Sedimentary Hosted Iron Ores

Sedimentary Hosted Iron Ores

13.13 Sedimentary Hosted Iron Ores ER Ramanaidou and MA Wells, CSIRO Earth Science and Resource Engineering, WA, Australia ã 2014 Elsevier Ltd. All ...

13MB Sizes 2 Downloads 342 Views


Sedimentary Hosted Iron Ores

ER Ramanaidou and MA Wells, CSIRO Earth Science and Resource Engineering, WA, Australia ã 2014 Elsevier Ltd. All rights reserved.

13.13.1 Introduction 13.13.2 Definition and Classification of Iron-Formation Precambrian Iron-Formation Occurrence and Distribution of Precambrian Iron-Formation BIF Ores Mineralogy Chemical Composition Beneficiation of BIF/GIF 13.13.3 Enriched BIF-Hosted Iron Ores Residual BIF-Hosted Iron Ore Martite–Goethite BIF-Hosted Iron Ore Martite Microplaty Hematite BIF-Hosted Iron Ore Genetic models Weathering of Martite–Goethite and Hematite BIF-hosted Iron Ores Geochemistry of Martite–Goethite and Martite-microplaty Hematite Iron Ores Major and minor elements Rare earth elements Detrital Iron Deposits 13.13.4 Ooidal Ironstones Introduction POI Genesis Marine ironstones Terrestrial ironstones Comparative Evaluation of POI-Derived Iron Ore Mineralogy and Composition Bog Iron Ores General occurrence setting and early utilization Bog iron ore exploitation Types of bog iron ores 13.13.5 Summary Acknowledgments References



Iron, the most abundant and unquestionably the most important element on Earth, is concentrated mainly in the core, though it is only the fourth most abundant element in the earth’s crust (Table 1). The name iron is from a Celtic word, isarnon. From the Latin ferrum, iron bears the chemical symbol, Fe; it is also known as fer in French, hierro in Spanish, ferro in Portuguese, Eisen in German, ja¨rn in Swedish, and Tieˇ in Chinese, to name a few. With an atomic number of 26 and an atomic weight of 55.845, Fe has four naturally occurring stable isotopes and is a transition metal. 56Fe has a larger mass defect than any other nuclide. It is, therefore, the ultimate end product of stellar nuclear fusion (De Laeter et al., 2003). Fe has a total of four valence states (Fe2þ, Fe3þ, Fe4þ, and Fe6þ) although the ferrous (Fe2þ) and ferric (Fe3þ) states are the most significant. Because iron is found in reduced or oxidized environments at or near the surface, the Fe2þ/ Fe3þ relationship is critical since Eh–pH conditions drive its solubility (or precipitation) and influence the elemental and

Treatise on Geochemistry 2nd Edition

313 315 316 317 318 319 319 320 320 321 321 322 324 329 331 332 332 332 334 334 335 335 337 339 343 343 343 344 348 349 349

mineralogical end-product association (Figure 1). Under oxidizing conditions, ferric oxides are highly insoluble and they persist mainly as goethite or hematite, staining rocks with reddish, yellowish and orange hues. Iron was known and exploited in prehistoric times and inscriptions from the earliest civilizations mention its use. The ‘Iron Age,’ starting in the first millennium BC, is the technological stage where iron replaced bronze in weapons. Early men knew about iron in meteorites but because they could not melt it (pure iron melts a 1537  C), it was only used to make small objects such as arrowheads by hammering. Around 4000 years ago in Anatolia, Hittite blacksmiths started to transform iron to produce what they called ‘good iron’ (Raymond, 1986). The iron was heated in burning charcoal incorporating carbon, thus, producing steel by smelting. An additional breakthrough was the quick cooling of the hot metal in water: the quenching effect. Although quenching improves the hardness of steel, it also makes it brittle, creating fine cracks. A further advance, tempering, showed that if quenched steel was reheated at 700  C for a short while and allowed to cool slowly, it loses its brittleness (Raymond,



Sedimentary Hosted Iron Ores

Table 1

Common elements in earth


Percentage of whole earth (by weight)

Percentage of crust (by weight)

Iron Oxygen Silicon Magnesium Nickel Sulfur Calcium Aluminium Sodium Potassium

35 30 15 14 2.4 1.9 1.1 1.1

5 46.6 27.7 2.1 0.01 0.05 3.6 8.1 2.8 2.6

1.2 1.0

Fe[3+] Fe(OH)[2+]



0.6 0.4 Eh (V)


0.2 Fe[2+]



−0.2 −0.4




−0.8 1



7 pH




Figure 1 Eh–pH diagram depicting the stability relations among phases in the system Fe–O–H at 298 K and 105 Pa total pressure. Modified from Takeno N (2005) Atlas of Eh-pH Diagrams: Intercomparison of Thermodynamic Databases. Japan: Geological Survey of Japan.

1986). These three processes are the foundations of the steel industry as we see it today. In 1855, with the development of the Bessemer process, steel production increased drastically. There are three forms of iron; wrought iron, cast iron, and steel. Wrought iron (low carbon, <0.1%C) is only moderately hard. Cast iron (high carbon, >1.8%C) is extremely strong. Steel (0.05 < %C < 2) contains a small amount of carbon in the iron structure and combines the best of both worlds. It can be cast into moulds and shaped when red hot, and holds an edge when it has been sharpened. The collection of ferroalloys with additional elements, such as manganese, silicon, nickel, tungsten, molybdenum, vanadium, cobalt, titanium, or chromium has grown vastly to

produce the advanced materials in use today. They include metal alloys for high tensile and stainless steel, extremely high temperature, high loads used in components for the aircraft, automotive, and general engineering industries such as propeller shafts and automobile parts. Today, steel is employed in large amounts for construction at all scales, such as large structural parts for buildings or tankers, extremely delicate and refined tools, or everyday utensils. The industrial power of a country fundamentally depends on its steel industry in peace or war. Steel is now produced in greater quantities than any other man-made material with the exception of concrete, which confirms its central importance in modern industrialized economies. Iron ore has been mined from igneous, metamorphic and sedimentary hosted deposits for the last three millennia. Most iron ores are mined from large open pits such as Carajas in Brazil and Mount Whaleback in Australia and from some rare underground mines, such as Kiruna in Sweden. Bog iron, a chemical or biochemical iron-rich sediment composed largely of goethite was used during the Iron Age in Scandinavia and the majority of the iron smelted during the time of the Vikings was derived from bog iron. The marine ooidal goethitic iron deposits of the ‘minette’ were mined to feed the furnaces of the Lorraine Basin (Bubenicek, 1960) until 1997. Currently, the only iron-rich ooidal deposits mined are the giant terrestrial West Australian channel iron deposits (CID) (Ramanaidou et al., 2003) and the more modest, riverine–lacustrine equivalents of Kazakhstan (Ionkov et al., 2011; Yanitzkii, 1960). In the twentieth century, iron ore was mostly mined from banded iron-formation (BIF)-hosted, hematite-rich deposits with high Fe grade (>60 wt%) from Australia, Brazil, Canada, India, Russia, South Africa, and the United States of America. Depletion of these direct shipping, BIF-hosted, hematite-rich ores in the USA, as a result of World War II ‘high-grading,’ led to the development of the BIF-hosted magnetite ores requiring beneficiation by removal of the gangue minerals to produce a magnetite concentrate. The current high demand for iron ore in China has seen a similar impact in Australia and the Iron Quadrangle in Brazil, with a new trend for the development of the socalled magnetite BIF-hosted deposits in Western Australia. At the start of the first decade of the twenty-first century, the majority of iron ore production took place in China, Australia, Brazil, and India, while Russia, Ukraine, and South Africa had annual ore production rates between 50 and 100 Mt (Table 2). The ratio of iron ore content to crude ore production over the 2004–2011 period (Table 2) provides an estimate of the relative amount of fresh BIF and BIF-hosted iron ores: the higher the iron concentrate-crude ore ratio, the higher the content of direct shipping ore (DSO). There are four types of iron ore products: (1) high-grade Direct Shipping Ores, formed using crushing and screening methods to produce lump and fines with grades of 55–65% Fe; (2) Concentrates, magnetite ores with 30% Fe, which, following crushing, grinding, and magnetic separation, produce a concentrate with >60% Fe; (3) Pellets, formed from fines, concentrates, or ground ore, mixed with a binder and fired in a grate kiln and (4) Sinter, an agglomerated product formed by firing iron ore fines and coke plus limestone, produced as a furnace feed. In the last 8 years, iron ore production has increased dramatically, fuelled by the fierce demand from China’s steel

Sedimentary Hosted Iron Ores

Table 2


Worldwide iron ore production (U.S. Geological Survey, 2012)










Crude ore

Iron content

Iron content/crude ore

Australia Brazil Canada China India Iran Kazakhstan Mauritania Mexico Russia South Africa Sweden Ukraine United States Venezuela Other countries Total

234 262 29 320 121 18 20 10 11 97 39 22 66 55 19 37 1360

262 281 30 420 152 19 19 11 12 97 40 23 69 54 21 42 1552

275 318 34 601 181 26 22 11 11 102 41 23 74 53 22 67 1861

299 355 33 707 202 32 24 12 11 105 42 25 78 52 21 47 2045

342 355 31 824 220 32 23 11 12 100 49 24 73 54 21 47 2218

394 300 32 880 245 33 22 10 12 92 55 18 66 27 15 43 2244

433 370 37 1070 230 28 24 11 14 101 59 25 78 49 14 48 2591

480 390 37 1200 240 30 24 11 14 100 55 25 80 54 16 50 2806

35 000 29 000 6300 23 000 7000 2500 3000 1100 700 25 000 1000 3500 6000 6900 4000 12 000 166 000

17 000 16 000 2300 7200 4500 1400 1000 700 400 14 000 650 2200 2100 2100 2400 6200 80 150

0.49 0.55 0.37 0.31 0.64 0.56 0.33 0.64 0.57 0.56 0.65 0.63 0.35 0.30 0.60 0.52 0.48

mills. In 2011, China imported almost two-thirds of the world’s total iron ore exports and produced about 60% of the world’s pig iron, bringing worldwide iron ore production and price to unparalleled levels. This chapter concentrates on sedimentary hosted iron ores that represent 90% of the currently mined iron ores. Of the 220 iron ore mines in production in 2011, ninety percent were sedimentary hosted with the remaining ten percent covering magmatic and skarn deposits, such as Savage River in Australia, El Romeral in Chile, La Perla in Mexico, Marcona in Peru, Faleme in Senegal, and Kiruna and Malberget in Sweden. The key objective of this chapter is to cover sedimentary hosted iron deposits. Whether sedimentary hosted iron deposits can be considered as iron ore is not unconditional but is underpinned by the geological, geographical, economical, social, technical and political parameters at a given time. Today (in 2013), high-grade iron ores must have excellent chemical grade, that is, more than 55% Fe, less than 5% of SiO2 and Al2O3, and less than 0.075% P, hold superior metallurgical properties, and be free of environmentally damaging elements such as chlorine and sulfur, to name a few. Iron ores with chemical composition and metallurgical characteristics outside of these boundaries can still be mined profitably but will attract financial penalties. The bulk of iron ore mined today is extracted from Precambrian iron-formation and a prominent part of the chapter is dedicated to these deposits. Ooidal iron ores mined in large volumes (>120 Mt annually) in Western Australia and in Kazakhstan are also described, as well as those of past importance. Finally, the bog iron deposits are covered because of their historical significance.

13.13.2 Definition and Classification of Iron-Formation The current accepted definition for iron-formation is derived from James (1954, 1966). In discussing the North American deposits, he defined iron-formation as a sedimentary rock with more than 15% of Fe. He classified these sediments into two

types; (1) the cherty iron-formation of Precambrian age and (2) the non-cherty oolitic iron rocks or ‘ironstones’ of Phanerozoic age. James (1983) suggested that the division between cherty and non-cherty iron-formations was the result of custom rather than logic. Following the widely used shorthand introduced by Trendall and Blockley (1970), James (1983) was more specific in defining BIF as “a lithological term for a chemical sediment consisting of a thinly layered or laminated rock in which chert (or its metamorphic equivalent) alternates with layers that are composed mainly of iron minerals; the iron content (as Fe) is typically in the range 20–35% and SiO2 is in the range 40–50%.” Granular iron formation or GIF is often described under the term BIF although the essential difference between the two types has been documented for a long time. GIF has been defined as well-sorted chemical sands (Clout and Simonson, 2005). It is important to note that the vast range and conflicting concepts of iron-formation are often based on parochial studies and are applied to new developments elsewhere without adequate study. Trendall (1983) suggested that the term iron-formation (hyphenated) be used as a “lithological and stratigraphical term for iron-rich sedimentary rocks ‘whose principal chemical characteristic is an anomalous high content of iron“ thus returning to the concept of James (1954) by including ‘iron stones’ in ‘iron-formation.’ Others such as Stanton (1972) and Young (1989), have remained with the traditional ‘iron-formation’ and ‘iron stones,’ the latter including the volcano-sedimentary iron deposits and the bog iron deposits. Dimroth (1977) proposed a classification of iron-rich sediments into three types; (1) detrital chemical sediments encompassing the cherty iron formation (GIF and BIF); oolitic type and the aluminous iron-formation (Minette-type); (2) ironrich shales including pyritic shales (sulfide iron-formation) and the siderite-rich shales (clay ironstones, coal ironstones), and (3) ‘other types’ that consist of iron-rich laterites, bog iron ores, manganese nodule and oceanic iron crusts, iron-rich mud; and lastly, placers of magnetite, hematite, or ilmenite sands. Zitzmann and Neumann-Redlin (1977) have proposed an extremely comprehensive iron ore classification scheme based


Sedimentary Hosted Iron Ores

on the ‘genetic concept’ for all European iron ore deposits, which contains all iron ore types and not just the sedimentaryassociated types; it includes the following ten deposit types; (1) Liquid magmatic; (2) Intrusive magmatic; (3) Contact metasomatic; (4) Hydrothermal with two subtypes, metasomatic and vein; (5) Volcano-sedimentary; (6) Marine-sedimentary with four subtypes, oolitic, detrital, placers, and oolitic-detrital; (7) Continental-sedimentary with three subtypes, limnal, fluvial, and blackbands; (8) Residual; (9) Metamorphic banded ore, and (10) polymetamorphic skarns. Kimberley (1989a,b) proposed an iron-formation classification based on six paleoenvironmental groups: (1) Shallow-volcanic-platform iron-formation (SVOP-IF); (2) Metazoan-poor, extensive, chemical-sediment rich, shelf-sea iron-formation (MECS-IF); (3) Sandy, clayey, and oolitic, shallow island-dotted-sea iron-formation (SCOSIF); (4) Deep-water iron-formation (DWAT-IF); (5) Sandy, oolite-poor, shallow-sea iron-formation (SOPS-IF) and (6) Coal-swamp iron-formation. As Kimberley (1989a,b) stated, the classification for categorizing iron-formations is always subjective. The British Geological Survey (BGS) provides a cohesive classification for all sedimentary rocks (Hallsworth and Knox, 1999) that is not based on genesis but is purely descriptive. This classification consists of 11 categories and subgroups based on grain size. ‘Iron sediments and ironstones’ is one of the eleven categories that are defined as having more than 50% iron-bearing minerals. Iron sediments represent the unlithified iron-rich sediments and are divided into three types based on particle size. The classification of lithified iron-rich sediments or ironstones is based on textural information. In this chapter, for simplicity and clarity, the traditional separation of iron-formation and ironstone has been retained. As many authors (e.g., James, 1966; Kimberley, 1989a,b; Zitzmann and Neumann-Redlin, 1977) have stated – any universal classification of iron-formation that includes iron ore is arbitrary and subjective.

Precambrian Iron-Formation

The literature on Precambrian iron-formations is extremely well documented with publications from Australia (Goode et al., 1983; Krapez, 1993; Krapez et al., 2003; Morris, 1983; Trendall, 1980, 1983; Trendall and Blockley, 1970); Brazil (Dorr, 1969, 1973; Rosiere et al., 2008; Spier et al., 2003); Mongolia (Ilyin, 2009); North America (Bekker et al., 2010; Cloud, 1973, 1983; Dimroth, 1976; Dimroth and Chauvel, 1973; Gross, 1980; Gross et al., 1983; Gross and Zajac, 1983; Holland, 2005; James, 1954, 1983; Klein, 2005; Klein and Beukes, 1993; Lepp, 1987; Morey, 1983; Simonson, 1987, 2003); South Africa (Bekker et al., 2010; Beukes, 1973, 1983; Beukes and Gutzmer, 2008; Beukes and Klein, 1990); and Ukraine (Belevtsev et al., 1983). The earliest classifications of Precambrian iron-formations were defined for North America (e.g., Dimroth and Chauvel, 1973; Gross, 1965; Gross and McLeod, 1980; James, 1954). The American Precambrian iron-formations have been divided into three types, namely, the Lake Superior-type, the Algoma-type (Gross, 1965, 1980; Gross et al., 1983), and the Rapitan-type (Klein and Beukes, 1993). The Lake Superior-type is cratonic and generally structurally undeformed. The Algoma-type is, in

contrast, typically strongly folded, occurs in greenstone belts, and is interbedded with volcanic rocks, graywacke, turbidite, and pelitic sediments; the sequences are commonly metamorphosed (Gross, 1965, 1980; Gross et al., 1983). Ages of the Algoma-type range between 3.5 billion and a few hundred million years. The Neoproterozoic BIF of the Rapitan in Canada and the Urucum in Brazil rarely extends more than 10 km. It is limited to Phanerozoic orogenic zones (Ilyin, 2009) and is associated with glacial diamictites (Beukes and Gutzmer, 2008; Klein and Beukes, 1993). James (1954) has divided the iron-formations into four facies, namely, oxide, silicate, carbonate and sulfide, based on their dominant mineralogy. Goode et al. (1983) pointed out that the Nabberu Basin (Western Australia) represented the first area of extensive granular (Superior-type) iron formations to be recognized outside North America, contrasting with the laminated type of the Hamersley Province. Trendall and Blockley (2004) and Trendall and Morris (1983) recommended not using the Algoma/Superior classification as it has no clear sense and has led to confusion. For instance, the Dales Gorge member of the Brockman Iron Formation (no hyphen) in Western Australia was first classified as Algoma by Dimroth (1976), as Superior by Gross (1980), and later as Algoma by Gross (1991), and now as Lake Superior by Bekker et al. (2010). Texturally, iron-formations are also divided into two groups: (1) BIF is dominant in Archean to earliest Paleoproterozoic successions, whereas (2) GIF is much more common in later Paleoproterozoic successions. Trendall (2002) suggested that the terms BIF and GIF be considered as two lithological subtypes of iron-formation. Klein (2005) showed that most wellbanded IF (BIF) are older than 2.0 Ga and that the banding remains even after intense metamorphism, whereas the younger Lake Superior region and Labrador Trough IF are often granular (GIF) with decimeter-scale banding. The primary subdivision of Precambrian iron-formations of Clout and Simonson (2005) is based on banding and granularity derived from their original particle size. In this classification, BIF was deposited as a chemical mud whereas GIF formed as chemically derived sands (Goode et al., 1983). GIF is crudely banded and more discontinuous than BIF. Clout and Simonson (2005) based their classification on James (1954) with the four IF facies, namely, oxide, silicate, carbonate, and sulfide, but they mention that although James (1954) called them ‘sedimentary facies,’ they are more associated with metamorphic facies based on the dominant mineralogy. GIF have dominant oxide or silicate mineralogy while IF also include the carbonate facies. The Sulfide facies is rarely encountered. They also point out that, as with all sedimentary rocks, textures and structures are more important than mineralogy for understanding genesis. Beukes and Gutzmer (2008) further developed the definition of Precambrian iron-formation by proposing a textural classification to encompass all occurrences of Precambrian iron-formation based on three components (1) the granules; (2) the matrix; and (3) the microcrystalline quartz (or chert) giving rise to the GIF; the micritic ironformation (MIF), and lastly, the banded micritic ironformation. Trendall (1965) and Trendall and Blockley (1970) proposed a now widely accepted convenient terminology for BIF banding based on the scale involved – macrobanding

Sedimentary Hosted Iron Ores

(meter scale), and mesobanding (millimeter to centimeter), with microbanding (sub- to 5 mm) applied specifically to a varve-related type, and microlaminae (sub-mm) (Ewers and Morris, 1981; Morris, 1993). Occurrence and Distribution of Precambrian Iron-Formation

Abundance of BIF relative to Hamersley Group BIF as maximum

Precambrian iron-formations occur at specific times in Earth history. Klein (2005), following Klein and Gole (1981), depicts the relative volumes of BIF/GIF deposited on Earth (Figure 2). With the exception of rare occurrences centered at 3.8 Ga (Isua, Greenland), 0.7 Ga (Rapitan in Canada; Urucum in Brazil, and Damara in Namibia), and 0.4 Ga (Devonian of Western Siberia in Kalugin, 1973), the bulk of BIF deposition occurred between 3.5 and 1.8 Ga with a maximum around 2.5 Ga bearing the largest BIF of the Hamersley Group in Western


Australia; the Itabira Group in Minas Gerais; and the Transvaal Group in South Africa. Bekker et al. (2010) provide a schematic geographical distribution of the Precambrian iron-formations with estimated figures for the sizes (Figure 3). The genesis of Precambrian iron-formation (BIF/GIF) is still strongly debated and, thus, a large volume of the literature has been devoted to the topic, of which the most recent include (Ayres, 1972; Barley et al., 1998; Bekker et al., 2004, 2010; Beukes, 1973; Beukes and Gutzmer, 2008; Braterman and CairnsSmith, 1987; Cloud, 1968, 1973; Drever, 1974; Franc¸ois, 1986; Goodwin, 1982; Holland, 1984, 2005; Huston and Logan, 2004; James, 1983; Klein and Beukes, 1992; Konhauser et al., 2002; Laberge et al., 1987; Lepp, 1987; McConchie, 1987; Morris, 1993; Morris and Horwitz, 1983; Robbins et al., 1987; Simonson, 2003; Trendall, 1983; Trendall and Blockley, 1970). The styles of deposition of iron-formations have changed through time and most authors agree that these changes are connected to the

Itabira Group BIF, Quadrilatero Carajas Formation, Ferrifero, Minas Gerais, Brazil Brazil Hamersley Canadian greenstone belts; Group, W.A. Transvaal Supergroup, S. Africa Yilgarn Block, W.A. Lake Superioir Region, USA Nova Lima Krivoy Rog series, Ukraine Group BIF, Brazil Zimbabwe; S. Africa; Ukraine; Venezuela; Brazil; Western Australia

Labrador Trough, Canada Frere Formation IFs, Nabberu Basin, Western Australia

Isua, West Greenland







Rapitan Group, Canada; Urucum Region, Brazil; Damara Supergroup, Namibia Western Siberia, Russia



Ga Figure 2 Relative abundance of Precambrian and Neoproterozoic BIF including the main BIF areas. Modified from Klein C (2005) Some Precambrian banded iron-formations (BIFs) from around the world: Their age, geologic setting, mineralogy, metamorphism, geochemistry, and origins. American Mineralogist 90: 1473–1499.

Deposit size >100 000 Gt 10 000–50 000 Gt 1000–10 000 Gt 1–150 Gt Age Neoproterozoic Mesoproterozoic Middle-late Paleoproterozoic Early Paleoproterozoic/Neoarchean Mesoarchean-Neoarchean Figure 3 Occurrences of Precambrian iron-formations. Modified from Bekker A, Slack JF, Planavsky N, et al. (2010) Iron formation: The sedimentary product of a complex interplay among mantle, tectonic, oceanic, and biospheric processes. Economic Geology 105: 467–508. With permission from the Society of Economic Geologists.


Sedimentary Hosted Iron Ores

global evolution of Earth. The latest paper on iron-formation genesis (Bekker et al., 2010) argues that iron-formation depositions are linked to time periods when large igneous provinces (Barley et al., 1998) formed and relate BIF genesis to volcanic sediments redistributed by turbid currents with later silica replacement to form the chert horizons (Krapez et al., 2003). For instance, around 2.6 Ga, the deposition of the largest Precambrian iron-formation coincides with the construction of large continents that impacted ‘the heat flux at the core–mantle boundary’ as well as the rise of oxygen in the atmosphere (Bekker et al., 2010). The iron-formations deposited during the Neoproterozoic are correlated to intense magmatic activity and global glaciations, such as the ‘Snowball Earth’ period.

BIF Ores

BIF per se currently provide 50% of iron ore worldwide (Table 2). Synonyms for BIF include the ‘banded jaspilites’ (a)

of Australia (Figures 4 and 5); the ‘itabirites’ of Brazil (Figure 6); the ‘banded hematite quartzite or BHQ’ of India; the ‘taconite’ of the North American Lake Superior; the ‘ironstone’ of South Africa and the quartz-banded ore in Scandinavia (Stanton, 1972). Some of these terms are now obsolete but the use of itabirite is still thriving in Brazil. Fifty percent of the iron ore deposits hosted in BIF are located in China and to some extent in North America. China represents around 80% of the BIF mined globally with a yearly production of 1200 Mt. Other projects ready to come on line include the Serra Azul in the Iron Quadrangle in Brazil (Amorim and Alkmim, 2011); the Project Magnet in South Australia and the George Palmer and Karara Magnetite Deposits in Western Australia. According to Flint and Preston (2010), 15 Gt of BIF exist in the Hamersley Province of Western Australia. However, the vast bulk of BIF will not be mined despite its easy access until economic necessity steps in.

Mesobanding in BIF - Dales Gorge, Karijini National Park

Magnetite oxidising to martite and kenomagnetite



BIF varves

10 mm M-mpIH ore at OB 18 Figure 4 (a) Mildly weathered DGM BIF at Fortescue Falls, Dales Gorge, Karijini National Park. The face illustrates typical mesobanding, showing mainly chert and magnetite bands here, with the magnetite ranging from fairly fresh to oxidized with pressure-induced podding in some bands. Strongly oxidized carbonate–silicate bands appear at the top of the photo. Note also the fine regular ‘microbanding’ of some chert mesobands (magnified in (b)) with the equivalent superbly preserved sequence in ore (c). Reproduced from Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia - A critical review. Australian Journal of Earth Sciences 58: 417–451

Sedimentary Hosted Iron Ores

1m Figure 5 West Australian banded iron-formation from the Fortescue Falls.

ripidolite, ferri-annite, and various carbonates, such as the siderite and dolomite-ankerite series. Fibrous riebeckite, also called crocidolite or blue asbestos, is found in the BIF of Hamersley and Transvaal Provinces. Blue asbestos was mined in Wittenoom in the Hamersley’s from 1947 but was stopped in 1966 because it had ceased to be profitable. Regional metamorphism impacts on the liberation of the minerals and sometimes on the behavior of the pelletization process (Han, 2004). Low grade metamorphism that generates Mg carbonates, such as siderite and magnesiosiderite, can be an impediment to MgO content control as well as increasing heat intake, decreasing pellet induration, and throughput (Johnson et al., 2007). The increase in metamorphic grade is demonstrated by mineralogical changes with the appearance of high-temperature minerals such as amphiboles (cummingtonite, grunerite, tremolite, ferri-actinolite), almandine, and, finally, pyroxene and olivine (Figure 7). With increasing metamorphic grade, the particle size of quartz and magnetite also increases (Dorr, 1969). An increase in grain size and mineralogical changes were also observed by Pires (1995) from the Western greenschist facies to the Eastern amphibolitic facies in the Quadrila`tero Ferrı`fero in Brazil. The increase in grain size of magnetite with metamorphic grade should improve its liberation during processing. However, some plants might not be able to grind large magnetite grains, which can lead to a diminution in throughput (Johnson et al., 2007). The amount of hematite can also increase with metamorphism; for instance, in the fresh BIF of the Capanema mine, the amount of magnetite is not dominant and most of the primary iron oxides are hematitic (Ramanaidou, 2009). Some of these fresh BIF are currently mined in Brazil as ‘itabirite.’ 10 cm Figure 6 Brazilian banded iron-formation from the Capanema mine in Minas Gerais.


