UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mineralogy and crystal structures of barium silicate minerals from Fresno County, California Basciano, Laurel Christine 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0016.pdf [ 5.94MB ]
Metadata
JSON: 831-1.0099479.json
JSON-LD: 831-1.0099479-ld.json
RDF/XML (Pretty): 831-1.0099479-rdf.xml
RDF/JSON: 831-1.0099479-rdf.json
Turtle: 831-1.0099479-turtle.txt
N-Triples: 831-1.0099479-rdf-ntriples.txt
Original Record: 831-1.0099479-source.json
Full Text
831-1.0099479-fulltext.txt
Citation
831-1.0099479.ris

Full Text

MINERALOGY AND C R Y S T A L STRUCTURES OF BARIUM SILICATE MINERALS FROM FRESNO COUNTY, CALIFORNIA by L A U R E L CHRISTINE BASCIANO B.Sc. Honours, SSP (Geology), Queen's University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Earth and Ocean Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1999 © Laurel Christine Basciano, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of !PcX,rU\ a^/icJ OreO-^ Scf&PW The University of British Columbia Vancouver, Canada Date OeC S /79 DE-6 (2/88) Abstract The sanbornite deposits at Big Creek and Rush Creek, Fresno County, California are host to many rare barium silicates, including bigcreekite, UK6, walstromite and verplanckite. As part of this study I described the physical properties and solved the crystal structures of bigcreekite and UK6. In addition, I refined the crystal structures of walstromite and verplanckite. Bigcreekite, ideally BaSi 20 5-4H 20, is a newly identified mineral species that occurs along very thin transverse fractures in fairly well laminated quartz-rich sanbornite portions of the rock. It post dates the other associated barium silicates and represents either a later primary phase from intruded fluids or a product of alteration of pre-existing Ba-rich minerals, possibly sanbornite. It is colourless and forms poorly developed crystalline masses parallel to the fracture direction. There are two perfect cleavages {010} and {001}. Bigcreekite is biaxial negative, with indices of refraction a 1.537(2), p 1.538(2), y 1.541(2); X = b, Y = a, Z = c and 2V m e a s = 59.2(5)°, 2V c a l c = 60.1 °. The crystal structure of bigcreekite was solved in space group Prima to R = 3.65%, with cell parameters a 5.038(6), b 9.024(3), c 18.321 (6) A, and Z = 4. The empirical formula of bigcreekite (based on 9 O) is (Ba,. 0Na 0 . 0 1) 1 . 1Si,. 9 9Hg . 0 2O 9 . D c a l c = 2.739 g/cm3 and D r a e a s = 2.66(3) g/cm3. Bigcreekite is a hydrous chain silicate containing four-membered rings which form chains of silica tetrahedra, parallel to [100] and staggered in the [001] direction. Water molecules fill the large spaces between the rows of silicon tetrahedra. Bigcreekite has similarities to sanbornite and gillespite. Bigcreekite was named for Big Creek, California, its type locality. UK6, ideally (Si, Al) 4Ba 3C10 9(Cl, H 2 0) 4 , is a newly identified mineral species that is light blue-grey, with one perfect cleavage on {0001}, and forms irregular masses up to 10 mm ii enclosed in parallel-bedded sanbornite-quartz rock. UK6 is uniaxial negative, with indices of refraction e 1.594(2), co 1.642(2). The crystal structure of UK6 was solved in space group P63mc to R = 4.85%, with cell parameters a 5.2432(7), c 29.859(6) and Z = 2. The empirical formula of UK6 (based on 7 cations and 2 H 2 0 per formula unit) is (Baj 9 7, N a 0 0 3 , Cao 0 1) 3 0 1(Si 2 6 3 , Al , 3 4 , Ti0.03)4ClO9(Cl, 6 1 , 2H 2 0, F 0 0 5 ) 3 6 6 , and D c a l c = 3.709 g/cm3. The UK6 structure is based on double tetrahedral layers, [T 4O g]„„, consisting of six-membered rings, with three layers of barium polyhedra connecting the tetrahedral layers. UK6 is part of the Monteregianite-(Y)-Wickenburgite series (Struntz classification) and is structurally and chemically similar to cymrite. The crystal structure of walstromite was originally solved by Glasser & Glasser (1968) in space group P1 using photographic methods to an R index of 16 %. It was redetermined in space group PJ with unit cell parameters a 6.733(1), b 9.608(2), c 6.685(1) A , a 69.64(3), p 102.29(3), y 96.89(3)°, Z = 2 to an R index of 3.8 %. The empirical formula of walstromite (based on 9 anions per formula unit) is (Ca, 9 4 7 , Na^,,), 9 5 8 Ba, 0 2 1 (S i 2 9 9 5 , A l 0 0 2 3 ) 3 0 1 8 O 9 which is similar to the formula determined by Glasser & Glasser (1968). The walstromite structure consists of trigonal rings of silica tetrahedra arranged in layers on (101) with barium and calcium atoms lying approximately halfway between the silicon tetrahedral rings, perpendicular to the a axis. Walstromite is isostructural with margarosanite, as predicted by Glasser & Glasser (1968). The crystal structure of verplanckite was originally solved by Kampf et al. (1973) in space group P6/mmm to an R index of 10.2 %. It was redetermined in P6/mmm with unit cell parameters a 16.300(2) c 7.169(1) A, Z = 3 to an R index of 4.0 %. The empirical formula of verplanckite (based on 12 Si and Al per formula unit) is (Ba ] 2 7 6 , C a ^ Na 0 2 1 ) 1 2 9 8 (Mn 4 5 4 , T i u 9 , Mgo.o3> Feo.2)6.i7(Siii.66> A l 0 3 4 ) I 2 0 3 1 7 8 (C1 2 3 6 4 , F 0 4 9 ) 2 4 1 3 , which is significantly different than analyses iii obtained by Alfors et al. (1965) and Kampf et al. (1973) due to the large variation in chemistry. Verplanckite is a framework structure consisting of four-membered tetrahedral rings and square pyramid polyhedra which form open hexagonal rings, with barium polyhedra located along the inner edges of the rings. There is a large degree of disorder in the verplanckite structure. Verplanckite is similar to the zeolite group of minerals with a very open framework. Table of Contents Abstract ii Table of Contents v List of Tables viii List of Figures ix Acknowledgments xii 1.0 Introduction 1 1.1 Geochemistry of Barium 1 1.2 Barium Coordination 4 1.3 Geological Occurrence of Barium Minerals 5 1.3.1 Vein and Cavity-Filling Barite Deposits 5 1.3.2 Residual Barite Deposits 5 1.3.3 Bedded Barite Deposits 6 2.0 Area Geology 8 2.1 Introduction 8 2.2 Regional Geology 11 2.3 Local Geology and Mineralogy 15 2.2.1 Sanbornite assemblage 18 2.2.2 Quartz and Walstromite Assemblage 23 3.0 Experimental Procedure 29 3.1 Crystal selection and preparation 29 3.2 Data collection 30 3.3 Structure Solution by Patterson Methods 32 3.4 Structure Refinement 34 3.5 Errors 37 3.6 Programs and Procedure 37 3.6.1 MISSYM 38 3.6.2 STR UCTURE TIDY 38 3.7 Electron-probe Microanalysis 39 3.8 Infrared Spectroscopy 40 3.9 Bond Valence 40 3.10 Gladstone Dale Rule 41 4.0 Bigcreekite: New Mineral Description 43 v 4.1 Introduction 43 4.2 Occurrence 43 4.3 Physical and Optical Properties 45 4.4 Chemical Composition 46 4.41 Electron-microprobe analysis 46 4.42 Infrared analysis 49 4.5 X-ray Crystallography and Crystal-Structure Determination 49 4.51 Experimental 49 4.52 Structure solution and refinement 51 4.6 Description of the Structure 55 4.6 Discussion 61 5.0 UK6: New Mineral Description 67 5.1 Introduction 67 5.2 Occurrence 67 5.2 Physical and Optical Properties 69 5.4 Chemical Composition 69 5.5 X-ray Crystallography and Crystal-structure Determination 72 5.5.7 Unit Cell and Space Group Determination 72 5.5.2 Experimental 78 5.53 Crystal Structure Solution and Refinement 78 5.6 Description of the Structure 80 5.7 Discussion 92 6.0 Walstromite: Refinement 98 6.1 Introduction 98 6.2 Previous Work 98 6.3 Electron-probe Microanalysis 100 6.4 X-ray Crystallography and Crystal-Structure Refinement 103 6.41 Experimental 103 6.42 Structure solution and refinement 106 6.5 Description of the Structure Based on Earlier Work 107 6.6 Description of the Structure Based on This Work 107 6.7 Discussion 117 7.0 Verplanckite: Refinement 123 7.1 Introduction 123 7.2 Previous Work 123 7.3 Electron-probe Microanalysis 124 7.4 X-ray Crystallography and Crystal-Structure Refinement 127 7.41 Experimental 127 7.42 Structure solution and refinement 130 7.5 Description of the Structure Based on Earlier Work 134 vi / 7.6 Description of the Structure Based on This Work 134 7.7 Discussion 139 8.0 Conclusion 144 8.1 Bigcreekite, UK6, Walstromite and Verplanckite 144 8.2 Future Research 146 9.0 References 148 vii List of Tables 2.1 Minerals found in sanbornite deposits at Fresno County, California. 9 2.2 Ages of deposits (Hinthorne 1974) 12 4.1 X-ray powder-diffraction data for bigcreekite (Pers. comm., A Roberts). 47 4.2 Electron-probe microanalysis of bigcreekite. 48 4.3 Miscellaneous information for bigcreekite. 50 4.4 Atomic parameters for bigcreekite. 52 4.5 ' Selected interatomic distances (A) and angles (°) for bigcreekite. 53 4.6 Bond-valence arrangement in bigcreekite. 54 5.1 Electron-probe microanalysis of UK6. 73 5.2 Miscellaneous information for UK6. 79 5.3 Atomic parameters for UK6. 81 5.4 Selected interatomic distances (A) and angles (°) for UK6. 82-83 5.5 Bond-valence arrangement in UK6. 84 6.1 Cell dimensions of walstromite. 101 6.2 Atomic coordinates for walstromite (Glasser & Glasser 1968) 102 6.3 Electron-probe microanalysis of walstromite. 104 6.4 Miscellaneous information for walstromite. 105 6.5 Final atomic parameters for walstromite. 108 6.6 Selected interatomic distances (A) and angles (°) for walstromite. 109-110 6.7 Bond-valence arrangement in walstromite. 111 7.1 Atomic coordinates for verplanckite (Kampf etal. 1973). 125 7.2 Selected interatomic distances (A) for verplanckite (Kampf et al. 1973). 126 7.3 Electron-probe microanalysis of verplanckite. 128 7.4 Miscellaneous information for verplanckite. 129 7.5 Final atomic parameters for verplanckite. 131 7.6 Selected interatomic distances (A) and angles (°) for verplanckite. 132 7.7 Bond-valence arrangement in verplanckite. 133 viii List of Figures 2.1 Sanbornite localities in the Western Metamorphic Belt subprovince of the 13 Sierra Nevada. After Hinthorne (1974). 2.2 Location of sanbornite deposits in Fresno County, California. BC locality 16 in Figure 2.1. 2.3 Location of sanbornite deposits along Big Creek and Rush Creek, and 17 mining claims Esquire #1 and Esquire #7. After Alfors et al. (1965). 2.4 Backscattered electron micrograph showing UK6 (C). Scale bar 200 urn. 20 2.5 Backscattered electron micrograph of an altered barite grain. Scale bar 200 urn. 21 2.6 Backscattered electron micrographs. A - barite needles filling open cavities 22 in celsian, B - Barite and fine grained mixture of silica and BaO filling dissolution cracks in sanbornite. Scale bars 200 um. 2.7 Backscattered electron micrographs. A - Spherules of silica formed along 24 transverse fractures in sanbornite, B - Fine grained mixture of silica and BaO filling dissolution cracks in sanbornite. Scale bars 200 um. 2.8 Backscattered electron micrograph showing celsian grains enclosed in 25 wollastonite. Scale bar 200 urn. 2.9 Backscattered electron micrograph of a witherite vein. Scale bar 200 um. 27 2.10 Backscattered electron micrograph showing alteration of wollastonite. 28 Scale bar 200 um. 4.1 Vein of bigcreekite (arrow) in quartz-sanbornite rock. Scale bar 1 cm. 44 4.2 (a) Coordination of barium in bigcreekite. (b) Barium polyhedron in 56 bigcreekite. 4.3 The crystal structure of bigcreekite, showing the chains of silicon tetrahedra 57 and barium atoms parallel to (100). 4.4 (a) Four-membered tetrahedral rings in bigcreekite, which form tetrahedral 58 chains along [100]. (b) Silicon tetrahedra forming 4-membered rings in gillespite. 4.5 Silica chains along [100] with a repeat distance of two tetrahedra with barium 59 atoms between the chains. ix 4.6 A sheet of barium polyhedra in bigcreekite, responsible for the perfect 60 cleavage on (001). 4.7 Bigcreekite structure projected onto (100) showing voids containing hydrogen 62 atoms. 4.8 Crystal structure of krauskopfite. 63 4.9 Crystal structure of sanbornite. 65 4.10 Crystal structure of gillespite. 66 5.1 Mass of UK6 (arrow) in quartz-sanbornite rock. Scale bar lcm. 68 5.2 Photomicrograph of UK6 (center of field, third-order interference colours). 70 Large surrounding grain with second-order interference colours is sanbornite; the grain to the left with first-order white-grey interference colours is celsian. Plane-polarized light, crossed polars. Field of view is 5.1 mm. 5.3 Backscattered electron micrograph of a zoned UK6 grain. White dots and 71 numbers refer to electron-microprobe analyses given in Table 5.1. Scale bar 100 um. 5.4 Zero-level precession photograph taken perpendicular to (0110). 75 5.5 Zero-level precession photograph taken along an a axis. Note the 76 absence of hhlhl when / = In. 5.6 Zero-level precession photograph taken along a. 11 5.7 (a) Coordination ofBa(l) in UK6. (b)Ba(l) polyhedron in UK6. 85 5.8 (a) Coordination of Ba(2) in UK6. (b) Ba(2) polyhedron in UK6. 86 5.9 (a) Coordination of Ba(3) in UK6. (b) Ba(3) polyhedron in UK6. 88 5.10 Perspective view of the crystal structure of UK6. 89 5.11 Crystal structure ofUK6 projected onto (010). 90 5.12 Crystal structure of UK6 projected onto (001). 91 5.13 Perspective view of the crystal structure of UK6, showing the layers of 93 barium polyhedra bonding to the tetrahedral layers. 94 5.14 Perspective view of Ba(2) and Ba(3) polyhedral sheets, bonded by Cl(l) atoms along [001] with Ba(l) atoms occupying spaces between them in UK6. 5.15 (a) Ba(2) polyhedral layer in UK6 projected onto (0001), showing the 95 location of Cl/0(1) atoms involved in H 2 0 complexes and the spaces containing the hydrogen atoms, (b) Perspective view of Ba(3) polyhedral layer in UK6, showing the location of Cl/0(2) atoms involved in H 2 0 complexes x and the spaces containing the hydrogen atoms. 5.16 Perspective view of the crystal structure of cymrite. 97 6.1 Mass of walstromite (arrow) in quartz-sanbornite rock. Scale bar 1 cm. 99 6.2 Crystal structure of walstromite (Glasser & Glasser 1968). 112 6.3 (a) Coordination of barium in walstromite. (b) barium polyhedron in walstomite. 113 6.4 Coordination of Ca(l) in walstromite, forming a distorted square antiprism. 115 6.5 The walstromite structure projected on (101), showing the layers of 116 alternating rings of silica tetrahedra and the positions of the barium atoms and calcium atoms. 6.6 Walstromite structure projected on (101) showing the (100) cleavage. 119 6.7 Projection on (010) showing chains of barium polyhedra and Ca(2) 120 octahedra. 6.8 Projection of the walstromite structure on (101) showing a sheet of 121 Ca polyhedra with barium in the open spaces. 6.9 Crystal structure of walstromite (top) and margarosanite (bottom). 122 7.1 Coordination of Ba(l) in verplanckite. 136 7.2 (a) Coordination of Ba(2) in verplanckite. (b) Ba(2) polyhedron in verplanckite. 137 7.3 (a) Coordination of Xin verplanckite. (b) X polyhedron in verplanckite, 138 forming a square pyramid. X- Mn, Ti, Fe, Mg and Na. 7.4 (a) Four-membered tetrahedral rings in verplanckite. (b) pin-wheels formed 140 by X polyhedra in verplanckite, where X = Mn, Ti, Fe, Mg and Na. 7.5 The verplanckite structure proj ected on to (010). 141 7.6 The verplanckite structure proj ected on to (001). 142 xi Acknowledgments Many people helped me while I was working on this thesis, for which I am grateful. First, I thank Dr. Lee Groat for a great project and for supervising it. I thank Gail Dunning for the opportunity to work with barium silicates and for sending me the samples. I thank the members of my advisory committee, Dr. Mati Raudsepp and Dr. Greg Dipple for their valuable comments. I additionally thank Mati for spending so much time editing my thesis and for helping with the electron-probe microanalyses. I thank Dr. Joel Grice for collecting the CCD data and Dr. Robert Gault for doing electron-probe microanalyses of bigcreekite. I also thank Dr. James Hinthorne for sending me a copy of his Ph.D. thesis. I would like to thank Brent and Vanessa for wasting lots of my time and for helping me a little bit ©. The basement would not have been nearly as fun without you. Other good friends I have made at UBC and would like to thank are: Cari, Jen, Emily, Heike, Mike, Layne, Eugene, Steve, and Sarah. I really enjoyed playing hockey with the Stress Tensors and playing ultimate with Those Damn Bastards and I thank both teams for this. I thank Tom for being very supportive, understanding and inspiring, I will never forget it. Finally, I thank my family for their concern and support. xn 1.0 Introduction The sanbornite deposits at Big Creek and Rush Creek, Fresno County, California host many rare barium silicates. Fresno County is the type locality for nine of these barium silicates. Additional rare species from this region have been recognized and may also prove to be new. Walstromite, verplanckite, bigcreekite (previously UK22) and unknown mineral 6 (UK6) are part of this series of barium silicates. Walstromite and verplanckite were described in detail by Alfors et al. (1965) and their crystal structures were solved by Glasser & Glasser (1968) and Kampf et al. (1973), respectively. Bigcreekite and UK6 have not been previously described. The purpose of this study is to characterize bigcreekite and UK6 by describing their physical properties and solving their crystal structures, and to refine the crystal structures of walstromite and verplanckite. Crystal structure solution and refinement were done with the use of single crystal X-ray crystallography. 1.1 Geochemistry of Barium Barium is the twenty-first most abundant element in the lithosphere. It is a soft silvery-grey metal that resembles lead. The atomic number of barium is 56 and it has an atomic weight of 137.327(7). It is part of group 2 on the periodic table, the alkali earth metals. Ba 2 + in the crystal structure of barium minerals is surrounded by O2", OH", Cl" and H 2 0 or a halogen as nearest neighbors. The bond-type is predominantly ionic. Barium is the largest divalent cation except for Ra 2 + and therefore replaces or is replaced isostructurally by other large cations (Dawson 1985). For a heavy element barium is quite abundant in the earths crust. Barium, like the other alkaline earth metals, is predominantly lithophile in character. It is among the most abundant trace elements in the upper lithosphere. Barium is highly concentrated in the earth's crust as compared to its abundance in stony meteorites. The scarcity of barium in chondrites is comparable to the scarcity of barium in the pyroxenites and peridotites of the earth, which consist largely of silicates of magnesium and divalent iron. The average content in the earth's igneous rocks was calculated by von Englehardt (1936) to be 480 ppm. The average content of barium in sedimentary rocks is 430 ppm., which is in good agreement with that of the igneous rocks (Goldschmidt 1954). Although barium is relatively abundant in igneous rocks, it only very exceptionally forms minerals. The bulk of barium is concealed in the rock-forming minerals of the igneous rocks. Barium chemically resembles strontium and calcium which are also alkaline earth metals, but their manner of occurrence in igneous rocks is significantly different. Strontium substitutes for calcium in rocks of igneous and sedimentary origin, whereas barium does not. Barium instead extensively substitutes for potassium. Barium does not substitute for calcium while strontium does, due to the similarity in ionic size (Rankama & Sahama 1950): Ion Ionic Radius (A) Ca 2 + 1.06 Sr* 1.18 Ba 2 + 1.38 K + 1.33 The Ba 2 + ion is too large to substitute for the Ca 2 + ion in mineral structures, and for the 2 same reason cannot substitute for M g 2 + and Fe 2 +. The difference in charge between barium and potassium causes barium to be captured in the first fractions of potassium minerals during fractional crystallization. The relationship of barium and potassium in fractional crystallization of potassium feldspars from igneous magmas is analogous to the relationship between calcium and sodium in the crystallization of plagioclases (Goldschmidt 1954). The plagioclase series contains very small concentrations of barium, but even within this very narrow range there is a distinct decrease towards the sodic end of the series. The preference of barium for potassium feldspars and for early porphyritic potash feldspars is marked, as is the scarcity of barium in potassium feldspars from late pegmatites. It has been shown by Noll (1934) and von Engelhardt (1936) that the content of barium in potassium feldspar depends on the temperature of formation. The highest content of barium is in the alkalic rocks (syenites and nephalene syenites), because of the barium in the potassium feldspars (Goldschmidt 1954). There is only one pure barium feldspar that occurs in nature, celsian, Ba(Al 2Si 20 8), which is extremely rare. It has a very restricted paragenesis and most often occurs in association with manganese deposits formed in hydrofhermal systems. Barium is enriched in the manganese-rich oxides because the Mn(OH) 4 sol is negatively charged and attracts cations. Barium can also be concentrated in minerals of the mica and apatite groups. It is dominantly found in those minerals that are formed by late magmatic pneumatolytic or hydrothermal deposits. Most barium minerals are concentrated in hydrothermal deposits. The content of barium is low in granite pegmatites, but tends to become concentrated during the last stage of magmatic crystallization, and forms minerals in hydrothermal rocks (Goldschmidt 1954). In this respect barium behaves like calcium and strontium. The bulk of 3 these metal ions are removed from the melt during the main stage of crystallization but become strongly concentrated after the pegmatitic-pneumatolytic stage. Some hydrothermal barium minerals are: barite Ba(S04), a very common constituent in many metaliferous veins; witherite, BaC0 3 ; barylite, BeBa(Si207); and the zeolites: brewsterite, harmotome, and edingtonite. The hydrothermal solutions may also extract barium from the surrounding rocks. The resulting solutions are enriched in barium in regard to potassium, as they are in calcium with respect to sodium. This is because the component with the higher melting point is preferably attacked and dissolved. 