The potential performance of steel industry aggregates for the removal of heavy metals from degraded water

Recently it was speculated that the next World War will be over water. There are dire forecasts that many third world nations are going to run out of water as a result of climate change, while a number of first world countries, like parts of Europe and certain areas of the USA, are already experiencing unusual drought conditions. There is also growing alarm about the damage that degraded water is doing to our environment, either through the degradation of lakes and rivers through an over abundance of nutrients, or the contamination of marine silts and sediments with heavy metals that can, and in some cases already have, entered the food chain.

This paper is about the ability of the slag industry's aggregate to clean up water that has become degraded with heavy metals, either before it is discharged or before it is put to some other beneficial use such as irrigation.

Impervious surfaces are a major contributor to urban storm water (also known as surface water run-off) impacts. Storm water in turn, has been identified as a major contributor to water quality degradation, as it can have significant concentrations of harmful pollutants that can adversely affect the receiving aquatic environment. Materials that can remove harmful pollutants and can be incorporated into storm water treatment devices offer part of the urban storm water solution. Previous research into the use of iron and steel furnace slag aggregates as water filtration media has demonstrated high adsorptive capacity for various metals and phosphorus and the removal of fine particulates in the source solution.

Encouraged by previous research and some promising results obtained with slag aggregate filter beds in New Zealand, The Australasian (Iron and Steel) Slag Association (ASA), commissioned a study into the effectiveness of common iron and steel slag aggregates to remove contaminates typically found in storm water run-off, while maintaining their hydraulic performance. The study also provides data on the potential environmental effects and effectiveness of different iron and steel slags produced in New Zealand and Australia, for remediation of storm water pollution.

This paper provides an overview of the ASA commissioned study, as well as describing some of the in-field results observed with filter beds in New Zealand. A full copy of the ASA study maybe obtained by contacting the ASA, the address of which is listed in the references.¹

Introduction

The ability of iron and steel industry slag aggregates to remove phosphorus and certain heavy metals from degraded water has been shown in previous small-scale studies over many years e.g. Yamada in 1986 and Mann in 1997.^2,3 The international use of these materials for this purpose has, however, been sporadic for a variety of reasons.

In New Zealand, iron making or 'melter slag' has been successfully used as a filtration media for waste water treatment (targeting phosphorous and suspended solids), as well as for the treatment of steel mill site storm water and the removal of phosphorous from dairy farm run-off.⁴

The ASA felt that although the New Zealand work had produced promising results, it had been based on that country's unique iron sand-sourced steel industry aggregates, rather than the more conventional iron ore or scrap-based process routes and therefore, a more in-depth study was required if the work was to have international significance. To this end Landcare Research, a Crown Research Agency in New Zealand, was asked to undertake a study into the potential for a number of Australasian slags to remove heavy metals from storm water. Landcare Research was chosen to carry out the work as it had already conducted a number of studies into the use of novel materials for the treatment of storm water and as a result, had a database with which to compare the results of this new work.

Methods

An assessment of the environmental impact of iron and steel slag aggregates was carried out by a desktop search of scientific literature and via the internet. The resulting data was collated and potential environmental contaminants from iron and steel slag were identified.

Six different types of iron and steel slag were supplied by ASA members (Table 1) and were tested under laboratory conditions as a storm water treatment medium. At the request of the ASA, Landcare was asked to subject the aggregates to a wide range of heavy metals, many of which are not common, or are only present in small quantities in storm water. Experience in Australia in dealing with Environmental Protection Agencies (EPAs), on slag related leachate issues dictated the inclusion of the additional elements noted in this table. This 'M17' list is therefore an Australian classification and not an international one (Table 2).

Table 1: Slag origins, type and code used in the filtration study.

Table 2: The 'M17' metals.

The ability of the slags to remove contaminants was investigated by irrigating filter media columns with artificial storm water containing the phosphorus, nitrogen and the list of M17 'metals' and measuring the concentration in the resulting leachate. The usefulness of iron and steel slag products as a treatment media in storm water treatment devices was assessed, taking into account the results of the laboratory study and published literature on the use of slag materials in other types of treatment system.

