This report deals with brominated dioxins (PBDD/Fs); how they are formed and emitted and what
that leads to in terms of levels in the environment. The report focuses on the situation in Sweden,
but it includes a comprehensive review of studies and data from the whole world. Furthermore, the
report includes a tentative estimation of the total amount of PBDD/Fs that is present and potentially
may be emitted from products and materials in the Swedish society, as well as a discussion on how
these emissions can be minimized.
The PBDD/Fs is a group of unintentionally produced contaminants that are analogous to the more
well‐known chlorinated dioxins (PCDD/Fs), but that contain bromines instead of chlorines. There are
also mixed brominated‐chlorinated dioxins (PBCDD/Fs), although these are only touch upon briefly in
this report. Due to the structural similarities, the PBDD/Fs, PCDD/Fs and PBCDD/Fs share many
properties with each other, including low water solubility and volatility and long environmental halflives.
This means that PBDD/Fs will stick to particles in the environment and stay there for a long
time, usually several years. The similarities also apply to the toxicity, which generally is very high, and
according to the current recommendations the toxic equivalency factors (TEFs) developed for the
PCDD/Fs can also be used for the PBDD/Fs.
Like the PCDD/Fs, PBDD/Fs may be formed in all kinds of combustion processes provided that
bromine is present in some form, which today usually is the case. The bromine content in materials,
waste and fuels has thus increased drastically since brominated flame retardants (BFRs) were
introduced on the market in the 1970s. In the combustion process, the PBDD/Fs can either be
formed from basic elements such as carbon, hydrogen and bromine, or from small precursor
molecules. While the former ‘de novo’ pathway often is the dominating process for PCDD/Fs, the
precursor pathway seems to be more important for the PBDD/Fs. This is because many BFRs, and
especially the polybrominated diphenylethers (PBDEs), constitute far advanced precursors of
PBDD/Fs, and especially of polybrominated dibenzofurans (PBDFs). As a result, mainly PBDFs are
formed in most incineration processes, and the yields increase during poor combustion conditions
when the density of precursors is high.
Furthermore, PBDD/Fs may also be formed during thermal stress of BFR mixtures or BFR containing
materials and products. This means in practice that PBDD/Fs are formed already when the BFR
mixture is produced and that the PBDD/F content steadily increases as the BFRs are mixed into
polymers, when the polymers are refined into materials and products, and when the products are
disposed of and recycled by various processes. In addition, PBDD/Fs may also be formed through
photolytic transformation of BFRs, which may occur as BFR treated products are exposed to sunlight
for instance. In all these processes, the transformation of PBDEs to PBDFs is by far the most
important pathway, and when PBDEs are present all other formation pathways are negligible.
However, recently it has been recognized that PBDD/Fs, and in this case mainly polybrominated
dibenso‐p‐dioxins (PBDDs), may be formed naturally in some marine environments via biological and
photolytical pathways.
The emission from incineration processes are both a result of the release of PBDD/Fs that already are
present in the material being combusted and of the formation of PBDD/Fs in the actual process. Both
these processes are favored by poor combustion conditions, which are more prominent during open
Sources and levels of PBDD/Fs in the Swedish environment
3
burning activities than during controlled incineration in large scale facilities. As a consequence the
emissions measured from incineration facilities are relatively low, while the emissions measured in
connections to open burning activities and accidental fires can be extremely high. However, the
emissions also depends on the fuel and its content of BFRs. Higher emission have therefore been
measured from industrial (IWI) and hazardous waste incinerators (HWI) and from metallurgical
processes using electronic waste as one of its feed stocks, as compared to municipal solid waste
incinerators (MSWI). Furthermore, extremely high PBDD/F emissions have been measured during
open burning of electronic waste (e‐waste), which is carried out as a recycling activity in some
developing countries, and also may occur during accidental fires in e‐waste recycling facilities for
instance.
Relatively large emissions of PBDD/Fs have been measured in connection to production facilities for
BFRs and facilities that are using BFRs to treat products, although there are no such data available for
Swedish facilities. Nevertheless, high PBDD/F levels have been measured in PBDE mixtures as well as
in materials treated with PBDEs, both in Sweden and in other countries. These PBDD/Fs constitute a
threat to human health and the environment as the materials are used, disposed of and recycled
since the PBDD/Fs may be emitted during the whole life cycles of the materials. Not least the
recycling may lead to large emissions of PBDD/Fs, and especially during informal e‐waste recycling
activities. However, large emissions have also been measured in controlled recycling facilities, and in
these cases the emissions seem to be connected to the dust released from the interior of the e‐waste
during the dismantling processes. This may lead to extensive exposure of the recycling workers at
least.
