You are on page 1of 11

Environmental Pollution 137 (2005) 305e315

www.elsevier.com/locate/envpol

Sampling technologies and air pollution control


devices for gaseous and particulate arsenic: A review
Lieve Helsen *
Department of Mechanical Engineering, Division of Applied Mechanics and Energy Conversion,
Katholieke Universiteit Leuven, Celestijnenlaan 300A, 3001 Heverlee, Belgium
Received 7 September 2004; accepted 28 January 2005

Submicron arsenic fumes are dicult to control in conventional air pollution control devices.
Abstract
Direct measurement of arsenic release requires a good sampling and analysis procedure in order to capture and detect the total
amount of metals emitted. The literature is extensively reviewed in order to evaluate the eciency of full eld-scale and laboratory
scale techniques for capturing particulate and gaseous emissions of arsenic from the thermo-chemical treatment of dierent sources
of arsenic. Furthermore, trace arsenic concentrations in ambient air, national standard sampling methods and arsenic analysis
methods are considered. Besides sampling techniques, the use of sorbents is also reviewed with respect to both approaches (1) to
prevent the metals from exiting with the ue gas and (2) to react or combine with the metals in order to be collected in air pollution
control systems. The most important conclusion is that submicron arsenic fumes are dicult to control in conventional air pollution
control devices. Complete capture of the arsenic species requires a combination of particle control and vapour control devices.
2005 Elsevier Ltd. All rights reserved.
Keywords: Arsenic; Sampling; Air pollution control device; Gaseous emissions; Particulate emissions; Literature review

1. Introduction
Since thermal processes that use arsenic containing
sources inherently lead to volatilisation of arsenic,
appropriate arsenic capturing devices have to be installed or arsenic capturing methods have to be applied.
These devices are said to be commercially available, but
very few successful tests have been carried out on
industrial scale for the specic case of thermal conversion of chromated copper arsenate (CCA)-treated wood.
In the frame of a study that aims at identifying the best
available thermochemical conversion technology (pyrolysis, gasication or (co-)combustion) for CCA impregnated wood waste (Helsen and Van den Bulck, 2005),
the literature is critically reviewed with respect to the
* Tel.: C32 16 322505; fax: C32 16 322985.
E-mail address: lieve.helsen@mech.kuleuven.ac.be.
0269-7491/$ - see front matter 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envpol.2005.01.025

available sampling methods and air pollution control


devices for gaseous and particulate arsenic.
Pyrolysis of wood is known to lead to very specic
submicron aerosol fumes. Standard methods for emission sampling were found to be not appropriate for
pyrolysis euents (Lachenal et al., 1989). Rapid cooling
of pyrolysis gases (heavy tar-like hydrocarbons) quickly
clogs the sampling line. Moreover, most experimental
studies with CCA-treated wood report arsenic mass
balances that are far below 100% (Pasek and McIntyre,
1993; Hirata et al., 1993; Tauw Infra Consult, 1987;
Lindroos, 1999). These low mass balances are suspected
to be due to incomplete sampling of the metals released.
Since the information available with respect to
sampling of arsenic released in a thermal treatment
plant for CCA-treated wood is rather limited, industrial
experience with other feedstock can be helpful in the
design of an appropriate arsenic sampling device or the

306

L. Helsen / Environmental Pollution 137 (2005) 305e315

application of an appropriate arsenic capturing method.


Coal combustion, waste incineration and metallurgical
operations generate hot ue gases containing As.
Therefore, the literature is critically reviewed with
respect to all sampling procedures and air pollution
control devices for gaseous and particulate arsenic,
regardless of the original feedstock used. It should be
kept in mind that CCA-treated wood contains a thousand times more arsenic than for example coal does and
thermal conversion of wood may give rise to sticky tar
compounds. The detailed description of the mechanisms
responsible for arsenic release during thermal conversion of arsenic containing feedstock falls beyond the
scope of the present paper, but can be found elsewhere
(Helsen et al., 2003a).
When discussing arsenic sampling technologies and
air pollution control devices, the measurement of trace
arsenic concentrations in ambient air, national standard
sampling methods and arsenic analysis methods should
also be looked at since they often appear in national or
international emission legislation aimed at controlling
the air quality.
This paper is organised as follows: in the next section
an extensive literature review concerning sampling
methods and air pollution control devices for gaseous
and particulate arsenic is given, while in the nal section
the most important conclusions are summarised.

2. A critical literature review


Direct measurement of arsenic release requires a good
sampling and analysis procedure in order to capture and
analyse the total amount of metals emitted. Therefore, the
literature is extensively reviewed with respect to sampling
techniques for gaseous and particulate arsenic, as well as
capturing methods for arsenic. This literature review is
divided into ve subsections: CCA-treated wood, other
feedstock, trace arsenic concentrations in ambient air,
arsenic analysis, and national standard methods.
2.1. CCA-treated wood
Thermal conversion of CCA-treated wood leads to
percentages of arsenic volatilised ranging between 8%
and 95% of the arsenic content in the wood waste,
depending on the reactor temperature, the residence time
at a given temperature, the air ow rate, the oxygen partial
pressure, the pressure, the extended period of ash heating,
the chlorine and sulfur content and the impregnation
process (Dobbs et al., 1978; Wilkins and Murray, 1980;
Kramer et al., 1985; McMahon et al., 1985; Tauw Infra
Consult, 1987; Marutzky, 1990; Hirata et al., 1993; Pasek
and McIntyre, 1993; Kitamura and Katayama, 2000;
Syrjanen and Kangas, 2001; Hata et al., 2003; Helsen and
Van den Bulck, 2003).

