Kimberlites Associated with the Lucapa Structure, Angola
Kimberlites associated with the Lucapa structure, Angola
Sandra Elvira Robles Cruz
ADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tdx.cat) ha estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX. No s’autoritza la presentació del seu contingut en una finestra o marc aliè a TDX (framing). Aquesta reserva de drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita de parts de la tesi és obligat indicar el nom de la persona autora.
ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tdx.cat) ha sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR. No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR (framing). Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la persona autora.
WARNING. On having consulted this thesis you’re accepting the following use conditions: Spreading this thesis by the TDX (www.tdx.cat) service has been authorized by the titular of the intellectual property rights only for private uses placed in investigation and teaching activities. Reproduction with lucrative aims is not authorized neither its spreading and availability from a site foreign to the TDX service. Introducing its content in a window or frame foreign to the TDX service is not authorized (framing). This rights affect to the presentation summary of the thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate the name of the author. Kimberlites associated with the Lucapa structure, Angola
Sandra Elvira Robles Cruz
Kimberlites associated with the Lucapa structure, Angola
by Sandra Elvira Robles Cruz
BIENNIUM 2007-2008 Ciencies de la Terra
PhD. Thesis ACADEMIC DISSERTATION
Departament de Cristal·lografia, Mineralogia i Dipòsits Minerals Facultat de Geologia Universitat de Barcelona 2012
1
Supervisors: Dr. Joan Carles Melgarejo Draper
Universitat de Barcelona
Dr. Salvador Galí Medina
Universitat de Barcelona
Dr. Mónica Escayola
CONICET-IDEAN
Committee: Dr. José Mangas Viñuela (President)
Universidad de Las Palmas de Gran Canaria
Dr. Joaquín A. Proenza F. (Secretary)
Universitat de Barcelona
Dr. M. Pura Alfonso Abella (Comm. Member)
Universitat Politècnica de Catalunya
Dr. Maite García Vallès (Alternate) Dr. Fernando Gervilla L. (Alternate)
Universitat de Barcelona Universidad de Granada
Cover: View from northwest of the Catoca mine, Angola.
2
ABSTRACT
Six kimberlite pipes within the Lucapa structure in northeastern Angola have been investigated using major and trace element geochemistry of mantle xenoliths, macro- and megacrysts.
Geothermobarometric calculations were carried out using xenoliths and well-calibrated single crystals of clinopyroxene. Geochronological and isotopic studies were also performed where there were samples available of sufficient quality.
Results indicate that the underlying mantle experienced variable conditions of equilibration among the six cites. Subsequent metasomatic enrichment events also support a hypothesis of different sources for these kimberlites. The U/Th values suggest at least two different sources of zircon crystals from the Catoca suite. These different populations may reflect different sources of kimberlitic magma, with some of the grains produced in U- and Th-enriched metasomatized mantle units, an idea consistent with the two populations of zircon identified on the basis of their trace element compositions.
Calculated temperature and pressure from xenoliths are less scattered than T-P data calculated from single crystals. The calculated northeastern Angola paleogeotherm is consistent with a single value for the CA and the CU79 kimberlites. The differences in T-P values between these kimberlites may reflect the different way each kimberlite sampled the lithosphere. The lithospheric thickness calculated from the northeastern Angola paleogeotherm yielded 192 km.
This research shows that the absence of fresh Mg-rich ilmenite in the Catoca kimberlite (one of the largest bodies of kimberlite in the world), as well as the occurrence of Fe3+-rich ilmenite, do not exclude the presence of diamond in the kimberlite. This is a new insight into the concept of ilmenite and diamond exploration, and leads to the conclusion that compositional attributes must be evaluated in light of textural attributes.
3
The tectonic setting of northeastern Angola was influenced by the opening of the South Atlantic
Ocean, which reactivated deep NE–SW-trending faults during the early Cretaceous. The new interpretation of a kimberlitic pulse during the middle of the Aptian and the Albian, which provides precise data on the age of a significant diamond-bearing kimberlite pulse in Angola, will be an important guide in future exploration for diamonds. These findings contribute to a better understanding of the petrogenetic evolution of the kimberlites in northeastern Angola and have important implications for diamond exploration.
Keywords: kimberlite; Angola; ilmenite; garnet; clinopyroxene; diamond; zircon; xenolith, mantle, Lucapa.
4
RESUMEN
Kimberlitas asociadas a la estructura Lucapa fueron estudiadas mediante geoquímica de elementos mayoritarios y elementos traza tanto en xenolitos del manto, como en macro- y megacristales provenientes de seis chimeneas kimberlíticas localizadas en el noreste de Angola. Cálculos geotermobarométricos se realizaron utilizando xenolitos del manto y cristales individuales de clinopiroxeno bien calibrados. Estudios geocronológicos e isotópicos se realizaron en aquellos casos donde se contaba con muestras de buena calidad disponibles.
Los resultados indican que el manto subyacente experimentó diferentes condiciones de equilibrio.
Eventos posteriores de enriquecimiento metasomático también apoyan la hipótesis de diferentes fuentes para estas kimberlitas. Los valores de U/Th sugieren al menos dos fuentes diferentes para los cristales de circón provenientes de la kimberlita de Catoca. Estas poblaciones diferentes puede reflejar diversas fuentes de magma kimberlítico, donde algunos de los granos podrían haberse producido en unidades del manto metasomatizadas y enriquecidas en U y Th, una idea que es coherente con las dos poblaciones de circón identificados con base en composiciones de elementos traza.
Los valores de temperatura y presión calculados a partir de xenolitos muestran menor dispersión que los datos TP calculados a partir de cristales individuales. La paleogeoterma calculada para las kimberlitas de CA y CU79 se ajusta a un solo rango de valores. En general, las diferencias en los valores de PT entre estas kimberlitas pueden reflejar la forma diferencial como cada kimberlita muestrea la litosfera. El espesor de la litosfera calculado a partir de la paleogeoterma es de 192 km para el noreste de Angola.
Esta investigación también demuestra que la ausencia de ilmenita fresca rica en Mg en la kimberlita de Catoca (una de las kimberlitas más grandes del mundo), así como la presencia de ilmenita rica en Fe3+ no excluye la presencia de diamantes en dicha kimberlita. Esta es una nueva
5 visión sobre el concepto de ilmenita en la exploración de diamantes, y conduce a la conclusión de que los estudios de composición deben estar acompañados de caracterizaciones texturales.
El ambiente tectónico en el noreste de Angola fue influenciado por la apertura del Océano
Atlántico Sur, lo cual reactivó profundas fallas con tren NE-SW durante el Cretácico temprano. La nueva interpretación de un pulso kimberlítico durante la mitad del Aptiense y Albiense proporciona datos precisos sobre la edad de un pulso kimberlítico diamantífero muy significativo en Angola, esta información será una guía importante para futura exploración de diamante. Estos resultados también contribuyen a una mejor comprensión de la evolución petrogenética de las kimberlitas en el noreste de
Angola y tienen importantes implicaciones para la exploración de diamante.
Palabras clave: kimberlita; Angola; ilmenita; granate; clinopiroxeno; diamante; circón; xenolito, manto, Lucapa.
6
TABLE OF CONTENTS
ABSTRACT...... 3
RESUMEN…...... 5
TABLE OF CONTENTS...... 7
LIST OF ORIGINAL PUBLICATIONS AND PARTICIPATION OF SERC IN EACH PUBLICATION……...... 8
PREFACE...... 10
CHAPTER 1 - INTRODUCTION...... 11 1.1 Kimberlites…………...... 11 1.2 Diamond production from kimberlites...... 14 1.3 Diamond production in Angola……...... 16 1.4 Geology of Northeastern Angola...... 17 1.5 The aim of the thesis...... 19 1.6 Methodology...... 21 1.7 Structure of the thesis...... 24
CHAPTER 2 – REVIEW AND RESULTS OF ORIGINAL PUBLICATIONS……………………………………………………………..………...... 25 2.1 Paper I...... 25 2.2 Paper II………………...... 26 2.3 Paper III……………...... 27 2.4 Paper IV……………...... 28 2.5 Paper V……………...... 28 2.6 Paper VI……………...... 30
CHAPTER 3 –DISCUSSION...... 31 3.1 The SCLM beneath Angola and implications for diamond exploration...... 31 3.2 Heterogeneous mantle and metasomatism revealed by subsolidus reactions in ilmenite...... 32 3.3 Diamond potential and regional comparison among diamondiferous and barren kimberlites...... 34 3.4 Future research...... 36
CHAPTER 4 – MAIN CONCLUSIONS...... 37
ACKNOWLEDGMENTS...... 39
REFERENCES...... 41
ORIGINAL PUBLICATIONS...... 49
RESUMEN DE LA TESIS EN ESPAÑOL...... 111
7
LIST OF ORIGINAL PUBLICATIONS AND PARTICIPATION OF SERC IN EACH PUBLICATION
This thesis includes the following six publications:
Paper I. Robles-Cruz, S., Watangua, M., Melgarejo, J.C., Galí, S., 2008. New Insights into the
Concept of Ilmenite as an Indicator for Diamond Exploration, Based on Kimberlite Petrographic
Analysis. MACLA - Revista de la Sociedad Española de Mineralogía, September No. 9, 205-206.
Published.
Paper II. Robles-Cruz, S.E., Watangua, M., Melgarejo, J.C., Gali, S., Olimpio, A., 2009. Contrasting compositions and textures of ilmenite in the Catoca kimberlite, Angola, and implications in exploration for diamond. Lithos 112S, 966-975. Published.
Paper III. Robles-Cruz, S., Lomba, A., M., Melgarejo, J., Galí, S., Olimpio, A., 2009. The Cucumbi
Kimberlite, NE Angola: Problems to Discriminate Fertile and Barren Kimberlites. MACLA - Revista de la Sociedad Española de Mineralogía, September No.11, 159-160. Published.
Paper IV. Robles-Cruz, S.E., Escayola, M., Melgarejo, J.C., Watangua, M., Galí, S., Gonçalves, O.A.,
Jackson, S., 2010. Disclosed data from mantle xenoliths of Angolian kimberlites based on LA-ICP-MS analyses, in: Acta Mineralogica-Petrographica. Abstract Series, Vol. 6, pp. 553. Published.
Paper V. Robles-Cruz, S.E., Escayola, M., Jackson, S., Galí, S., Pervov, V., Watangua, M., Gonçalves,
O.A., Melgarejo, J.C., 2012. U–Pb SHRIMP geochronology of zircon from the Catoca kimberlite,
Angola: Implications for diamond exploration. Chemical Geology 310-311, 137-147. Published.
Paper VI. Robles-Cruz, S.E., Melgarejo, J.C., Galí, S., Escayola, M., 2012. Major- and trace-element compositions of indicator minerals that occur as macro- and megacrysts, and of xenoliths, from kimberlites in northeastern Angola. Minerals, Special Issue "Advances in Economic Minerals".
Officially accepted for publication.
8
S.E. Robles-Cruz’s contribution to the multi-authored paper was:
Papers I, III, and IV, she participated in the fieldwork and sampling. She carried out the petrography studies, SEM imaging, mineral chemistry analyses (microprobe and LA-ICP-MS), processing, and writing the papers.
Paper II, she participated in the fieldwork and sampling. She carried out the petrography studies,
SEM imaging, mineral chemistry analyses, processing, and writing the manuscript for the most part.
Paper V, she participated in the fieldwork and sampling. She carried out the petrography studies,
SEM imaging, mineral chemistry analyses, LA-ICP-MS analyses, preparation of samples for
SHRIMP analyses, processing and interpretation of raw data from LA-ICP-MS and SHRIMP analyses, and writing the manuscript.
Paper VI, she participated in the fieldwork and sampling. She carried out the petrography studies,
SEM imaging, mineral chemistry analyses, LA-ICP-MS analyses, preparation of samples for Sm/Nd analyses, processing and interpretation of data, and writing the manuscript.
9
PREFACE
Kimberlites are one of the most fascinating types of rocks from the Earth. They are complex rocks and provide significant information about the mantle. As well, the study of kimberlites contributes to a better understanding of the evolution of the planet. The study of kimberlites also has economic relevance, since they can trap diamonds during their ascent.
I began my Ph.D. Project in 2008 at the Departament of Cristal·lografia, Mineralogia i Dipòsits
Minerals, Facultat de Geologia, Universitat de Barcelona, with the financial support of a 3-year FI grant and then a BE 6-month grant, both sponsored by the Departament d'Educació i Universitats of the Generalitat de Catalunya and European Social Fund.
This project was the continuation of the Diploma de Estudios Avanzados (DEA) I presented in
2007 under the supervision of Professor Joan Carles Melgarejo i Draper. The PhD research project was directed by Prof. Joan Carles Melgarejo i Draper and Prof. Salvador Galí, both professors from the Department of Cristal·lografia, Mineralogia i Dipòsits Minerals department, Facultat de Geologia,
Universitat de Barcelona. Dr. Monica Escayola from CONICET-IDEAN Instituto de Estudios
Andinos, Laboratorio de Tectónica Andina, Universidad de Buenos Aires also participated as co- advisor of this Ph.D. thesis. The Ph.D. project was supported by the projects CGL2005-07885/BTE and CGL2006-12973 of Ministerio de Educación y Ciencia (Spain).
10
CHAPTER 1 – INTRODUCTION
1.1 Kimberlites
Kimberlites are relatively rare rocks of great scientific and economic importance. The name
“kimberlite” name was proposed by Professor Henry Carvil Lewis in 1887 and since then that is how the rock has been known as (Lewis, 1887; 1888). Lewis described the rock as a type of volcanic breccia, a porphyritic mica-bearing peridotite (Mitchell 1995). The name followed the type-locality rules of nomenclature at that time; it was named after the locality Kimberley, South Africa. Two large groups of kimberlites (group I and II) were introduced by Smith (1983), based on isotopic studies.
Smith et al. (1985) and Skinner (1986, 1989) proposed that kimberlites could be divided in these two distinct groups: group I (kimberlites sensu stricto), and group II (orangeites, phlogopite-rich
“kimberlites”). Later, several studies clearly established that group I and II “kimberlites” are mineralogically and geochemically quite distinct, and group II rocks have closer affinities to lamproites than to group I kimberlites (Mitchell, 1995, and references therein).
Kimberlites, also known as group I kimberlites (Mitchell 1995), are defined as volatile-rich
(dominantly CO2) potassic ultrabasic rocks that usually show a distinctive inequigranular texture as a result of the presence of crystals (macro- and megacrysts) and xenoliths inside a fine-grained matrix
(Clement and Skinner, 1985; Mitchell, 1986). The mineralogy of kimberlites is very variable and complex. Mega- and macrocrysts are mainly composed of olivine, magnesian ilmenite, Cr-poor titanian pyrope, diopside, phlogopite, enstatite, and Ti-poor chromite; where olivine macrocrysts are a characteristic component except in fractionated kimberlites (Mitchell 1995). Some kimberlites may also contain diamond. Mantle and crustal xenoliths can be also present in kimberlites. The fine- grained matrix may include a second generation of primary euhedral-to-subhedral olivine, monticellite, phlogopite, perovskite, spinel, apatite, and serpentine (Mitchell, 1995). It has been also reported (Kamenetsky et al., 2004) that the groundmass is extremely enriched (at least 8 wt.%) in
11 water-soluble alkali chlorides, alkali carbonates, and sulfates (proportion 5:3:1), and commonly shows immiscibility textures between these phases.
Kimberlites occur as pipe intrusions (Figure 1.1) with an upper crater facies, intermediate diatreme facies, and deep hypabyssal facies (Clement and Skinner, 1985). These facies were produced by explosive emplacement under volcanic and subvolcanic conditions. Crater facies rocks are divided into lavas, pyroclastic rocks, and resedimented volcaniclastic rocks; kimberlite diatremes are cone shaped, composed of clasts of cognate or xenolithic origin with or without matrix, and classified as
“tuffisitic kimberlite” and “tuffisitic kimberlite breccia” (Clement 1982; Clement and Skinner, 1985;
Mitchell 1995); and hypabyssal kimberlites comprise the root zones of diatreme and occur as dikes and sills (Mitchell 1995, and references therein).
The study of xenoliths, megacrysts (crystals greater than 1 cm in their maximum dimension) and macrocrysts (0.5-10 mm) from kimberlites play an important role in the understanding of the characteristics of the mantle and the kimberlite petrogenesis itself. Minerals such as pyrope and eclogitic garnet, chrome diopside, Mg-rich ilmenite, chromite and, to a lesser extent, olivine in superficial materials (tills, stream sediments, loam, etc.) are one of the most important tools, other than bulk sampling, to assess the diamond content of a particular pipe (Pell, 1998), consequently they are called indicator minerals.
Kimberlites are preferentially associated with cratons worldwide (Figure 1.2). Diamondiferous kimberlites have been reported as Proterozoic to Tertiary in age, with diamond crystals that vary from early Archean to as young as 990 Ma (Pell 1998). In 1995 there were already 5000 kimberlites identified, and 10% of them were diamondiferous (Janse and Sheahan 1995).The first kimberlite was discovered in 1869 in South Africa where the first diamond from primary deposit was found. Three years later kimberlites were recognized as primary deposit for diamond (Janse and Sheahan 1995).
Diamond, however, is not genetically related to kimberlites, but rather it is a xenocryst that is formed in the upper mantle. Diamond in kimberlites can be found as sparse xenocrysts or diamondiferous xenoliths hosted by intrusives emplaced as subvertical pipes or resedimented volcaniclastic and pyroclastic rocks deposited in craters (Pell 1998). Most of the natural diamond crystals come from peridotite and in less proportion (33%) from eclogitic sources (Stachel and Harris, 2009, and
12 references therein). The research of kimberlites and natural diamond from kimberlites had a significant impetus since the 1st International Kimberlite Conference in 1973.
Figure 1.1 Idealized diagram of a kimberlite magmatic system (after Mitchell 1995)
Mitchell (1986) defined kimberlites using the typomorphic assemblage of primary minerals and emphasizing their petrologic characteristics as: “Kimberlites are inequigranular alkalic peridotites containing rounded and corroded megacrysts of olivine, phlogopite, magnesian ilmenite and pyrope set in fine-grained groundmass of second generation euhedral olivine and phlogopite together with primary and secondary (after olivine) serpentine, perovskite, carbonate (calcite and/or dolomite) and spinels. The spinels range in composition from titaniferous magnesian chromite to magnesian ulvöspinel-magnetite. Accessory minerals include diopside, monticellite, rutile and nickeliferous sulphides. Some kimberlites contain major modal amounts of monticellite”.
