The astrophysical r-process generates around half the abundance of elements heavier than iron, yet precisely where or how this process occurs remains a topic of intense inquiry. Detailed simulations studying the sensitivity of nuclear properties performed in recent years have shown that among the various experimental quantities entering into modeled r-process abundances, atomic mass differences between isotopes hold the greatest sway. In particular, masses near the abundance peaks at N=82 (A~130) and N=126 (A~195), as well as in the rare-earth region (A~165), are particularly influential. Despite these facts, the latter two regions' masses are largely unmeasured. Understanding the formation of one of these r-process hallmarks, the rare-earth abundance peak, could shed light on the astrophysical sites because this peak is uniquely sensitive to underlying nuclear properties, particularly to nuclear binding energies which have so far been largely derived from theoretical mass models.We have performed precise atomic mass measurements of 24 neutron-rich rare-earth isotopes using the JYFLTRAP double Penning trap mass spectrometer at the University of Jyväskylä in Finland. The atomic masses of 14 of them have been experimentally determined for the first time, while the precisions for all have been significantly improved. Changes in two-neutron separation and neutron pairing energies show systematic deviations from theoretical mass model predictions. Their impact on the calculated r-process abundances in this region are significant, resulting in a smoother overall pattern with less odd-even staggering.Finally, a new experimental facility in development at Argonne National Laboratory will allow for Penning trap mass spectrometry on a swath of N=126 isotopes previously inaccessible, and of critical importance for understanding the r-process. One essential component for this facility to efficiently purify its radioactive ion beams, a multi-reflection time-of-flight mass spectrometer (MR-TOF), has been built and commissioned at the University of Notre Dame. We have performed the off-line commissioning of the MR-TOF utilizing, for the first time, ion bunches produced using a Bradbury-Nielsen gate. We found that this bunching method results in large initial efficiency losses as the ions loop in the MR-TOF relative to bunches that would be produced by a radiofrequency cooler and buncher. Nevertheless, the off-line commissioning has shown that the MR-TOF can meet the new facility's demands in terms of resolving power.