Composition and thermal structure of the lithospheric mantle beneath kimberlite pipes from the Catoca cluster, Angola

Tectonophysics 530–531 (2012) 128–151

Contents lists available at SciVerse ScienceDirect

Tectonophysics journal homepage: www.elsevier.com/locate/tecto

Composition and thermal structure of the lithospheric mantle beneath kimberlite pipes from the Catoca cluster, Angola I.V. Ashchepkov a,⁎, A.Y. Rotman b, S.V. Somov b, V.P. Afanasiev a, H. Downes c, A.M. Logvinova a, S. Nossyko d, J. Shimupi e, S.V. Palessky a, O.S. Khmelnikova a, N.V. Vladykin f a

Institute of Geology and Mineralogy, SD RAS, acad. Koptyug avenue 3, Novosibirsk 630090, Russian Federation Central Science and Research Geology and Prospecting Institute of the Stock Company, “ALROSA”, Chernyshevsky Str,7, Mirny, 678170, Russian Federation Department of Earth and Planetary Sciences, Birkbeck University of London, London, UK d Catoca Mining Society, Angola e Endiama Co., Angola f Institute of Geochemistry SD RAS, Favorsky str. 1a, Irkutsk, 66403, Russian Federation b c

a r t i c l e

i n f o

Article history: Received 27 December 2010 Received in revised form 8 December 2011 Accepted 9 December 2011 Available online 22 December 2011 Keywords: Mantle lithosphere Thermobarometry Geochemistry Angola Garnet Kimberlite

a b s t r a c t Garnet, clinopyroxene and ilmenite xenocrysts from three Angolan kimberlite pipes belonging to the Catoca cluster (Angola Caquele, Camitongo I and II, and Catoca) from the SW part of the Congo–Kasai craton, reveal similar features which suggest a similarity of mantle structure. PT estimates for pyropes, Cr-diopsides and picroilmenites reveal similar geothermal conditions of ~ 37–40 mW/m2. This is slightly higher than the values determined for the Catoca pipe. Higher temperature conditions ~ 45 mW/m2 were determined for low-Cr pyroxenes and omphacites. The similar general mineralogy and suggested mantle lithology, as well as reconstructed layering of the sub-continental lithospheric mantle (SCLM), are similar for Camitongo I–II as well as for Caquele and Catoca pipes. Heating at depths of 7.5–4.5 GPa (240–140 km) is a general feature of the SCLM beneath the field. The high temperature trend for low-Cr and hybrid pyroxenes from the base of the SCLM up to 30 GPa (100 km) represents the PT path of the protokimberlite melts. PT conditions for ilmenites mainly correspond to colder conditions of crystallization in wall rocks and the outer parts of magmatic channels. Individual geochemical features of the minerals for each SCLM suggest pervasive metasomatism in lower part of the SCLM. Clinopyroxene trace element patterns from the Caquele pipe reveal a lherzolitic affinity; they are LILE-enriched with Ba peaks due to phlogopite melting, while those from Camitongo I–II show Ta–Nb enrichment and Pb troughs. The ilmenite trends trace the mantle column from deep to shallow mantle, evolving to Fe-ilmenites due to advanced AFC of protokimberlite magma that also produced abundant Fe-rich clinopyroxenes. The rise of calculated fO2 correlates with the position of protokimberlites. Comparison with the thermal gradient derived from peridotitic inclusions from Catoca cluster is lower than for Lesotho possibly related to the thicker lithospheric roots beneath the Congo–Kasai craton. © 2011 Elsevier B.V. All rights reserved.

1. Introduction African kimberlites and their deep-seated inclusions have been extensively investigated (Bell et al., 2003; Cartigny et al., 1999; Gibson et al., 2008; Nixon, 1973, 1987; Nixon et al., 1981; Rawlinson and Dawson, 1979; Stachel & Harris, 2008; Winterburn et al., 1990; Viljoen et al., 1992; Viljoen et al., 2009 and references therein). Petrology of the sub-cratonic lithospheric mantle (SCLM) beneath Africa has been reconstructed using xenoliths and xenocrysts (Aulbach et al., 2002; De Bruin, 2005; McCammon et al., 2001; Rogers and Grütter, 2009; Rawlinson and Dawson, 1979; Shee and Gurney, ⁎ Corresponding author at: Koptyug ave 3, Novosibirsk 630090, Russian Federation. Tel. + 7 383 333 14 14. E-mail address: [email protected] (I.V. Ashchepkov). 0040-1951/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.tecto.2011.12.007

1979; Viljoen et al., 1992; Viljoen et al., 2009). Combined thermobarometric and geochemical studies, taking into account the seismic studies of the African lithosphere, allow us to deduce not only the lateral and vertical variations of the SCLM but also to suggest the variations of the petrographic and petrochemical composition with depth in the Kaapvaal and some other African cratons (Griffin et al., 2003a,b; Griffin et al., 2004; Gurney et al., 1979a,b; O'Reilly et al., 2009). The evolution of protokimberlite magma has been studied using megacryst compositions (Batumike et al., 2009; Boyd et al., 1984), though it has been suggested that megacrysts also reflect the composition of the deep mantle (Bell and Moore, 2004; Boyd et al., 1984) and represent the deepest SCLM layers. However, their isotopic signatures suggest a more complex origin for megacrysts, probably crystallized from highly contaminated plume melts that have interacted with the SCLM (Davies et al., 2001). The existence of multiple

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

15ο0’E

Zambia 20

ο

0’E Mbuji-Mayi

N

* **

+++

Luanda ++++

*

* +

*+* ** * ** *+ * Caquele *+* * * * + ******Camitongo **** + * * I- II + Catoca * * ++ + ++

*

*

+ ++++ +++ ++ + + + + + ++ + + +

Namibe

*

++

**

Kimberlite pipes this work Kimberlite pipes Carbonatites

10ο0’S

Saurimo

+ +

+

Luena

Angola

* *

Zair

15ο0’S

+ ++

*

+* *

*** *

Namibia

Fig. 1. Scheme of the kimberlite pipes location in Angola (after Egorov et al., 2007).

129

megacryst compositional groups suggests a difference in source magmas (Dawson and Stephens, 1975) and polystage origin and possibly polybaric crystallization of protokimberlite melts. Many studies have been devoted to diamonds from African kimberlites (Boyd and Danchin, 1980; Deines et al., 1984; Kopylova et al., 1997; McDade and Harris, 1999; Tsai et al., 1979). Angola is the third most productive region for diamonds in Africa and the world. Nevertheless information about the large kimberlite fields in Angola is rather scarce (Boyd and Danchin, 1980; Egorov et al., 2007; Eley et al., 2008; Sobolev et al., 1990) compared with South Africa. Approximately 700 kimberlites have now been found in Angola. Kimberlite pipes Camafuca Camazambo, Camutue and Catoca (Ganga et al., 2003; Robles-Cruz et al., 2009; Zuev et al., 1988) are among the largest and least eroded in the world. They are of the great interest not only for mining geology but also for petrologists because they give the youngest and most representative information about the composition of the upper mantle beneath Central Africa (Leost et al., 2003; van Achterbergh et al., 2001) which has been studied less than in the Kaapvaal craton (Gregoire et al., 2003; Gregoire et al., 2003; Nixon and Boyd, 1973; Moore and Lock, 2001; Moore and Lock, 2001; van Achterbergh et al., 2001, etc.) and other parts of the African continent. The Congo–Kasai craton has one of the deepest SCLM roots in Africa extending to 400 km (O'Reilly et al., 2009). The kimberlite belt of Angola crosses the southern outer margin of the craton and the thicker SCLM is located

Table 1 Major and trace element composition of the minerals from the concentrate of Caquele pipe, Angola. Ilm SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO Na2O K2O NiO Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U Sc V Co Cu Ni Rb Sr Y Zr Nb Cs

46.18 0.55 0.44 45.49 0.2 6.95

0.21 100.02 148 2.17 6.37 1.28 6.86 2.00 0.59 1.97 0.27 1.37 0.22 0.48 0.06 0.29 0.03 1.04 0.24 1.58 0.52 0.012 17 154 23 40.2 263 0.49 99 5.4 7.1 1.18 0.01

Cr–Di 1

Cr–Di 2

Cr–Di 3

Gar1

Gar2

Gar3

Gar4

Gar5

Gar6

Gar7

55.01 0.32 3.31 3.08 3.24 0 14.3 16.78 3.89 0.03 0.05 100.01 2154 6.38 23.30 4.23 21.4 5.81 1.91 5.60 0.72 3.98 0.60 1.31 0.14 0.77 0.09 17.02 0.25 1.20 0.60 0.219 127 682 47 14.7 377 20.96 400 13.8 111.2 4.61 0.123

55.55 0.28 2.18 0.95 3.72 0 17.82 17.51 1.91 0.04 0.05 100.01 2.28 4.5 16.8 3.13 16.2 4.44 1.25 3.98 0.49 2.77 0.39 0.87 0.09 0.44 0.06 3.26 0.09 1.15 0.12 0.013 42 360 42 32.9 278 0.07 202 9.2 19.8 0.36 0.004

55.42 0.33 2.03 1.12 3.61

41.74 0.05 20.39 5.69 7.89 0.35 20.98 2.95 0.11 0.05

40.49 0.25 12.98 13.31 6.99 0.24 18.5 6.46 0.11 0.05

41.33 0.05 19.4 5.49 8.97 0.38 18.27 5.37 0.11 0.05

41.62 0.16 18.24 6.34 8.99

41.63 1.01 15.85 7.68 7.98

42.43 0.68 20.05 2.78 7.28

100.2 0.67 0.04 0.29 0.11 1.18 0.87 0.46 2.01 0.41 3.28 0.72 2.00 0.30 2.10 0.33 1.43 0.02 0.31 0.03 0.009 76 156 31 4.6 78 0.04 0.42 18.5 25.8 0.17 0.003

99.38 0.35 0.05 0.51 0.18 1.77 1.44 0.62 2.80 0.54 4.34 0.88 2.49 0.38 2.45 0.36 2.61 0.01 0.38 0.01 0.007 68 162 26 5.2 81 0.04 0.54 23.2 42.4 0.10 0.0007

99.42 1.83 0.046 0.33 0.11 1.36 0.88 0.40 2.03 0.41 3.47 0.73 2.28 0.34 2.20 0.34 1.77 0.03 0.38 0.02 0.014 65 145 29 7.3 75 0.05 0.55 19.2 27.9 0.19 0.004

18.02 5.79 0.05 0 0.02 99.23 0.56 0.045 0.28 0.10 2.30 0.63 0.20 0.70 0.09 0.57 0.11 0.37 0.07 0.55 0.12 0.62 0.04 0.59 0.05 0.073 104 182 21 9.5 14 0.12 0.25 3.0 10.3 0.08 0.002

17.78 7.14 0.10 0 0.05 99.22 7.28 0.072 0.32 0.10 1.00 0.72 0.36 1.60 0.34 2.79 0.61 1.92 0.30 2.01 0.32 1.29 0.03 0.26 0.021 0.020 69 149 30 3.6 78 0.32 0.73 15.2 22.1 0.22 0.009

21.25 4.59 0.09 0.01 0.02 99.18 1.07 0.132 0.39 0.14 1.37 1.21 0.57 2.92 0.63 5.21 1.18 3.69 0.58 3.73 0.62 2.58 0.04 0.70 0.018 0.010 94 199 37 6.4 102 0.06 0.67 29.9 40.6 0.25 0.005

42.4 0.97 17.78 6.74 7.49 0.24 20.11 5.06 0.11 0.05

18.13 18.21 2 0.03 0.08 100.96 302 0.53 1.13 0.21 1.47 0.70 0.22 0.73 0.12 0.83 0.14 0.41 0.06 0.48 0.10 0.23 0.08 0.55 0.14 0.356 104 133 24 23.8 78 1.54 8.38 3.9 5.8 0.78 0.083

99.17 1.03 0.068 0.52 0.18 1.70 1.56 0.67 3.29 0.75 6.35 1.41 4.59 0.70 4.81 0.78 3.43 0.05 0.44 0.03 0.015 97 204 41 13.0 96 0.06 1.46 36.1 52.5 0.52 0.004

Zirc1

Zirc2

SiO2

31.92

32.01

ZrO2

66.04

66.09

CaO

0.016

0.038

P2O5

1.29 99.266 26.3 0.081 0.84 0.07 0.51 1.15 0.51 2.80 0.74 8.26 2.16 8.22 1.56 13.9 1.89 1926 2.33 1.22 4.5 12.7 177 0.2 0.02 5.07 0.44 0.221 0.6 54.5 35,540 2.36 0.007

1.02 99.158 55.0 0.343 0.69 0.07 0.26 0.737 0.191 1.11 0.36 4.25 1.31 5.52 1.03 7.4 1.39 635 1.75 2.04 3.59 8.90 80.2 1.05 0.24 23.12 14.44 0.277 2.7 35.8 8717 2.35 0.011

Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U Sc V Co Cu Ni Rb Sr Y Zr Nb Cs

130

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

northward (Fishwick, 2010; O'Reilly et al., 2009). Such deep SCLM roots like the Slave (Aulbach et al., 2007) or Siberian craton usually reveal a colder thermal regime (Pedersen et al., 2009) within the SCLM. In this paper we present petrological information obtained from mineral concentrates from the dense kimberlite cluster located in Lunda Sul province near the Catoca pipe, which is the fourth largest diamond-rich diatreme in the World. The nearby pipes Caquele (Kakele) and Camitongo I–II also contain diamonds (Zinchenko, 2008). Reconstructions of the PT conditions and mantle sequence beneath these pipes will be compared with those obtained by monomineral thermobarometry (Ashchepkov et al., 2010) for other kimberlite provinces in Africa (Lesotho, Botswana, Namibia etc.). The geochemical features of the pyrope garnets (Gar), clinopyroxenes (Cpx) and ilmenites (Ilm) will also be compared with those obtained from South Africa (Gregoire et al., 2003) and the DR Congo (Batumike et al., 2009; Pivin et al., 2009).

2. Location Angolan kimberlites (Egorov et al., 2007; Ganga et al., 2003; Pettit, 2009; Reis, 1972; Roman'ko et al., 2005; Sobolev et al., 1990) are distributed across the vast Congo–Kasai craton. Most of them are

located in a NE–SW trending belt (Fig. 1). Some kimberlite fields form separate groups scattered in the western part of Angola, 50–150 km from the coast. In this paper we give new information for xenocrysts from kimberlites from the Lunda Sul province kimberlite cluster which includes many pipes. We describe here material from Camitongo I and II pipes (8° 56′S, 19° 7′E) and Caquele (Kakele) pipe (8.28 S, 20.23 E), located 12 km south of Calonda where the Catoca kimberlite pipe is located (9°24 S, 20° 18′E). The Catoca cluster kimberlite are practically uneroded (Pervov et al., 2011). The age of zircon from Catoca is 118 Ma (Robles-Cruz et al., 2009) which is the youngest in the NW part of the kimberlite belt (Rogers and Grütter, 2009).

3. Samples All the data were obtained from mineral concentrates — both naturally derived from the laterite soil and also from drilling samples. We used mainly garnets, ilmenites and clinopyroxenes to establish the variation of geochemistry and major elements. The grains used had typical dimensions of 0.5–1 mm and some larger grains and intergrowths were considered as microxenoliths to determine the mineral associations. The grains were mostly taken from the sandy tuff kimberlites. All of them are rounded, possibly during eruption in the hot explosion cloud (Tables 1, 2, and 3).

