Environ Sci Pollut Res (2015) 22:14621–14628

DOI 10.1007/s11356-014-3104-4

12TH IHPA FORUM AND SELECTED STUDIES ON POPS

Dioxin formation and control in a gasification–melting plant

Katsuya Kawamoto & Haruo Miyata

Received: 14 January 2014 / Accepted: 26 May 2014 / Published online: 5 June 2014
# Springer-Verlag Berlin Heidelberg 2014

Abstract We investigated dioxin formation and removal in a compounds for plant monitoring was further evaluated. The
commercial thermal waste treatment plant employing a gasi- levels of these surrogates did not exceed values corresponding
fication and melting process that has become widespread in to 0.1 ng TEQ/m3N dioxins established from former tests. This
the last decade in Japan. The aim was to clarify the possibility further substantiated that surrogate measurement therefore can
of dioxin formation in a process operation at high tempera- well reflect dioxin levels.
tures and the applicability of catalytic decomposition of di-
oxins. Also, the possible use of dioxin surrogate compounds Keywords Catalytic bag filter . Catalytic decomposition .
for plant monitoring was further evaluated. The main test Dioxin formation . Incineration plant . Gasification and
parameter was the influence of changes in the amount and melting process . Surrogate organic halogen compounds
type of municipal solid waste (MSW) supplied to the thermal
waste treatment plant which from day to day operation is a
relevant parameter also from commercial perspective. Here Introduction
especially, the plastic content on dioxin release was assessed.
The following conclusions were reached: (1) disturbance of Municipal waste incineration is associated with several im-
combustion by adding plastic waste above the capability of portant environmental problems related to combustion and
the system resulted in a considerable increase in dioxin con- discharge of residue. In addition to the emission of inorganic
tent of the flue gas at the inlet of the bag house and (2) bag air pollutants such as hydrochloric acid and nitrogen oxides
filter equipment incorporating a catalytic filter effectively (NOx) or heavy metals, also hazardous organic
reduced the gaseous dioxin content below the standard of micropollutants such as polychlorinated dibenzo-para-di-
0.1 ng toxic equivalency (TEQ)/m3N, by decomposition and oxins (PCDDs), polychlorinated dibenzofurans (PCDFs),
partly adsorption, as was revealed by total dioxin mass bal- and dioxin-like polychlorinated biphenyls (dl-PCBs) (together
ance and an increased levels in the fly ash. Also, the possible referred as dioxins in this paper) are of concern and have
use of organohalogen compounds as dioxin surrogate attracted a great deal of attention. Historically high levels of
emissions from hazardous waste incinerators (Holmes et al.
1994, 1998; Liem et al. 1991; Malisch et al. 1999) but also
Responsible editor: Philippe Garrigues
municipal waste incinerators (Takeda and Takaoka 2013) have
Electronic supplementary material The online version of this article resulted in the contamination of surrounding soils via atmo-
(doi:10.1007/s11356-014-3104-4) contains supplementary material,
which is available to authorized users. spheric deposition of dioxins or releases from wet scrubbers.
Soils have also been polluted by dioxins by the disposal of
K. Kawamoto (*)
contaminated ashes from incineration processes (Holmes et al.
Graduate School of Environmental and Life Science, Okayama
University, 3-1-1 Tsushima-naka, Kita-ku, Okayama 700-8530, 1994, 1998; Pless-Mulloli et al. 2001; Watson 2001; Weber
Japan et al. 2008). Such soil contamination has resulted in the
e-mail: kkawamoto@okayama-u.ac.jp contamination of food products such as meat, milk, or eggs,
whose adulteration through the food chain and the ultimate
H. Miyata
Formerly, Sanki Engineering Co., Ltd., Seiroka Tower, Akashi-cho, exposure to humans highlights the need for an assessment of
Chuo-ku, Tokyo 110-0015, Japan areas around waste incinerators or metal industries known to
14622 Environ Sci Pollut Res (2015) 22:14621–14628

10000 Municipal waste incinerators

have emitted high levels of dioxins (Holmes et al. 1994, 1998; Industrial waste incinerators
Liem et al. 1991; Malisch et al. 1999; DiGangi and Petrlík Small-scale waste incinerators, etc.
2005). On the other hand, the current use of persistent organic Others

Emission amount (g-TEQ/y)

