Registration Dossier

Environmental fate & pathways

Adsorption / desorption

Currently viewing:

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Endpoint:
adsorption / desorption: screening
Type of information:
read-across based on grouping of substances (category approach)
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Remarks:
Well documented, scientifically sound study performed with methodology similar to OECD 106. No data on the behavior of tungsten trioxide in the environment are available. Adsorption data for tungsten metal are appropriate for read-across for this endpoint as the soluble species released are expected to be similar for each of the compounds, and are thus expected to behave similarly in the environment. For more details refer to the attached description of the read across approach.
Justification for type of information:
REPORTING FORMAT FOR THE CATEGORY APPROACH
1. HYPOTHESIS FOR THE ANALOGUE APPROACH: The hypothesis is that properties are likely to be similar or follow a similar pattern because of the presence of a common metal ion, in this case tungstate.
2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES):
Source: Tungsten metal
Target: Tungsten Carbide
3. CATEGORY APPROACH JUSTIFICATION: See Annex 1 in CSR
4. DATA MATRIX: See Annex 1 in CSR
Reason / purpose:
read-across: supporting information
Qualifier:
equivalent or similar to
Guideline:
OECD Guideline 106 (Adsorption - Desorption Using a Batch Equilibrium Method)
Deviations:
yes
Remarks:
-detailed characterization of the test substance was not provided
GLP compliance:
not specified
Type of method:
batch equilibrium method
Media:
soil
Specific details on test material used for the study:
Details on properties of test surrogate or analogue material (migrated information):
Not Applicable
Radiolabelling:
no
Test temperature:
25 degrees C
Analytical monitoring:
yes
Details on sampling:
- Concentrations: high (5 mg/L, nominal) and low (1 mg/L)
- Sampling interval: after 3 week contact time
- Sample storage before analysis: no data
Details on matrix:
COLLECTION AND STORAGE
- Geographic location: Eglin Airforce Base, Florida
- Collection procedures: no data
- Sampling depth (cm): no data
- Storage conditions: no data
- Storage length: no data
- Soil preparation (e.g.: 2 mm sieved; air dried etc.): no data


PROPERTIES OF SOIL MATRICES
The two soil types used were designated as "sand" and "clay". The "sand" sample was collected from the upper layer (A-horizon) of the Lakeland series soil. The "clay" soil sample was collected from the sandy clay loam sub surface layer (Bt-horizon) of the Lucy series. The Lakeland series "sand" soil was generally light tan in color. Quartz grains of sand size were the main component of the Lakeland series sample. These grains were sparingly coated by clays. The sample also contained approximately 1% organic fragments. The Lucy series soil was also dominated by quartz grains but these grains had more extensive clay coatings colored red by iron oxides.

PROPERTIES OF WATER MATRICES
Three types of artificial waters were used in the experiments: surface water, groundwater, and sea water.
See Table 1 in other information on Materials and Methods section for water characterization.
Details on test conditions:
TEST CONDITIONS
- pH: 4.5, 6.5, and 8.5
- Suspended solids concentration: no data



TEST SYSTEM
- Type, size and further details on reaction vessel: no data
- Water filtered (i.e. yes/no; type of size of filter used, if any): yes, no data
- Amount of soil/sediment/sludge and water per treatment (if simulation test): no data
- Soil/sediment/sludge-water ratio (if simulation test): no data
- Number of reaction vessels/concentration: one (per set of unique conditions)
- Measuring equipment: no data
- Method of preparation of test solution: metals were dissolved in the water matrices and then contacted with soil
- Other: The samples were periodically agitated to ensure equilibration of metals with soil. After a contact time of three weekes, the water was separated from the soil, filtered, and prepared for metal analysis.
Computational methods:
Scatter plots of Kd versus pH were used to evaluate the characteristic trends of each metal in different environments.
Adsorption and desorption constants:
The Kds for tungsten increased with decreasing pH. At pH 4.5 Kd values ranged approximately from 100-31,600; at pH 6.5 Kd values ranged from approximately from 10-3160; and at pH 8.5 Kd values ranged approximately from 3.16-100 (values presented in figure format in study).

In general, the trend for Kds as related to water type was as follows for tungsten: Kd(suface water)> Kd(ground water)>Kd(sea water), regardless of pH.

In general, the trend for Kds as related to soil type was as follows for tungsten: Kd(clay)> Kd(sand), regradless of pH.

