Assessment of ecotoxicological risks of element leaching from pulverized coal ashes

This thesis describes the consequences of the disposal of the combustion residues of coal, especially the uptake of elements from such residues and their effects on various organisms. The effects on benthic organisms in fresh and in seawater are considered in the first two parts. The third part looks at the uptake of elements from coal residues and their effect on the growth of plants and worms.

The central theme is the combustion residue known as pulverized fuel ash (PFA), or 'flyash'. Coal is a product of natural origin that was formed from plant remains in the period between 265 and 203 million years ago. During the coal forming process the elements from plants become 100 to 1000 times more concentrated. In all probability, further concentration occurred through airborne deposition and the salts from sea and freshwater. The elements present will be concentrated a further 10 times during combustion in a boiler. This means that following combustion, the unburned elements are re-released to the environment, whether or not they are bound to flyash particles. Some elements are released directly in the gaseous phase, other elements condense on the surfaces of the flyash spheres and another fraction is bound up in the matrix of these particles. The question is whether an adequate estimate can be made, from the ecotoxicological point of view, as to what effect on the environment PFA will have when used in unbound applications.

The elements in PFA can be divided into macro-elements which occur in relatively high concentrations (g/kg) and trace elements which occur at concentrations of less than I mg/kg. It is largely these trace elements which reach the environment through leaching from PFA and accumulate in organisms. In The Netherlands, however, discharge of PFA in unbound form (not as cement/concrete 'stabilized' products) is not a concern at the moment. It may possibly be (re)considered as an option in the future, at which point in time, knowledge of possible environmental effects will be essential. At present, nearly all PFA is used by the cement industry. Additionally, a small quantity is used in pellet form as artificial gravel in the concrete industry.

The anion forming elements are especially important from the point of view of environmental protection. In order of leaching potential, these elements are in order of highest to lowest: selenium>molybdenum>>wolfram>vanadium>antimony>arsenic>chrome. Accumulation of these elements by organisms in excessive concentrations can lead to undesirable effects. Such effects can be increased (biornagnification) in organisms which are higher in the food-chain.

Part 1
In the first part of this thesis research in the marine environment is described with organisms which are indigenous to intertidal sedimentary biotopes, i.e. the lugworm Arenicola marina , the ragworm Nereis virens , the baltic tellin Macoma balthica and the cockle Cerastodermaedule. The accumulation of such elements is described in the above species and their survival during a 90- days exposure to 100% PFA and a 50% PFA mixture with sand. A sand control and a reference of moderately contaminated harbour dredged sediment from Rotterdam harbour were also used. In addition, the effect of a daily dose of PFA was examined in an effort to simulate regular disposal. At the end of this period, the organisms were sampled and the elements zinc, arsenic, chrome copper, nickel, cadmium antimony and selenium were determined in animal tissue and sediment. In the ragworm experiment, young worms were used to examine the colonisation of disposed PFA. Moreover, the effects of a 230-days exposure to a PFA substrate is described in the cockle. Alteration in the reproductive processes, tissues and organs was assessed in these animals. Parasites were also examined, as organisms living in sub-optimal conditions are often more strongly parasitised.

Results
The concentration of the elements measured in the PFA substrates appeared to have been reduced by 10% during the test period. However, it was clear that the arsenic and selenium concentrations had not been reduced to the extent expected (<5% only), as R is known that both elements are easily leached. The accumulation patterns in the lugworm demonstrate that arsenic in particular is accumulated (a factor of 15 in 50% PFA 150% sand and a factor of 5 in 100% PFA on basis of tissue concentrations). This is probably caused by the binding of arsenic to sediment particles which are eaten by the lugworm in its sedimentary diet. Accumulation in the cockle was only a factor of 2 above the sedimentary levels. The small accumulation levels in the cockles from the 230-days experiment were unexpected. The accumulation factor for arsenic was a factor of 3 in the baltic tellin. The concentration of zinc in the baltic tellin was remarkably high in animals exposed to all sediment types, including the control, at>550 mg/kg (dry weight). However, these concentrations do not appear to be abnormal and existed at the start of the experiments. From this it can be seen that the 'lifestyle' of the organisms used in bioassays can strongly determine the accumulation of the anionic elements. Where the cationic elements copper, chrome and nickel are concerned, no unexpected accumulation patterns were found; only zinc was accumulated. Several elements were analyzed after 4, 8 and 12 weeks exposure in the ragworm experiment. Here also, the arsenic concentration increased up to week 8 (from 23 to 76 μg/g in 100% PFA, on basis of tissue concentrations), after which a decrease was observed after12 weeks. It is interesting to note that in the ragworm, selenium was slowly but steadily accumulated and no equilibrium was reached after 12 weeks exposure. The copper concentration in ragworms increased by a factor of three. The other cation forming elements zinc, chrome and nickel did not however accumulate.