Laboratory experimental design

The experiment was carried out under laboratory conditions. The slag materials were chosen and supplied by the ASA (Table 1). X-ray fluorescence (XRF) analysis of the slag samples was carried out by the Environmental Engineering Laboratories at New Zealand Steel using industry standard procedures.

Saturated hydraulic conductivity of the filter media was measured on repacked 100mm cores using the constant head method (Figure 1).⁵ A head of 0mm was set for the disk permeameters (Figures 2-3) to allow direct reading of K_sat. Five litres of each filter media was weighed and then packed into 150mm diameter columns made of high density polyethylene.

The filter media were tested for their contaminant retention capacity by applying six applications of artificial storm water using a peristaltic pump to control flow rate. Artificial storm water was used instead of natural storm water to facilitate consistency between applications. The alternative, stored storm water, is not stable, with solid and dissolved fractions of contaminants changing over time.⁶ The metals Cu and Zn are common contaminants of concern in urban storm water, present in the dissolved form.⁷

Similarly the nutrients P and N are considered 'typical' problem contaminants in urban storm water. The other contaminants in the M17 suite were requested by the ASA. The concentrations tested were similar to those reported along arterial roads in Australia⁸ and New Zealand⁹ or, for those contaminants lacking local measurements, from The International Storm water BMP Database (2005). No particulates were added to the artificial storm water.

Results and discussion

The following are abbreviated results from the full study and concentrate on the key elements normally found in storm water.

Environmental impact
Desktop and internet research produced a number of papers that have examined the potential toxicity of iron and steel industry aggregates¹⁰ and their potential impact on human health and ecological risk.¹¹ A general concensus throughout the examined papers was, that although slag has elevated total concentrations of trace metals, very little is likely to leach into the environment and that there is a minimal hazard to both the environment and health.

pH
It is preferable that the pH in the leachate remains between pH 5 and 9 as some metals, such as aluminium, are more toxic at both low and high pH. The six slag filter media reduced the acidity so that the pH in the leachate was more than that applied as storm water. Three of the slag tested (MS1, EAF1 and BF1), met the Australian and New Zealand guidelines for fresh and marine water quality (ANZECC and ARMCANZ 2000), of between pH 5.0 and 9.0. However, these are guidelines for pristine waters, not limits for contaminated water. Neither the Food Standards agencies of Australia and New Zealand, nor the US EPA has set limits for the pH of drinking water. The importance of acidity is that it affects the availability and toxicity of many trace metals in water. The pH of the slag material had a clear impact on many of the metal concentrations measured in this study, with slag with pH close to neutral being differentiated from those which were alkaline. At higher pH values, many metals will precipitate out of solution as metal hydroxides or other salts, reducing toxicity. However, high pH values can impact on aquatic biota.

Hydraulic conductivity
High hydraulic conductivity is preferable as it allows treatment of greater volumes of water. The hydraulic conductivity for all six slags was greater than could be supplied (K_sat > 100,000mmh^-1), and these slags have the potential to pass large volumes of storm water. Hydraulic conductivity is greater than that typical of unconsolidated gravel (11250mmh^-1).¹² Therefore, no practical hydraulic restrictions are expected from the slags until and unless they become clogged.

Retention of heavy metals
All but one of the slags significantly retained aluminium (Table 3), the exception being the BO1. The levels of aluminium released by this slag were however, not expected to create a significant environmental impact as they were below 18mg per litre that have been measured in bottled water (FSANZ press release 2004). Dilution in the receiving environment would reduce concentrations further.

Tables 3-9: Leachate retention data of selected M17 metals through slag samples.

All slags retained cadmium although MS1 and BF showed a steady decline in retention over the six runs (Table 4). Both these slags had a near neutral pH. Only the 100% scrap based EAF slag was unable to retain chromium (Table 5), possibly due to this element being present in the scrap feed stock. Copper is one of the most prominent contaminants of concern in road run off and all of the slags tested manage to retain this element to very high levels (Table 6). Historically, lead has been one of the most prominent contaminants of concern on road run-off, although the introduction of unleaded fuels has greatly reduced the presence of lead in road run-off storm water. All slags tested retained lead to a high degree (Table 7).

Although all slags tested managed to retain zinc, MS1 and BF1 with their near neutral pH showed a steady decline in retention over the six runs, while those with higher pH values showed strong retention potential, as zinc is more strongly retained at high pH levels (Table 8). Zinc retention by New Zealand Steel's slag filters over a number of months has, however, shown no noticeable drop off in performance and is discussed below.

