Topological aspects of chemically significant polyhedra

This paper considers both static and dynamic properties of chemically significant polyhedra. Static properties of polyhedra consider relationships between the numbers and degrees/sizes of polyhedral vertices, edges, and faces; polyhedral symmetries; and numbers of topologically distinct polyhedra of various types. Dynamic properties of polyhedra involve studies of polyhedral isomerizations from both macroscopic and microscopic points of view. Macroscopic aspects of polyhedral isomerization can be described by graphs called topological representations in which the vertices correspond to different permutational isomers and the edges to single degenerate polyhedral isomerization steps. Such topological representations are presented for isomerizations of polyhedra having five, six, and eight vertices. Microscopic aspects of polyhedral isomerizations arise from consideration of the details of polyhedral topology, such as the topological aspects of diamond-square-diamond processes. In this connection, Gale diagrams are useful for describing isomerizations of five- and six-vertex polyhedra, including the Berry pseudorotation of a trigonal bipyramid through a square pyramid intermediate and the Bailar or Ray and Dutt twists of an octahedron through a trigonal prism intermediate.     

Antisymmetry and stability of water systems. I. Planar cyclic clusters

All topologically distinct configurations of planar cyclic water clusters consisting of three to six molecules are calculated. The symmetry of configurations is analyzed using an additional operation of antisymmetry that changes the directions of all hydrogen bonds. It is concluded that the concept of antisymmetry and the presence of similar in properties but inequivalent “configurations-antipodes” reflects a new fundamental feature of water systems, namely, the internal molecular asymmetry.     

Chemical applications of topology and group theory. 26. The sextuple diamond-square-diamond rearrangement of the icosahedron through a cuboctahedron intermediate

The details of the symmetry factoring of the graphs corresponding to the icosahedron and the cuboctahedron are presented. Such symmetry factoring procedures use the sequence of two-foldC ₂ and threefoldC ₃ elementsC ₂ xC ₂ x CZ x C3 to give disconnected graphs having eigenvalue spectra similar to those of the original polyhedra but with components having only one and two vertices. In addition, the same symmetry factoring sequence is used to determine the eigenvalue spectrum of an intermediate in the sextuple diamond-square process for conversion of the icosahedron to the cuboctahedron.This paper is dedicated to Professor Frank Harary in recognition of his pioneering work in areas of graph theory closely related to chemical problems. For part 25 of this series, see ref. [1].     

Chemical applications of topology and group theory. XXI: Chirality in transitive skeletons and qualitative completeness [1]

This paper unifies the following ideas for the study of chirality polynomials in transitive skeletons: (1) Generalization of chirality to permutation groups not corresponding to three-dimensional symmetry point groups leading to the concepts of signed permutation groups and their signed subgroups; (2) Determination of the total dimension of the chiral ligand partitions through the Frobenius reciprocity theorem; (3) Determination of signed permutation groups, not necessarily corresponding to three-dimensional point groups, of which a given ligand partition is a maximum symmetry chiral ligand partition by the Ruch-Schönhofer partial ordering, thereby allowing the determination of corresponding chirality polynomials depending only upon differences between ligand parameters; such permutation groups having the point group as a signed subgroup relate to qualitative completeness. In the case of transitive permutation groups on four sites, the tetrahedron and polarized square each have only one chiral ligand partition, but the allene and polarized rectangle skeletons each have two chiral ligand partitions related to their being signed subgroups of the tetrahedron and polarized square, respectively. The single transitive permutation group on five sites, the polarized pentagon, has a degenerate chiral ligand partition related to its being a signed subgroup of a metacyclic group with 20 elements. The octahedron has two chiral ligand partitions, both of degree six; a qualitatively complete chirality polynomial is therefore homogeneous of degree six. The cyclopropane (or trigonal prism or trigonal antiprism) skeleton is a signed subgroup of both the octahedron and a twist group of order 36; two of its six chiral ligand partitions come from the octahedron and two more from the twist group. The polarized hexagon is a signed subgroup of the same twist group but not of the octahedron and thus has a different set of six chiral ligand partitions than the cyclopropane skeleton. Two of its six chiral ligand partitions come from the above twist group of order 36 and two more from a signed permutation group of order 48 derived from the P₃[P ₂] wreath product group with a different assignment of positive and negative operations than the octahedron.     