1.2 Barium Coordination Barium is a large ion and therefore must fit into large spaces within a crystal structure. As mentioned previously, it will substitute for other large ions. Barium typically has high coordination number and a mean Ba-0 distance of approximately 3 A. This relatively long bonding distance gives rise to weak Ba-0 bonds. In comparison, a typical strong bond is the Si-0 bond which is approximately 1.64 A in length. The strength of an ionic bond, which can be measured by the melting temperature, is inversely proportional to the bond length. Therefore the shorter the bond, the stronger it is. Another effect that bond strength has on a material is the hardness. The hardness of BaO is only 3.3 on Mohs hardness scale, which is low, due to the high length of the Ba-0 bond (Rankama & Sahama 1950). The coordination of barium is typically between 8 and 12. The higher the coordination number, the longer the average bond length. 1.3 Geological Occurrence of Barium Minerals Barite and witherite are the chief commercial sources of the element barium and the most abundant of the barium minerals. The protolith of the sanbornite deposits is either a barite or witherite deposit (Hinthorne 1974). Barite and witherite can be found alone or with fluorite, celestite, quartz, galena, and sphalerite (Harben & Bates 1984). The three most important types of barite deposit are vein and cavity-filling deposits, residual deposits and bedded deposits (Brobst 1983). 1.3.1 Vein and Cavity-Filling Barite Deposits Vein and cavity-filling deposits form along faults, joints, bedding planes, breccia zones and solution channels and in various sink structures (occurring mostly in limestone). The barite in this deposit type is dense, grey to white and associated with metallic ores such as pyrite, chalcopyrite, galena, and sphalerite. The distribution of these deposits is scattered and irregular, and they range in thickness from a few centimeters to a meter and in length from tens to hundreds of meters. The most likely source for vein and cavity filling deposits is low-temperature hydrothermal solutions. 1.3.2 Residual Barite Deposits Residual deposits occur as unconsolidated material and are formed by weathering of preexisting deposits. They consist of barite in the form of loose fragments embedded in residual 5 clay. The residual barite is white, translucent to opaque and occurs in mammillary, fibrous or dense fine-grained masses. The barite in these deposits ranges from microscopic crystals to boulders weighing hundreds of kilograms. The barite is associated with small amounts of pyrite, galena, and sphalerite which occur on the surface of the barite grains. The rest of the deposit consists of incompletely weathered rock fragments and yellow, red or brown illitic clays. The size and shape of residual barite deposits vary greatly, from very large (several hundred kilometers) to very small (meters). The shape of the original deposit determines the shape of the residual deposit. The barite may be disseminated in the residual deposit or it may be concentrated near the contact with bedrock. Both the barite and the clay are derived from weathering of the underlying rock, generally dolomite. The source of the barite most likely is epigenetic hydrothermal vein material left after the host rock was dissolved. When aragonite is converted to calcite or dolomite, Ba can no longer fit into the crystal structure and is excluded. Any sulfate in the pore fluids will then fix the barium ions as barite 1.3.3 Bedded Barite Deposits Bedded deposits include those in which barite occurs as a principal mineral or cementing agent in stratiform bodies in a layered sequence of rock. The barite layer in these deposits is generally fine grained and dark-grey to black. These deposits are a few centimeters thick to hundred of meters thick and can extend over many kilometers. The beds of barite are commonly interbedded with dark chert and siliceous shale and siltstone. Some barite-rich zones are more than 50 meters thick. Many small-scale sedimentary textures and structures have been described 6 in these deposits (Zimmerman 1969; Zimmermann & Amstutz 1964). Small amounts of clay minerals and pyrite are common in bedded deposits, whereas carbonate minerals are rare. The most likely genesis of bedded barite deposits is submarine-exhalative, in which barium ions derived from submarine hot springs combine with sulfate in the seawater to form a barite precipitate (Harben & Bates 1984). This would account for the sedimentary features and hydrothermal minerals. Submarine hot springs generating barite are known as white smokers. In 1984, 85% of the barite produced in the United States came from Nevada (Harben & Bates 1984), approximately 300 miles east of the Sanbornite deposits in California. Much of the Nevada barite occurs as bedded deposits and occurs in a northeast-trending zone up to 60 miles wide, stretching for approximately 300 miles across the central part of the state. The distribution of barite mineralization is coextensive with the Antler Orogenic Belt of thrust faulting that began in the Late Devonian. This activity moved the siliceous rocks and bedded barite almost 100 miles eastward to their present location (Harben & Bates 1984). The origin of the bedded barite deposits was summarized in Papke (1984); they are of exhalitive sedimentary origin, formed by solutions that rose along fissures in the ocean floor and discharged as submarine hot springs. Anaerobic sulfate-reducing bacteria produced sulfide ions that reacted with metallic ions and hydrogen to form small amounts of pyrite and other metallic sulfides. The process of barite deposition was interrupted periodically by sealing of the conduits, and interbeds of siliceous rocks were deposited during these intervals. With increased temperature and pressure the conduits were re-opened and deposition by barite resumed. 2.0 A r e a Geology 2.1 Introduction Seven new barium silicate minerals from the sanbornite deposits of eastern Fresno County, California, have been discovered and described: fresnoite, krauskopfite, macdonaldite, muirite, traskite, verplanckite, and walstromite (Stinson & Alfors 1963, 1964; Alfors & Stinson 1965; Alfors et al. 1965). Alforsite, a new barium end-member of the apatite group, was described by Newberry et al. (1981). Sanbornite was first described from Trumbull Peak, Mariposa County, California by Rogers (1932) and its crystal structure was solved by Douglass (1958). There are many other new barium silicates presently being studied. Names and formulas of all minerals found in the sanbornite deposits of Fresno County are given in Table 2.1. These minerals represent a mineralogically significant occurrence due to their unusual chemical associations. In 1957, the California Public Service Laboratory of the Division of Mines and Geology received a rock sample containing taramellite, gillespite and sanbornite from Fresno County, California from G.M. Landers. As he could not be located, the deposit could not be readily found. Shortly after, another sample was sent from Rush Creek, Fresno County by Robert E. Walstrom (REW). In April 1961, the deposit had been located and mining claims were established by REW and are still retained by him. Later in 1961 the original location of the sample was determined to be a road cut along Big Creek. Similar deposits are known from Chickencoop Canyon, Tulare County and Incline, Mariposa County, both in California; and from Baja California, Mexico. Sanbornite was first 8 T A B L E 2.1 MINERALS FOUND IN SANBORNITE DEPOSITS A T FRESNO COUNTY, CALIFORNIA alforsite Ba5(P04)3Cl anandite (Ba,K)(F+2,Mg)3(Si)Al,Fe)401o(0,OH)2 aragonite CaC0 3 barite BaS04 bazirite BaZrSi309 benitoite BaTiSi 30 9 bigcreekite BaSi2O5-4H20 calcite CaC0 3 celsian BaAl 2Si 20 8 chalcopyrite CuFeS2 diopside CaMgSi206 edingtonite BaAl2Si3O10-4H20 fluorapatite Ca5(P04)4F fresnoite Ba 2TiSi 20 8 galena PbS gillespite BaFe+2Si40,o hyalophane (K,Ba)Al(Si,Al)308 krauskopfite BaSi2O4(OH)2-2H20 macdonaldite BaCa4Sii6O36-10H2O mineral 10 Ba, Al, Si, Cl, 0 mineral 21 Si, O, H mineral 27 Ba, Fe, Si, Ca, P, Mn, 0 molybdenite MoS2 montmorillonite (Na,Ca)o. 3(Al,Mg) 2Si 40 1 0(OH) 2-«H 20 muirite Ba1 0Ca2MnTiSi1 003o(OH,Cl,F)1 0 opal S i0 2 -«H 2 0 pellyite Ba2Ca(Fe+2,Mg)Si6017 powellite CaMo04 pyrite FeS2 pyrrhotite Fe,.xS quartz Si0 2 sanbornite BaSi 20 5 scheelite CaW0 4 spalerite ZnS titantaramellite Ba4(Ti,Fe+ 3,Fe+ 2,Mg)4(BSi602 7)02Clx tobermorite (Ca,Ba)5Si6(OH)2016-3H20 traskite Ba9Fe+2Ti2(Si03)12(OH,Cl,F)2-6H20 tremolite Ca2(Mg,Fe+ 2)5Si802 2(OH)2 UK6 (Si,Al)4Ba3C109(Cl,H20)4 uraninite u o 2 verplanckite Ba2(Mn,Fe+2,Ti)Si206(0,OH,Cl,F)2-3H walstromite BaCa 2Si 30 9 witherite BaC0 3 wollastonite CaSi03 described by Rogers (1932) from the Incline, Mariposa County location. Sanbornite has also been noted in a pyroxene-garnet-quartz skarn deposit in the Yukon Territory as a minor constituent (Montgomery 1960). 2.2 Regional Geology Hinthorne (1974) suggested that the four sanbornite deposits of California and Baja California are in close association. These deposits are located in the Western Metamorphic Belt subprovince of the Sierra Nevada and occur along a line approximately parallel to the regional strike of the belt (Figure 2.1). The deposits are roof pendants in a granodiorite pluton. As a sedimentary origin for the barium in the sanbornite deposits is most probable (Hinthorne 1974), it was proposed by Hinthorne (1974) that the deposits represent a single stratigraphic horizon, or at least had similar depositional environments. This hypothesis would allow a tentative correlation between the various roof pendants containing the sanbornite bearing rocks. Gastil and Phillips (1971) suggested that the metasedimentary strata at or near all four sanbornite localities were originally deposited in a narrow basin parallel to the present trend of the Western Metamorphic Belt. However, age determinations for the metamorphosed sediments near the Sierra Nevada sanbornite localities do not confirm this hypothesis; see Table 2.2. These age determinations suggest that the sedimentary rocks had a wide range of ages before the emplacement of the earliest Sierra Nevada batholith rocks (Hinthorne 1974). Each roof pendant will have to be dated individually before the ages of the sanbornite-containing strata can be determined. 11 00 H l-H O PH CO o cj B o w h-t ffl < CJ K ca4 O L5» bo en S •S C3 O O ON cd ^ ft o 8 :§ s I OT ?3 CJ >H O PH HH CS 1/3 CD P H o o CJ CJ 7J c tu r— c3 ON o . - i W 3 d cd T j cj a, <+-< o o lH CJ O c CJ o PH CJ in o OB So PQ c CJ CJ CJ co T j T j CO 5H TJ cd in C VO ON co T j O CJ a CO Cd CJ >~> co O O . H CJ 1=3 O S "cj CN IT! m S i—l cj cd CJ CJ s CJ .s =3 w CJ O TJ O <D +H CO co o =3 XI s a 2 o o co r/. T j cd m ON CO o CJ B < G cd ' § C fd co o u m P H 41 OT r- c3 o cj T j <U - H cj +^  o ^ 3 cj S CJ cd I 3 t ° S £ CO cj co fi » o C S o < a •• CM o co cj o o I Pi 7 o 1 2 Figure 2.1. Sanbornite localities in the Western Metamorphic Belt subprovince of the Sierra Nevada. After Hinthorne (1974). Hinthorne (1974) proposed that the introduction of barium into the sedimentary host rocks could be the result of one or more of the following alternatives: 1. The barium was originally present in the sedimentary rocks as a carbonate or sulfate mineral. 2. The barium was introduced from an unknown source before metamorphism, but after deposition of the host sediments. 3. The barium was introduced from the plutons during intrusion and metamorphism (metasomatism). Hinthorne (1974) determined that the barium must have been introduced into the sedimentary sequence before metamorphism, either as a primary sedimentary constituent or as a diagenetic replacement in the original sequence. There is no evidence that the barium was introduced by the granitic intrusion. Aplite dikes cut through the sanbornite rocks but they contain less than 0.05% BaO (Hinthorne 1974). The average BaO content of granitic rocks is 0.048%; therefore these dikes do not contain excess barium. The original mineral assemblage in the sediments was generally quartz-witherite. The sanbornite and associated minerals were formed as a result of thermal metamorphism of this assemblage, immediately preceding and/or accompanying the intrusion of the nearby granitic rocks. Hinthorne (1974) concludes that the most probable sequence of events leading to the formation of the four similar occurrences of sanbornite rocks was: 1. Sedimentary barite (or witherite) precipitated locally as part of an abyssal sedimentary sequence in a possibly continuous narrow basin off the west coast of Mexico and California within the late Paleozoic to Early Jurassic. 14 2. Diagenetic activity then dissolved the barite and reprecipitated the barium as witherite. 3. Burial of the sediments was followed by one or more periods of severe deformation and contact metamorphism by the intrusion of a large granodiorite pluton. The temperature of metamorphism was between 440 and 600 °C (Hinthorne 1974). The probable pressure during the formation of the barium silicates was greater than 2.0 kb with an upper limit at 3.9 kb (Hinthorne 1974). 4. The massive barite, witherite and hydrated barium silicates formed recently by ground water deposition in the pores and joints of the quartz-sanbornite rocks. 2.3 L o c a l Geology and Minera logy The Fresno County localities occur along a narrow 5 km belt on the western slope of the Sierra Nevada, approximately 90 km by road northeast of Fresno (Figure 2.2). The sanbornite bearing rocks are in two meta-sedimentary roof pendants along Rush Creek (Esquire #1) and on the eastern border of an area of meta-sedimentary rocks along Big Creek (Esquire #7) (Figure 2.3). The sanbornite deposits consist of gneissic metamorphic rocks, composed mainly of variable amounts of sanbornite and quartz, and minor amounts of pyrrhotite and diopside. The sanbornite-quartz rock forms conformable tabular bodies up to 13m thick within foliated quartzite in a migmatite zone at or within 100 m of the contact with a large granodiorite pluton. The trend of the outcrop of the sanbornite-bearing quartzite is sub-parallel to the granodiorite contact. Some fragments of the sanbornite-bearing rock have broken off and are incorporated into the locally massive quartz veins at the margin of the pluton. The quartzite is conformably inter-bedded with 15 Figure 2.2. Loca t ion of sanbornite deposits in Fresno County, California. B C locality in Figure 2.1. 16 Big Creek Esquire # 1 Granodiorite Metamorphics Sanbornite Outcrop / Creek / Road Figure 2.3. Locat ion of sanbornite deposits a l o n g Big Creek a n d Rush Creek a n d mining c l a ims Esquire #1 a n d Esquire #7. After Alfors et al. (1965). 17 steeply-dipping mica schist, amphibole schist, and wollastonite-bearing quartzite (Alfors et al. 1965). The metamorphic rocks have undergone multiple deformations resulting in at least three superimposed sets of folds (Hinthorne 1974). Macdonald (1941) and Krauskopf (1953) have mapped the area in reconnaissance. There is a gradational change in the quartz to sanbornite ratio in both the Big Creek and Rush Creek localities. The changes are mainly dependent on local bulk compositions. Therefore, the sanbornite-quartz rock can be mapped as units with high quartz and sanbornite content. This layered structure of the deposit indicates that bedded barite deposits from an exhalitive sedimentary origin is the most probable protolith for the quartz-sanbornite rocks, similar to the barite deposits in Nevada. 2.2.1 Sanbornite assemblage Sanbornite is the dominant phase and appears to be one of the first minerals to have crystallized together with quartz, diopside and celsian during the first metamorphic event. Sanbornite is white to colourless, translucent to transparent, and forms anhedral plates that are up to 4 cm across and 1 cm in thickness. It has a poikiloblastic texture with inclusions of diopside and celsian. The anhedral diopside and celsian inclusions in the sanbornite are approximately 100 mm in size. The sanbornite grains are surrounded by anhedral diopside and celsian grains, approximately 200 um in size, which form aggregates roughly 1 cm across. In these aggregates are less abundant, subhedral to euhedral grains of titantaramellite. Less abundant are individual anhedral grains of pyrrhotite, chalcopyrite and alforsite. 18 Unknown mineral 6 (UK6) formed prior to sanbornite, celsian and diopside. It is found within sanbornite and celsian grains (Figure 2.4). UK6 forms anhedral grains that are approximately 60 um in diameter with a perfect cleavage in one direction. It has been altered to hyalophane along its cleavage and fractures. Witherite is closely associated with hyalophane in the cleavage of UK6. Crosscutting UK6 are veins of clotted needles, possibly verplanckite, approximately 40 urn in diameter. At the edges of the sanbornite grains are single grains of altered barite (Figure 2.5). In thin section (plane-polarized light) the altered barite grains are opaque and iron staining can be seen around the nearby celsian grains. The grains have some remnants of barite, altered from the outside inwards by iron oxide. Iron oxide veins cut across the grains and the enclosed diopside and chalcopyrite, with a further episode of veining by barite cutting across the iron oxide veins. Sanbornite has been extensively altered with numerous BaO and Si0 2 veins parallel to and against cleavage. Dissolution of sanbornite has occurred along and across the cleavage, leaving large vugs. Filling many of the dissolution spaces is a very fine-grained mixture of silica and BaO. An influx of barite occurred after the initial formation of sanbornite. Barite needles formed in open veins and vugs (Figure 2.6a). Massive barite veins cut through sanbornite grains parallel to and across cleavage. (Figure 2.6b). Veins of massive barite also surround and cut through grains of celsian, diopside and altered barite grains. Fine grained silica and BaO surround barite veins which run parallel to the cleavage of sanbornite (Figure 2.6b). There has been subsequent fracturing of the sanbornite grains perpendicular to cleavage causing micro-faults across the barite veins. 19 Figure 2.4. Backsca t te red e lect ron mic rograph showing UK6 (C). A - cels ian, B - sanbornite, C - UK6, D - hya lophane , E - diopside, F - traskite or titantaramellite, G - barite, H - witherite, I - fine gra ined mixture of sil ica a n d BaO, J - verplanckite. S c a l e bar 200 |u,m. 20 Figure 2.5. Backsca t te red e lect ron mic rograph of a n al tered barite grain. A - barite grain, a l tered from the outside inwards to iron oxides. Iron oxides are dark c o l o u r e d a n d subsequent barite veining is white. B - Silica from al tered sanbornite C - Diopside with a rim of barite D - Cha lcopyr i t e E - Barite vein Sca l e bar 200 u r n 21 Figure 2.6. Backscattered electron micrographs. A - Barite needles filling open cavities in celsian. B - Barite (lighter colour) and fine grained mixture of silica and BaO (darker colour) filling dissolution cracks in sanbornite. Scale bars 200 urn. Parallel to cleavage, open veins are partially filled with spherules of silica and dendritic barium and calcium oxides (Figure 2.7a). The spherules are 5 um in diameter and the dendrites are ~10 pm long. The spherules are possibly the new mineral UK21, which has been found at Site 2, where it occurs sparingly along transverse fractures that occasionally develop between large sanbornite cleavages. UK21 is a hydrous silicate, forming white, flat radiating crystals. Many sanbornite grains have been completely altered to fine-grained silica. Remnant barite veins indicate that the silicification occurred after the emplacement of barium. Within the silicified area there are isolated pieces of sanbornite, partially altered to hyalophane, and individual diopside grains. Celsian grains are not present. There is a gradual loss of barium in the fine grained silica from the edges of sanbornite grains causing zonation along the margins of the silicified area (Figure 2.7b). 