A comprehensive study of heat transfer and pressure drop of refrigerant R404A during condensation and supercritical cooling at near-critical pressures inside a 9.4 mm tube was conducted. Investigations were carried out at five nominal pressures: 0.8, 0.9, 1.0, 1.1 and 1.2 x Pcrit. Heat transfer coefficients were measured using a thermal amplification technique that measures heat duty accurately while also providing refrigerant heat transfer coefficients with low uncertainties. For condensation tests, local heat transfer coefficients and pressure drops were measured for the mass flux range 200 < G < 800kg/m2-s in small quality increments over entire vapor-liquid region. For supercritical tests, local heat transfer coefficients and pressure drops were measured for the same mass flux range as in the condensation tests for temperatures ranging from 30--110°C. For both phase-change condensation and supercritical cooling, frictional pressure gradients were calculated by separating the deceleration component due to momentum change from the measured pressure gradients.;During condensation, the effect of reduced pressure in heat transfer is not very significant, while this effect is more pronounced in the pressure gradient. Flow regime transition criteria by Coleman and Garimella (2003) were used to designate the prevailing flow regimes for a given combination of mass flux and quality. The condensation data collected in the present study were primary in the wavy and annular flow regimes. For supercritical cooling, the sharp variations in thermophysical properties in the vicinity of the critical temperature were found to have substantial effect on heat transfer coefficients and pressure drop. Based on the characteristics of the specific work of thermal expansion (contraction), the data from the supercritical tests were grouped into three regimes: liquid-like, pseudo-critical transition and gas-like regimes.;Flow regime-based heat transfer and pressure drop models were developed for both condensation and supercritical cooling. For condensation, the overall heat transfer model predicts 89% of the data within +/-15% while the overall pressure drop model predicts 96% of the data within +/-15%. For supercritical cooling, the heat transfer model predicted 73% of the data within +/-25% while the pressure gradient model predicts 90% of the data within +/-15%.