The Science

Algae are members of a group of predominantly aquatic photosynthetic organisms. Algae have many types of life cycles. They range in size from microscopic Micromonas species to giant kelps that reach 60 meters (200 feet) in length. Their photosynthetic pigments are more varied than those of plants. Their cells have features not found among plants and animals. They are oxygen producers and serve as the food base for almost all aquatic life.

(adapted from https://www.britannica.com/science/algae)

The algae of greatest concern in fresh water are the cyanobacteria, commonly called blue-green algae. Strictly speaking, cyanobacteria are bacteria rather than algae, but they photosynthesize, live in the same waters, and form the same kinds of blooms, so most people, and most public-health advisories, treat “blue-green algae” and “cyanobacteria” as the same thing. This page does the same: where it speaks of algae, it includes cyanobacteria.

The Basic Concepts

For an end use such as potable water, the reactor serves as a pre-treatment ahead of a drinking water plant, most likely as a reactor at a fixed location at the plant’s intake. There, the reactor treats all of the water going into the plant, not only a surface layer. That does not mean all of the water in the reservoir: only the part being withdrawn for drinking water use is treated, and the rest of the reservoir is left as it is. For drinking water, we do not want any cyanobacteria or algae or their toxins in the water. The species does not matter. The aim is to reduce all of them to below the limits that are considered the maximum safe amounts, and preferably to the lowest levels possible taking economic reality into account. For drinking water utilities, toxins entering an intake can force costly emergency treatment or, in severe cases, a do-not-drink advisory. Toledo, Ohio, in 2014 is the classic example. Independently of the water withdrawn for drinking, reactors carried by vessels can also be used on the reservoir itself, to mitigate a bloom at the surface for reasons of health or safety.

For uses such as health and safety we definitely want to mitigate the harmful algae and their toxins. Public health authorities usually begin by looking at the water, but visual inspection cannot tell whether a bloom contains toxins: samples must be sent to a laboratory for analysis, and some cyanobacteria can reach high cell counts without forming visible surface scums (Health Canada, 2020). We do not aim to treat the beneficial algae. Our aim is to reduce only the species of concern if at all possible. We just want to make the water safe and to protect the health of people and animals that come in contact with the water. Usually that means making blooms “go away.”

(Health Canada, Cyanobacteria and their Toxins in Recreational Water: Guideline Technical Document for Public Consultation, August 2020, https://www.canada.ca/en/health-canada/programs/consultation-cyanobacteria-toxins-recreational-water/document.html)

Algae are beneficial as well as detrimental, so we may not want to mitigate everything in the water

For aquaculture operators toxin exposure and oxygen depletion during a bloom can cause rapid, large-scale stock loss. Skin contact and accidental ingestion during swimming or boating in recreational water bodies are common exposure routes, especially for children and pets. For uses such as recreation and tourism and for aquaculture, we want to mitigate only the harmful algae and their toxins, but we specifically do not want to harm the other useful algae. In these uses, we are being asked to “thread the needle”: mitigate the specific harmful species and avoid harming the helpful species of algae.

Published research suggests that this can be done. Oxidants such as hydrogen peroxide have been found in some studies to affect cyanobacteria more strongly than green algae, so that green algae made up a larger share of the community after treatment (Yang and colleagues, Lusty and Gobler, and Papadimitriou and colleagues, summarized further down this page). A 2025 paper by Prof. Christopher J. Gobler and colleagues at Stony Brook University sums up that body of work: cyanobacteria are more sensitive to hydrogen peroxide than green algae and diatoms, which allows the cyanobacteria to be targeted selectively. Their own experiments show why the dose has to be matched to the bloom: a dose low enough to be largely harmless to other organisms reduced a light bloom of the cyanobacterium Dolichospermum, but a dense bloom needed doses high enough to harm zooplankton and other small animals. Our 2021 results, also summarized below, were early work with a benchtop reactor. Our current reactors are digitally controlled: the intensity of each treatment component (UV-C light, air bubbles, sonication and ozone) and the duration of the treatment can be adjusted, with the aim of mitigating the harmful species while sparing the others.

What are the toxins and what dangers do they pose?