2.2.2 Quartz and Walstromite Assemblage Walstromite grains are quite large (>1 cm) and are closely associated with quartz rather than sanbornite. Walstromite has a distinct cleavage in one direction and occurs as individual grains disseminated in the sanbornite quartz rock. Surrounding these large grains are anhedral quartz grains, ~2 mm in diameter. There are abundant quartz inclusions in the large walstromite grains, indicating that quartz formed before walstromite. Abundant tabular euhedral to anhedral celsian grains are enclosed in walstromite (Figure 2.8). They are approximately 80 um across and 150 pm in length. Many celsian grains contain 23 Figure 2.7. Backscattered electron micrographs A - Spherules of silica formed along transverse fractures in sanbornite. 1 - sanbornite? 2 - silica, 3 - celsian. B - Fine grained mixture of silica and BaO filling dissolution cracks in sanbornite. 1 - sanbornite, 2 - silica and BaO, 3 - silica. Scale bars 200 um, K 24 Figure 2.8. Backscattered electron micrograph showing celsian grains enclosed in walstromite. A - walstromite, B - celsian, C - silica partially altered to wollastonite D - rhodonite, E - muirite. Scale bar 2 0 0 um. 25 inclusions of quartz which are partially altered to wollastonite. Other inclusions in the walstromite are small anhedral grains of muirite, sphalerite and bustamite. All of these are approximately 30 u,m in diameter. The standard formula for muirite is Ba 1 0(Ca,Mn,Ti) 4Si 8O 2 4(OH,Cl,O) 1 2-4H 2O, but the X site in the muirite from this deposit contains little or no Mn or Ti (Khan & Baur 1971). There are also muirite crystals surrounding the walstromite. They are euhedral with faint cleavages at 90° to one another. The grains are approximately 50 u.m in diameter. Walstromite also contains an unidentified tabular mineral containing Ba, Si, Al , Cl , and O with trace amounts of Mn and Fe. As well as the quartz surrounding walstromite grains there is also wollastonite, muirite, celsian, bustamite, chalcopyrite, and diopside. The bustamite is least abundant. All are anhedral, rounded grains, ranging in size from 20 to 200 um. Accessory minerals include bazirite, which is a zirconium barium-silicate, and a mineral that is probably titantaramellite but possibly traskite. It is subhedral and ~ 500 u,m in diameter. All of these minerals occur as individual grains, but more often as clots of numerous minerals together. Celsian and similar minerals have extensive veining of witherite (Figure 2.9). The witherite veins cut across the 90° cleavage of celsian and are approximately 20 u.m across. Celsian surrounding some witherite veins has been altered to hyalophane. Rare grains of UK6 are present, enclosed in celsian grains; these appear similar to UK6 grains in the sanbornite. After the formation of the above mentioned minerals there was extensive alteration. The quartz around walstromite is extensively altered to wollastonite as well as some quartz inclusions in walstromite and celsian. The wollastonite in some areas has then been altered to barium and calcium oxides (Figure 2.10). 26 Figure 2.9. Backscattered electron micrograph of a witherite vein. A - walstromite, B - witherite, C - celsian, D - UK6 with hyalophane and silica along its cleavage, E - diopside. Scale bar 200 um. 27 Figure 2.10. Backscattered electron micrograph showing alteration of wollastonite. A - walstromite, B - muirite, C - wollastonite, D & E - BaO and CaO. Scale bar 200 u m 3.0 Experimental Procedure 3.1 Crystal selection and preparation The size of the crystal should be approximately less than 50 x 50 x 50 pm to minimize absorption of X-rays and more than 10 x 10 x 10 pm to maximize the intensity of the reflections. As the minerals examined in this study contain barium, a very high absorber of X-rays, crystal sizes should ideally be at the lower end of the range. However, the poor crystallinity of some samples necessitated using larger crystals. Crystals must not be twinned, as twinning complicates the solution of the structure. Although programs have been developed to deal with twinning, it still causes great difficulties. More importantly, the fragment chosen must not be an aggregate of crystals or have any inclusions. These conditions significantly decrease the quality of the data, leading to a less accurate and precise structure. The absence of twinning, inclusions or aggregation in crystals of bigcreekite, walstromite and verplanckite was confirmed with a polarizing microscope. The crystal of walstromite was ground into a sphere in an air-driven, circular abrasion grinder to better enable a model to be calculated for the absorption correction as the absorption of X-rays is uniform in all directions. Crystals of bigcreekite and verplanckite were not ground into spheres because their low hardness caused them to disintegrate in the grinder. An approximately cube shaped crystal of bigcreekite was chosen and an approximately spherical crystal of verplanckite was found. A cleavage fragment of UK6 was used. UK6 forms as aggregates, is poorly crystalline and is generally 29 twinned. Of many tests, only one crystal of UK6 that diffracted X-rays and did not appear to be twinned was found; its morphology was platy and no attempt at shaping it was made. The crystals were mounted with a random orientation on glass fibres using "five-minute" epoxy. The dimensions and origins of the crystals used in this study are given in Chapter 4 for bigcreekite, Chapter 5 for UK6, Chapter 6 for walstromite and Chapter 7 for verplanckite. 3.2 Data collection Data for all samples were collected on a Siemens P3 automated four-circle diffractometer equipped with a molybdenum-target X-ray tube (operated at 55 kV, 35 mA) and a graphite crystal monochromater mounted with equatorial geometry. A Siemens P4 instrument with a charge-coupled device (CCD) detector at the University of Manitoba was used to collect an additional set of data for UK6. The following procedure was used to determine the unit cell parameters. Using the four-circle diffractometer, rotation photographs were taken. Twenty-five reflections that were relatively strong and with as high a 20 value as possible were chosen from a rotation photograph. The points chosen were distributed in more than one octant of the reciprocal lattice to lower the standard deviations of the unit cell parameters. Once a number of reflections were described in terms of measured setting angles, they were used to define two matrices, conventionally called U and B. B uses quantities related to the direct and reciprocal cell constants to convert reflection o indices to parameters (in A) in reciprocal cell space. U relates these reciprocal space coordinates to the coordinate system of the diffractometer to give values that can be manipulated by rotations 30 about the four axes. The reflections wi l l then be centred accurately in the detector aperture and the setting angles wi l l be measured, the probable cell parameters w i l l be calculated and the orientation matrix required to collect a data set w i l l be determined. The precision of these measurements were enhanced by least-squares analysis. The unit cell parameters of U K 6 were determined with the C C D detector. U p to 25 reflections were located by an automatic routine which systematically searches reciprocal space for strong diffraction peaks. A repetitive process which uses various scan and aperture types to locate peak centers was then used to determine accurately the positions of these reflections. From the 25 scattering vectors, and their sum and difference vectors ± sJ} a set of known reciprocal lattice points were compiled. A primitive unit cell consistent with these lattice points and its corresponding orientation matrix was then determined and replaced with those of a higher symmetry lattice type i f warranted by the results of cell reduction programs. Using the four-circle, intensity data for all samples were collected in the 0-20 scan mode, using ninety-six steps with a scan range from [20 (MoKoc!)-! .! °] to [20 (MoKcc^+ l .T ] and a variable scan rate between 0.5 and 29.3 7 m i n depending on the intensity of an initial one-second count at the centre of the scan range. Backgrounds were measured for half the scan time at the beginning and end o f each scan. The stability o f the crystal was monitored by collecting two standard reflections every 50 measurements. After the intensity data collection, data was collected for the absorption correction. Reflections uniformly distributed with regard to 20 were measured at 5 ° intervals of t|/, which is the azimuth angle corresponding to rotation of the crystal about its diffraction vector, from 0 to 355° , after the method of North et al. (1968). These data were used to calculate the absorption 31 correction, which was then applied to the entire data set. The data were also corrected for Lorentz, polarization, and background effects, and were averaged and reduced to structure factors. Data and absorption collection statistics for each crystal and any deviations from this procedure are described in the individual chapters of each mineral. 3.3 Structure Solution by Patterson Methods The intensity of the diffracted X-rays and their geometrical factors are used in solving crystal structures. An electron density map is calculated from a Fourier Transform series, using F(hkl), the structure factor, which represents the combined scattering of X-rays for all atoms in the unit cell. F(hkl) is a complex number and therefore gives a magnitude, | Fmeas(/z&/) |, and a phase, (j){hkl). F(hkl) c a l c u l a t e d and F(hkl)m e a s u r e d are given below: FmeasihkT) - .^ intensity of diffracted X - rays x geometrical factors where f- is the atomic scattering factors of atom j ; x, y and z are the fractional coordinates of the atom; and a, b, and c are the unit cell dimensions. Using the F m e a s the electron density, p(xyz), can be determined using a Fourier series: 32 p(xyz) = —]T X X F(hkP)o,os 2z(hx+ ky+ lz)-h k-oo I where FmeasihkT) = J 7 I is the intensity of the measured reflection and V is the volume. (p(hkl) is not measurable, it is either 0° or 180°, or positive or negative, respectively. This causes what is known as the phase problem in X-ray determination of crystal structures. There are several methods of overcoming the phase problem: 1. The structure has to be roughly known, so that Fca!c(hkl) = Fm e a s(hkl) and qb(hkl), then least squares refinement is used until there are no more changes in qb(hkl). However, in the case of bigcreekite and UK6 there is no starting structure. 2. Patterson function, a heavy atom method, which uses vector methods. 3. Direct methods which is based on statistical relationships between the phases. As bigcreekite, walstromite, verplanckite and UK6 all contain barium, a heavy atom, the Patterson method was used. The Patterson method is a Fourier series which can be calculated directly from the experimental intensity data, not requiring the phase information. The result is not interpreted as a set of atomic positions, but rather as a collection of interatomic vectors all taken to a common origin. The 3-dimensional Patterson function The u, v, and w are replacing x, y, and z and are vectors now. The Patterson function explores the same field, the unit cell, as the electron density function, but instead gives a map of the is: P(u,v,w)= — ^ J ] 2 ] | F ( M / ) | 2 C O S 2 ^ ( / Z W + kv+ lw) h k I 33 interatomic vectors. In a structure with N atoms per unit cell, each atom forms a vector with the remaining N- l atoms, giving N(N-1) non-origin peaks. The Patterson unit cell is the same size and shape as the crystal unit cell, but it has to accommodate N 2 rather than N "peaks". The peaks in Patterson space tend to overlap when there are many atoms in the unit cell, causing difficulty when interpreting the electron density map. A heavy atom in the structure has a larger Patterson peak height and is easier to locate. The heavy atoms can be used as the trial structure and the lighter atoms can be found around it. With a trial structure the | Fcalc(/zA7) | and the phase (pcah(hkT) can be calculated. The structure factor can be expressed in terms of known (c) and unknown (u) atoms: Fcaic(hkl)= Fc(hkl)+ F»(hkl) As more of the structure becomes known, the unknown structure factors approach the correct calculated structure factors and the phase angle approaches the unobservable but required value cj)(hkl), called partial structure phasing. If the model includes atoms in reasonably accurate positions, the atoms should appear in the correct positions, and additional atoms should be revealed by the presence of peaks in sensible positions. 3.4 Structure Refinement Once a trial structure has been obtained, the refinement process can be carried out. The parameters refined are the scale factor and positional and thermal parameters for each atom. The three fractional coordinates for each atom are refined except when an atom is on a special position 34 or the origin of the unit cell must be fixed. The number of thermal parameters included in the refinement depends on the description of thermal motion that is used. If an atom is permitted only spherical or 'isotropic' thermal motion, then there is one thermal parameter for the atom included in refinement. The scattering factors used for each atom are corrected for thermal motion according to the equation: Where u 2 is the mean square displacement of the atom from its average position and f is the atomic scattering factor. When the thermal motion of an atom is described as an ellipsoid (anisotropic) there are six thermal parameters for the atom that are included in the refinement. The thermal motion correction applied to the scattering factors for this atom is: The atoms in the structures of bigcreekite, UK6, walstromite and verplanckite were refined as anisotropic. Scattering curves and anomalous dispersion coefficients were taken from Cromer and Mann (1967) and Cromer and Liberman (1970). The parameters were refined using full-matrix least squares methods. The minimized function is: / = f° exp - 2 f f 2 X E * > ; ' j 35 where k is used to scale the calculated to measured structure factors and a weight w is given to each reflection. In initial stages of refinement, each reflection is assigned unit weight; as convergence is achieved, a final weighting scheme is used. In this work, the weighting scheme is chosen such that uniform averages of wA 2 for all ranges of F m e a s are obtained. Weak reflections are classed as unobserved, because they are not measured accurately, and are therefore excluded from the refinement process. When the trial structure that is being refined is not complete an electron density map is calculated. A differential AF or difference Fourier synthesis is calculated: where pm and pc are measured and calculated electron densities. The result is a map of residual electron density not accounted for in the trial structure. Once heavy atoms had been located by Patterson methods and refined as a trial structure, a difference Fourier will yield the positions of the remaining non-hydrogen atoms. The accuracy of the structure model can be measured by the agreement of calculated and measured structure factors. This is referred to as the residual discrepancy factor, R-factor or weighted R-factor: ce" h k I hkl c o s 2 # ( / 2 x + ky+ Iz 36 1 Fc 2) 2 2 J For a well refined structure model, the value of R approaches a small value (approximately 3%), corresponding to the errors in both the experimental data and the model. 3.5 Errors With the method of least squares refinement, both the new values for the refined parameters and their errors are obtained. The error, a(p), for a parameter p is dependent on the number of observations and parameters, the precision of the data, and the agreement of observed with calculated structure factor magnitudes. Errors in the atomic coordinates have been used to calculate the errors in bond lengths and angles that appear in the tables of Chapters 4, 5, 6 and 7. 3.6 Programs and Procedure The Siemens SHELXTL PC system of programs were used throughout this study. The procedure for the structure solution and refinement of each crystal was different and is detailed in their respective chapters. Generally, for structure solution, the heavy atoms of the minerals (barium and chlorine), were located using the Patterson method. This was followed by inputting the rest of the lighter elements. The lowest R-factor possible was achieved by the following steps: setting all cations and anions anisotropic, restricting observed data to larger o(F) values, omitting 37 individual reflections, using a weighting factor, and refining atoms versus a substituting element. Substituting elements were determined with electron-probe microanalysis. 3.6.1 MISSYM The program MISSYM (Le Page, 1987) was used after structure solution and refinement to confirm there is no missing symmetry. Using atomic positions, the program is able to detect overlooked symmetry and quasi-symmetry. Inaccurate data or disorder of atoms may lead to incorrect results. MISSYM uses default errors of 1° , for the obliquity of metric symmetry elements, and 0.25 A for atomic positions. This serves to create a range that safeguards against slightly inaccurate data but it also may create problems by causing the detection of extra symmetry elements in a correct structure. The program is not able to prove the presence of symmetry elements nor is it able to discriminate symmetry from quasi-symmetry and so it is up to the user to examine the experimental data to confirm the reality of these symmetry elements. No missing symmetry was found in bigcreekite, walstromite, verplanckite or the final structure of UK6. There was missing symmetry in the initial stages of the structure solution of UK6 which helped lead to the final space group used. 3.6.2 STRUCTURE TIDY The program STRUCTURE TIDY (Gelato and Parthe, 1987) was used before the final refinement of bigcreekite, UK6 and verplanckite to standardize the atomic positions according to 38 the crystallographic conventions proposed by Parthe and Gelato (1984). This program is useful for the comparison of similar structures. The input consists of the minerals space group, unit cell parameters, and atomic positions. The program's output is a reordered list (based on Wyckoff letters) of standardized atomic positions and a standard space group setting. The origin and orientation of the structure is chosen as the setting with the minimum standardization parameter, where, N is the number of representative atoms in the asymmetric unit and x, y, z are the atomic coordinates. The program STRUCTURE TIDY was not used with walstromite as it is a refinement. The setting used was from the original crystallographic description. STRUCTURE TIDY was used with verplanckite because it resulted in a lower R-value than the settings in the original crystallographic description. 3.7 Electron-probe Microanalysis Chemical analyses of bigcreekite were done by Dr. Bob Gault at the Canadian Museum of Nature, located in Ottawa Ontario, on a JEOL 733 electron microprobe using Tracor Northern 5500 and 5600 automation, operating in the wavelength-dispersion mode, with the following operating conditions: accelerating voltage 15kV; beam current 20 nA; peak count time 20 s; 39 background count time 10 s; beam diameter 10 um. Data reduction was done with a conventional ZAF routine in the Tracor Northern T A S K series of programs. Chemical analysis of UK6, walstromite and verplanckite were done on a fully automated C A M E C A SX-50 electron microprobe, operating in the wavelength-dispersion mode, with the following operating conditions: accelerating voltage 15 kV; beam current 20 nA; peak count time 20 s; background count time 10 s; beam diameter 5 \im. Data reduction was done with the 'PAP' 4>(pZ) method (Pouchou and Pichoir, 1985). 