A comprehensive range of minerals occurring in ironformations is listed in Table 3. Klein (1983, 2005) and Miyano and Klein (1983) distinguish three metamorphic grades to classify the metasediments that are the Precambrian iron-formations: diagenetic to very low, medium-grade, and high-grade metamorphism (Figure 6). The South African BIF of the Kaapvaal Craton and the Western Australian BIF of the Hamersley Province are good examples of very low metamorphism (Trendall, 1983). The former region has experienced burial temperatures ranging from 100 to 150  C (Miyano and Klein, 1983) whereas the latter has undergone greenschist metamorphism (200–300  C) according to Klein and Gole (1981). The North American Negaunee, Biwabik, Gunflint, and the Sokoman iron-formations have also experienced late diagenetic or very low-grade metamorphic events. The minerals encountered include, by order of importance, chert, magnetite, hematite, and stilpnomelane (Figures 7–10) with additional minnesotaite, greenalite, pyrite and rare chamosite,


Chemical Composition

The chemical composition of Precambrian iron-formations from Australia, Brazil, South Africa, and North America (Table 4) shows that silica varies between 30% and 52%, alumina is below 5% and generally less than 2%, and iron varies between 20% and 40% and is generally in magnetite (see values of Fe2þ and Fe3þ) or hematite. Elements such as calcium and magnesium can reach 10% in the structure of carbonates, silicates, and phosphates. Sodium and potassium are associated with silicates. Phosphorus, a deleterious element during the steel making process, generally occurs as apatite Ca5 (PO4)3(OH, F, Cl). Overall, as Klein (2005) commented on Precambrian iron-formations, “they are clean sediments devoid of detrital input.” However, rare earth elements (REE) have been widely used to understand the genesis of iron-formations as it is assumed that, as occurs in modern equivalents, REE are coprecipitated with iron oxides (Bekker et al., 2010; Derry and Jacobsen, 1990; Frei et al., 2008; Klein and Beukes, 1989; Olivarez and Owen, 1989; Ruhlin and Owen, 1986). For example, REE abundances in combination with petrography were used in genetic modeling of the Hamersley Province iron-formation (Morris, 1993). This author suggested that BIF were deposited by a combination of surface currents precipitating the silicarich mesobands and upwelling from mid ocean ridges producing the iron-rich mesobands (Morris and Horwitz, 1983).


Sedimentary Hosted Iron Ores

Table 3

Mineralogy of iron-formation and iron ore





FeO Fe2O3




Fe2þ1y (Fe2þFe3þ1þ2/3y[ ]y/3)O4












(Fe2þ,Fe3þ,Al,Mg,Mn)2(Si, Al)2O5(OH)4

Named in 1845, magnetite is derived from the name of a Greek shepherd, Magnes, on Mt Ida after noting that iron nails in his shoe and the iron ferrule of his staff were attracted to a rock (Gaines et al., 1997) Named in 1927 from the first syllables of MAGnetite and HEMatite, in reference to the magnetism of the former and the composition of the latter (Gaines et al., 1997) Intermediate phase that forms during the topotactic oxidation of magnetite to maghemite, with the term kenomagnetite applied to transitional phases of the magnetite-maghemite series (Greenwood and Gibb, 1971; Kullerud et al., 1969) Named in antiquity from the Greek, ‘haimatitis,’ in reference to the ‘blood-like color’ of the powder Martite: Term used to describe hematite pseudomorphs after magnetite Platy and microplaty hematite: Terms introduced by Morris (1980, 1985) to describe fine platy hematite crystals around 50–100 mm in size found in iron ore Named in 1806 after German poet, philosopher and naturalist Johann Wolfgang von Goethe (1749–1832) (Gaines et al., 1997). In the past, the term ‘limonite’ has been synonymous with goethite (Nickel and Nichols, 1991) Many types of goethites are found in iron ore; they include: (1) the goethite pseudomorphs after gangue minerals such as chert, carbonate, and silicates (Morris, 1980, 1985); (2) ochreous goethite or ‘limonite’ – soft to medium hard, with micropores, yellow in color with a chalky appearance, and; (3) vitreous goethite – black to dark brown in color, hard, and glassy with a conchoidal fracture Known since antiquity, the name is derived from the Greek, ‘pyr,’ for fire as sparks can be produced when crystals are struck (Gaines et al., 1997) Name derived from the Greek word sideros, ‘iron.’ It is a valuable iron mineral, since it is 48% iron and contains no sulfur or phosphorus. Both magnesium and manganese commonly substitute for the iron Named after the municipality of Chamoson, in Switzerland. Chamosite is the Fe2þ end member of the chlorite group. A hydrous aluminium silicate of iron, which is produced in an environment of low to moderate grade of metamorphosed iron deposits, as gray or black crystals in Oodial iron stones Berthierine is used as the term to describe Fe-rich 1:1 trioctahedral (serpentine group) silicates of general formula with a 0.7 nm basal spacing (Taylor, 2005; Young, 1989)

Source: Ramanaidou ER, Wells M, Belton D, Verral M, and Ryan C (2008) Mineralogical and microchemical methods for the characterization of high-grade BIF derived iron ore. Reviews in Economic Geology 15: 129–156.

Although many elements including REE are important for genesis, only some are critical for the evaluation of iron ore and include iron, silica, alumina, phosphorus, sulfur, and, locally, chlorine, calcium, magnesium, sodium, titanium, and loss on ignition (LOI). Examples of worldwide iron-formation bulk chemistry (Table 4) show that silica and Fe are dominant with various amounts of Ca, Mg, Na, and K based on the presence of carbonates or silicates. Fe2þ and Fe3þ ratios were given where possible as they indicate the relative proportion of magnetite and hematite (although some Fe occurs in silicates and carbonates). Iron grade is the first parameter to determine the value of iron ore. Other elements such as Si, Al, and P are deleterious as their presence will affect the performance of downstream processing (Table 5). Alumina, sulfur, and phosphorus values of concentrates from Chinese mines derived from BIF/GIF show that these elements vary drastically, in particular S and P (Table 6). Trace elements in Precambrian iron-formations are generally low (Table 7).

Beneficiation of BIF/GIF

Mineralogy and grain size are two of the most critical parameters to beneficiate BIF. Understanding the speciation of iron is

important as Fe can also occur as silicates and carbonates. Upgrading of a magnetite BIF/GIF to ore requires beneficiation including pulverization, generally to less than 20–75 mm, to liberate the minerals and magnetic separation sometimes followed by flotation to remove the quartz, and hydrocyclone to collect the ultrafines in which the gangue minerals are removed to generate a magnetite concentrate. Hematite BIF/GIF upgrading involves more complex beneficiation as hematite is not magnetic. It also requires grinding as well as wet high-intensity magnetic separation, and hydrocyclone to remove the ultrafines, spirals, and reverse flotation.


Enriched BIF-Hosted Iron Ores

Classification of the different types of high-grade iron ores is complex as they are the result of multiple, superimposed hypogene and supergene processes. These various enrichments are well documented but consensus on formation processes, in particular for the hematite ores (martite-microplaty hematite), has not been reached and a strong and healthy debate still rages.

Sedimentary Hosted Iron Ores


We have subdivided the enriched BIF-hosted iron ores into two main types: (1) the bedded iron deposits or BID that provide the bulk of the high-grade iron ore, and (2) the minor detrital iron deposits or DID that are the result of the breakdown of BID. BID include both additive and subtractive aspects and they have been classically classified into three subtypes:

80% consist of martite–goethite ore. It is important to note that the martite–goethite and hematite ores are often affected by weathering; this weathered blanket is commonly referred to as ‘soft ore’ or ‘blue dust’ in Brazil and West Africa as opposed to the unweathered ‘hard ore.’ Soft ores currently represent a large proportion of mining operations. Residual iron ores represent only a small component of DSO.

1. Residual iron ore 2. Martite–goethite ores 3. Hematite (martite-microplaty) ores

These enriched BIF-hosted iron ores and, in particular, the hematite iron ores, are widespread in places such as Australia, Brazil, Guinea, India, and South Africa and represent the main direct shipping ores (Richardson, 2004). In Australia, the hematite iron ores represent only 20% of the BID DSO whereas

Grade of metamorphism Medium


Biotite zone

Diagenetic Early Late Chert Quartz ‘Fe3O4 + H2O’ Magnetite

Garnet zone

High StauroliteSillimanite kyanite and zone kyanite zone

‘Fe(OH)3’ Hematite Greenalite Stilpnometane Ferri-annite Talc - minnesotaite Fe - chlorite (ripidolite) Dolomite - ankerite Calcite Siderite - magnesite Riebeckite Cummingtonite - grunerite (anthophyllite) Tremolite - ferroactinolite (hornblende) Almandine Orthopyroxene Clinopyroxene

Residual BIF-Hosted Iron Ore

Supergene lateritic weathering processes generate iron enrichment by dissolution of the least stable minerals and reprecipitation of secondary iron minerals in the upper part of a formation and, as a consequence, the parental textures are usually obliterated. Weathering of BIF can generate high quality, residual iron ores by dissolving the gangue components (blue dust) by lateritic alteration (Beukes et al., 2003; Chicarino Varaja˜o et al., 1997; Dorr, 1973; Freyssinet et al., 2005; Melfi et al., 1988; Ramanaidou, 2009; Rios et al., 2004; Rosiere et al., 2002, 2008; Roy and Venkatesh, 2009; Spier et al., 2003; Taylor et al., 2001). In the Brazilian Capanema mine, Ramanaidou (1989, 2009) demonstrated that in situ chemical weathering of BIF in high rainfall areas resulted in the leaching of the BIF gangue and oxidation of magnetite to martite, and the development of a thick and well-defined lateritic blanket made up of consecutive weathering horizons (Figure 11). These include, from the base, (1) fresh BIF or itabirite (2) a desilicified horizon where the quartz is being progressively dissolved, (3) a friable hematite-rich horizon with domains similar to ‘blue dust,’ (4) a goethite-rich cemented horizon with partial loss of primary texture consisting of many generations of aluminous goethite and hematite covered by (5) ferruginous duricrust also called ferricrete or, locally, ‘canga.’ The weathering process affecting the BIF is subtractive and the iron enrichment is relative as the gangue material, principally quartz, is dissolved and the primary iron oxides remain. The primary texture is significantly destroyed as weathering intensifies, with the end product a surficial ferruginous duricrust or canga with textural characteristics ranging from massive, vermiform, botryoidal, and nodular to ooidal.

Fayalite Figure 7 Relative stabilities of minerals in metamorphosed ironformations as a function of metamorphic zones. Modified from Klein C (2005) Some Precambrian banded iron-formations (BIFs) from around the world: Their age, geologic setting, mineralogy, metamorphism, geochemistry, and origins. American Mineralogist 90: 1473–1499.

Martite–Goethite BIF-Hosted Iron Ore

The martite–goethite BIF-hosted iron ores are very common in Australia, including Western and South Australia, and are also found in India and South America. They represented the bulk of the DSO products exported from Australia in 2011 and will

4 cm Figure 8 Microphograph of BIF with typical layers of iron oxides (white) and gangue material in reflected light.


Sedimentary Hosted Iron Ores



0.1 mm Figure 9 Microphograph of BIF with raft of magnetite (Mt), in a chert (or microquartz) and stilpnomelane (St) matrix (reflected light).



0.1 mm Figure 10 Microphograph of BIF with raftS of magnetite (Mt), with chert (or microquartz) and carbonate (Cb) (reflected light).

very likely remain so for the next few decades. These deposits are hosted in the Archean Marra Mamba Iron Formation (Figure 12) with the iron ore mines of Marandoo, Mining Area C, West Angelas, Hope Downs, OB29, Christmas Creek as well as in the Proterozoic Brockman Iron Formation with the iron ore mines of Paraburdoo Eastern Range, Tom Price Section 6 (Figure 13), Opthalmia Range, and Rhodes Ridge (Figure 12). These martite–goethite ores are also found in the Northern Pilbara (e.g., Abydos and Pardoo) as well as in the Yilgarn (e.g., Koolyanobbing, Mt Jackson). They were first defined by Morris (1980, 1985). He showed that one of the main characteristics of the martite–goethite ores is that the original texture of the BIF from the micro- to the macroscale is fully preserved and not destroyed by weathering or metamorphism. While primary hematite is unchanged, primary magnetite has oxidized to martite and the gangue minerals such as chert, silicates, and iron-rich carbonates are pseudomorphed by goethite or, in part, leached out. The martite–goethite ore grades range between 58% and 63% Fe and P values vary between 0.05% and 0.2% P, with the Dales Gorge iron ores generally richer in P than the Marra Mamba iron ores (Morris and Kneeshaw, 2011). Petrological studies (Ramanaidou et al., 2008) demonstrated that there is a compositional variation of goethite pseudomorphs and that the composition of goethite reflects the minerals it has

replaced. Each type of goethite pseudomorph after chert, silicates, or carbonates displays a specific range of alumina, phosphorus, and silica. Goethite pseudomorphed after carbonates has the highest amount of Al, P, and Si; goethite pseudomorphed after chert contains the least amount of Al, P, and Si, while goethite pseudomorphed after silicates contains intermediate amounts of Al, P, and Si (Ramanaidou et al., 2008). The genesis of martite–goethite ore (Figure 14) has been explained in detail by Morris and Kneeshaw (2011) and is reproduced verbatim here: “Most of the deep-seated BIFhosted ferric ores extend to depths well beyond the likely reach of oxygenated water, in part due the original high content of Fe2þ in BIF. Fe3þ is insoluble under most natural conditions. Thus, to have concentrated in its present position by mimetic goethite replacement of the BIF matrix minerals, iron must have been transported in the soluble Fe2þ form in ground water, followed by oxidation to Fe3þ in the anode area with immediate precipitation as Fe3þ hydroxyoxides by hydrolysis. This in turn liberated the replaced gangue elements to the flowing aquifer systems and out into the surface drainage. Because magnetite and its kenomagnetite-maghemite derivatives are excellent conductors of electrons, seasonallycontrolled atmospheric effects in the sub-outcrop can drive massive electrochemical cells in BIF at depth. Solution of iron from surface BIF, probably organically mediated, leaves a permeable residue of friable silica locally known as ‘denatured BIF.’ This readily erodes, exposing more BIF for subsurface reaction. Fe2þ is transferred in groundwater through pores and fractures to the reacting zone at depth and precipitated as Fe3þ, pseudomorphing chert (Morris and Fletcher, 1987), carbonates, and silicates by iron oxyhydroxide, with partial oxidation of magnetite to kenomagnetite during this phase and, eventually, to martite in the weathering stages. The main oxidative process is by electron exchange and no free O2 is required in the reacting zone during genesis. Ionic exchange via groundwater completes the reaction. Thus, the ore body grows upward as rapid erosion removes the friable exBIF surficial silica residue forming a topographic low that potentially increases the groundwater flow through the system. The enrichment process ceases when there is no more BIF to erode, when climate changes occur, or if tectonic movement intervenes to initiate burial. The iron ore ridges of today do not reflect the early genetic conditions, and the amount of erosion needed during genesis is trivial compared to the vast amount of erosion of the surrounding rocks that resulted in the current reverse topography of the deposits.“

Martite Microplaty Hematite BIF-Hosted Iron Ore

The martite-microplaty hematite BIF-hosted iron ores are prevalent worldwide where they represent the main DSO. In the Hamersley Province, Western Australia, martite-microplaty type iron ores represent 20% of the total DSO such as the deposits of Mount Whaleback and Mount Tom Price (Figure 12). These deposits display hard, medium, and soft hematite ores, with the soft ores equivalent to the blue dust ores, which are formed by strong leaching of martite-microplaty hematite ores (Clout, 2002; Morris, 1985). The porosity varies from a few percent to up to 50% in Mount Tom Price and

Table 4

SiO2 Al2O3 Fe2O3 FeO MgO CaO Na2O K2O H2Oþ H2O CO2 TiO2 MnO 0.44 P2O5 C S FeS2 SO2 LOI Total Fe as Fe2O3 Fe3þ Fe2þ Total Fe Fe3þ/Fe2þ

Chemistry (wt%) of Precambrian iron-formations 1






















42 0.3 22 15.3 2.88 6.7 0.13 0.14 1.1 0.12 8.37 0.06 0.14

43.51 0.36

44.34 0.89 29.23 13.42 2.3 1.78 0.53 1.26 0.98 0.17 4.62 0.05 0.17

49.5 2.4 19.9 14.4 2.9 3.5 0.22 0.61 2 0.6 3.1 0.14 0.07

40.85 0.41 30.52

42.5 0.4 22.5 16.9 3.78 5.81 0.43 0.15 1.3 0.17 5.25 0.2 0.12

45.9 2.23 3.4 27.2 4.95 1.46 0.26 0.54 3.58 1.26 7.73 0.12 0.12

51.1 1.2 5.2 23.4 4.25 1.86 0.19 0.41 2.86 0.21 6.47 0.06 0.07

38.8 2.5 26.8

45.78 2.04 27.77

43.15 0.09 46.37

41.5 0.09 57.7

47.3 1.07 43.1

5.14 5.65 0.03 0.03

1.1 2.63 0.32 0.02

2.85 1.75 0.04 0.02

0.26 0.05 0.01 0.02

0.03 0.02 0.01 0.005

5.68 3.03 0.33 0.27

14.3 0.09 0.24

4.47 0.01 0.05

53.63 4.74 29.44 3.24 2.15 <0.01 0.08 1.77 3.42 0.95 0.03 0.24 0.04

40.6 0.1 58.7

0.02 0.06

51.63 1.14 16.87 17.68 3.47 0.05 0.16 2.67 1.4 0.1 4.83 0.07 0.27

45.76 0.09 49.17

2.69 1.65 0.04 0.02

29.6 0.13 2.71 36.58 3.51 0.89 0.08 0.23 1.02 0.05 24.75 0.01 dtm.

34.3 0.87 56.6

5.21 2.85 0.02 0.28

46.86 0.43 24.68 17.19 2.58 1.49 0.16 0.10 0.57 0.08 5.81 0.05 n.

51.68 0.11 39.79

3.8 8.2 0.09 1.9 2.7

42.84 0.2 31.95 15.38 1.89 1.7 0.42 0.25 0.92 4.12

0.07 0.05 0.09

n.m 0.09 0.3

n.m 0.01 0.42

0.07 0.02 0.05

0.58 0.017 0.02

n.m 0.4 0.05

0.18 0.01

0.18 0.09

0.095 0.11 0.09

0.049 0.67 0.47

0.11 2.1 0.7

0.14 0.1 0.45

















99.97 49 22.37 12.00 34.36 1.86

4.06a 94.31 46.37 32.46 n.a 32.46 n.a

n.a n.a 96.64


14.34 99.1 27.77 19.44 n.a 19.44 n.a


n.a 0.04a 99.82

n.a 0.04a 99.442

n.a n.a 101.72

39.62 n.a 39.62 n.a

34.42 n.a 34.42 n.a

41.09 n.a 41.09 n.a

40.39 n.a 40.39 n.a

30.17 n.a 30.17 n.a

0.06 0.03 0.14

99.47 38.7 15.40 11.93 27.33 1.29

43.83 3.03 1.81 0.03 0.06

5.81 0.03 0.07 0.2 0.07 0.12

98.81 43.83 n.a 34.19 34.19 n.a

4.63 10.26 0.196 0.179

12.09 0.016 0.269 0.06 0.102

16.5 0.1 0.3


100.05 43.99 20.46 10.47 30.93 1.95

99.76 35.7 13.93 11.23 25.16 1.24

11.84a 87.492 30.5 21.36 n.a 21.36 n.a

99.805 41.3 15.75 13.18 28.93 1.19

99.939 33.6 2.38 21.22 23.60 0.11

100.19 31.2 3.64 18.25 21.89 0.20

102.38 26.9 18.76 n.a 18.76 n.a





0.05 n.a 102.92

17.25 13.36 30.61 1.29

1.89 28.43 30.32 0.07

11.79 13.74 25.53 0.86

20.58 2.52 23.1 8.17

27.85 n.a 27.85 n.a

1 Iron-formation at base of Wittenoom Dolomite (Davy, 1975); 2 Brockman iron-formation, Dales Gorge Member (Ewers and Morris, 1981); 3 Brockman iron-formation, Joffre Member (Trendall and Pepper, 1977); 4 Weeli Wolli iron-formation; 5 Marra Mamba iron-formation (Ewers and Morris, 1981); 6 Lower part of Marra Mamba iron-formation (Klein and Gole, 1981); 7 Lower part of Marra Mamba iron-formation (Klein and Gole, 1981); 8 Lower part of Marra Manba iron-formation (Davy, 1975); 9 Average for S bands within the Dales Gorge Member (Trendall and Pepper, 1977); 10 Average for 13 macrobands (Trendall and Pepper, 1977); 11 Average S macroband composition, Dales Gorge Member (Ewers and Morris, 1981); 12 Average BIF macroband compositions, Dales Gorge Member (Ewers and Morris, 1981); 13 Brockman iron-formation, Dales Gorge Member, BIF2 macroband; 14 Brockman iron-formation, Dales Gorge Member, S13 macroband; 15 Brockman iron-formation, Dales Joffre Member; 16 Boolgeeda iron-Formation; 17 Kuruman-Penge iron-formation (Klein and Beukes, 1992); 18 Urucum iron-formation (Klein, 2004); 19 Brockman iron-formation (Klein and Beukes, 1992); 20 Sandur iron-formation (Gutzmer et al., 2008); 21 Carajas iron-formation (Figueiredo e Silva et al., 2008; Klein and Ladeira, 2002); 22 Superior-type BIF (McClung, 2006). a LOI < CO2, probably from the presence of magnetites that gain weight during the ignition.


Sedimentary Hosted Iron Ores

Table 5

Effect of deleterious elements on process performance of iron making


Tolerance threshold


Deleterious effect





Increases slag amount and limestone used, reduces shrinkage and blowhole formation (bad casting) Increases melt viscosity, impediment to tap off liquid slag



Quartz, primary silicate, clay minerals (kaolinite and smectite) Clay minerals (kaolinite and smectite), gibbsite, Al-substituted goethite, and hematite Apatite, P-rich goethite



BIF minerals, organic S

Cl Mg Ti

<500 ppm Varies <1%

BIF minerals Mg-Carbonates Ilmenite

Removal cost, increases steel brittleness, increases fluidity Increases SOx emission, requires higher MgO to partition S in slag Increases dioxine emission Decreases physical strength of pellet and sinter Decreases physical strength of pellet and sinter

Source: Clout JMF (1998) The effects of ore petrology on downstream processing performance. In: Mine to Mill. Melbourne: The Australasian Institute of Mining and Metallurgy; Clout, 2006.

Table 6 Some examples of deleterious elements in Chinese BIF concentrates Mine

Al2O3 (%)

S (%)

P (%)

Tangshan Jidong Shaogang Qianan Jinan Local Benxi Nanfen Tianjin Hanx Anshan Qidashan Anshan Gongchangling Taiyuan Ekou Taiyuan Local Wuhan Local Meishan Local Manshan Aoshan Manshan Dondshan Manshan Tao

0.50 – 0.83 0.51 0.50 0.71 1.29 0.77 2.53 1.39 – 1.51 3.07 0.89

0.048 0.06 0.112 0.018 0.205 0.59 0.52 0.146 0.128 0.259 0.58 0.16 0.55 0.33

0.022 0.022 – 0.016 – 0.03 0.02 0.023 0.04 0.019 0.36 0.54 0.84 0.014

Mount Whaleback deposits (Clout, 2002; Taylor et al., 2001). The microplaty hematite, made of a three-dimensional (3D) network of 10–300 mm size plates, is haphazardly oriented (Morris, 1980). According to Morris (1980) and Morris and Kneeshaw (2011), in Australia, martite-microplaty hematite ores are characterized by a preserved initial macro-, meso- and even microbanding of the original BIF (Figure 15). At a petrological scale, Morris and Kneeshaw (2011) argue that the only difference between the martite–goethite ore and the martitemicroplaty hematite is in the matrix. The martite mesobanding is identical in both types, but the mimetic goethitic texture found in the martite–goethite ore consists of secondary microplaty hematite (Figures 16 and 17). Importantly, Morris (2012) showed that secondary microplaty hematite occurs in many different environments, and not just in BIF-hosted iron ore deposits. In significant ore bodies underneath the weathering and supergene-enriched zones of the Mount Tom Price hematite deposit, Dalstra and Guedes (2004) and Thorne et al. (2004) have described chlorite-hematite-talc alteration in BIF and iron ore along the Southern Batter fault. In the Southern Ridge ore body and the North deposit at Tom Price, Thorne et al. (2004, 2008) identified a large (100 m) hydrothermal halo consisting

of a paragenetic succession started by the ore proper with martite-microplaty-apatite grading into a hematite-ankeritemagnetite zone and finally to magnetite-siderite-Fe-silicates. A similar paragenesis was detected beneath the weathering horizon in almost all large BIF iron ore deposits worldwide (e.g., Taylor et al., 2001; Thorne et al., 2004). Particular examples include the South African iron ore deposit of Thabazimbi, Pic de Fon in Guinea (Cope et al., 2008), and Itabira in Brazil (Rosiere and Rios, 2004). The Koolyanobbing deposits in the Yilgarn craton of Western Australia also show significant hydrothermal alteration zones compatible with hypogene hydrothermal processes. In the Carajas iron ore mine in Brazil, Figueiredo et al. (2008) and Lobato et al. (2008) define four types of hematite – martite, microplaty, anhedral-subhedral and, finally, euhedraltabular. Hard hematite ores are found at the Sishen mine in South Africa where Beukes et al. (2002a) described three types of ores, accounting for 63% (massive type), 20% (massive laminated) and 17% (conglomeratic). At Thabazimbi in South Africa, in addition to the hematitemartite ores, Lobato et al. (2008) described a hydrothermal paragenetic succession including dolomite-hematite, calcitehematite, and talc-hematite, with Beukes et al. (2002a) describing a carbonate protore deposited at around 150  C as a precursor of the hematite ore. Genetic models Three models that describe the genesis of martite-microplaty hematite ores have been proposed by many authors and include syngenetic, supergene mimetic-metamorphism, and hypogene models. These models have to be integrated into a 3D crustal depth framework; the (modern) hypogene model was applied to iron ore underneath the weathering and supergene-enriched zone. The most updated hypogene model actually uses the term ‘modified.’ The syngenetic model implies that chert-free, iron-rich ‘facies’ have been deposited concomitantly with the BIF. This model was first proposed by Chamberlin (1883), followed by Harder and Chamberlin (1915), Baldwin and Gross (1967), Schorscher (1982) and Hoefs et al. (1982). Recently, Lascelles (2002, 2007, 2008) has again supported the syngenetic model.