Total
pollutants and similar chemicals (Scheringer et al. 2012) in
products and articles results that such waste should not be 1000

recycled or disposed but requires destruction capacity such as

thermal waste treatment plants (Weber et al. 2013).
The introduction of dioxin regulations in Japan including a
limit of 0.1 ng toxic equivalency (TEQ)/m3 in the flue gas 100
from newly constructed incineration plants has resulted in a
drastic decrease in dioxin air emissions in Japan (Fig. 1)
(Ministry of the Environment 2014). Dioxins have been re-
duced in the last two decades not only by advanced flue gas 10
treatment but also by improvement of combustion control and 1996 1998 2000 2002 2004 2006 2008 2010 2012

proper treatment of incineration residues (Yoshida et al. 2009; FY

Hagenmaier 1991). Dioxins are in particular generated during Fig. 1 Historical changes in total emissions of dioxins in Japan
the gas-cooling process and in the gas-cleaning equipment
even in modern BAT incineration plants. Dioxins are formed
here by de novo synthesis during the degradation of carbona- over the last decade. The main processes in this system are an
ceous materials (e.g., soot and poly-aromatic hydrocarbon initial pyrolysis/gasification and then a secondary melting step
(PAH)) in the presence of chlorine and metals in the temper- that is triggered by high-temperature combustion. Because the
ature range of ca. 200 to 550 °C (Addink and Olie 1995; process has a very high-temperature melting process (usually
Weber et al. 2001a, b). Dioxin release is usually reduced to around 1,300 °C), it is considered that dioxins are completely
low levels by dust abatement using bag filter operated at decomposed. However, dioxins can still form in the flue gas
temperatures below 200 °C. The substitution of old electro- cooling catalyzed by fly ash. Fly ash from melting furnace
static precipitators operated in the de novo temperature win- differs from conventional one by a very low carbon content
dow to bag filters operated at lower temperatures has resulted but contains a higher level of catalytic metals (Kawamoto and
in a large decreases in dioxin emissions (Kawamoto 2008). Mabuchi 2001). The dioxin formation characteristic of this
The mechanisms of flue gas cleaning by bag filters are the thermal waste treatment system with this different cooling and
following: (1) better condensation of gaseous pollutants at fly ash condition has not been described for their dioxin
lower operating temperatures (around 150 to 200 °C) than formation potential.
with electric precipitators, (2) fine dust collection (to sub- The approaches to control dioxin formation and release
micrometer range), and (3) enhancement of physical–chemi- can include the optimization of the thermal treatment/
cal removal on the filter, whereby the deposited layer is incineration step including waste feeding and mixing or
composed of fly ash and alkali reagent along with activated the optimized flue gas cleaning. Within this regime, also
carbon powder, which works effectively as adsorbents, in the rapid signaling of increasing dioxin concentration might be
mentioned low temperature range (Inoue and Kawamoto a parameter for optimizing plant performance. For this, a
2005; Chang et al. 2009). surrogate flue gas index of organic halogens has recently
Dioxin formation during the cooling process may result in been used for flue gas monitoring (Kawamoto et al. 2003;
two types of problem. Formation can increase pollutant con- Watanabe et al. 2010; Reinmann et al. 2006, 2010; Nakui
centrations in the fly ash in the dry-basis flue gas cleaning et al. 2011).
processes which are most popular in modern incineration In this study, we investigated the dioxin formation and
plants. This might require additional treatment of the ash removal in a commercial thermal waste treatment plant
because in Japan, the dioxin content of the dust collected in using gasification and melting processes. By an appropri-
the bag filter is regulated to no more than 3 ng TEQ/g ate sampling strategy, we gained insights into dioxin com-
(Ministry of the Environment of Japan 1999) in the final position changes in the flue gas and ash samples with a
disposal. The second problem is that if substantial amounts total dioxin balance. The testing covered the influences of
of dioxins are formed, then large amounts of adsorbents such changes in the mixing ratio of air supplied to the furnace
as activated carbon are needed to reduce their emission to and the amount of municipal solid waste (MSW) supplied
below the regulation limit of 0.1 ng TEQ/m3N for stack to the substitute incinerator including also the change of
emissions. waste composition in respect to plastic content. Further-
A new type of thermal waste treatment plant applying a more, the relationships between dioxin and surrogate con-
gasification and a melting stage has been introduced in Japan centrations were measured and are discussed.
Environ Sci Pollut Res (2015) 22:14621–14628 14623

