ebook img

Reduction Method for One-loop Tensor 5- and 6-point Integrals Revisited PDF

0.12 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Reduction Method for One-loop Tensor 5- and 6-point Integrals Revisited

Reduction Method for One-loop Tensor 5- and 6-point Integrals Revisited Theodoros Diakonidis 9 Deutsches Elektronen-Synchrotron,DESY,Platanenallee 6, 15738 Zeuthen,Germany 0 0 A complete analytical reduction of general one-loop Feynman integrals with five legs 2 for tensors up to rank R=3 and six legs for tensors up to rank 4 is reviewed [1]. An n elegantformalismwithextensiveuseofsignedminorswasdevelopedforthecancellation a of leading inverse Gram determinants. The resulting compact formulae allow both for J a study of analytical properties and for efficient numerical programming. Here some 8 special numerical examples are presented. 2 ] 1 Introduction h p - The Feynman integrals for reactions with up to four external particles have been system- p atically studied and evaluated in numerous studies. It is needed to be mentioned here the e h seminalpapers[2]and[3]andtheFortranpackagesFF[4]andLoopTools[5],whichevidently [ show the situation so far. The treatment of Feynman integrals with a higher multiplicity than four becomes quite involved if questions of efficiency and stability become vital, as it 1 happens with the calculational problems related to high-dimensional phase space integrals v 5 over sums of thousands of Feynman diagrams with internal loops. 5 What is reviewedhere is anapproachwhichreduces the tensor integralsalgebraicallyto 4 sums over a small set of scalar two-, three- and four-point functions, which are assumed to 4 be known. To accomplishthis methods ofref.[6,7]areused. The presentgoalis to provide . 1 compact analytic formulas for the complete reduction of tensor pentagons and hexagons to 0 scalarmasterintegrals,whicharefree ofleadinginverseGramdeterminants. For astudy of 9 gauge invarianceand ofthe ultraviolet(UV) andinfrared(IR) singularity structure ofa set 0 : of Feynman diagrams, it is evident that a complete reduction is advantageous, and it may v also be quite useful for a tuned, analytical study of certain regions of potential numerical i X instabilities. r The numerics are obtained with two independent implementations, one made in Mathe- a matica, andanother one in Fortran. The Mathematica programhexagon.mwith the reduc- tionformulaeismadepubliclyavailable[8],seealso[9]forashortdescription. Fornumerical applications,one hasto link the packagewitha programfor the evaluationofscalarone- to four-point functions, e.g. with LoopTools[5, 10, 4], CutTools [11, 12], QCDLoop [13]. 2 Useful Notations Itis usefulto introducethe notationforthe loopintegralsandalsoforcertaindeterminants that occur in the recurrence relations and their solutions. The one-loop, N-point tensor integrals of rank R in d-dimensional space-time are defined as, ddk k ...k I(N) (d;ν ,...,ν )= µ1 µR (1) µ1...µR 1 N iπd/2Dν1...DνN Z 1 N LCWS/ILC2008 with propagatordenominators D =(k−q )2−m2+iǫ. (2) j j j Thedeterminantofan(N+1)×(N+1)matrix,knownasthemodifiedCayleydeterminant is defined as: [14], 0 1 1 ... 1 1 Y Y ... Y 11 12 1N (cid:12) (cid:12) () ≡ (cid:12) 1 Y12 Y22 ... Y2N (cid:12) , (3) N (cid:12) (cid:12) (cid:12) .. .. .. .. .. (cid:12) (cid:12) . . . . . (cid:12) (cid:12) (cid:12) (cid:12) 1 Y Y ... Y (cid:12) (cid:12) 1N 2N NN (cid:12) (cid:12) (cid:12) (cid:12) (cid:12) with coefficients (cid:12) (cid:12) Y =−(q −q )2+m2+m2, (i,j =1...N). (4) ij i j i j All other determinants appearing are signed minors of () , constructed by deleting m N rowsandm columns from() , andmultiplying with a signfactor. They will be denoted by N j j ··· j 1 2 m ≡ (−1)Pl(jl+kl) k k ··· k (cid:18) 1 2 m (cid:19)N rows j ···j deleted sgn sgn 1 m , (5) {j} {k} columns k ···k deleted (cid:12) 1 m (cid:12) (cid:12) (cid:12) where sgn and sgn are the signs(cid:12)(cid:12) of permutations that sort t(cid:12)(cid:12)he deleted rows j ···j {j} {k} 1 m and columns k ···k into ascending order. 