3.8 Infrared Spectroscopy Infrared spectra were collected for bigcreekite and UK6 by Elizabeth Moffatt at the Canadian Conservation Institute. Experimental details and results for bigcreekite are given in Chapter 4. 3.9 Bond Valence After the final phase of structure refinement and electron-probe microanalysis, bond valence analysis was done for each mineral. Bond lengths are related to bond valence because the bond-valence summation for a particular atom is assumed to be distributed between the bonds it forms, making them equal to the valence number of that atom. This provides a check of the "correctness" of the structure and bonding model. The valence of an atom, V f , is defined by the summation of bond valences, Vy, of bonds between the central atom i and its surrounding atoms, j : 40 I vs = Vi The bond-valence parameter used from Brese & O'Keefe (1991) is described by: where, R ; j is a bond-valence parameter between pairs of atoms in observed crystal and molecular structures, dy is the bond length, and b is a 'universal' constant equal to 0.37 A (Brese & O'Keefe, 1991 from Brown & Altermatt, 1985). The bond-valence parameters from Brese & O'Keefe (1991) were used to generate the bond-valence tables for each crystal. 3.10 Gladstone-Dale Rule The Gladstone-Dale relationship (compatibility index) is a measure of the compatibility of the mean index of refraction, density and chemical composition. It is used to determine the internal consistency of data for a compound. The compatibility index (1- K p / K c ) is a numerical value which fits into a scale. K p and K c were described by Jaffe (1956) as follows: Where K is the specific refractive energy of the mineral; ka are the specific refractive energies of the components of the mineral; p a are the weight percent of the components; n is the mean index 41 of refraction of the mineral; and d is the measured taken from Mandarino (1981), as well as the scale Compatibility Index ± 0 . 0 0 0 to ± 0 . 0 1 9 ± 0.020 to ± 0.039 ± 0.040 to ± 0.059 ± 0.060 to ± 0.079 > ± 0.079 The compatibility index was calculated for the ne\ UK6 in the future. density of the mineral. The values for ka were used in this thesis: Category Superior Excellent Good Fair Poor v mineral bigcreekite and will be calculated for 42 4.0 Bigcreekite: New Mineral Description 4.1 Introduction Bigcreekite was first discovered by Robert E. Walstrom in 1980 in blast rock derived from the Big Creek road construction in the 1960's in eastern Fresno County, California. Bigcreekite was recognized as a new mineral, containing Ba, Si and traces of Ca (Appleman, 1983). Bigcreekite was named for Big Creek, California, it's type locality (Pers. comm., G. Dunning). General details of the experimental methods used in this chapter are found in Chapter 2. 4.2 Occurrence There are two known occurrences of bigcreekite, both within sanbornite deposits. The first occurrence is at the site of the Esquire # 7 claim along Big Creek in eastern Fresno County, California. In this area, the sanbornite deposits consist of gneissic metamorphic rocks which are composed mainly of variable amounts of sanbornite and quartz and minor amounts of pyrrhotite, diopside and other rare barium minerals (Alfors et al. 1965). These metasediments have been cross-fractured in places, allowing later fluids to deposit additional minerals or alter earlier ones. Bigcreekite formed along very thin (less than 0.5 mm) transverse fractures in fairly well laminated quartz-rich sanbornite portions of the rock. It is colourless and forms poorly developed crystalline masses parallel to the fracture direction (Figure 4.1). Bigcreekite appears to have formed during or after fracturing of the metamorphosed sediments. It postdates the other 43 Figure 4.1. Vein of bigcreekite (arrow) in quartz-sanbomite rock. Scale bar 1 cm. 44 associated barium silicates and may represent either a later primary phase from intruded fluids or a product of alteration of pre-existing Ba-rich minerals, possibly sanbornite or others within the laminated metamorphosed sediments. The second occurrence of bigcreekite is from Trumbull Peak, Mariposa County, California, where it occurs as thin seams cutting quartz-sanbornite-gillespite rock associated with titantaramellite, pyrrhotite, celsian, alforsite and witherite. Other rare barium containing silicates that are associated with the bigcreekite and the sanbornite are: muirite, verplanckite, celsian, benitoite, titantaramellite, fresnoite, and uvarovite. No other minerals have been noted along the thin fractures other than bigcreekite. 4.3 Physical and Optical Properties Bigcreekite occurs as poorly developed crystalline masses, millimeters in length, parallel to the fracture direction. There are two perfect cleavages {010} and {001}. It has a tabular habit, elongated along [100] direction. The cleavable masses are brittle, white to colourless, non-fluorescent, with vitreous to pearly lustre and a white streak. Bigcreekite has an approximate Moh's hardness of 2 to 3. The density, 2.66(3) g/cm3, was measured by suspension in bromoform. It is slightly low compared to the calculated density, 2.739 g/cm3, due to the small size of the crystals and the perfect cleavage, causing air to be trapped on the crystal surfaces and within the crystals. Bigcreekite is biaxial positive with the indices of refraction a 1.537(2) P 1.538(2) y 1.541(2) for 590 nm; 2V m e a s = 59.2(5)°, 2V c a l c = 60.1 °; dispersion is moderate, r<v. No 45 pleochroism was seen. The optical orientation is X = b, Y = a, Z = c. Application of the Gladstone-Dale rule (Mandarino, 1981) gives a compatibility index of 0.017, which is considered to be excellent. Andy Roberts from the Geological Survey of Canada collected X-ray diffraction data for bigcreekite (Table 4.1). Unit cell data refined from the powder data are: a = 5.038(6) A, b = 9.024(3) A, c = 18.321(6) A. 4.4 Chemical Composition 4.41 Electron-microprobe analysis Electron-probe microanalyses were done with the following operating conditions: excitation voltage 15 kV, beam current 20 nA, peak count time 20 s, background count time 10 s and beam diameter 5 pm. For the analyzed elements, the following standards and X-ray lines were used: albite (NaKa), celestite (SrLoc), sanbornite (SLKoc, BaZ,a), diopside (Ca£a) , sanidine (KKa), and phlogopite (F^a). The following elements were analyzed for but were not detected: Ca, Sr, and F. Based on 3 cations per formula unit, the empirical formula (Ba[ oNaooj), ,Si, 9 9 H 8 0 2 O 9 was determined from an average of 4 analyses of bigcreekite with H 2 0 calculated by stoichiometry. The ideal formula for bigcreekite is BaSi 20y4H 20. To determine the amount of water in the structure of bigcreekite, structural information was needed. Table 4.2 gives the results for the electron-probe microanalyses. Severe burn-up of the sample occurred during the electron-microprobe analyses, resulting in the high wt. % totals obtained. The presence of H 2 0 was confirmed by infrared-spectrometry. 46 TABLE 4.1 X-RAY POWDER-DIFFRACTION DATA FOR BIGCREEKITE (Pers. comm., A. Roberts) 'est. d(meas.) d(calc) hkl 'est. ^(meas.) d(calc.) hkl 30 9.189 9.161 002 30 1.845 1.847 119 100 5.068 5.058 013 3 1.816 1.815 046 3 4.861 4.858 101 20 1.732 1.731 053 20 4.575 4.580 004 3 1.709 1.709 235 30 4.418 4.414 102 1.707 146 20 4.265 4.277 111 1.698 0210 85 4.54 4.048 022 20 1.695 1.695 208 5 3.875 3.886 103 1.692 151 30 3.572 3.569 113 3 1.663 1.665 218 30 3.398 3.395 015 25 1.637 1.638 0111 3 3.299 3.306 121 1.637 153 3 3.206 3.214 024 3 1.620 1.619 303 5 3.153 3.155 122 3 1.604 1.599 139 15 3.056 3.054 006 3 1.578 1.586 228 45 2.974 2.968 031 1.568 321 10 2.814 2.815 115 10 1.557 1.558 1111 60 2.706 2.710 124 3 1.540 1.541 155 20 2.614 2.611 106 15 1.525 1.528 245 5 2.558 2.557 131 1.527 0012 30 2.512 2.519 200 1.527 305 2.508 116 10 1.503 1.504 060 5 2.43 2.426 210 10 1.485 1.488 324 40 2.327 2.325 035 1.485 156 10 2.291 2.290 008 3 1.460 1.467 250 75 2.257 2.260 126 1.542 316 20 2.191 2.191 042 5 1.446 1.448 149 5 2.138 2.139 222 25 1.424 1.425 157 10 2.07 2.069 223 1.422 0410 25 2.042 2.042 028 3 1.399 1.397 334 5 2.01 2.009 142 20 1.391 1.390 1212 15 1.974 1.975 037 3 1.373 1.373 2111 3b 1.928 1.931 230 3 1.348 1.349 327 1.921 231 30 1.304 1.304 159 1.890 232 1.303 166 25 1.884 1.887 1.878 109 144 114.6 mm Debeye-Scherrer powder camera, Cu radiation, Ni filter (I CuKa = 1.54178 A), visually estimated intensities; b = broad line, not corrected for shrinkage and no internal standard, indexed with a = 5.038, b = 9.024, c = 18.324 A 47 TABLE 4.2. ELECTRON-PROBE MICROANALYSES OF BIGCREEKITE Weight % Pt. 1 Pt. 2 Pt. 3 Pt. 4 Ave. Na 2 0 0.26 0.09 0.00 0.07 0.105 CaO 0.00 0.08 0.04 0.00 0.03 SrO 0.02 0.02 0.00 0.03 0.02 BaO 46.59 48.56 50.39 49.97 48.88 S i 0 2 37.59 39.42 37.41 38.22 38.16 HzO* 22.53 23.48 22.87 23.17 23.013 SUM 106.99 111.65 110.71 111.46 110.20 Atomic Proportions Pt. 1 Pt. 2 Pt. 3 Pt. 4 Ave. Na+ 0.027 0.009 0.000 0.007 0.01 Ca 2 + 0.00 0.004 0.002 0.000 0.00 Sr2* 0.001 0.001 0.000 0.001 0.00 Ba 2 + 0.972 0.972 1.036 1.014 1.00 Si 4 + 2.001 2.014 1.962 1.978 1.95 H+ 8.000 8.00 8.000 8.000 8.00 o 2- 8.987 9.009 8.962 8.975 8.98 Note: analyses are normalized on a basis of 3 cations per formula unit. 'Determined by stoichiometry. 4.42 Infrared analysis The infrared spectrum of bigcreekite was obtained using a Bomem Michelson MB-120 Fourier-transform spectrometer with a diamond-anvil cell microsampling device. H 2 0 is present in the structure of bigcreekite seen from the infrared spectra (Pers. comm., E. Moffatt). 4.5 X-ray Crystallography and Crystal-Structure Determination 4.51 Experimental The crystal used in this study is from Esquire # 7, Big Creek, Fresno County, California. The unit cell was derived using 50 reflections in the range 9.49 to 49.59° 20. Cell dimensions were refined using the method of least squares (Table 4.3). A full sphere of reflections (9549 measurements, exclusive of standards) were collected from 3 to 60° 20. Thirty-six of the reflections were rejected due to systematic absence violations and 574 reflections were rejected due to inconsistent equivalents, leaving 1274 reflections of which 8 are suppressed. Fourteen strong reflections uniformly distributed with regard to 20 were measured at 5° intervals of i|/ (the azimuthal angle corresponding to rotation of the crystal about its diffraction vector) from 0 to 355° , after the method of North et al. (1968). Of the data collected (1015 reflections) there were 27 reflections rejected due to bad backgrounds, leaving 988 reflections that were used to calculate an absorption correction. The merging R index for the ij/-scan data set decreased from 14.6% before the absorption correction to 3.6% after the absorption correction. This correction was then applied to the entire data set; minimum and maximum transmissions were 0.36 and 0.84 respectively. The data were also corrected for Lorentz, polarization and background effects, 49 T A B L E 4.3. M I S C E L L A N E O U S INFORMATION FOR BIGCREEKITE a (A) 5.0453(8) Rad/mono MoKa/graphite b 9.044(1) c 18.366(5) Total reflections 9549 V(A3) 838.0(3) Unique reflections 1274 Space group Pnma Rjnt % 4.7 Z 4 F 0 > 4 a (F 0) 1244 Crystal size (mm) 0.2 x 0.2 x 0.4 R (observed) % 3.5 u. (MoKa; mm"1) 5.04 mm"1 Rw (observed) % 9.6 fi = I(|F0|-|Fc|)/I|f%| averaged and reduced to structure factors. Of the 1274 unique reflection, 1244 were classed as observed [F0 ^ 3 a (F0)]. 4.52 Structure solution and refinement Miscellaneous collection and refinement data are given in Table 4.3. The convention c<a<b is not followed for the sake of comparing bigcreekite to other barium sheet silicates. The mean value of | E 2 - 11 was found to be 0.921, which implies a centrosymmetric space group. Patterson techniques were used to solve the structure due to the presence of the barium. The structure was refined in Pnma to an R index of 8.9% for an isotropic displacement model. Conversion to anisotropic displacement factors for all of the atoms in the structure resulted in convergence at R and Rw indices of 3.5% and 9.6% respectively (R = 3.65% for all 1274 data). Addition of an isotropic extinction correction did not improve the results. Positional coordinates and anisotropic and equivalent isotropic displacement factors are given in Table 4.4. Selected interatomic distances and angles are given in Table 4.5, and a bond-valence analysis in Table 4.6. 51 N 0 CO T - CD CO. CO, CO 2 -CO i n CD CO CD i n CO m T — CD CM CM CO § o 00 C\f If) CM o O o 1 CO CO i n co CM, co" r -o CD — CQ CO CD, 00 i n , co CM CD i n CO CM, c\T CO ^ — T i n 1 CM 5 o 1 CNT CO o o o CM CO T — oo r -CO* X — in" CM CO CO CO m CM CO CD x— CD O CO CN? CO* LpT 00 ^ CO i n CM 00 o 00 CM i n CO CM CO •sr OT CM, CO 00, CD CN? CM, CM CO CM CNI c\T CD Co" CO CM O) CO CD cn ^_ CM, CM, CM, 00 OT CO 00 o T CD 00 CO CO o CO •ST i n o CM CD O CM i n CNI CM CM o 00 d d d d d d d d o ' 5-CM CM 00 CD O O CD d d d d • 1 N J CO CO m i n CD 3T CD o m -—-in CD T — CO o o 00 CD o x— CO CD m T — T — CM CM CO CNI o d d d d d d d CM o co O O m co O O T A B L E 4.5. S E L E C T E D INTERATOMIC DISTANCES (A) A N D A N G L E S (°) FOR BIGCREEKITE Ba-0(1)&0(1) f 2.807(3) 0(1)g-Ba-0(2)i 53.86(6) Ba-0 (1 )a&0(1)g 3.279(3) 0(1)g-Ba-0(5)c 60.00(5) Ba-0(2)i & 0(2)j 2.710(2) 0(1)g-Ba-0(5)d 122.25(5) Ba-0(5)c 2.865(3) 0(1)g-Ba-0(6)j 61.88(6) Ba-0(5)d 2.947(3) 0(2)h-Ba-0(2)i 117.90(8) Ba-0(6)j 2.861(4) 0(5)c-Ba-0(1) 69.76(6) <Ba-0> 2.918 O(5)o-Ba-O(1)f 69.76(12) 0(5)c-Ba-0(2)h 113.84(5) Si-0(2) 1.584(2) O(5)o-Ba-O(2)i 113.84(5) Si-0(3)c 1.638(2) 0(5)c-Ba-0(6)j 70.79(12) Si-0(3) 1.641(2) 0(5)d-Ba-0(1) 67.18(6) Si-0(4) 1.610(1) 0(5)d-Ba-0(1)f 67.18(6) <Si-0> 1.618 0(5)d-Ba-0(2)h 94.08(5) 0(5)d-Ba-0(2)i 94.08(5) 0(1)-Ba-0(2)h 76.85(8) 0(5)d-Ba-0(5)c 120.48(12) 0(1)-Ba-0(2)i 157.81(7) 0(5)d-Ba-0(6)j 168.73(14) 0(1)a-Ba-0(1) 59.93(8) 0(6)j-Ba-0(1) 120.22(9) 0(1)a-Ba-0(1)f 125.43(3) 0(6)j-Ba-0(1)f 120.22(9) 0(1)a-Ba-0(2)h 53.86(6) 0(6)j-Ba-0(2)h 80.24(7) 0(1)a-Ba-0(2)i 141.79(6) 0(6)j-Ba-0(2)i 80.24(7) 0(1)a-Ba-0(5)c 60.00(5) <0-Ba-0> 96.36 0(1)a-Ba-0(5)d 122.25(5) 0(1)a-Ba-0(6)j 61.88(6) 0(4)-Si-0(2) 111.88(14) 0(1)f-Ba-0(1) 84.63(13) 0(3)c-Si-0(2) 109.70(10) 0(1)f-Ba-0(2)h 157.81(7) 0(3)-Si-0(2) 114.17(10) 0(1)f-Ba-0(2)i 76.85(8) 0(3)c-Si-0(4) 108.98(11) 0(1)g-Ba-0(1) 125.43(3) 0(3)-Si-0(4) 107.91(12) 0(1)g-Ba-0(1)f 59.93(8) 0(3)-Si-0(3)c 103.80(6) 0(1)g-Ba-0(2)h 141.79(6) <0-Si-0> 109.41 Note: <fv1-<)>> denotes the mean metal-ligand distance (A). Equivalent positions: a = -x, -y, -z + 1; b = -x + 1, -y, z - Y2; c = x - Y2, y, -z + !4; d = x + V2, y, -z + Vi\ e = x + 1 / 2 , y, -z + 3/ 2; f = x, -y + V2, z; g = -x, y + !4, -z + 1; h = -x + V2, -y, z + Y2; i = -x + Vi, y + !4, z + V2, j = x - V2, y, -z + %. T A B L E 4.6. B O N D - V A L E N C E * A R R A N G E M E N T IN B IGCREEKITE Si Ba Total 0(1) 0.247 x 2 i 0.316 0.069 x 2 i 0(2) 1.114 0.321 x 2 i 1.436 0(3) 0.955 1.918 0.963 0(4) 1.039 x 2 i 2.077 0(5) 0.169 0.381 0.211 0(6) 0.214 0.214 Total 4.071 1.870 Calculated from the curves of Brese & O'Keeffe (1991). 4.6 Description of the Structure There are two distinct cation sites in the structure of bigcreekite. The atom at the Ba site, at special position 4(c) (x, lA, z), is coordinated by nine oxygen atoms, forming a very distorted pyritohedron (Figure 4.2). The Ba-0 distances range from 2.710 to 3.279 A (mean 2.918 A), and the O-Ba-0 angles vary from 53.86 to 168.73° (mean 96.36°). The polyhedral volume is 44.37 A 3 . The electron-microprobe data, refined site occupancy, and bond valence analysis confirm that the site is completely filled with Ba. The atom at the Si site, at position 8(d) (x,y,z), is coordinated by four oxygen forming a distorted tetrahedron. The Si-0 distances range from 1.584 to 1.641 A (mean 1.618 A), and the O-Si-0 angles range from 103.8 to 114.17° (mean 109.41 °). The polyhedral volume is 2.17 A 3 , with a tetrahedral angle variance of 12.5 and a mean tetrahedral quadratic elongation of 1.003. The electron-microprobe data, refined site occupancy, and bond valence analysis confirms that the site is completely filled with Si. Bigcreekite is a hydrous sheet silicate containing four-membered rings, which form chains of silica tetrahedra [Si 20 5], parallel to the a axis and staggered along c, with a repeat distance of two tetrahedra. Figure 4.3 shows the crystal structure of bigcreekite. Atoms 0(3) and 0(4) connect the tetrahedra to form the four-membered rings (Figure 4.4a), which are connected along the a axis by 0(3) atoms (Figure 4.5). The fourth atom in the tetrahedra, 0(2), connects the tetrahedral layer to the barium polyhedral layer. The Ba layers contain Ba0 9 polyhedra, which form sheets on (001), resulting in perfect cleavage in this direction (Figure 4.6). The Ba polyhedra are joined together by four 0( 1) atoms, 55 Figure 4 ,2b. Bar ium po lyhedron in b igcreeki te. 56 Figure 4.3. The crystal structure of b igcreeki te, showing the cha ins of s l icon te t rahedra a n d ba r i um a t o m s (dark spheres) paral le l to (100). The light spheres a re oxygen a toms. 57 Figure 4 .4b. Sil icon te t rahedra forming 4 - m e m b e r e d rings in gil lespite. 0(2) 0(3) 7 \ 4 o(4; Sil ica cha ins a l o n g [100] with a repea t d i s t ance of two te t rahedra with ba r ium a t o m s (spheres) b e t w e e n the cha ins . Figure 4.6. A sheet of ba r ium po l yhed ra in b igcreeki te , responsib le for the per fec t c l e a v a g e o n (001). 60 one 0(5) atom and one 0(6) atom, sharing two edges. Two 0(2) atoms connect each Ba polyhedra to two different rows of Si tetrahedra. Atoms 0(1), 0(5) and 0(6) are part of H 2 0 complexes and some weak hydrogen bonding may affect 0(2). There are large spaces between the rows of silica tetrahedra along [100], where the H atoms are most likely located, causing the perfect cleavage on (010) (Figure 4.7 ). This was confirmed with bond valence analyses, which are given in Table 4.6. 4.6 Discussion Bigcreekite is very similar to krauskopfite, BaSi 20 4(OH) 2 • 2H 20. They are identical in hand specimen and both form as thin vein minerals in sanbornite-bearing rock, but krauskopfite is closely associated with macdonaldite, whereas bigcreekite is not. Structurally they are somewhat similar. Krauskopfite is a single-chain silicate with the barium atoms located between the chains (Coda et. al., 1967) (Figure 4.8). Bigcreekite is also a chain silicate, but with double chains. The barium atoms are also located between the chains of Si tetrahedra. The repeat distance of the Si tetrahedra is five, running parallel to the c axis in krauskopfite, while bigcreekite tetrahedral chain repeat distance is two, parallel to a. Two vertices of the Si tetrahedra in krauskopfite are oxygen atoms and the other two are OH groups, while there are no water complexes bonded to the tetrahedra in bigcreekite. Like bigcreekite, krauskopfite also has two perfect cleavages, on (010) and (100). Bigcreekite also has similarities to sanbornite, BaSi 20 5 , and gillespite, BaFeSi 4O 1 0, in composition and occurrence. Sanbornite is a silicate sheet structure with the sheets on (001) and 61 •a b (010) c leavage Figure 4.7. Bigcreeki te structure p ro jec ted on to (100) showing voids con ta in ing hyd rogen a toms . A toms 0(5), 0(6) a n d 0(1) a re part of water c o m p l e x e s . Dark spheres a re ba r ium a toms , light spheres a re o x y g e n a t o m s a n d the te t rahedra a re si l icon te t rahedra. 62 a barium polyhedral sheet lying between the silica tetrahedral sheets (Figure 4.9). Each sheet consists of continuously linked, distorted, 6-membered rings of Si0 4 tetrahedra (Douglass, 1958). Gillespite is also a silicate sheet structure made up of linked, four-membered rings of Si0 4 tetrahedra (Pabst, 1943). Figure 4.10 shows the structure of gillespite and Figure 4.4b shows the four membered tetrahedral rings. Barium atoms lie between the sheets and are associated with oxygen atoms of both adjacent sheets, as in sanbornite. Iron atoms in gillespite are also between the silicon tetrahedral sheets, but are only bonded to one of the sheets. In the structure of bigcreekite, barium atoms are not bonded to both the upper and lower silicon sheets. Along the a and b axes, Ba atoms alternate bonding with the upper and lower silicon sheets. The largest difference between bigcreekite, sanbornite and gillespite is that bigcreekite contains water. There are water complexes lying between the rows of silicon tetrahedra, bonded to the Ba polyhedra. This causes a substantial increase in the cell dimensions b and c compared to sanbornite and causes a doubling of the cell volume. The unit cell parameters of sanbornite are: a = 4.63 A b = 7.69 Ac - 13.53 A, cell volume = 482 A and the cell parameters of bigcreekite are: a = 5.0453 A b = 9.044 A c = 18.366 A, cell volume = 838.0 A. The water in the structure of bigcreekite causes it to break down easily. It visibly de-waters quickly under vacuum, causing the crystals to alter to a fibrous mass. 64 Figure 4.9. Crystal structure of sanborni te. Tetrahedra a re si l icon te t rahedra, dark spheres a re ba r ium a toms . 