Table 2 Major and trace element composition of the minerals from the concentrate of Camitongo I pipe, Angola. Mineral

Gar1

Gar2

Gar3

Gar4

Cr–Di

Cr–Di

Cr–Di

Cr–Di

Cr–Di

Cr–Di

SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO Na2O K2O NiO

41.09 0.23 19.91 4.19 6.86 0.29 20.66 4.95 0.05 0 0.09 98.32 0.02 0.012 0.084 0.034 0.31 0.31 0.18 0.753 0.126 1.042 0.232 0.71 0.112 0.701 0.125 0.162 0.005 0.012 0.004 0.007 32.4 52.0 12.7 0.2 10.7 0.057 0.081 6.19 5.2 0.036 0.001

40.64 0.16 18.59 6.02 7.72 0.33 20.13 4.6 0.06 0 0.08 98.33 0.02 0.004 0.059 0.026 0.29 0.25 0.13 0.47 0.098 0.653 0.137 0.379 0.069 0.505 0.082 0.149 0.008 0.023 0.001 0.003 29.7 67.9 12.5 0.5 17.3 0.053 0.106 3.99 3.8 0.043 0.001

41.43 0.06 19.74 4.69 8.82 0.38 18.96 4.55 0.06 0.01 0.06 98.76 0.03 0.004 0.047 0.019 0.18 0.16 0.08 0.361 0.068 0.601 0.142 0.429 0.069 0.474 0.066 0.25 0.005 0.025 0.001 0.002 26.0 56.8 11.1 0.3 17.0 0.099 0.056 3.54 5.0 0.033 0.002

40.97 0.05 18.83 5.96 8.18 0.35 18.76 5.73 0.03 0 0.07 98.94 0.19 0.046 0.085 0.032 0.34 0.25 0.14 0.56 0.11 0.81 0.14 0.38 0.053 0.347 0.049 0.393 0.006 0.011 0.002 0.003 31.8 70.8 10.3 0.4 19.8 0.022 0.122 4.131 7.8 0.052 0.001

53.78 0.1 2.34 2.58 2.55

53.52 0.07 3.16 3.14 2.41

54.32 0.22 2.91 1.24 2.77

55.25 0.2 2.95 1.37 2.63

54.97 0.18 1.64 2.9 2.13

54.75 0.13 0.86 2.31 3.04

15.79 17.49 3.43 0.01 0.1 98.17 0.20 1.05 3.56 0.72 3.6 1.05 0.32 0.89 0.12 0.729 0.102 0.23 0.032 0.153 0.016 0.28 0.004 0.084 0.01 0.002 9.1 109 19.4 2.4 250 0.017 66.3 2.56 2.0 0.079 0.0007

15.61 16.71 3.81 0.01 0.13 98.57 0.18 4.59 12.5 2.18 10.6 3.11 0.82 1.83 0.11 0.39 0.043 0.072 0.007 0.03 0.00 2.77 0.03 0.459 0.144 0.039 44.5 425 11.9 0.3 188 0.033 185 1.03 24.3 0.214 0.0001

15.66 17.69 3.23 0.02 0.12 98.18 4.19 3.55 11.1 2.23 11.8 3.20 1.00 3.17 0.40 2.32 0.36 0.74 0.087 0.43 0.05 6.91 0.084 0.239 0.184 0.029 56.4 446 24.8 1.1 352 0.58 185 8.14 47.1 1.212 0.009

15.61 18.54 2.95 0 0.1 99.6 1.32 1.79 5.73 1.06 4.9 1.25 0.39 1.18 0.15 0.80 0.156 0.33 0.041 0.19 0.03 0.38 0.016 0.142 0.026 0.008 14.2 134 27.6 3.5 381 0.059 92.24 3.30 3.4 0.591 0.003

15.27 19.36 2.41 0.01 0.15 99.02 0.72 29.0 81.2 12.08 52.6 9.44 2.59 7.08 0.70 3.34 0.43 0.73 0.072 0.29 0.03 3.7 0.203 1.409 0.453 0.056 57.2 554 23.4 0.4 388 0.106 734 9.58 68.5 1.312 0.004

15.05 20.34 1.93 0.03 0.11 98.55 0.46 14.9 43.0 5.87 23.2 3.18 0.80 2.02 0.22 1.17 0.17 0.32 0.034 0.15 0.02 3.7 0.012 0.981 0.329 0.094 40.2 623 30.6 1.1 180 0.081 458 3.48 33.3 0.337 0.004

Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U Sc V Co Cu Ni Rb Sr Y Zr Nb Cs

Ilm

Ilm

Ilm

50.99 0.32 1 36.79 0.4 9.3

52.96 0.23 0.11 36.74 1.01 8.18

49.31 0.39 1.16 36.56 1.39 10.16

0.09 98.89 0.02 0.002 0.004 0.006 0.013 0.002 0.0007 0.003 0.0008 0.003 0.0009 0.002 0.001 0.003 0.001 8.8 69.3 0.011 0.0004 0.028 7.10 432 19.7 2.6 15.4 0.067 0.023 0.003 115 534.2 0.002

0.09 99.31 0.97 0.006 0.006 0.0006 0.01 0.008 0.002 0.008 0.001 0.006 0.001 0.004 0.002 0.004 0.001 5.9 61.6 0.14 0.014 0.02 6.17 303 17.7 3.1 19.1 0.074 0.144 0.01 90.2 498.6 0.002

0.08 99.06 0.18 0.016 0.027 0.002 0.009 0.004 0.002 0.006 0.0006 0.003 0.0007 0.003 0.0004 0.006 0.002 4.5 27.8 0.011 0.0001 0.013 3.07 214 12.8 2.3 14.8 0.119 0.023 0.005 75.1 267.1 0.004

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

131

Table 3 Major and trace element composition of the minerals from the concentrate of Camitongo II pipe Angola.

SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO Na2O K2O NiO Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U Sc V Co Cu Ni Rb Sr Y Zr Nb Cs

Gar1

Gar2

Gar3

Gar4

Gar5

Gar6

Gar7

Gar8

Cr–Di1

Cr–Di2

Cr–Di3

Cr–Di4

Ilm1

Ilm2

Ilm3

Zirc

Zirc

41.98 0.91 17.22 7.11 6.65 0.34 20.45 5.09 0.13 0.01

42.58 0.07 21.27 3.18 8.28 0.5 19.65 4.67 0.05 0

42.77 0.01 21.11 6.16 6.67 0.31 21.1 3.47 0.03 0.02

41.72 0.5 18.02 7.23 6.76 0.35 19.91 5.4 0.09 0

42.17 0.76 21.49 1.49 8.87 0.27 21.3 4.21 0.12 0

42.06 0.07 19.34 6.01 6.87 0.36 20.1 5.57 0.03 0

41.98 0.31 20.47 3.34 7.1 0.28 21.37 4.66 0.07 0

42.71 0.07 19.03 6.07 6.99 0.4 20.36 5.62 0.03 0.01

54.39 0.47 3.17 0.27 5.93 0.16 16.22 15.94 2.42 0.03

54.36 0.29 2.5 2.52 2.75 0.11 14.41 19.03 2.75 0.01

54.32 0.28 2.16 3.81 2.43 0.1 14.37 17.73 3.07 0.04

54.31 0.53 3.22 1.01 3.8 0.12 17.13 15.5 2.55 0.02

43.74 0.07 3.99 43.91 0.32 6.44

45.84 0.74 0.4 42.85 0.21 8.93

46.98 0.15 1.71 40 0.5 11.37

32.91

32.85

65.12

65.24

0.011

0.017

99.89 1.22 0.28 1.49 0.11 0.64 1.0 0.4 2.4 0.67 7.48 1.96 6.94 1.24 9.97 1.25 3144 2.88 0.77 7.95 14.5 286 0

99.25 0.05 0.004 0.03 0.01 0.21 0.7 0.2 0.4 0.06 0.51 0.13 0.39 0.06 0.45 0.08 0.23 0.00 0.03 0.00 0.004 35 52 11 0.13 7.3 0.0 0.06 3.02 3.79 0.25 0.002

101.65 0.10 0.008 0.08 0.03 0.302 0.6 0.2 0.7 0.17 1.53 0.39 1.26 0.18 1.02 0.14 0.65 0.01 0.05 0.01 0.011 27 65 11 0.50 23.1 0.0 0.15 9.84 11.04 0.04 0.003

99.98 16.90 25.3 68.43 11.13 50.8 11.2 2.9 8.6 1.00 5.51 0.82 2.02 0.25 1.44 0.20 5.41 0.40 0.99 0.48 0.069 107 470 16 0.41 272 0.4 688 21.86 97.69 2.56 0.005

101.68 0.00 0.003 0.01 0.01 0.127 0.3 0.1 0.5 0.09 0.67 0.17 0.56 0.11 0.77 0.14 0.25 0.00 0.02 0.00 0.002 29 58 12 0.08 3.8 0.04 0.04 4.55 3.42 0.03 0.002

100.41 0.01 0.004 0.03 0.01 0.19 0.3 0.1 0.5 0.08 0.62 0.13 0.47 0.08 0.63 0.12 0.28 0.00 0.02 0.00 0.004 47 77 13 0.17 5.6 0.08 0.05 3.99 5.61 0.03 0.0004

100.58 0.11 0.0007 0.00 0.00 0.002 0.1 0.0 0.0 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.03 0.00 0.01

101.29 1.77 0.091 0.18 0.04 0.375 0.4 0.1 0.7 0.15 1.40 0.34 1.08 0.19 1.26 0.22 0.35 0.01 0.03 0.01 0.014 55 83 15 0.22 7.5 0.15 0.76 8.90 8.05 0.10 0.002

99 0.37 22.5 63.25 10.42 50.14 9.7 2.7 7.4 0.93 5.33 0.82 2.13 0.27 1.79 0.26 3.13 0.43 0.77 0.22 0.028 101 415 14 0.41 260 0.13 708 22.17 67.29 1.52 0.002

98.73 0.11 2.212 8.51 1.95 11.53 3.9 1.2 3.7 0.47 2.44 0.33 0.73 0.08 0.39 0.04 4.59 0.18 0.20 0.03 0.011 42 396 28 1.51 116 0.05 208 8.19 89.2 0.79 0.0006

98.31 0.34 2.371 7.62 1.32 6.588 1.7 0.5 1.4 0.17 0.82 0.11 0.22 0.03 0.13 0.02 0.57 0.01 0.22 0.02 0.003 19 189 23 2.63 398 0.03 132 2.77 4.9 0.15 0.073

98.19 30.53 26.84 79.76 13.79 66 13.7 3.7 10.3 1.18 6.11 0.83 1.77 0.21 1.07 0.14 10.09 0.11 1.15 0.59 0.13 132 549 15 0.99 212.2 0.28 903 20.16 79.5 1.72 0.003

1.18 99.221 14.2 0.5 17.8 0.36 2.43 1.31 0.45 1.88 0.37 3.05 0.85 2.82 0.52 4.86 0.63 2798 0.74 0.20 1.32 3.89 2730 3.51

1.25 99.357 0.04 0.006 0.84 0.04 0.45 1.27 0.50 2.61 0.71 7.58 2.13 8.80 1.609 15.19 2.20 2294 1.59 0.82 6.09 16.58 302. 0.11

1.37 1.33 0.253 4.8 23.3 69,571 0.90 0.002

0.277 1.156 0.328 0.1 57.2 71,444 1.28 0.078

0.44 1.4 0.0 0.49 51.27 72,803 3.76 0.004

0.0002 0 0 0 0.01 0.5 0.04 0.02 0.00 0.00 0.002

4. Analytical methods The microprobe analyses were produced using the CamebaxMicro in IGM SB RAS, Novosibirsk. All analyses were made in wavelengthdispersive (WDS) mode using a variety of natural reference minerals and synthetic glasses for calibration (Lavrentyev et al., 1987). The accelerating voltage was 15 kV and a focused beam of 15 or 20 nA was used. Reduction procedure using “Karat program” (Lavrent'ev and Usova, 1994) was applied to the analyses. The relative standard deviation does not exceed 1.5%; the precision was close to 0.02– 0.01% for minor elements. About 150–300 grains of each type of mineral were taken from each of the 4 pipes. After estimations of the variations for major components, the representative varieties were analyzed by LAM-ICP method in Analytical Center of UGM SD RASc using the LAM ICP MS (Finnigan Element with YAG Nd 266 LaserProbe laser system) (isotope analyst S.V. Palessky) by rastering the beam over of the surface of the polished grains to exclude fractionation and increase the intensity during laser ablation. Fluctuation of the plasma was monitored by 129Xe from Ar gas, the concentration 40Ca for garnets and clinopyroxenes, and Ni and V for ilmenites, were used for the comparison with the EPMA data and normalization. The laser spot diameter did not exceed 10–20 μm with the scanning of the surface — 30 scans for samples. The concentration of 33 trace elements were determined using 2–3 isotopes for the most complicated elements (48 isotopes together) and compared to chose the best values with the lower relative standard

0.08 98.55 0.02 0.015 0.05 0.00 0.021 0.024 0.005 0.008 0.001 0.006 0.002 0.004 0.002 0.006 0.0008 2.85 26.16 0.013 0.0007 0.01 4.98 185 25.9 6.3 96 0.008 0.058 0.001 35.5 180 0.0004

0.08 99.05 0.07 0.010 0.008 0.005 0.011 0.011 0.003 0.008 0.002 0.004 0.0003 0.002 0.0005 0.011 0.002 2.53 87.8 0.034 0.002 0.009 5.00 235 24.9 3.8 101 0.001 0.033 0.0004 36.8 649 0.001

0.08 100.79 0.28 0.13 0.14 0.01 0.052 0.019 0.0003 0.012 0.002 0.005 0.001 0.005 0.0008 0.003 0.001 6.58 100.2 0.023 0.0005 0.034 8.58 184 13.2 1.8 15 0.003 0.116 0.007 97.9 802 0.003

deviations and producing smoother REE patterns. Scanning time for each grain was close to 3 min. The compositions of the minerals from concentrates are given in the Supplementary data. 5. Major element chemistry of the minerals 5.1. Garnets Pyropes from three kimberlite pipes from the Catoca cluster have similar compositions. Several garnets from Camitongo I–II pipes are richer in Cr compared with published results (Egorov et al., 2007) from other fields. Most of the compositional variation is concentrated within the lherzolite field in the Cr2O3–CaO discriminant diagram (Grütter et al., 2004; Sobolev et al., 1973; Sobolev, 1980) reaching 14 wt.% Cr2O3 for those from Caquele and Camitongo kimberlite pipes (Fig. 2) which is higher than the Cr content of garnets (Gar) from Chicuatite, Chikolongo, Palue, Viniaty and Ochinjau pipes located within the NE zone that controls the location of most Angolan kimberlites (Egorov et al., 2007). Deviations to the harzburgite field (Burgess and Harte, 1999; Griffin et al., 1999a,b; Griffin et al., 2002; Grütter et al., 2004; Schulze, 2003; Sobolev et al., 1973) were found only starting from 12% wt.% Cr2O3 and several data points of sub-calcic garnets were found within the 5–7 wt.% Cr2O3 interval (Tables 1–3). For the Caquele (Kakele) kimberlite pipe, the major trend is restricted by 10 wt.% Cr2O3 which is close to values found for garnets from

132

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 12

12

Camitongo II

Camitongo I CaO

CaO

8

8

4

4

0

0

0

2

4

6

8

10

12

14

0

2

4

Cr2O3

6

8

10

12

14

12

14

Cr2O3 12

12

CaO

CaO

Catoca

10

Caquele

8

8 6

Kamitongo2

4

4

2 0

0 0

2

4

6

8

10

12

14

0

2

Cr2O3

4

6

8

10

Cr2O3

Fig.2. Discrimination diagram for garnets from Angolan kimberlite pipes Caquile, Camitingo 1, Camitingo II, Catoca (Sobolev et al., 1973).

Catoca (Sobolev et al., 1990). The Catoca pipe (Correia and Laiginhas, 2006) contains a large amount of harzburgitic and dunitic garnets, constituting about half of the whole population. Na-bearing eclogitic garnets (Grütter and Quadling, 1999; McCandless and Gurney, 1989; Sobolev and Lavrent'ev, 1971; Sobolev et al., 1998) are more typical for Camitongo I pipe than for Camitongo II and were not found in the studied separates from the Caquele pipe 5.2. Clinopyroxenes The compositions of Cr-diopsides in the concentrates of the three studied pipes are similar and nearly overlap in the variation diagram (Fig. 3). However the amount of the different diopside groups in each pipe differs greatly. Three major groups of clinopyroxenes (Cpx) include Cr-diopsides, low-Cr, Fe-, Al-rich Cpx which are typical of megacrystalline suites in kimberlites (Gurney et al., 1979a; Kopylova et al., 2009; Moore and Belousova, 2005; Rogers and Grütter, 2009 e.g.) and eclogitic omphacites (Beard et al., 1996; Boyd and Danchin, 1980; McCandless and Gurney, 1989; Viljoen et al., 2005). Most of the typical Cr-diopsides with 2–3 wt.% FeO correspond to compositions found in the middle part of the lithospheric mantle sections worldwide, with 1–3.5 wt.% Na2O and b4 wt.% Cr2O3 (Boyd et al., 2004; Sobolev, 1977). They are relatively low in Al2O3 (0.5 to 4 wt.%). Typical Cr-diopsides in the 2–3.7 wt.% FeO range show increasing Na2O, FeO and TiO2, which is common for hydrous metasomatism. As the trends continue, the TiO2 concentrations rise with the FeO content and Cr2O3 decreases from 5 wt.% but the most Fe-rich compositions are divided from the main trend by a break. Fe-rich varieties of clinopyroxenes may be subdivided into three groups according to Al2O3 content. A minority of the low-Ti compositions with high sodium contents are typical of eclogites, plotting in

the Groups B–C field in MgO–Na2O diagram. Some of them (C) are typical omphacites with 7–4 wt.% Na2O and 7–12 wt.% Al2O3. There many transitional Cr-bearing varieties with high FeO (~5 wt.%) and Al2O3 content which correspond to metasomatites. The Camitongo II pipe contains fewer Cr-rich garnets and a high abundance of Fe-rich pyroxenes. Cr-rich varieties reveal metasomatic features. In Camitongo I pipe the relatively scarce eclogitic garnets correspond to the rarity of typical Na-, Al-rich omphacites (Dawson and Stephens, 1975; Morimoto et al., 1988; Sobolev, 1977). The most Fe-rich (to 8 wt.% FeO) and low-Cr pyroxenes with high TiO2 contents are close to pigeonite compositions but with relatively high Na2O contents (2–4 wt.%) common for many kimberlites worldwide. The CaO component for low-Cr clinopyroxenes increases together with FeO, which is typical for fractionation and for megacryst associations (Ashchepkov et al., 2011; Moore and Belousova, 2005). The Ca-rich, lower-Cr compositions are close to those of the garnet-bearing megacrystalline pyroxenites which contain ilmenites or are in intergrowths with megacrystalline ilmenites (Kopylova et al., 2009; Rodionov et al., 1988). In contrast the concentrate from Camitongo II pipe contains abundant eclogitic pyroxenes as well as those from pyroxenites and omphacites. The concentrate from Caquele pipe also contains abundant eclogitic omphacites, pigeonitic to augitic pyroxene, Cr-metasomatites and peridotitic Cr-diopsides. Typical peridotitic orthopyroxenes and rare richterite were found in the concentrate from Camitongo I pipe. 5.3. Ilmenites Picroilmenites from the three pipes have highly variable compositions overlapping only partly in the high-MgO field (Fig. 4). In the variation diagram TiO2–MgO–FeO they display long trends typical

