Interactions of Silica with Iron Oxides: Effects on Oxide Transformations and Sorption Properties

CA9500441 AECL EACL

AECL-11257, COG-l-95-023

Interactions de la silice avec les oxydes de fer : Effets sur les transformations des oxydes et les propriétés de sorption

Peter Taylor

August 1995 août s f'V VUL. <- • rti i: AECL-11257 COG-I-95-023

AECL

INTERACTIONS OF SILICA WITH IRON OXIDES: EFFECTS ON OXIDE TRANSFORMATIONS AND SORPTION PROPERTIES

Peter Taylor

Whiteshell Laboratories Pinawa, Manitoba ROE 1LO 1995 AECL-11257 COG-I-95-023

INTERACTIONS OF SILICA WITH IRON OXIDES: EFFECTS ON OXIDE TRANSFORMATIONS AND SORPTION PROPERTIES

Peter Taylor

ABSTRACT

This report is a review of the literature on the adsorption of silica species on iron oxides and oxyhydroxides, and its effects on the adsorption of other species and on oxide interconversion reactions. The information is discussed briefly in the contexts of nuclear waste disposal and boiler-water chemistry.

Whiteshell Laboratories Pinawa, Manitoba ROE 1LO 1995 AECL-11257 COG-I-95-023

INTERACTIONS DE LA SILICE AVEC LES OXYDES DE FER EFFETS SUR LES TRANSFORMATIONS DES OXYDES ET LES PROPRIÉTÉS DE SORPTION

par

Peter Taylor

RÉSUMÉ

Le présent rapport constitue le compte rendu des publications sur l'adsorption des espèces de silices sur les oxydes de fer et les oxyhydroxydes, et de ses effets sur l'adsorption d'autres espèces et les réactions d'interconversion d'oxydes. Ces données sont abordées brièvement dans le contexte du stockage permanent des déchets nucléaires et de la chimie de l'eau de générateur de vapeur.

Laboratoires de Whiteshell Pinawa (Manitoba) ROE 1LO 1995 AECL-11257 COG-I-95-023

CONTENTS Page

1. INTRODUCTION

2. REVIEW OF SILICA/IRON OXIDE INTERACTIONS 2

2.1 ADSORPTION OF SILICA(TE) ON IRON OXIDE SURFACES 2 2.2 INFLUENCE OF SILICA(TE) ON IRON OXIDE SORPTION PROPERTIES 3 2.3 INFLUENCE OF SILICA(TE) ON IRON OXIDE INTERCONVERSIONS 4 2.3.1 Stability of Natural and Synthetic Ferrihydrites 4 2.3.2 Other Natural Iron Oxide/Silicate Interactions 6 2.3.3 Goethite Synthesis 6 2.3.4 Lepidocrocite Synthesis and Alteration 6 2.3.5 Alteration of Maghemite to Hematite 7 2.4 TECHNICAL APPLICATIONS 7 2.5 OTHER OXIDE SYSTEMS 8

3. CONCLUSIONS

ACKNOWLEDGEMENTS

REFERENCES AECL-11257 COG-I-95-023

1. INTRODUCTION

The Fe-Q-H-,Q system is ccmolicated by the existence of sever?.! met?.st?.blc solid phases in addition to magnetite and hematite. Important examples include maghemite (y-Fe2O3), goethite (a-FeOOH), lepidocrocite (y-FeOOH), and ferrihydrite (approx. 5Fe2O3-H?O); several less common phases are also known. The chemistry of these phases has been reviewed by Schwertmann and Cornell (1991), Langmuir (1969), Taylor and Owen (1993), Blesa and Matijevic (1989), Taylor (1987), Waychunas (1991) and Greenland and Mott (1978). These reviews emphasize the preparation, thermodynamic stability, redox chemistry, interconversion chemistry, crystallography, mineralogy and soil chemistry of these materials, respectively. Interconversion between these phases is slow at low temperatures (<100°C), and many of them can occur in soils and weathered rocks, and also as low-temperature corrosion products on structural steels. Numerous chemical species, notably orthophosphate, orthosilicate and carboxylates, are known to influence the course of interconversion of these phases (Blesa and Matijevic 1989, Schwertmann and Cornell 1991). The situation is much simpler at elevated temperatures (>100°C), where the hydrous oxides become less important, and maghemite is probably the most persistent metastable phase (Taylor and Owen 1993 and references therein).

The high capacity of fine-grained iron oxide minerals to adsorb both cationic and anionic species from aqueous solution makes them important in immobilizing a variety of soil nutrients and environmental contaminants (Parfitt 1978); many examples are cited in Section 2. Iron oxide minerals may help to retard the migration of radionuclides from a nuclear fuel waste disposal vault to the biosphere (Davison et al. 1994, Vandergraaf et al. 1993, Vandergraaf and Ticknor 1994). In addition, iron minerals such as magnetite and biotite are expected to be important in the restoration of reducing conditions in a waste vault; the nature of the Fe(III) oxides associated with Fe(II) minerals, such as biotite and magnetite, is important in estimating the redox potential that will be attained (Johnson et al. 1994, Taylor and Owen 1993). Since dissolved silica (as Si(OH)4 and OSi(OH)3") occurs at concentrations around 2 x 10"4 mol/dm3 in deep granitic groundwaters (Gascoyne 1988), it is important to understand how it may affect both the sorption properties and the redox chemistry of iron oxides in a waste disposal system. For example, it might account for the relatively low affinity for iodate adsorption by a silica-containing natural goethite, as compared with a pure, synthetic hematite (Ticknor and Cho 1990).

Silica is also present at low concentrations in the feedwater systems of steam generators (SGs) at nuclear power generating stations. As well as contributing directly to SG fouling in some cases, silica may influence the alteration, transport and deposition of the iron oxide corrosion products (e.g., lepidocrocite, magnetite and hematite) that are major contributors to SG fouling (Moskvin et al. 1979, 1981; Sawicki and Brett 1993; Sedov et al. 1978, 1980).