Combustion trials (Lindroos, 1999) performed with


CCA-treated wood in an existing incineration plant,
where the normal gas cleaning system is extended with
a venturi scrubber, show that the arsenic emission is too
high (2.8 mg Nm3). Furthermore, the use of electrostatic precipitators (ESP) does not suce to keep arsenic
emission below the European target value of 0.5 mg
Nm3 (for the sum AsCCrCCuCSbCPbCCoCMnC
Ni) (Lindroos, 2002). It was concluded that a special
design is needed for the gas cleaning, usually a combination of wet and dry methods. The latter have the
disadvantage that as far as occupational hygiene is
concerned, the treatment of dry impure arsenic trioxide
dust is dicult. Scrubbing of the gases is said (Syrjanen,
1999) to be of special importance since most of the
arsenic (60e90%) is released as very small particles.
However, venturi scrubbers are generally not ecient in
capturing small particles (1 mm in diameter or less)
(Barton et al., 1992).
Wu et al. (Wu and Biswas, 1993; Wu et al., 2000;
Wu and Barton, 2001; Iida et al., 2004) proposed the
introduction of mineral sorbents in the combustion
system as a solution for both the problem of metal
emission and the problem of metal leaching from the
combustion ash. Equilibrium analyses (Wu and Barton,
2001; Iida et al., 2004) were conducted to predict potential
sorbent materials for chemically adsorbing the heavy
metals during the combustion of CCA wood. Alkaline
earth metals and alkali metals were found to be eective
for capturing arsenic by forming metal-sorbent compounds. However, sulfur and chlorine present in the
combustion system may interfere with the performance of
the sorbents due to the formation of chlorides and sulfates
that deplete available sorbent for the metals. Ca(OH)2
and Na2CO3 were then examined in an experimental
combustion system for their performance (Iida et al.,
2004). The addition of these sorbents signicantly
enhanced the retention of the metals in the ash, especially
for arsenic. Regarding the leaching Ca(OH)2 was shown
to be very eective, except for chromium, while Na2CO3
resulted in very high leaching levels of all metals. The
changes in retention and leaching are due to the formation
of new metal-sorbent compounds.
A lot of combustion experiments with CCA-treated
wood were performed on lab-scale, the resulting mass
balances for arsenic never reaching 100%. Five of these
studies, each using dierent sampling systems, are listed in
Table 1. Helsen et al. (2003b) developed a sampling and
analysis method for arsenic emissions during pyrolysis of
arsenic containing wood on lab-scale. Validation was
carried out through tube furnace experiments using both
CCA-treated wood and arsenic trioxide powder as
input. For the working conditions considered (pyrolysis
at 350  C for 20 min with a nitrogen ow rate of
100 Nl h1) the combination of a quartz cooling tube
and a cellulose ester membrane lter impregnated with

L. Helsen / Environmental Pollution 137 (2005) 305e315


Table 1
Overview of combustion experiments with CCA-treated wood
performed on lab-scale, with their corresponding ue gas cleaning
equipment and resulting arsenic mass balance
Reference

Flue gas cleaning equipment

As mass
balance

Tauw Infra
Consult, 1987

MulticycloneCcellulose
nitrate lter

!50%

Pasek and
McIntyre, 1993

Two bubblers:

20e90%

emptyC2% NaOH
or glass beadsC
2% monoethanolamine
Corneld et al.,
1993

Glass bre lterC5 washing


bottles
H2OeNaOHeNaOHeHCleHCl

NA

Hirata et al.,
1993

Three spiral traps, chilled


with ice water (0  C),
dry ice-methanol (60  C) and
liquid N2 (190  C)

50e80%

McMahon
et al., 1985

glass transfer tubeCglass


bre lterC
TBAH impregnated
cellulose lter

61e104%

NA, As mass balance is not available (non-signicant amount


captured in washing bottles and the metals retained were all in
particulate form); TBAH, tetra-n-butylammonium hydroxide.

a Na2CO3eglycerol solution was sucient to capture the


arsenic. Two extra impingers (the rst one containing 50
ml HNO3 1 M in the case of CCA-treated wood and 50
ml NaOH 0.1 M in the case of As2O3 and the second one
containing 50 ml NaOH 0.1 M) were added downstream
of the lter section as backup either in case of lter
failure or to check whether all the arsenic released is
captured by the cooling tube and lter. The arsenic
distribution over the dierent units of the sampling train
was determined by inductively coupled plasma mass
spectrometry (ICP-MS), after dissolving the lters and
backup pads in HNO3 and H2O2.
Characterisation of the resultant combustion ash or
charcoal residue has been performed by Helsen and
Van den Bulck (1998), Hata et al. (2003) and Iida et al.
(2004) by scanning electron microscopy coupled to energy
dispersive X-ray analysis (SEM-EDX), transmission
electron microscopy (TEM) or X-ray diraction (XRD).
Both Cu arsenates and Cr arsenates have been identied.
It can be concluded that additional investments are
needed for better gas cleaning systems, tuned in to the
type of burner, gasier or pyrolyser, and for measurements to control emissions.
2.2. Other feedstock
2.2.1. Arsenic oxide
In commercial installations particulate matter is often
separated at an elevated temperature (circa 200  C) at

307

which other impurities, such as arsenic oxide, are


appreciably volatile. Wouterlood and Bowling (1979)
studied the removal of arsenic oxide vapour from hot ue
gases on the laboratory scale, by passing the gas stream
through a bed of dierent adsorbent particles. Physical
methods seem attractive because arsenic oxide is a known
catalyst poison, which implies that it is strongly and
selectively adsorbed on particulate surfaces. Inert materials such as glass beads, glass wool, and y ash were
found to be virtually ineective; basic and acid chemicals
worked to a limited extend. Surface-active agents such as
activated carbons, silica gel, and molecular sieve 13!
adsorbed arsenic oxide vapour strongly.
In an attempt to capture the particulate and volatile
arsenic compounds at the same time, Walsh et al. (1977)
tested an impregnated lter sampling system. An
aqueous glycerol and polyethylenimine polymer solution
was used to impregnate Whatman lters. Laboratory
studies showed that for volatile arsenic trioxide the total
collection eciency of three impregnated lters in line
averages approximately 96% at ow rates from 15 to 23
l min1. Preliminary test results indicated that tetra-nbutylammonium hydroxide (TBAH) is apparently a very
ecient collection substrate for arsenic trioxide vapours.
The eciency of lter sampling for arsenic in the atmosphere has also been studied by Appel et al. (1984) by
comparing glass bre lters, Teon lters and cellulose
lters impregnated with TBAH. Only the impregnated
lters provided measurable As levels, but still not a 100%
collection.
While multiple species of inorganic arsenic (particulate form, arsenic trioxide vapour and arsine) coexist
in industrial environments, traditional air sampling
methods quantify only total inorganic arsenic. The
National Institute for Occupational Safety and Health
(NIOSH) conducted laboratory experiments and eld
studies to develop a technique to separate mixtures of
particulate inorganic arsenic compounds and arsine, and
to detect and quantify arsenic trioxide vapour (Costello
et al., 1983). This study concluded that in the presence of
heated arsenic sources, such as battery manufacturing
plants, monitoring solely with conventional (untreated)
lters will underestimate arsenic exposure because
arsenic trioxide vapour is not eciently collected.
Furthermore, the individual particles collected on the
lters were spherically shaped with a median circular
area equivalent diameter less than 1 mm. Arsenic was not
detectable in these particles, indicating that the arsenic
concentration in the particles was less than 1% of the
total mass.
Demange et al. (1992) studied the eciency of several
lters and concluded that dierent methods perform
well as long as impregnated lters are used. They
developed a sampling train consisting of a prelter to
retain particulates and As2O3 vapour, followed by two
quartz bre lters to trap arsine (Demange et al., 2000).