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 10

Caquele

10

clo

pB s ou gite r G lo ec

6

4

pC

ou Gr

8

p ou

A

ec

4

A s p ou gite r G clo e

2

8

12

14

16

18

4

20

MgO %

8

4

2

4

12

14

16

18

B p s ou gite r G lo ec

6

A p s ou gite r G clo e

4

20

4

MgO %

4

4

6

8

6

10

12

14

16

18

20

MgO %

12

8

4

0.6

0.4

TiO2 %

TiO2 %

0.4

2

4

6

0.2

2

4

6

2

4

6

0.4

0.2

0 2

4

6

Na2O %

4

2

2

4

0 2

6

6

4

6

6

6

4

4

Al2O3 %

TiO2 %

10

2

0

Na2O %

e

ou Gr

0.6

0.2

2

4

2

4

6

Cr2O3 %

2

3

2 1

1

2

4

FeO %

6

2 1

0

0

2

4

3

3

Cr2O3 %

8

8

6

0.6

4

6

12

Al2O3 %

Al2O3 %

10

pC

g clo

2

Na2O %

6

Camitongo I

s

ite

8

Cr2O3 %

4

10

pB s ou gite r G lo ec

6

Gr

12

Camitongo II

s

te

gi

lo

2

Na2O %

Na2O %

G

e pC rou

lo ec

Na2O %

8

es

git

es git

133

0 2

4

6

2

FeO %

4

6

FeO %

Fig. 3. Variation diagram for clinopyroxenes from Angolan kimberlite pipes.

for kimberlites (Wyatt et al., 2004). The trend of ilmenites (Ilm) from the Caquele pipe is close to that described for the Catoca kimberlite (Robles-Cruz et al., 2009). Ilmenites from Catoca (Robles-Cruz et al., 2009) reveal four different groups including high-Mg types common for kimberlites and diamond-bearing associations (Pokhilenko et al., 1976), Fe-rich ilmenites (enriched in hematite), Mn-ilmenites and low-Mg ilmenites (Schulze et al., 1995) as is common for other regions. Some Mg-rich ilmenites were porous and surrounded by Fe-rich varieties. Such intergrowths were not detected in the studied kimberlites, although the compositional varieties of ilmenites from Catoca and Caquele pipe are close. Variations of MgO (6–14 wt.% MgO) in ilmenites from Caquele and Camitongo I–II pipes are typical for cratonic kimberlites (Schulze et al., 1995; Wyatt et al., 2004) (Fig. 3). The whole population of ilmenites can be divided into: a) an Mg-rich array similar to

that found in the Premier pipe (Gaspar and Wyllie, 1984), b) an intermediate one (8–10 wt.% MgO) typical for the Camitongo I–II pipes and c) a separate group of more Fe-rich ilmenites containing essential admixtures of hematite and magnetite molecules which give them magnetic properties. The latter group is typical only for the Caquele pipe. Ilmenites of this group is found Catoca pipe (Robles-Cruz et al., 2009). They occur rarely among ilmenites from Camitongo I–II pipe where lower Fe concentrations are characteristic. The trends of the ilmenites from Caquele and the Camitongo I and II are complementary and together they form a long trend similar to ilmenites from Southern African pipes like Monastery and others (Gurney et al., 1979a; Moore et al., 1992). The greatest differences for ilmenites are in Cr2O3 concentrations which are higher for the metasomatites (Ashchepkov et al., 2010; Ashchepkov et al., 2011). Such concentrations were found for the

134

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

Caquele

Camitongo I

Ti O2 %

60.0

0 %Fe2O3

50.0

50.0

20 %Fe2O3

20 %Fe2O3

20 %Fe2O3

40.0

40.0

40.0

60 %Fe2O3

0.8

40 %Fe2O3

40 %Fe2O3

40 %Fe2O3

0

4

MnO %

80 %Fe 2 O3 100 %Fe O 2 3

8

12

16

20

MgO %

30.0 1.0

60 %Fe2O3

0

0.8

4

MnO %

80 %Fe 2 O3 100 %Fe O 2 3

8

12

16

20

MgO %

30.0 1.0

0.6

0.6

0.6

0.4

0.4

0.2

0.2

0.2

0.0 70

0.0 70

35

40

45

50

55

FeO %

60

35

40

45

50

55

FeO %

50

50

40

40

30

30

30

20

20 45

50

1.2

50

0.8

0.4

0.4

0.4

35

40

45

50

55

TiO2 %

0.0 0.430

16

20

45

50

55

45

50

55

35

45

50

55

50

55

Al2O3 %

1.2

0.8

V 2O5 %

40

55

Al2O3 %

0.8

0.0 0.6 30

35

12

MgO %

20 45

55 1.2

80 %Fe2 O3 100 %Fe O 2 3

8

FeO %

60

50

4

MnO %

0.0 70

40

Al2O3 %

60 %Fe2O3

0

0.8

0.4

60

Ti O2 %

60.0

0 %Fe2O3

0 %Fe2O3

50.0

30.0 1.0

Camitongo II

Ti O2 %

60.0

40

V 2O5 %

45

50

55

0.0 0.3 30

35

40

V 2O5 %

TiO2 %

0.3

TiO2 %

0.2

0.4 0.2

0.1

0.2

0.1 0.0

0.0

0.0 35

40

45

50

55

NiO %

0.2

35 0.2

40

45

50

0.0 16

35

40

45

50

55

0.0 16

35

40

45

50

55

0.0 16

12

12

8

8

8

4

4

4

35

Cr2O3 %

40

45

50

55

0 6.0

35

40

45

50

55

0 6.0

4.0

4.0

2.0

2.0

2.0

0.0 35

40

TiO2 %

45

50

55

0.0 30

40

45

50

55

35

40

45

50

55

40

45

50

55

Cr2O3 %

Cr2O3 %

4.0

30

35

MgO % 12

0 6.0

45

0.1

MgO %

MgO %

40

NiO %

0.2

0.1

0.1

35

55

NiO %

0.0 35

40

45

50

55

30

35

TiO2 %

TiO2 %

Fig. 4. Variation diagram for picroilmenites from Angolan kimberlite pipes.

Camitongo II pipes in the Ti-Mg-rich part of the trend and for ilmenites from Caquele pipe where they show a bi-modal distribution with peaks near 55 and 40 wt.% TiO2. Several orthopyroxenes from the concentrate show low Fe# ~ 7, typical of those derived from the depleted mantle, and one relatively Fe-rich is close in Fe#16 to the low-Cr Cpx. A single Cr-rich (1.63 wt.% Cr2O3) richterite, with 5.14 wt.% Na2O and 2.26 wt.% K2O, was also found among the minerals from this pipe. Phlogopites are frequent in the concentrates from all the pipes.

6. Thermobarometry The PT conditions for the minerals from the concentrates were determined from clinopyroxenes, according to recently published barometers (Ashchepkov et al., 2010; 2011) and, for comparison, the Nimis and Taylor (2000) barometer which is applicable mainly to low-Fe Cr-diopsides. Temperatures were calculated with the Nimis and Taylor (2000) thermometer. Monomineral barometers for garnets and ilmenites (Ashchepkov et al., 2010) in combination

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

135

0

0

Lesotho

1

T oC

1 Sp

2 3 4

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1.Opx 2.OPXDia 3.CpxNiTa00 4.CpxAsh10 5.CpxAsh10Ecl 6.CpxAsh10Dia 7.CpxAsh10PXt 8.GarAsh10 9.Gar Ash10 Dia 10.ChrAsh10 11. Chr Ash10Dia 12. Ilm Ashch10 13. BrKo90

2

Gr SEA

3

Grap

hite Diam on d

4 45 mw/m2

5

5 6

6

1

P(Gpa)

600

800

1000

40 mw/m2

1200

2

3

3

4

4

5

5

6

6 7 P(Gpa)

8

8

8 0.05

1400

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

2

P(Gpa)

35 mw/m2

8

-ΔLogFO2

7

7

7

Variationa of Cpx, Opx, Gar, Chr, Ilm

0.10

0.15

0.20

0.25

0.30

0.35 0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 5. PTXF plot: PT–X(Fe#, CaO, Cr2O3, Al2O3)–fO2 (PTXF) plots using xenoliths data set for Lesotho (Nixon and Boyd, 1973; Smith, 1999) pipes obtained using monomineral thermobarometry.orthopyroxene: 1. T°C (Brey and Kohler, 1990, Opx)(12) -P (GPa) (McGregor, 1974)(28). 2. the same for the diamond inclusions; clinopyroxene: 3. P (GPa) and T°C (Nimis and Taylor, 2000). 4. T°C (Nimis and Taylor, 2000) corrected-P(GPa)-(Ashchepkov et al., 2010); 5. The same for pyroxenites; 6. The same for eclogites, for temperatures (Krogh, 1988); 7. The same for diamond-bearing eclogites. Garnet: 8. T°C (O'Neill and Wood, 1979, monomineral)-P (GPa) (Ashchepkov et al., 2010). 9. The same for diamond inclusions; chromite: 10.T°C (O'Neill and Wall, 1987), monomineral-P (GPa) (Ashchepkov et al., 2010); 11. the same for diamond inclusions; 12. ilmenite: T°C (Taylor et al., 1998)P(GPa) (Ashchepkov et al., 2010-Ilm); 13. T°C and P (GPa) (Brey and Kohler, 1990, Opx).

with the single grain versions of O'Neill and Wood (1979) and Taylor et al. (1998) thermometers were also used. These methods reproduce in general the combined PT diagram for SCLM beneath Lesotho but yield some higher temperature points compared to previous diagrams based on pyroxenes (Nixon and Boyd, 1973) (Fig. 5) and add details to the PT plots for individual pipes. They also replicate the PT plots obtained with the Brey and Kohler (1990) and orthopyroxene estimates (McGregor, 1974) for the best known localities of mantle xenoliths in Africa (Bell et al., 2003) (see Discussion). We also calculated oxygen fugacity conditions for ilmenites and rare chromites using the monomineral version of the Taylor et al. (1998) oxygen barometers with the calculation of Fe#Ol according to the previous work (Ashchepkov et al., 2010). The monomineral version of the Gar–Ol–Opx method (Gudmundsson and Wood, 1995) obtained with regression and the new Cpx method constructed by the cross correlations of the Fe 3 + in Cpx with the oxygen fugacity values obtained for garnets were used for the additional characterization of the mantle SCLM section. The configurations of the obtained geotherms are specific for each studied Angolan pipe (Figs. 6–9). Clinopyroxene geotherms for SCLM beneath pipes from the Catoca cluster show a typical inflection starting at 5.5–6.0 GPa (180–200 km). The regular break at 6.0 GPa (200 km) and 1200 °C as determined by Nixon and Boyd (1973) for Lesotho (Fig. 5) was not detected because the dispersion in PT estimates is rather high. PT estimates for garnet and low-Fe clinopyroxenes produce colder geotherms for the SCLM in the western part of Angola in the southeastern part of the Congo–Kasai craton compared to those for South

Africa (Bell et al., 2003) and Namibia (Franz et al., 1996; 1997; Leost et al., 2003). Unlike the previous versions of garnet thermobarometry (Griffin and Ryan, 1996; Griffin et al., 1989; Ryan et al., 1996), our equation (Ashchepkov, 2006; Ashchepkov et al., 2010) gives pressures up to 8.0 GPa (270 km). Relatively cold geotherms with gradients higher than conductive (Pollack and Chapman, 1977) were obtained both with garnets and clinopyroxenes and even for some metasomatic Cr-rich ilmenites (Figs.6–10). Mantle xenocrysts from Caquele (Fig. 6) and Camitongo I–II generally give PT conditions along the 35–38 mW/m 2 geotherm (Figs. 7–8). The garnet geotherm for the SCLM beneath the Catoca pipe (Fig. 9) is very similar. These thermal conditions are similar or slightly higher than those determined for the Siberian craton (38 mW/m 2) (Boyd et al., 1997) and the Slave craton (Aulbach et al., 2007; Russell and Kopylova, 1999) etc. The lower temperature conditions were determined mainly for Cr-rich pyroxenes and slightly higher for garnets. The PT points for several orthopyroxenes according to the Brey and Kohler (1990) Opx thermometer and McGregor (1974) barometer give the same PT gradient as garnets and Cr-diopsides. The pronounced inflection of the geotherm for these pipes corresponds to depths of 6.5– 7.5 GPa (210–245 km) which are deeper than those determined for SCLM beneath South Africa. Garnets from Caquele give a continuous rather cold PT path above 5.0 GPa (165 km). Eclogitic lenses are detected near 3.0 GPa (100 km) and beneath 50 to 60 GPa (165–200 km). The high temperature (HT) pyroxenites and hybrid peridotites are common from the base of the SCLM preferentially to 4.0 GPa (130 km). Ilmenites show two separate PT trends. The high PT group corresponds to Mg-rich ilmenites 0

0

Caquele

1

T oC

1 Sp

2 3 4 5 6 7

Gr SEA Grap hite Diam

ond

Opx McGr74 Cpx As10 Cpx NiTa00 Cpx As10 Pxt Cpx As10 Ecl Garn As10 Chr As10 Ilm As10

P(Gpa)

600

45 mw/m2

800

1000

1

40 mw/m2

1400

-ΔLogFO2 1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

2

3

3

3

4

4

4

5

5

5

6

6

6

7 P(Gpa)

1200

Variationa of Cpx, Opx, Gar, Chr, Ilm

2

7 35 mw/m2

8

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

7 P(Gpa)

8

8 0.05

0.10

2

0.15

0.20

0.25

0.30

0.35 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 6. PTXF plots using xenocrysts from Caquele pipe obtained using monomineral thermobarometry.

136

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 0

0 T oC

Camitongo I

1

1 Sp

2 3

Gr SEA Grap

hite Diam on

4 5 6 7

1. 2. 3. 4. 5. 6. 7. 8.

d

Opx McGr74 Cpx As10 Cpx NiTa00 Cpx As10 Pxt Cpx As10 Ecl Garn As10 Chr As10 Ilm As10

P(Gpa)

600

45 mw/m2

800

1000

1

40 mw/m2

1200

Variationa of Cpx, Opx, Gar, Chr, Ilm

-ΔLogFO2 1

1. CaOin Gar 1.1.CaO CaO ininGar Gar 2. 22O 33 3in 2.2.Al in Opx Opx AlAl O 2O in 3. 22O 33 3 Cr Cr 3.3.Cr O in Cpx Cpx 2O TiO 4. in Chr TiO Chr 22 2ininChr 4.4.TiO in 5. Ilm Cr 22O 33 3 O 5.5.Cr Cr ininIlm Ilm 2O

2

2

3

3

3

4

4

4

5

5

5

6

6

6 7 P(Gpa)

8

8

1400

2

7

7 P(Gpa)

35 mw/m2

8

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

0.05

0.10

0.15

0.20

0.25

0.30

0.35 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 7. PTXF plots using xenocrysts data set for Camitongo 1 pipe obtained using monomineral thermobarometry. Symbols are the same as in Fig. 5.

(Robles-Cruz et al., 2009) and creates a stepped plot from 7.0 to 4.0 GPa (235–130 km) while the lower temperature group traces the 35–38 mW/m 2 geotherms. The highest temperature dispersion occurs from 5.0 to 4.0 GPa (165–130 km), coinciding with the PT conditions for low-Cr pyroxenites. Fe-rich varieties create a rather low temperature path in the shallow part of the mantle at 3.0– 2.0 GPa (100–65 km) which joins the 35–45 mW/m 2 geotherms. The SCLM beneath all the studied pipes but mostly Camitongo I–II (Figs. 7–8) reveal heating to 45 mW/m 2 at 7.0–5.5 GPa (230–170 km) and the HT path determined for the few ilmenites and low-Cr clinopyroxenes (both pigeonites and augites and even omphacites) trace this gradient to lower pressure conditions. The PTX diagram shows that this trend is formed by Cpx with different Fe#. The most Ferich compositions are HT eclogites which for the SCLM column

beneath the Caquele pipe practically coincides in PT conditions with augites, whereas for the Camitongo I pipe they yield pressures of 1.5–2.5 GPa (50–75 km) close to the garnet–spinel boundary for peridotites. The SCLM beneath these pipes can be divided into two parts. PT estimates for clinopyroxene xenocrysts and garnet from Camitongo I show wide variation in the lower part of mantle section and major heating near 4.0 GPa (130 km). Those from Camitongo II show three geotherms and a stepped HT path near 45 mW/m 2 conductive geotherm divided into four parts from the base of the SCLM up to 2.0 GPa (65 km). The PT estimates for the Caquele pipe reveal a close set of pressures for garnets, clinopyroxenes and ilmenites in the range of 3.5–7.0 GPa (115–230 km). A LT path is given by the peridotitic minerals starting from 5.5 GPa (170 km) and a HT one

0

0

Camitongo II

1

T oC

1 Sp

2 3 4 5 6 7

Gr SEA Grap hite Diam

ond

Opx McGr74 Cpx As10 Cpx NiTa00 Cpx As10 Pxt Cpx As10 Ecl Garn As10 Chr As10 Ilm As10

P(Gpa)

600

45 mw/m2

800

1000

1

1200

-ΔLogFO2 1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

2

3

3

3

4

4

4

5

5

5

6

6

6 7 P(Gpa)

8

1400

2

7 P(Gpa)

40 mw/m2

Variationa of Cpx, Opx, Gar, Chr, Ilm

2

7 35 mw/m2

8

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

8 0.05

0.10

0.15

0.20

0.25

0.30

0.35 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 8. PTXF plots using xenocrysts data set for Camitongo 2 pipe obtained using monomineral thermobarometry. Symbols are the same as in Fig. 5.

0

0

Catoca

T oC

1

1 Sp

2 3

Gr SEA G ra p

hite Diam on

4 5 6 7

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1. 2. 3. 4. 5. 6. 7. 8.

d

Opx McGr74 Cpx As10 Cpx NiTa00 Cpx As10 Pxt Cpx As10 Ecl Garn As10 Chr As10 Ilm As10

P(Gpa)

35 mw/m2

8 600

45 mw/m2

800

1000

1200

1

1400

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

1

2

2

3

3

3

4

4

4

5

5

5

6

6

6

7

7 8 0.05

0.10

2

7 P(Gpa)

P(Gpa) 40 mw/m2

-ΔLogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

0.15

0.20

0.25

0.30

8 0.35 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 9. PTXF plots using xenocrysts data set for Catoca pipe obtained using monomineral thermobarometry. Symbols are the same as in Fig. 5.

