nLab 2d TQFT



Quantum field theory


physics, mathematical physics, philosophy of physics

Surveys, textbooks and lecture notes

theory (physics), model (physics)

experiment, measurement, computable physics



A 2-dimensional TQFT is a topological quantum field theory on cobordisms of dimension 2.


When formulated as an (only) “globally” as 1-functors on a 1-category of cobordisms (see at FQFT for more), then 2d TQFTs have a comparatively simple classification: the bulk field theory is determined by a commutative Frobenius algebra structure on the finite dimensional vector space assigned to the circle (Abrams 96).

However, such global 2d TQFTs with coefficients in Vect do not capture the 2d TQFTs of most interest in quantum field theory, which instead are “cohomological quantum field theories” (Witten 91) such as the topological string A-model and B-model that participate in homological mirror symmetry.

These richer 2d TQFTs are instead local TQFTs in the sense of extended TQFT, i.e. they are (∞,2)-functors on a suitable (∞,2)-category of cobordisms (see at FQFT for more), typically on “non-compact” 2-d cobordisms, meaning on those that have non-vanishing outgoing bounary. As such they are now classified by Calabi-Yau objects in an symmetric monoidal (infinity,2)-category (Lurie 09, section 4.2). For coefficients in the (∞,2)-category of (∞,2)-vector space (i.e. A-∞ algebras with (∞,1)-bimodules between them in the (∞,1)-category of chain complexes), these theories had been introduced under the name “TCFT” in (Getzler 92, Segal 99) following ideas of Maxim Kontsevich, and have been classified in (Costello 04), see (Lurie 09, theorem 4.2.11, theorem 4.2.14).

2d TQFT (“TCFT”)coefficientsalgebra structure on space of quantum states
open topological stringVect k{}_kFrobenius algebra AAfolklore+(Abrams 96)
open topological string with closed string bulk theoryVect k{}_kFrobenius algebra AA with trace map BZ(A)B \to Z(A) and Cardy condition(Lazaroiu 00, Moore-Segal 02)
non-compact open topological stringCh(Vect)Calabi-Yau A-∞ algebra(Kontsevich 95, Costello 04)
non-compact open topological string with various D-branesCh(Vect)Calabi-Yau A-∞ category
non-compact open topological string with various D-branes and with closed string bulk sectorCh(Vect)Calabi-Yau A-∞ category with Hochschild cohomology
local closed topological string2Mod(Vect k{}_k) over field kkseparable symmetric Frobenius algebras(SchommerPries 11)
non-compact local closed topological string2Mod(Ch(Vect))Calabi-Yau A-∞ algebra(Lurie 09, section 4.2)
non-compact local closed topological string2Mod(S)(\mathbf{S}) for a symmetric monoidal (∞,1)-category S\mathbf{S}Calabi-Yau object in S\mathbf{S}(Lurie 09, section 4.2)

Filtrations of the moduli space of surfaces

The following study of the behaviour of 2-dimensional TQFTs in terms of the topology of the moduli spaces of marked hyperbolic surfaces is due to Ezra Getzler. It provides a powerful way to read off various classification results for 2d QFTs from the homotopy groups of the corresponding modular operad.

A A_\infty-monoid objects

Let CoreCore(FinSet) be the core of the category of finite sets. Under union of sets this is a symmetric monoidal category. Then for CC any monoidal category, a symmetric monoidal functor

Φ:Core(FinSet)C \Phi : Core(FinSet) \to C

is a commutative monoid in CC.

Let now CC be a category with weak equivalences, then we can speak of a lax symmetric opmonoidal functor

Φ:Core(FinSet)C \Phi : Core(FinSet) \to C

if the structure maps

Φ(n+m)Φ(m)Φ(n) \Phi(n+m) \stackrel{\simeq}{\to} \Phi(m) \otimes \Phi(n)
ϕ(m)Φ(1) m \phi(m) \stackrel{\simeq}{\to} \Phi(1)^{\otimes m}

are weak equivalences.

Segal called these “Δ\Delta-objects”. Since Carlos Simpson they are called Segal object?s.

