Figure 1

(a) Schematic of the microwave measuring setup, and sample with coplanar waveguide transmission line, with metal film shown as black. (b) Real part of change in diagonal conductivity from that at Landau filling $\nu =1$, $\mathrm{Re}[\Delta {\sigma }_{xx}(f)]$, vs frequency, $f$, for $\nu$ increasing from 1 (bottom trace) to 1.16 (top trace), in equal steps of 0.02. Successive traces are offset by $2\mu \mathrm{S}$ from each other for clarity. (c) $\mathrm{Re}[\Delta {\sigma }_{xx}(f)]$, vs $f$, for $\nu$ decreasing from 1 (bottom trace) to 0.84 (top trace), in equal steps of 0.02. Successive traces are offset by $2\mu \mathrm{S}$ from each other.Reuse & Permissions

Figure 2

$\mathrm{Re}[\Delta {\sigma }_{xx}(f)]$, here the change in real part of the diagonal conductivity from that at Landau filling $\nu =1/3$, vs frequency $f$ Successive traces were taken at intervals of 0.002 in $\nu$, and are offset by $0.1\mu \mathrm{S}$ from each other for clarity. (a) $\nu$ from $1/3$ (bottom trace), to 0.313 (top trace) (b) $\nu$ from $1/3$ (bottom trace), to 0.353 (top trace).Reuse & Permissions

Figure 3

(a) The resonance peak frequencies $f_{\mathrm{pk}}$ near filling factor $\nu =1$, vs $\nu$ (lower axis) and quasiparticle-hole density, $\tilde{n}$ (upper axis). (b) Solid circles: $f_{\mathrm{pk}}$ near $\nu =1/3$ vs $\nu$ (lower axis) and $\tilde{n}$ (upper axis); triangles: $f_{\mathrm{pk}}$ vs $\tilde{n}$ on upper axis only, near $\nu =1$, and with the sample tilted to 63°. (c) $S/f_{\mathrm{pk}}$ vs $\nu$, near $\nu =1$, where $S$ is integrated spectrum $\mathrm{Re}({\sigma }_{xx})$ vs $f$. (d) $S/f_{\mathrm{pk}}$ vs $\nu$, near $\nu =1/3$. Dashed lines in (c) and (d) are predicted $S/f_{\mathrm{pk}}$ calculated from quasiparticle charge density as described in the text.Reuse & Permissions