- Email: [email protected]
Leaching behaviour of synthetic aggregates
Waste Materials in Construction G.R. Woolley, J.J.J.M. Goumans and P.J. Wainwright (Editors) 2000 Elsevier Science Ltd.
695
Leaching behaviour of synthetic aggregates H.A. van der Sloot a, D. Hoede a, D.J.F Cresswell b and J.R Barton b. a E C N - Soil & Waste Research, P.O. Box 1, 1755 ZG Petten, The Netherlands b School of Civil Engineering, Woodhouse Lane, Leeds LS2 9JT, UK
In the framework of EU project Utilising innovative kiln technology to recycle waste into synthetic aggregate (BRST-CT98-5234), the leaching behaviour of synthetic aggregates has been studied to assess its environmental compatibility in the various stages of its use. Since the conditions are very different for the different uses, the assessment calls for a variety of different leaching conditions. The pH dependence test is used to cover important differences in pH environment to which the materials are exposed to as well as for an assessment of the buffering capacity of the material. Synthetic aggregate features a low buffer capacity, which makes it sensitive to externally imposed pH conditions. Utilisation and storage exposed to acidic conditions needs to be avoided. The pH dependence test and column leaching test are consistent. The CEN TC 154 method appears provide systematically low values due to the arbitrary selection of test conditions. Synthetic aggregate studied to date will not adversely affect the concrete in its service life. The main issue for aggregate use is the recycling and the "end of life" condition, when the material becomes construction debris. Not metals, but oxyanions, such as Cr VI and Mo are most relevant under these conditions. A concise test has been applied to assess crucial aspects of leaching for different production mixes.
1. I N T R O D U C T I O N In the framework of EU project Utilising innovative kiln technology to recycle waste into synthetic aggregate (BRST-CT98-5234), a new kiln design with a more efficient energy use is applied, which has more flexibility to process different waste materials as feedstock. The starting materials and the synthetic aggregates produced at laboratory scale at the School of Civil Engineering (Leeds) and at full scale facility in the south of England were tested for their leaching behaviour to assess their environmental compatibility in the various stages of their use. This includes storage prior to use, use as an aggregate in concrete, its recycling as construction debris in the same application, its reuse in other applications (e.g. road base construction) and its end-of-life situation (ultimate disposal). In all of these phases of use, the release of constituents from the material needs to be addressed to ensure environmental compatibility. Since the conditions are very different for the different uses, the assessment calls for a variety of different leaching conditions. Monolith leaching tests for bound applications [1,2] and percolation test [3,4] for several others. The pH dependence test [5] is used to cover important differences in pH environment to which the materials are exposed to as well as for an assessment of the buffering capacity of the material, which provides insight
696 in the sensitivity of the materials to extemally imposed conditions. In the framework of CEN TC 154, a compliance test for aggregates has been drafted, which contains an arbitrary choice of conditions unrelated to potential environmental impact. This method has been tested as well to verify its performance relative to the more elaborate evaluation.
2. E X P E R I M E N T A L
WORK
2.1 M a t e r i a l s
A range of materials have potential for use as feedstock for the aggrelgate production: granite washings, paper sludge, contaminated river sediment, MSWI bottom ash and shredder waste. The suitability of these materials in preparing pellets meeting technical specifications is studied at the School of Civil Engineering (Leeds). A key aspect is the water uptake of the pellets. The starting materials - sediments, municipal solid waste incinerator bottom ash (MSWI BA), shredder waste, milling residues, paper sludge - have been tested using the pH dependence test to identify the main leaching features of the starting materials for later comparison with the behaviour of pellets produced. Synthetic aggregate pellets, produced by firing in an innovative kiln at lab-scale from several combinations of sediment, municipal solid waste incinerator bottom ash, shredder waste, milling residues and paper sludge, and aggregates resulting from full scale aggregate production have been tested using the pH dependence test and the concise leaching test. On limited number of samples (due to large material amount required) the TC 154 method has been applied. On an aggregate sample from granite washings all tests including a column test on the intact pellets has been carried out. From a limited number of aggregate samples concrete specimen have been prepared and tested using the tank leaching test [1], a compliance tank leaching test [2] and the pH dependence test [5]. This allows an evaluation of the material in different stages of use, such as storage prior to use, use as an aggregate in concrete, its recycling as construction debris in the same application, its reuse in other applications (e.g. road base construction) and its endof-life situation (ultimate disposal). 2.2 M e t h o d s
The leaching tests selected for the study are: - pH static leach test [5]: This test provides information on the pH sensitivity of leaching
behaviour of the material. The test consists of a number of parallel extractions of a material at a liquid/solid ratio (L/S) 10 (1/kg) during 48 hours at a series of preset pH values. Since pH is one of the main leaching controlling parameters, the information can be used to evaluate the repeatability in testing (resulting from measurement at steep concentration - pH slopes) and to provide information on the sensitivity to pH in specific field scenarios. The acid neutralization capacity (ANC) derived from the test is a useful property in this respect. For material characterization this has been proven to be a very useful method [6,7]. The method is standardized in two experimental modes by CEN TC 292 Working Group 6. - Column leaching test with similarities to NEN 7343 [3] and a percolation test is developed at European level in CEN TC 292 WG6 [4], a column test in which 7 eluate fractions are collected within the range of L/S = 0.1-10 1/kg. The total test duration is approximately 21 days. The leachant is demineralized water (DMW). The test material is applied as received (around 1 cm Q) and upflow (14 ml/hr) is applied through a column waste height of 28 cm and a diameter of 10cm. - NEN 7345 Tank leach test [1]: In this test the specimen is subjected to leaching in a closed
697 tank. The leachant is renewed after 8 hours, 1, 2.25, 4, 9, 16, 36, 64 days at a leachant to product volume ratio (L/V) of approximately 5. The results are expressed in mg/m2. This test is a procedure to evaluate the release from monolithic material by predominantly diffusion control (e.g. exposure of structures to external influences). The distinction is necessary as the transport limitations set by a solid form result in a significantly lower environmental impact than derived from crushed material. This condition is valid as long as the product retains its integrity. To assess the behaviour after disintegration or demolition of monolithic forms, the information obtained in the pH static leach test is very relevant, as in this situation the pH is likely to change to more neutral conditions. - Compliance monolith leach test [2]. Although still in development by CEN TC 292 WG2 the main test features are: the specimen is leached after vacuum saturation with demineralised water at an liquid to area ratio of 5, eluates are produced after 6, 24 and 48 hours and subsequently analysed. - Compliance test for aggregates as developed in CEN TC 154 Aggregates[8]. This method consists of a leaching of the aggregate at a liquid to solid ratio of 10 (1/kg) for 24 hours. For this purpose the aggregate is placed on a grid above a magnetic stirrer. No particle size requirements are provided to limit the contribution of fines. - Concise leaching test [9], which has been proposed to address the main factors relevant to leaching from waste. It consists of four extractions for 24 hours at L/S-1 and L/S=I 0 at the material's own pH and two subsequent extractions at L/S=I 0 under pH control at pH=4 and neutral pH (or mild alkaline pH, when the material is neutral by itself). 2.3 Data treatment
The pH dependence test data are used as a basis of reference for data from other tests [6]. The data on the column percolation test are given as a function of the L/S. The relevant data from the compliance tests are inserted in these respective graphs. In the graphs, the regulatory criteria for granular materials (Category I and Category II of the Building Materials Decree [10]) are inserted for evaluation of critical nature of constituents. The data from the pH dependence test are used for geochemical modelling of potential solubility controlling minerals (not addressed here). The release from the intact pellets is addressed taking into account the diffusion from the interior of the pellets. For this purpose measurements were extended in one series up to several days of leaching. The release from concrete specimen is addressed to assess the contribution of aggregates to the overall material properties during both service life, recycling and "end of life" conditions.
