ELEMENTARY CRYSTALLOGRAPHY An introduction to the fundamental geometrical features of crystals M. J. Buerger Professor of Mineralogy and Crystallography Massachusetts Institute of Technology John Wiley & Sons, Inc, New York Chapman & Hall, Limited, LondonAbout the author . . . M, J. Buerger has been teaching in the fields of mineralogy, petrography, and crystallography at The Massachu- setts Institute of ‘echnology for over twenty-five years. Since 1944 he has been professor of mineralogy and ctystallography. He also carned three degrees at MIT: B.S. in mining engi- neering (1925), M.S. in geology (1927), and Ph.D. in mineralogy (1929). In 1948 he was visiti fessor of _ crystallography Faculty of Philosophy at the Univer- sity of Rio de Janeiro. The Geological Society of America, of which he was Vice-President in 1948, named Professor Buerger reci- pent of the Arthur L. Day Medal in 1951. He has served as President of the Mineralogical Society of America, the American Society for X-Ray and Electron Diffraction, and the Crystallo- graphic Society of America. He is also a member of the National Academy of Sciences, Professor Buerger is author of X-Ray Crystallography (Wiley, 1942). He is also co-editor of International Tables Jor X-Ray Crystallography, and of the international crystallographic journal, Zeinschrift fiir Kristallographie. About the book ... This is the first single book to pre- sent a rational introduction to the vast and interesting science of crystallog- raphy. Elementary Crystallography in- troduces the student to the science by offeting much more than a collection of facts and conclusions. It provides a penetrating study of the symmetry properties of crystals, which are: basic to all other phases of crystallography. Thus it gives the student the firm foundation in symmetry theory that he needs for any further intelligent study of crystals or of the parts of other sciences which deal with crystals. ‘The fungamental ideas concernin, crystal synimetry are a fascinating kin of solid geometry that can easily be understood by the reader with a back- ground of high school mathematics. ‘The author develops this philosophy in easy steps that anyone can readily grasp. BAHOSSELEMENTARY CRYSTALLOGRAPHYCopyricut © 1956 BY M. J. BUERGER All Rights Reserved This book or any part thereof must not be reproduced in ony form ‘without ‘the writen permission ofthe publisher Library of Congress Catalog Card Number: 55-9511 PRINTED IN THE UNITED STATES OF AMERICADedicated to Helen EvansPreface Nearly every mineralogy, metallurgy, chemistry, and physics textbook now contains a short section on crystallography. These abbreviated treatments can hardly be expected to present more than some of the conclusions of the science. Something was plainly needed to explain the reasoning leading to these conclusions. This book was written to supply a fuller account of the subject. Crystallography is a large science, and it is impossible to discuss all of it adequately in a single volume. This volume is concerned with the symmetry properties of crystals, which are fundamental to all other phases of crystal- lography. It is necessary to have a firm foundation in symmetry theory in order to make any further intelligent study of crystals or of the parts of other sciences which deal with crystals. The subject matter of this volume falls naturally into three sections. About the first half of the book is devoted to a rational development of the mega- scopically observable symmetries of crystals. A second part of the book is devoted to a discussion of the internal symmetries of crystals. A third section contains some advanced material. The fundamental ideas concerning crystal symmetry are inherently simple. They require no complicated mathematics, and can be understood by anyone whose background includes high school mathematics. In fact, they fall into the category of an interesting kind of solid geometry. Accordingly, elementary methods are used in most of the book, But, as a kind of appendix on ad- vanced methods, the book ends with three chapters which are concerned with group theory and its application to symmetry. The interested student who wishes to study this elegant device for formulating crystal symmetry may be expected to have the background necessary for really appreciating group theory by the time he reaches this point. I am indebted to several sources for permission to use illustrations. Pro- fessor Clifford Frondel of Harvard University kindly permitted me to borrow Figs. 10, 39, 40, 41, 44, 79, 103, 106, 129, 130, 159, 160, 167, 168, 171, 172, and 173 of Chapter 10 from Dana’s System of Mineralogy. Figures 9, 45, 46, 64, 65, 107, 108, 111, 114, 115, 116, 117, 124, 138, 139, 140, 141, viiviii Preface 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 163, and 164 of Chapter 10 are taken from Dana's Textbook of Mineralogy, and Figs. 16 and 17 of Chapter 11 are taken from Honess’ Etch Figures, with the kind per- mission of John Wiley & Sons. I am indebted to Professor Cornelius Hurlbut of Harvard University for the use of Figs. 1, 3, and 4 of Chapter 11 from his revision of Dana’s Manual of Mineralogy, and to Dr. H. E, Buckley for the use of Fig. 18 of Chapter 11. Dr. W. L. Bond supplied me with a sketch from which Fig. 9 of Chapter 11 was prepared. All other figures were drawn especially for this book. Mr. R. Ellis of Raytheon Manufacturing Co. was good enough to permit me to take photographs of his etched germanium spheres shown as Fig. 10 of Chapter 11. I am indebted to my colleague, Professor Avery Morton, for supplying modern equivalents for the names of organic compounds found in the older literature. Dr. Leonid V. Azaroff and Mrs. Sylvia Garvin kindly undertook the preparation of the index. Professors J.D. H. Donnay, I. Fankuchen, and E. M, Loebl each read a portion of the page proofs and made suggestions for corrections. Finally, it is a pleasure to acknowledge the great debt I owe to two able assistants who took charge of putting different parts of the copy into final form. Mrs. Evelyn Gower started with my penciled draft and cheerfully saw it through three successively improved stages of typescript. Mr. Nobukazu Niizeki prepared most of the drawings from my rough penciled drafts and incidentally made many suggestions leading to improvement. It is certain that without the wholehearted help of these two assistants the book would have taken two years longer to produce. M. J. BUERGER February 1956Contents List oF TABLES CHAPTER 1 ‘THe CRYSTALLINE STATE AND ITS RELATION TO OTHER STATES OF MATTER CHAPTER 2 Reperition THEORY : Geometrical operations required for repetition Nature of right and left Periodic repetitions Relation between repetition and symmetry Symmetry elements Permissible angular intervals Crystals and symmetry Exercises CHAPTER 3 ‘THE TRANSLATION PERIODICITY OF CRYSTALS Representation of translational periodicity by a lattice Vector representation of a lattice Conjugate translations Nonprimitive cells Notations for the rational features of a crystal Rational features Notation for rational lines Notation for points Notation for rational planes Lemma 1 Lemma 1 Lemma 2 Exercises CHAPTER 4 THE BASIC ROTATIONAL SYMMETRIES OF CRYSTALS Possible rotational repetitions Proper rotations Improper rotations Decomposition of improper rotation axes into other symmetry elements wna 4N 1 ix 23 23 23 2B 27 27x Contents meven, »/2 odd, ie, N= 4N +2 meven, 1/2 even, ie. n = 4N Operations of the first and second sorts ‘The limitations of a set by translation periodicity Possibility of simultaneous rotational symmetries in different directions Combination of rotations Euler's construction Permissible intersection angles Notation of combinations Permissible combinations Exercises CHAPTER 5 DERIVATION OF THE CRYSTAL CLASSES Importance of axial combinations Principles for discovering the permissible crystal symmetries The 32 crystal classes, Coincident rotation and rotoinversion axes Permissible combinations of proper and improper rotations Derivation of the crystal classes The monaxial classes ‘ The polyaxial symmetry combinations Exercises CHAPTER 6 COMBINATIONS OF SYMMETRY OPERATIONS INVOLVING REFLECTIONS AND INVERSIONS Fundamental combinations Two reflections Three reflection: Reflection and inversion Summary of relations : Alternative derivation of the crystal classes Principles of derivation The Schoenflies notation Derivation of the crystal classes Correspondence between Schoenflies and international symbols Exercises CHAPTER 7 ‘THE SYMMETRICAL PLANE LATTICES The crystallographic plane point groups Combination of a rotation and a perpendicular translation Combination of the rotation axes with a plane lattice General principles Some results of combining axes and plane lattices ‘The number of nonequivalent axes per cell Lattice types consistent with plane symmetries of the first sort Lattice types consistent with plane symmetries of the second sort 35 35 37 58 59 59 59 ot 66, Oo) 69 nm 2 n n B 5 79 79 79Contents xi The five plane lattices 83 Exercises 83 CHAPTER & THE SYMMETRICAL SPACE-LATTICE TYPES 84 Principles of derivation 84 Resolution of a space lattice into a sequence of plane lattices 84 General procedure 84 Symmetries to be considered 85 Designation of space-lattice types 85 Derivation of the space-lattice types 87 Symmetry 1 87 Symmetry 2 88 Symmetry 222 90 Symmetry 4 : 1 Symmetry 3 92 Symmetry 6 94 Symmetry 23 94 ‘Alternative discussion of symmetry 23, 96 The 14 space lattices 98 Exercises : : 98 CHAPTER 9 ‘Tue, COORDINATE SYSTEMS OF CRYSTALLOGRAPHY 99 The six symmetrical cells 99 Labeling and orientation of axes 102 Peculiarities of the crystal systems . 