Microbiological testing of circulating water by the membrane filtration method is a cornerstone practice for controlling biofouling, microbiologically influenced corrosion (MIC), and waterborne pathogens such as Legionella in cooling towers and closed-loop systems. By concentrating microorganisms from a measured water volume onto a 0.45 µm membrane filter and culturing them on selective media, this technique delivers quantitative, standards-compliant results aligned with methods such as ISO 8199, ISO 9308-1, ASTM D5246, and GB/T 5750.12. This technical guide explains the underlying principle, sample types, target organisms, step-by-step analytical procedures, quality control requirements, and typical industrial applications of membrane filtration-based microbiological analysis for circulating water.
Technical Overview
The membrane filtration (MF) method is a culture-based microbiological enumeration technique in which a measured volume of water is drawn under vacuum through a sterile membrane filter—typically 0.45 µm pore size and 47 mm in diameter—that retains bacteria and other microorganisms on its surface. The membrane is then transferred face-up onto a sterile absorbent pad saturated with liquid culture medium, or directly onto solid agar. After incubation at a defined temperature for a defined period, each retained cell or cell cluster multiplies into a visible colony, and the colonies are counted and reported as colony-forming units (CFU) per unit volume of the original sample.
For circulating water systems—open recirculating cooling circuits, closed chilled or hot water loops, and industrial process cooling water—the MF method is particularly valuable because it can process relatively large sample volumes (typically 1 mL to 100 mL, with serial dilution as required), achieving detection limits as low as 1 CFU per 100 mL. This sensitivity is far superior to that of single-step pour plate procedures and is essential for verifying biocide dosing effectiveness, monitoring biofilm precursor organisms, and demonstrating compliance with microbiological control targets. The method is harmonized with internationally recognized procedures, including ISO 8199 (general requirements and guidance for microbiological examinations by culture), ISO 6222 (enumeration of culturable microorganisms), ISO 9308-1 (coliform bacteria and Escherichia coli by membrane filtration), ISO 16266 and ASTM D5246 (Pseudomonas aeruginosa), and, in Chinese practice, GB/T 5750.12 together with the GB/T 14643 series for industrial circulating cooling water.
The principal technical considerations are sample turbidity, which can clog the membrane and mask colonies; the physiological condition of the organisms, since sublethally injured cells stressed by biocides may require resuscitation or less selective media; and sample freshness, because microbial populations in nutrient-rich cooling water can shift significantly within hours. Experienced analysts manage these issues through serial dilution, replicate filtration of multiple volumes, buffered rinse steps, and rapid transport-to-laboratory protocols, all executed under a documented quality management system in accredited third-party laboratories or qualified in-house facilities.
Testing Samples
Open Recirculating Cooling Water (Cooling Tower Water)
This is the most microbially challenging matrix: warm, aerated, sunlit, and continuously enriched by airborne dust, organic matter, and process leakage. Total bacterial counts frequently range from 10³ to 10⁶ CFU/mL. Samples are collected from the tower basin or cold deck, the circulation pump suction, and points downstream of biocide injection to evaluate kill efficiency. Multiple sampling points across the loop help localize contamination sources.
Closed-Loop Circulating Water
Chilled water, hot water, and glycol-based closed loops are nominally low-biomass systems. Testing verifies loop integrity, confirms that nitrite or organic corrosion-inhibitor programs have not been degraded by nitrifying or sulfate-reducing bacteria, and detects the anaerobic populations responsible for under-deposit pitting corrosion. Samples are drawn from drain valves on return headers or expansion tanks after an adequate flushing period to remove stagnant line water.
Make-Up and Supplementary Water
The feed water entering the circuit establishes the microbiological baseline for the entire system. Municipal make-up requires neutralization of residual chlorine or chloramine with sodium thiosulfate before testing, while well or surface sources may carry iron bacteria, fungi, and high natural organic loads that seed the loop with fouling organisms.
Industrial Process Circulating Water
Petrochemical, refinery, steel, and pulp-and-paper cooling circuits often carry hydrocarbon traces, oil, fibers, or mill residues that select for hydrocarbon-degrading bacteria and filamentous fungi. These samples are typically turbid and nutrient-rich, requiring dilution and multi-volume filtration to obtain countable membranes.
Deposits and Biofilm-Related Samples
Although the MF method targets liquid samples, sludge or slime scraped from tower fill, distribution decks, and heat-exchanger tube sheets can be homogenized in sterile buffered diluent, diluted, and filtered by the same technique. These supplementary samples locate biofilm reservoirs that continually reseeding the bulk water despite satisfactory planktonic counts.