Chemical applications of topology and group theory

This paper characterizesforbidden polyhedra, which are polyhedra with fewer than 9 vertices which cannot be formed using only the 9s,p, andd atomic orbitals. In this connection polyhedra are of particular interest if their symmetry groups are direct product groups of the typeR × C′ _s in whichR is a group containing only proper rotations andC′ _s is eitherC _s orC _i in which the non-identity element is an inversion center or a reflection plane which is called theprimary plane of the groupR ×C′ _s . Using this terminology polyhedra of the following types are shown always to be forbidden polyhedra: (1) Polyhedra having 8 vertices, such direct product symmetry point groups, and either an inversion center or aprimary plane fixing either 0 or 6 vertices; (2) Polyhedra having a 6-fold or higherC _n rotation axis. However, these conditions are not necessary for a polyhedron to be forbidden since in addition to one 7-vertex polyhedron and ten 8-vertex polyhedra satisfying one or both of the above conditions there are two forbiddenC _3v 8-vertex polyhedra which satisfy neither of the above conditions. For part 15 of this series see reference 1.     

The Chirality of Icosahedral Fullerenes: a Comparison of the Tripling (leapfrog), Quadrupling (chamfering), and Septupling (capra) Transformations

Fullerene polyhedra of icosahedral symmetry have the midpoints of their 12 pentagonal faces at the vertices of a macroicosahedron and can be characterized by the patterns of their hexagonal faces on the (triangular) macrofaces of this macroicosahedron. The numbers of the vertices in fullerene polyhedra of icosahedral symmetry satisfy the Goldberg equation v=20(h ²+hk+k ²), where h and k are two integers and 0 <h≥ k≥ 0 and define a two-dimensional Goldberg vector G = (h, k). The known tripling (leapfrog), quadrupling (chamfering), and septupling (capra) transformations correspond to the Goldberg vectors (1, 1), (2, 0), and (2, 1), respectively. The tripling and quadrupling transformations applied to the regular dodecahedron generate achiral fullerene polyhedra with the full I _h point group. However, the septupling transformation destroys the reflection operations of the underlying icosahedron to generate chiral fullerene polyhedra having only the I icosahedral rotational point group. Generalization of the quadrupling transformation leads to the fundamental homologous series of achiral fullerene polyhedra having 20 n ² vertices and Goldberg vectors (n, 0). A related homologous series of likewise achiral fullerene polyhedra having 60 n ² vertices and Goldberg vectors (n, n) is obtained by applying the tripling transformation to regular dodecahedral C₂₀ to give truncated icosahedral C₆₀ followed by the generalized operations (as in the case of quadrupling) for obtaining homologous series of fullerenes. Generalization of the septupling (capra) transformation leads to a homologous series of chiral C_20m fullerenes with the I point group and Goldberg vectors G=(h, 1) where m=h ²+h+1.     

Codes in platonic, Archimedean, Catalan, and related polyhedra: a model for maximum addition patterns in chemical cages

The notion of d-code is extended to general polyhedra by defining maximum sets of vertices with pairwise separation > or =d. Codes are enumerated and classified by symmetry for all regular and semiregular polyhedra and their duals. Partial results are also given for the series of medials of Archimedean polyhedra. In chemistry, d-codes give a model for maximal addition to or substitution in polyhedral frameworks by bulky groups. Some illustrative applications from the chemistry of fullerenes and boranes are described.     

Endohedral nickel and palladium atoms in metal clusters: analogy to endohedral noble gas atoms in fullerenes in polyhedra with five-fold symmetry

Nickel and palladium atoms with their closed-shell d(10) electronic configurations are encapsulated in the icosahedral clusters [Ni@Ni(10)E(2)(CO)(18)](4-)(E = Sb, Bi, Sb[rightward arrow]Ni(CO)(3), CH(3)Sn and n-C(4)H(9)Sn) and the geometrically related pentagonal antiprismatic cluster Pd@Bi(10)(4+) found in Bi(14)PdBr(16). Such endohedral d(10) atoms in pentagonal antiprismatic clusters are donors of zero skeletal electrons and interact only weakly with the atoms in the surrounding polyhedron so that they may be regarded as analogous to endohedral noble gases in fullerenes such as He@C(60). On the other hand, endohedral nickel and palladium atoms in 10- and 11-vertex flattened deltahedral bare metal clusters of group 13 metals without five-fold symmetry, such as Ni@E(10)(10-) found in Na(10)NiE(10)(E = Ga, In) and Pd@Tl(11)(7-) found in A(8)Tl(11)Pd (A = Cs, Rb, K), interact significantly with the cluster atoms, particularly those at the flattened vertices of the deltahedron. The role of endohedral d(10) atoms Ni and Pd in polyhedra with five-fold symmetry as "pseudo-noble-gases" can be related to their positions at the "composite divide" of the "Metallurgists' Periodic Table" proposed by H. E. N. Stone on the basis of alloy systematics as well as the equivalence of the five d orbitals in polyhedra with five-fold symmetry.     

