Cost-Effective Treatment of TMAH Wastewater

The disposal of TMAH (tetramethylammoniumhydroxide) wastewater from photoresist developer solutions is costly and toxic. To better address TMAH waste management, researchers from NEC Corp. (Kawasaki, Japan) and Nippon Shokubai Co. Ltd. (Himeji, Japan) developed a treatment method that is considerably less expensive than industrial disposal, while offering practical benefits over biological treatment methods and environmental benefits over conventional oxidation.

Using a two-step treatment, the method first decomposes the TMAH to trimethylamine (TMA) and ammonia, then decomposes the TMA using a selective oxidation catalyst to form primarily N₂, H₂ O and CO₂ . The treatment system's overall footprint is about one-sixth that of a biological treatment system and is tolerant to many treatment conditions. Though the catalytic oxidation system costs ~2.3× that of a biodegradation system, operation is simpler. The researchers performed long-term treatment of actual wastewater using this technique and reported their findings in the August issue of the IEEE Transactions on Semiconductor Manufacturing.

Many semiconductor fabs choose to dilute TMAH wastewater and treat it by biodegradation, but this process requires large, exclusive systems, which also generate N₂O, a greenhouse gas. Another alternative uses combustion followed by conventional oxidation, but this process also produces N₂O. For this reason, many fabs pay to treat TMAH wastewater as industrial waste. However, with increasing use of TMAH and tighter restrictions on total nitrogen content in wastewater, the need for a cost-effective, efficient treatment method has become more critical.

The NEC/Nippon Shokubai method uses pyrolysis to thermally decompose the TMAH ((CH₃)₄NOH) to TMA ((CH₃)₃N) and methanol (CH₃OH) at 300°C. The selective oxidation process follows, using a base metal series catalyst for nitrogenous compounds (Pd-V₂O₅-WO₃ supported on TiSi). Unlike Pt-based catalysts supported on Al₂O₃, the base metal series catalyst minimizes the concentration of harmful gases including NO_x, N₂O and CO.

To test the technique, the engineers condensed the developer wastewater to 29 g/L and sprayed it into the thermal decomposition section of the reactor (300°C) at 24 mL/hr, introducing air at 120 mL/hr. Selective oxidation was performed in an electric furnace that was packed with 60 cm³ of prefilter (4 layers of γ-Al₂O₃) and 28 cm³ of catalyst, and reacted at 275-350°C. The prefilter effectively removed the Si, P and S byproducts, which can significantly degrade catalyst performance. The researchers arrived at an optimum reaction temperature of 325°C, since lower temperatures resulted in NH₃ leakage and higher temperatures promoted NO_x generation.

Results showed the catalytic process converted 91 ppm of TMA to <0.1 mg/L TMAH, 1400 mg/L CH₃OH, 2900 mg/L TMA (as N) and 13,000 mg/L TOC. Results of the selective oxidation, compared with conventional oxidation, demonstrated greater conversion of nitrogen constituents to N₂ gas, rather than N₂ O and NO_x. Selective oxidation led to concentrations of 5 ppm NO_x (vs. 300-600 ppm), <6 ppm N₂ O (vs. 255 ppm) and 2 ppm CO (vs. 10 ppm).

The researchers then tested a large-scale system. For a factory that discharges 4 m³ /day of wastewater with 1% TMAH, the running cost was estimated at about one-ninth that of disposing the wastewater and about 2.3× that of a biological system. However, unlike the biological treatment system, the selective oxidation system proved capable of handling high concentrations of TMAH wastewater with better tolerance to fluctuations in TMAH concentration. The required footprint is 34 m² vs. 200 m² and cost is ¥4600/m³ vs. ¥2000/m³. Disposing of the equivalent amount of industrial waste would cost ¥40,000/m³.