The product from the burning of a combination of carbonaceous materials.

Long-term toxicity to aquatic invertebrates

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Study period:

1 September 2010 to 22 September 2010

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: A GLP compliant, guideline experimental study.

Qualifier:

according to guideline

Guideline:

OECD Guideline 211 (Daphnia magna Reproduction Test)

Deviations:

no

GLP compliance:

yes

Details on sampling:

- Test water was sampled and pH, hardness and dissolved oxygen were measured by the study laboratory. Water quality analysis including DOC, COD, conductivity, alkalinity, suspended solids, N, pesticides, PCBs and metals was conducted in a separate non-GLP laboratory. Metal analysis of four test solutions (3.2 mg/l, 18 mg/l and two controls) were conducted in another non-GLP laboratory.
- Sample storage conditions before analysis: Corresponding to the storage condition of test material.

Details on test solutions:

PREPARATION OF TEST SOLUTION

This study was run with a dilution water control (DWC) and nominal loading concentrations of 3.2, 5.6, 10, 18 and 32 mg/L.

The test solutions were prepared by the direct addition of test substance to dilution water in a volumetric flask and the volume made up to 1 or 2 litres. The test solutions were prepared the day before the renewal and then stirred overnight using magnetic followers. Prior to use the solutions were cooled to 20 ± 1ºC, allowed to settle out for 15 ± 5 minutes and then filtered through a 0.45 μm membrane filter.

All test solutions were clear and colourless in appearance.

Test organisms (species):

Daphnia magna

Details on test organisms:

TEST ORGANISM
- Common name: water flea
- Source: Continuous laboratory cultures at Brixham Environmental Laboratory
- Age of parental stock (mean and range, SD): parental stock 15±1 days old
- Age at study initiation: <24 hours
- Feeding during test
- Food type: a mixed algae diet of Chlorella vulgaris, strain CCAP 211/12 and Pseudokirchneriella subcapitata, strain CCAP 278/4.
- Amount: each test vessel received a nominal 0.06 mg C day-1 of each algal suspension
- Frequency: daily

ACCLIMATION
- Type and amount of food: The D. magna cultures were fed on a mixed algae diet of Chlorella vulgaris, strain CCAP 211/12 and Pseudokirchneriella subcapitata, strain CCAP 278/4. The D. magna cultures were fed daily ad libitum depending on age and density of the culture.
- Feeding frequency: daily depending on age and density of the culture
- Health during acclimation (any mortality observed): The test organisms and the culture from which they were obtained showed no evidence of disease before the test period.

QUARANTINE (wild caught)
- not applicable

METHOD FOR PREPARATION AND COLLECTION OF EARLY INSTARS OR OTHER LIFE STAGES:
D. magna <24 hours old, obtained from a single culture vessel, were used for testing. The parent animals were 15  1 days old and had been maintained with a twice weekly renewal of reconstituted water medium since birth.

Test type:

semi-static

Water media type:

freshwater

Total exposure duration:

21 d

Hardness:

The hardness (as CaCO3 mg L-1) of the dilution water control and highest test concentration ('on' and 'off') were measured on 3 occasions during the test. An additional measurement of hardness occurred for the highest test concentration on Day 5 after 100% mortality occurred in this concentration. See table 1.

Test temperature:

19.4 - 20.4°C

pH:

pH of 'on' solutions 7.7 - 8.7
pH of 'off' solutions 7.6 - 8.6

Dissolved oxygen:

Dissolved oxygen concentration of 'on' test solutions (mg L-1) 6.2 to 8.2
Dissolved oxygen concentration of 'off' test solutions (mg L-1) 8.8 to 9.8

Nominal and measured concentrations:

Nominal: 3.2, 5.6, 10, 18 and 32 mg/L and control (0)

Details on test conditions:

TEST SYSTEM
- Test vessel: Glass beakers
- Type (delete if not applicable): closed; covered with loose fitting glass lids
- Material, size, headspace, fill volume: glass, each containing 80 ml of test solution, providing a depth of approximately 50 mm
- Aeration: no
- Renewal rate of test solution (frequency/flow rate): three times weekly on each Monday, Wednesday and Friday throughout the study
- No. of vessels per concentration (replicates): 10 replicates per test concentration