Materials and methods increase in plastic content, wastes containing plastics from
crushed bulky items such as furniture were added to the usual
Thermal treatment plant investigated MSW at two different rates. These wastes were known to
contain relatively high share of plastic/polymers above
The flow of the incineration plant and the measurement points 20 %, while the precise plastic content was not determined
is shown in Fig. 2. The plant employs a fluidized bed-type because of quantification difficulties at the plant. The bag
gasifier in which the temperature is around 600 °C and a high- filter was operated at a temperature of 190 °C.
temperature melting furnace that is operated at about
1,300 °C. The plant has a capacity of 70 t MSW per day in Flue gas and ash sampling
two systems. The capacity is rather small since this plant was
constructed to serve a small city with a population of about Experiments on gas and ash measurement of dioxins and
62,000. Combustibles in the MSW are first gasified in the related PAHs were conducted in the incineration plant five
gasification reactor by contact with fluidized silica sand. Then, times (once a day for 5 days). Flue gas was sampled both
the mixture of gas, char, and noncombustibles emitted from before and after the bag filter and over a period of 4 h, from
the gasification process is introduced in a gyrating motion into about noon to 1600 h every day. During flue gas measure-
the melting furnace. The gas emitted from the melting process ment, fly ashes were sampled several times from the bottom of
is cooled in two apparatuses by the addition of water. After it the bag filter, and a composite sample was made to ensure a
has been cooled, the flue gas is cleaned in the bag filter and representative sample with an average dioxin content of the
further in the selective catalytic reduction (SCR) equipment. testing period. Dioxin surrogates comprising organic halogen
This bag filter contains a dioxin decomposition catalyst in the compounds such as gaseous chlorobenzenes in the flue gas
fil ter material to degrade semi-volatil e gaseous were also quantified after the bag filter. This surrogate sam-
micropollutants such as dioxins within the dust collection step pling was performed simultaneously by using two methods in
(Babcock-Hitachi 2014). The TiO2-based V2O5 catalyst is order to evaluate the different measurement methodologies.
impregnated into the fabric of the filter at a rate of 500 g/m2 Gas samples were collected by using two kinds of adsorption
of the filter felt. tube sets and then quantified by using different methods. In
this study, monitoring of organic halogens in flue gas samples
after bag filtration has been conducted over the whole series of
Plant operation 5 days of dioxin measurement experiments.

Measurements were conducted under conditions that took into Analysis

account such factors as the proportions of air injected at
different points and changes in the nature of the MSW input. Dioxins in the flue gas and fly ash samples were analyzed
Five runs were conducted (Table 1). The conditions used in according to the Japanese Industrial Standard JIS K 0311 by
run 1 were the standard operation conditions. The experimen- using high-resolution gas chromatography and high-
tal parameter “ratio of injected air” represented the ratio of the resolution mass spectrometry (HRGC–HRMS). Then, the
air injected before the melting furnace inlet to that injected at congener distribution and material balance of PCDDs and
the inlet. Assessment of this parameter was designed to inves- PCDFs were investigated. PAHs were also quantified by using
tigate the effect of changes in the amount of preinlet air on the the same HRGC–HRMS in order to assess major products of
mixing of gas in the furnace and on further combustibility in incomplete combustion in the flue gas and fly ash.
the chamber. The values of this parameter were changed in In the sampling of surrogate organic halogens, gaseous
runs 2 and 3. In runs 4 and 5, the type of MSW supplied to the compounds were continuously sampled with an adsorption
gasifier was partly changed. To investigate the influence of an tube at 100 °C every 30 min. This detection method was based
on the atomic absorption of halogen atoms at specific wave-
600ºC
lengths by using atomic emission spectrometry employing
1,300ºC 750ºC
MSW Fluidized bed Secondary White plume
radiofrequency helium plasma discharges (Watanabe et al.
Melting Gas
gasification combustion prevention 2010). In the other method of manual sampling, gas was
furnace furnace chamber air preheater cooler
drained off and separately collected, and organics in the gas
370ºC 195ºC 190ºC 210ºC were captured by two activated carbon tubes in series. The
Combustion Catalytic SCR
air preheater
Gas cooler
bag filter (De-Nox)
Stack carbon material was then combusted in an analytical furnace
఑Gas/ash in the laboratory, and the hydrochloric and hydrobromic acids
Injection of lime
Fly ash sampling point formed were quantified with a titration detector (Kawamoto
Fig. 2 Flow diagram of the incineration plant and gas and fly ash et al. 2003). Here, the organic halogens quantified by the first
sampling points method are referred to as organic chlorine (O-Cl) and organic
14624 Environ Sci Pollut Res (2015) 22:14621–14628

Table 1 Experimental conditions

Run Ratio of MSW used Amount of waste Proportion of waste with
no. injected air thrown into the high percentage of plastics
gasifier in average added to usual waste (%)
Preinlet Inlet (kg/h)
(%) (%)

1 40 60 Usual 1,225 –
2 32 68 Usual 1,400 –
3 50 50 Usual 1,490 –
4 40 60 Addition of waste containing a 1,155 10
high percentage of plastics
5 40 60 Addition of waste containing a 1,205 17
high percentage of plastics
at upper limit of designed
value
MSW municipal solid waste