With the lugworm, a high mortality was found in all substrates, i.e. 96% in the 100% PFA and 55.5% in the control. The lugworm appears sensitive to transplantation and 100% PFA is absolutely unacceptable as a substrate in which to burrow. By contrast, the baltic tellin showed a low mortality (20% in 100% PFA) and a wild population existed in the substrate, originating from the Wadden Sea. Cockle mortality reached 40% in the 100% PFA exposure and 15% in the control. Unexpectedly, only slight microscopic changes in the reproductive tissues were observed. In the experiment with the ragworms young animals with an average weight of 200 mg (ca. 2 cm in length) were used and an appreciable mortality of 50 - 60% was seen in the first weeks of the experiment. This mortality was to be expected as the delicate young animals were placed in the tanks by hand at the start of the experiment.

Conclusions
The general conclusion concerning the leaching of elements from PFA and subsequent uptake in the benthic fauna is that a species dependent differential situation occurs. Arsenic and selenium are the most prominent elements. A second conclusion is that PFA caused a physical effect which was reflected in an altered behaviour of the organisms examined. Even in a substrate of 50% PFA / 50% sand after 60 days none of the burried worms came to the surface, indicating severe stress. The consequences of this are that, a radical change in the species composition of the benthic fauna will take place upon dumping PFA in the marine environment and recovery will take years. A lesser sedimentation through deposition over larger areas will reduce the mortality of species or populations, but effects on species composition will still occur.

Part 2
The experiments described in Part 2 were carried out in flow chambers with the painters mussel Unio pictorum. This mussel species was chosen because R crawls freely in the upper layers of sediment unlike other sessile bivalve species such as the Zebra mussel (Dreissena polymorpha). The painters mussel moves by sticking out their foot, which then changes shape to form an anchor with which the mussel then pulls itself along. The mussels are rather active in summer and remain permanently open, but during the winter, dig themselves in.

The growth of mussels in PFA substrates was compared with data from Rhine sediment. Additionally, growth was compared with tagged mussels which were followed for several years in the River Linge, a tributary of the Rhine. Behaviour was examined with a laboratory version of the Mussel Monitor ^r, where activity was measured as the opening and closing actions of the valves. In a simultaneous research project at the University of Utrecht with a similar mussel species in aquaria without any substrate, the mussels remained closed and only opened for short periods. The specific question examined during these experiments was whether the mussels showed an abnormal behavioural pattern in PFA substrates and did this influence the accumulation of metals.

The accumulation of the metal cadmium is examined in a separate study. Although cadmium only occurs in very small amounts in PFA, it was none the less chosen for this accumulation experiment, because much experience had already been gained with cadmium in laboratory experiments at the University of Utrecht. The expectation was that accumulation would be considerably different in the laboratory than in semi-field experiments due to the large difference in opening 1 closure behaviour.

The accumulation of selenium and the occurrence of effects on the development of eggs in U. pictorum was examined. Arsenic and selenium leach readily from PFA, whereby arsenic can be strongly accumulated in contrast to selenium. Selenium (Se) is taken up only very slowly, but bioaccumulation and biornagnification does occur. Selenium is concentrated in proteins and especially in the developing oocytes which form a target organ. The literature indicates that some fish species in selenium rich waters in California die out at concentrations of 15 μg/l Se. An important fact is that Se can occur as different forms or 'species', e.g. selenite [Se( IV)], selenate [Se (VI)] and organo-selenium compounds. It is usually the Se (IV) form which occurs in PFA leachates, ie. this species which accumulates more rapidly than Se (VI). The accumulation of Se was examined in three experiments with dosed Se (IV).