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: Elendt's M4 D. magna medium
- Total organic carbon: 0.64 mg/L
- Solids suspended @ 105ºCa: <3 mg/L
- Metals: 682.66 μg/L
- Highest organophosphorous pesticides: <0.01 μg/L
- Highest organochlorine pesticides: 0.119 μg/L
- Alkalinity as CaCO3: 35 mg/L
- Ca/mg ratio: <0.1 μg/L
- Conductivity: 643 μS/cm

OTHER TEST CONDITIONS
- Adjustment of pH:
- Photoperiod: photoperiod of 16 hours light: 8 hours dark with 20 minute dawn and dusk transition periods was provided
- Light intensity: 10.5 μ Einsteins m-2 s-1

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

18 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

growth

Remarks:

Length

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

18 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: Survivability

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

32 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: Survivability

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

3.2 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

reproduction

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

5.6 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

reproduction

Duration:

21 d

Dose descriptor:

EC50

Effect conc.:

7 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

reproduction

Remarks on result:

other: CL 5.4 – 7.8 mg/L

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

3.2 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: Overall; all biological parameters taken into account (survival, reproduction, length)

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

5.6 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: Overall; all biological parameters taken into account (survival, reproduction, length)

Details on results:

100% mortality occurred in the 32 mg/L loading concentration by Day 5 of the study. Therefore this concentration was discontinued for the remainder of the study and excluded from the statistical analysis.

A statistically significant (p <0.05) reduction in reproduction was found between the control and the 5.6, 10 and 32 mg/L loading concentrations.
The overall No Observed Effect Concentration (NOEC) of mixed ash, based on nominal loading concentrations was 3.2 mg/L.

- Mortality of parent animals: no at <= 18mg/L
- No. of offspring produced per day per female: see. table 2.
- Body length and weight of parent animals: see table 3.
- Other biological observations: On Day 9 of the study, one individual in the 10 mg/L loading concentration was observed to be pale in comparison to the other replicates in this concentration. Aborted eggs were observed to be present in four replicates of the 5.6 and 10 mg/L loading concentrations, and in 7 replicates of the 18 mg/L loading concentration on Day 9 of the study. Aborted eggs were also observed in all replicates of the 10 and 18 mg/L loading concentrations on Day 19 of the study. No other symptoms of toxicity were observed in any of the surviving D. magna during the study.

Results with reference substance (positive control):

- Results with reference substance valid?
- Relevant effect levels:
- Other:

Reported statistics and error estimates:

No significant difference in length was found between any of the treatments and the control. Therefore, based on nominal concentrations, the NOEC and LOEC for length was 18 mg/L.
A statistically significant (p <0.05) difference (reduction) in reproduction, was found between the control and the 5.6, 10 and 18 mg/L loading concentrations. However, no significant difference was found between the control and the 3.2 mg/L concentration.

Observation of mortalities and symptoms of toxicity

Mortalities of P₀ generation D. magna by replicate are given in Table 2. 100% mortality had occurred in the 32 mg/L loading concentration by Day 5 of the study, so this concentration was terminated from the remainder of the study. On Day 9 of the study, one individual in the 10 mg/L loading concentration was observed to be pale in comparison to the other replicates in this concentration. Aborted eggs were observed to be present in four replicates of the 5.6 and 10 mg/L loading concentrations, and in 7 replicates of the 18 mg/L loading concentration on Day 9 of the study. Aborted eggs were also observed in all replicates of the 10 and 18 mg/L loading concentrations on Day 19 of the study.

No other symptoms of toxicity were observed in any of the surviving D. magna during the study. Therefore, based on nominal concentrations, the NOEC for survival was 18 mg/L and the LOEC for survival was 32 mg/L.

There were no mortalities in the dilution water control after 21 days; this satisfies the guideline validity criteria of <=20% at the end of the test.

Reproduction data

The total numbers of offspring (F1) obtained from each replicate vessel are given in Table 2. Also shown are the means, standard deviations and totals number of offspring for the control and each of the test concentrations.

