Rejection Rate Repeatability Testing: Verifying Membrane Separation Consistency

Rejection rate repeatability testing verifies that membrane filtration products deliver consistent solute retention across repeated measurement cycles, providing objective evidence of product quality, process stability, and test method reliability. This technical article explains the principles, sample types, test items, methods, standards, and applications of rejection rate repeatability testing for microfiltration, ultrafiltration, nanofiltration, and reverse osmosis membranes.

Technical Overview

Rejection rate (also called retention rate or salt rejection) is the most important performance indicator for a separation membrane. It quantifies the percentage of a specific solute, particle, or microorganism that the membrane removes from the feed stream. The rejection rate is calculated using the formula R = (1 − Cp / Cf) × 100%, where Cp is the concentration of the target species in the permeate and Cf is the concentration in the feed. A higher rejection rate indicates stronger retention capability and better separation selectivity.

A single rejection rate measurement, however, cannot confirm whether a membrane or a test method performs reliably. Rejection rate repeatability testing addresses this gap by performing multiple consecutive measurements under strictly identical conditions, meaning the same operator, the same apparatus, the same laboratory, and a short interval between replicates, as defined in ISO 5725 for precision of measurement methods. The results are evaluated statistically to determine whether the rejection performance is stable and reproducible within a defined tolerance.

Repeatability testing serves two complementary purposes. First, it is a quality control tool for membrane manufacturers, confirming that production lots meet consistent retention specifications before release. Second, it is a method validation tool that identifies instability arising from inadequate membrane preconditioning, pressure or temperature fluctuation, feed concentration drift, membrane compaction, fouling, or uncalibrated analytical instruments. Key statistical outputs include the mean rejection rate, standard deviation, relative standard deviation (RSD, also known as coefficient of variation), and the repeatability limit r in accordance with ISO 5725-2 and ISO 5725-6, with guidance for interlaboratory precision studies provided by ASTM E691.

Relevant reference frameworks include ASTM D6161 for membrane filtration terminology, ASTM D4194 for operating characteristics of reverse osmosis and nanofiltration devices, ASTM E1343 for molecular weight cut-off evaluation of flat sheet ultrafiltration membranes, ASTM F838 for bacterial retention of membrane filters, GB/T 32360 and related Chinese national standards for ultrafiltration membrane test methods, and JIS K 3821 for ultrafiltration membrane testing.

Testing Samples

Flat Sheet Membrane Coupons

Flat sheet samples of microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO) membranes are the most common test format. Coupons of a defined effective area are installed in laboratory cross-flow or stirred cells, allowing controlled evaluation of rejection rate repeatability against model solutes such as salts, neutral organics, or proteins.

Spiral Wound Elements and Membrane Modules

Complete spiral wound elements, tubular modules, and hollow fiber modules can be tested on standard element test rigs. Repeatability testing at the element level verifies not only the membrane material but also the fabrication quality, feed spacer configuration, glue line integrity, and sealing consistency across replicate runs.

Hollow Fiber Membranes

Hollow fiber bundles and mini-modules, widely used in water purification, membrane bioreactors, and hemodialysis, are tested in both outside-in and inside-out flow modes. Repeat challenge tests confirm stable cut-off behavior and detect fiber defects or potting inconsistencies.

Pleated Filter Cartridges and Membrane Discs

Sterilizing-grade filters (typically 0.22 µm) and depth filter cartridges are evaluated through repeated particle or bacterial retention challenges, supported by bubble point integrity checks between cycles to correlate retention and integrity data.

Ceramic and Polymeric Tubular Membranes

Ceramic ultrafiltration and microfiltration membranes used in harsh chemical or high-temperature environments are tested for rejection repeatability with appropriate markers, confirming stable performance after repeated thermal and hydraulic cycling.

Testing Items

A typical rejection rate repeatability test program includes the following items:

• Single-cycle rejection rate of the designated challenge solute, reported as a percentage for each replicate run.
• Mean rejection rate calculated across the full set of replicate measurements.
• Standard deviation and relative standard deviation (RSD) of the replicate rejection data.
• Repeatability limit r, derived at a 95% confidence level in accordance with ISO 5725-6.
• Pure water permeability or flux consistency measured in parallel, since flux drift often explains rejection variability.
• Salt rejection for NF and RO membranes using NaCl, MgSO4, Na2SO4, or CaCl2 feed solutions.
• Molecular weight cut-off verification for UF membranes using polyethylene glycol (PEG) or dextran marker solutions.
• Protein rejection using bovine serum albumin (BSA) or equivalent model proteins for bioprocess membranes.
• Particle or latex bead retention for MF membranes, and bacterial retention challenge per ASTM F838 where applicable.
• Pressure, temperature, feed concentration, recovery ratio, and conductivity or TOC stability checks throughout the test sequence.