Table 7 1 Ag As Au B Ba Be Bi Cd Ce Cl Co Cr Cu Dy Er Eu F Ga Gd Ge Ho La Li Lu Mo Nb Nd Ni Pb Pr Rb Sb Sc Sm Sn Sr Tb Te Th Tm U V W Y Yb Zn Zr



Trace element chemistry (ppm) of Precambrian iron-formations 2




























<2 <10

1 3

1 14

0.2 6

<1 7

0.9 8

1.3 32

<1 4

<1 10

<5 3

1 14

0.9 8

1.3 32


2 12

25 113

15 90

28 88

18 69

49 263

14 50

47 140

11 30

25 113

18 69

49 263

1 0.5

3 0.3

3 0.3

<2 0.1

3 0.3

4 0.3

<2 0.1

<2 0.1

3 0.3

3 0.3

4 0.3

140 25 10 10

90 2 14 101

170 2 10 80

80 5 114 41

78 1 9 96

130 6 30 120

80 4 110 45

80 6 120 35

90 2 14 101

78 1 9 96

130 6 30 120









7 0.66 0.62 0.17

15.83 n.a. n.a. 0.33

n.a. n.a. n.a. 0.114

6.45 0.3 0.18 0.14

28.41 0.25 0.11 0.13

12 0.66 0.62 0.17


0.18 2.29 0.1

n.a. n.a. 6.47 0.33

n.a. n.a. 1.76 0.091

0.33 0.06 1.28 0.01

0.27 0.06 2.24 0.01

60 1

128 <1

110 1

179 3

120 <0.5

157 3.8

150 2

220 4









340 <5

128 <1

120 <0.5

157 3.8




0.55 0.37 2.16

4.84 8.53 51 n.a. n.a. 56.5

n.a. n.a. 20 n.a. n.a. 20

0.6 1.17 38.03 1.35 0.25 1.02

4.2 1.36 13.08 5.49 0.31 28.44

3 2.68 18 5 0.38 36

0.17 0.34

3.97 1.47

0.2 0.258

1.02 2.02

0.14 0.19

6.4 0.4

3.45 0.09

71.17 0.31

68 0.046

1.38 0.05

7.03 0.05

25 0.12

0.06 0.09 0.02 10

0.88 n.a. 0.17 n.a.

0.093 n.a. 0.08 n.a.

2.02 3.02 0.15 25.75

0.19 n.a. 2.76 34.21

3 0.09 1.7 26

9.35 0.58 25.4 2.5

26.67 2.02 15.1 53

n.a. 0.495 n.a. 20

2.48 0.18 3.75 1.55

5.84 0.07 24.13 11.05

25 4.3 26 38




















1.1 0.55 0.29









20 6









16 2

<1 4




<1 8



<1 10

<1 5

<5 40

20 <5

12 42

10 60

63 13

9 38

21 54

80 11


31 1

50 1

37 <1 <5

16 1.2

83 1


5 <1 <5

4 10

5 2

22 53

5 55

2 36

22 35

20 115




<1 <5




40 15

<5 <10

12 42

9 38

21 54

20 <1 <5

60 <1 5


31 1

16 1.2

83 1

2 30

2 45

22 53

22 35

20 115


7.32 1.63 29.34 2.25 0.38 2.15


0.984 0.974 0.156 <2


2 23 64 5 10 10






0.111 1.6

0.116 2.26




0.099 1.62






















0.658 14






























46 30

53 50

30 16






0.935 11 120




49 38























1 Kuruman-Penge iron-formation (Klein and Beukes, 1992); 2 Urucum iron-formation (Klein, 2004); 3 Brockman iron-formation (Klein and Beukes, 1992); 4 Sandur iron-formation (Gutzmer et al., 2008); 5 Carajas iron-formation (Figueiredo e Silva et al., 2008; Klein, 2004); 6 Superior type BIF (McClung, 2006); 7 Iron-formation at base of Wittenoom Dolomite; 8 Mount Sylvia iron-formation; 9 Brockman iron-formation, Dales Gorge Member; 10 Brockman iron-formation, Joffre Member (Trendall and Pepper, 1977); 11 Weeli Wolli iron-formation; 12 Brockman iron-formation, Dales Gorge Member, Average BIF macroband (Trendall and Pepper, 1977); 13 Brockman iron-formation, Dales Gorge Member, Average S-band (Trendall and Pepper, 1977); 14 Weeli Wolli iron-formation (Upper part); 15 Weeli Wolli iron-formation (Lower part); 16 Marra Mamba Iron iron-formation, lower part (Davy, 1975); 17 Brockman Iron-formation, Dales Gorge Member; 18 Brockman Iron-formation, Dales Gorge Member, Average BIF macroband (Trendall and Pepper, 1977); 19 Brockman iron-formation, Dales Gorge Member, Average S-band (Trendall and Pepper, 1977); 20 Pacific ocean water (Masuda and Ikeuchi, 1979); 21 Average East Pacific Rise crest sediment (Piper and Graef, 1974).; 22 Average shallow water manganese nodule (Piper and Graef, 1974).; 23 Average Archean Jasper-magnetite iron-formation, Temagami, Canada (Fryer, 1977); 24 Average Archean oxide facies iron formation, Mary River, Canada (Fryer, 1977); 25 Average Karelian oxide facies, Finland (Laajoki and Lavikainen, 1977); 26 Average Sokoman oxide facies, Labrador (Fryer, 1977); 27 Average Sokoman silicate-carbonate facies, Labrador (Fryer, 1977); 28 Average enriched Sokoman oxide facies, Labrador (Fryer, 1977); 29 Rapitan pisolitic-cherty iron-formation, Snake River, Canada (Fryer, 1977).


Sedimentary Hosted Iron Ores

Capanema iron mine


Weathering sequences 1






200 m

P1 Weathering sequence 1 2

P2S 3





Loose horizon Semi indurated macronodules Indurated macronodules Canga (C)


Friable yellow horizon Weathered hematite horizon (WH) Hematite A horizon (HA) brm Hematite B horizon (HB) Soft Itabirite horizon (SI) Itabirite (It) or BIF Weathered dolerite

Figure 11 Weathering profile of the Capanema mine. Modified from Ramanaidou ER (2009) Genesis of lateritic iron ore from banded iron-formation in the Capanema mine (Minas Gerais, Brazil). Australian Journal of Earth Sciences 56: 605–620.


Sedimentary Hosted Iron Ores


Brockman iron formation Marra Mamba Iron formation M-MplH zone Mines Brockman IF and Prospects MMIF, respectively

Pannawonica Newman Carnarvon


22° S

Wittenoom Barrett-Lennard DID


Brockman Syncline

Turner Syncline

Wyloo Dome

Cloud Break

Tom Price

Hardey Syncline


×Newman 0


50 km

Turee Creek Syncline 118° E

120° E

Figure 12 Location of the martite–goethite and martite microplaty hematite ores in the Hamersley Province (WA). Modified from Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58: 417–451.

Figure 13 Dales Gorge Section 6 martite–goethite iron ores in the Hamersley Province (WA).

The supergene mimetic-metamorphism model first requires the generation of a martite–goethite supergene ore (Morris, 1980, 1985; Morris and Fletcher, 1987) and subsequently, burial induced metamorphism. The martite-microplaty hematite ore genesis follows two steps (1) genesis of a martite– goethite deposit at around 2 Ga followed by (2) low-grade regional metamorphism with temperatures of around 80–100  C that transforms the goethite matrix into

microplates of hematite. Hence, the difference between martite–goethite supergene ores and martite-microplaty hematite ores is seen only in the matrix (Morris, 1980). Part of the matrix goethite was not transformed into microplaty hematite but was nevertheless changed into a recrystallized goethite. Subsequently, the latter was at times partially or totally dissolved (Morris and Kneeshaw, 2011). Hypogene models have been proposed in the past by Dorr (1965), Griffin (1980), Gross (1965), Guild (1953), Kneeshaw (1975), Pomerene (1964), and Sims (1973), to name a few. These models fell out of favor in the 80s, but hypogene genetic models were again brought to life in the 90s. The recent literature on hypogene models has focused on unweathered ore whereas the original hypogene models often investigated weathered ore. All hypogene models have specific timing and fluid characteristics. For example, Li et al. (1993), Martin et al. (1998), and Powell et al. (1999) proposed a hydrothermal model, which stated that hot fluids generated by extensional deformation leached out the silica leaving behind the iron oxides. Barley et al. (1999), Hageman et al. (1999), and Thorne et al. (2004) introduced a hypogene-supergene genetic model, based mainly on fluid inclusions and stable isotope analysis, for the martite-microplaty hematite ore at Hamersley’s Mount Tom Price. The primary hypogene alteration brines (200  C) caused the replacement of the fresh BIF by a magnetite-siderite-Fe silicate alteration that also involved dissolution of chert. The second hydrothermal brines (350  C) produced a hematite-ankerite-magnetite paragenesis and,

Sedimentary Hosted Iron Ores

Electrochemical cell

1. Simplified Mt. Whaleback deposit

Cathode 4e− + O2 + 2H2O


e− BIF



Projected genesis surface



Ionic exchange via groundwater


Initial fluid BIF JOF


Present ground level



Electrical conductor (magnetite)

lt z o

JOF Ore M-mplH

Anode Fe2+ Fe3+ + e−



100 m 100 m

After Morris et al. (1980)

Transfer of Fe to anode

Fe3+ + organic

Fe2+ H 2O

2. Residual friable silica from leached BIF

Leached Fe2+ BIF Ore

(b) H4SiO4 to drainage

3. Quartz → goethite (Morris and Fletcher, 1987)

Transfer of Fe to anode Erosion of loose silica H 2O

Seasonal changes drive Redox cycling

Leached BIF

Reducing Ore


Hydrous ferrous silicate gel (Å)


Fe2+ (c)


Transfer of Fe to anode

Fe2+ Goethite (porous)

H 2O

Fe3+ + H 2O

Amorphous* silica





* Silica solubility increases > 10 x


Final martite-goethite ore body now subject to weathering, erosion and hardcap formation

Formation of M-mplH ore Mild regional ‘metamophism’ thermal gradient (80–100 °C) Erosion and leaching M-mplH + g M-mplH

Hardcap BIF

Hardcap BIF Ore

Leaching M-mplH ore



Figure 14 Electrochemical model for the formation of supergene deep-seated mimetic iron ores. Modified from Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58: 417–451.

Sedimentary Hosted Iron Ores

Mt Whaleback ~1.8 Gt M-mplH ore



Mt Tom Price ~0.9 Gt M-mplH ore



Nearly every textural feature of BIF is preserved in these M-mplH ores







H mplH + voids

Chert/hematite varves

10 mm

0.5 mm

Figure 15 Tom Price martite-microplaty iron ore. All the textural features are preserved from the macrobanding to the microbanding scale. (a, b) Wellpreserved DGM macrobanding at Whaleback and Tom Price. (c) Fresh BIF. (d–f) Mesobanding and ‘microbanding’ in ore. The only real textural differences between the M-G and M-mplH ores are in the martite to ‘matrix’ ratio (see text) and in the matrix mineralogy. Modified from Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58: 417–451.

ultimately, the microplaty hematite. The third supergene (in a weathering sense) stage dissolved the apatite. Taylor et al. (2001) proposed a model based on an integrated structural-hydrothermal alteration and fluid chemistry study. The hydrothermal alteration mineralogy producing a magnetite-siderite-apatite BIF protore was followed by a heated meteoric water phase that transformed the siderite into microplaty hematite with ankerite and magnetite into martite. Guedes et al. (2002) and Dalstra and Guedes (2004) suggested that all the microplaty hematites worldwide were formed by a unique model that consists of high-temperature brines dissolving the silica of the BIF and precipitating carbonates to produce a magnetite-carbonate-apatite protore, and subsequent ancient meteoric water (penetrating the crust during the mineralization process in the Archean or Proterozoic) events. The temperature of the brines may have varied from 450  C at Krivoy Rog (Ukraine), to 350  C at Tom Price (Australia), and to less than 300  C in Carajas (Brazil). Figueiredo e Silva et al. (2008) also suggested a hypogene origin for Carajas (Brazil), Mount Tom Price (Australia), and Thabazimbi (South Africa). For Carajas, Figueiredo e Silva et al. (2008) and Lobato et al. (2008) defined three alteration zones: (1) distal with magnetite-calcite-quartz-pyrite, (2) intermediate with martite-microplaty hematite-quartz, and (3) proximal with hematite  carbonate  quartz. For Mount Tom Price, a model similar to that of Hageman et al. (1999) was proposed. At Thabazimbi, no hydrothermal magnetite was found and chert in ore zones was either replaced by microplaty hematite or leached.

Using fluid inclusion information, Rosiere and Rios (2004) described three types of hematite associated with tectonically favorable sites in Brazil. In the Carajas iron ore mine in Brazil, Figueiredo e Silva et al. (2008) also described hard hematite ores in the supergene-enriched zone and, based on a combination of petrology and oxygen, carbon, and sulfur isotope analysis of fluid inclusions, they link the martite, microcrystalline, and microplaty hematite crystals to hydrothermal alteration. Morris and Kneeshaw (2011) have reviewed the latest syngenetic and hypogene models systematically and highlighted the difficulties with the various models. A schematic summary of the conflicting models is illustrated in Figures 16 and 17. Weathering of Martite–Goethite and Hematite BIF-hosted Iron Ores Secondary weathering processes affect all preexisting iron ore deposits around the world. The effects of weathering processes include textural obliteration, dissolution, and precipitation processes. Many examples have been well documented in Australia where the martite-goethite and martite-microplaty hematite ores have been affected by weathering (Angerer and Hagemann, 2010; Clout and Simonson, 2005; Clout, 2002; Freyssinet et al., 2005; Harmsworth et al., 1990; Morris, 1985; Ramanaidou and Morris, 2010). A classical, three-tiered profile consisting, from bottom to top, of the (1) iron ore proper, (2) a ‘hydrated zone’ with a thickness varying from a few meters up to 75 m and (3) a 1–2 m hematite-rich carapace that


Sedimentary Hosted Iron Ores

Alteration stages in hypogene - meteoric ore genesis (Taylor et al., 2001) Stage 1 (hypogene): chert solution (~250 ⬚C) > magnetite-siderite-stilpnomelane-pyrite-apatite. Stage 2 (deep warm meteoric circulation): magnetite > martite; siderite > mplH + ankerite. Stage 3 (leaching stage): carbonate leached > porous M-mplH + apatite. Stage 4 (weathering): apatite + carbonate leached > M-mplH ore. (a)

Mt Tom Price Southern Ridge Deposit Section 13692 E

10 000 mN

11 000 mN

Depth of weathering Southern Batter Fault


Centre Pit Stages 1+2+3+4




DGM “Bruno’s Band”

DGM Stages

HG M-mplH ore - high grade - low P


M-mplH ore - high grade - high P Stage 1

Magnetite - High P

WD (b)


Southern Batter Fault Hydrothermal solution (~250 ⬚C)



JOF Joffre Member

Whaleback Shale Member DGM Dales Gorge Member

Footwall Zone Mt Mcrae Shale Mt Sylvia Formation WD


Wittenoom Formation Dolerite

Modified from Taylor et al. (2001)


200 m Proximal M-mplH + Mt + apatite Intermediate H + Mt + ankerite + chlorite Distal Mt + siderite + stilpnomelane

Modified from Thorne et al. (2005)

Figure 16 Hypogene genetic modelling. Modified from Taylor D, Dalstra HJ, Harding AE, Broadbent GC and Barley ME (2001) Genesis of high-grade hematite orebodies of the Hamersley Province, Western Australia. Economic Geology 96, 837-873; Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58, 417–451.

has the textural feature of the primary iron ore (Harmsworth et al., 1990; Morris, 1985). The hydrated zone is dominated by goethite as either vitreous or ochreous types that destroy the original BIF textures. The end-result of this process is a textureless, goethite-rich duricrust. Along the Hamersley scarp, in the Marillana and Koodaiderie martite–goethite ores, the weathering mantle is as thick as 75 m, whereas in the martitemicroplaty hematite ores of Tom Price and Whaleback it averages 30 m (Morris, 1985). Clout (2002) showed that secondary processes affecting an existing iron ore include the dissolution of martite and microplaty hematite and the precipitation of vitreous goethite in the hydration zone. In the upper hydration zone, the development of cavities by dissolution and silicification, and precipitation of manganese minerals were widespread just below the carapace. Clout (2002) also described

the development of dehydration zones with the precipitation of colloform hydrohematite, which generally alternates with low-temperature colloform goethite and hematite. This is in agreement with similarly alternating goethite and hematite found in Brazil (Ramanaidou, 1989) and in the Yilgarn (Anand and Paine, 2002). The hydrated zone on martite– goethite iron ores can vary in thickness from 10 to 25 m and is generally better developed than that on martite-microplaty hematite ores, which is <10-m thick (Thorne et al., 2008). Angerer and Hagemann (2010) have described the occurrence of weathering processes highlighted by strong goethitization and destruction of primary textures at the Western Australian iron martite–goethite mine of Koolyanobbing. Weathering of BIF-hosted Indian iron ores has been well documented (Banerji, 1977; Dunn, 1937; Krishnan, 1954).

Sedimentary Hosted Iron Ores


Banded iron-formation (BIF) Magnetite

[± Hematite]





[± Hematite]





Carbonates Silicates


Residual ores


Fe hydroxyoxides +H2O Goethite


Martite-goethite (M-G) ores

Exposure - erosion, dehydration, groundwater leaching and reprecipitation of goethite

Regional ‘metamorphism’ ~ 80–100 °C (G → mplH + g) M-mplH + g ores

Leach g

M-mplH ores Main ores below hardcap

Hardcap Dehydration


Dissolution and reprecipitation

Hydrated zone Mt M oG G, g mplH

BIF Post ore changes

BIF texture preserved, variably leached, hard to friable, or goethite infilled




M-G and M-oG ores


Coarsening + Mt


Magnetite Martite (hematite pseudomorphs after Mt) Ochreous goethite Major, minor, goethite Microplaty with other secondary hematite

Processes Products

Figure 17 Schematic of supergene and supergene-metamorphic genetic modeling for the major BIF-hosted iron ores of the world. Modified from Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58: 417–451.

Weathering profiles were recorded by Morris and Trendall (1988) at the Bailadilla mines in Chhatisgarh where a thin, highly weathered carapace blankets a thick goethitic horizon developed on heavily leached hematite or ‘blue dust.’ In the Joda-Barbil region of Orissa in India, weathering processes have strongly affected primary iron ores and are manifested by the precipitation of many generations of goethite including vitreous, ochreous, and colloform types. Mukhopadhyay et al. (2008) commented that all Indian iron deposits were subjected to chemical weathering that resulted in (1) an enrichment of BIF and the production of blue dust ores and (2) an intense goethitization of the BIF-hosted iron ores. Brazilian examples are also reported by Guild (1953), Dorr (1965); Melfi et al. (1988), Ramanaidou (1989), Rosiere and Chemal (1991), Chicarino Varaja˜o (1994), Ramanaidou et al. (1996), Pires et al. (2005), Beukes et al. (2002a), and Spier et al. (2003), with an extremely well-developed weathering profile of up to 200 m depth. According to Spier et al. (2003), 15% of the total reserves of the Brazilian Quadrila´tero Ferrı´fero consists of hard hematite occurring in the Caue´ ironformation, and the remaining 85% is weathered and highly leached, and is locally called ‘soft ore’ or ‘blue dust.’

In West Africa and, in particular, in Guinea, Cope et al. (2008) described the development of a 1-m-thick, surficial goethitic carapace capping a 10–30 m weathered horizon where the primary mineralogy is, at times, obliterated by secondary precipitation of goethite and clay minerals. The weathering process affecting the primary iron ores can obliterate the primary features of the original ore and can add confusion to the comprehension of primary iron ore genesis. Supergene iron ores – General misconceptions regarding the term supergene led Ramanaidou and Morris (2010) to redefine the meaning of ‘supergene iron ores’ by returning to the original definition that included all ‘from above’ processes, specifically to include weathering. The majority of iron ore geologists equate the term ‘supergene iron ore’ with lateritic weathered iron ore. Supergene is defined as being formed from above, in contrast to hypogene, meaning genesis from below. Geochemistry of Martite–Goethite and Martite-microplaty Hematite Iron Ores Whole rock geochemical data for hematite-rich iron ores can be extremely useful for establishing the conditions of


Sedimentary Hosted Iron Ores

formation as well as for providing unique fingerprints for these ores (Gutzmer et al., 2008). Major and minor elements As expected, all the samples have high Fe content (>93 wt% Fe2O3), followed by silica and alumina with values ranging from 0.5 to 1.5 wt% (Table 8). MgO and CaO are generally very low (<0.06 wt%) except for the Maremane and Nauga East deposits. K is higher in Maremane-a, Urucum, and Thabazimbi. Mn is slightly elevated in Thabazimbi. P is very low in Mount Tom Price and ranges between 0.07 and 0.08 wt % at Sandur and Maremane-a. All other deposits show high levels of phosphorus. The compositional ratios between the BIF and the associated iron ores (Gutzmer et al., 2008) demonstrate the intensity of the changes that occurred during enrichment. Silica, Mg, and Ca show the greatest level of depletion indicating almost complete dissolution of quartz, silicates, and carbonates. It is important to note that silica is the major component in BIF and that this drastic reduction required a very powerful mechanism that replaced around 70% of the initial rock. Although alumina is generally extremely enriched in the rock to up to 13 times, the absolute alumina content is still very low (1.41%). For iron, the enrichment is merely double. However, volumetrically, Fe occupied 30% in the fresh BIF and, hence, a huge amount of iron was brought to the system. As for silica, the mechanism responsible for this big increase must have operated at a phenomenal scale. The most deleterious elements in Australian and Brazilian martite–goethite and hematite-rich iron ores are silica, alumina, and phosphorus. Generally, silica and alumina range, respectively, between 2.5–4% and 1–4%. Brazilian ores can be very low in alumina (<1%). Low phosphorus iron ores (<0.08%) are saleable without penalty; some Vale ores have extremely low P values (<0.03%). LOI is an indication of the amount of goethite in the ore: the higher the LOI, the higher the goethite content in martite–goethite iron ores such as West Angelas, MAC, and Cloud Break. Fines generally have a higher amount of goethite. The problems of dealing with high-P iron ore in general and the strategies used to mitigate elevated P contents have relied mainly on blending with low-P iron ore (e.g., Wells and Ramanaidou, 2011). However, a better long-term strategy is through beneficiation of the ore (e.g., Clout and Simonson, 2005) where the application or development of new or novel processing techniques may be effective in removing P and other deleterious elements such as Al and Si (e.g., Wells and Ramanaidou, 2011). For example, high temperature, 1000–1200  C roasting of P-bearing iron ore followed by leaching has been shown to significantly reduce the P content (e.g., Gooden et al., 1974) as cited by Dukino et al. (2000). However, such a high-temperature, energy-intensive process is currently not economically viable. A recently developed low-temperature, water/caustic leaching method has shown considerable success in reducing not only the P content, but also amounts of Si and Al in the initial ore (Edwards et al., 2011; Fischer-White et al., 2009). More recent studies have focused on a bioleaching/bioflotation approach that may potentially remove P from iron ore (Dwyer et al., 2011). While these studies have focused mainly on the high-P iron formation BIF-hosted ores of the Pilbara region, recent studies have reported the potential of a similar, low-temperature reductive

roasting technique applied to the Lisakovsk CID (Ionkov et al., 2011). The development of such techniques and their application on a wider scale may aid the economic viability of the comparatively high-P marine ironstones (and also ‘Hi-P Brockman ores’). Rare earth elements According to Gutzmer et al. (2008), there is a clear enrichment of light rare earths (Table 8) in all the supergene iron ores such as Maremane and Urucum, and the fractionation of light rare earths could be used as a fingerprint for differentiating supergene and hypogene iron ores. However, a larger dataset covering more deposits is required to provide genuine trends.