308

L. Helsen / Environmental Pollution 137 (2005) 305e315

The prelter was a quartz bre lter impregnated with


sodium carbonate. The successive quartz bre lters
were impregnated with silver nitrate. The lters were
dissolved in nitric acid, prior to analysis by ICP atomic
emission spectrometry (ICP-AES).
2.2.2. Other arsenic containing wastes
On an industrial scale arsenic containing wastes are
incinerated in a rotary kiln incinerator (for example at
the waste treatment company INDAVER in Belgium).
The gas cleaning equipment consists of a combination of
a boiler (for heat recovery), an electrostatic precipitator,
a four-steps wet scrubber (using lime and sodium
hydroxide) and an injection of activated carbon or
brown coal. The percentages of arsenic retained in the
dierent units are presented in Table 2 (Wauters, 1998).
From Table 2 it is clear that an ecient lter
(electrostatic precipitator) does not succeed in capturing
all the arsenic. Circa 5.4% of the As originally present in
the waste passes the electrostatic lter and is captured in
the wet scrubber by absorption and/or chemisorption.
It is not known, however, whether the percentages of
arsenic retained in the dierent units would be the same
or totally dierent if the As content in the waste stream
is much higher, for example when CCA-treated wood is
used as input.
An arsenic closure problem is often seen in medical
waste incinerators, fed with arsenic containing waste in
the form of operating and delivery room materials such
as cloth, blue plastic and rubber items and sterilisation
wrap. A study conducted in a rotary kiln incinerator by
the U.S. EPA showed the total mass balance of arsenic
ranging from 38% to 73% of the input arsenic material
(Chesworth et al., 1994). Chesworth et al. (1994) used
a co-ow laminar diusion ame system to study the
fate of arsenic and found that the arsenic species, predominantly in aerosol form, are captured by a Gore-tex
lter, which may act as a condensation surface for
species during sampling.
Hinshaw (1994) studied the behaviour and control of
metals in a hazardous waste incinerator using synthetic
wastes spiked with for example concentrated As2O3
aqueous solution. By using an air pollution control
system, consisting of a spray dryer, a cooler/condenser,
a venturi scrubber, a bae absorber and a mist eliminator, the metal stack emission rate was highly dependent
on the dierential pressure across the venturi scrubber,
but less dependent on the chlorine feed concentration.

However, the behaviour and control of the metals in the


gases entering the air pollution control system was highly
sensitive to the chlorine level.
Venkatesh et al. (1996) evaluated various mineral
sorbents constituting a spectrum of alumino-silicate
compounds and attapulgite clay (a magnesium hydroxide
silicate compound) for retention and immobilisation
of arsenic and other heavy metals. A mixture of the
mineral sorbent, PVC powder (to examine the inuence of
chlorine) and metals spike (200 mg As/kg, added as
arsenic oxide) was incinerated and the resultant solid
product was analysed for metal retention and leachability. Combining the dual criteria of limiting metal vaporisation and reducing leachability, and the additional
criteria that the toxicity characteristic leaching procedure
(TCLP) extract metal concentrations be below regulatory
levels, kaolin and attapulgite clay appear to have the most
potential as sorbents.
2.2.3. Coal
Coal combustion is estimated to account for 2e5%
of the total arsenic emissions from anthropogenic
sources globally (Sterling and Helble, 2003). The
vaporised arsenic will ultimately partition between the
vapour phase and the particulate y ash, which leads to
the enrichment of arsenic in the smallest particles. Aside
from its potential to be a hazardous air pollutant if
emitted to the atmosphere, arsenic, in the form of
gaseous arsenic oxide (As2O3), is poisonous for selective
catalytic reduction (SCR) catalyst; therefore, its presence is detrimental to the operation of SCR units for
NOx control. As2O3 solidies on both active and nonactive sites in the vanadium-based catalysts that are used
in SCRs, which reduces the activity of the catalyst.
Electrostatic precipitators (ESPs) are often used to
capture the y ash in the ue gas. However, the ESP is
usually installed after the SCR unit and therefore does
not oer a solution for the catalyst poisoning problem.
The 80% capture rule-of-thumb, which assumes that
each row of the ESP collects circa 80% of the y ash in
the ue gas entering the row, is a common assumption in
the utility industry. The actual collection eciency will
vary according to many variables, such as combustion
conditions, and ash amount and type. Sakulpitakphon
et al. (2003) found that the y ash arsenic content seems
to be within the range of the feed coal arsenic content,
which suggests that most of the arsenic is captured on y
ash. The arsenic concentration in the y ash increases

Table 2
Arsenic (%) retained in the dierent parts of the INDAVER rotary kiln incineration plant (Wauters, 1998)

Collected in
Temperature (  C)
As (wt%)

Bottom ash

Boiler ash

Fly ash

Sludge

Exhaust gas

Rotary kiln
1120
14.6

Boiler
280e240
5

Electrolter
260
75

Scrubber
250e60
5.4

!60
0.02

309

L. Helsen / Environmental Pollution 137 (2005) 305e315

markedly with decreasing particle size (Davison et al.,


1974; Maenhout et al., 1993). This dependency on
particle size is explained by a mechanism involving high
temperature volatilisation followed by preferential
condensation or adsorption onto the smallest particles.
Since thermodynamic equilibrium calculations do not
take into account the formation of small particles
(dicult to collect fume), nor kinetic eects, maximum
theoretical removals are calculated (Frandsen et al.,
1997; Blander et al., 1998; Sandelin and Backman,
2001). However, in a system of multiple metals,
thermodynamic equilibrium analysis may identify the
dominant forms of the metallic and radical species,
which can provide useful information in the development of more accurate kinetic models (Wu and Biswas,
1993).
Sorbent injection is a very attractive method to reduce
arsenic emission during coal combustion (Gullett and
Raghunathan, 1994; Jadhav and Fan, 2001; Wu and
Barton, 2001; Sterling and Helble, 2003). The most
important sorbents studied with respect to their eectiveness to reduce arsenic emissions during coal combustion are summarised in Table 3. Arsenic reacts, while
still in the vapour state, at high combustion temperatures, with various sorbents to form larger particles that
can be collected eectively by particulate collection
devices. The sequestering action of the sorbents reduces
the vapour form and/or ne particle form of the metal
(Gullett and Raghunathan, 1994). These sorbents can be
y ash, activated carbon or mineral material. Hydrated
lime (Ca(OH)2) and limestone (CaCO3) are found to be
very eective. While Ca is responsible for the reaction of
As with these solids, it is the availability of active Ca sites
at the surface of these solids that determines the rate of
reaction (Sterling and Helble, 2003). At temperatures
below 600  C tricalcium orthoarsenate (Ca3As2O8) is
formed, while temperatures between 700 and 900  C
give rise to the formation of dicalcium pyroarsenate
(Ca2As2O7), which is unstable and therefore responsible
for a decrease in As capture at higher temperature (Jadhav
and Fan, 2001). Sterling and Helble (2003), however,
reported a maximum capture of As with calcium oxide at
1000  C. They concluded that any calcium present in y
ash can react with arsenic oxide vapour and capture the
metal in water-insoluble forms of the less hazardous
As(V) oxidation state. Seames and Wendt (2000) suggest
the formation of AseCa reaction products in supermicron particles in the post-combustion zone. Cocombustion of municipal sewage sludge with coal allows
the arsenic to be reactively scavenged by calcium (Seames
et al., 2002). Coal alone contained insucient calcium to
accomplish this same scavenging eect.
2.2.4. Copper ores
In the metallurgic industry, the purication of copper
ores also results in the volatilisation of arsenic. Since