65 Figure 4.10. Crystal structure of gil lespite, The te t rahedra a re si l icon te t rahedra, the large dark spheres a re ba r ium a t o m s a n d the smal l dark spheres a re iron a toms . 66 5.0 UK6: New Mineral Description 5.1 Introduction Unknown mineral 6 (UK6) was discovered by Robert E. Walstrom in 1964 at the Esquire #1 claim in Fresno County, California (Figure 3.3). The exposed site containing UK6 measures approximately 1 x3m. Rare grains of UK6 have also been found at Esquire #7 claim along Big Creek. General details of the experimental methods used in this chapter are found in Chapter 2. The details of the structure and formula of UK6 given here are at the time of writing; there may be revisions at a later date. 5.2 Occurrence UK6 occurs as irregular masses up to 10 mm in diameter, enclosed in parallel-bedded sanbornite-quartz rock (Figure 5.1). It formed prior to sanbornite and quartz but the lack of iron in UK6 indicates that it probably formed later than traskite, titantaramellite and unknown mineral 10 (Pers. comm., G. Dunning). UK6 is associated with sanbornite, traskite, titantaramellite, fresnoite, celsian, witherite, macdonaldite, pyrrhotite, unknown mineral 10, unknown mineral 21, alforsite, and quartz. Small inclusions of unknown mineral 10 have been observed as cores within UK6. There are also inclusions of celsian and witherite in UK6. 67 Figure 5.1. Mass of UK6 (arrow) in quartz-sanbornite rock, Scale bar 1 cm. 68 5.2 Physical and Optical Properties UK6 is light blue-grey (Figure 5.1) and has a well developed {0001} cleavage (Figure 5.2). The cleavable masses are brittle, non-fluorescent and have a vitreous lustre and white streak. UK6 has an approximate Moh's hardness of 3 and a calculated density of 3.709 g/cm3. The measured density of UK6 could not be determined due to small crystal size, poor crystallinity and numerous inclusions. UK6 is uniaxial negative with indices of refraction co = 1.642(2) and e = 1.594(2) (for X = 589 nm); dispersion is slight, r<v. There is no noticeable pleochroism and extinction is parallel to the {0001} cleavage. Figure 5.2 shows UK6 in thin section under crossed polars. 5.4 Chemical Composition Electron-probe microanalyses were done with the following operating conditions: excitation voltage 15 kV, beam current 20 nA, peak count time 20 s, background count time 10 s, and beam diameter 5 pm. For the analyzed elements, the following standards and X-ray lines were used: barite (BaLoc), SrTi0 3 (SrZoc), diopside (SLKa, MgKa, CaKa), synthetic rhodonite (MnATa), scapolite (CLKa), grossular (ALKa), albite (NaKa), synthetic fluorphlogopite (FKa), rutile (YiKa) and synthetic fayalite (FeKa). The following elements were analyzed for but were not detected: Mg, Mn, Sr, and Fe. Compositional variation occurs very rarely within grains of UK6 (Figure 5.3); lighter areas have approximately 0.5 wt. % less Al than the surrounding darker areas. Other than rare zoning, there is essentially no compositional variation between grains. Calculation 69 Figure 5.2. Photomicrograph of UK6 (center of field, third-order interference colours). Large surrounding grain with second-order interference colours is sanbornite; the grain to the left with first-order white-grey interference colours is celsian. Plane-polarized light, crossed polars. Field of view 5.1 mm. Figure 5.3. Backscattered electron micrograph of a zoned UK6 grain. White dots and numbers refer to electron-microprobe analyses given in Table 5.1, Scale bar 100 um. 71 from an average of fifteen analyses of UK6 gave the empirical formula (based on 7 cations and 2 H 2 0 per formula unit): ( B a ^ , N a ^ , Cao 0 1) 3 0 1(Si 2 6 3 , Al; 3 4 , Tioo^ClO^Cl, 6 1 , 2H 20, F 0 0 5 ) 3 6 6 . The ideal formula is (Si, Al)4Ba3C109(Cl, H 20) 4 . Table 5.1 gives the results for the electron-probe microanalyses. Because of chlorine and oxygen substitution and the presence of H 2 0 , structural information was used in solving the formula of UK6. 5.5 X-ray Crystallography and Crystal-structure Determination The crystal used in this study is from Big Creek, Fresno County, California. There was difficulty deterrnining the symmetry of UK6 due to equipment limitations and poor sample quality. Data were collected using a serial four-circle diffractometer at The University of British Columbia and using a diffractometer equipped with a charge-coupled device (CCD) detector at The University of Manitoba. The symmetry was then confirmed with precession photography. 5.5.1 Unit Cell and Space Group Determination Using the serial four-circle diffractometer at The University of British Columbia 50 reflections in the range 9.49 to 49.59° 20 were centered. From these reflections, a unit cell was derived and then refined using the method of least squares to give a hexagonal unit cell: a = 5.252(2), c = 14.948(7) A. A full sphere of reflections was collected according to this unit cell. After preliminary structure solution, the possible space groups determined were P3 m 1, P3, or P31 m. Using the program MISSYM (Le Page, 1987), it was determined that with these space groups the trial structures had 72 CD => LL o CO CO >-< o Cd o LU m O cc a. I z O cc I— o LU _ l LU LO LU _J CO < x: CD CU > CO CD CO o •xt o LO CM a> CD CM CD LO LO o LO CD io < O 00 d CM d d i ^ LO d •xt d • csi • d o LO x— CO LO 00 LO o CO CM CD 00 cn o o CD LO CD •xt o CM CD CD LO CL d oo d CM d d m d •xt d • CM t o •xt x— CO CM •xt CM LO o LO CM oo o 00 o •xt CD CO o CO •xt CL o cri d CM d d LO d CM •xt d • CM i o co LO CM •xt CM o 00 CM CO CM 1^  CT> •xt CD LO o o •xt CM o ai d CM d d LO d - •xt d • CM • -*— o Cxi •xt CD CD CM •xt o CM 00 CM o O CO LO CO o CM CD CD 0. o CO d CM d d i< LO d d • CM i d o T— CO IX- •xt •xt o •xt CM LO CM CM CO a> LO LO o oo LO •xt CO -1—» Q_ o 00 d CM d d 1 ^ LO d •xt d i CM i d o O o CO 1^  CO CO o CM CO •xt CM CO O CO LO o CM CD CO CD -t—» Q. d 00 d CM d d l< LO d d • CM i d o T— CD o 00 LO 00 00 LO o CM CD CO o oo CO LO o CD CD 1^  cn a! d oo eri d d l< LO d •xt d • CM i ai CT) 00 CM 00 LO •xt CD LO o CO co LO •xt oo CD LO LO o CD CM a! o 00 ai d d 00 in d T— •xt d • CM i T— o T— i< o CD •xt CD CM o •xt CM CM cn LO CD CO LO LO O CO CO CT) CL d cd cri d d LO d •xt d • CM i d o T— CD CO o CD LO LO oo •xt o cn CM •xt 5; •xt LO LO o LO LO o a! d oo ai d d i< LO d ^r' d • CM • d o LO o •xt LO oo CO CO o CD CM 00 LO •xt CM LO CD O LO CD oo 00 CL d oo cri d d 1 < LO d T— •xt d • csi 1 CD CD •xt o LO o CD LO o CN CD CD co CN a> LO oo o cn CD LO CM CL d 00 ai d d l< LO d - •xt d • csi • d o CO ix-o CM o CO o LO o oo CM a> 1^  CO co cq o LO LO o 00 CD 00 o CL d od d CM d d LO d •<t d • CM i d o csi CM LO oo 00 CO o CD CM CT> o io LO LO LO o CM CD ^t LO CL d 00 ai d d 00 LO d •xt d • c\i i d o o> o io CD CO o CO CM LO oo cn CO cn CM LO 00 o CD CD LO CM CL d 00 ai d d LO d ^~ •xt d • CM i d o O CM < Z CO o _J < o CO O < O CM o H O < CD LL _ i O * O CN X n O _ l o II o _ l < H O H CO cz g •e o o. e CL O 0. o E o < CO •xt cn t— CO LO o 00 o CO CD O o cn o CD o cn d T— CM d d CM d c\i •xt d CO •xt CM CM •xt CO O CO o CO CO o o CD o LO O O d CM d d csi d CM •xt CO T - CM O T CD d •<- CM O O CM d d CO o d CM •>-CD O "xt CO •xt CO •xT CO •xt CO CD T -CD O |x- -r-CD O CD CD CM CO CD CM CO OO CO LO T - CM CO CO •xt CO CO CM O T-LO CD CM CD CM CM CO CO CO |x-CD CM O d CM O O d o o CD O csi d CM •>-CD O o o CO o o o CO T -O CD T - CM O O CM CM LO o cn CM O O CM CM CO o cn CM O O CM CM o O CM rx-<n O CM o CO CM O o o d co co a> o a> CM O O CM o o O O CO CO CM o o d co CO CM o o O CO o o o o d d co CM T -o o O O CO CO o LO CD o o cn CD d csi •xt d LO o CO CD o o o o d Cxi •xt T— CO o 00 LO o o CO o d csi •xt LO o CO LO o o CD CD d cxi •xt T _ UO o LO o o CO o d csi -xt T— •<t o LO CD o o CD CD d Cxi •xt d T— LO o •xt CD o o CM <n d csi •xt d LO o CD o o LO d Cxi •xt d LO o LO LO o o o d csi •xt T— CD o •xt CD o o •xt CD d Cxi •xt d oo o a> CD o o CD d c\i *t d CO o 00 CD o o LO CJ) d CM •xt d LO o CD LO o o CD d cxi •xt d 00 o oo CD o o in d CM •xt d Z < W O h < CO LL O X O CD E o o 'o "K >< TO CD 'E 1_ 0 -t—> <D Q E o CD C L O CX" X CM T J C CO CO c o CO o o CO CO CO CO c o ~o CD co E o (0 1 CD CO CD to CO cz CO CD 4—> O 73 missing symmetry, most notably a six-fold axis. Additionally, bond distances (e.g., Si-O) were not structurally appropriate. As many structures are twinned resulting in apparent hexagonal symmetry (Pers. comm., J. Grice), data for UK6 were collected again with orthorhombic symmetry at The University of Manitoba. Using a diffractometer equipped with a charge coupled-device detector (CCD), an orthorhombic unit cell, a = 5.2377(2), b = 9.0972(4), c = 29.859(1) A, was determined. Due to equipment limitations at The University of British Columbia, only half of the unit cell had been collected along the c axis giving a smaller unit cell and half of the needed data. The space group for the larger unit cell was determined to be Cmcm. A full sphere of reflections was collected according to this unit cell (details of the experimental procedure are given in section 5.5.2). Using the program MISSYM (Le Page, 1987), it was determined that the trial structure using Cmcm was missing six fold symmetry, making it evident that UK6 is hexagonal. The orthorhombic unit cell was transformed to a o hexagonal unit cell: a = 5.243(1), c = 29.859(6) A, with the transformation matrix 1, 0, 0; -0.5 ,0.5 ,0; 0 ,0, 1. The space group of UK6 was determined to be P63mc and confirmed by precession photography. A zero-level photograph taken perpendicular to (0110) shows the true distance between o reflections along c* , giving c = 29.59 A (Figure 5.4). A zero level photograph taken along [100] showed that hhlhl is missing when / = 2n, causing a doubling of the distance between reflections parallel to c (Figure 5.5). The length of a was determined from a zero-level photograph taken along c (Figure 5.6). The space group of UK6 was determined by hki I being present in all orders, hhlhl missing when l-2n and hhOl missing in all orders. These systematic absences give the following possible space groups: P63/mmc, P62c, P63mc, P3 le and P3\c. Of these space groups P63mc 74 hh2h0* Figure 5.5. Zero-level precession photograph taken along an a axis. Note the absence of hh2hl when I = 2n. 76 / \ a* Figure 5.6. Zero-level precession photograph taken along c 77 resulted in the lowest R value. Using the program MISSYM (Le Page 1987), no missing symmetry was found in the structure of UK6 with space group P6imc. 5.5.2 Experimental Using a CCD detector, X-ray diffraction data were collected by Dr. Joel Grice at The University of Manitoba. A number of reflections in the range of 4.48 to 30.01 ° 20 were centered. From this the unit cell was derived and refined using the method of least squares to give orthorhombic cell dimensions which were transformed to a hexagonal unit cell (Table 5.2). A full sphere of reflections (17381 measurements, exclusive of standards) was collected. Three of the reflections were rejected due to systematic absence violations and 168 were rejected due to inconsistent equivalents, leaving 919 reflections of which 99 were suppressed. Of the 919 unique reflections, 957 were classed as observed [F0 ^ 3o (F0)]. 5.53 Crystal Structure Solution and Refinement Miscellaneous collection and refinement data are given in Table 5.2. The mean value of | E 2 -11 is 0.964, which implies a centrosymmetric space group. Patterson techniques were used to solve the structure because of the presence of the barium. For an isotropic displacement model, the structure was refined in P63mc to an R index of 6.85% . Conversion to anisotropic displacement factors for all atoms resulted in convergence at R and indices of 4.85% and 14.58% respectively (R = 6.25% for all 919 data). Addition of an isotropic extinction correction did not improve the results. The site 78 T A B L E 5.2 M I S C E L L A N E O U S INFORMATION F O R UK6 a (A) 5.243(1) Rad/mono MoKa/graphite c 29.859(6) V{A3) 710.83 Total reflections 17381 Space group P63mc Unique reflections 1159 Z 2 Rint % 15.7 Crystal size (mm) 0.4 x 0.4 x 0.1 Fo > 4 a ( F 0 ) 957 u, (MoKa; mm"1) 5.04 mm"1 R (observed) % 4.85 Rw (observed) % 14.58 R = Z ( I ^ H ^ I ) / I > o l Rw = (Zw(\F0\-\Fe\)2 iZwFo2]0-6^^ occupancies of chlorine and oxygen sites were refined, significantly improving the results. Due to poor sample quality, oxygen sites were not precisely determined. Positional coordinates and anisotropic and equivalent isotropic displacement factors are given in Table 5.3. Selected interatomic distances and angles are given in Table 5.4, and a bond-valence analysis is given in Table 5.5. Bond valence analysis is not reliable due to the slight inaccuracy of oxygen positions, and was used with discretion. 5.6 Description of the Structure There are seven distinct cation sites in the structure of UK6; three contain barium and four contain silicon and aluminum. Ba(l) is at special position 2(b) (Vb, %, z) and Ba(2) and Ba(3) are at special positions 2(a) (0, 0, z). All Si/Al sites are at special positions 2(b). There are eight distinct anion sites. One is a chlorine site at special position 2(a); two sites contain both chlorine and oxygen, with Cl/0(1) at special position 2(b) and Cl/0(2) at special position 6(c) (x, x, z) and five contain oxygen, with 0(1) and 0(2) at special positions 6(c), and 0(3), 0(4) and 0(5) at special positions 2(b). The site occupancy of the Cl/O sites were refined giving site occupancies of: 42% Cl and 58% O at Cl/0(1); and 47% Cl and 53% O at Cl/0(2). Ba(l) is coordinated by twelve anions, three Cl(l) atoms, three 0(5) atoms, three Cl/0(1) atoms and three Cl/0(2) atoms, forming a distorted hextetrahedron (Figure 5.7). The Ba(l)-anion distances range from 2.787 to 3.673 A (mean 3.133 A) and the polyhedral volume is 72.607 A3. Ba(2) is coordinated by ten anions, one Cl(l) atom, six 0(2) atoms, and three Cl/0(1) atoms (Figure 5.8). The Ba(2)-anion distances range from 3.224 to 3.297 A (mean 3.257 A) and the polyhedral volume is 80 N CU CO 65s I f ? If) CD CD CNT CO T — CO CO CNI [104) 122) 158) 111) CO in IT) CO CO T — CN O CD m in CO CD CD CO CO CM in CM O 00 CO o CD CD in r»-o CD CO CNI CO 19441 CO CN CO CO CO CQ co~ o CM CO, in T — CNI CO o CO CN in o o o o o o o o o CD o o o CN ^ ^ ^ ^ , — ^ 00 m , , ^ — v , — v CD CO CO r CD x— 00 T — CO in CO, CO, in. 3, CD CO, CM, CM, 0C3, G> co' o ocT CO o in oo" O) oT CM CD O CO r— o o CN CN CD •rf CD 00 CD CD T — in in T — T — m T — CO CN , — . ^ . „ — v m CD , ^ s CO 00 CD m o in CD CO in T — CD co 00 00 CN •sr CO in T — CD 00 T - CD T — CN oo "<t 00 00 CO CD m 00 o T — (—\ CN T — CD OO CD CN in CN CO CD T — CNI T - 1 > CO CO t— CO CO o in ^ — v , s , ^ , ^ , , CD , , CD ^ , v v T — o CD , — ^ , , , . 00 O CD CM OO CO CO in CO CO o 2- CD, ^ ~ CD, T — co 52- CN CN, CM, x -•5— o oo o CO in CNT CO T — CN T T in CD CM CD 00 00 T — CN 00 o OO o CM CO oo in m 00 , , , s , ^ , v , v CO v CD s—s — s , — ^ T — o CD s ^ s v s ^ , ^ 00 o CD CN 00 CO CO m CO CD o in CD, h-. CD, T — CO CO, CN CM CM^  T — 2-CO (O r— in o co" o o CO m CNT o CO T — N- CN t— T — •sf in CD CM CD oo oo T — CN 00 o CO o CM CO 00 in r- in CO ^ , ^ in. CO CN, m 5, CO 3, CO, CN in. -^^  CO CO C\T CO CNT in a f in CM m CO CD CD r-- oo o CM o m m x— ^ — O o CD CD o CD o o 1^ - CO CN r>- m m o in CO T — CO CD 00 o o CO CN r-- o o d d d d d d d d d d d d d d CO CM CO CM CO CM CO CM CO CM CO CO CO CO CO ^ CO CO in ' r-- CD CD d d d i ^ ^ ^ — ^ CO, CO in CN o O m m in d d d CO CM CO <N CO-CO CQ CV c-r CM § T - CM o o CO CN CO co O O CO CM CO m O T A B L E 5.4 S E L E C T E D INTERATOMIC DISTANCES (A) A N D A N G L E S (°) FOR UK6 Ba(1)-CI/0(1)p& Cl/0(1)q &CI/0(1) Ba(1)-CI(1)l&CI(1)k &CI(1)m Ba(1)-0(6)l & 0(6)m & 0(6)n Ba(1)-CI/0(2)l & Cl/0(2)m & Cl/0(2)n <Ba(1)-anion> 2.787(7) 3.029(1) 3.042(16) 3.673(4) 3.133 Ba(2)-CI/0(1)a & CI/0(1)b&CI/0(1) Ba(2)-0(2) & 0(2)a & 0(2)o & 0(2)p & 0(2)r & 0(2)s Ba(2)-CI(1) <Ba(2)-anion> 3.224(3) 3.266(12) 3.297(12) 3.257 Ba(3)-CI/0(2) & Cl/0(2)p & Cl/0(2)a & Cl/0(2)r & Cl/0(2)s & CI/0(2)o Ba(3)-CI(1)k Ba(3)-0(1)a & 0(1)o & 0(1)p & 0 ( 1 ) r & 0 ( 1 ) s & 0 ( 1 ) <Ba(3)-anions> 2.840(3) 3.127(12) 3.342(9) 3.0936 T(1)-0(4) T(1) -0(2)&0(2)q& 0(2)s <T(1)-0> 1.540(38) 1.728(31) 1.680 T(3)-05 T(3)-0(2)t & 0(2)u & 0(2)m <T(3)-0> 1.591(35) 1.603(31) 1.603 T(2) -0(1)&0(1)q& 0(1)s T(2)-0(4) <T(2)-0> 1.653(25) 1.689(38) 1.658 T(4)-0(1) t&0(1)u& 0(1)m T(4)-0(5) <T(4)-0> 1.553(25) 1.709(37) 1.594 0(4)-T(1)-0(2)s 0(4)-T(1)-0(2)q 0(4)-T(1)-0(2) 0(2)s-T(1)-0(2)q 0(2)s-T(1)-0(2) 0(2)q-T(1)-0(2) 114.49(86) 114.49(86) 114.49(86) 104.02(99) 104.02(99) 104.02(99) 0(5)-T(3)-0(2)u 0(5)-T(3)-0(2)t 0(5)-T(3)-0(2)m 0(2)u-T(3)-0(2)t 0(2)u-T(3)-0(2)m 0(2)t-T(3)-0(2)m 114.80(1.09) 114.80(1.09) 114.80(1.06) 103.65(1.26) 103.65(1.24) 103.65(1.23) <0-T(1)-0> 109.25 <0-T(3)-0> 109.23 0(1)q-T(2)-0(1)s 109.85(68) 0(1)t-T(4)-0(1)u 109.57(92) 0(1)q-T(2)-0(1) 109.85(68) 0(1)t-T(4)-0(1)m 109.57(92) 0(1)q-T(2)-0(4) 109.09(70) 0(1)t-T(4)-0(5) 109.37(93) 0(1)s-T(2)-0(1) 109.85(68) 0(1)u-T(4)-0(1)m 109.57(91) 0(1)s-T(2)-0(4) 109.09(69) 0(1)u-T(4)-0(5) 109.37(93) 0(1)-T(2)-0(4) 109.09(69) 0(1)m-T(4)-0(5) 109.37(92) <0-T(2)-0> 109.47 <0-T(4)-0> 109.47 Note: <M-<|)> denotes the mean metal-ligand distance (A). Equivalent positions: a = x - 1, y - 1, z; b = x, y - 1, z; c = x - 1, y, z; d = x + 1, y, z; e = x, y + 1, z; f = x + 1, y + 1, z; g = -x, -y, z - 1 / 2 ; h = -x, -y + 1, z - !4; i = -x + 1, -y + 1, z - 1/2; j = -x, -y, z + Y2; k = -x, -y + 1, z + 72; I = -x + 1, - y + 1, z + 1/ 2; m = -x + 1, - y + 2, z + Y2\ n = -y, x - y, z; o = -y + 1, x - y, z; p = -y + 1, x - y , z; q = -y + 1, x - y + 1, z; r = - x + y, -x, z; s = - x + y, -x + 1, z; t = y, -x + y + 1, z + 1/ 2; u = x - y, x, z + y 2; v = -x + y + 1, -x + 1, z. 83 T A B L E 5.5 B O N D - V A L E N C E * A R R A N G E M E N T IN UK6 T(1) T(2) T(3) T(4) Bad ) Ba(2) BaO) Total 0(1) 0.925 1.212 0.058 2.253 0(1) 0.925 1.212 0.058 0(1) 0.925 1.212 0.058 0(1) 0.058 0(1) 0.058 0(1) 0.058 0(2) 0.753 1.067 0.072 1.963 0(2) 0.753 1.067 0.072 0(2) 0.753 1.067 0.072 0(2) 0.072 0(2) 0.072 0(2) 0.072 0(3) 1.255 0.839 2.093 0(4) 1.093 0.795 1.889 0(5) 0.131 0.393 0(5) 0.131 0(5) 0.131 Cl(1) 0.400 0.194 1.700 Gl(1) 0.400 Cl(1) 0.400 Cl/0(1)1 0.042 0.127 Cl/0(1)1 0.042 Cl/0(1)1 0.042 Cl/0(1)2 0.111 0.544 Cl/0(1)2 0.111 Cl/0(1)2 0.111 CI/0(2)1 0.151 0.132 0.414 CI/0(2)1 0.151 0.132 CI/0(2)1 0.151 0.132 CI/0(2)1 0.132 CI/0(2)1 0.132 CI/0(2)1 0.132 CI/0(2)2 0.323 0.281 0.885 CI/0(2)2 0.323 0.281 CI/0(2)2 0.323 0.281 CI/0(2)2 0.281 CI/0(2)2 0.281 CI/0(2)2 0.281 Total 3.513 3.613 4.294 4.429 3.227 1.083 3.130 Calculated from the curves of Brese & O'Keeffe (1991). 1 Oxygen atom; 2 Chlorine atom Figure 5.7b. Ba(l) polyhedron in UK6. 85 Figure 5.8b. Ba(2) polyhedron in UK6. 57.421 A 3 . Ba(3) is coordinated by 13 anions, one Cl(l) atom, six 0(1) atoms and six Cl/0(2) atoms forming a hexagonal prism and pyramid (Figure 5.9). The Ba(3)-anion distances range from 2.840 to 3.342 A (mean 3.094 A) and the polyhedral volume is 68.005 A 3 . The electron-microprobe data and refined site occupancy confirmed that these sites are completely filled with barium. Atoms at the tetrahedral sites are coordinated by four oxygen atoms. T(l)-0 distances range from 1.540 to 1.728 A (mean 1.680 A), and the 0-T(l)-0 angles range from 104.02 to 114.49° (mean 109.25°). T(2)-0 distances range from 1.653 to 1.689 A (mean 1.658 A), and the 0-T(2)-0 angles range from 109.09 to 109.85° (mean 109.47°). T(3)-0 distances range from 1.591 to 1.603 A (mean 1.603 A) , and the 0-T(3)-0 angles range from 103.65 to 114.80° (mean 109.23°). T(4)-0 distances range from 1.553 to 1.709 A (mean 1.594 A), and the 0-T(4)-0 angles range from 109.37 to 109.57° (mean 109.47°). Polyhedral volumes for T(l), T(2), T(3) and T(4) are: 2.410, 2.338,2.086, and 2.073 A 3 , respectively. T(l), T(2), T(3) and T(4) have tetrahedral angle variances of 33.933, 0.147,36.230, and 0.0434 and mean tetrahedral quadratic elongations of 1.008,1.000,1.008, and 1.003, respectively. From the results of electron-probe microanalyses, it was concluded that the T sites contain both silicon and aluminum. As bond-valence analysis is unreliable, it could not be determined at which T-sites the silicon or aluminum are located. UK6 is a layer silicate with barium atoms between the tetrahedral layers (Figure 5.10). Figures 5.11 and 5.12 depict the UK6 structure projected onto (010) and (001), respectively. The UK6 structure is based on double tetrahedral layers, [T 40 8]m o o. Tetrahedral sheets are formed by two polar tetrahedral sheets, [T 20 5]C O„, bonded by their free apices. Each sheet is characteristic of mica and consists of six-membered rings. The tetrahedral layers are on (001) and the individual tetrahedral sheets point either along [001] or [001 ]. 