For these reasons, I have reviewed the literature on the interactions of iron oxides with silica. This review is neither fully comprehensive nor very critical in outlook. Rather, it is intended to introduce the reader to the somewhat diffuse body of literature on this topic. The review focuses on experimental observations of silicate adsorption and its effects on the sorption properties and phase-alteration chemistry of iron oxides.

Non-specific adsorption (outer-sphere complexation) of ions within the charged double-layer on solid surfaces is controlled by the surface charge of the solid. The charge on most oxides in simple aqueous systems is controlled by the acid-base chemistry of surface oxide and hydroxide species. Protonation of these surface species in acidic solutions yields a net positive surface charge, whereas deprotonation, hydroxylation, or both, results in a net -2- AECL-11257 COG-I-95-023 negative surface charge in alkaline solutions. The pH point of zero charge (pHpzC), where the populations of positive and negative surface sites are equal, is an important parameter in determining sorption and other surface properties of oxides. At pH values below pHpzc, non- specific anion adsorption tends to be favoured, whereas at pH>pHpzc, cation adsorption occurs. For most iron (III) oxides, the value of pHpzC is about 8.

Specific adsorption (inner-sphere complexation) involves direct bonding between the adsorbing species and the oxide surface, an example being the Fe-O-Si(OH)3 bonding that has been proposed to exist in adsorbed silicate species on iron oxide surfaces. This type of adsorption is not directly controlled by surface charge, e.g., anion adsorption may persist at pH values well above pHp^. Nonetheless, the surface charge may influence the kinetics of specific adsorption by controlling the concentration of physisorbed intermediates, and the adsorption may in turn influence the surface charge properties. Thus, specific adsorption of anions tends to increase negative charge under conditions near pHPZC, whereas adsorbed cations tend to drive pU^c towards the value for the hydroxide of the cation in question (Parks 1965). Also, specific adsorption displays a pH dependence resembling that for non- specific adsorption, because of competition for surface sites between FT and other cations, and between OH" and other anions. Several excellent books are available that present detailed discussion and theoretical treatment of inorganic adsorption in aqueous systems (Anderson and Rubin 1981, Davis and Hayes 1986, Hochella and White 1990, Stumm 1992).

Throughout this report, the terms dissolved silica, silicic acid and silicate are used more or less interchangeably (which is also the case in the literature reviewed). Below pH 9, the predominant form of dissolved silica in aqueous solution is silicic acid, Si(OH)4 (or H4SiO4). Above pH 9, deprotonated forms such as OSi(OH)3" (or H3SiO4") become important, in accordance with expression [1].

+ Si(OH)4 * H + OSi(OH)3- (pKa = 9.7 at 25°C) (1) Oligomeric or colloidal silicate species may be present in supersaturated solutions, which can be formed experimentally by neutralization of alkaline silicates. Oligomeric species are also important in concentrated, alkaline silicate solutions. In most of the situations discussed below, "silica" or "silicate" are assumed to bind on oxide surfaces through Fe-O-Si bonds, as discussed in Section 2.1.

2. REVIEW OF SILICA/IRON OXIDE INTERACTIONS

2.1 ADSORPTION OF SILICA(TE) ON IRON OXIDE SURFACES

Much of the literature about silicate adsorption on iron oxide surfaces is concerned with soil minerals (especially ferrihydrite), and the influence of adsorbed silica on the behaviour of other anions that are important nutrients (e.g., phosphate and organic anions) or pollutants (e.g., selenate, selenite, and chromate). Parfitt (1978) reviewed much of the earlier literature on the adsorption of anions (including silicate) by soil minerals. As he noted, silicic acid adsorption by hydrous iron and aluminum oxides has been recognized since the 1960s (e.g., Beckwith and Reeve 1963, 1964; Kingston et al. 1967; Kingston and Raupach 1967; Jones and Handreck 1963; McKeague and Cline 1963a,b). This adsorption reaches a peak around pH 9, which corresponds to the pK for the first dissociation step of silicic acid [1]. -3- AECL-11257 COG-I-95-023

More recently, Balistrieri and Chao (1990) observed a plateau over the pH range 7.5 to 10, rather than a definite peak near pH 9, for silicate adsorption on ferrihydrite. However, this result was obtained at a much lower surface coverage (ça. 10"7 mol/m ), compared with about 2.5 x 10"6 mol/m2 in the work by Kingston et al. (1972). Sigg and Stumm (1980) measured silicate adsorption on goethite over a wide pH range; they observed substantial adsorption all the way from pH 3 to 10 and a plateau between about 7 and 9. These observations are consistent with other reports on specific adsorption of anions (Kingston 1981). Commonly, adsorption increases as the pH is increased up to the pKa value for the conjugate acid of the anion in question, then decreases at still higher pH values, because of increasing competition with OH" for surface sites.

Yokoyama et al. (1980) examined the adsorption of silicic acid on freshly prepared "iron(III) hydroxide", and also observed a broad peak in the amount of Si adsorbed around pH 9. They paid particular attention to polymerization of the adsorbed silicate species, either with one another or with aqueous silicate species. They found that the degree of polymerization, as well as the quantity of silicic acid adsorbed, reached a maximum at pH 9.

Reviewing the work of Kingston et al. (1972), Parfitt (1978) noted that borate, molybdate and silicate are each adsorbed by ligand exchange on goethite surfaces, and cited evidence from infrared spectroscopy for the formation of unidentate surface complexes of the =Fe-OSi(OH)3 type; surface complexes of this sort with a variety of oxyanions were proposed by Parfitt and Russell (1977) and Harrison and Berkheiser (1982). Ligand competition for specific surface sites has been inferred from the behaviour of phosphate and humus, which both tend to displace silicate, or suppress its adsorption (Obihara and Russell 1972, Kafkafi et al. 1970, Wada and Inoue 1974). Balistrieri and Chao (1987, 1990) reported the following sequences for competing sorption between selenite and other anions: (a) on goethite, phosphate > silicate > citrate > molybdate > bicarbonate/carbonate > oxalate > fluoride > sulfate; (b) on ferrihydrite, phosphate > silicate > molybdate » fluoride > sulfate.

