Abstract
The temperature evolution of CO2 during full-bore pipe depressurization events is relevant for risk assessment,
e.g., for CO2 transport in the context of CO2 capture and storage (CCS). We analyse and model the temperature
evolution for four different initial temperatures between 𝑇0 = 4.6 ◦C and 𝑇0 = 40 ◦C at supercritical pressures.
All the experiments showed an analogous temperature evolution, reaching similar minimum temperatures along
the pipe. The warmer the initial temperature, the earlier dry-out and faster temperature recovery was observed.
Also, the coldest experiment showed evidence of formation of more dry ice.
We employ the homogeneous equilibrium model (HEM) with different heat-transfer correlations and a two-fluid model (TFM) with different slip models in order to understand the observed data. The results indicate
that the heat transfer changes significantly with different thermodynamic states for the CO2. Also, the HEM
and the TFM with a RELAP-type friction model performed reasonably well at the outlet and at the closed end
of the pipe, but none of the tested models were able to fully describe the strong spatial and temporal gradients observed along the pipe during the depressurization.
e.g., for CO2 transport in the context of CO2 capture and storage (CCS). We analyse and model the temperature
evolution for four different initial temperatures between 𝑇0 = 4.6 ◦C and 𝑇0 = 40 ◦C at supercritical pressures.
All the experiments showed an analogous temperature evolution, reaching similar minimum temperatures along
the pipe. The warmer the initial temperature, the earlier dry-out and faster temperature recovery was observed.
Also, the coldest experiment showed evidence of formation of more dry ice.
We employ the homogeneous equilibrium model (HEM) with different heat-transfer correlations and a two-fluid model (TFM) with different slip models in order to understand the observed data. The results indicate
that the heat transfer changes significantly with different thermodynamic states for the CO2. Also, the HEM
and the TFM with a RELAP-type friction model performed reasonably well at the outlet and at the closed end
of the pipe, but none of the tested models were able to fully describe the strong spatial and temporal gradients observed along the pipe during the depressurization.