Detrital Iron Deposits

Detrital iron deposits or DID have been defined by Morris (1994) in the Hamersley Province of Western Australia where they represent a small resource of ca 500 Mt with only minimal production (Morris and Ramanaidou, 2007). For example, the Brockman Syncline No. 2 Detritals Deposit was mined from 1992 to 1998 (Butt et al., 2001). The lump extracted from this deposit was blended with lower lump iron ores to produce a saleable product. West Australian DID are generally lumpy, hematite-rich, and low in phosphorus (Morris, 1994). DID also exist in Brazil, for example, in the Rio Pardo area in Minas Gerais. Two DID mines were described from Chamakpur and Inganharan in Northern Orissa, India, by Mohapatra et al. (2008). In the Sishen iron ore mine in South Africa, Beukes et al. (2002b) described a detrital, karstic conglomeratic type resulting from the breakdown of the laminated ore that accounted for 17% of the total reserves of the 1.7 Gt deposit. The Barrett-Lennard deposit north of the Wyloo dome in Western Australia represents a 2 Mt DID basinal deposit of conglomeratic, martite-microplaty hematite fragments in the Proterozoic Mt formation, and other sporadic fragments and minor concentrations have been noted in this unit (Dalstra and Guedes, 2004; MacLeod, 1966; Morris, 1985). Lateritic weathering of the bedded iron ores in the Hamersley Ranges results in a three-tiered alteration sequence: a surface 2 m hematite-rich low phosphorus carapace, overlying the hydrated zone comprising mainly alumininous goethite that grades into an altered ore sequence in which goethite leaching and partial redeposition are prominent. Physical weathering and subsequent breakdown have affected the surface horizons and DID typically formed by the accumulation of colluvial products along the valley margins and into the valleys (Freyssinet et al., 2005; Mohapatra et al., 2008; Morris, 1994). According to Berkman and Mackenzie (1998), DID occur as shallow blankets of outwash scree deposited in structural depressions adjacent to iron formation escarpments. A typical cross-section of a DID includes at the top, immature detritals consisting of unsorted BIF clasts overlying mature detritals mainly composed of hematite-rich, subrounded clasts with pisolitic components increasing into the

Table 8

Geochemistry of martite–goethite and martite-microplaty hematite iron ores from selected iron ores


Maremane-A (Gutzmer et al., 2008)

Maremane-B (Gutzmer et al., 2008)

Urucum (Gutzmer et al., 2008)

Mount Tom Pricea (Ridley, 1999)

Thabazimbi (Gutzmer et al., 2008)

Carajas (Figueiredo e Silva et al., 2008)

Sandur (Gutzmer et al., 2008)

Nauga East (Gutzmer et al., 2008)

SiO2(%) TiO2(%) Al2O3(%) Fe2O3(%) MnO(%) MgO(%) CaO(%) Na2O(%) K2O(%) P2O5(%) La(ppm) Ce(ppm) Pr(ppm) Nd(ppm) Sm(ppm) Eu(ppm) Gd(ppm) Tb(ppm) Dy(ppm) Y(ppm) Ho(ppm) Er(ppm) Tm(ppm) Yb(ppm) Lu(ppm)

0.04 2 10.91 2.42 0.13 0.05 0.04 1.33 4.33 0.78 6.81 7.67 10.39 10.25 9.38 4.41 4.29 3 1.55 0.44 0.78 0.61 0.78 0.76 0.7

0.04 2.5 12.82 2.42 0.13 0.05 0.04 1.33 2 1.33 8.12 8.17 14.42 14.23 11.97 5.12 4.85 3.33 1.79 0.5 0.89 0.76 0.78 0.93 0.8

0.05 0.89 0.9 1.7 0.07 0.01 0.02 0.07 5.1 0.57 3.46 3.15

0.06 4 10.56 1.95 0.4 0.01 0.02 1.25 0.5 0.23 2.25 2.49 n.a n.a 2.83 2.46 n.a 3.91 n.a n.a n.a n.a n.a 1.74 1.65

0.04 2.75 4.94 2.34 0.83 0.002 0.01 0.33 6.67 1.33 0.99 0.99 1.18 1.15 1.26 0.82 1.2 1 0.71 0.61 0.78 0.74 0.67 0.81 0.7

0.01 n.a 7.22 1.65 n.a n.a n.a n.a n.a

0.04 1.5 12 1.63 1 0.08 0.4 1 2 2 2.61 3.32 2.96 3.3 0.48 2.86 3.89 3.2 2.63 1.96 2.5 2.06 0.06 1.67 1.67

0.01 0.5 3.27 2.45 0.1 0 0.06 n.a 0.67 1.78 0.77 1.05 1.2 1.29 1.76 1.41 1.42 1.56 1.36 0.82 1.06 0.94 0.96 1.16 1.1

1.95 0.04 1.2 96.4 0.04 0.24 0.2 0.04 0.13 0.07 15.6 24.7 3.95 16.7 3.19 0.75 2.36 0.27 1.02 4.11 0.14 0.38 0.07 0.44 0.07

1.89 0.05 1.41 96.2 0.04 0.24 0.21 0.04 0.06 0.12 18.6 26.3 5.48 23.2 4.07 0.87 2.67 0.3 1.18 4.72 0.16 0.47 0.07 0.54 0.08

2.16 2.08 1.71 0.99 0.52

0.24 0.81 0.78

1.76 0.08 0.78 96.2 0.03 0.01 0.05 0.02 0.1 0.17 22.4 34.9 4.89 18.4 3.06 0.57 2.07 0.31 2 13.8 0.47 1.4 0.24 1.64 0.26

2.55 0.04 0.95 95.7 0.02 0.03 0.03 0.05 0.01 0.05 3.96 5.57 0.7 3.03 0.73 0.28 0.99 0.18 1.16 9.53 0.28 0.85 0.17 0.86 0.15

2.03 0.06 0.54 93.3 0.25 0.01 0.06 0.01 0.2 0.12 2.27 3.2 0.45 1.87 0.43 0.14 0.66 0.09 0.47 5.73 0.14 0.46 0.06 0.47 0.07

2.64 3.95 3.52 3.63 9.32 6.46 12 12.6 13 3.05 13.83 14.36 n.a 20.57 17

0.61 n.a 0.65 95.3 n.a n.a n.a n.a n.a n.a 5.92 10 1.09 4.93 1.77 0.84 3.24 0.63 3.25 17.8 0.83 1.58 0.25 1.44 0.17

1.53 0.03 1.2 95.6 0.02 0.02 0.02 0.01 0.04 0.08 3.34 5.74 0.74 3.86 0.97 0.4 1.02 0.16 0.79 4.87 0.15 0.36 0.17 0.3 0.05

0.48 0.01 0.36 97.5 0.03 0.02 0.36 n.d 0.02 0.16 0.77 3.37 0.46 2.1 0.6 0.24 0.78 0.14 0.9 7.67 0.19 0.58 0.09 0.67 0.11

n.a., not analyzed; n.d., not detected. Geochemistry of martite-hematite iron ores from selected iron ore deposits. Average of ratios obtained between individual martite-hematite ore samples and the average protolith composition (italics). a Detection limits recorded are applicable to data sets from Mount Tom Price as well as major elements from the Maremane dome (A ¼ XRF data) and all other data sets (B ¼ ACME Laboratories) Modified from Gutzmer J, Chisonga BC, Beukes NJ, and Mukhopadhyay J (2008) The geochemistry of banded iron formation-hosted high-grade hematite-martite iron ores. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 157–183. Littleton, CO: Society of Economic Geologists.


Sedimentary Hosted Iron Ores

valleys. DID may be unconsolidated or are cemented as a hematite conglomerate. Both types may co-exist in the same deposit. DID fragments range from coarse sand to boulder-sized material. The fragments are generally thinly coated by goethite accretionary layers (cortex), often dehydrated to hematite. The smaller the particles, the larger the ratio of coating compared to primary nucleus. Magnetic minerals include residual kenomagnetite-maghemite after magnetite, or as a result of the reductive transformation of goethite during wild fires.

Number of deposits 5 10 15



−100 Cretaceous



Ooidal Ironstones

−200 Triassic


The enigmatic nature of ooidal ironstones – their global distribution, occurrence throughout the geological time scale, distinctive textural and mineralogical characteristics, and apparent paradox of a common ferriferous silicate mineralogy suggestive of formation under simultaneously oxic and anoxic conditions (Kimberley, 1994; Maynard, 1986) – has stimulated and driven research for nearly 170 years (Petra´nek and Van Houten, 1997). Despite their long history of investigation, a working definition for what constitutes an ironstone was agreed upon only in the latter half of the twentieth century, following the work of the International Geological Correlation Program-Project 277 ‘Phanerozoic Ooidal Ironstones (POI),’ 1988–92 (IGCP-277) (Taylor, 2005). In this chapter, the convention of Petra´nek and Van Houten (1997) is followed; they defined ooidal ironstones “. . .as distinctive non-cherty, sandy clayey siliclastic or siliclastic-carbonate sedimentary rock with more than 5% per cent of ferruginous ooids and more than 15 per cent of iron.” This definition includes both marine and terrestrial ironstones, but it is not intended to include granular iron-formations. This definition includes or incorporates the paleoenvironmental classification scheme as proposed by Kimberley (1981) where marine ooidal ironstones were defined as a sandy, clayey, and oolitic shallow-inland-sea iron formation, abbreviated to SCOS-IF. However, since the work of the IGCP-277, this scheme is not favored. In addition, the use of the historical terms ‘Minette’ and ‘Clinton’ to describe an ironstone facies distinction is now discouraged (Young, 1989), and the terms have fallen into disuse (Taylor, 2005). Ooidal ironstones occur throughout the geological time scale. Proterozoic ooidal ironstones are known and include the 2200 Ma occurrences in South Africa (Lower Proterozoic Timeball Hill Formation – Schweigart, 1965), in Western Australia (Turee Creek Formation – Petra´nek and Van Houten, 1997), and in Russia (Kuetsja¨rvi and Kolasjoki Sedimentary Formations, Kola Peninsula – Bekker et al., 2010). Later occurrences at 1400 Ma were reported in northern Australia, such as Roper Bar in the Northern Territory and Constance Range in Queensland, (Canavan, 1965; Edwards, 1958; Ferenczi, 2001; Trendall, 1973). Ooidal ironstones also occur in northern China (Sinian System), in North America (Gunflint, Mesabi Range), and in the Sokoman (Petra´nek and Van Houten, 1997). However, the vast majority of ooidal ironstones accumulated throughout the Phanerozoic Eon from the midCambrian to Recent (Petra´nek and Van Houten, 1997).

Permian −300 Carboniferous

Devonian −400 Silurian

Ordovician −500 Cambrian −600 Figure 18 Stratigraphic distribution of POI. From Van Houten FB and Bhattacharyya DP (1982) Phanerozoic oolitic ironstones – Geologic record and facies model. Annual Review of Earth and Planetary Sciences 10: 441–457. Reproduced with permission of Annual Review Inc.

An evaluation of the stratigraphic distribution of ooidal ironstones highlights two main episodes of formation (Figure 18): from the Ordovician to late Devonian (a period of 170 my) and from the late Triassic to the middle Cenozoic, another 170 million year period (Van Houten and Bhattacharyya, 1982; Young, 1989). This is discussed in a little more detail in Section, which deals with the genesis of marine ironstones. Compared to the voluminous literature dealing with the genesis and occurrence of marine ooidal ironstones (Bayer, 1989; Kimberley, 1978, 1981, 1989a,b, 1994; Maynard, 1986; Mu¨cke and Farshad, 2005; Petra´nek and Van Houten, 1997; Stanton, 1972; Taylor, 2005; Van Houten, 1992; Van Houten and Arthur, 1989; Van Houten and Bhattacharyya, 1982; Young, 1989; Zitzmann and Neumann-Redlin, 1977), there is very little research reported on terrestrial ooidal ironstones. Until recently, terrestrial ironstones have at best received only cursory mention by most ironstone researchers. This is somewhat surprising considering some early recommendations calling for a comparison of terrestrial and marine examples to better elucidate their origin

Sedimentary Hosted Iron Ores

(Taylor, 1992). Yet, the Kazakhstan ironstones were known very early on (Yanitzkii, 1960) while the Pilbara ironstones have been known since the early 1960s (e.g., see early references such as MacLeod, 1966, in Ramanaidou et al., 2003), with mining of these deposits (Robe River) beginning in the early 1970s (Ramanaidou et al., 2003). It is only in the last 10 years or so, that coupled with the decline in reserves of high grade, BIF-hosted iron ore, have these deposits come under intense study, mainly by Australian researchers (Dalstra et al., 2010; Danisˇ´ık et al., 2011; Haest et al., 2011, 2012a, b; Heim et al., 2006; Kepert et al., 2010; MacPhail and Stone, 2004; Morris and Ramanaidou, 2007; Ramanaidou and Morris, 2010; Ramanaidou et al., 2003; Storkey et al., 2010). Terrestrial ironstones, particularly the fluvial ooidal ironstones of the Pilbara region, Western Australia, now comprise an economically significant resource (e.g., McGregor et al., 2010), with potential for the discovery of additional deposits as highlighted by the recent delineation of other CID in the region (Dalstra et al., 2010; Kepert et al., 2010). As discussed in the following section that deals with bog iron ores, similar fresh-water ooidal or ‘lake ore’ deposits are known, for example, in the African great lakes, such as Lake Chad (Lemoalle and Dupont, 1973), and may be regarded as modern day equivalents of marine ironstones (Taylor, 2005). However, these deposits are not economically viable and are not discussed further in relation to ironstones. Examples of other terrestrial ironstones are known and include occurrences of continental (fluviatile and lucastrine) deposits in northwestern Nigeria (Kogbe, 1978) and goethitic ooids in the Lower Cretaceous Wealden sediments in South East England (Taylor, 1992). Recently, numerous discrete Fe-ooidal beds associated with abundant fresh-water mollusk fossil debris, as part of the dominantly lucastrine, transitional-alluvial 2.5–3.5 Ma (mid-Miocene) Nyakabino Formation, were described in the synrift-sediment exposures in the KisegiNyabusosi area, Albert Rift, Uganda (Roller et al., 2010). However, it is the fluvial ooidal deposits of the Pilbara region of Western Australia and the northern Turgai and Aral’sk districts of Kazakhstan that are perhaps the best known and most significant examples. Furthermore, in a review of 39 Cenozoic ooidal ironstone deposits, Van Houten (1992) concluded that ironstones of marine origin were economically unimportant, and that only the West Australian and Kazakhstan fluviatile ironstones were economically significant. The Kazakhstan ironstones have been broadly described by Zitzmann (1977) and Ramanaidou et al. (2003), who provide a summary of earlier, predominantly Russian studies of these deposits of fluviatile, deltaic-lucastrine, and lucastrine origin. An additional overview from more recent publications (Golubovskaya, 2003; Kokal et al., 2008) is provided later. However, continental/fluvial ooidal ironstones are well known in the Pilbara region of Australia as CID, which is now the commonly accepted term (Ramanaidou and Morris, 2010; Ramanaidou et al., 2003). The Australian CID yield a high quality iron ore with resources of tens of Gt. For example, the Robe and Marillana Formations alone account for nearly 9000 Mt. The potential for future discoveries in the region is highlighted by the recent discovery of the Solomon East CID with a current ironstone resource estimate of 1.5 Gt at 55.8 wt% Fe, 2.8 wt% Al2O3, 6.4 wt% SiO2, 0.07 wt% P, and 10.4 wt%


LOI (Kepert et al., 2010), and the Caliwingina deposit with a current CID resource estimate of 1.6 Gt at 57.7 wt% Fe and 0.1 wt% P (Dalstra et al., 2010). Mining of these deposits is expected to commence within the next few years, with, for example, mining of the Solomon East CID expected in 2013 (Kepert et al., 2010). Though comparatively smaller than other WA CID, mining operations recently began at the Nullagine Iron Ore Project, which represents a new CID mine in the WA Pilbara (Gale et al., 2012). Located  140 km north of Newman, the Nullagine mine has a current total resource of 103 Mt at 54.1 wt% Fe, 3.58 wt% Al2O3, 4.29 wt% SiO2, 0.014 wt% S, 0.018 wt% P, and 12.4 wt% LOI (Gale et al., 2012). As stated by Ramanaidou et al. (1991) and cited by McGregor et al. (2010): “the channel iron deposits and the marine ironstone ooids are distinctly different in character” but “there are many analogies and the accretion mechanisms may be similar.” For example, Kimberley (1978, p222) noted the presence of the well-preserved, cellular structure of wood fragments replaced by hematite and siderite in the Paz de Rio ironstone, Columbia. ‘Preserved’ wood fragments were also noted in the lower ironstones of the Bida area, middle Niger Valley, Nigeria (Adeleye, 1973). Sedimentary features indicative of deposition in a fluviatile system, such as channel scours and fills, cross-bedding, reworked horizons, slumped strata, and graded bedding have been reported in the Yandi CID (Stone et al., 2002) and are comparable to the sedimentary textures observed in marine ironstones. Hence, CID of alluvial origin (Ramanaidou et al., 2003) may be included with marine ironstones as part of a larger group termed Phanerozoic ooidal ironstones, POI. The aim of this chapter is, in part, to demonstrate and describe the general characteristics of POI of both marine and terrestrial origin, using representative examples of more wellknown deposits that have either been exploited already (e.g., Minette – marine ironstone) or are currently being mined (e.g., Robe, Yandi – terrestrial ironstone), and to compare/contrast the mineralogical and compositional characteristics of the iron ores. One important goal is to highlight the significance and economic importance of terrestrial ironstones or CID, as they will be referred to hereafter, which have been seriously overlooked in past investigations. In the following sections, an overview of the genesis of each ooidal ironstone type is first presented followed by an evaluation of the mineralogical and compositional characteristics of marine and terrestrial iron ore.

POI Genesis Marine ironstones Marine ooidal ironstones are characterized by the occurrence of ooids and/or pisoids as thin units or beds, typically <1 m thick (Taylor, 2005), as part of an (idealized) upward coarsening cyclothem, from marine basal black shales/mudstones, through siltstones and sandstones, to the ironstones at the top of the sequence, often associated with marine transgressive/ regressive cycles (Bayer, 1989; Petra´nek and Van Houten, 1997; Taylor, 2005). Despite an extensive and intensive history of investigation, the origin of marine ironstones is generally not well understood. Factors controlling ironstone distribution, sources of Fe and mechanisms of ooid formation, which may vary from


Sedimentary Hosted Iron Ores

formation to formation, are still the subject of much debate, with a number of competing mechanisms or hypotheses for their formation. For example, Kimberley (1981) lists 7 models and 13 mechanisms to account for the method of Fe transport and modes of iron concentration, respectively, in marine ironstones. Similarly, Young (1989) outlines nine mechanisms by which ooids are thought to form including mechanical accretion, reworking of pedogenic allochems, intraformational growth, crystallization from ferruginous gel precursors, and mineralization and replacement of calcareous ooids. Microorganisms, such as bacteria, fungi, and primitive algae, are also considered to have played a part in the formation of ooids and Fe-coated grains (Bayer, 1989; Burkhalter, 1995; Petra´nek and Van Houten, 1997; Preat et al., 2000; Samala et al., 2011; Taylor, 2005). Recently, Timofeeva and Golov (2010) summarized the similar role of microorganisms in the formation of ferro-manganiferous nodule formation in pedogenic environments. Regardless of the actual mechanism/s involved, once formed, a degree of mechanical sorting and physical accumulation of ooids is required for development of the ironstone formation to occur (Bayer, 1989; Young, 1989). Classical ironstone deposits include the Ordovician Wabana Formation of Newfoundland, the Silurian Clinton Group of the central and southern Appalachians, and the early Jurassic ‘Minette’ Lorraine ironstone of northeastern France and southwestern Luxembourg. As summarized by Taylor (2005), the work of the IGCP 277 project concluded that ironstone formation occurs through a complex interaction of a number of factors, such as seawater salinity and depth, source and availability of Fe, climatic conditions, atmospheric CO2/O2 contents, diagenetic processes and tectonism, with the local hydrodynamic conditions and topography of the land and seafloor exerting the dominant influence (Taylor, 2005). These factors may have operated at all time scales ranging from the short term (e.g., 400 000 years, Milankovich cycle), through the medium (e.g., 32 Ma) and longer terms (eras) (Van Houten and Arthur, 1989; Young, 1989). The distinctly bimodal stratigraphic distribution of marine ironstones (Figure 18) with peak periods of formation during the Ordovician and Jurassic has been associated with a number of factors including the generation of abundant organics during mild climatic conditions, high rates of detrital sedimentation matter, high global sea levels, and high degrees of continent dispersion and seafloor spreading (Bekker et al., 2010; Petra´nek and Van Houten, 1997; Van Houten and Arthur, 1989; Van Houten and Bhattacharyya, 1982; Young, 1989). However, climate is not considered a major factor in the formation of ironstones as, for example, most Phanerozoic marine ironstones are restricted to between 10 S and 70 N (Taylor, 2005; Van Houten and Bhattacharyya, 1982). Part of the reason for the lack of consensus and diversity of genetic models for ironstone and ooid formation has been the lack of a simple, contemporary analog against which older marine ironstones can be compared and used to guide understanding of the conditions that led to the formation of more ancient deposits. The comparatively recent discovery of a body of unconsolidated iron ooids and pisoids forming in a shallow-marine volcanic setting at Mahengetang, Indonesia (Heikoop et al., 1996), has been used as evidence for a volcanic-exhalative origin, with possible input from volcanic

ash, for Fe-ooid formation in ancient POI (Sturesson et al., 2000). The close association of Middle-Jurassic ironstones in the Baltoscandian region to volcanic ash beds and REE distribution patterns (Greenwood and Gibb, 1971) in iron-rich (goethitic) ooids were similarly taken as evidence of a volcano-clastic input that may have contributed to formation of these older ironstones (Sturesson, 2003; Sturesson et al., 1999). Though Bekker et al. (2010) note that many ironstones do not contain volcanic ash beds as evidence against possible volcanic ash input, diagenetic alteration and reworking has affected all ironstones to varying degrees and, hence, any textural or mineralogical indications of ash input would most likely have been lost. Thus, establishing a link for volcanic ash input may be difficult to prove or disprove. Recently, a broad secular link between marine ironstone formation and anoxic events (e.g., VMS emplacement, OM-rich shale deposits,) has been suggested (Bekker et al., 2010; Van Houten and Arthur, 1989) where hydrothermally sourced Fe (as well as P and Mn) was introduced to shallow sea/continental shelf margins (Bekker et al., 2010; Van Houten and Arthur, 1989). This lends weight to, and revives, one of the previously proposed models of ironstone formation involving a volcanicexhalative, deep weathering mechanism (Kimberley, 1981, 1989a,b, 1994). One of the difficulties in unraveling the origin of ironstones is that the now prevailing mineralogy represents a diagenetic and postdiagenetic alteration of an initial mineral assemblage (Taylor, 2005). In marine ironstones, ooids typically comprise alternating concentric bands of Fe-oxides (typically goethite) and berthierine sheaths (Petra´nek and Van Houten, 1997; Taylor, 2005). Goethite often appears as an alteration product of berthierine or it may also occur as a ‘primary’ phase (Young, 1989). Other Fe-bearing phases of importance in marine ironstones, include chamosite and siderite, while hematite, magnetite, and pyrite are usually of minor importance (Taylor, 2005). Deleterious phases in marine ironstones include carbonates (e.g., calcite, aragonite, dolomite, and ankerite) and phosphate minerals, such as carbonatefluroapatite, Ca5(PO4,CO3)3. F or ‘francolite’ and vivianite, Fe32þ(PO4)2.8H2O, are common and can be major components (Taylor, 2005). Indeed, marine ironstones typically show a P content of about 1 wt% P2O5, and may grade laterally into phosphorite deposits (Kimberley, 1994; Madon, 1992). Berthierine is considered to be one of the first formed, prediagenetic phases before any significant compaction or alteration of the ooidal mass had occurred (Young, 1989). Indeed, kaolinite may have acted as the precursor to the formation of berthierine, with the introduction of reduced Fe along with small amounts of Mg reacting under suboxic conditions leading to berthierine (Van Houten and Arthur, 1989). Berthierine is used as the term to describe Fe-rich 1:1 trioctahedral (serpentine group) silicates of general formula (Fe2þ,Fe3þ, Al,Mg,Mn)2(Si,Al)2O5(OH)4 with a 0.7 nm basal spacing (e.g., Taylor, 2005; Young, 1989). Chamosite, as described in Table 3, is the term used to denote an Fe2þ-rich, 2:1 trioctahedral chlorite with a 14 nm basal spacing and an end-member formula of (Fe52þ,Al)(Si3Al)O10(OH8). Chamosite and berthierine are very similar chemically and, indeed, berthierine may be diagenetically transformed to chamosite at temperatures in the range of 120–160  C and a depth of 3 km

Sedimentary Hosted Iron Ores

Alluvium - Recent CID - Late Tertiary Mesozoic and Upper Palaeozoic Middle Precambrian Granitoids Post Wyloo Group Precambrian Formations Upper Wyloo Group Formations Lower Wyloo Group Formations Mt Bruce Supergroup and Jeerinah Formation


Fortescue Group below Jeerinah Formation Pilbara Craton Granitoids Pilbara Craton Supracrustal rocks


20⬚00' S

100 km


Port Hedland




Major road River


21⬚00' S Marble Bar

Ro be R.



5 22⬚00' S

22⬚00' S Fortesc


ue R.




3 Tom Price



23⬚00' S

23⬚00' S Paraburdoo Newman

24⬚00' S 115⬚30' E

117⬚00' E

118⬚30' E

120⬚00' E

24⬚00' S 121⬚30' E

Figure 19 CID distribution in the Hamersley Province with the Robe and Marillana palaeochannels (modified from Ramanaidou ER and Morris RC (2010) Comparison of supergene mimetic and supergene lateritic iron ore deposits. Applied Earth Science 119: 56–59). Boxed outlines define the main CID provinces in the Hamersley Province: (1) Robe Rive and Bungaroo CID, (2) Central Hamersley (Caliwingina, Serenity-Cabbage Gum Bore, Solomon East), (3) Rocklea Dome, Beasley River CID, (4) Yandi CID, and (5) Nullagine Joint Venture.

(Taylor, 2005; Young, 1989). This occurrence of a common ferriferous silicate mineralogy and its association with Feoxides, such as goethite and hematite, is one of the confounding aspects of marine ironstones that has long puzzled researchers. Terrestrial ironstones The term channel iron deposit or CID encompasses all terrestrial Fe-rich oolitic and pisolitic ironstone deposits of fluviatile, deltaic-lucastrine, and lucastrine origin and includes the more widely known Hamersley Province (Western Australia) and Kazakhstan deposits, as well as lesser known occurrences in the northern Yilgarn Craton, Western Australia (e.g., Ramanaidou et al., 2003). In view of the recently published detailed literature, which is focused particularly on the Australian CID, only a brief overview of their genesis is provided here. Instead, the reader is referred to the previous articles, for example, by Harms and Morgan (1964), MacLeod (1966), and more recently by Ramanaidou et al. (2003), Morris and Ramanaidou (2007), Ramanaidou and Morris (2010), and the references contained therein, for a more detailed discussion of ironstone deposits in the Hamersley Province. Briefly, in the Hamersley Province, CID occurs within the Poondano Formation, the Robe Formation, the Marillana

Formation, and in the newly discovered Caliwingina system of the central Hamersley (Figure 19), which may be grouped as a single type because of their mineralogical and chemical similarities (McGregor et al., 2010). CID occupy meandering paleochannels incised within the mature and lithologically variable Hamersley surface (e.g., Ramanaidou and Morris, 2010; Ramanaidou et al., 2003). Recent incision by younger drainage systems has led to varying degrees of relief or topographic inversion, particularly in the western Robe River drainage system (Figure 19), resulting in CID now being generally preserved as low-lying mesas (Dalstra et al., 2010). These variably eroded deposits vary between <1 and 100 m in thickness, with channel widths of generally <1 km but varying up to 5 km (Ramanaidou and Morris, 2010; Ramanaidou et al., 2003). Fossil plant and pollen assemblages (Casuarinaceae and Myrtaceae species) in basal, organic-rich claystones provide an Early Oligocene constraint for the Marillana CID (MacPhail and Stone, 2004), while (U–Th)/He dating of Yandi CID matrix goethite gives an age range of 5–14 Ma for the latest phase of CID cementation (Heim et al., 2006). More recent (U–Th)/He dating of Robe River CID ore gave a Late Oligocene to Late Miocene age (11.6–18.3 Ma) comparable to that of the Yandi CID (Danisˇ´ık et al., 2011). The CID ore is typically ooidal and porous with minor peloids, with both commonly comprising a hematitic core,


Sedimentary Hosted Iron Ores


H Wood

3 cm

Wood (a)

1 mm





1 mm


1 mm

Figure 20 Polished surface of a channel iron deposit sample showing: (a) the heterogeneous nature of the components that include ooids and pisoids, with larger and more irregular peloids and ferruginised wood fragments. (b) A range of granules, including goethitic and hematitic (H) wood fragments. (c) Matrix-rich sample with pelletoids showing varied cracking in the hematitic (H) nuclei. (d) A range of simple and complex nuclei. Note the radial cracking in the cortex (bottom right), and the variations in the amount of matrix in the section. Modified from Figure 4 of Ramanaidou ER, Morris RC, and Horwitz RC (2003) Channel iron deposits of the Hamersley Province, Western Australia. Australian Journal of Earth Sciences 50: 669–690.

enveloped by a concentrically zoned goethite-rich cortex supported in a goethitic matrix (e.g., Ramanaidou et al., 2003). A variety of facies can be found within the CID – (1) granular with local intraformational conglomerate, (2) bedded, and (3) altered types including conchoidal, leached, and surface types. The granular facies is the only type that typically meets the specification to be considered as ore (Ramanaidou and Morris, 2010). The slightly older, Middle Oligocene Kazakhstan CIDs, as summarized by Ramanaidou et al. (2003), occupy lowlying river valleys with the largest occurrence at Lisakovsk (or Lisakovskiy). The Lisakovsk CID being up to 2–8 km in width and extending for a length of 100 km, with ore horizons averaging 8 m in thickness, has total reserves of 2800 Mt and is comparable in size and extent to some individual Pilbara Region CIDs (Ramanaidou et al., 2003). Goethitized and hematitized fossil wood is preserved, respectively, in the matrix or in the nucleus (Figure 20) and is a diagnostic and characteristic feature of CID (Morris and Ramanaidou, 2007; Ramanaidou et al., 2003). For example, the upper main ore zone of the Yandi CID may contain up to 15% fossil wood (Stone et al., 2002). Fossil wood fragments occurring as rare tree twigs and branches, and stems of plants have been reported in the Kerch Tobacco ores (Tsipurskii and Golubovskaya, 1989). In addition, preserved fragments of plant debris including woody stems and branches are also common in the Lisakovsk CID (Golubovskaya, 2003). The Robe

paleochannel is the longest in the region and contains CID partly preserved along a distance of nearly 150 km, whereas the Yandi CID deposits are hosted within the Marillana paleochannel, for nearly 90 km of its length (Ramanaidou and Morris, 2010; Ramanaidou et al., 2003; Storkey et al., 2010). Several genetic models to account for the formation of the Hamersley CID have been proposed over the years, summarized initially by MacLeod (1966), and recently broadly grouped into three types (Dalstra et al., 2010). Two models are similar and involve precipitation of iron oxides by oxidation of Fe2þbearing groundwater with CID forming as a bog ore equivalent or via the replacement of a precursor channel fill. The third and currently preferred model invokes sheetwash accretion of ferruginous, pedogenic ooids derived from a deep ferruginous regolith (only in part over BIF) with essentially riverine ‘mud-bankderived’ peloids as channel fill together with intraformational cementation of the iron oxyhydroxide CID matrix (Morris and Ramanaidou, 2007). The variable association of a magnetic ferric/ferrous oxide phase, most likely kenomagnetite within the cores and cortices of ooids and peloids was used by Morris and Ramanaidou (2007) as evidence for pedogenic formation of the Fe-rich pelletoids prior to or during deposition in the channels. Combustion of surficial organic matter during wildfires is thought to locally induce a temporary reducing atmosphere where iron oxides are partially reduced to kenomagnetite, which gradually oxidizes to maghemite (g-Fe2O3) or inverts to hematite (Morris, 1993; Morris and Ramanaidou, 2007).

