. Quantum Gravitational Contributions to Gauge Theories . .. . . . Yong Tang(汤勇) Institute of Theoretical Physics, Chinese Academy of Sciences. 中国科学院理论物理研究所 2011-4-1 Y.Tang (ITP-CAS) Quantum Gravitational Contributions 1 / 27 Outline . .1. Introduction . .2. Diagrammatical Calculation . .3. Background Field Method . .4. Vilkovisky-DeWitt Formalism . .5. Summary Y.Tang (ITP-CAS) Quantum Gravitational Contributions 2 / 27 Reference S. P. Robinson and F. Wilczek, Phys. Rev. Lett. 96, 231601 (2006). A. R. Pietrykowski, Phys. Rev. Lett. 98, 061801 (2007). D. J. Toms, Phys. Rev. D 76, 045015 (2007), Phys. Rev. Lett. 101, 131301(2008), Phys. Rev. D 80, 064040(2009). D. Ebert, J. Plefka and A. Rodigast, Phys. Lett. B660, 579(2008). Y. Tang and Y. L. Wu, Comm. Theo. Phys. 54, 1040(2010)[arXiv:0807.0331v2], arXiv:1012.0626. ······ Y.Tang (ITP-CAS) Quantum Gravitational Contributions 3 / 27 Introduction Introduction Einstein gravity theory is not renormalizable from the perspective of quantum field theory. Quantum effects of gravity become significant only at Planck scale, and it may be possible to treat it as an effective field theory at low energy scale. Robinson and Wilczek claimed that quantum gravity can correct gauge couplings with power-law running and render all gauge theories asymptotically free. Later, further analysis by other authors suggested that the previous result was gauge dependent. Y.Tang (ITP-CAS) Quantum Gravitational Contributions 4 / 27 Diagrammatical Calculation The Lagrangian . We begin with the action of Einstein-Yang-Mills theory . [ ] ∫ . √ 1 1 µα νβ a a 4 S = d x −g 2 R − g g Fµν Fαβ (1) κ 4 . .. . a where R is Ricci scalar and Fµν is the Yang-Mills fields strength Fµν = ∇µ Aν − ∇ν Aµ − ig[Aµ , Aν√ ]. With the minus-dimension coupling constant κ = 16πG. Usually, one expands the metric tensor around a background metric ḡµν and treats graviton field as quantum fluctuation hµν propagating on the background space-time, ḡµν . gµν = ḡµν + κhµν , gµν = ḡµν − κhµν + κ2 hµα hαν + ... (2) √ √ 1 1 1 −g = −ḡ[1 + κh − κ2 (hµν hµν − h2 )...] (3) 2 4 2 Y.Tang (ITP-CAS) Quantum Gravitational Contributions 5 / 27 Diagrammatical Calculation Gauge-fixing condition Let us set ḡµν = ηµν , where ηµν is the Minkowski metric. hµν is interpreted as graviton field, fluctuating in flat space-time. The lagrangian can be arranged to different orders of hµν or κ. In the de Donder harmonic gauge 1 χµ = ∂ν hµν − ∂ µ hνν = 0 2 Graviton propagator has a simple form Pµνρσ (k) = G i νρ µσ [g g + gµρ gνσ − gµν gρσ ] k2 (4) For simplicity, in the following, the metric gµν is understood as η µν . Y.Tang (ITP-CAS) Quantum Gravitational Contributions 6 / 27 Diagrammatical Calculation Faddeev-Popov factor Gauge fixing condition generally is accompanied with ghost. For instance, gauge condition χα [gµν ] = 0 introduces ghost with its action as Lgh = c̄α Qα β cβ , (5) where Qα β is the Faddeev-Popov factor, Qα β = δχα . δβ With harmonic gauge, the ghosts do not interact with other field, we can ignore them. Y.Tang (ITP-CAS) Quantum