bromine (O-Br), and those quantified by the latter are referred TEQ/m3N) probably because of post-furnace synthesis during
to as total organic halogens (TOX). gas cooling. All measurements at the outlet of the bag filter
were below 0.1 ng TEQ/m3N with an average mean value of
0.082 ng TEQ/m3N (standard deviation was 0.010 ng TEQ/
Results and discussion m3N), demonstrating that the catalytic filter would be enough
for reaching the 0.1 ng TEQ/m3N regulation limit. The mea-
Dioxin and PAH measurements and their relationship to plant surement study was conducted at 18 months after the thermal
operation waste treatment plant began to operate, and this fact suggests
that the catalytic effect of the bag filter could last long time.
Measurement data for dioxins in flue gas and fly ash are The dioxins were further substantially reduced by the SCR
shown in Table 2. However, the detailed data of PCDD, performed by the catalytic de-NOx system as shown in Fig. 1.
PCDF, and dl-PCB are given in the Online Resource. We The values were finally below 0.01 ng TEQ in the final
found that the concentration of dioxins in the gas at the outlet exhaust to the atmosphere which was in agreement with
of the catalytic bag filter was in each run below 0.1 ng TEQ/ measurements from the operator over the preceding 4 years
m3N with a bag filter inlet concentration of around 2.0 ng (average was 0.0062 ng TEQ/m3N and the standard deviation
TEQ/m3N in runs 1 to 4. Also, levels in the fly ash were quite was 0.0055 ng TEQ/m3N).
low (between 0.11 and 0.24 ng TEQ/g). In run 5, the dioxin Increasing the content of plastic waste with a high calorific
concentrations in the inlet gas (11 ng TEQ/m3N) and in the fly value above the feeding capacity of the thermal waste treat-
ash (3.0 ng TEQ/g) were by far the highest of all runs. ment plant reduced the stability of combustion. From the
The measurements demonstrate that the dioxin decompo- calorific value data for the usual MSW incinerated at the plant
sition catalyst performed very well. The average percentage and from the mixing ratio of the plastic-rich waste in run 5, the
removal of dioxin concentration from the gases in runs 1 to 4 heat value of the waste supplied in this run was roughly
was 94 and 96 % for the toxic equivalent. Although the flue estimated to be about 12,000 kJ/kg. This value was beyond
gas sampling between the gasification and melting step could the designed maximum value for the plant (10,600 kJ/kg).
not be done due to the actual difficulty in the flue gas sam- This was seen with the CO concentrations of run 5 with
pling, the gas entering the bag house was assumed to have 100 ppm, while runs 1 to 4 had CO levels of 26, 9, 30, and
relatively high dioxin concentrations (between 1.7 and 2.3 ng 38 ppm, respectively. These demonstrate that incomplete

Table 2 Measurement data for

dioxins Run no. Gas entering BF Fly ash Gas exiting BF Ratio of dioxin removal
(ng/m3N at 12 % O2) (ng/g) (ng/m3N at 12 % O2) from gas (%)

1 140 (2.3) 12 (0.17) 7.0 (0.071) 95 (97)

2 120 (1.7) 9.3 (0.11) 7.3 (0.078) 94 (95)
Numbers in parentheses are toxic 3 140 (2.0) 11 (0.13) 8.9 (0.098) 94 (95)
equivalents (ng TEQ/m3 N or ng 4 130 (2.0) 20 (0.24) 8.5 (0.076) 93 (96)
TEQ/g)
5 760 (11) 390 (3.0) 7.7 (0.085) 99 (99)
BF bag filter
Environ Sci Pollut Res (2015) 22:14621–14628 14625

combustion occurred especially in run 5. The conditions of Therefore, the PAH concentration at the outlet gas of the bag
run 4 in which a certain amount of high calorific waste fraction filter was low.
has been added were estimated to be roughly equivalent to the
maximum design value. This explains that the CO and dioxin Dioxin mass balance and composition profile
levels in run 4 were similar to those in runs 1 to 3. The changes
in the proportions of air injected before the inlet of the melting The point of mass balance analysis of dioxins around the bag
furnace and at the inlet port did not affect the dioxin concen- filtration process is usually the point at which these com-
tration, as shown in the data from runs 2 and 3 compared with pounds are removed from gas entering the bag filter and
those under the standard conditions of run 1 (Tables 1 and 2). moved to the fly ash collected in the filter (Yasuda et al.
The results of PAH measurements in Table 3 also show a 2009). However, the most marked difference in the measure-
difference between run 5 and other runs. Two points can be ment data in this study was that the catalytic filter efficiently
made. The first is that the total amount of PAHs in the fly ash reduced dioxin concentrations through decomposition at
in run 5 was high (170 ng/g dry), at 6.8 times than in run 1. 190 °C operation compared to the general de-NOx tower
This trend was similar to that with the dioxin data. Four reduction. In this removal, decomposition of gaseous compo-
compounds—phenanthrene, naphthalene, pyrene, and fluo- nents is the basic process that occurs, and the ratio of dioxins
ranthene—predominated. However, the PAH gas concentra- decomposed may differ according to the distribution of flue
tions in run 1 at the inlet of the bag filter were generally higher gas dioxin homologs and congeners between the gaseous and
than those in run 5: For example, the naphthalene concentra- particulate components.
tion was 6,400 ng/m3N at 12 % O2 in run 1 and 3,300 ng/m3N We calculated the mass balances of dioxins around the
in run 5. This result differed from that for the dioxins. The catalytic bag filtering process in runs 1 and 5 (Fig. 3). The
difference in gaseous PAH concentration and fly ash concen- amounts of dioxins in gases and fly ashes were calculated
tration might be explained by the higher adsorption quality of from measured concentrations in those media. The decompo-
fly ash in run 5 with the lower combustion quality. The sition amount was simply calculated as the difference between
removal efficiencies in the bag filter were almost the same the inlet amount and the amounts in the outlet gases and fly
(96.2 % in run 1 and 95.5 % in run 5). These measurement ash combined. The amount of dioxins in the outlet gas in run 1
data show that gaseous PAHs are decomposed with high was only 0.38 μg TEQ/h (3.2 %), compared with 12 μg TEQ/
efficiency by the catalytic filter or catalyst as has been shown h in the inlet gas, while the amount of dioxins in the fly ash
in laboratory experiments (Weber et al. 1999, 2001a, b). was 4.0 μg TEQ/h (33 % of the inlet amount). Therefore, the

