Determination of polychlorinated dibenzo-p-dioxins and dibenzo-furans in solid residues from wood combustion by HRGC/HRMS

Article abstract of DOI:10.1016/S0045-6535(99)00332-X

PCDD/PCDF were determined in solid samples from wood combustion. The samples included grate ashes, bottom ashes, furnace ashes as well as fly and cyclone ashes. The solid waste samples were classified into bottom and fly ash from native wood and bottom and fly ash from waste wood. For each of the four classes concentration distribution patterns from individual congeners, the sums of PCDD/PCDF and the international toxicity equivalents (I-TEQ) values are given. The I-TEQ levels of fly ash from waste wood burning can be approximately up to two thousand times higher than the values from fly ashes of natural wood. The I-TEQ levels in bottom ashes from waste wood combustion systems are as low as the corresponding ashes from the combustion of native wood. Grate ash samples from waste wood combustion systems with low carbon burnout show high levels of PCDD/PCDF.

Full text of DOI:10.1016/S0045-6535(99)00332-X

Chemosphere 40 (2000) 641±649
Determination of polychlorinated dibenzo-p-dioxins and
dibenzo-furans in solid residues from wood combustion by
HRGC/HRMS
a,
*
b
b
b
Samuel Wunderli , Markus Zennegg , Ivan Samuel Dole ꢀz al , Erika Gujer ,
Urs Moser , Max Wolfensberger , Philip Hasler , Dominik Noger ,
b
b
c
a
d
Christoph Studer , Georg Karlaganis
d
a
Section of Chemistry, Swiss Federal Laboratories for Materials Testing and Research (EMPA), Lerchenfeldstr. 5, CH-9014 St. Gallen,
Switzerland
b
Section of Organic Chemistry, Swiss Federal Laboratories for Materials Testing and Research (EMPA), Ueberlandstrasse 129,
CH-8600 D u bendorf, Switzerland
c
Verenum, Langmauerstrasse 109, CH-8006 Z u rich, Switzerland
Swiss Agency for the Environment, Forests and Landscape (BUWAL), Bern, Switzerland
d
Received 11 May 1999; accepted 26 August 1999
Abstract
PCDD/PCDF were determined in solid samples from wood combustion. The samples included grate ashes, bottom
ashes, furnace ashes as well as ¯y and cyclone ashes. The solid waste samples were classi®ed into bottom and ¯y ash
from native wood and bottom and ¯y ash from waste wood. For each of the four classes concentration distribution
patterns from individual congeners, the sums of PCDD/PCDF and the international toxicity equivalents (I-TEQ) values
are given. The I-TEQ levels of ¯y ash from waste wood burning can be approximately up to two thousand times higher
than the values from ¯y ashes of natural wood. The I-TEQ levels in bottom ashes from waste wood combustion systems
are as low as the corresponding ashes from the combustion of native wood. Grate ash samples from waste wood
combustion systems with low carbon burnout show high levels of PCDD/PCDF. Ó 2000 Elsevier Science Ltd. All
rights reserved.
Keywords: Polychlorinated dibenzo-p-dioxins (PCDD); Polychlorinated dibenzo-furans (PCDF); Solid waste; Ash; Wood combustion;
Electro®lter; Fly ash; Incineration
P
ABR:
sum of all quanti®able congeners. The limit of detection ( 6 ) is assumed to be 3 times the peak-to-peak noise of the mass trace
combined measurement
uncertainty (Anon, 1993b); MWI municipal waste incinerator
1
0 times the peak width near the expected peak; I-TEQ international toxicity equivalent value (Safe, 1992); u
c
1
. Introduction
Wald und Landschaft (BUWAL), Bern, with the aim of
conducting an actual survey concerning the quantities of
solid wastes and the concentrations of polychlorinated
dibenzodioxins and dibenzofurans (PCDD and PCDF)
contained therein. The solid wastes investigated were by-
products originating from various processes of the Swiss
technological environment. The solid wastes were slags,
ashes, ®lter dusts and ®lter cakes from the di€erent
processing steps of municipal and hazardous waste in-
This investigation represents a part of a project di-
rected by the Swiss authorities Bundesamt f u r Umwelt,
*
Corresponding author. Tel.: +41-71-274-7785; fax: +41-71-
74-7785.
