Effects of Bubble Interactions on Pool Boiling Heat Transfer

This thesis investigates whether geometry controlled bubble coalescence can measurably affect the boiling curve and the heat transfer coefficient (HTC) during saturated pool boiling of water at 1 atm on Si/SiO₂ chips. The experimental platform uses 20 × 20 mm², 380 µm thick silicon heaters with a hydrophilic SiO₂ top layer and nominally identical micro-cavities (diameter 75 µm). Three geometric degrees of freedom are isolated while keeping fluid, chemistry, and procedures fixed: intra-pair spacing C2C (300–1500 µm), inter-pair spacing P2P (3000–7000 µm), and cavity depth (5, 50, 100 µm). Quantitative boiling curves q″(ΔT) and HTC=h(ΔT) are reconstructed from electrical measurements; high-speed videos recorded at matching operating points provide regime-level interpretation. Repeatability is demonstrated on the same chip and across nominally identical chips, and a flat-surface baseline falls within the expected smooth-surface envelope. A consistent picture emerges. (i) Reducing C2C primarily affects the low-to-medium heat flux branch: tighter spacing keeps more sites active even at low heat flux and lowers the superheat required to exit natural convection, increasing HTC; once all sites are active, curves converge. (ii) P2P imprints the medium-to-high heat flux branch: too small P2P (≈ 3000 µm here) promotes long-lived, coalesced vapor structures near the chip’s center that impair rewetting and soften the high-flux slope, whereas larger P2P clears vapor more quickly. (iii) Cavity depth acts across the full range: 5 µm underperforms 50–100 µm because shallow pits retain vapor less reliably after departure; beyond ≈ 50 µm, added depth yields diminishing returns, and conduction-distance effects are sub-kelvin and secondary. The results condense into layout-oriented guidance: use small C2C to accelerate activation; avoid overly small P2P at high flux; and choose depths ≥ 50 µm to secure stable re-activation, all within the stated operating envelope.