13
Figure 1.2 Kimberlites and major diamond mines worldwide after Janse (2007) and Eckstrand et
al. (1995).
There are three large cratons in Africa: the South African, the West African, and the Central
African cratons. Most of the world’s active major kimberlite diamond mines are located in South
Africa, Botswana, Zimbabwe and Swaziland, on the South African craton which includes the Kalahari
Archon. The West African Craton includes the Man Archon and the Eburnean Proton. The Central
African craton includes two important archons: the Lunda-Kasai (Angola and Congo) and the
Tanzanian (Janse and Sheahan 1995). In Angola, kimberlite pipes and dykes are distributed in the northeast, central, and southwest part of the country. Most of the diamondiferous kimberlites in
Angola are concentrated in clusters in the northeastern area. There are also several alluvial mining areas in Lunda and Cuango, Angola (Llusià et al., 2005).
1.2 Diamond production from kimberlites
Worldwide diamond production from kimberlites is not easy to track since not all values are published and sometimes when they are published they may vary from one publication to another.
The Kimberley Process Certificate Scheme (KPCS) that monitors world rough diamond trade came into effect on 1 January 2003 (Read and Janse, 2009). This was the first real attempt to at least restrain
14 trade in conflict diamonds and to provide surveillance of the rough diamond trade from producers to merchants. Diamond production like other commodities (e.g., gold) depends on demand. Botswana,
Russia, Canada, South Africa, and Angola (in this order) were the top five diamond producing countries by value responsible of the 83% of the total world production in 2009. This represented the
65% of the total diamond production by weight in 2009, since DRC and Australia were in the top five producers by weight but they produce diamonds low in value (Read and Janse, 2009). Figure 1.3 shows the total diamond production for the main 38 kimberlites worldwide until 2009. Currently, the top five diamond producing countries by value are Botswana, Russia, Canada, South Africa and
Angola.
Figure 1.3 Diamond production worldwide (after Janse, 2007; Read and Janse, 2009)
15
1.3 Diamond production in Angola
Kimberlites in Angola are important not only because most of them are slightly eroded so their crater facies are well preserved, but also because there are large and high-grade kimberlites with significant potential diamond reserves (Khar'kiv et al., 1992). The first kimberlite in Angola, the
Camafuca-Camzambo pipe, was discovered in 1947 (De Andrade 1954). However, the first kimberlite that came into production in 1997 was the Catoca pipe, which was discovered in 1985. The civil war in Angola, between 1961 and 2002 (Blore, 2004), hampered progress in the country. The guerrillas controlled the richest diamond provinces and mined them illegally in part to fund their activities. The
National Union for the Total Independence of Angola (UNITA) and the Revolutionary United Front
(RUF), both acted against the international community's objectives of restoring peace in Angola
(Blore 2007). It is clear that “informal” diamond production was much higher than the official values e.g., UNITA’s smuggled production is estimated to have been worth close to $1 billion in 1996 (Janse
2007, and references therein). Angola was the first country to implement a full certificate of origin for diamond exports (at the beginning of 2000) following United Nations sanctions on UNITA’s diamond trading in 1998 and the beginning of investigation into illegal diamond trading in 1999 (Blore, 2004).
The goal of this certificate was to verify the exclusion of conflict diamonds. After 2000 and especially once the civil war ended up in 2002, the mining activities accelerated in Angola. The Catoca pipe passed from 2 Mct/year in 2000 to produce 6.7 Mct/year in 2007 (Read and Janse, 2009). Diamond production in Angola represents the 1% of the gross domestic product (GDP) of Angola (Bermúdez-
Lugo 2004). Estimated reserves in Angola are of 50 million carats in kimberlite pipes (Partnership
Canada Africa, 2004). Ore reserves in the Catoca pipe are given as 84 million tonnes to yield 60 million carats to 150 m depth or as 270 million tonnes to yield 195 Mct to 600 m depth (Read and
Janse, 2009). The mining in Camafuca pipe started in 2007 (low cost operation dredging the river bed) to recover 200,000 ct/yr for five years on a reserve of 13 Mct, which are contained in fluvial mud and sand grading into highly weathered kimberlite (Read and Janse, 2009). The Camatchia–Camagico mine in Angola is developed on two kimberlite pipes, where a reserve of 80 Mct has been estimated
(Read and Janse, 2009). Figure 1.3 includes the diamond production from Angolan kimberlites.
16
1.4 Geology of northeastern Angola
Angola is endowed with mineral resources that are the result of a relatively complex geological history. The study of its cratonic lithospheric mantle is important both for the light it sheds on the physical behaviour of old continents, as well as in contributing to our understanding of Angola's mineral potential.
Angola geology can be represented by three main stages (De Carvalho et al., 2000; Guiraud et al.,
2005, Figure 1.4): (1) An important Archean orogeny, registered by the Central Shield, Cuango Shield and Lunda Shield, most of them composed of gabbro, norite and charnockitic complexes, which constitute the Angolan basement. (2) Three main Proterozoic cycles, Eburnean-Paleoproterozoic,
Kibaran-Mesoproterozoic, and Pan-African-Neoproterozoic; being the Eburnean the most important and characterized by complex volcanosedimentary groups, gneisses and migmatites, granites and syenites. This regional Paleoproterozoic event was followed by the Kibaran cycle, which was related to extensional events that occurred on the border of Congo craton and that later generated clastic- carbonatic sequences and local basic magmatism. The Pan-African orogeny was associated with the development of Gondwana and leaded the generation of fold belts and granitic intrusions. The activation of zones of lithospheric weakness, especially major fault zones, favoured the subsequent break-up of Gondwana. (3) The deposition of Phanerozoic sedimentary sequences resting unconformably on previously eroded surfaces (Pereira et al., 2003). The subsequent break-up of
Gondwana, during the Jurassic to Cretaceous, between 190 and 60 Ma (e.g., Jelsma et al., 2004), caused the development of basins that are associated with deep fault systems in Angola. These fault systems facilitated the emplacement of alkaline, carbonatitic, and kimberlitic magmas (Pereira et al.,
2003).
The Lower Cretaceous regional extension determined the development of deep faults and grabens with trends NE-SW and NW-SE. The Lucapa structure is in the first group (trend NE-
SW).The northeastern part where most of the diamondiferous kimberlites in Angola are found, whereas the southwestern zone comprises important occurrences of undersaturated alkaline rocks and carbonatites (Reis, 1972). More minor kimberlite fields are found in the SW Angola (Egorov et al.,
2007).
17
Figure 1.4 Location map of the area of study. Geological map of northeastern Angola (after
De Araujo et al., 1988; De Araujo and Perevalov, 1998; De Carvalho et al., 2000; Egorov et al.,
2007). Abbreviations: Quaternary (QQ), Cenomanian (CE), Albian (AB), Permian (PP),
Carboniferous (CC), Undifferentiated (Undiff.), Group (Gp), Formation (Fm), sandstone (Sst),
conglomerate (Cgl), limestone (Lst), marlstone (Mrls), argillaceous limestone (ArgLst), claystone
(Clst), granite (Gr), gabbro (Gb), quartzite (Qzt), schist (Sch), granodiorite (Grdr), dolerite (Do),
amphibolite (Am), gneiss (Gns), carbonatites (Cbt), nephelite (Nph), syenite (Syt), ijolite (Ijt),
pyroxenite (Pxt), anorthosite (Ant), troctolite (Trt), Norite (Nrt), epidotite (Epd), granulite (Gnt),
eclogite (Ecl).
Kimberlites from southern Africa, North America, and Russia show similar ages between them. They show alternating periods of abundance or scarcity of kimberlite magmatism (Figure 1.5),
18 especially during Cenozoic/Mesozoic (Heaman et al., 2003; Jelsma et al., 2009). The coincidence of kimberlite occurrences with trans-lithosphere discontinuities may be a result of thermal perturbations.
Such conditions were favoured during rifting and the eventual supercontinent breakup, when a majority of these kimberlites were generated (Heaman et al., 2003). The geological configuration in
Angola, which was consistent with the aforementioned conditions associated with kimberlite generation, apparently set a tectonic control on the presence of kimberlites in Angola. Synsedimentary continental sediments (Calonda Formation) filled the Lucapa structure. The Lucapa structure is an old corridor from an oceanic transform (White et al., 1995), which has been active since Paleoproterozoic
(Jelsma et al., 2009), and is characterized by deep-seated faults associated with carbonatites and kimberlites. The Calonda Formation can also contain diamonds in paleoplacers; alluvial diamonds are found in placers associated with rivers passing across all these diamondiferous areas.
Figure 1.5 Three types of tectonic settings related to kimberlite magmatism. (a) Gondwana
assemblage during Pan-African orogeny. (b) Incipient rifting. Dark gray represents the Karoo basins.
(c) Trans-continental lithospheric discontinuities (gray lines) that have been reactivated, as tectonic
triggers, during the continental extension; and concomitant magmatism (dashed line) in Southern
Africa and South America. The white diamonds represent schematic groups of kimberlites and related
rocks (after Jelsma et al., 2009).
1.5 The aim of the thesis
The Lucapa structure has several hundreds of kimberlites (Figure 1.6). To date, there is no official information about all of them. The knowledge of kimberlites in Angola is low and according
19 to ENDIAMA (2012) only the 40% of the mining resources have been evaluated. Currently, there are
167 mining projects in Angola, 15 of them are active and the most important ones are: Muanga, Alto
Cuilo, Dala, Nhefo, Lunda-Nordeste, Cacuala, and Gango. The kimberlites under current production are: Catoca in Lunda Sul, and Camatchia, Camafuca, Camatue, and Camazuanza in Lunda Norte.
There is no detailed production information available for most of them.
Figure 1.6 Distribution of kimberlites in Angola (including data from Perevalov et al., 1992;
Egorov et al., 2007)
Some mineralogical studies have been carried out in the Catoca pipe to determine the diamondiferous potential of this kimberlite (Ganga et al., 2003; Kotel’nikov et al., 2005). However, there are important questions to address in terms of genesis and evolution of kimberlites in northeastern Angola. Kimberlite magma is considered as derived from the mantle of the Earth at a depth of more than 150 km (Dawson, 1980; Haggerty, 1995). Mantle xenoliths provide information about the subcratonic mantle and lithosphere, as well as melts and fluids associated with mantle
20 metasomatism. The analyses of indicator minerals provide information about the oxygen fugacity conditions, favorable conditions to sample and preserve diamond, and evolution of the kimberlite itself.
Research that I carried out during the “Trabajo de Investigación Tutelado” (TRT - Treball de
Recerca Tutelat) during 2007, established that ilmenite macrocrysts from the Catoca kimberlite exhibit different grades of replacement below 200 m depth, and are almost not visible above this level. This research builds upon the 2007 TRT research to determine if the composition of ilmenite from the Catoca kimberlite indicates one (or alternately multiple) recrystallization events, and explores the relationship between the different types of ilmenite and presence/preservation of diamond in this area. To determine this information, a regional study of kimberlites in the northeastern part of Angola was undertaken using sampling collected from drill cores of kimberlites in the Lunda, Catoca, and Muanga areas.
Another major objective of this thesis is to propose a profile that provides information about the mantle beneath the northeastern Angola based on the study of xenoliths, mega- and macrocrysts from six kimberlites. As well, an evaluation of the conditions that has an influence on diamond distribution along the area of study, based on petrographic and geochemical studies of barren (kimberlite without diamond presence) and diamondiferous kimberlites. Unfortunately, the comparison with barren kimberlites has been hampered because samples from barren kimberlites were the more altered and poor in fresh mantle xenoliths and indicator minerals.
1.6 Methodology
This research was developed in different phases: 1) field work, 2) sampling preparation, 3) analyses, and 4) discussion and writing of manuscripts.
1.6.1 Field work
There was preliminary field work in 2005 when samples from the Catoca kimbelite were collected by J.C. Melgarejo and his research group. Then in 2006, I started my “Trabajo de
Investigación Tutelado” (TRT - Treball de Recerca Tutelat) and I used those samples to start getting
21 an idea about the kimberlites from Angola and the results were presented to obtain the Diploma d'Estudis Avanzats (DEA, Robles-Cruz, 2007). In fall 2007, new field work occurred during which I collected the samples for the Ph.D. thesis. Specifically, about 750 drill cores and heavy-mineral concentrate samples from seventeen kimberlites were obtained (Table 1.1). Only six of the seventeen kimberlites containing samples of good quality were selected to carry out analyses: Catoca (CA),
Tchiuzo (TZ), Anomaly 116 (An116), Alto Cuilo-4 (AC4), Alto Cuilo-63 (AC63), and Cucumbi-79
(CU79).
Presence of diamonds Province Contract Kimberlite Borehole (YES/NO) LUNDA NORTE LUEMBA Tchiuzo 34 YES LUNDA NORTE LUEMBA Tchiuzo 44 YES LUNDA NORTE LUEMBA Tchiuzo G10 YES LUNDA NORTE LUEMBA Tchiuzo G18 YES LUNDA SUL CATOCA Catoca 0335 YES LUNDA SUL CATOCA Catoca 0536 YES LUNDA SUL CATOCA Catoca 033/35 YES LUNDA SUL CATOCA Catoca 044/35 YES LUNDA SUL CATOCA Catoca 77/35 Unknown LUNDA SUL CATOCA Catoca CA135 YES LUNDA SUL CATOCA Catoca CA336 YES LUNDA SUL CATOCA Catoca CA515 YES LUNDA SUL CATOCA Catoca CA535 YES LUNDA SUL CATOCA Catoca CA538 YES LUNDA SUL CATOCA Anomaly CAT-116 116 Some prospectivity LUNDA SUL CATOCA Camitongo 28 Some prospectivity LUNDA SUL LAPI Kambundu 216 NO LUNDA SUL ALTO CUILO Alto Cuilo 1 1 NO LUNDA SUL ALTO CUILO Alto Cuilo 16 11 Some prospectivity LUNDA SUL ALTO CUILO Alto Cuilo 254 5 YES LUNDA SUL ALTO CUILO Alto Cuilo 4 4 Some prospectivity LUNDA SUL ALTO CUILO Alto Cuilo 5 5 NO LUNDA SUL ALTO CUILO Alto Cuilo 63 6 YES LUNDA NORTE MUANGA Cucumbi 45 5 NO (not tested) LUNDA NORTE MUANGA Cucumbi 72 MFD07 YES LUNDA NORTE MUANGA Cucumbi 76 MFD03 NO LUNDA NORTE MUANGA Cucumbi 79 MFD01 YES LUNDA NORTE MUANGA Cucumbi 8 MFD06 NO LUNDA NORTE MUANGA Cucumbi 80 2 NO (not tested) Table 1.1 List of kimberlites sampled for this Ph.D. thesis
1.6.2 Sampling and preparation
This phase started in February 2008, when samples arrived from Angola after passing all the authorization process, and finished in March 2009, before I went to the Geological Survey of Canada
22 to carry out analyses. Kimberlites are very delicate rocks and they need to be prepared properly (low vacuum, no water, and diamond powder for polishing), otherwise they become useless. Unfortunately, some of the samples when revised had to be rejected (bad sample preparation). A set of thin (30 μ) and gross (80-100 μ) sections, and probes was obtained. The classification of the kimberlite textures was followed after Mitchell (1986; 1995), Pearson et al. (2007, and references therein), and Scott
Smith (2012).
1.6.3 Analytical methods
Samples were studied under the optical transmitted and reflective light microscope at the
Department of Cristal·lografia, Mineralogia i Dipòsits Minerals – Faculty of Geology, in order to pick up the best and representative samples to carry out the different type of analyses.
Petrographic studies were performed with a Scanning Electron Microscope – Environmental
Scanning Electron Microscope (SEM-ESEM) with an acopled EDS using BSE (Backscattered electrons) images to identify compositional heterogeneities in the samples (previously dried at 60°C during 24 hours for thin sections and during 72 hours for gross sections, cleaned blowing away powder, and then carbon coating). ESEM Quanta 200FEI, XTE325/D08395 was used to carry out these analyses. High vacuum conditions were preferred to get a precision of less than 0.5 μ in the spot. This equipment has a LINK EDS, which is made up by a Si (Li) crystal, with a Be window. This configuration allows determination of all the elements from Be to U.
Mineral chemistry of major elements were carried out using a Cameca SX-50 microprobe (see parameters at PAPER II) and a JXA JEOL-8900L microprobe (see parameters at PAPER VI). Mineral chemistry of trace elements were accomplished using laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), see parameters in the PAPER V and PAPER VI.
Geochronological U-Pb analyses were conducted using a Sensitive High Resolution Ion
Microprobe II (SHRIMP II), see parameters at PAPER V. Additional Sm/Nd isotopes analyses were carried out using a Thermo Finnigan Triton thermo-ionization mass spectrometer (TIMS).
23
1.7 Structure of the thesis
This PhD thesis presents the results divided in chapters based on the main findings of the research. The second chapter is a revision of the original publications written as part of this Ph.D. thesis. The third chapter will integrate all these analyses and results in a general discussion about the studied kimberlites. Finally, we will present the main conclusions of this research in the chapter fourth and we will suggest the main directions of the future research about kimberlites in Angola. The original publications are included at the end of this thesis.
24
CHAPTER 2 – REVIEW AND RESULTS OF ORIGINAL
PUBLICATIONS
2.1 Paper I
“New Insights into the Concept of Ilmenite as an Indicator for Diamond Exploration, Based on
Kimberlite Petrographic Analysis”
Biannual national journal: MACLA - Revista de la Sociedad Española de Mineralogía,
September No. 9, 205-206.
ISSN: 1885-7264.
Paper I is based on the preliminary findings about ilmenite from the Catoca kimberlite,
Angola. It is a continuation of the “Trabajo de Investigación Tutelado” (TRT – “Treball de Recerca
Tutelat”) carried out by SERC during 2007, and a comparison with ilmenite from the Cucumbi-79 kimberlite. Paper I also includes the regional setting of the area of research. Textural evidences of ilmenite indicate a different complex history of growth in the crystals. Six textural types of ilmenite were identified in Catoca and three compositional types of ilmenite.
The composition of the ilmenite from Catoca is the result of a set of replacement processes with rich fluids in Mg and Mn affecting an oxidized primary ilmenite in a higher or lower grade. These fluids are reducing, especially those rich in Mn. "Picroilmenite" has traditionally been interpreted as an indicator of kimberlite associations, as well as an indicator of low fO2, which is necessary for the preservation of diamond. Although Catoca and Cucumbi are diamondiferous kimberlites, they show that Mg ilmenite is clearly a late replacement product, and the grade of replacement of the primary grains is very variable. Therefore, this paper illustrates that the absence of magnesian ilmenite in a kimberlite does not appear to be a convincing argument to exclude the presence of diamonds. This is a new insight into the concept of ilmenite in diamond exploration.
