hg was lower in FA than in
the other residues. The reason for this is the low effi-
ciency of the electrostatic precipitator in removing Hg.
The wet scrubber effectively removes Hg and transfers it
to the sludge, and, thus, Hg is present in FAS. The actual
SD APC residue came from an SD flue gas cleaning sys-
tem in which activated carbon was injected prior to the
fabric filter for removal of Hg.
Mass Balances
During the Ferrox process, a substantial amount of the eas-
ily soluble salts was removed. Table 4 shows the removal
in percent of the initial content. More than 90% of Cl was
removed, and more than 70% of Na and K was removed.
The addition of sulfate during the process affected the
Figure 2 shows the distribution of selected elements
in anticipated minerals before and after treatment. Ana-
lytical data were available for the total content of most
726 Journal of the Air & Waste Management Association
Volume 52 June 2002

Lundtorp, Jensen, and Christensen
the largest mass loss because of the initial higher con-
tent of soluble salts (e.g., CaCl2).
Mass balances were set up for treatment of each type
of APC residue (SD, FA, and FAS) with respect to dry mat-
ter, water, and a range of elements. Figure 3 shows, as an
example, the mass balance for FA in terms of dry matter,
water, Pb, and Cl. The figure is based on masses measured
during the pilot plant experiment, determination of dry
matter content, and contents of Pb and Cl in the dry
matter. The water used for washing of the treated mate-
rial on the filter press was recycled for use in the follow-
ing batch and was, therefore, not accounted for in the
mass balance. The mass balance fit within 6% for dry
matter, which is considered to be acceptable. The balance
for water fit within 1%, which might be ascribed to loss
of water during aeration. The balances for Pb and Cl fit
within 5 and 13%, respectively. Figure 3 emphasizes the
objective of the process as the salts (e.g., Cl) left the treat-
ment plant primarily with the wastewater and the trace
metals (e.g., Pb) stayed in the stabilized solid residues.
The general picture, shown in Figure 3 for FA, was similar
for SD and FAS.
Leaching from Solids
The result for Pb, Cd, Cr, and Cl of the column leaching
test, the batch leaching test, and the availability test are
shown in Figure 4. Table 5 shows the accumulated release
at an L/S of 10 L/kg in the batch leaching test for a range
of elements. A comparison of the leaching from untreated
and treated residues shows that the treatment reduced
leaching of Pb by several orders of magnitude. The batch
leaching test for SD and FA showed the largest reductions
in leaching caused by the treatment, up to a factor of
36,000 at an L/S of 2 L/kg for Pb. The result of the column
test indicates that the treatment reduced the release of Pb
between 250 and 1000 times for all three residues. A com-
parison of the batch test and the column test indicates
that the Pb release during the batch leaching test gener-
ally was higher than that of the column test.
Figure 2. Anticipated mineralogical composition of untreated and
treated residues based on the content of elements.
balance of S. No major change in S was observed for FA
and FAS, while SD showed an increase in S caused by the
amounts of FeSO4 and H2SO4 used. The removal of salts
from the residues during treatment resulted in an over-
all mass loss of 7–14%, as can be seen from Table 4. Cal-
culation of the mass loss was based on the content of Al,
Cd, Co, Cr, Cu, Mg, Mn, Ni, P, Pb, Ti, Zn, and Zr in the
solid residue before and after treatment, assuming that
these elements remained in the solids and significant
concentrations were not lost to the wastewater. SD had
Table 4. The removal of salts during the treatment based on the content in the resi-
dues before and after treatment and the loss of weight during the treatment. All figures
are in % (w/w).
Element
SD
FA
FAS
Ca
31
99
87
82
–84
3
15
98
81
74
–3
2
7
92
79
72
5
Cl
K
Na
S
Si
–1
7
Dry matter
14
7
Figure 3. Mass balance with respect to dry matter (DM), water, Pb,
and Cl for treatment of one batch of FA.
