Liquid–vapor mercury isotope fractionation was investigated under equilibrium and dynamic conditions. Equilibrium evaporation experiments were performed in a closed glass system under atmospheric pressure between 0 and 22°C, where vapor above the liquid was sampled at chemical equilibrium. Dynamic evaporation experiments were conducted in a closed glass system under 10 −5 bar vacuum conditions varying (1) the fraction of liquid Hg evaporated at 22°C and (2) the temperature of evaporation (22–100°C). Both, residual liquid and condensed vapor fractions were analyzed using stannous chloride CV-MC-ICP-MS.Equilibrium evaporation showed a constant liquid–vapor fractionation factor (α 202/198 ) of 1.00086±0.00022 (2SD, n=6) within the 0–22°C range. The 22°C dynamic evaporations experiments displayed Rayleigh distillation fractionation behavior with liquid–vapor α 202/198 =1.0067±0.0011 (2SD), calculated from both residual and condensed vapor fractions. Our results confirm historical data (1920s) from Brönsted, Mulliken and coworkers on mercury isotopes separation using evaporation experiments, for which recalculated δ 202 Hg′ showed a liquid–vapor α 202/198 of 1.0076±0.0017 (2SD). This liquid–vapor α 202/198 is significantly different from the expected kinetic α 202/198 value ((202/198) 0.5 =1.0101). A conceptual evaporation model of back condensation fluxes within a thin layer at the liquid–vapor interface was used to explain this discrepancy. The δ 202 Hg′ of condensed vapor fractions in the 22–100°C temperature range experiments showed a negative linear relationship with 10 6 /T 2 , explained by increasing rates of exchange within the layer with the increase in temperature.Evaporation experiments also resulted in non-mass-dependent fractionation (NMF) of odd 199 Hg and 201 Hg isotopes, expressed as Δ 199 Hg′ and Δ 201 Hg′, the deviation in ‰ from the mass fractionation relationship with even isotopes. Liquid–vapor equilibrium yielded Δ 199 Hg′/Δ 201 Hg′ relationship of 2.0±0.6 (2SE), which is statistically not different from the one predicted for the nuclear field shift effect (Δ 199 Hg/Δ 201 Hg≈2.47). On the other hand, evaporation under dynamic conditions at 22°C led to negative anomalies in the residual liquid fractions that are balanced by positive anomalies in condensed vapors with lower Δ 199 Hg′/Δ 201 Hg′ ratios of 1.2±0.4 (2SD). This suggests that either magnetic isotope effects may have occurred without radical chemistry or an unknown NMF process on odd isotopes operated during liquid mercury evaporation.