Nonbonding orbitals in fullerenes: nuts and cores in singular polyhedral graphs

A zero eigenvalue in the spectrum of the adjacency matrix of the graph representing an unsaturated carbon framework indicates the presence of a nonbonding pi orbital (NBO). A graph with at least one zero in the spectrum is singular; nonzero entries in the corresponding zero-eigenvalue eigenvector(s) (kernel eigenvectors) identify the core vertices. A nut graph has a single zero in its adjacency spectrum with a corresponding eigenvector for which all vertices lie in the core. Balanced and uniform trivalent (cubic) nut graphs are defined in terms of (-2, +1, +1) patterns of eigenvector entries around all vertices. In balanced nut graphs all vertices have such a pattern up to a scale factor; uniform nut graphs are balanced with scale factor one for every vertex. Nut graphs are rare among small fullerenes (41 of the 10 190 782 fullerene isomers on up to 120 vertices) but common among the small trivalent polyhedra (62 043 of the 398 383 nonbipartite polyhedra on up to 24 vertices). Two constructions are described, one that is conjectured to yield an infinite series of uniform nut fullerenes, and another that is conjectured to yield an infinite series of cubic polyhedral nut graphs. All hypothetical nut fullerenes found so far have some pentagon adjacencies: it is proved that all uniform nut fullerenes must have such adjacencies and that the NBO is totally symmetric in all balanced nut fullerenes. A single electron placed in the NBO of a uniform nut fullerene gives a spin density distribution with the smallest possible (4:1) ratio between most and least populated sites for an NBO. It is observed that, in all nut-fullerene graphs found so far, occupation of the NBO would require the fullerene to carry at least 3 negative charges, whereas in most carbon cages based on small nut cubic polyhedra, the NBO would be the highest occupied molecular orbital (HOMO) for the uncharged system.     

On the enumeration of phase diagrams

The topological isomorphism of polyhedra with trivalent vertices and solid-liquid diagrams of three-component systems allows the problem of constructing the complete set of the topological types of diagrams with a given set of characteristics to be reduced to the problem of the generation of marked cubic graphs, which are the Schlegel projections of polyhedra. The problem of the enumeration of possible topological types of melting diagrams containing M binary and N ternary stoichiometric congruently melting compounds is considered. Relations between the topological characteristics of such diagrams are given. The total number of topologically different types of melting diagrams with one binary and one ternary congruently melting compounds was found to be 64.     

Exocyclobutadieneite

This communication describes a novel structural-type that could be the basis for a potentially new allotrope of C. The novel structural-type is called “exocyclobutadieneite”, and it is thus named for the 1,3-dimethylenecyclobutane generating fragment that the lattice is based upon. It is a 3-,4-connected network consisting of slightly distorted tetrahedral vertices, and slightly distorted pairs of trigonal planar vertices. The lattice can be derived from a known mineral structure called Cooperite (PtS or PdO) by a topologically isomorphic substitution of trigonal planar atom pairs, for square planar vertexes, in the parent Cooperite unit cell. As such, the new pattern bears a distinct counterpoint relationship with its sibling structural-type called the glitter lattice, which has already been described by the authors in several other papers. And whereas glitter is generated by a topologically isomorphic substitution of trigonal planar atom pairs for the square planar vertices in the Cooperite unit cell, in a fashion that extends the unit cell vertically along the crystallographic c-axis, the exocyclobutadieneite structure is generated, in counterpoint, by such an isomorphic substitution that extends the unit cell horizontally along the a- and b-axes. Both the resulting glitter and exocyclobutadieneite structural-types possess AB $_{2}$ stoichiometry (where A is a tetrahedral vertex and B is a trigonal planar vertex) and both occur in the tetragonal symmetry space group P4 $_{2}$ /mmc (#131). The networks differ, however, in the special positions adopted by the vertices in the resultant unit cells, and in their respective topology, as is evidenced by consideration of the Wells point symbols and Schläfli symbols for the 2 tetragonal networks. These differences are illustrated further in the course of the discussion to follow.     

