PCDD/F inhibition by prior addition of urea to the solid fuel in laboratory experiments and results statistical evaluation

Article abstract of DOI:10.1016/S0045-6535(00)00248-4

The objectives of this work were to study the effect of urea addition to the fuel before the combustion on the suppression of PCDD/F emissions, to examine the influence of urea to PCDD/F isomer patterns by employing statistical analysis techniques and to determine the most effective method of urea addition. Urea in the form of powder and as an aqueous solution was mixed with RDF and the fuel mixtures were incinerated in a lab-scale reactor at 1000°C. PCDD/F emitted during combustion experiments were collected in a quartz wool filter downstream the reactor unit. Analysis and quantification of PCDD/F showed that urea significantly affected PCDD/F emissions. Combustion of RDF containing 10% w/w urea resulted in PCDD/F emissions lower than 8 I-TEQ ng/g RDF, while the corresponding average value from single RDF combustion was 17 I-TEQ ng/g RDF. The PCDD/F prevention capacity of urea was independent from the method of urea addition to the fuel, as similar results were obtained after the addition of urea in the solid and in the aqueous phases. Results assessment by statistical methods showed that isomer patterns remained almost stable and were not affected by the method of urea addition.

Full text of DOI:10.1016/S0045-6535(00)00248-4

Chemosphere 42 (2001) 737±743
PCDD/F inhibition by prior addition of urea to the solid fuel
in laboratory experiments and results statistical evaluation
a,
b
b
b
b
P. Samaras *, M. Blumenstock , D. Lenoir , K.-W. Schramm , A. Kettrup
a
Laboratory of Solid Fuels and Environment, Chemical Process Engineering Research Institute, 6th km. Harilaou Thermi str., 57001
Thermi-Thessaloniki, Greece
b
Institute of Ecological Chemistry, GSF National Research Centre for Environment and Health, Ingolstadter Landstr. 1, D-85764,
Neuherberg, Germany
Abstract
The objectives of this work were to study the e€ect of urea addition to the fuel before the combustion on the
suppression of PCDD/F emissions, to examine the in¯uence of urea to PCDD/F isomer patterns by employing sta-
tistical analysis techniques and to determine the most e€ective method of urea addition. Urea in the form of powder and
as an aqueous solution was mixed with RDF and the fuel mixtures were incinerated in a lab-scale reactor at 1000°C.
PCDD/F emitted during combustion experiments were collected in a quartz wool ®lter downstream the reactor unit.
Analysis and quanti®cation of PCDD/F showed that urea signi®cantly a€ected PCDD/F emissions. Combustion of
RDF containing 10% w/w urea resulted in PCDD/F emissions lower than 8 I-TEQ ng/g RDF, while the corresponding
average value from single RDF combustion was 17 I-TEQ ng/g RDF. The PCDD/F prevention capacity of urea was
independent from the method of urea addition to the fuel, as similar results were obtained after the addition of urea in
the solid and in the aqueous phases. Results assessment by statistical methods showed that isomer patterns remained
almost stable and were not a€ected by the method of urea addition. Ó 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Combustion; PCDD/F; Emission reduction; Prevention; Inhibition
1. Introduction
combustion sources has been under study for more than
two decades, and today it is known that these com-
pounds are formed through heterogeneous reactions
taking place in the downstream regions (200±400°C) of
chlorine-containing ¯ue gases waste combustion pro-
cesses. Furthermore, it has been proved that metal ions
in the ash, primarily copper, play an important role in
the formation mechanisms acting as catalysts (Stieglitz
et al., 1989; Wehrmeier et al., 1998b).
The incineration of solid wastes has become an at-
tractive alternative method over their disposal to
dumping sites, o€ering the advantages of lowering the
volume of solid wastes and of energy recovery (Bemtgen
et al., 1995). However, the incineration processes are
major sources of potential emissions of hazardous sub-
stances into the environment. Among these compounds,
PCDD/F are of considerable concern as they exhibit
extremely toxic properties. Formation of PCDD/F from
Two approaches have been developed for the control
of PCDD/F emissions to the environment. The ®rst one
refers to the ``end of pipe'' treatment of emissions. The
so-called secondary measure approach is focused on the
removal of PCDD/F from ¯ue gases after their forma-
tion by means of gas cleaning devices. The second ap-
proach, the so-called primary measure technique,
includes measures and actions for controlling the
* Corresponding author. Tel.: +30-31-498-250; fax: +30-31-
498-190.
E-mail address: samaras@alexandros.cperi.certh.gr (P.
Samaras).
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 2 4 8 - 4

