Phân tích Nhiệt động lực học Khoáng vật từ Cerro Galan và Cerro Blanco, Tây Bắc Argentina

THERMODYNAMIC ANALYSIS OF MINERALS FROM CERRO GALAN AND CERRO BLANCO, NW ARGENTINA by JOAO CARLOS LUNA GONZALEZ (Under the Direction of ALBERTO PATINO DOUCE) ABSTRACT Cerro Galan and Cerro Blanco are two large Plio-Pleistocene calderas from the Andes of NW Argentina located within 120 km of each other. Large ash flows erupted from both volcanoes are relatively-well preserved and minimally eroded. Multiple sections of ignimbrites from both volcanoes were sampled and form the basis of this study. Their mineralogical assemblage— biotite + plagioclase + sanidine + Fe-Ti oxides + quartz + apatite—makes them well suited for the estimation of pre-eruptive intensive variables (temperature, pressure) in large silicic sub-volcanic magma chambers.

The presence of minerals that incorporate F, Cl, and OH such as apatite and biotite also makes them ideal for studying the relationship between these fluids species as function of vertical location and temperature within the pre- eruptive magma chamber. Results indicate both calderas were at a minimum pressure of 2 kbar based on an equilibrium temperature of about 700 0C and 4% H2O saturation. INDEX WORDS: Cerro Galan, Cerro Blanco, Feldspar thermometry, Fe-Ti oxides, Halogens, Thermodynamics THERMODYNAMIC ANALYSIS OF MINERALS FROM CERRO GALAN AND CERRO BLANCO, NW ARGENTINA by JOAO CARLOS LUNA GONZALEZ B., Barry University, 2010 A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE ATHENS, GEORGIA 2013 © 2013 JOAO CARLOS LUNA GONZALEZ All Rights Reserved THERMODYNAMIC ANALYSIS OF MINERALS FROM CERRO GALAN AND CERRO BLANCO, NW ARGENTINA by JOAO CARLOS LUNA GONZALEZ Major Professor: ALBERTO PATINO Committee: DOUG CROWE MICHAEL RODEN Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia May 2013 DEDICATION This thesis is dedicated to my family and fiancée Hannah James for the constant support and words of encouragement. I would like to specially thank my father Ignacio Luna for his unwavering support and words of wisdom.

iv ACKNOWLEDGEMENTS I would like to acknowledge my advisor Alberto Patino Douce for all the help and guidance he has provided me during my graduate career. Thank you for your patience and for pushing me to become a better scientist. I would also like to acknowledge my committee members Doug Crowe and Michael Roden for their input on my research. Chris Fleisher is thanked for his help with microprobe analyses.

v TABLE OF CONTENTS Page ACKNOWLEDGEMENTS. v LIST OF TABLES. vii LIST OF FIGURES. viii CHAPTER 1 Introduction.

