Elemental analyses of pine bark and wood in an environmental study

Abstract

Bark and wood samples were taken from the same individuals of Scots pine (Pinus sylvestris L.) from a polluted area close to a Cu-Ni smelter in Harjavalta and from some relatively unpolluted areas in western Finland. The samples were analysed by thick-target particle induced X-ray emission (PIXE) after preconcentration by dry ashing at 550 degrees C. The elemental contents of pine bark and wood were compared to study the impact of heavy metal pollution on pine trees. By comparison of the elemental contents in ashes of bark and wood, a normalisation was obtained. For the relatively clean areas, the ratios of the concentration in bark ash to the concentration in wood ash for different elements were close to 1. This means that the ashes of Scots Pine wood and bark have quite similar elemental composition. For the samples from the polluted area the mean concentration ratios for some heavy metals were elevated (13-28), reflecting the effect of direct atmospheric contamination. The metal contents in the ashes of pine bark and wood were also compared to recommendations for ashes to be recycled back to the forest environment. Bark from areas close to emission sources of heavy metal pollution should be considered with caution if aiming at recycling the ash. Burning of bark fuel of pine grown within 6 km of the Cu-Ni smelter is shown to generate ashes with high levels of Cu, Ni as well as Cd, As and Pb.

Full text

Elemental analyses of pine bark and wood in an
environmental study
K.-E. Saarela
a
, L. Harju
a,
*, J. Rajander
a,b
, J.-O. Lill
b,c
, S.-J. Heselius
b
,
A. Lindroos
d
, K. Mattsson
e
a
Laboratory of Analytical Chemistry, Process Chemistry Group, A
˚bo Akademi University, Biskopsg. 8, FIN-20500 A
˚bo, Finland
b
Turku PET Centre, Accelerator Laboratory, A
˚bo Akademi University, Porthansg. 3, FIN-20500 A
˚bo, Finland
c
Department of Physics, A
˚bo Akademi University, Porthansg. 3, FIN-20500 A
˚bo, Finland
d
Department of Geology and Mineralogy, A
˚bo Akademi University, Domkyrkotorget 1, FIN-20500 A
˚bo, Finland
e
Department of Biology, A
˚bo Akademi University, BioCity, Artillerig. 6, FIN-20500 A
˚bo, Finland
Received 6 January 2004; received in revised form 22 September 2004; accepted 24 September 2004
Abstract
Bark and wood samples were taken from the same individuals of Scots pine (Pinus sylvestris L.) from a polluted area close
to a Cu–Ni smelter in Harjavalta and from some relatively unpolluted areas in western Finland. The samples were analysed by
thick-target particle induced X-ray emission (PIXE) after preconcentration by dry ashing at 550 8C. The elemental contents of
pine bark and wood were compared to study the impact of heavy metal pollution on pine trees. By comparison of the elemental
contents in ashes of bark and wood, a normalisation was obtained. For the relatively clean areas, the ratios of the concentration
in bark ash to the concentration in wood ash for different elements were close to 1. This means that the ashes of Scots Pine
wood and bark have quite similar elemental composition. For the samples from the polluted area the mean concentration ratios
for some heavy metals were elevated (13–28), reflecting the effect of direct atmospheric contamination. The metal contents in
the ashes of pine bark and wood were also compared to recommendations for ashes to be recycled back to the forest
environment. Bark from areas close to emission sources of heavy metal pollution should be considered with caution if aiming at
recycling the ash. Burning of bark fuel of pine grown within 6 km of the Cu–Ni smelter is shown to generate ashes with high
levels of Cu, Ni as well as Cd, As and Pb.
D2004 Elsevier B.V. All rights reserved.
Keywords: Pine bark; Pine wood; Ashes; Heavy metals; PIXE; Atmospheric pollution
1. Introduction
The emission of heavy metals to the environment
is one of the most serious environmental problems
and the content of these elements tends to increase,
0048-9697/$ - see front matter D2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2004.09.043
* Corresponding author. Tel.: +358 2 2154425; fax: +358 2
2154479.
E-mail address: lharju@abo.fi (L. Harju).
Science of the Total Environment 343 (2005) 231 – 241
www.elsevier.com/locate/scitotenv

e.g. in soils. Heavy metals and other pollutants are
taken up to a varying degree by different plants and
also by different parts of a plant species. Different
methods have been applied to monitor heavy metal
emissions. Mosses have been widely used for large
scale regional monitoring of airborne emission of
these pollutants in northern Europe (Ru
¨hling et al.,
1992). Tree bark has been found to be a useful
bioindicator for monitoring airborne pollution
(Schulz et al., 1999; Harju et al., 2002). The sorption
and desorption of Cd and other heavy metals by pine
bark has been studied in more detail by Al-Asheh
and Duvnjak (1997).
