Application context and engineering objectives

Microfluidic devices use lithographically defined channels, chambers, mixers, filters and ports, often fabricated in glass, silicon, polymer or bonded multilayer structures. SU-8 is a negative epoxy-based photoresist widely used where relatively thick, mechanically robust patterned structures are required. Microfluidic process flows can require a thick film, multiple coating passes, high-aspect-ratio structures or coating on substrates that are not ideal for conventional spin processing. These characteristics make atomized deposition worth evaluating in selected development settings, but SU-8 rheology, solvent content, film stress and exposure requirements make process qualification essential.

The engineering objective is not simply to deposit liquid on a surface. It is to form a functional resist layer whose thickness, continuity, adhesion and chemical state are compatible with the complete lithography sequence. Ultrasonic atomization is one candidate deposition route: a vibrating atomizing surface breaks a supplied liquid into droplets that can be transported toward a moving substrate. The process can be attractive for non-contact delivery, controlled low-flow operation and selected geometries, but its suitability depends on the formulation, substrate, target film and production constraints. It should be compared with spin, slit, meniscus, dip and conventional spray coating using representative test structures.

Material characterization before equipment trials

Record the resist supplier, product grade, lot, solids content, solvent system, viscosity, surface tension, density, recommended filtration and storage conditions. Obtain the current technical and safety data sheets. Do not assume that two resists with a similar name or nominal viscosity behave identically. Temperature changes viscosity and evaporation; solvent composition can change wetting and drying; ageing and contamination can alter the atomization response. Establish a controlled conditioning procedure and measure fluid temperature at the point of dispense.

Wetted components must be compatible with the complete formulation, not merely its main solvent. Check tubing, seals, fittings, reservoirs and cleaning agents against chemical supplier data and the facility’s approved materials list. Avoid unverified material substitutions. Filter only where the resist supplier permits it and select a filter chemistry and pore size that do not remove functional components or alter the formulation. Degassing may help when bubbles are observed, but it too must be evaluated for volatile mixtures.

Substrate preparation and wetting control

Substrate cleanliness and surface energy often dominate early coating results. Define a repeatable cleaning sequence for the actual substrate material and prior process history. The sequence may involve compatible detergent, high-purity rinsing, controlled drying, dehydration or a qualified surface-activation step. It must not be copied blindly from another substrate: glass, silicon, polymer films, metal layers and patterned stacks respond differently to chemical and plasma exposure.

Measure contact angle at defined locations and at a controlled interval after cleaning. Contact angle is a useful indicator of surface condition, but it does not by itself predict final film uniformity. Excessive wetting can produce edge accumulation or drainage; insufficient wetting can cause islands, pinholes or dewetting. Record the time between cleaning, activation and coating, as airborne organic contamination and surface recovery can change the result.

Atomization and deposition variables

Build a parameter map rather than searching for a single magic setting. Candidate variables include liquid flow, ultrasonic drive condition, carrier-gas flow where used, nozzle-to-substrate distance, traverse speed, raster pitch, pass overlap, substrate temperature and number of passes. Their effects are coupled. A higher flow may raise deposited mass but also increase local pooling. A longer stand-off can broaden the footprint and increase exposure to ambient disturbances. A slower traverse may improve loading but can create wet bands if the solvent does not evaporate uniformly.

First characterize the spray footprint on a sacrificial coupon. Measure its width and deposited-mass profile at several positions and distances. Then choose a raster pitch that gives controlled overlap. Confirm stage straightness, acceleration behaviour and repeatability at the edges of the coating area. Keep the nozzle and substrate parallel where the application requires it, and define a consistent start-up and shut-down sequence so that transient flow does not mark the active area.

Film formation and drying

Droplets must arrive, spread, merge and lose solvent in a coordinated way. If droplets dry before coalescing, a rough or granular surface may form. If the wet loading is excessive, the film can develop ridges, runs, edge build-up or drying marks. Evaporation rate depends on solvent volatility, local airflow, substrate temperature, droplet size and the time between passes. Do not assume that a visually smooth wet surface will remain uniform after drying.

Use the resist manufacturer’s bake guidance as the starting point. Assess hotplate uniformity, substrate contact, ramp conditions and exhaust. Changes to bake temperature or duration can alter residual solvent, film stress, adhesion and photosensitivity. Any departure from the recommended profile should be treated as a controlled experiment with a documented reason and downstream lithography checks.

