Cosmological Topology In Paris 1998 Topologie Cosmologique A . - CORE

1y ago
14 Views
2 Downloads
649.96 KB
37 Pages
Last View : 14d ago
Last Download : 3m ago
Upload by : Joao Adcock
Transcription

Cosmological Topology in Paris 1998, 14 December 1998, Obs. de Paris, eds V. Blanlœil & B.F. RoukemaCosmological Topology in Paris 1998Topologie cosmologique à Paris 1998Editors: Vincent Blanlœil1 & Boudewijn F. Roukema21Institut de recherche mathématique avancée, Université Louis Pasteur et CNRS,7 rue René-Descartes, F-67084 Strasbourg Cedex, France2Inter-University Centre for Astronomy and Astrophysics,Post Bag 4, Ganeshkhind, Pune, 411 007, India (boud@iucaa.ernet.in)blanloei@math.u-strasbg.fr, boud.roukema@obspm.frLe 11 octobre 2000AbstractQuel est, ou pourrait être, la topologie globale de la partie spatiale de l’Univers ? L’Univers entier(précisément, l’hypersurface spatiale de celui-ci) est-il observable ? Les mathématiciens, les physiciens et lescosmologistes observationnels ont des approches différentes pour aborder ces questions qui restent ouvertes.What is, or could be, the global topology of spatial sections of the Universe? Is the entire Universe (spatialhypersurface thereof ) observable ? Mathematicians, physicists and observational cosmologists have differentstrategies to approaching these questions which are not yet fully answered.Un atelier international d’une journée a eu lieu à Paris pour les chercheurs de ces domaines complémentaires,pour introduire leurs sujets et pour présenter des revues autant que les derniers résultats de leurs travaux.Mathématiciens, astronomes et physiciens y ont participés. L’atelier a été organisé dans le cadre du PNC(programme national de cosmologie).Merci à tous les participants, à ceux qui ont présenté leur travaux ainsi qu’à ceux qui ont écouté et participéaux débats. Merci aussi à l’Observatoire de Paris, l’Institut d’Astrophysique de Paris et le PNC pour leur aide.Ce compte rendu est constitué de trois articles théoriques : celui de Ratcliffe et Tschantz sur un objetmathématique utile pour la gravité quantique, l’instanton gravitationnel; une exploration des liens éventuelsentre une constante cosmologique non-nulle et la topologie cosmique par Lachièze-Rey; et e Costa et Fagundesont présenté un potentiel V (φ) qui pourrait donner naissance à un univers multi-connexe, hyperbolique etcompact; et de quatres articles observationnels : une revue par Roukema; un rappel par Wichoski que vu lesdifficultés pratiques des méthodes statistiques à 3-D et à 2-D, la recherche des images topologiques de notrepropre Galaxie ne doit pas être oublié; un éclairci par Inoue sur le vif débat actuellement en cours concernantles analyses des données du fond diffus cosmique de COBE pour les modèles hyperboliques et compactes; et unrésumé de la méthode de la reconnaissance des schémas des taches par Levin & Heard.Nul peut prévoir en ce moment quels éléments théoriques et observationnels seront les plus importants,même si chacun de nous a ses propres intuitions. . .Cet atelier de décembre 1998 a suivi le premier du septembre 1997 à Cleveland, et comme il vient d’y avoirdeux séances parallèles sur la topologie cosmique et les 3-variétés hyperboliques à la rèunion Marcel GrossmannIX à Roma en juillet 2000, la continuation d’un développement rapide et soutenu est promise. . .1

