Azimuthally Sensitive Hanbury Brown–Twiss Interferometry in $\mathrm{A}\mathrm{u}+\mathrm{A}\mathrm{u}$ Collisions at $\sqrt{s_{NN}}=200\mathrm{G}\mathrm{e}\mathrm{V}$

We present the results of a systematic study of the shape of the pion distribution in coordinate space at freeze-out in $\mathrm{A}\mathrm{u}+\mathrm{A}\mathrm{u}$ collisions at BNL RHIC using two-pion Hanbury Brown–Twiss (HBT) interferometry. Oscillations of the extracted HBT radii versus emission angle indicate sources elongated perpendicular to the reaction plane. The results indicate that the pressure and expansion time of the collision system are not sufficient to completely quench its initial shape.

Figure 1

Squared HBT radii using Eq. (1) relative to the reaction plane angle for three centrality classes. The solid lines show allowed [24] fits to the individual oscillations.

Figure 2

Squared HBT radii relative to the reaction plane angle for four $k_T$ ($\mathrm{G}\mathrm{e}\mathrm{V}/c$) bins, 20%–30% centrality events. The solid lines show allowed [24] fits to the individual oscillations.

Figure 3

Fourier coefficients of azimuthal oscillations of HBT radii vs number of participating nucleons, for three $k_T$ ($\mathrm{G}\mathrm{e}\mathrm{V}/c$) bins. Left panels: means (0th-order FC) of oscillations; right panels: relative amplitudes (see text for details). Larger participant numbers correspond to more central collisions.

Figure 4

Source eccentricity obtained with azimuthally sensitive HBT (${\varepsilon }_{\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}}$) vs initial eccentricity from a Glauber model (${\varepsilon }_{\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}}$). The most peripheral collisions correspond to the largest eccentricity. The dashed line indicates ${\varepsilon }_{\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}}={\varepsilon }_{\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}}$. Uncertainties on the precise nature of space-momentum correlations lead to 30% systematic errors on ${\varepsilon }_{\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}}$ [30].