nLab string scattering amplitude



String theory

Quantum field theory




In perturbative string theory scattering amplitudes are defined as in quantum field theory, but with n-point functions of 1-dimensional worldline theories (Feynman diagrams) replaced by those of worldsheet 2d CFTs.

graphics grabbed from Jurke 10



The amplitudes are thought (see the commented references below) to come out term-wise (for each “loop order” hence for each genus and number of punctures of (super-)Riemann surfaces) finite (at least UV-finite), hence renormalized: the higher string oscillations may be seen as providing canonical counterterms for the massless excitations in the effective field theory. In this sense string theory provides a UV-completion of these effective field theories (supergravity coupled to Yang-Mills theory).


The full perturbation series is the sum of all these (finite) contributions over the genera of Riemann surfaces (the “loop orders”). This sum diverges, even if all loop orders are finite. Notice though, that a non-trivial perturbative QFT is not supposed to have a finite radius of convergence of its scattering amplitudes, since that would imply convergence also for negative coupling constant, which is physically unreasonable. (For the bosonic string the perturbation series has apparently been explicitly shown not to be Borel resummable.) For more on this see at non-perturbative effect and string theory FAQ – Is the divergence of the pertubation series fatal?.


A string scattering amplitude is called UV-finite at a given loop order (genus of a Riemann surface Σ\Sigma with nn marked points/string insertions) if the correlation function ϕ 1,,ϕ n Σ\langle \phi_1, \cdots, \phi_n\rangle_{\Sigma} is finite for every single such Riemann surface. The actual string amplitude at order (g,n)(g,n) though is the averaging of this over all possible conformal structures on Σ\Sigma, hence the integration of the correlation function, as a function on the conformal structure, over the compactified moduli space of conformal structures g,n\mathcal{M}_{g,n} (a Deligne-Mumford stack).

If also this integral is finite, hence if the total measure on the moduli space of conformal structures is finite, then one says the amplitude is IR-finite.

This distinction between UV-finiteness and IR-finiteness is not always highlighted in all of the articles below. All authors argue that the string is UV-finite to all order. The IR-finiteness is only discussed much more recently at low loop order.

IR non-finiteness is not physically fatal. For instance if a perturbative theory of quantum gravity develops a cosmological constant perturbatively, then the perturbation series will be IR-divergent, signifying the fact that background spacetime without cosmological constant is no longer a solution to the quantum-corrected equations of motion. Nevertheless, these potential IR-divergences seem to be absent for the superstring perturbation series. For the cosmological constant case this can already be seen from the fact that the effective QFT of type II supergravity etc. does not admit a cosmological constant, for that would violate supersymmetry.

Open-closed scattering duality and KLT relations

The open/closed string duality implies certain relations in string scattering amplitudes that in the point-particle limit induces relations between scattering amplitudes in Yang-Mills theory and in gravity. These are the KLT relations in QFT. See in particular Mafra-Schlotterer 18a, Mafra-Schlotterer 18b, Mafra-Schlotterer 18c.

Twistor string amplitudes and MHV amplitudes in Yang-Mills theory

The scattering amplitudes in twistor string theory induce the MHV amplitudes in (super-)Yang-Mills theory. See at string theory results applied elsewhere in the section Application to QCD – Scattering amplitudes.

pp-Adic formulation

The Veneziano amplitude (open bosonic string tree-level scattering) has an equivalent formulation as the inverse product over all prime numbers pp of an amplitude computed not by an integral in the real but in the p-adic numbers. For other open string amplitudes this holds up to some regularization. This is the topic of p-adic string theory, see there for more details.




A comprehensive account of the superstring S-matrix may be obtained from combining the general idea presented in

  • Joseph Polchinski, section 12.5 of vol 2 of String theory, Cambridge Monographs on Mathematical Physics (2001)

with the technical details laid out in

Survey of the tree level string scattering amplitudes includes

See also

On string scattering amplitudes in view of the S-matrix bootstrap:

  • Andrea Guerrieri, Joao Penedones, Pedro Vieira, Where is String Theory? (arXiv:2102.02847)

For more references see also at string theory results applied elsewhere.