3. RESULTS 3.1 Original wastes Granite w a s h i n g s - Granite washings are not critical from a leaching point of view for any of the parameters measured (about 28) according the Dutch Building materials Decree [ 10]. P a p e r sludge - Paper sludge is characterized by a very high TOC (dissolved organic carbon) and a relatively high TIC (carbonate). The elements Se, Mo, Ba and Cu only at pH > 10 exceed critical regulatory limits [10]. The material has a strong buffering capacity, as the amount of acid needed to increase the pH one pH unit is quite significant (2 Mol/kg). S h r e d d e r waste - The buffer capacity of shredder waste is relatively small. This implies that the material is sensitive to both acid and base influences. For instance in case of Zn, a small change in acid addition will lead to a significant increase in Zn leachability (more than
698 1 order of magnitude for 0.1 Mol/kg). The elements Cu, Cd, Mo, Sb, Zn and Pb exceed the regulatory criteria [ 10] over a large portion of the pH range studied. Cr and Sn become critical at high pH, whereas Ni and Co become critical at low pH. TOC shows a characteristic increase for many materials containing organic matter (increase of TOC with increasing pH). H a r b o u r s e d i m e n t - A small change in acid addition leads to a quite significant change in Zn leachability. The acid amount required to reach a pH below 5 is substantial. The agreement between CEN test data and the pH stat data are generally good. A duplicate analysis also leads to repeatable results, unless the concentrations become low and analyti,'al detection starts to play a role. For various elements it is clear that a relatively small change in pH may lead to very significant changes in leachability. Ni, Co and Zn show several orders of magnitude change within one to 2 pH units. M S W I n c i n e r a t o r bottom ash - Much information is available on this material [6,11,12], which indicates that Cu, Mo, C1 and Sb are potentially critical. In comparison with leaching data on synthetic aggregates produced from mixtures of wastes, generally significantly lower leachability is observed for many constituents. 3.2 Characterization of synthetic aggregate produced from waste The synthetic aggregates produced from one or a combination of the ,various waste streams have been tested for their leaching behaviour with several leaching tests, of which the pH dependence test proves to be the most versatile one. This test provides besides information on the pH dependence information on the acid neutralization capacity. In figure 1 the acid/base neutralization capacity for two typical aggregates is given. The sensitivity of this material to external pH influences is considerable as the acid neutralization capacity is quite low. When the aggregate is used in concrete for gravel replacement, this property has no further consequences. In storage exposed to rain, however, undesired leaching might occur. {
o.1 Synthethic Aggregates 0.08 ~ 0
0.06
Own pH
0.02 .
9
0
.o.0
-
.. ;..
, ..;
,
l',
Figure 1. Acid/base neutralization for synthetic aggregates illustrating the ]tow buffer capacity9 Two aggregate samples (a mixture of different wastes and a mixture of granite washings and paper sludge) have been subjected to a variety of different characterization leaching tests. This is done to set a reference for the behaviour of synthetic aggregate against which future
699 properties of materials to be produced with different secondary materials can be compared. The relation between the different tests needs to established beforehand to avoid surprises later. Based on the aggregate leaching results a first identification of critical components has been made relative to criteria from regulatory practice in the Netherlands (Building Materials Decree [10]). This regulation has both criteria for granular materials as well as for intact monolithic products against which the results obtained can be judged. The pH dependence test after size reduction (< 2 mm) of the two types of synthetic aggregates are compared to give a first indication of variability due to variation in source materials. In general, the leaching behaviour of both aggregates is quite comparable with some minor deviations (figure 2). The leaching curves as a function of pH are very consistent, which can be regarded as a characteristic for aggregate, which in its production process goes through a specific temperature - time profile. In figure 2 the leaching as function of pH is compared with the results of a column experiment on intact aggregate pellets (upflow) and a column experiment in downflow to simulate intermittent wetting/drying cycles. In general, the cumulative leached amounts as observed in the column leach tests follow a pattern, that corresponds to solubility (slope 1 in release/LS plot) for Ba and Zn. For Mo, the potential available fraction is depleted in L/S=2. Apparently, Mo is readily solubilized in the aggregate pores and subsequently washed out. The diffusion from the interior of the particles is apparently not a limitation (sufficiently porous and thus rapid exchange, see also discussion in section 3.3). The results for other samples of aggregate tested with a concise testing protocol[9], which are also indicated in figure 2, suggest that there is a consistent behaviour between aggregates produced from different starting materials. The aggregate production process tends to impose common characteristics in the leaching behaviour of aggregates with only individual deviations depending on the level of certain elements in the starting materials. The pH static leaching test data and the column leaching test results are internally consistent, as the cumulative leached amount at the L/S=I 0 in the column corresponds generally well with the pH dependence test (L/S=I 0) data at the corresponding pH. This feature allows translation of leaching test data from percolation tests to other exposure conditions at other pH conditions. The data also provide for a comparison between synthetic aggregate and the starting materials. Several elements - Li, Ca, K, Mn, Ca, Sr and Ba are incorporated in the silicate matrix in the sintering process. S is lost from the material by volatilization, thus leading to a lower leachability. Most other show little difference or are somewhat increased, such as Cr, V and Co. The acid~ase neutralization capacity between starting material and aggregates produced is markedly different. Due to the sintering process the neutral pellets have a very low buffer capacity. The degradation of calcite and the incorporation of Ca in the silicate structure are reasons for this change. The concise test offers a good option for an optimized quick testing protocol for quality control of the aggregate production process. This procedure leads to results within a few days. Several crucial aspects of leaching are addressed : solubility control, wash out, pH sensitivity over a wide pH range. Some specific properties can be tested on-site (so-called on site verification tests). These may include a simple test for water uptake by vacuum saturation and a measurement of conductivity.
700
Figure 2. Leaching of Ba, Mo and Zn from aggregate as a function of pH (left) and as a function of L/S (right). Two types of triangles point at column data for intact (10mm) and size reduced aggregate pellets (< 4 mm). Slope=l points at solubility control.