There was no clear relationship between the intial metal concentration and the sorption coefficient over the concentration ranges tested, which may indicate that the range of initial metal concentrations tested falls on the linear part of the adsorption isotherm.

It should be noted that at Kd values greater than 100, the differences in values were significantly affected by the error in the analysis of the ver small concentration of metals remaining in solution.

Recovery of test material:
No data
Concentration of test substance at end of adsorption equilibration period:
Not explicitly provided
Concentration of test substance at end of desorption equilibration period:
Not applicable
Conclusions:
The Kds for tungsten increased with decreasing pH.
Executive summary:

No adsorption/desorption data of sufficient quality are available for tungsten carbide (target substance). However, terrestrial bioaccumulation data are available for tungsten metal (source substance), which will be used for reading across. Due to similar water solubility and toxicity for the target substance compared to the source substance, the resulting read across from the source substance to the target substance is appropriate. In addition, read across is appropriate because the classification and labelling is similar for the source substance than the target substance, the PBT/vPvB profile is the same, and the dose descriptors are, or are expected to be, lower for the source substance. For more details, refer to the read-across category approach included in the Category section of this IUCLID submission and/or as an Annex in the CSR.

Endpoint:
adsorption / desorption
Remarks:
other: partitioning in the environment
Type of information:
read-across based on grouping of substances (category approach)
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Remarks:
The sampling process was well documented and sampling, metal analysis, and media characterization was orchestrated under scientifically sound principles. No data on the behavior of tungsten trioxide in the environment are available. Adsorption data for tungsten metal are appropriate for read-across for this endpoint as the soluble species released are expected to be similar for each of the compounds, and are thus expected to behave similarly in the environment. For more details refer to the attached description of the read across approach.
Justification for type of information:
REPORTING FORMAT FOR THE CATEGORY APPROACH
1. HYPOTHESIS FOR THE ANALOGUE APPROACH: The hypothesis is that properties are likely to be similar or follow a similar pattern because of the presence of a common metal ion, in this case tungstate.
2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES):
Source: Tungsten metal
Target: Tungsten Carbide
3. CATEGORY APPROACH JUSTIFICATION: See Annex 1 in CSR
4. DATA MATRIX: See Annex 1 in CSR
Reason / purpose:
read-across: supporting information
Principles of method if other than guideline:
Tungsten partition coefficients for water-sediment were derived using paired environmental monitoring samples of tungsten in water and sediment collected in various parts of the EU.
GLP compliance:
no
Type of method:
other: adsorption desorption based on paired sediment-water samples
Media:
sediment
Radiolabelling:
no
Test temperature:
Ambient temperature
Details on sampling:
-875 samples taken
- Each stream sediment sample was comprised of material taken from 5-10 points over a stretch of 250-500 m along the stream
- Sites were located at least 100 m upstream of roads and settlements.
-Sampling was started from the stream water sampling point, and the other sub-samples were collected up-stream. A composite sediment sample was made from sub-samples taken from beds of similar nature.
-Running stream water was collected from the small, second order, drainage basins (<100 km2) at the same site as the stream sediment.
-In dry terrain, such as Southern Europe, streams had no running water for most of the year. Hence, the sampling, whenever possible, was carried out during the winter and early spring months.
-Sub-samples of stream water were separately collected from each site: unfiltered water for major IC ion analysis, filtered water (0.45 μm) for ICP-MS and ICP-AES analysis, and DOC analysis.
-Sampling during rainy periods and flood events was avoided. The water sample was collected from the first, lowermost stream sediment sampling point.
-The pH and EC were measured, and alkalinity estimated by titration at the site.
Phase system:
other: Kd sediment-water value
Type:
other: Kd sediment-water value
Value:
28 395
Remarks on result:
other: 10th percentile
Phase system:
other: Kd sediment-water value
Type:
other: Kd sediment-water value
Value:
140 000
Remarks on result:
other: median
Phase system:
other: Kd sediment-water value
Type:
other: Kd sediment-water value
Value:
700 000
Remarks on result:
other: 90th percentile
Adsorption and desorption constants:
Kds were not specified in this study but data from this study were used to calculate Kd sediment-water values: (see the statistical derivation in endpoint summary) 28,395 (10th percentile), 140,000 (median), 700,000 (90th percentile)
Conclusions:
Using the data presented in this study, the Kd sediment-water, as calculated from paired stream sediment and stream water samples were as follows: 28,395 (10th percentile), 140,000 (median), 700,000 (90th percentile)
Executive summary:

No adsorption/desorption data of sufficient quality are available for tungsten carbide (target substance). However, terrestrial bioaccumulation data are available for tungsten metal (source substance), which will be used for reading across. Due to similar water solubility and toxicity for the target substance compared to the source substance, the resulting read across from the source substance to the target substance is appropriate. In addition, read across is appropriate because the classification and labelling is similar for the source substance than the target substance, the PBT/vPvB profile is the same, and the dose descriptors are, or are expected to be, lower for the source substance. For more details, refer to the read-across category approach included in the Category section of this IUCLID submission and/or as an Annex in the CSR.

Endpoint:
adsorption / desorption
Remarks:
adsorption
Type of information:
read-across based on grouping of substances (category approach)
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Remarks:
Well-documented, scientifically sound study that uses acceptable methodology, although the study has significant deviations from standard batch equilibrium guidelines. Only limited data on the behavior of tungsten metal in soil are available; therefore, read-across to sodium tungstate adsorption are used for read-across for this endpoint. Read-across is approriate for this endpoint as the soluble species released are expected to be similar for each of the compounds, and are thus expected to behave similarly in the environment. For more details refer to the attached description of the read across approach.
Justification for type of information:
REPORTING FORMAT FOR THE CATEGORY APPROACH
1. HYPOTHESIS FOR THE ANALOGUE APPROACH: The hypothesis is that properties are likely to be similar or follow a similar pattern because of the presence of a common metal ion, in this case tungstate.
2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES):
Source: Sodium Tungstate
Target: Tungsten Carbide
3. CATEGORY APPROACH JUSTIFICATION: See Annex 1 in CSR
4. DATA MATRIX: See Annex 1 in CSR
Reason / purpose:
read-across: supporting information
Qualifier:
no guideline followed
Principles of method if other than guideline:
In this study, solubilization and sorption of various tungstate species with a model soil was investigated in four sets of experiments. All leachates were processed through a 0.45-um filter to obtain dissolved constituents and removed suspended soil particles.

In the first set of experiment, to determine the effect of polymerization on sorption and thus dissolved phase concentrations, the partition coefficients for four commercially available tungsten compounds (sodium tungstate dihydrate, sodium phosphotungstate, sodium polytungstate, and tungstosilicic acid n-hydrate) on clean test soil in deionized water matrix were determined. All sorption partition coefficient studies had a minimum equilibration time of 3 days, with some experiments performed to 4 months.

In a second experiment, the effects of pH, phosphate (at various pH values), ionic strength, and humic acid on the tungstate Kd value were determined after an equilibration of 14-days.

In a third experiment, partition coefficients for the two most abundant metal oxides in the Grenada Loring soil (Al and Fe) were determined to elucidate tungstate’s affinity for pure metal oxides.

In order to examine the affect of soil-aging on the speciation of extractable, soluble tungsten compounds a fourth experiment was performed. In this experiment, the leachate obtained from extracting tungsten metal-spiked soil, after the soil had been aged for one year, with deionized water for 168 hrs. The leachate solution was then used as the spike solution for clean Grenada Loring soil to determine the effective partitioning for the tungsten species (undetermined) extractable from the soil.

GLP compliance:
no
Type of method:
batch equilibrium method
Media:
soil
Radiolabelling:
no
Test temperature:
no data
Analytical monitoring:
yes
Details on sampling:
- Sampling interval: equilibration times of 3 days or 4 months
- Sampling storage: up to 1 year
- Sample storage before analysis: no data
Details on matrix:
COLLECTION AND STORAGE
- Geographic location: Grenada Loring Soil was collected from the Brown Loam Experimental Station (Learned, MS)
- Collection procedures: The silty loam soil was collected with a front-end loader after the top 12 cm were removed to eliminate unwanted vegetation.
- Sampling depth (cm): > 12 cm
- Storage conditions: no data
- Soil preparation: < 1 cm sieved