25
2.2 Paper II
“Contrasting compositions and textures of ilmenite in the Catoca kimberlite, Angola, and implications in exploration for diamond”
Monthly international journal: Lithos 112S, 966-975.
ISSN: 0024-4937. Impact Factor: 3.246 (2011), 5-Year Impact Factor: 3.691 (Thomson Reuters,
2012). Journal in the Science Citation Index (SCI).
Paper II describes a detailed systematic petrographic characterization of the different types of ilmenite from the Catoca kimberlite. The Catoca kimberlite is emplaced in the northeastern part of the
Lucapa structure. The paper focuses on compositional and textural variations in ilmenite from drill- core material, in the hope of elucidating events before and during the emplacement of the kimberlitic magma. We characterize four main compositional variants of ilmenite, with enrichments in Fe3+, Mg,
Mn and nearly stoichiometric ilmenite, in seven textural classes of ilmenite, and distinguished crystals of variable size, ranging from micro- to megacrysts.
Most ilmenite is found to derive, through a complex process, from replacement of Fe3+-rich
3+ ilmenite, presumably originating by mantle metasomatism at a relatively high fO2. This Fe -rich ilmenite reacted with fluids under reducing conditions, producing Mg-rich ilmenite. The Mn-rich ilmenite is produced by interaction with a late CO2-rich fluid. The Mg-rich ilmenite is here clearly a minor phase and a late product of replacement. The absence of fresh Mg-rich ilmenite and the occurrence of Fe3+-rich ilmenite do not seem to be convincing arguments to exclude the presence of diamond crystals in a kimberlite.
The ilmenite macro- and megacrysts are assumed to be produced by disaggregation of ilmenite- bearing xenoliths (mainly relatively oxidized and metasomatized mantle peridotites and minor carbonatites). The subsequent reaction under disequilibrium conditions with kimberlite-derived fluids produced the replacement of the above macro- and megacrysts by secondary Mg-rich ilmenite. Late subsolidus reactions with the fluids associated with the kimberlite, also in disequilibrium conditions, produced the replacement of the early ilmenite types by highly reduced Mn-rich ilmenite. The enrichment in Nb of this late ilmenite (and in the ilmenite of the matrix), as well as its intimate
26 association with carbonates of Ba and Sr, indicate the interaction of the ilmenite crystals with a CO2- rich fluid.
This work proposes a new understanding of the connection between the search of ilmenite in diamond exploration: compositional attributes must be evaluated in light of textural attributes.
Although Catoca is a diamondiferous kimberlite, most of its ilmenite compositions are strongly oxidized and poor in Cr and Mg. Therefore, an important conclusion of Paper II is that the absence of
Mg-rich ilmenite in a kimberlite, or the absence of its corresponding placers, do not appear to be a convincing argument to exclude the occurrence of economic deposits of diamond in a kimberlite.
2.3 Paper III
“The Cucumbi Kimberlite, NE Angola: Problems to Discriminate Fertile and Barren
Kimberlites”
Biannual national journal: MACLA - Revista de la Sociedad Española de Mineralogía,
September No.11, 159-160.
ISSN: 1885-7264.
Paper III focuses on the petrography and composition of samples from the Cucumbi kimberlite.
The garnet compositions from Cucumbi-79 are plotted using the diagram of Grütter et al. (2004). The compositions plot into the graphite domain, out of the diamondiferous field harzburgitic G10 facies.
Based solely on this criterion the kimberlite would be classified as barren. However, the
Cucumbi kimberlite is diamondiferous. Similar problems were found in the Catoca pipe when using the composition of ilmenite or the composition of garnets. Therefore, the paper concludes that the garnet diagrams can be used to verify the minimum level of diamond content, but some kimberlites may contain diamond samples from deeper sources and that this should be taken into consideration when using these diagrams to assess the potential of kimberlite fields.
It is also important to mention that new diagrams have been proposed (i.e., Grütter et al.,
2006, McLean et al., 2007), where they integrate several attributes at once, and they seem to be a better tool for exploration.
27
2.4 Paper IV
“Disclosed data from mantle xenoliths of Angolan kimberlites based on LA-ICP-MS analyses”
National journal: Acta Mineralogica-Petrographica. Abstract Series, Vol. 6, pp. 553.
Published by the Department of Mineralogy, Geochemistry and Petrology, University of Szeged,
Hungary.
ISSN 0324-6523.
This paper presents preliminary observations of the type of xenoliths found in the Catoca and
Cucumbi-79 diamondiferous kimberlites, and the first set of analyses of Laser Ablation-Inductively
Coupled Plasma-Mass Spectrometry (LA-ICP-MS) from xenoliths. Two main different trends for garnet can be identified in the Catoca kimberlite based on Rare Earth Element (REE) patterns.
Eclogitic garnet has “normal” normalized Rare Earth Element (REEN, McLean et al., 2007) patterns, whereas garnet from lherzolite xenoliths usually has “sinusoidal” REEN patterns and rarely “normal”
REEN patterns. Clinopyroxene from eclogitic associations is Light REE (LREE) enriched. Garnet from the lherzolite xenoliths is characterized by a LREE-enrichment, a maximum around the LREE-
Heavy REE (HREE) limit and flat HREE.
Unlike in Cucumbi-79, garnet from lherzolite xenoliths presents “normal” patterns with lower
REE values. Garnet from phlogopite-rich xenoliths presents “normal” patterns, but their values are significantly (about 10x chondritic value) lower. Only clinopyroxene from phlogopite-rich xenoliths exhibits higher values in LREE than the same xenoliths in the Catoca pipe.
Data indicate that the mantle sampled by these two kimberlites might have been under different equilibration conditions and different degrees of metasomatism.
2.5 Paper V
“U–Pb SHRIMP geochronology of zircon from the Catoca kimberlite, Angola:
Implications for diamond exploration”
Semimonthly international journal: Chemical Geology 310-311, 137-147.
ISSN: 0009-2541. Impact Factor: 3.518 (2011), 5-Year Impact Factor: 4.063 (Thomson Reuters,
2012). Journal in the SCI.
28
Paper V presents the first age determinations of zircon from the diamondiferous Catoca kimberlite in northeastern Angola, the fourth largest kimberlite body in the world. The U–Pb ages were obtained using a Sensitive High Resolution Ion Microprobe II (SHRIMP II) on zircon crystals derived from tuffisitic kimberlite (TK) rocks and heavy-mineral concentrates from the Catoca kimberlite.
The SHRIMP results define a single weighted mean age of 117.9±0.7 Ma (Mean square weighted deviation MSWD=1.3). More than 90% of the results indicate a single age population. There is no evidence for variable ages within single crystals, and no diffusional profiles are preserved. These data are interpreted as the maximum age of the kimberlite eruption at Catoca. The U/Th values suggest at least two different sources of zircon crystals. These different populations appear to indicate different sources of kimberlitic magma, with some of the grains produced in U- and Th-enriched metasomatized mantle units.
This understanding is consistent with the two populations of zircon identified, based on REE abundances determined by LA-ICP MS analyses in this paper. One population originated from a depleted mantle source with low total REE (less than 25 ppm), and the other was derived from an enriched source, likely from the mantle or a carbonatite-like melt with high total REE (up to 123 ppm).
The tectonic setting of northeastern Angola has been influenced by the opening of the south
Atlantic, which reactivated deep NE–SW-trending faults during the early Cretaceous. The eruption of the Catoca kimberlite correlates with these regional tectonic events. The Calonda Formation (Albian–
Cenomanian age) is the earliest sedimentary unit that incorporates eroded material derived from the diamondiferous kimberlites. Thus, the age of the Catoca kimberlite eruption is restricted to the time between the middle of the Aptian and the Albian. This new interpretation will be an important guide in future exploration for diamonds because it provides precise data on the age of a diamond-bearing kimberlite pulse in Angola.
29
2.6 Paper VI
“Major- and trace-element compositions of indicator minerals that occur as macro- and megacrysts, and of xenoliths, from kimberlites in northeastern Angola”
Quarterly international journal: Minerals, Special Issue "Advances in Economic Minerals"
(submitted revised version of the manuscript).
ISSN: 2075-163X. Peer-reviewed open access journal. Published by the Multidisciplinary Digital
Publishing Institute (MDPI).
Paper VI compares the major- and trace-element compositions of olivine, garnet, and clinopyroxene that occur as single crystals (142 grains), with those derived from xenoliths (51 samples) from six kimberlites in the Lucapa area, northeastern Angola: Tchiuzo, Anomaly 116,
Catoca, Alto Cuilo-4, Alto Cuilo-63, and Cucumbi-79.
The samples were analyzed using electron probe microanalysis (EPMA) and LA-ICP-MS.
The results suggest different paragenetic associations for these kimberlites in the Lucapa area.
Compositional overlap in some of the macrocryst and mantle xenolith samples indicates a xenocrystic origin for some of those macrocrysts. The presence of mantle xenocrysts suggests a possibility diamond being present. Geothermobarometric calculations were carried out using EPMA data from xenoliths applying the program PTEXL.XLT. Additional well calibrated single-clinopyroxene thermobarometric calculations were also applied.
Results indicate the underlying mantle experienced different equilibration conditions.
Subsequent metasomatic enrichment events also support a hypothesis of different sources for the kimberlites. These findings contribute to a better understanding of the petrogenetic evolution of the kimberlites in northeastern Angola and have important implications for diamond exploration.
30
CHAPTER 3 – DISCUSSION
3.1 The SCLM beneath Angola and implications for diamond exploration
The characterization of the sub-continental lithospheric mantle (SCLM) is important for identifying the evolution of continents and their mineral potential (Pearson and Wittg, 2008). In particular, the mantle xenolith suite and certain garnet and clinopyroxene xenocrysts provide information regarding the composition and structure of the SCLM. In cases where fresh xenoliths are poor or absent, xenocrysts are very useful. Although they do not provide information as precise as xenoliths, xenocrysts can give a statistically reliable sample of the underlying mantle (Schulze, 1995).
Figure 3.1 Schematic model comparing diamondiferous and barren kimberlites from northeastern
Angola (modified after Haggerty, 1986; Mitchell, 1986; Mather et al., 2011). Lithosphere-
asthenosphere boundary (LAB), graphite (G), diamond (D).
31
The calculated temperature and pressure from xenoliths (PAPER VI), define a single paleogeotherm value for the CA and the CU79 kimberlites, and yielded a lithospheric thickness of
192 km calculated based on this paleogeotherm. A quantitative comparison between Angola lithosphere and geotherms from Bultfontein and Finsch kimberlites in southern Africa indicates a slightly cooler (steeper) paleogetherm for Angola than the paleogeotherms calculated from southern
Africa. This is consistent with the map of the lithospheric thickness of Southern Africa from shear wave velocities (Preistley and McKenzie, 2006), which indicates a thickness >180 km.
3.2 Heterogeneous mantle and metasomatism revealed by subsolidus reactions in ilmenite
Ilmenite from kimberlites in northeastern Angola are very particularly important because it is a common mineral that provides relevant information about their chemical environment and exhibit
2+ 3+ variable Fe :Fe ratios. This ratio is significantly influenced by fO2 conditions, and these conditions can be used to determine the preservation or destruction of diamond. The detailed petrographic study of the Catoca kimberlite (PAPER II) suggests a complex history for the ilmenite nodules. The diversity in textures and composition reflects the paragenetic position of ilmenite in the kimberlite
(accessory in xenoliths, macro- and megacrysts, matrix) and the replacement processes. We propose that most if not all of the ilmenite nodules are produced by disaggregation of ilmenite-bearing metasomatized peridotite xenoliths.
The composition of the early ilmenite is unusual because of its high Fe3+ contents. Similar Fe3+- rich compositions, although rare in kimberlites, have also been found in the Koidu kimberlite, in
Sierra Leone (Tompkins and Haggerty, 1985). The ilmenite from the Catoca pipe is even more strongly oxidized, indicating crystallization under relatively high fO2 conditions. Ilmenite is also replaced along small discontinuities, both in the grain borders and along internal surfaces by Mg-rich ilmenite. The replacement of the Fe3+-rich ilmenite by Fe2+- and Mg-rich ilmenite is indicative of a trend toward more reducing conditions (Haggerty and Tompkins, 1983). This type of sequence is similar to the so-called ilmenite magmatic trend (Haggerty et al., 1977; Pasteris, 1980; Schulze,
32
1984). However, the textural patterns attributable to replacement at Catoca, along grain borders, cracks or other discontinuities, strongly suggest the action of a fluid rather than a magma. It is difficult to ascertain the timing and place of this replacement. Certainly it was produced before kimberlite emplacement, because some nodules broken during the explosive processes are not replaced in the broken corners.
The replacement of the Fe3+-rich ilmenite by Fe2+- and Mn-rich ilmenite is also indicative of a trend toward strongly reducing conditions (Haggerty and Tompkins, 1983). Accordingly, these compositions could follow the kimberlite reaction trend of Haggerty et al. (1977), producing enrichment in Fe2+. The most significant aspects in this process are the strong enrichments in Mn and
HFSE. Similar enrichments have been interpreted in other kimberlites worldwide as produced by crystallization at the expense of a late-stage fraction of melt (Tompkins and Haggerty, 1985,
Chakhmouradian and Mitchell, 1999). In the Catoca case, two facts suggest the deposition of this ilmenite under the influence of a CO2-rich fluid phase: a) the intimate association of this Mn-rich ilmenite along with calcite, witherite, barytocalcite and strontianite; b) the development of this mineral association filling small fractures. In fact, the late stages of kimberlite emplacement are developed under the influence of CO2-rich fluids (Head and Wilson, 2008), which are also responsible for the alteration of host rocks in many kimberlite fields worldwide (Smith et al., 2004); Agee et al.
(1982) also attributed the formation of Mn-rich ilmenite in the Elliott County kimberlite, Kentucky
(USA) to Ca-enriched late fluids.
The composition of this replaced ilmenite is similar to that of the fine-grained euhedral ilmenite crystals found in the kimberlite matrix. Analogous trends have already been described in other kimberlite fields, but in the hypabyssal facies (i.e. Hunter et al., 1984). Similar textures and compositions in the groundmass are not rare in kimberlites. Tompkins and Haggerty, 1985;
Chakhmouradian and Mitchell, 1999 interpreted this type of ilmenite as produced by primary magmatic crystallization in the matrix of the kimberlite. In all these cases, however, Mn-rich ilmenite is produced in late events in the paragenetic sequence at Catoca, and in many cases contains other groundmass minerals such as perovskite and spinel (Tompkins and Haggerty, 1985). Although Mn-
33 rich ilmenite could be produced during magmatic crystallization, we contend that it could also be produced during late hydrothermal processes, during serpentinization. In fact, pyrophanite can be produced during serpentinization of ultrabasic rocks, where it appears as a late mineral in the paragenetic sequence (Mücke and Woakes, 1986; Liipo et al., 1994).
In any case, all of the ilmenite fractions in the specific Catoca kimberlite are quite different from those found in the carbonatitic xenoliths at Catoca. In the case of ilmenite from carbonatitic xenoliths, the growth of ilmenite takes place during the early stages of magmatic crystallization, and there is no evidence of replacement of a precursor ilmenite. Moreover, the crystals are distinct from the other variants of ilmenite in being extremely poor in Mg and Cr and the richest in Nb, thus defining a particular class, more similar to ilmenite found in carbonatites (Gaspar and Wyllie, 1983;
1984).
The existence of many varieties of ilmenite at Catoca has significant implications for mineral exploration. Magnesium-rich ilmenite has traditionally been interpreted as an indicator of kimberlite associations, as well as an indicator of low fO2, which is necessary for the preservation of diamond
(Garanin et al., 1997; Van Straaten et al., 2008). However, the Fe3+-rich ilmenite in the Catoca kimberlite represents more than 70% of the volume of the grains, and compositions fall into the domains of “no preservation of diamond” according to the diagram of Gurney and Zweistra (1995).
Moreover, these compositions of ilmenite are Mg- and Cr-poor, and hence using other criteria for discrimination among fertile and barren kimberlites (i.e., Haggerty, 1995). Based on this understanding the Catoca kimberlite could be expected to be barren. Although Catoca is a diamondiferous kimberlite, Mg-rich ilmenite here is clearly a product of late replacement, and the extent of replacement of the primary grains is very variable. This means, textural relations must be taken into account in the application of discriminates based on composition.
3.3 Diamond potential and regional comparison among diamondiferous and barren kimberlites
The interpretation of a maximum age for the kimberlitic eruption at 118 ± 1 Ma (PAPER V) is consistent with the idea than cretaceous kimberlites in Angola are expected to be younger than the carbonatites and alkaline rocks found in the Lucapa structure (Jelsma et al., 2012). Cretaceous
34 kimberlitic events of similar age have been reported in the São Francisco craton (Brazil), the
Kaapvaal craton (South Africa and Botswana), and the Congo-Kasai craton (the Democratic Republic of Congo), which were all part of Gondwanaland (e.g., Batumike et al., 2007; Jelsma et al., 2009).
Systems of deep faults present in these cratons probably were the focus of thermal perturbations and injection of melt.
Our interpretation of 118 ± 1 Ma for the maximum age of the kimberlitic eruption in Catoca, which is associated with a NE-SW tectonic trend (Lucapa structure), reinforces the hypothesis of
Jelsma et al. (2009) that 120 Ma (Aptian age) kimberlites are preferentially associated with NE-SW tectonic trends, whereas 85 Ma (Santonian age) kimberlites are emplaced in E-W lineaments. Our finding of an Aptian age for the maximum age of the kimberlitic eruption in Catoca is also consistent with a single model for the magmatic province, which would have extended over what is now southeastern Brazil and southwestern Africa, coincident with the opening of the South Atlantic Ocean
(Hawkesworth et al., 1992, 1999; Guiraud et al., 2010). The extensional tectonic setting, rifting, and opening of the South Atlantic during the Early Cretaceous (Pereira et al., 2003; Jelsma et al., 2009) and the reactivation of deep-seated fault systems probably contributed to lithospheric heating (mantle upwelling) and, ultimately, to kimberlitic magmatism in Angola.
