- Email: [email protected]
Products of steel slags an opportunity to save natural resources
Waste Management 21 (2001) 285±293
www.elsevier.nl/locate/wasman
Products of steel slags an opportunity to save natural resources H. Motz *, J. Geiseler Forschungsgemeinschaft EisenhuÈttenschlacken e. V., Bliersheimer Strasse 62, 47229 Duisburg, Germany
Abstract In Germany, and in the most industrial countries, the use of blast furnace and steel slags as an aggregate for civil engineering, for metallurgical use and as fertiliser has a very long tradition. Since the introduction of the basic oxygen steel making furnace (BOF) process and the electric arc furnace (EAF) process the German steel industry started extensive research on the development of ®elds of application for BOF and EAF slags. These investigations have been mainly performed by Forschungsgemeinschaft EisenhuÈttenschlacken e. V. (FEhS), the Research Association for blast furnace and steel slags. Today steel slags are well characterised and long-term experienced materials mainly used as aggregates for road construction (e.g. asphaltic or unbound layers), as armourstones for hydraulic engineering constructions (e.g. stabilisation of shores), and as fertiliser for agriculture purposes. These multifarious ®elds of application could only be achieved because the steelworks in¯uence the quality of slags by a careful selection of raw materials and a suitable process route. Furthermore, subsequent procedures like a treatment of the liquid slag, an appropriate heat treatment and a suitable processing have been developed to ensure that the quality of steel slags is always adequate for the end use. Depending on the respective ®eld of application, the suitability of steel slags has to be proven by determining the technical properties, as well as the environmental compatibility. For this reason test methods have been developed to evaluate the technical properties especially the volume stability and the environmental behaviour. To evaluate the volume stability a suitable test (steam test) has been developed and the results from laboratory tests were compared with the behaviour of steel slags under practical conditions, e.g. in a road. To determine the environmental behaviour leaching tests have been developed. In the meanwhile most of these test methods are drafted or already accepted as a CEN standard and are used for a continuous quality control. Usually the suitability of steel slags is stated by ful®lling the requirements of national and/or international standards and regulations. Based on these standards and regulations in Germany in 1998 about 97% of the produced steel slags have been used as aggregates for road construction (e.g. as surface layer, road base and sub base for high tracked roads), ways, earthworks, and armourstones for hydraulic structures. Consistent to the successful long-term experience not only products of steel slags but also products of blast furnace slags have been eliminated from the European Waste Catalogue and the European Shipment of Waste Regulation of the European Community, as well as from the lists of OECD for transfrontier movements by the decision of the OECD-Council from 21 September, 1995. # 2001 Published by Elsevier Science Ltd. Keywords: Slags; Steelslags; Metallurgical slags; Slag application; Technical properties of slags; Quality assurance of slags; Leaching behaviour of slags; Environmental regulations
1. Introduction In Europe and most other continents there is a great demand for aggregates mainly from civil engineering industry, especially in the ®eld of road and concrete constructions as well as for hydraulic purposes like stabilisation of river banks. But there are also the declared targets of the European Community [1].
* Corresponding author. Tel.: +49-2065-994510; fax: +49-20659945-0. E-mail address: [email protected] (H. Motz).
1. To protect the environment and to improve its quality. 2. To serve the protection of the human health. 3. To guarantee a cautious and ecient use of the natural resources. The last statement of the declared targets holds us responsible to save natural resources by using industrial co-products and to increase their utilisation rate wherever their technical and environmental properties are suitable concerning the relevant application ®eld. Regarding the situation in Germany (Fig. 1) about 675 million tons of natural aggregates like gravel, sand
0956-053X/01/$ - see front matter # 2001 Published by Elsevier Science Ltd. PII: S0956-053X(00)00102-1
286
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
and properties of steel slags as an aggregate for road construction and hydraulic structure. In addition, the quality control system which has been established to consider technical and environmental aspects during the recent years will be explained. 2. General properties of steel slag aggregates 2.1. Chemical and mineral composition
Fig. 1. Production and utilisation of natural aggregates and industrial co-products.
and other processed rocks are produced and requested by the industry per year. Contrary to this balanced situation with natural aggregates, the utilisation rate of about 174 million tons industrial co-products is much lower. Industrial co-products are, e.g. slags and coal mining residues (81 million tons), recycling materials, e.g. from demolition of concrete buildings (50 million tons) and other excavated materials from the building site (43 million tons). Their utilisation rate diers due to the technical and environmental properties of the relevant material between 10 and 100%. On average it is only 48%. Therefore, in Germany and in the other European countries the respective industry is concentrated on increasing the utilisation rate of industrial co-products and recycling materials [2,3]. The steel industry has traditionally produced co-products which have been successfully used in many ®elds of application. A good example are the steel slags from Basic-Bessemer or Thomas process. Since 1880 these slags have been used as a phosphatic fertiliser. Since then the steel industry has investigated their slags continuously, taken care of suitable processing and if necessary modi®ed the iron and steelmaking processes to get slag products which ful®l the requirements of the speci®c standards and regulations. This work has mainly been done by Forschungsgemeinschaft EisenhuÈttenschlacken e. V. (FEhS) Ð the German research institute for blast furnace and steel slags. In Europe every year nearly 12 million tons of steel slags are produced. Owing to the intensive research work during the last 30 years, today about 65% of the produced steel slags are used on quali®ed ®elds of application. But the remaining 35% of these slags are still dumped. It will need further intensive research work to decrease this rate as far as possible [4]. Contrary to the European ®gures the dumping rate of steel slags in Germany is only 7%. So 93% are used for dierent ®elds of application [3]. On the basis of the German experience with steel slags Ð these are basic oxygen furnace (BOF) slags and electric arc furnace (EAF) slags Ð this paper reports on the characteristics
The BOF- and EAF-slags from dierent sources within Europe, as characterised by their chemical and mineral compositions plus their technical properties, are generally comparable and independent of their producers. Dierences arise from the use of dolomite rather than lime with the eect of a higher MgO-content in the slag. Table 1 shows the range of the chemical composition of BOF- and EAF-slags. BOF- and EAF-slags are calciumsilicatic with a range of CaO between 42 and 55%, and a range of SiO2 between 12 and 18%. EAF-slags comprise CaO between 25 and 40% and 12 to 17% SiO2. Their MgO-content may be higher due to the reactions with the refractory lining. The main mineral phases of BOF- and EAF-slags are dicalciumsilicate, dicalciumferrite and wustite. Dicalciumsilicate is stabilised by the P2O5-content of the slag which prevents a disintegration of the slag known from other slags. The content of free lime and free MgO is the most important component for the utilisation of steel slags for civil engineering purposes with regard to their volume stability. In contact with water these mineral phases will react to hydroxides. Depending on the rate of free lime and/or free MgO this reaction causes a volume increase of the slag mostly combined with a disintegration of the slag pieces and a loss of strength. Therefore, the volume stability is a key criterion for using steel slags as a construction material. 2.2. Technical properties Static and dynamic forces and the environmental strains like rain, heat, freeze and thaw require adequate long-term behaviour under these conditions for all constructions which are built today. Therefore, the technical properties of processed aggregates which are used for these constructions are of fundamental importance. The most important properties are: 1. bulk density; 2. shape; 3. resistance to fragmentation (resistance to impact and crushing); 4. strength; 5. water absorption;
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
6. resistance to freezing and thawing; 7. volume stability; and 8. resistance to abrasion and polishing. During the past 30 years FEhS has investigated these properties of both, processed steel slag aggregates and established natural aggregates. Table 2 shows the mean values of selected properties derived from BOF- and EAF-slags, in comparison to granite and ¯int gravel. According to Table 2 steel slags can be processed to aggregates of high quality comparable with those of natural aggregates. The high bulk density of steel slags >3.2 g/cm3 quali®es steel slags as a construction material for hydraulic engineering purposes. In Germany about 400 000 tons per year are used as aggregates for the stabilisation of river banks and river beds against erosion. All other technical properties listed in Table 2 are comparable or even better than those of natural aggregates. In particular the high level of strength described by the impact and crushing value and additionally the rough surface texture are predominant. Both properties together with a high polished stone value (PSV) and a binder adhesion >90% qualify these slags as aggregates for high tracked road layers especially for asphaltic surface layers. More than 25 years ago in Germany test roads were built using steel slags as an aggregate for unbound and bituminous bound mixtures. The test results con®rm these good properties to the present day. The asphaltic layers had been resistant to deformation, rutting and polishing over this long-term period.
