Directed flow of particles is an important feature seen in heavy-ion collisions and is a sensitive probe of the equation of state (EoS) of the matter produced in the collisions. Model calculations have also predicted that directed flow could be a sensitive probe of the softening of the EoS associated with a first order phase transition. Directed flow of protons and anti-protons are also of interest as they offer sensitivity to both the contributions from the transported quarks and also the medium generated component from the produced quarks. Measurements of proton and net proton directed flow from BES-I have shown that there is a non-monotonous dependence on collision energy. We will present measurements of the directed flow of protons and antiprotons from the collision energies of 7.7, 9.2, 11.5, 14.5, 17.3, 19.6, and 27 GeV Au+Au collisions, using high statistics BES-II data from STAR. We will also present a decomposition of proton directed flow into a medium generated component and a component (v₁ excess) attributed to transported protons. The v₁ excess component is found to show a simple scaling between collision energies of 200 GeV to ~10 GeV, but to break scaling below that. The new results have significantly reduced uncertainties and also allow differential measurements in centrality and transverse momentum. Measurements will be compared to different model calculations and implications to the understanding of the QCD phase structure and EoS of the medium will be discussed.

Figure 1: Proton (filled circles) and Anti-Proton (open circles) directed flow with respect to rapidity for central (0-10%) mid-central (10-40%) and peripheral (40-80%) collisions for BES I (black) and BES II (red) energies. 

Figure 2: Excess Proton v₁ vs. normalized rapidity for all BES II collider energies as well as the BES I Excess Proton v1 results for $\sqrt{s_{NN}}=200 GeV$ for 10-40% centrality. We see that from 200 GeV down to 14.6 GeV all measurements fall along the same slope, breaking scale for 11.5 GeV and lower. The fit line is an odd third order polynomial of 14.6,17.3,19.6,27 and 200 GeV datapoints.

 
Figure 3: Proton, Anti-Proton, and Excess v₁ slope with respect to normalized rapidity versus collision energy at 10-40% centrality. We see excess v₁ scales with beam rapidity from 200 GeV down to at least 14.6 GeV. The JAM Mean Field Model seems to qualitatively describe the behavior more accurately than JAM Cascade. The JAM Mean Field is close to data at 14.6 and 19.6 GeV, but cannot simultaneously reproduce the result at 7.7 GeV. Extending to lower energies, there is clear breaking of scaling at or above 7.7 GeV with the BES-II dataset analysis, indication change in medium and collision dynamics at 7.7 GeV.

Conclusion:
In this work we have completed the precision measurement of the proton and antiproton directed flow as a function of rapidity for all the collider BES II energies, significantly increasing the accuracy over the BES I measurements. We see that looking at the excess v₁ of protons gives us a constant slope with respect to normalized rapidity from 200 GeV down to 14.6 GeV, and that this scaling deviates from this constant value at 11.5 GeV and below, indicating a change in the medium or collision dynamics. When compared to mean field calculations, the data has a lower slope than predicted even for the softest equation of state. This new variable gives us a new avenue to probe the equation of state.