Ruihong Electronics LCD photoresist is manufactured in two positive-tone formulations, RHLCD-100A and RHLCD-200B, for thin-film-transistor array layers, black matrix patterns, and column spacer arrays in active-matrix liquid crystal displays. The liquid composition combines a partially acrylated phenolic resin matrix, a diazonaphthoquinone photoactive compound, and propylene glycol monomethyl ether acetate as primary solvent. Representative viscosity at 25 °C measured with a cone-and-plate rheometer per DIN 53019-2:2008 is 6.2 mPa·s ±0.8 mPa·s. Solid content after oven drying at 105 °C for 2 h is 19.5 wt% ±1.0 wt%. For a 1.20 µm ±0.05 µm soft-bake film spin-coated at 2,000 rpm on a 300 mm square glass substrate, line/space resolution is 1.5 µm using a g-line stepper with numerical aperture 0.45 at exposure energy 80 mJ/cm² ±10 mJ/cm². Filtration through a 0.2 µm polytetrafluoroethylene membrane reduces point-of-use particle counts to 12 particles/mL or fewer for particles ≥ 0.5 µm, evaluated by laser particle counting calibrated with NIST-traceable latex spheres. Adhesion to indium tin oxide and alkali-free glass after postbake meets 4B or better using ASTM D3359-23 method B cross-cut tape testing.

In production, the resist is dispensed through a 0.1 µm point-of-use fluoropolymer filter onto static or slow-rotation glass substrates. Soft-bake is performed on a proximity hot-plate or enclosed oven coater with plate temperature uniformity ±0.5 °C, maintained at 88 °C to 93 °C for 90 s. For RHLCD-100A array layers, vacuum contact dry and post-exposure bake are omitted; refractive-index matching at 1.63 at 365 nm reduces standing-wave amplitude at the resist–substrate interface. Development in 2.38 wt% tetramethylammonium hydroxide at 23 °C ±0.5 °C for 60 s ±5 s yields sidewall angles of 78° to 84°. A deionized water rinse and spin-dry precedes final postbake at 220 °C for 30 min. The developer puddle process uses a low-foam 6-nozzle dispense pattern on a 6G TEL CLEAN TRACK ACT-8 developer; edge bead removal with PGMEA occurs before soft-bake at 2.0 mm to 3.0 mm from the substrate perimeter.

On a 6G coater load lock, substrates are not exposed to ambient longer than 10 min between adhesion promotion and resist coating. Relative humidity above 60% at 23 °C lowers glass surface contact angle and shifts dynamic wetting spread by 1.5 mm to 3.0 mm, which can alter edge bead removal width. The resist is stored at 5 °C to 25 °C; containers are equilibrated to cleanroom temperature for 8 h before opening to prevent condensation. In high-volume production, batch-to-batch viscosity variation below 0.3 mPa·s is required to hold spin-coating thickness variation within 3.5% across an 1,850 mm × 1,500 mm substrate.

What Distinguishes RHLCD-100A from Diazo–Novolak Color Filter Resists?

Compared with conventional diazo–novolak color filter resists, RHLCD-100A exhibits lower nitrogen evolution during postbake because the matrix resin is partially acrylated with tetrahydrofurfuryl acrylate. This modification raises optical transmittance at 550 nm to 94.5% ±0.5% when measured per ISO 13468-2:2019. The photoactive compound loading is 7.0 wt% ±0.3 wt% on solids, lower than the 9.0 wt% typically found in older color filter resists, which reduces dark erosion in unexposed areas to 12 nm/min in 2.38% tetramethylammonium hydroxide. Table 1 compares representative properties of RHLCD-100A to a generic diazo–novolak positive resist and a generic acrylic black matrix formulation.

The lower photoactive compound loading and modified resin polarity of RHLCD-100A narrow exposure latitude to ±10%, tighter than the ±15% typical of diazo–novolak systems. This characteristic prioritizes optical transmission and outgassing control over broad exposure latitude, making the formulation suitable for TFT array layers above 1.5 µm resolution and for color filter black matrices where volatile condensate is a contamination risk. Users shifting from diazo–novolak resists should re-qualify stepper dose uniformity because linewidth linearity degrades more sharply below 1.2 µm nominal size. Published data for sub-micrometre RHLCD-100A dimensions on 0.45 numerical aperture tools is limited.

Against acrylic black matrix resists, RHLCD-200B reduces pattern collapse because of lower molecular polarity and greater oxygen plasma etch resistance. Isothermal thermogravimetric weight loss at 220 °C over 30 min is below 0.5 wt%, compared with 1.2 wt% for a conventional carbon-loaded acrylic black matrix. Optical density at 550 nm is 3.2 per µm ±0.2 per µm, measured on a densitometer calibrated per ISO 5-3:2009, which is lower than 4.0 per µm achievable with heavily carbon-loaded acrylic systems. RHLCD-200B therefore is applied where resolution and outgassing control outweigh maximum optical density.

