<span class="mw-page-title-main">Ogawa88a</span>
Fabrice P. Lauss𝕪s onsensus Web

Time-of-flight measurement of excitonic polaritons in a GaAs/AlGaAs quantum well. K. Ogawa, T. Katsuyama and H. Nakamura in Appl. Phys. Lett. 53:1077 (1988).  What the paper says!?

This is an important and seminal paper in the physics of polaritons (here as waveguide polaritons), which predates Weisbuch's 1992 report of cavity polaritons[1] by several years, with many relevant results, including a possible indirect observation of line splitting.

The main result is the first observation of exciton-polaritons in quantum wells: «This letter presents the first experimental verification of QWEP's in a GaAslAIGaAs quantum well.» Previously, 2D polaritons were observed as surface polaritons (at the surface of ZnO crystals[2]):

Two-dimensional propagation of excitonic polaritons has been observed as surface excitonic polaritons in ZnO crystals.10,11 In semiconductor quantum wells, however, no experimental evidence for excitonic polaritons has been obtained, although a number of works have dealt with optical properties of two-dimensional excitons in quantum wells.12 Quantum well excitonic polaritons (QWEP's) have only been theoretically studied.13

where Ref. 13 is Ref. [3] and with further explanations that:

This is probably because all low-temperature optical measurements of quantum wells have been performed with incident light propagating perpendicular to the quantum well layers. Quantum well EP's will form in light propagation parallel to the layers,13 since a translational motion of the quantum well excitons along the layers is possible.

They infer a doublet structure from the fact that light and heavy holes seem to couple in an intermediate absorption line:

The broad line lying at 1.625 eV between the above two lines is active in each polarization and is probably composed of hh and lh exciton absorption lines. It is, therefore, suggested that both hh and lh excitons have a doublet line structure in the optical absorption spectrum and that the higher energy part of the hh exciton doublet unexpectedly overlaps with the lower energy part of the lh exciton doublet.

One explanation is: (the other of a "binary fluctuation" is deemed «somewhat unnatural»)

the doublets are produced by the splitting of the exciton lines resulting from QWEP formation. Generally, excitonic polaritons have two dispersion curves corresponding to the upper and lower polariton branches.1 Optical absorption takes place in the polariton picture when incident photon energies are dissipated through inelastic scattering via an excitonphonon interaction.l7

Although just a suggestion, this provides an earlier strong-coupling splitting than Weisbuch's.[1]

Besides absorption, they also measure time-of-flight propagation and observe considerable slow-down of the wavepacket propagation when hitting a polariton resonance.

From this, in principle, they could reconstruct the dispersion (cf. their Eq. (1)), but don't in this paper (they do it in Ref. [4])

This work is further discussed by S. Jorda.[5]

They cite Hopfield's Ref. [6] which is not the standards one, but a good one to understand their "spatial dispersion" discussion. They have a follow-up paper, focusing on polarization,[7] which is more famous although less fundamental (but published in PRL).

References