In semiconductor ultrapure water (UPW) systems, the core objective of reverse osmosis (RO) membrane integrity verification is to confirm two critical conditions:
1.The brine-side seals and central tube joints of membrane elements remain intact, with no cross-leakage from the concentrate side to the permeate side;
2.The membrane sheet itself is free of physical damage, cracking, desalination layer peeling or oxidative degradation, and has no defects allowing direct solute penetration.
Below are standardized integrity verification methods tailored for on-site semiconductor UPW operating conditions, sorted by testing accuracy and applicable scenarios. This document clarifies the judgment criteria and compliant operation key points for each test procedure.
I. Vacuum Decay Test (VDT) – Offline / Online During Shutdown, Highest Precision
This test is the industry standard for factory outgoing inspection, warehouse re-inspection of membrane elements, and full vessel integrity testing under system shutdown conditions. It features the highest detection sensitivity among all verification methods available, capable of accurately identifying tiny pinholes on membrane sheets, minor O-ring leakage, and sealing defects at central tube joints. It fully meets the stringent requirements of semiconductor ultrapure water systems for high cleanliness and zero leakage.
Test Principle
Fully drain all residual water from a single membrane element or full membrane vessel. Connect vacuum equipment to the permeate side, evacuate to a negative pressure of −80 kPa or higher, then close the valve for pressure holding. If leaks exist in the membrane sheet or sealing components, ambient air will infiltrate the permeate side through defective points, resulting in continuous vacuum decay of the system. The integrity of the membrane element can be judged based on the pressure drop magnitude over a fixed test duration. Complete drying of the membrane cavity is not required; thorough wetting and full drainage serve as the core precondition before testing.
Acceptance Criteria (for standard 8040 RO Membranes)
Evacuate to −90 kPa, hold steady pressure for 5 minutes:
1.Total pressure drop ≤ 3 kPa: The membrane element and sealing assembly are intact and pass inspection;
2.Pressure drop between 3 kPa and 10 kPa: Minor defects exist, requiring recheck and troubleshooting;
3.Pressure drop ＞ 10 kPa: Obvious leakage or membrane sheet damage is confirmed; the element is prohibited from being put into service in ultrapure water systems.
Core Operating Key Points
1.Membrane elements must be fully wetted and drained prior to testing. Dry membrane elements shall never be tested directly. The dense polymer layer of dry membranes contains inherent microscopic pores, which will cause false vacuum decay and lead to misjudgment. Full wetting closes native micro-pores, leaving only leakage channels from physical defects detectable.
2.Perform air tightness pre-inspection on the entire connecting pipeline, fittings and valves before testing to eliminate interference from external system leakage, ensuring test data solely reflects the condition of the membrane element body and its seals.
3.Calibrated vacuum gauges and high-precision timers must be used. All operations shall strictly comply with ASTM D6908 Standard Practice for Integrity Testing of Water Filtration Membrane Systems to guarantee accurate and compliant test data.
II. Pressure Decay Test (PDT) – On-Site Single Vessel Online Troubleshooting, High Field Efficiency
This method is adopted for targeted online troubleshooting of suspicious single membrane vessels after abnormal system operation, enabling rapid localization of O-ring cross-leakage and membrane sheet damage. Membrane element disassembly is not required, making it the core on-site fault targeting method. Its detection sensitivity is slightly lower than Vacuum Decay Test (VDT), and it can identify medium-to-large leakage defects.
Test Principle
Isolate the target suspicious membrane vessel independently and fully drain all internal residual water. Introduce clean medium (high-purity nitrogen is the first choice; dry oil-free compressed air may be used as a compliant alternative) from the permeate side, apply pressure slowly and stabilize it. The integrity of the membrane vessel can be judged by the magnitude of pressure decay over a fixed test duration.
Acceptance Criteria
Stably pressurize to 1.0 bar, hold pressure and stand for 10 minutes:
·Pressure drop ≤ 0.1 bar: The membrane elements and sealing assemblies inside the vessel are intact and qualified;
·Pressure drop ＞ 0.1 bar: Clear leakage points are confirmed.
Core Operating Key Points
1.Pressurization from the feed side or concentrate side is strictly prohibited! Reverse osmosis membranes feature an asymmetric structure. Pressurization on the feed side will extrude the membrane sheet backward, which easily causes irreversible damage to the desalination layer, membrane sheet collapse, and permanent failure of membrane elements.
2.The test gas source must pass through multi-stage precision filtration for oil removal, water removal and dust removal to prevent impurities from contaminating the membrane desalination layer and triggering secondary performance degradation, complying with the cleanliness requirements of semiconductor ultrapure water.
3.The pressurization rate shall be slow and steady. Instant high-pressure impact is forbidden to avoid stress deformation of membrane vessels, joints and sealing components, which would compromise test accuracy.
III. Single Vessel Permeate Conductivity Profiling Method – Online Non-Destructive, Rapid Fault Vessel Localization
This non-destructive rapid screening method allows system operation without shutdown. It is the preferred routine operation and preliminary fault diagnosis tool, capable of quickly locking abnormal membrane vessels and narrowing the troubleshooting scope, which satisfies the continuous operation demands of semiconductor ultrapure water systems.
Test Principle
Maintain the system at rated stable operating conditions. Use a portable high-precision conductivity meter to collect permeate water samples from each membrane vessel one by one, record permeate conductivity data of all vessels and compare against the average value. Membrane sheet damage or sealing cross-leakage will lead to high-salinity concentrate water infiltrating the permeate side, resulting in an abnormal sharp rise in permeate conductivity.
Acceptance Criteria
If the permeate conductivity of a single vessel is 50% or higher above the system average value, the vessel shall be judged to have sealing cross-leakage, sharp drop of membrane salt rejection rate or membrane sheet damage, and marked as a priority troubleshooting target.