287
These good experiences encouraged the steel industry and road administration to start further research work. The aim was to investigate the suitability of steel slags for noise reducing asphaltic layers. These layers are mainly characterised by a very high void content of nearly 20% by volume. Therefore, only aggregates with a high resistance to abrasion and polishing are required. In 1993 a new test road was built in order to vary the aggregates used for the production of the surface layer. The test road was divided into three sections. One section was built with 100% LD-slag, one with aggregates as a mixture of 50% LD-slag and 50% diabase and the third one for comparison with 100% diabase aggregates. Fig. 2 shows the development of the skid resistance measured on the top of the surface layer with the ``sideway forced coecient routine investigation machine Ð SCRIM'' from 1993 to 1998. It is obvious that there is no decrease of the SCRIM-values since ®nishing the test road in 1993. On the contrary, the SCRIM-values of all sections arise slowly. Section 1 and Section 3 with LD-slag and diabase, respectively, show the same level of SCRIM-values. This means that there are no signi®cant dierences between the abrasion and polishing behaviour of LD-slag and diabase as a well proved natural aggregate. But all these pretentious applications of steel slags are only practicable if there is sucient volume stability as the decisive criterion. To guarantee a sucient volume stability the steel works employ dierent methods of slag treatment. The most applicable ones are the following:
Table 1 Range of chemical composition of BOF- and EAF-slags (%)a Component
CaO
SiO2
Al2O3
MgO
MnO
P2O5
Fetotal
Ca0free
BOF-slag low MgO-content BOF-slag high MgO-content EAF-slag low MgO-content EAF-slag high MgO-content
45±55 42±50 30±40 25±35
12±18 12±15 12±17 10±15
<3 <3 4±7 4±7
<3 5±8 4±8 8±15
<5 <5 <6 <6
<2 <2 <1.5 <1.5
14±20 15±20 18±28 20±29
<10 <10 <3 <3
a
H. Motz and J. Geiseler.
Table 2 Technical properties of processed BOF- and EAF-slags, in comparison with established natural aggregatesa Characteristics 3
Bulk density (g/cm ) Shape Ð thin and elongated pieces (%) Impact value (%/wt.) Crushing value (%/wt.) 10% ®nes (KN) Polishing (PSV) Water absorption (%/wt.) Resistance to freeze-thaw (%/wt.) Binder adhesion (%) a
H. Motz and J. Geiseler.
BOF-slag
EAF-slag
Granite
Flint gravel
3.3 <10 22 15 320 58 1.0 <0.5 >90
3.5 <10 18 13 350 61 0.7 <0.5 >90
2.5 <10 12 17 260 48 <0.5 <0.5 >90
2.6 <10 21 21 250 45 <0.5 <1 >85
288
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
Fig. 2. Development of SCRIM-values on asphaltic surface layer built with LD-slags.
1. Determination of the free lime content by analysing samples from the liquid slag out of the converter and pouring slags of high or low free lime content into separate pits. To determine the free lime content the ethylene glycol method or statistical formulas on the basis of the mineralogical composition may be used. In Germany a limit of 4% for the free lime content is mostly used as a separation criterion. But this method is only applicable if the steel making process works without the addition of dolomite lime and if steel slags are produced with a low MgO content. 2. Weathering of the solid slags outside the slag pits for a certain period of time at free atmosphere to transform the free lime into calciumhydroxide. In some cases the slag is also sprayed with water to accelerate the hydration of free lime and/or MgO. But weathering takes time, needs a lot of storage space and produces ®ne grained slag which is hardly applicable as a construction material. 3. Treatment of the liquid slag outside the steelmaking process in a separate slag pot, so that the steel quality will not be in¯uenced by this slag treatment and the steelmaking process is not extended. Together with Thyssen Krupp Stahl AG, the FEhS developed such a method by laboratory tests and tests in the steel shop [5]. The process works on the basis of the injection of oxygen and dry sand into the liquid slag. Thus, additional heat is generated to keep the slag liquid and to dissolve the sand. The CaO/SiO2 ratio is reduced and the free lime is dissolved and chemically bound. In 1996 the treatment equipment was installed in line of converter no. 1 of the steel shop of Thyssen Krupp Stahl AG, Duisburg, since then it has worked successfully. The free lime content of all slags determined produced by this treatment show amounts <1% by weight. Thus, they do not show any volume increase. Free MgO is supposed to react in the same way as free lime.