87 Figure 5.10. Perspective view of the crystal structure of UK6. Tetrahedra are Si/AI, blue spheres are Ba, green spheres are Cl, red spheres are Cl/O and white spheres are O. 89 Figure 5.11. Crystal structure of UK6 projected onto (010). Tetrahedra are Si/AL blue spheres are Ba, green spheres are Cl, red spheres are Cl/O. and white spheres are O. 90 Figure 5.12. Crystal structure of UK6 projected onto (001). Tetrahedra are Si/AI, blue spheres are Ba, green spheres are Cl, red spheres are Cl/O. and white spheres are O. 91 Three layers of barium polyhedra connect the tetrahedral layers (Figure 5.13). Edge-sharing Ba(2) atoms form sheets that are bonded by 0(2) atoms to the tetrahedra pointing along [001]. Edge-sharing Ba(3) atoms form sheets that are bonded by 0(1) atoms to the tetrahedra pointing along [OOT], and Ba(l) polyhedra are located between the Ba(2) and Ba(3) polyhedral sheets. Ba(2) and Ba(3) polyhedral sheets are bonded by Cl(l) atoms along [001] with Ba(l) atoms occupying the spaces between them (Figure 5.14). It is suspected from electron-probe microanalysis and bond valence analysis that the oxygen in the Cl/0(1) and Cl/0(2) positions are part of H 2 0 complexes. The H 2 0 complexes are located in spaces between Ba(2) polyhedra and Ba(3) polyhedra (Figure 5.15a. and 5.15b.). The layering of tetrahedra and barium polyhedra is responsible for the perfect {0001} cleavage of UK6. 5.7 Discussion UK6 is part of the Monteregianite-(Y)-Wickenburgite series (Struntz classification), which are phyllosilicate minerals with tetrahedral double layers and related structures (VI1 l/H.38-60). With in this group, UK6 is structurally and chemically very similar to cymrite, BaAl 2Si 2(0,0H) 8-H 20. Cymrite is monoclinic, pseudohexagonal with cell dimensions of a = 5.346(2), b = 37.05(2), c = 7.698(1) A, and /?= 90° (Bolotina et al, 1991). Cymrite was first described by Runnells (1964) as hexagonal colourless plates and the powder pattern was indexed according to the hexagonal system. He noted that cymrite has perfect {0001} cleavage and is uniaxial negative. Cymrite's crystal structure was first solved by Drits et al. (1975) and refined by Bolotina et al. (1991). 92 c Figure 5.13. Perspective view of the crystal structure of UK6, showing the layers of barium polyhedra (blue) bonding the tetrahedral layers (yellow). c Ba(2) polyhedral sheet Ba(3) polyhedral sheet Figure 5.14. Perspective view of Ba(2) and Ba(3) polyhedral sheets, bonded by Cl(l) atoms along [001 ] with Ba(l) atoms occupying spaces between them in UK6. 94 Spaces containing H atoms Figure 5.15a. Ba(2) polyhedral layer in UK6 projected onto (0001), showing the location of Cl/0(1) atoms involved in H 20 complexes and the spaces containing the hydrogen atoms. c Spaces containing H atoms Figure 5,15b. Perspective view of Ba(3) polyhedral layer in UK6, showing the location of CI/0(2) atoms involved in H 20 complexes and the spaces containing the hydrogen atoms, 95 Like UK6, cymrite consists of double tetrahedral layers, [T 40 8]m o o. Instead of three layers of barium polyhedra connecting the tetrahedral layers, there is one layer of barium polyhedra (Figure 5.16). In the cymrite structure, the Al and Si tetrahedra are fully ordered over the positions of the tetrahedral networks. In the double layer, one network is completely filled with AI and the other network is completely filled with Si. The Al network is formed by geometrically regular six-membered rings and in the silicate layer the tetrahedra are rotated about the normal to their bases. Al and Si ordering was determined from bond lengths and bond-valence analysis. There may be Al and Si ordering in UK6 but this can not be determined due to inaccurate oxygen positions. Water complexes in cymrite are located between the tetrahedral rings, while in UK6 the water is located in the barium polyhedral layers. 96 c Figure 5.16. Perspective view of the crystal structure of cymrite. Tetrahedra are Si and AL dark blue spheres are barium and larger, lighter blue spheres are water. 97 6.0 Walstromite: Refinement 6.1 Introduction Walstromite was discovered by Robert E. Walstrom during the course of a project begun in 1957 by the Public Service Laboratory of the Division of Mines and Geology of the State of California. The properties of walstromite were briefly described by Stinson & Alfors (1964); a more detailed description was given by Alfors et al. (1965). Walstromite occurs as single grains disseminated in sanbornite-quartz rock, particularly in quartz-rich areas, and in wollastonite-bearing quartzite that contains little or no sanbornite. Grain size is generally very coarse (8 cm3) and most grains contain rounded inclusions of celsian and quartz. Walstromite is white to colourless (Figure 6.1), has a white streak and fluoresces dull pink under short wavelength ultraviolet light and bright pink under long wavelength ultraviolet light. Walstromite is associated with sanbornite, quartz, wollastonite, celsian, taramellite, pyrrhotite, pyrite, witherite, fresnoite and UK10. Details of the experimental methods used are given in Chapter 2. 6.2 Previous Work The chemical formula, (Bao 9 6 , Sr0 0 2) 0^(Caj 0 4 , Mno.02)2.o6Si2.8908 g3, was proposed by Alfors et al. (1965) for walstromite from Fresno County, California based on chemical analysis done by DC arc emission spectrographic methods. Their results are calculated to 100% from mstrument totals. 98 99 Ca2BaSi 3 0 9 was synthesized by Eskola (1922) in a study of the silicates of strontium and barium, who thought it to be hexagonal. Glasser & Glasser (1961) also studied the synthetic analogue of walstromite and concluded it was triclinic. Alfors et al. (1965) studied walstromite and its synthetic analogue and reported them to have similar optical properties. They determined the space group to be either P or P1. Glasser & Glasser (1961) chose a different unit cell for walstromite than that chosen by Alfors et al. (1965), corresponding to the three shortest translational distances. The transformation matrix is given in Alfors et al. (1965). Alfors et al. (1965) thought that there may be small but real differences in the structures of synthetic and natural walstromite due to the small differences in unit cell. Glasser & Glasser (1968) determined the space group of walstromite to be P1 with Z = 2 from Weissenberg photographs taken with MoKa radiation; the intensities were estimated visually to determine the structure. The cell-setting used by Alfors et al. (1965) was used in their study. Table 6.1 lists the unit cell dimensions measured by Glasser & Glasser (1961), Alfors et al. (1965), Glasser & Glasser (1968) and for this study. No corrections for absorption or extinction were made by Glasser & Glasser (1968) and the final R-factor was 16%. The structure was solved in projection only and because of barium (a heavy absorber) the oxygen atom positions were not well determined. The atomic coordinates determined by Glasser & Glasser (1968) are given in Table 6.2. 6.3 Electron-probe Microanalysis Electron-probe microanalyses were done with the following operating conditions: excitation voltage 15 kV, beam current 20 nA, peak count time 20 s, background count time 10 s, and beam diameter 5 pm. For the analyzed elements, the following standards and X-ray lines were used: barite 100 TABLE 6.1 CELL DIMENSIONS OF WALSTROMITE 1 2 3 4 a (A) 6.71(8) 6.743(5) 6.73(3) 6.733(1) 6(A) 6.73(3) 9.607(5) 9.61(6) 9.608(2) c(A) 9.61(7) 6.687(5) 6.72(3) 6.685(1) «(°) 88.37 69.85 69.65 69.64(3) PO 111.05 102.23 102.33 102.29(3) Y(°) 102.33 96.11 96.9 96.89(3) Sources: 1 Glasser & Glasser (1961), 2Alfors etal. (1965), 3 Glasser & Glasser (1968), 4 this study. T A B L E 6.2 ATOMIC COORDINATES FOR WALSTROMITE ( G L A S S E R & G L A S S E R 1968) Site X y z Ba 0.049 0.848 0.323 Ca(1) 0.272 0.507 0.763 Ca(2) 0.435 0.831 0.935 Si(1) 0.096 0.222 0.145 Si(2) 0.235 0.484 0.284 Si(3) 0.442 0.196 0.511 0(1) 0.236 0.251 -0.027 0(2) -0.098 0.144 0.102 0(3) 0.042 0.366 0.212 0(4) 0.366 0.556 0.089 0(5) 0.125 0.580 0.389 0(6) 0.352 0.365 0.494 0(7) 0.613 0.238 0.368 0(8) 0.517 0.084 0.765 0(9) 0.238 0.130 0.389 102 (Bala), SrTi0 3 (SrZa), diopside (SiKa, MgKa, CaKa), synthetic rhodonite (MnKa), scapolite (ClKa), grossular (AlKa), albite (NaKa), phlogopite (FKa), and fayalite (FeKa). Strontium and chlorine were not detected. There is essentially no compositional variation in walstromite, between grains or within individual grains. Based on 9 anions, the empirical formula (Ca, 9 4 7 , Nao011), 9 5 8 Ba, 021(Si2995, A l 0 0 2 3 ) 3 0 1 8 O 9 was determined from an average of 5 analyses of walstromite. Table 6.3 gives the results for the electron-probe microanalyses as well as the analysis done by Alfors etal. (1965). The results of the microprobe are slightly different than those obtained previously by DC arc emission spectrography (Alfors et al. 1965). In this study, the totals of the analyses were low and Sr was not detected, while Alfors et al. (1965) normalized to 100% from instrumental totals and found 0.53 wt. % SrO. Analyses from this study also show higher amounts of Al and lower amounts of Ca than Alfors et al. (1965). Due to the low analytical totals obtained in this study it is probable that there are very small amounts of H 2 0 or (OH)" are present in the structure. 6.4 X-ray Crystallography and Crystal-Structure Refinement 6.41 Experimental The crystal used in this study is from Esquire #7, Big Creek, Fresno County, California. The unit cell was derived using fifty reflections in the range 9.49 to 49.59° 20. Cell dimensions were refined using the method of least squares (Table 6.4). A full sphere of reflections (4616 measurements, exclusive of standards) were collected from 3 to 60° 20. None of the reflections were 103 Ill O cm \-co _ J I u. O co LU CO > < O DC O LU CQ o or a. i Z o cc O LU I LU co CD LU _l CQ < CD LU CD 1 0.07 39.6 26.1 33.3 0.07 0.26 39.38 23.95 34.27 co o CD CN CD CO 00 o o O cri CO CO CN CO o CN cn O CD CT) o o oi CO CO CN CO CD o CN LO CO CO O 00 XT d o CT) CO CN CO o CO CN LO OJ CD r^ -CT> o o CT) CO CO CN CO CO LO o 1^  CN O CD CO o O CT) CO CN CO O CN CO z CO o t N < eg o CO CaO BaO CO LO o o CO CO CT) o r-~: CT) LO cn CO CO CT) o LO CT) co 00 CT) SUM UJ LO g o CL O i 0. o 'E o < y co CM CO LO 00 CM CT) CN CD o o o> CT) O CT) o d CN T LO CM CO CNJ LO CO T— CN o CT) o o o CT) o cn d d CO T ~ LO LO •"3- CM o 00 T— CM 00 LO CO o o o CT) CT) o o d d CM T — CD CT) o CD h- •sr o CN CT) LO CN o o o CT) CT) O o d d CN T _ T _ CD o CO 1^  CM CO CD T— CM o CO T— oo o o o CT) o CT> d d CO T— T ~ LO LO CM o o CM CT) LO CN o o o CT) CT> O o d d CM T ~ T — CD CO CO CO z < CO O CQ IO CO cu Q_ in c o 'c co cn rz o cu N "co E i_ o £= CU I— TO in CU in « 8 CO £= 0 o ^ °- co J= o 5 r (0 3 w J5 CU Z> S E x £ o a 104 T A B L E 6.4 M I S C E L L A N E O U S INFORMATION FOR WALSTROMITE a (A) 6.733(1) Rad/mono MoKoc/graphite b 9.608(2) c 6.685(1) Total reflections 4616 a ( ° ) 69.64(3) Unique reflections 2326 P 102.29(3) Rint% 4.5 y 96.89(3) F 0 > 4 a ( F 0 ) 2172 V(A 3 ) 395.65(3) R (observed) % 3.8 Space group PT Rw (observed) % 12.1 Z 2 Crystal size (mm) LI (MoKct; mm"1) 5.04 mm"1 R-H{\Fo\-\Fc\)'H\F0\ rejected due to systematic absence violations but 457 reflections were rejected due to inconsistent equivalents, leaving 2326 reflections of which 130 are suppressed. Fourteen strong reflections uniformly distributed with regard to 20 were measured at 5 ° intervals of ij; (the azimuthal angle corresponding to rotation of the crystal about its diffraction vector) from 0 to 355 °, after the method of North et al. (1968). A total of 1008 reflections were collected, of which zero reflections were rejected, were used to calculate an absorption correction. The merging R index for the i[;-scan data set decreased from 3.7% before the absorption correction to 2.2% after the absorption correction. This correction was applied to the entire set of data; minimum and maximum transmissions were 0.036 and 0.068 respectively. The data were also corrected for Lorentz, polarization and background effects, averaged and reduced to structure factors. Of the 2326 unique reflections, 2172 were classed as observed [F0 > 4o (F0)]. 6.42 Structure solution and refinement Miscellaneous collection and refinement data are given in Table 6.4. The cell setting used in Glasser & Glasser (1968) is followed in this study for reasons of comparison. The mean value of | E 2 - 11 is 0.814 , which does not imply either a cenfrosymmetric or a non-cenfrosymmetric space group. As barium is present, Patterson techniques were used to solve the structure. For the isotropic displacement model the structure was refined in space group P T to an R index of 5.4%. Conversion to anisotropic displacement factors of all the atoms in the structure resulted in convergence at R and R„, indices of 3.8% and 12.1% respectively (R = 4.2% for all 2326 data). Addition of an isotropic extinction correction did not improve the results. Positional 106 coordinates and anisotropic and equivalent isotropic displacement factors are given in Table 6.5. Interatomic distances and angles are given in Table 6.6, and the results of bond-valence analysis are given in Table 6.7. The results of the bond valence analysis suggest that there is no water in the structure, but there may be small amounts of hydrogen bonded to 0(3), 0(4), 0(6), and 0(8). 6.5 Description of the Structure Based on Earlier Work Glasser & Glasser (1968) described the structure of walstromite as having (Si309)~6 rings arranged in layers parallel to (101) with Ca atoms approximately halfway between the rings (Figure 6.2). The Ba atoms are approximately coplanar with the oxygen atoms. The bond lengths given by Glasser & Glasser (1968) were not very meaningful because the oxygen positions were not well determined, but they noted O-Si-0 angles vary. The Si tetrahedra are distorted so that three of that O-Si-0 angles are less than 109.4°. Ca(l) is coordinated by six unshared and two shared oxygen atoms, forming an irregular square antiprism. The shared atoms are bonded to two silicon atoms. Ca(2) is bonded to six unshared oxygens, forming a slightly irregular octahedron. Barium is bonded to eight oxygen, six unshared and two shared. 6.6 Description of the Structure Based on This Work There are six distinct cation sites in the structure of walstromite, all at general positions. The site containing barium is coordinated by eight oxygen atoms (Figure 6.3a. and 6.3b.). Ba-0 distances range from 2.859 to 3.102 A (mean 2.837 A). The polyhedral volume is 37.42 A 3 . The electron 107 LU O or h-CO O LL CO LU I -LU < 0_ o o < _l < LO CO LU _ J CO < to * 5 0) CO LO CO, CO- CO, CO CD IS-, 52- IS-, IS-, I S--, c\T CD CO iff LO OS' CO oo' CO 00 CO o CD ccf CD ^F CD erf in CNT IS- CO IS-o IS- CD iff Is-CO CO in IS) IS) CNT CO~ CNT CO co CO co co CO CD IS-I •xt IS-i IS- CNI in i IS) IS) 00 If) O hj- OO |Xp IS- in •xt • IS) •xt CO CD 1 cp CO in CO co CNT CNT CN? CNT CN? CO co" co~ 5 X — CD •xt LO in CD IS) m I S -m a> CO CD CO CNI •xt CN IS) CD •xt in CD oo CD 00 CO CO co' iff CNI T— CNI Is-00 •<t 00 CD CD •xt Is-"xf o CO GO CO o 3, 3, iff CNI CD CNI T— iff CD CNI Ccf CO CNI CNI CNI cf C O CO C M C M , ro' o CO CM CO C O -CO t - CNI o o CO o -xt in O O CD O Is-o CNI feSS^SS^^^^-oco'roiffo' ^ ^ ^ ^ " ^ c o i o i s - c 3 ) i s - L n r S - T - j i s -L n i s ~ - c » ' s T L n c o ^ I V _ o ^ . O C O 0 ) O O O T ~ ' ^ T - T - ' r " T - T - T - C M T - C M t - T - C M ^ -£ 2 S 3 S raoinoof^ioocM ^ ^ ^ ^ ^ ^ T - T - T - T - T - C M T - C M T -i n - ^ ^ - v ^ ^ - v ^ - v - - v ^ ^ ^ - v ^ - v - - ^ ^ ^ ^ T - c N ^ c o c o c o c o i n i n c n a s c o i s - o ) 0 c M c o - x t o o r - L n c n o o o o c M ' x r o c D c o C O h - O J C M m t - C O l O t - ^ - O C N l T - C O C p d d d d d d d d d d d d d d d co CN}, CM 3^  3^  5, 3, 3^  T— CM T— CD CO cf 00 CO of iff CM CO CD oo T— CO T— •<t CO T— T— o 00 CO CO CO O CM 00 CJ) CM CM in CM Is- •xi" o in CO CO in oo •xt CNI CO 00 Is- oo in m Is-d d d d d d d d d d d d d d , , , , , _ , _ o ^ ^ , , CD , _ CM, CM CM CM in in in in in. CO, Ln, CO LO, T— erf 00 co' CD of co oo iff LO Is- CO CD o CD o m CM CM CO •xt •xt Is- co CO •xt OO CO in o •xt CO CD Is- CO CM CM "xt o CM CO o CM •xt co CD d d d d d d d d d d d d d d CO CD o o T J CU S> CO co CU _2 CO > T5 C CO 108 T A B L E 6.6 S E L E C T E D INTERATOMIC DISTANCES (A) A N D A N G L E S (°) F O R WALSTROMITE a) Silicon Tetrahedra Present study (A) "(A) Present study (°) T) Si(1)-0(2)a 1.584(4) 1.61 0(1)e-Si(1)-0(2)a 111.49(21) 110 Si(1)-0(1)e 1.596(4) 1.58 0(3)l-Si(1)-0(2)a 113.04(19) 115 Si(1)-0(3)l 1.663(4) 1.69 0(3)l-Si(1)-0(1)e 113.03(20) 120 Si(1)-0(9)a 1.674(4) 1.71 0(9)a-Si(1)-0(2)a 106.94(20) 107 <Si(1)-0> 1.629 1.65 0(9)a-Si(1)-0(1)e 0(9)a-Si(1)-0(3)l 109.51(19) 102.20(20) 106 97 <0-Si(1)-0> 109.37 109 Si(2)-0(5) 1.577(4) 1.66 0(4)l-Si(2)-0(5) 117.88(21) 125 Si(2)-0(4)l 1.596(4) 1.65 0(6) -Si(2)-0(5) 106.99(21) 102 Si(2)-(6) 1.674(4) 1.59 0(6)-Si(2)-0(4)l 111.32(20) 114 Si(2)-0(3)l 1.688(4) 1.72 0(3)l-Si(2)-0(5) 107.72(20) 104 <Si(2)-0> 1.634 1.66 0(3)I-Si(2)-0(4)I 0(3)l-Si(2)-0(6) 110.75(20) 100.73(19) 109 98 <0-Si(2)-0> 109.23 109 Si(3)-0(8)h 1.585(4) 1.68 0(7)g-Si(3)-0(8)h 114.64(21) 114 Si(3)-0(7)g 1.598(4) 1.59 0(6)-Si(3)-0(8)h 111.01(20) 114 Si(3)-(6) 1.672(4) 1.76 0(6)-Si(3)-0(7)g 108.43(20) 107 Si(3)-0(9)a 1.676(4) 1.62 0(9)a-Si(3)-0(8)h 110.21(20) 112 <Si(3)-0> 1.633 1.66 0(9)a-Si(3)-0(7)g 0(9)a-Si(3)-0(6) 110.45(20) 101.26(19) 110 99 <0-Si(2)-0> 109.33 109 b) Cation Polyhedra Present study (A) *(A) Present study (A) *(A) Ba-0(5) 2.554(4) 2.56 Ca(1)-0(1)g 2.334(4) 2.39 Ba-0(2) 2.709(4) 2.67 Ca(1)-0(7)q 2.400(4) 2.39 Ba-0(7)q 2.716(4) 2.72 Ca(1)-0(4)m 2.420(4) 2.33 Ba-O( l ) 2.796(4) 2.79 Ca(1)-0(4) 2.441(4) 2.49 Ba-0(2)f 2.859(4) 2.67 Ca(1)-0(5) 2.472(4) 2.38 Ba-0(9) 2.933(4) 2.99 Ca(1)-0(3)k 2.657(4) 2.63 Ba-0(9)i 3.031(4) 3.07 Ca(1)-0(6) 2.799(4) 2.78 Ba-0(8) 3.102(4) 3.06 Ca(1)-0(5)g 2.857(4) 2.79 <Ba-0> 2.837 2.82 <Ca(1)-0> 2.547 2.52 Ca(2)-0(2)i 2.304(4) 2.34 Ca(2)-0(7)q 2.318(4) 2.30 Ca(2)-0(8)c 2.334(4) 2.36 Ca(2)-0(1)d 2.358(4) 2.33 Ca(2)-0(8)j 2.437(4) 2.37 Ca(2)-0(4) 2.477(4) 2.50 <Ca(2)-0 2.371 2.37 Note: <M-<|>> denotes the mean metal-ligand distance (A). Equivalent positions: a = x, y -1, z ; b = x - 1 , y, z ; c = x, y, z + 1; d = x + 1, y, z + 1; e = -x, -y + 1, - z ; f = -x, -y + 2, -z; g = -x, -y + 1, -z + 1; h = -x + 1, -y + 1, -z + 1; i = - x, -y + 2, -z + 1; j = -x + 1, -y + 2, -z + 1; k = -x, -y + 1, -z + 2; I = x, y, z - 1 ; m = -x + 1, -y + 1, -z + 2; n = x - 1 , y, z - 1; o = -x + 1, -y + 2, -z + 2; p = x, y + 1, z; q = x + 1, y, z. * Glasser and Glasser (1968) no T A B L E 6.7 B O N D - V A L E N C E * A R R A N G E M E N T IN WALSTROMITE Si(1) Si(2) Si(3) Ba Ca(1) Ca(2) Total 0(1) 1.078 0.255 0.371 0.348 2.052 0(2) 1.114 0.322 0.402 2.053 0(2) 0.215 0(3) 0.900 0.841 0.155 1.896 0(4) 1.079 0.294 0.252 1.902 0(4) 0.278 0(5) 1.135 0.490 0.255 1.971 0(5) 0.090 0(6) 0.874 0.878 0.106 1.857 0(7) 1.073 0.316 0.310 0.387 2.087 0(8) 1.111 0.371 1.763 0(8) 0.281 0(9) 0.874 0.869 0.176 2.053 0(9) 0.135 Total 3.967 3.929 3.931 1.909 1.859 2.041 Calcu la ted from the curves of Brese & O'Keeffe (1991). Figure 6.2. Crystal structure of walstromite (Glasser & Glasser 1968) 112 Figure 6.3b. Barium polyhedron in walstromite. microprobe data, refined site occupancy, and bond-valence analysis confirm that the site is completely filled with barium. Calcium occupies two sites, Ca(l) and Ca(2). The Ca at Ca(l) is coordinated by eight oxygen atoms forming a distorted square antiprism (Figure 6.4). The square faces of the antiprism are bisected, forming high angle domes. Ca(l)-0 distances range from 2.334 to 2.857 A (mean 2.547 A). The polyhedral volume is 28.16 A . The Ca at Ca(2) is coordinated by six oxygen atoms fonriing a distorted octahedron. The Ca(2)-0 distances range from 2.304 to 2.477 A (mean 2.371 A). The o •> polyhedral volume is 17.362 A , with an octahedral angle variance of 54.762 and a mean octahedral quadratic elongation of 1.017. The electron-microprobe data, refined site occupancy, and bond-valence analysis confirm that the sites are completely filled with calcium. There are three tetrahedrally coordinated sites containing silicon, Si(l), Si(2) and Si(3). The Si(l)-0 bond distances range from 1.584 to 1.674 A (mean 1.629 A). The Si(2)-0 bond distances range from 1.577 to 1.688 A (1.634 A) and the Si(3)-0 bond distances range from 1.585 to 1.676 A (mean 1.633 A). The polyhedral volumes are 2.205,2.220, and 2.215 A3 for Si(l), Si(2) and Si(3) respectively. Their respective tetrahedral angle variances are 17.748,19.978, and 32.252; and their mean tetrahedral quadratic elongations are 1.005,1.005, and 1.008. According to the electron-microprobe data and bond valence analysis, there is a very minor amount of substitution of aluminum for silicon in the Si(l), Si(2) and Si(3) sites. The silica tetrahedra form trigonal rings arranged in layers on (101). The free apices of the tetrahedra all face in the same direction in each individual ring, with the direction of each ring alternating. The rings are not directly on top of each other, but are offset by approximately 3 A along [010], forming layers which accounts for the (010) cleavage (Figure 6.5). The arrangement of the 114 Figure 6.4. Coordination of Ca( l ) in walstromite, forming a distorted square antiprism. Figure 6.5. The walstromite structure p ro jec ted o n (101), showing the layers of a l ternat ing rings of si l ica te t rahedra a n d the positions of the bar ium a toms (dark spheres) a n d c a l c i u m a t o m s (light spheres). 