THE MINERALOGY AND macemmor- THE; ‘ mM-BA IRONORE Thesis far the Degree of M. S. MICHIGAN STATE UNIVERSH‘Y FODEE KROMAH 1971 WW 3 WWW ‘ 848 8706 W 9 / ‘ LI BRA R Y Michigan Sta to University ABSTRACT THE MINERALOGY AND PETROGRAPHY OF THE NIMBA IRON ORE BY Fodee Kromah About 300 million tons of high-grade hematite occur in Precambrian iron formation (itabirite) in the Nimba Series. The Nimba Series is a 1,500-meter thick sequence of schists, phyllites and iron formation. The Nimba Series has two units; a lower unit, called the Valley group is composed of metavolcanic rocks (basal rocks and amphibole schists); and an upper unit, the Ridge group consisting of metasedimentary rocks (phyllites and itabirites). Granitic rocks of 2.5 billion years to 3.6 billion years occur to the southeast of the Nimba Ridge where they formed straight contact with the Nimba itabirites. The major structural feature of the Nimba Range is a central anticline underlying the Seka Valley to the north and Central Gbahm to the south. This anticline is flanked by two synclines underlying the North Gbahm Ridge to the north, and the Yiti Valley and west slope of Gbahm Ridge to the south. The whole anticline-syncline structure is Fodee Kromah plunging to the southeast at a low angle. Only one major fault has been observed in the area and this is to the southeast where the contact of the granites and itabirites has been inferred as a fault. All folds are more or less isoclinal. There are two major types of iron ore, the blue ore (mainly hematite) containing an average of 67.8 per- cent Fe, and the brown ore (hematite and goethite) con- taining 65.5 percent Fe at depth and 63.5 percent Fe at the surface. Each type of ore has its own soft, medium, and hard varieties, however, the soft ores are the pre- dominant type as regards physical quality. Genetically all ores have been formed from leach- ing of silica from the iron formation. The iron formation apparently formed along an epi-continental landmass in a shallow basin during Precambrian time under conditions relevant to tropical climate. Quartz, clays and phosphorous (occurring in phosphatic minerals) are the main impurities in the ore deposits. THE MINERALOGY AND PETROGRAPHY OF THE NIMBA IRON ORE BY Fodee Kromah A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology 1971 ACKNOWLEDGMENT S The author wishes to thank the following people without whose assistance the completion of this study would have been more or less impossible: Mr. John Berge, Chief Geologist, Liberian American- Swedish Mining Company (LAMCO), and members of his staff for their assistance during the field work; the Liberian American-Swedish Mining Company for permission to use materials collected, and for financial assistance in mailing of ore and rock specimens. Dr. E. A. Nyema Jones, Mr. Thomas Sherman and Mr. Cletus Wotorson, all of the Liberian Geological Survey for giving me additional information on the geology of the Nimba area, and for making the library of the Liberian Geological Survey available to me while I was in Liberia. Dr. Harold B. Stonehouse for his propitious help and guidance as chairman of my thesis committee; Dr. James Trow for serving on the committee, and for his constructive statements of the manuscript; Dr. Bennett Sandefur for serving on the committee. The author also wishes to acknowledge the very helpful services of Miss Esther Parker for typing and proofreading of the manuscript. ii NT‘"‘ U43“; ". . I! 8" 3t ; 2 F 5g * k v Ei”; 3‘ w“ i ’1 i g E L;- 3. ‘1‘) 'r in: I; z" ""‘I lgéé -::\..d‘;l W TABLE OF CONTENTS Chapter Page I. INTRODUCTION . . . . . . . . . . 1 Previous Study. . . . . . . . . 3 Method of Study . . . . . . . . 5 II. THE STRATIGRAPHY OF THE NIMBA AREA. . . 6 Yekapa Series . . . . . . . . . 8 Nimba Series . . . . . . . . . 8 Basal Rocks . . . . 9 Seka Valley Amphibole .Schist . . . 9 Gbahm Ridge Phyllite . . . . . . 10 Nimba Itabirite. . . . . . . . 11 Mount Alpha Phyllite . . . . . . 16 Granitic Rocks . . . . . . . . 16 III 0 PETROGMPHY O O O O O O O O O 0 l7 Metavolcanic Rocks . . . . . . . l7 Metasedimentary Rocks . . . . . . l9 Gbahm Range Phyllite. . . . . . . 20 Mt. Alpha Phyllite . . . . . . . 21 IV. STRUCTURE OF NIMBA MINE AREA. . . . . 26 V. THE NIMBA IRON ORES. . . . . . . . 29 Blue Hematite Ores . . . . . . . 29 The Brown Ores. . . . . . . . . 35 Ore Minerals . . . . . . . . . 41 Magnetite. . . . . . . . . . 41 Hematite . . . . . . . . . . 42 Goethite . . . . . . . . . . 42 Quartz. . . . . . . . . . . 42 iii Chapter Page VI. ORIGIN OF THE NIMBA IRON ORES . . . . . 48 VI I 0 CONCLUSION O O O O O O O C I D O 5 3 LIST OF REFERENCES . . . - . . . . . . . 56 iv LIST OF TABLES Table Page 1. Summary of the Nimba Series (After Berge, 1968) O O C O O O O O O O O O O 7 Figure 11. 12. 13. LIST OF FIGURES Map of Liberia, Location of Nimba Mine Area . Nimba Itabirite, Hematite, Goethite, Magnetite and Silica. White is Granular Quartz . . Thin Section of Nimba Itabirite Showing Quartz (White) and Iron Ore Minerals (Black)-- Hematite, Magnetite and Goethite . . . . Medium Hard Itabirite--Well Developed Small Small Scale Folding . . . . . . . . Thin Section of Amphibole Schist Showing Prismatic Grains of Hornblende with Poorly Developed Terminations . . . . . . . Thin Section of Mt. Alpha Phyllite Showing Corroded and Cracked Grain of Garnet. . . Garnet in Mt. Alpha Phyllite . . . . . . Pyrites (Black) Occurring in Mt. Alpha Phy11ite O O O C O O O I O O O 0 Geologic Cross Sections. . . . . . . . Blue Hard Ore, Mainly Hematite and Highly Recemented . . . . . . . . . . . Grains of Hematite (White) Cemented by Quartz (Dark Gary) in Recemented Blue Ore . . . Siliceous Blue Medium Hard Ore, Mainly Hematite and Silica . . . . . . . . Siliceous Blue Biscuit, Mainly Hematite and Silica. O O O O O O O O O O O 0 vi Page 13 14 15 18 22 23 24 27 3O 32 33 34 Figure 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Page Siliceous Blue Ore, Mainly Hematite (H) and Quartz (Q). Note Late Quartz (q) Crystal- lizing Between Earlier Formed Quartz; Also Islands of Quartz in Hematite (Partial Replacement of Quartz by Hematite) . . . . 