The treatment of slag chosen in steel works generally depends on the requirements for the volume stability which are dierent depending on their application. For certain applications no restrictions on the volume stability are necessary, e.g. unpaved roads, parking areas and dams. But if steel slags are used in unbound and bound layers of roads the volume stability must be within certain limits. The question arises how to measure the volume stability and which requirements have to be ful®lled. For fresh and separated steel slags with a low MgO content which have not been weathered the free lime content determined by the ethylene glycol method may be used as a criterion for the volume stability. In Germany, experience has found that steel slags with a free lime content up to 7% may be used in unbound layers and up to 4% in asphaltic layers. But for partially wheathered slags this method does not work because the ethylene glycol method does not dier between hydrated and unhydrated free lime. Furthermore, slags with a higher MgO content can contain not only free lime but also free MgO. Therefore, the determination of the free lime content is not sucient for the assessment of the volume stability. At present there is no reliable quick test method to determine the content of free MgO as a decisive criterion in the steel shop. When using slags for road construction hydration develops mainly from the reaction between water vapour and free CaO and/or free MgO. Therefore, in Germany several test methods have been developed and tested to get information about the expansion capacity of steel slags. FEhS started these investigations in 1978 with autoclave tests, but, the test conditions Ð high temperature of 215 C and high pressure of 21 bar Ð are extremely unrealistic. Additionally, the investigations came to the conclusion that using autoclave conditions mineral reactions occur which have never been found in the ®eld under natural conditions. After that, water
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
storing tests were carried out with the result that there appeared wash o eects of free lime and/or free MgO during water storage. Therefore, the tested samples very often simulated a satisfying volume stability which was not veri®ed when other test conditions were chosen e.g. storage in a humidity chamber. This was the beginning of the development of the steam test. The principle of the steam test is uncomplicated. As shown in Fig. 3 a compacted slag specimen with a de®ned grain size distribution 0/22 mm is subjected to a ¯ow of steam of about 100 C in a steam unit at ambient pressure. By these means, the necessary moisture for the reaction of free lime and free MgO is continuously conveyed to the test sample. The volume increase caused by this reaction is read o from a dial test indicator at the surface of the specimen. The increase in volume is given as the result, calculated in per cent by volume of the compacted slag specimen. From 1980 the steam test has been used in Germany as a method for the quality control for steel slags relating to the factory production control and the third party control. Today it is a routine test which during recent years has helped to avoid damages caused by unstable steel slag aggregates. Meanwhile, the steam test is accepted by European countries as a harmonized test method and has been incorporated into European aggregate standards as a test method for steel slags. It is part of EN 1744-1 ``Tests for chemical properties of aggregates Ð chemical analysis'' [6], which was published ocially by CEN in 1998. Fig. 4 shows a typical test result achieved by using the steam test. It shows the development of the volume increase of a LD-slag in respect to time. The curve is typical for all steel slags. Initially the rate of the volume increase is high but decreases with time almost to zero. The slope of the curve depends on the slag type, mainly
289
on the free lime and/or the free MgO content. The curve statistically can be calculated by a formula to predict the ®nal volume increase. The tests which have been completed during the last 10 years have shown that for LD-slags with a low MgO content <5%, 24 h testing time is sucient because after that time the rate of the volume increase has nearly ®nished. LD-slags with a higher MgO content >5% and EAF-slags have to be tested for 168 h because the hydration of MgO takes more time than that of free lime. In the meantime the steam test method has been discussed on the European level by the responsible task group 8 of TC 154 ``Aggregates''. This task group tried to ®nd suitable requirements as a suggestion for the product standards developed in the subcommittees 3 and 4 of TC 154. It took a long time to develop a harmonised version which considers the dierent experiences regarding steel slags, national specialities of treating slags, and the philosophy of road construction.
Fig. 4. Typical volume increase of steel slags tested by the steam test.
Fig. 3. Principle of the steam test.
290
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
The result of the discussions which was ®nally accepted by TC 154 is given in Table 3. Dierent categories of requirements VA to VD have been developed due to the maximum expansion. These categories can be chosen by the European countries on the basis of their national experience with steel slags and their individual ®elds of application. The class VD states that there is no necessity of testing the volume stability of slags if the national experience con®rms a satisfactory performance record. The categories given in Table 3 are part of the aggregates standards drafted by TC 154 [7,8]. According to German experiences steel slag aggregates are suitable for unbound layers and asphalt layers if the expansion does not exceed the requirements for category VA. Steel slag aggregates, therefore, have a suitable volume stability if the expansion is limited to 5 and 3.5% respectively. When slags are used for hydraulic structure purposes as armourstones free CaO and/or free MgO will react with water as a liquid. Therefore, a separate expansion test has been developed for this purpose by the Expert Group ``Armourstone'' of SC 4 of TC 154. The Expert Group developed the boiling test on the basis of a similar test method which has been used in The Netherlands for many years. Using this test method, 20 slag pieces with a diameter of 50±150 mm have to be placed in a bath of boiling water and left there for 8 hours. After this time the loss of mass of the slag pieces has to be determined. Just as for steel slag aggregates for road construction, the Expert Group ``Armourstones'' had to establish suitable requirements for armourstones too. Again it took a long time to come to harmonised requirements because there were dierent experiences to calculate and to interpret the test results. As a result the Expert Group established two categories due to the loss of mass after ®nishing the boiling test as it is given in the following: . Category A: Maximum four of the 20 test pieces are allowed to show more than 0.5% loss of mass
. Category B: The total loss of mass of all test pieces shall not exceed 5%. The boiling test method is described in draft prEN 13383-2 Armourstone Ð Test methods [9] and the requirements in draft prEN 13383-1 Armourstone Ð Speci®cation [10]. Both drafts have been presented to the European countries in order to ask for comments in 1999. 2.3. Environmental behaviour When industrial co-products are used as an aggregate water resource implications must be taken into account particulary from the durability aspect. The assessment basically of the environmental compatibility of aggregates as a building material is not determined by the content of environmentally relevant elements in the solid material but by the potential leaching behaviour. Therefore, the major element analysis and the mineral composition as given by Table 1 are essential but not sucient for an evaluation of environmental impact. To get information about the eect on the ground water and soil, it is of more interest to know the concentrations of those environmentally relevant components which can be leached out. To simulate the leaching of aggregates in a laboratory during the past 10 years in Germany and other European countries many leaching tests methods have been developed. Watching the developing work especially that of TC 292 ``Wastes'' it has to be pointed out that aggregates that are well known and have been used for a long time and which are continuously quality controlled need not be tested by so-called characterisation tests. It is sucient to control their leaching behaviour continuously by a short-term compliance test. At present the only leaching technique which is standardised in Germany is the DEV-S4 method described in DIN 38 414 [11]. This technique works with a con-
Table 3 Maximum expansion values for steel slag aggregatesa Type of steel slag
BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO>5.0%) BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO>5.0%) BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO> 5.0%) BOF-slag/EAF-slag (MgO45.0%) a
H. Motz and J. Geiseler.
Maximum expansion (% by volume) Aggregates for bituminous bound mixtures
Aggregates for unbound mixtures
3.5 3.5 6.5 6.5 10 10 No requirement
5 5 7.5 7.5 10 10
Testing time (h)
Category
24 168 24 168 24 168
VA VB VC VD
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
stant overhead rotation and places mechanical stress on the material so that new fracture surfaces and varying degrees of abrasion are liable to occur. Therefore, this method is unsuitable for describing the leaching characteristic of aggregates and so a new leaching technique was developed on the basis of existing tank methods. This leaching method is given in Fig. 5. It diers from the DEV-S4 method essentially in the fact that the leaching sample is not moved but only the water. This has the advantage that additional strain on the investigated material is avoided. Furthermore, this technique also enables larger (hardened) test samples to undergo leaching. This technique has been suciently tested during the last 5 years as a routine investigation procedure. Therefore, it was suggested as an European test method for leaching aggregates. In the meantime the tank leaching test was harmonised and drafted by task group 8 of TC 154 as part 3 of the EN 1744 [12]. The draft of EN 1744-3 was presented to the European countries as CEN-Enquiry at the end of 1999. In Germany the tank leaching test is ocially used as a controlling method for factory production and third party control when slags are applied as aggregates for road construction and hydraulic structures. The leaching test results have shown that except for the pH value, which is aected by the partial solution of the slag lime, and the closely-related electric conductivity, the leaching of other elements is insigni®cant in terms of environmental impact. Chromium as a mineral component may occur in somewhat higher amounts. But the concentrations in leachates are low because the chromium ions are bound within stable crystalline phases. All other concentrations of heavy elements are very low and not relevant under environmental aspects. Organic substances do not exist in slags because of the high melting temperature of about 1600 C. In Germany steel slags processed to aggregates for road construction and hydraulic structures have to be analysed by leaching tests twice a year. The concentra-
291
tions of the leachate have to ful®l the requirements of relevant parameters listed in Table 4. If the requirements given in Table 4 are met, then steel slag aggregates are applicable for all uses in road constructions and hydraulic structures. 3. Quality control as a basis for a successful application In order to ensure a permanent satisfying quality control of all slag products which are used for the building site, the German slag producers and processors control their products continuously by a factory production control. The performance and the results are certi®ed regularly by a third party control institute twice a year. This controlling system was established by the German association for quality control of metallurgical slags Ð GuÈtegemeinschaft EisenhuÈttenschlacken e.V. (GGEHS) Ð together with the German road and waterway administration 30 years ago. The production control requirements were adapted continuously to the latest development in quality control systems. The factory production control includes the: 1. production management; 2. process control; 3. inspection, calibration and testing of the equipment; 4. inspection and testing of the slag products; and 5. handling and delivery. The third party control institute evaluates the factory production control and controls the products twice a year by a separate investigation concerning all relevant technical and environmental properties like volume stability and leaching behaviour. As an example for road construction products Table 5 gives a summary of the general properties and suitable test methods performed regularly by factory production control and third party control. If all elements of the factory production control system and those of the third party control conform to the requirements the producer will be certi®ed by the Table 4 German requirements for concentrations of leachates using steel slag aggregatesa Parameter
Steel slag Road construction Hydraulic structure
Relevant aggregate size
all in
pH-Value 10±13 Electric conductivity (mS/m) 500 Chromium (mg/l) 0.03 Fig. 5. Tank leaching test apparatus.