Gravitational Contributions 7 / 27 Diagrammatical Calculation β function and Counterterms Using the traditional Feynman diagram calculations, we can get the β function by evaluating two and three point functions of gauge fields. These Green functions are generally divergent, so counter-terms are needed to cancel these divergences. The relevant counter-terms to the β function are Tµν = iδab Qµν δ2 , Tµνρ = gfabc Vµνρ qkp δ1 µν Q ≡ qµ qν − q2 gµν νρ µ ρµ ν µν ρ Vµνρ qkp ≡ g (q − k) + g (k − p) + g (p − q) (6) The β function is given by β(g) = gµ Y.Tang (ITP-CAS) ∂ 3 ( δ2 − δ1 ) ∂µ 2 Quantum Gravitational Contributions (7) 8 / 27 Diagrammatical Calculation Feynman Diagrams At one loop level, diagrams are listed below. Also, we only have to keep the quadratic divergences, because logarithmic ones will only contribute to high order operator and then not to β function. (a) (c) Y.Tang (ITP-CAS) (b) (d) Quantum Gravitational Contributions (e) 9 / 27 Diagrammatical Calculation Two-point The two point functions from Fig. 1(a) and Fig. 1(b) are found in terms of ILIs to be [ ∫ [ ] (a)µν 2 T = 2κ dx Qµν I2 + q2 (3x2 − x)I0 ] νρ µρ ρσ µ ν µν 2 µν +q qρ I2 + q qρ I2 − g qρ qσ I2 − q I2 (M2q ) T(b)µν = −3κ2 Qµν I2 (0) where we have defined (8) ∫ 1 , 2 k − M ∫ kµ kν 2 I2µν (M ) ≡ d4 k 2 (k − M2 )2 I2 (M ) ≡ 2 Y.Tang (ITP-CAS) d4 k 2 Quantum Gravitational Contributions (9) 10 / 27 Diagrammatical Calculation Three-piont Three point functions from Fig.1(d) and 1(e) are found, when keeping only the quadratically divergent terms, to be { (d)µνρ 2 T = igκ − Vµνρ qkp I2 (0) + [( ) ∫ ρσ µσ µν νρ ρ µν µ νρ dx g qσ I2 − g qσ I2 + q I2 − q I2 (M2q ) ( ) µσ µ νρ ν ρµ νρ ρµ νσ + g kσ I2 − g kσ I2 + k I2 − k I2 (M2k ) ( ) ]} ρσ ρµ νσ µν ν ρµ ρ µν 2 + g pσ I2 − g pσ I2 + p I2 − p I2 (Mp ) T(e)µνρ = 3igκ2 Vµνρ qkp I2 (0) (10) with M2q = x(x − 1)q2 . Contraction is performed by using FeynCalc. Y.Tang (ITP-CAS) Quantum Gravitational Contributions 11 / 27 Diagrammatical Calculation Regularization schemes To handle with the divergent loop momentum integral, we need a regularization scheme. The lesson from quantum field theory tells us two guide rules. Regularization schemes Cut-off R Dimensional R Loop R Y.Tang (ITP-CAS) Gauge invariance × √ √ Quantum Gravitational Contributions Divergence behaviour √ × √ 12 / 27 Diagrammatical Calculation Cut-off Regularization Now we shall apply the different regularization schemes to the divergent ILIs. In cut-off regularization, when keeping only quadratically divergent terms, one has 1 I2Rµν = gµν I2R , 4 I2R ' i µν 2 g Λ 16π 2 (11) The resulting two and three point functions are (a+b)µν (a)µν (b)µν (d)µνρ (e)µνρ Tcutoff ≡ Tcutoff + Tcutoff [ [ ]] ∫ 1 i i 3 i µν 2 2 2 2 ≈ 2Q κ dx Λ + Λ − Λ =0 2 16π 2 16π 2 2 16π 2 (d+e)µνρ ≡ Tcutoff + Tcutoff ≈ 0 Tcutoff which agrees with the result obtained by Ebert et.al. Y.Tang (ITP-CAS) Quantum Gravitational Contributions 13 / 27 Diagrammatical Calculation Dimensional Regularization In dimensional regularization, where I2R (0) = 0 and R I2µν = 21 gµν