Fig. 4: Ergotropy and information entropy dynamics: experimental results. | npj Quantum Information

Fig. 4: Ergotropy and information entropy dynamics: experimental results.

From: Single-atom energy-conversion device with a quantum load

Fig. 4

a,b, Experimentally derived occupation probabilities pn of the load, on initial preparation and after Nc = 8 cycles, for the stroke (I) resetting parameter \({p}_{{\rm{D}}}^{{\mathcal{A}}}=0.5\). c, The (approximated) passified density matrix of the load, obtained from b. The occupation probabilities of the load in b are rearranged to go in descending order with increasing n number. This yields the passified load probability distribution \({\tilde{p}}_{n}\) as a function of n. d, Normalized ergotropy (energy per ω) \({{\mathcal{E}}}_{{\rm{L}}}\) of the load versus Nc, with \({p}_{{\rm{D}}}^{{\mathcal{A}}}=0.5\). The red continuous line shows the exact numerical simulation value. The black dashed line is the approximate ergotropy using only the diagonal elements of the reduced density matrix of the load. The shows the experimental data. The green arrows relate the measured probability distributions depicted in a,b to the corresponding value of ergotropy in d. e, Information entropy \({{\mathcal{S}}}_{{\rm{L}}}\) of the load versus time for the exact numerical estimate (red continuous line), an approximate numerical estimate considering only the diagonal elements of the density matrix (black dashed line), and the experimentally evaluated information entropy values after two, four, six and eight cycles (blue squares), for \({p}_{{\rm{D}}}^{{\mathcal{A}}}=0.32\). f, The same as e, but for the refrigerator cycle, and for \({p}_{{\rm{D}}}^{{\mathcal{A}}}=0.5\). Here the experimental data are given by the blue triangles. The quantities in this figure are derived from the same data used in Fig. 3, and the error bars represent one sigma standard deviation. As in Fig. 3, the reduced χ2 depends on the number of data points in the fit for a given number of cycles, which ranges from 25 to 200. The phase error used is 0.1, estimated from >150 replicates.

Back to article page