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

located near the 45 mW/m 2 geotherm. Garnets and pyroxenes from Catoca give PT conditions which in the lower part of the SCLM are close to the 36 mW/m 2 geotherm but this PT trend is steeper than the conductive geotherm and crosses the 40 mW/m 2 geotherm and diamond stability boundary (Kennedy and Kennedy, 1976) near 3.5 GPa (140 km), forming a sub-adiabatic branch. Judging by the PT conditions and the frequency of the points lying within the diamond stability field, all of the studied the pipes should be diamondiferous. The Catoca (Fig. 9) and Caquele (Fig. 6) pipes are diamond deposits and are currently being excavated. The Camitongo I–II pipes have a higher abundance of such grains in their concentrate but show a relatively low content of omphacites from eclogites and sub-Ca garnets. The combined data set shows that all PT estimates for the xenocrysts from all pipes of the Catoca cluster are complementary and together reproduce the structure of the mantle column, including common peridotites and a vein system represented by the PT points for the low-Cr pyroxenites and ilmenites as well as the hot eclogites and Fe-rich Cr-diopsides. Clinopyroxene data show several paths on the P–Fe# plots. The ilmenites coincide in Fe# values with the intermediate trend, while the most Fe-rich ilmenites coincide with the path for the high temperature eclogites. They form the layer from 6.0 to 4.5 GPa (145 km). Oxygen fugacity values (fO2) for minerals form SCLM beneath Catoca cluster reveal rather regular variations. The fO2 for ilmenites are close to the quartz–fayalite–magnetite (QMF) buffer being higher for the Fe-rich samples to + 1 Δ QMF. This is higher in general than for the peridotitic mantle of Africa (Woodland and Koch, 2003) and Slave Craton (McCammon and Kopylova, 2004) (see Discussion). The high values correspond to the magmatic systems which due to self oxidation may produce values even higher than QMF buffer. In SCLM beneath Caquele pipe there is a correlation of fO2 for garnet and clinopyroxenes. They form several (6) deviating sub-isobaric lines to the very low fO2 in separate levels. In the SCLM beneath the Caquele and Camitongo I and II pipes, ilmenites trace the line the diamond stability field (DSF) (McCammon et al., 2001) in the lower part of the lithospheric mantle and they quickly rise with decreasing pressure in the upper part. This is more typical for Caquele pipe and less so for the SCLM beneath the Camitongo I pipe. The accompanying high fO2 are found for some hybrid Fe–Cr–Na-rich clinopyroxenes in separate parts of the SCLM. These levels are individual for each SCLM. In the lower part of the mantle sections the high fO2values are calculated for the Cpx which are close in chemistry to those from sheared peridotites and for typical eclogites. The Cpx fO2 show also the higher values in middle part of mantle section usually associated with the pyroxenite layer (Pokhilenko et al., 1999).

137

7. Geochemistry of the minerals Trace element concentrations were determined by LAM ICP-MS for clinopyroxenes, garnets, ilmenites and zircons. Tables 1–3 all clinopyroxenes from the studied pipes correspond to the garnet facies with Gd/Ybn varying from 3 to 1 and La/Smn b 1 ratios which negatively correlate. These parameters vary from one pipe to another. The inclinations of the REE patterns are more variable and steeper for clinopyroxenes from Camitongo I pipe. Only one flattened REE spectrum in Caquele pipe (Fig. 11) corresponds to spinel peridotite. The clinopyroxenes spider diagram patterns in each pipe reveal specific features. Those from Caquele have humped rounded REE patterns similar to each other, typical for enriched mantle (Gregoire et al., 2003; Richardson et al., 1985) with modal abundances close to the primitive mantle. Most of them have very high Ba and one shows a U peak like those found for phlogopites in metasomatites (Gregoire et al., 2003). This is not typical for the clinopyroxenes from the other pipes. Two spiderdiagrams have small positive Pb anomalies, whereas the others with higher concentrations display negative ones. They reveal Zr-minima. Samples most enriched in trace elements show Hf peaks possibly related to metasomatites with rutile. Cpx from spinel facies have a U peak. The high peaks for the Ba probably relate to the high abundance of eclogites in the mantle column. Clinopyroxenes from Camitongo I pipe reveal three types of patterns (Fig. 12). Two correspond to Cpx from typical primitive garnet peridotites similar to those from the Caquele pipe. They are depleted in LILE–HFSE components, have strong Pb enrichment (as well as the other clinopyroxenes from this pipe) and display a strong Zr minimum. Two more enriched samples (REE ~ 10 relative to PM) reveal small enrichment in Th, U compared to the Nb–Ta troughs, which may be the sign of carbonatite metasomatism. Clinopyroxenes with REE enrichment factors near 100–50 PM are characterized by higher Th–U peaks, Zr troughs and may relate to metasomatism associated with zircon precipitation. Cpx from Camitongo II are divided into two groups. One has a moderate REE (~10) level and nearly primitive trace element spectrum with Ba depletion and Zr minima for those with lower trace element concentration. The three others are much more enriched (~100 PM) with HREE depletions. They reveal Ba–Th–U humps, Pb troughs and nearly flat HFSE except for Zr which is low as for most samples. Garnets from the three pipes differ in REE levels. Those from Caquele have higher HREE levels close to 10/PM (primitive mantle), one displays a flexed pattern with depletion in MHREE. All are moderately enriched in LILE–HFSE and have Pb and Hf peaks. Those from Camitongo I have lower HREE ~ 1 and reveal strong U–Rb peaks and deep Ba–Sr minima. Garnets from Camitongo II have

0

0

T oC

Catoca cluster together

1

1 Sp

2 3 4 5 6 7

Gr SEA Grap hite Diam ond 45 mw/m2

Opx McGr74 Cpx As10 Cpx NiTa00 Cpx As10 Pxt Cpx As10 Ecl Garn As10 Chr As10 Ilm As10

P(Gpa)

600

800

1000

1200

1

1400

1

1. CaOin Gar 1.1.CaO CaO ininGar Gar 2. Opx 22O 33 3in AlAl O ininOpx Opx 2.2.Al 2O in 3. 22O 33 3 Cpx Cr 3.3.Cr O ininCpx Cpx Cr 2O 4. TiO 22 2in Chr TiO ininChr Chr 4.4.TiO in 5. Ilm Cr 22O 33 3 5.5.Cr ininIlm Ilm Cr O 2O

2

3

3

4

4

4

5

5

5

6

6

6

7 P(Gpa)

40 mw/m2

-ΔLogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

2

7 35 mw/m2

8

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

7

8 0.10

3

P(Gpa)

8 0.05

2

0.15

0.20

0.25

0.30

0.35 0.0

8

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 10. PTXF plots using xenoliths data set for all the analyses for Catoca kimberlite cluster obtained using monomineral thermobarometry. Symbols are the same as in Fig. 5.

138

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

Caquele pipe Clinopyroxenes 100.00

10.00

10.00

Sample/PM

Sample/C1

100.00

1.00

0.10 La

Ce

Pr

Nd

100.00

0.10 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

100.00

Garnets

10.00

10.00

Sample/PM

Sample/C1

1.00

1.00

0.10

1.00

0.10

0.01

Ce Nd La Pr

Ilmenites

100.00

0.01 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

1000.00

100.00

10.00

Sample/PM

Sample/C1

10.00

1.00

1.00

0.10

0.01

0.10

Ce Nd La Pr

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

Zircons 1000.00

100.00

100.00

10.00

Sample/PM

Sample/C1

10.00

1.00

1.00

0.10

0.01

0.10 La

Ce Nd Pr

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Fig. 11. Rare earth patterns and trace element spidergrams for minerals from the Caquele pipe.

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

139

Camitongo I Cpx 100.00

100.00

10.00

Sample/C1

Sample/PM

10.00

1.00

0.10

1.00 0.01

0.10

Ce Nd La Pr

100.00

0.00

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

Garnets

Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

100.00

10.00

Sample/PM

Sample/C1

10.00

1.00

1.00

0.10

0.10 0.01

0.01 La

Ce Nd Pr

Ilmenites

1.00

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

1000.00 100.00 10.00

Sample/C1

Sample/PM

0.10

1.00 0.10

0.01 0.01 0.00 0.00 Ce Nd La Pr

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Fig. 12. Rare earth patterns and trace element spidergrams for minerals from the Camitongo I pipe.

moderate HREE and low LREE (Fig. 13) with an inflected REE pattern in general. They display a Rb peak higher than U and a generally elevated left part of the spider diagram, small Pb and Hf peaks and strong Ba–Sr dips. Zircons from two pipes in general display rather typical elevated trace element patterns with HREE enrichment, strong Ce, Th–U, Zr–Hf peaks and saw-like spidergrams. Three zircons from Camitongo II pipe display fan-like rotations of the REE patterns around Sm. The Ce anomaly is much higher for the pattern that is lower in REE in general. Ilmenites from different pipes also differ in trace element spectra. The rather elevated spiderdiagrams from Caquele pipe reveal high HFSE peaks and no Y anomaly. They display high Nb–Ta and relatively high Zr–Hf and Pb peaks and low LILE and Sr concentrations. Ilmenite from Camitongo I display gently inflected W-shaped REE distributions (~0.2 PM) and LREE enrichment typical for melts saturated in fluids. Very strong Nb–Ta and lower Pb and Zr–Hf peaks relate to the HT

metasomatites or group I (Ashchepkov et al., submitted). Ilmenites from Camitongo II reveal mainly W-shaped REE patterns but more inclined with LREE enrichment. They display strongly enriched Nb–Ta but lower Zr–Hf spiderdiagrams, some with small Pb peaks. Comitongo I–II ilmenites also display Y dips and are probably related to percolation of Al-depleted peridotites from the lithosphere base or percolation of the parental melt through the depleted source. 8. Discussion 8.1. Interpretation of mantle composition beneath western part of Congo–Kasai craton General mineral chemistry of the garnets and pyroxenes from the south-western part of the Congo–Kasai craton reveals rather enriched compositions which are in general similar to most localities in South Africa (Batumike et al., 2009; Bell et al., 2003; Boyd and Danchin,

140

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

Camitongo II Cpx 100.00

100.00

10.00

Sample/C1

Sample/PM

10.00

1.00

1.00 0.10

0.10

Ce Nd La Pr

100.00

0.01

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

100.00

Garnets

10.00

Sample/PM

10.00

Sample/C1

Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

1.00

1.00

0.10

0.10 0.01

0.01

Ce Nd La Pr

Ilmenite

1.00

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

1000.00 100.00 10.00

Sample/C1

Sample/PM

0.10

1.00 0.10

0.01 0.01 0.00 0.00

Ce Nd La Pr

0.00

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Zircon 10000.00

100.00

1000.00 100.00

Sample/PM

Sample/C1

10.00

1.00

10.00 1.00 0.10

0.10 0.01 0.01

Ce Nd La Pr

Sm Gd Dy Er Yb Eu Tb Ho Tm Lu

0.00 Cs Ba U Ta Ce Pr Sr Hf Eu Dy Y Yb Rb Th Nb La Pb Nd Sm Zr Gd Ho Er Lu

Fig. 13. Rare earth patterns and trace element spidergrams for minerals from the Camitongo II pipe.

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

1980; Correia and Laiginhas, 2006; Gregoire et al., 2003; Leost et al., 2003; Moore et al., 1992; Nixon et al., 1981; van Achterbergh et al., 2001; Viljoen et al., 2009). Chemistry of minerals from all three pipes characterize the parental mantle peridotites as being rather fertile Cpx-rich with abundant metasomatic associations including phlogopites and richterites which are found in concentrates. The mantle columns include many eclogites because omphacites and orange garnets are common in concentrates, and there are many Na–Al–Fe-rich Cr-diopside pyroxenites which are usually the result of the hybridization due to melting of eclogites and mixing with mantle material. The amount of metasomatites in the mantle should also be rather high, judging by the geochemistry of the minerals with peaks in Ba, Sr and U as well as Fe–Cr pyroxenes. This is also typical for many pipes from South Africa (Gregoire et al., 2002) including Kimberly (Gregoire et al., 2003; Simon et al., 2007), Finsch (Gibson et al., 2008) etc. The amount of the depleted associations is much higher for the enormous Catoca pipe compared with the other localities in the same kimberlite cluster. This is a typical feature for Yakutia where the largest pipes in the cluster usually contain more depleted material, corresponding to large dunite–harzburgite melt percolation channels (Ashchepkov et al., 2004; Ashchepkov et al., 2010). 8.2. Thermal structures and construction of the mantle columns beneath Catoca kimberlite cluster The PTX diagrams allow us to determine the layering of the mantle using the fluctuation of Fe# of all the minerals. The general division of the SCLM in this part of Angola is near 5.0 GPa (165 km) (Fig. 10). In the HT base it is possible to suggest 3–4 smaller layers and also not less than 3 additional peridotite horizons in the relatively low PT interval. The stepped HT Cpx paths reveal 3–4 sharp intervals including the heated SCLM base. The P–fO2 diagram shows that the primary layering of the SCLM beneath the Catoca cluster include at least 6 large units and about 10 small ones which are marked by arrays of Gar and low-Fe Cpx fO2 from −5 log units to QMF. For the SCLM beneath the individual pipes, some arrays are inclined and rising to the lower values reflecting melt migration and oxidation. But together they form a rather continuous layer (Fig. 10). The high temperature trend for low-Cr and hybrid pyroxenes from the lithosphere base to 2.5 GPa (100 km) represents the PT path of the protokimberlite melts. Ilmenite PT conditions mainly correspond to the colder conditions of crystallization in wall rocks and outer parts of channels than those found for the clinopyroxenes. It is likely that the structure of the SCLM is highly dependent on the abundance and composition of the eclogites. Most Fe-rich eclogite lenses were detected from near 3.0–3.5 GPa (100–115 km) and 5.0– 6.5 GPa (165–180 km). Clinopyroxenes with intermediate Fe# values

141

were detected above and below these lenses and probably reflect hybrid reactional layers formed by mixing of peridotites and eclogites. They are typical for many mantle sections worldwide. There is special type of the metasomatism with the simultaneous increase of Fe, Al, Na and very often Cr which possibly is result of the interaction of mantle melts with eclogites. This is typical for the mantle beneath the Caquele pipe and regions like the Alakite field in Yakutia (Ashchepkov et al., 2004) and for Kelsey Lake −1 pipe in Colorado (Ashchepkov et al., 2001). The amount of relatively low temperature eclogites in SCLM beneath this pipe is also higher. Judging by high Fe#s, the eclogites from Caquele relate to the primary metasomatic eclogites type I (Gréau et al., 2011). The lower Fe# eclogites are higher temperature and correspond to the higher temperatures marking the PT path of the protokimberlites and reactional lower temperature varieties. Xenocrysts from each pipe represent part of the whole path of the protokimberlite melts. They have some similar features such the general layering location of the eclogite lenses the HT branches referred to the channels of protokimberlites and surrounding metasomatites. Together they compile the whole structure of the SCLM (Fig. 10). The lithosphere beneath the Caquele pipe may be divided into three intervals from the top down to 40 GPa. The most enriched clinopyroxene compositions occur within the 40–50 GPa interval where pyroxenites are common in the mantle section (Pokhilenko et al., 1999). The layer at ~ 4.5–5.0 GPa (155–165 km) is traced by LT eclogites. Ilmenite trends commonly stop near these levels (Ashchepkov et al., 2010). But beneath the Catoca cluster, they appear again at the top of the mantle, together with HT eclogites and pyroxenites. Two more heated layers may be roughly determined at 5.0–7.0 GPa (230–245 km) and near 7.0–7.5 GPa (230–245 km), where two more eclogitic lenses are present. The Cr-poor pyroxenites trace the trend of ilmenite compositions in the upper part but in the lower part they belong to Cr-bearing metasomatites. More detailed layering within the SCLM beneath Camitongo I–II pipes is apparent from the PT and PTX trends which have inflections at the boundaries of 4–5 layers within the 7.0–4.0 GPa (230– 130 km) in the lower interval and 3–4 layers in the upper part. Mantle xenocrysts from the Camitongo I pipe represent 2–3 layers in the upper part of the mantle section down to 4.0 GPa (130 km). The peridotitic part is probably similar to those suggested beneath Caquele pipe. Cr- and Fe-rich metasomatites should be abundant within the 3.5–5.5 GPa (115–180 km) interval, because here the PT points in the P–Fe# diagram for ilmenites and garnets show high scattering of Fe# values, coinciding in one cloud. Abundant low-Cr pyroxenites and hybrid peridotites and rare eclogites also relate to this interval. The low-Cr ilmenite fractionation trend traces the magmatic system probably responsible for heating the mantle. The more Fe-rich pyroxenites (Fe# > 0.15) may be derived from more 0

0 T oC

Kao

1

1 Sp

2 3 4

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

Gr SEA Grap hite Diam ond 45 mw/m2

5 6 7 35 mw/m2

8 600

800

1000

1200

1400

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

1

2

3

3

3

4

4

4

5

5

5

6

6

6

7 P(Gpa)

40 mw/m2

-ΔLogFO2

2

7 P(Gpa)

1

Variationa of Cpx, Opx, Gar, Chr, Ilm

7 P(Gpa)

8

8

8 0.05

0.10

2

0.15

0.20

0.25

0.30

0.35 0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 14. PTXF plots using xenoliths data set for Kao (Lesotho) pipe (MacGregor, 1979) obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5.