a
H. Motz and J. Geiseler.
<60 mm >60 mm 11 80 0.03
10 60 0.03
292
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
Table 5 Quality control of steel slag propertiesÐtest frequenciesa Property
Test frequency
Test method
Factory production control 1 2 3 4 5 6 7 8 9 10 11 a
Petrogra®c characteristics Particle density Particle size distribution Particle shape Organic matter Resistance to fragmentation Resistance to freeze-thaw Water absorption Volume stability Resistance to polishing Leaching properties
1 per week 1 per week 1 per week 1 per day 1 per week 1 per week 2 per year
Third party control 1 2 2 2 2 2 1 1 2 1 2
per 2 years per year per year per year per year per year per 2 years per 2 years per year per 2 years per year
DIN EN 932-2/3 prEN 1097-6 prEN 933-10 DIN EN 933-3/4 DIN EN 1744-1 DIN EN 1097-2 DIN 52104 prEN 1097-6 DIN EN 1744-1 prEN 1097-8 prEN 1744-3
H. Motz and J. Geiseler
material because of their advantageous technical properties. In order to con®rm these practical experiences, more than 25 years ago test roads were built in agreement with steel producers and the road building administration. Thus steel slags were used not only for unbound layers like road bases and sub-bases, but also for bituminous bound layers like surface layers. All test roads have been built according to the German standards concerning aggregates and dimensioning urban roads and highways. Furthermore, all test roads have been divided into several sections which gave the possibility of investigating steel slags in comparison to established natural aggregates like crushed gravel or arti®cial aggregates like blast furnace slag. By building test roads the following results could be achieved:
Fig. 6. Quality mark of the GuÈtegemeinschaft EisenhuÈttenschlacken e.V.
GGEHS. After that he is permitted to mark his delivery tickets with the quality mark of the GGEHS (Fig. 6) as a sign of a successful certi®cation. 4. Summarised experiences with the application of steel slags Steel slags have been produced and used successfully in dierent European countries as a road construction
1. The crushed and rough surface of processed steel slag aggregate mixtures provides a stronger bearing capacity directly after compaction and higher than when using other aggregates. 2. There is no in¯uence of heavy rain on the bearing capacity of unbound layers built with steel slag mixtures. 3. The carbonatic solidi®cation leads to an increase of bearing capacity. 4. The aggregate mixtures are permanently stable if the requirements for the volume stability have been ful®lled. 5. The asphaltic surface layers remain permanently plain even under heavy trac. 6. The resistance to polishing of asphaltic surface layers remains on a high level over a long-term period. 7. Roads built with steel slags as an unbound or bituminous bound aggregate do not in¯uence the environment by leaching.
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
In the ®eld of hydraulic structures steel slags are mostly used for: 1. 2. 3. 4.
dams and dikes; stabilisation of river bottoms; re®lling of erosion areas on river bottoms; and stabilisation of river banks.
Usually aggregate sizes >10 mm are used to prevent an erosion of ®ne particles. Mainly the high density, the high level of strength and abrasion as well as the rough texture of processed steel slag aggregates ensure a longterm resistance to dynamic forces coming from waves and river ¯ow. The long-term behaviour of hydraulic constructions made by steel slags has been investigated by FEhS together with German river authorities regarding: 1. the volume stability; 2. the eects of leaching; and 3. the eects on the river fauna. Summarised the results show that the properties of steel slag armourstones are comparable to those of established natural stones like basalt. The engineering and ecological properties of steel slags are accepted in many countries and so steel slags are widely used as a construction material. In Germany the ®elds of road- and waterway construction have made a very signi®cant contribution to the present utilisation rate of about 93%. Thus, the use of steel slags saves existing resources of natural aggregates.
293
References [1] Vertrag zur GruÈndung der EuropaÈischen Wirtschaftsgemeinschaft EWG-Vertrag Artikel 130 r, Ziele der Umweltpolitik. [2] Krass, K. et al. Anfall, Aufbereitung und Verwertung von Recycling-Baustoen und industriellen Nebenprodukten im Wirtschaftsjahr 1997. Strasse und Autobahn 9, 1999. [3] Fachverband EisenhuÈttenschlacken e.V. Statistische Erhebung der Erzeugung und Produktion von EisenhuÈttenschlacken in den Jahren 1945±1990. [4] Forschungsgemeinschaft EisenhuÈttenschlacken e. V. Statistische Erhebung der Erzeugung und Produktion von EisenhuÈttenschlacken in Europa im Jahr 1996. [5] KuÈhn, M., Drissen, P., Geiseler, J., A new BOF slag treatment technology. 2. European Oxygen Steel Making Congress from 13±15 October 1997. [6] DIN EN 1744-1. PruÈfverfahren fuÈr chemische Eigenschaften von GesteinskoÈrnungen, Teil 1: Chemische Analyse (Mai 1998). DIN Deutsches Institut fuÈr Normung e.V., Berlin. [7] prEN 13043. Aggregates for bituminous mixtures and surface dressings for roads, air®elds and other tra®cked areas. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [8] prEN 13242. Aggregates for unbound and hydraulically bound materials for use in civil engineering work and road construction. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [9] prEN 13383-2. Armourstone Part 2: Test methods. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [10] prEN 13383-1. Armourstone Part 1: Speci®cation. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [11] DIN 38414. Deutsche Einheitsverfahren zur Wasser-, Abwasserund Schlammuntersuchung, Schlamm und Sedimente (Gruppe S)±Bestimmung der Eluierbarkeit mit Wasser (S4). DIN Deutsches Institut fuÈr Normung e.V., Berlin [12] prEN 1744-3. Tests for chemical properties of aggregates Part 3: preparation of eluates by leaching of aggregates. DIN Deutsches Institut fuÈr Normung e.V., Berlin.