Table 3 PAHs in flue gases (ng/m3 N at 12 % O2), with gas removal ratios and fly ash contents (ng/g dry)

Run 1 Run 5

Inlet of BF Fly ash Outlet of BF Removal ratio (%) Inlet of BF Fly ash Outlet of BF Removal ratio (%)

Naphthalene 6,400 5.7 340 94.7 3300 42 170 94.8

Acenaphthylene 330 0.29 0.70 99.8 12 0.83 0.20 98.3
Acenaphthene 16 5.3 0.69 95.7 10 1.0 0.71 92.9
Fluorene 39 1.5 3.4 91.3 21 0.38 4.4 79.0
Phenanthrene 900 2.6 16 98.2 460 51 6.5 98.6
Anthracene 38 0.28 2.6 93.2 17 3.1 1.1 93.5
Fluoranthene 490 2.1 1.0 99.8 170 28 1.3 99.2
Pyrene 1,400 5.1 0.37 100 130 33 1.0 99.2
Benzo[a]anthracene N.D. 0.05 0.20 – 3 0.20 0.10 96.7
Chrysene 20 0.29 0.13 99.4 13 1.4 0.46 96.5
Benzo[b]fluoranthene 18 0.20 0.06 99.7 13 0.72 0.15 98.8
Benzo[k]fluoranthene N.D. 0.04 N.D. – 3 0.15 0.05 98.3
Benzo[a]pyrene N.D. N.D. N.D. – 3 0.26 0.08 97.3
Indeno[1,2,3-cd]pyrene 6 0.05 N.D. 100 2 0.13 0.03 98.5
Benzo[g,h,i]perylene 190 1.7 N.D. 100 42 4.9 0.11 99.7
Dibenzo[a,h]anthracene N.D. N.D. N.D. – N.D. 0.005 N.D. –
Total 9,800 25 370 96.2 4200 170 190 95.5

BF bag filter, N.D. no data

14626 Environ Sci Pollut Res (2015) 22:14621–14628

Run 1 Decomposition : 8.1 difference is probably the higher adsorption quality of the fly
ash in run 5 with higher carbon content compared to the other
Inlet :12
runs with better combustion quality.
BF Outlet : 0.38
The fate of dioxins can be discussed on the basis of changes
in the distribution patterns of PCDD/F congeners in the gas
and the fly ash. Figure 4 shows the changes in homolog
Fly ash : 4.0 pattern in runs 1 and 5. The pattern of the inlet gas (gas and
ash) in run 1 shows that the lower chlorinated PCDFs were the
Decomposition : 14 major congeners at standard operation. The flue gas after bag
Run 5
Inlet:69
filtration had the same pattern. Similar homolog distributions
were observed in runs 2 to 4. However, the fly ash contained
BF Outlet : 0.54 PCDD constituents at levels similar to those of PCDFs. This
indicates that PCDD remains in fly ash more than PCDF or
PCDD was newly formed on the ash. In contrast, the inlet flue
gas in run 5 was composed of PCDFs consisting mainly of
higher chlorinated congeners but also with relatively small
Fly ash：55
amounts of PCDDs. Although the dioxin pattern in the fly ash
Fig. 3 Mass balance of dioxins around the bag filter (BF) process differed considerably between runs 1 and 5, the distribution
(numbers show dioxin flow rates in μg TEQ/h) pattern in the outlet gas from the catalyst was almost the same.
The reason is that the lower chlorinated PCDFs have a higher
volatility and are mainly present in the gas phase. In addition,
amount decomposed was estimated from these data to be PCDDs are better degraded by catalytic filters compared to
8.1 μg TEQ/h (close to 68 % of inlet amount). Also, the PCDF (Weber et al 2001b).
dioxin amounts in the fly ash from runs 2 to 4 ranged from Examination of the pattern of PCDD/Fs observed in the
17 to 39 % of the amounts in the inlets, indicating that about fly ash of run 5 revealed the predominance of PCDD
one third of the dioxins in the inlet flue gases was adsorbed to congeners—especially the more highly chlorinated
particles and retained in the ash under these operation condi- ones—and was quite different from that of run 1. This
tions. However, for run 5, most of the dioxins were adsorbed finding suggests that formation of highly chlorinated
on to the fly ash: 79 % of the dioxins in the inlet gas were PCDD congeners dominated the fly ash in run 5 during
retained by the fly ash. Therefore, the percentage decomposi- the gas-cooling process between the combustion air pre-
tion of gaseous dioxins decreased for run 5 to 20 %, compared heater and gas cooler under conditions in which the con-
with 61 to 83 % in runs 1 to 4. The major reason for this tact time may have been increased (Fig. 2).