Table 3
Sorbents studied with respect to their eectiveness to reduce arsenic
emissions during coal combustion
Reference

Sorbent

Comments

Sterling and
Helble, 2003

Ca compounds
in y ash:
CaO

CaO is most effective


(T: 600e1000  C)
Reaction product is Ca
arsenate
Ca oxide sulfation
is a problem

2CaO$SiO2
CaO$SiO2
Jadhav and
Fan, 2001

Activated carbon

Fly ash
Minerals
Silica

Alumina

Kaolinite

Minerals: lower
emission and products
of lower leachability
interaction CaOCAs2O3
depends on T
!600  C: amount captured
increases with T and
Ca3As2O8 is formed
700e800  C: amount
captured decreases with
T and Ca2As2O7 is formed
No signicant
effect of SO2 or HCl

Emathlite
Lime
Soda ash
Wu and
Barton, 2001

Al-, Fe-, Li-,


Na-, K-, Be-,
Mg-, Ca-based
sorbents

Afnity to adsorb As
(500e1750 K): LiOKO
NaOAlOCaOMgO
FeOBe
Effectiveness may be reduced
when Cl or S is present
Products with
high alumina content: low
leaching
high silica content: high
leaching

Gullett and
Raghunathan,
1994

Hydrated lime

As capture is function of
sorbent type and injection
mode (1000e1150e1300  C):
kaoliniteOlimestoneO
hydrated lime
Not a simple condensation
mechanism, the particles
external surface area or
pore diffusion is important
to the capture mechanism

Limestone
Kaolinite

Bauxite

the aim of this purication process is to eliminate as


much arsenic as possible to obtain a pure copper
product, the release of arsenic to the gas phase (and
subsequent capture) is highly favourable. The company
Umicore (Olen, Belgium), for example, makes use of
a gas cleaning system consisting of a combination of a
boiler (for energy recovery), injection of lime, injection
of activated carbon or brown coal, dilution with air until
the nal temperature has reached 95  C and a baghouse
lter. The arsenic distribution over the dierent units is
not known, but the resulting arsenic concentration in

310

L. Helsen / Environmental Pollution 137 (2005) 305e315

the exhaust gas is around 0.47 mg Nm3, which is below


the limit specied in the emission legislation (1 mg
Nm3). This low concentration is reached by the
addition of lime, the presence of a thin dust layer on
the lter and mixing the gas stream with air, resulting in
a dilution. However, the absolute amount of arsenic
emitted into the atmosphere is signicant (295.6 kg
year1) (Geenen, 1999). The amount of arsenic collected
in the dust reached 745 kg in 1998 (VMM, 1998).
Chemical analysis of copper smelter ue dust reveals the
presence of arsenite (As(III)), arsenate (As(V)), arsenic
sulde and As2O3 (Eatough et al., 1979).
In metallurgical processes arsenic disposal is shifted
from formation of calcium arsenate to formation of
insoluble ferric arsenate compounds. Lime precipitation,
which was widely used in the past, is being abandoned
as a result of strong evidence showing that calcium
arsenate compounds decompose slowly in contact with
atmospheric CO2 to form calcium carbonate and soluble
arsenic acid. Scorodite (FeAsO4$2H2O) has several
advantages over arsenical ferrihydrite as a disposal
compound including a lower iron demand, a higher
density and a greater thermodynamic stability (Riveros
et al., 2001).
At the copper line of Outokumpu Harjavalta Metals
(Pori, Finland), which is the major copper producer in
Finland and manufacturer of CCA preservatives, a trial
has been carried out to recycle CCA-treated timber
in the copper smelting process. The energy content of
the wood was used in the ash smelting process, copper
and arsenic were recovered in the CCA preservative
manufacturing plant, burning gases were fed to the
sulphuric acid plant and chromium is subjected to a slag
treatment process. How the arsenic was recovered is not
specied (Nurmi and Lindroos, 1994). A second trial
investigates heavy metal emission during gasication of
CCA-treated wood and subsequent combustion of the
synthesis gas for district heating (Nurmi, 1996). The
arsenic emission varied between 1120 and 1640 mg h1,
corresponding to 63e92 g of As for a 56 h run, being at
maximum 0.1% of the arsenic originally present in the
wood. The amount of arsenic collected in the condensing water and the demister were respectively 14.2 kg and
30 g, while 8.8 kg of As remained in the ash. However,
the arsenic mass balance is less than 30%, illustrating
again that accurate sampling of arsenic is not an
easy job.
2.3. Trace arsenic concentrations in ambient air
A variety of sampling and analytical techniques
have been used to measure trace metal concentrations in
air including impingers, electrostatic precipitators and
lters. Some sampling techniques used to measure trace
metal (in particular As) concentrations in ambient air
are listed in Table 4, together with their eciencies.

Filter eciencies seem to decrease if lower arsenic


concentrations have to be measured.
Smith (1979) derived an equation to calculate the
overall eciency of a sampling train consisting of a series
of impingers. He concluded that the amount caught by
the last impinger should be close to zero to obtain a high
overall eciency. The presence of sample in the last
impinger indicates that some unknown amount has
passed through the train entirely. Smith emphasised
the need to seek the sampling conditions and train length
that will produce a nil indication in the last impinger. Li
and Pupp (1979) investigated the eciency of water as
a collection medium for arsenic oxide vapour. They
concluded that water and dilute (2%) sodium hydroxide
are equally eective collection media for arsenic trioxide
vapour. Three Greenburg-Smith impingers, each lled
with 250 ml of water, cooled with ice, are satisfactory
for the collection of arsenic trioxide vapour. No vapour
was lost from such a sampling train at ow rates up to
20 l min1. However, up to 15% of the collected arsenic
may be absorbed on the walls of the impinger train;
this amount will not dissolve in water unless the train
is thoroughly washed with sodium hydroxide.
Over recent years, lters have been favoured over
other measurement methods due to their high eciency
for collection of small particles (Maggs, 2000). Dierent
lter media have been considered for the determination
of trace metals in air, such as glass bre, quartz bre,
membrane lters and Teon lters. An important
consideration in selecting the most appropriate lter
medium for collection of trace elements in particulate
matter is the background metal concentration of the
lter material (blank value). The blank values of arsenic
in the dierent lter media are given in Table 5, together
with some other properties.
Impregnated lters are found to be able to capture
simultaneously both particulate arsenic and arsenic
oxide vapour. The latter is collected by reaction with
the chemical agent on the impregnated lter, e.g. for
a sodium carbonate impregnated backup pad (Maggs,
2000) the following stoechiometric reaction takes place:
As2 O3 CNa2 CO3 /2NaAsO2 CCO2

2.4. National standard methods


Besides the American NIOSH standards (NIOSH,
1984, 1985) and the MDHS 41/2 (MDHS, 1995) for the
sampling of arsenic in air (see Table 4), a German
standard VDI 3868 (VDI, 1994) and a Dutch draft NVN
2817 (NNI, 1996) have been published. The German
standard has been developed to sample the total amount
of particulates and lter passing matter (metals, metalloids and their compounds), while the Dutch draft
describes a sampling and analysis method for the

