1 Introduction

Algae, as aquatic organisms, grow also in wastewater due to the availability of large amounts of nutrients (mainly nitrogen and phosphorus), macro- and microelements and carbon dioxide from the process of organic matter degradation (Prajapati et al. 2013; Rawat et al. 2011). Together with other microorganisms, organic matter, and mineral matter, they form flocs which are characterised by multimodal size distribution and varied strength (Fettweis et al. 2014). Determination of the size distribution of those particles in wastewater allows the choice of a suitable method of their removal (Bawiec et al. 2017; Kuśnierz and Wiercik 2016; Huang et al. 2010). That parameter, apart from density and porosity, is one of the factors in the estimation of the ability of flocs for sedimentation or flotation (Bawiec et al. 2017; Wu and He 2012; Barbusiński and Kościelniak 1995).

Researchers are searching for the relationship between the size of flocs and the growth of microalgae (Kuśnierz and Łomotowski 2015). The growth dynamics of algae determines the conditions of their cultivation and the harvesting method. The method of separation of microalgae is the greatest challenge and generates high costs in cultivation systems (Prajapati et al. 2013). With suitable sizes and density of aggregates, algae can be separated with the methods of sedimentation, flotation, or filtration (Bawiec et al. 2018; Prajapati et al. 2013; Rawat et al. 2011). During their growth, microalgae concentrate in colonies. Thanks to the produced extracellular substances in the form of polymers, their adhesive properties increase (Kuśnierz and Łomotowski 2015; Fettweis et al. 2014; Jarvis et al. 2005). The process of algae combining to form larger aggregates can be initiated by a single cell to which other cells adhere as a result of turbulent motions in water. That process can be described using the modified Avrami equation relating to the precipitation of solids from a solution (crystallisation process) (Bawiec et al. 2018; Kuśnierz and Łomotowski 2015; Tran et al. 2013). The Avrami model, introduced in 1940, assumes sphericity of precipitated particles and is based on two processes: nucleation and growth. Nucleation is responsible for the appearance of very small particles in the solution, the so-called crystallisation nuclei, which then grow as a result of combining with other particles, forming larger aggregates after a specific period of time (t) (Bawiec et al. 2018; Kuśnierz and Łomotowski 2015; Tran et al. 2013). According to Tran et al. (2013), the “nucleation-growth” mechanism can be successfully applied for the description of the phenomenon of biological colonisation.

Laser diffraction is one of the methods of analysis of suspension particle sizes (Kelly and Kazanjian 2006). It is commonly used in many branches of industry and also in laboratory studies (Brar and Verma 2011; Zhenhua et al. 2001; Black et al. 1996), due to its simplicity, high reproducibility, reliability, broad range of measurement, and speed of analysis (Xu 2015; Wu and He 2012; Kelly and Kazanjian 2006; Zhenhua et al. 2000). In addition, the method is not destructive in relation to the analysed particles (Zhenhua et al. 2001; Zhenhua et al. 2000) and also allows real-time observation of changes in the structure and size of suspensions (Xu 2015; Brar and Verma 2011), which is important in the analysis of flocs containing microorganisms. These features of laser analysis of particle size give that method an advantage over the microscopic image analysis (Brar and Verma 2011). The microscopic methods are effective and more accurate. However, they are complicated and time-consuming methods, and also require specialist knowledge in the field of phycology (Kuśnierz and Łomotowski 2015). Particle size distribution in the analysed material can be characterised on the basis of laser light dispersion with the use of a laser granulometer (Kuśnierz and Wiercik 2016). However, due to the geometric diversity of flocs, the use of a laser granulometer for the measurement of their size does not provide results as accurate as those obtained by means of a microscope (Jennings and Parslow 1988). The operation of instruments analysing diffraction patterns is based on an optical model which assumes sphericity of the analysed particles (Brar and Verma 2011; Zhenhua et al. 2000; Black et al. 1996; Jennings and Parslow 1988). Non-spherical particles (such as the majority of flocs) can cause deviation of diffraction patterns, due to which their sizes obtained as a result of analysis of those patterns may differ from the actual sizes (Zhenhua et al. 2000; Jennings and Parslow 1988). Therefore, can the laser diffraction technique be complementary in relation to microscopic analysis?