Elemental concentrations have been reported both
for stem wood and bark of pine trees (Young and
Guinn, 1966; Fossum et al., 1972; Basham and
Cowling, 1976; Lo¨vestam et al., 1999). Metals are
taken up by stem wood from soil and soil water, and
an enrichment of the inorganics occurs towards the
outer parts (branches and the bark) of the tree. The
interpretation of the results from analysis of wood can
be difficult, because wood is a very heterogeneous
material. Already within individual tree rings there are
large seasonal variations in the elemental concen-
trations (Harju et al., 1996; Lo¨vestam et al., 1999).
Many elements also appear in low concentrations.
Several analytical techniques have been suggested
for the chemical analysis of inorganic constituents in
wood related materials (Ivaska and Harju, 1999).
Most commonly used are wet-chemical methods like
atomic absorption spectrometry (AAS), atomic emis-
sion spectrometry (AES) and inductively coupled
plasma mass spectroscopy (ICP-MS). These methods
demand digestion of the samples prior to the
analysis. The elemental content of stem wood of
different tree species has been determined using these
methods (Young and Guinn, 1966; Keitanniemi and
Visapa¨a¨, 1978; Hamm et al., 1986). Inductively
coupled plasma mass spectrometry has also been
applied to chemical analyses of bark samples after
dissolution of the samples (Ma et al., 2000; Bellis et
al., 2002). Laser ablation ICP-MS has been used by
Narewski et al. (2000) for direct determination of
elements in tree bark samples. Particle-induced X-ray
emission (PIXE) analysis of pine bark has been used
for the study of long-term environmental changes by
Raunemaa et al. (1987), who determined relative
concentrations of Si, S, Cl, K, Ca, Mn and Fe. The
PIXE technique has also been applied by Esch et al.
(1996),Harju et al. (1997) and Saarela et al. (2002)
in studies of trunk wood.
Knowledge of the elemental content of stem wood,
bark and other tree related materials are of great
environmental and industrial interest. In the present
work, the elemental contents of both bark and stem
wood of pines were determined with the thick-target
PIXE method. Samples were taken from different
areas in Finland. The wood and the bark from the
same stem were analysed. Analyses were performed
on the ashed material.
2. Experimental
2.1. Sampling
Samples of Scots pine (Pinus sylvestris L.) were
taken from Harjavalta, an area with a high load of
airborne heavy metal pollution, and from two other,
non-industrialised, areas in western Finland (Fig. 1).
A Cu–Ni smelter has been active in Harjavalta for
over 50 years. The annual emissions of Cu, Ni, Zn, Pb
and Cd from the smelter have been reported by
McEnroe and Helmisaari (2001). According to Euro-
pean moss studies (Ru
¨hling et al., 1992) the smelter is
one of the main emission sources of heavy metals like
copper and nickel in Finland. Southerly winds are
reported to be somewhat dominant in the area
(Salemaa et al., 2001). The sampling sites in the
present study were about 6 km northeast of the smelter
in Harjavalta. The relatively unpolluted sites are
represented by Nagu in the Turku archipelago and
by the four sites; Kronoby, Vimpeli, Pyh7j7rvi and
Piippola in Ostrobothnia (Fig. 1).
The samples were taken during forest felling, by
cutting a 5–10 cm thick disc of the stem at a height of
about 0.5 m above the ground. In the laboratory the
samples were cleaned with a stream of air and air-
dried. In order to obtain concentrations on a dry
matter basis the samples were further dried in an oven
at 105 8C. As has been earlier shown (Harju et al.,
1996, 2002) both stemwood and bark are chemically
very heterogeneous materials. Therefore the labora-
tory sampling of wood was made according to a
sectorial pattern (Harju et al., 1997) in order to obtain
a representative sample of the whole stem. The bark
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241232

segments outside the wood sectors were also analysed
(see Section 3.2).
2.2. Dry ashing
Bark and wood samples can be directly analysed
with thick-target PIXE. However, the concentrations
of many elements in bark and wood are low and the
elements are unevenly distributed. For these reasons
the materials were dry-ashed as described in detail
earlier (Harju et al., 1997; Lill et al., 1999). The wood
and bark samples (10–20 g) were dry-ashed in a
Vulcan 3–130 programmable oven by slowly increas-
ing the temperature to 550 8C. The samples were kept
Fig. 1. Map of the sampling sites in western Finland.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241 233

at this temperature for at least 12 h. At this temper-
ature Hg and also volatile elements like the halogens
are partly lost. The ashed samples were stored in
desiccators until analysed. Biological certified refer-
ence materials, CRMs, (1–3 g) were dry-ashed in the
same way. Homogeneous materials for analyses were
obtained by shaking of the ashes before target
preparation.
