Figure 1

(a) Experimental setup. Two pump beams ${\mathbf{k}}_{\mathbf{p}}$ and $-{\mathbf{k}}_{\mathbf{p}}$ are focused on the microcavity sample MC. The microcavity emission is collimated by a lens ($f=50\mathrm{mm}$). Two movable screens $S1$ and $S2$ placed in the far field plane (FFP) eliminate the transmitted pump beams. The upper part of the emission is detected by photodiode PD1, the lower part by photodiode PD2. The photocurrent sum and difference are then spectrally analyzed. (b) Representation in $k$ space of the expected scattering processes. Correlated pairs are connected by arrows.Reuse & Permissions

Figure 2

(a) Far-field emission in the TE polarization for a total pump power of: (a) 3 mW, (b) 19 mW, (c) 27 mW, (d) 39 mW. Cavity-exciton detuning $\delta =0$. At low pump power (a)–(b), light intensity is roughly uniform on the elastic circle (except for the two pump spots, lying approximately on the horizontal diameter). At high pump power (c)–(d), scattered polaritons accumulate around the ${\mathbf{k}}_{\mathbf{y}}$ diameter.Reuse & Permissions

Figure 3

Optical power of the upper (U) and lower (L) beam as a function of the pump power. Cavity-exciton detuning $\delta =0$. Inset (a) the zones over which the upper and lower emissions are, respectively, integrated are shown. Inset (b) log-log plot of the upper beam emission as a function of pump power.Reuse & Permissions

Figure 4

(a) Four wave mixing optical power as a function of the pump power. The threshold value of the pump power and the average emitted power differ from those reported in Fig. 3, because the two sets of data were taken in slightly different experimental conditions. (b) Normalized intensity correlation $C_{\mathbf{k},-\mathbf{k}}$ as a function of the pump power. (c) Noise of photocurrent sum $S_{+}$ and of photocurrent difference $S_{-}$ (normalized to the standard quantum limit) as a function of the pump power. In all curves, dashed lines are guides to the eye.Reuse & Permissions