www.elsevier.nl/locate/wasman
Products of steel slags an opportunity to save natural resources H. Motz *, J. Geiseler Forschungsgemeinschaft EisenhuÈttenschlacken e. V., Bliersheimer Strasse 62, 47229 Duisburg, Germany
Abstract In Germany, and in the most industrial countries, the use of blast furnace and steel slags as an aggregate for civil engineering, for metallurgical use and as fertiliser has a very long tradition. Since the introduction of the basic oxygen steel making furnace (BOF) process and the electric arc furnace (EAF) process the German steel industry started extensive research on the development of ®elds of application for BOF and EAF slags. These investigations have been mainly performed by Forschungsgemeinschaft EisenhuÈttenschlacken e. V. (FEhS), the Research Association for blast furnace and steel slags. Today steel slags are well characterised and long-term experienced materials mainly used as aggregates for road construction (e.g. asphaltic or unbound layers), as armourstones for hydraulic engineering constructions (e.g. stabilisation of shores), and as fertiliser for agriculture purposes. These multifarious ®elds of application could only be achieved because the steelworks in¯uence the quality of slags by a careful selection of raw materials and a suitable process route. Furthermore, subsequent procedures like a treatment of the liquid slag, an appropriate heat treatment and a suitable processing have been developed to ensure that the quality of steel slags is always adequate for the end use. Depending on the respective ®eld of application, the suitability of steel slags has to be proven by determining the technical properties, as well as the environmental compatibility. For this reason test methods have been developed to evaluate the technical properties especially the volume stability and the environmental behaviour. To evaluate the volume stability a suitable test (steam test) has been developed and the results from laboratory tests were compared with the behaviour of steel slags under practical conditions, e.g. in a road. To determine the environmental behaviour leaching tests have been developed. In the meanwhile most of these test methods are drafted or already accepted as a CEN standard and are used for a continuous quality control. Usually the suitability of steel slags is stated by ful®lling the requirements of national and/or international standards and regulations. Based on these standards and regulations in Germany in 1998 about 97% of the produced steel slags have been used as aggregates for road construction (e.g. as surface layer, road base and sub base for high tracked roads), ways, earthworks, and armourstones for hydraulic structures. Consistent to the successful long-term experience not only products of steel slags but also products of blast furnace slags have been eliminated from the European Waste Catalogue and the European Shipment of Waste Regulation of the European Community, as well as from the lists of OECD for transfrontier movements by the decision of the OECD-Council from 21 September, 1995. # 2001 Published by Elsevier Science Ltd. Keywords: Slags; Steelslags; Metallurgical slags; Slag application; Technical properties of slags; Quality assurance of slags; Leaching behaviour of slags; Environmental regulations
1. Introduction In Europe and most other continents there is a great demand for aggregates mainly from civil engineering industry, especially in the ®eld of road and concrete constructions as well as for hydraulic purposes like stabilisation of river banks. But there are also the declared targets of the European Community [1].
* Corresponding author. Tel.: +49-2065-994510; fax: +49-20659945-0. E-mail address: [email protected] (H. Motz).
1. To protect the environment and to improve its quality. 2. To serve the protection of the human health. 3. To guarantee a cautious and ecient use of the natural resources. The last statement of the declared targets holds us responsible to save natural resources by using industrial co-products and to increase their utilisation rate wherever their technical and environmental properties are suitable concerning the relevant application ®eld. Regarding the situation in Germany (Fig. 1) about 675 million tons of natural aggregates like gravel, sand
0956-053X/01/$ - see front matter # 2001 Published by Elsevier Science Ltd. PII: S0956-053X(00)00102-1
286
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
and properties of steel slags as an aggregate for road construction and hydraulic structure. In addition, the quality control system which has been established to consider technical and environmental aspects during the recent years will be explained. 2. General properties of steel slag aggregates 2.1. Chemical and mineral composition
Fig. 1. Production and utilisation of natural aggregates and industrial co-products.
and other processed rocks are produced and requested by the industry per year. Contrary to this balanced situation with natural aggregates, the utilisation rate of about 174 million tons industrial co-products is much lower. Industrial co-products are, e.g. slags and coal mining residues (81 million tons), recycling materials, e.g. from demolition of concrete buildings (50 million tons) and other excavated materials from the building site (43 million tons). Their utilisation rate diers due to the technical and environmental properties of the relevant material between 10 and 100%. On average it is only 48%. Therefore, in Germany and in the other European countries the respective industry is concentrated on increasing the utilisation rate of industrial co-products and recycling materials [2,3]. The steel industry has traditionally produced co-products which have been successfully used in many ®elds of application. A good example are the steel slags from Basic-Bessemer or Thomas process. Since 1880 these slags have been used as a phosphatic fertiliser. Since then the steel industry has investigated their slags continuously, taken care of suitable processing and if necessary modi®ed the iron and steelmaking processes to get slag products which ful®l the requirements of the speci®c standards and regulations. This work has mainly been done by Forschungsgemeinschaft EisenhuÈttenschlacken e. V. (FEhS) Ð the German research institute for blast furnace and steel slags. In Europe every year nearly 12 million tons of steel slags are produced. Owing to the intensive research work during the last 30 years, today about 65% of the produced steel slags are used on quali®ed ®elds of application. But the remaining 35% of these slags are still dumped. It will need further intensive research work to decrease this rate as far as possible [4]. Contrary to the European ®gures the dumping rate of steel slags in Germany is only 7%. So 93% are used for dierent ®elds of application [3]. On the basis of the German experience with steel slags Ð these are basic oxygen furnace (BOF) slags and electric arc furnace (EAF) slags Ð this paper reports on the characteristics
The BOF- and EAF-slags from dierent sources within Europe, as characterised by their chemical and mineral compositions plus their technical properties, are generally comparable and independent of their producers. Dierences arise from the use of dolomite rather than lime with the eect of a higher MgO-content in the slag. Table 1 shows the range of the chemical composition of BOF- and EAF-slags. BOF- and EAF-slags are calciumsilicatic with a range of CaO between 42 and 55%, and a range of SiO2 between 12 and 18%. EAF-slags comprise CaO between 25 and 40% and 12 to 17% SiO2. Their MgO-content may be higher due to the reactions with the refractory lining. The main mineral phases of BOF- and EAF-slags are dicalciumsilicate, dicalciumferrite and wustite. Dicalciumsilicate is stabilised by the P2O5-content of the slag which prevents a disintegration of the slag known from other slags. The content of free lime and free MgO is the most important component for the utilisation of steel slags for civil engineering purposes with regard to their volume stability. In contact with water these mineral phases will react to hydroxides. Depending on the rate of free lime and/or free MgO this reaction causes a volume increase of the slag mostly combined with a disintegration of the slag pieces and a loss of strength. Therefore, the volume stability is a key criterion for using steel slags as a construction material. 2.2. Technical properties Static and dynamic forces and the environmental strains like rain, heat, freeze and thaw require adequate long-term behaviour under these conditions for all constructions which are built today. Therefore, the technical properties of processed aggregates which are used for these constructions are of fundamental importance. The most important properties are: 1. bulk density; 2. shape; 3. resistance to fragmentation (resistance to impact and crushing); 4. strength; 5. water absorption;
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
6. resistance to freezing and thawing; 7. volume stability; and 8. resistance to abrasion and polishing. During the past 30 years FEhS has investigated these properties of both, processed steel slag aggregates and established natural aggregates. Table 2 shows the mean values of selected properties derived from BOF- and EAF-slags, in comparison to granite and ¯int gravel. According to Table 2 steel slags can be processed to aggregates of high quality comparable with those of natural aggregates. The high bulk density of steel slags >3.2 g/cm3 quali®es steel slags as a construction material for hydraulic engineering purposes. In Germany about 400 000 tons per year are used as aggregates for the stabilisation of river banks and river beds against erosion. All other technical properties listed in Table 2 are comparable or even better than those of natural aggregates. In particular the high level of strength described by the impact and crushing value and additionally the rough surface texture are predominant. Both properties together with a high polished stone value (PSV) and a binder adhesion >90% qualify these slags as aggregates for high tracked road layers especially for asphaltic surface layers. More than 25 years ago in Germany test roads were built using steel slags as an aggregate for unbound and bituminous bound mixtures. The test results con®rm these good properties to the present day. The asphaltic layers had been resistant to deformation, rutting and polishing over this long-term period.