2.3. Target preparation
Target pellets were prepared using a commercial
pelletising device and consisted mainly of graphite
with the ashed sample pressed on the front surface of
the pellet. The following bio-fuel CRMs from
Swedish University of Agricultural Sciences, Ume3,
Sweden were used for the calibration and evaluation
of the PIXE analyses: Wood Fuel (Pine), NJV 94-5,
Energy Forest (Salix), NJV 94-3 and Energy Grass
(Phalaris arundinaceae L.), NJV 94-4. Other CRMs
used were Pine Needles (1575) from NIST (National
Institute of Science and Technology, Gaithersburg,
MA, USA) and Beech Leaves from BCR (Community
Bureau of Reference, Brussels).
2.4. PIXE analysis
The prepared samples were irradiated with a 3
MeV proton beam for about 10 min each. The protons
were produced with the 2bo Akademi MGC-20
cyclotron. The set-up is described in more detail
elsewhere (Lill et al., 1999). The integrated charge on
the sample was determined indirectly by monitoring
the light emission in air induced by the proton beam
(Lill, 1999). The X-rays emitted during the irradiation
were measured with an Intrinsic Germanium Planar
(IGP) detector and the obtained spectra were analysed
off line using the GUPIX software (Maxwell et al.,
1989). A typical X-ray spectrum obtained by PIXE
analysis of an ashed stemwood sample of pine from
the polluted area in Harjavalta is shown in Fig. 2. The
precision expressed as relative standard deviation,
RSD, of the PIXE method is 1–2% (Lill et al., 1993)
and of the dry ashing ca 2% (Saarela et al., 1995).
Fig. 2. X-ray spectrum of ash of pine bark from Harjavalta. Note the clear peaks from the trace metals silver and cadmium to the right in the
spectrum.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241234

When analyzing wood related materials with PIXE
combined with dry ashing, typical limits of detection
(LOD) for heavy metals are 0.02–0.1 mg/kg (Lill et
al., 1998).
3. Results and discussion
3.1. Elemental content in pine wood
Results obtained in this work from analyses of
individual tree discs of Scots Pine by the thick-target
PIXE method are given in Tables 1–3. The ash
contents are also included in Tables 1 and 2. Mean
elemental concentrations and relative standard devia-
tions are given both for wood and bark samples.
The ash percentages (0.23–0.33%), reflecting the
total content of inorganics in wood, are relatively
constant in pinewood for all three areas studied (Table
1). The Ni concentrations are somewhat higher in the
wood samples taken from Harjavalta compared to the
samples from the two reference areas. The Cu
concentrations are only slightly higher for the
Harjavalta area. No clear trend was observed for Zn.
The concentrations of the two main mineral
elements K and Ca are lowest in the samples from
Harjavalta (Table 1). The same is also true for Rb and
Sr. This could be interpreted as an antagonistic effect
from other cations including hydrogen ions in acid
precipitation in the area. Derome and Lindroos (1998)
give inhibition of mineralization of litter and displace-
ment from cation exchange sites as the reason for the
observed decrease of exchangeable Ca, Mg and K in
the organic forest soil layer near the Harjavalta
smelter. They also gave the same explanation for
earlier findings by Nieminen and Helmisaari (1996) of
low levels of Ca, Mg and K in pine needles at the
same sampling spots.
Analytical data for stem wood of some of the pines
have earlier been reported by us (Harju et al., 1997;
Saarela et al., 2002). Results from analyses of these
wood samples are also included in this work, in order
to obtain the elemental concentration ratios in bark
ash to wood ash. Several stem wood samples from the
same pine discs have been dry-ashed and analyzed
with good agreement.
The elemental content of stem wood reflects
mainly the concentrations of metal ions in soil and
soil water, from which the metal ions can be taken up
by the tree at the root membranes. Usually the
precipitation of atmospheric pollutants, e.g. heavy
metals, leads to some degree of soil pollution. It can
Table 1
Elemental concentrations in wood samples of Scots pine
Element Nagu Ostrobothnia Harjavalta
Average S.D. nAverage S.D. nAverage S.D. n
Ash [%] 0.27 0.03 3 0.33 0.07 8 0.23 0.03 3
P 112 65 3 58 24 8 64 7 3
S 111 36 3 62 22 8 84 17 3
K 620 212 3 596 246 8 448 79 3
Ca 689 167 3 789 136 8 487 110 3
Ti blq 1.4 1.3 6 blq
Mn 32 10 3 83 36 8 58 13 3
Fe 9.7 6.8 3 12.9 15.0 8 1.9 1.0 3
Ni 0.21 0.06 3 0.13 0.06 8 0.34 0.18 3
Cu 0.59 0.10 3 0.86 0.21 8 0.96 0.14 3
Zn 11.7 8.3 3 5.6 2.3 8 7.8 3.2 3
As blq blq 0.01 0.01 2
Rb 3.8 1.5 3 2.4 1.4 8 1.0 0.1 3
Sr 2.8 0.4 3 2.9 0.9 8 1.4 0.8 3
Ag blq blq 0.13 0.12 3
Cd blq blq 0.48 0.13 3
Ba 4.5 6.1 3 3.3 3.5 8 1.4 1.6 2
Pb 0.10 0.07 3 blq 0.08 0.06 3
All concentrations are given in mg/kg dry weight. Blg = below limit of quantification.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241 235

be concluded that airborne heavy metal pollution is
not easily reflected in stem wood data.