Sedimentary Hosted Iron Ores

Alternative models of CID genesis and formation of ooids/ pisoids have been proposed and include sheetwash accretion of pisoliths formed by the action of raindrop impacts on dry soils (Lascelles, 2007), and genesis from an older precursor, greensand protore derived from volcanic ash (Schwann, 2009). However, these models require further analyses before they can be considered as acceptable alternatives to the currently held view. Comparative Evaluation of POI-Derived Iron Ore Mineralogy and Composition Petra´nek and Van Houten (1997) described 366 POI occurrences with a brief description of their stratigraphic and petrologic records as a contribution to the IGCP-277 Project. More recently, McGregor et al. (2010) completed an updated database of nearly 400 POI deposits worldwide to better evaluate the characteristics of both marine and terrestrial ironstones in terms of their location, ore endowment, mineralogy, chemistry, and age. Building on the work of these authors, the following discussion compares the mineralogy and composition of marine deposits such as the Kerch, Paz de Rio, Salzgitter, Frodingham, Cleveland, and Lorraine deposits to the terrestrial Lisakovsk and Pilbara CID ores. General deposit characteristics, including their age, location, and composition are compiled in Tables 9 and 10 from data reported by Mu¨cke and Farshad (2005), Slater and Highley (1977), Neumann-Redlin et al. (1977), and Horon (1977). A first evaluation of ironstone ore composition in Table 10 highlights the varied and much lower average Fe contents of the marine ores (from 24.9 wt% for Cleveland to 45.3 wt% for Lisakovsk) and Khazakhstan ore compared to the Pilbara CID ores (57.1–58.8 wt% Fe), and the higher gangue element (SiO2, Al2O3, and P) content of the former ores compared to the latter (Table 10). Furthermore, Al2O3 and SiO2 contents in the marine ironstones are between 2 and 6 times greater than in the Pilbara CID (Table 10). The exception is the Lisakovsk CID ore where the average Al2O3 content is similar to that of the Pilbara CID (Table 10). Furthermore, phosphorus contents of the marine and Lisakovsk ores are between 10 and 25 times greater than that of the Pilbara CID ores (Table 10). Indeed, marine ironstones can contain very high and varied P contents such as the Cambrian–Ordovician marine ironstone in northern Wales, with P contents ranging between 0.3 and 4.9 wt% P (Kholodov and Butuzova, 2001; Zitzmann, 1978). Pilbara CID ore mineralogy is dominated by hematite and goethite, commonly containing Al and Si within the goethite structure. Kenomagnetite-maghemite is variably associated within ooidal cortices and cores, and represents the main magnetic phase in these deposits (e.g., Morris, 1994; Ramanaidou et al., 2003). Gangue mineralogy is dominated mainly by kaolinite occurring in voids and as fine layers alternating within goethite and hematite in pelloids, with other waste phases including halloysite, gibbsite, boehmite, illite and Ca/ Fe-smectite (Ramanaidou et al., 2003). The distribution of aluminosilicates, such as kaolinite, in Robe River CID ore was mapped using electron microprobe wavelength dispersive spectroscopy (WDS), where coincidental concentrations of Al and Si (Figure 21) were used to infer the presence of kaolinite, mainly


in the ooid cortex (Ramanaidou et al., 2003). However, these waste phases typically comprise <5% of the ore proper, which accounts for the comparatively low gangue element (Si and Al) concentration in the Pilbara CID ore (Table 10). The poorer quality of the marine ironstones and the Lisokovsk ores is due to their varied mineralogy compared to the Hamersley CIDs. Mineralogy of the Salzgitter, Frodingham, and Lorraine ironstones is dominated by goethite or ‘limonite,’ chamosite, and siderite within a fine-grained, clayey carbonate and chamositic matrix (e.g., Neumann-Redlin et al., 1977; Slater and Highley, 1977), whereas the Cleveland ore consists of rare, chamositic-fringed ooids in a mainly gray, sideritic and chamositic mudstone (Slater and Highley, 1977). At Paz de Rio, ooids are composed of goethite and hematite in approximately equal proportions, with (Ca, Mg)-siderite and K-bearing chamosite as the main silicate phase (Kimberley, 1980). For the Kerch and Lisakovsk ores, Fe ranges from 20 to 52 wt% (Zitzmann, 1977) whereas for the limonitic Lisakovsk ooidal ore, Fe contents vary from 30.4 to 52.5 wt% (e.g., Golubovskaya, 2003). Kerch brown ores comprise smectite, ‘hydrogoethite,’ Mn-hydroxides, siderite, and Fe–Ca phosphates, with a matrix comprising mainly ‘hydrogoethite’ (Golubovskaya, 2001). The mineralogy of the ooids of the Lisakovsk CID ore is dominated by ‘hydrogoethite’ with ‘amorphous’ material; quartz and prehnite fragments often form the ooid core (Golubovskaya, 2003). In less well-sorted ores, the matrix comprises a varied mixture of quartz, smectite, mica, kaolinite, and goethite, whereas in better sorted ores, it includes birnessite, goethite, and minor mica (Golubovskaya, 2003; Tsipurskii and Golubovskaya, 1989). The variable ore Fe grade at Lisakovsk relates to the varied ooid/quartz mixture (Golubovskaya, 2003), though the ore is amenable to upgrading by wet jigging and gravity-magnetic beneficiation methods so that ore with an average Fe content of 38 wt% can be upgraded to produce a concentrate with an average Fe content of 49.0 wt% (Kokal et al., 2008). Though ‘hydrogoethite’ is regarded as important at both the Kerch and Lisakovsk deposits, neither Golubovskaya (2001) nor Golubovskaya (2003) provide compositional data for this phase and it is likely this may be taken as a synonym for ‘limonite.’ The occurrence of Al and Si in CID ores is commonly associated with the iron oxide mineralogy, mainly goethite, which is considered to contain a few weight percent Al and Si within the goethite structure (Ramanaidou et al., 2003). Unitcell dimensions calculated for goethite and hematite in Fe-rich material from the Uche mine, Paz de Rio (Fajardo et al., 2004), were comparable to unit-cell dimensions of standard, Al-free goethite and hematite. Similarly, based on XRD evidence, Maynard (1986) concluded that the oolitic goethite in all 14 Proterozoic to Cretaceous ironstones examined in their study was free of structural aluminum. The low levels or absence of Al in goethite in marine ironstones has been used, in part, as evidence against a reworked terrestrial source for the occurrence of ooids in marine deposits (Collin et al., 2005; Maynard, 1986; Samala et al., 2011). However, in the Minette ironstone, ooidal goethite contains on average 7–10 mol% Al (Siehl and Thein, 1989), though it has been argued that the elevated Al (and Cr) contents reported in the Minette may represent a lateritic residual concentration of a detrital component that was reworked during ironstone formation (Maynard, 1986). The formation

Table 9

Name, age, location, and general characteristics of representative examples of POI iron ore






Production (year)

Av. Fe (wt%)

Total Fe endowment (Mt)

Operational (Y/N)




Middle Oligocene

100–120 km SW of Kustanay, northern Tugai Lowland, Kazakhstan





Zitzmann (1978)

Robe River


Late Eocene-Miocene

Mesas A and J

32 Mt (started 1992)



BHPB Yandicoogina RTIO Yandicoogina Nullagine IOP


Late Eocene-Miocene

16 km south-west of Pannawonica, Western Australia 90 km north-west of Newman, Western Australia

949 Mt (295 Mt @>38% Fe) 2850 Mt


2360 Mt



Late Eocene-Miocene


Late Eocene-Miocene

McGregor et al. (2010) McGregor et al. (2010) MacGregor et al. (2010) BC Iron (2011)




Paz de Rio




Late Eocene-Early Miocene Early Cretaceous







Early Jurassic (Early Lias) Early Jurassic (Middle Lias) Early-Mid Jurassic

BHPB Yandi RTIO Yandi

140 km north of Newman, Western Australia Kerch Peninsula, Azov iron ore basin Azov-Black Sea, Russia Paz de Rio-Sabanalarga region, Columbia


55 Mt



53.7 Mt (started 1998) N/A (started Feb. 2011) 70 Mt (total)


850 Mt


Paz de Rio






77.3 Mt (1938–1976) 259.7 Mt (1859–1976) 375.1 Mt (1854–1964) 900 Mt (Fe) (1871–1997)


450 Mt



1408 Mt


28 (26–31)

223 Mt



641 Mt



28 km SW to 38 km S Brunswick, Germany SW Kingston upon Hull, Lincolnshire, England East Teessdie, North Yorkshire, England


Minette Basin, north-eastern France




Zitzmann (1978) Kimberley (1980) Zitzmann (1978) Zitzmann (1978) Zitzmann (1978) Zitzmann (1978)

Note: The term endowment is used in a broad sense and includes all categories of reserves, resources, and estimates, with their associated differences in confidence, compiled from current data. Exceptions to this are the Australian ironstones where reported deposit tonnages are JORC compliant.

Table 10

Average composition (wt%) for selected marine and terrestrial POI Kerch (Brown ore) (Golubovskaya, 2001)

SiO2 TiO2 Al2O3 Fe MnO MgO CaO P Na2O K 2O H2O CO2 C S LOI Oxide Tot.

28.4 0.31 6.68 32.5 1.22 1.01 1.55 1.02 0.52 0.81

Kerch (Brown ore Av) (Zitzmann, 1977, p. 357)

Paz de Rio (Kimberley, 1980)



5 37.7 2.32 1 1.75 1

5.39 42.7 0.22 0.53 1.58 1.06

Salzgitter (Mu¨cke and Farshad, 2005, p. 251 Profile 1)

11.3 0.18 6.14 40.7 0.19 1.21 7.44 0.47 0.19 0.25 0.97

Frodingham Scrunthorpe (Mu¨cke and Farshad, 2005)

26.2 0.48 9.26 27.4 0.47 2.69 5.42 0.19 0.25 1.13 1.25

Cleveland Ironstone (Mu¨cke and Farshad, 2005)

10.5 0.22 5.39 24.9 0.54 3.69 15.56 0.42 0.38 0.43 0.56

Lorraine (Mu¨cke and Farshad, 2005)

15.3 0.22 4.81 37.2 0.32 1.52 7.90 0.66 0.10 0.23 0.93

0.14 0.07 10.9 99.5


0.07 12.1 N/A

14.2 100.9

14.8 100.0

28.6 99.4

14.7 100.2

Lisakovsk (Oolitic ore) (Golubovskaya, 2003)

SiO2 TiO2 Al2O3 Fe MnO MgO CaO P Na2O K2O H2O CO2 C S LOI Oxide Tot.

17.7 0.61 1.70 45.3 0.28 0.35 0.81 0.30 0.46 0.25 2.07 1.49 0.26 12.6 100.1

Pilbara DSO Sinter Fines (<6.3 mm) (METS, 2011)

Nullagine Pilbara DSO (Gale et al., 2012)

Yandi (BHPB)

Yandi (Rio Tinto)

Robe River





1.40 57.7

1.51 58.8

2.50 57.1

1.97 56.9

0.05 0.04 0.04

0.09 0.04 0.05

0.16 0.51 0.03





0.011 10.2 N/A

0.008 9.7 N/A

0.015 10.0 N/A

0.011 12.1

Fe and P contents for the marine and Lisakovsk ironstones were recalculated from the Fe2O3 and FeO contents listed for these deposits. Conventional industry practice reports Fe and P in elemental abundance (wt%), while all other assays are reported as oxide wt%. N/A, not available. Note: Compositional data for the Frodingham, Cleveland, Saltzgitter, and Lorraine ironstones was taken from Mu¨cke and Farshad (2005) measured as whole-rock XRF analyses. Comparison of this data against earlier ironstone ore assay reports for the Frodingham and Cleveland deposits (e.g., Slater and Highley, 1977), Saltzgitter (e.g., Neumann-Redlin et al., 1977), and for Lorraine (e.g., Horon, 1977) confirmed the data of Mu¨cke and Farshad (2005) as representative of previous ore assay results. Data for the Paz de Rio ironstone was averaged from three profiles covering the lateral extremes and central region of the orebody (Kimberley, 1980). At the Kerch deposit, the Tobacco and Brown ores are the most important, with exploitation focused only on the Brown ore (Zitzmann, 1977) – data for the Kerch Brown ores were taken from Golubovskaya (2001). At Lisakovsk, ore is distinguished as either limonite or limonite-siderite-leptochlorite types (Zitzmann, 1977) – data for only limonitic oolitic ore (e.g., Golubovskaya, 2003) are presented. Averaged element abundances (wt%) for Fe and P are presented, while Al and Si contents are presented as their respective oxides as is current industry practice.


Sedimentary Hosted Iron Ores

Al (wt%) 2.2

Fe (wt%) 69

0 1 mm

0.0 1 mm P (wt%) 0.7

Si (wt%) 1.8

0.0 1 mm

0.0 1 mm

Figure 21 Wavelength dispersive spectroscopic (WDS) x-ray mapping of the Fe (b), Si (c) and Al (c) distribution of a high-P (0.118 wt%) Robe River CID ore. Coincidental elevated concentrations of Al and Si (yellow-red colours) infer the presence of kaolinite within the ooid cortex and matrix. ‘Hot-spots’ of elevated P abundance (red circles) correspond to isolated occurrences of apatite. Element distribution maps were obtained at a 3 mm-step resolution using a JEOL 8500F hyperprobe operating at 15 kV and 128 nAm. Scale bar is 1 mm. Further details are provided in Ramanaidou et al. (2008).

and occurrence of Al-bearing goethite in Fe-rich ooids or pisoids in terrestrial weathering profiles is well known (Nahon et al., 1980). The elevated P contents of the marine ironstones and of the Lisakovsk CID are linked to the presence of discrete P-bearing phases such as chlorapatite (e.g., Paz de Rio – Kimberley, 1980), and Fe/Mn and Fe–Ca phosphates, such as vivianite and kerchenite (or kertschenite) Fe2þFe23þ(PO4)2(OH)2.6H2O, mitridatite Ca2Fe33þ(PO4)3O2.3H2O, bosphorite (or santabarbaraite), Fe33þ(PO4)2(OH)3.5H2O, (e.g., Kerch – Zitzmann, 1977; Golubovskaya, 2001), and switzerite (Mn,Fe)3(PO4)2.7H2O, ludlamite (Fe,Mg,Mn)(PO4)2.4H2O, and phosphoferrite (Fe2þ, Mn)3(PO4)2.3H2O (e.g., Lisakovsk – Ionkov et al., 2011). In the absence of any discrete, P-bearing phases in ironstones, previous work has noted an association between P and Fe. Kimberley (1980) reported a close correlation between P, Ca, and Fe, because of the presence of apatite. However, at Lisakovsk,

Golubovskaya (2003) concluded that there was no association between Fe and P. At Lisokovsk, P is uniformly distributed throughout the ‘limonitic’ ooids and is associated with other gangue phases, rendering the ore impotent to beneficiation using physical methods (e.g., Kokal et al., 2008). Similarly, in a study of more than 40 marine and terrestrial ‘limonitic’ ironstones, Kholodov and Butuzova (2001) did not distinguish any clear trend between the contents of P and Fe. Formation of discrete phosphorous mineralization may, in part, be biologically derived, with Fe-phosphates such as strengite, FePO4.2H2O, and vivianite, Fe3(PO4)2, shown to be associated with biofilms and magnetotactic bacteria (e.g., Konhauser, 1998). For the Pilbara CID, P is known to be associated with Feoxides, particularly goethite, though the exact nature of this association (i.e., structural vs surface absorbed) is unknown. By comparison, in BIF-hosted high-P Brockman martite– goethite ores, current evidence supports a strong goethite-P

Sedimentary Hosted Iron Ores

association with P most likely present as a surface-adsorbed species in micropores or as an occluded phase within nanosized, intradomain regions within ochreous goethite crystals (Wells and Ramanaidou, 2011).

Bog Iron Ores


bog iron deposits in the Animas river watershed, Colorado, gave C dates of 3800–4010 BP (Yager et al., 2003). This was generally consistent with earlier dating where Fe-replaced wood fragments in ferricretes in the upper parts of the watershed had radiogenic 14C ages of 8840–9580 years BP and overlapped reported post-glaciation ages for wood fragments in cirque deposits in the area (Yager et al., 2003). 14 General occurrence setting and early utilization ‘Bog iron’ ores typically occur in low-lying areas such as swamps, marshes, or meadows and where drainage is sluggish or impeded (Stanton, 1972). Most of these deposits occur in northern hemisphere countries such as central Europe, Scandinavia, Russia, and northern America (e.g., Dake and Rolla, 1916; Stanton, 1972). Here, the deposition of Holocene age (c.5000 BP) glacial tills and sediments in glacially ‘graded’ terrains has resulted in water-logged or impeded drainage conditions. Iron concentrations form in the sluggish streams and the many small lakes and swamps that later develop in these post-glacial terrains (Stanton, 1972). However, more recent examples have been described in mountain spring-related deposits in New Zealand (Childs et al., 1986). One of the most notable exceptions of bog-related deposits not associated with paleoglacial terrains are the iron deposits or ‘pseudogossans’, as they were first described at Mesa de los Pino´s and Cerro de los Vacas, Rio Tinto, Huelva Province, Spain (Dake and Rolla, 1916; Finlayson, 1910). Initially studied in the late 1880s, oxidation of the Cu-bearing, pyritic ores at Rio Tinto acted, as cited by Bateman (1927), as the ferrous iron source, with subsequent oxidation resulting in significant iron oxide deposition occurring in a swampy lake or marsh. These mainly hematite-rich deposits were reported to contain Miocene plant debris (Finlayson, 1910). Drainage rejuvenation and landscape dissection through tectonic uplift finally resulted in the formation of perched valley deposits of significant iron oxide accumulation. For example, dissection by the Rio Agrio river has formed the perched Mesa de los Pino´s and Cerro de los Vacas deposit, which at Mesa de los Pino´s is  1 km long, 100 m wide, and up to 8 m thick (Bateman, 1927). More recently, these perched deposits or terraces have been the focus of rock magnetic remanence carrier studies using anisotropy of magnetic susceptibility to characterize the magnetic fabric of both the true gossans and ‘displaced’ gossans in the Rio Tinto district (Essalhi et al., 2011). Occurrences such as the Rio Tinto deposits aside, countries regarded as the most important bog iron producers include Germany (McMillan and Schwertmann, 1998), Poland (Kaczorek and Sommer, 2003), Belgium (De Geyter et al., 1985; Landuydt, 1990; Stoops, 1983), Denmark (Breuning-Madsen et al., 2000; Postma, 1977), Norway, Sweden, Finland, Russia (e.g., references as cited by Postma, 1977) and USA, including localities in Canada (e.g., Ontario – Evans et al., 1978; Moore, 1910) and North America (e.g., Colorado – Theobald et al., 1963; Yager et al., 2003); Maryland – (Maryland – Bricker et al., 2003), and New Jersey – (New Jersey – Braddock-Rogers, 1930; Crerar et al., 1981). Dating of bog iron ore deposits is based largely on the ‘paragenetic’ (stratigraphic) associations in which the iron oxide accumulations occur and there are few examples of attempts to date the bog iron deposits directly. However, recent radiogenic 14 C dating of wood fragments enveloped in stream-developed, Bog iron ore exploitation Bog iron deposits occupy a significant place in early human history, with their initial exploitation known since the PreRoman iron age (c.500 BC). The majority of the iron smelted during the time of the Vikings was derived from bog iron. One of the earliest recorded use of bog ores in Europe is the occurrences in Upper Silesia, southern Poland, where exploitation had been known prior to the fourteenth century (Czaja, 2001). In Denmark, bog ores were known from early (c.1895–1946) European and Russian reports, as cited by Postma (1977), and the Holocene bog ores in the Nete Valley, Antwerp, have been known since the mid 1840s and exploited for iron until as recently as the 1950s (De Geyter et al., 1985). Bog iron deposits at a number of North American localities (e.g., Maryland, New Jersey) were large enough to support smelting operations over extended periods. On the Atlantic coast, early colonial use of bog ores in the basins of the Mullica and Great Egg Harbor rivers, Pines region, New Jersey, supported an extensive iron ore industry from about 1700, which continued for some 150 years (Braddock-Rogers, 1930). Early commercial use of bog ore in dried form as building material is also documented (Braddock-Rogers, 1930). In addition, indigenous tribes in the area are reported to have used the dried, powdered form of the ore as war paint (Braddock-Rogers, 1930). To the southwest of Maryland, along the Nassawango Creek on the Delmarva Peninsula, smelting of bog ore (using oyster and clam shells as the flux and charcoal for fuel) occurred at Snow Hill, resulting in the production of up to 700 tons per year over the period 1830–50 (Bricker et al., 2003). In Colorado, bog iron has been known in various counties (e.g., Summit County) since the late 1800s, and Theobald et al. (1963) cite early work by Chauvenet (1890) who provided the first description of bog ore in the area. In the spring-related or ‘seepage’ bog ores of north New Zealand, Childs et al. (1986) cites earlier reports from the late 1870s of the Taranaki Maori people digging ochre out from swamps along the Waiwakaiho River and using the dried form for ceremonial and practical purposes, such as paint for homes, canoes, and personal adornment. The longevity of some of these early smelting operations was, in part, due to the ‘regenerative’ nature of the deposits with Fe-oxide mineralization being replenished over relatively short periods. This has been noted at a number of bog ore deposits worldwide. For example, the Nassawango Creek bog ores, as previously described, were reported to regenerate after only a few years (Bricker et al., 2003). At other localities, for example, the regeneration of bog ores at La-a-la Tortue, Quebec, reportedly occurred after 8–10 years, whereas Swedish lake ores were apparently replenished after 25 years (Moore, 1910). However, problems with the high P and S contents of some bog ores – the earlier cited analysis reported up to 10% P in the bog ore at Nassawango Creek (Bricker et al.,


Sedimentary Hosted Iron Ores

2003) – coupled with the discovery of larger, better quality iron ore deposits elsewhere resulted in production ceasing, for example, in Maryland, by the mid-1850s (Braddock-Rogers, 1930; Bricker et al., 2003). Types of bog iron ores Continuing the earlier work by Harder (1919), Stanton (1972) divides bog iron ore deposits into two main types, namely, Lake ores and Marsh or peat ores. Lake ores Lake ores, which include volcanic lake systems such as those in Japan (Stanton, 1972), are well known in eastern and northern America and in Scandinavia (Halbach, 1976; Schwertmann et al., 1987). Lake ores as ferro-manganiferous nodules have been reported in the Russian lakes of the Saint Petersburg region (Chukhrov et al., 1982). Related lake ore deposits are also known in the central African great lakes (Johnson, 1984). Well-sorted ooids, averaging  0.25 mm in size and comprising smectite (montmorillonite), goethite, and silica are forming off the present Charli river delta that feeds into the shallow (3 m) deep waters of Lake Chad, Nigeria (Lemoalle and Dupont, 1973). Similar smectite (nontronite), opaline ‘limonite,’ and vivianite ooidal and pisolitic deposits are forming at greater depths (<250 m) in Lake Malawi, eastern Malawi (Mu¨ller and Fo¨rstner, 1973). They are texturally and mineralogically comparable to the pisoidal and related (i.e., crust and ‘penny’) ore types forming in Finnish lakes (Halbach, 1976; Schwertmann et al., 1987). Indeed, the mineralogical and textural characteristics of these ore types are more akin to those of ooidal ironstones than to bog ores (see later) and may represent a present day, fresh-water analog of older ooidal ironstones (Taylor, 2005). More recently, spheroidal hematite-concretions were reported to be actively forming in the hypersaline, acidic ( pH 4), shallow subsurface sediments of Lake Brown, Western Australia (Bowen et al., 2008). Radiogenic 14C dating of organic-rich sediments entrapping the hematitic ooids indicated ages of <3000 years, showing that ooidal formation may occur under modern conditions (Bowen et al., 2008). Bearing in mind that while these deposits may cover an extensive area – the ooidal deposits at Lake Chad are distributed over an area of some Table 11

Oxide Fe2O3 MnO2 TiO2 CaO K2O P2O5 SiO2 Al2O3 MgO Na2O H2Oþ Total

2700 km2 with a corresponding Fe content of 30  106 tons (Lemoalle and Dupont, 1973) – they are not economically viable. Stream/spring bog ores The occurrence of large bog-like ochreous deposits in small streams associated with some springs fed by vents on (andesitic) volcanoes (e.g., Mt Egmont and Mt Ruapehu) on the North Island of New Zealand (Childs et al., 1982, 1986; Henmi et al., 1980) may represent a related or similar lake bog ore variant. At some locations, these deposits can be >1 m thick and up to 1600 m in length (e.g., Childs et al., 1982). Detailed analyses identified these deposits as comprising mainly 0.1–0.5 mm-sized aggregates of fine, spherical particles of ferrihydrite between 3 and 10 nm in diameter (Childs et al., 1986; Henmi et al., 1980), and, in some cases, very poorly ordered goethite (Childs et al., 1982). The very high surface areas of up to 600 m2 g1 measured for these short-range ordered, hydrous Fe oxides account for the high levels of silica, up to 22.0 wt% SiO2, and P up to 1.74 wt% (Table 11), which are most likely present as adsorbed species. The identification of ‘hisingerite’ (an Feallophane), with the tentative formula Fe2Si2O5(OH)42H2O, by Childs et al. (1982) in these deposits may simply be a misidentification of ferrihydrite with high levels of surfaceadsorbed silica. However, it has been reported on the basis of shifts in XRD line positions and from the increased thermal stability of ferrihydrite with increasing silica content, which inhibits conversion to hematite, that Si may replace Fe within the structure of ferrihydrite to contents as high as 7.5 wt% Si (Boyd and Scott, 1999; Carlson and Schwertmann, 1981; Vempati and Loeppert, 1989). Major chemistry and isotopic water analysis indicated that the waters issuing from these vents were meteoric in origin and have been carbonated from a volcanic source (Childs et al., 1982). Dissolved solids were thought to originate from andesite through which the water passes and accounted for the high Si content of >10 wt% (dried material – see Table 11). Ferrous iron in solution (up to 30–40 g m3) undergoes abiotic oxidation upon exposure, leading to the formation of an ochreous deposit (Childs et al., 1982). Development of Fe-rich sediments up to 3 m thick occurs in the bays of the volcanic islands of Nea Kameni and Palea Kameni (Schroll, 1978). A range of

Composition of stream-related bog ores Henmi et al. (1980)

Childs et al. (1982)

PC863 78.7

A1 72.1 0.15 0.02 1.19

A2 66.1 0.24 0.08 0.94

B 67.2 0.15 0.02 1.78

1.74 12.3 0.09 0.18

0.66 18.3 1.09 0.61

11.1 98.87

11.1 99.07

0.10 0.38

10.5 0.05 0.14 8.58 98.45

PC538/6 73.7 0.07 0.01 1.46 0.02 0.82 10.7 0.16 0.54 0.06 8.95 96.52

Childs et al. (1986)

0.21 17.0 0.78 0.23

C 57.5 4.77 0.03 1.90 0.02 0.09 21.0 2.24 0.40

D 58.9 0.30 0.05 2.18 0.02 0.11 22.0 2.88 0.36

11.7 99.06

13.1 101.04

12.7 99.56

PC1020 77.1 0.09 0.03 0.29

PC1022 63.9 1.65 0.07 0.94

PC1019 74.3 0.57 0.02 0.35

0.85 10.0 0.98 0.15 0.32 10.1 99.91

0.21 17.5 1.44 0.20

0.87 12.4 <0.01 0.07 0.34 10.6 99.58

14.0 99.87

PC863 78.8 0.10 0.38

10.5 0.07 0.13 8.58 98.52

Henmi et al. (1980): Based on oven-dried material; Childs et al. (1982): Percentage by weight of 110  C-dry material. H2O LOI1000  C for 1 h; Childs et al. (1986): Percentage by weight of 110  C-dry material. H2O LOI1000  C for 1 h.

