Filtration Sterilization: Types, Mechanism, and Uses
The filter that couldn't catch everything, and how that failure revealed viruses for the first time. Depth vs. membrane filters, pore sizes, and why "sterile filtered" doesn't always mean pyrogen-free.
The filter that failed, and what that failure revealed
In the 1880s, Charles Chamberland, working in Louis Pasteur's laboratory (the same lab that gave the world the pressurized-steam autoclave), developed a porcelain filter fine enough to strain bacteria out of a liquid entirely. It became the standard tool for producing bacteria-free filtrates, useful for purifying drinking water and for isolating whatever infectious agent might be responsible for a disease, on the assumption that once the bacteria were filtered out, nothing infectious should remain behind.
In 1892, Dmitri Ivanovsky was studying tobacco mosaic disease, a condition devastating tobacco crops, and ran crushed, infected plant sap through a Chamberland filter fine enough to catch every known bacterium. The filtered liquid, by every existing standard, should have been rendered harmless. Instead, when applied to healthy tobacco plants, it still caused disease. A few years later, Martinus Beijerinck repeated and extended this work and proposed something genuinely new: the infectious agent wasn't a bacterium being missed by the filter, it was something smaller than any known bacterium, something that could pass straight through a filter fine enough to stop everything else. He called it a "contagium vivum fluidum," a living, soluble contagion. It would later be recognized as the first clearly identified virus.
The very fact that made Chamberland's filter reliable for bacteria, its fixed, fine pore size, was exactly what let something even smaller slip through undetected. That same principle, a filter is only as good as its pore size relative to what you're trying to catch, is still the central idea behind filtration sterilization today, and still has real limits worth knowing about.
Figure: Filtration Sterilization
Filtration is the preferred method of sterilizing heat-sensitive liquids and gases without exposing them to denaturing heat. Unlike every other method in this cluster, filtration doesn't kill anything at all, it simply physically removes contaminating microorganisms. It is the method of choice for sterilizing antibiotic solutions, toxic chemicals, radioisotopes, vaccines, and carbohydrates, all of which are heat-sensitive.
The liquid or gas is passed through a filter, a device with pores too small for microorganisms to pass through but large enough to allow the liquid or gas itself through.
Figure: Relative size of human cells, bacteria, and viruses (Image source: Patrice D. Cani)
The selection of a filter for sterilization must account for the size range of the contaminants to be excluded. The most commonly used filter is composed of nitrocellulose and has a pore size of 0.22 μm. Bacteria range from 0.3 to 0.5 μm, while viruses range from 20 nm to 0.36 μm. A 0.22 μm filter retains all bacteria and bacterial spores, but not all viruses, exactly the gap that made Ivanovsky and Beijerinck's discovery possible.
Solutions for intravenous use are made pyrogen-free using filtration. Heat sterilization can kill the organisms in such solutions, but heat-resistant endotoxins (lipopolysaccharide from gram-negative bacterial cell walls) may still remain and cause fever. Filtration only removes the organism itself; if the organism was already dead and had released endotoxin before filtration, that endotoxin passes straight through the filter along with the liquid.
Working Mechanism of Filtration Sterilization
Filters work by physically trapping particles larger than the pore size and by retaining somewhat smaller particles through electrostatic attraction between the particles and the filter material. Besides porosity, other factors influencing filtration efficiency include:
- The electric charge of the filter
- The electric charge carried by the organisms
- The nature of the fluid being filtered
Filtration of liquids is accomplished either by pulling the solution through a cellulose acetate or cellulose nitrate membrane under vacuum (negative pressure) or by forcing the solution through the filter using positive pressure above the fluid.
Filtration of air is accomplished using high-efficiency particulate air (HEPA) filters, designed to remove organisms larger than 0.3 μm from isolation rooms, operating rooms, and biological safety cabinets.
Filter sterilization handles heat-sensitive materials, but the handling and disposal of the used filter still follows the foundational safety and waste-disposal practices outlined in microbiology laboratory safety rules. Review those practices before working with infectious filtrates.