A one-day international workshop, supported by the PNC (Programme National de la Cosmologie), was heldin Paris for members of the different disciplines to introduce their respective subjects and present both reviewsand up-to-date research methods and results. Mathematicians, astronomers and physicists were welcomed.Thank you to all the participants, both to those who presented work and those who listened and participatedin the discussion. Thank you also to the Observatoire de Paris, the Institut d’Astrophysique de Paris, and thePNC.In this volume, we have three theoretical articles: that of Ratcliffe & Tschantz about a mathematical objectwhich should be useful for quantum gravity, the gravitational instanton; an exploration of possible links betweena non-zero cosmological constant and cosmic topology by Lachièze-Rey; and e Costa & Fagundes presented a potential V (φ) which could give birth to a multiply connected, compact hyperbolic universe; and four observationalarticles: a review by Roukema; a reminder by Wichoski that given the practical difficulties of statistical methodsin 3-D and in 2-D, the search for topological images of our own Galaxy should not be forgotten; some very interesting comments by Inoue on the lively debate presently underway regarding compact hyperbolic model analysesof the COBE cosmic microwave background data; and a summary of the method of spot pattern recognition byLevin & Heard.Noone can predict which theoretical and observational elements will be the most important, even if each ofus has his or her own intuition. . .This Dec 1998 workshop followed the first in Cleveland in Sep 1997, and as two parallel sessions on cosmictopology and hyperbolic 3-manifolds have just taken place at the Marcel Grossmann IX meeting in Roma inJuly 2000, continued rapid development and excitement in the field is the safest prediction to make for the nearfuture. . .Comité d’organisation et scientifique/ Organising and Scientific Committee:Boud Roukema, Vincent Blanloeil, Jean-Pierre Luminet, Gary MamonSome additional useful links are provided at the electronic site of the proceedings html.Participants:Bajtlik, Stanislaw, Nicolas Copernicus Astronomical Center, Warsaw, PolandBlanlœil, Vincent, IRMA, Univ de Strasbourg I, FranceBois, Eric, Observatoire de Bordeaux, FranceCélérier, Marie-Noëlle, DARC, Observatoire de Paris-Meudon, FranceFagundes, Helio, IFT, Univ. Estadual Paulista, BrazilGausmann, Evelise, Instituto de Fisica Teorica, Univ. Estadual Paulista, BrazilInoue, Kaiki Taro, Yukawa Institute for Theoretical Physics, Kyoto, JapanLachièze-Rey, Marc, CEA, Saclay, FranceLehoucq, Roland, Service d’Astrophysique de Saclay, FranceLevin, Janna, Astronomy Centre, Univ Sussex, United KingdomLuminet, Jean-Pierre, DARC, Observatoire de Paris-Meudon, FranceMadore, John, Université de Paris Sud, FranceMamon, Gary, IAP, Paris, FranceMarty, Philippe, Institut d’Astrophysique Spatiale, Orsay, FrancePierre, Marguerite, Service d’Astrophysique du CE Saclay, FrancePogosyan, Dmitri, CITA, Toronto, CanadaRatcliffe, John G., Vanderbilt University, Tennessee, USARoukema, Boud, IUCAA, Pune, IndiaUzan, Jean-Phillippe, Univ of Geneva, SwitzerlandVan Waerbeke, Ludovic, CITA, Toronto, CanadaWeeks, Jeff, Canton NY, USAWichoski, Ubi, Brown University, USA2

Programme:Chair: Luminet09h00: Jean-Pierre Luminet (DARC, Observatoire de Paris - Meudon)Cosmological Topology: Opening Remarks09h05: Jeff Weeks (Canton NY, USA)(1) Deducing topology from the CMB; (2) The structure of closed hyperbolic 3-manifolds09h55: Dmitri Pogosyan (CITA, Toronto)Some work on hyperbolic 3-manifolds and COBE data10h00: John Madore (Univ Paris Sud)Topology at the Planck Length10h30-11h00: coffee break11h00: John G. Ratcliffe (Vanderbilt University)Gravitational Instantons of Constant Curvature11h30: Marc Lachize-Rey (CEA, Saclay)The Physics of Cosmic Topology12h00: Boudewijn Roukema (IUCAA, Pune)Observational Methods, Constraints and Candidates12h30: Marguerite Pierre (Service d’Astrophysique, CEA, Saclay)X-ray Cosmic Topology13h00-14h00: LunchChair: Fagundes14h00: Helio Fagundes (IFT, Univ Estadual Paulista)Creation of a Closed Hyperbolic Universe14h30: Ubi Wichoski (Brown Univ, USA)Topological Images of the Galaxy15h00: Jean-Phillippe Uzan (Univ of Geneva)Cosmic Crystallography: the Hyperbolic Case15h30-16h00: coffee break16h00: Kaiki Taro Inoue (Yukawa Institute for Theoretical Physics)CMB anisotropy in a compact hyperbolic universe16h30: Janna Levin (Astronomy Centre, Univ Sussex)How the Universe got its Spots17h00: Stanislaw Bajtlik (Copernicus Center, Warsaw)Applying Cosmo-topology: Galaxy Transverse VelocitiesModerator: Roukema17h30: (all participants)General Discussion18h00: Close3

