Primary breakup in prefilming airblast nozzles is governed by gas–liquid shear and three-dimensional swirling airflow on the external liquid sheet. Understanding the breakup behavior from reservoir detachment to the onset of secondary breakup is ess...
Primary breakup in prefilming airblast nozzles is governed by gas–liquid shear and three-dimensional swirling airflow on the external liquid sheet. Understanding the breakup behavior from reservoir detachment to the onset of secondary breakup is essential for accurate spray prediction. However, previous studies have mainly employed simplified two-dimensional nozzles due to optical limitations, which restricts quantitative observation of the continuous axial evolution of breakup in realistic 3D configurations. Furthermore, the behavior of ligaments and non-spherical droplets formed after film rupture has not been sufficiently investigated.
In this study, a three-dimensional prefilming airblast nozzle with dual swirlers was examined while varying swirl vane angles (0°, 30°, 45°) and chamber pressure difference (3%, 5%). The breakup region within 20 mm downstream of the nozzle lip was divided into the sheet breakup, primary breakup and secondary breakup regions. High-speed camera imaging, PDPA, Spraytec, and shadowgraphy system measurements were used to quantify breakup modes, breakup length, droplet size and distribution, morphology, and gas–liquid relative motion.
In the sheet breakup region, the liquid film did not disintegrate instantaneously but underwent a repetitive process consisting of film accumulation, partial detachment, and subsequent rupture. During this process, multiple breakup modes were observed to coexist, including perforation, sheet disruption, ligament formation, and shear stripping. The breakup length showed a clear saw-tooth–type temporal fluctuation rather than a steady value, while its time-averaged length decreased almost linearly as the vane angle increased. In addition, a spanwise periodic structure with a characteristic spacing of approximately 4 mm was consistently identified, suggesting that sheet breakup follows a repeatable three-dimensional instability pattern rather than occurring in a random manner.
In the primary breakup region, the dominant fragmentation mechanism varied with operating conditions. Ligament breakup was frequently observed, whereas bag–rim disintegration and collision-induced fragmentation became prominent under specific flow conditions. Based on the downstream evolution of droplet size, a predictive model for D10 was developed, yielding a high correlation coefficient (R² ≈ 0.96). Across all test cases, the measured droplet size distributions were well represented by a log-normal function.
In the secondary breakup region, droplet shape parameters such as eccentricity, minor-to-major axis ratio, and shape factor decreased monotonically with axial distance, indicating gradual relaxation toward a spherical geometry. Evaluation of the Weber number revealed a high breakup potential in the near nozzle region, while downstream droplets were predominantly characterized by We < 12. This trend allowed the termination of secondary breakup to be identified based on Weber-number criteria.
Overall, the present experiments show that sheet breakup, primary atomization, and subsequent droplet shape stabilization in a three-dimensional prefilming airblast nozzle are not independent processes but are continuously connected through the combined effects of gas–liquid shear, local film thickness, and swirl intensity. The observed breakup-length behavior, spatial periodicity, droplet size prediction, and Weber-number-based breakup persistence provide physically grounded information that can be directly utilized for CFD initialization and spray modeling in practical airblast atomizer designs.