36 Lateritic Ironstone, Mainly Goethite, Commonly Interbedded with Blue Hard Ore . . 37 Hydration of Hematite (H) to Goethite (Gray) in Brown Ore . . . . . . . . . . . 38 Hydration of Hematite (White) to Goethite (Light Gray). Quartz is Dark Gray (w-- Probably Micro Fracture Serving as Pathway for Water?). . . . . . . . . . . . 39 Hematite (White) Occurring Together with Quartz (Dark Gray) . . . . . . . . . 4O Hematite (White) Replacing Magnetite (Gray). . 43 Hematite (White) Nearly Completely Replacing Magnetite (Dark Gray) . . . . . . . . 44 Freehand Sketch Showing Hematite (h) Replacing Magnetite (m): Magnification = 175X . . . 45 Freehand Sketch of Well-Developed Grains of Hematite (h): Magnification = 175x. . . . 45 Hematite (Martite) Grain (White) Replaced Magnetite (Gray) and Also in the Process of Being Replaced by Goethite (Black) . . . . 46 Depositional Zones in Hypothetical Basins in Which Iron Compounds are Being Precipitated (After James, 1954) . . . . . . . . . 49 Stability Fields of Iron Oxides as Functions of Eh and pH at 25 degrees C and 1 Atmosphere Total Pressure. Dashed Line Indicates Lower Stability Limit of Magnetite in Presence of Metastable Water (After Garrels and Christ, 1965). . . -. . . . . . . . . . . 51 vii CHAPTER I INTRODUCTION The Nimba Range, about 20 miles long, of which 8 miles are in Liberia and 12 miles in Guinea, is located about 200 miles northeast of Monrovia along the inter- national boundary with Guinea and the Ivory Coast (Figure l). The ore bodies, contained along two parallel bands of Precambrian itabirite iron formation, form a steep-sided ridge, the slopes of which in some places are parallel to the dip at inclinations of up to 70°. The. gentler slopes, lower down on either side of the crest, are underlain by phyllites, and the relatively flat-lying surrounding area is underlain by gneisses. The Nimba Ridge which contains the main ore body trends southwest— northeast. Other topographic features include the north— northeast trending Gbahm Ridge, the relatively broad Seka Valley lying between the North Gbahm and Nimba Ridges, the narrow, deep and steep sided Yiti Valley between the Junction-Gbahm Ridge and South Nimba Ridge, and the north- east trending Nimba Ridge. The high point is 1,384 meters above sea level at Guesthouse Hill on the Nimba Ridge, and the highest point 1 PO SIERRA lEONE N '1‘ I I I 510 39.0 Milo: 4 N J Scale L3 000 000 LIBERIA FIGURE I Figure l.--Map of Liberia, location of Nimba mine area. on the Gbahm is 1,350 meters. Maximum relief in the area is 750 meters. The terrain is irregular, and the two ridges are steep-sided below the crests. The mine is located 7k° north of the equator among tropical rain forests. The average annual rainfall is about 120 inches, mainly from violent cloudbursts which occur most frequently between the months of April and October. The highest precipitation in 24 hours since 1957 (when records were first kept) was 6.55 inches, but a rainfall of 1.5 inches within the space of one-half hour is common during the rainy season. Average daily tempera- ture during the dry season are 75-80 degrees Fahrenheit, with extremes of 40 and 90 degrees Fahrenheit. Previous Study The Nimba Iron Ore was discovered in the Nimba Mountains in the latter part of 1955. In July 1960, the Liberian American-Swedish Mining Company (LAMCO) came into existence and was given concession rights to explore and develop the iron ore deposits of the Nimba-Gbahm Ridge. The first geologic reconnaissance mapping over the Nimba-Gbahm area was conducted by LAMCO geologists Lackshewitz and Tremaine. Their results are included in an unpublished report issued in 1962. According to Berge (1968), the Lackshewitz and Tremaine report includes an impressive number of stream valleys and traverse lines mapped by pace and compass with altimetric vertical control; a stratigraphic and structural interpretation was reported with no "concrete proposal" made. The foremost problem was transportation, and access to the ore deposits was by foot only. Furthermore, outcrops were sparse, and those which could be mapped were deeply weathered. De- tailed petrographic work was not carried out at that time. Since the beginning of production in 1963, ore, low grade iron formation, and waste rock have been con-4 tinuously mapped by the LAMCO Geology Section. Extensive core drilling has been carried out both for production and for engineering studies. The result has been the compi- lation of much incomplete data pertaining not only to the ores, but also to the petrology, structure and strati- graphy. Petrographic studies of thin sections of rocks collected by Lackshewitz and Tremaine together with samples of ores collected during mining have been studied by LAMCO geologists to some extent. In 1966, John Berge, Chief Geologist of LAMCO, published an incomplete paper entitled "Genetical Aspects of the Nimba Iron Ores." He attributes the origin of the brown ore to "lateric processes" but gives no explanation to the genesis of the blue ore. In the final analysis, he concludes that "the problem of blue ores is not yet solved." Method of Study In the summer of 1969, I spent two months in Liberia doing field work with John Berge and members of the LAMCO Geology Section collecting rock and ore samples together with available information of the Nimba—Gbahm area. Thin sections of the rocks and ores have been made and studied extensively together with polished sections of the ores. These results together with field obser— vations and available data, have led to an explanation of the origin of the Nimba Iron Ore. CHAPTER II THE STRATIGRAPHY OF THE NIMBA AREA John Berge, chief geologist of LAMCO, is responsi- ble for most of the regional stratigraphy and structure