Contents1 Gravitational Instantons of Constant Curvature,1.1 Introduction . . . . . . . . . . . . . . . . . . . . . .1.2 Spherical and Flat Gravitational Instantons . . . .1.3 Hyperbolic Gravitational Instantons . . . . . . . .1.4 Noncompact Hyperbolic Gravitational Instantons .John G. Ratcliffe,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 Topology, the vacuum and the cosmological constant,2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . .2.1.1 Topology and cosmology . . . . . . . . . . . . . .2.1.2 Characteristic lengths . . . . . . . . . . . . . . .2.2 Topology and vacuum energy . . . . . . . . . . . . . . .2.2.1 Quantum fields in non Minkowskian space-time .2.2.2 Topological Casimir effect . . . . . . . . . . . . .2.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . .3 Creation of a Closed Hyperbolic Universe,.Steven T. Tschantz 5. . . . . . . . . . . . . 5. . . . . . . . . . . . . 6. . . . . . . . . . . . . 7. . . . . . . . . . . . . 11Marc Lachièze-Rey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .S. S. e Costa and H. V.Fagundes4 Observational Methods, Constraints and Candidates, Boudewijn4.1 Introduction: a spectrum of differing observational approaches . . . .4.1.1 3-D methods . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.1.2 2-D methods . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.1.3 Consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.2 Comparison of different approaches . . . . . . . . . . . . . . . . . . . .4.3 Candidates versus constraints . . . . . . . . . . . . . . . . . . . . . . .4.4 Conclusion and suggestions for the future . . . . . . . . . . . . . . . .F. Roukema. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .171717171818191921.22222223232324245 Topological Images of the Galaxy, U. F. Wichoski275.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275.2 The search for topological images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 Comments on the constraints on the topology of compact low-density universes,Taro Inoue7 Topological Pattern Formation,KaikiJanna Levin and Imogen Heard3134List of Figures123456789A fundamental domain for the half-twisted 3-torus . . . . . . . . . . . . . . . . . . . . . . . . . .A chain of adjacent sides of a regular 120-cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . .A fundamental domain for the Davis manifold cross-section . . . . . . . . . . . . . . . . . . . . .A hyperbolic right-angled rhombic dodecahedron . . . . . . . . . . . . . . . . . . . . . . . . . . .Potential V (φ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Plots of likelihoods of Thurston models relative to Einstein-de Sitter models. . . . . . . . . . . .The correlation of every point on the sky with its opposite in the finite Thurston manifold. . . .The correlation of one point on the sky with the rest of the sphere in the Thurston space. Thereis a tri-fold symmetry apparent in the middle of the sphere. . . . . . . . . . . . . . . . . . . . .The back of fig. ?. The tri-fold symmetry is again apparent with the three-leaf pattern in themiddle of the sphere. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47910122132353536

List of Tables123Cross-sections of the Ratcliffe-Tschantz hyperbolic 4-manifolds . . . . . . . . . . . . . . . . . . . 13The orientable, Ratcliffe-Tschantz, hyperbolic 4-manifolds . . . . . . . . . . . . . . . . . . . . . . 14Summary of methods and observational results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