3.2. Elemental content in pine bark
Lo¨vestam et al. (1999) found, by scanning
micro-PIXE, that Fe, Cu and Zn were concentrated
to a few grains in a studied pine bark sample. We
also performed a PIXE-scan on pine bark. The
sampled piece represented the thickest part of the
bark (22 mm) with cracks down to inner bark
(about 2 mm thick layer) appearing on both sides.
The distribution of some elements is shown in Fig.
3. Due to the complexity in the physical structure
and topography of bark and to the strong variations
in the elemental content in the different bark layers,
whole bark samples were analysed in this work.
Preconcentration and increased homogeneity was
obtained by dry ashing and thorough shaking of the
ashes.
The mean ash contents of bark from the 3 regions
studied were 1–3% (Table 2) and varied more than the
ash content of the wood in the same pine trees. The
concentrations of Ni, Cu and Pb in pine bark from
Harjavalta were much higher than in the samples from
the two reference areas (Nagu and Ostrobothnia). The
Ni and Cu concentrations obtained for the samples
from Harjavalta exceeded the highest concentrations
reported by Schulz et al. (1999) in pine bark from
differently polluted areas in Northern, Central and
Eastern Europe. The Zn concentrations were of the
same order for all three sampling areas. Also, As
could be determined in the bark samples from
Harjavalta. Concentrations of Cd and Ag could be
quantified in bark (but not wood) samples from Nagu.
The Cd is easily related to the use of fossil fuels—a
main source of emission of this element (Narodo-
slawsky and Obernberger, 1996). The Ag found is
more puzzling.
Pine bark samples from Ostrobothnia contained
surprisingly high levels of some heavy metals (Fe,
Mn, Zn) and also S, compared to samples from the
two other areas. As was the case for wood, the
concentrations of K, Ca, Rb and Sr were clearly lower
in the bark samples from Harjavalta, than in those
from Nagu and, especially, Ostrobothnia.
Lippo et al. (1995) have compared the use of bark,
mosses and lichens as bioindicators in a nationwide
study of atmospheric heavy metal pollution in Fin-
land. They found that bark had the lowest concen-
tration of metals and did not reveal regional
differences.
Table 2
Elemental concentrations in bark samples of Scots pine
Element Nagu Ostrobothnia Harjavalta
Average S.D. nAverage S.D. nAverage S.D. n
Ash [%] 1.31 0.45 3 3.04 0.94 8 0.92 0.31 3
P 382 225 3 746 240 8 164 97 3
S 345 70 3 853 204 8 329 99 3
K 1610 922 3 3170 1540 8 651 317 3
Ca 7230 4790 3 12,700 6130 8 4460 2920 3
Ti blq 7.2 6.3 6 3.3 1.5 2
Mn 31 9 3 432 202 8 71 34 3
Fe 33 12 3 162 138 8 147 102 3
Ni 0.60 0.23 3 1.71 1.43 8 18 5 3
Cu 3.5 0.2 3 3.4 1.5 8 89 40 3
Zn 15.0 4.3 3 45.0 19.4 8 43.3 37.4 3
As blq blq 0.92 0.21 3
Rb 10.9 5.0 3 19.5 14.7 8 1.6 1.0 3
Sr 12.3 1.6 3 22.0 6.5 8 8.9 4.7 3
Ag 1.0 1.1 3 blq blq
Cd 1.00 0.02 2 0.80 0.10 3 0.92 0.45 2
Ba blq 73 77 8 17 20 3
Pb 1.0 0.5 3 blq 9.1 10.0 3
All concentrations are given in mg/kg dry weight.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241236

3.3. Elemental concentration ratios for bark ash to
wood ash
Barnes et al. (1976) have reported concentration
ratios for bark to wood for the elements Cu, Pb and Zn
in Norway Spruce and Poplar Lombardy. The
variations in the ratios were quite large from 0.8 to
120. Also, in our study the concentration ratios for
bark to wood for most elements are clearly above
unity due to the higher ash content of bark.However,
by compensating for the differences in ash content and
using the elemental concentrations ratios for bark ash
to wood ash instead, the ratios given in Table 3 are
obtained.