Silicate adsorption on soils has been fitted to both Langmuir and Freundlich types of isotherm (Obihara and Russell 1972, Wada and Inoue 1974). More recently, the constant capacitance model, developed by Stumm, Schindler and their collaborators, has been used successfully to represent specific adsorption of several anions, including silicate, on goethite and other hydrous iron oxide surfaces (Goldberg 1985, Sigg and Stumm 1980, and references therein). The appeal of the constant capacitance model is that it explicitly defines surface species and equilibria. An obvious criticism - or at least a source of unease - is that this model typically uses a large number of fitting parameters (complexation constants) to describe some rather broad and visually featureless adsorption envelopes. Nevertheless, this model has recently been used with success to describe the time- and pH-dependence of silicate adsorption on ferrihydrite and goethite, in one of the most detailed experimental studies to date on this subject (Hansen et al. 1994).

2.2 INFLUENCE OF SILICA(TE) ON IRON OXIDE SORPTION PROPERTIES

Anderson and Benjamin (1985) reported that the pHpzC of ferrihydrite fell from about 8 to 4 with increasing Si content. This increased the solid's affinity for cadmium, i.e., the pH of the adsorption edge decreased, but little effect was noted on the adsorption of Zn, Cu or Co on ferrihydrite. The affinity of the solid for selenite decreased with increasing Si content, in such a way that the relationship between the quantity of selenite adsorbed and the value of (pH - pHPZC) remained essentially unchanged. The authors noted: "The sorption experiments emphasize the very different effects of the average surface charge on adsorption of oxyanions -4- AECL-11257 COG-I-95-023 and hydrolysable cations. These electrostatic effects seem to be very important in the former cases, while they are almost totally absent in the latter. Evidence supporting this generalization...[was obtained]. ,from competitive adsorption studies (Benjamin 1983, Benjamin and Bloom 1981, Davis 1984)"."

In a related study, Anderson and Benjamin (1990) investigated the sorption behaviour of a mixed suspension of silica and ferrihydrite. They observed that the adsorption of metallic ions (Cd2+, Ag+, Zn2+) was almost unaffected by the presence of silica, whereas for both selenite and phosphate, the pH region of the adsorption edge shifted about one unit in the acid direction when silica was present in addition to ferrihydrite. They concluded that "The experimental results... are consistent with a scenario involving partial SiO2 dissolution and subsequent sorption of silicate on the Fe(OH)3... Soluble silica apparently competes for surface sites with phosphate and selenite, but has little influence on the adsorption of silver, cadmium or zinc".

Schwertmann and Fechter (1982) examined a series of both natural and synthetic ferrihydrite specimens with different Si content and found that their pHPZC values were also considerably lower (5.3 to 7.5) than for Si-free synthetic material (ca. 8).

Zachara et al. (1987) investigated the adsorption of chromate on amorphous iron oxyhydroxide, including the effects of silica. They concluded that "like inorganic carbon, 2 H4SiO4(aq) depresses CrO4 " adsorption. A larger effect occurs if the Fe2O3-H2O(am) is aged with H4SiO4 before chromate addition. The aging induces a systematic decrease in chromate adsorption up to 168 h, after which time additional aging up to 672 h has no further influence. The decrease in chromate adsorption induced by H4SiO4 reflects the adsorption of silica and its effects on the iron surface".

2.3 INFLUENCE OF SILICA(TE) ON IRON OXIDE INTERCONVERSIONS

Many researchers have commented on the influence of silica (or silicate) on the conversion of the less stable (more soluble) iron(III) oxides and hydrous oxides, such as ferrihydrite and Jepidocrocite, to the more stable forms, goethite and hematite. Many of these references are more or less anecdotal, and few systematic investigations have been reported.

2.3.1 Stability of Natural and Synthetic Ferrihydrites

Chukhrov et al. (1973) reported the inhibition of hematite crystallization from natural ferrihydrite by adsorbed silica, and similar findings were reported by Carlson and Schwertmann (1981). Karim (1984) prepared ferrihydrites containing up to 1.85 wt.% Si (3.9% SiO2) by oxidation of aqueous FeCl2 solution, following a procedure described by Schwertmann and Thalmann (1976). The presence of silica in the initial solution inhibited the formation of lepidocrocite, in favour of ferrihydrite, and the silica incorporated in the ferrihydrite inhibited its subsequent thermal transformation to hematite under dry conditions. Whereas an Si-free ferrihydrite, prepared by hydrolysis of ferric nitrate solution, transformed to hematite at 350 to 450°C, the corresponding transformation of ferrihydrite containing 1.48% SiO2 occurred at 650 to 750°C. This result is consistent with observations by Carlson and Schwertmann (1981) on the thermal alteration of a natural siliceous ferrihydrite. These authors concluded that Fe-O-Si bonds were implicated in the stabilization of ferrihydrite, but they did not comment further on the transformation mechanism. It would appear that the presence of Si blocks the reconstruction and ordering of ferrihydrite to form hematite, and that hematite growth then occurs only at temperatures high enough for Si to diffuse out of the Fe-O structural framework. -5- AECL-11257 COG-I-95-023

Vempati et al. (1990) have reported similar findings (i.e., inhibition of hematite formation) on both dry and aqueous thermal alteration of Si-containing ferrihydrite. After 36 h at pH 12 and 91°C, ferrihydrite with Si/Fe molar ratios below 0.05 was converted to hematite (with smaller quantities of goethite), whereas no transformation was observed with ferrihydrites having Si/Fe - 0.10. Dry, Si-free ferrihydrite transformed to hematite in 2 h at 300°C, but the transformation temperature increased dramatically with increasing Si content, up to 800°C at an Si/Fe molar ratio of 0.25. Additional detail of the suppression of ferrihydrite transformation to goethite or hematite by silica is furnished by Vempati and Loeppert (1989).