PROPERTIES
- Soil texture
- % sand: 3
- % silt: 72
- % clay: 26
- Horizon: no data
- Soil taxonomic classification: no data
- Soil classification system: no data
- Soil series: no data
- Soil order: no data
- pH: 6.7
- Organic carbon (%): 0.7
- CEC (meq/g): 0.075 (cation), 0.025 (anion)
- Carbonate as CaCO3: no data
- Insoluble carbonates (%): no data
- Extractable Cations: no data
- Special chemical/mineralogical features: no data
- Clay fraction mineralogy: no data
- Moisture at 1/3 atm (%): no data
- Bulk density (g/cm3): no data
- Biomass (e.g. in mg microbial C/100 mg, CFU or other): no data
Details on test conditions:
TEST CONDITIONS
- Buffer: no data
- pH: no data
- Suspended solids concentration: no data
- Other: no data

TEST SYSTEM
- Type, size and further details on reaction vessel: no data
- Measuring equipment: Tungstate analysis was performed using an Agilent (Palo Alto, CA) 1100 HPLC interfaced to the PerkinElmer Elan 6000 ICP–MS with a cross-flow pneumatic nebulizer.
- Test performed in closed vessels due to significant volatility of test substance: no
- Test performed in open system: yes
Computational methods:
The Kd values were determined following the linear Freundlich model with associated linear fit correlation.
Key result
Sample No.:
#1
Type:
Kd
Value:
284 L/kg
Remarks on result:
other: 3-days
Key result
Sample No.:
#2
Type:
Kd
Value:
845 L/kg
Remarks on result:
other: 4-months
Adsorption and desorption constants:
Adsorption coefficient (Kd) for tungstate in non-aged soil, extraction with deionized water:
3 d: 284 L/kg
4 mo: 845 L/kg
Details on results (Batch equilibrium method):
In the experiment designed to examine the effects of tungstate polymerization on the Kd value, tungstate showed a marked increase in Kd values over the 4 month study, (Kd 3 days= 284 L/kg; Kd 100 days = 845 L/kg) suggesting that it interacts with the soil to become less mobile with time, either through precipitation or sorption mechanisms or further changes in speciation. Kds for the other compounds tested were lower with the 3-day and 100-day Kds ranging from 92-112 L/kg and 97-500 L/kg, respectively. In the experiment designed to examine the effects of tungstate polymerization on the Kd value, phosphotungstate showed a marked increase in Kd values over the 4 month study (112 and 500 L/kg at 3 d and 4 mo, respectively), suggesting it interacts with the soil to become less mobile with time, either through precipitation or sorption mechanisms or further changes in speciation. The polytungstate and tungstosilicate compound Kd values remained relatively unchanged over the 4-month Kd study (92 and 135 L/kg and 103 and 97 L/kg for polytungstate and tungstosilicate at 3 d and 4 mo, respectively), suggesting that they do not undergo significant geochemical changes in this system over the time period studied.

In the second experiment, which determined the effects of pH, phosphate, ionic strength and humic acid content on the tungstate Kd, the data suggest that increased pH (either from sodium hydroxide or phosphate, both at pH = 11) decreased the Kd value (the Kd values at pH 3, 7 and 11 were 414, 141, and 62 L/kg, respectively, compared to 673 L/kg for deionized water), as would be expected with an anionic compound (i.e. surface exchange sites become negatively charged). Humic acid and sodium sulfate as concomitant compounds had little effect on tungstate Kd values in comparison to the deionized water system (863 L/kg for Na2SO4 pH=7.0, and 749 L/kg for 10 mg/L humic acid). This was likely due to the fact that tungstate has a higher affinity for sorption sites than these compounds, which may not be the case with phosphate, or the humic substance behaves as a surfactant to the soil particle surface. Additionally, substantial amounts of naturally occurring salts will dissolve from the soil into the deionized water system, essentially making the ionic strength of the deionized water system similar to the sodium sulfate and humic acid matrices. Because the Grenada Loring soil has substantial amounts of oxide-forming metals (1.1% Al, 1.5% Fe 0.06% Mn), it is likely that some combination of these phases will influence tungsten sorption processes.

In order to elucidate tungsten's affinity for pure metal oxides, the Kd values were obtained for the most abundant metal oxides in the soil. Aluminum and iron (III) oxides have tungstate Kd values of approximately 1970 and 1780 L/kg, respectively, using the same procedure previously described for the soil experiments. In comparison, arsenic oxyanion compounds have partition coefficients on freshly precipitated aluminum, iron, and manganese oxyhydroxides that range from <1 to over 200 L/kg depending on pH conditions, suggesting tungstate sorption to metal oxide phases will be extensive in this system.

