Mineralogy has many applications in the business of waste and waste disposal:

1. • precipitation of minerals and adsorption of elements and compounds to mineral surfaces can serve to immobilize pollutants or concentrate them for recycling

2. • release of elements and compounds can occur by the converse processes of dissolution and desorption

3. • minerals can act as impermeable barriers between pollutants and the environment

4. • minerals can be used as analogues to determine the long-term stability of engineered materials used for the encapsulation of waste.

The Applied Mineralogy Group of the Mineralogical Society organized a meeting ‘The mineralogy of waste and waste disposal’ held at the British Geological Survey, Keyworth, Nottingham, on 25th September 2000. The meeting attracted members of both the academic (Earth Science, Chemistry and Material Science departments) and industrial communities. A wide variety of papers was presented dealing with such diverse issues as nuclear waste disposal and the recycling of phosphorus from wastewater. Seven of the papers are collected here.

Land south of Lille, contaminated by heavy metals from local metal refining, has become the subject of intensive research. Topsoil from this area is used in the current work to investigate the growth behaviour of Spring wheat (Triticum aestivum L., cv. Paragon) when amended with organo-zeolitic fertilizer. Research has shown that soil substrates amended with ammoniated zeolitic tuff promote large populations of nitrifying bacteria which, as the result of ensuing enzyme reactions, produce available nitrogen together with H+ ions. It appears that the proton activity promotes cation mobilization allowing plant uptake to reach levels which, to a large extent, satisfy the nutritional requirements of the plant; only in the case of Zn is the level far exceeded. Very large differences in plant morphology occur when plants are grown in the amended substrates and the resulting development of large dense root systems, which have the ability to bind soil particles, are particularly important in their ability to reduce surface erosion and pollution from run-off.

Extensive use of cement and concrete is envisaged in the construction of geological disposal facilities for radioactive wastes. The hyperalkaline porefluids typical of groundwaters that have reacted with these materials have the potential to react chemically with other engineered barrier components such as bentonite, potentially degrading their performance. Analcime, NaAlSi2O6.H2O, has been identified from previous modelling and experimental studies as a potential alteration product of bentonite.

Laboratory experiments to investigate the stability of analcime under hyperalkaline porefluid conditions have been performed. Experiments used both batch and fluidized bed equipment at 25, 50, 70 and 90°C in K-based pH buffer solutions, both under- and over-saturated with respect to analcime. Results from dissolution experiments demonstrate that release of Na was greatly enhanced (by up to a factor of thirty) over that for Si and Al, particularly at pH 10 and 11. However, enhanced release of both Na and Al occurred in the batch experiments at pH 12–13. Near stoichiometric dissolution was observed in fluidized bed experiments under steady-state conditions at 70°C. Sodium was removed from the analcime structure by ion exchange for K, without involving dissolution and re-precipitation of the analcime framework. Scanning electron microscopy of reacted analcime grains showed that some grains had pronounced cracks parallel to original cleavage traces. These cracks were a result of volume decrease due to the substitution of K for Na ions and water molecules in the analcime structure to form leucite, KAlSi2O6.

Synthesis of the dissolution data shows that the rate of dissolution increased with increasing temperature in the range 25–70°C and with pH at each temperature. Absolute rates of dissolution ranged from 10−10 mol m−2 s−1 at pH 9.5 at 25°C to 10−7 mol m−2 s−1 a pH 12 at 70 and 90°C. The rate of dissolution at any temperature was pH-dependent, such that the rate could be described by k (aH+)n, where k is the rate constant and n is −0.3 at 25°C, −0.4 at 50°C, −0.6 at 70°C and −0.7 at 90°C. Attempts to measure the growth rate of analcime in supersaturated solutions at 70 and 90°C were unsuccessful, although a limiting rate at 70°C, pH 10 was calculated to be 4 × 10−11 mol m−2 s−1, roughly 100× less than the rate of dissolution under the same conditions.

These results imply that any trace amounts of analcime in bentonite will be converted to leucite by reaction with cement fluids with a high K/Na ratio. In some instances, leucite may thus incorporate K+ in preference to other phases (e.g. illite, K-feldspar) during alteration of bentonite by cement porefluids.