738
P. Samaras et al. / Chemosphere 42 (2001) 737±743
PCDD/F formation, such as optimisation of the com-
bustion processes i.e., temperature of furnace and post-
furnace sections, air ¯ow, turbulence, and residence time
(Kaune et al., 1995). This approach also includes the
addition of some selected compounds in the post-com-
bustion zone that aims to block the formation of PCDD/
F and their precursors. This concept is known as inhi-
bition and di€ers from the ¯ue gas cleaning methods.
Inhibition deals with the prevention of PCDD/F for-
mation, while the latter methods remove PCDD/F al-
ready formed.
catalysts and the inhibitor involving the formation of
stable inactive compounds or strongly bonded organo-
metallic complexes.
In most of the studies presented here, the experiments
were performed under laboratory scale conditions,
where ¯y ash samples were impregnated with the in-
hibitors and the reactants, and were heated to the de-
sired temperature. However, in full-scale studies, the
inhibitors were injected into the ¯ue gas at the post-
combustion zone. By such a method, PCDD/F inhibi-
tion potential is a€ected by several operating parameters
such as the location and the temperature of the inhibitor
injection, the contact surface area between the active
metal sites and the inhibitor molecules and the concen-
tration of the compound. In addition, the unadsorbed/
unreacted inhibitors present in the ¯ue gas could also
cause problems, as some inhibitors are toxic themselves
or can be transformed into other organic/inorganic
chemicals which are toxic/corrosive (Buekens and Hu-
ang, 1998). Furthermore, the introduction of an organic
compound in the post-combustion zone could result in
high CO concentration in the stack gases, due to prob-
lems during the combustion process. Preliminary studies
on this inhibition technique, applied on a technical scale
reactor, showed the diculties and the problems of this
method (Lenoir et al., 1989).
Various compounds have been tested as possible
PCDD/F inhibitors. One group of inhibitors includes
basic compounds such as NH3, CaO, NaOH, KOH, and
Na2CO3 (Vogg et al., 1987; Naikwadi and Karasek,
1989; Karasek et al., 1992). These inhibitors might
change the acidity of the ¯y ash surface. Another group
of inhibitors includes compounds that are likely to form
some kind of complexes with the transition metal spe-
cies, responsible for catalysing the PCDD/F reaction
pathways. Sulphur compounds such as Na2S, Na2S2O3,
CS2, thiophene, methyl mercaptan, SO3 and SO2 are
capable of reducing the PCDD/F concentration (Kar-
asek et al., 1988; Lindbauer et al., 1992, 1993; Addink

et al., 1993; Ogawa et al., 1996; Ruokojarvi et al., 1998).
In these cases, a postulated role of sulphur is to react
with the copper catalytic sites, altering their form and
presumably their ability to produce Cl2 through the
Deacon process reaction (Grin, 1986; Gullett et al.,
1992). A second postulated role of sulphur is to undergo
homogeneous reactions, converting the primary chlori-
nating agent, Cl2, into a form (HCl) less likely to un-
dergo aromatic substitution reactions forming PCDD/F
precursors (Tuppurainen et al., 1998). Furthermore,
functionalised amines such as ethanolamine, EDTA,
dimethylamine, and triethanolamine can also be e€ective
as inhibitors (Dickson et al., 1989; Karasek et al., 1992;
In this study, we performed a series of experiments
for the prevention of PCDD/F formation by mixing a
dioxin-preventing compound (urea) initially with the
raw solid fuel, prior to the combustion. Urea was se-
lected for a number of reasons such as low cost and
non-toxic properties. In addition, urea with its amino-
functional group may form some surface nitrogen
complexes with the metal species (such as imine and cy-
anides), resulting in the irreversible deactivation of the
catalytic sites (Lippert et al., 1991; Buekens and Huang,
1998). Furthermore, the addition of urea prior to the
combustion should not su€er from increased CO emis-
sions, observed during the addition of an inhibitor in the
post-combustion zone, as the CO group in the urea
molecule will be oxidised to CO2 in the high temperature
reaction zone, under conditions of complete combustion.
The objectives of this work were to study the e€ect of
fuel pre-mixing on the reduction of PCDD/F emissions,
using a simple method of pre-treatment and a low cost
material, to investigate the in¯uence of urea to PCDD/F
homologue pro®les and isomer patterns by means of
statistical analysis methods and to determine the most
e€ective way of urea addition to the fuel.

