Skip to main content
Log in

\(2d\,\) Fu–Kane–Mele invariant as Wess–Zumino action of the sewing matrix

  • Published:
Letters in Mathematical Physics Aims and scope Submit manuscript

Abstract

We show that the Fu–Kane–Mele invariant of the 2d time-reversal invariant crystalline insulators is equal to the properly normalized Wess–Zumino action of the so-called sewing-matrix field defined on the Brillouin torus. Applied to 3d, the result permits a direct proof of the known relation between the strong Fu–Kane–Mele invariant and the Chern–Simons action of the non-Abelian Berry connection on the bundle of valence states.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. All bundle gerbes and line bundles considered below come equipped with a Hermitian structure and a Hermitian connection and their isomorphisms are assumed to respect those structures.

References

  1. Alvarez, O.: Topological quantization and cohomology. Commun. Math. Phys. 100, 279–309 (1985)

    Article  ADS  MathSciNet  MATH  Google Scholar 

  2. Carpentier, D., Delplace, P., Fruchart, M., Gawȩdzki, K.: Topological index for periodically driven time-reversal invariant 2D systems. Phys. Rev. Lett. 114, 106806 (1–5) (2015)

  3. Carpentier, D., Delplace, P., Fruchart, M., Gawȩdzki, K., Tauber, C.: Construction and properties of a topological index for periodically driven time-reversal invariant 2D crystals. Nucl. Phys. B 896, 779–834 (2015)

    Article  ADS  MATH  Google Scholar 

  4. Chatterjee, D.S.: On gerbes. Cambridge University thesis (1998)

  5. De Nittis, G., Gomi, K.: Classification of “quaternionic” Bloch-bundles: topological quantum systems of type AII. Commun. Math. Phys. 339, 1–55 (2015)

    Article  ADS  MathSciNet  MATH  Google Scholar 

  6. Fiorenza, D., Monaco, D., Panati, G.: \(\mathbb{Z}_2\) invariants of topological insulators as geometric obstructions. Commun. Math. Phys. 343, 1115–1157 (2016)

    Article  ADS  MathSciNet  MATH  Google Scholar 

  7. Freed, D., Moore, G.: Twisted equivariant matter. Ann. H. Poincaré 14, 1927–2023 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  8. Fu, L., Kane, C.L.: Time reversal polarization and a \({Z}_{2}\) adiabatic spin pump. Phys. Rev. B. 74(1–13), 195312 (2006)

  9. Fu, L., Kane, C.L., Mele, E.J.: Topological insulators in three dimensions. Phys. Rev. Lett. 98(1–4), 106803 (2007)

  10. Gawȩdzki, K.: Topological actions in two-dimensional quantum field theory. In: ’t Hooft, G., Jaffe, A., Mack, G., Mitter, P., Stora, R. (eds.) Non-Perturbative Quantum Field Theory, pp. 101–142. Plenum Press, New York, London (1988)

  11. Gawȩdzki, K.: Bundle gerbes for topological insulators. arXiv:1512.01028 [math-ph]

  12. Gawȩdzki, K., Reis, N.: WZW branes and gerbes. Rev. Math. Phys. 14, 1281–1334 (2002)

    Article  MathSciNet  MATH  Google Scholar 

  13. Gawȩdzki, K., Suszek, R.R., Waldorf, K.: WZW orientifolds and finite group cohomology. Commun. Math. Phys. 284, 1–49 (2008)

    Article  ADS  MathSciNet  MATH  Google Scholar 

  14. Gawȩdzki, K., Suszek, R.R., Waldorf, K.: Bundle gerbes for orientifold sigma models. Adv. Theor. Math. Phys. 15, 621–688 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  15. Kane, C.L., Mele, E.J.: \(\mathbb{Z}_2\) topological order and the quantum spin Hall effect. Phys. Rev. Lett. 95(1–4), 146802 (2005)