103 Cubic (isometric) system 103 ‘Tetragonal system 103 Hexagonal system 103 Rhombohedral axes 106 The orthorhombic system 106 The monoclinic system 107 ‘The triclinic system 107 Exercises un CHAPTER 10 CrystAL FoRMs 112 Isogonal relations between space groups and point groups 112 Repetition by space group and point group 412 The external surfaces of crystals 114 Crystal morphology 115 Form theory 115 i 417 discussion of forms 118. The triclinic system : 119 Chest. 119. ‘Mineral representatives 120 Other inorganic representatives 120 Organic representatives 120xii Class T Mineral representatives Other inorganic representatives Organic representatives “The monoclinic system Class 2 Mineral representatives Other inorganic representatives Organic representatives Class m Mineral representatives Other inorganic representatives Organic representatives Class = al representatives Other inorganic representatives Organic representatives ‘The orthorhombic system Class 222 Mineral representatives Other inorganic representatives Organic representatives Class mm2 Mineral representatives Other inorganic representatives Organic representatives Class 2 22 Mineral representatives Other inorganic representatives Organic representatives The tetragonal system Cast Mineral representatives Other inorganic representatives Organic representatives Chass t eral representatives Other inorganic representatives Organic representatives 4 Class = eral representatives Other inorganic representatives Organic representatives Class 422 Mineral representatives Other inorganic representatives Organic representatives Class 4mm : Mineral representatives Other inorganic representatives Organic representatives Contents 120 12 121 121 124 123 123, 123, 124 124 124 124 124 126 126 127 127 127 128 129 130 130 130 131 131 131 131 131 132 132 134 136 136 136 136 136 137 137 137 137 138 138 138 138 138 138 139 139 139 139 140Contents Class 32m Mineral representatives Other inorganic representatives Organic representatives Clas Mineral representatives Other inorganic representatives Organic representatives The hexagonal system Class 3 Mineral representatives Other inorganic representatives Organic representatives Chss3 : Mineral representatives Other inorganic representatives Organic representatives Class 32 Mineral representatives Other inorganic representatives Organic representatives Class 3m Mineral representatives Other inorganic representatives Organic representatives, Class 32 m Mineral representatives, Other inorganic representatives Organic representatives Class 6 Mineral representatives Other inorganic representatives Organic representatives Bee Class 6 = = Class = Mineral representatives Other inorganic representatives Organic representatives Class 622 Mineral representatives Other inorganic representatives Organic representatives Class 6mm ‘Mineral representatives Other inorganic representatives Organic representatives Class 6m2 = — m2 Mineral representatives Other inorganic representatives Organic representatives 140 141 ut Mt ui 142 142 142 142 146 146 146 146 49 149 150 150 150 150 151 151 151 151 152 152 152 153 153 154 154 154 154 134 154 155 155 155 155 156 156 156 157 157 157 158 158 158 159 159 150xiv Mineral representatives Other inorganic representatives Organic representatives ‘The isometric system Class 23 Mineral representatives Other inorganic representatives Organic representatives Class 432 a Class = 3 = Mineral representatives Other inorganic representatives Organic representatives Class 430m Mineral representatives Other inorganic representatives Organic representatives en Class 432 nm Mineral representatives Other inorganic representatives Organic representatives Exercises CHAPTER 11 PRACTICAL DETERMINATION OF POINT-GROUP SYMMETRY ‘Theory of symmetry determination from morphology Determination of symmetry by dissolution methods Dissolution forms Etch figures ‘Symmetry and optical properties Refractive indices and symmetry Optical activity Piezoelectricity effect Pyroelectric effect Diffraction symmetry CHAPTER 12 ‘THE ALGEBRA OF OPERATIONS Representation of combinations by products Exponents The inverse of a product Theorem 1 ‘The transformation of one operation by another Theorem 2 Condition for permutability Theorem 3 Meaning of the conjugate product ‘Theorem 4 Theorem 5 Contents 159 159 159 160 160 162 164 164 164 164 164 166 166 166 166 167 167 167 167 168 168 168 168 169 170 176 176 77 183, 185 136 187 189Contents CHAPTER 13 SPACE GROUPS ISOGONAL WITH THE MONAXIAL POINT GROUPS Combinations of rotations and translations Combination of a rotation and a perpendicular translation Theorem 1. Combination of a rotation and a parallel translation ‘Theorem 2 Combination of a rotation and a general translation Importance of screw axes Possible values of the rotational component of screw operations Possible values of the translational component of screw operations The permissible screw axes and their designations Derivation of the parallel-axial space groups Principles of derivation Space groups isogonal with point group 1 Space groups isogonal with point group 2 Space groups isogonal with point group 3 Space groups isogonal with point group 4 ‘Space groups isogonal with point group 6 Résumé of the space groups isogonal with the monaxial point groups CHAPTER 14 ‘THe DINEDRAL SPACE GROUPS Combinations involving nonparallel rotations Combination of operations of two intersecting screws Effect of displacing an axis, Combination of operations of two nonintersecting 2-fold screws Combination of operations for any two nonintersecting screws Derivation of the space groups with nonparallel axes General principle Conventions in representing space groups Space-group symbols. ‘Space groups isogonal with 222 Space groups isogonal with point group 32 Space groups isogonal with point group 422 Space groups isogonal with point group 622 Résumé of the dihedral space groups CHAPTER 15 ‘THe ISOMETRIC AXIAL SPACE GROUPS Space groups isogonal group 23 Space groups isogonal with point group 432 ‘Theory of derivation. Space groups derivable from P23, 123, and F23 Space groups derivable from P2,3 and 12,3 Résumé of isometric axial space groups CHAPTER 16 ‘SPACE GROUPS CONTAINING OPERATIONS OF THE SECOND SORT Combinations of operations of the second sort with translations Combination of a reflection and perpendicular translation xv 199 199 199 199. 200 200 201 202 203 206 206 209 209 213, 218 224 228 229 229 229 231 232 233 234 234 234 235 236 242 250 262 269 210 270 277 277 281 285 289 290 290 290Xvi Combination of an inversion and a translation . Combination of a rotoreflection and a translation parallel to the axis, Combination of a reflection and a parallel translation, Combination of a reflection and a general translation Importance of glides ‘Symbols for reflection planes and glide planes ‘The combination of a glide and a general translation Combination of a screw and a perpendicular glide Procedure in deriving complex space groups Triclinic space groups Point group 1 Monoclinic space groups Point group m Point group — Orthorhombie space groups Point group mm2 Lattice P Lattice 1 Lattice C Lattice 4 Lattice F Résumé of space groups isogonal with mm. 222 Point group = = — eee nm me 222 Resumé of space groups isogonal with = — Tetragonal space groups : Point group 4 Point group * Lattice P Lattice Z Point group 4mm. Lattice P Lattice 7 Résumé of space groups inogonal wi th 4mm Point group 42m Lattice P Lattice 2 Résumé of space groups isogonal with 72m Point ae int group = : Lattice P Lattice 7 422 Résumé of space groups isogonal with = = — pace groups isogonal with == ‘Trigonal space groups Point group 3 Lattice P Contents 291 292 292 294 294 295 296 296 296 298 298 299 299 302 307 307 309 310 317 321 323 329 329 330 340 340 349 331 352 352 304 398, 398 398Contents Lattice R Résumé of space groups isogonal with 3 