Volume 52 June 2002
Journal of the Air & Waste Management Association 727

Lundtorp, Jensen, and Christensen
Figure 4. Release of Pb, Cd, Cr, and Cl from treated and untreated residues during column test, CEN test (two-stage batch test at L/S 2 and 10),
and availability test.
For untreated FA, the higher pH during the batch
leaching test compared with the column test could be an
explanation, because Pb is amphoteric and the release,
therefore, is higher at high pH. For SD, the pH was the
same in the two tests, and, hence, the difference in leach-
ing must be caused by other factors (e.g., time and physi-
cal conditions). The availability of Pb (see Figure 4) was
less than 20% of the Pb content in the residues (see Table
3). The availability of Pb from the untreated FAS was al-
most 1 order of magnitude lower than that of the treated
residue (see Figure 4). This suggests that the content of
sulfide-containing compounds in the sludge from the wet
scrubber wastewater treatment (added to precipitate trace
metals) is, to some extent, persistent during the availabil-
ity test. A lower availability of Cd from the untreated FAS
is likewise observed (see Figure 4). Of the available por-
tion, only 0.0005–0.0009% of the Pb is leachable from
the treated residues under realistic conditions (the col-
umn test). This indicates that the expected leaching of Pb
is very low compared with the actual content of Pb in the
residues. The release of the untreated residues SD and FA
during the batch leaching test is close to the availability,
whereas the release of treated SD and FA during the batch
leaching test is several orders of magnitude lower.
During the availability test, only a small portion of
the iron is released from the treated residues [SD: 0.2%
(w/w); FA: 2.8% (w/w); FAS: 0.2% (w/w)] compared with
the amount added during the treatment (see Table 1).
Hence, the added iron is not leachable. Pb is presumably
attached on the surface of the iron oxides and, therefore,
728 Journal of the Air & Waste Management Association
Volume 52 June 2002

Lundtorp, Jensen, and Christensen
Table 5. Result of the CEN test as accumulated release at an L/S of 10 L/kg.
Residue
SD
FA
FAS
Untreated
Treated
Untreated
Treated
Untreated
Treated
pHa
Spec. Cond.a
12.52
33.5
11.29
3.2
12.69
20.7
11.23
3.3
12.05
15.7
11.17
3.6
mS/cm
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Ca
Cl
77,000
180,000
<0.2
6400
1700
<0.1
18,000
85,000
<0.1
6000
1400
<0.5
10,000
59,000
<0.4
7100
4300
0.5
Fe
K
27,000
<4
850
32,000
<6
1500
<5
27,000
<6
1100
< 5
Mg
Na
S
<5
26,000
3300
0.063
0.012
0.10
1000
4800
0.054
<0.002
3.7
27,000
6200
0.074
0.13
1500
5200
0.032
<0.002
9.5
23,000
7200
0.084
0.032
3.7
1300
4600
0.082
0.0065
1.1
As
Cd
Cr
Cu
Hg
Mo
Ni
2.8
3.7
<0.01
0.002
2.6
0.32
<0.01
0.0009
2.4
0.25
0.023
0.005
1.9
0.013
2.6
0.0089
3.0
0.017
3.6
0.054
730
<0.008
0.048
0.18
<0.01
450
0.052
0.068
0.091
<0.02
48
0.022
0.17
0.19
Pb
Zn
67
29
27
aAt an L/S of 10.
available for leaching at low pH. This assumption is sup-
ported by the fact that the available fraction of Pb in the
treated residues is on the same order of magnitude as
that of the untreated residues. If Pb was confined inside
an iron oxide structure, it presumably would only be re-
leased as the iron is released. This is consistent with the
study performed by Sørensen et al.,13 who found Pb to
be associated with the particle surface of an SD APC resi-
due and an FA treated with the Ferrox process. Further-
more, Sørensen et al.13 mention that, in general, Pb2+
cannot be incorporated in crystalline iron oxides because
of the larger radius of the Pb2+ ion. However, it is pos-
sible that Pb is associated with another phase formed
during the treatment process. Figure 5 shows the leach-
ing of Pb as a function of pH as determined in the pH-
static leaching test. From this figure, it is obvious that,
although pH has an effect on the Pb release, the stabili-
zation (wash out of complexing ligands, addition of iron
oxides) itself accounted for a reduction in Pb release of
up to 4 orders of magnitude.