142

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 0

0 T oC

Letseng

1

1 Sp

2 3 4

Gr SEA Grap

hite Diam o

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

nd 45 mw/m2

5 6

P(Gpa)

600

800

1000

40 mw/m2

1200

1

1.1.CaO CaOininGar Gar 2.2.AlAl 2O 3 3ininOpx Opx 2O 3.3.CrCr 2O 3 3ininCpx Cpx 2O 4.4.TiO 2 2ininChr Chr TiO 5.5.CrCr in Ilm 2O 3 2O3 in Ilm

2

3

3

3

4

4

4

5

5

5

6

6

6

2

7

7 P(Gpa)

P(Gpa)

35 mw/m2

8

-ΔLogFO2

2

7

7

1

Variationa of Cpx, Opx, Gar, Chr, Ilm

8

8

1400

0.05

0.10

0.15

0.20

0.25

0.30

0.35 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 15. PTXF plots using xenoliths data set for Letseng la Terai (Lesotho) pipe (Moore and Lock, 2001; Woodland and Koch, 2003) obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5.

Fe-rich melts, which were probably plume-related. Eclogites situated in the lowermost part near 7 GPa (225 km) are more Fe-rich (up to Fe# ~ 30 and even more). For the Camitongo II pipe, sharper layering is seen on the P–Fe#Ol diagram. The upper part from 1.5 to 3.5 GPa (50–120 km) splits into two layers. The sequence from the top to 2.5 GPa (60 km) contains pyroxenites and lherzolites. The level near 3.5 GPa (115 km) is traced by the abundant Cr-bearing heated peridotites which are not essentially Fe-rich. The lower part of mantle column contain the rising and evolving magmatic system probably in the form of a channel. On the P–Fe# diagram, Fe-rich ilmenites and Cr-poor clinopyroxenes create a joint trend of coeval Fe# growth with decreasing pressures. Probably this was an AFC process because Cpx as well as the ilmenites are contaminated with Cr increasing upward. They probably were surrounded by the metasomatites because Fe-rich Cr-bearing clinopyroxenes became more ferriferous than garnets displaying typical fertilization signs. The lower part from 7.5 to 5.5 GPa (250–180 km) is represented by heated rocks forming a convective branch and surrounding the lower magma chamber crystallizing ilmenites and lower-Cr pyroxenites. Thermal gradients for the mantle columns of most Southern African kimberlite pipes mainly correspond to the 40 mW/m 2 geotherms but some depleted rocks including diamond-bearing associations reveal colder conditions approaching the 35 m/Wm 2 gradient. The same has been determined for some of the coldest eclogites in the lithospheric mantle beneath the Lesotho kimberlites (Fig. 5). This means that the South Africa lithosphere was heated probably during the Upper Proterozoic–Paleozoic and Mesozoic time by the numerous plumes which caused the breakup of Rodinia and Gondwana. The thicker lithosphere of the Congo–Kasai craton,

reaching 400 km in the central parts (O'Reilly et al., 2009), was the reason for the colder thermal state determined for the studied pipes. The Mesozoic mantle geotherms from Angola are colder than the Mesozoic (Figs. 5, 14–20) and Proterozoic (Figs. 22–23) geotherms for South and East Africa. 8.3. Ilmenite trends as indicators of the development of the feeder system for kimberlite eruptions Long ilmenite compositional trends from South Africa kimberlites are suggested to have been formed by fractionation (Griffin et al., 1997; Moore et al., 1992), but the minor components reveal more complicated trends and origin. The concentration of minor components (NiO, MnO, Al2O3, V2O5 and Cr2O3) is regulated by the different phases that accompanied the ilmenites during AFC fractionation of the parental melts. For ilmenites, Al2O3 is controlled by the presence of garnet and chromite, NiO by sulfides and olivine, and Cr2O3 by chromite etc. These phases may play different roles as assimilants or precipitates. We assume that the ilmenite fractionation trend in kimberlites was produced in the long vertical feeding system in the mantle (Ashchepkov et al., 2010), similar to the augite trend in alkali basalts (Ashchepkov et al., 2011). PT data for pyroxenes coexisting with ilmenites from Monastery (Gurney et al., 1979a; Haggerty et al., 1979) show that this was a polybaric trend. An earlier model suggested the presence of large pegmatite-like bodies in the lithosphere base where the megacrysts crystallized (Boyd and Danchin, 1980; Griffin et al., 1997; Moore, 1987). The increase of fO2 in the upper part of the mantle section is regarded as the reflection of differentiation with preferential olivine

0

0 T oC

1

Finsch

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1 Sp

1

-ΔLogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm 1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

1

2

2

3

3

3

4

4

4

5

5

5

5

6

6

6

6

2 3 4

Gr SEA Grap hite Diam

ond

7

7 35 mw/m2

8 600

7

45 mw/m2

P(Gpa)

800

1000

1200

40 mw/m2

1400

8

P(Gpa)

2

7 P(Gpa)

8 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 16. PTXF plots using xenoliths data set for Kimberly (Lesotho) pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Simon et al., 2007; Jacob et al., 2009; Katayama et al., 2009).

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

143

0

0 T oC

Kimberley

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1

1 SEA Sp

2

1

Variationa of Cpx, Opx, Gar, Chr, Ilm

4 5 6 7 8

Grap hite Diam ond 1.Opx 2.OPXDia 3.CpxNiTa00 4.CpxAsh10 5.CpxAsh10Ecl 6.CpxAsh10Dia 7.CpxAsh10PXt 8.GarAsh10 9.Gar Ash10 Dia 10.ChrAsh10 11. Chr Ash10Dia 12. Ilm Ashch10 13. BrKo90

2

2

3

3

3

4

4

4

5

5

5

6

6

6

600

35 mw/m2

800

1000

40 mw/m2

1200

2

7

7 45 mw/m2

P(Gpa)

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

Gr

3

-ΔLogFO2

7 P(Gpa)

P(Gpa)

8

8

1400

0.05

0.10

0.15

0.20

0.25

0.30

0.0

8 2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 17. PTXF plots using xenoliths and diamond inclusions data set for Finsch (Lesotho) pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Appleyard et al., 2004; Gibson et al., 2008; Gurney and Switzer, 1973; Shee et al., 1982; Viljoen et al., 1992).

The simplest crystallization scheme is suggested for Camitongo II. The HT (P) conditions correspond to HT metasomatites produced by interaction of hot plume magma with the peridotite substrate. The differentiation assemblages are represented by olivine and minor garnet as well as ilmenite itself, according to modeling calculations. But natural ilmenite assemblages are mainly garnet pyroxenites (Rodionov et al., 1988). Ilmenite peculiarities are rather specific for the kimberlite fields (Sobolev, 1980) but the difference between the

precipitation and self oxidation of the magmatic system due to hydrogen migration. Interaction of the eclogites which have rather high fO2 in the lower part of the mantle section is probably responsible for the additional oxidation of the mantle hybrid veins. The dense clots of high-Fe Cpx in different levels probably mark the intermediate magmatic chambers or large veins. They trace the boundaries of the primary mantle layering and the traps for mantle melts (Tappe et al., 2007).

0

0 T oC

1

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1

DeBeers Pool SEA Sp

2

1

4

Gr a phit e Dia mo nd 45 mw/m2

5 6 7

2

2

3

3

3

4

4

4

5

5

5

6

6

6

8 600

800

1000

7

P(Gpa)

35 mw/m2

40 mw/m2

1200

P(Gpa)

8

8

8

1400

2

7

7 P(Gpa)

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

Gr

3

- ΔLogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

0.05

0.10

0.15

0.20

0.25

0.30

0.35 0.0

8.0

-6.0 -4.0 -2.0 0.0

Fig. 18. PTXF plots diamond inclusions data set for De Beers Pool (Kaapvaal craton) pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Phillips et al., 2004; Banas et al., 2009).

0

0 T oC

1

Jagersontein

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1 Sp

2

1

Variationa of Cpx, Opx, Gar, Chr, Ilm

- Δ LogFO2 1

2

2

2

3

3

3

4

4

4

5

5

5

6

6

Gr

3 4

Gra p Dia hite mo nd

SEA

5

45 mw/m2

6

6

7

7 P(Gpa)

35 mw/m2

8 600

800

1000

1200

P(Gpa)

40 mw/m2 8

1400

7

7

P(Gpa)

8

8

0.05

0.10

0.15

0.20

0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 19. PTXF plots using xenoliths and diamond inclusion data for Jagersfontein (Kaapvaal) pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after (Aulbach et al., 2007; Burgess and Harte, 1999; Jacob et al., 2003; 2005; Tappert et al., 2005; Viljoen et al., 2005; Winterburn et al., 1990).

144

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 0

0 T oC

1

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

Orapa Sp

2 3 4

1

1

1

2

2

2

3

3

3

4

4

4

5

5

5

6

6

6

Gr SEA

Grap hite Diam ond

5

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

45 mw/m2

6 7

600

800

1000

P(Gpa)

P(Gpa)

35 mw/m2

8

1400

8

8

40 mw/m2 8

1200

7

7

7

P(Gpa)

0.05

0.10

0.15

0.20

0.25

0.30

0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 20. PTXF plots using xenoliths and diamond inclusions data set for Orapa pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Deines et al., 2009: Stachel et al., 2004).

populations of closely located pipes Camitongo I–II is very high. For the Camitongo II pipe, dissolution of a Cr-rich phase (probably clinopyroxene and chromite) in the protokimberlite magma was negligible. For Camitongo I, the trend is more extended. For the mantle column beneath the Caquele pipe, two levels of the protokimberlite magma fractionation are evident. The HT (P) line is close to that of Camitongo I with minor garnet fraction which is seen not only on PT diagrams but also in the rapid decrease of Al. The upper level is characterized by a new differentiation trend possibly in a long channel at 30–2.0 GPa (65 km) and wall rock interaction accompanied by crystallization of olivine and garnet in different proportions. Thus the branching system seems to be the most probable reason for the variation of ilmenite and clinopyroxene in the middle part of the mantle section. It seems that enrichment in chromium was mainly regulated by dissolution of Cr-bearing phases from the peridotite substrate surrounding the feeding system. Zonation and difference in the compositions of the porous and hard grains (Robles-Cruz et al., 2009) is common for ilmenites. Zonation in Cr may be of different types (Schulze et al., 1995). It should be noted that ilmenites crystallized within and near the magmatic channels. Small xenocrysts or an ilmenite mush may have been transported by the magma to the upper horizons and were cemented by the Fe-rich ilmenite aggregates which could explain the sharp zonation. Trace elements for ilmenites from Caquele relate to the metasomatites (W-shaped and low concentration) and to differentiation (LREE-enriched from melts which precipitated garnets) while for those from the Camitongo I relate mainly to the metasomatites. The

hydrous metasomatites occur mainly within the middle part of the mantle section. The REE patterns of the garnets and clinopyroxenes which show an inflection in Eu–Gd probably result from hybridization with melts contaminated by eclogites. This suggestion is proved by the existence of HT omphacites. The difference between the reconstructed mantle sections beneath Camitongo I–II and Caquele was probably formed by capture at different stages of the kimberlite magmatism which reflected the development of the deep-seated magmatic system. Not only the vein system but also surrounding peridotites changed rapidly. Megacrysts from Camitongo I should be related to the first phase of development of the magmatic system. At the next stage of development of the protokimberlite system of channels and chambers, it was split into the lower and upper sections. In the next stage, shown by megacrysts from Camitongo II, differentiated protokimberlite melts migrated to a higher level at 4.0 GPa (130 km) within the mantle section and differentiated, producing wall rock and vein metasomatites and Caenriched pyroxenitic associations and some depleted peridotites. The last stages relate to the mantle columns beneath the Catoca and Caquele pipes when a large amount of melt created the shallow magmatic system. Interaction of a large volume of flowing volatile-rich melt (Stachel and Harris, 2008) within the veins in peridotites was probably the reason for the high degree of depletion, creation of dunites served as melt conductor (Braun and Kelemen, 2002) and diamond formation. Diamond growth may be explained by interaction of the oxidized essentially carbonatitic melt with the reduced mantle peridotites. Such processes should be especially effective during interaction with those eclogites which initially contained carbon.

0

0 T oC

1

Letlhakane

1 Sp

2 3 4

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

Gr SEA Grap hite Diam ond 45 mw/m2

5 6

P(Gpa)

35 mw/m2

8 600

800

1000

1200

1400

1

2

3

3

3

4

4

4

5

5

5

6

6

6

P(Gpa)

8

8

8 0.05

0.10

2

7

7 P(Gpa)

40 mw/m2

1. 1. CaO CaO in in Gar Gar 2. 2. Al Al22O O33 in in Opx Opx 3. O33 in in Cpx Cpx 3. Cr Cr22O 4. 4. TiO TiO22 in in Chr Chr O in 5. Cr 2 3 Ilm 5. Cr2O3 in Ilm

2

7

7

1

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

0.15

0.20

0.25

0.30

0.35 0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 21. PTXF plots using xenoliths and diamond inclusions data set for Letlhakane pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (van Achterbergh et al., 2001; Deines and Harris, 2004; Stiefenhofer et al., 1997).

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

145 0

0 T oC

1

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

Premier SEA Sp

2

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

1

1

2

2

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

Gr

2

3

Grap hite

3

3

3

4

Diam ond

4

4

4

5

5

5

5

6

6

6

6

7 45 mw/m2

7 P(Gpa)

35 mw/m2

8 600

800

1000

P(Gpa)

1400

8

8

40 mw/m2 8

1200

7

7

P(Gpa)

0.05

0.10

0.15

0.20

0.25

0.30

0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 22. PTXF plots using xenoliths and diamond inclusions data set for Premier pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Gregoire et al., 2005; Gurney et al., 1985; Tsai et al., 1979; Viljoen et al., 2009).

8.4. Comparison of the Angolan SCLM structure and compositions with SCLM of other African cratons Using the same scheme of monomineral thermobarometry and the Opx–Gar–Cpx thermobarometry after Brey and Kohler (1990), we calculated a series of diagrams for the SCLM beneath the different pipes with diverse ages of kimberlite eruptions in Africa. Monomineral thermobarometry gives more detailed structures for the SCLM beneath the different pipes from Lesotho (Boyd and Nixon, 1975; Smith, 1999) (Figs. 5, 14–17) than the published straight line 40 mW/m 2 geotherm (Gibson et al., 2008) (Fig. 17) obtained with the Brey and Kohler (1990) formulation or inflected geotherms (Boyd, 1973) calculated using Opx–Cpx thermobarometry (McGregor, 1974; MacGregor, 1979 etc.). For example, combinations of the PT estimates for xenoliths from different pipes show that the inflection at the base of the lithosphere corresponds to different levels (5.5 or 6.5 GPa) (180–230 km) which possibly reflect an agedependent perturbation in the base of the lithosphere. Most of them relate to the several episodes of Mesozoic activity from 85 to 145 Ma (Woodhead et al., 2009). The interval 120–145 Ma is close to the time of kimberlite magmatism in NE Angola. The SCLM beneath some pipes reveals nearly straight line or slightly split 42–23 mW/m 2 geotherms, for example for Letseng (Fig. 15) (Moore and Lock, 2001) (Fig. 15) and Kao (MacGregor, 1979) (Fig. 14) pipes. But PT calculations for Opx and Cpx and garnets show additional horizons near 7.0 GPa (230 km). Also, the well documented diagram obtained by polymineral methods (Brey and Kohler, 1990 ) for the Kimberly SCLM (Fig. 16) (Jacob et al., 2009; Katayama et al., 2009; Simon et al., 2007) reveals two additional layers corresponding to 6.0–6.5 GPa (200–215 km) and 7.0–7.5 GPa

(230–245 km). For the SCLM beneath the Finsch pipe (Appleyard et al., 2004; Gibson et al., 2008; Gurney and Switzer, 1973; Lazarov et al., 2009; Shee et al., 1982; Viljoen et al., 1992) (Fig. 17), monomineral thermobarometry shows that the geothermal gradient splits into two parts. Instead of single 40 mW/m 2 gradient documented by Gibson et al. (2008), we also obtained a cold branch which is well established by the Opx thermobarometry for diamond inclusions near the graphite–diamond (Gr–Di) boundary and extends downward by the PT points obtained with garnet thermobarometry for diamond inclusion giving the convective branch from 6.5 to 8.0 GPa (215–270 km). PT path for diamond-bearing eclogites (Appleyard et al., 2004; Smith et al., 1991) also split the geothermal path into a HT path which continues down to 6.5 GPa (215 km) by the PT for chromites and an intermediate PT path at ~ 3.8 GPa (125 km), corresponding to several points for Opx and Cpx and Brey and Kohler (1990) estimates which were also shown by Lazarov et al. (2009). This splitting into several PT paths is also found for the SCLM sampled by the Jagersfontain pipe (Jacob et al., 2003; 2005; Pyle and Haggerty, 1998; Tappert et al., 2005; Winterburn et al., 1990) (Fig. 19). The Opx-based PT estimates for peridotites show an inclined extended PT path parallel to the graphite–diamond boundary line (Gr–Di) (Kennedy and Kennedy, 1976) in the lower part of the mantle section and located near 40 mW/m 2 conductive geotherm and forming 4 groups from 2.0 to 6.7 GPa (66–220 km). But Cpx-based thermobarometry produced further splitting, forming three branches when completed by the PT points for the Fe-rich Cpx and omphacites and in addition for pyrope diamond inclusions. The PT points for eclogites plot between 5.5 and 4.0 GPa (170–130 km) and in addition from 6.7 to 5.5 GPa (220– 170 km), revealing the subadiabatic continuation rising up from 35 0

0 T oC

1

1

Roberts Victor Sp

2 3 4

Gr SEA

Grap hite Diam ond

45 mw/m2

5 6

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm 1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

1 2

2

2

3

3

3

4

4

4

5

5

5

6

6

6

35 mw/m2 40 mw/m2 7

7

8

8 600

800

1000

1200

1400

7

7 P(Gpa)

P(Gpa)

P(Gpa)

8

8 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 23. PTXF plots using xenoliths and diamond inclusions data set for Roberts Victor pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Gréau et al., 2011; Jacob et al., 2003; Jacob et al., 2005; Sautter and Harte, 1990).