Toxin class	Typically produced by	Primary concern	Accepted limits (guideline values)
Microcystins	Microcystis and related cyanobacteria	Liver toxicity; the most commonly monitored freshwater cyanotoxin	Drinking water (EPA 10-day health advisory): 0.3 µg/L for children under 6; 1.6 µg/L for older children and adults. Recreational water (EPA): 8 µg/L.
Cylindrospermopsin	Cylindrospermopsis and related genera	Liver and kidney effects; can persist after the bloom is visually gone	Drinking water (EPA 10-day health advisory): 0.7 µg/L for children under 6; 3.0 µg/L for older children and adults. Recreational water (EPA): 15 µg/L.
Saxitoxins	Certain freshwater and marine cyanobacteria and dinoflagellates	Neurotoxic; associated with paralytic shellfish poisoning	No EPA value. WHO: 3 µg/L in drinking water (acute exposure); 30 µg/L in recreational water.
Domoic acid	Pseudo-nitzschia (marine diatoms)	Neurotoxic; accumulates in shellfish and forage fish	No drinking-water value. In shellfish: 20 µg/g (20 ppm), the regulatory limit set by Health Canada in 1989 after the 1987 Prince Edward Island poisoning, and adopted by the U.S. FDA as its action level in 1993 (30 ppm in Dungeness crab viscera).
Anatoxins	Various cyanobacteria genera	Fast-acting neurotoxic effects; short environmental persistence	No EPA value. WHO (anatoxin-a, provisional reference value): 30 µg/L in drinking water (short-term exposure); 60 µg/L in recreational water.

(Sources: U.S. Environmental Protection Agency, Drinking Water Health Advisories for microcystins and for cylindrospermopsin, 2015; U.S. EPA, Recommended Human Health Recreational Ambient Water Quality Criteria or Swimming Advisories for Microcystins and Cylindrospermopsin, 2019; World Health Organization, Guidelines for Drinking-water Quality and Guidelines on Recreational Water Quality, cyanotoxin background documents, 2020; U.S. Food and Drug Administration, Fish and Fishery Products Hazards and Controls Guidance; for domoic acid, Campbell et al. (2024), “Life Course Considerations in Environmental Health: Developmental Neurotoxicity of Domoic Acid at Doses Below Acute Effect Levels in Adult Humans,” Birth Defects Research 116, e2419, https://doi.org/10.1002/bdr2.2419, and Pulido (2008), “Domoic Acid Toxicologic Pathology: A Review,” Marine Drugs 6, 180–219, https://doi.org/10.3390/md6020180. The drinking-water and recreational values are health advisories and guidelines, not enforceable federal limits; states may set their own, and Health Canada’s recreational guideline for total microcystins is 10 µg/L.)

Summing up what the science tells us

Different cyanobacterial taxa and different toxins can exhibit substantially different susceptibility to the same physical or chemical treatment.
Consequently, a treatment procedure that is effective for one cyanobacterium or one toxin may be substantially less effective for another.
Therefore, treatment selection and optimization should take into account the cyanobacterial population actually present in the water.
Consequently, characterization of what is in the water is an important prerequisite to selecting an appropriate mitigation procedure.

The evidence for each of these points is set out further down this page, under “The dive into the DEEP end of the pool.”

The Bottom Line: “One size DOES NOT fit all.”

Eget Liber can work with you to identify what is in the water you are dealing with and what is causing problems. Eget Liber can then discuss with you the design of a custom SFTS reactor for treating the situation you face.

The dive into the DEEP end of the pool . . . (where scientific types will feel right at home.)

Each section below is closed to keep the page short. Click a section’s title to open it, and click it again to close it.

Different algae species need different treatments: the evidence from 6 studies (click to open or close)

In work with Dr. Dominique Derminio and Prof. Gregory Boyer of SUNY College of Environmental Science and Forestry, we have shown that different algae species react differently to treatments. For example, Dolichospermum (formerly Anabaena) was reduced by every treatment that included UV-C, while Microcystis was reduced by the combined treatment, by sonication alone and aeration alone.