The release of Cd from both treated and untreated
residues is low (see Figure 4). The higher pH in the leachate
from the untreated residues favors a low release, because
Cd release is solubility-controlled by Cd hydroxide (solu-
bility decreases with increasing pH). Despite this, the
treatment further reduces the leaching of Cd. The Cd con-
centrations in the eluate from the columns with treated
SD and FA were less than the detection limits (0.05 µg/L),
and the release was reduced by more than 2 or 3 orders of
magnitude in the initial leachate (L/S ~0.1 L/kg). The re-
lease of Cd from FAS showed a slight increase after treat-
ment compared with the untreated residue. However, in
all cases, the concentrations were low (untreated FAS resi-
dues: <0.1 µg/L, treated FAS residue: 0.2–0.6 µg/L).
The pH-static leaching test (see Figure 5) clearly
showed that treatment decreases the Cd release. The avail-
able fraction of Cd in both the untreated and treated resi-
due (see Figure 4) is on the same order as the total content
of Cd in the untreated residues, which indicates that Cd
is not bound within the added iron because iron is not
released during the availability test of the treated residues.
Cd is more likely associated with Ca compounds (sulfates
and carbonates), the surface of iron oxides, or as a sepa-
rate phase (Cd hydroxides and Cd carbonates). This is also
consistent with the previously mentioned study performed
by Sørensen et al.13 However, the amounts of Cd leached
from the treated residues during the column test were only
a small fraction of the available amount (between 0.0001
and 0.0013%), suggesting that the leaching of Cd is very
low compared with the content of Cd in the residues.
Furthermore, the lower leaching from the treated residue
suggests the release of Cd from the treated residue is con-
trolled by sorption to the iron oxides.
The treatment does not reduce the release of Cr. For
SD and FA, the release from the treated residue is higher
than that from the untreated residue, especially in the col-
umn test (see Figure 4). This could be a feature of the test
setup, because gas was formed in columns with untreated
Volume 52 June 2002
Journal of the Air & Waste Management Association 729

Lundtorp, Jensen, and Christensen
Figure 5. Release of Cd, Cr, and Pb as a function of pH. Result of a pH-static test, which is a single batch test at an L/S of 10 L/kg and fixed pH.
FA and, to some extent, SD. The gas consisted primarily
of H2 but contained a small amount of CH4. This indi-
cates that there were reducing conditions in these closed
columns with untreated residue, which could cause a re-
duction of Cr(VI) to the less soluble Cr(III). Release of Cr
from FAS is only slightly higher than from the treated
residues. There is no clear trend in the result from the
pH-static leaching tests (see Figure 5). At pH 9, which could
be a realistic pH in a landfill situation, the release of Cr is
slightly higher from treated residues. Cr is present as an
oxyanion. The leaching of other oxyanions (i.e., As, Mo)
shows similar tendencies (see Table 5), suggesting that the
treatment had only limited effect on oxyanions in general.
The figures in Table 5 indicate that treating the residues
reduced the release of other trace metals, including Cu, Hg,
and Zn, in addition to the previously mentioned trace met-
als. For Cu and Zn, the results of the pH-static leaching test
(data not shown) support this conclusion. However, the
results from the pH-static leaching test and the column
leaching test with respect to Hg (data not shown) show no
difference between treated and untreated residue.
that the available fractions are 1–2 orders of magnitude
smaller in the treated residue than in the untreated resi-
due. The release of Cl was consequently also reduced by
1–2 orders of magnitude. The removal of Cl during the
Ferrox treatment leads to a reduction in the release of Cl
from the treated residues. Table 5 shows that the release
of Na, K, and Ca was also reduced after the treatment,
whereas the release of S was only slightly reduced from
FA and FAS and slightly increased from SD. The results
for S were expected, because S was added as FeSO4 as a
part of the treatment and in highest quantities in the case
of SD (see Table 1). A comparison of the achieved results
with the laboratory studies reported in Jensen et al.8 and
Lundtorp et al.9,10 shows that the leaching properties are
the same even though the process has been developed with
changes in both the process and the scale. This indicates
that the leaching properties of treated residues are very ro-
bust and only slightly affected by variations in the process.