Point group 3m Lattice P Lattice R Resumé of space ‘groups isogonal with 3m iat orp SS Point group 3 Hexagonal space groups ir a Point group 6 = = a 3 Point group 6m2 =~ m2 ’oint group ; 6 Point group — int group — Point group 6mm 6 Point group Isometric space groups Point group 43m : te Point group = 3 2 : 4 Point group © CHAPTER 17 Résumé OF SPACE GROUPS AND THEIR DERIVATION ‘A look back at space-group derivation Tabulation of the space symmetries ‘Symbols for the various symmetries ‘The total number of space symmetries and their distribution ‘The enantiomorphic pairs CHAPTER 18 EQUIVALENT POSITIONS IN SPACE GROUPS Sets of equivalent positions Space group Pmm2 Space group P| General positions and special positions Space group Pmm2 Space group P& Space group P4r Space group P4 Utilization of equipoint information Tabulation of equipoint information Determination of the number of molecules per unit cell Limitation on the locations of molecules CHAPTER 19 ‘THe DETERMINATION OF SPACE GROUPS AND THE UTILIZATION OF SPACE-GROUP IN- FORMATION ‘The background for space-group determination xvii 398 398 398, 398 399) 399 399 400 400 401 408 408 419 420 420 42t 429 440 440 441 456 457 457 an 4xviii Space-group determination Determination by abnormal spacings Ambiguities in the subnormal-spacing method Quantitative methods ‘The importance of space-group study CHAPTER 20 Grovr THEORY . Fundamental ideas about groups Preliminary ideas The group postulates Postulate 1 Postulate 1a Postulate 2 Postulate 3 ‘The multiplication table of a group Powers of operations Permutability Conjugate products ‘Transforms Theorem Theorem of permutability Abelian groups Conjugates Conjugate elements Theorem Sets of conjugate elements Invariant elements Subgroups, Basic ideas Cosets ‘The order of a subgroup ‘Theorem Conjugate subgroups Invariant subgroups Group products ‘Theorem Theorem Theorem Isomorphic groups Simple isomorphism General isomorphism ‘The transform of a group by i own elements CHAPTER 21 Grour THEORY APPLIED TO POINT SYMMETRIES Groups of rotations about an axis General aspects of groups of rotations The groups of improper rotations n odd neven, n/2 odd meven, n/2 even Contents 472 472 473 474 474 415 475 415 ATS 475 493 493 493 494 494 495, 496Contents Subgroups of the cyclical groups of proper rotations Subgroups of the cyclical groups of improper rotations The crystallographic cyclical rotation groups Groups of rotations about several axes Products of two proper rotation groups {2|{2} = 222, also designated Ds, V, or Q [2| 3} = 32, also designated Ds {2} [4] = 422, also designated Di {2} {6} = 622, also designated De ‘The isometric rotation groups. {3}{2|{2} = 23, also designated T [3}12]12}12) = 432, also designated 0 Groups of the second sort Point groups of the second sort General considerations Groups not containing ¢ Groups containing operation i CHAPTER 22 GROUP THEORY APPLIED TO SPACE SYMMETRIES Groups of translations Space groups and their symmetry elements Quotient groups ‘Theorem Homomorphism Infinite groups Construction of space groups Subgroups of space groups DEX. xix 496 496 497 498 498 499 S01 302 302 302 503 306 508 508 308 su 515 318 516 516 Si7 518 519 521 524 525List of CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. CHAPTER Table 1 Table 2. Table 3. Table 4. CHAPTER Table 1. Table 2. Table 3. Table 4. CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. CHAPTER Table 1 Table 2. CHAPTER Table 1. Table 2, tables 4 The rotoreflection axes The rotoinversion axes Correspondence between rotoreffection and rotoinversion axes Solutions of M = 2 cos « for permissible periods of crystallographic axes cos W + cos Ucos V sin Usin V cos W + cos Ucos V Crystallographic solutions of cos. = S221” + ex U cos V Data for crystallographic solutions of cos w = Geometry of the permissible nontrivial combinations of rotations Angles between certain rational directions in the isometric system 5 Reduction of coincident rotation and rotoinversion symmetries The monaxial crystal classes, The polyaxial classes conforming to PPP or PIT ‘The polyaxial classes conforming to both proper and improper combina tions 6 Classification of symmetry operations ‘The 32 crystal classes Some characteristics of Daa and Tz groups Correspondence between Schoenflies notation an 7 The rotations resulting from combining the operations of a 3-fold axis with the nonequivalent translations of the cell The rotations resulting from combining the operations of a 4-fold axis with the nonequivalent translations of the cell ‘The rotations resulting from combining the operations of a 6-fold axis with the nonequivalent translations of the cell ‘The numbers and locations of nonequivalent axes The plane-lattice types consistent with the 10 plane point groups 8 Derivation of the space-lattice types Designations of the general space-lattice types 9 The crystal systems Possible sets of interaxial angles in the triclinic system xxi 49 31 52 53 60 66 67 78 75 78 83 100 itxxii CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. CHAPTER Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. CHAPTER Table 1. Table 2. CHAPTER Table 1 Table 2. Table 3. Table 4. List of Tables 10 Forms occurring in the triclinic system Forms occurring in the monoclinic system Forms occurring in the orthorhombic system Forms occurring in the tetragonal system : Forms occurring in the classes of the hexagonal system having a 3 or 3 axis Forms occurring in the classes of the hexagonal system having a 6 or 6 axis Forms occurring in the isometric system n Distribution of forms in the triclinic, monocli tems : Distribution of forms in the tetragonal system Distribution of forms in the hexagonal system Distribution of forms in the isometric system Optical properties of the crystal classes Piezoelectric properties of the crystal classes Crystal classes arranged according to certain distinguishing physical properties and orthorhombic sys- 13 Permissible crystallographic screw axes Classification of screws according to number of threads Combinations of axes and lattices to be made in deriving the parallel- axial space groups Nonequivatent translations of cells, Results of combining 3, 33, or 32 with R Results of combining 4, 41, 42, or 4s with P Results of combining 4, 41, 42, or 43 with Results of combining 6, 61, 62, 6:, 6s, or 6s with P 16 Symbols used for glide planes Combination of possible glide planes with body-centering translation Combinations of 2-fold screws and parallel glides Space groups isogonal with mm2 based upon P Combinations of 2-fold screws with parallel d-glides (separation, s = a/4) Combinations of 2-fold screws with parallel d-glides (separation, s = a/8) Combinations of 4-fold screws and parallel glides Combinations of a 4 axis and perpendicular screws aw Résumé of the space groups Distribution of the space symmetries 18 Equipoints of space group Pm? Equipoints of space group P4 Equipoints of space group P4r Equipoints of space group P4 119 123 128 133, 147 148 163 172 173, 174 175 184 186 190 204 206 207 208 217 219 222 225 295 300 308 310 322 326 367 370 442 458List of Tables CHAPTER Table 1. CHAPTER Table 1 Table 2. Table 3. Table 4. Table 5. Table 6. CHAPTER Table 1 19 Determination of the nature of the screw axis isogonal with a 4-fold axis of the point group a. ‘Subgroups of the crystallographic cyclical groups of proper rotations Subgroups of the crystallographic cyclical groups of improper rotations Derivation of the crystallographic point groups of the second sort not containing operation ¢ : : The 11 crystallographic point groups of the first sort, and their subgroups ‘The 10 crystallographic point groups of the second sort not containing the operation é, and their subgroups The 11 crystallographic point groups of the second sort containing i, and their subgroups 22 Determination of translations associated with space groups isogonal with mm? based upon P xxiii 473, 496 497 310 511 512 513 523| + The crystalline state and its relation to other states of matter It is usual to regard matter as existing in three states, designated gaseous, liquid, and solid. With an increasing knowledge of the structure of matter it has become plain that a more rational subdivision of matter is into gaseous, liquid, and crystalline states, In order to appreciate the distinctive features of crystals, itis worthwhile to compare the crystalline state with the other states of matter. The distinctive structural feature of a gas is that it is composed of particles (atoms or molecular groups of atoms) which are substantially independent of one another except for occasional collisions. According to Avogadro's law, a mole of gas contains 6.03 x 10% molecules. Under standard conditions, a mole of gas occupies a volume of 22.4|iters. This is equal to 22.4 x 10° cc, or (22.4 x 10%) x 10% A%.