Cornell et al. (1987) showed that the transformation of ferrihydrite to goethite and hematite in aqueous alkaline media is strongly inhibited by silicate species. At low silicate concentrations (<10'4 mol/dm3), ferrihydrite transformed to hematite by internal rearrangement, rather than to goethite by dissolution and reprecipitation. At higher concentrations, transformation of the ferrihydrile was suppressed completely. For example, 10"3 mol/dm3 silica suppressed the transformation of ferrihydrite for 3 to 4 months at pH 13 and 70°C. Co-precipitated silica was also found to be more effective than added aqueous silica in the inhibition of ferrihydrite transformation. Similarly, Anderson and Benjamin (1985) found that Si added before or shortly (<2 h) after the precipitation of ferrihydrite inhibited its transformation to goethite, whereas later addition (>24 h) resulted in the formation of hematite as well as goethite.

Cornell et al. (1989) reviewed the crystallization of ferrihydrites, including the effects of various adsorbates, including silica, on the transformation. They noted that "additives that retard [the crystallization of ferrihydrite to goethite] and also promote formation of hematite indirectly... [include] silicate species". They emphasized that silicate species are regarded as ligands rather than substituents because only very low levels are incorporated into the goethite lattice. Citing Rubio and Matijevic (1979), they state that species such as phosphate and silicate, which form multinuclear surface complexes, stabilize an oxide, whereas polydentate, mononuclear surface species weaken the metal-oxygen bonds in the lattice and so promote dissolution. Cornell et al. (1987) found that ellipsoidal crystals of hematite, with a granular appearance resembling ferrihydrite, can be produced in the presence of oxalate or silicate species. They suggested that this shape is determined by anion adsorption on the parent ferrihydrite, rather than the growing hematite.

Parfitt et al. (1992) have postulated a submicroscopic domain structure for natural siliceous ferrihydrite, in which silicate is bound to, and forms bridges between, the surfaces of ordered ferrihydrite domains with dimensions of about 1 nm. In support of this structural model, Parfitt et al. (1992) state that phosphate readily displaces only a small fraction of the silicate from natural ferrihydrite, although a slow reaction continues over time (Parfitt 1989). Parfitt et al. (1992) pointed out that similar surface coatings have been suggested to exist in some natural goethite specimens (Smith and Eggleton 1983). The latter authors suggested that "It is possible that the silica found in these three botryoidal [fibrous] goethite samples might be adsorbed to the surface of individual needle-like crystals. The surface:volume ratio shows good general agreement between the amount of silica observed (-2%) and the calculated amount predicted as a monolayer between the needles (-3%). Cornwall goethite does not have fibrous morphology and is free of silica". Herbillon and Tran Vinh An (1969) also proposed a diphasic structure for synthetic hydrous SiO2/Fe2O3 precipitates. As well as influencing the stability of ferrihydrite, additions of silica can also affect the morphology of the crystallization product. Bye and Howard (1971) noted that co-precipitation with silica caused goethite needles to form spherulitic aggregates. Cornell and Giovanoli (1987) found that the morphology of goethite, grown from ferrihydrite at pH 12.5 and 70°C -6- AECL-11257 COG-I-95-023 over 50 h, changed from acicular cr>'stals -450 x 40 nm under Si-free conditions to shorter and thicker crystals with faceted ends in the presence of 1 mol.% Si [presumably Si/(Fe+Si)], and to short acicular, pseudohexagonal and bipyramidal crystals with 9.6 mol.% Si. Citing Carlson and Schwertmann (1981), they noted that adsorbed Si stabilizes ferrihydrite and strongly retards its dissolution, whereas dissolved Si hinders nucleation of gcethite.

Some additional observations on the effects of silica on transformation of ferrihydrite are noted in Section 2.3.3.

2.3.2 Other Natural Iron Oxide/Silicate Interactions

Although, as this report shows, aqueous silica can interact strongly with iron(III) oxides, there are no known iron(III) silicate compounds, either hydrated or anhydrous, in the Fe2O3-SiO2- H2O system. However, iron(III) is commonly present in more complex silicate minerals and synthetic materials, often in full or partial structural substitution for magnesium or aluminum, and less commonly for gallium or barium (Ghose 1978).

It is commonly suggested that clay minerals are coated with iron oxides, which then control the surface chemistry of the clay; however, Greenland and Mott (1978) commented as follows:

"The rather frequently made suggestion that the kaolinite particles, which form the main constituent of the clay fraction of many tropical soils, are coated with iron oxides to give the soils their red colour is probably not always correct. Electron micrographs show the iron oxide to be present in many such soils as small, discrete particles (probably ferrihydrite)... Although coatings of iron hydroxide can be precipitated onto kaolinite particles under suitable (acid) conditions... similar features are not normally seen in soil clays, although in some hydromorphic soils where iron hydroxides may be seasonally dissolved under reducing conditions and reprecipitated under oxidizing conditions, precipitates of iron hydroxide may envelop clay particles."

2.3.3 Gocthite Synthesis

Goethite is most conveniently prepared by aging a fresh precipitate of ferrihydrite at 70°C in alkaline solution (~3 mol-dm"3 KOH). Schwertmann and Cornell (1991) noted that "Preparation fof goethite] must be carried out in polyethylene vessels because with glass vessels and strongly alkaline media some Si dissolves. Silicate retards the conversion of ferrihydrite to goethite and may also be retained by the goethite". They refer to an earlier investigation by Atkinson et al. (1968), and also cite Cornell et al. (1987), who found that "The transformation [of ferrihydrite to hematite and/or goethite] can be blocked by small amounts of adsorbates such as silicate, phosphate and a range of organics and also by co- precipitated Al".