No offspring were produced by the P₀ D. magna in the 32 mg/L loading concentration before the 100% mortality in this loading concentration. In all other loading concentrations, all P₀ D. magna had released their first offspring by Day 8, and all surviving D. magna had completed at least five broods by the end of the study. The mean number of offspring produced in the control, per surviving parent, at the end of the test was 127, exceeding the minimum of 60 specified in the guideline. The residual standard deviation (coefficient of variation) around the mean number of living offspring per parent in the control was 9.6%, below the maximum of 25% recommended in the guideline.

A statistically significant (p<0.05) difference (reduction) in reproduction, was found between the control and the 5.6, 10 and 18 mg/L loading concentrations. However, no significant difference was found between the control and the 3.2 mg/L concentration. Therefore, the NOEC for reproduction was 3.2 mg/L and the LOEC was 5.6 mg/L. The median effect concentration (EC50) was determined to be 7.0 mg/L with 95% confidence intervals 5.4 – 7.8 mg/L.

Table 2 Reproduction data

Nominal loading conc of mixed ash (mg/L)	Total number of offspring per P0 D. magna										Mean number of offspringper P0D. magna	Standard deviation	Total number of offspring	Mean number of offspringas % of control
	1	2	3	4	5	6	7	8	9	10
Control	132	134	132	120	130	120	144	114	126	114	127	9.6	1266	-
3.2	123	123	125	116	122	133	91	127	135	134	123	12.7	1229	97
5.6	55	22	84	94	79	97	78	85	71	88	75*	22.2	753	59
10	40	46	35	39	49	23	39	35	31	33	37*	7.4	370	29
18	38	37	30	30	14	14	32	31	16	21	26*	9.3	263	20
32	M	M	M	M	M	M	M	M	M	M	-	-	-	-

*Statistically significant difference between treatment and control (p<0.05)

M=Mortality

Length data

The lengths of the surviving P₀ D. magna at the end of the test are given in Table 3 along with the mean and standard deviation for the control and test concentrations.

No significant difference in length was found between any of the treatments and the control. Therefore, based on nominal concentrations, the NOEC and LOEC for length was 18 mg/L.

Table 3 Adult length measurements

Nominal loading conc of mixed ash (mg/L)	Length in graticule scale divisions (Day 21)per P0 D. magna										Mean length	Standard deviation	Mean
	1	2	3	4	5	6	7	8	9	10
Control	54	55	53	52	51	51	54	53	52	52	53	1.3	4.4
3.2	51	55	53	55	54	54	52	52	56	52	53	1.6	4.5
5.6	55	53	52	52	53	55	52	52	52	55	53	1.4	4.4
10	56	53	49	53	54	55	52	52	53	54	53	1.9	4.4
18	55	54	54	53	54	54	54	51	54	53	54	1.1	4.5
32	M	M	M	M	M	M	M	M	M	M	-	-	-

^$Length in mm (1 mm = 12 scale divisions).

No significant difference between treatment and solvent control (p <0.05).

M = Mortality.

Validity criteria fulfilled:

yes

Remarks:

There were no mortalities in the dilution water control after 21 days; this satisfies the guideline validity criteria of <=20% at the end of the test.

Conclusions:

Taking all biological parameters into account (survival, reproduction, length) and based on nominal loading concentrations, the overall NOEC of Ash was 3.2 mg/L, and the LOEC was 5.6 mg/L.
The median effect concentration (EC50), based on reproduction was determined to be 7.0 mg/L with 95% confidence intervals 5.4 – 7.8 mg/L.

Executive summary:

Chronic toxicity of Ash was determined in a GLP compliant study according to OECD Guideline 211 (Daphnia magna Reproduction Test). In the test, young, less than 24 h old D. magna were exposed to the test substance for 21 days in a semi-static test. Due to low water solubility of Ash, test solutions were filtrated to form Water Accommondable Fractions (WAF) of Ash. The nominal concentrations of test solutions were 0 (control), 3.2, 5.6, 10, 18 and 32 mg/L. Mortality of the P0 generation was recorded daily for each test vessel. Observations were made daily from Day 0 for the presence of live and dead offspring (termed F1 generation) in each vessel.At the end of the test, the length (apex of helmet to base of spine) of each surviving P0 Daphnia was measured. 100% mortality occurred in the 32 mg L-1 loading concentration by Day 5 of the study. Therefore this concentration was discontinued for the remainder of the study and excluded

from the statistical analysis. Taking all biological parameters into account (survival, reproduction, length) and based on nominal loading concentrations, the overall NOEC of Ash was 3.2 mg/L, and the LOEC was 5.6 mg/L. The median effect concentration (EC50), based on reproduction was determined to be 7.0 mg/L with 95% confidence intervals 5.4 – 7.8 mg/L.

