To the Editor —

In a recent Article published in Nature Photonics1, Sich et al. reported the observation of bright polariton solitons in a semiconductor microcavity. As part of their analysis, the authors discussed the response time of polariton solitons in comparison with the response time of what they called “light-only solitons in semiconductor cavity lasers” or “pure-light cavity solitons in wide-aperture semiconductor lasers”, citing work of Barland et al.2, Pedaci et al.3 and Ackemann et al.4

In this correspondence, we want to remark that the cavity solitons studied by Barland et al.2 and Pedaci et al.3, and reviewed by Ackemann et al.4, are actually composite structures consisting intrinsically of both a light and a material component, and that referring to them as 'pure light' is therefore not appropriate.

Indeed, there is of course no nonlinear optics without a medium and any modulation of the light field will have a counterpart in terms of material variables, be it electronic states, Zeeman states, twist angles of liquid-crystal directors, carrier populations or coherences. Depending on the relative timescales involved, the material dynamics might be adiabatically eliminated, making this connection less obvious but not less real.

In the specific case of semiconductor microcavities referred to by Sich et al.1, the relevance of the material component has been envisaged since the very first studies of cavity solitons in semiconductor systems2,5, and the impact of material timescales on cavity solitons nucleation and motion has actually been discussed by Pedaci et al.3 and Ackemann et al.4, among others. For instance, dedicated numerical analyses6 have shown that forcing a spatial separation of the material and optical components of a cavity soliton causes it to disappear, further demonstrating the intrinsic 'light–matter' nature of cavity solitons in semiconductor microcavity models. Perhaps even more significantly, this composite nature has been exploited in experiments that involve controlling cavity solitons by optically manipulating their material component7.

In view of these observations, we believe that the terms 'light only' and 'pure-light' do not accurately reflect the nature of semiconductor cavity solitons and are therefore not helpful in framing a quantitative or qualitative comparison with polariton solitons.