E-mail address: samuel.wunderli@empa.ch (S. Wunderli).
2
0
045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 3 3 2 - X

6
42
S. Wunderli et al. / Chemosphere 40 (2000) 641±649
cinerators and wood combustion facilities, solid wastes
from metal industry, slags from battery recycling, solid
waste from the paper industry, ash from cremation and
tissue incineration from hospitals, ash from open ®res
and sludge from road dust samplers. The data were
acquired from more than 100 samples, including ap-
proximately 30 samples from many types of wood
combustion, which provide the basis for the future cal-
culation of the corresponding PCDD/PCDF ¯ow in
these wastes. The samples investigated and presented in
this paper originated mainly from the di€erent furnace
parts of burner systems for wood combustion in which
natural, residual or waste wood was incinerated. The
samples were electro®lter ashes, ¯y ashes, bottom ashes
and grate ashes.
about the type of chemicals and their concentrations
which are contained in such residues (Noger et al.,
1996). Dioxins and some heavy metals are present in
nearly all types of ashes from wood burning and have
most relevance to the decisions on how to eciently
optimise the safe disposal of the residues (Oehme and
M u ller, 1995; Noger et al., 1996; BUWAL, 1993;
Hasler and Nussbaumer, 1994, Pohlandt and Mar-
utzky, 1994).
The combustion of wood in Switzerland is mainly
accomplished by using both log wood boilers and au-
tomatic wood chip furnaces. Manually fed log wood
boilers are more widely used than automatic units. The
primary fuel for log wood boilers is uncontaminated
native wood. Burning of waste wood is not permitted.
Table 1 summarises the number and types of actual
Swiss wood burning systems. Mainly small-scale com-
bustion systems are operated.
The main primordial sources for the unintended
formation of PCDD/PCDF are thermal and chemical
industrial processes (Fiedler and Van den Berg, 1996;
Ballschmitter and Bacher, 1996; Oehme, 1998). Wood
burning is always accompanied by the unwanted pro-
duction of PCDD/PCDF. They can be found in the solid
ashes and in the waste gas stream. The amount of
PCDD/PCDF that will be formed is dependent on the
quality of wood burned.
The combustion of waste wood requires a more ad-
vanced ash handling system and furnace technology
(Nussbaumer et al., 1998). Furthermore, it is prohibited
to burn waste wood in furnaces with a capacity below
350 kW. Waste wood furnaces are primarily used as
process heat units. Fuel gas cleaning systems are re-
quired such as an electrostatic precipitator or fabric ®l-
ters to limit the heavy metal and particulate emissions.
Table 2 shows the estimated amount and classi®cation
of wood burned in di€erent types of combustion sys-
tems. The quantity of wood used in Switzerland for
energy production annually amounts to approximately
In Switzerland, the annual wood burning per capita
amounts to about 320 kg (Neue Z u rcher Zeitung,
1
996). 380 000 t of waste wood are produced annually.
A more intensive use of wood to produce energy is a
component of the present Swiss energy policy. Poten-
tially, about 6% of the heat energy production of
Switzerland could be supplied by wood combustion
3
2.1 million m of wood resulting in 500 GWh of energy.
(
VHe, 1991; Noger et al., 1996). In 1993 the combus-
6
m of wood resulted in an estimated
In Switzerland more than 60% of the energy from
wood is produced in small scale log wood boilers with
untreated wood as a fuel. The overall combustion of
wood generates approximately 24 000 t of ash annually
(Table 3).
3
tion of 2.1 ´ 10
amount of 24 000 t of ash. An approximate mass
fraction of 1% remains as ash from wood burning. This
compares to a mass fraction of 5±6% of ash remaining
from the burning of Miscanthus, (family Graminae).
This plant is discussed as a possible alternative renew-
able energy source.