“Preliminary Treatments to Combat cHABs Using UV-C, Sonication, Ozone, and Aeration on Cyanobacterial Cultures and Lake Water,” Dominique Derminio, Jason Dean, Andrew Bishuk, and Gregory Boyer (Keuka College, Eget Liber, Inc., and SUNY-ESF) (presented November 15, 2021 at the Eastern Analytical Symposium)

Purcell, Parsons and Jefferson exposed four algae of different shapes and cell walls to ultrasound at frequencies from 20 to 1,144 kHz: two cyanobacteria, Microcystis aeruginosa (single cells) and Aphanizomenon flos-aquae (filaments); a green alga, Scenedesmus subspicatus (colonies); and a diatom, Melosira sp. (filaments). Aphanizomenon (up to 99% of cells removed) and Melosira (up to 83%) proved highly susceptible. Microcystis and Scenedesmus were only reduced by 16% and 20%, respectively, even with energy inputs as high as 133 kWh/m³. Ultrasound did, however, reduce the photosynthetic activity of Microcystis by 65%, which the authors took to mean that ultrasound may control the growth of a bloom even where it does not remove the bloom. They attributed the differences to cell shape (filamentous or not), cell-wall structure (silica or peptidoglycan), and the presence of gas vacuoles.

(Summarized from Purcell, D., Parsons, S. A. & Jefferson, B. (2013), The influence of ultrasound frequency and power, on the algal species Microcystis aeruginosa, Aphanizomenon flos-aquae, Scenedesmus subspicatus and Melosira sp., Environmental Technology 34, 2477–2490. https://doi.org/10.1080/09593330.2013.773355)

These results and the 2021 results above may appear to conflict, since the 2021 work found that sonication alone reduced Microcystis, while Purcell and colleagues classed Microcystis as not susceptible to ultrasound. The two studies used different strains of Microcystis and measured different things under different conditions. The 2021 work applied ultrasound at 28 kHz and low power, and measured the growth of the culture over five days, by cell count and chlorophyll-a. Purcell and colleagues measured the fraction of cells removed, at frequencies of 20 to 1,144 kHz, and they also found that ultrasound reduced the photosynthetic activity of Microcystis by 65%, which is itself a sign of suppressed growth. The outcome depends on the species, on the treatment conditions, and on what is measured, which is the point of this page.

Zamyadi and colleagues ozonated water from five cyanobacterial bloom events in Lake Champlain, at 2 to 5 mg/L of ozone for up to 10 minutes. The blooms contained Dolichospermum, Aphanizomenon, Microcystis and Pseudanabaena, at 197,000 to 1,282,000 cells/mL. Ozonation lysed (broke open) 41% to 80% of the cells and reduced the toxins released from the cells to below detection limits. Microcystis was the genus least affected by ozone.

(Summarized from Zamyadi A, Coral LA, Barbeau B, Dorner S, Lapolli FR, Prévost M. Fate of toxic cyanobacterial genera from natural bloom events during ozonation. Water Res. 2015 Apr 15;73:204-15. https://doi.org/10.1016/j.watres.2015.01.029. Epub 2015 Feb 4. PMID: 25682048.)

Yang and colleagues tested hydrogen peroxide (H₂O₂), a chemical algaecide, on four cyanobacteria in the laboratory and on a dense bloom in an aquaculture pond. In the laboratory, the filamentous Dolichospermum, Cylindrospermopsis and Planktothrix were about ten times more sensitive than the single-celled Microcystis: the concentration of H₂O₂ that reduced the algae by half (the EC₅₀) averaged 0.41 mg/L for the three filamentous genera, against 5.06 mg/L for Microcystis. In the pond, 1.3 mg/L or more eliminated Planktothrix, and 6.7 mg/L or more eliminated Microcystis. At 6.7 mg/L, microcystin fell and the diversity of the phytoplankton increased, with relatively small effects on zooplankton; 20 mg/L harmed zooplankton the most. Green algae (chlorophytes) were favored over the toxic cyanobacteria.

(Summarized from Yang Z, Buley RP, Fernandez-Figueroa EG, Barros MUG, Rajendran S, Wilson AE. Hydrogen peroxide treatment promotes chlorophytes over toxic cyanobacteria in a hyper-eutrophic aquaculture pond. Environ Pollut. 2018 Sep;240:590-598. https://doi.org/10.1016/j.envpol.2018.05.012. Epub 2018 May 12. PMID: 29763862.)