311

L. Helsen / Environmental Pollution 137 (2005) 305e315


Table 4
Overview of sampling techniques used to measure trace metal (in particular As) concentrations in ambient air
Reference

Metal compound

Sampling system

Eciency

Roberts et al., 1977

As
concO1 mg m3

Glass bre lter


Glass bre lter
Nuclepore
AA Millipore

17e65%
46.7e97%
99.2e99.8%
98.2e99.9%
Lower

conc!1 mg m3
Lao et al., 1974

As2O3

Glass bre lter


FilterCGreenburgeSmith impinger

Low (10e24%)
Higher

Crecelius and
Sanders, 1980

part. AsCAs2O3 vapourCAsH3

Dry impingerCTeon lter


C2 impingers (NaOH, H2O, HNO3 or NH4OH)

High for part. As


O97% for As2O3
!1% for AsH3
Low for As2O3
(44C29)% AsH3
Backup for AsH3
Effective for As2O3

C2 silvered quartz bead traps


C4 stage charcoal trap
Maggs, 2000

part. AsCinorg.
As As2O3 vapour

Cellulose ester membrane lter


CNa2CO3 impregnated backup pad

NA

MDHS, 1995

part. AsCinorg. As
As2O3

Cellulose ester membrane lter


CNa2CO3 impregnated backup pad

NA
Complies EN 482

NIOSH, 1984

part AsCAs2O3

Na2CO3 impregnated membrane


lterCbackup pad

93%

NIOSH, 1985

AsH3

Solid sorbent tube with activated


coconut shell charcoal

89%

NA, eciency is not available; MDHS, Methods for the Determination of Hazardous Substances.

determination of the content of the elements arsenic,


cadmium, chromium, copper, mercury, lead, zinc and
their alloys, in particulate matter and in the gas phase
(sampling fume, process and exhaust gases). Both
methods extract a part of the gas volume isokinetically
from the exhaust stream and force this part through a
two-stage device, comprising a lter system that retains
the particulate matter and an absorption system in the
form of washing bottles. At least three gas washing
bottles are connected in series and lled with suitable
absorbent solutions. Dierent (untreated) lter materials and dierent absorbent solutions can be used. Metal
Table 5
Properties and blank values of arsenic (ng cm2) in the dierent lter
media (Maggs, 2000)
Filter type

Blank value Properties

Glass bre

40e60

Borosilicate glass, blank values can be


variable, low ow resistance

Quartz bre 0.5e5

Quartz glass, lower blank values for many


elements compared to glass bre lters,
similar ow resistance to glass bre lters,
total digestion is easier than glass bre
lters, relatively expensive

Membrane

0.1e4

Cellulose acetate or cellulose nitrate,


low blank values, high ow resistance
(increased under high humidity
conditions), prone to electrostatic charge

Teon

0.3

Very low blank values, high ow


resistance, relatively expensive

analysis is performed by using hydride generation


atomic absorption spectrometry (AAS) or oven AAS
in the case of VDI and AAS or inductively coupled
plasma atomic emission spectrometry (ICP-AES) in the
case of NVN. As a control the last washing bottle is
analysed separately; the amount of arsenic found in this
bottle should be less than 10% of the total mass. Using
the VDI standard there have been made positive
experiences with a series of waste incinerators and
furnaces using fossil fuels.
The U.S. EPA Reference Methods (EPA, 1993) have
been set up to provide industry with a reasonable and
reliable means of demonstrating compliance with applicable regulations. Among them are U.S. EPA Method 108
for the determination of particulate and gaseous arsenic
emissions and U.S. EPA Method 108A for the determination of arsenic in ore samples from non-ferrous
smelters. According to the U.S. EPA Method 108 particulate and gaseous arsenic emissions are withdrawn
isokinetically from the source and collected on a glass
mat lter and in a water sampling train, consisting of four
impingers containing respectively water, water, empty
and water absorption media.
Furthermore, the Intersociety Committee Methods
of Air Sampling and Analysis (Lodge, 1988) published
a lot of methods for ambient air sampling and methods
for chemicals in air of the workplace and in biological
samples. With respect to arsenic sampling they provide
the following methods: Method 302 Determination of
As content of atmospheric particulate matter using

312

L. Helsen / Environmental Pollution 137 (2005) 305e315

membrane or glass bre lters, and Method 804 As, Se


and Sb in urine and air by hydride generation and AAS
using membrane lters for the air samples.
The International Standards Organisation (ISO) is
engaged in the development of ISO 15202, which will
describe a generic method for the determination of
metals and metalloids in airborne particulate matter by
ICP-AES (Butler and Howe, 1999). However, an
international standard method for the determination
of gaseous and/or particulate arsenic in emitted gas/
aerosol streams has not yet been developed.

2.5. Arsenic analysis


In dust analysis, the analytical step is generally
preceded by an appropriate preparation of the sample
in order to dissolve the collected particulate matter or
extract the metals to be analysed. The dissolution can be
achieved by wet digestion (ultrasonic agitation, hotplate digestion or microwave assisted) or dry ashing.
There is risk of losing part of the metal and/or
contaminating the sample solution if the number of
intermediary stages in the extraction process increases.
Alternative sample introduction systems could be: laser
ablation (LA) or electrothermal vaporisation (ETV).
After dissolution, analysis is carried out using a number
of dierent techniques: ion chromatography (IC),
ultraviolet spectrometry (UV), electrochemical methods
(EQ) such as dierential pulse polarography, atomic
absorption spectrometry (AAS), inductively coupled
plasma mass spectrometry (ICP-MS), inductively coupled plasma atomic emission spectrometry (ICP-AES),
laser plasma atomic emission spectrometry (LP-AES)
and X-ray uorescence spectrometry (XRF). Both the
minimum detection limit (determined as that concentration which produces a signal equivalent to 2 or 3
times the standard deviation of the blank signal for
aqueous solutions) and the relative error (based on the
analysis of standards) are important characteristics of
the analytical technique.
Analysis of the solid product resulting from the
thermal process (charcoal or ash) is usually performed
by one of the following methods: X-ray diraction
(XRD), electron spectroscopy for chemical analysis
(ESCA), scanning electron microscopy coupled to energy
dispersive X-ray analysis (SEM-EDX), transmission
electron microscopy (TEM), Fourier transform infrared
spectroscopy (FTIR), electron spin resonance spectroscopy (ESR), particle induced X-ray emission (PIXE),
electron induced X-ray emission (EIXE) or neutron
activation analysis (NAA).
A detailed description and comparison of these
analytical techniques is beyond the scope of the present
paper. Good reviews have been made by Burguera and
Burguera (1997) and Vanhaecke and Moens (1999), who

described trends in trace element determination and


speciation. Since trivalent arsenic is considerably more
toxic than pentavalent arsenic, speciation studies are
very important.

