Diamond detectors for direct detection of sub-GeV dark matter

We propose using high-purity lab-grown diamond crystal for the detection of sub–giga electron volt dark matter. Diamond targets can be sensitive to both nuclear and electron recoils from dark matter scattering in the mega-electron-volt and above mass range as well as to absorption processes of dark matter with masses between sub–electron volts to tens of electron volts. Compared to other proposed semiconducting targets such as germanium and silicon, diamond detectors can probe lower dark matter masses via nuclear recoils due to the lightness of the carbon nucleus. The expected reach for electron recoils is comparable to that of germanium and silicon, with the advantage that dark counts are expected to be under better control. Via absorption processes, unconstrained QCD axion parameter space can be successfully probed in diamond for masses of order 10 eV, further demonstrating the power of our approach.

Figure 1

Band structure of diamond in the reduced-zone scheme, reproduced from Ref. [17].

Figure 2

Estimated mean free path using the Callaway model fitted to experimental data [55, 56] and normalized to the same geometry ($l_b=2.6\mathrm{mm}$) and surface quality ($P=4\times 10^{-4}$). The model is described in Ref. [55], with the modification that the specular reflection probability $P$ is a constant rather than a function of phonon frequency.

Figure 3

Energy resolution scaled according to Eq. (12) using the parameters in Table 3. These are shown along with the scaling relations for the given volume and efficiency as a function of $T_c$ (solid line for design A, dashed for design B, and dot-dashed for design C). Also shown are the resolution demonstrated in a tungsten TES photon detector [62] and a Si detector with tungsten QET readout [67]. The corrected resolution of the Si detector for upgraded readout electronics is also shown, compared to its ideal scaling relation (dotted red line).

Figure 4

Projected reach at 95% C.L. for absorption of kinetically mixed dark photons with mass greater than 1 meV. The solid black curves indicate the expected reach for a kilogram-year exposure of diamond. Projected reach for germanium and silicon [45], Dirac materials [10], polar crystals [12], molecules [82], and superconducting aluminum [9] targets are indicated by the dotted curves. Constraints from stellar emission [83, 84], DAMIC [85], SuperCDMS [72], and Xenon [84] data are shown by the shaded orange, green, purple, and blue regions, respectively.

Figure 5

Projected reach at 95% C.L. for absorption of axionlike particles. The reach of a kilogram-year exposure of diamond is shown by the solid black curve. The reach for semiconductors such as germanium and silicon [45] and superconducting aluminum [9] targets is depicted by the dotted magenta, purple, and gray curves, respectively. Stellar constraints from Xenon100 data [88] and white dwarves [87] are indicated by the shaded red and orange regions, respectively. Constraints from loop-induced couplings to photons are presented in the shaded blue and green regions [89, 90]. The QCD axion region of interest is indicated in shaded gray.

Figure 6

Recoil spectra for ${\overline{\sigma }}_e=10^{-37}{\mathrm{cm}}^2$ of silicon (dashed), germanium (dotted), and diamond (solid) for $m_{\chi }=10\mathrm{MeV}$ (blue) and 1 GeV (green). The top and bottom plots are for a momentum-independent and a momentum-dependent form factor respectively.

Figure 7

Projected reach at 95% C.L. for electron recoils in silicon (red), germanium (blue), and diamond (green) detectors with one e-h pair threshold (top) and two e-h pair threshold (bottom) for 1 g-day of exposure (solid) and 1 kg-yr of exposure (dashed), for a momentum-independent form factor. The current exclusion limits from SENSEI [73], CDMS HVeV [72], and XENON10 [2] are also depicted for comparison. The parameter space corresponding to producing the observed dark matter relic density via standard freeze-out in a minimal dark sector model is indicated by the solid thick mustard curve labeled “freeze-out” (see Ref. [92] for more details).

Figure 8

Projected reach at 95% C.L. for electron recoils in silicon (red), germanium (blue), and diamond (green) detectors with one e-h pair threshold (top) and two e-h pair threshold (bottom) for 1 g-day of exposure (solid) and 1 kg-yr of exposure (dashed), for a momentum-dependent form factor. The current exclusion limits from SENSEI [73], CDMS HVeV [72], and XENON10 [2] are also depicted for comparison. The parameter space corresponding to producing the observed dark matter relic density via standard freeze-in in a minimal dark sector model is indicated by the solid thick mustard curve labeled freeze-in (see Ref. [92] for more details).

Figure 9

Nuclear recoil projected reach at 95% C.L. for He (blue), diamond (green), Si (red), and Xe (cyan) for energy thresholds of 1 eV (left panel) and 10 meV (right panel). The dashed lines are for a gram-day exposure, while the dot-dashed lines are for a kilogram-year exposure. The yellow region indicates the neutrino floor for He [15], and the brown region indicates the neutrino floor for C (computed according to the formalism in Ref. [79]). Also shown in gray are the current best limits on NR dark matter interactions from $\nu \mathrm{CLEUS}$ [94], CRESST-III [36, 96], CDMSLite [97], Darkside-50 [98], and Xenon-1T [99] for comparison.