Neither membrane filtration nor chemical disinfection is a complete answer to virus removal on its own. Reverse osmosis and tight ultrafiltration membranes physically block viruses because the particles are simply too large to fit through the pores, achieving very high log-removal rates. Chlorine and UV disinfection work differently: they don’t remove the virus particle at all, they inactivate it, breaking down its ability to infect a host cell. The EPA has required a “multi-barrier” combination of both approaches in U.S. drinking water treatment for decades, because each one covers gaps the other leaves open.
If you’ve ever wondered why a municipal treatment plant runs filtration and disinfection in sequence instead of picking one, this is why. And if you’re speccing equipment for a commercial or industrial site handling process water, wastewater reuse, or a private water source, understanding the difference changes how you design the system.
How Do Membranes Actually Stop a Virus?
Membranes stop viruses by size exclusion. The pore openings in the membrane are physically smaller than the virus particle, so the virus can’t pass through, regardless of its chemical structure or resistance to disinfectants.
Most waterborne viruses range from about 0.02 to 0.3 microns in diameter. Membrane filtration comes in a tiered lineup based on pore size:
– Microfiltration (MF): roughly 0.1 micron pores. Removes bacteria and protozoa like Cryptosporidium and Giardia, but many viruses are still small enough to slip through.
– Ultrafiltration (UF): roughly 0.01 micron pores. Tight enough to capture most viruses along with everything MF removes.
– Nanofiltration (NF): roughly 0.001 micron pores. Removes nearly all viruses plus a significant share of dissolved organics and multivalent salts.
– Reverse osmosis (RO): roughly 0.0001 micron pores. At this scale, viruses aren’t just filtered, they’re rejected along with dissolved salts, meaning RO is listed by the EPA among the technologies capable of removing both bacteria and viruses from water.
The math is straightforward once you line up the numbers: a virus at 0.02 microns is roughly 200 times larger than an RO membrane’s effective pore opening. It has nowhere to go.
How Does Chemical Disinfection (Chlorine and UV) Compare?
Chlorine and UV don’t remove viruses physically. They inactivate them in place, and the two methods have different strengths depending on the pathogen.
Chlorine is highly effective against most viruses and bacteria but comparatively weak against chlorine-resistant protozoa like Cryptosporidium. UV works almost in reverse: it’s very effective against Cryptosporidium and Giardia at economical doses, but requires a much higher dose to reliably inactivate viruses. The EPA’s own guidance puts a number on it: a UV dose of 186 mJ/cm² is required for 4-log (99.99%) inactivation of viruses, a considerably higher dose than what’s needed for the protozoan targets UV is typically sized around.
That mismatch is exactly why plants rarely rely on UV alone for virus control. Many systems are engineered to hit Cryptosporidium and Giardia targets with UV, then lean on chlorine, chloramine, or another chemical step to close the gap on virus inactivation. Neither disinfectant removes the dead virus particle or its fragments from the water; it stays in solution, inactivated but physically present, unless a filtration step downstream picks it up.
Which Method Gets a Higher Virus Log-Removal Rate?
Properly sized RO and tight UF membranes deliver some of the highest achievable virus log-removal rates because they remove the physical particle rather than depending on contact time, dose, or water chemistry. Chemical disinfection can reach comparable inactivation levels, but only when dose, contact time, and water quality all line up correctly, and it does nothing for particles that shield a virus from the disinfectant.
This is the practical distinction that gets lost in marketing copy: log-removal (filtration) and log-inactivation (disinfection) are not identical guarantees. A membrane failure is usually a mechanical, verifiable event, a torn fiber, a compromised seal. A disinfection failure can be quieter: a cloudy patch of water, an under-dosed contact chamber, an organism that happens to be more resistant than the dose was calibrated for. Turbidity and organic matter can shield viruses from both chlorine and UV exposure, cutting effective performance without any equipment “failing” in a way that trips an alarm.
Why Does the EPA Require Both, Not Just One?
The EPA’s multi-barrier approach to drinking water requires layered protection, source control, filtration, and disinfection, because no single barrier is reliable enough on its own across the full range of pathogens and real-world conditions a treatment plant encounters. It’s a principle shared by drinking water regulators across Canada, the UK, Germany, and several other countries, not a uniquely American standard.
The Ground Water Rule requires 4-log treatment of viruses for systems drawing from groundwater under the influence of surface water, and CDC surveillance still finds virus-linked outbreaks tied to drinking water systems, norovirus in particular has repeatedly been identified as a leading cause of waterborne outbreaks in CDC’s outbreak reporting going back decades. That track record is the reason regulators didn’t settle on “pick a method.” Filtration catches what disinfection might miss on any given day, whether that’s an under-treated pocket of water, a resistant strain, or a batch of high-turbidity source water. Disinfection catches free virus particles or fragments that could pass through an imperfect membrane. Stacked together, the combined barrier is meaningfully more reliable than either technology carrying the job alone.