Cosmological Topology in Paris 1998, 14 December 1998, Obs. de Paris, eds V. Blanlœil & B.F. Roukema1Gravitational Instantons of Constant CurvatureJohn G. Ratcliffe, Steven T. TschantzDepartment of Mathematics, Vanderbilt University,Nashville, Tennessee 37240, U.S.A.1999 Mathematics Subject Classification. Primary 51M20, 53C25, 57M50, 83F05Key words and phrases. Flat manifold, hyperbolic manifold, gravitational instanton, totally geodesichypersurface, 24-cell, 120-cellAbstracta closed orientable Riemannian 4-manifold M 2MRcalled the gravitational instanton of the real tunnelinggeometry. The instanton M admits a reflection map θthat is an orientation reversing involution which fixesthe totally geodesic submanifold Σ and permutes thetwo portions MR . According to Gibbons, the involution θ plays a crucial role in the quantum theorybecause it allows one to formulate the requirement of“reflection positivity”.In the standard example of a real tunneling geometry, the instanton M is the unit 4-sphere S 4 and Σis the unit 3-sphere S 3 thought of as the equator ofS 4 . The 3-sphere S 3 is the simplest model of the universe that is isotropic and the 4-sphere S 4 is the onlygravitational instanton that is isotropic.The Riemannian manifolds that are locallyisotropic are the manifolds of constant sectional curvature k. For simplicity, a Riemannian manifold of constant sectional curvature is usually normalized to havecurvature k 1, 0, or 1. A Riemannian manifold ofconstant sectional curvature k 1, 0, or 1 is calleda hyperbolic, Euclidean, or spherical manifold, respectively. Euclidean manifolds are also called flat manifolds. We shall assume that a hyperbolic, Euclidean, orspherical manifold is connected and complete. We shallalso assume that a manifold does not have a boundary unless otherwise stated. Then a hyperbolic, Euclidean, or spherical n-manifold is isometric to the orbitspace X/Γ of a freely acting discrete group of isometries Γ of hyperbolic, Euclidean, or spherical n-spaceX H n , E n , or S n , respectively. A discrete group Γof isometries of H n or E n acts freely on H n or E n ifand only if Γ is torsion-free.In this paper, we classify all closed flat 4-manifoldsthat have a reflective symmetry along a separating totally geodesic hypersurface. We also give examples ofsmall volume hyperbolic 4-manifolds that have a reflective symmetry along a separating totally geodesichypersurface. Our examples are constructed by gluingtogether polytopes in hyperbolic 4-space.1.1IntroductionIn a recent paper [3], G.W. Gibbons mentioned that theexamples of minimum volume hyperbolic 4-manifoldsdescribed in our paper [7] might have applications incosmology. In this paper, we elaborate on our examples and introduce some new examples. In particular,we construct an example which answers in the affirmative the following question posed by G.W. Gibbonsat the Cleveland Cosmology-Topology Workshop. Canone find a closed hyperbolic 4-manifold with a (connected) totally geodesic hypersurface that separates?In this paper a hypersurface is a codimension one submanifold. We begin by describing the geometric setupof real tunneling geometries.According to Gibbons [3], current models of thequantum origin of the universe begin with a real tunneling geometry, that is, a solution of the classical Einsteinequations which consists of a Riemannian 4-manifoldMR and a Lorentzian 4-manifold ML joined across atotally geodesic spacelike hypersurface Σ which servesas an initial Cauchy surface for the Lorentzian spacetime ML . In cosmology, Σ is taken to be closed, thatis, compact without boundary, and in accordance withthe No Boundary Proposal one usually takes MR tobe connected, orientable, and compact with boundaryequal to Σ.Given this setup one may pass to the double 2MR MR MR by joining two copies of MR across Σ. This is6