701 In comparison with regulatory criteria, the metals (such as Zn shown here) and Ba are of less concern than oxyanions, such as Mo (shown here), Sb and Se. For this evaluation, the relevant pH domain in application is from neutral to alkaline. The latter pH is relevant for the incorporation in concrete. When the aggregate is applied in unbound form, its sensitivity to external pH influences needs to be taken into account, which may put certain restrictions on its use in acidic environments. The increased leaching of oxyanions from sintered products is consistent with earlier observations on related materials -Lytag [ 13] and sintered bricks[ 14]. 3.3 Role of diffusion from aggregate pellets in a leaching test A key issue in the discussion on the leaching of pellets in a leaching test focuses on the particle size and the diffusion of constituents from the interior of the particles into the surrounding solution. The internal porosity is the crucial parameter here. The results from an up-flow column experiment (diameter 10 cm, height 28 cm, flow rate 14 ml/hr) on uniform aggregate (diameter around 10 ram, aggregate porosity 15 %) is compared with pH dependence test on the size-reduced aggregate (broken to < 2 mm, 24 hrs, L/S=I 0, own pH). In table I the results are given for the elements A1, B, Ca, K, Li, Mg, Mo, P, S, Si and Sr, which are accepted on the basis of sufficient analytical sensitivity. In the case of Ba (sufficient sensitivity), the pH difference between column and pH dependence test is important (steep gradient). Interpolation would lead to a better agreement for Ba. After log transformation (table II) the elements that are accepted before transformation also pass, but in addition elements at relatively low concentrations, such as Co, Cr, Cu, Fe, Mn, Se and Zn show a good agreement. The rational behind the log transformation is that at low concentrations a larger tolerance is acceptable. A factor 2 difference in concentration at a level of 0.01 mg/kg is the same!! This can only be shown after log transformation. The elements Cd, Pb, Ni and Sb are too close to the detection limit to be able to draw conclusions and therefor can not contribute positively to a conclusion, but do not challenge it either. At the percolation rates applied diffusion from the interior of uniform particles up to 10 mm is fast enough in the way the column is operated (particularly the flow rate of 15 ml/hr) to provide the same end results as a batch test with size-reduced material at LS=10. Only, when the porosity of the material is low (e.g molten slag) a more significant effect of delayed release can be expected and flow rate becomes a factor. However, under that condition the question arises, if extending the running time of a column test is useful or another way of determining the contribution of diffusive release should be addressed (compacted granular leach test [ 15]). In a comparison of the release at LS=10 for the column test (10 mm pellets) with the release from the same material broken to less than 2 mm at the corresponding pH (LS=I 0) in the pH dependence test, no significant difference is noted for several elements. By applying a log transformation, it is clear that the agreement even extends to the elements leached in low concentrations. The effect of crushing the aggregate has little effect on the leaching of elements at higher liquid to solid ratios (L/S > 2 1/kg) suggesting that the porous nature of the aggregate only leads to a relatively small delay in release. To address long term environmental impact size reduction is the better alternative for testing than running the test longer to reach a stable end point. This will also improve the repeatability of the test results.
702 Table I. Comparison of leaching from intact particles and equilibrium test data. Parameter Column (mg/kg) Intact ANC (mg/kg) Broken Ratio Column/ANC Aggregate (d= 10mm, aggregate (d <2 mm) uniform size) Cumulative L/S=10 release at L/S=I 0 pH 9.01 9.58 A1 3.58 2.99 1.20 B 4.47 4.69 0.95 Ca 131 164 0.80 K 7.9 11.1 0.71 Li 0.27 0.29 0.94 Mg 49.0 42.7 1.15 Mo 0.86 0.88 0.98 P 1.98 1.74 1.14 S 106 125 0.85 Si 26.7 24.7 1.08 Sr 0.32 0.35 0.91 Average 0.98 Std 0.15 N 11 Table II. Comparison of log transformed data. Log column Log pH stat (mgNg) (mg/kg) A1 0.554 0.476 B 0.651 0.671 Ca 2.117 2.215 Co -1.892 -1.699 Cr -1.690 -2.095 Cu -1.767 -1.843 Fe -1.528 -1.572 K 0.896 1.044 Li -0.568 -0.539 Mg 1.690 1.630 Mn -2.645 -2.301 Mo -0.067 -0.057 Na 1.390 1.683 P 0.297 0.240 S 2.026 2.097 Se -1.114 -0.878 Si 1.427 1.393 Sr -0.497 -0.456 Zn -1.906 -1.847 Average Std N
Ratio log columrLqog pH stat 1.165 0.969 0.9'56 1.114 0.807 0.9'59 0.972 0.858 1.054 1.037 1.149 1.175 0.826 1.234 0.966 1.270 1.024 1.090 1.032 1.032 0.128 1'9
703
3.4 Comparison of different leaching test methods To assess the performance of the compliance leaching tests, as developed in CEN TC 154 Aggregates, a comparison is made between tests on intact pellets and size reduced pellets. The following methods have been applied: column test on intact pellets to quantify the ultimate release at a longer time scale, batch tests using a rollertable on intact pellets for 24 up to 96 hours, pH dependence test on pellets size reduced to < 2 mm and the CEN TC 154 method. The comparison is complicated due to the different end pH in TC 154 method and the other tests. From the column test and batch test up to 96 hours contact time, it can be concluded that very similar results are obtained (table III). These data also correspond well with the equilibrium- based pH stat test data on size reduced pellets (<2 mm) after 24 hours at the same pH as column and batch test. This implies that a similar comparison between TC 154 and pH stat at the corresponding pH could be expected. However, the TC 154 data are in all cases significantly lower than the pH stat data at corresponding pH. For a second type of aggregate, the same has been observed. Apparently, the test conditions as specified in the TC 154 method are not stable as diffusion from the particles is not completed in 24 hours. Since it only takes a few days longer to reach more stable conditions, which would also be more appropriate in relation to assessing environmental impact, two options exist to modify the procedure. Either use a longer contact time for the test or apply size reduction to speed up the process. The ruggedness and reproducibility of the TC 154 test will be negatively affected by the present testing conditions. Table III.Comparison of size-reduced pellets with intact pellets in different test conditions. Aggregate 1 Aggregate 2 Column Batch pH stat pH stat TC 154 pH stat TC 154 L/S=10 T=96 hrs 24 hrs L/S=10 24 hrs 24 hrs 24 hrs Intact, pH L/S=10 L/S=10 24 hrs Intact L/S=10 Intact Parameter -- 9.3 -8.5 pH = 9.5 pH=9.5 pH=8.3 pH=8.3 pH=9.5 pH=9.8 Ba 0.11 0.033 0.035 0.18 0.014 0.91 0.38 Ca 130 125 164 365 55 666 63 K 7.9 10.8 11 19 2.1 30 2.7 PO4 (P) 1.9 0.91 1.7 1.1 0.54 1.6 0.065 SO4 (S) 106 100 124 119 48 144 35 Mo 0.86 0.61 0.88 0.58 0.50 0.079 0.055
3.5 Utilization of aggregates in concrete. The main intention for use of the aggregate is in the construction of light-weight concrete. The environmental evaluation is based on a scenario approach as described in ENV 12920 [ 16]. In this application, the leaching process is largely governed by the chemical environment that is imposed by the cement. Based on other work [17], the leaching of metals is hardly an issue in leaching from cement-based products. The main parameters in release from cementbased products are oxyanions, such as Cr VI, V, Mo, and in case of application in drinking water sector A1. A crucial issue for synthetic aggregates, is the extent to which the construction debris resulting from cement-based products containing these synthetic aggregates pose undesired long term effects in leaching after the leaching conditions imposed by the cement are no longer maintained. Recycling of concrete containing synthetic aggregate in new cement-based products will not likely lead to unacceptable release in the service life of
704
such new products. The significant pH change, as a result of the carbonation of the highly alkaline matrix upon degradation or forced size reduction, is important in this respect. The evaluation of cement-based products after their service life in unbound form should be based on size reduced material, which due to the enhanced exposure to the atmosphere is more strongly carbonated and as such has a lower pH than normal cement mortar or concrete. The changes brought about by this combination of effects are covered largely by the pH dependence leach test [5]. The range of leachability conditions that construction debris (including synthetic aggregate) can cover in their "second life" are reflected by the leachability behaviour going from pH 13 (porewater pH of cement) to neutral conditions pH 8 (calcite controlled system). This evaluation is illustrated in Figure 3. In the,' case of Cr the leachability of the cement aggregate mix (after size reduction to < 2 mm) is clearly dominated by the cement itself as the leaching data for reference mortar and synthetic aggregate mortar coincide (Cr leaching from aggregate is low). In the case of Mo, the leaching behaviour of crushed synthetic aggregate mortar is changed after incorporation in cement. Here the aggregate does increase the release level over that imposed by the cement iL,;elf. The data as obtained for size reduced material exceed the present regulatory limits :for construction materials in the "end of life" stage. The size range of construction debris can range from very coarse material (to be judged rather as a monolith) to fine material, which is best judged as a granular material. Here the latter is chosen, which is a worst case approach. Work to assess the intermediate particle size ranges and their contribution to the overall leaching process is underway. 100
10
Mo oi-a--
10
...........................