146

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 0

0 T oC

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1

1

1

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

Venetia Sp

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

2

2

2

3

3

3

4

4

4

5

5

5

5

6

6

6

6

2 3 4

Gr SEA Grap hite Diam ond

7

7 35 mw/m2

8 600

800

1000

40 mw/m2

1200

P(Gpa)

P(Gpa)

8

1400

7

7

45 mw/m2

P(Gpa)

8

8 0.05

0.10

0.15

0.20

0.25

0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 24. PTXF plots using xenoliths and diamond inclusions data set for Venetia pipe obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Aulbach et al., 2002; Denies et al., 2001; Hin et al., 2009; Richardson et al., 2009; Viljoen et al., 1999).

to 38 mW/m 2 geotherm and crossing the Gr–Di line. The additional groups of PT conditions were estimated for garnet inclusions forming a cold geotherm from ~5.5 to 6.0 GPa to the graphite–diamond (GrDi) boundary. The high variation of thermal gradients in the SCLM beneath Kaapvaal are proved by the data for Opx diamond inclusions from De Beers Pool (Phillips et al., 2004) obtained with Opx-based barometry (Brey and Kohler, 1990; McGregor, 1974) (Fig. 18). The ancient thermal conditions correspond to the Archean time for diamond inclusions (Banas et al., 2009) when the lithophere was thicker (see Fig. 27). The P–fO2 diagrams allow us to see details of the structure of the mantle sections and to mark the location of the intermediate magmatic chambers. There are at least 4 such level in the SCLM beneath the pipes in central Lesotho (Nixon and Boyd, 1973) (Fig. 5) including Matsoku, Letsing la Terrae, Kao and some other kimberlite diatremes. The diagrams for the SCLM beneath separate pipes like Letseng and Kao (Figs. 14–15) show that location of Fe-rich and fO2-rich horizons in SCLM beneath each pipe are coinciding. But they are individual for each mantle section. A similar diagram for the SCLM beneath Kimberly mine shows a rather complex structure with the location of high fO2 values in least at 6 levels which are coinciding with the level of hearing (Katayama et al., 2009). Most Cpx values are higher than those for Gar, reflecting refertilization in all levels of mantle column. The diagram for Finch (Fig. 17) show three separate levels for the diamond inclusions located within the DSF. A similar diagram for the De Beers Pool SCLM (Fig. 18) shows a linear fO2 trend from the base to 4.0 GPa where the garnets have pyroxenitic features. The fO2 of the omphacites from Jagersfontein pipe show the four separated levels for eclogites. The refertillization

trend for common lherzolites locates near and above the DSF (Fig. 19). The SCLM beneath Orapa (Aulbach et al., 2002; Cartigny et al., 1999; Stachel et al., 2004) (Fig. 20) reveals roots extending to 8.0 GPa (270 km) generally heated to 40 mW/m2 or more. Similar conditions were determined for the SCLM beneath the Letlhakane pipe (Fig. 21) (Stienfenhofer et al., 1997; Deines and Harris, 2004; van Achterbergh et al., 2001) located 40 km from Orapa. The fO2 values for the diamond eclogites and inclusions show that the eclogites are more oxidized and probably are related to magmatic processes, at least near the pyroxenite lens (4 GPa), while the garnets were formed at more reduced conditions. The Mg-rich chromites mark very similar conditions in all mantle columns (50–65 kbar and −4 to −2 log units relative to QMF). The geotherms for the Proterozoic pipes like Premier in the Kaapvaal craton (Viljoen et al., 2009) including data for diamond inclusions (Deines et al., 1984) (Fig. 22) give in general higher temperature geothermal conditions above the 45 mW/m 2 geotherm. The SCLM base is near 7.0 GPa (235 km) and even 8.0 GPa (270 km) as documented by the Opx estimates. The PT conditions for pyrope diamond inclusions (Tsai et al., 1979) in general correspond to lower T conditions and trace mainly the ~ 38 mW/m 2 geotherm. The omphacitic diamond inclusions are in general higher T, giving an adiabatic path rising near the 45 mW/m 2 geotherms but general peridotitic estimates also trace the 40 mW/m 2 geotherm (Stachel et al., 2004). But several PT conditions determined for Opx, Chr and pyropes are located near the ~ 36 mW/m 2 geotherm. PT conditions for omphacite diamond inclusions and eclogites from Roberts Victor SCLM (Hatton and Gurney, 1979; Jacob et al., 2003;

0

0 T oC

1

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1

Namibia SEA Sp

2

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

4

Gra phit e Dia mo nd

5

1

2

2

3

3

3

4

4

4

5

5

5

6

6

Gr

3

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

2

45 mw/m2

6

6

P(Gpa)

35 mw/m2

8 600

800

1000

1200

P(Gpa) 40 mw/m2

1400

7

7

7

7

P(Gpa)

8

8 0.05

0.10

0.15

0.20

0.25

0.30

0.350.0

8 8.0

-6.0 -4.0 -2.0 0.0

Fig. 25. PTXF plots using xenoliths and diamond inclusions data set for pipes form Namibia obtained using monomineral (Ashchepkov et al., 2010) and polymineral (Brey and Kohler, 1990) thermobarometry. Symbols are the same as for Fig. 5. Data after: (Franz et al., 1996; 1997; Harris et al., 2004; Leost et al., 2003; Schmädicke et al., 2011).

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

147

0

0

Guinea

T oC

1

Fe# Ol in equilibrium with Cpx, Opx, Gar, Chr, Ilm

1 SEA Sp

- Δ LogFO2

Variationa of Cpx, Opx, Gar, Chr, Ilm

1

1

1. CaO in Gar 2. Al2O3 in Opx 3. Cr2O3 in Cpx 4. TiO2 in Chr 5. Cr2O3 in Ilm

2

2

2

3

3

3

4

4

4

5

5

5

5

6

6

6

6

2

Gr

3 4

Grap hite Diam ond

P(Gpa)

8 600

800

45 mw/m2 35 mw/m2

1000

1200

40 mw/m2

1400

7

7

7

7

P(Gpa)

P(Gpa)

8

8

8 0.05

0.10

0.15

0.20

0.25

0.30

0.0

2.0

4.0

6.0

8.0

10.0 -6.0 -4.0 -2.0 0.0

Fig. 26. PTXF plots using xenocrysts data for Guinea obtained using monomineral (Ashchepkov et al., 2010) thermobarometry. Symbols are the same as for Fig. 5.

those from Venetia pipe, showing a linear trend to the level separating the lower and upper parts of the lithospheric mantle sections possibly marking the diapiric trend. Most pyroxenes form a dense cluster −3 to 0 log units ΔQMF and the trend of the increasing fO2 in upper part. We can conclude from these comparisons that the deeper levels of the SCLM beneath Africa, estimated according to PT from mantle inclusions, are quite variable. Proterozoic pipes reflect HT conditions corresponding to SCLM heating by hot plume melts which crystallized hot pyroxenites and HT (remelted) eclogites. The generally heated mantle lithosphere beneath Lesotho shows the 40 mW/m 2 gradients obtained with Opx-based barometer (Boyd and Nixon, 1975; McGregor, 1974) and most other methods. But ancient diamond-bearing dunites and harzburgites still retain LT conditions corresponding to the ancient gradients, possibly due to the lack of clinopyroxenes in the sub-calcic rocks. The double or even triple 0

T oC 1

SEA

2

3

Grap hit Diam

4

e

ond

b a th Ma n tle A d ia

2005) nearly coincide with 40 mW/m2 geotherm, while the others show hotter conditions (Fig. 23). The fO2 for the eclogites in SCLM of Roberts Victor show two tendencies — the lowest values ~−4 log units coincide with the values determined for the pyropes. But most of them are located just near the line of DSF together with the most lherzolitic Cpx and garnet values which are slightly lower. In the SCLM beneath the Roberts Victor pipe the common eclogites are more oxidized than diamond-bearing. Colder and deeper conditions are revealed by mantle xenoliths from the Zimbabwe craton. It represents one of the ancient depleted cores according to the model of O'Reilly et al. (2009). The central part of the craton is more depleted (Smith et al., 2009) and shows colder conditions while the outer are possibly highly modified by melt percolation (Griffin and O'Reilly, 2007). Geotherms for the SCLM beneath the Venetia pipe (Fig. 24) (Aulbach et al., 2002; Hin et al., 2009; Richardson et al., 2009) in the Zimbabwe craton also show three geotherms reaching depths of 7.5 GPa (240 km). The oxidation state and trend of the eclogites and omphacites from diamond inclusions of Venetia pipes are similar, showing the increasing incline line from SCLM base to pyroxenite lens and the scattering to the lower values within 5–7 GPa interval. In Namibia (Boyd et al., 2004; Franz et al., 1996; 1997; Harris et al., 2004) (Figs. 25–26) the PT estimates for peridotites from the SCLM mainly correspond to conditions close to the Gr–Di boundary. Even eclogites from Gibeon province show the same pressures (Schmädicke et al., 2011) but lower temperatures. Some PT points for diamond inclusions of OPx, Omph and pyrope mark the level near 7.0 GPa (230 km) and deeper (Fig. 22). They are found in diamonds from a placers and this is a question for what time they (Harris et al., 2004). It was suggested that perturbations took place in an intermediate magmatic chamber (Franz et al., 1997). Another possibility is hydraulic fracturing and deformation at the top and near the magmatic chambers which sometimes took place in several levels. But the presence of the deeper material may suggest an additional mechanism like upward focused flows (Scambelluri et al., 2010). This process may be accompanied by the movement of the solid- and fluid mixed matter within the contacts of the protokimberlite channels which are possibly represented by the polymict peridotites (Pokhilenko et al., 2009). One can suggest more widescale upwelling in the permeable zones (Padrón-Navarta et al., 2010) similar to the processes that produced the uprise of Ronda or Beni Bousera peridotites (Crespo et al., 2006). This solves the problem between rather shallow pressure conditions of xenoliths which definitely reflect adiabatic ascent and more deep-seated roots corresponding to eclogites, dunites and pyroxenites found mainly as diamond inclusions (Leost et al., 2003; Harris et al., 2004). For the Angolan pipes the HT branch is marked by Cr-poor associations of pyroxenes tracing the PT branch of the protokimberlite channels. The SCLM values of the garnets from Namibia are very similar to

5

Upper SCLM 45 mw/m2

Lower SCLM

6

Shearing

40 mw/m2

7

8

MBL

9

35 mw/m2 10

11

P(GPa)

33 mw/m2

12

13 600

800

1000

1200

1400

Fig. 27. Schemes of determinations of lithosphere thickness by crossing with mantle adiabat A) (Artemieva, 2009). Combined data for Finsch, DeBeers Pool, Kao, Letseng la Terai pipes. See Figs. 14,15,16,18.

148

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

geotherms (Figs. 16–25) marked by diamond inclusions possibly are evidence of destruction of the lithospheric keel from 270 to 230 km (See Fig. 27). The off-craton kimberlites (Boyd et al., 2004; Franz et al., 1996; 1997; Leost et al., 2003 ) have passed through less stable lithosphere which possibly allowed diapiric ascent of peridotites heated by the protokimberlite melts from the base of the lithosphere. But the deeper level still exists at the base of SCLM near 6.0 GPa (200 km) and deeper. The monomineral thermobarometry possibly yields splitting for garnet lherzolites due to disequilibrium between garnets and pyroxenes (Fraser and Lawless, 1978). The Congo–Kasai craton reveals thicker and colder mantle lithosphere than the SCLM beneath Lesotho. But the composition of the mantle should be relatively less depleted than the Udachnaya lithosphere in Yakutia (Boyd et al., 1997; Ionov et al., 2010) and other northern continents. Recent mantle seismic tomography data give evidence about the polyasthenospheric mantle structure (Fishwick, 2010). This explains why some pipes contain xenocrysts from levels deeper than the location of sheared peridotites corresponding to 5.5–6.5 GPa (170– 215 km). Of course this boundary has a physical nature and reflects the stress-related zone obtained locally possibly due to magmatic fracturing (Katayama et al., 2009). The polyasthenospheric model is proved by monomineral thermobarometry not only for the Kaapvaal craton and presence of rare ultra-deep xenoliths (Macdougall and Haggerty, 1999). It possible also to suggest this for the Congo–Kasai craton because the common peridotites according to pyrope garnets are documented to 50–55 GPa (165–180 km) (Egorov et al., 2007), but judging by the xenocrysts with 14–15 wt.% Cr2O3 deeper peridotite roots should also exist. The PT conditions for the mantle xenocrysts from Guinea (Nikolenko et al., 2010) also show a great SCLM thickness and similar PT gradients and close to 36 mW/m 2 for estimates from the garnets which show several trends of heating. They are close to the PT trends for the ilmenites and Fe-rich Cr-bearing pyroxenes. Ti-chromites correspond to the same level. The fO2 values of the garnets show several increasing trends, possibly marking five separate levels in the lower part of the lithospheric mantle section. Clinopyroxene values in general indicate refertilization, yielding values slightly lower and sub-parallel to the DSF fluctuating near − 2 log units ΔQMF and chromites mark slightly higher values between the ilmenite and CPx arrays. The higher scatter is found in the upper part of the SCLM. The thick lithospheric roots like those beneath the Congo–Kasai craton (O'Reilly et al., 2009) correspond to a colder geotherm than those determined for Lesotho. But even in such cratons the rifting zones and peripheral so-called off-craton (Franz et al., 1996) areas show heated geotherms. The PT data for peridotites are located near the Gr–Di boundary (Boyd et al., 2004), possibly due to diapiric upwelling. Nevertheless, colder and deeper relics of lithospheric mantle can be found as diamond inclusions. Monominineral thermobarometry in this case marks the advective branches formed most likely by diapiric upwelling or by interaction with plume melts usually starting from the 4.0 or 7.0 GPa (130–230 km). 9. Conclusions 1. Mantle lithosphere beneath the NE part of the Congo–Kasai craton is composed largely of fertile peridotite according to the lherzolitic trend of garnets. Only the mantle column beneath the largest Catoca pipe contains abundant sub-calcic garnets from dunite and harzburgite, which are the magma conduits and are responsible for the high diamond potential. 2. The thick lithospheric mantle beneath the SW part of the Congo– Kasai craton is colder than in South Africa showing a PT gradient in the SCLM close to 36–38 mW/m 2.

3. The mantle sections beneath the Camitongo I and II pipes represent two stages of development of protokimberlite melts with migration of the magma differentiation level from the lower part to 4.0 GPa (130 km) which was influential in capturing the mantle xenoliths. The more depleted and eclogite-rich mantle column beneath the Catoca and Caquele pipes is the reason for their higher diamond potential. Acknowledgments To ALROSA Company which provided the concentrate of heavy minerals. To the staff of EPMA and ICP MS laboratories of the IGM SB RAS. Appendix A The statistical between regression obtained from the work (Gudmundsson and Wood, 1995) and corrections for the temperature and pressure justified by the comparisons obtained with the Ol–Sp and Ilm–Ol oxybarometers (Taylor et al., 1998) allow to estimate the fO2 (Δ log QMF) by following simple equations: 



F5 ¼ Fe3Ga=FeGar; Fo2 ¼ 2030:2 F5ˆ3−1061:4 F5ˆ2  þ 190:89 F5−12:644
  Fo2 ¼ Fo2−0:075 P þ ðT0−500Þ=2500−0:5 0:55 For clinopyroxene the cross calibration of Fe 3 + in CPx and fO2 for Gar allow to receive the following regression. Fo2 ¼ −186:71  Fe3Cpx  2 þ 48:617  Fe3Cpx−2:3262 Fo2 ¼ Fo2 þ ðT0−500Þ=3500−0:01  P Fo2 ¼ ðFo2−0:5Þ  0:9 For orthopyroxene the cross calibration of Fe 3 + in OPx and fO2 for Gar allow to receive the following regression. Fo2 ¼ 23:882  Fe3Opx  ðFeOpx  15Þ  2−1:8805 Fo2 ¼ Fo2 þ ððT0−400Þ=1000Þ  ðFeOpx  20Þˆ 2−0:0175  P Fo2 ¼ ðFo2  ðFeOpx  15Þ  2−0:9  P=70Þ  0:9 Fo2 ¼ ðFo2−0:5Þ  0:9

Appendix B. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.tecto.2011.12.007. References Appleyard, C.M., Viljoen, K.S., Dobbe, R., 2004. A study of eclogitic diamonds and their inclusions from the Finsch kimberlite pipe, South Africa. Lithos 77, 317–332. Artemieva, I.M., 2009. The continental lithosphere: reconciling thermal, seismic, and petrologic data. Lithos 109, 23–46. Ashchepkov, I.V., 2006. Empirical garnet thermobarometry for mantle peridotite. Russian Geology and Geophysics 47, 1071–1085. Ashchepkov, I.V., Vladykin, N.V., Mitchell, R.H., Coppersmith, H., Garanin, V.G., Saprykin, A.I., Khmelnikova, O.S., Anoshin, G.N., 2001. Mineralogy of the mantle xenocrysts from the Colorado kimberlite KL-1. Revista Brasileira de Geociencias 31 (4), 545–552. Ashchepkov, I.V., Vladykin, N.V., Nikolaeva, I.V., Palessky, S.V., Logvinova, A.M., Saprykin, A.I., Khmel'nikova, O.S., Anoshin, G.N., 2004. Mineralogy and geochemistry of mantle inclusions and mantle column structure of the Yubileinaya kimberlite pipe, 395. Doklady RAN. ESS, Alakit Field, Yakutia, pp. 517–523. Ashchepkov, I.V., Pokhilenko, N.P., Vladykin, N.V., Logvinova, A.M., Kostrovitsky, S.I., Afanasiev, V.P., Pokhilenko, L.N., Kuligin, S.S., Malygina, L.V., Alymova, N.V.,