Phosphorous is undesirable in storm water since, along with nitrogen, it can promote algal growth. All six slag aggregates retained phosphorous although the EAF slag tested showed the lowest results (Table 9).

Summary

• All six slags tested met the estimated removal rate targets for some but not all of the 'M17' metals;
• Five of the slags met the phosphorous removal rate;
• None met the nitrogen removal rates;
• The impact of pH on removal efficiency was noticeable, the more alkaline slags being more effective, while the near neutral slag's performance started to drop off.
• Although this study gives a potential indicator for similar slag types, industry experience over many years would suggest that slags from individual sites must be tested to ascertain their respective performance properties, as different sources of iron ore, coal or furnace additives can have a considerable impact on their water treatment potential.

Similarly, differences can be noted with the actual performance of slag aggregate filter beds in the field, as opposed to laboratory experiments using artificial mixes. For example, results obtained from the melter slag aggregate storm water filters at New Zealand Steel have shown a retention level of approximately 75% over a period of eight months, compared to the laboratory performance of 48% noted in Table 10.

Table 10: Heavy metal reduction in New Zealand waste water treatment plants.

Other field results in New Zealand

Melter slag aggregates have been used as filtration media for waste water treatment plants (WWTP) in New Zealand since 1993.⁴ Although their primary function is to reduce the level of suspended solids by polishing the water prior to discharge, the retention of phosphorous and certain heavy metals was detected shortly after the first installation.

Subsequent monitoring has shown that while the retention of phosphorous dropped off after some five years, a continued reduction of heavy metals has been sustained. The performance of two WWTP in this regard is shown in Table 11. One litre grab samples were obtained from each plant's inlet and outlet in August 2006 and analysed by New Zealand Steel's Environmental and Process Engineering Laboratories using standard ICP procedures.

The long term performance of slag aggregates as a filtration media is of considerable importance to water treatment engineering consultants and their clients when considering relative treatment options. The above results perhaps give some indication that these materials have the potential to deliver a satisfactory service life approaching 15 years. It must also be remembered that slag aggregates are treating these elements in solution as well as in suspension. Sand, one of the most commonly used filtration aggregates, has been shown to remove contaminants from run-off but to be unable to retain them as the contaminants are 'trapped' in the inter-pore spaces and are easily washed out by then next flush of water.¹³

Conclusions

All the slag aggregates tested in this study have potential as storm water filter media. They are a potential substitute for sand in filtration devices and in infiltration devices such as filter strips. In these laboratory experiments, the ability of slag was tested to remove the most biologically available and most difficult fraction of contaminants to remove – the dissolved fraction. Other potential uses are treatment of landfill leachate, agricultural, domestic and industrial waste water treatment and treatment of acid mine drainage.

Within certain limitations, slag aggregates have the potential to make a significant contribution to the treatment of degraded water, perhaps as stand alone filters where there is sufficient room for the filter bed footprint, or in 'treatment trains', in conjunction with proprietary storm water treatment devices, or in addition to conventional filter aggregates such as sand. In the latter case their role is one of being able to deliver a final polish, reducing residual metals that remain in solution. The principal limitation is pH with some slags, care being required to ensure that biota in the receiving water body will not be adversely effected if there is insufficient dilution of the discharge effluent.

It is recommended that further study by the international slag community should be undertaken to gain a better understanding of long term filter performance and into potential alkaline treatment devices to offset high pH discharges. The treatment of degraded water is rapidly becoming a key issue for federal, state and municipal authorities and presents a significant opportunity and market for slag based aggregates over their naturally quarried equivalents.

Comments

This paper began with a rather somber statement about water, a commodity that we all take for granted and indeed one which is a very essential ingredient to life on this planet.

The 'Alchemist' newsletter in 2006 reported that European scientists have devised a simple filter to remove arsenic salts from contaminated drinking water, an insidious health threat to countless people who rely on wells for drinking water, particularly on the Indian sub- continent. The filter uses an iron oxide-coated sand. As New Zealand Steel's chief chemist commented, 'guess what slag is'!