of the area. The stratigraphy of the area is divided into the Yekepa Series and the Nimba Series (Table l), which have their equivalent in other parts of Africa, but no correlation has been made with respect to rocks in other areas. G. W. Leo (1966), has reported a program of geo- chronology of Liberian rocks at Massachusetts Institute of Technology undertaken in cooperation with the United States Geological Survey and the Liberian Geological Survey. The chief aim of the program is to incorporate ages of Liberian rocks into a comprehensive dating program covering much of West Africa and northeastern South America, in the hope of finding supporting evidence for the hypothesis of continen- tal drift. Twenty-eight age determinations by the Rb-Sr method from 10 separate localities have been made of granitic rocks collected in western Liberia and gneises from the south end of the Nimba Range. The ages fall in the general range of 2.5 billion years to 3.6 billion years, which establishes the Precambrian shield of Liberia 6 mxoon owmmnmnuas can Amumumaoamcoo Nuumsqv mmufluuumno mxoou ammmm macho mm m musmfiflpmm cauflamm Ha > can msomumoamo cmumamoumpcw sua3 sofluflmommo oeummHo mxoou A0fluwmm©cm umflnom maonflngfim mEom paw Emwcmoao> manmnoumv oacmoao> moaam> mxom |||||||||||||||||||||||||| 5|"II"'I-'III"II|'Il"|l.lnf'lll|'l'l.""'III'II'II.I-.U'II'II||"'II| mcoflumamo mcofiuwocoo mnaospmn Inmucw oeuwamm suw3 Hops: cofluamommc mwuflaamnm muwumm muflaahsm “msflhmzumm3 Hmofifimno Iwuflzmmumunuumso mmpflm Esmnw mcofiuflosoo mcHNfioflxo coaumEHom macawfi moons cowuflmoamo Goya ODMOHHflm pmpmHmo msoum “mauumspmmz Hmonsmno numucn can; munuflnmuH munuanmufi wasnz wmuflm mucmfiwvwm mmoN muwaamcm nuumsv mcwm cam mamau “mmma muwuoanonmuwofiumm mama mo soflmonm H80flcmnomz nmufluoflnnwuuwso ouaaawnm mamas .uz sofluflmommo mo mcofluwocoo momma xoom cowumfihom muonw .Ammma .mmumm umummv mmfiumm mnsfiz man no sumassmuu.a mamas as one of the more ancient provinces known on earth (Leo, 1967). Age determinations of the gneisses from the south end of the Nimba Range show that the last period of crystallization or recrystallization was more than 2.5 billion years ago (Hurley, 1967). Yekapa Series The Yekapa Series comprises the bedrock underlying the broad valley between the Nimba-Gbahm Ridge on the east and the Takadeh-Yuelliton Ridge to the west (Plate 1). There are no outcrops of the Yekapa Series in the area, however, as projected from outcrops outside the area, at least two rock types have been recognized--acid rocks which are usually strong foliated assemblages of plagio- clase, microcline, quartz and biotite, and the basic rocks which are foliated hornblende amphibolites. Nimba Series The Nimba Series has been divided into two groups and five formations. These rocks underlie the Nimba and Gbahm Ridges and the Seka and Yiti Valleys, and continue to the northeast in the Republic of Guinea. Table 1 illustrates the stratigraphy of the area. Since information about the stratigraphy and structure of the Nimba-Gbahm-Range is available only through mining operations, the relation between the Yekepa and Nimba rocks is not well known. In fact, only one observation resembling a contact between the Yekepa and Nimba Series has been made. This is at the base of the northeast slope of Nordhund Hill in the north-central part of the area. Here heavily weathered Yekepa gneisses and sheared quartzite of the Nimba Series appear in adjacent cuts along the LAMCO railroad. It cannot be determined here if there exists an unconformity between the two, but the foliation in the two weathered rock cuts is parallel. Basal Rocks Present information about the basal rocks is far from adequate, however, they consist of quartzites and ultrabasic rocks, whose thickness and mutual relationship have not been determined. Stratigraphically these rocks are considered to be the oldest of the Nimba Series. A railroad cut north- east of Nordhund Hill has exposed this rock which appears to be sheared quartzite. Hand specimens show that it is a greenish, strongly foliated, quartz-pebble rock, with the pebbles elongated to the foliation suggesting that the rock may have been a conglomerate. Berge (1968) reports that a thin section shows about 80 percent quartz and the remainder is fine mica. About half the quartz is fine- grained and the texture is cataclastic. Seka Valley Amphibole Schist The lower-most unit of the Nimba Series to be given a formation status is the Seka Valley amphibole lO schist. Its contact with the underlying basal unit of the Nimba Series has not been observed, but it is presumed to be gradational. The schist has a thickness of about 700 meters and a characteristic topography comprising most of the northwest slope of the Gbahm Ridge, as well as the lower Seka Valley slopes of the Nimba and Gbahm Ridges. The topography is one of gently sloping spurs protruding far out from the main Nimba and Gbahm Ridges, varying from 750 to 920 meters in altitude; alternating with these spurs are deep, steep-sided valleys which also are per- pendicular to the main ridge (Berge, 1968). The rock is deeply weathered on these spurs, and includes the amphibole schist and chlorite-amphibole schist members of the Gbahm Ridge, and gabbro diorite, tremolite schist, and talc- tremolite schist units occurring in the Seka Valley. Gbahm Ridge Phyllite The lowest unit in the Ridge group is the Gbahm Ridge phyllite which is mostly a graphite-quartz rock. The shallow dip of the formation, but also the fact that it is difficult to determine the contact in weathered rock, led Lackshewitz and Tremaine (1962), to assert that this unit covers much of the Seka Valley slope of the Gbahm Ridge. Outcrops of this formation can be seen as a folded silvery-grey phyllites in road cuts along the Gbahm Ridge portion of the main mountain road. In con- trast to the yellow-orange weathering products of the ll Seka Valley amphibole schist, Lackshewitz