The the overall Kd value for all dissolved, extractable tungsten species obtained using leachate from the extraction of aged, tungsten-metal spiked soil, after 3 days of equilibration was 110 L/kg (R2 = 0.98), which is almost a factor of 3 lower than the value obtained for tungstate, but close to the values obtained for polytungstate and tungstosilicate 284 L/kg for tungstate, 92 L/kg for polytungstate, and 103 L/kg for tungstosilicate) obtained in the first set of experiments. The Kd value measured for the geochemical form of tungstate in the tungsten spiked soil closely matches those of commercially available polymer tungsten compounds, indicating that a large portion of soluble tungsten in the Grenada Loring soil exists as polymeric species.

Elemental analysis of the soil yielded the following concentrations in mg/kg: aluminum (10,900); barium (81); calcium (925); iron (14,900); potassium (778); magnesium (1336); manganese (660); sodium (85); phosphorous (455); sulfur (172); titanium (37); and tungsten (12). In the tungsten metal-spiked soil the tungsten content was 7080 mg/kg.

Conclusions:
When extracted in deionized water, the mobility of tungstate in the soil decreased from 3 days to 4 months with Kd values increasing from 284 to 845 L/kg. Kd values of tungstate (measured after 14-days) decreased with increasing pH with values of 414, 141 and 62 L/kg at pH 3.0, 7.0 and 11.0; however, in the presence of sodium sulfate and humic acid the Kd values were similar to that obtained for deionized water (673 L/kg), 863 and 749 L/kg, respectively, likely due to the fact that tungstate has a higher affinity for sorption sites than these compounds. For the tungsten species extracted from the tungsten metal-spiked soil that had aged for 1 year, the overall Kd value obtained after 3 days of equilibration was 110 L/kg.
Executive summary:

No adsorption/desorption data of sufficient quality are available for tungsten carbide (target substance). However, adsorption/desorpion data are available for sodium tungstate (source substance), which will be used for reading across. Due to lower water solubility and lower toxicity for the target substance compared to the source substance, the resulting read across from the source substance to the target substance is appropriate as a conservative estimate of potential toxicity for this endpoint. In addition, read across is appropriate because the classification and labelling is more protective for the source substance than the target substance, the PBT/vPvB profile is the same, and the dose descriptors are, or are expected to be, lower for the source substance. For more details, refer to the read-across category approach included in the Category section of this IUCLID submission and/or as an Annex in the CSR.

Endpoint:
adsorption / desorption: screening
Type of information:
read-across based on grouping of substances (category approach)
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Remarks:
Well documented study similar to OECD Test Guideline 106 with acceptable deviations. Limited data on the behavior of tungsten metal in soil are available; therefore, read-across to sodium tungstate adsorption are used for read-across for this endpoint. Read-across is approriate for this endpoint as the soluble species released are expected to be similar for each of the compounds, and are thus expected to behave similarly in the environment. For more details refer to the attached description of the read across approach. No GLP-compliance information was included in the peer review publication, however the study has adequate and reliable coverage of the key parameters foreseen to be investigated in the corresponding test methods and adequate and reliable documentation of the study is provided.
Justification for type of information:
REPORTING FORMAT FOR THE CATEGORY APPROACH
1. HYPOTHESIS FOR THE ANALOGUE APPROACH: The hypothesis is that properties are likely to be similar or follow a similar pattern because of the presence of a common metal ion, in this case tungstate.
2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES):
Source: Sodium Tungstate
Target: Tungsten Carbide
3. CATEGORY APPROACH JUSTIFICATION: See Annex 1 in CSR
4. DATA MATRIX: See Annex 1 in CSR
Reason / purpose:
read-across: supporting information
Qualifier:
equivalent or similar to
Guideline:
OECD Guideline 106 (Adsorption - Desorption Using a Batch Equilibrium Method)
Deviations:
yes
Remarks:
: 1) CaCl2 solution not used; 2)samples were settled gravimetrically and not by centrifugation
GLP compliance:
not specified
Remarks:
The study has adequate and reliable coverage of the key parameters foreseen to be investigated in the corresponding test methods and adequate and reliable documentation of of the study is provided.
Type of method:
batch equilibrium method
Media:
soil
Radiolabelling:
no
Test temperature:
no data
Analytical monitoring:
yes
Details on sampling:
- Concentrations: 5, 10, 15, 20, 25, 50, 75, 100, 150 and 200 mg metal/L
- Sampling interval: 1, 20, 40, 60, 80 and 100 days
- Sample storage before analysis: no data
Details on matrix:
COLLECTION AND STORAGE
- Geographic location: no data
- Collection procedures: no data
- Sampling depth (cm): no data
- Storage conditions: no data
- Storage length: no data
- Soil preparation: no data