Testing Items
Membrane filtration protocols for circulating water typically quantify the following indicator and problem organisms:
- Heterotrophic plate count (HPC, total viable bacteria count) — the primary trending parameter for overall microbial biomass and biocide program performance in cooling circuits.
- Total coliforms, thermotolerant (fecal) coliforms, and Escherichia coli — indicators of external contamination ingress, make-up water quality failure, or cross-connection with sanitary systems.
- Pseudomonas aeruginosa — a biofilm pioneer organism and opportunistic pathogen highly tolerant of quaternary ammonium biocides.
- Legionella species, including Legionella pneumophila — cooling towers are recognized amplifiers of this pathogen; enumeration follows ISO 11731-based culture with concentration, selective pretreatment, and membrane filtration or spread plating onto GVPC agar.
- Fungi (yeasts and molds) — significant in timbered towers (wood rot), paper machine circuits, and organic-rich loops.
- Sulfate-reducing bacteria (SRB) — principal drivers of microbiologically influenced corrosion and hydrogen sulfide generation; enumerated by most-precise liquid dilution (MPN) techniques per GB/T 14643.2 when membrane countable growth is impractical.
- Iron bacteria, slime-forming bacteria, and nitrifying bacteria (ammonia- and nitrite-oxidizing) — associated with fouling deposits and accelerated depletion of nitrite-based corrosion inhibitors.
Interpretation benchmarks vary by system design and jurisdiction. As general industrial guidance aligned with Chinese circulating cooling water design codes such as GB 50050, total viable counts in open recirculating systems are commonly maintained below 10⁵ CFU/mL, with many operators targeting 10⁴ CFU/mL or lower; fungi are typically controlled below a few tens of CFU/mL; and Legionella action levels for cooling towers generally fall between 100 and 1,000 CFU/L depending on local health authority requirements. All results should be assessed against the site-specific water management plan and trending history rather than a single fixed threshold.
Testing Methods
The complete MF determination comprises five stages: sampling, filtration, incubation, colony counting with confirmation, and calculation with reporting. Each stage is governed by strict sterility, traceability, and quality assurance requirements.
Sampling and Transport
- Collect samples in pre-sterilized borosilicate or single-use PETG bottles; add sodium thiosulfate (typically to 18 mg/L final concentration) when an oxidizing biocide residual is present, so that bactericidal action stops at the moment of sampling.
- Flush sampling lines and valves before collection, fill bottles to at least 80% capacity leaving headspace for mixing, and label with sample identity, location, date, time, and biocide status.
- Transport samples at 2–8 °C protected from light, or as specified by the applicable standard; analyze preferably within 6 hours and never beyond 24 hours, with Legionella samples prioritized for earliest analysis per ISO 11731 guidance.
Filtration and Incubation
- Assemble a pre-sterilized funnel unit on a vacuum manifold, moisten the membrane with sterile buffered dilution water, and filter one or more measured volumes (for example 100 mL, 10 mL, and 1 mL, plus decimal dilutions for high-biomass cooling water) to ensure at least one countable membrane.
- Rinse the funnel walls with three successive 20–30 mL portions of sterile buffered water to distribute retained organisms evenly across the membrane and wash away inhibitory residues.
- Transfer each membrane to its appropriate medium and incubate: R2A agar or plate count agar for HPC (20–28 °C for 5–7 days for low-nutrient recovery, or 36 °C for 40–48 hours per specific method); chromogenic coliform agar, m-Endo, or m-Tergitol media for coliforms and E. coli (36 ± 1 °C for 21–24 hours; 44 °C for thermotolerant coliforms); CN (mPA-CN) agar for Pseudomonas aeruginosa (36 ± 1 °C for 40–48 hours); GVPC agar for Legionella (36 ± 1 °C for up to 10 days, following acid or heat pretreatment of concentrated samples); and rose bengal or malt extract agar for fungi (20–25 °C for 3–7 days).
Counting, Confirmation, and Reporting
- Count typical colonies under 5–15× magnification with cool transmitted light; membranes carrying roughly 20–200 colonies provide optimal statistical precision, while confluent growth or counts far exceeding the countable range are reported as too numerous to count (TNTC).
- Confirm presumptive colonies with follow-up tests: oxidase reaction, pigment formation, and casein hydrolysis or subculture for P. aeruginosa; β-glucuronidase and lactose fermentation for E. coli; cysteine growth requirement and latex agglutination or molecular confirmation for Legionella isolates.
- Calculate results as CFU per 100 mL = (colonies counted × dilution factor × 100) ÷ volume filtered in mL, applying the rounding and detection-limit conventions of ISO 8199; samples yielding zero colonies are reported as less than 1 CFU per the total volume examined.