Mark W. Lusty and Prof. Christopher J. Gobler, of Stony Brook University’s School of Marine and Atmospheric Sciences in Southampton, New York, performed experiments using water from four diverse, eutrophic lakes in New York to study the effects of hydrogen peroxide (H₂O₂) on cyanobacteria and non-target members of the microbial community. H₂O₂ was administered at 4 mg/L. A combination of fluorometry, microscopy, flow cytometry, and high throughput DNA sequencing were used to quantify the effects on eukaryotic and prokaryotic plankton communities. The addition of H₂O₂ resulted in a significant reduction in cyanobacteria levels in nearly all experiments (10 of 11), reducing their relative abundance from, on average, 85% to 29% of the total phytoplankton community with Planktothrix being highly sensitive, Microcystis being moderately sensitive, and Cylindrospermopsis being most resistant. Concurrently, eukaryotic algal levels increased in 75% of experiments. The bacterial phyla Actinobacteria, cyanobacteria, Planctomycetes, and Verrucomicrobia were most negatively impacted by H₂O₂, with Actinobacteria being the most sensitive. The ability of H₂O₂ to reduce, but not fully eliminate, cyanobacteria from the eutrophic water bodies suggests it may not be an ideal mitigation approach in high biomass ecosystems.

(Adapted from Lusty, M.W.; Gobler, C.J. The Efficacy of Hydrogen Peroxide in Mitigating Cyanobacterial Blooms and Altering Microbial Communities across Four Lakes in NY, USA. Toxins 2020, 12, 428. https://doi.org/10.3390/toxins12070428)

Papadimitriou and co-workers observed that hydrogen peroxide could be a promising tool against cyanobacteria and cyanotoxins. They evaluated the effect of hydrogen peroxide on phytoplankton from two hypertrophic waterbodies in Greece. The effect of hydrogen peroxide on concentration of microcystins found in the waterbodies was also studied. Treatment with 4 mg/L hydrogen peroxide was applied to water samples originated from the waterbodies and cyanobacterial composition and biomass, phycocyanin, chlorophyll-a, and intra-cellular and total microcystin concentrations were studied. Cyanobacterial biomass and phycocyanin was reduced significantly after the application of 4 mg/L hydrogen peroxide in water treatment experiments while chlorophytes and extra-cellular microcystin concentrations were increased. Raphidiopsis (Cylindrospermopsis) raciborskii was the most affected cyanobacterial species after treatment of the water of the Karla Reservoir in comparison to Aphanizomenon favaloroi, Planktolyngbya limnetica, and Chroococcus sp. Furthermore, Microcystis aeruginosa was more resistant to the treatment of Pamvotis lake water in comparison with Microcystis wesenbergii and Microcystis panniformis. Different effects of hydrogen peroxide treatment were observed among cyanobacterial genera as well as among cyanobacterial species of the same genus. Different effects could be the result of the different resistance mechanisms of each genus or species to hydrogen peroxide. Hydrogen peroxide could be used as a treatment for the mitigation of cyanobacterial blooms in a waterbody; however, the biotic and abiotic characteristics of the waterbody should be considered.

(Adapted from Papadimitriou, T.; Katsiapi, M.; Stefanidou, N.; Paxinou, A.; Poulimenakou, V.; Laspidou, C.S.; Moustaka-Gouni, M.; Kormas, K.A. Differential Effect of Hydrogen Peroxide on Toxic Cyanobacteria of Hypertrophic Mediterranean Waterbodies. Sustainability 2022, 14, 123. https://doi.org/10.3390/su14010123)

Different toxins need different treatments: the evidence from 4 studies (click to open or close)

Onstad and colleagues measured how fast ozone and hydroxyl radicals (•OH) react with three cyanotoxins: microcystin-LR (MC-LR), cylindrospermopsin (CYN) and anatoxin-a (ANTX). At pH 8, the second-order rate constants for reaction with ozone were (4.1 ± 0.1) × 10⁵ M⁻¹ s⁻¹ for MC-LR, about 3.4 × 10⁵ M⁻¹ s⁻¹ for CYN, and about 6.4 × 10⁴ M⁻¹ s⁻¹ for ANTX. The rates for CYN and ANTX depend strongly on pH: about (2.5 ± 0.1) × 10⁶ M⁻¹ s⁻¹ for CYN above pH 8, and for ANTX about (8.7 ± 2.2) × 10⁵ M⁻¹ s⁻¹ above pH 9 but only (2.8 ± 0.2) × 10⁴ M⁻¹ s⁻¹ below pH 8. Hydroxyl radicals react far faster with all three toxins; at pH 7 the rate constants were (1.1 ± 0.01) × 10¹⁰ M⁻¹ s⁻¹ for MC-LR, (5.5 ± 0.01) × 10⁹ M⁻¹ s⁻¹ for CYN, and (3.0 ± 0.02) × 10⁹ M⁻¹ s⁻¹ for ANTX. During ozonation of natural water, hydroxyl radicals contributed most to the oxidation of ANTX, less to CYN, and least to MC-LR. Ozone attacks primarily the parts of each molecule responsible for its toxic effects, which the authors concluded means that ozone selectively detoxifies these cyanotoxins.