2.3.4 Lepidocrocite Synthesis and Alteration

Karim (1986) investigated the effect of dissolved silica on the oxidation of ferrous chloride solutions; he found that it suppresses the formation of "green rust" (a precursor of lepidocrocite), and instead favours ferrihydrite as the ultimate oxidation product. Thus, silica appears to favour (or at least permit) the formation of ferrihydrite, as well as suppressing its alteration. Similarly, earlier work by Schwertmann and Thalmann (1976) showed that dissolved silica prevented the formation of lepidocrocite from aqueous ferrous chloride, and favoured the formation of ferrihydrite instead. -7- AECL-11257 COG-I-95-023

Schwertmann and Taylor (1972) investigated the alteration of lepidocrocite to goethite, and concluded that silicate ions retard the nucleation of goethite; the effect was less important at higher temperatures, and disappeared if silicate was added after nucleation had taken place.

2.3.5 Alteration of Maghemite to Hematite

Swaddle and Oltmann (1980) found that the presence of silica, originating from Pyrex glassware used in the synthesis of maghemite, retarded the subsequent hydrothermal alteration of the maghemite to hematite. This product was only 7% converted to hematite after 3 h in water at 187°C, whereas 97% conversion occurred with silica-free maghemite under similar conditions. The silica contamination also favoured alteration of the maghemite to goethite, rather than hematite, in some experiments performed in NaOH or KOH solution (1 mol-dm"3, 150 to 175°C).

Swaddle and Oltmann (1980) considered four modes by which silica might inhibit the alteration of maghemite to hematite:

(a) aqueous complexation of iron(III) by silicate;

(b) chemisorption of silicate on the maghemite to form a dissolution-resistant coating;

(d) elimination or coating of hematite nuclei by silicate.

Although the evidence was not clearcut, Swaddle and Oltmann (1980) concluded that mode (b) was the most important factor inhibiting hematite formation.

In reference to the work of Swaddle and Oltmann (1980), Blesa and Matijevic (1989) comment: "The influence of silica most probably involves adsorption of silicate species through surface complexation reactions, exerting a strong effect on the nucleation stage of both oc-Fe2O3 and a-FeOOH; the former nuclei seem to be comparatively more readily destabilized by silicate ions".

Taylor and Owen (1993) reported similar observations on the influence of silica (and phosphate) on the conversion sequence magnetite -> maghemite -4 hematite. They found that silica, introduced as either aqueous colloidal or powdered vitreous SiO2, inhibited the formation of hematite by a dissolution/precipitation mechanism, and enhanced the persistence of maghemite as a metastable oxidation product, in aqueous media at temperatures between 100 and 200°C. Above 200°C, conversion of maghemite to hematite proceeded by a solid- state mechanism.

2.4 TECHNICAL APPLICATIONS

The affinity of silica for iron oxide surfaces has been utilized in the preparation of magnetic silica dispersions, i.e., silica particles with a magnetite core. The initial coating of silica has been applied by adsorption of aqueous silicate species (Philipse et al. 1994).

Fujita et al. (1990) have shown that dissolved silica can markedly reduce the corrosion rate of powdered iron in water at 100°C, apparently by the formation of thin protective layers on the iron/oxide surface. The quantity of silica, relative to the surface area of iron, required for this protective effect was estimated to be 3.9 x 10"'° mol Si/cm2 Fe. -8- AECL-11257 COG-I-95-023 2.5 OTHER OXIDE SYSTEMS

This report does not explore the literature on silicate adsorption on materials other than iron oxides. There are indications, however, that aluminum oxides and a variety of crushed or precipitated metal hydroxides (Mg(OH)2, Ni(OH)2, Co(OH)2) have a marked affinity for silica (Kingston et al. 1972, McKeague and Cline 1963b).

3. CONCLUSIONS

There is an extensive, but somewhat diffuse literature on the interaction of silicate species with iron oxide surfaces. Silicate is strongly adsorbed on the surfaces of various iron oxides and oxyhydroxides, with a maximum in surface coverage near pH 9. The adsorption of silica is slow, in comparison with many ionic adsorption processes, and may take several weeks to approach equilibrium at ambient temperature. This may complicate investigations of the effects of silicate on the adsorption of other species, or on oxide transformation processes. In general, silicate adsorption appears to have a pronounced effect on the adsorption of other anions (displacing the limiting pH for anion adsorption to lower values), but has little effect on cation adsorption. Low levels of silicate can also profoundly influence the course of iron oxide interconversion reactions, either by stabilizing precursor phases against dissolution or inhibiting the nucleation of product phases. Similar behaviour may occur in other oxide systems, notably those of other first-row transition elements and aluminum, but they appear to have received much less attention than iron oxide systems.

Silica is ubiquitous in natural groundwater systems, and may therefore influence the role of oxide minerals in immobilizing environmental contaminants, including radionuclides in a nuclear waste vault. It may also affect the formation, transport and accumulation of iron- oxide corrosion products in steam-generating systems, such as the secondary cooling circuits of CANDU nuclear generating systems. There is published evidence that low concentrations of silica can inhibit iron corrosion, but it is not clear whether the overall effect of silica on boiler fouling would be beneficial or detrimental.

It is possible that adsorbed silica on iron oxide particle surfaces may play a role in sludge consolidation and hardening, either by modifying the alteration kinetics of precursor phases (lepidocrocite and hematite) to magnetite deposits, or by a mechanism similar to the cementing of goethite particles in some natural formations (see Section 2.3.1). It should be noted, however, that in CANDU reactor systems with high Si content in steam-generator deposits, much of the Si is present as separate phases, rather than being intimately associated with iron oxides (Gonzalez 1986a, 1986b; Turner et al. 1993). It has been suggested that such silicate phases may be involved in sludge hardening.