Quality Assurance
- Run sterility blanks at the start and end of each filtration session, verify media growth-promotion and selectivity with characterized reference strains, document membrane lot verification, and maintain duplicate or replicate analyses on a defined proportion of samples.
- Support the entire process with chain-of-custody records, calibrated balances and incubators, and participation in interlaboratory proficiency testing consistent with an ISO/IEC 17025 quality system, whether performed in-house or by accredited third-party laboratories.
Applications
HVAC and Data Center Cooling Systems
Weekly-to-monthly MF-based HPC trending validates oxidizing and non-oxidizing biocide programs in commercial cooling towers, while periodic Legionella enumeration supports water management plans, tower cleaning verification, and due-diligence documentation for building owners and operators.
Power Generation and Petrochemical Plants
Condenser and service cooling circuits rely on MF monitoring to protect heat-exchanger efficiency, control sessile and planktonic populations, and target SRB suppression programs that prevent under-deposit pitting and sulfide fouling of critical alloy surfaces.
Steel, Metallurgy, and Pulp and Paper
Direct-contact and spray cooling loops, rolling-mill emulsion circuits, and paper machine white-water systems use membrane filtration counts to schedule slime-control chemical dosing and to verify cleanliness after boil-outs and mechanical cleaning campaigns.
Food, Beverage, and Pharmaceutical Facilities
Process cooling water, CIP rinse loops, and purified water distribution systems benefit from the low detection limits of MF with R2A medium, where slow-growing oligotrophic bacteria at levels below the reach of rapid indicator methods can still compromise product quality and hygiene zoning.
Compliance, Audits, and Incident Response
Standards-aligned MF results generated by accredited third-party laboratories are routinely used to satisfy regulatory inspections, customer audits, contractual key performance indicators, and forensic investigation following suspected contamination or Legionella events, providing legally defensible quantitative data with full method traceability.
FAQ
Why is membrane filtration preferred over pour plate for circulating water microbiology?
Membrane filtration concentrates organisms from volumes far larger than the 1 mL typically embedded in a pour plate, lowering the detection limit to approximately 1 CFU per 100 mL. It also removes dissolved treatment chemicals and residual biocide through rinse steps, improving recovery of stressed organisms, and it allows the analyst to select target-specific media for each organism group on a single filtered sample series.
What sample volume should be filtered?
The volume is chosen to yield roughly 20–200 colonies on the membrane. Clean make-up water may permit 100 mL or more, while heavily loaded open cooling water with counts above 10⁵ CFU/mL often requires 1 mL, 0.1 mL, or decimal dilutions. Filtering a series of volumes (for example 100 mL, 10 mL, and 1 mL) in parallel guarantees a countable result.
How quickly must samples be analyzed after collection?
Preferably within 6 hours and no later than 24 hours, with refrigerated transport at 2–8 °C. Because circulating water is nutrient-rich, die-off or regrowth during storage distorts counts, so rapid delivery to the laboratory and prompt filtration are essential for representative results.
What results indicate a microbiological control problem?
Rising HPC trends exceeding 10⁴–10⁵ CFU/mL in open circuits, detection of coliforms or Pseudomonas aeruginosa, SRB positivity in closed loops, or Legionella above the site action level (commonly 100–1,000 CFU/L) all warrant investigation of biocide residual and dosing control, tower fill cleanliness, stagnant legs, and biofilm reservoirs.
How are turbid or biocide-laden samples handled?
High turbidity is managed by serial dilution and multi-volume filtration; if membranes clog before an adequate volume passes, spread plate or MPN methods provide an alternative. Oxidizing biocide residuals are neutralized at sampling with sodium thiosulfate, and non-oxidizing residues are addressed through buffered rinses and, where appropriate, neutralizing broth validated for the specific chemistry in use.
How often should a circulating system be tested?
Open recirculating systems are typically monitored for HPC weekly during warm seasons and at least monthly otherwise, with Legionella testing quarterly plus after cleaning events, treatment failures, or construction. Closed loops are generally tested quarterly to semi-annually unless trends dictate otherwise, and all intervals should be tightened following any change in chemistry or operation.
Does a zero colony result guarantee the water is free of microorganisms?
No. Culture-based MF detects only organisms able to grow on the chosen medium under the chosen incubation conditions. Viable-but-nonculturable cells, intracellular Legionella within amoebae, and organisms sequestered in biofilms may escape detection. Robust monitoring programs therefore complement MF results with rapid methods such as ATP luminometry, direct microscopy, or qPCR where a fuller risk picture is required.