116 tetrahedral rings also accounts for the cleavage on (011). Barium and calcium atoms lie approximately halfway between the silicon tetrahedral rings, perpendicular to the a axis, producing the cleavage on (100) (Figure 6.6). Ca(l) atoms lie in the (010) plane halfway between the plane with Ca(2) and Ba atoms. Barium polyhedra run in chains parallel to the c axis, with two edge sharing Ca(2) octahedra occupying spaces between the chains. (Figure 6.7). Alternately, Ca(l) edge sharing polyhedra form infinite chains along [101], with two Ca(2) edge sharing octahedra between the chains along [010]. This forms a sheet of Ca polyhedra on (101 ), with barium atoms in the open spaces of the Ca sheet (Figure 6.8). Along [101 ] the Ca sheets do not lie directly on top of one another, but are offset by the angle between the b axis and the ac plane. The three-membered rings tetrahedral rings are bonded to the Ca(l) chain. 6.7 Discussion It was predicted by Glasser & Glasser (1964) and Glasser & Glasser (1968) that the structure of walstromite is isostructural to that of margarosanite, PbCa2Si309. The structure of margarosonite was solved by Freed & Peacor (1969). It consists of planes of tetrahedral sites, which form three-membered rings, alternating with planes of calcium sites between sheets of close-packed oxygen atoms parallel to (101). Ca(l) polyhedra form an infinite edge-sharing chain parallel to [101] and Pb and Ca(2) sites alternate along the edge of this chain. Walstromite and margarosanite are in fact isostructural, with identical structural features (Figure 6.9). They both belong to the group of 117 I compounds based on X 3 0 9 rings. It was suggested originally by Glasser & Glasser (1961) that walstromite was part of this group and this was verified by Glasser & Glasser (1966). 118 Cam \ 3* \ ] oo 1, oo \ Q j/^ 0 <— Ca(l) 0 0 QQ «— Ca(2) 1 \ 0 4 QJ 1 JO 1 j O 1 j O 0 O \ 0 0 (100) c leavage Figure 6.6. Walstromite structure p ro jec ted o n (101) showing the (100) c l e a v a g e . The po l yhed ra a re si l ica te t rahedra, dark spheres a re ba r ium a t o m s a n d the light spheres a re c a l c i u m a toms . No te al ternat ing C a ( l ) a n d Ca(2) layers. 119 Figure 6.7. Project ion o n (010) showing cha ins of ba r ium po l yhed ra (dark) a n d Ca(2) o c t a h e d r a (light). 120 Figure 6.8. Project ion of the walstromite structure o n (101) showing a sheet of C a po l yhed ra with ba r ium in the o p e n s p a c e s . The C a ( l ) po l yhedra a re dark, the Ca(2) po l yhedra a re light a n d the dark spheres a re ba r ium a toms . 121 Figure 6.9. Crystal structure of walstromite (top) and margarosanite (bottom). Si tetrahedra are yellow, Ba atoms are dark blue, C a atoms are light blue, Pb atoms are pink. 122 7.0 Verplanckite: Refinement 7.1 Introduction Verplanckite was discovered in the Fresno County sanbornite deposits in 1962 by Stewart Agrell of Cambridge University. The properties of verplanckite were briefly described by Alfors & Stinson (1965); a more detailed description was given by Alfors et al. (1965). Verplanckite is extremely rare, occuring in sanbornite-quartz rock as radial masses of brownish-orange prismatic crystals up to 3 mm long and 1 mm wide and as disseminated grains, locally concentrated in thin layers. It forms translucent elongated hexagonal crystals with a number of prism and pyramid forms, most notably the {1120} prism (Alfors & Stinson 1965). Verplanckite is closely associated with: quartz, sanbornite, celsian, diopside, taramellite, pyrrhotite, pyrite, fresnoite, muirite and traskite. Details of the experimental methods used are given in Chapter 2. 7.2 Previous Work 6 Alfors et al. (1965) determined that verplanckite is hexagonal with a possible space group of P62m, P6m2, P6mm, P6/mmm or P622. Cell parameters were determined to be a = 16.35(2), c 7-17(2) A. The chemical formula, (Ba^, Cag 0 2) 2 0 4(Mno 7 2, Fe 0 1 5 , T i 0 1 3 , M g ^ ) , 0 2 (Si 2 0 1 , A W W C V i ^ O o ^ ' OH, 1 7 , C l 0 5 8 ) 2 0 0 • 2.87H20, was proposed by Alfors et al. (1965) for verplanckite from Fresno County, California based on chemical analyses done by DC arc emission spectographic methods. Their results are calculated to 100% from instrument totals. 123 Kampf et al. (1973) did electron-probe microanalysis on a crystal of verplanckite from Fresno County, California which gave the chemical formula B a u 2 7 M n 4 0 6 T i ! 5 0 Fe 0 2 9 Si 1 2 O 3 6 O 2 2 0C1 8 8 4 (H 2 O) 7 0 5 . which is similar to the formula obtained by Alfors et al. (1965), except for lower water and higher chlorine contents. Kampf et al. (1973) solved the structure of verplanckite with a four-circle diffractometer using crystal-monochromatitized AgAxc radiation. No absorption correction was applied to the observed data. The space group was determined to be P6lmmm with cell parameters a = 16.398(10) and c 7.200(4) A. Using an isotropic displacement model, a discrepancy index of R = 10.2% was achieved for all observed reflections. The atomic coordinates determined by Kampf et al. (1973) are given in Table 7.1, and selected atomic distances are given in Table 7.2. 7.3 Electron-probe Microanalysis Electron-probe microanalyses were done with the following operating conditions: excitation voltage 15 kV, beam current 20 nA, peak count time 20 s, background count time 10 s, and beam diameter 5 pm. For the analyzed elements, the following standards and X-ray lines were used: barite (BaLct), SrTi0 3 (SrLa), diopside (SiKa, MgKa, CaKa), synthetic rhodonite (MnKa), scapolite (ClKa), grossular (AlKa), albite (NaKa), phlogopite (FKa), and fayalite (FeKa). Strontium was not detected. There is large compositional variation between grains of verplanckite. Si0 2 wt. % ranges from 18.72 to 19.76%, BaO wt% ranges from 53.13 to 54.44% and Cl wt % ranges from 21.55% to 24.85%. Based on twelve Si and atoms per formula unit, the empirical formula, (Ba 1 2 7 6 ,Ca 0 0 1 j Na 0 2 1 ) 1 2 9 8 (Mn 4 5 4 , Ti, 1 9 , Mgo 0 3, Fe 0 2 ) 6 , 7 (Si u 6 6 , A l 0 3 4 ) , 2 O 3 ] 7 8(C1 2 3 6 4 , F 0 4 9 ) 2 4 ) 3 , was 124 TABLE 7.1 ATOMIC COORDINATES FOR VERPLANCKITE (Kampf etal. 1973) Site X y z Ba(1) 0.3484(3) 0 0 Ba(2) 0.2164(2) 0.4328 Vi X 0.2606(4) 0.5212 0 Si(1) 0.4440(4) 0.8880 0.274(2) Cl(1) V, % Vi 0.3 x Cl(1) 0.142(2) 0.284 0.118(6) 0.6 x Cl(2) 0.301(2) 0 Vi Od) 0.160(1) 0.499(1) 0.197(3) 0(2) Vi 0 0.206(5) 0(3) 0.441(2) 0.882 Vi 0(4) = (0, OH) v3 2/3 0 Note: X= Mn, Ti and Fe. T A B L E 7.2 S E L E C T E D INTERATOMIC DISTANCES (A) FOR V E R P L A N C K I T E (Kampf et al. 1973) Si-0(1) x2 1.59(2) Si-0(2) 1.66(2) Si-0(3) 1.63(2) Ave. 1.62 X - O ( l ) x4 2.07(2) X-0(h4) 2.07(2) Ave. 2.07 Ba(1)-0(1) x4 2.92(2) Ba(1)-0(2) x2 2.92(2) Ba(1)-CI(2) x4 3.11(3) Ba(1)-CI(3) x2 3.68(1) Ave. 3.11 Ba(2)-0(1) x4 2.79(2) Ba(2)-0(3) x2 3.20(4) Ba(2)-CI(1) 3.32(1) Ba(2)-CI(2) x2 3.46(5) Ba(2)-CI(3) x2 3.10(3) Ave. 3.09 Note: X= Mn, Ti and Fe 126 determined from an average of 15 analyses. Table 7.3 gives the results for the electron-probe microanalyses. The results of the electron-probe microanalysis are substantially different from the results obtained by Alfors et al. (1965) and Kampf et al. (1973). In this study there was more Cl and minor H 2 0. The results from Alfors et al. (1965) and Kampf et al. (1973) were also differing in the amount of Cl and H 2 0 indicating a large variation in chemistry in verplanckite. The high totals obtained in this study (average 103.47 Wt%) indicates that there was probably H 2 0 loss during analysis. 7.4 X-ray Crystallography and Crystal-Structure Refinement 7.41 Experimental The crystal used in this study is from Esquire #7, Big Creek, Fresno County, California. The unit cell was derived using fifty reflections in the range 9.49 to 49.59° 20. Cell dimensions were refined using the method of least squares (Table 7.4). A full sphere of reflections (3683 measurements, exclusive of standards) was collected from 3 to 60° 20, of which eleven reflections were rejected due to bad backgrounds. None of the reflections were rejected due to systematic absence violations, but two were rejected due to inconsistent equivalents leaving 996 unique reflections, of which 157 were suppressed. Fourteen strong reflections uniformly distributed with regard to 20 were measured at 5 ° intervals of t|/ (the azimuthal angle corresponding to rotation of the crystal about its diffraction vector) from 0 to 355°, after the method of North et al. (1968). A total of 1146 reflections, of which none were rejected, were used to calculate an absorption correction. The 127 •a CO CO LO r - o m OI O CO O c-o CM o CD 00 CO o t— CM o T— t o d d oS d CM 00 d d CO d CO d IO CO LO CM o CO CO CM o CD CM CO o CM LO CO ^— o 03 o 00 CO o o CO CD . — T— l"~ o d d CO d CM co d d d CM o LO co IO CM o f- o CO CO o CD CO CO o CD o LO CT) 1^  T— o LO o CD o •* o CD CO o T— T— CT) o d d oi d CM cri d d CO d CO d IO CO LO CM O LO CM IO o o LO o O OI 00 CM LO ^— o CO LO o LO CO CO o 00 T— o r-o d d cri d CM CO d d CO CD CM d LO co LO CM o CNI o 00 CD o LO CO o o CT) IO CM CD CT) * — o o CD o o CO CM IO 00 CO o d d cri d CM cri d d CO d d CM LO CM T o CO o 00 o CM LO LO o o CM CO O o * — o O 5 CO o CO CT) o d d oS d CM 00 d d d d CO LO CM • T o "*"-o o •a- io CM CO CT) CM o CO o CD CT) CM o M" 00 o Nf r- CO o r~ CM CO CM CT) d d d CO d CM 00 d d CO d CO d LO* CM LO CM 1 o oo o CO oo o OO cn CO o CO LO CD •<t o T— o •<»• IO o OO OI CO o CO CM r- T— o d d cri d CM 00 d d CO d CM O LO CO IO CM 1 1 o CM o LO CO CM 05 o CO CO IO LO CO CM o •* o o CD O CM o cn 00 o CD CO o d d cri d CM cri d d CO d d LO T— LO CM • 1 o o IO o CM CM CO O) o 00 CO CO o> M" 1— o IO o CD CO CO o •f r- 1— o OI r-o d d cri d c\i 00 d d CO d CM d CM IO CM T O 00 in CM ai CM IO o CD o 00 1-CM o T ~ o LO CO •<i- o CM CM o CM o ci d d CT) d CM CO d d d CO d LO T— LO CM • 1 o o> CO 03 LO o r- 00 CD o t- LO CO CO CO o m co LO o LO CO O) 1— T— CM o d d cri d CM CO d d CO d CM d LO CO T— LO CM i 1 O * — r- CD CO LO CM CO o co CNJ o CT) o CO o CO CM !•» CO d d d CO d CM 00 d d CO d CO o LO CO LO CM o LO CO cn h- o CD o CO CM O O) r~ o CO CM o co r-. CO o T— CM T— o CM CD d d d cri d CM 00 d d CO d CO d LO CM LO CM O CM CM CM CO CM CO o o o CO CM CO O CM o o r- o CO o CO CM CO CO OI d d d co d CM cri d d CO d CO d LO CO T— LO CM 1 1 o LO CM CO o LO CO CT) o CO CM CO CO CM o •t 00 o LO CD CO o O) CO CO CO LO d d d CO d CM CO d d CO d CO d LO CO LO CM • 1 o CO _l O CM o o CM O CM O O O O LL CL TA < _i o < O Z LU < _l II II o Z ^ < (/) O I- S LL co m U_ O o O 1-co c g CL g E o < CM CO o CO OD CD o cn •9 LO o CM o o CD cn CO 00 CO cn CM cn d d d d d d CM d CO CM CO d CO LO LO o CM o o CO o o o cn o NJ-o o CM CO o CO CT) CO LOCO LO CO CO d d d d d d CM d CM CM CO d CO LO LO o CM o o CM CO CO CD o o o CM CT) LO CM CM o o f-LO 00 CD O CO co cn CO d d d d d d CM d CO CM CO d CO LO LO CM o cn LO o o CO co o o •>!• CO LO CO CO IO LO o CM CT) CO d d d d d d CM d csi CM CO d CO LO o o CO CO CO o o r- CO LO CM o o CO CO •9-LO CO LO CT) CM o CT) CO d d d - d d d CM d CM CM CO d CO •* LO CM o o CO CO cp o 00 LO CO CM o o Nf 00 i-~ LO LO CO LO 00 cn LO CM d d d ^ d d d CM d CM CM CM CO d CO LO LO Nf CM o o CM CO CO CO o CD CD o o O LO CO "3- CD CO CM LO d d d - d d d CO d CM CO CO CD LO CT) o o co CO CO o o O CM LO CO o o CD OO OI 00 IN- CD LO CM CM d d d d d d CM d CM CM CO d CO LO LO LO CM o CO o CO o CM CM o r- LO o o O LO CM CM OI CM CM t>-co o d d d d d d CO d IO CM CO CO LO CM CO o CM CO CO CO o CO IO o CM o o CD LO CO LO O CM CO CD OI OI d d d - d ••t d d CM d CM CM CM CO d CO •<r IO 00 CM o CM CO 00 CD o CT) COLO CO CM o o CT) cn CO CM O CT) CT) CO o d d d d d d CM d CM CO CO CD LO CM CM CM CO CO cp o o 00 CO cn CM o o CT) o CD O CO CO CO p CO d d d - d d d CM d co CM CO CO CD LO o CM CD o CO CO cq o LO 1^ LO o o CO CO CO LO O CO LO CM CO o CO LO d d d d d d CM d CM CO CO CD LO CO CO o CO CM CM o cn OI •* 00 o o CD CO CM "3- 00 CO LO CO 00 d d d d d d CM d CO CM CO d CO IO LO CM CM O CM CO 00 CO CM o 00 CM |N- co o o LO LO LO o CM CO l> CD CO CM LO d d d - d d d CO d LO CM CO CO I-* LO o CO CM O CM CO CT) CD COo OI LO o CM o o I--CD 00 CO LO CO cn o d d d d d d CO d CM CO CO LO i S c O + CM + ^ C M C M C M , , Z 5 < O ) O I - 2 L L O ) ( D L . O O CO CO c O < c TO 3 E £ i CD T3 C CO CO CM o CO CO CO n co c o 73 0) N lo E i o c CO L_ ro CO CD CO CO c CO aj o T A B L E 7.4 M I S C E L L A N E O U S INFORMATION F O R V E R P L A N C K I T E a (A) 16.300(2) Rad/mono MoKa/graphite c 7.169(1) V{AZ) 1649.65(3) Total reflections 3683 Space group PQImmm Unique reflections 996 Z 3 Rjnt % 3.6 Crystal size (mm) 0.3 x 0.3 x 0.7 mm F 0 > 4 a (F 0) 821 p ( M o K a ; mm"1) 5.04 mm"1 R (observed) % 4.0 Rw (observed) % 12.0 R = Z ( | F „ H r 7 c l ) / Z l ^ o l Rw=[Zw(\FA-\Fc\)2/^F02r,w^ merging R index for the i|/-scan data set decreased from 2.44% before the absorption correction to 1.3% after the absorption correction. This correction was applied to the entire set of data; minimum and maximum transmissions were 0.872 and 0.770 respectively. The data were also corrected for Lorentz, polarization and background effects, averaged and reduced to structure factors. Of the 996 unique reflections, 821 were classed as observed [F0 ^ 4o (F0)]. 7.42 Structure solution and refinement Miscellaneous collection and refinement data are given in Table 7.4. The cell setting used in Kampf et al. (1973) is not followed in this study because the standard cell setting given by STIDY (Gelato and Parthe, 1987) led to a more accurate structure. The cell was transformed from the one used in Kampf et al. (1973) with the transformation matrix 0,0,0; 0,0,0; 0,0, Vz. The mean value of | E 2 - 11 is 0.939, which implies a centrosymmetric space group. As barium is present, Patterson techniques were used to solve the structure. For the isotropic displacement model the structure was refined in space group P6lmmm to an R index of 14.7% . Conversion to anisotropic displacement factors of all the atoms in the structure resulted in convergence at R and R„ indices of 4.0% and 12.0% respectively (R = 5.4% for all 996 data). Addition of an isotropic extinction correction did not improve the results. Positional coordinates and anisotropic and equivalent isotropic displacement factors are given in Table 7.5. Interatomic distances and angles are given in Table 7.6, and the results of bond-valence analysis are given in Table 7.7. As well as the electron-probe microanalysis results, the bond-valence analysis suggest that there is no H 2 0 in the structure of 130 CD cz c75 It o o o N CD -*—« GO o o o o OT o oo o CO o o o o o o o d CO d IT) o o o •xt CO CO CN T— T— CO CO LO CO CO 00 LO" •xt co CO CO CNI CO T— CO csf CO CNI CD CM o CNI co •xt d o CD CM d ro CN 00 •xt o Is-•xt CO CM, "ro CQ CD o o •—-LO co" CO CO oo OJ oo 00 CM oo CM CD 55 55 CN LO CD LO o CD CM O CO LO" CM LO LO OT OT CO •xt CD CD CD O •xt CM co~ OT CD CNT CO CM CO CO CM CM CM co LO" LO CM CM O CM O CD LO LO CO O •xt CD •xt OT 00 OT LO CM •x? CO O LO CM CO 00 O OT LO LO CO of LO CM CM O 5" 00 OT LO oo Is-0O CD LO o CD •xt LO ^ oo" O —^^  Is- •xt T— CD oo CO 5, CM CD of CJ) CM 00 1^  o CD O T— CO CO CD CD CO •xt OT Is-co OT CM co" CO CO 00 OT Jz" OT O co •xt CD Is- co CO CO OT CD CO O o d CD CO co x " CM • x " O cxTx d cxTx CJ) oo co T— T— T— •xt CO o d T— CM co~ s •x s—' o O o o OT O CM CO Is-LO O O CM T- -Xt LO •xt Is-CM •xt Is-Is-00 LO o ° CO LO LO CM OT OT •xt CO CM CM IO o CNT |xf CD OT CO Is- T— T— T— CO CO CN o CM CO CM •xt CO Is-•xt OT OT co OT m 00 CO CM, Is- 00 m CM CD T— o x— Is- LO" OT CO 00 CD •xt OT CO T— CO Is- CD CO CO CM CD CO CD CM CO jx^ CO LO CM ^ ^ ,—v CM ,— O 00 CM, T— oo' Cxf LO CO •xt CM CM co ^— CO — ^— T— o OT T— CO CM d d o d co" ixT CM LO •xt OT •xt T— o d d O , s. CO x? x^_ -^ 00 T— CO i^ Is-Is- m in CM LO d d d 55 o CM o CO o ro Z TJ C ro cn CD LL II X TJ 0 CD L-ro to o _=j ro > TJ C ro T A B L E 7.6 S E L E C T E D INTERATOMIC DISTANCES (A) A N D A N G L E S (°) FOR V E R P L A N C K I T E Si-0(1)u &0(1)y Si-0(2) Si-0(3) <Si-0> 0(1)y-Si-0(1)u 0(2)-Si-0(1)u 0(2)-Si-0(1)y 0(3)-Si-0(1)u 0(3)-Si-0(1)y 0(3)-Si-0(2) <0-Si-0> Ba(1)-0(1), 0(1)x, 0(1)q &0(1)z Ba(1)-CI(1)l&CI(1)m Ba(1)-0(2)o&0(2)m Ba(1)-CI(5)cc, Cl(5)l, Cl(5)m & Cl(5)t Ba(1)-CI(2) Ba(1)-CI(3)&CI(3)s <Ba(1)-anion> 1.601(5) 1.617(2) 1.650(3) 1.617 116.41(32) 109.33(23) 109.33(23) 106.86(21) 106.86(21) 107.69(46) 109.41 2.758(4) 3.056(1) 3.209(5) 3.263(10) 3.312(1) 3.376(11) 3.112 X-CI(4) X-0(1), 0(1 )r, 0(1)aa &0(1)x <X-anion> 0(1)-X-CI(4) 0(1)r-X-CI(4) 0(1)r-X-0(1) 0(1)aa-X-CI(4) 0(1)aa-X-0(1) 0(1)aa-X-0(1)r 0(1)x-X-CI(4) 0(1)x-X-0(1) 0(1)x-X-0(1)r 0(1)x-X-0(1)aa <anion-X-anion> Ba(2)-0(1)w, 0(1 )e, 0 (1 )y&0(1 )v Ba(2)-0(3)j & 0(3) Ba(2)-CI(5) & Cl(5)r Ba(2)-CI(3)e, Cl(3)v, Cl(3)d & Cl(3)bb <Ba(2)-anion> 2.056(2) 2.076(4) 2.072 100.03(14) 100.03(14) 86.74(23) 100.03(14) 159.95(28) 89.78(23) 100.03(14) 89.78(23) 159.95(28) 86.74(23) 107.31 2.877(5) 2.907(4) 3.026(14) 3.111(6) 2.985 Note: <M-<|>> denotes the mean metal-ligand distance (A). X= Mn, Ti, Fe, Mg and Na. Equivalent positions: a = x, y, z + 1; b = -y + 1, x - y + 1, z; c = y, -x + y, -z; d = y, -x + y, -z + 1; e = -x + y, -x, z; f = -x + y, -x + 1, z; g = -x + y, -x, z + 1; h = x - y + 1, x, -z; i = x, y, z - 1; j = -x+ 1, -y, -z + 1; k = -y, x -y , z-1; I = -y, x -y , z; m = x -y , x, -z; n = x -y , x, -z + 1; o = y, -x + y - 1, -z; p = -x + 1, -y, z; q = x, y, -z; r = x, y, -z + 1; s = y, -x + y + 1, z; t = x - y, x, z; u = x - y + 1, x, z; v = -x + y, -x, -z + 1; w = y, x, -z + 1; x = -x + y, y, z; y = y, x, z; z = -x + y, y, -z; aa = -x + y, y, -z + 1; bb = y, -x + y, z; cc = -y, x - y, -z. T A B L E 7.7 B O N D - V A L E N C E * A R R A N G E M E N T IN V E R P L A N C K I T E Si(1) X Ba(1) Ba(1) Total 0(1) 1.064 0.462 0.282 0.205 2.013 0(1) 1.064 0.462 0.282 0.205 0(1) 0.462 0.282 0.205 0(1) 0.462 0.282 0.205 0(2) 1.020 0.083 2.205 0(2) 0.083 0(3) 0.999 0.189 2.376 0(3) 0.189 Cl(1) 0.152 0.304 Cl(1) 0.152 Cl(2) 0.141 0.425 Cl(3) 0.059 0.121 0.302 Cl(3) 0.059 0.121 Cl(3) 0.121 Cl(3) 0.121 Cl(4) 0.313 0.938 Cl(5) 0.054 0.103 0.212 Cl(5) 0.054 0.103 Cl(5) 0.054 Cl(5) 0.054 Total 4.146 2.159 2.077 1.887 Note: X= Mn, Ti, Fe, Mg and Na. Calcu la ted from the curves of Brese & O'Keeffe (1991). verplanckite. The formula determined from the structure solution is: B a n 7 7 ( M n 4 0 4 , T i 1 9 6 ) 6 ( S i , A l ) 1 2 0 3 6 C l 1 2 7 6 . 7.5 Description of the Structure Based on Earlier Work Kampf et al. (1973) described verplanckite as a three-dimensional framework composed of four-membered rings, (Si 4 0 ] 2 ) 8 " , of silica tetrahedra, and triple units of square pyramidal polyhedra around X sites, where X-Mn, T i and Fe. The Xpolyhedra are joined at a common apical oxygen atom. These triple units are connected to each other in the c direction by the silicate rings. The silicate tetrahedra and the X polyhedra are arranged in Chinese checkerboard fashion on (001), leaving a hexagonal void. Located along the edges of the void are Ba and C l atoms. The formal coordination of Ba(l) and Ba(2) is 12 and 11 respectively, but since some C l sites are partially occupied the effective coordination numbers are smaller. The void in verplanckite is comparable to that of the zeolites, but marginally larger. 