The investigation of three classes of wood (LRV,
1998) and two types of solid residues were the main
subject of this work: waste wood (painted, glued and
waste wood from construction and demolition of
buildings and furniture, excluding PVC coated waste
and impregnated wood; residual wood products (all types
of wood products which have been processed such as
chipboard and wood dust from machining); native wood
The solid residues from wood combustion can either
be used in industry as supplementary material such as
concrete production or ®ller material, in agriculture as
fertilisers or must be disposed of in waste dumps. Re-
search work is in progress to get more information
Table 1
Number and size of wood combustion systems used in Switzerland (1993) (Nussbaumer et al., 1998)
Combustion system
Log wood boiler
Heat output (kW)
Number of facilities in Switzerland
<30
>30
490 000
115 000
Automatic wood chip furnace
<70
70
1400
2400
>
Automatic urban waste wood chip furnace
>350
15

S. Wunderli et al. / Chemosphere 40 (2000) 641±649
643
Table 2
The estimated amount of wood used for energy production in Switzerland (1993; 10
6
3
m
of wood equals to 2500 GWh of energy)
(
Nussbaumer et al., 1995)
Combustion system
Native wood, residual wood
3
Waste wood
3
(m )
Total wood
3
(m )
(m )
Log wood boilers
Automatic wood chip furnaces
1 310 000
647 000
Combustion not allowed
110 000
1 310 000
780 000
Total
1 980 000
110 000
2 090 000
Table 3
Wood ash generated annually in Switzerland (1993) (Noger et al., 1996)
Combustion system
Log wood boiler
Heat power output (kW)
<30
30
<70
70
Wood ash generated (t/a)
6100
7200
400
>
Automatic wood chip furnace
>
6500
4000
Automatic urban waste wood chip furnace
Total
>350
±
24 000
(
natural wood without any treatment, including saw
dust, shavings and bark).
Pulverisette, Germany). The mean diameter of the
sample particles after grinding was 0.05±0.1 mm.
Grinding quartz sand followed by water and acetone
rinsing was used to clean the grinding equipment after
each sample. Subsamples of 10±50 g, depending on the
amount of PCDD/PCDF presumed to exist, were used
for the analysis after a homogenising step by turning
and shaking the glass bottles.
The classi®cation di€ers from other countries (e.g.,
Germany). The solid residues were classi®ed as ¯y ash
and bottom ash. Fly ash includes all solid material being
separated from the ¯ue gas stream by cyclones, elec-
trostatic precipitators (ESP), fabric ®lters and ceramic
®
lters. Bottom ash contains all solid residues (commonly
known as wood ash) which are collected after passing
through a moving grate or remaining in the furnace
compartment depending on the construction of the in-
cinerator. The bottom ash was sometimes partially
transformed into slag. From some incinerators only a
mixture of cyclone ash and ash from grate separation
was obtainable for analysis. Such sample materials were
classi®ed as ¯y ash.
Soxhlet extraction, clean-up and detection were per-
formed according to a method described in reference
(Dole ꢀz al et al., 1995). Soxhlet extraction instead of su-
percritical ¯uid extraction (SFE) using carbon dioxide/
acetone was chosen because the thimble volume was too
small for certain samples. Furthermore, no extraction
eciency study could be performed for all the di€erent
type of samples. A short description of the procedure is
given:
The air-dried ash was treated with diluted hydro-
chloric or sulphuric acid, the residue was ®ltered, wa-
shed free of acid and dried again. The aqueous phase
was extracted using ®rst methylene chloride and then re-
extracted with cyclohexane for the collection of the re-
maining methylene chloride. The air-dried solid phase
was then transferred to a pre-extracted thimble, the di-
luted multi spike solution was distributed over the solid,
covered by dry sodium sulphate and Soxhlet extracted
for 24 h using toluene. The organic phases were com-
bined and the volume reduced to about 1 ml by evap-
oration of the solvents at a reduced pressure. For the
multistage clean-up, a semi-automatic low pressure
chromatography system (Fluid management system,
FMS, Watertown MA, USA) with a combined silica
2
. Experimental
Representative sampling was performed according to
DIN and ISO standards (Sommer, 1979). In large fa-
cilities the sample object sometimes could not be ac-
cessed as a whole. In those cases subsamples were
collected periodically, mixed and divided later. Finally,
about 1 kg of sample was collected, ground in the lab-
oratory and stored in the darkness under argon at
�
20°C.
