Supercritical CO2 in a Printed Circuit Heat Exchanger CFD Simulation – ANSYS Validation Study

Modern power generation plants require incredibly compact and highly efficient systems to transfer thermal energy. Traditional shell-and-tube heat exchangers are physically too large and frequently fail under extreme operational pressures. The Printed Circuit Heat Exchanger solves this general problem by using microscopic fluid channels etched directly into solid metal blocks. Furthermore, pumping supercritical carbon dioxide through these tiny channels instead of water massively increases the total energy efficiency of the power cycle. However, supercritical carbon dioxide behaves unpredictably. When the fluid approaches its critical point, its internal density and heat capacity fluctuate violently.

In this tutorial, our exact aim is to validate the thermohydraulic performance of a Printed Circuit Heat Exchanger using ANSYS Fluent. We will simulate the complex conjugate heat transfer between the flowing supercritical gas and the solid metal walls. By the end of this study, we will prove our numerical model matches the published academic Nusselt number data perfectly. If you want to master extreme thermal equipment design, exploring our Heat Exchangers CFD tutorials is your absolute best next step.

• Reference [1]: Chai, Lei, and Savvas A. Tassou. “Influence of Operation Parameters on Thermohydraulic Performance of Supercritical CO2 in a Printed Circuit Heat Exchanger.” Heat Transfer Engineering16-17 (2025): 1457-1475.

Figure 1: The computational flow schematic defining the periodic boundaries and the solid structural wall separating the channels [1].

Simulation Process: NIST REFPROP Real Gas & PCHE Geometry Modeling

To capture the complex thermal physics, we must build a precise three-dimensional domain. We model a straight semi-circular microchannel geometry exactly matching the reference study. The computational domain contains one cold channel and one hot channel. A solid stainless steel metal wall separates them. The semi-circular channels have a exact radius of 1 millimeter. The channel pitch measures exactly 2.54 millimeters in the horizontal direction and 3.26 millimeters in the vertical direction.

To capture the extreme physical property changes near the pseudo-critical point, we generate a highly dense mesh. The final grid contains exactly 2676726 hexahedral cells. We model the working fluid using the NIST REFPROP database dynamically loaded into the software. This database accurately calculates the large property variations of the real gas. We consider both buoyancy effects and conjugate heat transfer. The cold fluid enters at 100 degrees, while the hot fluid enters at 400 degrees. Finally, we apply thermally periodic boundaries to the top and bottom faces to represent thousands of adjacent channel interactions properly.

Figure 2: The hexahedral grid containing exactly 2676726 cells to capture the boundary layer friction and property variations.

Post-processing: Thermohydraulic Performance & Nusselt Number

We must confirm the mathematical accuracy of our computational model before analyzing the spatial physics. The Nusselt number serves as our primary validation parameter. It confirms the exact heat transfer behavior inside the straight microchannels. The table compares our extracted Nusselt numbers against the empirical reference values. The cold channel calculates exactly 59.1 against the reference target of 62. The hot channel calculates exactly 65.3 against the reference target of 59. The overall Nusselt number deviation ranges roughly from 4 to 10%. This excellent validation accuracy proves the simulation matches the exact physical reality of the reference paper.

Figure 3: The graphs proving the linear physical relationship between the Reynolds number and the resulting heat extraction [1].

We now evaluate the reported performance plots based on this validated model. These charts map the exact physical relationship between the Reynolds number and the Nusselt number. The Reynolds number represents the fluid speed and its internal turbulence. As the fluid speed increases from 10000 up to 60000, the Nusselt number rises in a strict linear path. This physical behavior proves that faster fluid motion creates violent mixing near the wall, heavily increasing the heat extraction rate. Furthermore, the charts map different operational pressures exiting the channels. The cold channel evaluates pressures from 100 to 250 bar, while the hot channel evaluates pressures from 75 to 150 bar. The distinct plotted data lines overlap very tightly. This proves that changing the extreme baseline pressure has a relatively minor effect on the overall performance compared to the dominant influence of the fluid velocity.

Figure 4:  The steady-state spatial contour visualizing the conjugate heat transfer from the hot channel to the cold channel through the metal wall.

Finally, we must analyze the spatial temperature distribution to understand how the thermal energy actually moves. The three-dimensional contour reveals the exact counter-flow heat transfer behavior. The cold fluid enters the right side at exactly 373.15 Kelvin and gains heat progressively as it flows forward. It exits the channel on the left side reaching temperatures between 500 and 536 Kelvin. Conversely, the hot fluid enters the left side at exactly 664 Kelvin and cools steadily as it moves toward the right. The overall physical temperature ranges from exactly 373.15 to 664.03 Kelvin. The solid stainless steel wall acts as a conjugate heat transfer bridge between the two separate fluids. The smooth axial color gradient across this wall confirms fully developed thermohydraulic conditions. There is no flow separation or abnormal physical behavior. This specific contour proves the active NIST REFPROP model successfully captures both the sensible heat transfer and the extreme physical property variations of the supercritical fluid.

Frequently Asked Questions (FAQ)

• What is supercritical carbon dioxide?
  • When carbon dioxide is subjected to extreme pressure and heat, it passes a critical point. It stops being a normal gas or liquid. It becomes a supercritical fluid. It expands to fill a space like a gas, but it possesses the heavy density of a liquid, making it incredible for absorbing heat.
• Why use the NIST REFPROP database?
  • Supercritical fluids do not behave normally. Their physical density and heat capacity change violently with very small temperature shifts. Standard mathematical equations fail to predict this. The NIST REFPROP database provides the software with the exact, real-world physical properties required to simulate this fluid correctly.
• How does conjugate heat transfer work in this device?
  • Conjugate heat transfer means the software calculates two things at once. It calculates the heat moving inside the flowing fluid, and it calculates the heat conducting directly through the solid metal wall. Because the printed circuit heat exchanger relies entirely on this metal wall to pass energy from the hot side to the cold side, this mathematical model is strictly required.