Titan’s dense atmosphere of nitrogen with trace amounts of methane makes it one of the most chemically rich places in the solar system. Solar and cosmic radiation ionizes and dissociates these molecules, and overtime, the carbon, hydrogen and nitrogen recombine and form a range of complex organics known as tholins. Neish et al. (2010) showed that amino acids are produced when these tholins are exposed to liquid water. This biological potential is a primary motivation for the newly selected Dragonfly mission to Titan. Dragonfly will directly sample Titan’s ice to analyze organic molecules that are thought to have been exposed to liquid water. Frozen water melt ponds found in Titan’s impact craters are thus a prime target for Dragonfly’s astrobiological investigations. For Dragonfly to effectively analyze and sample the solidified melt sheet for prebiotic (or possibly biological) products, the team must understand the geologic history of these water-organic mixtures. In this work, we use the planetary ice model of Buffo et al. (2020) to track the concentrations of organic molecules within Titan’s melt ponds as they freeze. This model has previously been used to track saltwater freezing into terrestrial sea ice and the Europan crust. Here, we track the concentration evolution of HCN, a common molecule on Titan, within the forming ice and the residual melt of Titan’s impact melt ponds. In the model, impurity dynamics are driven by gravity drainage due to buoyancy differences. HCN is less dense than water, which suggests that it will be rejected from the floor of the melt pond while at its roof HCN will stay suspended and become trapped in the ice at a level equivalent to the bulk concentration of the underlying melt pond. The concentration will therefore increase with depth as rejected impurities concentrate in the melt. The ideal location to sample is thus the middle portion of the impact crater’s melt sheet. To access this part of the melt sheet, Dragonfly will ultimately be limited to ice that has been exposed through fluvial erosion by rivers. Our results will offer context on the time scales a sample may have remained liquid at any given depth.