3. Conclusions
From the literature review the following conclusions
are drawn:
1. A simple cooling and/or condensation process does
not succeed in collecting all the arsenic.
2. Since most of the arsenic is released as very small
particles, venturi scrubbers are generally not ecient in capturing arsenic.
3. Physical methods (adsorption on particulate surfaces) seem attractive but only surface-active agents,
such as activated carbons, silica gel, and molecular
sieves, adsorb arsenic oxide vapour strongly.
4. Filtration through a quartz bre, glass bre or
Teon lter collects the arsenic only partly; for
Nuclepore and Millipore lters, on the other hand,
high lter eciencies were reported; good collection
eciencies were reported for alkali impregnated
(cellulose ester membrane) lters; impregnation
(with an aqueous glycerol-polyethylenimine solution, sodium carbonate or TBAH) seems to be
necessary, resulting in a chemical reaction between
the arsenic compound and the chemical agent on the
lter.
5. The lower the arsenic concentration to be measured
(e.g. low concentrations in the ambient air), the
lower the lter eciency.
6. Also the combination of lters and impingers
(containing acids or alkali) seems to be ecient.
7. Water and dilute sodium hydroxide are equally
eective collection media for arsenic trioxide vapour
in impingers; however, some arsenic compounds
adsorbed to the impinger walls need to be thoroughly washed with sodium hydroxide.
8. In the case of a sampling train consisting of a series
of impingers, the amount caught by the last
impinger should be close to zero to obtain a high
overall eciency.
9. In big chimneys a small part of the total emitted
ow is isokinetically sampled and forced through
a heated lter and at least three impingers that
contain an acid solution.
10. To collect arsine (AsH3) a charcoal tube or a silver
nitrate impregnated lter is used.
11. In coal red power plants ne y ash particles that
are enriched with arsenic are collected in an ESP;
however, ESPs are found to be insucient to capture
all arsenic released when incinerating arsenic containing waste.

L. Helsen / Environmental Pollution 137 (2005) 305e315

12. Sorbent injection combined with ltration is a very


attractive and ecient method to reduce arsenic
emissions during coal combustion and metallurgical
operations; also for the combustion of CCA-treated
wood waste this option is under study now.
13. An international standard procedure for sampling
and analysis of gaseous and/or particulate arsenic
emissions does not exist; only national standards
and drafts, such as VDI 3868, NVN 2817, EPA
method 108 and NIOSH 7901, can be found.
14. A lot of analysis techniques are available to
determine the arsenic content of a sample.
Submicron arsenic fumes are thus dicult to control in
conventional particle control devices such as an
electrostatic lter or a baghouse, since they have their
minimum collection eciencies in the submicrometer
range. Complete capture of the arsenic species requires
a combination of particle control and vapour control
devices, such as a combination of an ecient lter and
a wet scrubber or the injection of a suitable sorbent to
trap arsenic vapours (arsenic oxides or arsine). Only if
the scrubber solution can be reused in the CCA
impregnation process, no additional waste streams are
created. The use of sorbents to capture the arsenic
vapours by physical and/or chemical means is a very
attractive alternative to scrubbers, particularly when the
sorbents are able to capture and bind the metals in the
bottom ash or charcoal, thereby preventing them from
exiting with the ue gas. The sorbents needed for this
scavenging eect can also be delivered by other waste
streams in a co-combustion plant. Another approach is
that the vaporised metals react with the sorbent particles
at elevated temperatures or heterogeneously condense
onto the sorbents as the ue gas cools. The goal of this
approach is to make sorbent particles available in the
ue gas, so that metals can react or condense, and
eectively be collected by air pollution control systems.
Lab-scale experiments have shown that impregnated
lters can collect both particulate arsenic and arsenic
oxide vapours simultaneously. However, these impregnated lter systems seem to be not feasible on an
industrial scale. For the same reason arsenic toxicants,
as recovered in 60  C and 90  C condensates are
impossibly collected in industrial incinerators. Therefore, the combination of lters and scrubbers or the
injection of sorbents seems to be the best available
technique at present.
The gas cleaning system ultimately installed should be
tuned in to the type of burner, gasier or pyrolyser used.
The problem of sticky tar compounds is characteristic to
pyrolysis systems, while gasication and combustion use
temperatures high enough to crack these heavy tar-like
hydrocarbons. Furthermore, a reducing environment
(e.g. in high temperature gasication) may convert the
arsenic in the feedstock completely to metallic arsenic,

313

which is much easier to capture than arsenic trioxide since


metallic arsenic does not go through a liquid phase upon
cooling and has a higher sublimation temperature than
arsenic trioxide (Lagoutte and Garnier, 2001). Since
trivalent arsenic is more toxic and mobile compared to
pentavalent arsenic, speciation studies are very important. From the literature it is known that in combustion y
ash, which is formed in an oxidising environment, arsenic
appears mainly in trivalent form. Since (re-)oxidation to
the pentavalent form only occurs under high pressure,
most of the arsenic released is expected to be in trivalent
state (as As2O3), and thus dicult to capture.

Acknowledgements
L. Helsen is a post-doctoral research fellow of the
Fund for Scientic Research of Flanders (Fonds voor
Wetenschappelijk Onderzoek d Vlaanderen) (Belgium).
The authors are grateful to the company ARCH Timber
Protection Limited (UK) for the nancial support of the
research work.

References
Appel, B.R., Tokiwa, Y., Hoer, E.M., 1984. Eciency of lter
sampling for arsenic in the atmosphere. Atmospheric Environment
18 (1), 219e222.
Barton, R.G., Clark, W.D., Seeker, W.R., 1992. Fate of metals in
waste combustion systems. In: Seeker, W.R., Koshland, C.P.
(Eds.), Incineration of Hazardous Waste: Toxic Combustion ByProducts. Gordon and Breach Science, Switzerland, pp. 327e342.
Blander, M., Milne, T., Dayton, D., Backman, R., Blake, D.,
Kuhnel, W., Linak, W., Nordin, A., Ljung, A., 1998. Equilibrium
chemistry of biomass combustion: A round-robin set of calculations using available computer programs and data bases. In:
Gupta, R., Wall, T.F., Baxter, L.L. (Eds.), Impact of Mineral
Impurities in Solid Fuel Combustion. Kluwer Academic.
Burguera, M., Burguera, J.L., 1997. Analytical methodology for
speciation of arsenic in environmental and biological samples.
Talanta 44, 1581e1604.
Butler, O.T., Howe, A.M., 1999. Development of an international
standard for the determination of metals and metalloids in
workplace air using ICP-AES: evaluation of sample dissolution
procedures through an interlaboratory trial. Journal of Environmental Monitoring 1, 23e32.
Chesworth, S., Yang, G., Chang, D.P.Y., Jones, A.D., Kelly, P.B.,
Kennedy, I.M., 1994. The fate of arsenic in a laminar diusion
ame. Combustion and Flame 98, 259e266.
Corneld, J.A., Vollam, S., Fardell, P., 1993. Recycling and disposal of
timber treated with waterborne copper based preservatives.
Presented at the 24th annual IRG meeting, Orlando, USA,
publication IRG/WP 93-50008.
Costello, R.J., Eller, P.M., Hull, R.D., 1983. Measurement of multiple
inorganic arsenic species. American Industrial Hygiene Association
Journal 44 (1), 21e28.
Crecelius, E.A., Sanders, R.W., 1980. Development of an arsenic
trioxide vapor and arsine sampling train. Analytical Chemistry
52 (8), 1310e1312.