Reference 2

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Reason / purpose for cross-reference:

reference to same study

Principles of method if other than guideline:

Bottom ash (aged for five years) was studied with sequential batch leaching from L/S 1 to L/S 3 using the modified EN 12457-3 (two stage batch test) method. The modified method included three subsequently steps of leaching at cumulative liquid to solid (L/S) ratios of 1, 2 and 3 L/kg. These eluates were i) tested with a subchronic ecotoxicity test with the harpacticoid copepod Nitocra spinipes, which has been the most robust and sensitive species in our previous studies on ash, ii) analyzed for total content of elements, and iii) analyzed with DGT filters to obtain bioavailable fractions of selected elements. Finally, a spiked solution, which elements and concentrations were selected on the basis of L/S 3 analytical results, was tested with N. spinipes and analysed accordingly.

GLP compliance:

no

Details on sampling:

- Concentrations: L/S 1, 2 and 3
- Sampling method:
- Sample storage conditions before analysis: ash samples and all eluates were stored at 4±2° C

Test organisms (species):

Nitocra spinipes

Details on test organisms:

Study design
Long-term ecotoxicity test, The Larval Development Ratio Test (LDR), The LDR endpoint (i.e. the ratio between the number of copepodites and total number of living animals at the end of the test) is recorded, together with mortality, after 5-7 days when ~ 50 % of the control animals have reached a copepodite stage.

Test type:

semi-static

Water media type:

brackish water

Total exposure duration:

7 d

Test temperature:

22 ºC

pH:

7.9 - 8.1

Nominal and measured concentrations:

Cu and K was tested at the following nominal concentration: Cu at 60, 120 and 240 μg/L, whereas K at 35, 105 and 315 mg/L.

The leaching followed the procedure of EN 12457-3 (two stage batch test), but was modified to include three subsequently steps of leaching at cumulative liquid to solid (L/S) ratios of 1, 2 and 3 L/kg. In each leaching step the material was mixed with the leachant (MilliQ-water) to an L/S of 1 L/kg and agitated for 24h. The solid residue was thereafter separated by filtration (0.45μm membrane filters) and the generated eluate divided into subsamples for the following analysis (chemical element analysis, DGT analysis and ecotoxicity tests).

Details on test conditions:

Baltic Sea brackish water

Duration:

7 d

Dose descriptor:

LOEC

Effect conc.:

ca. 35 mg/L

Nominal / measured:

nominal

Conc. based on:

dissolved

Basis for effect:

other: Larval development

Remarks on result:

other: Concentration of K.

Details on results:

Leaching
The leached concentrations of elements at the L/S ratios studied are shown in Table 8. For more information on the leaching study, see Section 4.8.

Based on a toxicity index (TI, table 9), the results of Pb and Zn in the studied eluates were very low and their contribution to toxic effects detected in an LDR-test can thus be regarded as insignificant. For Ca and Al it was close to 1, stabile for all studied eluates and thus indicated a risk of toxic effects. Here the calculated TI for Al was therefore misleading since it is well known that aluminum do not cause any toxic effects at pH 8 and Al had very low bioavailability at the prevailing pH. Estimated TI for Cu indicated no significant risk of toxicity and the value of the index significantly declined from the L/S 1 to L/S 3.
 