This paper provides a brief perspective on synthetic, phosphate-based waste forms for high level radioactive waste (HLW). Evidence in support of their long-term stability is then discussed by reference to the degradation of natural monazites with emphasis on the fate of released uranium, thorium and the rare earths (REE). It is apparent that the REE can be mobilized and fractionated at temperatures anticipated in a HLW repository (∼200°C). This provides an indication of the likely fate of the trivalent actinides (Am(III), Cm(III)) if incorporated in similar matrices. Thorium, though released on alteration of monazite, tends to re-concentrate locally in secondary, microcrystalline phases. In relative terms, U is readily removed from monazites. Although it can be re-concentrated in alteration products, the potential exists for substantial loss of U to groundwater. The findings of this research have important implications for the performance of radioactive waste disposal systems where there is a clear need for improved chemical data to describe the precipitation-dissolution of phosphate phases. It is concluded that monazite-like ceramics designed for the containment of HLW will retain tetravalent actinides but may release uranium in response to natural degradative processes.

Routine monitoring of landfill leachates has been extended to include characterization of suspended solids recovered by filtration. Calcite is consistently identified as a suspended solid, with less frequent reports of quartz and clays (kaolinite, illite, chlorite). Morphologically, calcite occurs as discrete grains, coatings on quartz sand and as microconcretions. Preliminary stable isotope data for seven samples generally show positive δ13C values (relative to PDB) up to +3.5‰ and δ18O values between −5 and −8‰, consistent with an origin through precipitation from leachate. Geochemical modelling of leachate compositions for the same samples indicates that the leachates are saturated with respect to calcite, and that the degree of supersaturation decreases for older samples. Mass balance considerations show that the proportions of methane and carbon dioxide observed for landfill gas do not reflect the amount of bicarbonate that is potentially available from the anaerobic decomposition of putrescible waste. Overall, putrescible waste has the potential to form a maximum of 1.9 g of calcite for every gram of waste, although values less than this are likely to be achieved in practice. From these differing lines of evidence, there can be no doubt that calcite precipitation should be expected to take place within landfill systems as an essential part of the waste degradation and stabilization process, and should be considered in modelling both gas evolution and carbon emissions.

The removal of phosphorus from wastewaters is becoming very common, to meet water quality targets, and avoid environmental problems related to eutrophication. At the same time, agricultural application of P-rich sewage biosolids is diminishing for reasons of logistics and of public pressure. As a result P from wastewaters is ultimately disposed of in landfills. Over the long term, phosphate ore reserves will become depleted. Recycling of P from wastewaters may thus be a realistic prospect if scientific and technical issues can be resolved. At the centre of the scientific problems lie considerations about optimizing phosphate precipitation as Ca phosphates. A number of Ca phosphate minerals exist, although by far the most common of these is hydroxylapatite. Precipitation kinetic considerations, however, suggest that other Ca phosphates (such as brushite, octacalcium phosphate, whitlockite, monetite or amorphous Ca phosphate) may initially precipitate, and later recrystallize into the most stable hydroxylapatite. This article reviews the complex precipitation mineralogy, chemistry and kinetics of Ca phosphates.

Some of the physical, chemical and mineralogical characteristics of Portland cement and related materials relevant to nuclear waste immobilization are defined. The ability to condition and maintain a high aqueous pH is undoubtedly the most important factor: it precipitates many species as hydrous oxides or hydroxides. However, in the longer term, many species – cationic as well as anionic – react with one or more cement components forming solubility-limiting phases. Progress on characterization of these phases is outlined. Many of the host phases have natural equivalents and this gives comfort in respect of their likely persistence over geological time. The emerging picture of immobilization in cement suggests that cement compositions can be tailored in terms of pH, Eh and internal chemistry so as to maximize immobilization potential. Nickel, uranium and chromium and chloride are used as examples.

Air pollution control (APC) facilities at waste incinerator plants produce large quantities of solid residues rich in salts and heavy metals. Heavy metals are readily released to water from the residues and it has, therefore, been found suitable to apply a rapid co-precipitation/adsorption process as a means to immobilize the toxic elements. In the ‘Ferrox process’, this immobilization is based on co-precipitation with an Fe(III) oxide formed by oxidation of Fe(II) by air in an aqueous slurry with the APC residue at alkaline pH. In this work we have undertaken a Mössbauer spectroscopy study of the Fe oxide phase formed by precipitation at room temperature and of the oxides present after heating to 600 and 900°C. The only Fe oxide observed in the Ferrox product at room temperature is a very poorly crystalline ferrihydrite. Analytical transmission electron microscopy showed that the main elements associated with the ferrihydrite are Si and Ca. Following heating to 600°C the oxide is still characterized as an amorphous Fe oxide, and it is probable that Si associated with the ferrihydrite is decisive in preventing crystallization. After the 900°C treatment a transformation into defect maghemite is observed. Reducing gases produced from carbon in the samples probably induces this transformation. It eases, thus, the reduction of Fe(III) and the consequent formation of magnetite that eventually oxidizes to maghemite during cooling in air.