ACKNOWLEDGEMENTS

This work was supported in part by the Canadian Nuclear Fuel Waste Management Program, under the auspices of the CANDU Owners Group. I thank P.V. Balakrishnan, D.J. Jobe, IJ. Muir and K.V. Ticknor for helpful comments on a draft of this report. -9- AECL-11257 COG-I-95-023

REFERENCES

Anderson, M.A. and A.J. Rubin (editors). 1981. Adsorption of inorganics at solid-liquid interfaces. Ann Arbor Science Publishers, Ann Arbor, MI.

Anderson, P.R. and M.M. Benjamin. 1985. Effects of silicon on the crystallization and adsorption properties of ferric oxides. Environmental Science and Technology 19, 1048-1053.

Anderson, P.R. and M.M. Benjamin. 1990. Constant-capacitance surface complexation model: Adsorption in silica-iron binary oxide suspensions. In Chemical Modeling of Aqueous Systems II. (D.C. Melchior and R.L. Bassett, editors). American Chemical Society Symposium Series No. 416, Chapter 21.

Atkinson, R.J., A.M. Posner and J.P. Quirk. 1968. Crystal nucleation in Fe(III) solutions and hydroxide gels. Journal of Inorganic and Nuclear Chemistry 30. 2371-2381.

Balistrieri, L.S. and T.T. Chao. 1987. Selenium adsorption by goethite. Soil Science Society of America Journal 51. 1145-51.

Balistrieri, L.S. and T.T. Chao. 1990. Adsorption of selenium by amorphous iron oyhydroxide and manganese dioxide. Geochimica et Cosmochimica Acta 54. 739-751.

Beckwith, R.S. and R. Reeve. 1963. Soluble silica in soils. I. The sorption of silicic acid by soils and minerals. Australian Journal of Soil Research 1, 157-168.

Beckwith, R.S. and R. Reeve. 1964. Soluble silica in soils. II. The release of monosilicic acid from soils. Australian Journal of Soil Research 2, 33-45.

Benjamin, M.M. 1983. Adsorption and surface precipitation of metals on amorphous iron oxyhydroxide. Environmental Science and Technology 17. 686-692.

Benjamin, M.M. and N.S. Bloom. 1981. Effects of strong binding of anionic adsorbates on adsorption of trace metals on amorphous iron oxyhydroxide. In "Adsorption from Aqueous Solutions" (P.H. Tewari, editor), Plenum Press, New York, NY, pp. 41-60.

Blesa, M.A. and E. Matijevic. 1989. Phase transformations of iron oxides, oxohydroxides, and hydrous oxides in aqueous media. Advances in Colloid and Interface Science 29, 173- 221.

Bye, G.C. and C.R. Howard. 1971. Examination by nitrogen adsorption of the thermal decomposition of pure and silica-doped goethite. Journal of Applied Chemistry and Biotechnology 21, 324-329.

Carlson, L. and U. Schwertmann. 1981. Natural ferrihydrites in surface deposits from Finland and their association with silica. Geochimica et Cosmochimica Acta 45. 421-429.

Chukhrov, F.V., B.B. Zvyagin, L.P. Ermilova and A.I. Gorshkov. 1973. New data on iron oxides in the weathering zone. Proceedings of International Clay Conference, Madrid, 1972. Vol. 1, pp. 397-402. Division de Ciencias, CSIC, Madrid. -10- AECL-11257 COG-I-95-023

Cornell, R.M. and R. Giovanoli. 1987. The influence of silicate species on the morphology of goethite (a-FcOOH) grown from ferrihydrite (5Fe2O3-9H2O). Journal of the Chemical Society, Chemical Communications, pp. 413-414.

Cornell, R.M., R. Giovanoli and P.W. Schindler. 1987. Effect of silicate species on the transformation of ferrihydrite into goethite and hematite in alkaline media. Clays and Clay Minerals 33, 21-28.

Cornell, R.M., R. Giovanoli and W. Schneider. 1989. Review of the hydrolysis of iron(III) and the crystallization of amorphous iron(III) hydroxide hydrate. Journal of Chemical Technology and Biotechnology 46. 115-134.

Davis, J.A. 1984. Complexation of trace metals by adsorbed natural organic matter. Geochimica et Cosmochimica Acta48, 679-691.

Davis, J.A. and K.F. Hayes (editors). 1986. Geochemical processes at mineral surfaces. Symposium Series Volume 323, American Chemical Society, Washington, DC.

Davison, C.C. et al. 1994. The disposal of Canada's nuclear fuel waste: the geosphere model for postclosure assessment. Atomic Energy of Canada Limited Report, AECL-10719, COG- 93-9, Appendix B.

Fujita, N., C. Matsuura and K. Ishigure. 1990. The effect of silica on hydrogen evolution and corrosion of iron in high-temperature water. Corrosion Science 46. 804-812.

Gascoyne, M. 1988. Reference groundwater composition for a depth of 500 m in the Whiteshell Research Area - Comparison with synthetic groundwater WN-1. Atomic Energy of Canada Limited Technical Record, TR-463. Available from SDDO, AECL Research Company, Chalk River, Ontario KOJ 1JO.

Gnose, S. 1978. Crystal chemistry [of iron]. In Handbook of Geochemistry (K.H. Wedepohl, executive editor), Springer-Verlag, Berlin. Vol. II/3, Chapter 26-A.

Goldberg, S. 1985. Chemical modeling of anion competition on goethite using the constant capacitance model. Soil Science Society of America Journal 49. 851-856.

Gonzales, F. 1986a. Analysis of Pickering NGS A Unit 2 boiler 10 sludge pile deposits. Ontario Hydro Research Report 86-65-K.