3 Geologic Setting and Petrology. 7 Summary of Investigation. 12 Results and Discussion. 24 3 Fe-Ti Oxides.

35 Results and Discussion. 60 Results and Discussion. 82 vi LIST OF TABLES Page Table 2.1: Section 1 Cerro Galan plagioclase compositions obtained by microprobe analyses .2: Section 1 Cerro Galan sanidine compositions obtained by microprobe analyses.3: Section 2 Cerro Galan plagioclase compositions obtained by microprobe analyses .4: Section 2 Cerro Galan sanidine compositions obtained by microprobe analyses.5: Cerro Blanco plagioclase compositions obtained by microprobe analyses .6: Cerro Blanco sanidine compositions obtained by microprobe analyses .1: Section 1 Cerro Galan ilmenite compositions obtained by microprobe analyses .2: Section 1 Cerro Galan magnetite compositions obtained by microprobe analyses .3: Section 2 Cerro Galan ilmenite compositions obtained by microprobe analyses .4: Section 2 Cerro Galan magnetite compositions obtained by microprobe analyses .5: Cerro Blanco ilmenite compositions obtained by microprobe analyses .6: Cerro Blanco magnetite compositions obtained by microprobe analyses .7: Section 1 Cerro Galan temperatures calculated at 1 and 5 kb .8: Section 2 Cerro Galan temperatures calculated at 1 and 5 kb .9: Cerro Blanco temperatures calculated at 1 and 5 kb.1: Section 1 Cerro Galan apatite compositions obtained by microprobe analyses .2: Section 2 Cerro Galan apatite compositions obtained by microprobe analyses .3: Cerro Blanco apatite compositions obtained by microprobe analyses .4: Section 1 Cerro Galan biotite compositions obtained by microprobe analyses .5: Section 2 Cerro Galan biotite compositions obtained by microprobe analyses .6: Cerro Blanco biotite compositions obtained by microprobe analyses. 66 vii LIST OF FIGURES Page Figure 1.1: Location map showing the extent of the high plateau of the Central Andes .2: Cross section through the Central Andes showing exaggerated topography and distribution of andesitic volcanoes, back arc centers and large silicic calderas .3: Regional view of the Cerro Galan and Cerro Blanco calderas .4: Distribution of the Cerro Galan ignimbrite .1: Ternary diagram of feldspar samples from section 1 of Cerro Galan .2: Ternary diagram of feldspar samples from section 2 of Cerro Galan .3: Ternary diagram of feldspar samples from Cerro Blanco .4: Feldspar ternary diagram modified from Fuhrman and Lindsley (1988) .5: Ternary diagram of feldspar samples from section 1 of Cerro Galan fitted to Hokada’s 2001 ternary feldspar model with calculated isotherms.6: Ternary diagram of feldspar samples from section 2 of Cerro Galan fitted to Hokada’s 2001 ternary feldspar model with calculated isotherms.7: Ternary diagram of feldspar samples from Cerro Blanco fitted to Hokada’s 2001 ternary feldspar model with calculated isotherms .8: SEM images of alkai-feldspar and plagioclase from section 1 of Cerro Galan.9: SEM images of alkai-feldspar and plagioclase from section 2 of Cerro Galan.10: SEM images of alkai-feldspar and plagioclase from Cerro Blanco .11: Pressure-temperature phase diagram showing the solidus curves for H2O-saturated and dry granite….12: H2O solubility in rhyolitic melts at 700 0C modified from Liu et al.13: Section 1 feldspar thermometry (A) Plus error and (B) Minus error .14: Section 2 feldspar thermometry (A) Plus error and (B) Minus error .15: Cerro Blanco thermometry (A) Plus error and (B) Minus error .1: Section 1 Cerro Galan Fe2+, Fe3+, and Ti in ilmenite vs.2: Section 1 Cerro Galan Fe2+, Fe3+, and Ti in magnetite vs.3: Section 2 Cerro Galan Fe2+, Fe3+, and Ti in ilmenite vs.4: Section 2 Cerro Galan Fe2+, Fe3+, and Ti in magnetite vs.5: Cerro Blanco Fe2+, Fe3+, and Ti in ilmenite vs.6: Cerro Blanco Fe2+, Fe3+, and Ti in magnetite vs.7: Real activity vs.

mole fractions for ilmenite-hematite (A & B) and magnetite-ulvospinel (C) solid solutions .8: Section 1 Cerro Galan ilmenite-magnetite pairs used for thermodynamic calculations.9: Section 2 Cerro Galan ilmenite-magnetite pairs used for thermodynamic calculations.10: Cerro Blanco ilmenite-magnetite pairs used for thermodynamic calculations .11: Scanning electron images of Fe-Ti oxides with alteration in Section 1 of Cerro Galan .12: Scanning electron images of Fe-Ti oxides with alteration in Section 2 of Cerro Galan .13: Scanning electron images of Fe-Ti oxides with alteration in Cerro Blanco .1: Section 1 Cerro Galan apatite anion concentration vs.2: Section 2 Cerro Galan apatite anion concentration vs.3: Cerro Blanco apatite anion concentration vs.4: Section 1 Cerro Galan biotite anion concentration vs.5: Section 2 Cerro Galan biotite anion concentration vs.6: Cerro Blanco biotite anion concentration vs.7: Section 1 Cerro Galan lnK values vs.8: Section 2 Cerro Galan lnK values vs.9: Cerro Blanco lnK values vs.10: SEM images of altered biotite grains from section 1 of Cerro Galan .11: SEM images of altered biotite grains from section 2 of Cerro Galan .12: SEM images of altered biotite grains from Cerro Blanco .13: SEM images of altered apatite grains from Section 1 of Cerro Galan .14: SEM images of altered apatite grains from Section 2 of Cerro Galan .15: SEM images of altered apatite grains from Cerro Blanco .16: Backscatter image and X-ray map from apatite grain in Section 1 of Cerro Galan .17: Comparison of anion concentrations between (A) Section 1 of Cerro Galan and (B) Cerro Blanco. 79 x CHAPTER 1 Introduction Previous Work: Cerro Galan Cerro Galan is a large volcanic complex in the Altiplano-Puna plateau of northwest Argentina and has been the focus of numerous studies. Some of the early workers include Sparks et al. (1985) and Francis et al.