~
t
i
t
! ......................
m
\ V--
o ~ . . . . _ _ _ ..........
E 0,1
Aggr T Mortar ref. Pv~rtar A g g r . T B M D Cat I B M D cat II
/
--4
i m:
Ii .l
[ I O-k--O. y.... I ...... ;-o/
9 ...............................
0
.l
0,1
/
.... o/
/ .oJ
~ - - o .... c ..............
Io I/
21t / "
/
1 t
J
, /I
m
/
~m
~
,
-0'
i
0,01
Cr
i /
:,
e
=
0,001 1
I 3
I 5
I 7
plt
i 9
lll
J
0,01 1 3r
5m
7I
9(
1(1
plt
Figure 3. Leaching data for crushed synthetic aggregate mortar in comparison with the leaching behaviour of the synthetic aggregate. BMD = Building Materials Decree [10]. The relevant pH window is given as a box.
4. CONCLUSIONS When waste leachability is compared with leaching data on synthetic aggregates produced from mixtures of wastes, generally significantly lower leachability is observed for many constituents.
705 Due to the quite low acid neutralization capacity, the sensitivity of aggregates to external pH influences is considerable. When the aggregate is used in concrete for gravel replacement, this property has no further consequences. When used in contact with acidic environments, however, undesired leaching might occur. The leaching curves as a function of pH are very consistent, which can be regarded as a characteristic for synthetic aggregate. Apparently, the process characteristics tend to produce a material with similar properties in spite of varying inputs. This can be used for QC of the production process with a concise test addressed the most crucial leaching aspects. In a comparison of the release at LS=I 0 for the column test on 10 mm pellets with the same material broken to less than 2 mm at the corresponding pH at LS=10 in the pH dependence test, no significant difference is noted for almost all elements. By applying a log transformation, it is clear that the agreement even extends to the elements leached in low concentrations. The test conditions as specified in the CEN TC 154 method are not stable as diffusion from the particles is not completed in 24 hours. The inherent instability, which is most pronounced for pellets with moderate to high porosity, can be avoided by applying a longer contact time for the test or by applying size reduction. Upon application of synthetic aggregates in cement-bound products, the effect of the aggregate leachability is generally suppressed, except for Mo that shows a mild increased leachability of the final product. The main issue for aggregate use in concrete is to ensure that its behaviour in the recycling and end of life stage does not adversely affect the long term leaching behaviour of cement products.
REFERENCES
1. Tank leaching test - NEN 7345 Determination of leaching from monolithic contruction materials and waste materials by means of a diffusion test. October 1994 2. Compliance leach test CEN TC 292 Working Group 2 (1996): Characterization of waste. Leaching. Compliance test for leaching of granular waste materials. Determination of the leaching of constituents from granular waste materials and sludges. Draft European Standard prEN 12457. 3. Leaching characteristics of solid earthy and stony building and waste materials; Leaching tests; Determination of the leaching of inorganic components from granular materials with the column test, Dutch standard NEN 7343, NNI, 1995. 4. Characterization leaching test: Percolation simulation test CEN TC 292 Working Group 6 Work item 292016 (in prep.) 5. Characterization leaching test: Influence of pH under steady state conditions. CEN TC 292 Working group 6: Work item 292015 (in prep.) 6. Harmonization of leaching/extraction tests , 1997. Studies in Environmental Science, Volume 70. Eds H.A. van der Sloot, L. Heasman, Ph Quevauviller, Elsevier Science, Amsterdam, 292 pp. 7. H.A. van der Sloot. Developments in evaluating environmental impact from utilization of bulk inert wastes using laboratory leaching tests and field verification. Waste Management 16 (1-3), 1996, 65-81. 8. Test for chemical properties of aggregates - Part 3: Preparation of eluates by leaching of aggregates, CEN TC 154 Aggregates, 1999.
706
9. D.S. Kosson, and H.A. van der Sloot, H.A. Integration of Testing Protocols for Evaluation of Contaminant Release from Monolithic and Granular Wastes. In: Waste Materials In Construction- Putting Theory into Practice. Studies in Environmental Science 71. Eds. J.J.J.M. Goumans, G.J. Senden, H.A. van der Sloot. Elsevier Science Publishers, Amsterdam, 1997, 201-216. 10. Building Materials Decree. Staatsblad van het Koninkrijk der Nederlanden, 1995, 567. l l. IAWG (International Ash Working Group; A.J.Chandler, T.T.Eighmy, J.Hartlen, O.Hjelmar, D.S.Kosson, S.E.Sawell, H.A.van der Sloot, J.Vehlow). 1997. Municipal Solid Waste Incinerator Residues . Studies in Environmental Science 67, Elsevier Science, Amsterdam, 974 pp. 12. J.A. Meima. Geochemical modelling and identification of leaching processes in MSWI bottom ash. Thesis University of Utrecht, 1997. 13. C.W. Versluijs, I.H. Anthonissen and E.A. Valentijn. Integrale evaluatie van deelonderzoeken Mammoet '85. Report 738504008.1990. 14. CROW Handboek Uitloogkarakterisering, Deel I, II and III, CROW, Ede, The Netherlands, 1994 - 1996. 15. Determination of the maximum leachable quantity and the emission of inorganic contaminants from granular construction materials and waste materials - The compacted granular leach test. Concept Dutch pre-standard NVN 7347 NNI (1994). October 1994. 16. Characterization of waste - Methodology guideline for the determination of the leaching behaviour of waste under specified conditions. PrENV, CEN/TC 292. CEN (1996). 17. H.A. van der Sloot. Characterization of the leaching behaviour of cement mortars to assess long term environmental behaviour during their service life and their recycling stage. ECN RX-98-026.