The grinding equipment consisted of a rotation type
mill. It had a swinging cylinder and a ¯at torus in a ¯at
cylindrical box all part made of tungsten carbide closed
with a rubber sealed tungsten carbide cover (Fritsch,

6
44
S. Wunderli et al. / Chemosphere 40 (2000) 641±649
column (acid/base), alumina column (ICN supe B) and
carbon celite column (PX 21) was used. For each sample
the whole set of columns was exchanged completely in
order to avoid any cross contamination. Toluene was
used to back¯ush the PCDD/PCDF from the carbon
column. After high level samples, the clean-up system
was additionally rinsed using di€erent solvents. No sig-
ni®cant memory signals from the tubing could be de-
tected. No sulphur removal step was necessary for any
of the investigated sample extracts.
The recovery values for all the individual 2,3,7,8-
chlorine substituted PCDD/PCDF ranged from 60% to
100% with a median value of 90%. The whole method
was tested in several certi®cation exercises conducted by
BCR for the analysis of PCDD/PCDF congeners in
di€erent candidate certi®ed reference materials. The
linearity of the mass spectrometric response was at least
four orders of magnitude and ranged from 100 ng to 10
pg of 2,3,7,8-chlorine substituted congener per injection.
The linearity of the mass spectrometric response was
determined by plotting the relative response factors
calculated from the sum of the peak areas of two mass
traces from the native and the sum of two peak areas of
mass traces from the labelled compounds versus the
mass injected for each 2,3,7,8-chlorine substituted con-
gener. A series of ®ve di€erent standards was used for
this purpose. The native standards were supplied from
BCR (Bureau Communautarie de R ꢂe f ꢂe rence, Geel,
GC-separation was achieved with a J&W DB-Dioxin
column (length 60 m, inner diameter 0.25 mm, ®lm
thickness 0.25 lm). Helium carrier gas with an inlet
4
pressure of 6.9 ´ 10 Pa was used. Gas chromatographic
separation was carried out on a Varian 3400 GC after
automatic splitless injection (splitless period 40 s) of 2±
3
ll of sample at 260°C injector temperature. The GC
was programmed starting at 140°C with a hold-time of
1
3
1
min. The temperature was then increased at a rate of
Belgium) and the
C12-labelled internal isotope stan-
1
1.43°C/min up to 220°C following by the second tem-
dards originated from CIL (Cambridge Isotope Labo-
ratories, MA, USA). The reference material used to
check the analytical method was CRM 429 (BCR ¯y ash
extract). The concentration values for most of the
congeners measured in the reference material CRM 429
agreed within the ranges of the certi®ed values.
perature rate of 2°C/min up to 260°C. The end tem-
perature of 260°C was hold 260°C for 62 min. Each
chromatographic run took 80 min. The Finnigan MAT
9
to 10 000 (10% valley de®nition).
5 HRGC/HRMS mass spectrometric resolution was set
‡
Electron impact ionisation EI at 70 eV was used.
The two most abundant ions of the molecular isotope
cluster were monitored with multiple ion detection
3
.1. Classes of di€erent wood ash samples
(
MID) at 0.8 s cycle time. In the ®rst three mass win-
A total of 37 samples were quanti®ed for PCDD/
PCDF. Table 5 shows I-TEQ values measured.
dows (18 masses per window monitored) the dwell time
was 31 ms for each mass and 75 ms in the forth mass
window (10 masses monitored).
The bar in Figs. 1±5 illustrate the concentration
distributions and indicate the minimum, 10% percentile,
median, 90% percentile and the maximum value for each
variable (cf. Fig. 1). Mostly the highest contribution to
the I-TEQ value derives from the 2,3,4,7,8-PeCDF.
OCDD, OCDF and 1,2,3,4,6,7,8-HpCDD are domi-
nating in the distribution of the 2,3,7,8-chlorine substi-
tuted congeners.
All the solvents used were from a single batch (Merck
quality: Supra Solv). They were used without further
pre-treatment.
3
. Results and discussion
A median value of 2±3% is found for 2,3,7,8-chlorine
substituted TCDD concentration compared to OCDD
(cf. Fig. 2). Fig. 3 demonstrates that the highest
contribution to the I-TEQ value stems from the 2,3,4,7,
8-PeCDF. Highest concentrations are found for OCDD,
1,2,3,4,6,7,8-HpCDD and 1,2,3,7,8,9-HpCDF. Fig. 4
shows that again the highest contribution to the I-TEQ
value originates most often from the 2,3,4,7,8-PeCDF.