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 Khmelnikova, O.S., Palessky, S.V., Nikolaeva, I.V., Karpenko, M.A., Stagnitsky, Y.B., 2010. Structure and evolution of the lithospheric mantle beneath Siberian craton, thermobarometric study. Tectonophysics 485, 17–41. Ashchepkov, I.V., André, L., Downes, H., Belyatsky, B.A., 2011. Pyroxenites and megacrysts from Vitim picrite-basalts (Russia): polybaric fractionation of rising melts in the mantle? Journal of Asian Earth Sciences 42, 14–37. Aulbach, S., Stachel, T., Viljoen, K.S., Brey, G.P., Harris, J.W., 2002. Eclogitic and websteritic diamond sources beneath the Limpopo Belt — is slab-melting the link. Contributions to Mineralogy and Petrology 143, 56–70. Aulbach, S., Pearson, N.J., O'Reilly, S.Y., Doyle, B.J., 2007. Origins of xenolithic eclogites and pyroxenites from the Central Slave Craton, Canada. Journal of Petrology 48, 1843–1873. Banas, A., Stachel, T., Phillips, D., Shimizu, N., Viljoen, K.S., Harris, J.W., 2009. Ancient metasomatism recorded by ultra-depleted garnet inclusions in diamonds from DeBeers Pool, South Africa. Lithos 112 (S2), 736–746. Batumike, J.M., Griffin, W.L., O'Reilly, S.Y., 2009. Lithospheric mantle structure and the diamond potential of kimberlites in southern D.R. Congo. Lithos 112 (S1), 166–176. Beard, B.L., Fraracci, K.N., Taylor, L.A., Snyder, G.A., Clayton, R.N., Mayeda, T.K., Sobolev, N.V., 1996. Petrography and geochemistry of eclogites from the Mir kimberlite, Yakutia, Russia. Contributions to Mineralogy and Petrology 125, 293–310. Bell, D.R., Moore, R.O., 2004. Deep chemical structure of the southern African mantle from kimberlite megacrysts. South African Journal of Geology 107, 219–223. Bell, D.R., Schmitz, M.D., Janney, P.E., 2003. Mesozoic thermal evolution of the southern African mantle lithosphere. Lithos 71, 273–287. Boyd, F.R., 1973. A pyroxene geotherm. Geochimica et Cosmochimica Acta 37, 2533–2546. Boyd, F.R., Danchin, R.V., 1980. Lherzolites, eclogites and megacrysts from some kimberlites of Angola. American Journal of Science 280A, 528–549. Boyd, F.R., Nixon, P.H., 1975. Origins of the ultramafic nodules from some kimberlites of northern Lesotho and the Monastery Mine, South Africa. Physics and Chemistry of the Earth 9, 432–454. Boyd, F.R., Nixon, P.H., Boctor, N.Z., 1984. Rapidly crystallised garnet pyroxenite xenoliths possibly related to discrete nodules. Contributions to Mineralogy and Petrology 86, 119–130. Boyd, F.R., Pokhilenko, N.P., Pearson, D.G., Mertzman, S.A., Sobolev, N.V., Finger, L.W., 1997. Composition of the Siberian cratonic mantle: evidence from Udachnaya peridotite xenoliths. Contributions to Mineralogy and Petrology 128, 228–246. Boyd, F.R., Pearson, D.G., Hoal, K.O., Hoal, B.G., Nixon, P.H., Kingston, M.J., Mertzman, S.A., 2004. Garnet lherzolites from Louwrensia, Namibia: bulk composition and P/ T relations. Lithos 77, 573–592. Braun, M.G., Kelemen, P.B., 2002. Dunite distribution in the Oman ophiolite: implications for melt flux through porous dunite conduits. Geochemistry, Geophysics, Geosystems (G-cubed), 2001GC000289. Brey, G., Kohler, T., 1990. Geothermometry in four phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378. Burgess, S.R., Harte, B., 1999. Tracing lithosphere evolution through the analysis of heterogeneous G9/G10 garnets in peridotite xenoliths, I: major element chemistry. In: Gurney, J.J., et al. (Ed.), Proceedings of the VII th International Kimberlite Conference. : Red Roof Design. Cape Town, South Africa, pp. 66–80. Cartigny, P., Harris, J.W., Javoy, M., 1999. Eclogitic, peridotitic and metamorphic diamonds and the problems of carbon recycling — the case of Orapa (Botswana). In: Gurney, J.J., Pascoe, M.D., Richardson, S.H. (Eds.), Proceedings of the VIIth International Kimberlite Conference, 1. Red Roof Design, South Africa, pp. 117–124. Correia, E.A., Laiginhas, F.A., 2006. Garnets from the Camafuca–Camazambo kimberlite (Angola). Anais da Academia Brasileira de Ciências 78, 309–315. Crespo, E., Luque, F.J., Rodas, M., Wada, H., Gervilla, F., 2006. Graphite–sulfide deposits in Ronda and Beni Bousera peridotites (Spain and Morocco) and the origin of carbon in mantle-derived rocks. Gondwana Research 9, 279–290. Davies, G.R., Spriggs, A.J., Nixon, P.H., 2001. A non-cognate origin for the Gibeon kimberlite megacryst suite, Namibia: implications for the origin of Namibian kimberlites. Journal of Petrology 42, 159–172. Dawson, J.B., Stephens, W.E., 1975. Statistical classification of garnets from kimberlite and associated xenoliths. Journal of Geology 83, 589–607. De Bruin, D., 2005. Multiple compositional megacryst groups from the Unitjiesberg and Witberg kimberlites, South Africa. South African Journal of Geology 108, 233–246. Deines, P., Harris, J.W., 2004. New insights into the occurrence of 13C-depleted carbon in the mantle from two closely associated kimberlites: Letlhakane and Orapa, Botswana. Lithos 77, 125–142. Deines, P., Gurney, J.J., Harris, J.W., 1984. Associated chemical and carbon isotopic composition variations in diamonds from the Finsch and Premier kimberlites, South Africa. Geochimica et Cosmochimica Acta 48, 325–342. Deines, P., Viljoen, F., Harris, J.W., 2001. Implications of the carbon isotope and mineral inclusion record for the formation of diamonds in the mantle underlying a mobile belt: Venetia. South Africa Geochimica et Cosmochimica Acta 65, 813–838. Deines, P., Stachel, T., Harris, J.W., 2009. Systematic regional variations in diamond carbon isotopic composition and inclusion chemistry beneath the Orapa kimberlite cluster, in Botswana. Lithos 112 (S2), 776–784. 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 Geology and Geophysics 48, 323–336. Eley, R., Grütter, H., Louw, A., Tunguno, C., Twidale, J., 2008. Luxinga kimberlite cluster (Angola) with evidence supporting the presence of kimberlite lava. Extended Abstracts of the Ninth International Kimberlite Conference, Frankfurt, Germany, 9IKC-A-00166.

149

Fishwick, S., 2010. Surface wave tomography: Imaging of the lithosphere–asthenosphere boundary beneath central and southern Africa? Lithos 120, 63–73. Franz, L., Brey, G., Okrusch, M., 1996. Steady state geotherm, thermal disturbances and tectonic development of the lower lithosphere underneath the Gibeon Kimberlite Province, Namibia. Contributions to Mineralogy and Petrology 126, 181–198. Franz, L., Brey, G.P., Okrusch, M., 1997. Metasomatic reequilibration of mantle xenoliths from the Gibeon Kimberlite Province, Namibia. Russian Geology and Geophysics 38 (1), 245–259. Fraser, D.G., Lawless, P.J., 1978. Palaeogeotherms — implications of disequilibrium in garnet lherzolite xenoliths. Nature 273, 220–222. Ganga, J., Rotman, A.Ya., Nosiko, S., 2003. Pipe Catoca, an example of the weakly eroded kimberlites from north-east of Angola. 8th International Kimberlite Conference Long Abstract. FLA312. 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. Contributions to Mineralogy and Petrology 85, 133–140. Gibson, S., Malarkey, J., Day, J.A., 2008. Melt depletion and enrichment beneath the western Kaapvaal Craton: evidence from Finsch peridotite xenoliths. Journal of Petrology 49, 1915–1929. Gréau, Y., Huang, J.-X., Griffin, W.L., Renac, C., Alard, O., O'Reilly, S.Y., 2011. Type I eclogites from Roberts Victor kimberlites: products of extensive mantle metasomatism. Geochimica et Cosmochimica Acta 75, 6927–6954. Gregoire, M., Bell, D.R., Le Roex, A.P., 2003. Garnet Lherzolites from the Kaapvaal Craton (South Africa): trace element evidence for a metasomatic history. Journal of Petrology 44, 629–657. Grégoire, M., Bell, D.R., Le Roux, A.P., 2002. Trace element geochemistry of glimmerite and MARID mantle xenoliths: their classification and relationship to phlogopitebearing peridotites and to kimberlites revisited. Contributions to Mineralogy and Petrology 142, 603–625. Grégoire, M., Tinguely, C., Bell, D.R., le Roex, A.P., 2005. Spinel lherzolite xenoliths from the Premier kimberlite (Kaapvaal craton, South Africa): nature and evolution of the shallow upper mantle beneath the Bushveld complex. Lithos 84, 185–205. Griffin, W.L., O'Reilly, S.Y., 2007. Cratonic lithospheric mantle: is anything subducted? Episodes 30 (1), 43–53. Griffin, W.L., Ryan, C.G., 1996. An experimental calibration of the “nickel in garnet” geothermometer with applications, by D. Canil: discussion. Contributions to Mineralogy and Petrology 124, 216–218. Griffin, W.L., Cousens, D.R., Ryan, C.G., Sie, S.H., Suter, G.F., 1989. Ni in chrome pyrope garnets: a new geothermometer. Contributions to Mineralogy and Petrology 103, 199–202. Griffin, W.L., Moore, R.O., Ryan, C.G., Gurney, J.J., Win, T.T., 1997. Geochemistry of magnesian ilmenite megacrysts from southern African kimberlites. Russian Geology and Geophysics 38 (2), 421–443. Griffin, W.L., Shee, S.R., Ryan, C.G., Win, T.T., Wyatt, B.A., 1999a. Harzburgite to lherzolite and back again: metasomatic processes in ultramafic xenoliths from the Wesselton kimberlite, Kimberley, South Africa. Contributions to Mineralogy and Petrology 134, 232–250. Griffin, W.L., Fisher, N.I., Friedman, J., Ryan, C.G., O'Reilly, S.Y., 1999b. Cr–pyrope garnets in the lithospheric mantle. I. Compositional systematics and relations to tectonic setting. Journal of Petrology 40, 679–704. Griffin, W.L., Fisher, N.I., Friedman, J.H., O'Reilly, S.Y., Ryan, C.G., 2002. Cr–pyrope garnets in the lithospheric mantle 2. Compositional populations and their distribution in time and space. Geochemistry, Geophysics, Geosystems 12. Griffin, W.L., O 'Reilly, S.Y., Abe, N., Aulbach, S., Davies, R.M., Pearson, N.J., Doyle, B.J., Kivi, K., 2003a. The origin and evolution of Archean lithospheric mantle. Precambrian Research 127, 19–41. Griffin, W.L., O'Reilly, S.Y., Natapov, L.M., Ryan, C.G., 2003b. The evolution of lithospheric mantle beneath the Kalahari Craton and its margins. Lithos 71, 215–241. Griffin, W.L., Graham, S., O'Reilly, S.Y., Pearson, N.J., 2004. Lithosphere evolution beneath the Kaapvaal Craton: Re–Os systematics of sulfides in mantle-derived peridotites. Chemical Geology 208, 89–118. Grütter, H.S., Quadling, K.E., 1999. Can sodium in garnet be used to monitor eclogitic diamond potential? In: Gurney, J.J., Gurney, J.L., Pascoe, M.D., Richardson, S.H. (Eds.), J. B. Dawson volume, Proceedings of the 7th International Kimberlite Conference. Red Roof Design, Cape Town, pp. 314–320. Grütter, H.S., Gurney, J.J., Menzies, A.H., Winter, F., 2004. An updated classification scheme for mantle-derived garnets, for use by diamond explorers. Lithos 77, 841–857. Gudmundsson, G., Wood, B.J., 1995. Experimental tests of garnet peridotite oxygen barometry. Contributions to Mineralogy and Petrology 119, 56–67. Gurney, J.J., Switzer, G.S., 1973. The discovery of garnets closely related to diamonds in the Finsch Pipe, South Africa. Contributions to Mineralogy and Petrology 39, 103–116. Gurney, J.J., Jacob, W.R.O., Dawson, J.B., 1979a. Megacrysts from the Monastery kimberlite pipe, South Africa. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the 2nd International Kimberlite Conference: American Geophysical Union, 2, pp. 227–243. Gurney, J.J., Harris, J.W., Rickard, R.S., 1979b. Silicate and oxide inclusions in diamonds from the Finsch kimberlite pipe. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Kimberlites, Diatremes and Diamonds, their Geology, Petrology and Geochemistry. American Geophysical Union, Washington, pp. 1–15. Gurney, J.J., Harris, J.W., Rickard, R.S., Moore, R.O., 1985. Inclusions in Premier Mine diamonds. Transactions of the Geological Society of South Africa 88, 301–310. Haggerty, S., Hardie, R.B., McMahon, B.M., 1979. The mineral chemistry of the ilmenite nodule associations from the Monastery Diatreme. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the Second International Kimberlite Conference: American

150

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151

Geophysical Union: The Mantle Sample: Inclusions in Kimberlites and Other Volcanics, Vol. 2, pp. 249–256. Harris, J.W., Stachel, T., Léost, I., Brey, G.P., 2004. Peridotitic diamonds from Namibia: constraints on the composition and evolution of their mantle source. Lithos 77, 209–223. Hatton, C.J., Gurney, J.J., 1979. Diamond graphite eclogite from the Roberts Victor Mine. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the 2nd International Kimberlite Conference: American Geophysical Union, 2, pp. 29–36. Hin, R.C., Morel, M.L.A., Nebel, O., Mason, P.R.D., van Westrenen, W., Davies, G.R., 2009. Formation and temporal evolution of the Kalahari sub-cratonic lithospheric mantle: constraints from Venetia xenoliths, South Africa. Lithos 112 (S2), 1069–1082. Ionov, D.A., Doucet, L.S., Ashchepkov, I.V., 2010. Composition of the Lithospheric Mantle in the Siberian Craton: new Constraints from Fresh Peridotites in the Udachnaya-East Kimberlite. Journal of Petrology 51, 2177–2210. Jacob, D.E., Schmickler, B., Schulze, D.J., 2003. Trace element geochemistry of coesitebearing eclogites from the Roberts Victor kimberlite, Kaapvaal Craton. Lithos 71, 337–351. Jacob, D.E., Bizimis, M., Salters, V.J.M., 2005. Lu–Hf and geochemical systematics of recycled ancient oceanic crust: evidence from Roberts Victor eclogites. Contributions to Mineralogy and Petrology 148, 707–720. Jacob, D.E., Viljoen, K.S., Grassineau, N.V., 2009. Eclogite xenoliths from Kimberley, South Africa — a case study of mantle metasomatism in eclogites. Lithos 112 (S2), 1002–1013. Katayama, I., Suyama, Y., Ando, S., Komiya, T., 2009. Mineral chemistry and P–T condition of granular and sheared peridotite xenoliths from Kimberley, South Africa: origin of the textural variation in the cratonic mantle. Lithos 109, 333–340. Kennedy, C.S., Kennedy, G.C., 1976. The equilibrium boundary between graphite and diamond. Journal of Geophysical Research 81, 2467–2470. Kopylova, M.G., Gurney, J.J., Daniels, L.M., 1997. Mineral inclusions in diamonds from the River Ranch kimberlite, Zimbabwe. Contributions to Mineralogy and Petrology 129, 366–384. Kopylova, M.G., Nowell, G.M., Pearson, D.G., Markovic, G., 2009. Crystallization of megacrysts from protokimberlitic fluids: geochemical evidence from high-Cr megacrysts in the Jericho kimberlite. Lithos 112 (S1), 284–295. Krogh, E.J., 1988. The garnet-clinopyroxene Fe–Mg thermometer — a reinterpretation of existing experimental data. Contributions to Mineralogy and Petrology 99, 44–48. Lavrent'ev, Yu.G., Usova, L.V., 1994. New version of KARAT program for quantitative X-ray spectral microanalysis. Zhurnal Analiticheskoi Khimii 5, 462–468. Lavrentyev, Y.G., Usova, L.V., Kuznetsova, A.I., Letov, S.V., 1987. X-ray spectral quant metric microanalysis of the most important minerals of kimberlites. Russian Geology and Geophysics 48 (5), 75–81. Lazarov, M., Woodland, A.B., Brey, G.P., 2009. Thermal state and redox conditions of the Kaapvaal mantle: a study of xenoliths from the Finsch mine, South Africa. Lithos 112 (S1), 913–923. Leost, I., Stachel, T., Brey, G.P., Harris, J.W., Ryabchikov, I.D., 2003. Diamond formation and source carbonation: mineral associations in diamonds from Namibia. Contributions to Mineralogy and Petrology 145, 12–24. Macdougall, J.D., Haggerty, S.E., 1999. Ultradeep xenoliths from African kimberlites: Sr and Nd isotopic compositions suggest complex history. Earth and Planetary Science Letters 170, 73–82. MacGregor, I.D., 1979. Mafic and ultramafic xenoliths from Kao kimberlite pipe. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the 2nd International Kimberlite Conference: American Geophysical Union, 2, pp. 156–172. McCammon, C.A., Kopylova, M.G., 2004. A redox profile of the Slave mantle and oxygen fugacity control in the cratonic mantle. Contributions to Mineralogy and Petrology 148, 55–68. McCammon, C.A., Griffin, W.L., Shee, S.R., O'Neill, H.S.C., 2001. Oxidation during metasomatism in ultramafic xenoliths from the Wesselton kimberlite, South Africa: implications for the survival of diamond. Contributions to Mineralogy and Petrology 141, 287–296. McCandless, T.E., Gurney, J.J., 1989. Sodium in garnet and potassium in clinopyroxene: criteria for classifying mantle eclogites. In: Ross, J. (Ed.), Their mantle/crust setting, diamonds and diamond exploration. : Kimberlites and Related Rocks, Vol 2. Blackwell, Carlton, pp. 827–832. McDade, P., Harris, J.W., 1999. In: Gurney, J.J., Gurney, J.L., Pascoe, M.D., Richardson, S.H. (Eds.), Syngenetic inclusion diamonds from Letseng-la-terai, Lesotho. Proceedings of the VIIth International Kimberlite Conference, 2. Red Roof Design, South Africa, pp. 557–565. McGregor, I.D., 1974. The system MgO–SiO2–Al2O3: solubility of Al2O3 in enstatite for spinel and garnet peridotite compositions. American Mineralogist 59, 110–119. Moore, A.E., 1987. A model for the origin of ilmenite in kimberlite and diamond: implications for the genesis of the discrete nodule (megacryst) suite. Contributions to Mineralogy and Petrology 95, 245–253. Moore, A., Belousova, E., 2005. Crystallization of Cr-poor and Cr-rich megacryst suites from the host kimberlite magma: implications for mantle structure and the generation of kimberlite magmas. Contributions to Mineralogy and Petrology 49, 462–481. Moore, A.E., Lock, N.P., 2001. The origin of mantle-derived megacrysts and sheared peridotites — evidence from kimberlites in the northern Lesotho–Orange Free State (South Africa) and Botswana pipe clusters. South Africa Journal of Geology 104, 23–38. Moore, R.O., Griffin, W.L., Gurney, J.J., Ryan, C.G., Cousens, D.R., Sie, S.H., Suter, G.F., 1992. Trace element geochemistry of ilmenite megacrysts from the Monastery kimberlite, South Africa. Lithos 29, 1–18.