695
Leaching behaviour of synthetic aggregates H.A. van der Sloot a, D. Hoede a, D.J.F Cresswell b and J.R Barton b. a E C N - Soil & Waste Research, P.O. Box 1, 1755 ZG Petten, The Netherlands b School of Civil Engineering, Woodhouse Lane, Leeds LS2 9JT, UK
In the framework of EU project Utilising innovative kiln technology to recycle waste into synthetic aggregate (BRST-CT98-5234), the leaching behaviour of synthetic aggregates has been studied to assess its environmental compatibility in the various stages of its use. Since the conditions are very different for the different uses, the assessment calls for a variety of different leaching conditions. The pH dependence test is used to cover important differences in pH environment to which the materials are exposed to as well as for an assessment of the buffering capacity of the material. Synthetic aggregate features a low buffer capacity, which makes it sensitive to externally imposed pH conditions. Utilisation and storage exposed to acidic conditions needs to be avoided. The pH dependence test and column leaching test are consistent. The CEN TC 154 method appears provide systematically low values due to the arbitrary selection of test conditions. Synthetic aggregate studied to date will not adversely affect the concrete in its service life. The main issue for aggregate use is the recycling and the "end of life" condition, when the material becomes construction debris. Not metals, but oxyanions, such as Cr VI and Mo are most relevant under these conditions. A concise test has been applied to assess crucial aspects of leaching for different production mixes.
1. I N T R O D U C T I O N In the framework of EU project Utilising innovative kiln technology to recycle waste into synthetic aggregate (BRST-CT98-5234), a new kiln design with a more efficient energy use is applied, which has more flexibility to process different waste materials as feedstock. The starting materials and the synthetic aggregates produced at laboratory scale at the School of Civil Engineering (Leeds) and at full scale facility in the south of England were tested for their leaching behaviour to assess their environmental compatibility in the various stages of their use. This includes storage prior to use, use as an aggregate in concrete, its recycling as construction debris in the same application, its reuse in other applications (e.g. road base construction) and its end-of-life situation (ultimate disposal). In all of these phases of use, the release of constituents from the material needs to be addressed to ensure environmental compatibility. Since the conditions are very different for the different uses, the assessment calls for a variety of different leaching conditions. Monolith leaching tests for bound applications [1,2] and percolation test [3,4] for several others. The pH dependence test [5] is used to cover important differences in pH environment to which the materials are exposed to as well as for an assessment of the buffering capacity of the material, which provides insight
696 in the sensitivity of the materials to extemally imposed conditions. In the framework of CEN TC 154, a compliance test for aggregates has been drafted, which contains an arbitrary choice of conditions unrelated to potential environmental impact. This method has been tested as well to verify its performance relative to the more elaborate evaluation.
2. E X P E R I M E N T A L
WORK
2.1 M a t e r i a l s
A range of materials have potential for use as feedstock for the aggrelgate production: granite washings, paper sludge, contaminated river sediment, MSWI bottom ash and shredder waste. The suitability of these materials in preparing pellets meeting technical specifications is studied at the School of Civil Engineering (Leeds). A key aspect is the water uptake of the pellets. The starting materials - sediments, municipal solid waste incinerator bottom ash (MSWI BA), shredder waste, milling residues, paper sludge - have been tested using the pH dependence test to identify the main leaching features of the starting materials for later comparison with the behaviour of pellets produced. Synthetic aggregate pellets, produced by firing in an innovative kiln at lab-scale from several combinations of sediment, municipal solid waste incinerator bottom ash, shredder waste, milling residues and paper sludge, and aggregates resulting from full scale aggregate production have been tested using the pH dependence test and the concise leaching test. On limited number of samples (due to large material amount required) the TC 154 method has been applied. On an aggregate sample from granite washings all tests including a column test on the intact pellets has been carried out. From a limited number of aggregate samples concrete specimen have been prepared and tested using the tank leaching test [1], a compliance tank leaching test [2] and the pH dependence test [5]. This allows an evaluation of the material in different stages of use, such as storage prior to use, use as an aggregate in concrete, its recycling as construction debris in the same application, its reuse in other applications (e.g. road base construction) and its endof-life situation (ultimate disposal). 2.2 M e t h o d s
The leaching tests selected for the study are: - pH static leach test [5]: This test provides information on the pH sensitivity of leaching
behaviour of the material. The test consists of a number of parallel extractions of a material at a liquid/solid ratio (L/S) 10 (1/kg) during 48 hours at a series of preset pH values. Since pH is one of the main leaching controlling parameters, the information can be used to evaluate the repeatability in testing (resulting from measurement at steep concentration - pH slopes) and to provide information on the sensitivity to pH in specific field scenarios. The acid neutralization capacity (ANC) derived from the test is a useful property in this respect. For material characterization this has been proven to be a very useful method [6,7]. The method is standardized in two experimental modes by CEN TC 292 Working Group 6. - Column leaching test with similarities to NEN 7343 [3] and a percolation test is developed at European level in CEN TC 292 WG6 [4], a column test in which 7 eluate fractions are collected within the range of L/S = 0.1-10 1/kg. The total test duration is approximately 21 days. The leachant is demineralized water (DMW). The test material is applied as received (around 1 cm Q) and upflow (14 ml/hr) is applied through a column waste height of 28 cm and a diameter of 10cm. - NEN 7345 Tank leach test [1]: In this test the specimen is subjected to leaching in a closed
697 tank. The leachant is renewed after 8 hours, 1, 2.25, 4, 9, 16, 36, 64 days at a leachant to product volume ratio (L/V) of approximately 5. The results are expressed in mg/m2. This test is a procedure to evaluate the release from monolithic material by predominantly diffusion control (e.g. exposure of structures to external influences). The distinction is necessary as the transport limitations set by a solid form result in a significantly lower environmental impact than derived from crushed material. This condition is valid as long as the product retains its integrity. To assess the behaviour after disintegration or demolition of monolithic forms, the information obtained in the pH static leach test is very relevant, as in this situation the pH is likely to change to more neutral conditions. - Compliance monolith leach test [2]. Although still in development by CEN TC 292 WG2 the main test features are: the specimen is leached after vacuum saturation with demineralised water at an liquid to area ratio of 5, eluates are produced after 6, 24 and 48 hours and subsequently analysed. - Compliance test for aggregates as developed in CEN TC 154 Aggregates[8]. This method consists of a leaching of the aggregate at a liquid to solid ratio of 10 (1/kg) for 24 hours. For this purpose the aggregate is placed on a grid above a magnetic stirrer. No particle size requirements are provided to limit the contribution of fines. - Concise leaching test [9], which has been proposed to address the main factors relevant to leaching from waste. It consists of four extractions for 24 hours at L/S-1 and L/S=I 0 at the material's own pH and two subsequent extractions at L/S=I 0 under pH control at pH=4 and neutral pH (or mild alkaline pH, when the material is neutral by itself). 2.3 Data treatment
The pH dependence test data are used as a basis of reference for data from other tests [6]. The data on the column percolation test are given as a function of the L/S. The relevant data from the compliance tests are inserted in these respective graphs. In the graphs, the regulatory criteria for granular materials (Category I and Category II of the Building Materials Decree [10]) are inserted for evaluation of critical nature of constituents. The data from the pH dependence test are used for geochemical modelling of potential solubility controlling minerals (not addressed here). The release from the intact pellets is addressed taking into account the diffusion from the interior of the pellets. For this purpose measurements were extended in one series up to several days of leaching. The release from concrete specimen is addressed to assess the contribution of aggregates to the overall material properties during both service life, recycling and "end of life" conditions.