287
These good experiences encouraged the steel industry and road administration to start further research work. The aim was to investigate the suitability of steel slags for noise reducing asphaltic layers. These layers are mainly characterised by a very high void content of nearly 20% by volume. Therefore, only aggregates with a high resistance to abrasion and polishing are required. In 1993 a new test road was built in order to vary the aggregates used for the production of the surface layer. The test road was divided into three sections. One section was built with 100% LD-slag, one with aggregates as a mixture of 50% LD-slag and 50% diabase and the third one for comparison with 100% diabase aggregates. Fig. 2 shows the development of the skid resistance measured on the top of the surface layer with the ``sideway forced coecient routine investigation machine Ð SCRIM'' from 1993 to 1998. It is obvious that there is no decrease of the SCRIM-values since ®nishing the test road in 1993. On the contrary, the SCRIM-values of all sections arise slowly. Section 1 and Section 3 with LD-slag and diabase, respectively, show the same level of SCRIM-values. This means that there are no signi®cant dierences between the abrasion and polishing behaviour of LD-slag and diabase as a well proved natural aggregate. But all these pretentious applications of steel slags are only practicable if there is sucient volume stability as the decisive criterion. To guarantee a sucient volume stability the steel works employ dierent methods of slag treatment. The most applicable ones are the following:
Table 1 Range of chemical composition of BOF- and EAF-slags (%)a Component
CaO
SiO2
Al2O3
MgO
MnO
P2O5
Fetotal
Ca0free
BOF-slag low MgO-content BOF-slag high MgO-content EAF-slag low MgO-content EAF-slag high MgO-content
45±55 42±50 30±40 25±35
12±18 12±15 12±17 10±15
<3 <3 4±7 4±7
<3 5±8 4±8 8±15
<5 <5 <6 <6
<2 <2 <1.5 <1.5
14±20 15±20 18±28 20±29
<10 <10 <3 <3
a
H. Motz and J. Geiseler.
Table 2 Technical properties of processed BOF- and EAF-slags, in comparison with established natural aggregatesa Characteristics 3
Bulk density (g/cm ) Shape Ð thin and elongated pieces (%) Impact value (%/wt.) Crushing value (%/wt.) 10% ®nes (KN) Polishing (PSV) Water absorption (%/wt.) Resistance to freeze-thaw (%/wt.) Binder adhesion (%) a
H. Motz and J. Geiseler.
BOF-slag
EAF-slag
Granite
Flint gravel
3.3 <10 22 15 320 58 1.0 <0.5 >90
3.5 <10 18 13 350 61 0.7 <0.5 >90
2.5 <10 12 17 260 48 <0.5 <0.5 >90
2.6 <10 21 21 250 45 <0.5 <1 >85
288
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
Fig. 2. Development of SCRIM-values on asphaltic surface layer built with LD-slags.
1. Determination of the free lime content by analysing samples from the liquid slag out of the converter and pouring slags of high or low free lime content into separate pits. To determine the free lime content the ethylene glycol method or statistical formulas on the basis of the mineralogical composition may be used. In Germany a limit of 4% for the free lime content is mostly used as a separation criterion. But this method is only applicable if the steel making process works without the addition of dolomite lime and if steel slags are produced with a low MgO content. 2. Weathering of the solid slags outside the slag pits for a certain period of time at free atmosphere to transform the free lime into calciumhydroxide. In some cases the slag is also sprayed with water to accelerate the hydration of free lime and/or MgO. But weathering takes time, needs a lot of storage space and produces ®ne grained slag which is hardly applicable as a construction material. 3. Treatment of the liquid slag outside the steelmaking process in a separate slag pot, so that the steel quality will not be in¯uenced by this slag treatment and the steelmaking process is not extended. Together with Thyssen Krupp Stahl AG, the FEhS developed such a method by laboratory tests and tests in the steel shop [5]. The process works on the basis of the injection of oxygen and dry sand into the liquid slag. Thus, additional heat is generated to keep the slag liquid and to dissolve the sand. The CaO/SiO2 ratio is reduced and the free lime is dissolved and chemically bound. In 1996 the treatment equipment was installed in line of converter no. 1 of the steel shop of Thyssen Krupp Stahl AG, Duisburg, since then it has worked successfully. The free lime content of all slags determined produced by this treatment show amounts <1% by weight. Thus, they do not show any volume increase. Free MgO is supposed to react in the same way as free lime.