The combined measurement uncertainty for individ-
ual congeners concentrations is estimated to be 20±30%
based on samples used in a BCR certi®cation exercise
and a comparison of three selected individual samples.
Comparison measurements of these three samples were
performed in the University of Ume ꢁa , Sweden. The re-
sults (cf. Table 4) agreed reasonably.
Table 4
Measurement comparison from the two laboratories, EMPA and University of Ume ꢁa , Sweden
Residue type
c
I-TEQ values ‹ u (ng/kg)
EMPA
University of Umeꢁa
Electro®lter ash from a modern MWI
Cyclone ®lter ash from residual wood
Grate ash from deciduous trees
1600 ‹ 400
2000 ‹ 500
7.1 ‹ 1.8
810
2500
6.3

S. Wunderli et al. / Chemosphere 40 (2000) 641±649
645
Table 5
Levels of I-TEQ values (ng/kg) in ¯y and bottom ash from the di€erent classes of wood; n is the number of samples analysed in that
class; median values instead of mean values are given
a
b
c
Residue
type
Waste wood
Residual wood
Native wood
Min
Max
Median (n)
Min
Max
Median (n)
Min
Max
Median (n)
Fly ash
Bottom
730
4.2
21 000
11
2800 (10)
6.1 (4)
18
±
6300
±
2000 (3)
± (0)
1.5
0.60
4.0
8.6
2.6 (6)
4.9 (9)
a
For the distributions (cf. Figs. 1±5), the three residual waste wood ¯y ash samples of this table were included in the waste wood ¯y ash
class as no distinct di€erence in I-TEQ values between ashes from waste wood and residual wood could be seen.
b
No Bottom ash samples were available.
The native wood bottom ash sample class was extended by 5 samples (cf. Fig. 2). One ash from a forest ®re (I-TEQ ˆ 9.9 ng/kg), two
c
charcoal samples from charcoal works (I-TEQ ˆ 4.9 and 8.9 ng/kg) and two samples from an open forest ®re with low and high carbon
burnout (I-TEQ ˆ 31 and 32 ng/kg).
Fig. 1. Native wood bottom ash samples. Distributions (strip box plots) for all the 2,3,7,8-chlorine substituted congeners concen-
trations, the sums for the each congener group and for the I-TEQ values.
The most prominent individual congeners are OCDD,
,2,3,4,6,7,8-HpCDD, 1,2,3,7,8-PeCDF, 2,3,4,7,8-
PeCDFand1,2,3,4,6,7,8-HpCDF.
oxins, whereas the concentrations of the 2,3,7,8-PCDFs
do not show this increase but have elevated levels for
the lower and higher chlorinated congeners. It is evi-
dent that the I-TEQ values from ¯y ash and bottom
ash originating from native wood are considerably less
di€erent than the values for waste wood where they
di€er by a factor of more than one thousand. Native
wood ¯y ash comprises either mixed grate and cyclone
ash or mixed furnace and cyclone ash or pure cyclone
ash. Waste wood ¯y ash comprises either ashes from
fabric ®lters, electrostatic precipitators, or cyclone
separators.
1
The distribution of I-TEQ-levels of bottom ash from
natural wood is comparable to that of the ¯y ash from
natural wood (Fig. 5), whereas the congener concen-
tration distribution patterns of the two classes, are not
similar. In Fig. 5 residual wood ash samples are in-
cluded in the waste wood ¯y ash class. Ash from waste
wood samples is heavily loaded compared to other
types of wood ash samples. For the 2,3,7,8-chlorine
substituted congener distributions see Figs. 1±4. Inter-
estingly Fig. 5 shows that native wood ¯y ash samples
have slightly reduced compared to the corresponding
bottom ashes. Generally the concentrations increase
with the degree of chlorination for the dibenzo-p-di-
Some typical patterns of the 2,3,7,8-chlorine substi-
tuted PCDD/PCDF in ¯y ash samples from waste wood
and bottom ash, as well as from the burning of natural
wood have been presented earlier by our group

6
46
S. Wunderli et al. / Chemosphere 40 (2000) 641±649
Fig. 2. Native wood ¯y ash samples.
Fig. 3. Waste wood bottom ash samples.