3. RESULTS 3.1 Original wastes Granite w a s h i n g s - Granite washings are not critical from a leaching point of view for any of the parameters measured (about 28) according the Dutch Building materials Decree [ 10]. P a p e r sludge - Paper sludge is characterized by a very high TOC (dissolved organic carbon) and a relatively high TIC (carbonate). The elements Se, Mo, Ba and Cu only at pH > 10 exceed critical regulatory limits [10]. The material has a strong buffering capacity, as the amount of acid needed to increase the pH one pH unit is quite significant (2 Mol/kg). S h r e d d e r waste - The buffer capacity of shredder waste is relatively small. This implies that the material is sensitive to both acid and base influences. For instance in case of Zn, a small change in acid addition will lead to a significant increase in Zn leachability (more than
698 1 order of magnitude for 0.1 Mol/kg). The elements Cu, Cd, Mo, Sb, Zn and Pb exceed the regulatory criteria [ 10] over a large portion of the pH range studied. Cr and Sn become critical at high pH, whereas Ni and Co become critical at low pH. TOC shows a characteristic increase for many materials containing organic matter (increase of TOC with increasing pH). H a r b o u r s e d i m e n t - A small change in acid addition leads to a quite significant change in Zn leachability. The acid amount required to reach a pH below 5 is substantial. The agreement between CEN test data and the pH stat data are generally good. A duplicate analysis also leads to repeatable results, unless the concentrations become low and analyti,'al detection starts to play a role. For various elements it is clear that a relatively small change in pH may lead to very significant changes in leachability. Ni, Co and Zn show several orders of magnitude change within one to 2 pH units. M S W I n c i n e r a t o r bottom ash - Much information is available on this material [6,11,12], which indicates that Cu, Mo, C1 and Sb are potentially critical. In comparison with leaching data on synthetic aggregates produced from mixtures of wastes, generally significantly lower leachability is observed for many constituents. 3.2 Characterization of synthetic aggregate produced from waste The synthetic aggregates produced from one or a combination of the ,various waste streams have been tested for their leaching behaviour with several leaching tests, of which the pH dependence test proves to be the most versatile one. This test provides besides information on the pH dependence information on the acid neutralization capacity. In figure 1 the acid/base neutralization capacity for two typical aggregates is given. The sensitivity of this material to external pH influences is considerable as the acid neutralization capacity is quite low. When the aggregate is used in concrete for gravel replacement, this property has no further consequences. In storage exposed to rain, however, undesired leaching might occur. {
o.1 Synthethic Aggregates 0.08 ~ 0
0.06
Own pH
0.02 .
9
0
.o.0
-
.. ;..
, ..;
,
l',
Figure 1. Acid/base neutralization for synthetic aggregates illustrating the ]tow buffer capacity9 Two aggregate samples (a mixture of different wastes and a mixture of granite washings and paper sludge) have been subjected to a variety of different characterization leaching tests. This is done to set a reference for the behaviour of synthetic aggregate against which future
699 properties of materials to be produced with different secondary materials can be compared. The relation between the different tests needs to established beforehand to avoid surprises later. Based on the aggregate leaching results a first identification of critical components has been made relative to criteria from regulatory practice in the Netherlands (Building Materials Decree [10]). This regulation has both criteria for granular materials as well as for intact monolithic products against which the results obtained can be judged. The pH dependence test after size reduction (< 2 mm) of the two types of synthetic aggregates are compared to give a first indication of variability due to variation in source materials. In general, the leaching behaviour of both aggregates is quite comparable with some minor deviations (figure 2). The leaching curves as a function of pH are very consistent, which can be regarded as a characteristic for aggregate, which in its production process goes through a specific temperature - time profile. In figure 2 the leaching as function of pH is compared with the results of a column experiment on intact aggregate pellets (upflow) and a column experiment in downflow to simulate intermittent wetting/drying cycles. In general, the cumulative leached amounts as observed in the column leach tests follow a pattern, that corresponds to solubility (slope 1 in release/LS plot) for Ba and Zn. For Mo, the potential available fraction is depleted in L/S=2. Apparently, Mo is readily solubilized in the aggregate pores and subsequently washed out. The diffusion from the interior of the particles is apparently not a limitation (sufficiently porous and thus rapid exchange, see also discussion in section 3.3). The results for other samples of aggregate tested with a concise testing protocol[9], which are also indicated in figure 2, suggest that there is a consistent behaviour between aggregates produced from different starting materials. The aggregate production process tends to impose common characteristics in the leaching behaviour of aggregates with only individual deviations depending on the level of certain elements in the starting materials. The pH static leaching test data and the column leaching test results are internally consistent, as the cumulative leached amount at the L/S=I 0 in the column corresponds generally well with the pH dependence test (L/S=I 0) data at the corresponding pH. This feature allows translation of leaching test data from percolation tests to other exposure conditions at other pH conditions. The data also provide for a comparison between synthetic aggregate and the starting materials. Several elements - Li, Ca, K, Mn, Ca, Sr and Ba are incorporated in the silicate matrix in the sintering process. S is lost from the material by volatilization, thus leading to a lower leachability. Most other show little difference or are somewhat increased, such as Cr, V and Co. The acid~ase neutralization capacity between starting material and aggregates produced is markedly different. Due to the sintering process the neutral pellets have a very low buffer capacity. The degradation of calcite and the incorporation of Ca in the silicate structure are reasons for this change. The concise test offers a good option for an optimized quick testing protocol for quality control of the aggregate production process. This procedure leads to results within a few days. Several crucial aspects of leaching are addressed : solubility control, wash out, pH sensitivity over a wide pH range. Some specific properties can be tested on-site (so-called on site verification tests). These may include a simple test for water uptake by vacuum saturation and a measurement of conductivity.
700
Figure 2. Leaching of Ba, Mo and Zn from aggregate as a function of pH (left) and as a function of L/S (right). Two types of triangles point at column data for intact (10mm) and size reduced aggregate pellets (< 4 mm). Slope=l points at solubility control.