CONCLUSION
The Ferrox process has been demonstrated successfully in
a semi-industrial scale on an SD APC residue, an FA, and
an FAS. For FA and FAS, 13–20 kg Fe/t was used in the
treatment. For SD, 49 kg Fe and 9 kg H2SO4 were used per
In Figure 4, Cl is included as a representative of the
salts. Figure 4 shows that almost all available Cl was re-
leased during batch leaching tests and column tests and
730 Journal of the Air & Waste Management Association
Volume 52 June 2002

Lundtorp, Jensen, and Christensen
4. van der Sloot, H.A.; Heasman, L.; Quevauviller, P. Harmonization of
Leaching/Extraction Tests. Studies in Environmental Science 70; Elsevier
Science B.V.: Amsterdam, The Netherlands, 1997.
5. Méhu, J.; Keck, N.; Navarro, A. In Proceedings of Waste Stabilization &
Environment 99, Lyon Villeurbanne, France, April 13–16, 1999; Société
Alpine de Publications: Grenoble, France, 1999.
6. NKK Corporation. Visit to Tokyo Edogawa Incineration Plant Built by
NKK Corporation in October 2000; Environmental Plant Sales Depart-
ment, NKK Corporation: Tokyo, Japan, 2000.
7. The Council of The European Union. Council Directive 1999/31/EC
of 26 April 1999 on the Landfill of Waste; Official Journal of the EC,
L 182, 16/07/1999, 0001-0019.
8. Jensen, D.L.; Lundtorp, K.; Christensen, T.H. Treatment of Incinera-
tor Air Pollution Control Residues with FeSO4: Laboratory Investiga-
tions of Design Parameters; Waste Manage. Research 2002, 20, 69-79.
9. Lundtorp, K.; Jensen, D.L.; Sørensen, M.A.; Mogensen, E.P.B.;
Christensen, T.H. Treatment of Waste Incinerator Air Pollution Con-
trol Residues with FeSO4: Concept and Product Characterization; Waste
Manage. Research 2002, 20, 80-89.
10. Lundtorp, K.; Jensen, D.L.; Sørensen, M.A.; Christensen, T.H. On-Site
Treatment and Landfilling of MSWI Air Pollution Control Residues;
J. Haz. Mat., submitted for publication, 2001.
11. Cornell, R.M.; Schwertmann, U. The Iron Oxides—Structure, Properties,
Reactions, Occurrence and Uses; VCH Verlagsgesellschaft GmbH:
Weinheim, Germany, 1996.
12. Benjamin, M.M.; Sletten, R.S.; Bailey, R.P.; Bennett, T. Sorption and
Filtration of Metals Using Iron Oxide-Coated Sand; Water Res. 1996,
30, 2609-2620.
13. Sørensen, M.A.; Bender-Koch, C.M.; Stackpoole, M.M.; Bordia, R.K.;
Benjamin, M.M.; Christensen, T.H. Effects of Thermal Treatment on
Mineralogy and Heavy Metal Behavior in Iron Oxide-Stabilized APC
Residues; Environ. Sci. Technol. 2000, 34, 4620-4627.
14. Solid Waste, Granular Inorganic Material: Compliance Batch Leaching Test;
Nordtest Method NT Envir 005; Nordtest: Espoo, Finland, 1998.
15. Compliance Test for Leaching of Granular Waste Materials; CEN-Test
Draft, prEN 12457; European Committee for Standardisation: Brus-
sels, Belgium, 1996.
16. Solid Waste, Granular Inorganic Material: Availability Test; Nordtest
Method NT Envir 003; Nordtest: Espoo, Finland, 1995.