Where Does Forward Osmosis Fit Into This Picture?
Forward osmosis (FO) uses the same size-exclusion principle as reverse osmosis, but it moves water across the membrane differently. RO forces water through the membrane with applied hydraulic pressure, working against the natural osmotic gradient. FO does the opposite: it uses a concentrated “draw solution” on one side of the membrane to pull water across naturally, following osmotic pressure instead of fighting it.
That difference matters for energy use and fouling. Because FO doesn’t rely on high applied pressure, it typically runs at lower energy input and tends to foul more slowly than pressure-driven membranes, which makes it attractive for difficult source waters, high-strength wastewater, or industrial streams where RO membranes would clog quickly. The catch is that FO isn’t a finished product at the end of the line, since the draw solution has to be regenerated and separated from the treated water afterward, adding process steps and cost that RO doesn’t need. In practice, RO remains the more common choice for straightforward potable and process-water applications, while FO tends to get selected for specialty cases: high-fouling wastewater, resource recovery, or situations where energy cost outweighs the added complexity of draw-solution recovery.
Both technologies share the same underlying virus-rejection mechanism. The membrane pore size, not the driving force behind it, is what determines whether a virus gets through.
How AMPAC USA Applies This to Commercial and Industrial Systems
We design and build membrane-based water treatment systems for commercial and industrial clients who can’t afford to guess on pathogen control, food and beverage processors, healthcare facilities, manufacturers, and municipalities needing dependable process or reuse water. In practice, that almost always means RO as the core barrier, sized to the source water’s turbidity and fouling potential, paired with pretreatment stages that protect the membrane and, where the application calls for it, a disinfection step layered in ahead of or behind the membrane train.
The design choice isn’t about picking RO over chlorine or UV. It’s about sequencing them correctly for the specific water source and end use, whether that’s a commercial reverse osmosis system sized for a mid-size facility, a full industrial reverse osmosis system built for high-volume process water, or a commercial reverse osmosis water purification setup focused on potable-grade output. A membrane that isn’t protected by proper pretreatment fouls fast and stops performing at its rated log-removal. A disinfection stage that isn’t dosed for the actual water chemistry on site underperforms quietly, without anyone noticing until testing catches it. Getting the combination right is most of the engineering work.
Frequently Asked Questions
Can reverse osmosis alone remove all viruses from water?
RO membranes reject viruses at a very high rate through size exclusion, and the EPA lists RO among the technologies capable of removing both bacteria and viruses. But no single treatment barrier is considered fail-safe on its own, which is why the EPA’s multi-barrier guidance still recommends pairing filtration with disinfection and proper pretreatment rather than relying on any one step in isolation.
Is UV disinfection enough to kill viruses in water?
UV can inactivate viruses, but it requires a substantially higher dose than what’s needed for protozoa like Cryptosporidium and Giardia, the EPA cites 186 mJ/cm² for 4-log virus inactivation. Many systems size UV around the protozoa target and add chlorine or another chemical step to reliably close the gap on viruses.
What’s the difference between virus “removal” and virus “inactivation”?
Removal means the virus particle is physically taken out of the water, which is what membrane filtration does. Inactivation means the virus is still present but chemically or physically disabled so it can no longer infect a cell, which is what chlorine and UV do. Both outcomes protect the end user, but they fail differently, which is a key reason regulators require layered barriers.
Does forward osmosis remove viruses as effectively as reverse osmosis?
Yes, in principle, because FO and RO membranes rely on the same pore-size-based rejection mechanism. FO uses osmotic pressure from a draw solution instead of applied hydraulic pressure, which lowers energy use and fouling risk, but it adds a draw-solution regeneration step that RO doesn’t require.
Why do commercial and industrial facilities need more than one treatment barrier?
A single barrier depends on ideal, unchanging conditions to keep performing at its rated level. Turbidity, fouling, dosing drift, and source water changes can all quietly degrade one method’s performance without triggering an obvious failure. Layering membrane filtration with disinfection, backed by correct pretreatment, gives a facility a backup barrier so one weak point doesn’t become a contamination event.
Sources: U.S. Environmental Protection Agency (Ground Water Rule, UV Disinfection Guidance Manual, drinking water treatment technology guidance); Centers for Disease Control and Prevention (Surveillance for Waterborne Disease Outbreaks Associated with Drinking Water, MMWR); Aquaporin (forward osmosis technical guidance); Safe Drinking Water Foundation; American Water Works Association multi-barrier treatment principles.
AMPAC USA engineers custom water purification systems for commercial, industrial, and emergency applications — from 500 GPD to multi-million GPD. Trusted by municipalities, military, and industry worldwide.