7.6 Description of the Structure Based on This Work There are three distinct cation sites in verplanckite, all at special positions. There are two sites containing barium, Ba( l ) and Ba(2). Ba(l) is at special position 6(1) (x, 2x, 0) and Ba(2) is at special position 6(k) (x, 0, Vz). Si is at special position 12(o) (x, 2x, z). There are eight anion sites in verplanckite, six of them are distinct. There are five sites containing chlorine, C l ( l ) , Cl(2), Cl(3), Cl(4) and Cl(5). A l l five chlorine sites are partially occupied. C l ( l ) and Cl(5) are 1.03 A apart, making it 134 probable that they are disordered with chlorine statistically occupying either site. C l ( l ) is in special position 60) (x, 0, 0); Cl(2) is in special position 2(c) (V3, %, 0); Cl(3) is in special position 12(o) (x, 2x, z); Cl(4) is in special position 2(d) (V3, %, V2); and Cl(5) is in special position 12(n) (x, 0, z). There are three distinct sites containing oxygen. 0(1) is in a general position; 0(2) is in special position 12(p) (x, y, 0); and 0(3) is in special position 6(i) (lA, 0, z). Ba(l) is formally coordinated by fifteen anions: four 0(1) atoms, two C l ( l ) atoms, two 0(2) atoms, four Cl(5) atoms, one Cl(2) atom and two Cl(3) atoms (Figure 7.1). The Ba(l)-anion distances range from 2.758 to 3.376 A (mean 3.112 A). The polyhedral volume is 72.66 A 3 . Since all C l sites are partially occupied and C l ( l ) and Cl(5) are possibly statistically disordered, the effective coordination of Ba(l) is 9.38. Ba(2) is formally coordinated by twelve anions: four 0(1) atoms, two 0(3) atoms, two Cl(5) atoms, and four Cl(3) atoms forming a pentahedral prism and dipyramids (Figure 7.2). The Ba(2)-anion distances range from 2.877 to 3.111 A (mean 2.985 A). The polyhedral volume is 60.96 A 3 . A s 1 all C l sites are partially occupied, the effective coordination of Ba(2) is 8.04. The electron-microprobe data and refined site occupancy confirmed there is minor Ca and N a substitution in Ba(l) and Ba(2) sites. X i s formally coordinated by five anions, one Cl(4) atom and four 0(1) atoms, forming a square pyramid (Figure 7.3). The X-anion distances range from 2.056 to 2.076 A (mean 2.072 A). The polyhedral volume is 6.74 A . The effective coordination o f X i s 4.6 with Cl(4) partially occupied. The electron-microprobe and refined site occupancy data confirmed that the X site is occupied by M n , Fe, M g , N a and T i . 135 Cl(5) Figure 7,1. Coordination of Ba( 1) in verplanckite. Figure 7.2b. Ba(2) polyhedron in verplanckite. 137 Figure 7.3b. X polyhedron in verplancktie, forming a square pyramid, where X = Mn, Ti, Fe, Mg and Na. There is one tetrahedrally coordinated site containing silicon. The S i -0 bond distances range from 1.601 to 1.650 A (mean 1.617 A). The polyhedral volume, tetrahedral angle variance and mean tetrahedral quadratic elongation are 2.16 A3,12.97 and 1.00, respectively. According to the electron-microprobe data, there is a minor amount of A l substitution for Si . Verplanckite is a framework structure consisting of four-membered tetrahedral rings (Figure 7.4a) and square pyramid polyhedra. The axes of the tetrahedral rings are along ax, a2 and a3, forming sheets on (001). The square pyramidal polyhedra are joined by a common apical oxygen, 0(4), forming pin-wheels (Figure 7.4b) on (001). The pin-wheels are connected to the silicate rings in the c direction (Figure 7.5). Open hexagonal rings are formed by the X polyhedra and tetrahedral rings along [001], with barium polyhedra along the inner edges of the rings (Figure 7.6). Cl(3) and Cl(5) atoms are bonded to B a atoms and are closest to the center of the void. The distance between opposite Cl(3) atoms in the void is 8.01 A and the distance between opposite Cl(5) atoms is 8.90 A . When the diameter of the C l atoms, 3.6 A , is subtracted, the size of the void is 4.49 and 5.3 A, respectively. Using the average of these two values the volume of the void in one unit cell is approximately 134.9 A3. Since both Cl(3) and Cl(5) are only partially occupied, the effective diameter of the void is larger, approximately 6.46 A , with an approximate unit cell volume of 235 A . This is slightly smaller than the volume found in Kampf et al. (1973), approximately 300 A3. 7.7 Discussion Verplanckite is similar to the zeolite group of minerals. It has a very open structure with a void which is large when compared to most zeolites. Faujasite, the most open framework of the zeolites, 139 Figure 7,4b. Pin-wheels formed by X polyhedra in verplanckite, where X = Mn, Ti, Fe, Mg and Na. Figure 7.5. The verplanckite structure projected on to (010), silicon tetrahedra are yellow, X polyhedra are violet, barium atoms are blue and chlorine atoms are green. X = Mn, Ti, Mg, Fe. 141 Figure 7.6. The verplanckite structure projectea on to (001), silicon tetrahedra are yellow, X polyhedra are violet, barium atoms are blue and chlorine atoms are green. X = Mn, Ti, Mg, Fe. Note Cl(3) and Cl(5) atoms, closest to the centre of the void. 142 has pore openings surrounded by 12-membered tetrahedral rings with free diameters of about 7.3 A (Baur 1964). Ion exchange properties of verplanckite could not be investigated due to the small sample size. A s mentioned above, the structure formula, B a n 7 7 ( M n 4 0 4 , T i , 9 6 ) 6 ( S i , A l ) 1 2 0 3 6 C l i 2 7 6 , and the formula determined by electron-probe microanalysis, (Ba 1 2 7 6 , Cao 0 1 Nao 2 1 ) 1 2 9 g ( M n 4 5 4 , T i , 1 9 , Mgo 0 3 , Fe 0 2 ) 6 1 7 ( S i , ! 6 6 , A l 0 3 4 ) i 2 0 3 1 7 8 ( C 1 2 3 6 4 , F 0 4 9 ) 2 4 1 3 differ slightly. Part of the difference could be caused by compositional variation between verplanckite crystals, as the structure crystal was not used in electron-probe microanalysis. There is substantially more C l in the microprobe chemical analysis. It is possible that small amounts of C l are substituting in O sites. A s in Kampf et al. (1973), H 2 0 could not be located in the structure of verplanckite. Kampf et al. (1973) suggested that H 2 0 and O H must be distributed over the free volume of the void, either statistically over many positions with very small occupancy factors, or freely floating through the framework, like it occurs in faujasite (Baur 1964). It is possible that this is the case in this refinement for H 2 0 and C l atoms. In this study and Kampf et al. (1973) a difference in B a content could not be resolved between structure results and microprobe results. The discrepancy could be removed by recalculating the microprobe analysis on the basis of 12 B a atoms, but this would result in a Si content of 12.8 atoms and there are no likely sites for the excess Si. The verplanckite structure has a large degree of disorder. There is a large amount of partial occupation of atom sites, disorder between sites and atoms that can not be located at all. A s in Kampf et al. (1973) there is also evidence in the Fourier synthesis for positional disorder of several of the atoms. In this study Cl(3), Cl(4) and Cl(5) have very high temperature factors indicating positional disorder (Table 7.5). 8.0 Conclusion 8.1 Bigcreekite, UK6, Walstromite and Verplanckite Bigcreekite formed along very thin transverse fractures in fairly well laminated quartz-rich sanbornite portions of the rock. It post dates the other associated barium silicates and represents either a later primary phase from intruded fluids or a product of alteration of pre-existing Ba-rich minerals, possibly sanbornite. It is colourless and forms poorly developed crystalline masses parallel to the fracture direction. There are two perfect cleavages {010} and {001}. Bigcreekite is biaxial negative, with indices of refraction a 1.537(2), p 1.538(2), y 1.541(2); X = b, Y = a, Z - c and 2 V m e a s = 59.2(5)°, 2 V c a l c = 60.1 °. The crystal structure of bigcreekite was solved in space group Pnma to R = 3.5%, with cell parameters a 5.038(6), b 9.024(3), c 18.321(6) A . D c a l c for the ideal formula is 2.739 g/cm3 and Z = 4. The empirical formula of bigcreekite (based on 9 O) is (Ba, 0Nao m \ ]Si, 9 9 H g 0 2 O 9 . Bigcreekite is a hydrous chain silicate containing four-membered rings, which form chains of silica tetrahedra, parallel to [100] and staggered in the [001] direction. The B a layer contain B a 0 9 polyhedra, which form sheets perpendicular to (001). Water molecules fill the large spaces between the rows of silicon tetrahedra. The structure of bigcreekite has similarities to sanbornite and gillespite. U K 6 is light blue-grey, with one perfect cleavage on {0001}, and forms irregular masses up to 10 mm enclosed in parallel-bedded sanbornite-quartz rock. U K 6 is uniaxial negative, with indices of refraction e 1.594(2), w 1.642(2). The crystal structure of U K 6 was solved in space group P63mc to R = 4.85%, with cell parameters a 5.2432(7), c 29.859(6), Z = 2. The empirical formula of U K 6 (based on 7 cations and 2 H 2 0 per formula unit) is 144 (Ba 2 . 9 7 ,Na 0 0 3,Ca 0 . 0 1 ) 3 . 0 1 (Si 2 . 6 3 ,Al 1 3 4 ,Tio.o3)4Cl l 0 9 (Cl , 6 1 ,2H 2 0, F 0 . 0 5 ) 3 . 6 6 . D c a l c = 3.709 g/cm 3. T h e U K 6 structure is based on double tetrahedral layers, [T4Og]„„,, consisting of six-membered rings, with three layers of barium polyhedra connecting the tetrahedral layers. U K 6 is part of the Monteregianite-(Y)-Wickenburgite series (Struntz classification) and is structurally and chemically similar to cymrite. The crystal structure of walstromite was refined in space group P1 with unit cell parameters a 6.733(1), b 9.608(2), c 6.685(1) A , a 69.64(3), p 102.29(3), y 96.89(3)°, Z = 2. The refined structure was similar to the original crystal structure of walstromite described by Glasser & Glasser (1968), but with the anion positions more accurately determined. The redetermination of the walstromite crystal structure gave a final R index of 3.8 % which is improved from the original value of 16 % obtained by Glasser & Glasser (1968). The empirical formula of walstromite (based on 9 anions per formula unit) is (Ca, 9 4 7 , Na,, o u ) , 9 5gBa, 0 2 1 ( S i 2 9 9 5 , A l 0 0 2 3 ) 3 0 1 8 O 9 which is similar to the formula determined by Glasser & Glasser (1968). The walstromite structure consists of trigonal rings of silica tetrahedra arranged in layers on (101) with barium and calcium atoms lying approximately halfway between the rings, perpendicular to the a axis. Walstromite is isostructural with margarosanite, as Glasser & Glasser (1968). The crystal structure of verplanckite was refined in space group P6/mmm with unit cell parameters a 16.300(2) c 7.169(1) A , Z = 3. Anion positions are significantly different than the original crystal structure description of verplanckite by Kampf et al. (1973). The basic structure is similar but in this study there are five chlorine positions and three oxygen positions compared to three chlorine positions and four oxygen positions in Kampf et al. (1973). The redetermination of the verplanckite crystal structure gave a final R index of 4.0 % which is improved from the original value of 10.2 % obtained by Kampf et al. (1973). The empirical formula of verplanckite (based on 12 Si and 145 A l per formula unit) is ( B a , 2 7 6 , C a ^ N a ^ , ) , ^ ! ^ ^ , T i , 1 9 , Mgo. 0 3, F e 0 2 ) 6 1 7 ( S i , , 6 6 , Alo 34)i203i.78(d23.645 ^0.49)24.13' which is significantly different than analyses obtained by Alfors et al. (1965) and Kampf et al. (1973) due to large variations in chemistry. Verplanckite is a framework structure consisting of four-membered tetrahedral rings and square pyramid polyhedra which form open hexagonal rings, with barium polyhedra located along the inner edges of the rings. The verplanckite structure has a large degree of disorder. There is partial occupation of atom sites, disorder between sites and atoms that could not be located at all. There is also evidence in the Fourier synthesis for positional disorder of several of the atoms. It was determined from the crystal structure analysis and the electron-probe analysis that small amounts of C l are probably substituting in O sites and H 2 0 and C l atoms may be distributed over the free volume of the void in verplanckite, either statistically over many positions with very small occupancy factors, or freely floating through the framework. Verplanckite is similar to the zeolite group of minerals with a very open framework. The void in verplanckite is large when compared to most zeolites. 8.2 Future Research During the course of this study possible future research became evident: 1) Confirm the protolith of the sanbornite deposits in Fresno County, California as well as the other sanbornite deposits in California and Baja, California. 2) Better determine the ages of the sanbornite deposits in California and Baja, California to establish i f they are from the same stratigraphic horizon or just the same depositional environment. 146 3) Further refine the crystal structure of verplanckite by splitting the anion positions with the large temperature factors, indicating positional disorder and adding a weighting factor. 4) Describe the other new barium silicate minerals from Fresno County, California and submit them to I M A to be voted on as new minerals. 5) Refine the barium silicate minerals with R indexes > 8 to better locate anion positions. 6) Determine a structural hierarchy of barium silicate minerals and perform a study of the geochemical conditions for the formation of barium silicate minerals and see i f there is a correlation between conditions of formation and the type of structure. For example: Do barium silicates with open structures form under low temperature conditions? 147 9.0 References A L F O R S , T. J. & STINSON, M . C . (1965): New minerals from Fresno County - H Calif. Div. Mines and Geology, Mineral Info. Serv. 18, 27-30. A L F O R S , T. J., STINSON, M . C , M A T T H E W S , R. A . , & P A B S T , A . (1965): Seven new barium minerals from eastern Fresno County, California. Am. Min. 50, 314-340. B A U R , W . H . (1964): On the cation and water positions in faujasite. Am. Min. 49, 697-704. B O L O T I N A , N . B . , R A S T S V E T A E V A , R .K. , A N D R I A N O V , V.I . & K A S H A E V , A . A . (1991): Refinement of modulated crystals: structure of cymrite. Sov. Phys. Crystallogr. 36, 190-194. B R E S E , N . E . & O ' K E E F E , M . (1991): Bond-Valence Parameters for Solids. Acta Cryst. B47, 192-197. B R O B S T , D . A . (1983): Barium Minerals. In Industrial Minerals and Rocks (S.J. Lefond ed.). Volume 1, 5 t h Edition. American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc. New York. 485-501. B R O W N , I.D. & A L T E R M A T T , D. (1985): Bond-valence parameters obtained from a systematic analysis of the inorganic crystal structure database. Acta Cryst. B41,244-247. C L A R K , L . D . (1964): Stratigraphy and structure of part of the western Sierra Nevada metamorphic belt. U.S. Geol. Survey Prof. Paper 410, 70 p. C O D A , A . , D A L N E G R O , A . & ROSSI, G . (1967): The crystal structure of krauskopfite. AttiAccad. Nazi. LinceiRend. Classe. Sci.fis., mat. natur. Al, 859-873. C R O M E R , D.T. & L E B E R M A N , D. (1970): Relativistic calculation of anomalous scattering factors for X-rays. J. Chem. Phys. 53,1891-1898. C R O M E R , D.T. & M A N N , B.J . (1967): X-ray scattering factors computed from numerical Hartree-Fock wave functions. Acta Crystallogr. A24, 321-324. D A W S O N , K . R . (1985): Geology of barium, strontium, and fluorine deposits in Canada. Geological Survey of Canada, Economic Geology Report 34 D O U G L A S S , R . M . (1958): The Crystal Structure of Sanbornite, B a S i 2 0 5 , ^ w . Min. 43, 517-536. 148 DRITS, V . A . , K A S H A E V , A . A . & S O K O L O V A , G . V . (1975): Crystal structure of cymrite. Kristallografiya. 20, 280-286. (trans. Sov. Phys. Cryst. 20, 171-175). E N G L E H A R D T , W . V O N (1936): Die Geochemie des Barium. Chemie der Erde, 10,187-. E S K O L A , P. (1922): The silicates of strontium and barium. Amer. J. Sci. 5th ser. 4, 331-375. F R E E D , R .L . & P E A C O R , D.R. (1969): Determination and refinement of the crystal structure of margarosanite, PbCa2Si 3 0 9 . Zeits. Krist. 128,213-228. G A S T I L , G. , & PFffLLlIPS R.P. (1971): A n alternative reconstruction for Mesozoic California. Geol. Soc. America, Abs. With Programs (Cordilleran sec), 3, no. 2, 122-123. G E L A T O , L . M . & P A R T H E , E. (1987): S T R U C T U R E T I D Y - a computer program to standardize crystal structure data. J. Appl. Crystallogr. 20, 139-143. G L A S S E R F.P. & G L A S S E R L.S.D. (1961): Crystallographic study of Ca2BaSi 3 0 9 . Zeits. Krist. 116, 263-265. G L A S S E R L.S .D. & G L A S S E R F.P. (1968): The Crystal Structure of Walstromite. Am. Min. 53,9-13. G L A S S E R L.S. & G L A S S E R F.P. (1966): The Crystal Structure of Walstromite, C a j B a S i A . Abstract, 1966 meeting International Mineralogical Association, Cambridge, England. G O L D S C H M I D T , V . M . (1954): Geochemistry. Oxford University Press, London. H A R B E N , P.W. & B A T E S , R .L . (1984): Geology of the Nonmetallics. First Edition. Metal Bulletin Inc. New York. H I N T H O R N E , J.R., (1974): The Origin of Sanbornite and Related Minerals. PhD. Thesis. University of California, Santa Barbara, U S A . J A F F E , H .W. (1956): Application of the rule of Gladstone and Dale to minerals. Am. Min. 41,757-777. JONES, D .L . & M O O R E , J.G. (1973): Lower Jurassic ammonite from the south-central Sierra Nevada, California. Jour. Research U.S. Geol. Survey. 1,453-458. K A M P F , A .R . , K H A N , A . A . & B A U R , W . H . (1973): Barium Chloride Silicate with an Open Framework: Verplanckite. Acta Cryst. B29,2019-2021. K H A N , A . A . & B A U R , W. H . (1971): Eight-Membered Cyclosilicate Rings in 149 Muirite. Science 173, 916-918. K R A U S K O P F , K . B . (1953): Tungsten deposits of Madera, Fresno, and Tulare Counties, California. Calif. Div. Mines and Geology. Spec. rept. 35. L E P A G E , Y . (1987): Computer derivation of the symmetry elements implied in a structure description. J. Appl. Crystallogr. 20,264-269. M A C D O N A L D , G . A . (1941): Geology of the western Sierra Nevada between the Kings and San Joaquin rivers, California. Univ. of Calif Publ, Bull. Dept. Geol. Sci., 26, 215-286. M A N D A R I N O , J.A. (1981): The Gladstone-Dale relationship: Part IV. The compatibility concept and its application. Can. Min. 19, 441-450. N E W B E R R Y , N . G . , E S S E N E , E.J. & P E A C O R , D.R. (1981): Alforsite, a New Member of the Apatite Group: the Barium Analogue of Chlorapatite. Am. Min. 66, 1050-1053. N O L L , W. (1934): Geochemie des Strontiums; mit Bemerkungen zur Geochemie des Bariums. Chemie der Erde. 8, 507-600. N O R T H , A .C .T . , PHILLIPS, D .C . & M A T H E W S , F.S. (1968): A semi-empirical method of absorption correction. Acta Crystallogr. A24, 351-359. P A B S T , A . (1943): Crystal structure of gillespite, BaFeSi 4 O 1 0 . Am. Min. 28, 372-390. P A P K E , K . G . (1984): Barite in Nevada. Nevada Bur. Mines and Geology. Reno, Nevada. P A R T H E , E. & G E L A T O , L . M . (1984): The standardization of inorganic crystal-structure data. Acta Cryst. A40, 169-183. P O U C H O U , J.L. & PICHOIR, F. (1985): P A P f(rZ) procedure for improved quantitative microanalisis. Microbeam Analysis, 1985,104-106. P U T N A M , G.W. & A L F O R S , J.T. (1965): Depth of intrusion and age of Rocky H i l l Stock, Tulare County, California. Geol. Soc. America Bull. 76, 357-364. R A N K A M A , K . & S A H A M A , T.G. (1950): Geochemistry. The University of Chicago Press. Chicago. R O G E R S , A . F . (1932): Sanbornite, a New Barium Silicate Mineral from Mariposa County, California, Am. Min. 17,161-172. R U N N E L S , D.D. (1964): Cymrite in a copper deposit, Brooks Range, Alaska. Am. Min. 49,158-165. 150 STINSON, M . C . & A L F O R S , J.T. (1963): Unusual minerals from Fresno County, California. Calif. Div. Mines and Geology, Mineral Info. Serv. 16 (1), 10-11. STINSON, M . C . & A L F O R S , J.T. (1964): New Minerals from Fresno County -1 , California. Calif Div. Mines and Geology, Mineral Info. Serv. 17, 235-238. Z I M M E R M A N , R . A . (1969): Stratabound barite deposits in Nevada. Mineralium Deposita. 4, 401-409. Z I M M E R M A N , R . A . & A M S T U T Z , G.C. (1964): Small scale sedimentary features in the Arkansas barite district. In Sedimentology and Ore Genesis, G.C. Amstutz ed. Volume 2, 157-163. 151 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099479/manifest

Comment

Related Items