Influent gas Influent gas

Concentration (ng/m 3N@12%O2)

Concentration (ng/m3N@12%O2)

80 200

60 150

40 100
Decompo- 50 Decompo-
20 sition Effluent gas sition
0
Concentration (ng/m 3N@12%O2)

0 2.5 Effluent gas

pC s
O s

pC s

O s
xC s

H Fs
Pe D s

P e DF s

F
D
H DD
D

H DF
F
H D

D
D

D
D
D
D
D

C
C

xC
C
C
C
C
CD s
O s

Hp Fs

O s
Hx Ds

Hx Fs
Pe Ds

Pe Fs

Te
F

Concentration (ng/m 3N@12%O2)

D

Te
Hp D
D

F
CD
CD

D
CD

2
D
D

D
D
D

2
C
C
C

C
C
C

Te
Te

BF
1.5 1.5

BF 1

0.5
1

0.5

Fly ash 0 Fly 0

ash
CD s
O s

Hp Fs

O s
Hx Ds

Hx Fs
Pe Ds

Pe Fs

F
D
CD s
O s

Hp Fs

O s
Hx Ds

Hp D
Hx Fs

2
Pe Ds

Pe Fs

F
F
D

CD
CD
Hp D
D

D
CD
D
D

D
D
D
CD
CD

D
CD
D

C
D

C
C

C
D

C
D

C
Concentration (ng/g)

Te
C

Te
C
C

C
C
C

150
Te
Te

Concentration (ng/g)

1.5
100
1

0.5 50

0 0
CD s
CD s

O s

Hp Fs
O s

Hp Fs

O s
Hx Ds
O s
Hx Ds

Hx Fs
Pe Ds

Pe Fs
Hx Fs
Pe Ds

Pe Fs

F
D
F
D

Hp DD
D

F
Hp D
D

CD
CD

CD
CD

D
CD
D
D
D
D

D
CD
D
D

D
D
D

C
C
C

C
C
C
C
C
C

C
C
C

Te
Te
Te
Te

Run 1 Run 5
Fig. 4 Changes in PCDD/F congener distribution patterns in runs 1 and 5
Environ Sci Pollut Res (2015) 22:14621–14628 14627

TOX（µg/m3N) 250
concentration of O-Cls was approximately 1 order of magni-
200
tude higher compared to O-Brs. The range of concentration
150
changes for both was not very large, indicating that no specific
100 high bromine-containing waste was introduced to the plant
15 0.1

DXN-TEQ（ng-TEQ/m3N)
during these measurement campaigns. Moreover, these con-
DXNs(ng/m3N@12%O2)

0.08
O-Cls/O-Brs（µg/m3N)

centrations were markedly lower than the TOX because of the

10
0.06 selective nature of the collection method. The fact that TOX
0.04 was observed in higher concentrations than the others sug-
5
0.02
gests that the halogenated organic compounds in the flue gas
before the stack were much more volatile and semi-volatile.
0 0
0 6 12 18 0 6 12 18 0 6 12 18 0 6 12 18 0 6 12 18 0 Since the catalytic bag filter effectively removed gaseous
Run1 Run 2 Run 3 Run 4 Run 5 dioxins, there was practically no difference in dioxin concen-
O-Cls O-Brs DXNs DXN-TEQ TOX
tration in the off-gas after the bag filter in the five runs
Fig. 5 Monitoring data for surrogate organic halogens and dioxins (Table 2). Therefore, it was not possible to establish a corre-
measured in each run
lation between the dioxin and surrogate organic halogen con-
centrations in these experiments. However, the data obtained
Surrogate organic halogen monitoring
in the runs can be used to monitor surrogates for the control of
dioxin emissions, because the O-Cl concentrations were lower
A rapid monitoring of the gaseous emission of dioxins and
than 10 μg/m3, where results of other studies have shown that
other hazardous compounds might be useful for the control of
the dioxin concentrations can also be regarded as low
such emissions, e.g., by increased carbon spray, and also
(Kawamoto et al. 2003; Watanabe et al. 2010; Nakui et al.
supports public awareness and acceptance of waste incinera-
2011). Also in these studies, the dioxin levels were statistically
tion plants. In this context, many studies have been conducted
correlated with the O-Cl concentration.
to establish rapid or continuous dioxin monitoring or sampling
With our limited data set, we found a correlation between
methods, or both, for the appropriate control of emissions
TOX concentration and the sum of the O-Cl and O-Br con-
from thermal waste treatment plants (Kawamoto et al. 2003;
centrations (Fig. 6). However, the concentration ranges of the
Reinmann et al. 2006, 2010; Yasuda et al. 2009; Watanabe
two data sets were quite different. This finding suggests that
et al. 2010; Nakui et al. 2011).
more volatile and semi-volatile compounds went through the
Monitoring data for surrogate organic halogens and five
catalytic bag filter, whereas dioxins were decomposed by
TOX data points are shown in Fig. 5. Dioxin levels for total
selective catalytic reactions. This is in agreement with the
concentration and toxic equivalents obtained from the daily
results from the laboratory test with TiO2-based dioxin oxida-
measurements are also plotted in the figure. The range of TOX
tion catalysts which showed that at the operation temperature
differs from those of the other surrogates. This is because
of the bag filter, PCDD/Fs are still removed and destroyed
volatile, semi-volatile, and further water-soluble organic hal-
with high efficiency while the removal rate of chlorobenzenes
ogen compounds are collected and analyzed in the TOX
(the major organohalogen compounds in incinerator off-gas)
measurement, whereas semi-volatile organic halogens are col-
at this temperature is low (Weber et al. 1999).
lected in the measurement of O-Cls and O-Brs owing to
selective sampling at a temperature of 100 °C. The