Types of Filters Used in Sterilization
Modern sterilizing filters fall into two main structural categories, both descendants of the same basic principle Chamberland's porcelain filter established:
Depth Filters
A depth filter is a fibrous sheet or mat made from a random array of overlapping paper or borosilicate (glass) fibers. Rather than a single, uniform pore size, a depth filter traps particles throughout a tangled network of fibers, the way debris gets caught in a maze rather than stopped by a single gate.
Uses of depth filters
- Filtration sterilization of air in industrial processes
- Simple depth filters in forced-air heating and cooling systems, trapping dust, spores, and allergens
- Biosafety applications, most notably in biological safety cabinets
HEPA filters
A typical high-efficiency particulate air (HEPA) filter is a single sheet of borosilicate glass fiber treated with a water-repellent binder, a specific type of depth filter. It is pleated to increase surface area and mounted inside a rigid frame. HEPA filters range in size from a few square centimeters (vacuum cleaners) to several square meters (biological containment hoods and room air systems). Control of airborne particulates with HEPA filters allows the construction of clean rooms and isolation rooms for quarantine and specialized diagnostic or research laboratories. HEPA filters remove 0.3 μm test particles with an efficiency of at least 99.97%, including most microorganisms, from the airstream.
Membrane Filters
Membrane filters are the most common type used for liquid sterilization in the microbiology laboratory. They are composed of high-tensile-strength polymers such as cellulose acetate, cellulose nitrate, or polysulfone, prepared as circular membranes about 150 μm thick, containing millions of microscopic pores of uniform diameter, unlike a depth filter's random fiber network, a membrane filter works more like a fixed sieve. Pore size is adjusted during the polymerization process to suit the application.
Membrane filter porosity ranges from 0.1 μm to 10 μm; the most commonly used sizes are 0.22 μm and 0.45 μm. Membranes are held in special holders and are often preceded by depth filters made of glass fiber, which remove larger particles that might otherwise clog the membrane filter. The solution is pulled or forced through the filter and collected in a previously sterilized container.
Uses of membrane filters
- Sterilization of fluid materials: pharmaceuticals, ophthalmic solutions, antibiotics, and other heat-sensitive solutions
- Identification and enumeration of microorganisms
Find more about the membrane filtration technique.
Why This Matters Clinically
- Filtration is essential for the modern biologics industry. Monoclonal antibodies, vaccines, and other protein-based therapeutics are frequently too heat-, radiation-, and chemical-sensitive for any other sterilization method, making membrane filtration the only practical option for producing a sterile final product.
- A "sterile filtered" solution is not automatically pyrogen-free. As noted above, endotoxin released by bacteria that died before filtration will pass straight through even a properly functioning 0.22 μm filter.
- Standard sterilizing-grade filters do not reliably exclude everything. Certain very small viruses and Mycoplasma species, which lack a rigid cell wall and can be smaller than 0.3 μm, can pass through a standard 0.22 μm filter. This is a genuine, documented source of contamination in cell culture and biologics manufacturing, and it's the direct modern descendant of exactly the gap Ivanovsky and Beijerinck exploited over a century ago.
Advantages of Filtration Sterilization
- Less capital-intensive than most physical sterilization methods
- Suitable for heat-sensitive liquids (infusions, vaccines, hormones)
- Large volumes of liquid can be filtered reasonably quickly
Limitations of Filtration Sterilization
- Only liquids and gases can be sterilized by this process
- Filters are expensive to replace, particularly nano-filters
- Material limitations affect efficacy, including breakage of glass filters, rupture of membrane filters, and absorption of filtrate by Seitz filters
- Clogging may occur
- Does not remove endotoxin, and does not reliably exclude the smallest viruses or Mycoplasma
How to Remember
Completing the "how each method kills" picture for the whole cluster, with filtration as the exception. Moist heat denatures, dry heat oxidizes, ETO alkylates, radiation damages DNA. Filtration doesn't kill anything at all, it's pure physical removal. This is exactly why it exists as a completely separate category: for materials nothing else can touch without destroying them.