The aim of the study was a comparison of granulometric and microscopic methods for the analysis of the dynamics of changes in the growth of algae in wastewater.

2 Materials and Methods

2.1 Microalgal Cultivation

The experimental set-up consisted of four glass tanks filled with biologically treated sewage from a hydroponic lagoon. In the central part of each of the tanks, a barrier was placed, in the form of plastic netting, filled with pine bark that had been washed, dried, and weighed. The bark served as substrate for the cultivation of algae, for the purpose of their subsequent separation. Apart from the bark, the wastewater was enriched with pure carbon dioxide and also illuminated with 5-W LED lamps emitting red and blue light which is suitable for the growth of plants and algae. The tanks were illuminated for 4 h a day (from eight p.m. to midnight). To accelerate the growth of algae, inoculum was introduced to the wastewater, originating from a laboratory culture. The inoculum was introduced in the form of suspension, in the amount of 250 ml per 11 l of wastewater in each of the tanks. Tank A1 was additionally equipped with a circulation pump to force the flow of wastewater through the barrier of the pine bark.

The cultivation of algae was conducted for 2 months, but after the first month, due to low concentration of mineral nitrogen in the wastewater, it was decided to supplement its level through the addition of ammonium nitrate (NH4NO3) and calcium nitrate (Ca(NO3)2 × 4H2O). Doses of the nitrogen salts applied were chosen so as to achieve total nitrogen concentration equal to 20.0, 40.0, and 60.0 mg dm−3 in tanks A2, A3, and A4, respectively. Tank A1 was left without any addition of nitrogen. The applied nitrogen salts ensured the supply of nitrogen in the form of ions NO3− and NH4+ which are easily assimilated by algae (Graham et al. 2009). A schematic of the experimental set-up is presented in Fig. 1.

Fig. 1
figure 1

Experimental set-up with the cultivation of algae in biologically treated sewage

Full size image

2.2 Analysis of Particle Size Distribution of the Biologically Treated Sewage

Analysis of samples of treated sewage in which algae grew was conducted with the use of a laser granulometer, “Mastersizer 2000,” Malvern Instruments Ltd. The instrument measures particle diameters within the range of 0.02–2000 μm. The samples of treated sewage used for the analysis were 1 dm3 in volume. Prior to measurements, the samples were homogenised using the “Hydro MU” wet sample dispersion unit equipped with a mechanical stirrer and a pump which pumped the stirred sample into the sample cell (Garbowski et al. 2017). During the passage of the sample through the sample cell, suspension particles are illuminated with laser light (red and blue light). As a result, light scattering by the particles, fluctuations are created that are received by detectors installed in the instrument (Bawiec et al. 2018; Brar and Verma 2011; Kelly and Kazanjian 2006).

Thanks to the software supplied by the manufacturer of the instrument, diffraction patterns were transformed by a suitable algorithm, which assumes sphericity of the analysed particles, into particle size distribution with calculation of the diameters of flocs in the sample (Garbowski et al. 2017; Kelly and Kazanjian 2006). Data obtained from the analysis of flocs in the treated sewage were presented as the volumetric share of particles with a given diameter in the analysed material. The next step in the interpretation of results of particle size distribution of the treated sewage was the calculation of the cumulative percentage from the particle size distribution and application of those data for the determination of the empirical coefficients in the Avrami equation.

