Semiconductor

Ultra-Pure Water (UPW) is the most heavily consumed process fluid in semiconductor manufacturing. A single 300 mm wafer requires approximately 8,000 liters of UPW — equivalent to one person's drinking water for a decade. But UPW is not simply "very clean water." From the first principles of physical chemistry, UPW is an aggressively active solvent driven by an enormous chemical potential gradient.

The root of this behavior lies in the Gibbs free energy of the system. Water at 18.2 MOhm-cm resistivity is a profoundly non-equilibrium state — it contains virtually no ions. When UPW contacts any material surface, the chemical potential gradient drives ions from the material into the water until solubility equilibrium is reached. This is not a question of "whether" ions will leach — it is a question of "how much."

This is why the semiconductor industry's acceptance criterion for UPW piping materials is "zero tolerance" rather than "below threshold." A 5 nm-node fab produces wafers worth tens of millions of dollars per day. A 1% yield loss caused by pipe-material leaching translates to hundreds of millions of dollars in annual losses. At this scale of consequence, pipe material selection transcends engineering economics and enters the domain of process reliability philosophy.

UPW piping faces a unique challenge: the fluid being transported is itself the most aggressive solvent in the system, while the downstream process is exquisitely sensitive to any contamination. From a materials-science perspective, the three candidate pipe families diverge at the atomic level:

UPW system failure in a semiconductor fab is not a "fix a leak and move on" problem. Any detectable change in UPW quality means that tens of thousands of wafers have already completed process steps in contaminated rinse water. From a root-cause analysis perspective, every UPW water-quality incident follows the same causal chain:

Pipe material selection compromise → Long-term trace leaching undetected → Wafer defect density gradually rising → Yield loss manifests in production data → Tens to hundreds of millions of dollars in damage already incurred

An even more insidious risk lies in the time-delay effect: the inner surface leaching of metallic pipes is a gradual, cumulative process. Water-quality data from the first 3-6 months after fab commissioning may appear entirely normal — because the pipe's inner-surface oxide layer has not yet been sufficiently attacked by UPW. Once the protective oxide layer is dissolved through prolonged contact, the high-purity metal substrate becomes exposed, and the leaching rate surges — just as the fab has entered full production. The cost of a shutdown-and-replace at that point is 5-10 times higher than pre-production testing and replacement.

The only way to eliminate this risk chain is to exclude materials with inherent remineralization risk at the design-selection stage — which is precisely the first-principles logic behind GRE-RTR becoming the preferred pipe material for semiconductor UPW systems.

Grounded in a deep first-principles understanding of semiconductor UPW systems, LEISA provides the following specialized testing services for non-metallic pipes in semiconductor applications. Our testing methodology follows a three-tier progressive logic — "first verify the material, then verify the product, finally verify the system" — ensuring that the pipe is clean before installation, rather than hoping to clean it afterward.