Addink et al., 1996; Ruokojarvi et al., 1998). A complex
reaction with the copper catalytic sites is the most likely
manner in which the functionalised amines or, more
probably, their pyrolytic products, work (Lippert et al.,
1991).
Among the nitrogen-containing compounds, urea has
been examined by several research teams for its PCDD/
F inhibiting ability, either by addition to the ¯y ash or
by injection in the form of water±methanol solutions in
the post-combustion zone (Dickson et al., 1989; Yli-
Kenturi et al., 1999). It was found that urea presents a
strong PCDD/F reduction capacity, reaching up to 90%,
when applied in lab- and pilot-scale units. However,
while considerable inhibition was observed in the par-
ticle phase where PCDD/F reduction was high (up to
80%), the gas phase PCDD/F reductions were modest,
2. Experimental

up to 30% (Ruokojarvi et al., 1999; Tuppurainen et al.,
The fuel used in this study was a refuse derived fuel
(RDF), produced from pre-treated municipal solid
wastes (Rethmann Plano GmbH). This fuel contained
1999). In these studies, inhibition activity of urea was
explained as a result of certain interactions between the

P. Samaras et al. / Chemosphere 42 (2001) 737±743
739
0.9 ‹ 0.08% total chlorine and 250 ‹ 29 mg copper/kg
RDF. An amide (urea, Merck) was used as the inhibit-
ing compound in a concentration of 10% of the total
sample amount. Urea was added in the fuel prior to the
combustion experiments by using two methods:
remaining ash was withdrawn from the quartz tube. The
outlet of the quartz tube was extracted for 2 h in toluene,
while the quartz wool ®lter was Soxhlet extracted with
toluene for 24 h. Both extracts from quartz wool ®lter
and quartz tube were then combined in a single solution.
Clean up of samples was performed by a procedure in-
volving application of the extracts in a silica column, an
acidi®ed silica column and an alox column. Identi®ca-
tion/quanti®cation of PCDD/F was performed using a
HRGC/HRMS (MAT 95 Finnigan), according to the
methods described in detail in previous work (Schramm
et al., 1995).
·
Addition of the appropriate amounts of urea in the
solid phase.
·
Preparation of an aqueous solution of urea and addi-
tion to the fuel, followed by mixture drying at 80°C
overnight.
A laboratory scale horizontal type reactor (Fig. 1)
was used, in order to achieve reproducible and easily
repeatable combustion experiments. The laboratory re-
actor consisted of a quartz tube with ID ˆ 4.5 cm and
length ˆ 140 cm. A quartz wool ®lter was placed
downstream of the quartz reactor and was used for solid
particles collection. In addition, two traps, each con-
taining 300 ml toluene, were immersed in an ice bath and
were located after the ®lter for the removal of PCDD/F
from the reaction gases. This system could be easily
cleaned between the combustion experiments in order to
minimise the in¯uence of previous tests (memory ef-
fects), and facilitated the achievement of same experi-
mental conditions.
Statistical evaluation of the data was performed by
multivariate analysis, using a statistical software pack-
age (STATISTICA 5.0). Data assessment commenced
by the construction of the analysis of variance tables
(ANOVA) in order to point out the e€ect of di€erent
urea forms on the emitted homologue patterns and to
determine the di€erences within the homologue patterns.
Analysis of variance accomplished by splitting the ob-
served variance and by testing for di€erences of statis-
tical signi®cance between the mean values.
Principal component analysis (PCA) was applied to
evaluate possible similarities in the composition of
PCDD/F in the samples and especially to group the
congeners. The primary intention of PCA was to obtain
an overview of the dominant ``patterns'' in a data table.
The results of PCA were visualised by graphs, where the
relationship between the emitted congeners was pre-
sented in the so-called score plots. Variables (congeners)
which gave similar results were located close to each
other in the score plots, while those which yielded di-
vergent results were located further apart.
Combustion experiments of RDF and RDF/urea
mixtures were performed under strictly similar experi-
mental conditions. The reactor was electrically pre-
heated to the desired temperature of 1000°C under a
¯ow of 1 l/min air. At the set temperature, about 5 g of
the fuel sample (RDF or RDF/urea) was introduced into
the centre of the heating zone and combustion took
place for 15 min. At the end of the combustion experi-
ment, the glass components were disconnected and the
Fig. 1. Schematic view of the laboratory scale combustion unit (1 ± inlet with activated carbon ®lter, 2 ± sampling boat, 3 ± quartz
wool ®lter, 4 ± toluene traps, 5 ± connection to ¯ow control unit and to vacuum pump).