Morimoto, N., Fabries, J., Ferguson, A.K., Ginzburg, I.V., Ross, M., Seifert, F.A., Zussman, J., Aoki, K., Gottardi, G., 1988. Nomenclature of pyroxenes. Mineralogical Magazine 52, 535–550. Nikolenko, E.I., Afanas'ev, V.P., Pokhilenko, N.P., 2010. Peculiarities of the composition of zoned picroilmenites from the Massadou field (Giunea) and Dachnaya pipe (Yakutia) kimberlites. Doklady Earth Sciences 434, 1386–1389. Nimis, P., Taylor, W., 2000. Single clinopyroxene thermobarometry for garnet peridotites. Part I. Calibration and testing of a Cr-in-Cpx barometer and an enstatite-in-Cpx thermometer. Contributions to Mineralogy and Petrology 139, 541–554. Nixon, P.H., 1973. Lesotho Kimberlites. Cape and Transvaal, Cape Town, pp. 48–56. Nixon, P.H., 1987. Mantle Xenoliths. Wiley, Chichester. 844 pp. Nixon, P.H., Boyd, F.R., 1973. Petrogenesis of the granular and sheared ultrabasic nodule suite in kimberlite. In: Nixon, P.H. (Ed.), Lesotho Kimberlites. Cape and Transvaal, Maseru, pp. 48–56. Nixon, P.H., Rogers, N.W., Gibson, I.L., Grey, A., 1981. Depleted and fertile mantle xenoliths from Southern African kimberlites. Annual Review of Earth and Planetary Sciences 9, 285–309. O'Neill, H.St.C, Wood, B.J., 1979. An experimental study of Fe–Mg-partitioning between garnet and olivine and its calibration as a geothermometer. Contributions to Mineralogy and Petrology 70, 59–70. O'Neill, H.St.C, Wall, V.J., 1987. The olivine orthopyroxene-spinel oxygen geobarometer, the nickel precipitation curve, and the oxygen fugacity of the Earth's upper mantle. Journal of Petrology 28, 1169–1191. O'Reilly, S.Y., Zhang, M., Griffin, W.L., Begg, G., Hronsky, J., 2009. Ultradeep continental roots and their oceanic remnants: a solution to the geochemical ‘mantle reservoir’ problem? Lithos 112 (S1), 1043–1054. Padrón-Navarta, J.A., Tommasi, A., Garrido, C.J., Sánchez-Vizcaíno, V.L., GómezPugnaire, M.T., Jabaloy, A., Vauchez, A., 2010. Fluid transfer into the wedge controlled by high-pressure hydrofracturing in the cold top-slab mantle. Earth and Planetary Science Letters 297, 271–286. Pedersen, H.A., Fishwick, S., Snyder, D.B., 2009. A comparison of cratonic roots through consistent analysis of seismic surface waves. Lithos 109, 81–89. Pervov, A., Somov, S.V., Korshunov, A.V., Dulapchii, E.V., Félix, J.T., 2011. The Catoca kimberlite pipe, Republic of Angola: a paleovolcanological model. Geology of Ore Deposits 53, 295–308. Pettit, W., 2009. Geophysical signatures of some recently discovered large (> 40 ha) kimberlite pipes on the Alto Cuilo concession in northeastern Angola. Lithos 112 (S1), 106–115. Phillips, D., Harris, J.W., Viljoen, K.S., 2004. Mineral chemistry and thermobarometry of inclusions from De Beers Pool diamonds, Kimberley, South Africa. Lithos 77, 155–179. Pivin, M., Féménias, O., Demaiffe, D., 2009. Metasomatic mantle origin for Mbuji–Mayi and Kundelungu garnet and clinopyroxene megacrysts (Democratic Republic of Congo). Lithos 112S, 951–960. Pokhilenko, N.P., 2009. Polymict breccia xenoliths: evidence for the complex character of kimberlite formation. Lithos 112 (S2), 934–941. Pokhilenko, N.P., Sobolev, N.V., Sobolev, V.S., Lavrentiev, Y.G., 1976. Xenoliths of diamond bearing ilmenite–pyrope lherzolites from the kimberlite pipe Udachnaya (Yakutia). Doklady Akademii Nauk SSSR 231, 438–442. Pokhilenko, N.P., Sobolev, N.V., Kuligin, S.S., Shimizu, N., 1999. Peculiarities of distribution of pyroxenite paragenesis garnets in Yakutian kimberlites and some aspects of the evolution of the Siberian craton lithospheric mantle. Proceedings of the VII International Kimberlite Conference. The P.H. Nixon volume, pp. 690–707. Pollack, H.N., Chapman, D.S., 1977. On the regional variation of heat flow, geotherms, and lithospheric thickness. Tectonophysics 38, 279–296. Pyle, J.M., Haggerty, S.E., 1998. Eclogites and the metasomatism of eclogites from the Jagersfontein kimberlite: punctuated transport and implications for alkali magmatism. Geochimica et Cosmochimica Acta 62, 1207–1231. Rawlinson, P.J., Dawson, J.B., 1979. A quench pyroxene–ilmenite xenolith from kimberlite: implications for pyroxene–ilmenite intergrowths. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the 2nd International Kimberlite Conference: American Geophysical Union, 2, pp. 292–299. Reis, B., 1972. Preliminary note on the distribution and tectonic control of kimberlites in Angola. The 24th International Geological Congress — Section 4, 276–281. Richardson, S.H., Erlank, A.J., Hart, S.R., 1985. Kimberlite-borne garnet peridotite xenoliths from old enriched subcontinental lithosphere. Earth and Planetary Science Letters 75, 116–128. Richardson, S.H., Pöml, P.F., Shirey, S.B., Harris, J.W., 2009. Age and origin of peridotitic diamonds from Venetia, Limpopo Belt, Kaapvaal–Zimbabwe craton. Lithos 112S, 785–792. Robles-Cruz, S.E., Watangua, M., Isidoro, L., 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 112 (S2), 966–975. Rodionov, A.S., Amshinsky, A.N., Pokhilenko, N.P., 1988. Ilmenite–pyrope wehrlite — a new type of kimberlite xenoliths paragenesis. Russian Geology and Geophysics 19/ 7, 53–57. Rogers, A.J., Grütter, H.S., 2009. Fe-rich and Na-rich megacryst clinopyroxene and garnet from the Luxinga kimberlite cluster, Lunda Sul, Angola. Lithos 112 (S2), 942–950. Roman'ko, E.F., Egorov, K.N., Podvysotskii, V.T., Sablukov, S.M., D'yakonov, D.B., 2005. A new diamondiferous kimberlite region in southwestern Angola. Doklady Biological Sciences: Proceedings of the Academy of Sciences of the USSR, Biological Sciences Sections/Translated from Russian 403A (6), 817–821. Russell, J.K., Kopylova, M.G., 1999. A steady state conductive geotherm for the north central Slave, Canada: inversion of petrological data from the Jericho kimberlite pipe. Journal of Geophysical Research 104, 7089–7101.

I.V. Ashchepkov et al. / Tectonophysics 530–531 (2012) 128–151 Ryan, C.G., Griffin, W.L., Pearson, N.J., 1996. Garnet geotherms: pressure–temperature data from Cr–pyrope garnet xenocrysts in volcanic rocks. Journal of Geophysical Research, B: Solid Earth 101, 5611–5625. Sautter, V., Harte, B., 1990. Diffusion gradients in an eclogite xenolith from the Roberts Victor kimberlite pipe: (2) kinetics and implications for petrogenesis. Contributions to Mineralogy and Petrology 105, 637–649. Scambelluri, M., Van Roermund, H.L.M., Pettkec, T., 2010. Mantle wedge peridotites: fossil reservoirs of deep subduction zone processes: inferences from high and ultrahigh-pressure rocks from Bardane (Western Norway) and Ulten (Italian Alps). Lithos 120, 186–201. Schmädicke, E., Okrusch, M., Rupprecht-Gutowski, P., Will, T.M., 2011. Garnet pyroxenite, eclogite and alkremite xenoliths from the off-craton Gibeon Kimberlite Field, Namibia: a window into the upper mantle of the Rehoboth Terrane. Precambrian Research 191, 1–17. Schulze, D.J., 2003. A classification scheme for mantle-derived garnets in kimberlite: a tool for investigating the mantle and exploring for diamonds. Lithos 71, 195–213. Schulze, D.L., Anderson, P.F.N., Hearn Jr., B.C., Hetman, C.M., 1995. Origin and significance of ilmenite megacrysts and macrocrysts from kimberlite. International Geology Review 37, 780–812. Shee, S.R., Gurney, J.J., 1979. The mineralogy of xenoliths from Orapa, Botswana. In: Boyd, F.R., Meyer, H.O.A. (Eds.), Proceedings of the 2nd International Kimberlite Conference: American Geophysical Union, 2, pp. 37–49. Shee, S.R., Gurney, J.J., Robinson, D.N., 1982. Two diamond-bearing peridotite xenoliths from the Finsch Kimberlite, South Africa. Contributions to Mineralogy and Petrology 81, 79–87. Simon, N.S.C., Carlson, R.W., Pearson, D.G., Davies, G.R., 2007. The origin and evolution of the Kaapvaal cratonic lithospheric mantle. Journal of Petrology 48, 589–625. Smith, D., 1999. Temperatures and pressures of mineral equilibration in peridotite xenoliths: review discussion and implication. In: Fei, Y., Bertka, C.M., Mysen, B.O. (Eds.), Mantle Petrology: Field Observations and High Pressure Experimentation: A tribute to Francis R. (Joe) Boyd: Geochemical Society Special Publication, 6, pp. 171–188. Smith, C.B., Gurney, J.J., Harris, J.W., Otter, M.L., Kirkley, M.B., Jagoutz, E., 1991. Neodymium and strontium isotope systematics of eclogite and websterite paragenesis inclusions from single diamonds, Finsch and Kimberley Pool, RSA. Geochimica et Cosmochimica Acta 55, 2579–2590. Smith, C.B., Pearson, D.G., Bulanova, G.P., Beard, A.D., Carlson, R.W., Wittig, N., Sims, K., Chimuka, L., Muchemwa, E., 2009. Extremely depleted lithospheric mantle and diamonds beneath the southern Zimbabwe Craton. Lithos 112 (S2), 1120–1132. Sobolev, N.V., 1977. Deep seated Inclusions in Kimberlites and the Problem of the Composition of the Upper Mantle. AGU, Washington, DC. 350 pp. Sobolev, N.V., 1980. Significance of picroilmenite for the localization of kimberlite fields. Russian Geology and Geophysics 24 (5), 149–151. Sobolev, N.V., Lavrent'ev, Y.G., 1971. Isomorphic sodium admixture in garnets formed at high pressures. Contributions to Mineralogy and Petrology 31, 1–12. Sobolev, N.V., Lavrentev, Y.G., Pokhilenko, N.P., Usova, L.V., 1973. Chrome-rich garnets from the kimberlites of Yakutia and their parageneses. Contributions to Mineralogy and Petrology 40 (1), 39–52. Sobolev, N.V., Mankenda, A., Kaminsky, F.V., Sobolev, V.N., 1990. Garnets from kimberlites of northeastern Angola and their composition — diamond potential relationship. Doklady. AN SSSR ESS 315, 1225–1229. Sobolev, N.V., Snyder, G.A., Taylor, L.A., Keller, A., Yefimova, E.S., Sobolev, V.N., Shimizu, N., 1998. Extreme chemical diversity in the mantle during eclogitic diamond formation: evidence from 35 garnet and 5 pyroxene inclusions in a single diamond. International Geology Review 40, 567–578.

151

Stachel, T., Harris, J.W., 2008. The origin of cratonic diamonds — constraints from mineral inclusions. Ore Geology Reviews 34, 5–32. Stachel, T., Viljoen, K.S., McDade, P., Harris, J.W., 2004. Diamondiferous lithospheric roots along the western margin of the Kalahari Craton: the peridotitic inclusion suite in diamonds from Orapa and Jwaneng. Contributions to Mineralogy and Petrology 147, 32–47. Stiefenhofer, J., Viljoen, K.S., Marsh, J.S., 1997. Petrology and geochemistry of peridotite xenoliths from the Letlhakane kimberlites, Botswana. Contributions to Mineralogy and Petrology 127, 147–158. Tappe, S., Foley, S.F., Stracke, A., Romer, R.L., Kjarsgaard, B.A., Heaman, L.M., Joyce, N., 2007. Craton reactivation on the Labrador Sea margins: 40Ar/39Ar age and Sr– Nd–Hf–Pb isotope constraints from alkaline and carbonatite intrusives. Earth and Planetary Science Letters 256, 433–445. Tappert, R., Stachel, T., Harris, J.W., Muehlenbachs, K., Ludwig, T., Brey, G.P., 2005. Diamonds from Jagersfontein (South Africa): messengers from the sublithospheric mantle. Contributions to Mineralogy and Petrology 150, 505–522. Taylor, W.L., Kamperman, M., Hamilton, R., 1998. New thermometer and oxygen fugacity sensor calibration for ilmenite amd Cr–spinel-bearing peridotite assemblage. In: Gurney, J.J. (Ed.), 7th IKC Extended abstracts, Cape Town. FLA891. Tsai, H.-M., Meyer, H.O.A., Moreau, J., Milledge, J., 1979. Mineral inclusions in diamond:, Premier, Jagerfontein and Finsch kimberlites, South Africa, and Williams mine, Tanzania. Proceedings of the 2nd International Kimberlite Conference. American Geophysical Union 2, 16–26. van Achterbergh, E., Griffin, W.L., Stiefenhofer, J., 2001. Metasomatism in mantle xenoliths from the Letlhakane kimberlites: estimation of element fluxes. Contributions to Mineralogy and Petrology 141, 397–414. Viljoen, K.S., Swash, P.M., Otter, M.L., Schulze, D.J., Lawless, P.J., 1992. Diamondiferous garnet harzburgites from the Finsch kimberlite, Northern Cape, South Africa. Contributions to Mineralogy and Petrology 110, 133–138. Viljoen, K.S., Phillips, D., Harris, J.W., Robinson, D.N., Gurney, J.L., Gurney, J.J., Pascoe, M.D., Richardson, S.H., 1999. Mineral inclusions in diamonds from the Venetia kimberlites, Northern Province, South Africa. Proceedings of the 7th International Kimberlite Conference, the J. Barry Hawthorne volume. Red Roof Design cc, Cape Town, pp. 888–895. Viljoen, K.S., Schulze, D.J., Quadling, A.G., 2005. Contrasting Group I and Group II eclogite xenolith petrogenesis: petrological, trace element and isotopic evidence from eclogite, garnet–websterite and alkremite xenoliths in the Kaalvallei kimberlite, South Africa. Journal of Petrology 46, 2059–2090. Viljoen, F., Dobbe, R., Smit, B., 2009. Geochemical processes in peridotite xenoliths from the Premier diamond mine, South Africa: evidence for the depletion and refertilisation of subcratonic lithosphere. Lithos 112 (S2), 1133–1142. Winterburn, P.A., Harte, B., Gurney, J.J., 1990. Peridotite xenoliths from the Jagersfontein kimberlite pipe: I. Primary and primary-metasomatic mineralogy. Geochimica et Cosmochimica Acta 54, 329–341. Woodhead, J., Hergt, J., Phillips, D., Paton, C., 2009. African kimberlites revisited: In situ Sr-isotope analysis of groundmass perovskite. Lithos 112 (S1), 311–317. Woodland, A.B., Koch, M., 2003. Variation in oxygen fugacity with depth in the upper mantle beneath the Kaapvaal craton, Southern Africa. Earth and Planetary Science Letters 214, 295–310. Wyatt, B.A., Baumgartner, M., Anckar, E., Grutter, H., 2004. Compositional classification of “kimberlitic” and “non-kimberlitic” ilmenite. Lithos 77, 819–840. Zinchenko, V.N., 2008. Morphology of diamonds from kimberlite pipes of the Catoca field, Angola. Geology of Ore Deposits 50 (8), 086–816. Zuev, V.M., Khar'kiv, A.D., Zinchuk, N.N., Mankenda, A., 1988. Weakly eroded kimberlite pipes of Angola. Russian Geology and Geophysics 29, 56–62.