Poncelet's Theorem in Finite Projective Planes and Beyond

Total Page:16

File Type:pdf, Size:1020Kb

Poncelet's Theorem in Finite Projective Planes and Beyond Research Collection Doctoral Thesis Poncelet's Theorem in finite projective planes and beyond Author(s): Kusejko, Katharina Publication Date: 2016 Permanent Link: https://doi.org/10.3929/ethz-a-010682687 Rights / License: In Copyright - Non-Commercial Use Permitted This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use. ETH Library Diss. ETH No. 23366 Poncelet's Theorem in finite projective planes and beyond A thesis submitted to attain the degree of DOCTOR OF SCIENCES of ETH ZURICH (Dr. sc. ETH Zurich) presented by Katharina Kusejko MSc Math. ETH Zurich born on 01.10.1989 citizen of Austria accepted on the recommendation of Prof. Dr. Norbert Hungerb¨uhler,advisor Prof. Dr. Alexander Pott, co-advisor 2016 "Perhaps I could best describe my experience of doing mathematics in terms of en- tering a dark mansion. One goes into the first room and it is dark, entirely dark. One stumbles around, bumping into the furniture. Gradually you learn, where each piece of furniture is. And finally, after six months or so, you find the light switch, you turn it on and suddenly, everything is illuminated and you can see exactly where you are." Andrew Wiles Zusammenfassung Die vorliegende Dissertation handelt von Kegelschnitten in endlichen projektiven Ebenen sowie von Kreisketten in endlichen M¨obiusebenen. Die Hauptresultate wur- den bereits in [22], [23], [26] und [24] ver¨offentlicht. Ein besonderer Fokus liegt auf dem Schließungssatz des franz¨osischen Mathematikers Jean-Victor Poncelet, der 1813 fur¨ Kegelschnitte in der reellen projektiven Ebene eine interessante Beobachtung zu Polygonzugen,¨ dessen Kanten tangential zu einem Kegelschnitt sind und dessen Ecken auf einem anderen Kegelschnitt liegen, forma- lisiert und bewiesen hat. Dieses klassische Resultat, welches zu Beginn der These im Detail erkl¨art wird, wurde in den letzten 200 Jahren wiederholt in verschiedenen Teilgebieten der Mathematik - unter anderem in der Analysis, der Geometrie und der Theorie elliptischer Kurven - aufgegriffen. Endliche projektive Ebenen, welche uber¨ einem endlichen K¨orper definiert werden, erfullen¨ eine endliche Version des Poncelet'schen Satzes. Diese Tatsache wird in meh- reren Kapiteln mit Hilfe unterschiedlicher Techniken, darunter kombinatorische, syn- thetische und algebraische Methoden, bewiesen. Im Falle jener Ebenen, welche nicht uber¨ endlichen K¨orpern konstruiert werden, ist die Poncelet'sche Schließungseigen- schaft nicht zwangsl¨aufig gegeben. Dies wird anhand endlicher projektiver Ebenen der Ordnung 9 im Detail erl¨autert. Weiterfuhrende¨ Fragen zum Schließungssatz von Poncelet, wie etwa ein Poncelet'sches Schließungskriterium, werden fur¨ Spezialf¨alle von Kegelschnittpaaren betrachtet. Dies liefert unter anderem einen neuen Zugang zu trigonometrischen Identit¨aten in endlichen projektiven Ebenen, fur¨ welche ein geometrischer Zugang zur Trigonometrie a priori nicht gegeben ist. Im Vordergrund des behandelten Spezialfalles stehen Kegelschnitte in Diagonalform, was in der Folge Fragen zur Diagonalisierbarkeit von Kegelschnittpaaren im Endlichen aufwirft. Mit Hilfe kombinatorischer Uberlegungen¨ werden Bedingungen fur¨ die Diagonalisierbar- keit pr¨asentiert und Unterschiede zum reellen Fall erkl¨art. Eine dem Poncelet'schen Satz verwandte Aussage ist jene des Schweizer Mathema- tikers Jakob Steiner, welche von Ketten sich beruhrender¨ Kreise handelt. Hier wird eine analoge Aussage fur¨ Steiner-Ketten in endlichen M¨obiusebenen, welche uber¨ endlichen K¨orpern konstruiert werden, im Detail studiert. Ahnlich¨ wie im Satz von Poncelet fuhren¨ auch hier die Berechnungen eines Schließungskriteriums fur¨ Kreis- ketten zu einer Interpretation trigonometrischer Identit¨aten in endlichen Ebenen. Durchgehend werden die jeweiligen Eigenschaften reeller und