High-bypass ratio turbofan engines are commonly employed in aircrafts. Their usage is essential to guarantee low specific fuel consumption, reduced CO2 emissions and low noise levels. Such modern aero-engines benefit from high efficiencies by operating at turbine inlet temperatures in excess of the melting point of the turbine components. To enable this, compressor air is supplied to the turbine for cooling and purging purposes. The reintroduction of the cooling air back into the mainstream flow is known to alter the flow field and to affect the aerodynamic performance of the turbine components. A component especially susceptible to the interaction between the mainstream and purge flow is the Turbine Center Frame, located between high-pressure turbine (HP) and low-pressure (LP) turbine. For ever higher bypass ratios, this turbine transition ducts need to be designed with axial lengths as short as possible and larger radial offsets to avoid engine weight penalties while at the same time maintaining aerodynamic performance. More detailed experience in the field of intermediate turbine ducts is needed to identify further opportunities to improve turbofan engine performance, including an indepth understanding of the interaction between mainstream and purge flows. This paper presents a Computational Fluid Dynamics (CFD) study of the effect of the purge flow temperature, and hence density, on the aerodynamic performance of an engine representative Turbine Center Frame (TCF). Several steady-state Reynoldsaveraged Navier-Stokes (RANS) simulations were conducted for varying purge flow temperatures using an in-house code called LINARS. Time-averaged five-hole-probe measurements acquired in the Transonic Test Turbine Facility (TTTF) at Graz University of Technology were used as inlet boundary conditions to impose an engine-relevant flow field. The results obtained from two reduced and two increased purge flow temperature conditions were compared to a reference case. The reference case results showed agreement with static wall pressure measurements, hence validating the simulation. Changing the purge flow temperature significantly affected the main flow locally as well as overall. The position and size of vortices in the TCF were changed under the presence of hotter or cooler purge flows. Additionally, a flow separation on the outer duct wall observed in the baseline case was suppressed in the cold-purged flow case. The cold-purged TCF showed a 28.8% lower total pressure loss than the hot-purged one. This indicates that a more aggressive TCF design may be feasible in a cold-purged operation.