Slags, steelmaking, converter

Water solubility

Description of key information

Slags are inorganic solid UVCB. Consequently, single data cannot represent the complex and varying processes leading to the release of traces of analytes from the solid material.
The key values for release of ions from ferrous slags are estimated to be 
heavy metals: insoluble 0.01 mg/l
calcium: moderately soluble (0.1-600 mg/L)
other alkaline and alkaline earth elements, aluminium: slightly soluble (0.1-100 mg/L)
Sulfate and thiosulfate: slightly to moderately soluble (0.1-1000 mg/L) 
The solubility of slag analytes decreases during aging of slag.

Key value for chemical safety assessment

Water solubility:

0.01 mg/L

at the temperature of:

20 °C

Additional information

Concentrations of important analytes in leachates (L/S 10/1, DIN 38414-4) of ferrous slags used for testing of ecological or toxicological properties of ferrous slags (Table compiled from 2009 and 2010 toxicological and ecotoxicological testing)

Leached substance	ABS	GGBS	BOS	EAF C	EAF S	SMS
Tested grain size (mm)	8 - 11	< 0.1	8 - 11	8 - 11	0 - 10	< 0.09
pH	11.2	11.6	11.8	11.4	12.2	11.9
Conductivity μS/cm	570	852	1695	599	5130	2060
COD (mg/L)	32	< 15			< 15	< 15
As (mg/L)	< 0.002	< 0.002	< 0.002	< 0.002	< 0.002	0.005
Ba (mg/L)	0.120	0.05	0.055	0.353	0.193	0.930
Ca (mg/L)	59.1	92.3	49.7	49.7	549	249
Cd (mg/L)	< 0.0005	< 0.0005	< 0.0005	< 0.0005	< 0.0005	< 0.0005
Co (mg/L)	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Cr total (mg/L)	< 0.001	0.002	< 0.001	< 0.001	0.005	0.004
Cr(VI) (mg/L)	< 0.010	< 0.010	< 0.010	< 0.010	< 0.010	< 0.010
Cu (mg/L)	< 0.002	< 0.002	0.010	< 0.002	< 0.002	< 0.002
Fe (mg/L)	0.010	< 0.01	< 0.010	< 0.010	< 0.010	0.059
Hg (mg/L)	< 0.0002	< 0.0002	< 0.0002	< 0.0002	< 0.0002	< 0.0002
Mn (mg/L)	0.006	< 0.0005	0.003	0.001	0.0009	0.015
Mo (mg/L)	0.007	0.003	0.058	0.020	0.047	0.008
Ni (mg/L)	< 0.002	< 0.002	< 0.002	< 0.002	< 0.002	< 0.002
Pb (mg/L)	< 0.002	< 0.002	< 0.002	< 0.002	0.003	< 0.002
Se (mg/L)	< 0.005	< 0.005	< 0.005	0.005	< 0.005	0.027
Tl (mg/L)	< 0.0005	< 0.0005	< 0.0005	< 0.0005	< 0.0005	< 0.0005
V (mg/L)	0.014	< 0.005	0.091	0.234	< 0.002	0.009
Zn (mg/L)	< 0.005	< 0.005	0.020	< 0.005	< 0.005	0.053
F (mg/L)	0.6	0.6	< 0.4	< 0.4	2.4	< 0.4
Cl (mg/L)	< 1.0	20	1.0	< 1.0	4.0	< 1.0
SO4 (mg/L)	52	8	3.0	3.0	134	< 1.0

Overall solubility

Only a small part of the inorganic material contained in slags is released (Larm 1998).

pH of leachates of ferrous slags

The pH of slag leachates is basic (pH 11.2 -12.2). The pH value may be due to

• firstly to free calcium oxide (Larm 1998)
• secondly to other alkaline and alkaline earth element oxides e.g. oxides of magnesium, potassium and sodium
• thirdly, under more acidic conditions, some silicates disintegrate and release e.g. calcium ions. As silicic acid is a week acid, dissolution of several calcium silicates leads to pH increases.
• fourthly aged slags become covered with a layer of carbonates which, as salts of a weak acid and strong bases (in the case of alkaline earth elements), form basic solutions.

Conductivity

The conductivity of slag leachates is mostly caused by the high mobility of the OH ions which are predominantly a result of the presence of free alkaline earth oxides/hydroxides

Heavy metals

Heavy metals occur, if at all, only in traces in ferrous slag leachates because the heavy metals are bound into crystal matrices (e.g. spinel) (Tossavainen and Lind 2005, Tossavainen et al. 2005) or absorbed in glass phase(s).