endlicher Ebenen ver- glichen, und vor allem erstaunliche Parallelen und kontraintuitive Unterschiede her- vorgehoben. Abstract This dissertation is mainly about conics in finite projective planes and chains of circles in finite M¨obiusplanes. The main results are already published in [22], [23], [26] and [24]. A particular focus lies on Poncelet's Theorem, a famous result by the French math- ematician Jean-Victor Poncelet. In 1813, he made an interesting observation about two conics in the real projective plane and a polygonal chain with vertices on one conic and the edges tangent to the other conic. This classical result, which is dis- cussed in detail at the beginning of the thesis, was studied repeatedly in the last 200 years in various areas of mathematics - such as analysis, geometry and in the theory of elliptic curves. Finite projective planes defined over a finite field satisfy a finite version of Pon- celet's Theorem. This is proven in several chapters of this thesis by using different mathematical tools, such as combinatorial, synthetic and algebraic methods. In the case of finite projective planes not defined over finite fields, the finite version of Poncelet's Theorem is proven to be wrong. This is shown by examining the finite projective planes of order 9. Further questions related to Poncelet's Theorem, such as a criterion for a polygonal chain to close up, are discussed for special cases of conics. This leads to a new approach of trigonometric identities in finite projective planes, for which a geometric interpretation of trigonometry is a priori not given. A particular focus lies on conics in diagonal form and therefore questions about the diagonalisation of conics in finite projective planes are discussed. Conditions on the simultaneous diagonalisation of conic pairs are presented by using combinatorial methods. These conditions are then compared to the real case. A result related to Poncelet's Theorem deals with chains of touching circles and is due to the Swiss mathematician Jacob Steiner. In this dissertation, we discuss a similar statement for Steiner chains in finite M¨obiusplanes constructed over finite fields. Again, by investigating a criterion for the chain of circles to close, we get an interpretation of trigonometric identities in finite M¨obiusplanes. Throughout, the properties of real and finite planes are compared, with an emphasis on surprising parallels and counter-intuitive differences. Contents 1 Introduction 1 2 Preliminaries 7 2.1 Finite projective planes . .7 2.2 Finite Fields . .9 2.3 The Desarguesian Plane . 10 2.4 Conics . 12 2.5 Pencil of conics . 13 3 Poncelet's Theorem 15 3.1 The classical version . 15 3.2 Interpretation in finite planes . 17 3.3 Degenerate Cases . 19 3.4 Examples . 21 4 Poncelet's Theorem in finite projective planes of order 9 25 4.1 Poncelet's Theorem in PG(2; 9)..................... 25 4.2 Poncelet's Theorem in the finite projective planes over S ....... 33 4.2.1 The miniquaternion near-field S ................. 33 4.2.2 The plane Ω . 35 4.2.3 The plane ΩD ........................... 38 4.2.4 The plane Ψ . 40 5 Some proofs for Poncelet's Theorem 45 5.1 A synthetic proof . 45 5.2 A combinatorial proof . 53 5.2.1 Preliminaries and preparation . 53 5.2.2 Poncelet's Theorem for triangles . 55 5.2.3 Proof of Poncelet's Theorem . 57 Contents 6 A Poncelet Criterion 63 6.1 Construction and properties . 63 6.2 Poncelet's Theorem for conics Ok .................... 64 6.3 Relations for pairs of conics . 66 6.4 A Poncelet Criterion . 72 6.4.1 Poncelet coefficients . 72 6.4.2 Poncelet polynomials . 80 6.5 Comparison to other methods . 82 6.5.1 Comparison to the Euclidean Plane . 82 6.5.2 Comparison to Cayley's Criterion . 85 7 Simultaneous Diagonalization 87 7.1 The disjoint case . 87 7.2 The nondisjoint case . 94 7.3 Summary . 98 8 Steiner's Theorem in the finite M¨obiusplane M(q) 99 8.1 Miquelian M¨obius planes . 100 8.2 Steiner's Theorem in M(q)........................ 