The geochronological studies (PAPER V) and geochemical studies suggest that the distribution of kimberlites in Angola is strongly influenced by the tectonic setting. The presence and preservation of diamond depends on the chemical conditions (e.g., fO2, metasomatism) and the rate of ascent of the kimberlite magma which traps and transports diamond to the crust. Haggerty (1986), already proposed that intra- and inter-kimberlite diamond grades differ because of the heterogeneous distribution of potential, differences in the sources, sorting of diamonds during entrainment, flow and mixing of different batches of kimberlites, and varying degrees of resorption of diamond in the ascent magma.
To date there is no information to validate the idea of mineralogical differences between diamond-bearing and diamond-free (barren) kimberlites. The tectonic configuration sets the favorable conditions for diamond presence (kimberlites that pass through craton roots), and the detailed petrographic study of indicator minerals, e.g., ilmenite, and xenoliths provide essential information to determine the conditions for the preservation of diamond.
35
3.4 Future research
Additional work that was not able to be included in this thesis since it is still in progress.
Includes work on the trace element compositions of ilmenites from diamondiferous (CA and CU79) and a barren kimberlite (CU76). The analytical part has been completed and I am currently working on the results and discussion of these data. It is anticipated this research will be ready in January
2013.
Some fluid inclusions in ilmenite were identified during the analyses of PAPER II. Then, ten representative samples of ilmenite with fluid inclusions from the CA, TZ, CU79, and AC4 were selected and analyzed by Dr. D. Kamenetsky. These data will be used for next publications.
Unfortunately, the set of samples of fresh xenoliths arrived after the author of this thesis
(SERC) carried out the analytical phase. These materials will be used for future Ph.D. thesis.
36
CHAPTER 4 – MAIN CONCLUSIONS
The main conclusions of this thesis are:
1. The presence and distribution of the studied kimberlites in northeastern Angola is influenced by
the tectonic setting, as we have been able to determine based on regional and geochronological
studies. The diamondiferous Catoca kimberlite is tectonically related to other Early Cretaceous
kimberlites associated with NE – SW lineaments in southwestern and southern Africa.
2. The maximum age of eruption of the Catoca kimberlite as being during the Aptian provides
precise data on the age of an important diamond-bearing kimberlite pulse in northeastern Angola
and should act as an important guide for diamond exploration.
3. The age of the Catoca kimberlite is restricted to between 118± 1 Ma (the maximum age for the
kimberlite eruption in Catoca) and 112 Ma, the beginning of deposition of diamondiferous clasts
in the Calonda Formation. The eruptive event for the Catoca kimberlite appears to have taken
place in this range of ages.
4. The preservation and differences in diamond grade among kimberlites is influenced by the fO2,
mixing of kimberlite batches, rate of ascent of the magma toward the crust, and different events of
metasomatism.
5. The composition of the Catoca ilmenite is complex, and the result of multiple processes. The
ilmenite macro- and megacrysts were likely produced by disaggregation of ilmenite-bearing
xenoliths (mainly relatively oxidized and metasomatized mantle peridotites and minor
carbonatites). The subsequent reaction under disequilibrium conditions with kimberlite-derived
fluids produced the replacement of the above macro- and megacrysts by secondary Mg-rich
ilmenite.
6. Late subsolidus reactions with the fluids associated with the kimberlite, also in disequilibrium
conditions, produced the replacement of the early ilmenite types by highly reduced Mn-rich
ilmenite. The enrichment in Nb of this late ilmenite (and in the ilmenite of the matrix), as well as
37
its intimate association with carbonates of Ba and Sr, can be interpreted in terms of an interaction
of the ilmenite crystals with a CO2-rich fluid.
7. New understanding in regards to the concept of ilmenite in diamond exploration is proposed. The
absence of Mg-rich ilmenite in a kimberlite or the corresponding placers does not appear to be a
convincing argument to exclude the occurrence of economic deposits of diamond.
8. Some of the zircon crystals from the Catoca kimberlite could have been produced in U–Th-
enriched metasomatized mantle units (MARID or glimmeritic suite assemblages), while others
have chemistries suggestive of a depleted asthenosphere source.
9. The CA and CU79 diamondiferous kimberlites indicate different sources and metasomatic events,
and the diamond present in each one may be derived from different protoliths.
10. The calculated northeastern Angola paleogeotherm is consistent with a single value for the CA
and the CU79 kimberlites. The differences in T-P values between these kimberlites may reflect
the different way each kimberlite sampled the lithosphere. The lithospheric thickness calculated
from the northeastern Angola paleogeotherm yielded 192 km.
38
ACKNOWLEDGMENTS
This doctoral thesis was funded by the projects CGL2005-07885/BTE and CGL2006-12973 of
Ministerio de Educación y Ciencia (Spain), and the AGAUR SGR 589 and AGAUR SGR 444 of
Generalitat de Catalunya. I received an FI pre-doctoral grant and a BE grant both sponsored by the
Departament d'Educació i Universitats de la Generalitat de Catalunya and European Social Fund. I thank ENDIAMA and the Sociedade Mineira de Catoca, LDA, especially Dr. Vladimir Pervov
(petrologist for Catoca), Prof. M. Watagua, and all the mine geologists from Catoca, Alto Cuilo and
Muanga, who allowed to acquire samples for this study as well as facilities during the mine trip. Also thanks to the Universidade Agostinho Neto (Dr. A.O. Gonçanvels) for facilitating the trips in Angola.
I acknowledge the Geological Survey of Canada, Ottawa, especially to Dr. Simon Jackson, for all his help during my six-month research visit and during the writing phase, also thanks to Dr. Bill
Davis who helped me with the revision and interpretation of the SHRIMP data. I also acknowledge the Electron Microprobe Laboratory, Department of Earth and Planetary Sciences, McGill University, especially to Mr. Lang Shi for assistance in the use of EPMA. I also express my thanks to the Serveis
Cientificotècnics de la Universitat de Barcelona for assistance in the use of SEM/ESEM-BSE-EDS analyses (E. Prats, R. Fontarnau†, Dr. J. García Veigas), and EMP (Dr. X. Llovet). Thanks to M.
Rejas (ICTJA) for assistance in separation of some samples.
I am grateful to Prof. Joan Carles Melgarejo Draper and Prof. Salvador Galí Medina who not only directed my Ph.D. thesis but also gave me all their support as good advisors and friends. I also wish to thank Dr. Mónica Escayola (also co-director) who established all the contacts to carry out the
LA-ICP-MS, SHRIMP, and Sm/Nd analyses at the Geological Survey of Canada, Ottawa, and at the
Pacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences,
University of British Columbia, Vancouver.
I am indebted to Dr. Robert Martin, emeritus professor at the Earth & Planetary Sciences
Department, McGill University, who reviewed all my manuscripts, arranged for me to acquire the
39
EPMA analyses at the McGill University, helped me and advised me through the whole thesis, and also corrected parts of this Ph.D. thesis. I acknowledge with gratitude the help and advice I received from Vicki Loschiavo. The interesting discussions I had with professors who are kimberlite experts during international conferences are greatly appreciated: they contributed to the development of this thesis. I also value the guidance I received from Prof. Joaquin Proenza during challenging moments of this thesis.
I want to express my enormous gratitude to my partner Rainer and our baby to be born, who are my inspiration for this thesis; also to my mother Perla, brother Wilson, and cousins Lupe and Charli, and to my friends: Rafael David, Hildebrando, Leonardo, Fernando, Andrea, Sebastien, Ignacio,
Jaume, Amaia, and Eder, who helped me a lot in different ways and at different stages during the thesis. Also thanks to the Recursos Minerals research group and to the Cristal·lografia, Mineralogia i
Dipòsits Minerals Department for all their help during these years. Finally, I dedicate this thesis to the memory of my grandparents, who always gave me energy, motivation, and courage to accomplish different goals.
40
REFERENCES
Agee, J.J., Garrison Jr., J.R., Taylor, L.A., 1982. Petrogenesis of oxide minerals in kimberlite, Elliott
County, Kentucky (USA). Am. Mineral. 67 (1–2), 28–42.
Batumike, J.M., O’Reilly, S.Y., Griffin, W.L., Belousova, E.A., 2007. U–Pb and Hf-isotope analyses
of zircon from the Kundelungu Kimberlites, D.R. Congo: Implications for crustal evolution.
Precambrian Res.156 (3-4), 195-225.
Bermúdez-Lugo, O., 2004. The Mineral Industry of Angola. In U.S.Geological Survey Minerals
Yearbook-2004. U.S.G.S.
Blore, S., 2004. The key to kimberlite internal diamond controls, seven case studies. Partnership
Africa Canada and Global Witness Publishing Inc., 1-20.
Blore, S., 2007. Diamond Industry Annual Review. Republic of Angola. Partnership Africa Canada,
1-24.
Chakhmouradian, A.R., Mitchell, R.H., 1999. Niobian ilmenite, hydroxylapatite and sulfatian
monazite: alternative hosts for incompatible elements in calcite kimberlite from
Internatsional'naya, Yakutia. Can. Mineral. 37 (5), 1177–1189.
Clement, C.R., 1982. A comparative geological study of some major kimberlite pipes in the Northern
Cape and Orange Free State. Thesis (Ph. D.), University of Cape Town.
Clement, C.R., Skinner, E.M.W., 1985. A textural-genetic classification of kimberlites. Trans. Geol.
Soc. S. Afr. 88, 403-409.
Dawson, J. B., 1980. Kimberlites and their Xenoliths. Anon. Germany: Springer-Verlag, pp. 252.
41
De Andrade, C.F., 1954. On the discovery of a kimberlite type of igneous rock in the diamondiferous
fields of Luanda. In Association des Services Géologiques Africains. Deuxième Partie. Questions
Diverses et Annexes. Fascicule XXI. Alger.
De Araujo, A.G., Perevalov, O.V., 1998. Carta de recursos minerais - Mineral Resources Map.
Ministério Geol. e Minas. Inst. Geol. Angola, Luanda.
De Araujo, A.G., Perevalov, O.V., Jukov R.A., 1988. Carta geológica de Angola - Geological Map of
Angola. Scale 1 : 1 000 000. Ministèrio da Indústria. Inst. Nac. Geol., Luanda.
De Carvalho, H., Tassinari, C., Alves, P.H., 2000. Geochronological review of the Precambrian in
western Angola: links with Brazil. J. Afr. Earth Sci. 31 (2), 383-402.
Egorov, K.N., Roman’ko, E.F. Podvysotsky, V.T., Sablukov, S.M, Garanin, V.K., D’yakonov, D.B.,
2007. New data on kimberlite magmatism in southwestern Angola. Russian Geol. Geophys. 48 (4),
323-336.
Ekstrand, O.R., Sinclair, W.D., and Thorpe, R.I., 1995. Geology of Canadian mineral deposit types:
Geological Survey of Canada, Geology of Canada, no. 8, and Geological Society of America's
Geology of North America, vol. P-1, 640 p.
ENDIAMA, 2012. Empresa Nacional de Diamantes de Angola, E.P. Accessed on June.
http://www.endiama.co.ao/endiama_operacoes_grupo.php.
Ganga, J., Rotman, A.Y., Nosiko, S., 2003. Pipe Catoca, an example of the weakly eroded kimberlites
from North-East of Angola. The 8th International Kimberlite Conference, Extended Abstract.
Garanin, V.K., Kudryavtseva, G.P., Posukhova, T.V., 1997. Indicators of diamond preservation in
kimberlite. In: Papunen, H. (Ed.), Mineral Deposits: Research and Exploration, Where Do They
Meet? In: Proceedings of the 4th Biennial SGA Meeting. Turku, Finland, 767–770.
Gaspar, J.C., Wyllie, P.J., 1983. Ilmenite (high magnesium, manganese, niobium) in the carbonatites
from the Jacupiranga Complex, Brazil. Am. Mineral. 68 (9–10), 960–971.
42
Gaspar, J.C., Wyllie, P.J., 1984. The alleged kimberlite-carbonatite relationship: evidence from
ilmenite and spinel from Premier and Wesselton Mines and the Benfontein Sill, South Africa.
Contrib. Mineral. Petrol. 85, 133–140.
Grütter, H.S., Gurney, J.J., Menzies, A.H., Winter, F., 2004. An updated classification scheme for
mantle-derived garnet, for use by diamond explorers. Lithos 77, 841–857.
Grütter, H.S., Latti, D., Menzies, A., 2006. Cr-saturation arrays in concentrate garnet compositions
from 1 kimberlite and their use in mantle barometry. J. Petrol. 47, 801–820.
Guiraud, R., Bosworth, W. Thierry, J. Delplanque A., 2005. Phanerozoic geological evolution of
Northern and Central Africa: an overview. J. Afr. Earth Sci. 43, 83–143.
Guiraud, M., Buta-Neto, A., Quesne, D., 2010. Segmentation and differential post-rift uplift at the
Angola margin as recorded by the transform-rifted Benguela and oblique-to-orthogonal-rifted
Kwanza basins. Mar. Pet. Geol. 27, 1040–1068.
Gurney, J.J., Zweistra, P., 1995. The interpretation of the major element compositions of mantle
minerals in diamond exploration. J. Geochem. Explor. 53 (1–3), 293–309.
Haggerty, S.E., 1986. Diamond genesis in a multiply-constrained model. Nature 320, 34 – 38.
Haggerty, S.E., 1995. Upper mantle mineralogy. J. Geodyn. 20 (4), 331–364.
Haggerty, S.E., Hardie III, R.B., McMahon, B.M., 1977. The composition of ilmenite nodule
associations from the Monastery diatreme: Proceedings of the 2nd International Kimberlite
Conference, vol. 2, pp. 249–256.
Haggerty, S.E., Tompkins, L.A., 1983. Redox state of Earth's upper mantle from kimberlitic
ilmenites. Nature 303 (5915), 295–300.
Hawkesworth, C.J., Gallagher, K., Kelley, S., Mantovani, M., Peate, D.W., Regelous, M., Rogers
N.W., 1992. Parana magmatism and the opening of the South Atlantic, in: Storey, B.C., Alabaster,
43
T., Pankhurst, R.J (Eds.), Magmatism and the causes of continental break-up. Geol. Soc. Spec.
Publ. No. 68, London, 221-240.
Hawkesworth, C.J., Kelley, S., Turner, S., Le Roex, A., Storey, B., 1999. Mantle processes during
Gondwana break-up and dispersal. J. Afr. Earth Sci. 28 (1), 239-261.
Hunter, R.H., Kissling, R.D., Taylor, L.A., 1984. Mid- to late-stage kimberlitic melt evolution:
phlogopites and oxides from the Fayette County kimberlite, Pennsylvania. Am. Mineral. 69, 30–
40.
Head, J.W., Wilson, L., 2008. Integrated model for kimberlite ascent and eruption. 9th International
Kimberlite Conference Extended abstract No. 9IKC-A-00284.
Heaman, L.M., Kjarsgaard, B.A., Creaser, R.A., 2003. The timing of kimberlite magmatism in North
America: implications for global kimberlite genesis and diamond exploration. Lithos 71, 153– 184.
Janse, A.J.A., 2007. Global rough diamond production since 1870. Gems and Gemology,
Gemological Institute of America, vol. 43, No. 2, 98–119.
Janse, A.J.A., Sheahan, P.A., 1995. Catalogue of worldwide diamond and kimberlite occurrences: a
selective and annotative approach. J. Geochem. Explor. 53, 73-111.
Jelsma, H.A., Barnett, W., Richards, S., Lister, G., 2009. Tectonic setting of kimberlites. Lithos 112S,
55-165.
Jelsma, H.A., de Wit, M.J., Thiart, C., Dirks, P.H.G.M., Viola, G., Basson, I.J., Anckar, E., 2004.
Preferential distribution along transcontinental corridors of kimberlites and related rocks of
Southern Africa. S. Afr. J. Geol. 107, 301-324.
Jelsma, H.A., Krishnan, U., Perrit, S., Preston, R., Winter, F., Lemotlo, L., v/d Linde, G., Armstrong,
R., Phillips, D., Joy, S., Costa, J., Facatino, M., Posser, A., Kumar, M., Wallace, C., Chinn, I.,
Henning, A., 2012. Kimberlites from Central Angola: a case study of exploration findings.
Extended Abstr. No. 10IKC-42, 10th International Kimberlite Conference, Bangalore, 1–5.
44
Kamenetsky, M.B., Sobolev, A.V., Kamenetsky, V.S., Maas, R., Danyushevsky, L.V., Thomas, R.,
Pokhilenko, N.P., Sobolev, N.V., 2004. Kimberlite melts rich in alkali chlorides and carbonates: a
potent metasomatic agent in the mantle. Geology 32, 845-848.
Khar'kiv, A., Levin, V., Mankenda, A., Safronov, A., 1992. The Camafuca-Camazambo kimberlite
pipe of Angola, the largest in the world. Int. Geol. Rev. 34 (7), 710-719.
Kotel'nikov, D.D., Zinchuk, N.N., Zhukhlistov, A.P., 2005. Stages of serpentine and phlogopite
transformation in the Catoca kimberlite pipe, Angola. Dokl. Earth Sci. 403 (6), 866–869.
Lewis, H.C., 1987. On diamandiferous peridotite and the genesis of the diamond. Geol. Mag. 4, 22-
24.
Lewis, H.C., 1988. The matrix of diamond. Geol. Mag. 5, 129-131.
Liipo, J.P., Vuollo, J.E., Nykänen, V.M., 1994. Pyrophanite and ilmenite in serpentinized wehrlite
from Ensilä, Kuhmo greenstone belt, Finland. Eur. J. Mineral. 6, 145–150.
Llusià Queral, R.M., Pereira Ginga, S., Galiano, A., Sebastião, P., Melgarejo Draper, J.C., Olimpio
Gonçalves, A., Proenza Fernandez, J., Morais, E., 2005. Aluviões asociados ao campo kimberlítico
da zona de Muanga (Lunda Norte, Angola). XIV semana Geoquímica/VIII Congresso de
Geoquimica dos Paises de Lingua Portuguesa, 351-354.
Mather, K.A., Pearson, D.G., McKenzie, D., Kjarsgaard, B., Priestley, K., 2011. Constraining the
depth and thermal history of cratonic lithosphere using peridotite xenolith and xenocryst
thermobarometry and seismology. Lithos 125, 729–742.
McLean, H., Banas, A., Creighton, S., Whiteford, S., Luth, R.W., Stachel, T., 2007. Garnet xenocrysts
from the Diavik mine, NWT, Canada: Composition, color, and paragenesis. Can. Mineral. 45,
1131–1145.
Mitchell, R.H., 1986. Kimberlites: Mineralogy, Geochemistry, and Petrology. Plenum Press, New
York, pp. 442.