* Thus, the volume occupied by a molecule is 22.4 x 10" 6.03 x 108 This is equivalent to one molecule in a cube whose edge is 737 x 10° = 33 A. Since the diameter of a molecule of a simple gas is of the order of two or three Angstrém units, it is evident that the space within a gas is very sparsely occupied. The molecules of the gas are in constant motion, owing to their temperature energy, and they course back and forth through their container, bouncing from its walls, and occasionally colliding and rebounding from one another. It is because of this very independence of the molecules from one another that gases can expand to an unlimited amount, and indeed can approximately obey the well-known perfect gas equation pV = nT. This law can be derived as the expected behavior of a system of mutually independent particles which have a negligible volume compared with the volume, Y, of their container. Thus, the salient characteristic of a gas is the near- independence of its particles and the consequent lack of influence of one particle on another. Ina liquid, the molecules are so close as to be in contact. For this reason the attraction of one molecule on its immediate neighbors is high. Indeed, the attraction is sufficiently great to hold neighboring molecules in contact so that, =37x 10 AY * A stands for Angstrom unit, One Angstrim unit is 10-#em. 12 Chapter 1 when a liquid is poured, it maintains a constant volume. Yet the molecules do not have permanent holds on one another, because thermal agitation has a sufficiently high energy to frequently lift a molecule through the field of attraction of another. Therefore, a liquid is structurally a mass of molecules so close to one another that each attracts its neighbors strongly,'yet not strongly enough that the bonds between molecules can be permanently maintained in the face of thermal agitation. Suppose, in the model of a liquid just considered, that thermal agitation is reduced to a point where its energy is insufficient to separate a pair of molecules which become bonded together. At temperatures below this, bonds between neighboring molecules are permanently maintained, and the collection of molecules assumes a rigidity due to the permanent bridgework of bonds through- out the collection. While the arrangement of such a collection of molecules may be either random or ordered, an ordered arrangement is one of lesser energy and therefore tends to develop. This ordered arrangement is the crystalline state. This book is concerned with the geometrical nature of this order and some of its consequences. From what has been said it is evident that the orderly arrangement of the molecules or atoms in crystals is a consequence of the dominance of the bonding of these units over other features of the aggregate. The order represents a state of low energy of the units with respect to this bonding. ‘Thermal agitation tends to disturb the order, and when the temperature is high enough so that the average thermal energy of a molecule exceeds its bonding energy, the molecules escape from one another’s influence, and the order is destroyed by melting.2 - Repetition theory One of the fundamental features of a pure chemical substance is that each molecule (or other unit) is a duplicate of all other molecules. From a geo- metrical point of view, each molecule is a copy, or repetition, of some arbitrarily chosen motif molecule of the mass. When a motif is repeated systematically the result is a periodic pattern. The periodic repetition may extend in one, two, or three dimensions. A systematic study of patterns proves that there are only 2 types of purely one-dimensional patterns, 17 types of purely two-dimensional patterns, and 230 types of three- dimensional patterns. Since crystals are regions of matter composed of the same kind of molecule systematically repeated, these must be arranged in one of the 230 three-dimensional pattern types. The geometrical features of crystals are consequently determined by three-dimensional pattern theory. Any study of crystallography must therefore have pattern theory as a foundation. Some of the geometrical aspects of crystallography can be appreciated, in anticipation, by informally examining patterns in two dimensions. It was stated above that there are 17 of these. This means that, if some general motif is selected, there are only 17 available ways of repeating it. Suppose that one selects a scalene triangle as an asymmetrical motif. The 17 possible ways of repeating it in two dimensions are shown in the illustration inside the covers of this book. That there are only these 17 ways will not be proved at this point. They are introduced here chiefly to illustrate the fundamental pattern charac- teristic of crystals. The labels of these patterns appear to be arbitrary, but they will turn out to be natural representations of the patterns as the theory of patterns is developed in this book. It will benefit the student very much to become pattern-conscious from this point on. He is advised to study all two-dimensional patterns that come under his notice. Such patterns will be found in wallpaper, floor-tiling, textile weaves, necktie ornamentation, etc. All such patterns (provided they are truly two- dimensional) will be found to fall in one of the 17 categories mentioned above. ‘They differ only in the motif selected for repetition. As the student begins his notice of plane patterns, many of their characteristics will be meaningless to him. As the study of this book proceeds, however, more and more of the characteristics of patterns will be recognized, until, finally, each one will come to be completely understood. 34 Chapter 2 Crystals conform to a set of pattern types like those of the inside-cover illustrations, only there are more available types because crystals are three- dimensional. Given a molecule, or other motif group of atoms, however, only 230 types of patterns are available as repetitions of the motif group. This book is largely devoted to finding these 230 types of space patterns by elementary means, to seeing what it is that limits them to this number, and to studying some of their characteristics, including resemblances and differences. This is the primary objective of elementary crystallography. Many of the inter~ mediate results found in the process of deriving the 230 space patterns are useful in their own right. In the chapters that follow, attention is drawn to these intermediate results in crystallography and to their usefulness. In this chapter some of the preliminary notions involved in repetition theory are discussed. This discussion is based on the fact that each possible pattern can be analyzed by considering what geometrical operations are necessary to repeat the motif molecule to become the numerous molecules of the crystal. Geometrical operations required for repe ion Consider the repetition of some asymmetrical figure, such as a comma, Fig. 1. The geometrical problem involved in repetition can be appreciated in the following formulation: If one has a rubber stamp located so that it prints the comma at location m of Fig. 1a, what movement must be given to the stamp in order to place it so that it can print the comma at location n of Fig. la? The required movement can be analyzed into two parts. Figure 1b shows that the stamp must be translated from one location to the other by an amount ¢, and Fig. Ic shows that it must also be rotated through an angle a. ‘Thus a com- bination of a translation and a rotation is sufficient to repeat a motif, (This is true in three dimensions as well as two dimensions.) This conclusion has a somewhat restricted application, because any asym- metrical figure (such as a hand, or an appropriate molecule) can exist in both right-handed and left-handed varieties. These are said to be enantiomorphous. On the other hand, a pair of duplicate right-handed figures, such as the two commas of Fig. la (or a pair of duplicate left-handed figures), are said to be congruent. This means that if one of the congruent right-handed figures is placed over the other, they “‘run together.” Figures which are thus related by either congruence or enantiomorphism are said to be equivalent. If the specifications for drawing a right-handed figure are given, the informa- tion is sufficient for drawing the corresponding left-handed figure. Thus the left-handed figure can be regarded as a kind of repetition of the corresponding right-handed figure. Suppose, then, that a left-handed figure is to be repeated froma right-handed model, Fig. 2a. What geometrical operations are required ? From Fig. 2b, c, and d it is evident that not only must the original figure beRepetition theory Fic. 1. Repetition of a congruent figure by a