Polariton Phenomena in Semiconductor Microcavities

Microcavities group members. February 2013.

Researchers

Academic Staff:	Maurice Skolnick David Whittaker Dmitry Krizhanovskii
Post-Docs:	Paul Walker François Fras Maksym Sich
PhD Students:	Magdalena Durska Lloyd Tinkler Jasmin Chana

Schematic of a Microcavity
Fig. 1: Schematic diagrams of planar microcavity structure with quantum wells embedded in the antinodes of confined photonic mode.

Fig. 2 (a, b): Dispersion curves of a planar microcavity. LP, UP: lower and upper polariton branches, on logarithmic scale in a) and linear in b) for non-resonant and resonant excitation respectively. a) corresponds to the formation of a quasiequilibrium BEC and b) to the OPO high density signal and idler polariton phases.

Semiconductor quantum microcavities (QMC) are one-dimensional planar structures grown by layer-by-layer epitaxial techniques. The cavity plays the role of a "defect" in a periodic stack of layers providing strong localisation of light along the growth direction, the so-called Fabry-Perot localised mode. Strongly magnified optical fields of the Fabry-Perot mode can interact with excitonic states of quantum wells (QWs) grown in the cavity. In the strong coupling regime this interaction leads to the creation of a new type of quasiparticle, the "microcavity polariton". Their optical properties are of great fundamental interest because these quasi-particles possess properties of light (photons) and matter (excitons) at the same time. The polaritons can undergo Bose-Einstein condensation to macroscopically occupied states in well defined regions of k-space. Reviews by the group can be found in Semiconductor Science and Technology 13, 645 (1998) [link], IEEE Sel Topics in Quantum Electronics 8, 1060 (2002) [link].

There are a few ways to excite polariton condensates using external laser sources. Firstly, MC structures can be excited at high energy well above exciton level (Fig. 2(a)), the free carriers which are created relax first into the lower energy polariton and, eventually, a polariton Bose-Einstein condensate at k around zero is formed at sufficiently high pump densities.

The second way to create a polariton condensate at k=0 is to use laser pumping resonant with the lower polariton branch shown in Fig. 2(b). Direct polariton-polariton scattering from the pump results in condensed signal and idler states, which form an optical parametric oscillator (OPO). The OPO signal has properties similar to BECs formed under nonresonant pumping, since its phase is independent of that of the pump.

Selected Results:

Polaritons in waveguides

Waveguide structure
Fig. 3. Sample layer structure (not to scale).

Polariton dispersion
Fig. 4. Angular profile of PL at low excitation power.

In this work, we directly observe the strong coupling between inorganic quantum well excitons and the guided mode of a planar film waveguide. Compared to using distributed Bragg reflectors (DBRs), optical confinement by total internal reflection (TIR) in principle provides for smaller losses and also larger Rabi splitting through better spatial overlap of optical field and quantum well(s). Waveguides naturally operate at large in-plane wavevectors and group velocities so that polaritons should propagate large distances within their lifetime. To couple light in and out of the structure, we employ a grating coupler, which consists of a periodic modulation of the guidecladding layer interface: see figure 3.

Where the dispersion of the electromagnetic mode crosses that of the excitons and the coupling strength is large enough, there is an avoided crossing of the modes, which is the signature of the strong coupling regime. Fig. 4(a) shows the angle-resolved PL spectrum for excitation and detection at the same region within a grating patterned area, which clearly show the anti-crossing. Observed Rabi splitting of 5–6 meV comes from only a single quantum well. The polaritons propagate with a group velocity of 26 µm/ps on resonance and with a lifetime of 11.4 ps.

References:

Appl. Phys. Lett. 102, 012109 (2013).

Bright polarion solitons

Fig. 5. A reconstruction of the soliton motion in plane of a microcavity.

Experimental scheme for soliton excitation
Fig. 6. Experimental scheme of the soliton experiment.

In this work we for the first time demonstrate bright 2D polariton solitons. The polariton solitons are localised nondiffracting wavepackets. They travel at speeds of approximately 1.7 µm/ps, and their size is of the order of 5 µm with corresponding broad spectrum in energy-momentum space. Fig. 5 shows a video of reconstruction of the motion of the soliton.

Polariton solitons are dissipative solitons. They are triggered and sustained by input lasers. The ‘Pump’ laser is focused onto a large spot (100 µm), whilst the pulsed writing beam is focused onto a small region (7 µm). Each pulse of the writing beam triggers a soliton which travels within the pump spot, as shown on Fig. 6.

References:

Nature Photon. 6, 50–55 (2012).