Sparks et al. (1985) were pioneers in identifying, describing, dating, and petrologically characterizing the numerous ignimbrites of the Galan complex. Francis et al. (1989) were among the first investigators to emphasize geochemical studies in order to elucidate the mechanism by which the Cerro Galan complex was developed.

Francis et al. (1989) concluded that the Cerro Galan caldera complex represents the development of a cordilleran plutonic complex in the Andean crust. They assert that basaltic magma provided the heat for triggering melting in the crust at mid to deep levels. Furthermore, they claimed that magma mixing generated the formation of the characteristic dacitic magmas erupted at Cerro Galan.

Recent investigators include Folkes et al. (2011) who building on the work of Francis et al. (1989) further emphasized the importance of crustal assimilation in the generation of the Cerro Galan ignimbrites. With the use of thermodynamics and traditional geochemical investigations Folkes et al.

(2011) concluded that magma generation in the Galan complex comes from two sources within the crust. This was inferred from mineral barometry, geothermometry, and major and trace element analyses of 2 types of pumice clasts—which they term as white and grey—found in the Cerro Galan ignimbrite. They assert that the white pumice clasts equilibrated in the upper 10 km of the crust (<3 kbar), while the grey pumice clasts originated from a deeper (up to 18 km in the crust), hotter source than the white pumice magma. Furthermore, building on the work of Hildreth and Moorbath (as cited in Skewes and Sterns, 1995) this investigation proposes the existence of a lower-crustal MASH (melting, assimilation, storage, and homogenization) in the Cerro Galan system, where magmas are geochemically buffered (this 1 buffering is a consequence of parental magmas being delivered to shallow crustal levels where MASH occurs prior to eruption), producing the underlying geochemical and isotopic signatures that characterize the compositionally similar ignimbrites in the Galan complex.

Another important recent study is that of Cas et al. (2011) who characterized the flow dynamics and eruption style of the latest ignimbrite—the Cerro Galan ignimbrite (CGI)—of the Cerro Galan complex as a low ‘boiling over’, fountain style eruption, with a relatively constant, sustained and high discharge rate. This investigation has important implications for post-emplacement magmatic processes including alteration and re-equilibration of mineral compositions. If the Cerro Galan and Cerro Blanco ignimbrites were not quenched effectively after eruption then minerals in these ash flow deposits may have altered.

If this was the case then compositions of these minerals would not be useful for inferring conditions of the magma chambers prior to eruption. Previous Work: Halogens applied to Cerro Galan Other studies which have influenced this investigation are those of Boyce and Hervig (2008) and Patino Douce and Roden (2006). Boyce and Hervig (2008) showed that OH and Cl growth zonation in apatite grains recorded a multistage magmatic history before eruption. Based on apatite growth rates, they attributed the zoning profiles to H2O degassing and recharge less than 400 days prior to eruption.

Similarly, Patino Douce and Roden (2006) demonstrate the utility of apatite as a probe for F, Cl, and OH volatiles in magmatic systems. These workers demonstrated the importance of using apatite equilibria in conjunction with other minerals –in this case merrillite—in order to calculate halogen and water fugacities in planetary magmas. Merrillite is not present in Cerro Galan or Cerro Blanco however the thermodynamic concepts utilized in the previously described study can be applied to the apatite-biotite phase equilibria in order to elucidate pre-eruptive conditions in the sub-volcanic magma chambers. Purpose of Investigation No previous studies have tried to use volatile compositions (halogens and water) in conjunction with thermodynamic calculations in order to elucidate the pre-eruptive magma conditions in the silicic magma chambers of Cerro Galan and Cerro Blanco.

This investigation sought to estimate pressure, 2 temperature, and volatile budget (F, Cl, and OH) in both calderas as a function of vertical location within the magma chambers in order to infer possible correlations and differences in the evolution of the individual magma chambers. To this end, compositions of apatite, biotite, Fe-Ti oxides, and feldspars (plagioclase and alkali-feldspar) were chosen from a mineral assemblage in both calderas consisting of plagioclase, sanidine, Fe-Ti oxides, biotite, quartz, apatite, and rare zircon.