translation and a rotation, Q Fic, 2. Repetition of an enantiomorphous figure by a reflection, translation, and rotation.6 Chapter 2 displaced by translation and rotation but it must also be reflected to the left- handed equivalent figure. The geometrical operations of translation, rotation, and reflection are thus sufficient to repeat a left-handed figure from a right- handed figure (or the reverse) in two dimensions. Nature of right and left Consider an asymmetrical plane figure such as a comma, Fig. 3a. The figure can be specified by a list of the coordinates of its outline. If the same list is applied to a coordinate system in which the positive direction of the X axis y y ay ¥, Fic. 3a, Fig. 36. is reversed, Fig. 3b, the left-handed equivalent figure develops instead of the original right-handed one in Fig. 3a. No movement of the figure in 3 can make it coincide with that of 3a. This is simply because no movement of the coordinate system X’Y of Fig. 3b can make it coincide with the system XY of Fig. 3a. The total number of possibilities of reversal of the positive directions of the coordinate axes are outlined in Fig. 4, It is evident from a study of these that reversal of X to X corresponds to a reflection across Y, that a reversal of ¥ to ¥ corresponds to a reflection across X, while a reversal of both Y and Y to X and ¥ corresponds to a 180° rotation about the origin. It can therefore be concluded that the reversal of one coordinate axis generates an enantio- morphous figure, whereas the reversal of two coordinate axes generates a congruent figure which can be brought into coincidence with the original figure by a rotation of 180°. In three dimensions, reversals of the positive directions of one, two, or three coordinate axes are possible. This gives eight combinations as outlined in Fig. 5. The consequences of reversals are therefore somewhat more complicated for three dimensions, and are epitomized in Fig. 6: (1) A reversal of one axis, Fig. 66, corresponds to a reflection in the plane of the other two. (2) A reversal of two axes, Fig. 6c, gives rise to a set of axes which can be brought back to the original set, XYZ, by a rotation of 180° about the unreversed axis. (3) A reversal of three axes, Fig. 6d, corresponds to an inversion of the three axes through theRepetition theory 7 y Y Fig. 4. orignal 1 reversed 2 reversed 3 reversed Sretiection ‘Srolaton Sinversion ‘bout 28 Chapter 2 origin point. The axes of Fig. 6c can be brought into coincidence with those of Fig. 6a by a rotation of 180° about Z. The axes of Fig. 6d can be brought into coincidence with those of Fig. 6b by a rotation of 180° about ¥. Thus Figs. 6a and 6c constitute one congruent set and Figs. 65 and 6d constitute another congruent set. These two sets cannot be rotated into each other in any way, and so are enantiomorphous. Thus, in three dimensions the enantiomorphic relation can be established either by a reflection or by an inversion. Periodic repetitions In a collection of like figures, the entire set can be reproduced from any one member by repetition. If the several repetition operations are not related to one another the result is a random aggregate. On the other hand, a special C= = 6 =6 Fic. 7. Repetition by a periodically repeated translation, 1. Kind of repetition can be generated by a continued application of the same operation, Thus, let the operation act on the first figure to produce the second figure, and let the same operation act on the second figure to produce a third, etc. In this case all the figures are related to the first by periodic compounding of the one operation, This results in a periodic array of figures. Such an array is recognized as an ordered array, and this order corresponds to the order observed in the arrangement of atoms in crystals. Any kind of operation can be used as a generator of a periodic repetition. Figures 7, 8, 9, and 10, respectively, show periodic repetition generated by repeated application of a translation 1, a rotation A,, a reflection m, and of a combination of translation and reflection, the combination being designated g. Using Fig. 8 as an example, note that each figure is related to its immediate neighbors by the operation 4, Each figure is related to its second neighbor by the operation repeated twice. This is usually written as a kind of “product” A+A,ic., 4%. Third neighbors are thus related by 4: A+ A = 43, etc, Note that, if one figure is regarded as the progenitor of the others, its first neighbor is produced by A, its second neighbor by 4%, its third by 4°, etc. The set ofRepetition theory Fic. 8. Repetition by a periodically repeated rotation through angle « about axis A. 6 Q 6 Q 6 Fic. 10. Repetition by a periodically repeated translation-reflection, g.10 Chapter 2 repeated figures is thus produced from the original figure by the set of operations A, A®, A’, --+. Tt is important to have a label for the “operation” which produces the first figure from itself. This is called the identical operation and is designated 1 (or sometimes J or E). Thus the complete set of operations 1, A, A2, AB, -~ + produces the complete set of figures from the first figure. There is a one-to-one correspondence between repeated figures and the operations that produce them. This complete set of operations is called a group. The word “group” has special mathematical connotations which will not be developed at this point. Relation between repetition and symmetry In a periodic repetition, such as in Fig. 8, the operation has been said to produce the second figure from the first, the third from the second, ete. A more general view is to regard the operation as acting on all space. In Fig. 8, then, the operation requires that all space (and therefore anything contained in it) be repeated at an angular interval «, The object is contained in space, so it is required to be repeated at an angular interval « as a second object. But the requirement of repeating all space at an angular interval « also requires that the second object be repeated again at an angular interval « to become a third object, etc. Thus, if an operation repeats space at an angular interval a, it is only consistent with a set of m (in this case, 4) symmetrically arranged objects separated by angular interval a. The operation is usually spoken of as a symmetry operation, since, acting on all space, it requires a symmetry of that space. Note that the operation of translation is also a kind of symmetry operation. Symmetry elements The geometrical locus about which a group of repeating operations acts is called a symmetry element. Thus the line normal to the page about which operations A,, 4.2, ++ all act is a symmetry element; in this specific case it is called an axis of rotational symmetry, or simply a rotation axis. In Fig. 9, the plane across which the reflection occurs is a symmetry element called a reflection plane, or a mirror, etc. Pert je angular intervals In a foregoing section the relation between repetition and symmetry was discussed. A consequence of that discussion is that any angular repetition interval « is allowable provided that periodic repetition at this interval eventuallyRepetition theory itt superposes the repeated object on the original object. Suppose that » repetitions at angle « complete this cycle. It immediately follows that n+a=2n, a) so that = 2n|n. Q) Accordingly, the allowable angular-repetition interval is any aliquot part of 2. Thus, to an angular-repetition interval « there correspond n repetitions per 27. The symmetry element corresponding to this « is called an n-fold axis, since the axis is associated with an n-fold duplication of any motif. Evidently 1-fold, 2-fold, 3-fold, - -- axes are possible in angular repetition. The short- hand notations for these axes are simply the numerals 1, 2, 3, « - - , respectively. Crystals and symmetry Since crystals are orderly, periodic arrangements of atoms, it follows that the basic geometrical theory of crystals is fundamentally concerned with the operations of repetition and their combinations. The remainder of this book is largely concerned with a development and discussion of the symmetry pro- perties of crystals. In the following chapters operations of repetitions are considered singly, and in combinations which prove to be useful in theoretical and practical crystallography. Exercises 1. Set up a “right-handed” axial coordinate system XY in a plane. Locate points having coordinates 2,1; 1,3; and 5,6, Construct a triangle by drawing lines between these points. Duplicate this exercise with a left-handed coordinate system and compare the triangles, Make a tracing of the triangle of the last drawing and try to superpose it on the triangle of the first drawing, Can you do this? Turn the tracing paper upside down and try to superpose it on the first drawing. 