45
Mitchell, R.H., 1995. Kimberlites, Orangeites, and Related Rocks. Plenum Press, New York, pp. 410.
Mücke, A.,Woakes, M., 1986. Pyrophanite: a typical mineral in the Pan-African Province of Western
and Central Nigeria. J. Afr. Earth Sci. 5 (6), 675–689.
Nimis, P., Taylor, W. R., 2000. Single clinopyroxene thermobarometry for garnet peridotites. Part I.
Calibration and testing of a Cr-in-Cpx barometer and an enstatite-in-Cpx thermometer. Contrib.
Mineral. Petrol. 139, 514–554.
Pasteris, J., 1980. The significance of groundmass ilmenite and megacryst ilmenite in kimberlites.
Contrib. Mineral. Petrol. 75 (4), 315–325.
Pearson, D.G., Canil, D., & Shirey, S.B., 2007. Mantle samples included in volcanic rocks: xenoliths
and diamonds. In Carlson, R.W. (ed.). The mantle and core. Treatise on Geochemistry, vol. 2,
Washington, Elsevier, 171-275.
Pearson, D.G., Wittig, N., 2008. Formation of archean continental lithosphere and its diamonds: the
root 15 of the problem. J. Geol. Soc. (London, U. K.)165, 1–20.
Pell, J., 1998. Kimberlite-hosted Diamonds. In Geological Fieldwork (1997), British Columbia
Ministry of Employment and Investment, Paper 1998-1, pages 24L-1 to 24L-4.
Pereira, E., Rodrigues, J., Reis B., 2003. Synopsis of Lunda geology, NE Angola: implications for
diamond exploration. Comun. Inst. Geol. Min. 90, 189-212.
Perevalov, O.V., Voinovsky, A.S., Tselikovsky, A.F., Agueev, Y.L., Polskoi, F.R., Khódirev, V.L.,
Kondrátiev, A.I., 1992. Geologia de Angola. Notícia explicativa da carta geológica a escala
1:1.000.000. Serviço Geológico de Angola. Lunda.
Preistley, K., McKenzie, D., 2006. The thermal structure of the lithosphere from shear wave
velocities. Earth Planet. Sci. Lett. 244, 285–301.
Read, G.H., Janse, A.J.A., 2009. Diamonds: Exploration, mines and marketing. Lithos 112S, 1–9.
46
Reis, B., 1972. Preliminary note on the distribution and tectonic control of kimberlites in Angola.
Abstr., 24th Int. Geol. Congr., Rep. Sess. 4, 276-281, Montreal.
Robles-Cruz, S.E., 2007. Textura y química de los minerales del grupo de la ilmenita de la kimberlita
de Catoca (Angola). Treball de Recerca Tutelat (TRT). Diploma d'Estudis Avanzats (DEA).
Universitat de Barcelona. Unpublished.
Schulze, D.J., 1984. Cr-poor megacrysts from the Hamilton Branch kimberlite, Elliott County,
Kentucky. In Kornprobst, J. (Ed.), Kimberlites II: TheMantle and Crust–Mantle Relationships. In:
Developments in Petrology 11B. Amsterdam, Elsevier, 97–108.
Schulze, D.J., 1995. Low-Ca garnet harzburgites from Kimberley, South Africa: Abundance and
bearing on the structure and evolution of the lithosphere. J. Geophys. Res. (Solid Earth), vol. 100,
No. B7, 12,513-12,526.
Scott Smith, B. H.; Nowicki, T. E.; Russell, J. K.; Webb, K. J.; Mitchell, R. H.; Hetman, C. M.;
Robey, J. V. A.; Skinner, E. M. W.; Robey, J. V., 2012. Kimberlite terminology and classification.
In Proceedings of the 10th Kimberlite Conf., Bangalore, India, 6-11 February, Abstract Volume,
30–31.
Skinner, E.M., 1986. Constrasting Group 1 and 2 kimberlite petrology: Towards a genetic model for
kimberlites. Fourth Int. Kimberlite Conf., Perth, Australia. Extended Abstracts, 202-204.
Skinner, E.M., 1989. Contrasting group I and II kimberlite petrology: Towards a genetic model for
kimberlites. In Ross et al. (1989) q.v., vol. 1, pp. 528-544.
Smith, C.B., 1983. Pb, Sr, and Nd isotopic evidence for sources of African Cretaceous kimberlites.
Nature 304, 51-54.
Smith, C.B., Gurney, J.J., Skinner, E.M.W., Clement, C.R., Ebrahim, N., 1985. Geochemical
character of southern African kimberlites: a new approach based on isotopic constraints. Trans.
Geol. Soc. S. Afr. 88, 267-280.
47
Smith, C.B., Sims, K., Chimuka, L., Duffin, A., Beard, A.D., Townend, R., 2004. Kimberlite
metasomatism at Murowa and Sese pipes, Zimbabwe. Lithos 76 (1–4), 219–232.
Stachel, T., Harris, J.W., 2009. Formation of diamond in the Earth’s mantle. J. Phys.: Condens.
Matter 21, 364206, 1-10.
Tompkins, L.A., Haggerty, S.E., 1985. Groundmass oxide minerals in the Koidu kimberlite dikes,
Sierra Leone, West Africa. Contrib. Mineral. Petrol. 91 (3), 245–263.
Thomson Reuters, 2012. Journal Citation Reports. Available online: http://admin-
apps.webofknowledge.com/JCR/JCR (accessed on 24 September).
Van Straaten, B.I., Kopylova, M.G., Russell, J.K., Webb, K.J., Scott Smith, B.H., 2008.
Discrimination of diamond resource and non-resource domains in the Victor North pyroclastic
kimberlite, Canada. J. Volcanol. Geotherm. Res. 17, 128–138.
White, S.H., de Boorder, H., Smith, C.B., 1995. Structural controls of kimberlite and lamproite
emplacement. J. Geochem. Explor. 53 (1-3), 245-264.
48
ORIGINAL PUBLICATIONS
PAPER I
Reprinted from MACLA - Revista de la Sociedad Española de Mineralogía, September No. 9. Robles- Cruz, S., Watangua, M., Melgarejo, J.C., Galí, S., 2008. New Insights into the Concept of Ilmenite as an Indicator for Diamond Exploration, Based on Kimberlite Petrographic Analysis.
macla nº 9. septiembre ‘08 revista de la sociedad española de mineralogía 205
New Insights into the Concept of Ilmenite as an Indicator for Diamond Exploration, Based on Kimberlite Petrographic Analysis / SANDRA ROBLES CRUZ (1*), MANUEL WATANGUA (2), JOAN CARLES MELGAREJO (1), SALVADOR GALI (1)
(1) Departament de Cristal·lografia, Mineralogia i Dipòsits Minerals. Facultat de Geologia. Universitat de Barcelona. Martí i Franquès s/n. 08028, Barcelona (España) (2) ENDIAMA, Major Kanhangulo 100, Luanda (Angola)
INTRODUCTION. diamond potential is currently being studied. The Catoca kimberlite is the This study presents results of the initial most important primary diamond phase of the research project, deposit in Angola, hosted by “Kimberlites associated to the Lucapa Precambrian rocks and covered by structure, Angola (Africa)”, within the Mesozoic-Cenozoic sedimentary deposits framework of a multilateral agreement (Janse et al, 1995). between the Faculty of Geology- Universitat de Barcelona, the Empresa PETROGRAPHY. Nacional de Diamantes de Angola and the Agostinho Neto University (Luanda- There are some minerals which are Angola). frequently associated to diamond inside kimberlites and they are used as fig 1. Intercumular ilmenite in peridotitic xenolith. The research is based on two sets of indicator minerals for the diamond core sampling down to 600 m deep. The exploration. The main indicator minerals first set comes from Catoca pipe and are: magnesian ilmenite (Pell, 1998), allowed us to identify complete crater garnet and chromite (Wyatt et al, 2004). and diatreme facies. The second one (18 However, for this instance we will focus kimberlites) comes from Cucumbi, on ilmenite since it is the first mineral Cacuilo, Tchiuzo, Alto Cuilo, Camitongo analyzed in 2007. and Kambundu, whose samples were gathered during fall 2008. Currently, we Diverse xenoliths, comprising lherzolite, are working on these sets of samples. eclogite, harzburgite, carbonatite, gneiss and amphibolite are distributed through The kimberlites are ultrabasic rocks with the Catoca and Cucumbi kimberlites. high content of volatiles mainly CO2, and Some shales and sandstones can be a typical inequigranular texture present in the upper part of this fig 2. Nodular xenocrysts of ilmenite with characterized by the presence of macro- Kimberlite. Accessory minerals and replacement. megacrysts which can be xenoliths or xenocrysts comprise garnet, zircon, Cr- xenocrysts embedded in a fine-grained rich diopside, amphibole, phlogopite, matrix (Mitchell, 1995; Benvie, 2007). chromite and several generations of These special rocks have a great ilmenite. Secondary minerals include importance, not only in scientific terms serpentine-group minerals being the since they add valuable information most abundant, calcite, barite, about lithospheric mantle but also barytocalcite, witherite and strontianite. because they can contain diamond. Based on optical petrographic studies REGIONAL SETTING. and BSE images from SEM-ESEM with EDS microanalysis, we have been able
The area of interest is localized in to discriminate up to six textural types of fig 3. Euhedral crystals of ilmenite in matrix. northeastern Angola (Africa), being ilmenite in Catoca and Cucumbi tectonically controlled by the Lucapa kimberlite: a) intercumular ilmenite in Zircon xenocrysts are partially replaced structure, a former rift (Guiraud et al., peridotitic xenoliths (Fig 1); b) anhedral by fine-grained baddeleyite, and at least 2005) of early Cretaceous that extends ilmenite in carbonatite xenoliths; c) two populations exist according to the NE-SW across Angola. Associated to this ilmenite unaltered megacrysts; d) trace element distribution. All of these structure there is a magmatic belt, nodular xenocryst of ilmenite with crystals are enriched in HREE, but with a which is composed by kimberlites different grades of replacement, some noticeable positive Ce anomaly, similar toward NE and carbonatites toward SW. of them with symplectitic textures (Fig. to that reported in zircon in a MARID At present, over 2000 kimberlites have 2); e) skeletal ilmenite; and f) euhedral xenolith from a southern African been identified in this structure and their crystals of ilmenite in matrix (Fig. 3). kimberlite (Dawson et al., 2001). The
palabras clave: kimberlita, diamante, ilmenita, manto, Angola. key words: kimberlite, diamond, ilmenite, mantle, Angola.
resumen SEM/SEA 2008 * corresponding author: [email protected] 206
crystals are not optically zoned, but environment. Magnesian ilmenite is also ilmenite in a higher or lower grade. there is a slight depletion in REE from enriched in Cr and Ni. More advanced These fluids are reducing, especially the core to the rim. replacement produces a symplectitic those rich in Mn. Picroilmenite has replacement of ilmenite II by ilmenite III. traditionally been interpreted as an MINERAL CHEMISTRY OF ILMENITE. indicator of kimberlite associations, as well as an indicator of low fO2, which is Every texture has been systematically necessary for the preservation of analyzed with EPMA which allowed us to diamond. Although Catoca and Cucumbi identify three compositional types of are diamondiferous kimberlites, they ilmenite (I, II and III). This combined show that Mg ilmenite is clearly a late technique –texture and composition replacement product, and the grade of analysis- has been suitable for analyzing replacement of the primary grains is zircon and garnet as well. very variable. Therefore the absence of magnesian ilmenite in a kimberlite does The primary ilmenite (type I) in not appear to be a convincing argument megacrysts (xenocrysts) is generally rich to exclude the presence of diamonds. in Cr and Fe3+. Its composition is similar Accordingly, this work proposes a new to the intercumulus crystals in peridotite insight into the concept of ilmenite in xenoliths. This ilmenite is replaced, in diamond exploration. the first instance, by magnesian ilmenite (type II). This process takes place along ACKNOWLEDGMENTS. microdiscontinuities (cleavage, border grains, contour subgranes, kink band This research is supported by the project planes, etc.), producing diffusive CGL2006-12973 of Ministerio de replacements. In a more advanced Educación y Ciencia (Spain), the AGAUR stage, symplectitic replacements occur, fig 4. MgTiO3-FeTiO3-Fe2O3 ratio for the different SGR 589 of Generalitat de Catalunya involving an early generation of types of ilmenite (I, II and III) discriminated for each and a FI grant sponsored by the texture. magnesian ilmenite (type II) at the Departament d’Educació i Universitats expense of Fe3+-rich primary ilmenite A late generation of ilmenite III (Mn-rich de la Generalitat de Catalunya i del Fons (from texture a to d ). A late generation ilmenite) is found rimming all the above Social Europeu. The authors of Mn-Nb-Zr-rich ilmenite (type III) cuts mentioned generations, and is strongly acknowledge the Serveis the previous ones. enriched in Nb, Ta, Zr, W, Hf, Th and U, Cientificotècnics de la Universitat de and poor in Mg and Fe3+. The Barcelona. Contrastingly, the late euhedral Mn-Nb- composition of this ilmenite is similar to Zr ilmenite crystals found in the that of the fine-grained euhedral REFERENCES. kimberlite matrix do not present any ilmenite crystals found in the kimberlitic evidence of replacement. This ilmenite matrix and also to that of the ilmenite Benvie, B.(2007): Mineralogical imaging of is poor in Mg and Fe3+. Their kimberlites using SEM-based techniques. crystals found in the carbonatitic Minerals Engineering, 20, 435-443. compositions are identical with the Mn- xenoliths. Crystals of Mn-rich ilmenite rich ilmenite produced during late Dawson, J.B., Hill, P.G., Kinny, P.D. (2001): (ilmenite type III) are not replaced or Mineral chemistry of a zircon-bearing, replacement stages of ilmenite zoned, and seem to have crystallized in composite, veined and metasomatised megacrysts. Compositions of Mn-rich equilibrium with the kimberlitic magma. upper-mantle peridotite xenolith from ilmenite are similar to those found in Both the late generations of ilmenite kimberlite. Contrib. Mineral Petrol., 140, carbonatite xenoliths. and the baddeleyite replacing zircon can 720-733. be produced by interaction of a Guiraud, R., Bosworth, W., Thierry, J., DISCUSSION AND CONCLUSIONS. Delplanque, A. (2005): Phanerozoic carbonate-bearing kimberlitic magma geological evolution of Northern and enriched in Mn and HFSE. The Textural evidences indicate a different Central Africa: An overview. Journal of replacement of Fe3+-rich ilmenite by Mg- African Earth Sciences, 43, 83-143. complex history of growth in the and Mn-rich ilmenite implies that the Janse, A.J.A. & Sheahan, P.A. (1995): xenocrysts. Unaltered megacryst early ilmenite was formed under Catalogue of world wide diamond and ilmenite (ilmenite type I) rich in Fe3+ (fig. oxidizing conditions in the mantle, and kimberlite occurrences: a selective and 4), indicates crystallization under high the lastest compositions of ilmenite annotative approach. Journal of fO2 conditions; this ilmenite contains Nb, were produced by reaction with the Geochemical Exploration, 53, 73-111. Cr, Ni and Ta in low contents. Its Mitchell, R.H. (1995): Kimberlites, Orangeites, kimberlitic magma. and related rocks. New York, Plenum composition is similar to those ilmenite Intercumular megacrysts that occur in Press, 410 pp. Megacrysts of ilmenite are frequently Pell, J. (1998): Kimberlite-hosted Diamonds, peridotite xenoliths. Hence, most of the present in diamondiferous kimberlites, in Geological Fieldwork 1997. British ilmenite xenocrysts seem to have been contrasting with ilmenite observed in Columbia Ministry of Employment and produced by disaggregation of mantle barren kimberlites. This might become a Investment 1998-1, 24L1-24L4. xenoliths. Ilmenite I is replaced along new guide in diamond exploration. Wyatt, B.A., Mike, B., Anckar, E., Grutter, H. discontinuities by magnesian ilmenite In conclusion, the composition of this (2004): Compositional classification of “kimberlitic” and “non-kimberlitic” (ilmenite type II); the elemental ilmenite is the result of a set of ilmenite. Lithos, 77, 819-840. distribution of Mg in these grains points replacement processes with rich fluids in to processes of replacement through Mg and Mn affecting an oxidized primary solid-state diffusion in a typical reducing ORIGINAL PUBLICATIONS
PAPER II
Reprinted from Lithos, 112S. Robles-Cruz, S.E., Watangua, M., Melgarejo, J.C., Gali, S., Olimpio, A., 2009. Contrasting compositions and textures of ilmenite in the Catoca kimberlite, Angola, and implications in exploration for diamond.
Lithos 112S (2009) 966–975
Contents lists available at ScienceDirect
Lithos
journal homepage: www.elsevier.com/locate/lithos
Contrasting compositions and textures of ilmenite in the Catoca kimberlite, Angola, and implications in exploration for diamond
Sandra E. Robles-Cruz a,⁎, Manuel Watangua b, Leonardo Isidoro b, Joan C. Melgarejo a, Salvador Galí a, Antonio Olimpio c a Departament de Cristal·lografia, Mineralogia i Dipòsits Minerals, Facultat de Geologia, Universitat de Barcelona, Martí i Franquès, s/n, E-08028, Barcelona, Spain b ENDIAMA, Major Kanhangulo, 100, Luanda, Angola c Departamento de Geologia, Faculdade de Ciências, Universidade Agostinho Neto, Av. 4 de Fevereiro 7, 815, Luanda, Angola article info abstract
Article history: The Catoca group-I kimberlite, the only currently active diamond-producing mine in Angola, was emplaced in Received 2 October 2008 the northeastern part of the Lucapa structure. We focus here on compositional and textural variations in Accepted 16 May 2009 ilmenite from drill-core material, in the hope of elucidating events before and during the emplacement of the Available online 25 June 2009 kimberlitic magma. We have characterized four main variants of ilmenite, with enrichments in Fe3+, Mg, Mn and nearly stoichiometric ilmenite, and in seven textural classes, and have distinguished crystals of variable Keywords: Ilmenite size, ranging from micro- to megacrysts. Most ilmenite is found to derive, through a complex process, from 3+ Kimberlite replacement of Fe -rich ilmenite, presumably originating by mantle metasomatism at a relatively high fO2. 3+ Diamond This Fe -rich ilmenite reacted with fluids under reducing conditions, producing Mg-rich ilmenite. The Mn- Fluid rich ilmenite is produced by interaction with a late CO2-rich fluid. The Mg-rich ilmenite is here clearly a Texture minor phase and a late product of replacement. The absence of fresh Mg-rich ilmenite and the occurrence of Composition Fe3+-rich ilmenite do not seem to be convincing arguments to exclude the presence of diamond crystals in a kimberlite. Compositional attributes must thus be considered with caution, and only in light of textural studies, in exploration programs. © 2009 Elsevier B.V. All rights reserved.
