Heat transfer mechanisms in supercritical CO2 flows near the critical point length scale effects
Abstract
Supercritical carbon dioxide is poised to revolutionize a range of Navy’s mission critical systems including portable power production unit, centralized coolant system, and standalone cooling device. However, lack of accurate predication tools (e.g., heat transfer coefficient correlations) and insufficient knowledge about the mechanisms controlling heat transfer processes are hindering its practical realization in key energy and cooling systems. Large discrepancies and inconsistencies between available correlations — which are mostly empirical and are based on adaptation of constant properties correlations, such as the Dittus-Boelter or Gnilienski correlations— exacerbate the situation. The proposed effort seeks to change this state-of-affairs through an in-depth experimental study, supported by some modeling, about internal flow spanning two orders of magnitude length scales (from ~200 ?m to ~25 mm), three orders of magnitude Reynolds numbers (from ~100 to ~105), six orders of magnitude Rayleigh number (from 25,000 to 5×1010), and heat fluxes from O(0.1 W/cm2) at the macro scale to O(100 W/cm2) at the micro scale. Detailed spatiotemporal fluid temperature and flow velocity inside the boundary layer and simultaneous surface temperature measurements will be performed to reveal the dominate heat transfer mechanisms. (Note that for internal fully developed flow, the entire cross-sectional area of the channel is inside the boundary layer.) The fluid temperature in the vicinity of the wall will be used to infer its thermal conductivity, and hence, the contribution of conduction to the heat transfer coefficient (convection = conduction + advection). The velocity field measurements will be leveraged to obtain the turbulent kinetic energy (TKE) level, which in turn, will provide a measure of flow mixing (i.e., advection). In general, it is expected that the turbulent kinetic energy level will be closely correlated with the heat transfer coefficient. A considerable change in the turbulent kinetic energy level will likely suggests a change in the dominant heat transfer mechanism. Quantitative measure of this relation will allow to quantify transitions between heat transfer mechanisms and to develop better pertinent heat transfer coefficient mechanistic correlations. However, in the presence of gravity, turbulence may be anisotropic as the eddies caused from near-wall viscous shearing may be formed differently from the eddies formed due to viscous shearing in the buoyant motion. As a result, a single value for TKE will not capture the full turbulence stress tensor, and distribution of TKE among different directions. For this purpose, for a select few cases, simultaneous measurement of fluctuating temperature across the cross-section will be performed along with all three components of velocity so that anisotropic nature of turbulence can be identified, and turbulent heat flux can be computed as well. While at the micro scale, the fluid temperature and flow field will not be measured, high-speed images coupled with spatiotemporal surface temperature measurements will allow to reveal key heat transfer processes, compare them to results from the larger scales, and develop accurate correlations pertinent at length scales in the range of 0.1 mm to 1 mm. The results from all these measurements with supplemental data/correlations from the literature will then be evaluated, consolidated, and broken down to separate heat transfer modes. They will then be reconstructed to develop transition criteria and mechanistic correlations that are independently validated
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 10, 2018
- Source ID
- N000141812362
Entities
People
- Subith Vasu Sumathi
Organizations
- Office of Naval Research
- United States Navy
- University of Central Florida Board of Trustees