250 Conclusions

We drew the following conclusions from our study in a

200
commercial gasification and melting process:
TOX (µg/m3)

1. The dioxin formation characteristics in the cooling line of

150
a high-temperature gasification melting system are com-
parable with those in a thermal waste treatment plant. The
system can guarantee low dioxin emissions if it is operat-
ed within the designed plant operation frame.
100
5 7 9 11 13 2. The disturbance of combustion caused by adding a high
O-Cls + O-Brs (µg/m3) share of plastic waste above the designed capacity result-
Fig. 6 Relationship between the concentrations of the surrogates (O-Cls ed in an increase in dioxin content of the flue gas at the
+ O-Brs) and TOX, as quantified by using two different methods bag filter inlet. However, the catalytic bag filter effectively
14628 Environ Sci Pollut Res (2015) 22:14621–14628

reduced the gaseous dioxin content below 0.1 ng TEQ/ Waste Management and Public Cleansing Law specially controlled
municipal solid waste
m3N, by decomposition and partly adsorption, as was
Ministry of the Environment: National Dioxins Emission Inventory
revealed by total dioxin mass balance and increased level (Summary) (2014) http://www.env.go.jp/en/headline/file_view.
in the fly ash. The fate of dioxins was discussed on the php?serial=443&hou_id=1743. Accessed Apr 2014
basis of changes in the distribution patterns of PCDD/F Nakui H, Koyama H, Takakura A, Watanabe N (2011) Online
measurements of low-volatile organic chlorine for dioxin
congeners in the gas and the fly ash.
monitoring at municipal waste incinerators. Chemosphere
3. To monitor the dioxins in flue gases by a simpler param- 85:151–155
eter, we quantified surrogate organic halogen compounds. Pless-Mulloli T, Edwards R, Päpke O, Schilling B (2001) PCDD/PCDF
The surrogate compound concentrations in this text did not and heavy metals in soil and egg samples from Newcastle allot-
ments. Report by the University of Newcastle
exceed values corresponding to 0.1 ng TEQ/m3N dioxins
Reinmann J, Kuch B, Weber R (2006) Continuous monitoring of unin-
established in former studies. Surrogate concentration there- tentionally formed POPs listed under the Stockholm.
fore reflected the dioxin levels quite well also in this study Organohalogen Compd 68:852–855
and might be used for a simplified monitoring approach. Reinmann J, Weber R, Haag R (2010) Long-term monitoring of
PCDD/PCDF and other unintentionally produced POPs—con-
cepts and case studies from Europe. Sci China Chem 53:
1017–1024
References Scheringer M, Strempel S, Hukari S, Ng CA, Blepp M, Hungerbühler K
(2012) How many persistent organic pollutants should we expect?
Atmos Pollut Res 3:383–391
Addink R, Olie K (1995) Mechanisms of formation and destruction of Takeda N, Takaoka M (2013) An assessment of dioxin contamination
polychlorinated dibenzo-p-dioxins and dibenzofurans in heteroge- from the intermittent operation of a municipal waste incinerator in
neous systems. Environ Sci Technol 29:1425–1435 Japan and associated remediation. Environ Sci Pollut Res 20:2070–
Babcock-Hitachi KK (2014) Catalytic fabric filters, http://www.bhk.co. 2080
jp/environment_e/decrease/catalyst/index.html. Accessed Apr 2014 Watanabe N, Kawamoto K, Asada S, Fujiyoshi H, Miyata H, Watanabe
Chang Y-M, Hung C-Y, Chen I-H, Chang C-T, Chen C-H (2009) G, Suzuki S (2010) Surrogate study for dioxins from municipal
Minimum feeding rate of activated carbon to control dioxin emis- waste incinerator in startup condition: applicability as a dioxin
sions from a large-scale municipal solid waste incinerator. J Hazard control indicator and an organohalogen emission warning. J Mater
Mater 161:1436–1443 Cycles Waste Manag 12:254–263
