Astrophysical and local constraints on string theory: Runaway dilaton models

One of the clear predictions of string theory is the presence of a dynamical scalar partner of the spin-2 graviton, known as the dilaton. This will violate the Einstein equivalence principle, leading to a plethora of possibly observable consequences which in a cosmological context include dynamical dark energy and spacetime variations of nature’s fundamental constants. The runaway dilaton scenario of Damour, Piazza, and Veneziano is a particularly interesting class of string theory inspired models which can in principle reconcile a massless dilaton with experimental data. Here we use the latest background cosmology observations, astrophysical and laboratory tests of the stability of the fine-structure constant, and local tests of the weak equivalence principle to provide updated constraints on this scenario, under various simplifying assumptions. Overall we find consistency with the standard $\mathrm{\Lambda }\mathrm{CDM}$ paradigm. We improve the existing constraints on the coupling of the dilaton to baryonic matter by a factor of 6 and to the dark sector by a factor of 2. At the one-sigma level the current data already exclude dark sector couplings of order unity, which would be their natural value.

Figure 1

Likelihood constraints in the various 2D planes for the runaway dilaton scenario in the linearized approximation. The black lines represent the one-, two-, and three-sigma confidence levels, and the colormap depicts the reduced chi-square of the fit. The value of the reduced chi-square at the best fit is ${\chi }_{\nu }^2=0.99$.

Figure 2

Same as Fig. 1, for the alternative parametrization including the effective dark energy equation of state. The value of the reduced chi-square at the best fit is ${\chi }_{\nu }^2=0.99$.

Figure 3

Likelihood constraints in the various 2D planes for the runaway dilaton scenario. Top, middle, and bottom plots respectively correspond to the dark coupling, matter coupling, and field coupling assumptions. The black lines represent the one-, two-, and three-sigma confidence levels, and the colormaps depict the reduced chi-square of the fits. The value of the reduced chi-square at the best fit is ${\chi }_{\nu }^2=0.99$ in all three cases.

Figure 4

One-dimensional posterior likelihoods for the full analysis in Fig. 3. Comparing the results of the dark coupling, matter coupling, and field coupling assumptions, the only significant differences occur for the dark energy coupling, ${\alpha }_V$.