2. Repeat question 1, using points with coordinates 1,1; 5,1; and 3,7. How do you explain the difference in the results obtained with the sets of coordinates given in questions Land 2? 3. Which patterns in the illustrations inside the covers have 2-fold axes? Which have 3-fold axes? Which have 4-fold axes? Which have S-fold axes? Which have 6-fold axes? Which have 7-fold axes? 4. Which patterns in the illustrations inside the covers can be divided symmettically by ‘one or more mirror planes?3: The translation periodicity of crystals In the Introduction it was pointed out that crystals are unique in being ordered. matter, and in Chapter 2 a specific notion of the nature of order was given. A study of the kinds of order and their combination will form a large part of the discussion in this book, but there is one kind which is common to all crystals, that is, translational periodicity. Crystals have translational periodicity in three dimensions. Conversely, a crystal may be defined as a region of matter within which the atoms are arranged in a three-dimensional translationally periodic pattern, This translational periodicity has geometrical consequences which will receive initial attention in this chapter. A digression is entered at this point to qualify the characteristic of translational periodicity in real crystals. Ifthe siructure ofa crystal is periodic when static, heat motion slightly disturbs the periodicity of thestructure at any one instant. Since heat motionis extremely rapid, however, the structure is still periodic if it is considered over any appreciable interval of time. A crystal is thus statistically periodic when considered at any real temperature, Instances are now recognized in which the entire pattern of a crystal is not periodic. In such instances the nonperiodic aspect of the structure is rendered so by extreme temperature motion of one or more kinds of atoms in the structure, or by a statistical situation which is the formal equivalent of it. For example, in crystals of the high-temperature form of Agl, the large iodine atoms form a periodic structure while the smaller silver atoms, under the influence of thermal energy, flow continuously through the interstitial channels between the iodine atoms. Statistically, over even a short period of time, this kind of pattern is triperiodic, and consequently it behaves geometrically in the same manner as a strictly periodic pattern. For ignation “crystal” is extended to include even matter which is only in three dimensions. Representation of translational periodicity by a lattice ‘The geometry consequent upon the translation periodicity of a pattern can be conveniently discussed if other features of the pattern are eliminated. Thus, Fig. 1a shows a section of a pattern which repeats at the translation interval, ¢. For purposes of studying the geometry of the repetition, rather than that of the motif which is repeated, the pattern can be represented by a row of points separated by interval t, Fig. 1b, Such a periodic sequence of points is called a linear lattice (or row). ‘The points of a linear lattice may be regarded as generated 12The translation periodicity of crystals 13 by the repeated application of a translation, f, to some initial point. Any point may be taken as the initial point. Ifa linear pattern having a repetition interval f, is itself periodically repeated by another translation f,, as indicated in Fig. 2a, a plane pattern is generated. The repetitional aspects of the pattern can be abbreviated by a two-dimensional array of points called a plane lattice (or net), Fig. 2b. Fic. 1a, Pattern produced by periodic 6 66 GG GG G _eepetition in one dimension and defined by translation, 1. Fic. 1b, One-dimensional lattice of pattern . in Fig. 1a. Finally, if a plane pattern having repetition intervals 1, and f, is periodically repeated by a third noncoplanar translation f,, a space pattern is generated, Fig. 3a. The geometrical characteristics of the pattern are conveniently represented by a three-dimensional collection of points known as a space lattice, Fig. 3b. When the term lattice is used without qualification, a space lattice is implied. 66666666 66666666 66666666 66666666 166666666 4 Fic. 2a, Pattern produced by periodic Fic. 2. Two-dimensional lattice of repetition in two dimension and defined pattern in Fig. 2a. by translations, f, and f, A lattice is thus an epitome of the pattern of translational repetition. If any motif is imagined as placed so that a selected point falls on one of the points of the lattice, then the lattice implies that an identical, parallel motif unit appears in the neighborhood of each point, and specifically located so that each motif unit then has an identical point at a lattice point. Except for the convenience of a particular problem, the lattice should not be14 Chapter 3 thought of as having a definite origin in a crystal, any more than a periodic wave has a definite origin. Rather the lattice points of any row are separated from one another by phase differences of 27, 4, 67,-+- ,n(2n) of the cycle of repetition. ESSESESS ESSSEESS EESSSSES $ESEESES SSSESSSS Grae Fic. 3a, Pattern produced by periodic Fi. 35. Three-dimensional lattice of repetition in three dimensions and pattern in Fig. 3a, defined by translations fy, ts, and f3. Vector representation of a lattice Any plane lattice can be regarded as a linear lattice having translation period 1,, Fig. 4a, periodically repeated by a second translation f,, as developed in the previous section. ‘The translations ¢, and f can be regarded as vectors. If any lattice point is chosen as an origin, the location of any other given lattice point enna Fic. 4a, Fic, 46, ‘The shaded parallelogram is a primitive cell. can be defined with respect to this origin lattice point by a vector 7, Fig, 4b. Obviously Tis a linear sum of vectors #; and /,, For example, in Fig, 4b, T= 2, + 3h ® and, in general, for any lattice point Tul toy 2The translation periodicity of crystals 15 where u and v are integers. The more general relation for three dimensions is evidently T= ut, + vy + wh, @ where 1, v, and w are integers, Let the symbol K stand for a collection. Then the complete set of vectors to all lattice points is T=kT = K{ui, + vt + wig} (u,v, w, integers), @) Let -7* stand for the point at the end of vector T. Then the point at the end of the vector (3) is “ut + vfy-+ wty. The collection of points constituting the space lattice T’ can then be represented as P= Ker} = K{ut, + vig + wig} (uw, v, w, integers). (6) Conjugate translations If %, % and f are chosen, the lattice is uniquely defined, according to (5). On the other hand, a given lattice does not uniquely fix a single set of trans- lations f, , and f,. Fundamentally this is because a linear combination of ,, ty, and f, may be used to define another vector, say ,'. Consider this feature in two dimension: In Fig, 46 the vector Tis specifically a T=m4+3y This relation implies that the plane lattice was regarded as a linear lattice having period 1,, repeated periodically at an interval fy. But, since the original row having interval f, is infinite, the plane lattice can also be thought of as a periodic row of interval ,, repeated periodically by a different translation 1,', Fig. 5. But 7,’ can be defined as a linear combination of #, and fq, specifically © @16 Chapter 3 If this value is substituted into (1), there results (8) Fic. 8. The shaded parallelogram is a primitive cel Any of the infinite number of translations, tq, which periodically repeats the row having interval t,, can be chosen as a second translation. Furthermore, any of the infinite number of row directions of the lattice could have been used as the first row, and any translation which periodically repeats this row could have been chosen as second translation. Two such “paired” translations are called conjugate translations. ‘Together they determine a parallelogram known as a primitive cell. A primitive cell is shaded in Fig. 4 and Fig. 5. Fria. 6. One lattice point is associated with each primitive cell. This can readily be seen by imagining the area occupied by the lattice to be partitioned into primitive cells, Fig. 6. If the outlines of the parallelograms are shifted slightly with respect to the lattice points, it is obvious that there is one lattice point in each cell. Every primitive cell of a lattice has the same area, This is because the area is the product of the period of the first row, namely #,, by the spacing between rows. The spacing is independent of the direction of fy.The translation periodicity of crystals 17 Nonprimitive cells The cell outlined by two arbitrarily chosen vectors between lattice points is not necessarily a primitive cell, as is illustrated