Results
The growth of mussels in PFA was somewhat retarded when compared to growth in river sediment (Rhine and Linge). The mussels had initial difficulties in burrowing into the PFA and were only comparable with the Rhine situation after 5 days. The individual mussels nearly always stood open and showed a high level of activity interspersed with short periods when the valves were closed.

In the cadmium experiment, mussels were exposed to four separate dosages of 50 ug/l. The accumulation of cadmium appeared to progress rapidly, and a plateau was reached after 3 weeks with kidney concentrations of ca. 400 μg/g (freeze dried weight); which means a bioconcentration factor (BCF) of 8000. No elimination of cadmium was observed during a consecutive depuration period of 29 weeks. The lack of any difference in accumulation in the presence or absence of a substrate was unexpected.

At the selenium experiments the effect of enriching PFA and Rhine sediment with selenium was examined. The highest concentrations were observed in animals exposed to the enriched Rhine sediment. Mussels were continuously exposed for 11 weeks to 50 μg/l of Se (IV) in order to examine the possible effect on the development of oocytes and embryos. Selenium accumulation had not reached a steady state after 11 weeks of exposure. Differences (not significant) were found in the synchronisation of glochidia production (larvae on the gills of female mussels) as well as in the average size of the oocytes at the end of the exposure period. Following exposures of 22 and 11 weeks to concentrations of 50 and 250 μg/l, an accumulation of 5 and 15 μg/g was observed, with BCFs of 100 and 60 respectively (freeze dried weight). The most unexpected result was that no differences were observed in the numbers of glochidia and that no deformations were found. This implies that no demonstrable (significant) effects were found in the chosen experimental setup. Se resembles sulphur in uptake behaviour and is built in proteins (eggs). The element is an essential nurient, but can be toxic at slightly higher concentrations than ambient. On the basis of literature data for the development of eggs in a perch species during a 32 week experiment, the expectation was that effects would indeed be found.

Conclusions
The accumulation of cadmium in U. pictorum occurred rapidly. The kidneys formed the the main target organ and a steady state is apparently reached after three weeks of exposure. Elimination of cadmium hardly occurr over a period of 28 weeks whithout dosing. No mortality was found during the exposure and during the elimination period. An important finding is that the presence or absence of substrate, which has a dear effect on behaviour (valve movement), did not influence the uptake of cadmium.

One apparent effect of selenium on oocytes, embryos and larval development is that the timing of the appearance of larvae in the gills of the females is altered. The consequence of this is that the mussels retain the larvae longer on the gills and release them later. Such longer retention might have consequences for the life cycle of this species through a delay in the settlement on fish and the ditribution of the larvae.

Main conclusion from this section is that PFA has a marked influence on the quality as a sediment by physical changes leading to abnormal behaviour and secondly through the leaching of elements.

Part 3
Part 3 is concerned with research carried out with plants and earthworms. Phytomonitoring as illustrated here with duckweed, is still a rather uncommon monitoring technique but admirably suited to the purpose of testing leachates from PFA. Experiments with the duckweed Lemna minor are described where the effects of conventional PFA, PFA from low NOx burners, bottom ash and coal gasification ash were examined. In order to be able to compare the observed effects properly, sediments from a canal and a river were sampled and tested also. At the same time different methods of producing leacheates were examined. A distinction is made between natural leacheates; and artificially produced leacheates, where element concentrations were in agreement with the 'cascade' leachate technique (successive leaching of 5 times). An image analysis technique was developed to measure the growth of duckweed in a simple and effective manner, in the various experiments. The effects of coal gasification slag (a type of bottom ash) were also examined. In the coal gasification process, the flammable gaseous components are first removed from the powdered coal with the aid of steam under high pressure and reducing conditions. The gas mixture thus produced is burned by conventional means. During this process, only a relatively small fraction of flyash is produced and is removed. The bulk of the residue is formed by coal gasification slag (CG slag) which remains after the gasification process. The growth of yellow nut sedge (Cyperus esculentus) and the elements which accumulate in the plant were examined in the laboratory. The effects on the earthworm Eisenia fetida were examined in a similar manner.