As a result of the great number of sources of PBDD/Fs, these contaminants are nowadays more or
less ubiquitous in the environment. PBDD/Fs have been found in air, soil, sediments, sewage water,
sludge, various animal and plant species, food and feed, indoor dust and humans. In air, the PBDD/F
levels are usually lower than the PCDD/F levels, but in some urban and industrial environments the
reverse profile can be seen. The highest PBDD/F levels in air are found outside e‐waste recycling
facilities and at informal e‐waste recycling sites. The levels are usually correlated with the PBDE
levels and the profiles are dominated by PBDFs. In Sweden, the air levels reported from urban
environments are usually lower than those reported from urban environments in Asia, but for rural
sites the levels are more similar. In fact, the levels in urban environments in Sweden are not that
different from the levels in Swedish rural environments.
In soil the highest PBDD/F levels are generally found at informal e‐waste recycling sites as mentioned
previously. However, similar levels have actually been found in connection to a Swedish recycling
facility that had been subject to an accidental fire. Otherwise, soil levels are generally elevated in
urban environments in comparison to rural and agricultural environments. Like for air, the PBDD/F
levels are generally correlated with the PBDE levels, and the profile is dominated by PBDFs, indicating
that the PBDD/Fs originate from PBDE treated materials. Lake sediments principally follow the same
trend as soil, with higher levels in urban and PBDE exposed areas and with a dominance of PBDFs.
However, marine sediments sometime show a different pattern. Hence, in shallow coastal waters
with high biological productivity and sunshine penetration, the levels of PBDDs may sometimes be
extremely high. This is because of natural formation of PBDDs in these environments.
Sources and levels of PBDD/Fs in the Swedish environment
4
When it comes to biota, food and feed, the PBDD/F levels are generally higher in fatty samples, such
as fatty fish, mussels, eggs, liver and carcass fat. However, the levels in fish and shellfish seem also to
be highly influenced by the presence of local sources in their living habitat, with natural sources
giving rise to the largest variations. Some fish and mussel populations along the Swedish coastline
have thus been found to contain extremely high levels of PBDDs. However, local PBDE related
sources may also give rise to elevated levels, which overall may result in mixed PBDD/F profiles in
some marine fish populations. Apart from these locally influenced fish and mussels, the PBDD/F
levels in marine species are generally lower than the PCDD/F levels. Still, the PBDD/Fs have been
found to contribute significantly to the total dioxin‐like toxicity in these species, which is serious
considering that some of them already contain close to acceptable levels of other dioxin‐like
compounds.
For indoor environments, PBDD/Fs have mainly been measured in dust. The levels vary considerably,
which is believed to be connected to a varying presence of PBDE containing materials in different
indoor environments. However, this correlation is not always easy to discern. Overall, somewhat
higher PBDD/F levels have been found in dust from public buildings, such as hotels and offices, than
in dust from residential houses. However, the highest PBDD/F levels reported for dust have been
found in a residential house in USA. In this house the levels were even higher than those recorded for
workshop floor dust at an e‐waste recycling site in China. Nevertheless, the PBDD/F levels are usually
correlated with the PBDE levels which support the theory that they have the same sources. In
addition, the PBDD/F profiles are usually dominated by PBDFs. In most cases the PBDD/F levels in
dust are higher than the PCDD/F levels, and when it comes to the total dioxin‐like toxicity the
PBDD/Fs contribute significantly. The levels measured in Swedish dust are approximately at the same
levels as in other countries.
In humans, PBDD/Fs have been measured in mother’s milk, adipose tissue and blood. In mother’s
milk the levels are in the same range all over the world, even when including mothers from an
e‐waste site in Vietnam and mothers from USA with much higher PBDE levels in their milk. This,
together with the fact that the PBDF profiles in mother’s milk generally are different from those
observed in dust, indicates that the uptake and metabolism of PBDD/Fs in humans are somewhat
selective. However, overall the PBDD/F levels in mother’s milk are lower than the PCDD/F levels and
their contribution to the total dioxin‐like toxicity seem to be fairly low. PBDD/Fs have also been
measured in human adipose tissue from Sweden and Japan. The levels seem to be somewhat lower
in Sweden than in Japan, but the PBDD/Fs were detected in all Swedish samples as well, verifying
their widespread distribution. The PBDD/F profiles where generally similar to those found in the milk
samples, supporting the theory of a selective uptake and metabolism. However, contribution of the
PBDD/Fs to total dioxin‐like toxicity seems to be higher in the adipose tissue than in the milk
samples. Finally, PBDD/Fs have been found in relatively high levels in blood from two fire fighters in
San Francisco.