Sedimentary Hosted Iron Ores iron minerals including amorphous Fe3þ hydroxide, goethite, siderite, framboidal pyrite, hisingerite, nontronite, and, likely, vivianite were identified. The presence of Fe-oxidizing bacteria, such as Gallionella ferruginea may act as nucleation sites for the crystallization of iron oxides (Schroll, 1978). A similar and related phenomenon associated with recent anthropogenic activity is the formation of surface Fe-rich precipitates at discharge sites of acid mine drainage (AMD) or acid rock drainage (Burgos et al., 2012; Burton et al., 2008; Marescotti et al., 2012), or other disturbed sites, such as mine pits and drained coastal lowland soils (Fitzpatrick et al., 1996; Kumpulainen et al., 2007). These anthropogenic bog ores, as they may be termed, are typically of limited extent with related deposits in Pennsylvania forming over an area of 6000 m2 and can be up to 2 m thick (Burgos et al., 2012). Iron oxidation in this strongly acidic (pH ¼ 2–4), sulfatebearing effluent may occur abiotically due simply to small changes in Eh/pH through mixing of AMD with unpolluted waters or is microbially mediated by acidophyllic, Fe2þoxidizing bacteria. The mineralogy of AMD Fe precipitates is influenced mainly by Eh, pH, and SO42 with a general mineral stability sequence with increasing pH in the range 2.0–4.4 of jarosite, schwertmannite, goethite, ferrihydrite, and amorphous Fe precipitates (Marescotti et al., 2012). In the presence of SO42, schwertmannite, Fe8O8(OH)82x (SO4)x with 1 < x < 1.75 (Bigham et al., 1990), forms as the dominant Fe3þ mineral over the pH range 2.5–4.0, with lesser amounts of poorly ordered goethite (Burgos et al., 2012; Knorr and Blodau, 2007). Schwertmannite is metastable with respect to goethite and transforms via hydrolysis over a period of months to years, with the rate of transformation to goethite being influenced by factors such as temperature, pH, concentration of SO42, and Fe2þ (Bigham et al., 1990; Knorr and Blodau, 2007; Paikaray and Peiffer, 2010; Peretyazhko et al., 2009). Subsequently, the surfaces of anthropogenic bog ores are typically rich in schwertmannite and become progressively more goethite rich with depth (Acero et al., 2006; Burgos et al., 2012; Peretyazhko et al., 2009). Marsh or peat bog ores Marsh or peat bog iron ores, as typified by many European occurrences, are found generally in sediments of the many small rivers in the low moors of glaciated terrains with moraines and outwash areas, and are also developed in hydromorphic, loamy, sandy to clayey alluvium and soils (e.g., De Geyter et al., 1985; Landuydt, 1990; Postma, 1977). Seasonal fluctuations in the water table give rise to periodic wetting and drying cycles, which typically produce an overlying oxidized zone in the upper 50–80 cm and an underlying reduced zone in the deeper parts of the soil profile (e.g., De Geyter et al., 1985; Landuydt, 1990; Stoops, 1983). These soils commonly have a high OM content and the bog ores are often associated with peat deposits (e.g., Landuydt, 1990; Postma, 1977). These bog iron ores, which can show Fe contents of up to 40–50 wt% (Breuning-Madsen et al., 2000; De Geyter et al., 1985; Kaczorek and Sommer, 2003), are often referred to as ‘soft’ bog ores (Table 12). In parts of these profiles where the porous microstructure enables good aeration, compacted or well-cemented layers of ‘hard’ bog ores can develop. They may be continuous or, with


the impact of anthropogenic activity associated with land use (i.e., plowing and draining), they may become mechanically disintegrated, forming as fragments or nodules (e.g., Kaczorek and Sommer, 2003). In the oxidized zone, goethite and ferrihydrite are the most important Fe-minerals, whereas in the reduced zone, siderite (FeCO3) or a mixed Fe/Mn-carbonate and vivianite are the most important Fe-bearing minerals identified. Vivianite and siderite also generally form the ground mass in the reduced zone (Landuydt, 1990; Stoops, 1983). The high P contents often reported for marsh bog iron ores, up to 6–8 wt% P2O5 (Table 12), can be associated with vivianite, but some of the P may have an anthropogenic origin being introduced through the use of fertilizers (Breuning-Madsen et al., 2000). The source of Fe in these bog ores may originate from the reduction of primary Fe-bearing minerals in the peat, such as goethite (McMillan and Schwertmann, 1998) or is introduced by groundwater flowing into the bog with Fe sourced from material through which it has flowed. For example, a possible primary source for the Belgian Campine soft bog ores (e.g., De Geyter et al., 1985; Landuydt, 1990; Stoops, 1983) is the underlying glauconiferous sands on which the bog ores are developed. Carbonate for siderite formation may have its origin from organic matter metabolized by bacteria under anoxic conditions or is introduced by groundwaters circulating through calcareous Alpine rocks (e.g., McMillan and Schwertmann, 1985; Postma, 1977). Blackband ironstones Described as a variant of the marsh bog ores, blackband ironstones are characteristically associated with Carboniferous and Permian coal seams (Stanton, 1972) as part of a lacustrine cyclothem system and underlie mudstone, siltstone, and seat-earth deposits (Taylor, 2005). Blackband ironstones typically comprise finally laminated sideritic ironstones alternating with organic-rich (>10%) laminae (Taylor, 2005). This is confirmed by the high FeO (38.0–50.9 wt%) and CO2 (29.9–32.7 wt%) and elevated C (1.4–5 wt%) contents reported for selected Blackband ironstones (e.g., Table 13). Blackband ironstones may be considered as ‘mature’ equivalents of peat or marsh bog ores (Taylor, 2005). Bacterial association in bog ores The occurrence of certain Fe- and Mn-mediating bacteria (in association with a range of ferromanganese deposits) has been known for some time with the more well-known species of bacteria such as Gallionella, Sphaerotilus, Leptothrix and Clonothrix described by the late-nineteenth century as reported by (Ghiorse, 1984). Biologically mediated Fe2þ oxidation and Fe3þ reduction has been reported in a range of aqueous marine and freshwater systems including lake sediments, aquifers, soils, deep-sea vents, and wetlands including bog iron deposits (Fortin and Langley, 2005; Konhauser, 1998; Timofeeva and Golov, 2010). Under oxic conditions, enzymatic metabolism by acidophilic bacteria (pH range 2–5) such as Thiobacillus ferrooxidans, Sulfolobus, Leptospirillum, Pedomicrobium, and Metallogenum (Fischer, 1988) and neutrophilic (pH 5–7) bacteria such as Gallionella ferruginea (e.g., Fischer, 1988; Konhauser et al., 2011) promotes oxidation of ferrous or Fe2þ to ferric or Fe3þ (Fortin and Langley, 2005; e.g., Figure 22). Under anoxic

Table 12 Oxide

FeO Fe2O3 MnO2 TiO2 CaO K 2O P2O5 SiO2 Al2O3 MgO Na2O H2Oþ C Total

Composition of marsh-related bog ores

De Geyter et al. (1985)

BreuningMadsen et al. (2000)

Kaczorek and Sommer, (2003)

W. Jutland 62.3









57.2 3.00

69.5 1.41

59.2 0.99

68.8 1.41

68.8 0.84

73.5 0.99

62.6 2.40

67.1 2.82

4.86 21.8

5.43 18.6

5.13 13.0

5.11 8.56

5.82 27.6

5.13 12.9

7.08 12.9

7.97 12.9

6.33 86.87

8.86 94.94

11.4 78.36

5.06 83.85

8.23 103.03

11.4 92.51

19.0 85.00

20.9 90.75




10a dark


40 dark






41.0 0.06 0 1.18

41.4 0.06 0 0.95

48.8 0.75 0 1.05

65.7 0.15 0 0.52

61.0 0.99

57.9 1.29

40.6 0.36

71.7 0.18

67.1 0.27

19.2 0.09

26.9 0.12





















3.48 10.5 0.93 0.12

























Sedimentary Hosted Iron Ores conditions, both oxidation of Fe2þ and reduction of iron oxides, through microbial anaerobic respiration, may occur (Fortin and Langley, 2005; Konhauser et al., 2011). However, at near-neutral pH, the role of biotic Fe2þ oxidation is uncertain as in the pH range 5–7, simple chemical or abiotic oxidation increases 100-fold for each unit-increase in pH (Schwertmann, 1988; Schwertmann and Taylor, 1989). The association of these bacteria, or ‘iron bacteria’ as they are commonly known (Ghiorse, 1984; Schwertmann Table 13 Oxide

Fe2O3 FeO MnO2 TiO2 CaO K2O P2O5 SiO2 Al2O3 MgO Na2O CO2 S SO3 FeS2 C H2Oþ H2O Total

Blackband ironstone composition Samples 1



1.49 38.0 1.51

50.9 0.58

0.42 43.1 2.07

29.9 0.55 0.8 10.0 5.57 3.37



0.33 5.1 2.35 0.3

1.18 8.67 4.47 2.09


31.8 0.18

32.7 0.18

5 2.7




0.06 1.42 1.47 0.74 94.93

Data from Stanton (1972). Samples from Yorkshire (1), Natal (2) and Ohio (3).

Figure 22 Biogenic Fe3þ mineralization during bacterial oxidation of Fe2þ by the nitrate-reducing Acidovorax strain BioFeN1. Fe3þ oxyhydroxide minerals of different crystallinity (orange globules, blue tabular crystals, brown aggregates) either partially or completely encrust bacterial cells (yellow). Reproduced with permission from Photolibrary (“© photolibrary. All rights reserved.”)


and Taylor, 1989), within bog iron and related deposits has long been documented (Braddock-Rogers, 1930; Crerar et al., 1979; Dake and Rolla, 1916; Inman, 1923; Murad, 1982). The study by Crerar et al. (1979) provided one of the first detailed descriptions of the bacteria identified (e.g., Thiobacillus ferrooxidans, Leptothrix ochracea, Crenothrix polyspora, Sidercapsa geminate, and Metallogenium sp.) within the New Jersey bog iron deposits and of the role they play in catalyzing Fe2þ oxidation. Recently, filamentous bacteria ( 1 mm in diameter) and other cell-like structures were identified in association with precipitates of jarosite and poorly ordered amorphous Fe oxides as coatings on river bed cobbles and preserved with goethite in <11 000–2.1 Ma terrace sequences of the Rio Tinto river system (Preston et al., 2011). These authors identified the bacterial filaments as similar to those of the acid-loving Leptothrix and Acidithiobacillus spp. Bog ore iron oxides – mineralogy Oxidation of ferrous-iron bearing solutions typically results in the formation of poorly ordered iron oxhydroxides such as ferrihydrite whether through biotic (Chukhrov, 1973; Chukhrov et al. 1976, 1977) or abiotic (e.g., Chukhrov et al., 1976 – Fe/Mn nodules; Vodyanitskii, 2010) pathways. Ferrihydrite is thermodynamically unstable and acts as the predominant precursor for transformation to goethite or hematite (e.g., Chukhrov et al., 1977; Pinney et al., 2009; Vodyanitskii, 2010). The next most stable ferric oxhydroxide after ferrihydrite is feroxyhyte, with the general formula d-FeOOH (Vodyanitskii, 2010), which may spontaneously transform into goethite (Chukhrov et al., 1976, 1977). In natural systems, feroxyhyte was first reported as a constituent in oceanic Fe–Mn nodules (Chukhrov et al., 1976, 1977) and then in nodules in Russian gley soils (Chukhrov et al., 1977). Later, feroxyhyte was identified in rusty precipitates in Finnish glaciofluvial sands and gravels (Carlson and Schwertmann, 1980). More recently, feroxyhyte was found in ferromanganese crusts along the mid-Atlantic ridge and on underwater slopes of the Sea of Okhotsk (Baturin and Dubinchuk, 2011; Baturin et al., 2012). Biotic manganiferous feroxyhyte has also been noted with filamentous bacterial remnants, and the presence of low levels of Mn (<5 wt%) has been reported to increase the chemical stability of feroxyhyte (Vodyanitskii, 2010). This may explain the occurrence of feroxyhyte in deep-sea Fe–Mn nodules as noted above. Ferrihydrite and feroxyhyte are discriminated principally by XRD analysis (e.g., Vodyanitskii, 2010). However, as ferrihydrite and feroxyhyte have related structures, d-spacings for a number of diffraction lines are near-coincident (Carlson and Schwertmann, 1980). Consequently, it is possible that in some bog ores, feroxyhyte has previously been misidentified as ferrihydrite (De Geyter et al., 1985; Kaczorek and Sommer, 2003). Considering the stabilizing effect of Mn on feroxyhyte, a proxy for its occurrence may be the presence of Mn as reported in some bog ores (e.g., up to 5% MnO – Postma, 1977), which may in part have been erroneously attributed to the occurrence of (poorly ordered) Mn oxides. Little is known about the transformation reactions of feroxhyte. However, the conditions that mediate the transformation of ferrihydrite to better ordered Fe3þ phases are far better


Sedimentary Hosted Iron Ores

understood with numerous investigations of natural ferrihydrite and synthetic systems documenting both inhibitory and promotional effects related to ferrihydrite transformation. For example, see the detailed reviews by Jambor and Dutrizac (1998) and Cornell and Schwertmann (2003). This transformation is significantly influenced by solution chemistry, such as pH, temperature, time, Fe2þ concentration, and the presence of inorganic anions and cations (Das et al., 2011; Fischer and Schwertmann, 1975; Hamzaoui et al., 2002; Liu et al., 2010; Murphy et al., 1976; Vodyanitskii, 2010). These show a strong affinity for adsorption with ferrihydrite because of their typically small particle size (e.g., 3–7 nm) and the resulting high surface area (e.g., 2–5  105 m2 kg1) of ferrihydrite (Schwertmann and Taylor, 1989). For comparison, Carlson and Schwertmann (1980) reported surface area values in the range of 143–305 m2 kg1 for synthetic feroxyhyte. Hence, the following discussion deals mainly with the transformation of ferrihydrite to better ordered iron oxides. Due to their prevalence in natural environments, the influence of P (as PO43) and Si (as SiO42) on the transformation of ferrihydrite to better ordered iron (oxyhydr) oxides, such as goethite and hematite, has been widely studied in artificial systems (Barro´n et al., 1997; Cornell et al., 1987; Galvez et al., 1999). In the pH range 3–9, transformation of synthetic ferrihydrite to more crystalline iron oxides, as measured by the proportion of ammonium oxalate-soluble Fe to total Fe, was inhibited for P/Fe (atomic ratios) of >2.5% (Galvez et al., 1999). Silicate is even more effective in inhibiting the transformation of ferrihydrite, with silicate concentrations as low as 104 M sufficient to stabilize ferrihydrite (Cornell et al., 1987; Doelsch et al., 2000), through the formation via ligandexchange of monomeric and trimeric surfaces bridging Si/Fecomplexes (Swedlund et al., 2010). It is known that the presence of aqueous Fe2þ can catalyze the transformation of ferrihydrite (and other metastable Febearing phases, such as schwertmannite, jarosite, and lepidocrocite) to more thermodynamically stable forms such as goethite (e.g., Jones et al., 2009; Pedersen et al., 2005, and the references contained therein). Reductive dissolution via electron transfer from surface-adsorbed Fe2þ and Fe2þ exchanged for Fe3þ at terminal surface sites solubilizes the ferrihydrite surface (Hansel et al., 2011). The reductively dissolved Fe3þ recrystallizes to more thermodynamically stable and more crystalline iron oxides, such as goethite (e.g., Cornell and Schwertmann, 2003; Hansel et al., 2011). However, the presence of surface-adsorbed PO43 (Larese-Casanova et al., 2010) or SiO42 (Jones et al., 2009) inhibits the reductive dissolution of ferrihydrite stabilizing its transformation by limiting direct surface adsorption of Fe2þ and, in the case of SiO42, by inhibiting polymerization of Fe3þ to more stable Fe3þ minerals (Jones et al., 2009). The presence of Al has been shown to exhibit similar inhibitory effects to the Fe2þ induced, reductive transformation to more stable Fe-oxyhydroxides (e.g., goethite), whether it is present as a surface-adsorbed species or as an isomorphous substituent in synthetic ferrihydrite (Hansel et al., 2011). These results are consistent with, and help to explain, the mineralogy typically reported for marsh bog ore and for the stream-hosted bog ore deposits. For both bog ore types, ferrihydrite is the predominant Fe3þ mineral, typically with minor

amounts of often very poorly ordered goethite and lepidocrocite (De Geyter et al., 1985; Kaczorek and Sommer, 2003), although, as previously discussed, because of the similarity in the XRD patterns between ferrihydrite and feroxyhyte, especially in the case of poorly ordered materials, there is the potential for the misidentification of ferrihydrite, leading to the over-reporting of its prevalence in natural systems. The presence of variably crystalline goethite/lepidocrocite detected in some bog ores may in part be related to seasonal variations in groundwater chemistry (i.e., cations, anions, and organics) and fluctuations in groundwater table levels. For example, in lake or stream bog ores, periodic fluctuations in Si concentration may see initial complexation with biogenically oxidized Fe3þ phases (i.e., ferrihydrite) and, thus, removal of aqueous Si. Peak activities in the concentration of Si and Fe2þ may or may not be coincidental, so that for some (extended) periods, biogenic Si/P-ferrihydrite is stabilized against transformation to more thermodynamically stable Fe3þ-minerals. As pH and the concentration of Fe2þ vary with time, the system comes under competing reaction kinetics, which drives the mineralogy to either less (ferrihydrite) or more (goethite/lepidocrocite) thermodynamically stable forms. In peat bogs, seasonal fluctuations in groundwater levels may expose deeper parts of the bog profile resulting in the partial oxidation of Fe2þ-bearing phases (siderite and vivianite) forming poorly ordered Fe(III) phases (e.g., De Geyter et al., 1985; McMillan and Schwertmann, 1998; Stoops, 1983).



Iron ore and its final product, steel, are arguably the most historically social and economically significant commodity of the last three millennia. Matched with the increased demand for higher grade material, changes in the economic viability of iron ore deposits have seen iron ore production shift from early, low-grade bog iron ore and marine ironstones to the gigantic BIF-hosted deposits and locally the terrestrial Phanerozoic ironstones. Despite their significance and the abundant literature devoted to these deposits, questions still remain regarding their genesis and their relationship with past global and/or local events. For instance, BIF geochemistry has been reasonably well studied, whereas only very few studies are concerned with the geochemistry of high-grade BIF-hosted deposits. China and the United States of America are still the dominant force in the production of magnetite-hosted BIF deposits, but Australia and Brazil are currently developing these deposit types. The BIF-hosted martite-microplaty hematite deposits represent the bulk of high-grade iron ores. The BIF-hosted martite-goethite and terrestrial ironstones, locally called CID, now hold primary importance in the Australian iron ore industry while the significance of the martite-microplaty hematite ores has decreased through depletion of natural reserves. Although the European and North American marine ooidal deposits have been geochemically well characterized in the past, their economic significance has decreased. It is the new, giant terrestrial ooidal deposits of Western Australia, which had previously received little international interest, that now represent economically significant examples of this iron ore

Sedimentary Hosted Iron Ores

type. Now, the developing Chinese economy and the predicted emergence of India will provide the impetus for the discovery and exploitation of new deposits and ore types worldwide, and will see the importance of iron ore as a foundation for economic growth to continue for years to come.

Acknowledgments The authors thank S. Hagemann, R. Morris, S. Scott, K. Taylor, and A. Trendall for their constructive reviews. We also express our gratitude to S. Morin-Ka, C. Anjou, L. Fonteneau, A. Vartesi, and T. Naughton for their tremendous help with the references and the drafting.

References Acero P, Ayora C, Torrento C, and Nieto JM (2006) The behavior of trace elements during schwertmannite precipitation and subsequent transformation into goethite and jarosite. Geochimica et Cosmochimica Acta 70: 4130. Adeleye DR (1973) Origin of ironstones, an example from the middle Niger Valley, Nigeria. Journal of Sedimentary Petrology 43: 709–727. Amorim LQ and Alkmim FF (2011) New Ore types from the caueˆ banded iron formation, Quadrilatero Ferrifero, Minas Gerais, Brazil – Responses to the growing demand. Iron Ore 2011, pp. 59–71. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Anand R and Paine M (2002) Regolith geology of the Yilgarn Craton, western Australia: Implications for exploration. Australian Journal of Earth Sciences 49: 162. Angerer T and Hagemann SG (2010) The BIF-hosted high-grade iron ore deposits in the Archean Koolyanobbing greenstone belt, western Australia: Structural control on synorogenic- and weathering-related magnetite-, hematite-, and goethite-rich iron ore. Economic Geology 105: 917–945. Ayres DE (1972) Genesis of iron-bearing minerals in banded iron formation mesobands in the dales gorge member, Hamersley group, western Australia. Economic Geology 67: 1214–1233. Baldwin AB and Gross WH (1967) Possible explanations for the localization of residual hematite ore on a Precambrian iron formation. Economic Geology 62: 95–108. Banerji AK (1977) On the Precambrian banded iron-formations and the manganese ores or the Singhbhum region, eastern India. Economic Geology 72: 90–98. Barley ME, Kerrich R, Krapez B, and Groves DI (1998) The 2.72–2.60 Ga bonzana: Metallogenic and environmental consequences of the interaction between mantle plumes, lithospheric tectonics and global cyclicity. Precambrian Research 91: 65–90. Barley ME, Pickard AL, Hagemann SG, and Folkert SL (1999) Hydrothermal origin for the 2 billion year old Mount Tom Price giant iron ore deposit, Hamersley province, western Australia. Mineralium Deposita 34: 784–789. Barro´n V, Ga´lvez N, Hochella MF Jr., and Torrent J (1997) Epitaxial overgrowth of goethite on hematite synthesized in phosphate media: A scanning force and transmission electron microscopy study. American Mineralogist 82: 1091–1100. Bateman AM (1927) Ore deposits of the Rio Tinto (Huelva) district, Spain. Economic Geology 22: 569–614. Baturin GN and Dubinchuk VT (2011) Mineralogy and chemistry of ferromanganese crusts from the Atlantic Ocean. Geochemistry International 49: 578–593. Baturin GN, Dubinchuk VT, and Rashidov VA (2012) Ferromanganese crusts from the sea of Okhotsk. Oceanology 52: 88–100. Bayer U (1989) Stratigraphic and environmental patterns of ironstone deposits. Geological Society, Special Publication 46: 105. Bekker A, Holland HD, Wang PL, et al. (2004) Dating the rise of atmospheric oxygen. Nature 427: 117–120. Bekker A, Slack JF, Planavsky N, et al. (2010) Iron formation: The sedimentary product of a complex interplay among mantle, tectonic, oceanic, and biospheric processes. Economic Geology 105: 467–508. Belevtsev YaN, Belevtsev RYa, and Siroshtan RI (1983) The Krivoy Rog Basin. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 221–251. Amsterdam: Elsevier. Berkman DA and Mackenzie DH (1998) Brockman No. 2 detritals (B2D) iron ore deposit. Monograph Series 22, pp. 375–380. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy.