Results
Duckweed appeared to be very suitable organism to monitor leacheates from PFA. It appears from the experiments that the anion forming elements cause the effects in the conventional PFA and the low NOx leacheates, while the cations cause the effects in the sediments from the Apeldoornsch Kanaal and the Rhine. The toxicity of PFA is considerably lower when compared to that of the Apeldoornsch Kanaal. This difference is mainly caused by zinc (>4000 μg/l which occurs in the leachate from the Apeldoornsch Kanaal; the 'no effect concentration' for duckweed is about 160 μg/l.

The order of toxicity for elements such as those found in separately executed tests with duckweed was: cadmium>copper>zinc,>arsenic (111)>seleniurn (IV)>boron>molybdenum.

The leaching of elements from coal gasification slag appeared to be minimal. The growth of the yellow nut sedge decreases with increasing concentration of coal gasification slag through lack of nutrients and not as a result of the toxic elements from CG slag. Addition of nutrients resulted in a marked improvement in growth. Growth in PFA was equally retarded, in this case because the fresh PFA contained boron which shows growth effects on plants above 60 μg/g. Accumulation was found to be restricted to boron and molybdenum. The physical effect of compaction of the PFA, through which the substrate becomes so hard as to prevent proper root growth, should also be considered as a serious effect. A clearly negative effect on the growth and mortality of the worm E. fetida was found in PFA and CG slag. The mortality was especially high (32%) in PFA; growth was only found in potting soil. Accumulation was only observed for arsenic. The availability of the anion forming elements from PFA was greater that from CG slag.

Conclusions
The contention that PFA in general should be evaluated as a chemical waste needs to be reconsidered in the light of the results presented here. At present, The Government of The Netherlands is considering a revision of this standpoint, through which PFA will be placed in a category with raw materials. The effects of conventional and low NOx PFA are related to the anion forming elements. The leaching of elements from CG slag appears to be very small. The sediments from the Apeldoornsch Kanaal show a far higher toxicity than PFA, which is caused by the high zinc concentrations. The effects of the sediments from the Rhine and the Apeldoornsch Kanaal are associated with the cation forming elements.

Final Conclusions
The main conclusion is that through the disposal of PFA, the greatest changes are caused by physical effects, as appears from the marine and freshwater studies. The acceptability of PFA as a substrate for benthic organisms is far worse that expected. Within the relatively short time period of the tests, accumulation of elements and acute effects appears to be far less important than expected. A reduction in number of oocytes in U. pictorurn was found prior to spawning. The later appearance of the larvae of U. pictorum point to possible effects at the population level. Where accumulation of elements is concerned, the anions arsenic and selenium are the most important elements in leacheates.

Phyto-monitoring of PFA leacheates and sediments from the Apeldoornsch Kanaal and Rhine with the duckweed (L. minor) shows that PFA has but a comparatively low toxicity. The leaching of elements from CG slag is minimal and the observed effects can be accounted for by a lack of nutrients. The main thrust of the "Bouwstoffenbesluit" which is in preparation at present and aims at preventing soil pollution, is the leaching of unbound and stabilized residues. For PFA in unbound form, this means that the leaching characteristics must be measured in a column test and the composition determined following Aqua Regia digestion. During digestion, the matrix is broken open and a composition spectrum arises which is far removed from the natural situation as illustrated by the duckweed experiments, for example. For the elements antimony, chrome, fluorine, molybdenum, selenium, vanadium and the sulphate component, the threshold value of the "Bouwstoffenbesluit" will in all probability not be achievable when based on the proposed column test. More knowledge is needed concerning the leaching of PFA in unbound and stabilized form under natural conditions, when considering the biologically available fraction. This means that more information is needed on the speciation of the elements in bioassay studies. In unbound form, the physical effects on the aquatic environment (substrate) and thereby on species and populations is an area where as yet, little is known. The possibilities for stabilized applications of non-usable wastes such as bottom ashes, and flyash from coal- and waste combustion, in 'artificial waste blocks' requires more attention. Applications can certainly be found in the form of artificial reefs, aimed at biotope improvement for sessile organisms and young fish.

In general, toxicological risks of unbound PFA deposition show to be less severe as initially assumed. However, physical effects for benthic organisms will be severe.