For samples from Nagu (in the Turku archipelago)
the concentration ratios for most elements in ashes
Fig. 3. Elemental concentration profiles for some elements in a pine bark sample. The sample is scanned from inner bark (left) towards outer
bark (right).
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241 237

were in the range 0.6–0.9. The ratio of 2.1 was
obtained for the major element Ca. This means that
we have an enrichment of Ca in bark compared to
wood. Zinc and Mn, which are essential trace
elements for the tree, show the low concentration
ratios of 0.3 and 0.2, respectively (Table 3). These
metals are effectively taken up from soil and soil
water. For the pine bark and the stemwood samples
taken from different sites in Ostrobothnia the concen-
tration ratios obtained for ashes were for most
elements in the range 0.5–1.5. The elevated concen-
trations obtained for some heavy metals in bark from
Ostrobothnia (Table 2) are thus normalized in the
comparison given in Table 3. Similar ratios were
obtained for K, Ca, Rb and Sr in samples from
Ostrobothnia and Nagu. The Ca ratios were around 2
for all tree sampling sites.
The last two columns in Table 3 show the results
calculated for the polluted area in Harjavalta. The
concentration ratios for ashes of bark and wood show
clear evidence of airborne pollution of heavy metals.
Concentration ratios from ca 13 to 28 were obtained
for Ni, Fe, Cu, As and Pb. The Zn ratio was 1.4 for the
pine samples from Harjavalta.
The concentration ratio for Cd is also close to 1.
This although the Cd concentrations in ashes of bark,
but also wood, from Harjavalta can be concluded to be
elevated. Cd is chemically closely related to the plant
essential element Zn, which is effectively taken up by
the tree. The especially high concentration in wood
ash indicates that plant-available Cd is present in the
soil at elevated levels on this sampling spot. The
reason for this could be found from a past with larger
and more uncontrolled atmospheric emission of
metals from the Cu–Ni smelter. Inorganic emission
has mainly been in the form of dust, which contains
Ni and Cu, and to a lesser extent Cd, Pb, Fe and Zn
(McEnroe and Helmisaari, 2001). Emissions peaked
during the 1980s but have been continuously reduced
during the last two decades due to more stringent
environmental control (McEnroe and Helmisaari,
2001). Silver in wood from Harjavalta (Ta ble 1)
could represent the same kind of bmemory effectQ.
Silver in the corresponding bark was below the limit
of quantification (LOQ).
Airborne heavy metal pollution will increase the
metal content in soil and soil water to some degree and
will thus also result in an increased uptake by the tree
to the stem wood (see 3.1). However, the uptake is
effectively different for elements, and this anthropo-
genic contribution is more difficult to monitor in the
bulk of the tree, the stem wood, than in bark which is
also affected by direct sorption from air. Comparing
concentrations in bark (ashes) to the concentrations in
the stem wood (ashes) of the same tree will, however,
be of help in identifying the inorganic atmospheric
pollutant. By this method variations in bark concen-
trations, due to local soil geochemistry or soil pollution
from other sources than the air, can be excluded.
3.4. Comparison with guideline values for forest
fertilisation
The National Board of Forestry in Sweden
recommends compensatory forest fertilisation by
using ash products that mainly originate from burning
of forest fuels (Anon., 2001). Amounts used should be
calculated to compensate for the mineral ions taken
out from the ecosystem with the biomass. At the same
time, the net input of toxic heavy metals with ash
products should not exceed the amounts removed by
logging. On the basis of these principles, and a
recommended maximum of 3 tons of ash/ha to be
spread, limits for heavy metals in the ash product is
given by the Swedish Forest Board (Anon., 2001). In
Table 4, the concentrations of some heavy metals in
Table 3
Elemental concentration ratios for bark ash to wood ash from pine
Element Nagu Ostrobothnia Harjavalta
Bark/
Wood
S.D. Bark/
Wood
S.D. Bark/
Wood
S.D.