1. Introduction expected to preserve diamond, whereas the presence of Fe3+-rich ilmenite could indicate oxidizing processes that could destroy crystals of Angola has become an important producer of diamond (Janse and diamond (Gurney et al., 1993; Gurney and Zweistra, 1995; Kostrovitsky Sheahan, 1995; Read and Janse, this issue), with a significant part of the et al., 2004, 2006; van Straaten et al., 2008). Other compositional production being obtained from the Catoca kimberlite, in Lunda Sul features of ilmenite are of more controversial origin. In particular, Mn- province, in northeastern Angola. Catoca, currently the unique active rich compositions, found in previous surveys of some Angolan kimberlite mine in Angola, is located in the Lucapa structure, a system of kimberlites (Llusià Queral et al., 2005a,b; Rogers and Grütter, this Cretaceous extensional faults trending NE–SW (Reis, 1972; De Carvalho issue) have been attributed to supergene processes; others propose et al., 2000; Guiraud et al., 2005). magmatic crystallization under reducing conditions (Hwang et al.,1994) For many years, the composition of ilmenite has been stressed as an or metasomatic processes in the mantle (Meyer and McCallum,1986). In exploration guide for diamondiferous kimberlites and placers (e.g., this investigation, we seek explanations of the real significance of Mitchell, 1989, 1995, 1997; Wyatt et al., 2004). It has been correlated as compositional and textural variations of ilmenite in kimberlites. We well with conditions in the mantle where the kimberlitic magmas used back-scattered electron (BSE) petrography with microanalysis originated (Haggerty and Tompkins,1983; Arculus et al.,1984; Haggerty, using energy-dispersion spectroscopy (EDS), quantitative powder X-ray 1991a,b; Gurney et al., 1993). Griffin and Ryan (1995) have suggested diffraction (PXRD), and quantitative chemical analyses using an that with some patterns of compositional evolution in ilmenite electron-microprobe (EMP) on a suite of 81 representative thin sections megacrysts, once can assess the occurrence of mechanisms of fractional and 19 probes from two core samples of Catoca pipe. With these new crystallization of single batches of magma associated with extensive datasets, we are able to shed new light on the origin of these unusual metasomatic alteration of the wallrocks, and hence destruction of compositions. diamond. Thus, reduced kimberlites bearing Mg-rich ilmenite would be 2. The Catoca kimberlite
2 ⁎ Corresponding author. Tel.: +34 9340 21344; fax: +34 9340 21340. The Catoca pipe outcrop, 639000 m , is found 30 km NNW of E-mail address: [email protected] (S.E. Robles-Cruz). Saurimo, the capital of Lunda Sul. Catoca can be classified as a group-I
0024-4937/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.lithos.2009.05.040 S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975 967
Fig. 1. Cross section of the Catoca kimberlite (adapted from Kriuchkov et al., 2000).
kimberlite (Mitchell, 1995). Complete crater and diatreme facies are microprobe analysis, allowed us to discriminate four main variants recognized (Ganga et al., 2003; Fig. 1), according to the classification of ilmenite based on compositional attributes: a) Fe3+-rich ilmenite, criteria of Clement and Skinner (1985),asmodified by Scott Smith et al. b) Mg-rich ilmenite, c) Mn-rich ilmenite, and d) near-ideal ilmenite 2+ 4 (2008), thus indicating the minimal extent of erosion of the kimberlite. (Fe Ti O3). These types can be easily distinguished using BSE images, Crater facies are found up to 230–270 m in depth, and are composed as the Mg-rich ilmenite displays the darkest shades, and the Mn-rich in the uppermost part by epiclastic sandstones with cross-stratification ilmenite is the lightest. On the other hand, up to seven textural classes of in the central part, and coarse debris rimming the crater. Altered crystals ilmenite have been established, based on their paragenetic position and of garnet and diopside may occur as accessory minerals in these degree of replacement: 1) intercumulus Fe3+-rich ilmenite grains in sediments. Ilmenite is rare in this unit. Most of the cement is ferruginous, metasomatized peridotite xenoliths, 2) anhedral ilmenite in carbonatite but in some areas, the sandstone has a calcareous-ferruginous cement. xenoliths; 3) homogeneous Fe3+-rich ilmenite present as macro- and In the lower part of the sequence, the content of volcaniclastic material megacrysts; 4) partially replaced ilmenite macro- and megacrysts; 5) increases, and the lower half of the crater facies become dominated by symplectitic ilmenite xenocrysts; 6) skeletal ilmenite crystals in a resedimented volcaniclastic kimberlite facies (unit RVK). pelletal matrix; 7) tabular Mn-rich ilmenite crystals in a kimberlite Below the crater facies, Ganga et al. (2003) described classic matrix. The suite of ilmenite that we examined contains grains of tuffisitic kimberlite facies (TK; Mitchell et al., this issue); in the variable size: microcrystals have 5–20 μm in diameter, being some as current terminology, this category could be named massive volcani- large as 50 μm; most macrocrysts have dimensions between 1 and clastic kimberlite (Sparks et al., 2006) or Wesselton-type volcaniclas- 10 mm; megacrysts are very rare and may exceed 2 cm. tic kimberlite (Scott Smith et al., 2008). Ganga et al. (2003) also The distribution of these ilmenite textural types is not homogeneous divided this unit into different subunits based on the size of the in the crater and diatreme kimberlite facies. Homogeneous ilmenite fragments, and pointed out the occurrence of abundant xenoliths macro- and megacrysts are found in all the kimberlite facies, but the derived from the host rocks, and the scarcity of xenoliths from the remainder of the textural variants are restricted to the diatreme facies, mantle. The extensive drilling allowed us to sample these units down mainly in the volcaniclastic kimberlite (below 250 m in depth). Partly to 609 m in depth. replaced megacrysts occur in the uppermost part of the diatreme facies, The diatreme facies are strongly altered all along the profile; olivine and strongly corroded macro- and megacrysts (in particular, those with is completely replaced by serpentine, calcite and saponite (Kotel'nikov a symplectitic texture) are only found below 350 m in depth. et al., 2005). Xenoliths of the host rocks are common, and comprise gneiss, amphibolite, granite, sandstone and shale. Mantle xenoliths are 3.1. Intercumulus grains of Fe3+-rich ilmenite in metasomatized also quite altered and include mainly metasomatic peridotite, and rarely peridotitic xenoliths eclogite as well as xenoliths of carbonatite. Xenocrysts encountered in the diatreme facies comprise G9 and G10 varieties of garnet according to The intercumulus grains of Fe3+-rich ilmenite are anhedral, 300– the classification of Grütter et al. (2004),zircon(partlyreplacedby 800 μm in diameter, and interstitially distributed among roundish baddeleyite), chromian diopside, amphibole, phlogopite and ilmenite. grains of olivine (Fig. 2A). The olivine is completely replaced by The matrix of the kimberlite contains lizardite, apatite, calcite, ilmenite serpentine, but the mesh texture typical of serpentinized olivine is and chromite; titanite, zirconolite, baddeleyite, barite, dolomite, with- clearly recognizable. The ilmenite is polycrystalline, and grains show erite, barytocalcite, strontianite, sulfides, identified by PXRD and EMP, polygonally annealed textures with curved borders and triple points. and minor minerals are widespread in the matrix and fill small veinlets. Similar polygonal textures in ilmenite in kimberlitic suites have been interpreted as formed by intense annealing of stressed ilmenite 3. Petrography of ilmenite (Mitchell, 1973; Haggerty et al., 1977; Tompkins and Haggerty, 1985). The intercumulus ilmenite is quite homogeneous in composition and Optical petrographic studies and back-scattered electron (BSE) rich in Fe3+ (as inferred from stoichiometry: see below), although it images taken with a SEM-ESEM with EDS microanalysis, coupled with may be partly replaced by Mg-rich ilmenite. 968 S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975
3.3. Homogeneous macrocrysts of Fe3+-rich ilmenite
These macrocrysts (0.2–2 cm across) are generally rounded with smooth surfaces, and they can be mono- or polycrystalline. The angular
Fig. 2. Primary grains of ilmenite. SEM image, mode BSE. (A) Intercumulus Fe3+-rich ilmenite (brighter) in peridotite xenolith. (B) Anhedral Mn-rich in carbonatite xenolith. (C) Ilmenite macrocryst without visible signs of replacement. Ilmenite (Ilm), phlogopite (Phl), zircon (Zrn), calcite (Cal), apatite (Ap).
3.2. Anhedral grains of ilmenite in carbonatite xenoliths Fig. 3. Partly replaced polycrystalline nodules of ilmenite. SEM image, mode BSE. Carbonatite xenoliths are rare (less than 7% of grains) in Catoca, and (A) Polycrystalline nodule of ilmenite (lightest) showing corrosion on the uppermost accessory ilmenite in them occurs as small crystals (50–90 μm) found side and replacement to Mg-rich ilmenite (slightly darker) along the subgrain borders. (B) Polycrystalline nodule of Fe3+-rich ilmenite corroded on one side and replaced along as inclusions in phlogopite crystals, or intergrown with calcite, apatite, subgrain borders and small cracks by Mg-rich ilmenite (darker) and Mn-rich ilmenite zircon and phlogopite (Fig. 2B). The ilmenite is usually rimmed at grain (pale gray to white). (C) Detail of a polycrystalline nodule of Fe3+-rich ilmenite. Ilmenite margins by Mn-rich ilmenite. (Ilm), serpentine (Srp). S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975 969 shape of monocrystalline grains is consistent with an origin by disag- they show an intense replacement, leading to a symplectitic texture; this gregation of polycrystalline grains (Fig. 2C). Homogeneous macrocrysts type of replacement is developed only at the border of the crystal of ilmenite or Mg-rich ilmenite are widely represented in kimberlites (Fig. 4A) or it can affect the whole crystal (Fig. 4B). The other mineral (i.e., Mitchell, 1973), but in the Catoca kimberlite, this textural type is originally present in these intergrowths has been fully replaced by very rare (less than 5% of grains). serpentine, but similar unreplaced textures in many kimberlites consist of pyroxene and ilmenite (i.e., Haggerty et al., 1977). These grains have 3.4. Partly replaced macro- and megacrysts of Fe3+-rich ilmenite complex patterns of replacement, with ilmenite strongly replaced by Mg-rich ilmenite; the zoned grains are finally overgrown by Mn-rich These coarse crystals (0.4–2 cm) are also rounded and they occur ilmenite (Fig. 4C). Manganese-rich ilmenite forms only a thin rim or usually in the core of pelletal lapilli; they present different grades of veinlet, and it is accompanied by small grains of barytocalcite, replacement (Fig. 3A, B, C). These megacrysts consist mainly of Fe3+- strontianite and baddeleyite. Most of the symplectitic replacement rich ilmenite similar in shape and composition to the last category. takes place in crystal discontinuities and is adapted to pre-existing However, they have been corroded and replaced by Mg-rich ilmenite features as deformation-induced kink bands or subgrains. along old surfaces and discontinuities. Some crystals may display freshly fractured borders, thus indicating that corrosion and replace- 3.6. Skeletal crystals of ilmenite in a pelletal matrix ment took place before the explosive processes in the kimberlite. These small crystals (80–150 μm) may be present in pelletal lapilli 3.5. Macrocrysts of symplectitic ilmenite and exhibit very irregular shapes (Fig. 4D). The core of the crystals is generally constituted by ilmenite, which can be partially replaced by Macrocrysts of Fe3+-rich ilmenite from the deepest parts of diatreme Mn-rich ilmenite. It is the least common variety of ilmenite in the Catoca facies are similar in shape and composition to two previous types, but kimberlite, and is found only in the uppermost part of the diatreme.
Fig. 4. Advanced replacements of ilmenite. SEM image, mode BSE. (A) An elongate crystal of Fe3+-rich ilmenite affected by kink-band deformation, partly replaced in both sides by Mg-rich ilmenite (slightly darker). It is possible to have symplectitic intergrown of three kinds of ilmenite by a replacement mechanism. The symplectitic intergrowth here is not primary. Intergrowths of ilmenite have previously been described in literature (Tompkins and Haggerty, 1984; Haggerty and Tompkins, 1984; Kostrovitsky and Piskunova, 1990; Haggerty, 1995). (B) Another crystal of Fe3+-rich ilmenite almost completely replaced by Mg-rich ilmenite. This intergrowth is probably of the same origin as in 4A. The gray domains are made of ilmenite–geikielite solid solution. (C) Detail of a symplectitic intergrown of ilmenite. Note that some grains of Mn-rich ilmenite are euhedral. Accompanying minerals include serpentine (Srp), calcite (Cal), witherite (Wth). Note the sharp contact between Mg-rich ilmenite and Mn-rich ilmenite. (D) Skeletal crystals of ilmenite showing similar replacements as in the symplectitic intergrowth. Ilmenite (Ilm), witherite (Wth). 970 S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975
3.7. Tabular crystals of Mn-rich ilmenite in the kimberlite matrix in Mg between Fe3+-rich ilmenite and Mg-rich ilmenite. As expected, there is a clear negative correlation between Ti and Fe3+, and the lowest Tabular crystals of Mn-rich ilmenite (1–10 μm in length), slightly Nb- values in Fe3+ are found in Mn-rich ilmenite, which produces a tight rich, are set in randomly oriented groups in the kimberlite groundmass group (Fig. 6B). However, there is a continuous trend among composi- (Fig. 5), and may be associated with euhedral grains of apatite and tions of Fe3+-rich ilmenite and Mg-rich ilmenite. The level of Fe2+ is chromite. They are internally homogeneous and unaltered, reflecting higher in Mg-rich ilmenite than in Fe3+-rich ilmenite, but it also tends to equilibrium with the kimberlite matrix, and are found only in the decrease in Mn-rich ilmenite owing to the substitution of Fe2+ for Mn2+ diatreme facies. Although rare, members of the ilmenite group have (Fig. 6C). The contents of Cr are quite variable, but they are higher in Mg- been found in the groundmass of other kimberlite pipes, either Mn- rich ilmenite and lower in Mn-rich ilmenite; they show a rough positive enriched and slightly Nb-enriched (Chakhmouradian and Mitchell, correlation with Mg. On the other hand, Mn and Fe3+ very clearly show 1999) or Mg-rich (Nielsen and Sand, 2008). an antithetic behaviour (Fig. 6D), which can be described as a trend of Fe3+ decrease (at low levels of Mn) followed by a trend of Mn increase 4. Composition of ilmenite (without Fe3+). The Zr contents are quite low in all the types of ilmenite. A negative correlation between Fe3+ and Mg can be seen, with a Representative grains of ilmenite from every class of texture were progressive increase in Mg from Fe3+-rich ilmenite toward Mg-rich selected using as reference the SEM-BSE images, and then analyzed ilmenite (Fig. 6E). Finally, the niobium content tends to increase where with a Cameca SX-50 microprobe, with four wavelength-dispersion the Mn content increases in Mn-rich ilmenite (Fig. 6F), as described in spectrometers. All ilmenite crystals were analyzed with an excitation other kimberlite fields (i.e., Chakhmouradian and Mitchell, 1999), voltage of 20 keV, beam current of 20.1 nA and a take-off angle of 40°. although the highest values are found in ilmenite from carbonatite We used the following standards, crystals and lines; periclase (Mg, xenoliths. Groundmass ilmenite is compositionally similar to the outer TAP Kα), orthoclase (Al, TAP Kα), diopside (Si, TAP Kα), wollastonite rim of ilmenite megacrysts.
(Ca, PET Kα), rutile (Ti, PET Kα), synthetic Cr2O3 (Cr, PET Kα), rhodonite (Mn, LIF Kα), hematite (Fe, LIF Kα), synthetic NiO (Ni, LIF 5. Discussion
Kα), synthetic ZrO2 (Zr, PET Lα), and metallic Nb (Nb, PET Mα). The ratio Fe2+/Fe3+ is calculated by stoichiometry. Ilmenite textures and composition are diverse in the Catoca At this phase of the research, we analyzed with EMP about 400 kimberlite, thus suggesting a complex history for the ilmenite points from 40 grains of 20 representative samples of a suite of 100 nodules. The diversity in textures and composition reflects primarily core samples (81 thin sections and 19 probes) from two boreholes of to the paragenetic position of ilmenite in the kimberlite (accessory Catoca Kimberlite. Datasets totalling less than 95% after charges were in xenoliths, macro- and megacrysts, matrix) and replacement balanced were rejected. Analyses were made along profiles in order to processes. evaluate progressive changes among the types of ilmenite. Ilmenite has been described in many kimberlites worldwide as an Correlations among the major and minor components in ilmenite of accessory mineral in many metasomatized peridotite xenoliths. Some the different types are shown in Fig. 6; selected results of electron- widespread examples comprise the MARID suite (Dawson and Smith, microprobe analyses from Catoca are given in Table 1 under the fol- 1977), MORID veins (i.e., Jones et al., 1982) and some metasomatized lowing headings: (1) intercumulus in peridotite xenolith, (2) anhedral garnet-bearing peridotites (Stiefenhofer et al., 1997; Kopylova et al., in carbonatite xenolith, (3) homogeneous macrocryst, (4) partly re- 1999). Some of the Catoca xenoliths can be included in the last placed macro- and megacryst, (5) symplectitic macrocryst, (6) skeletal category, and others have similarities with the ilmenite-bearing crystal, and (7) matrix. As can be seen in Fig. 6A, the Mn-rich ilmenite dunite xenoliths described by Kaminsky et al. (2002). plots outside the classic kimberlite domain in the TiO2–MgO plot of The macro- and megacrysts in kimberlites have been interpreted Wyattetal.(2004); moreover, there is a continuous trend of enrichment worldwide either as xenocrysts (Armstrong et al., 2004; Hearn, 2004)or as produced by primary magmatic crystallization (Moore,1987). At least at Catoca, the similarity in composition of ilmenite in the unreplaced parts of all the macro- and megacrysts and ilmenite from intercumulus positions in peridotite xenoliths suggest that the most if not all of the ilmenite nodules are produced by disaggregation of ilmenite-bearing metasomatized peridotite xenoliths. On the other hand, the composition of this early ilmenite is unusual because of its high Fe3+ contents. Similar Fe3+-rich compositions, although rare in kimberlites, have also been found in the Koidu kimberlite, in Sierra Leone (Tompkins and Haggerty, 1985), but at Catoca, the ilmenite is even more strongly oxidized,
indicating crystallization under relatively high fO2 conditions (Fig. 7A); this ilmenite also contains Nb, Cr and Ni in low contents. The second feature is that ilmenite is replaced along small discontinuities, both in the grain borders and along internal surfaces (cracks, twin planes, cleavages, kink bands), by Mg-rich ilmenite. The replacement of the Fe3+-rich ilmenite by Fe2+- and Mg-rich ilmenite is indicative of a trend toward more reducing conditions (Haggerty and Tompkins, 1983; Fig. 7A). This type of sequence is similar to the so-called ilmenite magmatic trend (Haggerty et al., 1977; Pasteris, 1980; Schulze, 1984). However, the textural patterns attributable to replacement at Catoca, along grain borders, cracks or other disconti- nuities, strongly suggest the action of a fluid rather than a magma. It is difficult to ascertain the timing and place of this replacement. Certainly it was produced before kimberlite emplacement, because Fig. 5. Euhedral platy crystals of Mn-rich ilmenite in matrix. Serpentine (Srp). The some nodules broken during the explosive processes are not replaced crystals are in equilibrium with the kimberlite matrix in the diatreme facies. in the broken corners. S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975 971
Fig. 6. Correlation between major elements or between major and minor elements for the main textural types of ilmenite. Atoms per formula unit (apfu). 972
Table 1 Chemical analyses of ilmenite crystals from Catoca.