Beukes NJ (1973) Precambrian iron-formations of southern Africa. Economic Geology 68: 960–1004. Beukes NJ (1983) Palaeoenvironmental setting of iron-formations in the depositional basin of the Transvaal supergroup, south Africa. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 131–209. Amsterdam: Elsevier. Beukes J, Dorland HD, Gutzmer J, Nedashi M, and Ohmoto H (2002a) Tropical laterites, life on land, and the history of atmospheric oxygen in the Paleoproterozoic. Geology 30: 491–494. Beukes NJ and Gutzmer J (2008) Origin and paleoenvironmental significance of major iron formations at the Archean-Paleoproterozoic boundary. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 5–47. Littleton, CO: Society of Economic Geologists. Beukes J, Gutzmer J, and Mukhopadhyay J (2002b) The geology and genesis of high-grade hematite iron ore deposits. Iron Ore 2002, pp. 23–29. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Beukes NJ, Gutzmer J, and Mukhopadhyay J (2003) The geology and genesis of high-grade hematite iron ore deposits. Transactions of the Institution of Mining and Metallurgy. Section B. Applied Earth Science 112: B18–B25. Beukes NJ and Klein C (1990) Geochemistry and sedimentology of a facies transition, from microbanded to granular iron-formation, in the early proterozoic Transvaal supergroup, south Africa. Precambrian Research 47: 99–139. Bigham JM, Schwertmann U, Carlson L, and Murad E (1990) A poorly crystallized oxyhydroxysulfate of iron formed by bacterial oxidation of Fe(II) in acid mine waters. Geochimica et Cosmochimica Acta 54: 2743. Bowen BB, Benison KC, Oboh-Ikuenobe FE, Strory S, and Mormile MR (2008) Active hematite concretion formation in modern acid saline lake sediments, Lake Brown, Western Australia. Earth and Planetary Science Letters 268: 52–63. Boyd T and Scott SD (1999) Two-XRD-line ferrihydrite and Fe–Si–Mn oxyhydroxide mineralization from Franklin Seamount, Western Woodlark Basin, Papua New Guinea. The Canadian Mineralogist 37: 973–990. Braddock-Rogers K (1930) The bog ore industry in South Jersey prior to 1945. Journal of Chemical Education 7: 1493–1519. Braterman PS and Cairns-Smith AG (1987) Iron photoprecipitation and the genesis of the banded iron-formations. In: Appel PWU and LaBerge GL (eds.) Precambrian Iron-Formations, pp. 215–245. Athens: Theophrastus Publ. and Pty. Co. Breuning-Madsen H, Ronsbo J, and Holst MK (2000) Comparison of the compositioni of iron pans in Danish burial mounds with bog iron and spodic material. Catena 39: 1–9. Bricker OP, Newell WL, and Simon NS (2003) Bog iron formation in the Nassawango watershed, Maryland. Gordon Conference on Catchment Science: Interactions of Hydrology, Biology and Geochemistry 7. Bubenicek L (1960) Recherches sur la constitution et la repartition des minerais de fer dans l’aalenien de lorraine. PhD Thesis, Faculte des sciences de Nancy, France. Burgos WD, Borch T, Troyer LD, et al. (2012) Schwertmannite and Fe oxides formed by biological low pH Fe(II) oxidation versus abiotic neutralization: Impact on trace metal sequestration. Geochimica et Cosmochimica Acta 76: 29–44. Burkhalter RM (1995) Ooidal ironstones and ferruginous microbialites: Origin and relation to sequence stratigraphy (Aalenian and Bajocian, Swiss Jura Mountains). Sedimentology 42: 57–74. Burton ED, Bush RT, Sullivan LA, and Mitchell DRG (2008) Schwertmannite transformation to goethite via the Fe(II) pathway: Reaction rates and implications for iron-sulfide formation. Geochimica et Cosmochimica Acta 72: 4551–4564. Butt A, Hawke P, and Flis M (2001) Of detritals, derivatives and determination? An example of detrital iron discovery. Exploration Geophysics 32: 185–188. Canavan F (1965) Iron ore deposits of Roper Bar. Geology of Australian ore deposits 1: 212–215. Carlson L and Schwertmann U (1980) Natural occurrence of feroxyhite (d’-FeOOH). Clays and Clay Minerals 28: 272–280. Carlson L and Schwertmann U (1981) Natural ferrihydrites in surface deposits from Finland and their association with silica. Geochimica et Cosmochimica Acta 45: 421–429. Chamberlin TC (1883) Geology of Winsconsin 1. Wisconsin: U. S. Geological Survey. Chauvenet R (1890) The iron resources of Colorado. Transactions of the Society of Mining Engineers of American Institute of Mining, Metallurgical and Petroleum Engineers, Incorporated (AIME) 18: 266–273. Chicarino Varaja˜o CA (1994) Evolution supergene de l’or riche en palladium de la mine de Maquine, Quadrilatere Ferrifere, Minas Gerais Bresil. PhD Thesis, Universite de droit, d’economie et des sciences d’Aix-Marseille, France. Chicarino Varaja˜o CA, Ramanaidou ER, Colin F, and Nahon D (1997) Geˆnese dos corpos de hematita compacta (‘Hard hematite ores’): Metasomatismo, sedimentac¸a˜o ou alterac¸a˜o supergenica? REM: Revista Escola Minas, Ouro Preto 50: 40–43.


Sedimentary Hosted Iron Ores

Childs CW, Downes CJ, and Wells N (1982) Hydrous iron oxide minerals with short range order deposited in a spring/stream system, Tongariro National park, New Zealand. Australian Journal of Soil Research 20: 119–129. Childs CW, Wells N, and Downes CJ (1986) Kokowai Springs, Mount Egmont, New Zealand: Chemistry and mineralogy of the ochre (ferrihydrite) deposit and analysis of the waters. Journal of the Royal Society of New Zealand 16: 85–99. Chukhrov FV (1973) On the genesis problem of thermal sedimentary iron ore deposits. Mineralium Deposita 8: 138–147. Chukhrov FV, Gorshkov AI, Berezovskaya VV, and Sivtsov AV (1982) Mineralogy of lake ores. International Geology Review 25: 833–837. Chukhrov FV, Gorshkova AI, Zvyagina BB, Yermilovaa LP, and Sapolnovaa LP (1977) Transformation of ferric compounds into iron oxides. International Geology Review 119: 766–774. Chukhrov FV, Zvyagin BB, Yermilova LP, and Gorshkov AI (1976) Mineralogical criteria in the origin of marine iron-manganese nodules. Mineralium Deposita 11: 24–32. Cloud P (1968) Atmospheric and hydrospheric evolution on the primitive earth. Science 160: 729–736. Cloud P (1973) Paleoecological significance of the banded iron-formation. Economic Geology 68: 1135–1143. Cloud P (1983) Banded iron-formation – A Gradualist’s dilemma. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems 6, pp. 401–416. Amsterdam: Elsevier. Clout JMF (1998) The effects of ore petrology on downstream processing performance. Mine to Mill. Melbourne: The Australasian Institute of Mining and Metallurgy. Clout JMF (2002) Upgrading processes in BIF-derived iron ore deposits: Implications for Ore genesis and downstream mineral processing. Iron Ore 2002, pp. 237–242. Melbourne, VIC: Australasian Institute of Mining and Metallurgy. Clout J (2006) Iron formation-hosted iron ores in the Hamersley province of western Australia. Applied Earth Science 115: 115–125. Clout J and Simonson BM (2005) Precambrian iron formations and iron formation-hosted iron ore deposits. Economic Geology 100: 643–679. Collin PY, Loreau JP, and Courville P (2005) Depositional environments and iron ooid formation in condensed sections (Callovian-Oxfordian, south-eastern Paris Basin, France). Sedimentology 52: 969. Cope IL, Wilkinson JJ, Boyce AJ, et al. (2008) Genesis of the pic fon iron oxide deposit, Simandou Range, Republic of Guinea, west Africa. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore 15, pp. 339–360. Littleton, CO: Society of Economic Geologists. Cornell RM, Giovanoli R, and Schindler PW (1987) Effect of silicate species on the transformation of ferrihydrite into goethite and hematite in alkaline media. Clays and Clay Minerals 35: 21–28. Cornell RM and Schwertmann U (2003) The Iron Oxides: Structure, Properties, Reactions, Occurrences and Uses. Weinheim: John Wiley and Sons. Crerar DA, Knox GW, and Means JL (1979) Biogeochemistry of bog iron in the New Jersey Pine barrens. Chemical Geology 24: 111–135. Crerar DA, Means JL, Yuretich RF, et al. (1981) Hydrogeochemistry of the New Jersey coastal plain 2. Transport and deposition of iron, aluminum, dissolved organic matter, and selected trace elements in stream, ground- and estuary water. Chemical Geology 33: 23–44. Czaja S (2001) Mining and hydrological transformations in Upper Silesia from the fifteenth to the nineteenth century. The Geographical Journal 167: 57–71. Dake CL and Rolla MA (1916) The formation an distribution of bog iron ore deposits. Transactions of the American Institute of Mining Engineers 53: 116–124. Dalstra H, Gill T, Faragher A, Scott B, and Kakebeeke V (2010) Channel iron deposits, a major new district around the Caliwingina Creek, central Hamersley Ranges, Western Australia. Applied Earth Science 119: 12–20. Dalstra HJ and Guedes SC (2004) Giant hydrothermal deposits with Mg-Fe metasomatism: A comparison of the Carajas, Hamersley, and other ores. Economic Geology 99: 1793–1800. Danisˇ´ık M, Ramanaidou ER, Evans NJ, et al. (2011) (U-Th)/He geochronology of channel iron deposits, Robe River, Hamersley Province, Australia – Implications for ore genesis. Iron Ore 2011, pp. 83–86. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Das S, Hendry MJ, and Essilfie-Dughan J (2011) Transformation of two-line ferrihydrite to goethite and hematite as a function of pH and temperature. Environmental Science and Technology 45: 268–275. Davy R (1975) A Geochemical Study of a Dolomite-BIF Transition in the Lower Part of the Hamersley Group. Perth, WA: Geological Survey of Western Australia D8. De Geyter G, Vandenberghe RE, Verdonck L, and Stoops G (1985) Mineralogy of holocene bog-iron ore in northern Belgium. Neues Jahrbuch fuer Mineralogie Abhandlungen 153: 1–17.

De Laeter JR, Bohlke JK, De Bievre P, et al. (2003) Atomic weights of the elements: Review 2000 – (IUPAC technical report). Pure and Applied Chemistry 75: 683–800. Derry LA and Jacobsen SB (1990) The chemical evolution of Precambrian seawater: Evidence from REEs in banded iron formations. Geochimica et Cosmochimica Acta 54: 2965–2977. Dimroth E (1976) Aspects of the sedimentary petrology of cherty iron-formation. In: Wolf KH (ed.) Handbook of Strata-Bound and Stratiform Ore Deposits: Part 2. Regional Studies and Specific Deposits, vol. 7, pp. 203–254. Amsterdam: Elsevier. Dimroth E (1977) Facies models 5. Models of physical sedimentation of iron formations. Geoscience Canada 4: 23–30. Dimroth E and Chauvel JJ (1973) Petrography of the Sokoman iron formation in part of the Central Labrador Trough, Quebec, Canada. Geological Society of America Bulletin 84: 111–134. Doelsch E, Rose J, Masion A, et al. (2000) Speciation and crystal chemistry of iron (III) chloride hydrolyzed in the presence of SiO4 ligands. 1. An Fe K-edge EXAFS study. Langmuir 16: 4726–4731. Dorr JVN (1965) Nature and origin of the high-grade hematite ores of Minas Gerais, Brazil. Economic Geology 60: 1–46. Dorr JVN (1969) Physiographic, stratigraphic and structural development of the Quadrilaˆtero Ferrıˆfero, Minas Gerais, Brazil. U. S. Geological Survey Professional Paper, 641-A. Dorr JVN (1973) Iron formations of South America. Economic Geology 68: 1005–1022. Drever JI (1974) Geochemical model for the origin of Precambrian banded iron formations. Geological Society of America Bulletin 85: 1099–1106. Dukino RD, England BM, and Kneeshaw M (2000) Phosphorus distribution in BIF-derived iron ores of Hamersley Province, Western Australia. Applied Earth Science 109: 168–176. Dunn JA (1937) Hydrothermal leaching of iron ores: Discussion. Economic Geology 32: 976–977. Dwyer R, Rea S, Bruckard W, and Holmes R (2011) Bio-beneficiation of Australian iron ores – Potential applications of indigenous bacteria for flotation, flocculation and phosphorous removal. Iron Ore 2011, pp. 397–402. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Edwards AB (1958) Oolitic iron formations in northern Australia. International Journal of Earth Sciences 47: 668–682. Edwards CI, Fisher-White MJ, Lovel RR, and Sparrow GJ (2011) Removal of phosphorous from Australian iron ores. Iron Ore 2011, pp. 403–412. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Essalhi M, Sizaret S, Barbanson L, et al. (2011) A case study of the internal structures of gossans and weathering processes in the Iberian Pyrite Belt using magnetic fabrics and paleomagnetic dating. Mineralium Deposita 46: 981–999. Evans LJ, Rowsell JG, and Aspinall JD (1978) Massive iron formations in some gleysolic soils of southwestern Ontario. Canadian Journal of Soil Science 58: 391–395. Ewers WE and Morris RC (1981) Studies of the Dales Gorge Member of the Brockman iron formation, Western Australia. Economic Geology 76: 1929–1953. Fajardo M, Pe´rez Alca´zar GA, Moreira AM, and Speziali NL (2004) Mo¨ssbauer and XRD comparative study of host rock and iron rich mineral samples from Paz de Rio iron ore mine in Columbia. Hyperfine Interactions 156: 395–402. Ferenczi P (2001) Iron ore, manganese and bauxite deposits of the Northern Territory. Department of Mines and Energy. Figueiredo e Silva RC, Lobato LM, Rosiere CA, et al. (2008) A hydrothermal origin for the jaspilite-hosted, giant Serra Norte iron ore deposits in the Carajas Mineral Province, Para state, Brazil. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 255–290. Littleton, CO: Society of Economic Geologists. Finlayson AM (1910) The pyritic deposits of Huelva, Spain: Part II, Ore deposits. Economic Geology 5: 403–437. Fischer WR (1988) Microbiological reactions of iron in soils. In: Stucki JW, Goodman BA, and Schwertmann U (eds.) Iron in Soils and Clay Minerals, pp. 715–748. Boston, MA: D. Reidel Publishing Company. Fischer WR and Schwertmann U (1975) The formation of hematite from amorphous iron(III) hydroxide. Clays and Clay Minerals 23: 33–37. Fischer-White MJ, Lovel RR, and Sparrow GJ (2009) Phosphorus removal from iron Ore with a low temperature heat treatment. Iron Ore 2009, pp. 249–254. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Fitzpatrick RW, Fritsch E, and Self PG (1996) Interpretation of soil features produced by ancient and modern processes in degraded landscapes: V. Development of saline sulfidic features in non-tidal seepage areas. Geoderma 69: 1–29. Flint D and Preston DJ (2010) WA Iron Ore Overview. Western Australia Department of Mines and Petroleum and Western Australia Department of State Development.

Sedimentary Hosted Iron Ores

Fortin D and Langley S (2005) Formation and occurrence of biogenic iron-rich minerals. Earth-Science Reviews 72: 1–19. Franc¸ois LM (1986) Extensive deposition of banded iron formations was possible without photosynthesis. Nature 320: 352–354. Frei R, Dahl PS, Duke EF, et al. (2008) Trace element and isotopic characterization of Neoarchean and Paleoproterozoic iron formations in the Black Hills (South Dakota, USA): Assessment of chemical change during 2.9–1.9 Ga deposition bracketing the 2.4–2.2 Ga first rise of atmospheric oxygen. Precambrian Research 162: 441–474. Freyssinet P, Butt CRM, Morris RC, and Piantone P (2005) Ore-forming processes related to lateritic weathering. Economic Geology 100: 681–722. Fryer BJ (1977) Trace element geochemistry of the Sokoman Iron Formation. Canadian Journal of Earth Sciences 14: 1598–1610. Gaines RV, Skinner HCW, Foord EE, Mason B, and Rosenzweig A (1997) Dana’s New Mineralogy. New York: John Wiley and Sons. Gale I, Lynch M, Peranec J, Van Anen J, and Ltd BI (2012) Nullagine iron ore project. The AusIMM News. Galvez N, Barron V, and Torrent J (1999) Effect of phosphate on the crystallization of hematite, goethite, and lepidocrocite from ferrihydrite. Clays and Clay Minerals 47: 304. Ghiorse WC (1984) Biology of iron- and manganese-depositing bacteria. Annual Review of Microbiology 38: 515. Golubovskaya EV (2001) Facies and geochemical features of the iron ore complex of the Kerch Peninsula. Lithology and Minerals Resources 36: 224–235. Golubovskaya EV (2003) Some geochemical features of iron ores from the Lisakov deposit. Lithology and Minerals Resources 38: 275–281. Goode ADT, Hall WDM, and Bunting JA (1983) The Nabberu basin of Western Australia. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 295–323. Amsterdam: Elsevier. Gooden JEA, Walker WM, and Allen RJ (1974) Amdephos – A chemical process for dephosphorisation of iron ore. Proceedings of the Second National Chemical Engineering Conference, pp. 38–49. St Lucia: University of Queensland. Goodwin AM (1982) Distribution and origin of precambrian banded iron formation. In: Inda HAV and Duarte FB (eds.) International Symposium on Archean and Early Proterozoic Geologic Evolution and Metallogenesis, pp. 457–462. Sao Paulo: Sociedade Brasileira de Geologia. Greenwood NN and Gibb TC (1971) Mossbauer Spectroscopy. London: Chapman and Hall. Griffin AC (1980) Structure and iron-ore deposition in the Archaean Koolyanobbing greenstone belt, Western Australia. In: Glover JE and Groves DI (eds.) Archaean Geology: Second International Symposium, pp. 429–438. Sydney, NSW: Geological Society of Australia. Gross GA (1965) Origin of Precambrian iron formations. Economic Geology 60: 1063–1065. Gross GA (1980) A classification of iron formations based on depositional environments. The Canadian Mineralogist 18: 215–222. Gross GA (1991) Geochemical data on iron-formations, pp. 57–59. Open-File Report, Geological Survey of Canada. Gross GA and McLeod CR (1980) A preliminary assessment of the chemical composition of iron formations in Canada. The Canadian Mineralogist 18: 223–229. Gross GA, Weber R, and Guerrero JC (1983) Tectonic systems and the deposition of iron-formation. Precambrian Research 20: 171–187. Gross GA and Zajac IS (1983) Iron-formation in fold belts marginal to the Ungava craton. In: Trendall AF and Morris RC (eds.) Iron-formation: Facts and Problems, vol. 6, pp. 253–294. Amsterdam: Elsevier. Guedes SC, Rosiere CA, Barley ME, and Lobato L (2002) Carbonate alteration associated with the Carajas high-grade hematite deposits, Brazil. Iron Ore 2002, pp. 62–66. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Guild PW (1953) Iron deposits of the Congonhas district, Minas Gerais, Brazil. Economic Geology 48: 639–676. Gutzmer J, Chisonga BC, Beukes NJ, and Mukhopadhyay J (2008) The geochemistry of banded iron formation-hosted high-grade hematite-martite iron ores. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 157–183. Littleton, CO: Society of Economic Geologists. Haest M, Cudahy T, Laukamp C, and Gregory S (2012a) Quantitative mineralogy from infrared spectroscopic data: II. Three-dimensional mineralogical characterization of the Rocklea Channel iron deposit, Western Australia. Economic Geology 107: 229. Haest M, Cudahy T, Laukamp C, and Gregory S (2012b) Quantitative mineralogy from infrared spectroscopic data: II. Validation of mineral abundance and composition scrpit at the Rocklea Dome channel iron deposit in Western Australia. Economic Geology 107: 209–228.


Haest M, Cudahy T, Laukamp C, et al. (2011) Characterisation of bedded and channel iron Ore deposit using CSIRO’s HyloggingTM systems. Iron Ore 2011, pp. 249–255. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Hageman S, Barley ME, Folkert SL, Yardley BW, and Banks DA (1999) A hydrothermal origin for the giant Tom Price iron ore deposit. In: Hughes FE (ed.) Mineral Deposits, Processes to Processing, pp. 41–44. Rotterdam: Balkema. Halbach P (1976) Mineralogical and geochemical investigations of Finnish lake ores. Bulletin of the Geological Society of Finland 48: 33–42. Hallsworth CR and Knox RWOB (1999) Classification of sediments and sedimentary rocks. British Geological Survey 3. Hamzaoui A, Mgaisi A, Megriche A, and Maaoui ME (2002) Kinetic study of goethite formation from ferrihydrite in alkaline medium. Industrial and Engineering Chemistry Research 41: 5226–5231. Han TM (2004) Effect of mineralogy on processing of low grade iron ores from the Negaunee iron-formation on Marquette range of the Lake Superior District. Proceedings and Abstracts – Institute on Lake Superior Geology. Meeting, pp. 66–67. Madison, WI: Institute on Lake Superior Geology. Hansel CM, Learman DR, Lentini CJ, and Ekstrom EB (2011) Effect of adsorbed and substituted Al on Fe(II)-induced mineralization pathways of ferrihydrite. Geochimica et Cosmochimica Acta 75: 4653. Harder EC (1919) Iron-Depositing Bacteria and Their Geologic Relations. Washington: U. S. Geological Survey 113. Harder EC and Chamberlin RT (1915) The geology of central Minas Gerais, Brazil. Journal of Geology 23: 385–424. Harms JE and Morgan BD (1964) Pisolitic limonite deposits in Northwest Australia. Proceedings of the Australasian Institute of Mining and Metallurgy 91–124. Harmsworth RA, Kneeshaw M, Morris RC, Robinson CJ, and Shrivastava PK (1990) BIF-derived iron ores of the Hamersley Province. In: Hughes FE (ed.) Geology of the Mineral Deposits of Australia and Papua New Guinea: Monograph 14, pp. 617–642. Parkville, VIC: The Australasian Institute of Mining and Metallurgy. Heikoop JM, Tsujita CJ, Risk MJ, Tomascik T, and Mah AJ (1996) Modern iron ooids from a shallow-marine volcanic setting: Mahengetang, Indonesia. Geololy 24: 759–762. Heim JA, Vasconcelos PM, Shuster DL, Farley KA, and Broadbent GC (2006) Dating paleochannel iron ore by (U–Th)/He analysis of supergene goethite, Hamersley Province, Australia. Geology 34: 173–176. Henmi T, Wells N, Childs CW, and Parfitt RL (1980) Poorly-ordered iron-rich precipitates from springs and streams on andesitic volcanoes. Geochimica et Cosmochimica Acta 44: 365–372. Hoefs J, Mueller G, and Schuster AK (1982) Polymetamorphic relations in iron ores from the Iron Quadrangle, Brazil: The correlation of oxygen isotope variations with deformation history. Contributions to Mineralogy and Petrology 79: 241–251. Holland HD (1984) The Chemical Evolution of the Atmosphere and Oceans, pp. 374–407. Princeton, NJ: Princeton University Press. Holland HD (2005) Sedimentary mineral deposits and the evolution of Earth’s near-surface environments. Economic Geology 100: 1489–1509. Horon O (1977) Les gisements de fer de la France. Les Gisements de fer d’Europe. In: Walther HW and Zitzmann A (eds.) The Iron Ore Deposits of Europe and Adjacent Areas, vol. 1, pp. 143–159. Hannover: Federal Institute for Geosciences and Natural Resources. Huston DL and Logan GA (2004) Barite, BIFs and bugs: Evidence for the evolution of the Earth’s early hydrosphere. Earth and Planetary Science Letters 220: 41–55. Ilyin AV (2009) Neoproterozoic banded iron formations. Lithology and Mineral Resources 44: 78–86. Inman OL (1923) Iron-depositing bacteria. Science 6: 13. Ionkov K, Gaydardzhiev S, Correa de Araujo A, et al. (2011) Dephosphorization of limonitic concentrate by roasting, acid leaching and magnetic separation. Iron Ore 2011, pp. 445–452. Melbourne, VIC: Australasian Institute of Mining and Metallurgy. Jambor JL and Dutrizac JE (1998) Occurrence and constitution of natural and synthetic ferrihydrite, a widespread iron oxyhydroxide. Chemical Reviews 98: 2549–2585. James HL (1954) Sedimentary facies of iron-formation. Economic Geology 49: 235–293. James HL (1966) Chemistry of the Iron-Rich Sedimentary Rocks. Reston, VA: U. S. Geological Survey. James HL (1983) Distribution of banded iron-formation in space and time. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 471–490. Amsterdam: Elsevier. Johnson TC (1984) Sedimentation in large lakes. Annual Review of Earth and Planetary Sciences 12: 179–204.


Sedimentary Hosted Iron Ores

Johnson RC, Scott GW, and Lukey HM (2007) Implications of mineraligy, gain size and texture on liberation and pellet quality of Great Lakes Iron Ore. Iron Ore 2007, pp. 109–111. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Jones AM, Collins RN, Rose J, and Waite TD (2009) The effect of silica and natural organic matter on the Fe(II)-catalysed transformation and reactivity of Fe(III) minerals. Geochimica et Cosmochimica Acta 73: 4409. Kaczorek D and Sommer M (2003) Micromorphology, chemistry, and mineralogy of bog iron ores from Poland. Catena 54: 393–402. Kalugin AS (1973) Geology and genesis of the Devonian banded iron-formation in Altai, Western Siberia and eastern Kazakhstan. Genesis of Precambrian Iron and Manganese Deposits, pp. 159–195. Paris: UNESCO. Kepert D, Clarke N, Simpson C, and Edwards D (2010) Discovery of the Solomon iron deposits, Hamersley Province, Western Australia. Applied Earth Science 119: 21–27. Kholodov VN and Butuzova GY (2001) Problems of iron and phosphorus geochemistry in the Precambrian. Lithology and Minerals Resources 36: 291–302. Kimberley MM (1978) Paleoenvironmental classification of iron formations. Economic Geology 73: 215–229. Kimberley MM (1980) Origin of oolitic iron formations – Reply. Journal of Sedimentary Petrology 50: 299–302. Kimberley MM (1981) Oolitic iron formations. In: Wolf KH (ed.) Handbook of Strata-Bound and Stratiform Ore Deposits: Part III. Regional Studies and Specific Deposits, vol. 9, pp. 25–76. Amsterdam: Elsevier Scientific Publishing Company. Kimberley MM (1989a) Exhalative origins of iron formations. Ore Geology Reviews 5: 13–145. Kimberley MM (1989b) Nomenclature for iron formations. Ore Geology Reviews 5: 1–12. Kimberley MM (1994) Debate about ironstone: Has solute supply been surficial weathering, hydrothermal convection, or exhalation of deep fluids? Terra Nova 6: 116–132. Klein C (1983) Diagenesis and metamorphism of Precambrian banded iron-formations. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 417–469. Amsterdam: Elsevier. Klein C (2004) Geochemistry and mineralogy of Neoproterozoic banded iron formation and their associated Fe and Mn deposit, with reference to the Griqualand West deposit, South Africa. Journal of African Earth Sciences 30: 1233–1244. Klein C (2005) Some Precambrian banded iron-formations (BIFs) from around the world: Their age, geologic setting, mineralogy, metamorphism, geochemistry, and origins. American Mineralogist 90: 1473–1499. Klein C and Beukes NJ (1989) Geochemistry and sedimentology of a facies transition from limestone to iron-formation deposition in the early Proterozoic Transvaal Supergroup, South Africa. Economic Geology 84: 1733–1774. Klein C and Beukes NJ (1992) Proterozoic iron-formations. In: Condie KC (ed.) Proterozoic Crustal Evolution, vol. 10, pp. 383–418. Amsterdam: Elsevier. Klein C and Beukes NJ (1993) Sedimentology and geochemistry of the glaciogenic late Proterozoic Rapitan iron-formation in Canada. Economic Geology 88: 542–565. Klein C and Gole MJ (1981) Mineralogy and petrology of parts of the Marra Mamba iron formation, Hamersley Basin, Western Australia. American Mineralogist 66: 507–525. Klein C and Ladeira EA (2002) Petrography and geochemistry of the least altered banded iron-formation of the Archean Caraja´s Formation, northern Brazil. Economic Geology 97: 643–651. Kneeshaw M (1975) Mount Whaleback iron orebody, Hamersley iron province. Geology of Australia and Papua New Guinea: Monograph 5, vol. 1, pp. 910–916. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Knorr KH and Blodau C (2007) Controls on schwertmannite transformation rates and products. Applied Geochemistry 22: 2006–2015. Kogbe CA (1978) Origin and composition of the ferruginous oolites and laterites of North-Western Nigeria. Geology Rundsch 67: 662–674. Kokal HR, Farias JM, and Chop SV (2008) A roasting-leaching demonstration plant for removal of phosphorus from iron ore. In: Young C, Taylor PR, Anderson CG, and Choi Y (eds.) Hydrometallurgy 2008 – 6th International Symposium, pp. 431–438. Littleton, CO: Society for Mining, Metallurgy and Exploitation Inc. (SME). Konhauser KO (1998) Diversity of bacterial iron mineralization. Earth-Science Reviews 43: 91–121. Konhauser KO, Hamade Y, Morris RC, et al. (2002) Could the bacteria have formed the Precambrian banded iron formations? Geology 30: 1079–1082. Konhauser KO, Kappler A, and Roden EE (2011) Iron in microbial metabolism. Elements 7: 89–93. Krapez B (1993) Sequence stratigraphy of the Archaean supracrustal belts of the Pilbara Block, Western Australia. Precambrian Research 60: 1–45.