  16. Murray, M.K.: Bundle gerbes. J. Lond. Math. Soc. 2(54), 403–416 (1996)

    Article  MathSciNet  MATH  Google Scholar 

  17. Panati, G.: Triviality of Bloch and Bloch-Dirac bundles. Ann. H. Poincaré 8, 995–1011 (2007)

    Article  MathSciNet  MATH  Google Scholar 

  18. Qi, X.-L., Hughes, T.L., Zhang, S.-C.: Topological field theory of time-reversal invariant insulators. Phys. Rev. B. 78(1–43), 195424 (2008)

  19. Rudner, M.S., Lindner, N.H., Berg, E., Levin, M.: Anomalous edge states and the bulk-edge correspondence for periodically driven two-dimensional systems. Phys. Rev. X. 3(1–15), 031005 (2013)

  20. Schreiber, U., Schweigert, C., Waldorf, K.: Unoriented WZW models and holonomy of bundle gerbes. Commun. Math. Phys. 274, 3164 (2007)

    Article  MathSciNet  MATH  Google Scholar 

  21. Witten, E.: Global aspects of current algebra. Nucl. Phys. B 223, 422–432 (1983)

    Article  ADS  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Krzysztof Gawȩdzki.

Additional information

K. Gawȩdzki: Directeur de recherche émérite.

Appendix

Appendix

We establish here (in the reversed order) the properties of the WZ amplitude \(\,\exp [\mathrm {i}S_\mathrm{WZ}(w)]\,\) claimed in Remark in Sect. 2.

First, for a smooth family of fields \(\,w_t(k)\,\) defined on \(\,\mathbb T^2\,\) such that \(\,w_t\circ \vartheta =-w_t^\mathrm{T}\,\) for all \(\,t\,\) (with \(\,\vartheta \,\) induced by \(\,k\mapsto -k\))  the well known formula for the derivative of the WZ action, easily following from the definition (2.10), gives:

$$\begin{aligned} \frac{_d}{^{dt}}S_\mathrm{WZ}(w(t))= & {} \frac{_1}{^{4\pi }}\int _{\mathbb T^2}\hbox {tr}\big ( (w^{-1}_t\partial _tw_t)(w_t^{-1}dw_t)^2\big )\nonumber \\= & {} \frac{_1}{^{4\pi }}\int _{\mathbb T^2}\vartheta ^* \hbox {tr}\big ((w^{-1}_t\partial _tw_t)(w_t^{-1}dw_t)^2\big )\nonumber \\= & {} \frac{_1}{^{4\pi }}\int _{\mathbb T^2} \hbox {tr}\big (((w^\mathrm{T}_t)^{-1}\partial _tw_t^\mathrm{T})((w_t^\mathrm{T})^{-1}dw_t^\mathrm{T})^2\big )\nonumber \\= & {} \frac{_1}{^{4\pi }}\int _{\mathbb T^2} \hbox {tr}\big (((w^\mathrm{T}_t)^{-1}\partial _tw_t^\mathrm{T})((w_t^\mathrm{T})^{-1}dw_t^\mathrm{T})^2\big )^\mathrm{T}\nonumber \\= & {} -\frac{_1}{^{4\pi }}\int _{\mathbb T^2} \hbox {tr}\big (((dw_t)w_t^{-1})^2(\partial _tw_t)w_t^{-1}\big )\nonumber \\= & {} -\frac{_1}{^{4\pi }}\int _{\mathbb T^2} \hbox {tr}\big ((w_t^{-1}\partial _tw_t)(w_t^{-1}dw_t)^2\big )\,=\,0, \end{aligned}$$
(7.1)

implying the invariance of the WZ amplitude \(\,\exp [\mathrm {i}S_\mathrm{WZ}(w)]\,\) under smooth deformations of \(\,w\,\) preserving the symmetry (2.5).

