From Ultrasonic Photoresist Droplets to a Continuous Film: Atomization, Coalescence and Drying
Ultrasonic spray coating changes how photoresist reaches a substrate, but it does not remove the physics of wet-film formation. Mechanical vibration creates droplets, and controlled delivery distributes them across a target. A continuous resist layer still requires those droplets to interact, spread and lose solvent under suitable conditions. The idea that there is no need for redistribution after landing is therefore incomplete. Ultrasonic Photoresist Coating works through a controlled balance between useful local leveling and undesirable larger-scale movement that can reduce coverage at difficult features.
Separate atomization from the finished resist layer
Liquid supplied to a vibrating atomizing surface forms droplets whose behavior depends on operating conditions and liquid properties. Nozzle frequency is relevant, but viscosity, surface tension, feed and vibration state also matter. A nominal frequency does not uniquely specify the droplets arriving on a wafer. Transport gas, flight distance and evaporation can change the delivered population. Measure or characterize the relevant behavior for the actual formulation rather than transferring a water-based droplet description to a solvent-rich photoresist.
Low-velocity delivery is not the same as zero movement
Guiding gas helps direct the mist, and its interaction with the substrate and topography can influence local access. Droplets may be diverted near structures or miss recessed surfaces. Once on the substrate, wetting and capillary effects remain. Gravity can matter for some orientations, but it is not the only source of motion, just as gravity is not the sole cause of spin-coating redistribution. A technical description should connect the actual transport and surface conditions rather than explain the process through one force alone.
Understand evaporation before impact
Solvent can leave a droplet during flight, changing its concentration and viscosity. This may help limit subsequent macroscopic flow, but excessive loss can prevent the deposited material from merging into a continuous film. The result may be roughness, isolated deposits or weakly attached particles. Longer distance or stronger drying is not automatically better. A useful trial investigates flight and substrate conditions together, with observations of the deposited and baked film rather than only the visible plume.
Why coalescence is necessary
Initially separated droplets must overlap and merge sufficiently to close gaps. Their spreading depends on substrate chemistry, residual solvent and neighboring deposits. If liquid mobility is too low, pinholes or an orange-peel texture can remain. If it is too high, material may migrate from a convex edge toward a recess. The acceptable balance depends on the intended resist thickness and geometry. A perfectly stationary droplet is not automatically a desirable precursor to a smooth, continuous photoresist coating.
Surface preparation changes this balance. Cleaning, dehydration and any appropriate adhesion treatment should be specified for the actual substrate and resist. Different surface materials on one wafer can wet differently, so a flat silicon reference may not represent a patterned oxide-and-metal structure. Record the interval between preparation and coating. Solvent-only controls or simple wetting trials can help separate incoming surface problems from atomization problems before a larger recipe study begins.
Control delivery and accumulation
For a simple raster, nominal dry solids per area scale with delivered flow, solids concentration and pass count, and inversely with travel speed and path spacing, subject to capture. This is a planning relationship, not a complete film model. It assumes a representative feed and repeated footprint. Edges, overspray, topographic area and local flow can invalidate a uniform interpretation. Calibrate against measured dry loading or thickness, and inspect critical locations separately. The number of passes alone cannot establish final thickness.
Pass interval changes the starting surface for the next pass
A second pass may land on a wet layer, a partly dried layer or a surface with substantially reduced solvent. These states can produce different merging and leveling behavior. Record the interval and the path sequence, including time spent returning or turning. A recipe that works on a small coupon may create different interpass timing on a larger panel. Transfer should preserve the relevant wet-film history or explicitly reoptimize it, rather than simply use the same flow and speed.
Use temperature and exhaust as process variables
The actual substrate temperature depends on support contact, heat capacity and evaporative cooling. A heater setting does not describe the whole surface during spray delivery. Exhaust must manage solvent vapor and stray aerosol, but its flow can also affect evaporation and deposition. Define the operating arrangement and check representative locations. Increasing temperature to remove pooling may create roughness elsewhere, so adjustments should be evaluated across the substrate and feature types rather than at the original failure location alone.
Recognize common defect mechanisms
Pinholes may come from inadequate coalescence, dewetting, particles or other local disturbances. Corner thinning may reflect wet-film redistribution or limited deposition access. Excess thickness near a recess can reflect both incoming material and subsequent flow. A defect image rarely identifies one cause by itself. Use controlled changes and staged inspection where practical. Check the liquid path for dried polymer, gels and contaminants; ultrasonic operation does not guarantee that all deposits are removed or that every resist remains stable during storage.
Separate dry-film quality from lithographic function
Soft bake changes solvent content and can alter thickness or stress. Exposure and development then test the resist’s chemical and mechanical suitability. A continuous as-coated film can still develop poorly if reformulation or processing changed its behavior. Inspect after each relevant step, using the same critical geometry. If the resist serves as an etch mask, include etch survival and pattern transfer. This prevents an attractive coating micrograph from being mistaken for a qualified semiconductor process.
Design measurements around competing outcomes
Measure planar thickness distribution, local corner and sidewall coverage, roughness and defect frequency with suitable methods. State sampling and uncertainty. A recipe may improve corner coverage while increasing planar roughness, so retain both responses. Use independent runs and resist preparations to assess repeatability. Predefine what constitutes an unacceptable pinhole or uncovered region according to the device function. Reporting only average thickness can obscure the very defect that motivated the switch from spin coating.
Develop a usable operating window
Start with resist compatibility and stable atomization, then investigate deposition and drying. Change variables through a structured plan so effects can be interpreted. Record liquid age, preparation, cleaning and startup state. Include pause-and-restart tests because solvent loss at the nozzle can change early samples. Confirm the selected window on representative topography and area. A process description should state how the film forms and where it fails, not promise that every droplet remains exactly in place under all conditions.
Keep representative samples from the beginning and end of an extended run to detect changing concentration or nozzle condition. If the same nominal program produces a different film after idle time, define a recovery procedure and verify it. A useful production-oriented result includes the first acceptable sample and the amount of material consumed before it, not only the stable portion of the run. This makes both the film-quality assessment and the material-use comparison more representative of practical operation.
Where possible, compare inspection at defined intermediate states: after deposition, after soft bake and after development. The point at which a defect appears can narrow its likely origin. A discontinuity present immediately suggests a different investigation from a film that lifts only during development. Use matched samples if the inspection is destructive. This staged evidence helps prioritize chemistry, drying or adhesion work without assuming that every visible failure originates in atomization.
Arrange a photoresist film-formation trial
Qualified manufacturers and R&D teams can contact Cheersonic at market2@cheersonic.com to discuss photoresist, catalyst coating or other functional coating projects and arrange a coating sample test. Review Cheersonic ultrasonic coating equipment solutions and provide resist identity, compatible solvent, substrate geometry, target thickness and observed defects. A trial linking atomization, coalescence and downstream lithography can identify a practical process window for the actual application.