740
P. Samaras et al. / Chemosphere 42 (2001) 737±743
ture zone, and only part of it escaped combustion or/and
possibly was converted to other active forms.
3. Results
High PCDD/F concentrations were found during the
The distribution of PCDD and PCDF concentrations
in the ¯ue gases is shown in Fig. 2 for the various
combustion experiments. As it is shown in this ®gure,
for RDF combustion, furans dominated over the diox-
ins and the ratio PCDD:PCDF was about 1:11. In the
RDF±urea combustion experiments, furans also pre-
vailed over the dioxins and the corresponding ratio
varied from 1:8 to 1:14.
single RDF combustion experiments, mainly extracted
from the solid particles, which were collected at the
outlet of the reactor and at the quartz wool ®lter. The
average I-TEQ value of PCDD/F measured was about
17 ng/g RDF. However, lower PCDD/F emissions were
measured during the combustion experiments of the
RDF±urea mixtures, than the corresponding values
from untreated RDF. The average I-TEQ value from
duplicate experiments was about 8 ng/g RDF when urea
was added in the solid phase, and slightly lower, about 4
ng/g RDF, when urea was added in the aqueous phase.
Small variations in the PCDD/F I-TEQ values were
observed for the combustion of RDF±urea mixtures,
where the results ranged from 1 to 11 ng/g RDF.
However, high variations deduced from the RDF com-
bustion experiments and the I-TEQ values ranged from
4 to 46 ng/g RDF. Taking into account the closely
controlled experimental conditions, these variations
could be attributed to di€erences in the elemental con-
centration of the raw fuel. For this reason, elemental
analysis was performed on ®ve RDF samples and it was
found that copper relevant concentration in the samples
varied about Æ11:5%, while the chlorine content varied
about Æ8%. As a result, the observed variations in
PCDD/F values could be attributed to variations in the
composition of the solid fuel. Moreover, space limita-
tions in the reactor system, dictated the use of small
amounts of sample, with properties which were not fully
representative of the bulk sample characteristics.
The homologue patterns of PCDD and PCDF
emitted during the combustion of RDF and RDF±urea
mixtures are shown in Figs. 3 and 4, respectively. As it is
shown in these ®gures, lower chlorinated homologues
prevailed over the higher ones.
Statistical analysis was performed on the resulting
data in order to ®nd any possible statistical signi®cant
correlation between the recorded emission data and the
fuel matrices. The ANOVA showed that there is no
signi®cant in¯uence (P-level <5%) of urea and of the
type of urea addition (solid or liquid) on the emitted
PCDD/F pro®les. This means that urea addition to the
The amount of urea used in this study was quite
large, about 10%. However, a very small amount of urea
should be enough to prevent PCDD/F formation by
blocking the copper active catalytic sites. Apparently, a
large amount of urea was burned in the high tempera-
Fig. 3. PCDD homologue patterns during combustion of RDF
and RDF mixtures.
Fig. 2. PCDD and PCDF distributions during RDF and RDF±
urea combustions. (RDF ± single component experiments,
RDF + Uw ± addition of urea in aqueous solution, RDF + Us ±
addition of urea in solid phase).
Fig. 4. PCDF homologue patterns during combustion of RDF
and RDF mixtures.