1.2Spherical and Flat Gravitational ric, over a connected, closed, orientable, flat (n 1)manifold N .InstantonsProof. Assume that M is flat. Then M is complete,The first observation to make about real tunneling geometries is that the Euler characteristic of a gravita- since M is compact. Hence we may assume M E n /Γwhere Γ is a freely acting discrete group of orientationtional instanton M is even, sincepreserving isometries of Euclidean n-space E n . Letχ(M ) χ(MR ) χ(MR ) χ(Σ) 2χ(MR ).φ : E n E n /Γ be the quotient map. Then φ is auniversal covering projection. Now φ 1 (Σ) is a totallyThere are only two spherical 4-manifolds, namely S 4 geodesic hypersurface of E n , since φ is a local isometry.and elliptic 4-space P 4 (real projective 4-space). Spher- Therefore φ 1 (Σ) is a disjoint union of hyperplanes ofical 4-space S 4 is the prototype for a gravitational in- E n .stanton whereas P 4 is not a gravitational instanton,The inclusion map ι : Σ M induces an injectionsince χ(P 4 ) 1.ι : π1 (Σ) π1 (M ) on fundamental groups, since eachWe next classify Euclidean (or flat) gravitational in- component of φ 1 (Σ) is simply connected. Choose astantons. In order to state our classification, we need component P of φ 1 (Σ). We may identify π (M ) with1to recall the definition of a twisted I-bundle. Let N be Γ and π (Σ) with the subgroup C of Γ that leaves P1a nonorientable n-manifold. Then N has an orientable invariant. The group Γ has cohomological dimension ndouble cover Ñ and there is a fixed point free, orien- and the group C has cohomological dimension n 1,tation reversing involution σ of Ñ such that N is the since M and Σ are aspherical manifolds. Now everyquotient space of Ñ obtained by identifying σ(x) with subgroup of finite index of an n-dimensional group isx for each point x of Ñ . Let I be a closed interval n-dimensional. Therefore the index of C in Γ is infinite.[ b, b] with b 0. Then σ extends to a fixed point The number of components of φ 1 (Σ) is the index of Cfree, orientation preserving involution τ of Ñ I de- in Γ. Therefore φ 1 (Σ) is a disjoint union of an infinitefined by τ (x, t) (σ(x), t). The twisted I-bundle B number of hyperplanes of E n . These hyperplanes areover N is the quotient space of Ñ I obtained by iden- all parallel, since any two nonparallel hyperplanes oftifying τ (x, t) with (x, t) for each point (x, t) of Ñ I. E n intersect.Then B is an orientable (n 1)-manifold with boundColor MR white and the rest of M black. Theary B Ñ and Ñ I is a double cover of B. Note boundary between the white and black regions of M isthat B is a fiber bundle over N with fiber I. If N Σ. Lift this coloring to E n via φ : E n M by coloris a Riemannian manifold, then Ñ has a Riemannian ing φ 1 (M ) white and the rest of E n black. Then theRmetric so that the double covering from Ñ to N is a lo- regions between the hyperplanes of φ 1 (Σ) are coloredcal isometry. We give I the standard Euclidean metric alternately white and black, since the coloring mustand Ñ I the product Riemannian metric. Then τ is change at each hyperplane of φ 1 (Σ). Let R be thean isometry of Ñ I and so B inherits a Riemannian component of φ 1 (M ) containing P . Then R is theRmetric, which we called the twisted product Rieman- closed white region bounded by two adjacent hypernian metric, so that the double covering from Ñ I planes P and Q of φ 1 (Σ). Hence R is a Cartesianto B is a local isometry. It is worth mentioning that product P I where I is a closed line segment runningtwisted I-bundles occur naturally in topology, since a perpendicularly from P to Q. Therefore R is simplyclosed regular neighborhood of a nonorientable hyper- connected, and so the inclusion map κ : M MRsurface N of an orientable manifold M is a twisted induces an injection κ : π (M ) π (M ) on funda 1R1I-bundle over N . We are now ready to state our clas- mental groups. Hence we may identify π (M ) with1Rsification theorem for flat gravitational instantons. We the subgroup A of Γ that leaves R invariant.shall state our theorem in arbitrary dimensions, sinceLet H be the hyperplane of E n midway betweenthe proof works in all dimensions.P and Q. Then H cuts I at its midpoint. Now eachelement of A maps I to a line segment running perpenTheorem 1 Let M be a connected, closed, orientable,dicularly from P to Q, since each element of A leavesRiemannian n-manifold that is obtained by doubling R P Q invariant and preserves perpendiculara Riemannian n-manifold MR with a totally geodesicity. Therefore A leaves H invariant. Hence H/A is aboundary Σ. Then (1) M is flat and Σ is connected ifhypersurface of MR R/A.and only if MR is a twisted I-bundle, with the twistedAssume first that Σ is disconnected. Then no eleproduct Riemannian metric, over a connected, closed,ment of A interchanges P and Q, and so A leaves bothnonorientable, flat (n 1)-manifold N ; and, (2) M isP and Q invariant. Hence A preserves the productflat and Σ is disconnected if and only if MR is a prodstructure R H I. Therefore R/A H/A I. Nowuct I-bundle, N I with the product Riemannian metsince A preserves the orientation of E n and preserves7