DiGangi J, Petrlík J (2005) The egg report. International POPs Watson A (2001) Comments on the “Report on the analysis of
Elimination Network (IPEN) http://english.arnika.org/ipep/reports/ PCDD/PCDF and heavy metals in soil and egg samples related to
GLOBAL_eggsreport.pdf the Byker incinerator”. http://www.greenpeace.org.uk/files/pdfs/
Hagenmaier H (1991) Destruction of PCDD and PCDF in refuse incin- migrated/MultimediaFiles/Live/FullReport/3188.pdf. Accessed
eration plants by primary and secondary measures. Chemosphere Apr 2014
23:1429–1437 Weber R, Sakurai S, Hagenmaier H (1999) Low temperature decompo-
Holmes SJ, Jones KC, Miller CE (1994) PCDD/PCDF contamination of sition of PCDD/PCDF, chlorobenzenes and PAHs by TiO2-based
the environment at Bolsover UK. Organohalogen Compd 24:373–377 V2O5-WO3 catalysts. Appl Catal B Environ 20:249–256
Holmes SJ, Green N, Lohmann R, Jones KC (1998) Concentrations of Weber R, Iino F, Imagawa T, Takeuchi M, Sakurai T, Sadakata M (2001a)
PCDD/PCDFs in soil around a point source. Organohalogen Compd Formation of PCDF, PCDD, PCB, and PCN in de novo synthesis
39:257–260 from PAH: mechanisms and correlation to fluidized bed incinera-
Inoue K, Kawamoto K (2005) Fundamental adsorption characteristics of tors. Chemosphere 44:1429–1438
carbonaceous adsorbents for 1,2,3,4-tetrachlorobenzene in a model Weber R, Plinke M, Xu Z, Wilken M (2001b) Destruction efficiency of
gas of an incineration plant. Environ Sci Technol 39:5844–5850 catalytic filters for polychlorinated dibenzo-p-dioxin and dibenzo-
Kawamoto K (2008) Experience of incinerator development in Japan, In: furans in laboratory test and field operation—insight into destruction
Seminar on thermal waste treatment in Hong Kong, http://www.epd. and adsorption behavior of semivolatile compounds. Appl Catal B
gov.hk/epd/english/news_events/events/files/K_Kawamoto.pdf. Environ 31:195–207
Accessed Apr 2014 Weber R, Gaus C, Tysklind M, Johnston P, Forter M, Hollert H, Heinisch
Kawamoto K, Mabuchi K (2001) Formation characteristics of H, Holoubek I, Lloyd-Smith M, Masunaga S, Moccarelli P, Santillo
polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, D, Seike N, Symons R, Torres JPM, Verta M, Varbelow G, Vijgen J,
and coplanar PCBs in fly ash from a gasification–melting process of Watson A, Costner P, Wölz J, Wycisk P, Zennegg M (2008) Dioxin
municipal solid waste. J Mater Cycles Waste Manag 3:38–47 POP-contaminated sites—contemporary and future relevance and
Kawamoto K, Imaizumi T, Fujiyoshi H, Tanaka M (2003) Application of challenges. Environ Sci Pollut Res 15:363–393
summary organic halogens for the management of dioxins in flue Weber R, Aliyeva G, Vijgen J (2013) The need for an integrated approach
gas (in Japanese). J EICA 8:31–41 to the global challenge of POPs management. Environ Sci Pollut
Liem AKD, Hoogerbrugge R, Kootstra PR, van der Velde EG, de Jong Res Int 20:1901–1906
APJM (1991) Occurrence of dioxins in cow’s milk in the vicinity of Yasuda K, Yoda I, Kawamoto K (2009) Feasibility of a monitoring
municipal waste incinerators and a metal reclamation plant in the system for detecting changes in dioxin concentrations of both in
Netherlands. Chemosphere 23:1675–1684 flue gas and fly ash in incineration process. J Mater Cycles Waste
Malisch R, Gleadle A, Wright C (1999) PCDD/F in meat samples from Manag 11:328–338
domestic farm animals and game. Organohalogen Compd 43:265–268 Yoshida H, Takahashi K, Takeda N, Sakai S (2009) Japan’s waste
Ministry of the Environment, Government of Japan (1999) http://law.e- management policies for dioxins and polychlorinated biphenyls. J
gov.go.jp/htmldata/S46/S46F03601000035.html, Regulations of Mater Cycles Waste Manag 11:229–243