314

L. Helsen / Environmental Pollution 137 (2005) 305e315

Davison, R.L., Natusch, D.F.S., Wallace, J.R., Evans, C.A., 1974.


Trace elements in y ash d dependence of concentration on particle
size. Environmental Science and Technology 8 (13), 1107e1113.
Demange, M., Vien, I., Hecht, G., Hery, M., 1992. Mise au point dune
methode de prele`vement du trioxyde de diarsenic. Cahiers de Notes
Documentaires 146, 63e70.
Demange, M., Elcabache, J.M., Grzebyk, M., Peltier, A., Proust, N.,
Thenot, D., Ducom, P., Fritsch, J., 2000. Phosphine sampling and
analysis using silver nitrate impregnated lters. Journal of
Environmental Monitoring 2, 476e482.
Dobbs, A.J., Phil, D., Grant, C., 1978. The volatilization of
arsenic on burning copper-chromium-arsenic (CCA) treated wood.
Holzforschung 32 (1), 32e35.
Eatough, D.J., Eatough, N.L., Hill, M.W., Mangelson, N.F.,
Ryder, J., Hansen, L.D., 1979. The chemical composition of
smelter ue dusts. Atmospheric Environment 13, 489e506.
EPA (Environmental Protection Agency), 1993. 40 CFR Ch. 1 (7-1-93
Edition), Pt. 61, App. B, Method 108 and 108A, pp. 238e249.
Frandsen, F.J., Helble, J.J., Erickson, T.A., Kuhnel, V., 1997.
Prediction of trace element partitioning in utility boilers, Presented
at the EPA/DOE/EPRI Combined Utility Air Pollutant Control
Symposium, Washington, DC.
Geenen, C., 1999. Umicore, Olen, personal communication.
Gullett, B.K., Raghunathan, K., 1994. Reduction of coal-based metal
emissions by furnace sorbent injection. Energy and Fuels 8 (5),
1068e1076.
Hata, T., Bronsveld, P.M., Vystavel, T., Kooi, B.J., De
Hosson, J.Th.M., Kakitani, T., Otono, A., Imamura, Y., 2003.
Electron microscopic study on pyrolysis of CCA (chromium,
copper and arsenic oxide)-treated wood. Journal of Analytical and
Applied Pyrolysis 68e69, 635e643.
Helsen, L., Van den Bulck, E., 1998. The microdistribution of copper,
chromium and arsenic in CCA treated wood and its pyrolysis
residue using energy dispersive X-ray analysis in scanning electron
microscopy. Holzforschung 52 (6), 607e614.
Helsen, L., Van den Bulck, E., 2003. Metal retention in the residue
after low-temperature pyrolysis of chromated copper arsenate
(CCA) treated wood. Environmental Engineering Science 20 (6),
569e580.
Helsen, L., Van den Bulck, E., 2005. Review of disposal technologies
for chromated copper arsenate (CCA) treated wood waste,
with detailed analyses of thermochemical conversion processes.
Environmental Pollution 134, 301e314.
Helsen, L., Van den Bulck, E., Van Bael, M.K., Mullens, J., 2003a.
Arsenic release during pyrolysis of CCA treated wood waste:
current state of knowledge. Journal of Analytical and Applied
Pyrolysis 68e69, 613e633.
Helsen, L., Van den Bulck, E., Cooreman, H., Vandecasteele, C.,
2003b. Development of a sampling train for arsenic in pyrolysis
vapours resulting from pyrolysis of arsenic containing wood waste.
Journal of Environmental Monitoring 5, 758e765.
Hinshaw, G.D., 1994. Behavior and control of metals in a hazardous
waste incinerator. Hazardous Waste and Hazardous Materials
11 (1), 93e109.
Hirata, T., Inoue, M., Fukui, Y., 1993. Pyrolysis and combustion
toxicity of wood treated with CCA. Wood Science and Technology
27, 35e47.
Iida, K., Pierman, J., Tolaymat, T., Townsend, T., Wu, C.-Y., 2004.
Control of chromated copper arsenate wood incineration air
emissions and ash leaching using sorbent technology. Journal of
Environmental Engineering 130 (2), 184e192.
Jadhav, R.A., Fan, L.S., 2001. Capture of gas-phase arsenic oxide by
lime: kinetic and mechanistic studies. Environmental Science and
Technology 35 (4), 794e799.
Kitamura, T., Katayama, H., 2000. Behavior of copper, chromium
and arsenic during carbonization of CCA treated wood. Mokuzai
Gakkaishi 46 (6), 587e595.