both sides of H in R, we deduce that A preserves orientation on H. Therefore N H/A is an orientable manifold and MR is the product I-bundle, N I, with theproduct Riemannian metric. Moreover, N is a closedmanifold, since MR and N are compact.Now assume that Σ is connected. Then there isan element α of A that interchanges P and Q. LetB be the subgroup of A that leaves both P and Qinvariant. Then B is a subgroup of A of index two.Now B preserves the product structure R H I.Therefore R/B H/B I. Now since A preservesthe orientation of E n , with B preserving both sidesof H in R and α interchanging both sides of H in R,we deduce that B preserves orientation on H and αreverses orientation on H. Therefore Ñ H/B is theorientable double cover of the nonorientable manifoldN H/A. Now A/B acts on Ñ I so that Ñ I/(A/B) R/A is a twisted I-bundle over H/A. ThusMR is a twisted I-bundle, with the twisted productRiemannian metric, over the nonorientable manifoldN . Now MR is compact and so its double cover Ñ Iis compact. Hence Ñ and N are compact, and so N isa closed manifold.Conversely, if MR is an I-bundle over a flat (n 1)manifold N , with either the product or twisted productRiemannian metric, and a totally geodesic boundary,then obviously M 2MR is flat.zyxFigure 1: A fundamental domain for the half-twisted3-torusN by Theorem 1. Here Σ is the orientable double coverof N . According to Theorem 3.5.9 of Wolf [8], if N isaffinely equivalent to N1 or N2 , then Σ is a flat 3torus, whereas if N is affinely equivalent to N3 or N4 ,then Σ is affinely equivalent to O2 . Thus only the firsttwo affine equivalence types of closed orientable flat3-manifolds are possible initial hypersurfaces for thecreation of a connected Lorentzian universe from a flatgravitational instanton.We call the closed orientable flat 3-manifold O2a half-twisted 3-torus because O2 can be constructedfrom a rectangular box, centered at the origin in E 3with sides parallel to the coordinate planes, by identifying opposite pairs of vertical sides by translations andidentifying the top and bottom sides by a half-twist inthe z-axis. See Figure 1. The first homology group ofO2 is ZZ ZZ 2 ZZ 2 . Therefore O2 is not topologicallyequivalent to a 3-torus. It is worth noting that O2 isdouble covered by a 3-torus. This is easy to see bystacking two of the boxes defining O2 on top of eachother.There are exactly 10 closed flat 3-manifolds up toaffine equivalence. Six of these manifolds are orientableand four are nonorientable. We shall denote the orientable manifolds by O1 , O2 , . . . , O6 and the nonorientable manifolds by N1 , N2 , N3 , N4 . As a referencefor closed flat 3-manifolds, see Wolf [8]. We shall takethe same ordering of the closed flat 3-manifolds as inWolf [8]. In particular, the 3-manifold O1 is a flat 3torus.Let M be a flat gravitational instanton. Then M isa connected, closed, orientable, flat 4-manifold that isobtained by doubling a flat Riemannian 4-manifold MRwith totally geodesic boundary Σ. Assume first thatΣ is disconnected. Then MR is a product I-bundleO I, with the product Riemannian metric, over aclosed orientable flat 3-manifold O by Theorem 1. Thisimplies that MR is just a straight tube with openingand closing end isometric to O. Here Σ MR isthe disjoint union of two isometric copies of O. Onecan interpret the geometry of MR as leading to thebirth of disjoint identical twin Lorentzian universes or,by reversing the arrow of time in one of the universes,as a collapse and subsequent rebirth of a Lorentzianuniverse.Assume now that Σ is connected. Then MR is atwisted I-bundle, with the twisted product Riemannian metric, over a closed nonorientable flat 3-manifold1.3Hyperbolic Gravitational InstantonsA hyperbolic gravitational instanton is a gravitationalinstanton that is a hyperbolic manifold. Thus a hyperbolic gravitational instanton is a closed, orientablehyperbolic 4-manifold, M , with a separating, totallygeodesic, orientable, hypersurface Σ which is the set offixed points of an orientation reversing isometric involution of M . As a reference for hyperbolic manifolds,see Ratcliffe [6]. Cosmologists are interested in small8