The treatment of slag chosen in steel works generally depends on the requirements for the volume stability which are dierent depending on their application. For certain applications no restrictions on the volume stability are necessary, e.g. unpaved roads, parking areas and dams. But if steel slags are used in unbound and bound layers of roads the volume stability must be within certain limits. The question arises how to measure the volume stability and which requirements have to be ful®lled. For fresh and separated steel slags with a low MgO content which have not been weathered the free lime content determined by the ethylene glycol method may be used as a criterion for the volume stability. In Germany, experience has found that steel slags with a free lime content up to 7% may be used in unbound layers and up to 4% in asphaltic layers. But for partially wheathered slags this method does not work because the ethylene glycol method does not dier between hydrated and unhydrated free lime. Furthermore, slags with a higher MgO content can contain not only free lime but also free MgO. Therefore, the determination of the free lime content is not sucient for the assessment of the volume stability. At present there is no reliable quick test method to determine the content of free MgO as a decisive criterion in the steel shop. When using slags for road construction hydration develops mainly from the reaction between water vapour and free CaO and/or free MgO. Therefore, in Germany several test methods have been developed and tested to get information about the expansion capacity of steel slags. FEhS started these investigations in 1978 with autoclave tests, but, the test conditions Ð high temperature of 215 C and high pressure of 21 bar Ð are extremely unrealistic. Additionally, the investigations came to the conclusion that using autoclave conditions mineral reactions occur which have never been found in the ®eld under natural conditions. After that, water
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
storing tests were carried out with the result that there appeared wash o eects of free lime and/or free MgO during water storage. Therefore, the tested samples very often simulated a satisfying volume stability which was not veri®ed when other test conditions were chosen e.g. storage in a humidity chamber. This was the beginning of the development of the steam test. The principle of the steam test is uncomplicated. As shown in Fig. 3 a compacted slag specimen with a de®ned grain size distribution 0/22 mm is subjected to a ¯ow of steam of about 100 C in a steam unit at ambient pressure. By these means, the necessary moisture for the reaction of free lime and free MgO is continuously conveyed to the test sample. The volume increase caused by this reaction is read o from a dial test indicator at the surface of the specimen. The increase in volume is given as the result, calculated in per cent by volume of the compacted slag specimen. From 1980 the steam test has been used in Germany as a method for the quality control for steel slags relating to the factory production control and the third party control. Today it is a routine test which during recent years has helped to avoid damages caused by unstable steel slag aggregates. Meanwhile, the steam test is accepted by European countries as a harmonized test method and has been incorporated into European aggregate standards as a test method for steel slags. It is part of EN 1744-1 ``Tests for chemical properties of aggregates Ð chemical analysis'' [6], which was published ocially by CEN in 1998. Fig. 4 shows a typical test result achieved by using the steam test. It shows the development of the volume increase of a LD-slag in respect to time. The curve is typical for all steel slags. Initially the rate of the volume increase is high but decreases with time almost to zero. The slope of the curve depends on the slag type, mainly
289
on the free lime and/or the free MgO content. The curve statistically can be calculated by a formula to predict the ®nal volume increase. The tests which have been completed during the last 10 years have shown that for LD-slags with a low MgO content <5%, 24 h testing time is sucient because after that time the rate of the volume increase has nearly ®nished. LD-slags with a higher MgO content >5% and EAF-slags have to be tested for 168 h because the hydration of MgO takes more time than that of free lime. In the meantime the steam test method has been discussed on the European level by the responsible task group 8 of TC 154 ``Aggregates''. This task group tried to ®nd suitable requirements as a suggestion for the product standards developed in the subcommittees 3 and 4 of TC 154. It took a long time to develop a harmonised version which considers the dierent experiences regarding steel slags, national specialities of treating slags, and the philosophy of road construction.
Fig. 4. Typical volume increase of steel slags tested by the steam test.
Fig. 3. Principle of the steam test.
290
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
The result of the discussions which was ®nally accepted by TC 154 is given in Table 3. Dierent categories of requirements VA to VD have been developed due to the maximum expansion. These categories can be chosen by the European countries on the basis of their national experience with steel slags and their individual ®elds of application. The class VD states that there is no necessity of testing the volume stability of slags if the national experience con®rms a satisfactory performance record. The categories given in Table 3 are part of the aggregates standards drafted by TC 154 [7,8]. According to German experiences steel slag aggregates are suitable for unbound layers and asphalt layers if the expansion does not exceed the requirements for category VA. Steel slag aggregates, therefore, have a suitable volume stability if the expansion is limited to 5 and 3.5% respectively. When slags are used for hydraulic structure purposes as armourstones free CaO and/or free MgO will react with water as a liquid. Therefore, a separate expansion test has been developed for this purpose by the Expert Group ``Armourstone'' of SC 4 of TC 154. The Expert Group developed the boiling test on the basis of a similar test method which has been used in The Netherlands for many years. Using this test method, 20 slag pieces with a diameter of 50±150 mm have to be placed in a bath of boiling water and left there for 8 hours. After this time the loss of mass of the slag pieces has to be determined. Just as for steel slag aggregates for road construction, the Expert Group ``Armourstones'' had to establish suitable requirements for armourstones too. Again it took a long time to come to harmonised requirements because there were dierent experiences to calculate and to interpret the test results. As a result the Expert Group established two categories due to the loss of mass after ®nishing the boiling test as it is given in the following: . Category A: Maximum four of the 20 test pieces are allowed to show more than 0.5% loss of mass
. Category B: The total loss of mass of all test pieces shall not exceed 5%. The boiling test method is described in draft prEN 13383-2 Armourstone Ð Test methods [9] and the requirements in draft prEN 13383-1 Armourstone Ð Speci®cation [10]. Both drafts have been presented to the European countries in order to ask for comments in 1999. 2.3. Environmental behaviour When industrial co-products are used as an aggregate water resource implications must be taken into account particulary from the durability aspect. The assessment basically of the environmental compatibility of aggregates as a building material is not determined by the content of environmentally relevant elements in the solid material but by the potential leaching behaviour. Therefore, the major element analysis and the mineral composition as given by Table 1 are essential but not sucient for an evaluation of environmental impact. To get information about the eect on the ground water and soil, it is of more interest to know the concentrations of those environmentally relevant components which can be leached out. To simulate the leaching of aggregates in a laboratory during the past 10 years in Germany and other European countries many leaching tests methods have been developed. Watching the developing work especially that of TC 292 ``Wastes'' it has to be pointed out that aggregates that are well known and have been used for a long time and which are continuously quality controlled need not be tested by so-called characterisation tests. It is sucient to control their leaching behaviour continuously by a short-term compliance test. At present the only leaching technique which is standardised in Germany is the DEV-S4 method described in DIN 38 414 [11]. This technique works with a con-
Table 3 Maximum expansion values for steel slag aggregatesa Type of steel slag
BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO>5.0%) BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO>5.0%) BOF-slag/EAF-slag (MgO45.0%) BOF-slag/EAF-slag (MgO> 5.0%) BOF-slag/EAF-slag (MgO45.0%) a
H. Motz and J. Geiseler.
Maximum expansion (% by volume) Aggregates for bituminous bound mixtures
Aggregates for unbound mixtures
3.5 3.5 6.5 6.5 10 10 No requirement
5 5 7.5 7.5 10 10
Testing time (h)
Category
24 168 24 168 24 168
VA VB VC VD
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
stant overhead rotation and places mechanical stress on the material so that new fracture surfaces and varying degrees of abrasion are liable to occur. Therefore, this method is unsuitable for describing the leaching characteristic of aggregates and so a new leaching technique was developed on the basis of existing tank methods. This leaching method is given in Fig. 5. It diers from the DEV-S4 method essentially in the fact that the leaching sample is not moved but only the water. This has the advantage that additional strain on the investigated material is avoided. Furthermore, this technique also enables larger (hardened) test samples to undergo leaching. This technique has been suciently tested during the last 5 years as a routine investigation procedure. Therefore, it was suggested as an European test method for leaching aggregates. In the meantime the tank leaching test was harmonised and drafted by task group 8 of TC 154 as part 3 of the EN 1744 [12]. The draft of EN 1744-3 was presented to the European countries as CEN-Enquiry at the end of 1999. In Germany the tank leaching test is ocially used as a controlling method for factory production and third party control when slags are applied as aggregates for road construction and hydraulic structures. The leaching test results have shown that except for the pH value, which is aected by the partial solution of the slag lime, and the closely-related electric conductivity, the leaching of other elements is insigni®cant in terms of environmental impact. Chromium as a mineral component may occur in somewhat higher amounts. But the concentrations in leachates are low because the chromium ions are bound within stable crystalline phases. All other concentrations of heavy elements are very low and not relevant under environmental aspects. Organic substances do not exist in slags because of the high melting temperature of about 1600 C. In Germany steel slags processed to aggregates for road construction and hydraulic structures have to be analysed by leaching tests twice a year. The concentra-
291
tions of the leachate have to ful®l the requirements of relevant parameters listed in Table 4. If the requirements given in Table 4 are met, then steel slag aggregates are applicable for all uses in road constructions and hydraulic structures. 3. Quality control as a basis for a successful application In order to ensure a permanent satisfying quality control of all slag products which are used for the building site, the German slag producers and processors control their products continuously by a factory production control. The performance and the results are certi®ed regularly by a third party control institute twice a year. This controlling system was established by the German association for quality control of metallurgical slags Ð GuÈtegemeinschaft EisenhuÈttenschlacken e.V. (GGEHS) Ð together with the German road and waterway administration 30 years ago. The production control requirements were adapted continuously to the latest development in quality control systems. The factory production control includes the: 1. production management; 2. process control; 3. inspection, calibration and testing of the equipment; 4. inspection and testing of the slag products; and 5. handling and delivery. The third party control institute evaluates the factory production control and controls the products twice a year by a separate investigation concerning all relevant technical and environmental properties like volume stability and leaching behaviour. As an example for road construction products Table 5 gives a summary of the general properties and suitable test methods performed regularly by factory production control and third party control. If all elements of the factory production control system and those of the third party control conform to the requirements the producer will be certi®ed by the Table 4 German requirements for concentrations of leachates using steel slag aggregatesa Parameter
Steel slag Road construction Hydraulic structure
Relevant aggregate size
all in
pH-Value 10±13 Electric conductivity (mS/m) 500 Chromium (mg/l) 0.03 Fig. 5. Tank leaching test apparatus.
