Hyperdiamond Feynman Checkerboard in 4-Dimensional Spacetime
Total Page:16
File Type:pdf, Size:1020Kb
View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by CERN Document Server quant-ph/950301 5 THEP-95-3 March 1995 Hyp erDiamond Feynman Checkerb oard in 4-dimensional Spacetime Frank D. (Tony) Smith, Jr. e-mail: [email protected] and [email protected] P.O.Box for snail-mail: P.O.Box 430, Cartersville, Georgia 30120 USA WWW URL http://www.gatech.edu/tsmith/home.html Scho ol of Physics Georgia Institute of Technology Atlanta, Georgia 30332 Abstract A generalized Feynman Checkerb oard mo del is constructed using a processed by the SLAC/DESY Libraries on 22 Mar 1995. 〉 4-dimensional Hyp erDiamond lattice. The resulting phenomenological mo del is the D D E mo del describ ed in hep-ph/9501252 and 4 5 6 quant-ph/9503009. PostScript c 1995 Frank D. (Tony) Smith, Jr., Atlanta, Georgia USA QUANT-PH-9503015 Contents 1 Intro duction. 2 2 Feynman Checkerb oards. 3 3 Hyp erDiamond Lattices. 10 3.1 8-dimensional Hyp erDiamond Lattice. :: :: ::: :: :: :: 12 3.2 4-dimensional Hyp erDiamond Lattice. :: :: ::: :: :: :: 14 3.3 From 8 to 4 Dimensions. : ::: :: :: :: :: ::: :: :: :: 19 4 Internal Symmetry Space. 21 5 Protons, Pions, and Physical Gravitons. 25 5.1 3-Quark Protons. : :: :: ::: :: :: :: :: ::: :: :: :: 29 5.2 Quark-AntiQuark Pions. : ::: :: :: :: :: ::: :: :: :: 32 5.3 Spin-2 Physical Gravitons. ::: :: :: :: :: ::: :: :: :: 35 1 1 Intro duction. The 1994 Georgia Tech Ph. D. thesis of Michael Gibbs under David Finkel- stein [11] constructed a discrete 4-dimensional spacetime Hyp erDiamond lat- tice, here denoted a 4HD lattice, in the course of building a physics mo del. The 1992 MIT Ph. D. thesis of Hrvo je Hrgovcic under Tommasso To oli [12] constructed a discrete 4-dimensional spacetime Minkowski lattice in the course of building a simulation of solutions to the Dirac equation and other equations. Hrgovcic's lattice is closely related to the 4HD lattice. Both the Gibbs mo del and the Hrgovcic simulation di er in signi cant ways from the D D E mo del constructed from 3 3 o ctonion matrices 4 5 6 in hep-ph/9501252 and quant-ph/9503009 . However, b oth theses have b een very helpful with resp ect to this pap er, which generalizes the 2-dimensional Feynman checkerb oard [8, 9] to a phys- ically realistic 4-dimensional spacetime. In this 4-dimensional 4HD lattice generalized Feynman Hyp erDiamond checkerb oard, the prop erties of the particles that move around on the checkerb oard are determined by the physically realistic D D E mo del constructed from 3 3 o ctonion matrices 4 5 6 in hep-ph/9501252 and quant-ph/9503009 . The 4-dimensional Hyp erDiamond 4HD checkerb oard is related to the 8-dimensional Hyp erDiamond 8HD = E in the same way 8 the 4-dimensional physical asso ciative spacetime is related to the 8-dimensional o ctonionic spacetime in the D D E mo del, 4 5 6 so the Hyp erDiamond 4HD Feynman checkerb oard is fundamentally the discrete version of the D D E mo del. 4 5 6 For more ab out o ctonions and lattices, see the works of Go e rey Dixon [3, 4, 5, 6]. 2 2 Feynman Checkerb oards. The 2-dimensional Feynman checkerb oard [8, 9] is a notably successful and useful representation of the Dirac equation in 2-dimensional spacetime. To build a Feynman 2-dimensional checkerb oard, start with a 2-dimensional Diamond checkerb oard with two future lightcone links and two past lightcone links at eachvertex. The future lightcone then lo oks like @I @ If the 2-dimensional Feynman checkerb oard is co ordinatized by the complex plane C : the real axis 1 is identi ed with the time axis t; the imaginary axis i is identi ed with the space axis x; and p p the two future lightcone links are (1= 2 )(1 + i) and (1= 2)(1 i). 2 2 In cylindrical co ordinates t; r with r = x , 2 2 2 2 the Euclidian metric is t + r = t + x and the Wick-Rotated Minkowski metric with sp eed of light c is 2 2 2 2 (ct) r =(ct) x . For the future lightcone links to lie on the 2-dimensional Minkowski lightcone, c =1. Either link is taken into the other link by complex multiplication by i. Now, consider a path in the Feynman checkerb oard. At a given vertex in the path, denote the future lightcone link in the same direction as the past path link by1,and the future lightcone link in the (only p ossible) changed direction by i. 3 1 i @I @ The Feynman checkerb oard rule is that if the future step at a vertex p oint of a given path is in a di erent direction from the immediately preceding step from the past, then the path at the p ointofchange gets a weightof im, where m is the mass (only massive particles can change directions),and is the length of a path segment. Here I have used the Gersch [10] convention of weighting each turn by im rather than the Feynman [8, 9] convention of weighting by+im, b ecause Gersch's convention gives a b etter nonrela- tivistic limit in the isomorphic 2-dimensional Ising mo del [10]. HOW SHOULD THIS BE GENERALIZED TO HIGHER DIMENSIONS? 