100 8.3 The plane M(5) .............................. 103 8.4 Existence and length of Steiner chains . 105 8.5 Generalization . 110 8.5.1 A M¨obiusinvariant for pairs of circles . 110 8.5.2 Transformation of non-intersecting circles into concentric circles111 8.5.3 General criterion for Steiner chains . 113 8.6 Comparison to the Euclidean plane . 113 Bibliography 117 Dank 121 Curriculum Vitae 123 Chapter 1 Introduction This thesis is about a branch of geometry called finite projective geometry. In a nutshell, projective geometry is like Euclidean geometry with additional infinite elements. Imagine standing on a train track and looking at the horizon. Although the tracks are parallel, they seem to meet at infinity. 1 This is exactly what projective geometry deals with. In particular, there are no parallel lines in a projective plane. This explains the idea of the term projective. Now for the finiteness. In our continuous world there are infinitely many points on every line. But one can describe a concept of geometry also by taking only finitely many points. In particular, a line will be a collection of finitely many points. This branch of geometry is relatively new and has for example applications in coding theory and cryptography. In this thesis, we are mainly interested in one particular property of a finite projective plane, namely the property of being a Poncelet plane. Jean-Victor Poncelet (1788 - 1867) [36] was the first mathematician who formalized projective geometry, and it was him, in 1813, to state one of the most beautiful results in this field. One version reads as follows. 1Picture taken from http://www.thoughtsupport.net/how-to-get-back-on-track/ 1 Introduction Theorem (Poncelet's Theorem). Let C and C0 be two conics generally situated in the real projective plane. If it is possible to find an m-sided polygon, m 3, such that the vertices lie on C0 and the sides are tangent to C, then there are≥ infinitely many other such m-sided polygons. Moreover, one cannot find such an m0-sided polygon for m = m0 for the same pair of conics C and C0. 6 Figure 1.1: Example of Poncelet's Theorem In Chapter 2, we will start with some preliminaries about finite projective planes.
Recommended publications
  • 1. Math Olympiad Dark Arts
    Preface In A Mathematical Olympiad Primer , Geoff Smith described the technique of inversion as a ‘dark art’. It is difficult to define precisely what is meant by this phrase, although a suitable definition is ‘an advanced technique, which can offer considerable advantage in solving certain problems’. These ideas are not usually taught in schools, mainstream olympiad textbooks or even IMO training camps. One case example is projective geometry, which does not feature in great detail in either Plane Euclidean Geometry or Crossing the Bridge , two of the most comprehensive and respected British olympiad geometry books. In this volume, I have attempted to amass an arsenal of the more obscure and interesting techniques for problem solving, together with a plethora of problems (from various sources, including many of the extant mathematical olympiads) for you to practice these techniques in conjunction with your own problem-solving abilities. Indeed, the majority of theorems are left as exercises to the reader, with solutions included at the end of each chapter. Each problem should take between 1 and 90 minutes, depending on the difficulty. The book is not exclusively aimed at contestants in mathematical olympiads; it is hoped that anyone sufficiently interested would find this an enjoyable and informative read. All areas of mathematics are interconnected, so some chapters build on ideas explored in earlier chapters. However, in order to make this book intelligible, it was necessary to order them in such a way that no knowledge is required of ideas explored in later chapters! Hence, there is what is known as a partial order imposed on the book.