Next, let us consider a change of the trivialization of the valence bundle

$$\begin{aligned} e_i(k)=\sum \limits _{i'}U_{i'i}(k)\,e'_{i'}(k) \end{aligned}$$
(7.2)

with unitary \(\,U(k)\).  For the sewing matrices, this gives

$$\begin{aligned} w_{ij}(k)= & {} \big \langle e_i(-k)|\theta e_j(k)\big \rangle = \sum \limits _{i',j'}\overline{U_{i'i}(-k)\,U_{j'j}(k)} \,\big \langle e'_{i'}(-k)|\theta e'_{j'}(k)\big \rangle \nonumber \\= & {} \sum \limits _{i',j'}U^{-1}_{\ ii'}(-k)\,U^{-1}_{\ jj'}(k)\,w'_{i'j'}(k)\quad \end{aligned}$$
(7.3)

so that \(\,w(k)=U^{-1}(-k)w'(k)(U^{-1}(k))^\mathrm{T}\,\) or

$$\begin{aligned} w'(k)=U(-k)w(k)U(k)^\mathrm{T}. \end{aligned}$$
(7.4)

The map \(\,\mathbb T^2\ni k\mapsto U(k)\in U(n)\,\) can be smoothly contracted to the one

$$\begin{aligned} \mathbb T^2\ni k\mapsto U_{n_1,n_2}(k)=\mathrm{diag}[\mathrm{e}^{\mathrm {i}(n_1k_1+n_2k_2)},1, \dots ,1], \end{aligned}$$
(7.5)

where \(\,n_1,n_2\in \mathbb Z\,\) are the winding numbers of \(\,\det {U(k)}\,\) along the basic cycles of \(\,\mathbb T^2\).  By the previous argument, it is enough to check the relation \(\,\exp [\mathrm {i}S_\mathrm{WZ}(w')] =\exp [\mathrm {i}S_\mathrm{WZ}(w)]\,\) for \(\,U(k)=U_{n_1,n_2}(k)\).  Besides, by an \(\,SL(2,\mathbb Z)\,\) change of variables \(\,k\),  one may achieve that \(\,n_1=0\).  Let \(\,\mathcal{D}\,\) be the unit disc in \(\,\mathbb C\).  Then \(\,\mathcal{B}=\mathcal{D}\times S^1\,\) is a 3-manifold with the boundary \(\,S^1\times S^1\cong \mathbb T^2\).  Since \(\,\det {w(k)}\,\) has no windings, there exists a smooth contraction

$$\begin{aligned}{}[0,1]\times \mathbb T^2\ni (r,k)\,\longmapsto \, W(r,k)\in U(n) \end{aligned}$$
(7.6)

such that \(\,W(1,k)=w(k)\,\) and \(\,W(r,k)=I\,\) for \(\,r\,\) close to zero. We shall identify \(\,W\,\) with a smooth map defined on \(\,\mathcal{B}\,\) by setting

$$\begin{aligned} W(r\mathrm{e}^{\mathrm {i}k_1},\mathrm{e}^{\mathrm {i}k_2})=W(r,k). \end{aligned}$$
(7.7)

Consider two other smooth maps \(\,V_{1,2}:\mathcal{B}\rightarrow U(n)\,\) given by

$$\begin{aligned} V_1(r\mathrm{e}^{\mathrm {i}k_1},\mathrm{e}^{\mathrm {i}k_2})=\mathrm{diag}[\mathrm{e}^{-\mathrm {i}n_2k_2},1,\dots ,1], \qquad V_2(r\mathrm{e}^{\mathrm {i}k_1},\mathrm{e}^{\mathrm {i}k_2})=\mathrm{diag}[\mathrm{e}^{\mathrm {i}n_2k_2},1,\dots ,1].\nonumber \\ \end{aligned}$$
(7.8)

