We present a novel implementation of supermassive black hole (SMBH) formation, dynamics, and accretion in the massively parallel tree+SPH code, ChaNGa. This approach improves the modeling of SMBHs in fully cosmological simulations, allowing for a more de- tailed analysis of SMBH-galaxy co-evolution throughout cosmic time. Our scheme includes novel, physically motivated models for SMBH formation, dynamics and sinking timescales within galaxies, and SMBH accretion of rotationally supported gas. The sub-grid parameters that regulate star formation (SF) and feedback from SMBHs and SNe are optimized against a comprehensive set of z = 0 galaxy scaling relations using a novel, multi-dimensional parameter search. We have incorporated our new SMBH implementation and parameter optimization into a new set of high resolution, large-scale cosmological simulations called Romulus. We present initial results from our flagship simulation, Romulus25, showing that our SMBH model results in SF efficiency, SMBH masses, and global SF and SMBH accretion histories at high redshift that are consistent with observations. We discuss the importance of SMBH physics in shaping the evolution of massive galaxies and show how SMBH feedback is much more effective at regulating star formation compared to SNe feedback in this regime. Further, we show how each aspect of our SMBH model impacts this evolution compared to more common approaches. Finally, we present a science application of this scheme studying the properties and time evolution of an example dual AGN system, highlighting how our approach allows simulations to better study galaxy interactions and SMBH mergers in the context of galaxy-BH co-evolution.

The sources that reionized the universe are still unknown, but likely candidates are faint but numerous galaxies. In this paper we present results from running a high resolution, uniform volume simulation, the Vulcan, to predict the number densities of undetectable, faint galaxies and their escape fractions of ionizing radiation, $f_\mathrm{esc}$, during reionization. Our approach combines a high spatial resolution, a realistic treatment of feedback and hydro processes, a strict threshold for minimum number of resolution elements per galaxy, and a converged measurement of $f_\mathrm{esc}$. We calibrate our physical model using a novel approach to create realistic galaxies at z=0, so the simulation is predictive at high redshifts. With this approach we can (1) robustly predict the evolution of the galaxy UV luminosity function at faint magnitudes down to $M_\mathrm{UV}$~-15, two magnitudes fainter than observations, and (2) estimate $f_\mathrm{esc}$ over a large range of galaxy masses based on the detailed stellar and gas distributions in resolved galaxies. We find steep faint end slopes, implying high number densities of faint galaxies, and the dependence of $f_\mathrm{esc}$ on the UV magnitude of a galaxy, given by the power-law: log $f_\mathrm{esc} = (0.51 \pm 0.04)M_\mathrm{UV} + 7.3 \pm 0.8$, with the faint population having $f_\mathrm{esc}$~35%. Convolving the UV luminosity function with $f_\mathrm{esc}$($M_\mathrm{UV}$), we find an ionizing emissivity that is (1) dominated by the faintest galaxies and (2) reionizes the universe at the appropriate rate, consistent with observational constraints of the ionizing emissivity and the optical depth to the decoupling surface tau_es, without the need for additional sources of ionizing radiation.