PROPERTIES
- Soil texture
- % sand: 81.9, 90.40 and 77.20 for soils SM(1), SM(2) and SM(3), respectively
- % silt: 18.10, 9.60 and 22.30 for soils SM(1), SM(2) and SM(3), respectively
- % clay: no data
- Horizon: no data
- Soil taxonomic classification: no data
- Soil classification system: no data
- Soil series: no data
- Soil order: no data
- pH: 5.74, 4.85 and 5.50 for soils SM(1), SM(2) and SM(3), respectively
- Organic carbon (%): 0.89, 0.57 and 0.83 for soils SM(1), SM(2) and SM(3), respectively
- CEC (meq/100 g): 4.50, 4.30 and 6.50 for soils SM(1), SM(2) and SM(3), respectively
- AEC (meq/100 g): 5.03, 2.69 and 4.25 for soils SM(1), SM(2) and SM(3), respectively
-BET surface area: 2.30, 1.98, and 2.80 for soils SM(1), SM(2) and SM(3), respectively
- Carbonate as CaCO3: no data
- Insoluble carbonates (%): no data
- Extractable Cations (Ca, Mg, Na, K, H): no data
- Special chemical/mineralogical features: no data
- Clay fraction mineralogy: no data
- Moisture at 1/3 atm (%): no data
- Bulk density (g/cm3): no data
- Biomass (e.g. in mg microbial C/100 mg, CFU or other): no data.
Details on test conditions:
TEST CONDITIONS
- Buffer: no data
- pH: no data
- Suspended solids concentration: no data
- Other: no data

TEST SYSTEM
- Type, size and further details on reaction vessel: no data
- Water filtered: no data
- Number of reaction vessels/concentration: 3
- Are the residues from the adsorption phase used for desorption: no
Computational methods:
- Adsorption and desorption coefficients (Kd): no data
- Freundlich adsorption and desorption coefficients: no data
- Slope of Freundlich adsorption/desorption isotherms: no data
- Adsorption coefficient per organic carbon (Koc): no data
- Regression coefficient of Freundlich equation: yes, these correlation coefficients were generated
- Other: A least squares fit was performed to determine the sorptive parameters of each soil-metal system according to the power law relation: Cs = KdCw^(1/n), where Kd is the distribution coefficient, and n is a constant specific to the soil-metal system. For each soil-metal system, the distribution coefficient and exponent were determined by nonlinear regression according to the Marquardt-Levenberg algorithm, using the software package Sig-maplot.
Key result
Sample No.:
#1
Type:
Kd
Value:
16.6 dimensionless
Remarks on result:
other: SM1: Kd 1-day
Key result
Sample No.:
#2
Type:
Kd
Value:
27.1 dimensionless
Remarks on result:
other: SM2: Kd 1-day
Key result
Sample No.:
#3
Type:
Kd
Value:
75.8 dimensionless
Remarks on result:
other: SM3: Kd 1-day
Key result
Sample No.:
#4
Type:
Kd
Value:
325.8 dimensionless
Remarks on result:
other: SM1 Kd 100-days
Key result
Sample No.:
#5
Type:
Kd
Value:
167.9 dimensionless
Remarks on result:
other: SM2 Kd 100-days
Key result
Sample No.:
#6
Type:
Kd
Value:
181.8 dimensionless
Remarks on result:
other: SM3: Kd 100-days
Adsorption and desorption constants:
SM1: Kd 1-day = 16.6 ± 2.1; Kd 100-day = 325.8 ± 20.6
SM2: Kd 1-day = 27.1 ± 5.5; Kd 100-day = 167.9 ± 8.7
SM3: Kd 1-day = 75.8 ± 5.5; Kd 100-day = 181.8 ± 13.5

Soil

Equilibration

Kd(including 95% CI)

n* (including 95% CI)