The "maze vs. sieve" analogy for depth vs. membrane filters: a depth filter is a maze, particles get lost and trapped somewhere inside a tangled network of fibers, with no single, fixed opening size. A membrane filter is a sieve, a fixed array of uniform holes; a particle either fits through or it doesn't.
Anchor for the hook: the filter that couldn't catch everything is exactly how we discovered a whole category of pathogens smaller than bacteria existed at all. The same gap in pore-size coverage that let a virus slip past Chamberland's filter in 1892 is the same reason a modern 0.22 μm filter can't be assumed to catch every virus or every Mycoplasma today.
Anchor for the pyrogen point: filtration removes the living organism, but if it was already dead and left its toxic "skeleton" (endotoxin) behind before filtration, that skeleton passes straight through the filter along with everything else.
Key exam facts in one table
| Fact | Detail |
|---|---|
| Mechanism | Physical removal (size exclusion + electrostatic retention); does not kill organisms |
| Standard sterilizing pore size | 0.22 μm |
| What it reliably retains | Bacteria (0.3–0.5 μm) and bacterial spores |
| What it may not retain | Some viruses (as small as 20 nm) and Mycoplasma species |
| Historical discovery enabled by a filtration gap | Ivanovsky (1892) and Beijerinck (1898), tobacco mosaic virus, the first identified virus |
| Depth filter | Random fiber network (paper/glass fiber); traps particles throughout its structure |
| Membrane filter | Fixed, uniform pore array (cellulose acetate/nitrate, polysulfone); works by size exclusion |
| HEPA filter efficiency | Removes 0.3 μm particles at ≥99.97% efficiency |
| Does not remove | Endotoxin/pyrogen from already-dead gram-negative bacteria |
| Best suited for | Heat-, radiation-, and chemical-sensitive liquids and gases: vaccines, biologics, antibiotic solutions |
Where Students Get Confused
- Assuming filtration kills microorganisms. It doesn't. It's the one method in this entire cluster that achieves sterility through physical removal alone, with no killing mechanism involved at all.
- Assuming "sterile filtered" means pyrogen-free. A filter removes the organism itself; it cannot remove endotoxin that was already released into solution before filtration.
- Assuming a 0.22 μm filter excludes all viruses and all possible contaminants. Some viruses and Mycoplasma species are small enough to pass through a standard sterilizing-grade filter.
- Confusing depth filters and membrane filters as the same thing. A depth filter traps particles throughout a random fiber network; a membrane filter has a fixed, uniform pore size and works purely by size exclusion.
References
- Madigan, M. T., Bender, K. S., Buckley, D. H., Sattley, W. M., & Stahl, D. A. (2018). Brock Biology of Microorganisms (15th ed.). Pearson.
- Willey, J. M., Sherwood, L. M., & Woolverton, C. J. (2016). Prescott's Microbiology (10th ed.). McGraw-Hill Education.
- Beijerinck, M. W. (1898). Über ein Contagium vivum fluidum als Ursache der Fleckenkrankheit der Tabaksblätter. Verhandelingen der Koninklijke Akademie van Wetenschappen te Amsterdam. (verify exact citation details in CMS before publishing — cited from memory)
- Ivanovsky, D. (1892). Concerning the mosaic disease of the tobacco plant.
Frequently Asked Questions
What is filtration sterilization?
Why is filtration used instead of heat for some materials?
What pore size is used for standard sterilizing filtration?
Can filtration remove all viruses?
What is the difference between a depth filter and a membrane filter?
Does filtration remove pyrogens (endotoxin)?
How efficient are HEPA filters?
How did filtration lead to the discovery of viruses?

Tankeshwar Acharya, MSc (Medical Microbiology)
Tankeshwar Acharya is an Assistant Professor in the Department of Microbiology at Patan Academy of Health Sciences (PAHS), Nepal, where he has been teaching and practicing clinical microbiology for over 14 years. He is the founder of Microbe Online, one of the leading free microbiology education resources on the web, covering bacteriology, mycology, parasitology, immunology, and clinical laboratory diagnostics written from direct experience in both the classroom and the diagnostic laboratory.