Gonzales, F. 1986b. Analysis of Bruce NGS Unit 3 Steam Generator 3 deposits. Ontario Hydro Research Report 86-82-K.

Greenland, D.J. and C.J.B. Mott. 1978. Surfaces of soil particles. In "The Chemistry of Soil Constituents" (D.J. Greenland and M.H.B. Hayes, editors), Wiley, Chichester, U.K., pp. 321- 353.

Harrison, J.B. and V.E. Berkheiser. 1982. Anion interactions with freshly prepared hydrous iron oxides. Clays and Clay Minerals 30. 97-102.

Hansen, H.C.B., B. Raben-Lange, K. Raulund-Rasmussen and O.K. Borggaard. 1994. Monosilicate adsorption by ferrihydrite and goethite at pH 3-6. Soil Science 158. 40-46. -11- AECL-11257 COG-I-95-023

Herbillon, AJ. and J. Tran Vinh An. 1969. Heterogeneity in silicon-iron mixed hydroxides. Journal of Soil Science 20. 233-235.

Kingston, F.J. 1981. A review of anion adsorption. In "Adsorption of Inorganics at Solid- Liquid Interfaces" (M.A. Anderson and A.J. Rubin, editors), Ann Arbor Science Publishers, Ann Arbor, MI. Chapter 2.

Kingston, F.J. and M. Raupach. 1967. The reaction between monosilicic acid and aluminium hydroxide. Australian Journal of Soil Research 5, 295-309.

Kingston, F.J., R.J. Atkinson, A.M. Posner and J.P. Quirk. 1967. Specific adsorption of anions. Nature 215. 1459-61.

Kingston, F.J., A.M. Posner and J.P. Quirk. 1972. Anion adsorption by gibbsite and goethite. I. The role of the proton in determining adsorption envelopes. Journal of Soil Science 23, 177-192.

Hochella, M.F., Jr. and A.F. White (editors). 1990. Mineral-water interface geochemistry, reviews in mineralogy, Volume 23. Mineralogical Society of America, Washington DC.

Johnson, L.H., D.M. LeNeveu, D.W. Shoesmith, D.W. Oscarson, M.N. Gray, R.J. Lemire and N.C. Garisto. 1994. The disposal of Canada's nuclear fuel waste: the vault model for postclosure assessment. Atomic Energy of Canada Limited Report, AECL-10714, COG-93-4.

Jones, L.H.P. and K.A. Handreck. 1963. Effects of iron and aluminum oxides on SiO2 in solution in soils. Nature 198. 852-853.

Kafkafi, U., M. Giskin and J. Hagin. 1970. Phosphate and silica adsorption and desorption from soils. Israeli Journal of Chemistry 8, 373-381.

Karim, Z. 1984. Characteristics of ferrihydrites formed by oxidation of FeCl2 solutions containing different amounts of silica. Clays and Clay Minerals 32, 181-184.

Karim, Z. 1986. Formation of ferrihydrite by inhibition of green rust structures in the presence of silicon. Soil Science Society of America Journal 50. 247-250.

Langmuir, D. 1969. The Gibbs free energies of substances in the system Fe-O2-H2O-CO2 at 25°C. U.S. Geological Survey Professional Paper 650-B, B180-B184.

McKeague, J.A. and M.G. Cline. 1963a. Silica in soils. Advances in Agronomy 15. 339- 396.

McKeague, J.A. and M.G. Cline. 1963b. Si in soil solutions. II. The adsorption of monosilicic acid by soil and other substances. Canadian Journal of Soil Science 43, 83-96.

Moskvin, L.N., A.A. Efimov, LA. Varovin, G.A. Usacheva, S.B. Tomilov and A.A. Petrov. 1979. Mb'ssbauer spectroscopic determination of phase composition of corrosion products of structural materials of primary circuit of RBMK-1000 reactor with neutral water conditions. Atomnaya Énergiya (English translation), 46, 28-31. - 12 - AECL-11257 COG-I-95-023

Moskvin, L.N., A.A. Efimov, E.V. Sosnovskaya, B.A. Gusev and S.B. Tomilov. 1981. Application of Môssbauer spectroscopy to the determination of the phase composition of finely dispersed corrosion products of reactor materials in a water coolant. Atomnaya Énergiya (English translation), 51. 383-385.

Obihara, C.H. and E.W. Russell. 1972. Specific adsorption of silicate and phosphate by soils. Journal of Soil Science 23, 105-117.

Parfitt, R.L. 1978. Anion adsorption by soils and soil materials. Advances in Agronomy 30. 1-50.

Parfitt, R.L. 1989. Phosphate reactions with natural allophane, ferrihydrite and goethite. Journal of Soil Science 40, 359-369.

Parfitt, R.L. and J.D. Russell. 1977. Adsorption on hydrous oxides. IV. Mechanisms of adsorption of various ions on goethite. Journal of Soil Science 28. 297-305.

Parfitt, R.L., S.J. van der Gaast and C.W. Childs. 1992. A structural model for natural siliceous ferrihydrite. Clays and Clay Minerals 40, 675-681.

Parks, G.A. 1965. The isoelectric points of solid oxides, solid hydroxides, and aqueous hydroxo complex systems. Chemical Reviews 65, 177-198.

Philipse, A.P., M.P.B. van Bruggen and C. Pathmamanoharan. 1994. Magnetic silica dispersions: preparation and stability of surface-modified silica particles with a magnetic core. Langmuir 10, 92-99.

Rubio, J. and E. Matijevic. 1979. Interactions of metal hydrous oxides with chelating agents. I. p-FeOOH and EDTA. Journal of Colloid and Interface Science 68, 408-421.

Sawicki, J.A. and M.E. Brett. 1993. Môssbauer study of corrosion products from a CANDU secondary system. Nuclear Instruments and Methods in Physics Research, B76. 254-257.