and Tremaine (1962) recorded outcrops of this phyllite in stream cuts also. The thickness of the phyllite is about 250 meters; it forms steep (about 45°-50°) slopes, in contrast to the gentle spurs on the amphibole schist. Berge (1968), has recorded weathering of this phyllite to a depth of at least 40 meters. Nimba Itabirite . The Nimba iron formation has been called an itabirite and it is 250-400 meters thick. Pure itabirite is a banded iron-formation containing alternating laminae of specular flakes of hematite and/or magnetite and granu— lar quartz (Figure 14). Itabirite has been defined by Dorr and Barbosa (1963) as follows: The term itabirite denotes a laminated, meta- morphosed, oxide-facies formation, in which the original chert or jasper bands have recrystallized into granular quartz and in which iron is present as hematite, magnetite, or martite. The quartz bands contain varied but generally minor quantities of iron oxide, the iron—oxide bands may contain varied but generally minor quantities of quartz. The term should not include quartzite of elastic origin with iron oxide cement even though such rocks are some- times grossly banded. It should only include rocks in which the quartz is megascopically recognizable as crystalline, in order to differentiate it from unmetamorphosed oxide-facies iron-formation. A certain amount of impurity in the form of a dolomite or calcite, clay, and the metamorphic minerals de- rived from these materials may be included, but these may never be dominant constituents over any notable thickness. Where they are, the rock term must be qualified by the use of appropriate mineral name as a qualifier (for example, dolomitic itabirite, a rock in which the dolomite largely takes the place of quartz). Rarely itabirite grades into ferruginous 12 chert which, when recrystallized, may look like low— grade itabirite, although commonly it is finer grained and whiter. To prevent confusion, a cutoff point of about 25 percent iron should be established. This figure is a practical one, as few itabirites are so lean in iron and most ferruginous cherts do not con- tain so much iron. Itabirite may grade into pure hematite through enrichment in iron or removal of quartz. The cutoff point might well be set at 66 percent because at and above this grade quartz is rarely concentrated in regular laminae. Fresh itabirites are gray to light brown but weather to shades of brown as a result of hydration of hematite to limonite. Berge (1968) has recognized four types of laminae: One consists of almost entirely equidimensional or elongate grains of equigranular quartz (white). A second type of lamina consists of equal proportions of equigranular quartz and iron oxide (magnetite and/ or hematite). A third type consists of coarse aggre- gates of highly strained quartz together with coarse grains of subhedral magnetite-hematite (black and white). Finally, a fourth type consists mainly of coarse elongated grains or subhedral masses of magnetite-hematite with fine quartz inclusions (black). Diffuse laminae range from 0.5 to 5 milli— meters in thickness. The Nimba itabirite consists of quartz, hematite and magnetite, with occasional amphibole and chlorite in silica-bearing laminae. Goethite is present not as a primary mineral, but results from chemical weathering of the iron formation. The itabirite has a dry Fe content of 40 percent, contains 40 percent iron oxides, and 60 percent quartz by volume. The Nimba itabirite forms ridges and comprises the backbone of the Nimba, Gbahm, and South Nimba Ridges, 13 Figure 2.--Nimba Itabirite, hematite, goethite, magnetite and silica. White is granular quartz. 1X 14 ) Figure 3.--Thin section of Nimba Itabirite showing quartz (white) and iron ore minerals (b1ack)--hematite, magnetite and goethite. 175x 15 Figure 4.--Medium hard itabirite--well developed small scale folding. 2x 16 jutting up sharply from the Gbahm Ridge phyllite on the northwest slope of the Nimba Ridge (Plate 1). Mount Alpha Phyllite The Mount Alpha Phyllite is the youngest unit of the Nimba Series. It is so named because a complete section of the formation is exposed in the isoclinal Nimba syncline on Mt. Alpha, between the Mt. Alpha and Northeast Extension ore bodies. The thickness of this formation varies from 50 to 100 meters. The Mt. Alpha Phyllite has a variable topography. At two locations it forms the summit along the range and in both cases the phyllite is covered by an iron capping. The Nimba itabirite is postulated as the source of the iron capping. Berge (1968) has observed that in both locations, the summits occur a few meters higher than hard cappings over small ore bodies on the southeast limb of the Nimba syncline. Granitic Rocks Granitic rocks (Precambrian granites, gneiss, schist) are known to occur to the southeast of the Nimba Ridge where they form straight contact with the Nimba itabirite. These rocks are coarse-grained and have gneissic foliation. Samples of rocks taken have been used mainly for age determination and no petrographic study has been made. CHAPTER III PETROGRAPHY Metavolcanic Rocks The Seka Valley amphibole schist (Figure 5), consists of hornblende and plagioclase (oligoclase) and accessory carbonate, quartz, garnet, biotite, apatite, and opaques. The average composition of the rock is as follows: Mineral % by volume hornblende 45 oligoclase 30 quartz 10 opaques 6 biotite 5. carbonates 3 The hornblende is blue-green to olive-green. The are prismatic with poorlydeveloped terminations. grains are anhedral to subhedral with symmetrical tion. Hornblende prisms up to two millimeters in are common. These laths are corroded and contain growths of feldspar and quartz. 17 crystals The extinc- length inter- 18 "I Figure 5.--Thin section of Amphibole schist showing prismatic grains of hornblende with poorly developed terminations. 