701 In comparison with regulatory criteria, the metals (such as Zn shown here) and Ba are of less concern than oxyanions, such as Mo (shown here), Sb and Se. For this evaluation, the relevant pH domain in application is from neutral to alkaline. The latter pH is relevant for the incorporation in concrete. When the aggregate is applied in unbound form, its sensitivity to external pH influences needs to be taken into account, which may put certain restrictions on its use in acidic environments. The increased leaching of oxyanions from sintered products is consistent with earlier observations on related materials -Lytag [ 13] and sintered bricks[ 14]. 3.3 Role of diffusion from aggregate pellets in a leaching test A key issue in the discussion on the leaching of pellets in a leaching test focuses on the particle size and the diffusion of constituents from the interior of the particles into the surrounding solution. The internal porosity is the crucial parameter here. The results from an up-flow column experiment (diameter 10 cm, height 28 cm, flow rate 14 ml/hr) on uniform aggregate (diameter around 10 ram, aggregate porosity 15 %) is compared with pH dependence test on the size-reduced aggregate (broken to < 2 mm, 24 hrs, L/S=I 0, own pH). In table I the results are given for the elements A1, B, Ca, K, Li, Mg, Mo, P, S, Si and Sr, which are accepted on the basis of sufficient analytical sensitivity. In the case of Ba (sufficient sensitivity), the pH difference between column and pH dependence test is important (steep gradient). Interpolation would lead to a better agreement for Ba. After log transformation (table II) the elements that are accepted before transformation also pass, but in addition elements at relatively low concentrations, such as Co, Cr, Cu, Fe, Mn, Se and Zn show a good agreement. The rational behind the log transformation is that at low concentrations a larger tolerance is acceptable. A factor 2 difference in concentration at a level of 0.01 mg/kg is the same!! This can only be shown after log transformation. The elements Cd, Pb, Ni and Sb are too close to the detection limit to be able to draw conclusions and therefor can not contribute positively to a conclusion, but do not challenge it either. At the percolation rates applied diffusion from the interior of uniform particles up to 10 mm is fast enough in the way the column is operated (particularly the flow rate of 15 ml/hr) to provide the same end results as a batch test with size-reduced material at LS=10. Only, when the porosity of the material is low (e.g molten slag) a more significant effect of delayed release can be expected and flow rate becomes a factor. However, under that condition the question arises, if extending the running time of a column test is useful or another way of determining the contribution of diffusive release should be addressed (compacted granular leach test [ 15]). In a comparison of the release at LS=10 for the column test (10 mm pellets) with the release from the same material broken to less than 2 mm at the corresponding pH (LS=I 0) in the pH dependence test, no significant difference is noted for several elements. By applying a log transformation, it is clear that the agreement even extends to the elements leached in low concentrations. The effect of crushing the aggregate has little effect on the leaching of elements at higher liquid to solid ratios (L/S > 2 1/kg) suggesting that the porous nature of the aggregate only leads to a relatively small delay in release. To address long term environmental impact size reduction is the better alternative for testing than running the test longer to reach a stable end point. This will also improve the repeatability of the test results.
702 Table I. Comparison of leaching from intact particles and equilibrium test data. Parameter Column (mg/kg) Intact ANC (mg/kg) Broken Ratio Column/ANC Aggregate (d= 10mm, aggregate (d <2 mm) uniform size) Cumulative L/S=10 release at L/S=I 0 pH 9.01 9.58 A1 3.58 2.99 1.20 B 4.47 4.69 0.95 Ca 131 164 0.80 K 7.9 11.1 0.71 Li 0.27 0.29 0.94 Mg 49.0 42.7 1.15 Mo 0.86 0.88 0.98 P 1.98 1.74 1.14 S 106 125 0.85 Si 26.7 24.7 1.08 Sr 0.32 0.35 0.91 Average 0.98 Std 0.15 N 11 Table II. Comparison of log transformed data. Log column Log pH stat (mgNg) (mg/kg) A1 0.554 0.476 B 0.651 0.671 Ca 2.117 2.215 Co -1.892 -1.699 Cr -1.690 -2.095 Cu -1.767 -1.843 Fe -1.528 -1.572 K 0.896 1.044 Li -0.568 -0.539 Mg 1.690 1.630 Mn -2.645 -2.301 Mo -0.067 -0.057 Na 1.390 1.683 P 0.297 0.240 S 2.026 2.097 Se -1.114 -0.878 Si 1.427 1.393 Sr -0.497 -0.456 Zn -1.906 -1.847 Average Std N
Ratio log columrLqog pH stat 1.165 0.969 0.9'56 1.114 0.807 0.9'59 0.972 0.858 1.054 1.037 1.149 1.175 0.826 1.234 0.966 1.270 1.024 1.090 1.032 1.032 0.128 1'9
703
3.4 Comparison of different leaching test methods To assess the performance of the compliance leaching tests, as developed in CEN TC 154 Aggregates, a comparison is made between tests on intact pellets and size reduced pellets. The following methods have been applied: column test on intact pellets to quantify the ultimate release at a longer time scale, batch tests using a rollertable on intact pellets for 24 up to 96 hours, pH dependence test on pellets size reduced to < 2 mm and the CEN TC 154 method. The comparison is complicated due to the different end pH in TC 154 method and the other tests. From the column test and batch test up to 96 hours contact time, it can be concluded that very similar results are obtained (table III). These data also correspond well with the equilibrium- based pH stat test data on size reduced pellets (<2 mm) after 24 hours at the same pH as column and batch test. This implies that a similar comparison between TC 154 and pH stat at the corresponding pH could be expected. However, the TC 154 data are in all cases significantly lower than the pH stat data at corresponding pH. For a second type of aggregate, the same has been observed. Apparently, the test conditions as specified in the TC 154 method are not stable as diffusion from the particles is not completed in 24 hours. Since it only takes a few days longer to reach more stable conditions, which would also be more appropriate in relation to assessing environmental impact, two options exist to modify the procedure. Either use a longer contact time for the test or apply size reduction to speed up the process. The ruggedness and reproducibility of the TC 154 test will be negatively affected by the present testing conditions. Table III.Comparison of size-reduced pellets with intact pellets in different test conditions. Aggregate 1 Aggregate 2 Column Batch pH stat pH stat TC 154 pH stat TC 154 L/S=10 T=96 hrs 24 hrs L/S=10 24 hrs 24 hrs 24 hrs Intact, pH L/S=10 L/S=10 24 hrs Intact L/S=10 Intact Parameter -- 9.3 -8.5 pH = 9.5 pH=9.5 pH=8.3 pH=8.3 pH=9.5 pH=9.8 Ba 0.11 0.033 0.035 0.18 0.014 0.91 0.38 Ca 130 125 164 365 55 666 63 K 7.9 10.8 11 19 2.1 30 2.7 PO4 (P) 1.9 0.91 1.7 1.1 0.54 1.6 0.065 SO4 (S) 106 100 124 119 48 144 35 Mo 0.86 0.61 0.88 0.58 0.50 0.079 0.055
3.5 Utilization of aggregates in concrete. The main intention for use of the aggregate is in the construction of light-weight concrete. The environmental evaluation is based on a scenario approach as described in ENV 12920 [ 16]. In this application, the leaching process is largely governed by the chemical environment that is imposed by the cement. Based on other work [17], the leaching of metals is hardly an issue in leaching from cement-based products. The main parameters in release from cementbased products are oxyanions, such as Cr VI, V, Mo, and in case of application in drinking water sector A1. A crucial issue for synthetic aggregates, is the extent to which the construction debris resulting from cement-based products containing these synthetic aggregates pose undesired long term effects in leaching after the leaching conditions imposed by the cement are no longer maintained. Recycling of concrete containing synthetic aggregate in new cement-based products will not likely lead to unacceptable release in the service life of
704
such new products. The significant pH change, as a result of the carbonation of the highly alkaline matrix upon degradation or forced size reduction, is important in this respect. The evaluation of cement-based products after their service life in unbound form should be based on size reduced material, which due to the enhanced exposure to the atmosphere is more strongly carbonated and as such has a lower pH than normal cement mortar or concrete. The changes brought about by this combination of effects are covered largely by the pH dependence leach test [5]. The range of leachability conditions that construction debris (including synthetic aggregate) can cover in their "second life" are reflected by the leachability behaviour going from pH 13 (porewater pH of cement) to neutral conditions pH 8 (calcite controlled system). This evaluation is illustrated in Figure 3. In the,' case of Cr the leachability of the cement aggregate mix (after size reduction to < 2 mm) is clearly dominated by the cement itself as the leaching data for reference mortar and synthetic aggregate mortar coincide (Cr leaching from aggregate is low). In the case of Mo, the leaching behaviour of crushed synthetic aggregate mortar is changed after incorporation in cement. Here the aggregate does increase the release level over that imposed by the cement iL,;elf. The data as obtained for size reduced material exceed the present regulatory limits :for construction materials in the "end of life" stage. The size range of construction debris can range from very coarse material (to be judged rather as a monolith) to fine material, which is best judged as a granular material. Here the latter is chosen, which is a worst case approach. Work to assess the intermediate particle size ranges and their contribution to the overall leaching process is underway. 100
10
Mo oi-a--
10
...........................