Testing Methods

Test Apparatus and Conditions

Testing is performed on a calibrated cross-flow filtration rig comprising a high-pressure feed pump, temperature-controlled feed tank, pressure regulation valves, flow meters, and online or offline permeate sampling points. Typical benchmark conditions follow recognized practice: RO membranes are commonly challenged with a 2000 mg/L NaCl solution at approximately 1.55 MPa (225 psi) and 25 °C, NF membranes with 2000 mg/L MgSO4 at approximately 0.6 to 0.75 MPa, and UF membranes with PEG or dextran markers at low pressure with defined recirculation flow. All instruments, including conductivity meters, TOC analyzers, UV-Vis spectrophotometers, and pressure transducers, must be calibrated before the test series begins.

Membrane Preconditioning

Before data collection, each membrane is thoroughly rinsed to remove preservatives and then compacted with deionized water at the test pressure until the flux stabilizes. Preconditioning removes transient compaction effects, which are a leading cause of poor repeatability in the first few measurement cycles.

Replicate Measurement Protocol

The rejection rate sequence is performed at least in triplicate, and preferably five to ten times, under nominally identical conditions. For each replicate, the system is allowed to reach steady state, feed and permeate samples are collected simultaneously, and the solute concentration is determined by the appropriate analytical technique:

• Conductivity measurement for single-salt feed solutions.
• Total organic carbon (TOC) analysis for organic marker solutions.
• Colorimetric or UV-Vis spectrophotometric methods for PEG, dextran, and protein assays.
• Particle counting or laser obscuration for latex bead challenges.
• Plate count or flow cytometry methods for bacterial retention tests.

Data Processing and Acceptance Criteria

For each replicate i, the rejection rate Ri is calculated as (1 − Cp,i / Cf,i) × 100%. The mean value, standard deviation, and RSD are then computed for the complete data set. Commonly applied acceptance criteria require an RSD of no more than 3% to 5% for salt rejection testing, with tighter limits for high-rejection RO membranes and looser limits for UF cut-off evaluation, unless stricter product specifications apply. If any replicate deviates beyond the repeatability limit, the test is investigated for leaks, membrane defects, instrument drift, or sampling error, and the sequence is repeated after corrective action.

Applications

Rejection rate repeatability testing is applied across the entire membrane value chain:

• Membrane manufacturing quality control and lot release testing, ensuring consistent retention performance across production batches.
• Research and development benchmarking, where new membrane chemistries are compared against reference materials under statistically validated conditions.
• Drinking water, brackish water, and seawater desalination projects, verifying that RO and NF elements meet contractual rejection guarantees.
• Biopharmaceutical processing, including tangential flow filtration, virus filtration, and sterile filtration validation for regulatory submissions.
• Food and beverage processing, such as dairy protein concentration, whey fractionation, juice clarification, and dealcoholization.
• Hemodialysis and medical device membranes, where consistent solute clearance is safety-critical.
• Ultrapure water production for the semiconductor and power industries.
• Measurement system analysis (gage R&R) studies and internal laboratory method validation, demonstrating that rejection test methods are fit for purpose.
• Failure analysis and customer dispute resolution, where repeatable retention data provide objective technical evidence.

FAQ

What is rejection rate repeatability testing?

It is a precision evaluation in which the rejection rate of a membrane is measured several times under identical conditions, and the replicate results are analyzed statistically to confirm stable, consistent retention performance. It answers the question of whether the same membrane, tested the same way by the same operator, yields the same rejection value every time.

How does repeatability differ from reproducibility?

Repeatability refers to measurement agreement when the same operator uses the same equipment in the same laboratory over a short time period. Reproducibility refers to agreement between different operators, different equipment, or different laboratories. Both concepts are defined in ISO 5725, and repeatability testing is typically performed first as the foundation for any reproducibility study.

How many replicates are required?

Most protocols specify a minimum of three valid replicates, while five to ten replicates provide a more robust statistical basis for calculating standard deviation and RSD. The exact number depends on the product specification, the applicable standard, and the customer agreement.

What level of RSD is considered acceptable?

For salt rejection on RO and NF membranes, an RSD within 3% to 5% is generally regarded as acceptable. High-rejection seawater RO membranes often demand tighter control, while UF molecular weight cut-off testing may tolerate slightly wider variation. Product-specific specifications always take precedence over general guidelines.

Which standards govern rejection rate repeatability testing?

The core precision framework is defined by ISO 5725 and ASTM E691. Membrane-specific test conditions draw on ASTM D4194 for RO and NF devices, ASTM E1343 for UF molecular weight cut-off, ASTM F838 for bacterial retention, ASTM D6161 for terminology, GB/T 32360 for ultrafiltration membrane test methods, and JIS K 3821 for ultrafiltration testing. Testing can be performed in-house or by accredited third-party laboratories following these frameworks.

What sample information should be provided before testing?

Clients should provide the membrane type and format (coupon, disc, element, or module), the target challenge solute and concentration, the desired test pressure, temperature, and recovery conditions, the applicable specification or acceptance criteria, and any preconditioning or cleaning requirements. A typical repeatability study is completed within three to ten working days depending on sample complexity and replicate count.

What are the main causes of poor rejection rate repeatability?

The most frequent causes include insufficient membrane compaction or preconditioning, temperature and pressure fluctuations, feed concentration drift during recirculation, membrane fouling between cycles, sampling timing errors, uncalibrated analytical instruments, and sealing leaks in the test cell. A systematic investigation of these factors usually restores repeatable results.