P 0.7 0.6 1.4 0.7 0.6 0.4
S 0.6 0.2 1.5 0.6 1.0 0.4
K 0.5 0.4 0.6 0.4 0.4 0.2
Ca 2.1 1.5 1.7 0.9 2.3 1.6
Ti 0.6 0.7
Mn 0.2 0.1 0.6 0.4 0.3 0.2
Fe 0.7 0.5 1.3 1.9 20 18
Ni 0.6 0.3 1.4 1.4 14 8
Cu 1.2 0.2 0.4 0.2 23 11
Zn 0.3 0.2 0.9 0.5 1.4 1.3
As 23 20
Rb 0.6 0.4 0.9 0.8 0.4 0.3
Sr 0.9 0.2 0.8 0.3 1.6 1.2
Cd 0.5 0.3
Ba 2.4 3.5 3.2 5.3
Pb 2.1 1.9 28 37
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241238

wood and bark ashes of pine from the three sampling
sites in this study are given, and compared with these
recommended values (Anon., 2001). The highest
concentrations of the elements in pine wood and bark
ashes are obtained for Zn. For all three sampling sites
studied the Zn concentrations are of the same order
(1000–5000 Ag/g) and below the upper limit given.
The elemental concentrations ratios (Table 3) for
Cu, Ni and Pb in samples from Harjavalta were an
order of magnitude higher than those for Zn and Cd
(around 1). Cu, Ni and Pb concentrations in pine bark
ash from Harjavalta were also much higher than the
guideline values given in Table 4. Concentrations of
Cd in both ashes of wood and bark from Harjavalta
exceeded those recommended.
Arsenic concentrations were also quantified in
ashes of all bark samples from Harjavalta. The mean
concentration of As in the ashes was 100 mg/kg,
which exceeds the recommended concentration, 30
mg/kg, for this element (Anon., 2001).
Also, the Cd content in the bark ash from Nagu
exceeds the guideline value. The Cd concentrations in
the wood samples from Ostrobothnia and Nagu were
below the limit of quantification of the analytical
method used. For the relatively non-polluted areas
Ostrobothnia and Nagu the concentrations of Cu, Ni
and Pb were below the concentration limits given in
Table 4. Only the Ni content (78 mg/kg) obtained for
wood ash from Nagu is slightly above the guideline
value.
When discussing the analytical data given in Table
4, in terms of potential use of pine wood and bark
ashes as fertilisers, one should keep in mind that
ashes studied in this work were obtained by dry
ashing at 550 8C in a laboratory oven. This temper-
ature was mainly used for analytical chemical
reasons. The ashes represent closely the total
inorganics of the samples studied with most elements
retained in the ash. The temperatures in biofuel
combustion plants are much higher and the ashes
produced are mainly divided into bottom ash and fly
ash. According to Obernberger et al. (1997), volatile
toxic heavy metals like Pb and Cd are enriched in fly
ash fractions while Hg—present in very low concen-
trations in the fuel—escapes almost completely with
flue gas (Obernberger et al., 1997). The majority of
less volatile heavy metals like Cu and Ni are bound
(and enriched) in the bottom ash (Obernberger et al.,
1997) For circulating fluidized bed (CFB) combus-
tion, where the combustion temperature is about 850
8C, enrichment in fly ash is obtained for As, Cd, Pb
and Zn (Steenari and Lindqvist, 1997). Lind et al.
(1999) have thoroughly studied the volatilisation and
distribution of heavy metals during CFB combustion
of forest residue.
As shown in this work total ashes of bark generally
have the same chemical composition as the corre-
sponding stem wood ashes. Ashes of bark from an
area with atmospheric emission of heavy metals,
though, contain much higher concentrations of these
elements (as seen in Table 4). In biofuel combustion
plants, ash fractions with considerably higher heavy
metal contents than reported for total ashes would be
formed.
4. Conclusions
Stemwood and bark are heterogeneous materials
and quite a large amount of sample material is thus
needed to obtain statistically representative analytical
results. In this work dry ashing of 10–20 g of sample
is combined with direct TTPIXE analyses of the
ashes. The ashes are easily homogenised and thus—in
addition to the enrichment—a homogeneous target is
obtained for the particle induced X-ray emission
Table 4
Concentrations (mg/kg) of heavy metals in ash of pine wood and bark and guideline values for bio ash [Anon., 2001] as forest fertilizers
Nagu Ostrobothnia Harjavalta Guideline values
Wood S.D. Bark S.D. Wood S.D. Bark S.D. Wood S.D. Bark S.D. Wood ash
Ni 78 22 46 18 39 19 56 47 147 77 1990 581 70
Cu 223 36 266 15 261 64 111 49 413 60 9700 4350 400
Zn 4370 3120 1150 330 1710 698 1480 637 3360 1390 4720 4080 7000
Cd 77 2 26 3 205 56 101 49 30
Pb 36 27 73 41 36 26 989 10100 300
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241 239

(PIXE) analyses. Due to the method used amounts and
composition of the total ash from the materials can
also be reported.
The concentrations obtained of most heavy
metals were relatively low and normally in the
range 1–100 mg/kg in the non-ashed wood and bark
samples studied. The ash percentages of pinewood
(0.22–0.33%) and bark (0.9–3.0%) are also impor-
tant parameters for characterisation of the materials.