Freundlichisotherm R2

Kd100/Kd1

SM1

1 day

16.6 ± 2.1

2.11 ± 0.11

0.9551

19.6

 

100 days

325.8 ± 20.6

2.87 ± 0.14

0.97

 

SM2

1 day

27.1 ± 5.5

3.18 ± 0.40

0.736

6.2

 

100 days

167.9 ± 8.7

2.85 ± 0.10

0.9854

 

SM3

1 day

75.8 ± 5.5

2.79 ± 0.12

0.9722

2.4

 

100 days

181.8 ± 13.5

3.91 ± 0.26

0.9374

 

 Table 1. Freundlich distribution coefficients and exponents for three soil-tungsten systems

Tungsten Kd100 results indicate that in some systems a longer equilibration time will produce a Kd that more closely predicts the behavior of the metals in the soil under equilibrium conditions. In this case, large increases in the distribution coefficient were observed between one and 100 days. The ratios of Day 100 to Day 1 distribution coefficients were 19.6, 6.2 and 2.4 for tungsten in SM1, SM2 and SM3, respectively.

For each of the soil–tungsten systems studied the 20-day distribution coefficient is significantly greater than the one-day distribution coefficient (>95% confidence). Following a 100-day equilibration, the observed distribution coefficients are not significantly greater than the 80-day distribution coefficient (Kd 80), though the Kd 100 is significantly (>95% confidence) greater than the Kd 60. This indicates that equilibrium may have been reached by the 100-day mark, but not before.

The major reactions governing the transfer of the dissolved W from the aqueous phase to the solid phase include: sorption–desorption reactions, dissolution–precipitation as a function of pH, and dissolution–precipitation as a function of redox environment. The extent to which these reactions occur varies, and is largely dependent on specific site conditions. This geochemical dependence is demonstrated by the varied increases in W Kd over the 100-day period for the three SM soils in this study.

Conclusions:
Tungsten requires a long stabilization period, as evidenced by the comparison of Freundlich distribution coefficients over a 100-day period.
Executive summary:

No adsorption/desorption data of sufficient quality are available for tungsten carbide (target substance). However, adsorption/desorpion data are available for sodium tungstate (source substance), which will be used for reading across. Due to lower water solubility and lower toxicity for the target substance compared to the source substance, the resulting read across from the source substance to the target substance is appropriate as a conservative estimate of potential toxicity for this endpoint. In addition, read across is appropriate because the classification and labelling is more protective for the source substance than the target substance, the PBT/vPvB profile is the same, and the dose descriptors are, or are expected to be, lower for the source substance. For more details, refer to the read-across category approach included in the Category section of this IUCLID submission and/or as an Annex in the CSR.

Description of key information

The following partitioning coefficients were statistically derived based on studies using appropriate methodology:
Kd soils (Griggs et al., 2009 and Bednar et al., 2008):
10th percentile: 44 L/kg
Median: 174 L/kg
90th percentile: 692 L/kg
Kd sediment (Salminen (ed.) et al., 2005):
10th percentile: 28,395 L/kg
Median: 140,000 L/kg
90th percentile: 700,000 L/kg

Statistical derivation of soil-water partitioning coefficient for forwarding on to risk assessment:

Two tungsten partitioning and mobility studies (Griggs et al., 2009 and Bednar et al., 2008) were selected to derive soil partition coefficients for tungsten compounds. These studies provide Kd estimates with various soil and solution characteristics, and therefore provide a range of values that may be found in the environment. In addition, they met the quality characteristics specified in the REACH guidance 7.13-2 and followed methods equivalent to the suggested test method (OECD 106 batch equilibrium method). They are briefly summarised below.

Griggs et al., 2009:The authors measured Kd values using the batch sorption method. Initial batch slurries were prepared with 10 g of soil and 100 mL of sodium tungstate solution. Soil-metal systems were prepared in triplicate using each of three natural silty sand soils and each of 10 stock concentrations. Supernatant was sampled at 24 hours and at 100 days. The Freundlich model was used. The authors observed dynamic sorptive behavior in tungsten and suggest that in Kd studies, a longer equilibration time may provide a more accurate reflection of tungsten mobility in subsurface environments.