Core Operating Key Points
1.Fully open the sampling valve before sampling and continuously drain for 3–5 minutes to completely empty stagnant water inside the sampling pipeline. This ensures the collected sample represents real-time actual permeate from the vessel and avoids data deviation caused by residual water in pipelines.
2.Synchronously record full-process operating parameters including feed pressure, feed temperature, total permeate conductivity and recovery rate. Operating condition fluctuations will affect the conductivity baseline, which shall be taken as auxiliary judgment basis.
3.Keep consistent sampling sequence and duration for all membrane vessels to prevent misjudgment induced by operating condition fluctuations.
IV. Probe Test Method – Online Non-Destructive, Precise Localization of Defective Single Membrane Elements
This method is applied after abnormal membrane vessels are identified via the conductivity profiling method to further pinpoint the defective single membrane element inside the vessel. It eliminates the need for full vessel disassembly and sampling inspection, serving as the only effective online non-destructive precision positioning technique available and greatly reducing maintenance workload.
Test Principle
A special sterile stainless steel sampling probe is smoothly inserted into the internal central permeate tube from the permeate end of the membrane vessel and advanced inward at a uniform speed. Water samples are continuously collected simultaneously and connected to a portable conductivity meter for real-time monitoring. Each membrane element inside the vessel produces permeate independently before converging into the central permeate tube. Leakage from a defective membrane element will trigger a stepwise abrupt change in conductivity of water samples at the corresponding depth, enabling accurate positioning of the faulty membrane element.
Core Operating Key Points
1.Standardized two-person operation: One operator advances the probe slowly and evenly to control the insertion speed strictly and prevent scratches on the inner wall of the central permeate tube and membrane element connections; the other monitors conductivity data in real time and records the probe insertion depth to match the position of each membrane element accurately.
2.The probe shall be thoroughly cleaned and pre-wetted with ultrapure water before use to remove all residual impurities and avoid contamination of the membrane system.
3.Maintain stable system operating conditions throughout the test. Adjustment of pressure and flow parameters is prohibited to ensure consistent and continuous test data.
V. Dye Test – Offline Destructive Test, Ultimate Verification of Membrane Sheet Damage
This is an offline destructive inspection method only applicable to disassembled suspicious defective membrane elements. It is used to ultimately confirm whether the membrane sheet has physical damage, cracking, desalination layer peeling or glue line cracking, and distinguish whether the root cause of failure lies in membrane sheet damage or sealing component failure.
Test Principle
Seal the permeate end of the membrane element with special sealing plugs. Introduce prepared low-concentration dye solutions (methylene blue, rhodamine solution) from the feed side and circulate and soak under low pressure. The dense desalination layer of intact membranes can completely block dye penetration without any staining marks. At positions with membrane breakage, desalination layer peeling or glue line cracks, the dye will penetrate the membrane layer and form visible stained spots on the support layer.
Acceptance Criteria
Clearly visible stained spots or streaks on the membrane surface and glue lines indicate definite defective points. The damage degree of the membrane element can be quantitatively evaluated by counting the number and distribution range of defects to determine whether the element is repairable or shall be scrapped directly.
Core Operating Key Points
1.The standard concentration of dye solution shall be controlled between 0.01% and 0.05%, and the test pressure shall be strictly limited to ≤1.0 bar. Strict pressure control prevents dye penetration through non-defective areas and ensures accurate test results.
2.The membrane element must be thoroughly rinsed with ultrapure water after testing. All flushing wastewater shall be collected and treated in a compliant centralized manner; direct discharge is prohibited to avoid dye contamination of the ultrapure water system and the environment.
3.This test is destructive. The structure and surface of the membrane element will be altered after testing, rendering it unavailable for reuse in the system. It is only used for qualitative fault tracing and shall not be adopted for online operational inspection.
VI. Online Particle / Turbidity Comparison Method – Continuous Auxiliary Monitoring & Fault Early Warning
This exclusive online continuous monitoring method for semiconductor ultrapure water systems only serves as a trigger basis for fault early warning and cannot independently judge membrane integrity. Accurate diagnostic methods such as conductivity profiling and probe tests are required to confirm faults.
Test Principle
Under normal operating conditions, intact reverse osmosis membranes can effectively intercept particulate matter and colloids in water, resulting in extremely low and stable particle counts and turbidity in permeate. Physical rupture or desalination layer peeling of membrane sheets allows particles and colloids to directly penetrate the membrane layer, causing an abrupt abnormal rise in permeate particle count and turbidity readings.
Application Scenarios
For daily real-time monitoring. Once abnormal fluctuations of monitoring data are detected, precise troubleshooting procedures including conductivity profiling and probe tests shall be immediately initiated to realize early detection and localization of faults.
Standardized Fault Diagnosis & Troubleshooting Workflow (Exclusive for Semiconductor Ultrapure Water)
1.Daily On-duty Monitoring: Track real-time operating data of total RO permeate conductivity, particle count and turbidity. Initiate troubleshooting if readings rise abnormally or fluctuate beyond limits.
2.Full-System Preliminary Screening & Localization: Perform single-vessel permeate conductivity profiling across the entire system to rapidly lock the faulty membrane vessel with abnormal data.
3.Precise Element Positioning: Conduct probe test on the identified abnormal vessel to pinpoint the single defective membrane element inside the vessel.
4.Ultimate Root Cause Qualification: Shut down and disassemble the faulty membrane element, then carry out vacuum decay test or dye test to finally distinguish the root failure cause: either physical damage/oxidative degradation of the membrane sheet itself, or cross-leakage of O-rings and joint seals.
5.Closed-Loop Disposal: Replace sealing parts, repair or scrap membrane elements according to root causes. Restore system operation only after retesting yields qualified results.