by the examples in Figs. 7a and 7b. The fundamental reason for this is that if the first row is taken as that row which has translation interval %, a second vector, i, can be chosen so that it relates the original row to every second parallel row of the lattice, Fig. 7a, or to SoG ee . © eH2 . row . moe Fic, 7a, Translations defining a double Fic. 76, Translations defining a triple cell. cell every third row of the lattice, Fig. 7b, etc. These cells contain a total of several lattice points per cell; the count consists one lattice point for the lattice point at the corners of the cell, plus the additional lattice points in its interior. Such cells are called multiple cells. ‘The translations in Fig. 7a outline a double cell; those of Fig. 76 outline a triple cell; etc. It might be supposed that multiple cells are useless, and should be avoided. There are occasions, however, when they are very useful, especially to gain the advantages of orthogonality. This will be developed in Chapter 8. It is customary to select an appropriate cell, either primitive or multiple according to the requirements of the case, and call this the unit cell of the crystal. The notions of this and the foregoing section can obviously be applied to three-dimensional lattices by the addition of one more translation. Notations for the rational features of a crystal Rational features. Those geometrical features of a crystal which are deter- mined by lattice points are called rational. Thus, a lattice point (in respect to another lattice point) is a rational point; a row of lattice points is a rational line; and a plane of lattice points is a rational plane. All other features are irrational. It is important to have appropriate notations for referring to these rational features, ‘These notations are developed in the following sections:18 Chapter 3 Notation for rational lines. If an origin point is chosen, then any other lattice point can be reached by a linear combination of the conjugate vectoi 8) Tut, + thy + why Each such line Fis determined by the two lattice points at the ends of the vector, and so is rational. This vector equation has the same form for all rational lines, and the only distinctive feature about a particular line is the set of co- efficients wow. ‘Thus the three integers ww are taken as the indices of the line defined by vector equation (3). To indicate that the sequence of three integers represents a line, they are placed between square brackets, thus: [wow]. Par- ticular examples are [124], [100], and [ITI]. The bar over the integer in the last example indicates that it is negative; i.e. [1I1] is a short, neat notation for (1, I, 1. Note that [222] defines the same line as [111]. For this reason it is customary to designate a line by the use of the three smallest permissible indices; icc, if the three indices should contain a common factor, it is removed and the resulting set of integers without a common factor is given. Notation for points. Since a vector from the origin lattice point defines the point at its end, the three integers that define the vector also define the point. It is currently customary to write these three numbers without parentheses or brackets. ‘The reason for this is that one of the major uses of the notation for a point is to specify a point within a unit cell. Points within the cell necessarily have fractional coordinates, and a sequence of three fractions is not readily confused with the notations for a direction or a plane, which are sequences of three integers. When, however, one wishes to specify a rational point, i.c., a lattice point, then some definite notation is necessary. There existed a limited practice of enclosing them in a pair of double square brackets, to indicate that this set of three numbers refers to a point, thus : [{uvw]]. A preferable notation, used in this book, is to place them between dots, thus -urw. This notation has advantages when symmetrical sets are to be designated. Notation for rational planes. To derive a reasonable notation for rational planes, the well-known intercept form of the equation of a plane can be used. Consider the analog of a plane in three dimensions, namely a line in two dimensions, Fig. 84. If the line intercepts the coordinate axes Y and Y at A units and B units respectively from the origin, then the equation of that line can be writtenThe translation periodicity of crystals 19 x= A; and when x=0, y = B.) Similarly the equation of a plane in three dimensions, in terms of its intercepts A, B, and C, Fig. 8, is oye AtBtC The numbers A, B, and C could be accepted as characteristics of the plane, and indeed this was done in the early days of crystallography. But there are advantages to manipulating (10) into another form. Wy A disadvantage of (10) is that the coefficients of x, y, and z are fractions. To convert them into integers, both sides can be multiplied by ABC, giving ae at ae yt ue 2 or BC + ACy + ABz = ABC. (12) The coefficients of (12) are now products of integers, For simplicity replace them by the integers* (10) y x é\ x Z Fic. 8a. Fic. 86. ABC a) h= BC, k=AC, (13) l= 4B, so that (12) has the simpler form hic + ky + lz = ABC. (14) This is the equation of the rational intercept plane, i.e., of the particular plane whose intercepts are A, B, and C translations on the X, Y, and Z axes, respectively.* * In order to display the general idea without unnecessary complication, it is assumed in this. discussion that no pairs of the integers , B, and C contain common factors. If they do, the ion is slightly more complicated; this complication is discussed in the next footnote. Cases in which 4, B, or C are zero or infinity are also excluded.20 Chapter 3 The intercept plane does not, in itself, have very useful properties, nor is it easy to find from its equation. To transform (14) into something more useful, it is desirable at this point to introduce two auxiliary theorems, or lemmae: Lemma. There are ABC equally spaced, identical planes, from the origin to the rational intercept plane (provided that a common factor does not occur in any of the pairs AB, AC, or BC). To prove this, first consider the somewhat simplified problem in two dimensions, Figs. 8a and 9, There are A translation vectors of length 4 along “Ald Fig. 9a. Fic. 95, Fic. 9e. the X axis from the origin to the intercept line, These 4 translations repeat the intercept line A times from the intercept plane to the origin, Fig. 9a. There are B translation vectors of length f, along the Y axis from the origin to the intercept line. ‘These B translations repeat the intercept line B times from the intercept line to the origin, Fig. 9b. It similarly repeats B times each of the other A lines of Fig. 9a. (The truth of this last statement is not as obvious as the previous one. It will be appreciated if a careful sketch is made of Figs. 8a, 9a, and 9b, utilizing the B of the f, translations in both forward and reverse directions to operate on the other A lines of Fig. 94.) Thus there are AB lines from the origin to the intercept line, Fig. 9c. If the reasoning is extended to three dimensions it then becomes evident that there are ABC planes from the origin to the intercept plane.* All these planes are identical since they are repetitions by translation of the plane through the origin point. * (Do not read this footnote until after finishing this chapter.) This is valid only so long as 4, B, and C are each prime to one another. If any two of these, say AB, contain a common factor q, then the first line in the XY plane to have rational inter- cepts has intercepts Aig and Big. ‘There are therefore (4/q) (Big) lines from the origin to this rational-intercept line; there are g times as many lines from the origin to the line having intercepts A and B on X and Y, that is (AJq) (Blq)q = ABiq lines. More generally, if A and B contain the highest common factor q, Bad fon Candd in in»The translation periodicity of crystals 21 Lemma 2. The equation ax + by + cz =m represents a plane m times as far from the origin as ax + by -+.cz To see this, consider ax + by + cz =m, (15) az + by + cz=1, 6), and let 2, y, and z be zero in pairs in both equations. It is then plain that (15) is m times as far from the origin as (16). Returning now to Lemma 1 and the equation (14), it is evident that (14) describes the ABCth plane from the origin. According to Lemma 2, the description of the first plane is he tky + le=1. ay This is the equation of the plane nearest to the origin. While (17) is specifically the equation of a particular plane, it is equivalent by translation to a whole stack of parallel, identical, equally spaced planes. The specific character of (17) for a particular plane is embodied in the coefficients A, k, and /. This set of three integers is called the indices of the plane. To indicate that a sequence of three integers represents a plane, they are placed in parentheses, thus: (k/). There is a simple interpretation of (hkl). To see this, recast (17) in intercept form, thus: oy, ih + Tk This is the intercept form of an equation of a plane whose intercepts are 1/h, 1k, and 1Jl. This signifies that the planes of the stack (hkl) divide the t, trans- lation into h parts, the ty translation into k parts, and the ty translation into | parts. 