    [Show full text]
  • Conic Sections
    Conic sections Conics have recurred throughout this book in both geometric and algebraic settings. Hence, I have decided to dedicate the final chapter to them. As the only conics appearing on IMO geometry problems are invariably circles, the results proved in this chapter are largely irrelevant. Nevertheless, the material is sufficiently interesting to be worthy of inclusion. Sections of cones With the obvious exception of the circle, the conics were first discovered by the Greek mathematician Menaech- mus who contemplated slicing a right circular cone C with a flat plane . Indeed, the term ‘conic’ is an abbrevia- tion of conic section . It is more natural to consider C as the double cone with equation x2 y2 z2. If cuts both cones, then the conic section is a hyperbola . If it cuts only one cone in a closed curve, it is an ellipse . The intermediate case, where the plane is inclined at exactly the same slope as the cone, results in a parabola . Observe that x2 y2 z2 is the equation of a projective circle; this explains why all conic sections are equivalent under projective transformations. We define a Dandelin sphere 5 to be a sphere tangent to both C (at a circle *) and (at a point F, namely the focus ). The plane containing * intersects at a line ", known as the directrix . 1. Prove that the directrix is the polar of the focus. For an arbitrary point P on the conic, we let P R meet * at Q. 2. Prove that P Q P F . 3. Let A be the foot of the perpendicular from P to the plane containing *.
    [Show full text]
  • Pappus Chain
    Pappus Chain ​ ​ Ruisi Ma & Yimeng Liu Background Properties Arbelos Center of the Circle Art: The term "arbelos" means All the centers of the shoemaker's knife in Greek. An arbelos is combined circles in the Pappus chain The World Necklace Mathematically inspired Poster with three semicircles which are shared with one of are located on a common ellipse, for the following ​ ​ (2015) the others, all on the same side of a straight line that reason. The sum of the distances from the nth circle contains their diameters. These three circles are also to the two centers U (Largest)and V (Second) of tangential from each other. (Weisstein, 2020) the arbelos circles equals a constant: (Weisstein, Cosmic Evolution: The Rise of Complexity in Nature by Eric J Original: Chaisson . The Arbelo was first introduced in the book of 2020) Lemmas.The book of Lemmas is Thābitibn Qurra's @koji_glass (2019) Coordinate ​ book attributed to Archimedes. It consists of fifteen If r = AC / AB then the center of the nth circle is: propositions on circles. From Wikipedia. Further Research: Pappus Chain 1. How kind of applications are there for the Starting with the circle P 1 pappus chain? tangent to the three (Weisstein, 2020) 2. How could the properties of the pappus semicircles forming the Distance from bottom to nth circle’s center is n chain also apply to the steiner chain? ​ 3. How to use the pappus chain to calculate the arbelos, construct a chain of tangent circles P i , all time radiums: ​ ​ ​ area in geometry? tangent to one of the second small interior circles Using inverse geometry theory, and to the largest exterior one.
    [Show full text]
  • Exploring Steiner Chains with Möbius Transformations
    ExploringExploring Steiner Chains Steiner with Chains Möbius with Transformations* Möbius∗ Kristian Kiradjiev AMIMA, University of Oxford In this article, we use circular gardens, Steiner’s porism and For the sake of simplicity, we will limit ourselves to simple Möbius transformations to construct Steiner chains of tangential closed chains, i.e., wrapping only once around the inner bound- circles. We then explore some interesting area optimisation prob- ing circle. In Figure 1, we show a simple closed Steiner chain, lems and touch on Soddy’s hexlet and the Duplin cyclide. consisting of n =7circles. A lot of fascinating properties have been discovered so far. For example, it is known that the centres of the circles in the chain lie either on an ellipse (or circle) when one of the bounding circles 1 Introduction lies within the other, or on a hyperbola if not. Also, the points of tangency between the circles in the chain happen to lie on a cir- teiner chains are a beautiful example in circle geome- cle [1]. More interestingly, using inversion, a feasibility criterion try. A Steiner chain is defined as a chain of n circles, each has been established in [1] for whether a closed Steiner chain is S tangent to the previous one and the next one, and also to supported for a given n and a pair of bounding circles. two given non-intersecting circles [1], which we will call bound- The problem we aim to tackle in this article, is somewhat the ing circles. We focus exclusively on Steiner chains, one of whose opposite: given n positive numbers, does there exist a pair of bounding circles lies within the other.
