Introduction

The nutrient pollution (caused by an overflow of nutrients, mainly nitrogen and phosphorous, from agriculture, fish-farming, sewage waste waters, etc.) has become a dominant problem in a large number of aquatic ecosystems (Conley 2009). These nutrient inputs cause negative effects upon the quality of surface waters worldwide. Eutrophication causes increases in the biomass of macrophytes, green algae, and cyanobacteria in lakes, reservoirs, swimming pools, etc. Scientists focus mainly on cyanobacteria due to their toxin production, as they represent serious health hazards to animals and humans (Best 2002; Smith 2002; Codd 2005).

The usage of algicides as the most common method in controlling noxious phytoplankton may cause toxicity towards non-target species, accumulation of metals in sediments, cell lysis, and release of intracellular toxins (de Oliveira 2004; Giacomazzi and Cochet 2004). One of the most promising control strategies is the flocculation of blooming phytoplankton species by different types of compounds: clays, aluminum sulfate, polyaluminum chloride, ferric chloride, slaked lime, and others. The main objective of these applications is to immobilize cyanobacterial cells by creating physico-chemical interactions between phytoplankton species and flocculant. It is assumed that after these types of treatments, the cell life cycle is disturbed and cyanobacteria are going to be slowly decomposed at the surface layer of sediments as a result of the flocculation process.

Cationic polyacrylamides (CPAM’s) are widely used flocculants in the water industry (Baade 1989) and have some advantages in comparison with alum, such as lower coagulant dose requirements, a smaller increase in the ionic load of treated water, and cost savings (Rout 1999; Bolto and Gregory 2007). The aim of our study was to find out the efficiency of these compounds to flocculate phytoplankton cells from the water column, and thus examine their applicability for water management.

Materials and methods

Flocculants used

Ten cationic polyacrylamide (CPAM) flocculants (each with different molecular weight and charge—see Table 1) were used in the study. Praestol® flocculants (655 BC-S, 859 BS, 658 BC-S, 854 BC-S, 858 BS) are made by Ashland (USA) and XPC flocculants by Kemira (Sweden). Table 1 shows the individual types of chemicals together with their characteristic properties. For the test, flocculants were dissolved in distilled water to reach a 500 mg/l stock solution. The stock solution was then used for the test within 24 h.

Table 1 Basic properties of flocculants used
Full size table

Phytoplankton tested

The experiment was carried out with a natural phytoplankton community sampled from the Brno reservoir in which cyanobacteria (dominated by Microcystis sp.), eukaryotic algae (with dominance of Scenedesmus sp., and Coelastrum sp.), and diatoms (mainly Fragilaria sp.) dominated. The quantity of each phytoplankton group in the sample (20 ml) was determined by a FluoroProbe fluorometer (BBE Moldaenke, Kiel, Germany), which is able to quantify the groups of green algae, diatoms, and cyanobacteria (Gregor et al. 2005) as micrograms of chlorophyll a (Chl a) per liter of water.

Flocculation assay

The experiment was performed in 200-ml flasks (in three replicates) with a sample volume of 150 ml. Each flocculant was added to the flask with a natural phytoplankton assemblage to reach the final concentrations of 0.5, 1, 3, and 5 mg/l. Three natural water samples for testing were obtained from the Splaviska pond (Brno–Chrlice, Czech Republic) (pH: 7.9; conductivity: 783 μS/cm−1). Tests were performed at a room temperature of 24°C. Samples with no flocculant addition were used as control. After the addition of the flocculant, the suspension was mixed in a magnetic stirrer (mixing rate was found to be 600 RPM) to homogenize the phytoplankton assemblage with the treatment material. The flocculation efficiency (%) was evaluated as the decrease of chlorophyll a concentration in the water column of the treated samples in relation to the control, measured after 1 h of exposition. The removal efficiency of flocculants was determined by EC50 values. The values of EC50 (the concentration required to cause 50% inhibition of growth) were determined using linear regression analysis of inhibition percentage versus exact concentration of the flocculant.

Photosynthetic activity assay

Subsamples of 20 ml for photosynthetic activity measurement were taken immediately after Chl a quantity determination (see Phytoplankton tested chapter) from each testing flask. Samples were taken and tested after the mixing of the water column to ensure the uniform evaluation of the flask’s contents. Photosynthetic activity of phytoplankton communities in floccules was measured with a FluorCam charge-coupled device (CCD) imaging fluorometer (PSI, Brno, Czech Republic). The CCD camera measures sequences of chlorophyll fluorescence images, and is able to detect the changes of photosynthetic activity of autotrophs. The slow complementary area (SCA) method (Gavel and Marsalek 2004), which characterizes the slow phase of Kautsky kinetics, was applied. This method also identifies other effects within the photosynthetic processes, such as those on membranes, proton pump, and ATP synthesis, and reports complex information on damage done to autotrophic organisms (Gavel and Marsalek 2004).