350x 19 The plagioclase occurs in equal xenoblastic grains. Grains are generally subhedral and laths are rarely developed. Plagioclase grains normally exhibit zoned extinction. Inclusions of quartz, feldspar and biotite are common. Quartz with rotating extinction occurs and is associated with carbonates. Garnets are (usually) rare, subhedral, and are associated with carbonates and feldspar in the matrix. Feldspar occurs in the fine- grained amphibole schist. These feldspars show some zoning. The zoning of feldspars, uneven grain size, texture and composition indicate that these rocks were originally fine-grained igneous rocks that may have been extrusive. Metasedimentary Rocks Sedimentary members of the Seka Valley amphibole schist appear as relatively coarse-grained amphibole- garnet-quartz, or fine-grained quartz-biotite schists with carbonates, and garnets. A garnet-amphibole rock occurring on the northwest slope of the Nimba Ridge consists of amphibole, garnet and quartz. Plagioclase is less abundant and quartz and biotite are usually more abundant than in metavolcanic rocks. The hornblende is blue-green and some non- pheochloric grains are present. Garnets are anhedral and are very much elongated parallel to foliation. Average composition shows: 20 Mineral % by volume hornblende 50 plagioclase 15 quartz 25 biotite 6 Opaques 3 garnet 1 In another composition there is a weak alignment of amphiboles. The amphibole is deep green to olive, subhedral, and may be poikilitic but has a more distinctly prismatic shape than that in the schist. Carbonate is euhedral, frequently twined and grains are elongated parallel to foliation. Quartz grains are euhedral; but show undulatory extinction. The mineralogy and laminated structure clearly suggest a metamorphosed calcareous sedi- ment. Gbahm Range Phyllite A sample of black phyllite from the Seka Valley slope of Gbahm Ridge consists of biotite, chlorite, musco- vite, quartz amphibole and feldspar. Another sample con- sists of fine-grained graphite, quartz, pyrite, and muscovite and is cemented by a matrix of finer quartz and mica. This rock has a composition of: 21 Mineral % by volume quartz ' A 85 sillimanite 6 iron ores 5 biotite 3 carbonates l The sillimanite forms slender prisms of fibrous radial aggregates, with a diagonal cleavage and parallel extinction, the grains are mostly euhedral. Biotite, only associated with sillimanite, has a deep red-brown color owing to staining with iron—ore minerals and is strongly pleochloric in shades of brown. The rock has a micro- scopic appearance of a quartzite (85% quartz). Two other samples 75 meters apart in tunnel 8 consist of fine- grained graphite, quartz, pyrite and muscovite (Berge, 1968). Mt. Alpha Phyllite The Mt. Alpha phyllite occurs as a lepidoblastic assemblage of fine-grained quartz, chlorite, biotite, garnet and pyrite, with sphene and apatite as accessories. Quartz grains are typically subhedral to euhedral with extinction mostly even. Chlorite is light green and has a blue interference color. Both tan and brown varieties of biotite are present. Pyrite occurs in subhedral elongate masses (Figure 8). Garnets are euhedral to sub- hedral but are cracked and corroded (Figures 6 and 7). 22 MAI Figure 6.--Thin section of Mt. Alpha phyllite showing corroded and cracked grain of garnet. 175X Figure 7.--Garnet in Mt. Alpha phyllite. 250x 24 Figure 8.--Pyrites (black) occurring in Mt. Alpha phyllite. 350x 25 Carbonates are present in small amounts, usually in tiny grains. The characteristic mineralogy and texture of these phyllites indicate that they have been recrystallized from pelitic sediments. The Nimba Series is a 1,500-meter thick sequence of schists, phyllites and iron formation. It comprises of a lower unit, called the Valley group which consists of metavolcanic rocks; and an upper unit, the Ridge group, which consists of sedimentary rocks which have been metamorphosed. CHAPTER IV STRUCTURE OF NIMBA MINE AREA. The main structural feature of the Nimba Range is a central anticline underlying the Seka Valley to the north, and central Gbahm to the south (Figure 9--profiles 5, 8 and 13). This anticline is flanked by two synclines underlying the North Gbahm Ridge to the north, and the Yiti Valley and west slope of Gbahm Ridge to the south. The whole syncline-antiCline-syncline structure is plung- ing to the southwest at a low angle. According to Berge (1968), the present structural interpretation stems from two lines of reasoning: Stratigraphic arguments are as follows: (1) Petro- graphic data indicate that the two large amphibole schist horizons are of similar composition. Hence, they are assumed to be of the same formation (Seka Valley amphibole schist). (2) Based on the similarity of contacts of the east and west itabirite horizons on the Nimba Ridge with the Mt. Alpha phyllite, it is assumed that these two itabirite horizons are parts of the same formation. The acceptance of these two assumptions leads to the conclusion that both ridges are underlain by cores of complete isoclinal folds, either anticlinal or synclinal. Structural arguments are as follows: (1) Obser- vations on drag folds in phyllite near iron ore con- tacts on Mt. Alpha suggest that the fold core under- lying the Nimba Ridge is synclinal and that the anti- clinal fold is to the northwest or underlying Seka Valley. (2) Linear striations on ore and itabirite laminae plunge southwesterly. 26 m5 CWT" CENTRALNW HILL GUEST HOUSE HILL. 27 . . pun Halli?! ..' - - - . mum! PHYLLI‘“ E 3 g f c i 3. 5 .54» PROFILE :3 mom man-mm RIDGE GEOLOGIC CROSS SECTIONS FIGURE 9 '0.