~
t
i
t
! ......................
m
\ V--
o ~ . . . . _ _ _ ..........
E 0,1
Aggr T Mortar ref. Pv~rtar A g g r . T B M D Cat I B M D cat II
/
--4
i m:
Ii .l
[ I O-k--O. y.... I ...... ;-o/
9 ...............................
0
.l
0,1
/
.... o/
/ .oJ
~ - - o .... c ..............
Io I/
21t / "
/
1 t
J
, /I
m
/
~m
~
,
-0'
i
0,01
Cr
i /
:,
e
=
0,001 1
I 3
I 5
I 7
plt
i 9
lll
J
0,01 1 3r
5m
7I
9(
1(1
plt
Figure 3. Leaching data for crushed synthetic aggregate mortar in comparison with the leaching behaviour of the synthetic aggregate. BMD = Building Materials Decree [10]. The relevant pH window is given as a box.
4. CONCLUSIONS When waste leachability is compared with leaching data on synthetic aggregates produced from mixtures of wastes, generally significantly lower leachability is observed for many constituents.
705 Due to the quite low acid neutralization capacity, the sensitivity of aggregates to external pH influences is considerable. When the aggregate is used in concrete for gravel replacement, this property has no further consequences. When used in contact with acidic environments, however, undesired leaching might occur. The leaching curves as a function of pH are very consistent, which can be regarded as a characteristic for synthetic aggregate. Apparently, the process characteristics tend to produce a material with similar properties in spite of varying inputs. This can be used for QC of the production process with a concise test addressed the most crucial leaching aspects. In a comparison of the release at LS=I 0 for the column test on 10 mm pellets with the same material broken to less than 2 mm at the corresponding pH at LS=10 in the pH dependence test, no significant difference is noted for almost all elements. By applying a log transformation, it is clear that the agreement even extends to the elements leached in low concentrations. The test conditions as specified in the CEN TC 154 method are not stable as diffusion from the particles is not completed in 24 hours. The inherent instability, which is most pronounced for pellets with moderate to high porosity, can be avoided by applying a longer contact time for the test or by applying size reduction. Upon application of synthetic aggregates in cement-bound products, the effect of the aggregate leachability is generally suppressed, except for Mo that shows a mild increased leachability of the final product. The main issue for aggregate use in concrete is to ensure that its behaviour in the recycling and end of life stage does not adversely affect the long term leaching behaviour of cement products.
REFERENCES
1. Tank leaching test - NEN 7345 Determination of leaching from monolithic contruction materials and waste materials by means of a diffusion test. October 1994 2. Compliance leach test CEN TC 292 Working Group 2 (1996): Characterization of waste. Leaching. Compliance test for leaching of granular waste materials. Determination of the leaching of constituents from granular waste materials and sludges. Draft European Standard prEN 12457. 3. Leaching characteristics of solid earthy and stony building and waste materials; Leaching tests; Determination of the leaching of inorganic components from granular materials with the column test, Dutch standard NEN 7343, NNI, 1995. 4. Characterization leaching test: Percolation simulation test CEN TC 292 Working Group 6 Work item 292016 (in prep.) 5. Characterization leaching test: Influence of pH under steady state conditions. CEN TC 292 Working group 6: Work item 292015 (in prep.) 6. Harmonization of leaching/extraction tests , 1997. Studies in Environmental Science, Volume 70. Eds H.A. van der Sloot, L. Heasman, Ph Quevauviller, Elsevier Science, Amsterdam, 292 pp. 7. H.A. van der Sloot. Developments in evaluating environmental impact from utilization of bulk inert wastes using laboratory leaching tests and field verification. Waste Management 16 (1-3), 1996, 65-81. 8. Test for chemical properties of aggregates - Part 3: Preparation of eluates by leaching of aggregates, CEN TC 154 Aggregates, 1999.
706
9. D.S. Kosson, and H.A. van der Sloot, H.A. Integration of Testing Protocols for Evaluation of Contaminant Release from Monolithic and Granular Wastes. In: Waste Materials In Construction- Putting Theory into Practice. Studies in Environmental Science 71. Eds. J.J.J.M. Goumans, G.J. Senden, H.A. van der Sloot. Elsevier Science Publishers, Amsterdam, 1997, 201-216. 10. Building Materials Decree. Staatsblad van het Koninkrijk der Nederlanden, 1995, 567. l l. IAWG (International Ash Working Group; A.J.Chandler, T.T.Eighmy, J.Hartlen, O.Hjelmar, D.S.Kosson, S.E.Sawell, H.A.van der Sloot, J.Vehlow). 1997. Municipal Solid Waste Incinerator Residues . Studies in Environmental Science 67, Elsevier Science, Amsterdam, 974 pp. 12. J.A. Meima. Geochemical modelling and identification of leaching processes in MSWI bottom ash. Thesis University of Utrecht, 1997. 13. C.W. Versluijs, I.H. Anthonissen and E.A. Valentijn. Integrale evaluatie van deelonderzoeken Mammoet '85. Report 738504008.1990. 14. CROW Handboek Uitloogkarakterisering, Deel I, II and III, CROW, Ede, The Netherlands, 1994 - 1996. 15. Determination of the maximum leachable quantity and the emission of inorganic contaminants from granular construction materials and waste materials - The compacted granular leach test. Concept Dutch pre-standard NVN 7347 NNI (1994). October 1994. 16. Characterization of waste - Methodology guideline for the determination of the leaching behaviour of waste under specified conditions. PrENV, CEN/TC 292. CEN (1996). 17. H.A. van der Sloot. Characterization of the leaching behaviour of cement mortars to assess long term environmental behaviour during their service life and their recycling stage. ECN RX-98-026.