(
Wunderli et al., 1996). The distribution of I-TEQ-levels
called wet ash from this tank was removed continu-
ously by a conveyor belt. An I-TEQ value of 17 000
ng/kg was determined in the ¯y ash (cf. Fig. 6(b))
obtained from the fabric ®lter of the same burner.
Both patterns of the 2,3,7,8-chlorine substituted
congeners from the ¯y and grate ash look similar, but
the PCDD/PCDF content of the ¯y ash is approxi-
mately four to ®ve times higher. In this speci®c case
the PCDD/PCDF content of the grate ash from waste
wood does not di€er from ¯y ash samples originating
of bottom ash from natural wood is comparable to that
of the ¯y ash.
A single grate ash sample (cf. Fig. 6(a)) from waste
wood with low carbon burn-out showed much higher
PCDD/PCDF levels. This grate ash contained a high
content of unburned carbonised pieces of wood that
means a low carbon burnout. The hot ash fell through
a grate into a water-containing tank from where the
ash cooled down instantly. The ash sediment, the so-

S. Wunderli et al. / Chemosphere 40 (2000) 641±649
647
Fig. 4. Waste wood ¯y ash samples.
Fig. 5. Distribution of the I-TEQ values in the ashes from the four classes of wood combustion.
Fig. 6. (a) Waste wood grate ash (I-TEQ ˆ 3300 ng/kg) from a 0.9 MW grate burner (same burner as in (b)) low carbon burn-out. (b)
Waste wood ¯y ash (I-TEQ ˆ 17 000 ng/kg) from a 0.9 MW grate burner equipped with fabric ®lter.

6
48
S. Wunderli et al. / Chemosphere 40 (2000) 641±649
from other waste wood incineration facilities. The wet
ash contained unburned carbon piece. Activated
carbon or household cokes are very ecient
adsorption materials for PCDD/PCDF (Dubinin,
References
Ballschmitter, K., Bacher, R., 1996. Dioxine: Chemie, Analytik,
Vorkommen, Umweltverhalten und Toxikologie der halo-
genierten Dibenzo-p-dioxine und Dibenzofurane. VCH,
Weinheim.
1
965, Hasler and Nussbaumer, 1995). The smallest
carbon particles originating from the solid carbonised
waste wood containing the PCDD/PCDF are trans-
ported by upstreaming hot air and become part of the
BUWAL (Bundesamt f u r Umwelt, Wald und Landschaft),
1993. Dioxinemissionen von Holzfeuerungen. Schriftenreihe
Umwelt Nr. 208, Bern, Switzerland.
¯y ash.
Dole ꢀz al, I.S., Segebarth, K.P., Zennegg, M., Wunderli, S., 1995.
Comparison between supercritical ¯uid extraction (SFE)
using carbon dioxide/acetone and conventional soxhlet
extraction with toluene for the subsequent determination
The degree of carbon burnout of the waste in¯uences
the PCDD/PCDF concentration in the ashes. When
waste wood is combusted a high carbon burnout yields
ashes with low PCDD/PCDF contents, whereas a low
carbon burnout produces ashes with high PCDD/PCDF
contents. An additional reason for increased levels of
PCDD/PCDF is the possible content of pentachloro-
phenol (PCP) in the waste wood. PCP is a well-known
precursor of PCDD/PCDF or which contains substan-
tial traces of PCDD/PCDF from the manufacturing
process. PCDD/PCDF will be formed when PCP is
subjected to a high temperature treatment. These factors
may have caused the unusual high I-TEQ values in the
grate ash.
of PCDD/PCDF in
Chemosphere 31 (9), 4013±4024.
a single electro®lter ash sample.
Dubinin, M.M., 1965. Theory of the bulk saturation or
microporous activated charcoals during adsorption of gases
and vapours. Russ. J. Phys. Chem. 39 (6), 697±704.
Fiedler, H., Van den Berg, M., 1996. Polychlorinated dibenzo-
p-dioxins polychlorinated dibenzofurans and related com-
pounds: update and recent developments. Environ. Sci.
Pollut. Res. 3, 122±128.
Hasler, Ph., Nussbaumer, T., 1994. Dioxin-und Furanemissi-
onen bei Altholzfeuerungen, Eidgen o ssische Druck-
materialzentrale Bern, Nr. 805.174 d.