Sulfur

Leachates of blast furnace slags may contain sulfate, thiosulfate, sulfide and polysulfides in the form of their calcium, sodium, and potassium salts. The concentration of reduced sulfur ions in the leachates decreases in the time course of leaching and in the presence of oxygen (LECES 1991).

Thiosulfate and sulfate were detected in the leachates of ABS. Thiosulfate (reported concentrations were calculated as sulfate) decreased from > 9 g/L to approximately 0.1 g/L after a dozen leaching cycles. Sulfate decreased from starting levels at approximately 1 g/L to approximately 0.1 g/L within 5 leaching cycles. Other sulfur compounds (e.g. sulfides) were not detectable (no limit of detection available). In outdoor lysimeter experiments sulfate and thiosulfate concentrations were one order of magnitude less, e.g. the leaching rates of thiosulfate from ABS decrease sharply from 320 mg/L to approximately 10 mg/L within some 5 months whereas sulfate leaching rates were only slightly decreased during 20 months of outdoor exposure (FEhS 2002).

The sulphate concentration was 0.168 g/L in ABS leachates, 0.030 g/L in GBS leachates, 0.070 g/L in SMS, and less than 0.02 g/L in the other ferrous slags (Geiseler 1998).

Halides

Only a small part of the trace substances present in solid slag is leached. F concentrations in leachates attain some mg/L in BOS and SMS (Geiseler 1998, 2000)

Cyanide

Several measurements were performed but cyanide could not be detected (Larm 1998)

Organics

Due to the formation of ferrous slags at temperatures > 1000 °C, formation of organics is limited and expected to be very low. Consecutively, the concentrations of organics in slag leachates are expected to be negligible. However, to increase the possibility of detection, TOC, hydrocarbons, EOX, sum of BTEX (benzene, toluene, ethylbenzenes, xylenes), sum of long chain hydrocarbons, sum of 16 PAH compiled by the US EPA (naphthalene, dihydroacenaphthalene, acenaphthalene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benz(a)anthracene, chrysene, benz(b)fluoranthene, benz(k)fluoranthene, benz(a)pyrene, dibenz(a,h)anthracene, benz(g,h,i)perylene and indeno(1,2,3,c,d)pyrene), and sum of 6 PCB (polychlorinated biphenyls) were measured in solid slags as a worst case for slag leachates. In general, no organics were detected (Oekoinstitut 2007).

COD (chemical oxygen demand)

The COD is due to sulfide and thiosulfate in the slags and is relevant only for cooled blast furnace slag. These sulfur compounds are oxidised in the COD test to yield sulfate (Larm 1998).

Comparison of leaching methods

For the assessment of inorganic (waste) materials the DIN 38414-4 DEV S4 leaching method was compared with pH4 stat method. The later simulates worst case conditions of acidic rain and yields the acid neutralization capacity (ANC).

Under the acidic conditions of the pH4 stat method, the mobility of the heavy metals is significantly higher than measured by the DIN 38414-4 DEV S4 method. However, it is possible to use natural rock and glasses containing heavy metals to make a comparative assessment of the heavy metal leachability of the inorganic matrix.

For most materials a very tight binding of heavy metals is observed which is similar to the properties of inert glasses. This is especially valid for Cr in ferrous slags. In most materials, lead is more mobile (leachable) than As, Ba, Cd, Co, Cr, Cu, Mn, Ni, Sb and Zn which are classified as less mobile.

The mobility of these trace elements can be controlled by the production method of the material e.g. slow or rapid cooling for the production of crystalline or vitreous phases, respectively. There are also differences between the leaching methods: In general, the DIN 38414-4 DEV S4 method yields lower heavy metal concentrations in the leachate of crystalline slags, the pH4 stat method of vitreous materials.

The mobility of heavy metals as measured by the pH4 stat method depends on the acid used for pH adjustment. The use of sulfuric acid instead of nitric acid increases the mobility of heavy metals as sulfate leads to an increased destruction of silicates. In contrast, the leaching of lead and barium is decreased by sulfuric acid as their insoluble sulfates are formed.

The grain size used in the pH4 stat method is of minor importance for the results of the leaching experiments. However, a comparable grain size distribution should be used, e.g. less than 10 mm.

A multivariate data analysis including data clusters was performed which suggests a structure for both leaching methods. For hydraulic engineering stones it was shown that the data of the 15 elements included in this study, can be reduced to 3 characteristic factors. For other construction materials, there are 2 factors each sufficient for the almost complete description of their properties. An assessment of these factors enables the recognition of distinct types as well as the classification of these materials into the LAGA classes.

Even under worst case conditions of the pH4 stat method, the leaching of the trace elements As, Ba, Cd, Co, Cr, Cu, Mn, Ni, Sb and Zn is not relevant for ferrous slags (LFOe Lower Saxony 2000).