From wet benches to cooling systems, factories use enormous amounts of water. A typical foundry produces millions of gallons per day. However, in this era of climate change, water supplies are becoming less reliable and municipal water systems are becoming more restrictive. For example, local governments may limit a factory’s ability to pump from the public water supply or supply only treated wastewater rather than potable water. Advanced water management is essential for business continuity.
Water management considers both the input water supply and the output waste stream, as well as the factory’s own ability to reuse wastewater. On the input side, factories may need to accept less-than-ideal sources such as wastewater or seawater. On the production side, factories are trying to recover both usable water and process chemicals. As Gradiant Technologies co-founder and COO Prakash Govindan explained, in both cases, the first step is for the factory to understand its water supply and segregate sources that require different levels of treatment. That’s it. Separating high-solids water, corrosive chemicals, etc. greatly improves water management efficiency. A factory may have as many as 15 different waste streams, each with different treatment requirements.
Segregating waste streams also facilitates chemical recovery. For example, Govindan said isopropanol is a strong candidate for chemical recovery efforts. It is much more expensive than water and is found in more industrial waste than metals such as copper.
Purification techniques such as reverse osmosis are more cost effective if recoverable chemicals are extracted and solids are removed first. Reverse osmosis works by forcing the water to be purified through a semi-permeable membrane. Contaminants remain on the pressurized side of the membrane while pure water diffuses. Reverse osmosis “can clean anything,” says Govindan, but the more difficult the cleanup task, the higher the cost and energy requirements.
To match remediation resources to incoming waste streams, Gradiant creates an optimized digital twin of the facility, supported by real-time sensor data from the actual facility. Machine learning models use both historical and real-time information to predict the need for membrane replacement or other maintenance. At a microelectronics facility in Singapore, Gradiant was able to recover nearly 90% of incoming wastewater, reducing the factory’s freshwater requirements accordingly.
Catherine Derbyshire
(all posts)
Katherine Derbyshire is technical editor at Semiconductor Engineering.
