Two uncharged mirrors very close together block the vacuum's longer wavelengths between them. The space between the plates then holds less energy than empty space: a real, measured negative energy density. It's the only laboratory source of negative energy we have, and it's tiny.
The line is the ideal-plate pressure |P| against the gap; the shaded bands show the ranges real experiments covered. Lamoreaux 1997 used a torsion pendulum with a sphere and plate over 0.6–6 µm (about 5% claimed, later revised for an electrostatic background) [3]. Mohideen & Roy 1998 used an atomic-force microscope over 0.1–0.9 µm (about 1% at the closest gap) [4]. Bressi et al. 2002 measured true parallel plates over 0.5–3 µm (about 15%) [5]. Real metals and temperature lower the force below the ideal line at the smallest and largest gaps.
Physics doesn't allow unlimited negative energy. For a free quantum field, Ford and Roman showed that the energy density, averaged over a sampling time t₀, can't fall below ρ̄ ≥ -3ħ / (32 π² c³ t₀⁴) (massless scalar, 4D, Lorentzian sampling) [6]. The longer the pulse, the weaker it has to be: negative energy has to be "paid back" quickly.
Casimir energy is static and boundary-made, so it isn't limited the same way. That's why warp-drive physicists look to it, and why its ~10⁻²¹ kg/m³ scale (at 1 µm) is the honest starting point for the next tab.
Each line is a published answer, with its status. Bars compare each requirement with the best negative energy we can picture making: a 100 m × 100 m Casimir plate pair at a 10 nm gap (one hectare, a stadium-sized array).
| design | needs | status |
|---|
Stars from 5 to 25 kpc orbit at 229 km/s near the Sun, and the speed falls only 1.7 km/s per kpc farther out [7]. Visible matter alone predicts the speed should fall much faster: a disk of 6.8×10¹⁰ M☉ (a = 3 kpc, b = 0.28 kpc) plus a 4.5×10⁹ M☉ bulge [8]. Adding an NFW dark halo (8×10¹¹ M☉, concentration 15.3, scale radius 16 kpc, the MWPotential2014 setup [8]) closes most of the gap. Slide the halo to zero and watch the prediction fall away from the stars.
3,000 stars start at the speeds we really measure. With the halo on, gravity holds them in orbit. Switch it off: only the visible mass pulls, and the outer stars, moving too fast for it, drift outward and escape. That runaway, which real galaxies don't show, is why astronomers infer dark matter. One frame = 2 million years.
Near the Sun, analyses find ρ ≈ 0.3–0.5 GeV/cm³ (global models) to 0.4–0.6 GeV/cm³ (local studies) [9]. The flux uses ~230 km/s and a 0.6 m² body. Status: nobody has directly detected a dark matter particle. The LZ experiment (4.2 tonne-years) saw no excess and excludes WIMP–nucleon cross sections down to 2.2×10⁻⁴⁸ cm² at 40 GeV [10]. Dark matter is weighed by its gravity; it hasn't been seen.
Computed by the Lanturn Gravity Lab from the formulas on this page and stored on a sha256 hash chain. Your browser re-checks every link below.