Vortices/antivortices in the polariton condensates

(Left) Real space image of signal emission at k=0. (Right) Interference pattern demonstrating vortex state.
Fig. 7. a) Real space image of signal emission at k=0. b) Interference pattern demonstrating vortex state.

Topological defects occur in many areas of science, including particle physics, cosmology, condensed matter physics and optics. Vortices are typical examples, which occur in bosonic condensates, such as dilute gases, liquid helium, superconductors and polariton BECs. We observed spontaneous vortex formation in the microcavity optical parametric oscillator (OPO) carrying finite Orbital Angular Momentum (OAM). We also demonstrate that a weak probe beam carrying OAM enables imprinting of a vortex state on to macroscopically occupied polariton states.

The OPO was excited using a laser beam without OAM. Figure 7(a) shows an image of the "signal" emission at k=0 in real space above threshold. At the edge of the emission a defect is formed with a dip in the emission intensity (region A). Figure 7(b) shows the interference pattern between the "signal" image in Fig. 7(a) and the same image inverted in so that region A is mixed with region B. The fork-like dislocation indicates the spontaneous formation of a quantised vortex with finite OAM (L=1) in the signal. The phase relationship between pump, signal and idler results also in an anti-vortex in the idler state of OAM L=-1.

References:	Phys. Rev. B 87, 081309(R) (2013);
	Phys. Rev. Lett. 104, 126402 (2010).

Polariton condensation in dynamic acoustic lattices

Polariton dispersion with and without SAW
Fig. 8. Folding of the polariton dispersion (right) as a result of periodic potential introduced by the surface acoustic wave compared to the unperturbed case (left).

Fig. 9. A shematic image of the condensate wires that are formed in potential minima created by the SAW.

In these experiments we show the influence of the tunable potential introduced by a surface acoustic wave (SAW) on the polariton system. In angularly resolved photoluminiscence(PL) experiments we show that the effect of a SAW is to create a lateral polaritonic superlattice which folds the polariton dispersion in to mini-Brillouin zones (MBZ) with the size of the SAW wavelength. This resultsin the formation of energetic bandgaps due to anticrossing of the folded levels at the boundaries of the MBZ (see Fig. 8). We show that they can be tuned in a simple manner by increasing the applied acoustic power.

Also in this work we demonstrate that the tunable potential introduced by a surface acoustic wave (SAW) on a homogeneous polariton condensate leads to fragmentation of the condensate into an array of wires which move with the acoustic velocity. Reduction of the spatial coherence of the condensate emission along the surface acoustic wave direction is attributed to the suppression of coupling between the spatially modulated condensates. Interparticle interactions observed at high polariton densities screen the acoustic potential, partially reversing its effect on spatial coherence - figure 9.

References:	Phys. Rev. B 87, 155423 (2013);
	Phys. Rev. B 86, 100301(R) (2012);
	Phys. Rev. Lett. 105, 116402 (2010);
	Physica E 42, 2548-2551 (2010).

Research Opportunities

We have two laboratories dedicated to research of microcavities. The labs are equipped with helium bath, helium flow and 'coldfinger' cryostats. Also, magmetic measurements can be conducted using our superconducting magnet and cryostat. Excitation of the samples can be done by our two Ti:Sapphire lasers or via our diode laser for work requiring single mode excitation. We also employ a pulsed laser which can work in ps or fs modes.

The setup allows for a wide range of excitation angles and detection angles. Our setup contains both a Mach-Zehnder interferometer for measuring coherence times of the samples as well as a Hanbury-Brown & Twiss interferometer for g² function measurements via two fast APDs. Our setup allows for automated images to be made, either in real space or k-space, via our spectrometers and PIXIS CCDs. Our spatial light modulator allows modulation of both phase and amplitude of the laser beams driving polariton condensates, with the ability to study various topological defects, such as vortices and vortex lattices.

One of our labs is equipped with Hamamatsu Streak Camera with single photon counting module. The setup allows recording real space and energy-momentum space images with a ps resolution. The setup is highly automated and it also incorporates automated Stokes polarimeter and actively stabilised interferometer.

Apart from laboratory facilities, our group benefits from close cooperation with modern clean room fabrication facilities at the III-V National Centre as well as University of Cambridge.

All researchers including PhD students regularly attend international and national conferences that are held all around the world in places such as Korea (QD2008, ICPS2010), Brazil (ICPS2008), France (Paris, OECS2011) and Switzerland (Zurich, ICPS2012). Additionally PhD students attend annual conferences and schools organised within the Polatom Network.