P. Samaras et al. / Chemosphere 42 (2001) 737±743
741
fuel resulted in PCDD/F emissions, where no statistical
signi®cant di€erences were observed between the group
variances. On the other hand, the internal group vari-
ances of the homologue pro®les showed a signi®cant
dependence of their means due to measured variance (P-
level <5%). Therefore, a coecient of determination (B-
value) similar to the regression analysis as the twin to the
ANOVA in multivariate statistics, could be calculated.
For the homologue pro®les, the observed B-value was
49.5%, which illustrated that more than 49% of the total
variance of the data set were due to the homologue
distribution.
From both Figs. 5 and 6, it was concluded that iso-
mers of same degree of chlorination could be grouped.
The homologues, i.e., all congeners with di€erent degree
of chlorination, showed di€erent behaviour in the PCA.
The similar explained variances for factors 1 and 2 in
the case of PCDD and PCDF, which are denoted in the
captions of Figs. 5 and 6, showed the validity of this
result for both classes of compounds. Furthermore, the
isomer pattern was not in¯uenced from the combustion
parameters and resulted in grouping of isomers with
the same degree of chlorination. The results from the
PCA score plots showed that the urea addition did
not in¯uence the combustion isomer pro®les, while
grouping of isomers might be an indication of di€er-
ent formation pathways of the homologues. These
results are in agreement with the work of other re-
searchers (Wehrmeier et al., 1998a; Blumenstock et al.,
2001) underlining the stability of isomer pro®les against
experimental changes and various combustion condi-
tions.
PCA was employed in this work in order to deter-
mine possible relationships between the chlorinated
congeners (tetra to octachlorinated PCDD/F). The
concentrations of all the congeners from the homologue
groups were used as variables, and the resulting score
plots from PCA are shown in Fig. 5 for dioxins and in
Fig. 6 for furans, respectively, illustrating the correlation
between the fuel mixtures and the homologue patterns.
These preliminary results show that the technique of
urea addition to the fuel prior to the combustion, might
be an innovative, e€ective and low cost prevention
technique of PCDD/F formation. The successful appli-
cation of this method in full-scale combustion units will
result in suppression of their cost, as the installation and
operation of expensive processes like activated carbon
adsorption or catalytic combustion may no longer be
required. However, from a practical point of view, the
transition of lab-scale experiments to pilot- and full-
scale units is not easy to be accomplished, as the for-
mation of chlorinated compounds is a€ected by the
properties of the fuel and especially by the combustion
conditions and the type of the incinerator. In addition
the proportion of urea used in this study was quite large.
Accordingly, the application of the results from this
work to full-scale plants may not be straightforward.
However, the primary idea deduced from this study is
the addition of a low cost inhibitor in the fuel before the
combustion, in order to prevent PCDD/F emissions and
to avoid the use of more complicated methods, where
inhibitors are injected in the post-combustion zone. The
proposed technique of pre-combustion addition seems
to be very promising, as it could be easily applied on an
industrial scale, requiring only minor modi®cations of
the fuel-feeding device in existing plants.
Fig. 5. PCA score plot of dioxins (explained variance for factor
1: 65% and for factor 2: 28%).
Clearly, further experiments are necessary in pilot-
and full-scale combustion reactors, in order to examine
the ability of urea or other low cost and non-toxic in-
hibitors to prevent emissions of toxic gases, to determine
the most ecient inhibitor concentration and to exploit
the mechanism of inhibition. The selection of the in-
hibitor should be made carefully, keeping in mind that it
should work, despite the harsh combustion environ-
ment, block the catalytic sites and not provide any un-
desirable by-products.
Fig. 6. PCA score plot of furans (explained variance for factor
1:65% and for factor 2:25%).