Metrology and process-window development

Use a measurement plan that spans the useful area, not just the centre. Depending on film and substrate, thickness can be evaluated by profilometry, ellipsometry, reflectometry or another validated method. Transparent, multilayer or optically complex substrates may require a suitable optical model. Optical microscopy helps identify pinholes, particles, streaks and edge defects. Record mean thickness, range, within-substrate non-uniformity, edge exclusion and defect density using definitions agreed before testing.

A small design of experiments can identify the most influential variables. Begin with a screening study, then refine the factors that materially affect the response. Use replicate substrates and include a control coated by the current baseline process. Record resist lot, fluid temperature, substrate preparation, dispense time, nozzle settings, motion path, bake conditions and metrology. A single successful coupon is a feasibility observation, not evidence of process capability.

Functional lithography validation

The coating process must be judged by patterned results. Evaluate adhesion, exposure latitude, critical-dimension variation, sidewall profile, residual scum, defectivity and compatibility with the downstream etch or lift-off sequence. Select test patterns that represent the real feature sizes, pattern density and topography. Include both dense and isolated features where relevant. For thick films, examine through-thickness exposure and development; for thin films, determine whether local thickness changes affect dose and focus margins.

Keep coating and lithography variables separated during initial qualification. If exposure dose, focus, bake and development are all changed at once, it becomes difficult to identify the source of an improvement or failure. Establish a baseline lithography recipe, then change one controlled factor or use a structured experimental design. Repeat promising conditions across different substrate positions and, eventually, different material lots.

Defect diagnosis, safety and scale-up

Pinholes may be associated with particles, poor wetting, bubbles or local dewetting. Streaks may point to motion errors, non-uniform overlap, unstable liquid feed or drying gradients. Edge build-up can be related to substrate geometry and surface energy. Thickness bands may indicate an unsuitable raster pitch or a non-uniform footprint. Diagnose defects by correlating their location and morphology with the motion path, nozzle orientation, substrate cleaning and measured film map.

Solvent-based photoresists require the engineering controls specified by the safety data sheet and facility EHS procedures. Use suitable ventilation, compatible closed liquid handling where appropriate, approved waste collection and required personal protective equipment. Consider flammability, vapour exposure, static control and spill response. Do not operate an open spray process outside a properly assessed enclosure. Verify that the full system, including exhaust and waste handling, is suitable for the actual chemistry.

Before scale-up, test repeatability across operators, days, substrate lots and equipment start-ups. Determine cleaning intervals, nozzle inspection criteria, fluid-change procedures and acceptable process drift. Assess cycle time, consumables, maintenance, material utilization and rework. A coating method that performs well on a small coupon may require a different motion strategy, enclosure or substrate support when the active area becomes larger.

Conclusion and sample-coating evaluation

Use a stepped development plan: flat coupons for film formation, patterned coupons for lithographic response, then representative microfluidic structures for bonding and flow tests. Follow the SU-8 supplier’s recommendations for prebake, exposure, post-exposure bake and development, and qualify any modified sequence independently.

Cheersonic invites qualified manufacturers, process engineers and research teams to discuss ultrasonic atomizing coating evaluations. To make a sample trial meaningful, provide the substrate material and dimensions, resist family, solvent system, target dry-film thickness, active coating area, downstream process and acceptance criteria. A representative test plan can then be used to assess feasibility without assuming unverified performance.

Explore ultrasonic coating solutions

See ultrasonic coating equipment solutions and ultrasonic atomizing coating technology.

Designing representative microfluidic test structures

A useful microfluidic test vehicle should reproduce the features that matter in the final device: channel widths, wall heights, aspect ratios, corner geometries, reservoirs and bonding lands. Include both isolated and dense structures because local pattern density can affect exposure, development and residual stress. Where a glass substrate is bonded to another glass or polymer layer, reserve test areas for evaluating surface cleanliness and bond strength after the full thermal and chemical sequence. Coating quality should be linked to device function rather than judged only from a flat-film thickness map.

For SU-8, maintain traceability of the time between coating, soft bake, exposure and post-exposure bake. Thick films can respond strongly to thermal history, so document hotplate calibration, substrate contact and cooling conditions. During development, monitor whether narrow channels clear fully and whether tall features remain attached. After bonding, use an appropriate leak or flow test and inspect for channel collapse, obstruction or delamination. This sequence reveals failure modes that a resist coupon alone cannot detect.