zeta function regularization



, ,

Surveys, textbooks and lecture notes

  • ,


, ,

    • , , , ,

      • ,

      • ,

      • ,

        • ,

      • and
    • Axiomatizations

          • ,
        • -theorem

    • Tools

      • ,

        • ,

        • ,
    • Structural phenomena

    • Types of quantum field thories

        • ,

        • , ,

        • examples

          • ,
          • ,
        • , , , ,

        • , ,

Theta functions

, , ,


  • ,

in physics

Complex geometry

, ,

  • ,




  • dim=1dim = 1: ,

    • dim=2dim = 2:



In the context of regularization in physics, zeta function regularization is a method/prescription for extracing finite values for traces of powers of Laplace operators/Dirac operators by

  1. considering ss-powers for all values of ss in the complex plane where the naive trace does make sense and then

  2. using analytic continuation to obtain the desired special value at s=1s = 1 – as for zeta functions.

Analytic regularization of propagators

One speaks of analytic regularization (Speer 71) or zeta function regularization (e.g. BCEMZ 03, section 2) if a Feynman propagator/Green's function for a bosonic field, which is naively given by the expression “Tr(1H)Tr\left(\frac{1}{H}\right)” (for HH the given wave operator/Laplace operator) is made well defined by interpreting it as the principal value of the special value at s=1s= 1

Tr reg(1H)pvζ H(1) Tr_{reg} \left(\frac{1}{H}\right) \coloneqq pv\, \zeta_H(1)

of the zeta function which is given by the expression

ζ H(s)Tr(1H) s \zeta_H(s) \coloneqq Tr\left( \frac{1}{H} \right)^s

for all values of ss \in \mathbb{C} for which the right hand side exists, and is defined by analytic continuation elsewhere.

Analogously the zeta function regularization of the Dirac propagator for a fermion field with Dirac operator DD is defined by

Tr reg(DD 2)pvη D(1) Tr_{reg} \left(\frac{D}{D^2} \right) \coloneqq pv\, \eta_D(1)

where η\eta is the eta function of DD.

Functional determinants

Notice that the first derivative ζ H \zeta^\prime_H of this zeta function is, where the original series converges, given by

ζ H (s)= n=1 lnλ n(λ n) s. \zeta_H^\prime(s) = \sum_{n = 1}^\infty \frac{- \ln \lambda_n}{ (\lambda_n)^s} \,.

Therefore the functional determinant of HH (Ray-Singer 71) is the exponential of the zeta function of HH at 0:

Det regHexp(ζ H (0)). Det_{reg} H \coloneqq \exp(- \zeta_H^\prime(0)) \,.

(see also BCEMZ 03, section 2.3)

Via the analytic continuation involved in defining ζ H(0)\zeta_H(0) in the first place, this may be thought of as a regularization of the ill-defined naive definition “ nλ n\prod_n \lambda_n” of the determinant of HH. As such functional determinants often appear in quantum field theory as what is called zeta function regularization.

Higher amplitudes

Accordingly, more general scattering amplitudes are controled by multiple zeta functions (…).


Of Laplace operator on complex torus and Dedekind eta function

For /(τ)\mathbb{C}/(\mathbb{Z}\oplus \tau \mathbb{Z}) a complex torus (complex elliptic curve) equipped with its standard flat Riemannian metric, then the zeta function of the corresponding Laplace operator Δ\Delta is

ζ Δ=(2π) 2sE(s)(2π) 2s(k,l)×(0,0)1|k+τl| 2s. \zeta_{\Delta} = (2\pi)^{-2 s} E(s) \coloneqq (2\pi)^{-2 s} \underset{(k,l)\in \mathbb{Z}\times\mathbb{Z}-(0,0)}{\sum} \frac{1}{{\vert k +\tau l\vert}^{2s}} \,.

The corresponding functional determinant is

exp(E Δ (0))=(Imτ) 2|η(τ)| 4, \exp( E^\prime_{\Delta}(0) ) = (Im \tau)^2 {\vert \eta(\tau)\vert}^4 \,,

where η\eta is the Dedekind eta function.

(recalled e.g. in Todorov 03, page 3)

Zeta regularization for divergent integrals

the zeta regularizatio method can be extended to include also a regularization for the divergent integrals a x mdx \int_{a}^{\infty}x^{m}dx which appears in QFT, this is made by means of the identity

a x msdx=ms2 a x m1sdx+ζ(sm) i=1 ai ms+a ms r=1 B 2rΓ(ms+1)(2r)!Γ(m2r+2s)(m2r+1s) a x m2rsdx\begin{array}{l} \int_{a}^{\infty }x^{m-s} dx =\frac{m-s}{2} \int_{a}^{\infty }x^{m-1-s} dx +\zeta (s-m)-\sum_{i=1}^{a}i^{m-s} +a^{m-s} \\ -\sum_{r=1}^{\infty }\frac{B_{2r} \Gamma (m-s+1)}{(2r)!\Gamma (m-2r+2-s)} (m-2r+1-s)\int_{a}^{\infty }x^{m-2r-s} dx \end{array}