22 0 The 2-dim future light-cone is the 0-sphere S = S = fi; 1g , with 1 representing a path step to the future in the same direction as the path step from the past, and i representing a path step to the future in a (only 1 in the 2-dimensional Feynman checkerb oard lattice) di erent direction from the path step from the past. The 2-dimensional Feynman checkerb oard lattice spacetime can b e represented by the complex numb ers C , with 1;i representing the two future lightcone directions and 1; i representing the two past lightcone directions. Consider a given path in the Feynman checkerb oard lattice 2-dimensional spacetime. 4 Atany given vertex on the path in the lattice 2-dimensional spacetime, the future lightcone direction representing the continuation of the path in the same direction can b e represented by1,and the future lightcone direction representing the (only 1 p ossible) change of direction can b e represented by i since either of the 2 future lightcone directions can b e taken into the other bymultiplication by i, + for a left turn and for a right turn. If the path do es change direction at the vertex, then the path at the p ointofchange gets a weightof im, where i is the complex imaginary, m is the mass (only massive particles can change directions), and is the timelike length of a path segment, where the 2-dimensional sp eed of light is taken to b e 1. Here I have used the Gersch [10] convention of weighting each turn by im rather than the Feynman [8, 9] convention of weighting by+im, b ecause Gersch's convention gives a b etter nonrela- tivistic limit in the isomorphic 2-dimensional Ising mo del [10]. For a given path, let C b e the total numb er of direction changes, and c b e the cth change of direction, and i b e the complex imaginary representing the cth change of direction. C can b e no greater than the timelikecheckerb oard distance D between the initial and nal p oints. The total weight for the given path is then Y Y C C im =(m) ( i)= (im) (1) 0cC 0cC The pro duct is a vector in the direction 1ori. 5 Let N (C ) b e the numb er of paths with C changes in direction. The propagator amplitude for the particle to go from the initial vertex to the nal vertex is the sum over all paths of the weights, that is the path integral sum over all weighted paths: X C N (C )(im) (2) 0C D The propagator phase is the angle b etween the amplitude vector in the complex plane and the complex real axis. Conventional attempts to generalize the Feynman checkerb oard from 2-dimensional spacetime to k -dimensional spacetime are based on the fact that the 2-dimensional future light-cone directions are 22 0 the 0-sphere S = S = fi; 1g. The k -dimensional continuous spacetime lightcone directions are k 2 the (k 2)-sphere S . 2 In 4-dimensional continuous spacetime, the lightcone directions are S . Instead of lo oking for a 4-dimensional lattice spacetime, Feynman 0 2 and other generalizers went from discrete S to continuous S for lightcone directions, and then tried to construct a weighting 2 using changes of directions as rotations in the continuous S , and never (as far as I know) got any generalization that worked. 6 THE D D E MODEL 4HD HYPERDIAMOND 4 5 6 GENERALIZATION HAS DISCRETE LIGHTCONE DIRECTIONS. If the 4-dimensional Feynman checkerb oard is co ordinatized by the quaternions Q: the real axis 1 is identi ed with the time axis t; the imaginary axes i; j; k are identi ed with the space axes x; y ; z ; and the four future lightcone links are (1=2)(1 + i + j + k ), (1=2)(1 + i j k ), (1=2)(1 i + j k ), and (1=2)(1 i j + k ). In cylindrical co ordinates t; r 2 2 2 2 with r = x + y + z , 2 2 2 2 2 2 the Euclidian metric is t + r = t + x + y + z and the Wick-Rotated Minkowski metric with sp eed of light c is 2 2 2 2 2 2 (ct) r =(ct) x y z .