The product map \(\,W'=V_1WV_2:\mathcal{B}\rightarrow U(n)\,\) is a smooth extension of \(\,w'(k)=U_{0,n_2}(-k)w(k)U_{0,n_2}(k)^\mathrm{T}\,\) to the interior of \(\,\mathcal{B}\,\) so that, by Witten’s prescription,

$$\begin{aligned} S_\mathrm{WZ}(w')&=\int _\mathcal{B}(W')^*H=\int _\mathcal{B}(V_1WV_2)^*H\nonumber \\&=\int _\mathcal{B}\Big (V_1^*H+ W^*H+V_2^*H+\frac{_1}{^{4\pi }}\,d\,\hbox {tr}\big ((V_1^{-1}dV_1)WV_2d(WV_2)^{-1}\nonumber \\&\quad + (W^{-1}dW)V_2dV_2^{-1}\big )\Big ),\quad \end{aligned}$$
(7.9)

where we applied twice the formula

$$\begin{aligned} (W_1W_2)^*H=W_1^*H+W_2^*H+\frac{_1}{^{4\pi }}\,d\,\hbox {tr}\big ((W_1^{-1}dW_1)W_2dW_2^{-1} \big ) \end{aligned}$$
(7.10)

holding for two \(\,U(n)\)-valued maps \(\,W_{1,2}\,\) on the same domain.  Since \(\,V_1^*H=0=V_2^*H\,\) for dimensional reasons,  we infer that

$$\begin{aligned}&S_\mathrm{WZ}(w')-S_{WZ}(w)\nonumber \\&\quad =\,\frac{_1}{^{4\pi }}\int _{\partial \mathcal{B}}\hbox {tr}\Big ((V_1^{-1}dV_1)W(V_2dV_2^{-1})W^{-1}\nonumber \\&\qquad -(V_1^{-1}dV_1)(dW)W^{-1}+(W^{-1}dW)V_2dV_2^{-1}\Big )\nonumber \\&\quad =\,\frac{_1}{^{4\pi }}\int _{\mathbb T^2}\hbox {tr}\Big (\mathrm{diag}[-\mathrm {i}n_2dk_2,0,\dots ,0] \,\,w\,\,\mathrm{diag}[-\mathrm {i}n_2dk_2,0,\dots ,0]\,w^{-1}\nonumber \\&\qquad -\mathrm{diag}[-\mathrm {i}n_2dk_2,0,\dots ,0]\,(dw)w^{-1} +(w^{-1}dw)\,\mathrm{diag}[-\mathrm {i}n_2dk_2,0,\dots ,0]\Big ).\nonumber \\ \end{aligned}$$
(7.11)

Changing the variables \(\,k\mapsto -k\,\) in the integral on the right hand side and using the symmetry (2.5) of \(\,w\,\) together with the invariance of trace under the transposition,  one shows that the integral in question is equal to its negative, hence it vanishes.

Finally, note that \(\,\widetilde{W}(r,k)=-W(r,-k)^\mathrm{T}\,\) is also a contraction of \(\,w(k)\,\), and as \(\,W\),  it may be regarded as defined on \(\,\mathcal{B}\).  Then

$$\begin{aligned} S_\mathrm{WZ}(w)=\int _\mathcal{B}\tilde{W}^*H=\int _\mathcal{B}(-W^\mathrm{T})^*H=-\int _\mathcal{B}W^*H=-S_\mathrm{WZ}(w) \end{aligned}$$
(7.12)

modulo \(\,2\pi \),  implying  that \(\,\exp [\mathrm {i}S_\mathrm{WZ}(w)] =(\exp [\mathrm {i}S_\mathrm{WZ}(w)])^{-1}=\pm 1\).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gawȩdzki, K. \(2d\,\) Fu–Kane–Mele invariant as Wess–Zumino action of the sewing matrix. Lett Math Phys 107, 733–755 (2017). https://doi.org/10.1007/s11005-016-0922-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11005-016-0922-y

Keywords

Mathematics Subject Classification

Navigation