“ n»- '9‘" we 28 Only one major fault has been observed in the area; this is to the southeast of the Nimba Ridge, where the structural relationship of the Nimba itabirites with the granites has been inferred as a fault. This stems from the rather straight baseline of the Nimba Ridge and geologically by the apparent contact of the Nimba itabirite with granitic rocks without the intervening Gbahm Ridge phyllite and Seka Valley amphibole schist, and by the observations of mylonitic rocks at the contact recorded by Lackshewitz and Tremaine (1962). Due to inadequate geologic mapping of the area, geophysical (magnetic) correlation is not possible. However, some magnetic contours are considerably dis- placed along this "fault" line. CHAPTER V THE NIMBA IRON ORES In general, the ore types in the Nimba area are divided into: (1) hard, intermediate and soft high grade blue hematite ores, (2) brown ore (mostly soft). This classification is based on the study of iron ores by Dorr and Barbosa (1963, p. C55) in Brazil and has since been applied to the Nimba Iron Ores--it is based on the esti- mated percentage of material that would, after mechanized mining, crushing and screening, be above one-half inch in minimum dimension: Type of Ore Percentages Greater Than (Physical) One-half Inch in Diameter Hard ore 100-75 Medium hard 75—50 intermediate Medium soft 50-25 Soft 25- 0 Blue Hematite Ores The blue hematite ores (Figure 10) in Nimba have an average of 67.8 percent iron. The blue ore is a steel- blue in color and has its hard, medium and soft Varieties. 29 30 *1 Figure 10.--Blue hard ore, mainly hematite and highly recemented. 1X 31 Within in the present mine area, the main ore body (esti- mated at 190 million tons) consists of blue ore. The average composition of the blue ore is: F8203 97.0% L.O.I.* 1.0% SiO2 1.5% A1203 0.5% P 0.4% The hard ore (Figure 10) is compact, tough and resistant to both mechanical and chemical weathering. It is low in gangue materials, especially silica. Silica is the cementing material and its dark gray color is quite evident between white hematite grains in reflected light (Figure 11). The medium (Figure 12) or intermediate varieties are composed of disaggregated and powdery hematite which is also resistant to weathering. The soft ore (Figure 13) is the predominant type ore as regards physical quality. Most of this is called biscuit ore--the term referring to thin wafer-like laminae of iron ore minerals which occur uniformly throughout the zones of soft ore. These laminae vary in thickness from less than 1 mm to 5 mm. Their main characteristics are softness and friability and their laminated structures *Lost on ignition (amount of water). 32 Figure ll.--Grains of hematite (white) cemented by quartz (dark gray) in recemented blue ore. (Polished Section, 350K) 33 Figure 12.--Si1iceous blue medium hard ore, mainly hematite and silica. 1X 34 Figure 13.—-Si1iceous blue biscuit, mainly hematite and silica. 1X 35 have been retained from the original iron formation. Soft ores are known to grade vertically or horizontally into medium hard to hard varieties which usually retain the laminations present in the soft ores. This type of blue ore is more siliceous than medium and hard varieties (Figure 14). The Brown Ores The brown ore (Figures 15, 16, 17 and 18) is com—' posed essentially of hydrated ferric oxides (Goethite- Fe203°HZO) and feric oxides (hematite-Fe O3) and has its 2 hard, medium and soft varieties. It is found underlying flat topped flat ridges with a mean elevation of about 1300 meters above sea level. The flat topped ridges are preserved intact by the hardcapping. The brown ores have been estimated at 110 million tons and occupy the N-74, Mt. Alpha and Northeast Extension ore bodies (Plate I). The average composition is as follows: Brown Ore Surface Brown Ore Fe203 93.7% 90.0% L.O.I.* 4.0% 5.5% SiO2 1.5% 0.8% A1203 0.8% 2.8% P 0.06% 0.07% *Lost on ignition (amount of water). 36 MAI I_- _ _-_ _- _ __ Figure l4.--Si1iceous blue ore, mainly hematite (H) and quartz (Q). Note late quartz (q) crystallizing between earlier formed quartz; also islands of quartz in hematite (partial replacement of quartz by hematite). (Polished Section; 350x) ' 37 Figure 15.-~Lateritic ironstone, mainly goethite, commonly interbedded with blue hard ore. 1X 38 "I Figure l6.--Hydration of hematite (H) to goethite (gray) in brown ore. (Polished Section; 350X) 39 Figure l7.--Hydration of hematite (white) to goethite (light gray). Quartz is dark gray (w--probably micro fracture serving as pathway for water?). (Polished Section; 175X) 40 Figure 18.--Hematite (white) occurring together with quartz (dark gray). (Polished Section; 350X) 41 The brown ores are characterized by vertical zoning. In the surface zones (surface brown ore), the ores contain a higher percentage of goethite and alumina. Silica and alumina may combine to form clay minerals and with excess alumina, gibbsite forms. The brown ores are found under ridge crests and are capped by a hard completely recemented ironstone in which the laminated Structure has been re— tained. Berge (1966) has estimated the depth of this hard capping at two to five meters. Clays are abundant in the surface zones and filling vugs and cavities which account for the increased alumina content in the surface zone. The silica content increases with depth, and the ore gradually becomes a soft itabirite at an average depth of 75 to 100 meters. The brown ore has an average Fe content of 65.5 percent at depth and 63.5 percent at the surface. Ore Minerals Magnetite Magnetite occurs in grains ranging in size from 2 millimeters to about 10 microns. Average grains are 0.1 to 0.2 millimeters. Magnetite-bearing grains are commonly euhedral to anhedral, coarse grains are elongated and subhedral to anhedral. 