Krapez B, Barley ME, and Pickard AL (2003) Hydrothermal and resedimented origins of the precursor sediments to banded iron formation: Sedimentological evidence from the early Palaeoproterozoic Brockman Supersequence of Western Australia. Sedimentology 50: 979–1011. Krishnan MS (1954) The iron ore resources of Asia. Indian Mining Journal 2: 1–9. Kullerud G, Donnay G, and Donnay JDH (1969) Omission solid solution in magnetite: Kenotetrahedral magnetite. Zeitschrift fuer Kristallographie 128: 1–17. Kumpulainen S, Carlson L, and Raisanen ML (2007) Seasonal variations of ochreous precipitates in mine effluents in Finland. Applied Geochemistry 22: 760. Laajoki K and Lavikainen S (1977) Rare-Earth elements in the Archean iron formation and associated schists in Ukkolanvaara, Ilomantsi, SE Finland. Bulletin of the Geological Society of Finland 49(1): 105–123. Laberge GL, Robbins EI, and Han TM (1987) A model for the biological precipitation of Precambrian iron-formations: A. Geological evidence. In: Appel PWU and LaBerge GL (eds.) Precambrian Iron-Formations, pp. 69–96. Athens: Theophrastus Publ. and Pty. Co. Landuydt CJ (1990) Micromorphology of iron minerals from bog ores of the Belgian Campine area. In: Douglas LA (ed.) Soil Micro-Morphology: A Basic and Applied Science, pp. 289–294. New York: Sub-commision of soil micromorphology of International Society of Soil Science. Larese-Casanova P, Haderlein SB, and Kappler A (2010) Biomineralization of lepidocrocite and goethite by nitrate-reducing Fe(II)-oxidizing bacteria: Effect of pH, bicarbonate, phosphate, and humic acids. Geochimica et Cosmochimica Acta 74: 3721. Lascelles D (2002) A new look at old rocks – And alternative model for the origin of in situ iron ore deposits derived from banded iron-formation. Iron Ore 2002, pp. 107–126. Melbourne, VIC: Australasian Institute of Mining and Metallurgy. Lascelles D (2007) A possible origin of pisolitic iron ore deposits from raindrops and sheetwash in a semi-arid environment. Iron Ore 2007, pp. 113–121. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Lascelles D (2008) Banded iron formation to high grade iron ore – The fallacy of supergene enrichment. Pacrim Congress 2008, pp. 367–372. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Lemoalle J and Dupont B (1973) Iron-bearing oolites and the present conditions of iron sedimentation in Lake Chad (Africa). In: Amstutz GC and Bernard AJ (eds.) Ores in Sediments, pp. 167–178. New York: Springer. Lepp H (1987) Chemistry and origin of Precambrian iron-formations. In: Appel PWU and LaBerge GL (eds.) Precambrian Iron-Formations, pp. 3–30. Athens: Theophrastus Publ. and Pty. Co. Li ZX, Powell CM, and Bowman R (1993) Timing and genesis of Hamersley iron-ore deposits. Exploration Geophysics 24: 631–636. Liu H, Ma M, Qin M, Yang L, and Wei Y (2010) Studies on the controllable transformation of ferrihydrite. Journal of Solid State Chemistry 183: 2045. Lobato LM, Figueiredo e Silva RC, Hageman S, Thorne W, and Zucchetti M (2008) Hypogene alteration associated with high-grade banded iron formation-related iron Ore. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 107–128. Littleton, CO: Society of Economic Geologists. MacLeod WN (1966) The geology and iron deposits of the Hamersley Range area, Western Australia, p. 117. Perth: Geological Survey of Western Australia Bulletin. MacPhail MK and Stone MS (2004) Age and palaeoenvironmental constraints on the genesis of the Yandi channel iron deposits, Marillana Formation, Pilbara, northwestern Australia. Australian Journal of Earth Sciences 51: 497–520. Madon MBH (1992) Depositional setting and origin of berthierine oolitic ironstones in the lower Miocene Terengganu Shale, Tenggol Arch, offshore peninsular Malaysia. Journal of Sedimentary Petrology 62: 899–916. Marescotti P, Carbone C, Comodi P, Frondini F, and Lucchetti G (2012) Mineralogical and chemical evolution of ochreous precipitates from the Libiola Fe-Cu-sulfide mine (Eastern Liguria, Italy). Applied Geochemistry 27: 577. Martin DM, Li ZX, Nemchin AA, and Powell CM (1998) A pre-2.2 Ga age for giant hematite ores of the Hamersley province, Australia. Economic Geology 93: 1084–1090. Masuda A and Ikeuchi Y (1979) Lanthanide tetrad effect observed in marine environment. Geochemical Journal 13: 19–22. Maynard JB (1986) Geochemistry of oolitic iron ores, an electron microprobe study. Economic Geology 81: 1473–1483. McClung CR (2006) Basin Analysis of the Mesoproterozoic Bushmanland Group of the Namaqua Metamorphic Province, South Africa. PhD Thesis, University of Johannesburg, South Africa. McConchie D (1987) The Geology and Geochemistry of the Joffre and Whaleback Shale Members of the Brockman Iron Formation, Western Australia. Athens: Theophrastus Publ. and Pty. Co.

Sedimentary Hosted Iron Ores

McGregor F, Ramanaidou ER, and Wells MA (2010) Phanerozoic ooidal ironstone deposits – Generation of potential exploration targets. Applied Earth Science 119: 60–64. McMillan SG and Schwertmann U (1998) Morphological and genetic relations between siderite, calcite and goethite in a Low Moor Peat from southern Germany. European Journal of Soil Science 49: 283–293. Melfi AJ, Trecases JJ, Carvalho A, et al. (1988) The lateritic ore deposits of Brazil. Sciences Ge´ologiques, Bulletin 41: 5–36. METS (2011) Iron Ore Processing Course, Mineral Engineering Technical Services Pty Ltd. (METS), Cranston L (presenter), Perth. Miyano T and Klein C (1983) Conditions of Riebeckite formation in the iron-formation of the Dales Gorge Member, Hamersley Group, Western Australia. American Mineralogist 68: 517–529. Mohapatra KB, Jena S, Mahanta K, and Mishra P (2008) Goethite morphology and composition in banded iron formation, Orissa, India. Resource Geology 58: 325–332. Moore EJ (1910) The occurrence and origin of some bog iron deposits in the district of Thunder Bay, Ontario. Economic Geology 5: 528–537. Morey GB (1983) Animikie basin, Lake Superior region, USA. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 13–67. Amsterdam: Elsevier. Morris RC (1980) A textural and mineralogical study of the relationship of iron ore to banded iron-formation in the Hamersley iron province of Western Australia. Economic Geology 75: 184–209. Morris RC (1983) Supergene alteration of banded iron-formation. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 513–534. Amsterdam: Elsevier. Morris RC (1985) Genesis of iron ore in banded iron-formation by supergene and supergene-metamorphic processes: A conceptual model. In: Wolf KH (ed.) Handbook of Strata-Bound and Stratiform Ore Deposits: Part II. Regional Studies and Specific Deposits, vol. 13, pp. 73–235. Amsterdam: Elsevier. Morris RC (1993) Genetic modelling for banded iron-formation of the Hamersley Group, Pilbara Craton, Western Australia. Precambrian Research 60: 243–286. Morris RC (1994) Detrital iron deposits of the Hamersley Province, 76R. CSIRO Exploration Geoscience Restricted Report. Morris RC (2012) Microplaty hematite – Its varied nature and genesis. Australian Journal of Earth Sciences 59: 411–434. Morris RC and Fletcher AB (1987) Increased solubility of quartz following ferrous-ferric iron reactions. Nature 330: 558–561. Morris RC and Horwitz RC (1983) The origin of the iron-formation-rich Hamersley Group of Western Australia – Deposition on a platform. Precambrian Research 21: 273–297. Morris RC and Kneeshaw M (2011) Genesis modelling for the Hamersley BIF-hosted iron ores of Western Australia – A critical review. Australian Journal of Earth Sciences 58: 417–451. Morris RC and Ramanaidou ER (2007) Genesis of the channel iron deposits (CID) of the Pilbara region, Western Australia. Australian Journal of Earth Sciences 54: 733–756. Morris RC and Trendall AF (1988) A model for the deposition of the microbanded Precambrian iron formations – Discussion. American Journal of Science 288: 664–669. Mu¨cke A and Farshad F (2005) Whole-rock and mineralogical composition of Phanerozoic ooidal ironstones: Comparison and differentiation of types and subtypes. Ore Geology Reviews 26: 227–262. Mukhopadhyay J, Gutzmer J, Beukes J, and Bhattacharya HN (2008) Geology and genesis of the major banded iron formation – Hosted high-grade iron ore deposits of India. In: Hagemann S, Rosiere C, Gutzmer J, and Beukes NJ (eds.) Banded Iron Formation-Related High-Grade Iron Ore, vol. 15, pp. 291–316. Littleton, CO: Society of Economic Geologists. Mu¨ller G and Fo¨rstner U (1973) Recent iron ore formation in Lake Malawi, Africa. Mineralium Deposita 8: 278–290. Murad E (1982) The characterization of geothite by Mo¨ssbauer spectroscopy. American Mineralogist 67: 1007–1011. Murphy PJ, Posner AM, and Quirk JP (1976) Characterisation of hydrolysed ferric ion solutions: A comparison of the effects of various anions on the solutions. Journal of Colloid and Interface Science 56: 312–319. Nahon D, Carozzi AV, and Parron C (1980) Lateritic weathering as a mechanism for the generation of ferruginous ooids. Journal of Sedimentary Petrology 50: 1287–1298. Neumann-Redlin C, Walther HW, and Zitzmann A (1977) The iron Ore deposits of the Federal Republic of Germany. In: Walther HW and Zitzmann A (eds.) The Iron Ore Deposits of Europe and Adjacent Areas, vol. 1, pp. 165–186. Hannover: Federal Institute for Geosciences and Natural Resources.


Nickel EH and Nichols MC (1991) Mineral Reference Manual. New York: Van Nostrand Reinhold. Olivarez AM and Owen RM (1989) REE/Fe variations in hydrothermal sediments: Implications for the REE content of seawater. Geochimica et Cosmochimica Acta 53: 757–762. Paikaray S and Peiffer S (2010) Dissolution kinetics of sulfate from Shwertmannite under variable pH conditions. Mine Water and the Environment 29: 263–269. Pedersen HD, Postma D, Jakobsen R, and Larsen O (2005) Fast transformation of iron oxyhydroxides by the catalytic action of aqueous Fe(II). Geochimica et Cosmochimica Acta 69: 3967–3977. Peretyazhko T, Zachara JM, Boily JF, et al. (2009) Mineralogical transformations controlling acid mine drainage chemistry. Chemical Geology 262: 169–178. Petra´nek J and Van Houten FB (1997) Phanerozoic Ooidal Ironstones, vol. 7. Prague: Cesky Geologicky Ustav. Pinney N, Kubicki JD, Middlemiss DS, Grey CP, and Morgan D (2009) Density functional theory study of ferrihydrite and related Fe-oxyhydroxides. Chemistry of Materials 21: 5727–5742. Piper DZ and Graef PA (1974) Gold and rare-earth elements in sediments from the East Pacific Rise. Marine Geology 17: 287–297. Pires FRM (1995) Textural and mineralogical variations during metamorphism of the Proterozoic Itabira iron formation in the Quadrilatero Ferrifero, Minas Gerais, Brazil. Anais da Academia Brasileira de Cieˆncias 67: 77–105. Pires FRM, Pasin DAB, Bitencourt BR, and Pilotto D (2005) Origin of the hard hematite deposits of Quadrilatero Ferrifero, Minas Gerais, Brazil: Field, textural and fluid inclusion studies. Iron Ore 2005, pp. 123–128. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Pomerene JB (1964) Geology and ore deposits of the Belo Horizonte, Ibirite, and Macacos Quadrangles, Minas Gerais, Brazil. Reston, VA: U. S. Geological Survey. Postma D (1977) The occurrence and chemical composition of recent Fe-rich mixed carbonates in a river bog. Journal of Sedimentary Petrology 47: 1089–1098. Powell CM, Oliver NHS, Li ZX, Martin DM, and Ronaszeki J (1999) Synorogenic hydrothermal origin for giant Hamersley iron oxide ore bodies. Geology 27: 175–178. Preat A, Mamet B, De Ridder C, Boulvain F, and Gillan D (2000) Iron bacterial and fungal mats, Bajocian stratotype (Mid-Jurassic, northern Normandy, France). Sedimentary Geology 137: 107. Preston LJ, Shuster J, Fernandeza-Remolar D, et al. (2011) The preservation and degradation of filamentous bacteria and biomolecules within iron oxide deposits at Rio Tinto, Spain. Geobiology 9: 233–249. Ramanaidou ER (1989) Genese d’un gisement lateritique. Evolution supergene des itabirites Proterozoiques de la mine de fer de Capanema. PhD Thesis, Poitiers, France. Ramanaidou ER (2009) Genesis of lateritic iron ore from banded iron-formation in the Capanema mine (Minas Gerais, Brazil). Australian Journal of Earth Sciences 56: 605–620. Ramanaidou ER, Horwitz RC, and Morris RC (1991) Channel iron deposits, 62R. CSIRO Exploration Geoscience Restricted Report. Ramanaidou ER and Morris RC (2010) Comparison of supergene mimetic and supergene lateritic iron ore deposits. Applied Earth Science 119: 56–59. Ramanaidou ER, Morris RC, and Horwitz RC (2003) Channel iron deposits of the Hamersley Province, Western Australia. Australian Journal of Earth Sciences 50: 669–690. Ramanaidou E, Nahon D, Decarreau A, and Melfi AJ (1996) Hematite and goethite from duricrusts developed by lateritic chemical weathering of precambrian banded iron formations, Minas Gerais, Brazil. Clays and Clay Minerals 44: 22–31. Ramanaidou ER, Wells M, Belton D, Verral M, and Ryan C (2008) Mineralogical and microchemical methods for the characterization of high-grade BIF derived iron ore. Reviews in Economic Geology 15: 129–156. Raymond R (1986) Out of the Fiery Furnace: The Impact of Metals on the History of Mankind. University Park, PA: The Pennsylvania State University Press. Richardson K (2004) Assessment of the clear hills iron deposits Bad Heart Formation. Energy Resources Conservation Board, Alberta Geological Survey. Ridley M (1999) Evidence for the Hydrothermal Origin of Iron Ore, Southern Ridge, Mt. Tom Price, Western Australia. BSc, University of Western Australia, Australia. Rios FJ, Rosiere C, and Anonymous (2004) Petrographic studies in opaque minerals (hematite, martite) using infrared microscopy. In: 32nd International Geological Congress, p. 184. Italy: International Geological Congress. Robbins EI, Laberge GL, and Schmidt RG (1987) A model for the biological precipitation of Precambrian iron formations: B. Morphological evidence and modern analogs. In: Appel PWU and LaBerge GL (eds.) Precambrian Iron-Formations, pp. 97–139. Athens: Theophrastus Publ. and Pty. Co.


Sedimentary Hosted Iron Ores

Roller S, Hornung J, and Hinderer M (2010) Middle Miocene to Pleistocene sedimentary record of rift evolution in the southern Albert Rift (Uganda). International Journal of Earth Sciences 99: 1643–1661. Rosiere CA and Chemal F (1991) Textural and structural aspects of iron ores from Iron Quadrangle, Brazil. Source, Transport and Deposition of Metals: Proceeding of the 25 Years Society for Geology Applied to Mineral Deposits Anniversary Meeting, Balkema, Nancy. Rosiere CA and Rios FJ (2004) The origin of hematite in high-grade iron ore based on infrared microscopy and fluid inclusion studies: The example of the Conceicao Mine, Quadrilatero Ferrifero, Brazil. Economic Geology 99: 611–624. Rosiere CA, Siemes H, Rios FJ, and Quade H (2002) Deformation controlled high-grade iron ore. Quadrennial IAGOD Symposium and Geocongress 2002. Rosiere CA, Spier CA, Rios FJ, and Suckau EV (2008) The itabirites of the Quadrilatero Ferrifero and related high-grade iron ore deposits: An overview. Economic Geology 15: 223–254. Roy S and Venkatesh AS (2009) Banded iron formation to blue dust: Mineralogical and geochemical constraints from the Precambrian Jilling-Langalata Deposits, Eastern Indian Craton. Applied Earth Science (Transactions of the Institution of Mining and Metallurgy, Section B) 118: 178–188. Ruhlin DE and Owen RM (1986) The rare earth element geochemistry of hydrothermal sediments from the East Pacific Rise: Examination of a seawater scavenging mechanism. Geochimica et Cosmochimica Acta 50: 393–400. Samala W, El Aref M, and Gaupp R (2011) Mineralogical and geochemical investigations of the Middle Eocene ironstones, El Bahariya Depression, Western Desert, Egypt. Gondwana Research 22: 717–736. Schorscher HD (1982) Quadrilatero Ferrifero – Minas Gerais State: Rio das Velhas greenstone belt and Proterozoic rocks. Excursions annex International Symposium on Archean and Proterozoic Geologic Evolution and Metallogenesis, Brazil. Schroll E (1978) Contribution to the mineralogy of the iron-rich mud sediments of Santorini, Greece. In: Doumas C (ed.) Second International Scientific Congress, pp. 333–342. London: Thera and the Aegean World. Schwann P (2009) An alternative genesis for pisolites and channel iron deposits in Western Australia. In: Iron Ore 2009, pp. 171–185. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Schweigart H (1965) Genesis of the iron ores of the Pretoria series, South Africa. Economic Geology 60: 269. Schwertmann U (1979) The influence of aluminium on iron oxides. V. Clay minerals as sources of aluminium. Soil Science 128: 195–200. Schwertmann U (1988) Goethite and hematite formation in the presence of clay minerals and gibbsite at 25 degrees C. Soil Science Society of America Journal 52: 288. Schwertmann U, Carlson L, and Murad E (1987) Properties of iron oxides in two Finnish lakes in relation to the environment of their formation. Clays and Clay Minerals 35: 297–304. Schwertmann U and Taylor RM (1989) Iron oxides. In: Dixon JB and Weed SB (eds.) Minerals in Soil Environments, vol. 1, pp. 380–438. Madison: Soil Science Society of America. Siehl A and Thein J (1989) Minette-type ironstones. In: Young TP and Taylor WEG (eds.) Phanerozoic Ironstones, vol. 46, pp. 175–193. London: The Geological Society. Simonson BM (1987) Early silica cementation and subsequent diagenesis in arenites from four early Proterozoic iron formations of North America. Journal of Sedimentary Petrology 57: 494–511. Simonson BM (2003) Origin and evolution of large Precambrian iron formations. Geological Society of America Special Paper 370: 231–244. Sims SJ (1973) The Belinga iron ore deposit (Gabon). Genesis of Precambrian Iron and Manganese Deposits, pp. 323–333. Paris: UNESCO. Slater D and Highley DE (1977) The iron ore deposits in the United Kingdom of Great Britain and Northern Ireland. In: Walther HW and Zitzmann A (eds.) The Iron Ore Deposits of Europe and Adjacent Areas, vol. 1, pp. 393–409. Hannover: Federal Institute for Geosciences and Natural Resources. Spier CA, Oliveira SMB, and Rosiere CA (2003) Geology and geochemistry of the Aguas Claras and Pico iron mines, Quadrilaˆtero Ferrıˆfero, Minas Gerais, Brazil. Mineralium Deposita 38: 751–774. Stanton RL (1972) Iron concentrations of sedimentary affiliation. In: Stanton RL (ed.) Ore Petrology, pp. 399–453. New York: McGraw-Hill Book Company. Stone MS, George AD, Kneeshaw M, and Barley ME (2002) Stratigraphy and sedimentary features of the Tertiary Yandi channel iron deposits, Hamersley Province, Western Australia. Iron Ore 2002, pp. 137–144. Melbourne, VIC: The Australasian Institute of Mining and Metallurgy. Stoops G (1983) SEM and light microscopic observations of minerals in bog-ores of the Belgian Campine. Geoderma 30: 179–186.

Storkey A, Doecke A, and Whaanga A (2010) Stratigraphy of the western channel iron deposits of the Marillana Creek operations, Western Australia. Applied Earth Science 119: 2–11. Sturesson U (2003) Lower Palaeozoic iron oolites and volcanism from a Baltoscandian perspective. Sedimentary Geology 159: 241–256. Sturesson U, Dronov A, and Saadre T (1999) Lower Ordovician iron ooids and associated oolitic clays in Russia and Estonia: A clue to the origin of iron oolites? Sedimentary Geology 123: 63–80. Sturesson U, Heikoop JM, and Risk MJ (2000) Modern and Palaeozoic iron ooids – A similar volcanic origin. Sedimentary Geology 136: 137–146. Swedlund JPJ, Hamid DRD, and Miskelly MGM (2010) Insights into H(4)SiO(4) surface chemistry on ferrihydrite suspensions from ATR-IR, diffuse layer modeling and the adsorption enhancing effects of carbonate. Journal of Colloid and Interface Science 352: 149–157. Takeno N (2005) Atlas of Eh-pH Diagrams: Intercomparison of Thermodynamic Databases. Japan: Geological Survey of Japan. Taylor KG (1992) Non-marine oolitic ironstones in the Lower Cretaceous Wealdon sediments of southeast England. Geological Magazine 129: 349–358. Taylor WEG (2005) Sedimentary rocks: Ironstones. Encyclopedia of Geology, vol. 5, pp. 97–107. Amsterdam: Elsevier. Taylor D, Dalstra HJ, Harding AE, Broadbent GC, and Barley ME (2001) Genesis of high-grade hematite orebodies of the Hamersley Province, Western Australia. Economic Geology 96: 837–873. Theobald PKJ, Lakin HW, and Hawkins DB (1963) The precipitation of aluminum, iron and manganese at the junction of Deer Creek with the Snake River in Summit County, Colorado. Geochimica et Cosmochimica Acta 27: 121–132. Thorne W, Hageman S, and Barley ME (2004) Petrographic and geochemical evidence for hydrothermal evolution of the North deposit, Mount Tom Price, Western Australia. Mineralium Deposita 39: 766–783. Thorne W, Hagemann S, Webb A, and Clout J (2008) Banded iron formation-related iron ore deposits of the Hamersley Province, Western Australia. Society of Economic Geologists 15: 197–221. Timofeeva YO and Golov VI (2010) Accumulation of microelements in iron nodules in concretions in soils: A review. Eurasian Soil Science 43: 401–407. Trendall AF (1965) Origin of Precambrian iron formations. Economic Geology 60: 1065–1070. Trendall AF (1973) Precambrian iron-formations of Australia. Economic Geology 68: 1023–1034. Trendall AF (1980) A progress review of the Hamersley basin of Western Australia. Geological Survey of Finland Bulletin 307: 113–131. Trendall AF (1983) The Hamersley basin. In: Trendall AF and Morris RC (eds.) Iron-Formation: Facts and Problems, vol. 6, pp. 69–129. Amsterdam: Elsevier. Trendall AF (2002) The significance of iron-formation in the Precambrian stratigraphic record. In: Altermann W and Corcoran PL (eds.) Precambrian Sedimentary Environments: A Modern Approach to Ancient Depositional Systems: Special Publication of the International Association of Sedimentologists, vol. 33, pp. 33–66. Boulder, CO: Geological Society of America. Trendall AF and Blockley JG (1970) The iron formations of the Precambrian Hamersley group, Western Australia: With special reference to the crocidolite. Bulletin – Geological Survey of Western Australia 119: 366. Trendall AF and Blockley JG (2004) Precambrian iron-formation. Developments in Precambrian Geology 12: 403–421. Trendall AF and Morris RC (1983) Developments in Precambrian Geology: IronFormation: Facts and Problems 6. Amsterdam: Elsevier. Trendall AF and Pepper RS (1977) Chemical composition of the brockman iron formation. Geological Survey of Western Australia 25. Tsipurskii SI and Golubovskaya EV (1989) Smectites of tobacco ores in Kerch deposits. Lithology and Minerals Resources 24: 149–161. Van Houten FB (1992) Review of Cenozoic ooidal ironstones. Sedimentary Geology 78: 101–110. Van Houten FB and Arthur MA (1989) Temporal Patterns Among Phanerozoic Oolitic Ironstones and Oceanic Anoxia, pp. 33–49. London: Geological Society of London. Van Houten FB and Bhattacharyya DP (1982) Phanerozoic oolitic ironstones – Geologic record and facies model. Annual Review of Earth and Planetary Sciences 10: 441–457. Vempati RK and Loeppert RH (1989) Influence of structural and adsorbed Si on the transformation of synthetic ferrihydrite. Clays and Clay Minerals 37: 273–279. Vodyanitskii YN (2010) Iron hydroxides in soils: A review of publications. Eurasian Soil Science/Pochvovedenie 43: 1244–1254. Wells MA and Ramanaidou ER (2011) Occurrence and mineralogical association of phosphorous in Australian bedded iron ore deposits. Iron Ore 2011, pp. 331–336. Melbourne,VIC: The Australasian Institute of Mining and Metallurgy.

Sedimentary Hosted Iron Ores

Yager DB, Church SE, Verplanck PL, and Wirt L (2003) Ferricrete, Manganocrete, and Bog Iron Occurrences with Selected Sedge Bogs and Active Bogs and Springs in Part of the Animas River watershed, San Juan County, Colorado. Denver: U. S. Geological Survey MF-2406. Yanitzkii HL (1960) Oligocene Oolitic Iron Ores of Northern Turgai and Their Genesis, vol. 37. Moscow: Academy of Sciences. Young TP (1989) Phanerozoic ironstones: An introduction and review. In: Young TP and Taylor WEG (eds.) Phanerozoic Ironstones, vol. 46, pp. ix–xxv. London: Geological Society.


Zitzmann A (1977) The iron ore deposits of the western U.S.S.R. In: Walther HW and Zitzmann A (eds.) The Iron Ore Deposits of Europe and Adjacent Areas, vol. 1, pp. 325–391. Hannover: Federal Institute for Geosciences and Natural Resources. Zitzmann A (1978) The Iron Ore Deposits of Europe and Adjacent Areas. vol. 2, pp. 386. Hannover: Federal Institute for Geosciences and Natural Resources. Zitzmann A and Neumann-Redlin C (1977) The genetic types of iron ore deposits in Europe and adjacent areas. In: Walther HW and Zitzmann A (eds.) The Iron Ore Deposits of Europe and Adjacent Areas, vol. 1, pp. 13–35. Hannover: Federal Institute for Geosciences and Natural Resources.