742
P. Samaras et al. / Chemosphere 42 (2001) 737±743
Gullett, B.K., Bruce, K.R., Beach, L.O., 1992. E€ect of sulfur
4. Conclusions
dioxide on the formation mechanism of polychlorinated
dibenzodioxin and dibenzofuran in municipal waste com-
bustors. Environ. Sci. Technol. 26 (10), 1938±1943.
Karasek, F.W., Dickson, L.C, Hutzinger, O., 1988. Incinera-
tion of waste materials. US Patent 4 (793), 270.
The following conclusions were deduced from the
analysis of PCDD/F emitted during combustion of RDF
and RDF±urea mixtures:
1. Urea addition to the fuel prior to combustion was ef-
fective in reducing the PCDD/F I-TEQ concentration
in the ¯ue gases, from 17 ng/g RDF to lower than 8
ng/g RDF.
Karasek, F.W., Naikwadi, K., Hutzinger, O., 1992. Suppres-
sion of dioxin production in the incineration of waste
material. US Patent 5 (113), 772.
Kaune, A., Lenoir, D., Nikolai, U., Schramm, K.-W., Kettrup,
A., 1995. Seven-year decreasing trend of PCDD/F emissions
from a waste incinerator showing importance of primary
measures. Organohalogen Compounds. 23, 477±480.
2. The PCDD/F preventing capacity of urea was inde-
pendent from the method of urea addition to the fuel.
Similar results were obtained after the addition of the
urea in the solid and in the water phases.
3. PCA showed that isomer patterns were not a€ected
by experimental conditions, i.e., by the combustion
of fuels with and without the addition of urea as an
inhibitor.

Lenoir, D., Hutzinger, O., Mutzenich, G., Horch, K., 1989.
PCDD/F: Verhinderung der Entstehung chlorierter Dib-

enzodioxine und -furane bei der Mullverbrennung-Inhibie-
rung der katalytischen Bildung. UWSF-Z. Umweltchem.

Okotox. 1 (4), 3±6.
Lindbauer, R.L., Wurst, F., Prey, T., 1992. Combustion dioxin
suppression in municipal solid waste incineration with
sulphur additives. Chemosphere 25 (7±10), 1409±1414.
Lindbauer, R.L., Wurst, F., Prey, T., 1993. PCDD/F emission
control by sulphur addition ± new results with high-S
lignite, SO2 and SO3. Organohalogen Compounds 12, 35±
38.
Acknowledgements
We would like to thank Mr. U. Hankeln from
Rethmann Plano GmbH for kindly o€ering the RDF
sample. This work was ®nancially supported from EU
by contract number ERBFMBICT 972409.
Lippert, T., Wokaun, A., Lenoir, D., 1991. Copper catalyzed
aryl dehalogenation reactions and their inhibition. Environ.
Sci. Technol. 25 (8), 1465±1471.
Naikwadi, K.P., Karasek, F.W., 1989. Prevention of PCDD
formation in MSW incinerators by inhibition of catalytic
activity of ¯y ash produced. Chemosphere 19 (1±6), 299±
304.
References
Addink, R., Paulus, R.W.L., Olie, K., 1996. Prevention of
polychlorinated dibenzo-p-dioxins/dibenzofurans formation
on municipal waste incinerator ¯y ash using nitrogen and
sulfur compounds. Environ. Sci. Technol. 30 (7), 2350±2354.
Addink, R., Paulus, R.W.L., Olie, K., 1993. Inhibition of
PCDD/F formation during de novo synthesis on ¯y ash
using N- and S-compounds. Organohalogen Compounds
12, 27±30.
Ogawa, H., Orita, N., Horaguchi, M., Suzuki, T., Okada, M.,
Yasuda, S., 1996. Dioxin reduction by sulfur component
addition. Chemosphere 32 (1), 151±157.

Ruokojarvi, P.H., Halonen, I.A., Tuppurainen, K.A., Tarha-
nen, J., Ruuskanen, J., 1998. E€ect of gaseous inhibitors on
PCDD/F formation. Environ. Sci. Technol. 32 (20), 3099±
3103.