42 Hematite Hematite occurs in grains ranging in size from 2 millimeters to about 10 microns. Hematite and magnetite may occur together, the former replacing the latter without regard to magnetic grain shape (Figures 19, 20, 21 and 22). In a second type the hematite (martite) selectively replaces magnetite (Figure 23). Fine hematite grains are subhedral to anhedral while coarser grains are euhedral to subhedral. Elongate hematite grains range from 0.03 millimeter to 0.2 millimeter in dimension. Goethite Goethite, the third iron mineral, results from the hydration of hematite, and is associated with hematite rather than magnetite (Figure 23). Goethite tends to be anhedral. The grain sizes range from 0.1 millimeter to 2 millimeters. Goethite in the brown ore is not primary and has been formed by hydration of hematite during chemi— cal weathering. Quartz Quartz grains range in size from 0.01 to 0.7 millimeters in equidimensional grains (Figure 14). Aver- age grain size is 0.1 millimeters with elongation up to .0.2 millimeters the quartz grains developed anhedral faces with goethite. Quartz has an undulatory extinction. Quartz cements hematite grains (Figure 11) and also forms secondary growth rims around earlier quartz grains 43 Figure l9.--Hematite (white) replacing magnetite (gray). (Polished Section; 175x) 44 Figure 20.--Hematite (white) nearly completely replacing magnetite (dark gray). (Polished Section; 175X) 45 Flame 21. Freehand dutch showing hematite (h) "Placing magnetiteIm)? magnification - wax. Figure 22. Freehand sketch of well-developed gains of hematite (h): magnification - I75 X. 46 Figure 23.--Hematite (martite) grain (white) re- placed magnetite (gray) and also in the process of being replaced by goethite (black). (Polished Section; 350X) 47 (Figure 14). In Figure 14, earlier formed quartz grains are surrounded by late quartz. This is suggestive of two periods of crystallization where late quartz has crystal- lized in open spaces between grains and also has corroded edges of earlier quartz grains. There are islands of quartz grains in hematite which may indicate partial replacement of quartz by hematite (Figure 14). CHAPTER VI ORIGIN OF THE NIMBA IRON ORES The most important and controversial issue of the origin of iron ores and iron-bearing sediments is the source of the iron. The iron in iron bearing sediments has been considered to be derived either from weathering of the adjacent landmass or to have been contributed by volcanism or by hydrothermal waters. Van Hise and Leith (1911) are responsible for re- lating the deposition of iron formation to volcanism. According to Pettijohn (1957, p. 458), the major factor which led these early workers to the concept of a volcanic source for the iron was the presumed inadequacy of ordi- nary weathering to supply solutions of the proper type for precipitation of iron in iron formation. Shallow basins are postulated as the receptacles for the iron-silica sediments. Woolnough (1941), has sug- gested that iron formation is an epicontinental sediment formed as a chemical precipitate from river waters that entered saline lakes or other closed basins. The annual temperature cycle of fresh-water lakes, with its effects 48 49 on water stratification and chemistry of the lake waters, under certain climatic conditions, provides a mechanism for rhythmic deposition of silica and iron (Hough, 1957). Sakamato (1950) attributed the iron and silica to mature weathering in the neighboring areas and noted that these materials must have been transported in different ways according to the seasons, in a manner similar to the development of distinct soil horizons in wet and dry climates. Sakamato (1950) further suggested that the iron was carried to lakes during the wet season when the waters were acidic; but in the dry season, precipitation of the iron and concomitant transportation of silica occurred due to a sharp rise in pH. Sea level .—I a _.1. 'fi fiv- fi-Q.“ , *d.‘ ~ I . 175'}; I‘L“\ ' _ . 019/ \\ A ‘ .. \M mr‘w/mm I \ - 7:105 swim _ ah . e/ 9. . .. $96} Figure 24.--Depositional zones in hypothetical basins in which iron compounds are being precipi- tated (after James, 1954). Recent work by James (1954) has indicated that deposition of each iron facies is largely controlled by the Eh and pH conditions of the environment, especially the oxidation-reduction potential. High values of Eh and pH promote the oxide minerals. James (1954) tends 50 to favor weathering and deposition in a restricted basin (Figure 24) as a more general mechanism of concentration; to account for the very abundant iron ores of the Pre- cambrian when thick vegetation could not have aided the solution of iron from the lands, Krauskopf (1967) sug- gests that the atmosphere in Precambrian times may have contained more carbon dioxide than at present, making surface waters more acid and hence more effective in breaking down silicate minerals. Krumbein and Garrels (1952) believe that the pH and Eh of the environment provide general controls on chemical sedimentation which include both inorganic and biochemical processes. They further emphasize that the amount of a particular mineral that will precipitate depends upon the amount of the constituents available but that a change in deposition from one mineral to another will not take place unless there is a change in the Eh or pH of the environment. Stability fields of iron oxides with respect to Eh and pH have been calcu- lated by Garrels and Christ (1965). These indicate that hematite is stable under a strongly oxidizing environment while magnetite is stable under mildly reducing and mildly oxidizing conditions (Figure 25). Polished sections of the blue ores indicate re- placement of magnetite by hematite. Hematite can be seen along the grain borders and the cracks, working inward with a smooth front which shows no dependence on the 51 \- \. \. \. \. \'\ + 0.0 -* ‘\_ — \.\ '\_oe ‘%\. O \ , \, 4}? '\ 0e \. \. \. \. \. + 0.4 "‘ V»— HEMATITE * WATER FezOs * H2 0 +0.! - - EH 0.0 -‘ ~02- -O.4 -‘ ~0.6-4 -C.0-4 Figure 25.--Stability fields of iron oxides as functions of Eh and pH at 25 degrees C and 1 atmosphere total pressure. Dashed line indicates lower stability limit of magnetite in presence of metastable water (after Garrels and Christ, 1965). 