Microscopic observations

Before and after the flocculation assays, the cyanobacterial colonies were observed by light microscope (Olympus BX 60, Japan). The reason for microscopic observations was to detect potential cell damage after the treatment by flocculants, and to examine the bounding of flocculants on phytoplankton assemblage (see Fig. 3).

Data analysis

One-way-ANOVA followed by a Tukey HSD post hoc test was conducted to analyze differences in removal efficiency related to different concentrations of each flocculant (4 variables) and differences in photosynthetic activity (1 variable). The differences among flocculants were made by statistical analysis (Tukey post hoc test) of EC50 values for each polyacrylamide. Statistical analyses were performed using statistical software Statistica 8 (StatSoft Inc., Tulsa, OK, USA), and P < 0.05 was considered to be significant.

Results

The results of the phytoplankton removal (measured as a decrease of green algae, diatom, and cyanobacteria content from the water column) as well as the inhibition of photosynthetic activity of treated phytoplankton assemblage are summarized in Figs. 1 and 2. Tested cationic polyacrylamides showed different efficiencies in the flocculation of phytoplankton community. All flocculants significantly reduced phytoplankton biomass at least at the highest doses tested (5 mg/l), while only half of them significantly reduced photosynthetic activity. Moreover, the most of the polyacrylamides decreased biomass at the doses of 3 mg/l. Based on the EC50 values, the most effective flocculants from the group of Praestol® acrylamides were 858 BS and 859 BS (see Table 2). The removal efficiency of these two flocculants was the same (Tukey HSD post hoc test, P < 0.05) for each particular phytoplankton group. On the other hand, XPC 6 and XPC 8 were found to be the most effective from XPC group. After statistical analysis (Tukey HSD post hoc test, P < 0.05), we have found that 858 BS, 859 BS, XPC6, and XPC 8 showed the same removal efficiency. All these polyacrylamides showed a strong removal ability (>80%) even at a concentration of 3 mg/l. On the other hand, at the lowest concentrations (0.5 and 1 mg/l), most of the flocculants were inactive and their flocculation effectiveness was weak (<25%).

Fig. 1
figure 1

Efficiency of XPC flocculants—changes in chlorophyll a concentration in the water column and photosynthetic activity (measured by SCA parameter) of treated phytoplankton species after 1 h of exposure. Bars represent the standard deviations. Symbols indicate for each groups of phytoplankton (cyanobacteria—crosses, green algae—full circles, diatoms—squares, and total amount of chlorophyll a—empty circles), the flocculant concentrations significantly different from the corresponding control. An asterisk (*) indicates significantly lower photosynthetic activity in comparison with the control variant

Full size image
Fig. 2
figure 2

Efficiency of Praestol flocculants—changes in chlorophyll a concentration in water the column and photosynthetic activity (measured by SCA parameter) of treated phytoplankton species after 1 h of exposure. Bars represent the standard deviations. Symbols indicate for each groups of phytoplankton (cyanobacteria—crosses, green algae—full circles, diatoms—squares, and total amount of chlorophyll a—empty circles), the flocculant concentrations significantly different from the corresponding control. An asterisk (*) indicates significantly lower photosynthetic activity in comparison with the control variant

Full size image
Table 2 Comparison of EC50 (mg/l) values and standard deviations for all tested flocculants and cyanobacteria, green algae, and diatoms
Full size table

Column graphs in Figs. 1 and 2 show the differences of photosynthetic activity in different concentrations of flocculants. Generally, it can be assumed that the decrease in photosynthetic activity was the strongest mainly in the concentration of 5 mg/l, while the photosynthesis in lower concentration of CPAMs (3 mg/l) was affected only in XPC 4 polyacrylamide. Moreover, microscope photographs (Fig. 3) indicated that cell walls were not damaged during and after the treatment.