a
H. Motz and J. Geiseler.
<60 mm >60 mm 11 80 0.03
10 60 0.03
292
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
Table 5 Quality control of steel slag propertiesÐtest frequenciesa Property
Test frequency
Test method
Factory production control 1 2 3 4 5 6 7 8 9 10 11 a
Petrogra®c characteristics Particle density Particle size distribution Particle shape Organic matter Resistance to fragmentation Resistance to freeze-thaw Water absorption Volume stability Resistance to polishing Leaching properties
1 per week 1 per week 1 per week 1 per day 1 per week 1 per week 2 per year
Third party control 1 2 2 2 2 2 1 1 2 1 2
per 2 years per year per year per year per year per year per 2 years per 2 years per year per 2 years per year
DIN EN 932-2/3 prEN 1097-6 prEN 933-10 DIN EN 933-3/4 DIN EN 1744-1 DIN EN 1097-2 DIN 52104 prEN 1097-6 DIN EN 1744-1 prEN 1097-8 prEN 1744-3
H. Motz and J. Geiseler
material because of their advantageous technical properties. In order to con®rm these practical experiences, more than 25 years ago test roads were built in agreement with steel producers and the road building administration. Thus steel slags were used not only for unbound layers like road bases and sub-bases, but also for bituminous bound layers like surface layers. All test roads have been built according to the German standards concerning aggregates and dimensioning urban roads and highways. Furthermore, all test roads have been divided into several sections which gave the possibility of investigating steel slags in comparison to established natural aggregates like crushed gravel or arti®cial aggregates like blast furnace slag. By building test roads the following results could be achieved:
Fig. 6. Quality mark of the GuÈtegemeinschaft EisenhuÈttenschlacken e.V.
GGEHS. After that he is permitted to mark his delivery tickets with the quality mark of the GGEHS (Fig. 6) as a sign of a successful certi®cation. 4. Summarised experiences with the application of steel slags Steel slags have been produced and used successfully in dierent European countries as a road construction
1. The crushed and rough surface of processed steel slag aggregate mixtures provides a stronger bearing capacity directly after compaction and higher than when using other aggregates. 2. There is no in¯uence of heavy rain on the bearing capacity of unbound layers built with steel slag mixtures. 3. The carbonatic solidi®cation leads to an increase of bearing capacity. 4. The aggregate mixtures are permanently stable if the requirements for the volume stability have been ful®lled. 5. The asphaltic surface layers remain permanently plain even under heavy trac. 6. The resistance to polishing of asphaltic surface layers remains on a high level over a long-term period. 7. Roads built with steel slags as an unbound or bituminous bound aggregate do not in¯uence the environment by leaching.
H. Motz, J. Geiseler / Waste Management 21 (2001) 285±293
In the ®eld of hydraulic structures steel slags are mostly used for: 1. 2. 3. 4.
dams and dikes; stabilisation of river bottoms; re®lling of erosion areas on river bottoms; and stabilisation of river banks.
Usually aggregate sizes >10 mm are used to prevent an erosion of ®ne particles. Mainly the high density, the high level of strength and abrasion as well as the rough texture of processed steel slag aggregates ensure a longterm resistance to dynamic forces coming from waves and river ¯ow. The long-term behaviour of hydraulic constructions made by steel slags has been investigated by FEhS together with German river authorities regarding: 1. the volume stability; 2. the eects of leaching; and 3. the eects on the river fauna. Summarised the results show that the properties of steel slag armourstones are comparable to those of established natural stones like basalt. The engineering and ecological properties of steel slags are accepted in many countries and so steel slags are widely used as a construction material. In Germany the ®elds of road- and waterway construction have made a very signi®cant contribution to the present utilisation rate of about 93%. Thus, the use of steel slags saves existing resources of natural aggregates.
293
References [1] Vertrag zur GruÈndung der EuropaÈischen Wirtschaftsgemeinschaft EWG-Vertrag Artikel 130 r, Ziele der Umweltpolitik. [2] Krass, K. et al. Anfall, Aufbereitung und Verwertung von Recycling-Baustoen und industriellen Nebenprodukten im Wirtschaftsjahr 1997. Strasse und Autobahn 9, 1999. [3] Fachverband EisenhuÈttenschlacken e.V. Statistische Erhebung der Erzeugung und Produktion von EisenhuÈttenschlacken in den Jahren 1945±1990. [4] Forschungsgemeinschaft EisenhuÈttenschlacken e. V. Statistische Erhebung der Erzeugung und Produktion von EisenhuÈttenschlacken in Europa im Jahr 1996. [5] KuÈhn, M., Drissen, P., Geiseler, J., A new BOF slag treatment technology. 2. European Oxygen Steel Making Congress from 13±15 October 1997. [6] DIN EN 1744-1. PruÈfverfahren fuÈr chemische Eigenschaften von GesteinskoÈrnungen, Teil 1: Chemische Analyse (Mai 1998). DIN Deutsches Institut fuÈr Normung e.V., Berlin. [7] prEN 13043. Aggregates for bituminous mixtures and surface dressings for roads, air®elds and other tra®cked areas. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [8] prEN 13242. Aggregates for unbound and hydraulically bound materials for use in civil engineering work and road construction. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [9] prEN 13383-2. Armourstone Part 2: Test methods. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [10] prEN 13383-1. Armourstone Part 1: Speci®cation. DIN Deutsches Institut fuÈr Normung e.V., Berlin. [11] DIN 38414. Deutsche Einheitsverfahren zur Wasser-, Abwasserund Schlammuntersuchung, Schlamm und Sedimente (Gruppe S)±Bestimmung der Eluierbarkeit mit Wasser (S4). DIN Deutsches Institut fuÈr Normung e.V., Berlin [12] prEN 1744-3. Tests for chemical properties of aggregates Part 3: preparation of eluates by leaching of aggregates. DIN Deutsches Institut fuÈr Normung e.V., Berlin.