7 +h as) then there are ABC/qrs planes from the origin to the rational intercept plane. ‘The more general version of Lemma 1 is then: Lemma la. There are ABCiqrs equally spaced, identical planes from the origin to the rational intercept plane, whereq is the HCF of A and B, r the HCF of Band C, ands the HCF of C and A. Under these circumstances the sequence of equations from (11) to (14) becomes: ABCigrs _, ABCIgr ABC — 7 7 (ita) BO A ee (12a) gs? qs!” ges qs Let h = BClgrs, - ‘cer | oa 1 = ABlrs, Then he = ABCIqrs. (14a) This is the equation of the plane having intercepts 4, B, and Con X, ¥, and Z. Since there are ABCjqrs planes from the origin to this plane, the equation of the first plane is ha + ky += 1 (17a)22 Chapter 3 This theory has been developed for planes within a periodic pattern as repeated byalattice. - The notations are also applicable to, and used for, the designations of the surface planes of crystals, which are also “rational.” This subject is further developed in Chapter 10. Exercises 1. Using a piece of ordinary graph paper, locate the points having integral coordinates, thus defining a plane lattice. In this system of points draw lines between points which outline (@ a primitive cell, (6) a double cell, (¢) a triple cel 2. Using the plane lattice defined as in question 1, draw a square primitive cell. Maintain- ing a particular line for a base, and maintaining the same altitude, choose another side so that the cell so determined has the shape of a rhombus. Choose another side so that the cell so determined has the shape of a parallelogram. Are these primitive or multiple cells? 3. Using the plane lattice defined as in question 1, draw the directions (11), 13], (31), (10). 4, Using the plane lattice defined as in question 1, draw a line whose intercepts are A = 2, B=3. Repeat this line by using the +7, —f, +7, and —f translations successively several times (in other words, shift the lines in directions and amounts defined by these vectors, ‘which are the edges of the square primitive cell of the plane lattice). When this has been done, a small area should be covered by a pattern of uniformly spaced lines. How many such Tines are there from a cell origin to the line which makes intercepts A = 2 and B= 3 away from it? Compare the result with Lemma 1. 5. Carry out question 4 for A = 3, B = 4 and compare the result with Lemma 1. 6. Carry out question 4 for A = 2, B = 4, and compare the result with Lemma 1a. 7, Make a set of translation-equivalent dots representing a plane lattice. Pick any two noncollinear but nonconjugate translations. Draw the cell based upon these, and determine its multiplicity.4 - The basic rotational symmetries of crystals Possible rotational repetitions Proper rotations. ‘The discussion in the last part of Chapter 2 showed that periodic angular repetitions can be described by means of symmetry elements which are known as rotation axes. In the pure angular repetition, a right handed object (say) is the original object, and all repeated objects are also right-handed. The rotational operation which repeats a congruent object from another object is termed a proper rotation, and the corresponding symmetry element is a proper rotation axis, ‘The repetition of a point by some of the n-fold proper rotation axes is illustrated in Fig. | Improper rotations. ‘There exists, however, another kind of rotational repetition, namely one in which the enantiomorphous object (and indeed all enantiomorphous space) is repeated. ‘The repetition operation is thus a combination of a rotation and a reversal of sense. From an initial right- handed object (for example) it repeats a left, a right, a left, --- , so that neighbor- ing objects are enantiomorphous, while alternate objects are congruent. The rotation concerned with this kind of repetition is called an improper rotation (in contrast to the pure, or proper, rotation), and the corresponding symmetry element is an improper rotation axis, It was shown in Chapter 2 that three-dimensional enantiomorphous objects can be derived from one another either by a reflection or by an inversion. Accordingly, the alternating rotational repetition can be defined either in terms of a combination of a rotation and reflection or in terms of a combination of a rotation and an inversion, Symmetry axes corresponding to these possibilities are called rotoreflection axes and rotoinversion axes respectively. Rotoreflection axes are assigned the shorthand labels 7, 3,3, +++, ie., #, while rotoinversion axes are assigned the labels 1, 2,3,---, i.e. a. Figures 2 and 3 illustrate the way rotoreflection and rotoinversion axes repeat a point. Since a reflection and an inversion are alternate ways of transforming an object into its enantiomorphous equivalent, it follows that the set of patterns of repetition of a point by rotoreffection (a few of which are shown in Fig. 2) is the same as the set of patterns of repetition of a point by rotoinversion (a few of which are shown in Fig. 3), although the arrangement and labeling of mare not necessarily the same. ‘The correspondence between rotoreflection and rotoinversion axes is listed in Table 3 and will be discussed later. 23Fo, 1. Repetition produced by the operations of the proper roation axes, m. pando-Fip.2.-Repsiton produced by the operation ofthe improper rotation ach Sold cise erent poins above the plane ofthe pupers small open cles vepesent pons below the plane of the papers sqoisdis fo soussouauts ouoyvjos 21809 24.1 stKY \ 7 NY i) INL OY Fo. 3, Repetition produced by the operations ofthe improper rotation axes, Fi. oO — \ Or fp uardoug,The basic rotational symmetries of crystals 27 Decomposition of improper rotation axes into other symmetry elements ‘Some improper rotations are unique whereas others can be decomposed into separate symmetry elements. The improper axes behave differently in this respect according as m is even or odd and according as n/2 is even or odd. This behavior is concerned with the placing of an enantiomorphous or a congruent point at angular separations of 2 and 7 from the original point. First note that, when 7 is even, the cycle of repetitions is completed after the total angle of 2a has accumulated. But when 1 is odd, a total angle of 47 is required to complete the cycle. This can be readily appreciated for #, Fig. 2. For 3, for example, 3 rotations of 2x/3 bring the repeated point at the angular position of the initial point, but the repeated point is an enantiomorphous one. ‘A second sequence of 2r is required to place a congruent point at the initial point. Further discussion can be codified in three categories: (These are illustrated in Fig. 4. In these diagrams it is convenient to think of the solid circles as being above the plane of the paper and the open circles an equal distance below.) n odd, i.e., n =4N + 1 (where N isan integer). As pointed out above, when nis odd a sequence of 2m brings a point repeated by fi (upper left of Fig. 4) in reflecting position below the initial point. This reflection relation holds for all points, ‘Thus ff behaves as if the pure rotation axis m had caused the angular repetition, and then a perpendicular mirror, m, had reflected all points. An fold axis and a perpendicular mirror is written ~. 2 The axis 4 can be compared with fi as follows: For every open circle in (Fig. 4, upper left), an open circle occurs on the opposite side of the axis (Fig. 4, lower left). ‘Thus the general relation of points is shown in the left column of Fig. 4. Thus, while the solid and open circles are related by m in the upper part of Fig. 4, they are related by an inversion i in the lower part. Thus f=n-i. n even, 11/2 odd, i.e., N = 4N + 2. In this case, since a is even, the cycle is complete after a rotation of 2x. The enantiomorphous points of fare separated by an interval z. However, the congruent points are separated by an interval 2 so that they are related by a pure rotation period of n/2. Furthermore, since n is even, there is a point at 7, but since »/2 is odd this point is enantio- morphous with the initial point. Thus every point has an opposite enantiomorphous point, ie., a point related by an inversion. Therefore Thus # = rr . a F715 ies the rotoinversion axis is equal to a pure rotation of half the order, combined with an inversion.