Moderate drought conditions currently involve 25% of the earth's surface. Scientists at the Hadley Centre in the UK – using a supercomputer climate model programme – have forecast that this will rise to 50% by 2100. The figure for extreme drought rise is expected to rise from 3% currently, to 30% by 2100. Vast numbers of human beings will be dramatically effected.

It is also reported that wealthy investors, particularly in the USA, are rapidly buying up water rights or the options thereto, leading one observer to comment that water in the 21st century will become what oil was in the 20th.

Even if these forecasts are only half right, clean water, or improved waste water for irrigation, will become a highly sought after commodity. This study undertaken by the ASA, and the work being carried out by scientists such as Dr Drizo at the University of Vermont and Dr Lena Johansen in Sweden, plus the modest work we have done with these materials in NZ, confirms that slag has a role to play in saving this precious resource.

Acknowledgements

To Mathew Taylor of Landcare Research who prepared the original report and oversaw the conduct of the experiment. To the Australasian (Iron and Steel) Slag Association for permission to draw extensively on the final report and to New Zealand Steel's Environmental and Process Engineering Laboratory for XRF analysis and ICP determination of water samples.

References

1. Australasian (Iron and Steel) Slag Association, PO Box 1194, Wollongong, NSW 2500, Australia. A copy of the full study "An assessment of Iron and Steel Slag for the Treatment of Storm water Pollution" may be found on the Association's web site at www.asa-inc.org.au/environment.

2. H. Yamada, M. Kayama, K. Saito and M. Hara, 'A Fundamental research on phosphate removal by using slag', Water Research, 1986, 20(5), p547–577.

3. R. A. Mann, 'Phosphorous adsorption and disposition characteristics of constructed wetland gravels and steel works by-products', Australian Journal of Soil Research, 1997, 35, p375–384.

4. W. S. Bourke, S. Bilby, D. P. Hamilton, R. McDowell, 'Recent water improvement initiatives using melter slag filter materials', New Zealand Waste Water Publications, EnviroNZ05 Conference, Auckland, New Zealand, 2005.

5. A. Klute, C. Dirksen, 'Hydraulic conductivity and diffusivity: laboratory methods', Methods of soil analysis, part 1: Physical and mineralogical methods. American Society of Agronomy – Soil Science Society of America, Agronomy Monograph 9, 1986, p687–734.

6. G. A. Burton, R. Pitt, Storm water effects handbook, 2001, Boca Raton, USA, CRC Press.

7. E. Shaver, R. Horner, J. Skupien, C. May, G. Ridley, 'Fundamentals of urban run-off management: Technical and institutional issues', North American Lake Management Society in cooperation with U.S. Environmental Protection Agency, 2005, 2nd Edition.

8. A. Kumar, M. Woods, A. El-Merhibi, D. Bellifemine, D. Hobbs, H. Doan, 'The toxicity of arterial road run-off in metropolitan Adelaide: stage 2', Final Report to Transport SA, June 2002, Adelaide, Transport SA.

9. M. D. Taylor, S. Pandey, 'Road run-off contaminant removal by a treatment wall constructed at the Hewletts Rd/Tasman Quay Roundabout, Mount Maunganui: final report', Landcare Research Contract Report 0405/136 to Bay of Plenty Regional Council, 2005, Whakatane.

10. D. M. Proctor, K. A. Fehling, E. C. Shay, J. L. Wittenborn, J. J. Green, C. Avent, R. D. Bigham, M. Connolly, B. Lee, T. O. Shepker, M. A. Zak, 'Physical and chemical characteristics of blast furnace, basic oxygen furnace, and electric arc furnace steel industry slag', Environmental Science and Technology, 2000, 34, p1576–1582.

11. D. M. Proctor, E. C. Shay, K. A. Fehling, B. L. Finley, 'Assessment of human health and ecological risks posed by the uses of steel-industry slag in the environment', Human and Ecological Risk Assessment, 2002, 8, 681–711.

12. E. Shaw, Hydrology in Practice, 1993, p128, Los Angeles, California, Chapman and Hall, 2nd Edition.

13. S. Clark, R. Pitt, R. Field, 'Storm water treatment at critical areas: evaluation of filtration media', Cincinnati, Ohio, National Risk Management Research Laboratory, Office of Research and Development, US Environmental Protection Agency, 1999, p1–96.