Texture 1 1 1 2 3 3 3 4 4 4 5 5 5 6 6 6 7 7 7 Ilm type I I II IV I I I I II II I II III I II III III III III Borehole 535 535 535 535 535 535 535 535 535 535 535 535 535 535 535 535 535 535 535 Depth (m) 350 350 350 451 409 409 409 350 350 451 350 451 350 350 350 350 350 350 350 Point 26Bsn68 26Bsn67 26Bsn73 36f57 409a32b 40932b21 40932c21 26Bn74 26Ai63 36g59 26Aj17 36a13 26Aj24 26Aq32 26Ae39 26Ae34 26Aa12 26Aa15 26An86 (wt.%) SiO2 0.00 0.04 0.04 0.33 0.04 0.00 0.03 0.00 1.26 0.03 0.05 0.05 0.11 0.05 0.05 0.17 0.43 1.00 0.18 TiO2 43.88 44.06 50.67 46.54 36.78 36.49 36.65 38.69 51.64 54.30 47.25 50.10 51.30 42.13 46.12 51.36 48.79 47.86 49.79 Al2O3 0.19 0.21 0.23 0.02 0.12 0.04 0.14 0.12 0.18 0.16 0.14 0.07 0.05 0.15 0.12 0.02 0.05 0.16 0.06 ..Rbe-rze l ihs12 20)966 (2009) 112S Lithos / al. et Robles-Cruz S.E. Nb2O5 0.26 0.33 0.27 3.03 0.34 0.48 0.43 0.40 1.20 0.21 0.51 0.25 0.11 0.30 0.24 0.07 2.54 1.59 0.96 ZrO2 0.12 0.14 0.17 0.06 –– – 0.10 0.17 0.02 0.00 0.00 0.02 0.15 0.08 0.06 0.01 0.12 0.08 Cr2O3 0.90 0.89 1.04 0.00 2.14 2.86 2.88 0.28 0.70 1.07 0.89 2.09 0.08 0.62 3.88 0.29 0.15 0.36 0.17 Fe2O3 19.99 18.93 8.68 5.33 28.59 28.19 27.56 28.43 9.85 6.20 12.44 9.91 1.30 20.51 13.39 0.23 1.26 2.81 1.73 FeO 28.76 27.56 30.30 41.04 27.35 27.40 27.72 28.26 16.65 23.76 30.20 27.76 43.40 26.93 25.65 33.02 37.37 35.88 37.77 MnO 0.20 0.24 0.27 0.68 0.15 0.18 0.22 0.18 2.06 0.44 1.88 0.39 2.68 0.23 0.25 13.08 6.45 6.35 6.81 MgO 5.87 6.66 8.43 0.15 3.74 3.71 3.63 3.57 16.43 13.79 5.78 9.47 0.06 6.07 8.74 0.07 0.13 0.75 0.15 NiO 0.08 0.07 0.08 0.03 0.05 0.04 0.03 0.05 0.04 0.08 0.08 0.10 0.03 0.07 0.09 0.01 0.01 0.03 0.00 CaO 0.03 0.02 0.04 0.21 0.01 0.00 0.00 0.01 0.00 0.02 0.06 0.00 0.03 0.00 0.02 0.03 0.21 0.50 0.09 Total 100.28 99.14 100.22 97.43 99.32 99.39 99.30 100.08 100.17 100.08 99.29 100.18 99.17 97.20 98.63 98.41 97.40 97.41 97.79
Cations on basis of three O atoms (apfu) Si 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.03 0.00 Ti 0.81 0.80 0.90 0.97 0.70 0.69 0.70 0.72 0.87 0.93 0.87 0.89 0.98 0.79 0.84 0.99 0.95 0.92 0.96
Al 0.01 0.01 0.01 0.01 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 – Nb 0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.03 0.02 0.01 975 Zr 0.00 0.00 0.00 0.00 –– – 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Cr 0.02 0.02 0.02 0.00 0.05 0.06 0.06 0.01 0.01 0.02 0.02 0.04 0.00 0.01 0.07 0.01 0.00 0.01 0.00 Fe3+ 0.35 0.37 0.15 0.01 0.54 0.54 0.52 0.53 0.17 0.11 0.23 0.18 0.02 0.39 0.24 0.00 0.02 0.05 0.03 Fe2+ 0.56 0.58 0.60 0.94 0.58 0.58 0.59 0.59 0.31 0.45 0.62 0.55 0.92 0.56 0.52 0.71 0.81 0.77 0.81 Mn 0.00 0.00 0.01 0.03 0.01 0.01 0.01 0.00 0.04 0.01 0.04 0.01 0.06 0.00 0.01 0.28 0.14 0.14 0.15 Mg 0.24 0.21 0.30 0.00 0.14 0.14 0.14 0.13 0.55 0.47 0.21 0.33 0.00 0.23 0.31 0.00 0.00 0.03 0.01 Ni 0.00 0.00 0.00 0.00 –– – 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ca 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.00 FeTiO3 57.44 59.49 60.91 96.41 58.00 58.00 59.00 59.90 31.47 45.69 62.94 56.12 92.93 56.85 54.17 71.72 84.38 79.79 82.23 Fe2O3 17.95 18.97 7.61 0.51 27.00 27.00 26.00 26.90 8.63 5.58 11.68 9.18 1.01 19.80 12.50 0.00 1.04 2.59 1.52 MnTiO3 0.00 0.00 1.02 3.08 1.00 1.00 1.00 0.00 4.06 1.02 4.06 1.02 6.06 0.00 1.04 28.28 14.58 14.51 15.23 MgTiO3 24.62 21.54 30.46 0.00 14.00 14.00 14.00 13.20 55.84 47.72 21.32 33.67 0.00 23.35 32.29 0.00 0.00 3.11 1.02 Total 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Textural types: (1) intercumulus in peridotite xenolith; (2) anhedral in carbonatite xenolith; (3) homogeneous macrocryst; (4) partly replaced macro- and megacryst; (5) symplectitic macrocryst; (6) skeletal crystal; (7) matrix. The Fe3+ is calculated by charge balance and stoichiometry. S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975 973
Fig. 7. Compositions of the different textural types of ilmenite in the Catoca kimberlite: (A) in terms of the geikielite (MgTiO3)–ilmenite (FeTiO3)–hematite (Fe2O3) diagram, after
Haggerty and Tompkins (1983); (B) in therms of the pyrophanite (MnTiO3)–ilmenite (FeTiO3)–hematite (Fe2O3) diagram.
On the other hand, a late generation of Mn-rich ilmenite is found primary magmatic crystallization in the matrix of the kimberlite. In all rimming all the above-mentioned generations, and is extremely poor these cases, however, Mn-rich ilmenite is produced in late events in in Mg and Fe3+. The replacement of the Fe3+-rich ilmenite by Fe2+- the paragenetic sequence at Catoca, and in many cases mantles other and Mn-rich ilmenite is also indicative of a trend toward strongly groundmass minerals such as perovskite and spinel (Tompkins and reducing conditions (Haggerty and Tompkins, 1983; Fig. 7A,B). Haggerty, 1985). Although Mn-rich ilmenite could be produced during Accordingly, these compositions could follow the kimberlite reaction magmatic crystallization, we contend that it could be also produced trend of Haggerty et al. (1977), producing enrichment in Fe2+. during late hydrothermal processes, during serpentinization. In fact, However, the most significant aspects in this process are the strong pyrophanite can be produced during serpentinization of ultrabasic enrichments in Mn and HFSE. Similar enrichments were interpreted in rocks, where it appears as a late mineral in the paragenetic sequence other kimberlites worldwide as produced by crystallization at the (Mücke and Woakes, 1986; Liipo et al., 1994). expense of a late-stage fraction of melt (Tompkins and Haggerty, 1985, In any case, all of the ilmenite fractions in kimberlite are quite Chakhmouradian and Mitchell, 1999). In the Catoca case, two facts different from those found in the carbonatitic xenoliths at Catoca. In suggest instead the deposition of this ilmenite under the influence of a this case, the growth of ilmenite takes place during the early stages of
CO2-rich fluid phase: a) the intimate association of this Mn-rich magmatic crystallization, and there is no evidence of replacement of a ilmenite in open cavities with calcite, witherite, barytocalcite and precursor ilmenite. Moreover, the crystals are distinct from the other strontianite; b) the development of this mineral association filling variants of ilmenite in being extremely poor in Mg and Cr and the small fractures. In fact, the late stages of kimberlite emplacement are richest in Nb, thus defining a particular class, more similar to ilmenite developed under the influence of CO2-rich fluids (Head and Wilson, found in carbonatites (Gaspar and Wyllie, 1983, 1984). 2008), whose are also responsible for the alteration of host rocks in The existence of many varieties of ilmenite at Catoca has significant many kimberlite fields worldwide (Smith et al., 2004); Agee et al. consequences in mineral exploration. Magnesium-rich ilmenite has (1982) also attributed the formation of Mn-rich ilmenite in the Elliott traditionally been interpreted as an indicator of kimberlite associations,
County kimberlite, Kentucky (USA) to Ca-enriched late fluids. as well as an indicator of low fO2, which is necessary for the preservation Furthermore, the composition of this replaced ilmenite is similar to of diamond (Garanin et al.,1997; van Straaten et al., 2008). However, the that of the fine-grained euhedral ilmenite crystals found in the Fe3+-rich ilmenite represents in the Catoca kimberlite more than 70% of kimberlite matrix. Analogous trends have been already described in the volume of the grains, and compositions fall into the domains of “no other kimberlite fields, but in the hypabyssal facies (i.e. Hunter et al., preservation of diamond” according the diagram of Gurney and Zweistra 1984). Similar textures and compositions in the groundmass are not (1995) Fig. 8. Moreover, these compositions of ilmenite are Mg- and Cr- rare in kimberlites. Tompkins and Haggerty, 1985; Chakhmouradian poor, and hence using other criteria for discrimination among fertile and and Mitchell, 1999 interpreted this type of ilmenite as produced by barren kimberlites (i.e., Haggerty, 1995); Catoca could be expected to be 974 S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975
Fig. 8. Representation of the compositions of ilmenite from the Catoca suite in the diagram of Gurney and Zweistra (1995). Note that most of the population of primary compositions of ilmenite from Catoca plots in poorly mineralized domains. barren. Although Catoca is a diamondiferous kimberlite, Mg-rich ilmen- Acknowledgments ite here is clearly a product of late replacement, and the extent of replacement of the primary grains is very variable. In other words, This research was supported by the projects CGL2005-07885/BTE textural relations must be taken into account in the application of and CGL2006-12973 of Ministerio de Educación y Ciencia (Spain), the discriminants based on composition. AGAUR SGR 589 of Generalitat de Catalunya and a FI grant sponsored by the Departament d'Educació i Universitats de la Generalitat de Catalunya and European Social Fund. We also thank ENDIAMA and the mine 6. Conclusions geologists, who kindly allowed us to acquire samples for this study and gave all facilities for the mine trip. The authors also acknowledge the The composition of the Catoca ilmenite is complex, as the result of Serveis Cientificotècnics de la Universitat de Barcelona for assistance in multiple processes. The ilmenite macro- and megacrysts are assumed to the use of SEM/ESEM-BSE-EDS analyses (E. Prats, R. Fontarnau†,Dr.J. be produced by disaggregation of ilmenite-bearing xenoliths (mainly García Veigas) and EMP (Dr. Xavier Llovet). An early version of the relatively oxidized and metasomatized mantle peridotites and minor manuscript was improved with the comments of two anonymous carbonatites). The subsequent reaction under disequilibrium conditions reviewers. Vicki Loschiavo and Prof. Robert F. Martin made further with kimberlite-derived fluids produced the replacement of the above improvements to the revised version. macro- and megacrysts by secondary Mg-rich ilmenite. Late subsolidus reactions with the fluids associated with the kimberlite, also in disequilibrium conditions, produced the replacement References of the early ilmenite types by highly reduced Mn-rich ilmenite. The enrichment in Nb of this late ilmenite (and in the ilmenite of the matrix), Agee, J.J., Garrison Jr., J.R., Taylor, L.A., 1982. Petrogenesis of oxide minerals in kimberlite, as well as its intimate association with carbonates of Ba and Sr, can be Elliott County, Kentucky (USA). American Mineralogist 67 (1–2), 28–42. interpreted in terms of an interaction of the ilmenite crystals with a CO - Arculus, R.J., Dawson, J.B., Mitchell, R.H., Gust, D.A., Holmes, R.D., 1984. Oxidation states 2 of the upper mantle recorded by megacryst ilmenite in kimberlite and type A and B rich fluid. spinel lherzolites. Contributions to Mineralogy and Petrology 85, 85–94. Although Catoca is a diamondiferous kimberlite, most of its ilmenite Armstrong, K.A., Nowicki, T.E., Read, G.H., 2004. Kimberlite AT-56: a mantle sample – compositions are strongly oxidized and poor in Cr and Mg. Therefore, the from the north central Superior craton, Canada. Lithos 77, 695 704. Chakhmouradian, A.R., Mitchell, R.H., 1999. Niobian ilmenite, hydroxylapatite and absence of Mg-rich ilmenite in a kimberlite or the corresponding placers sulfatian monazite: alternative hosts for incompatible elements in calcite kimberlite does not appear to be a convincing argument to exclude the occurrence from Internatsional'naya, Yakutia. The Canadian Mineralogist 37 (5), 1177–1189. of economic deposits of diamond. Accordingly, this work proposes new Clement, C.R., Skinner, E.M.W., 1985. A textural-genetic classification of kimberlites. Transactions of the Geological Society of South Africa 88, 403–409. insight into the concept of ilmenite in exploration for diamond; Dawson, J.B., Smith, J.V., 1977. The MARID (mica–amphibole–rutile–ilmenite–diopside) compositional attributes must be evaluated in light of textural attributes. suite of xenoliths in kimberlite. Geochimica et Cosmochimica Acta 41 (2), 309–323. S.E. Robles-Cruz et al. / Lithos 112S (2009) 966–975 975
De Carvalho, H., Tassinari, C., Alves, P.H., 2000. Geochronological review of the Kotel'nikov, D.D., Zinchuk, N.N., Zhukhlistov, A.P., 2005. Stages of serpentine and Precambrian in western Angola: links with Brazil. Journal of African Earth Sciences phlogopite transformation in the Catoca kimberlite pipe, Angola. Doklady Earth 31 (2), 383–402. Sciences 403 (6), 866–869. Ganga, J., Rotman, A.Y., Nosiko, S., 2003. Pipe Catoca, an example of the weakly eroded Kriuchkov, A., Sarychev, I., Vouiko, V., Chipoia, B., Antoniuk, B., 2000. Kimberlite pipe of kimberlites from North-East of Angola. The 8th International Kimberlite Con- Catoca. Geology, substantial analyses of the rocks and characteristics of diamonds. ference, Extended Abstract. Geoluanda 2000 Abstract volume, 91–92. Garanin, V.K., Kudryavtseva, G.P., Posukhova, T.V., 1997. Indicators of diamond Liipo, J.P., Vuollo, J.E., Nykänen, V.M., 1994. Pyrophanite and ilmenite in serpentinized preservation in kimberlite. In: Papunen, H. (Ed.), Mineral Deposits: Research and wehrlite from Ensilä, Kuhmo greenstone belt, Finland. European Journal of Exploration, Where Do They Meet? In: Proceedings of the 4th Biennial SGA Mineralogy 6, 145–150. Meeting. Turku, Finland, pp. 767–770. Llusià Queral, R.M., Pereira Ginga, S., Galiano, A., Sebastião, P., Melgarejo Draper, J.C., Olimpio Gaspar, J.C., Wyllie, P.J., 1983. Ilmenite (high magnesium, manganese, niobium) in the Gonçalves, A., Proenza Fernandez, J., Morais, E., 2005a. Aluviões asociados ao campo carbonatites from the Jacupiranga Complex, Brazil. American Mineralogist 68 (9–10), kimberlítico da zona de Muanga (Lunda Norte, Angola. XIV semana Geoquímica/VIII 960–971. Congresso de Geoquimica dos Paises de Lingua Portuguesa, pp. 351–354. Gaspar, J.C., Wyllie, P.J., 1984. The alleged kimberlite-carbonatite relationship: evidence Llusià Queral, R.M., Pereira Ginga, S., Proenza, J.A., José Ribamar, M.R., Sebastião, P., from ilmenite and spinel from Premier and Wesselton Mines and the Benfontein Olimpio Gonçalves, A., Melgarejo Draper, J.C., Galiano, A.C., Alves Morais, E., 2005b. Sill, South Africa. Contributions to Mineralogy and Petrology 85, 133–140. O campo kimberlítico de Muanga (Lunda Norte, Angola. XIV semana Geoquímica/ Griffin, W.L., Ryan, C.G., 1995. Trace elements in indicator minerals: area selection VIII Congresso de Geoquimica dos Paises de Lingua Portuguesa, pp. 347–350. and target evaluation in diamond exploration. Journal of Geochemical Exploration Meyer, H.O.A., McCallum, M.E., 1986. Mineral inclusions in diamonds from the Sloan 53 (1–3), 311–337. kimberlites, Colorado. Journal of Geology 94 (4), 600–612. Grütter, H.S., Gurney, J.J., Menzies, A.H., Winter, F., 2004. An updated classification for Mitchell, R.H., 1973. Magnesian ilmenite and its role in kimberlite petrogenesis. Journal mantle-derived garnet, for use by diamond explorers. Lithos 77, 841–857. of Geology 81, 301–311. Guiraud, R., Bosworth, W., Thierry, J., Delplanque, A., 2005. Phanerozoic geological Mitchell, R.H., 1989. Kimberlites: mineralogy, geochemistry and petrology. New York, evolution of Northern and Central Africa: an overview. Journal of African Earth Plenum Press. 