volume hyperbolic gravitational instantons because theprobability of creation of a hyperbolic gravitational instanton increases with decreasing volume. The volumeof a hyperbolic 4-manifold M of finite volume is proportional to its Euler characteristic χ(M ) and so the Eulercharacteristic is an effective measure of the volume ofa hyperbolic gravitational instanton. The closed orientable hyperbolic 4-manifold of least known volumeis the Davis hyperbolic 4-manifold [2], which has Eulercharacteristic 26.In his talk at the Cleveland Cosmology-TopologyWorkshop, G.W. Gibbons asked the question:be used to tessellate hyperbolic 4-space with 3, 4, or5 of the 120-cells fitted around each ridge respectively.The set of isometries of hyperbolic 4-space preservingone of these tessellations will be a discrete group; thequotient of hyperbolic 4-space under the action of atorsion-free subgroup of finite index in this group willgive a closed hyperbolic 4-manifold which can be realized by gluing together some number of copies of thecorresponding regular 120-cell. The Euler characteristic of the hyperbolic orbifold determined by a regular120-cell, with dihedral angle 2π/3, π/2, and 2π/5 is 1,17/2, and 26, respectively; their volumes are proportional to their Euler characteristic.A purely combinatorial search for manifolds basedon gluing one or two of the 120-cells with dihedral angle 2π/3 is essentially intractable. Searches for sidepairings meeting some simple restrictions have failed touncover small volume hyperbolic 4-manifolds based onthis smallest regular 120-cell. A manifold based on the120-cell with dihedral angle π/2 can only result from agluing of an even number of 120-cells. In fact, we haveconstructed two different manifolds by gluing just tworight-angled 120-cells. These have Euler characteristic17, are nonorientable, and do not seem to have the kindof totally geodesic hypersurfaces desired.Let P be a regular hyperbolic 120-cell with dihedral angles 2π/5. For simplicity, realize P in the conformal ball model of hyperbolic 4-space with center atthe origin and aligned so the center of a side lies alongeach of the coordinate axes, i.e., there are centers ofsides having coordinates (x1 , x2 , x3 , x4 ) ( r, 0, 0, 0),(0, r, 0, 0), (0, 0, r, 0), and (0, 0, 0, r) for an appropriate r. Then the four coordinate hyperplanes of E 4 ,given by xi 0, for i 1, 2, 3, 4, are planes of symmetry of P . A side-pairing map for P can be describedas a symmetry of P taking a side S to another side S 0followed by reflection in the side S 0 . Thus side-pairingmaps will be determined by the orthogonal transformations of E 4 that are symmetries of P .The side of P lying along the positive x4 -axis willbe referred to as the side at the north pole, the side onthe negative x4 -axis will be referred to as the side atthe south pole, and the hyperplane with x4 0, willbe referred to as the equatorial plane of P . There are30 sides of P centered on the equatorial plane and 12ridges lie entirely in this hyperplane. The intersectionof the equatorial plane with P is a truncated, hyperbolic, ultra-ideal triacontahedron.A triacontahedron is a quasiregular convex polyhedron with 30 congruent rhombic sides. As a referencefor the geometry of a triacontahedron, see Coxeter [1].In a triacontahedron five rhombi meet at each vertexwith acute angles and three rhombi meet at each vertex with obtuse angles. A hyperbolic ultra-ideal tria-Can one find a closed hyperbolic 4-manifoldwith a totally geodesic two-sided hypersurface that separates?It is well known that there are closed hyperbolic 4manifolds with two-sided totally geodesic hypersurfaces. As pointed out by Gibbons [3], if a two-sidedhypersurface Σ of a manifold M does not separate,then M has a double cover with a separating hypersurface consisting of two disjoint copies of Σ. Thusan affirmative answer to Gibbon’s question has beenknown for some time with Σ disconnected. See for example, §2.8.C of [4]. However, in Gibbon’s paper [3], heasks whether the creation of a single universe is possible from a hyperbolic gravitational instanton. Thus amore interesting question (and probably what Gibbonsreally wanted to ask at the workshop) is the question:Can one find a closed hyperbolic 4-manifoldwith a connected totally geodesic twosided hypersurface that separates?We will answer this question in the affirmative byconstructing a hyperbolic gravitational instanton Mwith a connected initial hypersurface Σ. The manifold M is most easily understood as the orientabledouble cover of a manifold specified by a side-pairingof the same regular