DGT-analysis
The precision of the DGT-measurements appeared to be very good when the DGT samplers were exposed to the same eluate. Also when the samplers were exposed to eluate from two separate eluate replicates (data for L/S 1) the precision was good. For Fe, Cr and Pb the precision was in the latter case not so good. For Fe and Cr, the labile concentrations and the fractions of the measured total concentrations increased from L/S 1 to L/S 3. Also for the other selected elements there was a variation, but the changes were less pronounced as compared to Fe and Cr. Aluminium was not expected to bind to humic substances and the low labile fractions confirmed earlier indications that Al measured in eluates from aged bottom ash at pH 8-9 were mainly present as colloids and were hence not bioavailable. The labile fraction for Cu only varied between 38 % and 45 %, although the measured total concentrations ranged from 58 to 178 mg/L. The analysis of DGT-samplers showed that labile concentrations and thus very likely also bioavailable fractions of different elements varied in different media. For Cu, which was one of the critical elements from an ecotoxicological point of view, the observed variation in labile fractions was relatively limited, ranging from 24 % to 45 %, but was still of such a magnitude that it could inflict on the bioavailability of the element in ecotoxicological tests.
 
Ecotoxicological testing
The only significant difference from a control group was on larval development in copepods exposed to the highest concentration (i.e. 40.5%) of the eluate generated at L/S 1. The calculated TI values indicated that there are four elements that could be responsible for the significant effects on larval development in the highest concentration of the L/S 3 eluate which are Ca, Al, Cu and K. The DGT-analysis shows that the bioavailability of Al is very low, and it is therefore unlikely that Al can explain the observed changes in toxicity from L/S 1 to 3. For Ca, on the other hand, the calculated TI values indicate that Ca is present at levels which should induce toxicity at all L/S ratios. However, these findings do not corroborate the findings of the toxicity of the eluates when exposed toN. spinipes. For Cu, none of the calculated TI values were above 1 but considering natural biological variation and that the TI value in the L/S 1 eluate was 0.56, the potential toxic contribution from copper cannot be neglected. Still, the element for which the toxicity trend over the full L/S range best supports the calculated TI values is K.
Compiled mortality and larval development responses from both the spiked eluates in this study and the previous test were K was diluted in natural brackish water, which was also used as control and dilution media in this study. The figure 3 clearly shows that, by slightly increasing the concentration concentration-response curves for the two experiments were almost identical for both mortality and larval development. As compared to the test with K dissolved in brackish water only, the LOEC (Lowest Observed Effect Concentration) was lower (35 mg/L instead of 105 mg/L) when K was dissolved in the L/S 3 eluate, which in itself was not toxic. However, the LOEC value for mortality was unaffected.
 

Reported statistics and error estimates:

Estimated TI for Cu indicated no significant risk of toxicity and the value of the index significantly declined from the L/S 1 to L/S 3. The only significant difference from a control group was on larval development in copepods exposed to the highest concentration (i.e. 40.5%) of the eluate generated at L/S 1.

Table 9. No Observed Effect Concentrations (NOEC) for Nitocra LDR together with calculated toxicity index (TI) at 40.5% dilution of leachates in tox-tests. Data for NOEC adopted from (Värmeforsk, 2011). NOEC (mg/l).

		Toxicitetsindex vid 40.5 % utspädning av lakvätska
	NOEC (mg/l	L/S 1	L/S 2	L/S 3
Al	0.09	0.90	0.88	0.94
Ca*	160	1.10	1.09	1.10
K*	35	1.62	0.51	0.06
Cu	0.06	0.54	0.34	0.27
Pb	0.27	0.00	0.00	0.00
Zn	0.24	0.05	0.05	0.04

* Värden för toxicitetsindex har justerats enligt Ekvation 1 med hänsyn till spädmedlets halt av ämnena. Spädmedlets halt har hämtats från Tab. 8 och har antagits vara ca 70 mg/L för K och 90 mg/L för Ca. Negativa toxicitetsindex indikerar att halten i testlösningen är lägre än i spädmedlet.

Table 8. Concentrations of compounds in leachates (sampled at L/S 1, 2 and 3) together with concentrations in solution used for dilution in the ecotox-tests. Samples were filtrated through 0.45 μm pore-filter prior analyses. For comparison, concentrations in natural water sampled at the location Landsortsdjupet in the Baltic See are also shown. This sample is assumed to be comparable with the origin of the water used for dilution (Askö).

