Entangling macroscopic diamonds at room temperature: Bounds on the continuous-spontaneous-localization parameters

A recent experiment [K. C. Lee et al., Science 334, 1253 (2011)] succeeded in detecting entanglement between two macroscopic specks of diamonds, separated by a macroscopic distance, at room temperature. This impressive result is a further confirmation of the validity of quantum theory in (at least parts of) the mesoscopic and macroscopic domain, and poses a challenge to collapse models, which predict a violation of the quantum superposition principle, which is bigger the larger the system. We analyze the experiment in the light of such models. We will show that the bounds placed by experimental data are weaker than those coming from matter-wave interferometry and noninterferometric tests of collapse models.

Figure 1

Experimental setup, schematic representation. A pump pulse, split by the beamsplitter BS1, is sent to the diamonds. The two separate Stokes modes produced through Raman transitions and exiting each diamond, are combined by the polarization beamsplitter PBS2 and reach the detector $D_s$. A probe pulse sent after a short time interval is sent through the diamonds to produce anti-Stokes photons, which are then combined on PBS3 and sent to the detectors $D_a$.

Figure 2

Schematic representation of the displacements of carbon atoms in one diamond. The two different colors represent the two different counteroscillating sublattices. Full colors denote the two sublattices at rest, while faded ones represent the two sublattices at the maximum relative distance.

Figure 3

Exclusion plot in the $\lambda -r_C$ parameter space. The red curve (“Diamonds”) shows the bound here computed from the experiment under consideration; the region marked in red is excluded with a high confidence level. For comparison we have included the upper bounds from x-ray experiment [17] (green), from nanocantilever experiments [18] (blue), and from KDTLI experiment [1] (purple). For reference, we have also included the GRW [4] values ($\lambda ={10}^{-16}{\mathrm{s}}^{-1}$, $r_C={10}^{-7}$ m) and the values proposed by Adler [13]: ($\lambda ={10}^{-8}{\mathrm{s}}^{-1}$, $r_C={10}^{-7}$ m) and ($\lambda ={10}^{-6}{\mathrm{s}}^{-1}$, $r_C={10}^{-6}$ m).