Fig. 3
figure 3

Photograph from light microscope—a Unflocculated Microcystis sp. colony; b Flocculate of phytoplankton with typical “shell” of bound polymer (XPC 6 in the concentration of 3 mg/l) on the surface of cyanobacterial colony and cell magnification (c)

Full size image

Discussion

Organic polyelectrolytes are synthetic or natural compounds widely used especially for the coagulation and flocculation of activated sludge from wastewater treatment plant (Nguyen 2008; Saveyn et al. 2008). These biodegradable chemicals (Cameron 2000; Nakamiya and Kinoshita 1995) usually replace the application of metal salts (mainly aluminum or iron salts). Because the application of alum substances (polyaluminum chloride, aluminum sulfate) is one of the most used techniques for phytoplankton removal, the use of polyelectrolytes could be an interesting alternative. The main advantages are mainly cost savings and biodegradability. To our knowledge, utilization for algae coagulation was investigated in only a few studies in the 1980s (Bilanovic et al. 1988; Haarhoff and Cleasby 1989), but not with the new cationic polyacrylamide compounds with higher charge density like we used in this study.

We suppose that eventual application of polyacrylamides would be done in the form of an aqueous solution from a boat on the surface of the treated water body. The concentration of flocculant for phytoplankton removal in a specific body of water should be between 3 and 5 mg/l. On the other hand, preliminary trials with target water and phytoplankton should be done before each application to evaluate the effect on site-specific algal communities.

The results reached by our experiments show that the most effective flocculants concentrations are between 3 and 5 mg/l, when up to 80% of the phytoplankton biomass was flocculated. These findings correspond well with the results of Bilanovic (1988), where the coagulate doses were found to be from 1 to 10 mg/l.

From the CPAMs used in our study, there was >80% Chl a removal with four polymers (XPC 6, XPC 8, 858 BS, and 859 BS) with respect to tested concentrations. These polyelectrolytes fall into the groups with high and very high MW and CD. It could be explained by this statement that the most effective polymers for bridging are those with high MW (Caskey and Primus 1986). Moreover, high positive CDs would easily bind on negatively charged surfaces of phytoplankton cells (see Fig. 4). Contrary, materials with less effective flocculation power in our trials also had very high molecular weights and very strong cationic charge. From this viewpoint, the results here indicate that flocculation efficiency is not due to high MW and CD at least alone, but other factors that might be acting as well. For now, these factors are unknown. However, generally, in many cases CPAMs can create much stronger floccules than those built up by metal (Al or Fe) salts (Bolto and Gregory 2007).

Fig. 4
figure 4

Scheme of polymer chain binding on flocculated particle (modified from Bolto and Gregory 2007)

Full size image

Although laboratory testing indicates promising results, the effects under natural conditions could be different and have some limitations. Even though we used natural water as a medium in our work and natural phytoplankton species, different types of conditions in a real ecosystem could decrease aggregation effectiveness. Also the pH, conductivity, irradiance, oxygen saturation, and interactions with dissolved organic matter can all change the flocculation pattern in natural water (Aksberg and Wagberg 1989; Lafuma and Durand 1989; Sutton and Sposito 2005). A limitation could also be caused by toxicity towards non-target species. In the future, studies in real ecosystems (for example in mesocosm experiments) should be conducted, for a better understanding of their properties in natural waters and the possible negative effects on other aquatic biota. In safety CPAMs data sheets, EC50s are usually presented as mildly toxic with a toxicity >10 mg/l for fish and >35 mg/l for daphnids. This suggest their potential usage in aquatic ecosystems, but more specific toxicity tests should be done on these organisms to discover their real toxic effects for aquatic biota, including benthic organisms which could be affected by settled polyacrylamides. Thus, another work should be focused on a potential production of toxins by cyanobacterial cells after treatment, but based on our finding (microscopic analysis), we suggest that cyanobacterial cells were not injured by flocculation. Cyanobacterial cells remained intact (see Fig. 3), which prevented the excretion of cyanotoxins into the water. This suggestion is in accordance with another study focused on flocculants and coagulant as lime or alum, which did not cause serious cell damage and a release of toxins in comparison with conventional algicides (Lam 1995).

Conclusions

In this paper, the efficiency of cationic polyacrylamides to flocculate cyanobacteria from water column was investigated. The most effective polyacrylamides (858 BS, 859 BS, XPC 6, and XPC 8) showed good removal ability (>80%) in concentrations of 3 mg/l or greater. Moreover, microscopic observation and photosynthetic activity pointed out that cell walls and metabolism activities were not damaged, and thus a release of toxins from cyanobacteria could not occur. The main advantages of these compounds are mainly cost savings and biodegradability, especially in comparison with widely used salts of aluminum.