Superstring scattering

A review of superstring scattering amplitudes is in the last section of (Staessens-Vernocke 10). A general discussion of the problem of superstring amplitudes is in

On analycity of the superstring S-matrix:

  • Corinne de Lacroix, Harold Erbin, Ashoke Sen, Analyticity and Crossing Symmetry of Superstring Loop Amplitudes (arXiv:1810.07197)

Survey of the presence and role of divergences includes

Discussion of superstring scattering amplitudes in terms of pure spinors (Berkovits superstring) with emphasis on KLT relations:

For more see also at superstring field theory, such as

  • Corinne de Lacroix, Harold Erbin, Sitender Pratap Kashyap, Ashoke Sen, Mritunjay Verma, Closed Superstring Field Theory and its Applications, International Journal of Modern Physics AVol. 32, No. 28n29, 1730021 (2017) (arXiv:1703.06410)

The 1-loop amplitudes in type II string theory have been discussed in

and for heterotic string theory in

  • David Gross, J.A. Harvey, E.J. Martinec and R. Rohm, Heterotic String Theory (II). The interacting heterotic string, Nucl. Phys. B267 (1986) 75.

The description of 2-loop amplitudes, including the Berezin integral over the super-moduli space of super Riemann surfaces in superstring theory:

Review of this work:

  • A.Morozov, NSR Superstring Measures Revisited, JHEP0805:086,2008 (arXiv:0804.3167)

Further development in:

The technical issue of the moduli space of super Riemann surfaces of higher genus (for higher loop string scattering amplitudes) is discussed in

Review of the scattering of massive modes in type IIB string theory:

  • Nicholas Agia, Massive Type IIB Superstrings Part I: 3- and 4-Point Amplitudes [arXiv:2309.11538]

See also:

Higher order corrections

  • Christopher Pope, Higher-order corrrections in String and M-theory and Generalized Holonomies, December 2006 (pdf)

On finiteness

Here is a commented list of references on the degreewise finiteness of string scattering amplitudes.

Finiteness of bosonic string scattering

Introductory lecture notes include

  • Wieland Staessens, Bert Vercnocke, Lectures on Scattering Amplitudes in String Theory, Lecture notes based on lectures given at the fifth Modave School on Mathematical Physics (August 2009) (arXiv:1011.0456)

Discussion of the term-wise finiteness of the bosonic string scattering amplitudes is in

  • L. Clavelli, S. T. Jones, Finiteness of the bosonic string in fewer than 26 dimensions, Phys. Rev. D 39, 3795–3797 (1989) (SPIRE)

There are also arguments in

  • Simon Davis, On the domain of string perturbation theory, Classical and Quantum Gravity, Volume 6, Issue 12, pp. 1791-1803 (1989) (web)

Finiteness of superstring scattering

Finiteness of heterotic and type II superstring nn-point functions in flat spacetime is argued for in

General finiteness of superstring amplitudes is discussed in

  • Stanley Mandelstam, The nn Loop String Amplitude: Explicit Formulas, Finiteness and Absence of Ambiguities, Phys. Lett. B277 (1992) 82. (inspire)

which shows that a certain divergence which could appear does not.


  • A. Restuccia, J. Taylor, Finiteness of type II superstring amplitudes, Physics Letters B, Volume 187, Issues 3–4, 26 March 1987, Pages 267–272

argues finiteness of the superstring amplitudes at each order.


Arguments for the finiteness of superstring scattering amplitudes to all loop order based on the Berkovits superstring-formulation have been promoted in

(In footnote 2 this article claims that the claimed proofs of the same statement by G.S Danilov in hep-th/9801013, hep-th/0312177 are not in fact proofs.)

Discussion of 2-loop amplitudes from holomorphy arguments is in

See also

Graviton scattering

Computation of graviton scattering amplitudes (perturbative quantum gravity):


Discussion via AdS/CFT beyond the SCFT planar limit, using the conformal bootstrap:


On point-particle limit via tropical geometry

Scattering amplitudes of highly excited strings:

  • Dimitri P. Skliros, Edmund J. Copeland, Paul M. Saffin, Highly Excited Strings I: Generating Function (arxiv:1611.06498)

Last revised on March 15, 2024 at 07:14:37. See the history of this page for a list of all contributions to it.