I2R , the two and three point functions are found to be ∫ (a+b)µν 2 µν TDR ≈ 4κ Q dxI2R (M2q ), (12) [ ∫ (d+e)µνρ 2 dx (gµν qρ − qµ gνρ )I2R (M2q ) TDR = 2igκ (13) ] ν ρµ νρ µ R 2 ρµ ν ρ µν R 2 +(g k − k g )I2 (Mk ) + (g p − p g )I2 (Mp ) where the regularized quadratic divergence in dimensional regularization behaves as the logarithmic one 2 −i M2q [ − γE + 1 + O(ε)] (14) I2R (M2q )|DR = − 2 16π ε Y.Tang (ITP-CAS) Quantum Gravitational Contributions 14 / 27 Diagrammatical Calculation Loop Regularization We now make a calculation by using loop regularization. R With the consistency condition I2µν = 12 gµν I2R , we obtain for two and three point functions [ ∫ 3 (a+b)µν 2 µν dx − I2R (0) + 2I2R (M2q ) TLR = 2κ Q 2 ] 2 2 R 2 + q (3x − x)I0 (Mq ) (15) [ ∫ 1 (d+e)µνρ 2 TLR = 2igκ dx Vµνρ I R (0) + (gµν qρ − qµ gνρ )I2R (M2q ) 2 qkp 2 ] ν ρµ 2 ρµ ν ρ µν R 2 νρ µ R + (g k − k g )I2 (Mk ) + (g p − p g )I2 (Mp ) (16) Y.Tang (ITP-CAS) Quantum Gravitational Contributions 15 / 27 Diagrammatical Calculation β function from which we can directly read off the two-point and three-point counter-terms δ2κ and δ1κ respectively(only keep the leading quadratically divergent part): [ ] [ ] 2 2 κ 2 1 2 κ 2 1 2 δ2 = κ M − µ , δ1 = κ M −µ 16π 2 c 16π 2 c we obtain the gravitational corrections to the gauge β function ∆β κ = −gκ2 Y.Tang (ITP-CAS) µ2 16π 2 Quantum Gravitational Contributions (17) 16 / 27 Diagrammatical Calculation Running of couplings gi 1.2 1.0 0.8 0.6 0.4 0.2 Μ 5 10 15 20 Log10 GeV Figure: An illustration of gravitational contributions to the running of gauge couplings in the MSSM Y.Tang (ITP-CAS) Quantum Gravitational Contributions 17 / 27 Background Field Method Background Field Method The method is used to calculate the effective action, Γ [ ] ∫ i i δΓ i i exp Γ[ϕ̄] = Dµ[ϕ] exp S[ϕ] − i (ϕ − ϕ̄ ) (18) ~ ~ δ ϕ̄ Γ[ϕ̄] = S[ϕ̄] − ln det Qαβ [ϕ̄] ) ( 1 1 i α + ln det S[ϕ̄],ij + K [ϕ̄]Kj [ϕ̄] (19) 2 2Ω α In Einstein-Gauge system, gµν = ḡµν + κhµν , Aµ = Āµ + aµ (20) coupling renormalization constant Zg is connected to the gauge field renormalization constant ZA = 1 + δA with 1/2 Zg ZA = 1 in Background-Field Gauge, then β function is ∂ ∂ 1 ∂ 0 βgκ = µ g = µ Z−1 δA (21) g g = gµ ∂µ ∂µ 2 ∂µ Y.Tang (ITP-CAS) Quantum Gravitational Contributions 18 / 27 Vilkovisky-DeWitt Formalism Vilkovisky-DeWitt Formalism The diagrammatic approach and traditional background field method share a gauge condition dependent problem. To overcome the gauge condition problem, we shall us Vilkovisky-DeWitt Formalism. The effective action is modified to [ ] ∫ i i δ Γ̂ exp Γ̂[ν i ; ϕ∗ ] = Dµ[ϕ] exp S[ϕ] − i (σ i (ϕ∗ ; ϕ) − ν i ) ~ ~ δν (22) where ν i ≡ hσ i (ϕ∗ ; ϕ)i and the world function, 1 σ[ϕ? ; ϕ] = (length of geodesic from ϕ? to ϕ)2 . 