1 m 3 Pentagons In this chapter final results are provided of the reduction concerning ranks up to 3. More about these can be found in [15]. For the scalar 5-point function the recursion relation for the limit of d=4 is, 5 1 0 E ≡I = Is, (6) 5 0 s 4 s=1(cid:18) (cid:19)5 0 X (cid:18) (cid:19)5 Similarly, for the tensor integral of rank 1 (vector) in the limit d→4 we obtain: 4 Iµ = qµI , (7) 5 i 5,i i=1 X 5 1 0 i I ≡E = − Is, (8) 5,i i 0 0 s 4 s=1(cid:18) (cid:19)5 0 X (cid:18) (cid:19)5 LCWS/ILC2008 The tensor integral of rank 2 can be written: 4 Iµν = qµqνE +gµνE , (9) 5 i j ij 00 i,j=1 X 5 5 E = S4,sIs+ S3,stIst, (10) ij ij 4 ij 3 s=1 s,t=1 X X s 1 1 5 0 0 s 5 t s E = − (cid:18) (cid:19)5 Is− Ist . (11) 00 2 0 s=1 s "(cid:18)0 s(cid:19)5 4 t=1(cid:18)0 s(cid:19)5 3 # 0 X s X (cid:18) (cid:19)5 (cid:18) (cid:19)5 Finally the tensor integral of rank 3 4 4 Iµνλ = qµqνqλE + g[µνqλ]E , (12) 5 i j k ijk k 00k i,j,k=1 k=1 X X 5 5 5 E = S4,sIs+ S3,stIst+ S2,stuIstu, (13) ijk ijk 4 ijk 3 ijk 2 s=1 s,t=1 s,t,u=1 X X X 5 5 5 E = S4,sIs+ S3,stIst+ S2,stuIstu. (14) 00k 00k 4 00k 3 00k 2 s=1 s,t=1 s,t,u=1 X X X All coefficients,also those not explicitly defined here S4,s,S3,st,S2,stu, S4,s, S3,st, S2,stu, ijk ijk ijk 00k 00k 00k S4,s, S3,st (see [15]), are free of the leading Gram determinants. ij ij 4 Hexagons If the external momenta of a hexagon are 4-dimensional, their Gram determinant vanishes: () =0, and a linear relation between the propagatorsD exists: 6 j 0 6 j 1 = (cid:18) (cid:19)6D . (15) j 0 j=1 X 0 (cid:18) (cid:19)6 Withthisrelation,anyhexagonintegralcantriviallybereducedtopentagons. Forexample, for the scalar hexagon, one obtains the well-known result [14]: 0 6 r I = (cid:18) (cid:19)6 Ir, (16) 6 0 5 r=1 X 0 (cid:18) (cid:19) 6 where the scalarpentagonIr onthe righthandsideis obtainedbyremovingliner fromthe 5 hexagonI . Inthesameway,tensorhexagonsofrankRcanbereducedtotensorpentagons 6 LCWS/ILC2008 Figure 1: Momenta flow used in the numerical examples for six- and five-point integrals. of rank R. However, it was noticed in ref. [6] that a reduction directly to tensor pentagons of rank R−1 is also possible: 6 Iµ1...µR = vµ1Iµ2...µR,r, (17) 6 r 5 r=1 X where 5 1 0 i vµ ≡ − qµ. (18) r 0 0 r i i=1(cid:18) (cid:19)6 0 X (cid:18) (cid:19)6 A more general proof of this property was given in ref. [16]. By substituting the reduction formulas for tensor pentagons into eq. (17), we can immediately express tensor hexagons in terms of scalar master integrals. In this way using the formulas of the previous section we can provide results for integrals up to 4th rank for the hexagons (see [15]). 5 Numerical results In order to illustrate the numerical results which can be obtained with the described ap- proach, a representative collection of tensor components will be evaluated, for some special cases which are not included in [9]. The kinematics are visualized in Figure 1. Fortheevaluationofthescalartwo-,three-andfour-pointfunctions,whichappearafter the complete reduction, we have implemented two numerical libraries: • For massive internal particles: Looptools 2.2 [5, 10]; • If there are also massless internal particles: QCDLoop-1.4[13]. the first one in the published package hexagon.m [8] and both of them in the Fortran implimentation. LCWS/ILC2008 p 5.0 0.0 0.0 5.0 1 p 5.0 0.0 0.0 – 5.0 2 p – 1.6554963633 1.2970338732 – 0.9062452085 – 0.4869198730 3 p – 3.8970139847 0.0528728505 – 2.5360890226 2.9584074987 4 p – 4.4474896520 – 1.3499067237 3.4423342311 – 2.4714876256 5 m =···=m =0.0 1 5 Table 1: The external four-momenta for the five-point functions; all internal and external particles are massless. ǫ0 1/ǫ E 0.49975096E-03+ i 0.12807271E-02 0.33696138E-03– i 0.64416161E-03 0 E1 – 0.50336057E-03– i 0.10928553E-02 – 0.34786666E-03+ i 0.54767334E-03 E12 – 0.11603164E-02– i 0.17552616E-02 – 0.60899168E-03+ i 0.12327007E-02 E122 – 0.43997517E-02– i 0.34454891E-02 – 0.10597882E-02+ i 0.36519758E-02 1/ǫ2 E – 0.15779987E-03+i 0.00000000E+00 0 E1 0.17432984E-03+i 0.00000000E+00 E12 0.39238082E-03+i 0.00000000E+00 E122 0.11624600E-02+i 0.00000000E+00 Table 2: Selected tensor components of five-point tensor functions with massless particles; kinematics defined in Table 1 (Cross checked with golem95 [17]). p 5.0 0.0 0.0 5.0 1 p 5.0 0.0 0.0 – 5.0 2 p – 0.7623942818 0.5390582570 – 0.5220507689 0.1346262645 3 p – 3.3298826057 – 1.0349623069 – 1.1048040197 2.9658690580 4 p – 2.8267285956 – 1.4136906402 2.3189438782 – 0.7838192500 5 p – 3.0809945169 1.9095946901 – 0.6920890895 – 2.3166760725 6 m =1.0, m =1.2, m =1.4, m =1.6, m =1.8, m =2.0 1 2 3 4 5 6 Table 3: The external four-momenta for the six-point functions; all external legs massless and the internal massive. F 0.54701021E-04– i 0.67031213E-04 0 F3 – 0.32082506E-04+ i 0.24545301E-03 F11 0.13862332E-04– i 0.12247788E-03 F112 – 0.22452724E-04– i 0.39826579E-04 F0121 0.15817785E-03+ i 0.26882173E-03 Table 4: Selected tensor components of six-point tensor functions produced by the phase space point of Table 3. LCWS/ILC2008 6 Acknowledgments WorksupportedtheEuropeanCommunity’sMarie-CurieResearchTrainingNetworksMRTN- CT-2006-035505“HEPTOOLS”andbySonderforschungsbereich/TransregioSFB/TRR9of DFG “Computergestu¨tzte Theoretische Teilchenphysik”. I would also like to thank my col- laborators J. Fleischer, J. Gluza, K. Kajda, T. Riemann, and especially J. B. Tausk for useful discussions. References [1] Presentation http://ilcagenda.linearcollider.org/materialDisplay.py?contribId=77&sessionId=18&materialId=slides&confId=2628. [2] Gerard’tHooftandM.Veltman. Scalaroneloopintegrals. Nucl. Phys.,B153:365–401, 1979. [3] G.PassarinoandM.Veltman. Oneloopcorrectionsfore+e− annihilationintoµ+µ− intheWeinberg model. Nucl. Phys.,B160:151,1979. [4] G.vanOldenborgh. FF:ApackagetoevaluateoneloopFeynmandiagrams. Comput.Phys.Commun., 66:1–15,1991. [5] T.HahnandM.Perez-Victoria. Automatizedone-loopcalculationsinfourandddimensions. Comput. Phys. Commun.,118:153,1999. [6] J. Fleischer, F. Jegerlehner, and O. Tarasov. Algebraic reduction of one-loop Feynman graph ampli- tudes. Nucl. Phys.,B566:423–440, 2000. [7] A. Davydychev. A simple formula for reducing Feynman diagrams to scalar integrals. Phys. Lett., B263:107–111, 1991. [8] SilesianUniv.,Katowice,webpagehttp://www.us.edu.pl/∼gluza/hexagon, DESY,webpagehttp://www-zeuthen.desy.de/theory/research/CAS.html. [9] T.Diakonidis,J.Fleischer,J.Gluza,K.Kajda,T.Riemann,andJ.B.Tausk. Onthetensorreduction ofone-looppentagons andhexagons. Nucl. Phys. Proc. Suppl.,183:109–115, 2008. [10] Thomas Hahn and Michael Rauch. News from FormCalc and LoopTools. Nucl. Phys. Proc. Suppl., 157:236–240, 2006. [11] AndrevanHameren,JensVollinga,andStefanWeinzierl.Automatedcomputationofone-loopintegrals inmasslesstheories. Eur. Phys. J.,C41:361–375, 2005. [12] GiovanniOssola,C.Papadopoulos, andRobertoPittau. CutTools: aprogramimplementingtheOPP reductionmethodtocompute one-loopamplitudes. JHEP,03:042, 2008. [13] R.K.EllisandGiuliaZanderighi. Scalarone-loopintegralsforQCD. JHEP,02:002,2008. [14] D.B.Melrose. ReductionofFeynmandiagrams. Nuovo Cim.,40:181–213, 1965. [15] Th.Diakonidisetal. Acompletereductionofone-looptensor5-and6-pointintegrals. 2008. [16] A.DennerandS.Dittmaier.Reductionschemesforone-looptensorintegrals.Nucl.Phys.,B734:62–115, 2006. [17] T. Binoth, J. Ph. Guillet, G. Heinrich, E. Pilon, and T. Reiter. Golem95: a numerical program to calculateone-looptensorintegralswithuptosixexternal legs. 2008. LCWS/ILC2008

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.