Kramer, J.E., Lustenhouwer, J.W.A., Van Weenen, J.C.,


Brinkkemper, M.A.C., 1985. Gebruik van afvalhout, Ministerie
van Volksverhuizing, Ruimtelijke Ordening en Milieubeheer,
Leidschendam, Nederland, Nr. 17.
Lachenal, J., Toledo, J.M., Mezerette, C., Vergnet, A.M., 1989.
Development and construction of a sampling line for wood
pyrolysis emissions. Proceedings of the International Conference
on Pyrolysis and Gasication, Luxembourg, pp. 462e467.
Lagoutte, A., Garnier, A., 2001. Procede de valorisation par
gazeication des bois impregnes reformes. Presented at the IRG
Symposium Environment and Wood Preservation, CannesMandelieu, France.
Lao, R.C., Thomas, R.S., Teichman, T., Dubois, L., 1974. Eciency
of collection of arsenic trioxide in high volume sampling. The
Science of the Total Environment 2, 373e379.
Li, K.C., Pupp, C., 1979. The collection of arsenic oxide vapor in
water-lled impingers. American Industrial Hygiene Association
Journal 40 (5), 439e442.
Lindroos, L., 1999. Recycling of impregnated timber part 2:
combustion trial. Presented at the 30th annual IRG meeting,
Rosenheim, Germany, publication IRG/WP 99-50132.
Lindroos, L., 2002. Balance of arsenic and recycling. Prepared for
the 33rd annual IRG meeting, Cardi, Wales, UK, publication
IRG/WP 02-50189.
Lodge, Jr., J.P. (Ed.), 1988. Methods of Air Sampling and Analysis,
third ed. Lewis, Florida, pp. 361e364 and pp. 572e578.
Maenhout, A.W., Kauppinen, E.I., Lind, T.M., 1993. Study of sizefractionated coal-combustion aerosols using instrumental neutron
activation analysis. Journal of Radioanalytical and Nuclear
Chemistry 167 (2), 259e269.
Maggs, R., 2000. A review of arsenic in ambient air in the UK.
Prepared for the Department of Environment, Transport and the
Regions, Scottish Executive SSE/AQ/1465, The National Assembly
of Wales.
Marutzky, R., 1990. Entsorgung von mit Holzschutzmitteln behandelten Holzern. Holz als Roh- und Werksto 48, 19e24.
McMahon, A.K., Bush, P.B., Woolson, E.A., 1985. Release of copper,
chromium and arsenic from burning wood with preservatives,
Presented at the 78th Annual Meeting of the Air Pollution
Association, Detroit, MI, paper 85-56.3.
Health and Safety Exexutive, 1995. MDHS (Methods for the
Determination of Hazardous Substances) 41/2 Arsenic and Inorganic Compounds of Arsenic (Except Arsine) in Air. In: Health
and Safety Laboratory MDHS, 41/2. HSE Books, Sudbury.
NIOSH, 1984. National Institute for Occupational Safety and Health
method 7901: Arsenic trioxide, as As.
NIOSH, 1985. National Institute for Occupational Safety and Health
method 6001: Arsine.
NNI (Nederlands Normalisatie-Instituut), 1996. Nederlandse voornorm, luchtkwaliteit. Uitworp door stationaire bronnen. Monsterneming en analyse voor de bepaling van de gehalten aan arseen,
cadmium, chroom, koper, kwik, lood en zink en hun verbindingen,
in zwevend stof en in de gasfase. NVN 2817, Kalfjeslaan 2, Postbus
5059, 2600 GB Delft.
Nurmi, A.J., 1996. Disposal of CCA treated waste wood by
combustion: an industrial scale trial. Presented at the 27th
Annual IRG Meeting, Guadeloupe, France, publication IRG/WP
96-50068.
Nurmi, A.J., Lindroos, L., 1994. Recycling of treated timber by copper
smelter. Presented at the 25th Annual IRG Meeting, Bali,
Indonesia, publication IRG/WP 94-50030.
Pasek, E.A., McIntyre, C.R., 1993. Treatment and recycle of CCA
hazardous waste. Presented at the 24th Annual IRG meeting,
Orlando, USA, publication IRG/WP 93-50007.
Riveros, P.A., Dutrizac, J.E., Spencer, P., 2001. Arsenic disposal
practices in the metallurgical industry. Canadian Metallurgical
Quarterly 40 (4), 395e420.

L. Helsen / Environmental Pollution 137 (2005) 305e315


Roberts, J.W., Pollock, R.D., Svoboda, M.J., Watters, H.A., 1977.
Ambient air sampling for total arsenic near the Tacoma copper
smelter. Fourth International Clean Air Congress, pp. 424e426.
Sakulpitakphon, T., Hower, J.C., Trimble, A.S., Schram, W.H.,
Thomas, G.A., 2003. Arsenic and mercury partitioning in y
ash at the Kentucky power plant. Energy and Fuels 17,
1028e1033.
Sandelin, K., Backman, R., 2001. Trace elements in two pulverized
coal-red power stations. Environmental Science and Technology
35 (5), 826e834.
Seames, W.S., Wendt, J.O.L., 2000. Partitioning of arsenic, selenium
and cadmium during the combustion of Pittsburgh and Illinois #6
coals in a self-sustained combustor. Fuel Processing Technology
63 (2e3), 179e196.
Seames, W.S., Fernandez, A., Wendt, J.O.L., 2002. A study of ne
particulate emissions from combustion of treated pulverized
municipal sewage sludge. Environmental Science and Technology
36 (12), 2772e2776.
Smith, J.R., 1979. Estimating overall sample train eciency. Journal of
the Air Pollution Control Association 29 (9), 969e970.
Sterling, R.O., Helble, J.J., 2003. Reaction of arsenic vapor species
with y ash compounds: kinetics and speciation of the reaction
with calcium silicates. Chemosphere 51 (10), 1111e1119.
Syrjanen, T., 1999. Recycling of impregnated timber, part 1: crushing,
combustion plants, amount, costs and logistics. Presented at the
30th Annual IRG meeting, Rosenheim, Germany, publication
IRG/WP 99-50131.
Syrjanen, T., Kangas, E., 2001. Recycling of pressure impregnated
timber and preservatives d incineration techniques. Presented at
the IRG Symposium Environment and Wood Preservation,
Cannes-Mandelieu, France.
Tauw Infra Consult B.V., 1987. Verbranding van afvalhout, deel 1.
project 51766.01, Nationaal Onderzoeksprogramma Hergebruik
van Afvalstromen (NOH), Nederland.
Vanhaecke, F., Moens, L., 1999. Recent trends in trace element
determination and speciation using inductively coupled plasma

315

mass spectrometry. Fresenius Journal of Analytical Chemistry 364,


440e451.
VDI, 1994. Determination of total emission of metals, metalloids,
and their compounds; manual measurement in owing, emitted
gases; sampling system for particulate and lter-passing material.
Verband Deutsche Ingenieure (VDI) Handbuch Reinhaltung der
Luft, Band 4, VDI 3868 Part 1.
Venkatesh, S., Fournier Jr., D.J., Waterland, L.R., Carroll, G.J., 1996.
Evaluation of mineral-based additives as sorbents for hazardous
trace metal capture and immobilization in incineration processes.
Hazardous Waste and Hazardous Materials 13 (1), 73e94.
VMM (Vlaamse Milieu Maatschappij), 1998. Emissies uit jaarverslagen, U.M. Olen. address: VMM, Alfons Van de Maelestraat 96,
9320 Erembodegem, Belgium.
Walsh, P.R., Duce, R.D., Fasching, J.L., 1977. Impregnated lter
sampling system for collection of volatile arsenic in atmosphere.
Environmental Science and Technology 11 (2), 163e166.
Wauters, G., 1998. Case study: the behaviour of heavy metals in
a waste incineration process. International Solid Waste Association
(ISWA) Yearbook 1997e1998. James and James, London.
Wilkins, E., Murray, F., 1980. Toxicity of emissions from combustion and
pyrolysis of wood. Wood Science and Technology 14 (4), 281e288.
Wouterlood, H.J., Bowling, K. McG., 1979. Removal and recovery of
arsenious oxide from ue gases. Environmental Science and
Technology 13 (1), 93e97.
Wu, C.Y., Biswas, P., 1993. An equilibrium analysis to determine
the speciation of metals in an incinerator. Combustion and Flame
93 (1e2), 31e40.
Wu, C.Y., Iida, K., Pierman, J., Tolaymat, T., Townsend, T., 2000.
Control of heavy metal emissions from incineration of CCAtreated woods. Presented at the 220th American Chemical Society
National Meeting, Washington, DC.
Wu, C.Y., Barton, T., 2001. A thermodynamic equilibrium analysis to
determine the potential sorbent materials for the control of arsenic
emissions from combustion sources. Environmental Engineering
Science 18 (3), 177e190.