Halt i lakvatten vid L/S 1, 2 och 3							Halt i spädningsmedia
		L/S 1 n=2	Stdv %	L/S2 n=2	Stdv %	L/S3	Spädningsmedia* ofiltrerad	Spädnings-media*	Stdv %	Landsort s-djupet**
Metaller
Ca	mg/l	526	2	520	2	525	89	96,8	4	92
Fe	mg/l	<0,007		<0,004		0,005	<0,004	0,0187	112	0,0016
K	mg/l	210	12	114,5	2	77,4	70,2	74,7	13	70,8
Mg	mg/l	62,8	2	44,45	1	38,2	224	237	6	223
Na	mg/l	1435	2	695	0	377	1940	1980	4	1847
Si	mg/l	2,855	3	2,845	3	2,63
Al	μg/l	200,5	1	196,5	5	209	5,68	10,4	50
As	μg/l	3,245	15	<2,7		<1	2,31	2,31	74
Ba	μg/l	42,4	1	32,5	4	29,5	38,4	37,4	2
Cd	μg/l	<0,095		<0,071		0,077	<0,05	<0,05		0,021
Co	μg/l	0,9945	12	0,826	26	0,592	<0,05	0,09	63
Cr	μg/l	6,215	1	3,32	0	2,11	<0,5	0,842	24
Cu	μg/l	79,4	16	50,3	3	39,3	2,67	3,55	13	0,55
Hg	μg/l	<0,02		<0,02		<0,02	<0,02
Mn	μg/l	15,85	2	25	3	27,4	2,5	1,59	82	1,7
Mo	μg/l	221,5	6	174	6	134
Ni	μg/l	5,41	4	5,09	25	3,41	5,86	7,14	25	0,69
Pb	μg/l	0,4115	10	<0,02		0,22	<0,20	0,24	24
Sb	μg/l	37,5	10	24,45	0	18,2
Se	μg/l	3,46	11	1,645	9	0,897
V	μg/l	1,05	0	1,055	6	0,888
Zn	μg/l	27,65	5	26,9	23	26,1	12,8	14,1	16	0,38
Övriga
Cl	mg/l	936	4	337	17	106	4730	3915	19
F	mg/l	<0,2		0,41		<0,2	<0,4
S-tot	mg/l	1220	1	822	12	741	185	188	6	161
SO4	mg/l	3470	5	2500	7	2180	661	550	17
DOC	mg/l	13,3	3	5,855	15	4,43	3,48	4,48	33	3 -5
DIC	mg/l	22,7	1	21,75	7	18,6				2***

*Data för spädmedium hämtad ur rapport av Värmeforsk (2011). Spädmedium är filtrerat och upphettat brackvatten hämtat utanför Stockholms universitets fältstation på Askö.

**Data gällande för Landsortsdjupet från Forsberg (2005) med undantag för DOC där data är hämtade från Österlund (2010) och är representativa för Bottniska viken och Gotlandsdjupet.

*** Förväntad halt i Havsvatten.

DGT-analysis

The precision of the DGT-measurements appears to be very good when the DGT samplers were exposed to the same eluate. Also when the samplers were exposed to eluate from two separate eluate replicates (data for L/S 1) the precision is good. For Fe, Cr and Pb the precision is in the latter case not so good. For Fe and Cr, the labile concentrations and the fractions of the measured total concentrations increase from L/S 1 to L/S 3. Also for the other selected elements there is a variation, but the changes are less pronounced as compared to Fe and Cr. Aluminium is not expected to bind to humic substances and the low labile fractions confirm earlier indications that Al measured in eluates from aged bottom ash at pH 8-9 are mainly present as colloids and are hence not bioavailable. The labile fraction for Cu only varies between 38 % and 45 %, although the measured total concentrations ranged from 58 to 178 mg/L. The analysis of DGT-samplers shows that labile concentrations and thus very likely also bioavailable fractions of different elements vary in different media. For Cu, which is one of the critical elements from an ecotoxicological point of view, the observed variation in labile fractions is relatively limited, ranging from 24 % to 45 %, but is still of such a magnitude that it could inflict on the bioavailability of the element in ecotoxicological tests.