(Summarized from Gretchen D. Onstad, Sabine Strauch, Jussi Meriluoto, Geoffrey A. Codd & Urs von Gunten (2007). “Selective Oxidation of Key Functional Groups in Cyanotoxins during Drinking Water Ozonation.” Environmental Science & Technology 41, 4397–4404. https://doi.org/10.1021/es0625327.)

Miao and colleagues ozonated two microcystins, microcystin-LR and microcystin-RR, at different ozone doses, identified the intermediate and final breakdown products by liquid chromatography–mass spectrometry (LC-MS), and tested the toxicity of the products both in vitro (inhibition of the enzyme protein phosphatase) and in vivo (acute toxicity). They concluded that ozonation is a promising technology for removing and detoxifying these cyanotoxins. The ozonation end-products showed no adverse effects in vitro or in vivo, indicating that ozonation can completely remove the toxicity of the microcystins.

(Summarized from Miao, H.-F., Qin, F., Tao, G.-J., Tao, W.-Y. & Ruan, W.-Q. (2010). “Detoxification and degradation of microcystin-LR and -RR by ozonation.” Chemosphere 79, 355–361. https://doi.org/10.1016/j.chemosphere.2010.02.024.)

Yan and colleagues traced how ozone breaks down cylindrospermopsin (CYN), a toxin produced by several cyanobacteria often found in lakes and reservoirs used for drinking water. Using ultra-high-performance liquid chromatography with high-resolution mass spectrometry, they found that ozone degrades CYN readily, into at least 36 transformation products. CYN itself was toxic to human liver cells (the HepG2 cell line), with a half-maximal inhibitory concentration (IC₅₀) of about 64.1 µM after 24 hours and 12.5 µM after 48 hours, but the mixture of its ozonation products showed no measurable toxicity to the cells. The authors concluded that ozone is an effective and practical way to reduce CYN in water treatment without forming overtly toxic products.

(Summarized from Yan et al. (2016). “Ozonation of Cylindrospermopsin (Cyanotoxin): Degradation Mechanisms and Cytotoxicity Assessments.” Environmental Science & Technology 50, 1437–1446. https://doi.org/10.1021/acs.est.5b04540.)

Prof. Christopher J. Gobler and colleagues at Stony Brook University’s School of Marine and Atmospheric Sciences in Southampton, New York, tested hydrogen peroxide (H₂O₂), at doses from 0 to 30 mg/L, on six strains of Dolichospermum. Different strains of this one cyanobacterium produce different toxins: saxitoxin, anatoxin-a, or microcystin. At 100,000 cells/mL, 4 mg/L of H₂O₂ significantly reduced cell numbers, photosynthesis, nitrogen fixation, and toxin levels in every strain. At ten times that density, much higher doses were needed, and some strains were unaffected even at 15 mg/L, a concentration known to harm zooplankton and invertebrates. Anatoxin-a was degraded at every dose and density. In the dense cultures, however, neither microcystin nor saxitoxin was significantly degraded after four days, even at 15 mg/L, although the number of cells fell. The authors caution that in a dense bloom, H₂O₂ may destroy the cells while releasing microcystin or saxitoxin, which then persists in the water and can enter the food web.

(Summarized from Gobler, C. J., Kramer, B. J., Lusty, M. W., Thraen, J. & McTague, S. (2025). “The ability of hydrogen peroxide (H₂O₂) to degrade saxitoxin-, microcystin-, anatoxin-, and non-toxin-producing strains of the harmful cyanobacterium, Dolichospermum.” Journal of Environmental Management 387, 125696. https://doi.org/10.1016/j.jenvman.2025.125696)

These results make, in a single organism, the point of this section: in the dense cultures, the same treatment degraded one of the three toxins (anatoxin-a) but not the other two (microcystin and saxitoxin). They also show that destroying the cells is not the same as destroying the toxins, since the cells may release their toxins into the water as they break apart. This is one reason the SFTS follows its first stage with ozone and then UV-C, which are intended to degrade toxins released from damaged cells.