Cosmological Topology in Paris 1998, 14 December 1998, Obs. de Paris, eds V. Blanl il & B.F. Roukema Cosmological Topology in Paris 1998 Topologie cosmologique a Paris 1998 Editors: Vincent Blanl il1 & Boudewijn F. Roukema2 1Institut de recherche math ematique avanc ee, Universit e Louis Pasteur et CNRS, 7 rue Ren e-Descartes, F-67084 Strasbourg Cedex, France

Related Documents:

1 aht cooling systems gmbh paris 145 ad, vs paris 175 ad, vs paris 145 ad, vs typ b823, b823b, b823n paris 175 ad, vs typ b824, b824b, b824n paris 185 ad, vs typ b824, b824b, b824n paris 210 ad, vs typ b829, b829b, b825n paris 250 ad, vs typ b826, b826b, b826n athen 175 ad, vs typ b842, b842b, b842n athen 210 ad, vs typ b844, b844b, b844n athen xl 207-7 ad, vs

Zeno's paradoxes and the cosmological argument * JAN DEJNOZKA United States Naval Academy, A nnapolis, MD 21402-5000 1. Introduction My aim in this paper is to show that the cosmological argument for the existence of God commits a rather trivial linguistic (or conceptual)

Descartes, Rene Discourse on Method . Cosmological Argument,” p. 80-89 Lecture: “Aquinas’ Cosmological Arguments” Handouts ER . 6 / 8 · An Aristotle Primer · Outline of Aquinas’ Cosmological Arguments 12-Jul 8-Arguments for the Existence of God Pojman Lecture .

A. Inductive arguments – cosmological Inductive proofs the concept of a posteriori. Cosmological argument: St Thomas Aquinas first Three Ways 1. motion or change 2. cause and effect 3. contingency and necessity The Kalam cosmological argument with reference to William Lane Craig (rejection of actual infinities and

1 Introduction to Combinatorial Algebraic Topology Basic Topology Graphs to Simplicial Complexes Posets to Simplicial Complexes 2 Permutation Patterns Introduction and Motivation Applying Combinatorial Algebraic Topology Kozlov, Dimitry. Combinatorial algebraic topology. Vol. 21. Springer Science & Business Media, 2008.

TOPOLOGY: NOTES AND PROBLEMS Abstract. These are the notes prepared for the course MTH 304 to be o ered to undergraduate students at IIT Kanpur. Contents 1. Topology of Metric Spaces 1 2. Topological Spaces 3 3. Basis for a Topology 4 4. Topology Generated by a Basis 4 4.1. In nitude of Prime Numbers 6 5. Product Topo

a topology on R, the standard topology on R. Thus the open sets are unions of open intervals. A closed interval [c,d] is then closed in this topology. A half-open interval ]a,b],a bis neither open nor closed for this topology. We shall verify later that and R are the only subsets of R which are both open and closed (see (1.9.1)).

characterization of automotive disc brake, Applied Thermal Engineering (2014), doi: 10.1016/ j.applthermaleng.2014.10.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is .