for the case of m=1 m=-1 although the harmonic series has a pole we can regularize by the 2 possibilities

n=0 1n+a=Ψ(a) \sum_{n=0}^{\infty} \frac{1}{n+a} = -\Psi (a) or n=0 1n+a=Ψ(a)+log(a) \sum_{n=0}^{\infty} \frac{1}{n+a} = -\Psi (a)+log(a) in particular
n=1 1n=γ \sum_{n=1}^{\infty} \frac{1}{n} = \gamma Euler-Mascheroni constant, and Ψ(a)=Γ(a)Γ(a) \Psi(a)= -\frac{\Gamma '(a)}{\Gamma (a)}

So within this reuglarization there wouldn’t be any UV ultraviolet divergence

Analytic torsion

The functional determinant of a Laplace operator of a Riemannian manifold acting on differential n-forms is up to a sign in the exponent a factor in what is called the analytic torsion of the manifold.

context/ θ\theta ζ\zeta (= of θ(0,)\theta(0,-)) L zL_{\mathbf{z}} (= of θ(z,)\theta(\mathbf{z},-)) η\eta
/ θ(z,τ)=Tr(exp(τ(D z) 2))\theta(\mathbf{z},\mathbf{\tau}) = Tr(\exp(-\mathbf{\tau} \cdot (D_\mathbf{z})^2)) as function of τ\mathbf{\tau} of Σ\Sigma (hence of ) and / z\mathbf{z}analytically continued of ζ(s)=Tr reg(1(D 0) 2) s= 0 τ s1θ(0,τ)dτ\zeta(s) = Tr_{reg}\left(\frac{1}{(D_{0})^2}\right)^s = \int_{0}^\infty \tau^{s-1} \,\theta(0,\tau)\, d\tauanalytically continued of in z\mathbf{z}: L z(s)Tr reg(1(D z) 2) s= 0 τ s1θ(z,τ)dτL_{\mathbf{z}}(s) \coloneqq Tr_{reg}\left(\frac{1}{(D_{\mathbf{z}})^2}\right)^s = \int_{0}^\infty \tau^{s-1} \,\theta(\mathbf{z},\tau)\, d\tauanalytically continued of in z\mathbf{z} η z(s)=Tr reg(sgn(D z)|D z|) s\eta_{\mathbf{z}}(s) = Tr_{reg} \left(\frac{sgn(D_{\mathbf{z}})}{ { \vert D_{\mathbf{z}} } \vert }\right)^s pvL z(1)=Tr reg(1(D z) 2)pv\, L_{\mathbf{z}}(1) = Tr_{reg}\left(\frac{1}{(D_{\mathbf{z}})^2}\right) / fermionic pvη z(1)=Tr reg(D z(D z) 2)pv\, \eta_{\mathbf{z}}(1)= Tr_{reg} \left( \frac{D_{\mathbf{z}}}{(D_{\mathbf{z}})^2} \right) / 12L z (0)=Z H=12lndet reg(D z 2)-\frac{1}{2}L_{\mathbf{z}}^\prime(0) = Z_H = \frac{1}{2}\ln\;det_{reg}(D_{\mathbf{z}}^2)
θ(z,τ)\theta(\mathbf{z},\mathbf{\tau}) of over J(Σ τ)J(\Sigma_{\mathbf{\tau}}) in terms of covering coordinates z\mathbf{z} on gJ(Σ τ)\mathbb{C}^g \to J(\Sigma_{\mathbf{\tau}})
for a (for ) and (in )
for a , (being the L zL_{\mathbf{z}} for z=0\mathbf{z} = 0 the ) L zL_{\mathbf{z}} of a z\mathbf{z}, expressible “in coordinates” (by ) as a finite-order (for 1-dimensional representations) and generally (via ) by an (for higher dimensional reps) \cdot
for \mathbb{Q} (z=0\mathbf{z} = 0)/ (z=χ\mathbf{z} = \chi a ) (being the L zL_{\mathbf{z}} for z=0\mathbf{z} = 0) of a z\mathbf{z} , expressible “in coordinates” (via ) as a (for 1-dimensional Galois representations) and generally (via ) as an


Original articles include

  • Eugene Speer, On the structure of Analytic Renormalization, Comm. Math. Phys. 23, 23-36 (1971) (Euclid)

  • D. Ray, Isadore Singer, R-torsion and the Laplacian on Riemannian manifolds, Advances in Math. 7: 145–210, (1971) doi:10.1016/0001-8708(71)90045-4, MR 0295381

Modern accounts and reviews include

See also

  • Andrey Todorov, The analogue of the Dedekind eta function for CY threefolds, 2003 pdf

Last revised on January 15, 2016 at 11:13:38. See the history of this page for a list of all contributions to it.