Bednar et al., 2008:Geochemical parameters are determined for tungstate species in a model soil that describe the potential for tungsten mobility. The Freundlich model was used. Soluble tungsten leached from a metallic tungsten-spiked soil after six to twelve months aging reached an equilibrium concentration >150 mg/L within 4 h of extraction with deionized water. Partition coefficients (Kds) determined for various tungstate and polytungstate compounds in the model soil suggest a dynamic system in which speciation changes over time affect tungsten geochemical behavior. Partition coefficients for tungstate and some poly-species have been observed to increase by a factor of 3 to 6 over a four-month period, indicating decreased mobility with soil aging.

The Kd values from these studies are presented in Table 4.3. A cumulative distribution function of the data is presented in Figure 1. In accordance with the REACH guidance 7.13-2, a probability distribution was fit to the data using statistical software (SYSTAT, version 12). The best fitting distribution was a lognormal (5.160,1.076). The goodness of fit statistic for this distribution (Shapiro Wilks) was 0.941 (p= 0.277), indicating that a lognormal distribution cannot be rejected. The median and percentiles of the distribution are shown in the table below. Non-parametric estimates were also calculated for the sake of comparison and also appear in the table below.

The lognormal results of Median=174, 10thpercentile=44, 90thpercentile =692 will be passed forward to the risk assessment.

A third study, Meijer et al., 1998, was considered for incorporation, but the measured Kds were only presented on charts from which they could not be easily extracted. Therefore, they could not be included in the cumulative distribution analysis. However, the extrapolated median summary Kds that were reported by Meijer et al. (200 L/kg for the surface soil zone at pH=5, 100 L/kg for the vadose zone at pH=5.5 and 100 L/kg for the saturated zone at pH=6) are in agreement with our results.

Statistical derivation of sediment-water partitioning coefficient for forwarding on to risk assessment:

Sediment partition coefficients (Kd) were derived for tungsten using paired field-measured concentrations in sediment and water from locations throughout Europe. Stream sediment and water concentrations were obtained from the FOREGS Geochemical Baseline Database (Salminen (Ed.) et al., 2005) (http://www.gsf.fi/foregs/geochem/). The sediment and water samples were collected simultaneously, allowing the measurements to be paired by the location identifiers in the database. This resulted in a dataset of 800 paired sediment and water concentrations. Sediment partitioning coefficients were calculated for each data pair as follows:

Kd=Cs/ Caq

where Cs = total concentration of test substance in the solid phase (mg/kg) and Caq = concentration of test substance in the aqueous phase (mg/L).

The detection limit (DL) for tungsten in water is reported in the database as 0.002 µg/L, and the accompanying documentation states that non-detected (ND) concentrations were reported at half the detection limit (0.001 µg/L). Of the 800 data pairs, 106 included non-detected water concentrations. Being in the denominator of the equation, small differences in the reported water concentrations have a large effect on the Kd estimates. Therefore, assessment of the distribution of the Kd values was conducted both including and excluding the 106 pairs, where ½ DL (0.001 µg/L) was used for the water concentration.

The cumulative distribution function of the 800 pairs is shown in Figure 2. This distribution could not be readily fit to any of the commonly used continuous probability distributions (normal, lognormal, etc.), however due to the large number of data pairs, it is very well characterized, and therefore it is appropriate to report empirically derived percentiles for this distribution.

Due to the additional uncertainty posed by the substitution of half the detection limit for water concentrations, it is suggested that the percentiles estimated from the smaller dataset (N=694) with the NDs excluded are the most appropriate to use for the risk characterization.The following results will be passed forward to the risk assessment: Median=140000,10thPercentile= 28395,90thPercentile=700000.

Key value for chemical safety assessment

Koc at 20 °C:
0

Other adsorption coefficients

Type:
log Kp (solids-water in sediment)
Value in L/kg:
28 395
at the temperature of:
25 °C

Other adsorption coefficients

Type:
log Kp (solids-water in soil)
Value in L/kg:
44
at the temperature of:
25 °C

Additional information

No data on the behavior of tungsten carbide in the environment are available. Adsorption data for tungsten metal and sodium tungstate are presented in this section. The soluble species released are expected to be similar for each of the compounds, and are thus expected to behave similarly in the environment. However, the amount of soluble species resulting from tungsten metal and sodium tungstate is different, with sodium tungstate being much more soluble. Therefore, data for sodium tungstate and tungsten metal are expected to adequately capture the range of mobility of tungsten carbide in the environment.

[LogKoc: 0.0]