The data summarized in this report suggest that most PBDD/Fs that we are exposed to, apart from
the naturally produced PBDDs, originate from the BFR treated materials we have in our society, and
particular those that are treated with PBDEs. A rough estimate, performed in the report, suggests
that such materials could hide several tons of PBDD/Fs only in Sweden. Furthermore, we are adding
some hundreds of kilos of new PBDD/Fs every year in BFR containing products we are importing. This
mainly includes electrical and electronic equipment (EEE), vehicles and construction materials. At the
Sources and levels of PBDD/Fs in the Swedish environment
5
same time BFR containing materials are constantly removed from the society as they reach their endof‐
life stage. When this happens the materials are either recycled, destructed or put on landfill, with
each process constituting a certain risk for further emissions.
Today in Sweden, most of the BFR containing waste is incinerated, but some may also end up on
landfills. This may for example be the case for the shredder light fraction (SLF) from the vehicle
fragmentation process, since this fraction is somewhat complicated to incinerate. It is estimated that
up to 0.5 tons of PBDD/Fs may end up on Swedish landfills every year, and that these landfills may
hide several tons of PBDD/Fs in total. For the waste fraction that is incinerated in authorized facilities
the destruction efficiency of the PBDD/Fs is relatively high, at least when considering the stack
emissions and the levels in the fly ashes. However, when it comes to the bottom ashes the PBDD/F
levels can still be relatively high, indicating that all PBDD/Fs in the original waste is not destroyed or
alternatively that new PBDD/Fs are formed in the process. It is estimated that the bottom ashes
produced in Swedish MSWIs every year contain almost 6 kg of PBDD/Fs.
Besides all PBDD/Fs that are hidden in products, materials and waste in our society, there is an even
larger amount of not‐yet‐formed PBDD/Fs in all BFRs (and particularly PBDEs) that are present in the
same and similar products, materials and waste in the society. If all this material would be subject to
some kind of incomplete combustion process, as a worst case scenario, it could potentially give rise
to an additional 200 tons of PBDD/Fs that would be emitted. To eliminate this risk, these materials
will have to be removed and subsequently destructed. However, in this context it should be noted
that the risk for emission often will increase as the material is removed from its original placement
and when it is being recycled and destructed.
To minimize the emissions of PBDD/Fs from waste material a key factor would be the
implementation of efficient and sensitive identification and separation technologies that are capable
of separating BFR containing materials from non‐BFR‐containing materials, so that each fraction can
be treated appropriately. There are several spectroscopic technologies available for this purpose, and
also technologies based on differences in density and electrostatic properties. However, none of
these technologies are alone capable of screening out the BFR containing materials, but need to be
used in combinations. Still, a 100% separation will not be achieved, and usually around 5% will end
up in the wrong fraction. As a consequence, recycled materials should never be used for sensitive
applications, such as toys and household products, and when it comes to the BFR containing fraction,
it should not be recycled at all. It could perhaps be used to make basic chemicals (like bromine) or
fuel for the industry, but otherwise it should be destructed.
Destruction of BFR containing materials may be accomplished in authorized incineration facilities.
The mixing ration of the BFR materials in other wastes/fuels should however be kept low (<5%) in
order to limit the amount of PBDD/F precursors and corrosive HBr in the combustion zone and the
flue gases. The incinerators should also have highly developed flue gas cleaning systems and plans for
how ashes (not least bottom ashes) should be handled. Other high temperature processes, such as
cement kilns and metal smelter may also be used to destruct BFR containing materials provided that
these have similar emission control systems as the authorized incinerators. If these demands are
fulfilled, metals smelters are preferably used for the metal containing BFR waste, e.g. PC‐boards,
since these facilities are capable of recovering the metals.
Sources and levels of PBDD/Fs in the Swedish environment
6
As an alternative to a complete destruction of the BFR containing materials it may also be possible to
extract the bromine or the intact BFRs from the materials, after which the materials can be recycled
or treated as non‐BFR materials and the bromine reused in the industry. A couple of such
technologies have been suggested and described for BFR containing plastics, but none have so far
been applied in full‐scale.
The very last option for BFR containing waste, as with other organic waste fractions, is landfilling.
Normally it should not be used at all, but if it for some reason has to be chosen as an alternative it
has to be done under certain controlled conditions to minimize leakage, emissions and exposure of
humans and animals. The landfills also have to be secured from accidental fires and be able to
maintain a high security even if the surrounding and the climate are changing. The short‐term risks
mentioned above, e.g. risk for leakage and fire, also have to be considered when BFR containing
materials are stored temporarily, while waiting for other treatments for instance. In such situations,
it would perhaps also be wise to limit the amount of BFR containing materials/wastes that can be
stored in the same area in order to minimize the damages caused if an accidental fire still would
occur