In order to investigate the range of the solid solution series in herzenbergite-teallite minerals, samples of different composition were synthesized. Herzenbergite-teallite minerals were synthesized by an evacuated silica glass tube method at 700°C. A linear relationship between cell dimensions, a, b and c and composition is established. Extension of solid solution to the Pb-rich portion of the system PbS-SnS is limited; the solid solution area is between Pb1.060Sn0.940S2 and SnS at 700°C. Teallite coexisting with galena was also synthesized by hydrothermal recrystallization at 300, 400 and 450°C. The compositions of teallite are Pb1.140Sn0.860S2 at 300°C, Pb1.114Sn0.886S2 at 400°C, and Pb1.124Sn0.876S2 at 450°C, respectively. Their compositions shift towards the PbS end-member from stoichiometric teallite. The cell dimensions of teallite, which was synthesized hydrothermally, follow the linear relationship between cell dimensions and composition established at 700°C.

The pseudomorphic replacement of glass shards by zeolite minerals is a common feature of volcanoclastic sediments. In the majority of cases the initial stage of this reaction is the alteration of the glass surface to a clay mineral or celadonite after which the bulk of the glass is altered to zeolite. This replacement feature is seen particularly well in glass of rhyolitic composition; the zeolite mineral usually being clinoptilolite. Volcanoclastic rocks of Oligocene age exposed in the Rhodope Massif, Bulgaria offered an opportunity to study this reaction experimentally as rocks containing unaltered glass shards are known to be close stratigraphic equivalents of zeolitized tuffs and in this respect are considered to be precursor rocks. Low-temperature hydrothermal reactions conducted on the unaltered glass, which had been clay coated in the laboratory, demonstrates the importance of the clay-glass interface. An hypothesis is put forward to explain this type of zeolitization process and a distinction is drawn between these rocks and other sediments in which zeolite minerals form from volcanic glass without the presence of a clay interface.

Electron and ion microprobe data on two samples of welshite from the type locality of Långban, Sweden, gave analytical totals of 99.38–99.57 wt.% and BeO contents of 4.82–5.11 wt.%, corresponding to 1.692–1.773 Be/20 O. Mössbauer and optical spectra of one of these samples gave [iv]Fe3+/ΣFe = 0.91, [vi]Fe2+/ΣFe = 0.09, and no evidence of Mn3+. The resulting formula for this sample is Ca2Mg3.8Mn0.62+Fe0.12+Sb1.55+O2[Si2.8Be1.7Fe0.653+Al0.7As0.17O18, and that for the second sample, Ca2Mg3.8Mn0.12+Fe0.12+Fe0.83+Sb1.25+O2[Si2.8Be1.8Fe0.653+Al0.5As0.25O18], is related by the substitution involving tetrahedral and octahedral sites: 0.59[vi,iv](Fe,Al)3+ ≈ 0.42[vi](Mg,Mn,Fe)2+ + 0.21([vi]Sb,[iv]As)5+, i.e. 3[vi,iv]M3+ = 2[vi]M2+ + [vi,iv]M5+. Welshite is distinctive among aenigmatite-group minerals in the high proportion of Fe3+ in tetrahedral coordination and is unique in its Be content, substantially exceeding 1Be per formula unit. Given the cation distributions in other minerals related to aenigmatite, we think it is reasonable to assume that at least one tetrahedral site is >50% occupied by Be and that one octahedral site is >50% occupied by Sb, so that welshite should be retained as a distinct species with its own name in the aenigmatite group.

Alkali feldspar (Or64.5Ab34.8An0.6) in a granite pegmatite from Hanazono, Kosei Town, Shiga Prefecture, southwest Japan, consists of two types of regions observable with the naked eye: one is colourless and transparent, and the other is pale pink and opaque. The colourless regions are clear, cryptoperthitic and almost free of micropores, and the pale pink regions are turbid, vein microperthitic and with many micropores visible under a microscope. The latter regions may have been changed from the former by the catalytic action of water. Integrated absorbance values were calculated for both types of regions from Fourier-transform infrared microspectra in the range 2500–4000 cm−1. These values were used as semi-quantitative parameters of water contents. The turbid regions are much more enriched in water compared to the clear regions. The chemical states of water in both regions were also evaluated. Water in the turbid regions is estimated to be present mainly as H2O inclusions in micropores and to be present to a lesser extent as structural OH groups. A small amount of water in the clear regions may be present as structural OH groups. The water distribution in the feldspar records the cooling history of hydrothermal reactions causing the coarsening of cryptoperthites to microperthites.

The mineral name dienerite should be allowed to slip into obsolescence. Dienerite AsNi3–x is probably nickelskutterudite As3–xNi caused by a typographical error in the chemical analysis.