    [Show full text]
  • MF-$0.75 HC Not Available from EDRS. PLUS POSTAGE Geometry
    DOCUMENT RESUME ED 100 648 SE 018 119 AUTHOR Yates, Robert C. TITLE Curves and Their Properties. INSTITUTION National Council of Teachers of Mathematics, Inc., Washington, D.C. PUB DATE 74 NOTE 259p.; Classics in Mathematics Education, Volume 4 AVAILABLE FROM National Council of Teachers of Mathematics, Inc., 1906 Association Drive, Reston, Virginia 22091 ($6.40) EDRS PRICE MF-$0.75 HC Not Available from EDRS. PLUS POSTAGE DESCRIPTORS Analytic Geometry; *College Mathematics; Geometric Concepts; *Geometry; *Graphs; Instruction; Mathematical Enrichment; Mathematics Education;Plane Geometry; *Secondary School Mathematics IDENTIFIERS *Curves ABSTRACT This volume, a reprinting of a classic first published in 1952, presents detailed discussions of 26 curves or families of curves, and 17 analytic systems of curves. For each curve the author provides a historical note, a sketch orsketches, a description of the curve, a a icussion of pertinent facts,and a bibliography. Depending upon the curve, the discussion may cover defining equations, relationships with other curves(identities, derivatives, integrals), series representations, metricalproperties, properties of tangents and normals, applicationsof the curve in physical or statistical sciences, and other relevantinformation. The curves described range from thefamiliar conic sections and trigonometric functions through tit's less well knownDeltoid, Kieroid and Witch of Agnesi. Curve related - systemsdescribed include envelopes, evolutes and pedal curves. A section on curvesketching in the coordinate plane is included. (SD) U S DEPARTMENT OFHEALTH. EDUCATION II WELFARE NATIONAL INSTITUTE OF EDUCATION THIS DOCuME N1 ITASOLE.* REPRO MAE° EXACTLY ASRECEIVED F ROM THE PERSON ORORGANI/AlICIN ORIGIN ATING 11 POINTS OF VIEWOH OPINIONS STATED DO NOT NECESSARILYREPRE INSTITUTE OF SENT OFFICIAL NATIONAL EDUCATION POSITION ORPOLICY $1 loor oiltyi.4410,0 kom niAttintitd.: t .111/11.061 .
    [Show full text]
  • Volume 10 (2010) 1–6
    FORUM GEOMETRICORUM A Journal on Classical Euclidean Geometry and Related Areas published by Department of Mathematical Sciences Florida Atlantic University FORUM GEOM Volume 10 2010 http://forumgeom.fau.edu ISSN 1534-1178 Editorial Board Advisors: John H. Conway Princeton, New Jersey, USA Julio Gonzalez Cabillon Montevideo, Uruguay Richard Guy Calgary, Alberta, Canada Clark Kimberling Evansville, Indiana, USA Kee Yuen Lam Vancouver, British Columbia, Canada Tsit Yuen Lam Berkeley, California, USA Fred Richman Boca Raton, Florida, USA Editor-in-chief: Paul Yiu Boca Raton, Florida, USA Editors: Nikolaos Dergiades Thessaloniki, Greece Clayton Dodge Orono, Maine, USA Roland Eddy St. John’s, Newfoundland, Canada Jean-Pierre Ehrmann Paris, France Chris Fisher Regina, Saskatchewan, Canada Rudolf Fritsch Munich, Germany Bernard Gibert St Etiene, France Antreas P. Hatzipolakis Athens, Greece Michael Lambrou Crete, Greece Floor van Lamoen Goes, Netherlands Fred Pui Fai Leung Singapore, Singapore Daniel B. Shapiro Columbus, Ohio, USA Man Keung Siu Hong Kong, China Peter Woo La Mirada, California, USA Li Zhou Winter Haven, Florida, USA Technical Editors: Yuandan Lin Boca Raton, Florida, USA Aaron Meyerowitz Boca Raton, Florida, USA Xiao-Dong Zhang Boca Raton, Florida, USA Consultants: Frederick Hoffman Boca Raton, Floirda, USA Stephen Locke Boca Raton, Florida, USA Heinrich Niederhausen Boca Raton, Florida, USA Table of Contents Quang Tuan Bui, A triad of similar triangles associated with the perpendicular bisectors of the sides of a triangle,1
    [Show full text]