Correction to: Nature Physics https://doi.org/10.1038/nphys4189, published online 3 July 2017.

We miscalculated the gravitational acceleration arising from the tungsten mass by a factor of about two. The correct value is agrav = (33 ± 3) nm/s2. The corrected anomalous acceleration is aanomaly = acyl - agrav = (41 ± 24) nm/s2. These values replace the erroneous ones of agrav = (65 ± 5) nm/s2 and aanomaly = (9 ± 24) nm/s2.

The corrected 95% confidence interval is –7 nm/s2 < aanomaly < 89 nm/s2. Using a one-tailed test to bound fifth-force interactions (which must be attractive for scalar fields with a universal matter coupling), we constrain anomalous accelerations aanomaly < 81 nm/s2 (95% confidence level).

For chameleon fields with Λ = Λ0 = 2.4 meV and n = 1, we exclude up to M < 1.7 × 10−3 MP (replacing M < 2.8 × 10−3 MPl); the gap to torsion pendulum constraints is fully closed for Λ > 6.0 meV (replacing 5.1 meV), and for µ = 0.1 meV, we rule out λ < 0.8 (replacing 1).

Except for these corrections, our conclusions remain unchanged. We replotted the exclusion plot (Fig. 3 in the original paper) with the corrected aanomaly value. Owing to the wide range of parameters considered in the plot, the correction is barely visible.

Fig. 1: Original and Corrected Fig. 3.
figure 1

Corrected Fig. 3 demonstrates constraints on screened chameleons (a, b) and symmetrons (c).

Full size image

We thank Philipp Treutlein, Jannik Wyss and Tilman Zibold for pointing out the issue to us and Cris Panda for re-creating the figure.