442 pp. Sciences 43, 83–143. Mitchell, R.H., 1995. Kimberlites, orangeites, and related rocks. New York, Plenum Press. Gurney, J.J., Helmstaedt, H., Moore, R.O., 1993. A review of the use and application of 410 pp. mantle mineral geochemistry in diamond exploration. Pure and Applied Chemistry Mitchell, R.H., 1997. Kimberlites, orangeites, lamproites, melilitites, and minettes: a 65 (12), 2423–2442. petrographic atlas. Almaz Press Inc, Ontario. Gurney, J.J., Zweistra, P., 1995. The interpretation of the major element compositions of Mitchell, R.H., et al., 2009, this issue. Tuffisitic kimberlites from the Wesselton Mine, mantle minerals in diamond exploration. Journal of Geochemical Exploration 53 South Africa: Mineralogical characteristics relevant to their formation. Proceedings (1–3), 293–309. of the 9th International Kimberlite Conference. Lithos 112S, 452–464. Haggerty, S.E., 1991a. Oxide textures - a mini-atlas. In: Lindsley, D.H. (Ed.), Oxide Moore, A.E., 1987. A model for the origin of ilmenite in kimberlite and diamond: Minerals: Petrologic and Magnetic Significance. Reviews in Mineralogy 25, implications for the genesis of the discrete nodule (megacryst) suite. Contributions 129–219. to Mineralogy and Petrology 95, 245–253. Haggerty, S.E., 1991b. Oxide mineralogy of the upper mantle. In: Lindsley, D.H. (Ed.), Mücke, A., Woakes, M., 1986. Pyrophanite: a typical mineral in the Pan-African Province oxide minerals: petrologic and magnetic significance. Reviews in Mineralogy 25, of Western and Central Nigeria. Journal of African Earth Sciences 5 (6), 675–689. 355–416. Nielsen, T.F.D., Sand, K., 2008. The Majuagaa kimberlite dike, Maniitsoq region, West Haggerty, S.E., 1995. Upper mantle mineralogy. Journal of Geodynamics 20 (4), Greenland: constraints on an Mg-rich silico-carbonatitic melt composition from 331–364. groundmass mineralogy and bulk compositions. The Canadian Mineralogist 46, Haggerty, S.E., Hardie III, R.B., McMahon, B.M., 1977. The composition of ilmenite nodule 1043–1061. associations from the Monastery diatreme: Proceedings of the 2nd International Pasteris, J., 1980. The significance of groundmass ilmenite and megacryst ilmenite in Kimberlite Conference, vol. 2, pp. 249–256. kimberlites. Contributions to Mineralogy and Petrology 75 (4), 315–325. Haggerty, S.E., Tompkins, L.A., 1983. Redox state of Earth's upper mantle from Read, G.H., Janse, A.J.A.(B.), 2009, this issue. Diamonds: Exploration, mines and kimberlitic ilmenites. Nature 303 (5915), 295–300. marketing. Proceedings of the 9th International Kimberlite Conference. Lithos 112S, Haggerty, S.E., Tompkins, L.A., 1984. Subsolidus reactions in kimberlitic ilmenites: 1–9. exsolution, reduction and the redox state of the mantle. In: Kornprobst, J. (Ed.), Reis, B., 1972. Preliminary note on the distribution and tectonic control of kimberlites in Kimberlites I: Kimberlites and Related Rocks. Proceedings of the 3rd International Angola: The 24th International Geological Congress - Section 4, pp. 276–281. Kimberlite Conference. Rogers, A.J., Grütter, H.S., 2009, this issue. Fe-rich and Na-rich megacryst clinopyroxene Head, J.W., Wilson, L., 2008. Integrated model for kimberlite ascent and eruption. 9th and garnet from the Luxinga kimberlite cluster, Lunda Sul, Angola. Proceedings of International Kimberlite Conference Extended abstract No. 9IKC-A-00284. the 9th International Kimberlite Conference. Lithos 112S, 942–950. Hearn Jr., B.C., 2004. The Homestead kimberlite, central Montana, USA: mineralogy, Schulze, D.J., 1984. Cr-poor megacrysts from the Hamilton Branch kimberlite, Elliott xenocrysts, and upper-mantle xenoliths. Lithos 77, 473–491. County, Kentucky. In: Kornprobst, J. (Ed.), Kimberlites II: The Mantle and Crust–Mantle Hunter, R.H., Kissling, R.D., Taylor, L.A., 1984. Mid- to late-stage kimberlitic melt Relationships. In: Developments in Petrology 11B. Amsterdam, Elsevier, pp. 97–108. evolution: phlogopites and oxides from the Fayette County kimberlite, Pennsylva- Scott Smith, B.H., Nowicki, T.E., Russell, J.K., Webb, K.J., Hetman, C.M., Harder, M., nia. American Mineralogist 69, 30–40. Mitchell, R.H., 2008. Kimberlites: Descriptive Geological Nomenclature and Hwang, P., Taylor, W.R., Rock, N.M.S., Ramsay, R.R., 1994. Mineralogy, geochemistry and Classification. 9th International Kimberlite Conference Extended abstract No. petrogenesis of the Metters Bore No. 1 lamproite pipe, Calwynyardah field, West 9IKC-A-00124. Kimberley Province, Western Australia. Mineralogy and Petrology 51 (2–4), Smith, C.B., Sims, K., Chimuka, L., Duffin, A., Beard, A.D., Townend, R., 2004. Kimberlite 195–226. metasomatism at Murowa and Sese pipes, Zimbabwe. Lithos 76 (1–4), 219–232. Janse, A.J.A., Sheahan, P.A., 1995. Catalogue of world wide diamond and kimberlite Sparks, R.S.J., Baker, L., Brown, R.J., Field, M., Schumacher, J., Stripp, G., Walters, A., 2006. occurrences: a selective and annotative approach. Journal of Geochemical Dynamical constraints on kimberlite volcanism. Journal of Volcanology and Exploration 53, 73–111. Geothermal Research 155 (1–2), 18–48. Jones, A.P., Smith, J.V., Dawson, J.B., 1982. Mantle metasomatism in 14 veined peridotites Stiefenhofer, J., Viljoen, K.S., Marsh, J.S., 1997. Petrology and geochemistry of peridotite from Bultfontein mine, South Africa. Journal of Geology 90, 435–453. xenoliths from the Letlhakane kimberlites, Botswana. Contributions to Mineralogy Kaminsky, F.V., Sablukov, S.M., Sablukova, L.I., Shchukin, V.S., Canil, D., 2002. Kimberlites and Petrology 127 (1–2), 147–158. from the Wawa area, Ontario. Canadian Journal of Earth Sciences 39 (12), Tompkins, L.A., Haggerty, S.E., 1984. Koidu Kimberlite Complex, Sierra Leone. Geological 1819–1838. Setting, Petrology and Mineral Chemistry. In: Kornprobst, J. (Ed.), Kimberlites I: Kopylova, M.G., Russell, J.K., Cookenboo, H., 1999. Petrology of peridotite and pyroxenite Kimberlites and Related Rocks. Proceedings of the 3rd International Kimberlite xenoliths from the Jericho kimberlite: implications for the thermal state of the Conference. mantle beneath the Slave craton, Northern Canada. Journal of Petrology 40 (1), Tompkins, L.A., Haggerty, S.E., 1985. Groundmass oxide minerals in the Koidu kimberlite 79–104. dikes, Sierra Leone, West Africa. Contributions to Mineralogy and Petrology 91 (3), Kostrovitsky, S.I., Alymova, N.V., Yakovlev, D.A., Serov, I.V., Ivanov, A.S., Serov, V.P., 2006. 245–263. Specific features of “picroilmenite” composition in various diamondiferous fields of Van Straaten, B.I., Kopylova, M.G., Russell, J.K., Webb, K.J., Scott Smith, B.H., 2008. the Yakutian province. Doklady Earth Sciences 406 (1), 19–23. Discrimination of diamond resource and non-resource domains in the Victor North Kostrovitsky, S.I., Malkovets, V.G., Verichev, E.M., Garanin, V.K., Suvorova, L.V., 2004. pyroclastic kimberlite, Canada. Journal of Volcanology and Geothermal Research 17, Megacrysts from the Grib kimberlite pipe (Arkhangelsk Province, Russia). Lithos 77 128–138. (1–4), 511–523. Wyatt, B.A., Mike, B., Anckar, E., Grütter, H., 2004. Compositional classification of Kostrovitsky, S.I., Piskunova, L.F., 1990. Two groups of symplectites from the same “kimberlitic” and “non-kimberlitic” ilmenite. Lithos 77, 819–840. Kimberlite pipe. Doklady Earth Science Sections 306 (3), 176–179. ORIGINAL PUBLICATIONS
PAPER III
Reprinted from MACLA - Revista de la Sociedad Española de Mineralogía, September No.11. Robles- Cruz, S., Lomba, A., M., Melgarejo, J., Galí, S., Olimpio, A., 2009. The Cucumbi Kimberlite, NE Angola: Problems to Discriminate Fertile and Barren Kimberlites.
macla nº 11. septiembre ‘09 revista de la sociedad española de mineralogía 159
The Cucumbi Kimberlite, NE Angola: Problems to Discriminate Fertile and Barren
Kimberlites / SANDRA ROBLES-CRUZ (1,*), ANDRÉ LOMBA (2), JOAN CARLES MELGAREJO (1), SALVADOR GALI (1), ANTONIO OLIMPIO GONÇALVES (3) (1) Departament de Cristal·lografia, Mineralogia i Dipòsits Minerals. Facultat de Geologia. Universitat de Barcelona. Martí i Franquès s/n. 08028, Barcelona (España) (2) ENDIAMA, Angola (3) Departamento de Geologia, Faculdade de Ciências, Universidade Agostinho Neto, Av. 4 de Fevereiro 7, 815, Luanda (Angola)
INTRODUCTION.
A classical key issue in exploration of diamondiferous kimberlites is the accurate use of typical diamond indicator minerals in order to discriminate among fertile and barren kimberlites. In fact, the conclusive criterion is the occurrence of diamond itself which proves the productivity of a given kimberlite. In a previous paper (Robles-Cruz et al., 2009), we pointed out that in the Catoca pipe, the use of ilmenite composition is not suitable to confirm the diamond grade. We have used a new set of samples in Cucumbi area to study the reliance of some of these parameters, in particular, the use of garnet composition as a guide in diamond exploration.
Cucumbi is located in Cacolo, Lunda Sul province, northeastern Angola. This area is notable because of the occurrence of diamondiferous kimberlites (fig. 1).
fig 1. General location of kimberlites in Angola. Modified after De Carvalho et al. (2000) and Egorov et al. Cucumbi samples exhibit crater facies (2007). along the first 100 m, characterized by volcanoclastic rocks, and diatreme composed by gabbro, norite and (alkaline, carbonatitic, kimberlitic) and facies, showing typical tuffisitic charnockitic complexes, which marginal basins. The Lower Cretaceous kimberlite (Mitchell et al., 2009). constitute the Angolan basement. regional extension determined the development of deep faults and grabens METHODOLOGY. (2) Three main Proterozoic cycles, with trends NE-SW and NW-SE. The Eburnean-Paleoproterozoic, Kibaran- Lucapa structure corresponds to the first Thin and polished sections were studied Mesoproterozoic, and Pan-African- group, and the NE part concentrates using transmitted and reflected optical Neoproterozoic; being the Eburnean the most of the diamondiferous kimberlites microscopy, followed by SEM-BSE-EDS most important and characterized by in Angola, including Cucumbi, whereas analysis. Chemical analyses were volcanosedimentary groups, gneisses, the southwestern zone comprises obtained with EPMA. migmatites, granites and syenites. important outcrops of undersaturated alkaline rocks and carbonatites (Reis, GEOLOGICAL SETTING. (3) Unconformably lying Phanerozoic 1972). Other minor kimberlite fields are sequences, which are the result of the found in the SW Angola (Egorov et al., Angola has a complex geological history Pangea formation and the consecutive 2007). that can be represented by three main breaking-up of Gondwana, that stages (De Carvalho et al., 2000; Fig. 1): contributed to the formation of rift PETROGRAPHY AND COMPOSITION basins associated to fault systems which (1) An important Archaean orogeny, later allowed the apparition of marine The Cucumbi samples exhibit all the registered by the Central Shield, Cuango sequences, the origin of the Karoo main characteristic features of Tuffisitic Shield and Lunda Shield, most of them Supergroup, intraplate magmatism Kimberlite TK (Figs. 2, 3). They are
palabras clave: Kimberlita, Mineral indicador, Diamante, Granate. key words: Kimberlite, Indicator mineral, Diamond, Garnet.
resumen SEM 2009 * corresponding author: [email protected] 160
generally massive, poorly sorted, clast- Garnet and clinopyroxene are usually de la Generalitat de Catalunya and supported rocks with the following main present as mega- and macrocrysts, and European Social Fund. We also thank components: anhedral olivine only rarely as part of xenoliths. ENDIAMA and the mine geologists, who macrocrysts, pseudomorphosed by Garnet composition is diverse. Using the kindly allowed us to acquire samples for serpentine and smectite; other mega- garnet classification of Grütter et al. this study and gave all facilities for the and macrocrysts as garnet, ilmenite, (2004), it may be stated (Fig. 4) that mine trip. The authors also acknowledge clinopyroxene, and phlogopite, whether some garnet derive from lherzolite (G9) the Serveis Cientificotècnics de la enclosed in a pelletal assemblage of and others from pyroxenite and eclogite Universitat de Barcelona for assistance serpentine or not, often pelletal lapilli, (G4, G5), only a few of them come from in the use of SEM/ESEM-BSE-EDS and an interclast groundmass in the uncommon, unusual or “polymict” analyses (E. Prats, Dr. J. García Veigas) matrix, mainly composed by serpentine, mantle lithologies. and EPMA (Dr. Xavier Llovet). less common by chlorite, smectite and calcite. The size and distribution of DISCUSSION AND CONCLUSIONS. REFERENCES. mega- and macrocrysts is chaotic (Figs. 2, 3). Using the diagram of Grütter et al. De Carvalho, H., Tassinari, C., Alves, P.H (2004) to plot the garnet compositions (2000): Geochronological review of the from Cucumbi, it should be pointed out Precambrian in western Angola: links with that all these compositions plot into the Brazil. Journal of African Earth Sciences 31 graphite domain, out of the (2), 383-402. diamondiferous field harzburgitic G10 Egorov, K.N., Roman’ko, E.F., Podvysotsky, V.T., Sablukov, S.M., Garanin, V.K., facies. Therefore, this kimberlite could D’yakonov, D.B. (2007): New data on be classified as barren using only that kimberlite magmatism in southwestern criterion. However, the Cucumbi Angola. Russian Geology and Geophysics kimberlite has proven to be 48, 323-336. diamondiferous. In fact, similar Grütter, H.S., Gurney, J.J., Menzies, A.H., problems were found in the Catoca pipe Winter, F. (2004): An updated classification when using the composition of ilmenite for mantle-derived garnet, for use by (Robles-Cruz et al., 2009) or the diamond explorers. Lithos 77, 841-857. Mitchell, R. H., Skinner, E. M., Scott-Smith, B. fig 2. Cucumbi, a diamondiferous drill hole. A typical composition of garnets. H. (2009): Tuffisitic Kimberlites: pattern of a tuffisitic kimberlite (TK) facies, with Mineralogical Characteristics Relevant to macrocrysts containing rounded pseudomorphosed Therefore, the garnet diagrams can be their Formation. Lithos, Special Issue 9IK., olivine xenocrysts, rounded ilmenite xenocrysts and used to verify the minimum level of in press crustal rock xenoliths, all set in a groundmass of serpentine and phlogopite. Image from the scanned diamond content, but some kimberlites Reis, B. (1972): Preliminary note on the thin section. may contain diamond samples from distribution and tectonic control of deeper sources. Hence, it should be kimberlites in Angola: The 24th taken into consideration when using International Geological Congress - Section 4, 276-281. these diagrams to assess the potential Robles-Cruz, S.E., Watangua, M., Isidoro, L., of kimberlite fields. Melgarejo, J.C., Galí, S., Olimpio, A. (2009): Contrasting compositions and textures of ilmenite in the Catoca kimberlite, Angola, and implications in exploration for diamond. Lithos, Special Issue 9IK., in press.
fig 3. Xenocrystals of olivine (pseudomorphosed by serpentine (Srp)), phlogopite (Phl), ilmenite (Ilm) in a serpentine groundmass. SEM image, mode BSE. fig 4. Classification of the Cucumbi garnets in a plot Cr2O3 versus CaO (wt.%), according with the Magnesian ilmenite (9-13 wt.% MgO) is compositional fields of Grütter et al. (2004). present as rounded mega- and macrocrysts (fig. 3), as part of xenoliths ACKNOWLEDGEMENTS. and as inclusions in phlogopite. In some cases macrocrysts of ilmenite are This research was supported by the partially replaced along the borders by projects CGL2005-07885/BTE and perovskite and spinel (Fig. 3). Ilmenite CGL2006-12973 of Ministerio de texture is usually either cumulus or Educación y Ciencia (Spain), the AGAUR homogenous. Symplectite textures are SGR 589 of Generalitat de Catalunya lacking in this kimberlite, in contrast and a FI-2006 grant sponsored by the with Catoca. Departament d’Educació i Universitats
ORIGINAL PUBLICATIONS
PAPER IV
Reprinted from Acta Mineralogica-Petrographica. Abstract Series, Vol. 6. Robles-Cruz, S.E., Escayola, M., Melgarejo, J.C., Watangua, M., Galí, S., Gonçalves, O.A., Jackson, S., 2010. Disclosed data from mantle xenoliths of Angolian kimberlites based on LA-ICP-MS analyses