Schwertmann, U. and R.M. Cornell. 1991. Iron oxides in the laboratory. VCH Verlag, Weinheim.

Schwertmann, U. and H. Fechter. 1982. The point of zero charge of natural and synthetic ferrihydrites and its relation to adsorbed silicate. Clay Minerals _T7, 471-476.

Schwertmann, U. and R.M. Taylor. 1972. Influence of silicate on the transformation of lepidocrocite to goethite. Clays and Clay Minerals 20, 159-164.

Schwertmann, U. and H. Thalmann, 1976. The influence of [Fe(II)], [Si] and pH on the formation of lepidocrocite and ferrihydrite during oxidation of aqueous FeCl2 solutions. Clay Minerals J_L 189-199.

Sedov, V.M., P.O. Krutikov, E.A. Konstantinov, A.V. Shul'gin, V.I. Ryabov, Yu.O. Zakharzhevskii, A.P. Eperin and V.G. Shevchenko. 1978. Corrosion products in main technological systems of atomic power plants with an RBMK-1000 reactor during operation. Atomnaya Énergiya (English translation), 46, 23-28. -13- AECL-11257 COG-I-95-023

Sedov, V.M., P.O. Krutikov, E.A. Konstantinov, V,A. Shishkunov and A.A. Afanas'ev. 1980. Môssbauer spectroscopy in^the phase-composition analysis of corrosion deposits in a nuclear power station. Atomnaya Énergiya (English translation), 46. 396-399.

Sigg, L. and W. Stumm. 1980. The interaction of anions and weak acids with the hydrous goethite (a-FeOOH) surface. Colloids and Surfaces 2, 101-117.

Smith, K.L. and R.A. Eggleton. 1983. Botryoidal goethite: a transmission electron microscope study. Clays and Clay Minerals 3J., 392-396.

Stumm, W. 1992. Chemistry of the Solid-Water Interface: Processes at the mineral-water and particle-water interface in natural systems. John Wiley & Sons, New York, NY.

Swaddle, T.W. and P. Oltmann. 1980. Kinetics of the magnetite-maghemite- hematite transformation, with special reference to hydrothermal systems. Canadian Journal of Chemistry 58, 1763-1772.

Taylor, P. and D.G. Owen. 1993. Oxidation of magnetite in aerated aqueous media. Atomic Energy of Canada Limited Report, AECL-10821, COG-93-81.

Taylor, R.M. 1987. Non-silicate oxides and hydroxides. In Chemistry of Clays and Clay Minerals (A.C.D. Newman, editor). Mineralogical Monograph No. 6. John Wiley and Sons, New York, NY, pp. 129-201.

Ticknor, K.V. and Y.-H. Cho. 1990. Interaction of iodide and iodate wuth granitic fracture- filling minerals. Journal of Radioanalytical and Nuclear Chemistry 140. 75-90.

Turner, C.W., K. Shamsuzzaman and R.L. Tapping. 1993. Factors affecting the consolidation of steam generator sludge. Atomic Energy of Canada Limited Report, AECL- 10759, COG-92-352.

Vandergraaf, T.T. and K.V. Ticknor. 1994. A compilation and evaluation of sorption coefficients used in the geosphere model of S Y VAC for the 1990 assessment of the Whiteshell Research Area. Atomic Energy of Canada Limited Report, AECL-10546, COG- 92-59.

Vandergraaf, T.T., K.V. Ticknor and T.W. Melnyk. 1993. The selection of a sorption data base for the geosphere model in the Canadian Nuclear Fuel Waste Management Program. Journal of Contaminant Hydrology H, 327-345.

Vempati, R.K. and R.H. Loeppert. 1989. Influence of structural and adsorbed Si on the transformation of synthetic ferrihydrite. Clays and Clay Minerals 37. 273-279.

Vempati, R.K., R.H. Loeppert, H. Sittertz-Bhatkar and R.C. Burghardt. 1990. Infrared vibrations of hematite formed from aqueous and dry-thermal incubation of Si-containing ferrihydrite. Clays and Clay Minerals 38, 294-298.

Wada, K. and A. Inoue. 1974. Adsorption of monomeric silica by volcanic ash soils. Soil Science and Plant Nutrition (Tokyo) 20, 5-15. - 14- AECL-11257 COG-I-95-023

Waychunas, G.A. 1991. Crystal chemistry of oxides and oxyhydroxides. In Oxide Minerals: Petrologic and Magnetic Significance. Reviews in Mineralogy 25 (D.H. Lindsley, editor). Chapter 2.

Yokoyama, T., T. Nakazato and T. Tarutani. 1980. Polymerization of silicic acid adsorbed on iron(III) hydroxide. Bulletin of the Chemical Society of Japan 53, 850-853.

Zachara, J.M., D.C. Girvin, R.L. Schmidt and C.T. Resch. 1987. Chromate adsorption on amorphous iron oxyhydroxide in the presence of major groundwater ions. Environmental Science and Technology 21, 589-594. AECL-11257 COG-l-95-023

Cat. No. / N° de cat.: CC2-11257E ISBN 0-660-16215-6 ISSN 0067-0367

To identify individual documents in the series, we have assigned an AECL- number to each. Please refer to the AECL- number when requesting additional copies of this document from

Scientific Document Distribution Office (SDDO) AECL Chalk River, Ontario Canada KOJ 1JO

Fax:(613)584-1745 Tel.: (613) 584-3311 ext. 4623 Price: A

Pour identifier les rapports individuels faisant partie de cette série, nous avons affecté un numéro AECL- à chacun d'eux. Veuillez indiquer le numéro AECL- lorsque vous demandez d'autres exemplaires de ce rapport au

Service de Distribution des documents officiels (SDDO) EACL Chalk River (Ontario) Canada KOJ 1JO

Fax:(613)584-1745 Tél.:(613)584-3311 poste 4623 Prix: A