2 Y.Tang (ITP-CAS) Quantum Gravitational Contributions 19 / 27 Vilkovisky-DeWitt Formalism The metric, Gij[ϕ] let Gij [ϕ] denote the metric of the field space, then at one-loop order, the effective action is given by Γ̂[ϕ̄] = S[ϕ̄] − ln det Qαβ [ϕ̄] ( ) 1 1 ,i i α + ln det ∇ ∇j S[ϕ̄] + χ [ϕ̄]χ,j [ϕ̄] 2 2Ω α (23) with ∇i ∇j S[ϕ̄] = S,ij [ϕ̄] − Γ̄kij S,k [ϕ̄]. Here the connection Γ̄kij is determined by Gij [ϕ]. The metric is chosen to be ( ) 1 1 µν ρσ 1 µ(ρ σ)ν Ggµν (x)gρσ (x0 ) = 2 |g(x)| 2 g g − g g δ(x, x0 ) (24) κ 2 GAµ (x)Aν (x0 ) = |g(x)| 2 gµν (x)δ(x, x0 ) 1 Y.Tang (ITP-CAS) Quantum Gravitational Contributions (25) 20 / 27 Vilkovisky-DeWitt Formalism Landau-DeWitt gauge The Vilkovisky-DeWitt formalism is applicable for any gauge condition. In Landau-DeWitt gauge, the calculation is much simpler. Landau-DeWitt gauge conditions (ω = 1) is determined by the gauge transformation and in U(1) gauge it reads 2 µ 1 (∂ hµλ − ∂λ h) + ω(Āλ ∂ µ aµ + aµ F̄µλ ), κ 2 µ χ = − ∂ aµ . χλ = (26) And in this case we can replace Γ̄kij with Γkij , where Γkij is the Christoffel connection determined by ∇i Gjk = 0 Y.Tang (ITP-CAS) Quantum Gravitational Contributions 21 / 27 Vilkovisky-DeWitt Formalism Parameter, ω and v ω is a parameter introduced for a comparison with the traditional background-field method. When ω = 0, it goes to harmonic gauge. Also a parameter, v, is introduced for the connection terms, ∇i ∇j S[ϕ̄] = S,ij [ϕ̄] − vΓkij S,k [ϕ̄] (27) At one-loop order with Landau-DeWitt gauge, the effective action is Γ[ϕ̄] = S[ϕ̄] − ln det Qαβ [ϕ̄] ( ) 1 1 i α i + lim ln det ∇ ∇j S[ϕ̄] + K [ϕ̄]Kj [ϕ̄] . (28) 2 Ω→0 2Ω α Use R I2µν = a2 gµν I2R Y.Tang (ITP-CAS) Quantum Gravitational Contributions (29) 22 / 27 Vilkovisky-DeWitt Formalism One-loop correction The total quadratically divergent one-loop gravitational contribution to the effective action is given ∫ ∫ 1 4 2 2 R1 Γ̂[Āµ ] = d xF̄ + κ CI2 d4 xF̄ 2 (30) 4 4 where the constant C is given by ] 4a2 − 1 ( [ C = v (2ζ − 1) − 4(κ2 ξ − 1) 8 ) 2 + 8(κ ξ − 1) − 16ω − 4 + ω(−1 + 6a2 ) (31) obtain the gravitational correction to the β function 1 ∂ βgκ = gµ δA , 2 ∂µ Y.Tang (ITP-CAS) δA ' −κ2 CI2R Quantum Gravitational Contributions (32) 23 / 27 Vilkovisky-DeWitt Formalism Results Note that we used the notation for the quadratic divergences, R I2µν = a2 gµν I2R . (33) For different regularization schemes, a2 is different. We list the results C v = 1, ω = 1, ζ → 0, ξ → 0 v = 0, ω = 0, ζ = 1/2, ξ = 1/κ2 Y.Tang (ITP-CAS) a2 = 1 4 a2 = 1 2 − 98 0 − 12 Quantum Gravitational Contributions 1 2 24 / 27 Vilkovisky-DeWitt Formalism gi 1.2 1.0 0.8 0.6 0.4 0.2 5 10 15 20 Log10@ Μ D GeV Figure: An illustration of gravitational contributions to the running of gauge couplings in the MSSM Y.Tang (ITP-CAS) Quantum Gravitational Contributions 25 / 27 Summary Summary We considered the gravitational contributions to the runnning of gauge couplings. Both the traditional Diagrammatical and Background Field Method give the same result that the gravity tends to render all gauge theories asymptotically free. However the result is gauge condition dependent. We present the gauge condition independent result in the framework of the Vilkovisky-DeWitt formalism. Y.Tang (ITP-CAS) Quantum Gravitational Contributions 26 / 27 Summary THANK YOU! Y.Tang (ITP-CAS) Quantum Gravitational Contributions 27 / 27