When interpreting analytical data for bark obtained
by, for examples, wet chemical methods, quite large
variations in the total content of inorganics (ash
percentage) have not been considered in environ-
mental studies. A comparison of elemental contents in
ashes of bark and wood, from the same tree, is a way
to identify and evaluate antrophogenic effects on
trees. In areas with no ongoing atmospheric metal
pollution the content of inorganics in both bark and
wood is basically dependent on uptake of the ions
from the soil. The contents of inorganic elements—
and thus the ash content—is many folds higher in bark
compared to the underlying wood. However, the
composition of elements in both materials is very
similar. For the reference areas studied the ratios of
concentration in bark ash to the concentration in wood
ash for elements are, thus, mostly quite close to 1. For
pine samples from the area close to the Cu–Ni smelter
the elemental concentration ratios of some heavy
metals were clearly elevated. The variations in
concentration quota for all areas studied are generally
high.
A comparison of the heavy metal content in
ashes of pine bark, with guideline values, showed
that bark ashes of trees 6 km from the Cu–Ni
smelter at Harjavalta contained concentrations of As,
Ni, Cu, Cd and Pb that clearly exceeded those
recommended for ashes to be used for forest
fertilisation. Also, the ashes of stem wood contained
concentrations of Cu, Ni and Cd that were close to
or exceeded the recommendations.
To follow the principle of no net input of toxic
elements to the forest ecosystem, and recommenda-
tions given for ashes on this basis, bark should be
used—as biofuels—with great attention to the origin
of the material. The use of bark from areas with
atmospheric heavy metal pollution easily generates
ashes that should not be circulated back to the forest
environment.
Acknowledgements
The financial support by the Rector and the
University Council of 2bo Akademi University is
greatly acknowledged. One of the authors (K-E. S)
acknowledges the financial support from the Society
of Swedish Literature in Finland. The authors thank
the technical staff of the 2bo Akademi Accelerator
Laboratory. We also thank TkL Kari Saari for
sampling in Ostrobothnia and Mr. B. Bjfrkqvist for
help with dry ashing.
References
Al-Asheh S, Duvnjak Z. Sorption of cadmium and other heavy
metal ions by pine bark. J Hazard Mater 1997;56:35 – 51.
Anon. Rekommendationer vid uttag av skogsbr7nsle och kom-
pensationsgfdsling. Svenska skogsstyrelsens Meddelande (an-
nouncement of the Swedish forest board) 2/2001; 2001. p. 1–16.
ISSN 1100-0295.
Barnes D, Hamadah MA, Ottaway JM. The lead, copper and zinc
content of tree rings and bark A measurement of local metallic
pollution. Sci Total Environ 1976;5:63 – 7.
Basham HG, Cowling EB. Distribution of essential elements in
forest trees and their role in wood deterioration. Beih Mater Org
1976;3:155 – 65 [Org. Holz].
Bellis DJ, McLeod CW, Satake K. Pb and
206
Pb/
207
Pb isotopic
analysis of a tree bark near Sheffield, UK recording historical
change in airborne pollution during the 20th Century. Sci Total
Environ 2002;289:169 – 76.
Derome J, Lindroos A-J. Effect of heavy metal contamination on
macronutrient availability and acidification parameters in forest
soil in the vicinity of the Harjavalta Cu–Ni smelter SW Finland.
Environ Pollut 1998;99:225 – 32.
Esch M, Hofmann D, Krebs G, Thierfelder C, Wunsch R, Maier R,
et al. Elemental composition of different types of wood. Nucl
Instrum Methods, B 1996;09/110:328 – 31.
Fossum T, Hartler N, Libert J. The inorganic content of wood. Sven
Papptidn 1972;75:305 – 9.
Hamm U, Geller A, Gfttsching L. Schwermetalle in holz, prim7rfaser-
rfaserstoffen, altpapier und papier. Das Papier 1986;40(10A);37 – 46.
Harju L, Lill J-O, Saarela K-E, Heselius S-J, Hernberg F-J, Lindroos
A. Study of seasonal variations of trace-element concentrations
within tree rings by thick-target PIXE analyses. Nucl Instrum
Methods, B 1996;109/110:536– 41.
Harju L, Lill J-O, Saarela K-E, Heselius S-J, Hernberg F-J. Analysis
of trace elements in trunk wood by thick-target PIXE using dry
ashing for preconcentration. Fresenius’ J Anal Chem 1997;358:
523 – 9.