When Edge Bead Removal Degrades Linewidth Uniformity across G6 Substrates

Edge bead removal with PGMEA on RHLCD-100A creates a low-viscosity zone at the bead edge. If solvent evaporation is incomplete before soft-bake, capillarity-driven flow pulls resist toward the edge during the first 10 s of heating. On a 6G substrate of 1,850 mm × 1,500 mm, this flow produces an edge-wide film thickness swell of 8 nm to 15 nm in the outermost 30 mm, causing critical dimension deviation of 0.08 µm to 0.15 µm in the first 50 mm from the edge. The critical control is a two-step edge bead removal: PGMEA dispense at 1.2 mL/s for 2 s, followed by high-speed spin at 1,500 rpm for 5 s before soft-bake. Lines that transfer the substrate directly from edge bead removal to a proximity hot plate without the intermediate high-speed spin show batch-to-batch standard deviation in edge critical dimension of 0.11 µm, compared with 0.04 µm when the intermediate spin step is used.

This effect is observable on a production coater with dynamic dispense and exhaust flow set to 0.3 m/s. Exhaust velocity below 0.2 m/s increases solvent redeposition at the bead edge. Hot plate temperature is profiled daily with a calibrated 9-point thermocouple array; plate-to-plate variation must not exceed 0.7 °C. Substrates soft-baked at 85 °C retain residual solvent above 8.0 wt%, and later vacuum baking creates solvent bursts that generate pinhole densities up to 5/cm² in a 1.0 µm film. At 95 °C, residual solvent falls below 2.0 wt%, and the dissolution inhibitor can crystallize at the substrate interface, producing undercut profiles and interfacial residues. The soft-bake window is therefore clamped at 88 °C to 93 °C, with exhaust damper position fixed at 45% ±5% to balance evaporation rate and air flow stability.

The resist is incompatible with ammonia and primary amine vapor. Substrates stored in a cassette previously used for amine-based adhesion promoter show surface pH above 9 and dark erosion increases to 45 nm/min in 2.38% tetramethylammonium hydroxide, causing scumming at the substrate–resist interface. Storage under humidity above 60% RH at 23 °C accelerates diazonaphthoquinone hydrolysis. Opened containers should be used within 72 h or pre-dried with molecular sieve desiccant. The product is also incompatible with plasma-cleaned silicon nitride having residual fluorine above 2 at%; an adhesion promoter free of secondary amines is required for these surfaces. These boundaries derive from production qualification runs on glass and silicon nitride substrates and are revalidated with supplier batch data under ISO 9001:2015 documentation procedures.

Solvent Retention, Soft-Bake Kinetics, and Vacuum-Dry Uniformity

Soft-bake behavior of RHLCD-100A is governed by residual PGMEA diffusion through a 1.20 µm cast film. Thermogravimetric analysis of an unpigmented film at 10 °C/min under nitrogen shows a single mass-loss event between 110 °C and 145 °C, with 19.5 wt% weight loss matching the stated solids content. On a proximity hot plate, solvent retention after 90 s at 90 °C is 4.0 wt% ±0.5 wt%, measured by extraction gas chromatography with an internal standard per ISO 11890-2:2020. Solvent retention below 2.0 wt% corresponds to a 37% drop in dissolution inhibition. Retention above 8.0 wt% leads to nitrogen blistering during postbake. The vacuum-dry step, where used before exposure, should be performed at 0.5 kPa for 30 s. Pressures below 0.2 kPa cause surface skinning, and pressures above 1.0 kPa are insufficient to remove edge solvents.

Film shrinkage after postbake is 12.0% ±1.5% thickness reduction, measured by spectroscopic ellipsometry at 550 nm before and after 220 °C postbake. This shrinkage is lower than the 18% ±2% observed for a comparable novolac resist under the same conditions. The partially acrylated phenolic network resists densification during thermal cure. For black matrix and column spacer applications, RHLCD-200B is applied over indium tin oxide or color filter layers at spin speeds from 1,200 rpm to 2,800 rpm, yielding dry film thickness from 1.0 µm to 2.5 µm. The spacer dot diameter after photolithography is 10 µm ±0.3 µm; after thermal reflow at 220 °C for 30 min, edge slope remains 72° ±4°.

Compliance with flat-panel display chemical specifications is summarized in Table 2. Metal ion limits are determined by inductively coupled plasma mass spectrometry after acid digestion per ISO 11885:2007 and EPA 6020B. The product is managed under REACH 1907/2006 for the EU market, and RoHS compliance is verified for lead, cadmium, mercury, and chromium VI below 100 ppm by weight per IEC 62321-5:2013. Halogen content is below 900 ppm total chlorine by combustion ion chromatography per EN 14582:2016. Certificates of analysis include lot-specific viscosity, solids, peroxide, and particle count.

In G6 black matrix production, the resist is dispensed through a 0.1 µm PTFE point-of-use filter, and total dispense volume per substrate is 38 mL ±3 mL for a 1.2 µm film. The black matrix layer uses the same developer chemistry as RHLCD-100A, reducing cross-contamination risk in integrated coat-develop lines. Process qualification for RHLCD-200B spacer-dot configuration under 6th-generation slot-die coating is limited in published literature; manufacturers should reproduce thickness and edge-erosion data on target equipment before full-volume deployment.