This page is a summary of the discussions during the Fractionalization Phenomena session.

Talk: X. L. Qi

How many mass terms can you generate on edge of QSH system from a 3 component magnetic field? Without disorder it seems like you only get two possible mass terms. With disorder the answer is unknown currently, but in that case you can probably get 3 .

It is difficult to measure exactly half charge (e/2) much easier to measure a charge difference of e/2.

Can you insert generic spin flux without closing the bulk energy gap? In general no, but maybe the set of points where the gap closes has measure zero.

Does the QSH edge have fractional excitations? Yes and no, if you externally impose a magnetic domain wall then no. If interactions are string enough to spontaneously break time-reversal symmetry then it is possible to have e/2 charged "domain-wall" excitations.

Compare spin-charge separation, with Kane and Fu work on Z_2 pumping.

What is quantized in quantum spin Hall?

Talk: C. Kallin

What is the dynamics of the relative phase? It is gapped by a Higgs mechanism.

Relation between S_z conservation and instantons.

There are 2 U(1) phases in each layer.

Mass generation for 1d fermions on edge given by coupling to a U(1) field.

Is there any coupling between spinons and a physical electro-magnetic field? Possibly a small coupling due to a finite system size?

The topological mott insulator paper, and this model start from interactions with no spin-orbit coupling and dynamically generate it. What about starting from interactions + spin orbit coupling and looking for phases of matter?

Talk: Y. Hatsugai Talk: Y. Oreg

Can you think about it as similar to QSH? Not exactly because fractional charge will depend on electron-electron interactions.

Talk: A. Vishwanath

What happens in the case \nu_0=1 but \nu_1\nu_2\nu_3= (000)? This is a strong topological insulator with dirac node near \Gamma point in k-space.

Can disorder and perturbations destroy the dislocations? Can shrink to zero size, but they will act like vortices.

Can you imagine a case where these defects proliferate?

Loops that are closed and isolated do not have exact zero modes.

Can you get more than one set of modes?

For quantum Hall they have to always have a partner defect, cannot exist alone (Fermion doubling problem).

Talk: X. L. QiTalk: C. KallinTalk: Y. HatsugaiTalk: Y. OregTalk: A. Vishwanath