Harju L, Saarela K-E, Rajander J, Lill J-O, Lindroos A, Heselius
S-J. Environmental monitoring of trace elements in bark of
Scots pine by thick-target PIXE. Nucl Instrum Methods, B
2002;189:163 – 7.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241240

Ivaska A, Harju L. Analysis of inorganic constituents. In: Sjfstrfm
R, Ale´n R, editors. Analytical methods in wood chemistry,
pulping and papermaking. Berlin7Springer Verlag; 1999
[chapter 10 in the book, 365 pp.].
Keitanniemi O, Visap77 A. Amounts and behavior of certain
chemical elements in kraft pulp manufacture: results of a mill
scale study. Pap Puu 1978:60507 – 22.
Lill J-O. Charge integration in external beam PIXE. Nucl Instrum
Methods, B 1999;150:114– 7.
Lill J-O, Saarela K-E, Hernberg F-J, Heselius S-J, Harju L. A
novel method for charge integration in external beam
TTPIXE Application to analyses of biological materials.
Nucl Instrum Methods, B 1993;83:387 – 93.
Lill J-O, Harju L, Saarela K-E, Lindroos A, Heselius S-J. Increased
sensitivity in thick-target particle induced X-ray emission
analyses using dry ashing for preconcentration. Anal Chim
Acta 1999;378:273 – 8.
Lind T, Valmari T, Kauppinen EI, Sfiris G, Nilsson K, Maenhaut W.
Volatilization of the heavy metals during circulating fluidized
bed combustion of forest residue. Environ Sci Technol 1999;
33:496 – 502.
Lippo H, Poikolainen J, Kubin E. The use of moss, lichen and pine
bark in the nationwide monitoring of atmospheric heavy metal
deposition in Finland. Water Air Soil Pollut 1995;85:2241 – 6.
Lfvestam NEG, Johansson EM, Johanson SAE, Pallon J. Scanning
proton microprobe analysis applied to wood and bark samples.
Nucl Instrum Methods, B 1999;49:490– 4.
Ma R, Bellis D, McLeod W. Isotopic analysis of uranium in tree
bark by ICP mass spectrometry A strategy for assessment of
airborne contamination. Anal Chem 2000;72:4878 – 81.
Maxwell JA, Campbell JL, Teesdale W. The Guelph PIXE software
package. Nucl Instrum Methods, B 1989;43:218– 30.
McEnroe NA, Helmisaari H-S. Decomposition of coniferous forest
litter along a heavy metal pollution gradient, south-west Finland.
Environ Pollut 2001;113:11– 8.
Narewski U, Werner G, Schultz H, Vogt C. Application of
laser ablation inductively coupledcoupled mass spectrometry
(LA-ICP-MS) for the determination of major, minor and
trace elements in bark samples. Fresenius’ J Anal Chem
2000;366:167 – 70.
Narodoslawsky M, Obernberger I. From waste to raw material—the
rout from biomass to wood ash for cadmium and other heavy
metals. J Hazard Mater 1996;50:157 – 68.
Nieminen T, Helmisaari H-S. Nutrient retranslocation in the foliage
of Pinus sylvestris L growing along a heavy metal pollution
gradient. Tree Physiol 1996;16:825 – 31.
Obernberger I, Biedermann F, Widmann W, Riedel R. Concen-
trations of inorganic elements in biomass fuels and recovery in
the different ash fractions. Biomass Bioenergy 1997;12:211– 24.
Raunemaa T, Hannikainen M, Kulmala M, Hari P. Pine bark PIXE
analysis A new method to study long-term environmental
changes. Nucl Instrum Methods, B 1987;22:473– 5.
Rqhling 2., Brumelis G, Goltsova N, Kvietkus K, Kubin E, Liiv S,
et al, 1992. Atmospheric heavy metal deposition in northern
Europe 1990. Nord, vol. 12. Copenhagen7Nordic Council of
Ministers; 1992.
Saarela K-E, Harju L, Lill J-O, Rajander J, Lindroos A, Heselius
S-J. Thick-target PIXE analysis of trace elements in wood
incoming to a pulp mill. Holzforschung 2002;56:380– 7.
Salemaa M, Vanha-Majamaa I, Derome J. Understorey vegetation
along a heavy-metal pollution gradient in SW Finland. Environ
Pollut 2001;112:339 – 50.
Schulz H, Popp P, Huhn G, St7rk H-J, Schqrmann G. Biomonitoring
of airborne inorganic and organic pollutants by means of pine
tree barks: I Temporal and spatial variations. Sci Total Environ
1999;232:49 – 58.
Steenari M-M, Lindqvist O. Stabilisation of biofuel ashes for
recycling to forest soil. Biomass Bioenergy 1997;13:39– 50.
Young HE, Guinn VP. Chemical elements in complete mature trees
of seven species in Maine. Tappi 1966;49:190 – 7.
K.-E. Saarela et al. / Science of the Total Environment 343 (2005) 231–241 241