Chemical applications of topology and group Theory. XXIV: Chiralization of chemically significant polyhedra [1]

Permutation group-theoretical methods are used to study the chiralization of achiral polyhedral skeletons with v vertices by successive ligand replacement. Starting from the fully symmetrical ligand partition (), such chiralization processes may be characterized either by the minimum number of ligand replacement steps m, or the minimum number of different kinds of ligandsi, required to destroy all improper rotations. These parameters are trivially related to the lowest degree chiral ligand partition(s) as determined by the subduction of the skeleton point group G into the corresponding symmetric groupS by the procedure of Ruch and Schönhofer. Two different chiralization pathways with different values ofm andi are found for the octahedron, cube, hexagonal bipyramid, and icosahedron. Many less symmetrical chemically significant polyhedra have the degree 2 ligand partition (v - 2, 2) as the lowest degree chiral ligand partition and thus have only one chiralization pathway. Such polyhedra include the bicapped tetrahedron, trigonal prism, capped octahedron, bisdisphenoid, square antiprism, 4, 4, 4-tricapped trigonal prism, 4-capped square antiprism, 4,4-bicapped square antiprism, and the cuboctahedron.     

Eigenvalue spectra and geometric transformation of polyhedra

Starting from one fullerene, the three geometric transformations, cap, face dual and edge dual produce series of carbon clusters and deltahedra. The geometric relation between these polyhedra implies that their topological matrices and eigenvalue spectra must be relative. We have developed a matrix algebra method to research some polyhedra with high symmetry and one kind, two or three kinds of equivalent vertices such as C₆₀(I_h), resolve their exact eigenvalues, and proved this point.     

Some aspects of the symmetry and topology of possible carbon allotrope structures

Elemental carbon has recently been shown to form molecular polyhedral allotropes known as fullerenes in addition to the familiar graphite and diamond known since antiquity. Such fullerenes contain polyhedral carbon cages in which all vertices have degree 3 and all faces are either pentagons or hexagons. All known fullerenes are found to satisfy the isolated pentagon rule (IPR) in which all pentagonal faces are completely surrounded by hexagons so that no two pentagonal faces share an edge. The smallest fullerene structures satisfying the IPR are the known truncated icosahedral C₆₀ of I _h symmetry and ellipsoidal C₇₀ of D _5h symmetry. The multiple IPR isomers of families of larger fullerenes such as C₇₆, C₇₈, C₈₂ and C₈₄ can be classified into families related by the so-called pyracylene transformation based on the motion of two carbon atoms in a pyracylene unit containing two linked pentagons separated by two hexagons. Larger fullerenes with 3ν vertices can be generated from smaller fullerenes with ν vertices through a so‐called leapfrog transformation consisting of omnicapping followed by dualization. The energy levels of the bonding molecular orbitals of fullerenes having icosahedral symmetry and 60n ² carbon atoms can be approximated by spherical harmonics. If fullerenes are regarded as constructed from carbon networks of positive curvature, the corresponding carbon allotropes constructed from carbon networks of negative curvature are the polymeric schwarzites. The negative curvature in schwarzites is introduced through heptagons or octagons of carbon atoms and the schwarzites are constructed by placing such carbon networks on minimal surfaces with negative Gaussian curvature, particularly the so-called P and D surfaces with local cubic symmetry. The smallest unit cell of a viable schwarzite structure having only hexagons and heptagons contains 168 carbon atoms and is constructed by applying a leapfrog transformation to a genus 3 figure containing 24 heptagons and 56 vertices described by the German mathematician Klein in the 19th century analogous to the construction of the C₆₀ fullerene truncated icosahedron by applying a leapfrog transformation to the regular dodecahedron. Although this C₁₆₈ schwarzite unit cell has local O _h point group symmetry based on the cubic lattice of the D or P surface, its larger permutational symmetry group is the PSL(2,7) group of order 168 analogous to the icosahedral pure rotation group, I, of order 60 of the C₆₀ fullerene considered as the isomorphous PSL(2,5) group. The schwarzites, which are still unknown experimentally, are predicted to be unusually low density forms of elemental carbon because of the pores generated by the infinite periodicity in three dimensions of the underlying minimal surfaces. This revised version was published online in July 2006 with corrections to the Cover Date.     

Symmetry properties of chemical graphs. IX. The valence tautomerism in the P73− skeleton

We investigate the intramolecular rearrangement of the P₇^3? ion in which P? P bonds are continuously broken and reestablished resulting in a dynamic structure with all seven phosphorus atoms being chemically equivalent. This leads to 7/3 = 1680 valence tautomers. The P₇^3? ion is a typical fluctuating molecule, distantly related to C₁₀H₁₀ bullvalene. In order to determine the symmetry of the process we first consider the problem of the construction of the rearrangement graph. Initial steps toward this goal are illustrated, and additional characterizations of the graph are derived after its construction by computer. A number of properties of this rather complicated graph having 1680 vertices and 2520 edges are discussed: in particular the occurrence of various cycles, the maximal distance between vertices, the count of neighbors at different distance, and finally its symmetry.     