Ecotoxicological testing

The only significant difference from a control group was on larval development in copepods exposed to the highest concentration (i.e. 40.5%) of the eluate generated at L/S 1. The calculated TI values (Table 9) indicated that there are four elements that could be responsible for the significant effects on larval development in the highest concentration of the L/S 3 eluate which are Ca, Al, Cu and K. The DGT-analysis shows that the bioavailability of Al is very low, and it is therefore unlikely that Al can explain the observed changes in toxicity from L/S 1 to 3. For Ca, on the other hand, the calculated TI values indicate that Ca is present at levels which should induce toxicity at all L/S ratios. However, these findings do not corroborate the findings of the toxicity of the eluates when exposed to N. spinipes. For Cu, none of the calculated TI values were above 1 but considering natural biological variation and that the TI value in the L/S 1 eluate was 0.56, the potential toxic contribution from copper cannot be neglected. Still, the element for which the toxicity trend over the full L/S range best supports the calculated TI values is K.

Compiled mortality and larval development responses from both the spiked eluates in this study and the previous test were K was diluted in natural brackish water, which was also used as control and dilution media in this study. The figure 3 clearly shows that, by slightly increasing the concentration concentration-response curves for the two experiments are almost identical for both mortality and larval development. As compared to the test with K dissolved in brackish water only, the LOEC (Lowest Observed Effect Concentration) is lower (35 mg/L instead of 105 mg/L) when K is dissolved in the L/S 3 eluate, which in itself was not toxic. However, the LOEC value for mortality was unaffected.

Validity criteria fulfilled:

yes

Remarks:

Yes, for all eluates the control mortality was well below 20%, which is the acceptable level for this test.

Conclusions:

Potassium (K), calcium (Ca) and aluminium (Al) were the most ecotoxicologically relevant metals. Of them, K was the most dominating. However, long-term environmental risks of K for N. spinipes were found to be negligible. Cu, Zn and Pb were found to be less problematic as eluating components. Notably, rapidly changing concentrations of K affected critical developmental stages of the crustacean N. spinipes. Although K seems to be responsible for the chief part of the toxicity, our observations indicate that the observed responses are in fact due to a cocktail effect.

Executive summary:

The Larval Development Ratio Test (LDR) was used to detect adverse effects on larval and juvenile development of the harpacticoid copepod N. spinipes in a non-GLP-study.

 

The project focused on the long-term environmental risks of bottom ash characterizing a relevant aged bottom ash (ca. 5 years old) from incineration. Leaching was performed using a sequential batch leaching approach from L/S 1 to L/S 3. The primary aim was to study if the ecotoxicity of aged bottom ash eluates decreased with increasing L/S ratio, and whether this could be confirmed with chemical analysis. The use of DGT-samplers in small-scale laboratory experiments seem to function well to trace changes in availability of cationic elements, and can be used to unravel mechanisms important for understanding observed toxic responses of complex pollution matrices, such as ash eluates.

 

As noted before in chemical and ecotoxicological investigations of ash, these findings show that K was dominating the toxicity of ash eluates. However, since the effects triggered by K in the crustacean N. spinipes were not lethal but rather related to difficulties to acclimatize to rapidly changing conditions, long-term environmental risks of K for this species were negligible. For characterization of ash, the toxic potential of K remained problematic.

Description of key information

Chronic toxicity of Ash was determined in a GLP compliant study according to OECD Guideline 211 (Daphnia magna Reproduction Test). Taking all biological parameters into account (survival, reproduction, length) and based on nominal loading concentrations, the overall NOEC of Ash was 3.2 mg/L, and the LOEC was 5.6 mg/L. The median effect concentration (EC50), based on reproduction was determined to be 7.0 mg/L with 95% confidence intervals 5.4 – 7.8 mg/L

Key value for chemical safety assessment

Fresh water invertebrates

Fresh water invertebrates

Effect concentration:

3.2 mg/L

Additional information

In a GLP compliant study according to OECD Guideline 211 (Daphnia magna Reproduction Test

NOEC of Ash was found to be 3.2 mg/L, and the LOEC was 5.6 mg/L. In a subchronic test with N. spinipes on aged bottom ash eluates the most ecotoxicologically relevant metals were K, Ca and Al. Instead, Cu, Zn and Pb did not cause concern as toxic components to N. spinipes. However, even K was found to be the most dominating element in the test eluates, long-term environmental risks of it for N. spinipes were found to be negligible.