There is also Jim Stasheff‘s notion of an A-infinity algebra, given in terms of associahedra K nK_n, which are (n2)(n-2)-dimensional polytopes.

There is naturally a filtration on these guys with

F 0K nF 1K n, F_0 K_n \subset F_1 K_n \subset \cdots \,,

where F 0K nF_0 K_n is the set of vertices, F 1K nF_1 K_n the set of edges, etc.

The collection

{S (K n)} \{ S_\bullet(K_n) \}

of simplicial realizations of the K nK_n form an sSet-operad PP.

For XX a simplicial category that is symmetric monoidal, a PP-algebra over an operad XX in CC is an A A_\infty-monoid object

S (K n)C (X n,X) S_\bullet(K_n) \to C_\bullet(X^{\otimes n}, X)

MacLane‘s coherence theorem says or uses that if CC is an n-category,

we may replace K mK_m here by the nn-filtration F nK mF_n K_m.

Closed 2d quantum field theory

Compactified moduli spaces of Riemann surfaces


(Σ,(z 1,,z n)) (\Sigma, (z_1, \cdots, z_n))

be a compact oriented surface with nn distinct marked points. Write

(Σ,(z 1,,z n)) \mathcal{H}(\Sigma, (z_1, \cdots, z_n))

for the moduli space of hyperbolic metrics with cusps at the (z i)(z_i).

We have

M(Σ,z)=(Σ,z)/Diff +(Σ,z) M(\Sigma, \vec z) = \mathcal{H}(\Sigma, \vec z)/Diff_+(\Sigma, \vec z)


M g,n=Τ g,n/Γ ng, M_{g,n} = \Tau_{g,n} / \Gamma^ng \,,

where Τ g,n\Tau_{g,n} is the Teichmüller space and Γ\Gamma the mapping class group.

Here we can assume that the Euler characteristic χ(Σwithout{z i})<0\chi(\Sigma without \{z_i\}) \lt 0 because otherwise this moduli space is empty.

Fenchel-Nielson coordinates on moduli space

We want to parameterize Teichmüller space by cutting surfaces into pieces with geodesic boundaries and Euler characteristic ξ=1\xi = -1. These building blocks (of hyperbolic 2d geometry) are precisely

  • the 3-holed sphere;

  • the 2-holed cusp;

  • the 1-holed 2-cusp;

  • the 3-cusp

Each surface of genus gg with nn marked points will have

  • 2g2+n2g - 2 + n generalized pants;

  • 3g3+n3 g - 3 + n closed curves.

The boundary lengths i +\ell_i \in \mathbb{R}_+ and twists t it_i \in \mathbb{R} of these pieces for

1i3g3+n 1 \leq i \leq 3g-3+n

constitute the Fenchel-Nielsen coordinates on Teichmüller space Τ\Tau.

Also use θ i:=t i/ i/\theta_i := t_i/\ell_i \in \mathbb{R}/\mathbb{Z}

This constitutes is a real analytic atlas of Teichmüller space. On MM this reduces to coordinates t i/ it_i \in \mathbb{R}/{\ell_i \mathbb{Z}}, and these constitute a real analytic atlas of moduli space.

Allow the lengths i\ell_i to go to 0, but keep the angles θ i\theta_i. The resulting space is a real analytic manifold with corners Τ^\hat \Tau (due to Bill Harvey?) and this constitutes a Borel-Serre bordification? of Τ\Tau.

The mapping class group Γ\Gamma still acts on Τ^\hat \Tau and the quotient M^\hat M is an orbifold with corners, inside which still sits our moduli space MM.

Kimura-Stasheff-Voronov: add a choice of directions at each nodal point in Σ\Sigma. This removes all automorphisms and hence we no longer have to deal with an orbifold.

This yields the classifying stack 𝒫 g,n\mathcal{P}_{g,n} for Γ g,n\Gamma_{g,n}

Then the collection

{𝒫 g,n} \{ \mathcal{P}_{g,n} \}

is a modular operad: the operad that describes gluing of marked surfaces at marked points together with the informaiton on how to glue marked points of a single surface to each other.