Hasler, Ph., Nussbaumer, T., 1995. Optimierung des Abscheid-
2
everhaltens von HCl, SO , und Dioxinen/Furanen in einem
Gewebe®lter nach einer Altholzfeuerung, Bundesamt f u r
Energiewirtschaft (BEW), 3003 Bern.
4
. Conclusions
ISO, 1993. Guide to the expression of uncertainty in measure-
ment. ISO, Geneva, Switzerland.
The I-TEQ-values of all the analysed solid residues
LRV (Luftreinhalte-Verordnung), 1998. Switzerland, 3000
Bern, (16 December 1985 and updated February 1998).
Neue Z u rcher Zeitung, 1996. 28 February, Nr. 49, 61±64,
Z u rich.
from the incineration of native wood are below 10 ng/
kg. For such ashes disposal problems are minor. The I-
TEQ-values determined in the ashes from combustion
of residual wood are not di€erent from those of waste
wood. The production of PCDD/PCDF from solid
waste wood material may be very signi®cant when
combustion is incomplete. In this case, the grate ash
can be heavily loaded with PCDD/PCDF. Generally,
considerably lower values, usually 4±10 ng/kg, are
found in urban waste wood bottom ashes. A high
carbon burnout is highly recommendable in order to
lower the PCDD/PCDF load in the grate ash and to
minimise the amount of ash. The produced PCDD/
PCDF adhere to the activity carbon matrix which is
formed in situ. When such carbon particles are left
unburned, increased levels of PCDD/PCDF are found
in the residue.
Noger, D., Felber, H., Pletscher, E., Hasler, Ph., 1996.
Verwertung und Beseitigung von Holzaschen, Bundesamt
f u r Umwelt, Wald und Landschaft (BUWAL), Bern,
Switzerland, Schriftenreihe Umwelt Nr. 269.
Nussbaumer, T., Good, J., Jenni, A., Koch, P. (Eds.), 1995.
Projektieren automatischer Holzfeuerungen, Impulspro-
gramm PACER, Bundesamt f u r Konjunkturfragen, EDMZ
Nr. 724.237 d, Bern.
Nussbaumer, T., Neuenschwander, P., Ph. Hasler, Jenni, A.,
B u ehler, R., 1998. Energy from Wood. Swiss Federal Oce
of Energy, 3000 Bern, Switzerland.
Oehme, M., M u ller, M., 1995. Levels and congener pattern of
polychlorinated dibenzo-p-dioxins and dibenzofurans in
solid residues from wood ®red boilers in¯uence of combus-
tion conditions and fuel type. Chemosphere 30 (8), 1527±
1539.
Oehme, M. (Ed.), 1998. Handbuch Dioxine. Spektrum Akade-
mischer Verlag, Heidelberg, Berlin.
Pohlandt, K., Marutzky, R., 1994. Concentration and distri-
bution of polychlorinated dibenzo-p-dioxins (PCDD) and
polychlorinated dibenzofurans (PCDF) in wood ash.
Chemosphere 28 (7), 1311±1314.
Acknowledgements
We gratefully acknowledge Charles Macmillan (Ge-
neva) for the improvement of the English and Peter
Schmid (EMPA, D u bendorf) for his scienti®c advices
and proof-reading.
Safe, S., 1992. Development validation and limitations of toxic
equivalency factors. Chemosphere 25, 61±64.
Sommer, K., 1979. Probenahme von K o rnigen Masseng u tern.
Probenahme, DIN 53'803, Teil 1 bis 4. Sampling procedures

S. Wunderli et al. / Chemosphere 40 (2000) 641±649
649
for inspection by attributes ISO259, Parts 1±3 Springer,
Berlin.
¯y and bottom ash from waste wood and natural wood
burned in small to medium sized wood ®ring facilities in
Switzerland. Organohalogen Compd. 27, 231±236.
VHe (Vereinigung f u r Holzenergie), 1991. Mit Holz heizen/
modern und umweltgerecht, Z u rich.
Wunderli, S., Zennegg, M., Dolezal, I.S., Noger, D., Hasler,
Ph., 1996. Levels and congener pattern of PCDD/PCDF in