Ruokojarvi, P., Aatamila, M., Tuppurainen, K., Halonen, I.,
Bemtgen, J.M., Hein, K.R.G., Minchener, A.J., 1995. Com-
bined combustion of biomass/sewage sludge and coals.
APAS-Clean Coal Technology Programme 1992±1994.
Institute for Process Engineering and Power Plant Tech-
nology, University of Stuttgart, Stuttgart.
Ruuskanen, J., 1999. The e€ect of chemical inhibitor on
PCDD/F concentrations in di€erent particle size fractions.
Organohalogen Compounds 40, 555±558.
Schramm, K.-W., Merk, M., Henkelmann, B., Kettrup, A.,
1995. Leaching of PCDD/F from ¯y ash and soil with ®re-
extinguishing water. Chemosphere 30 (12), 2249±2257.
Stieglitz, L., Zwick, G., Beck, J., Roth, W., Vogg, H., 1989. On
the de novo synthesis of PCDD/PCDF on ¯y ash of
municipal waste incinerators. Chemosphere 18 (1±6), 1219±
1226.
Blumenstock, M., Zimmermann, R., Schramm, K.-W., Kett-
rup, A., 2001. Low chlorinated benzenes, dibenzo-p-dioxins
and dibenzofurans as surrogate compounds for the PCDD/
F emissions in the ¯ue and stack gas of a hazardous waste
incinerator. Chemosphere, accepted for publication.
Buekens, A., Huang, H., 1998. Comparative evaluation of
techniques for controlling the formation and emission of
chlorinated dioxins/furans in municipal waste incineration.
J. Hazard. Mater. 62, 1±33.

Tuppurainen, K., Halonen, I., Ruokojarvi, P., Tarhanen, J.,
Ruuskanen, J., 1998. Formation of PCDDs and PCDFs in
municipal waste incineration and its inhibition mechanisms:
a review. Chemosphere 36 (7), 1493±1511.
Dickson, L.C., Lenoir, D., Hutzinger, O., Naikwadi, K.P.,
Karasek, F.W., 1989. Inhibition of chlorinated dibenzo-p-
dioxin formation on municipal incinerator ¯y ash by using
catalyst inhibitors. Chemosphere 19 (8/9), 1435±1445.
Grin, R.D., 1986. A new theory of dioxin formation in
municipal solid waste combustion. Chemosphere 15 (9±12),
1987±1990.

Tuppurainen, K., Aatamila, M., Ruokojarvi, P., Halonen, I.,
Ruuskanen, J., 1999. E€ect of liquid inhibitors on PCDD/F
formation. Prediction of particle-phase PCDD/F concen-
trations using PLS modelling with gas-phase chlorophenol
concentrations as independent variables.. Chemosphere 38
(10), 2205±2217.

P. Samaras et al. / Chemosphere 42 (2001) 737±743
743
Vogg, H., Metzger, M., Stieglitz, L., 1987. Recent ®ndings on
the formation and decomposition of PCDD/F in municipal
solid waste incineration. Waste Management Res. 5, 285±
294.
Wehrmeier, A., Lenoir, D., Sidhu, S.S., Taylor, P.H., Rubey,
W.A., Kettrup, A., Dellinger, B., 1998b. Role of copper
species in chlorination and condensation reactions of
acetylene. Environ. Sci. Technol. 32 (18), 2741±2748.

Yli-Kenturi, N., Ruokojarvi, P., Asikainen, A., Ruuskanen, J.,
Wehrmeier, A., Lenoir, D., Schramm, K.-W., Zimmermann,
R., Hahn, K., Henkelmann, B., Kettrup, A., 1998a. Patterns
of isomers of chlorinated dibenzo-p-dioxins as tool for
elucidation of thermal formation mechanisms. Chemo-
sphere 36 (13), 2775±2801.

Halonen, I., Hanninen, K., 1999. Urea as PCDD/F inhib-
itor in incineration of RDF. Organohalogen Compounds
41, 311±314.