17. Solid Waste, Granular Inorganic Material: Column Test; Nordtest Method
NT Envir 002; Nordtest: Espoo, Finland, 1995.
ton treated. The process produced a wastewater, which
had a low content of most trace elements, including As,
Cd, Cr, Cu, Hg, Ni, Pb, and Zn. For example, the concen-
tration of Pb was between 28 and 69 µg/L. During the
process, salts were removed from the treated residue with
the wastewater. Therefore, the wastewater had a high con-
tent of salt (e.g., Cl concentrations between 21 and 52 g/L).
The treated products have a lower content of soluble
salts compared with the untreated residues, and, hence,
the leaching is also drastically reduced (e.g., the Cl re-
lease was reduced by 90–99%). The content of heavy met-
als in the residue is not changed during the treatment,
because only insignificant amounts of metals are removed
with the wastewater. However, in the treated residues, the
release of many heavy metals is controlled by an interac-
tion with the iron oxides, and the leaching of heavy met-
als is significantly reduced. The leachability of Pb was
reduced 250–36,000 times depending on the type of resi-
due and the leaching test. The leachability of Cd, Cu, and
Zn was also substantially reduced. The Ferrox process has
been developed to a level where full-scale application is
possible. The process time, process controls, use of water,
and use of chemical have been optimized. The wastewa-
ter can be discharged easily and the treated residues can
be safely landfilled.
ACKNOWLEDGMENTS
The EU LIFE 99-programme (LIFE 99 ENV/DK000615),
I/S Amagerforbrænding, I/S Vestforbrænding, AV Miljø,
and Babcock & Wilcox Vølund Aps have provided finan-
cial support. Henrik Birch, DHI Water & Environment, is
gratefully acknowledged for his on-site assistance. Erhardt
P.B. Mogensen is acknowledged for his contribution to
the design of the plant and to the experimental planning.
Furthermore, the authors wish to thank Ph.D. student
Mette Abildgaard Sørensen, Ph.D. student Charlotte
Scheutz, Carina Aistrup, Susanne Kruse, Herluf
Riddersholm, Heidi Dvinge, and Bent Skov, Environment
& Resources DTU, Technical University of Denmark, for
their assistance in the analytical work.
About the Authors
Kasper Lundtorp (corresponding author) is an industrial
Ph.D. fellow employed at Babcock & Wilcox Vølund Aps.
He may be contacted at Babcock & Wilcox Vølund Aps,
Falkevej 2, DK-6705 Esbjerg Øst, Denmark (e-mail:
kal@volund.dk). Dorthe Laerke Jensen is an associate pro-
fessor at the Technical University of Denmark in Lyngby,
Denmark. Thomas Hojlund Christensen is a professor at
Environment and Resources DTU (former Department of
Environmental Science and Engineering) at the Technical
University of Denmark.
REFERENCES
1. Burnley, S.J.; Coleman, T.; Gronow, J.R. The Impact of the Landfill
Directive on Municipal Waste Management in the UK. In Proceedings
for Sardinia 99, Seventh International Waste Management and Landfill
Symposium, Vol. IV; Sardinia, October 4–8, 1999; CISA Publisher:
Cagliari, Italy, 1999; pp 279-285.
2. Offermann-Clas, C. The New EU Law on the Landfills of Waste. In
Proceedings for Sardinia 99, Seventh International Waste Management and
Landfill Symposium, Vol. IV; Sardinia, October 4–8, 1999; CISA Pub-
lisher: Cagliari, Italy, 1999; pp 263-270.
3. Chandler, A.J.; Eighmy, T.T.; Hartlén, J.; Hjelmar, O.; Kosson, D.S.;
Sawell, S.E.; van der Sloot, H.A.; Vehlow, J. Municipal Solid Waste In-
cinerator Residues. Studies in Environmental Science 67; Elsevier Science
B.V.: Amsterdam, The Netherlands, 1997.
Volume 52 June 2002
Journal of the Air & Waste Management Association 731