The enumeration of electron-rich and electron-poor polyhedral clusters

We present as dual processes the capping of closed triangulated polyhedra with apical atoms and the making of holes in such polyhedra either by puncture of the surface or by excision of atoms and their edges. These processes are shown to generate stable chemical species containing respectively less or more than 2n + 2 skeletal electrons. The former species are designated as electron-poor whereas the latter are called electron-rich. Pólya's enumeration method is used to enumerate the distinct ways of capping and excising the closed, triangulated polyhedra to yield systems containing from four to twelve vertices. For the enumeration of cappings the appropriate cycle index is that of the dual of the polyhedron being capped, whilst for the enumeration of the excisions the cycle index is that of the polyhedron being excised.     

Symmetry properties of graphs of interest in chemistry. II. Desargues–Levi graph

The Desargues–Levi graph represents important chemical transformations: (1) isomerization routes for some carbonium ion rearrangements, (2) isomerization of trigonal bipyramidal structures, and (3) some pseudorotations of octahedral complexes. The symmetry properties of this graph have not been fully investigated in the past. Using the concept of the smallest binary code, all permutations which form the symmetry operations in the graph are registered. The resulting symmetry group can be represented as the direct product of S₅ (the full symmetric permutation group on five objects) and C_i (the inversion in the center). There are 14 classes belonging to the following partitionings: 1²⁰(1), 1⁸2⁶(1), 1⁴2⁸(1), 1²3⁶(1), 1²3²6²(1), 2 6³(2), 2²4⁴(2), 2¹⁰(3), 5⁴(1), and 10²(1). The total of 240 symmetry operations are distributed among the above 14 classes as follows: 1, 10, 15, 20, 20, 20, 20, 30, 30, 15, 10, 1, 24, and 24, respectively. Since partitioning cannot uniquely characterize a class, it is suggested that the distance between vertices in a cycle be introduced as an additional parameter to discriminate among classes having identical partitioning. Also, a suggestion to a generalization of the Mulliken notation for irreducible representations of the point molecular groups valid for more versatile symmetry groups of graphs is indicated.     

The dimensionality and topology of chemical bonding manifolds in metal clusters and related compounds

The chemical bonding manifolds in metal cluster skeletons (as well as in skeletons of clusters of other elements such as boron or carbon) may be classified according to their dimensionalities and their chemical homeomorphism to various geometric structures. The skeletal bonding manifolds of discrete metal cluster polyhedra may be either one-dimensional edge-localized or three-dimensional globally delocalized, although two-dimensional face-localized skeletal bonding manifolds are possible in a few cases. Electron precise globally delocalized metal cluster polyhedra withv vertices have 2v + 2 skeletal electrons and form deltahedra with no tetrahedral chambers having total skeletal bonding manifolds chemically homeomorphic to a closed ball. Electron-rich metal cluster polyhedra withv vertices have more than 2v + 2 skeletal electrons and form polyhedra with one or more non-triangular faces, whereas electron-poor metal cluster polyhedra withv vertices have less than 2v + 2 skeletal electrons and form deltahedra with one or more tetrahedral chambers. Fusion of metal cluster octahedra by sharing (triangular) faces forms three-dimensional analogues of polycyclic aromatic hydrocarbons such as naphthalene, anthracene, and perinaphthenide. Fusion of metal cluster octahedra by sharing edges can be extended infinitely into one and two dimensions forming chains (e.g. Gd₂Cl₃) and sheets (e.g. ZrCl), respectively. Infinite extension of such fusion of metal cluster octahedra into all three dimensions leads to bulk metal structures. Unusual anionic platinum carbonyl clusters can be contructed from stacks of Pt₃ triangles or Pt₅ pentagons. The resulting platinum polyhedra appear to exhibit edge-localized bonding, supplemented by unusual types of delocalized bonding at the top and the bottom of the stacks. Superconducting ternary molybdenum chalcogenides and lanthanide rhodium borides consist of infinite lattices of electronically linked edge-localized Mo₆ octahedra or Rh₄ tetrahedra, leading naturally to the idea of porous delocalization in superconducting materials.     

Nonconvex polyhedra by repeated truncation of semiregular polyhedra

Some of the semiregular (Archimedean) polyhedra (1–13 in Table 1) afford on truncation polyhedra that contain vertices where the sum of planar degrees for the faces which meet at those vertices is equal to (for 17, 18, and 23 in Table 3) or higher than 360° (21, 22, 24–26 in Table 3). Therefore such polyhedra are nonconvex.