A 2-dimensional closed TQFT is an algebra over an operad over this in a simplicial category, in the above sense.

This involves either the de Rham complex on 𝒫 g,n\mathcal{P}_{g,n} or S (𝒫 g,n)S_\bullet(\mathcal{P}_{g,n}).


F k𝒫 g,n:={[Σ]|...} F_k \mathcal{P}_{g,n} := \left\{ [\Sigma] | ... \right\}

where Σ\Sigma has 2g2+nk\geq 2g-2+n-k spheres as components (after cutting along zero-length closed curves).

So for instance

  • F 0𝒫 g,nF_0 \mathcal{P}_{g,n} is the pants-decomposition;

  • F 1𝒫 g,nF_1 \mathcal{P}_{g,n} is decompositions into pants and one piece being the result of either gluing two pants to each other or of gluing two circles of a single pant to each other.

    This F 1..F_1 .. is a connected space, due to a theorem by Hatcher-Thurston. Notice This is equivalent to the familiar statement that a closed 2d TFT is a commutative Frobenius algebra.

  • F 2𝒫 g,nF_2 \mathcal{P}_{g,n} is the decomposition into pieces as before together with one two-holed torus or one five-holed sphere.

    This space has the space fundamental group as 𝒫 g,n\mathcal{P}_{g,n}.

    This is equivalent to the theorem by Moore and Seiberg about categorified 2-d TFT.


(Ezra Getzler)

The inclusion

F k𝒫 g,n𝒫 g,n F_k \mathcal{P}_{g,n} \hookrightarrow \mathcal{P}_{g,n}

is kk-connected.

Here a map XYX\to Y is kk-connected if

  • π 0(X)π 0(Y)\pi_0(X) \to \pi_0(Y) is surjective;

  • π i(X,x)π i(Y,f(x))\pi_i(X,x) \to \pi_i(Y,f(x)) is a bijection for i<ki \lt k and surjective i=ki = k.

This means precisely that the mapping cone is kk-connected.


Use the cellular decomposition of moduli space g,1\mathcal{M}_{g,1} following Mumford, Thurston, Harer, Woeditch-Epstein, Penner.

Some other versions of this:

F kΤ g,nΤ g,n F_k \Tau_{g,n} \to \Tau_{g,n}

is kk-connected.

One can also use Deligne-Mumford compactifications

F k¯ g,n¯ g,n F_k \bar \mathcal{M}_{g,n} \to \bar \mathcal{M}_{g,n}

and this is also kk-connected.

Open-closed case



The folklore result that global closed 2d TQFTs with coefficients in Vect are equivalent to commutative Frobenius algebra structures is proven rigorously in

The classification result for open-closed 2d TQFTs was famously announced and sketched in

Textbook account:

A picture-rich description of what’s going on:


The local (extended TQFT) version of 2d TQFT which captures the topological string was mathematically introduced under the name “TCFT”.

The concept is essentially a formalization of what used to be called cohomological field theory in

  • Edward Witten, Introduction to cohomological field theory, InternationalJournal of Modern Physics A, Vol. 6,No 6 (1991) 2775-2792 (pdf)

The definition was given independently by


  • Graeme Segal, Topological field theory , (1999), Notes of lectures at Stanford university. (web). See in particular lecture 5 (“topological field theory with cochain values”).

The classification of TCFTs (i.e. “non-compact” local (extended 2d TQFT)) by Calabi-Yau A-infinity categories is due to

following conjectures by Maxim Kontsevich, e.g.

  • Maxim Kontsevich, Homological algebra of mirror symmetry , in Proceedings of the International Congress of Mathematicians, Vol. 1, 2 (Zürich, 1994), pages 120–139, Basel, 1995, Birkhäuser.

The classification of local (extended) 2d TQFT (i.e. the “compact” but fully local case) is spelled out in

This classification is a precursor of the full cobordism hypothesis-theorem. This, and the reformulation of the original TCFT constructions in full generality is in

Last revised on August 21, 2023 at 07:23:16. See the history of this page for a list of all contributions to it.