52 crystal structure of the magnetite (Figures 19, 20, 21 and 22). When such relations are observed, there can be no reasonable doubt that the magnetite is the older and hematite replaces it. In another specimen (Figure 23), the replacement is dependent on the structure of the magnetite grain. In both types of replacement, it is evident that magnetite is the host. CHAPTER VII CONCLUSION It has been noted earlier that deposition of the various iron minerals is largely controlled by the Eh and pH of the environment. In the environment of deposition, iron in the form of colloids, or of true solution is supplied to the basin in wet seasons from a nearby con- tinental landmass. Ferrous ion is oxidizedto ferric ion due to aeration. Then in the dry seasons when we have less water, the pH of the lake water rises to neutral and ferric hydroxide is precipitated. Then silica is supplied to the basin in colloidal solution and is deposited as chert interlaminated with the iron oxides. This process is repeated upon the return of the wet season when there is a supply of fresh surface water and there is a return from neutral to acid again, in this stage silica is pre- cipitated again. When the pH approaches neutral or alkalinity (dry season) once more, silica and iron hydroxides precipitate together, thus forming layers of silica and iron hydroxides. This iron may have been derived from basic rocks, now metamorphosed to the Seka 53 54 Valley Amphibole Schist, exposed at the time of deposition. Magnetite has been formed at or near the surface as a sedimentary iron formation (itabirite) and has been altered to hematite (martite) (Figures 12 and 13) and complete leaching or replacement of the chert and sili- cates from the itabirites have given rise to soft blue ores. Medium to hard blue ores result from chemical weathering and regional metamorphism of the soft blue ores. All ores have been formed by the removal of silica from the itabirite by surface waters during the wet seasons. In the brown ores, goethite has resulted from the hydration of hematite. During this process, alumina from the sur- face has been added to the surface zones of the brown ore. Based on field observations and microscopic study, the following conclusions can be made: 1. The Nimba Iron Ores of Liberia are of Pre- cambrian age. 2. The ores had an original sedimentary origin. 3. That the original iron mineral was probably magnetite; this has been partially replaced by hematite and enriched by lateritic weather- ing. 4. The ores were highly metamorphosed--perhaps to the sillimanite grade. 55 The ores have been folded but less faulting has occurred. Weathering, related to the present surface, has altered the hematite to goethite; near the surface an increase in Al O and reduction 2 3 in the ore may result from intense in SiO2 laterization effects. Because of similarities with other Pre- cambrian iron ores, it may be assumed that the ores were formed from a lower grade iron formation by leaching of silica and/or enrichment of iron (Figures 3 and 4 strongly suggest this). Finally, Precambrian sandstones in the Nimba area have been altered to quartzites; chemical sediments (iron formation) have been crystal- lized to form itabirites; sediments that were primarily argillaceous have changed to phyllite or schist and volcanic rocks are now gneiss and schist. LIST OF REFERENCES Bastin, Edson S., 1950, Interpretation of ores textures: The Geological Society of America, Mem. 45. Berge, John W., 1966, Genetical aspects of the Nimba iron ores: Geological Society of Liberia, Bull., v. 1, p. 36-430 , 1968, A proposed structural and stratigraphic interpretation of the Nimba-Gbahm range area: Geological, Mining, and Metallurgical Society of Liberia, Bull., v. 3, p. 28-44. Cooke, S. R. B., Microscopic structure and concentrability of the important iron ores of the United States: Bureau of Mines, Bull. 391. Dorr, John Van N., and Barbosa, Aluizio Licinio de Miranda, 1963, Geology and ore deposits of the Itabira Dis- trict, Minas Gerais, Brazil: U.S. Geological Survey, Professional Paper 34l-C. Garrels, Robert M., and Christ, Charles L., 1965, Solutions, minerals and equilibria: New York, Harper and Row. Hough, J. L., 1958, Fresh-water environment of deposition of Precambrian banded iron formation: Jour. of Sedimentary Petrology, v. 28, p. 414-430. Huber, N. K., 1958, The environmental control of sedi- mentary iron minerals: Econ. Geology: v. 53, p. Hurley, P. M., et 31., 1967, Age investigations in Liberia, in VariatiEns in isotopic abundances of strontium, calcium, and argon and related topics: M.I.T. 1381-15, 15th Ann. Progress Rept., Dept. of Geol. and Geophys., Massachusetts Inst. of Technol., Cambridge, p. 1-6. 56 57 James, H. L., 1954, Sedimentary facies of iron formation: Econ. Geology, v. 49, p. 235-293. Krumbein, W. C., and Garrels, R. M., 1952, Origin and classification of chemical sediments in terms of pH and oxidation-reduction potentials: Jour. Geology, v. 60, p. 1-33. Krauskopf, K. B., 1967, Introduction to geochemistry: New York, McGraw-Hill Book Company. Lackschewitz, W., and Tremaine, J. W., 1962, The regional geology of Mt. Nimba, Liberia: Unpublished rept., LAMCO J. V. Operating Co. Leo, G. W., 1967, Gechronology program in Liberia: Geo- logical, Mining, and Metallurgical Society of Liberia, Bull., v. 2, p. 96. Lepp, Henry, and Goldich, Samuel S., 1964, Origin of Pre- cambrian iron formation: Econ. Geology, v. 59, Park, Charles F., and MacDiarmid, Roy A., 1964, Ore deposits: San Francisco, W. H. Freeman and Company. Pettijohn, F. J., 1957, Sedimentary rocks: New York, Harper and Row. Sakamato, Takao, 1950, The origin of the Pre-Cambrian banded iron ores: Am. Jour. Sci., v. 248, p. 449- 474. Turner, Francis J., and Verhoogen, John, 1960, Igneous and metamorphic petrology: New York, McGraw-Hill Book Company. Van Hise, C. R., and Leith, C. K., 1911, The geology of the Lake Superior iron region: U.S. Geol. Survey Monograph 52, p. 499-529. Winkler, H. G. F., 1967, Petrogenesis of metamorphic rocks: New York, Springer-Verlag New York, Inc. Woolnough, W. G., 1941, Origin of banded iron deposits-- a suggestion: Econ. Geology: v. 36, p. 465-489.