2.3 Application of the Avrami Equation for the Description of Particle Size Distribution of Treated Sewage

The Avrami equation used in the analysis of results of particle size distribution of treated sewage is a modification of a general equation describing the process of precipitation of crystals from water solutions (Kuśnierz and Łomotowski 2015). The equation used in the analysis is an empirical equation (Eq. 1) and assumes the form of:

$$ V(d)=1-\mathit{\exp}\left(-k\times {d}^n\right) $$
(1)

V is the volumetric share of particles with diameter (d) in analysed material (%); d is the diameter of particles in suspension (μm); k is the constant, dependent on the conditions of the process of aggregate formation, which takes into account the rate of nucleation and growth of solid particles (−); and n is the constant related with the geometry of the products of agglomeration of particles, which assumes values in the range of 1–4 (−).

Using that equation, a curve of the volumetric content of particles with a given diameter in relation to the volume of all particles present in the analysed material is obtained. The empirical coefficients k and n were determined with the use of the model of nonlinear estimation in the software “Statistica 13.3 PL.” The calculations were performed with the method of least squares, using the Levenberg–Marquardt algorithm (Kuśnierz and Łomotowski 2015). The accuracy of fit of nonlinear regression model was estimated on the basis of the value of the coefficient of determination (R), according to the adopted scale: 0.0–0.2 (not matching), 0.2–0.4 (poor matching), 0.4–0.7 (moderate matching), 0.7–0.9 (strong matching), 0.9–1.0 (very strong matching).

2.4 Sampling and Microscopic Analysis

Microscopic analysis of the biologically treated sewage was conducted at the beginning of the experiment (raw treated sewage), during the growth of algae biomass in the tanks (after 4 weeks of the experiment), and after the end of the experiment (after 8 weeks). Samples were collected from the central (mid-depth) layer of the sewage to ensure homogeneity of the mixture (lack of scum and sludge) and conformance with the granulometric analysis which also performed for samples collected from the central layer of the treated sewage.

Cyanobacteria and algae species were observed with a Nikon Eclipse TE2000-S digital microscope equipped with a Nikon DS-Fi1 camera. Four samples of plankton material were collected from each tank so that they were representative of the whole tank. Five microscopic slides were prepared from each sample, and quantitative and qualitative analyses were conducted on them. The identification was performed live and also on material preserved. The taxonomy of cyanobacteria and algae is based on Hoek et al. (1995).

Cyanobacteria and algae species were identified based on the following specialist studies: Starmach (1972); Ettl (1978); Ettl (1983); Komàrek and Fott (1983); Wołowski and Hindak (2005), Kadłubowska (1984); Hindak (1996); Komàrek, Anagnostidis (Komárek and Anagnostidis 1999; Komárek and Anagnostidis 2005); Bąk et al. (2012); Škaloud et al. (2018). During a quantitative analysis of species in the samples, the taxa were counted on slides. The so-called visible calculation units were counted and the following counted as calculation units: individual cells and 100-lm filaments. The calculations were conducted in order, along the parallel specimen lines, through moving the field of vision by one unit.

The quantitative content of particular taxa was determined under the microscope using modified Starmach’s scale (Starmach 1955), where 1 means the individual occurrence of a given species, 2 from 1 to 20 units on standard viewing surface, 3 from 1 to 30 visible units, and 4 dominant or water bloom (over 30 units in every field).

3 Results

3.1 Particle Size Distribution of the Treated Sewage

The charts (Figs. 2, 4, 5, and 6) present changes in solid particle size in the treated sewage in which algae grew. The results were processed based on the Avrami equation relating to the process of precipitation of solid particles from water solutions. Figure 2 presents the percentage share of particles of a given diameter in treated sewage in tank A1 in the course of the experiment.

Fig. 2
figure 2

Cumulative volume frequency of particles of a given diameter in tank A1

Full size image

After the initial 2 weeks, there was a significant drop in particle diameters relative to those in the raw treated sewage (percentile d(0.9) ≈ 99 μm), which is visible in Fig. 7. At that time in tank A1, dominated particles have diameters < 8 μm, which means 12-fold decrease of particle diameter in relation to the raw treated sewage. The coefficient of determination for the applied nonlinear estimation varies in the range of R = 0.996–1.000.

After 4 weeks, an abundant growth of algae was noted in the treated sewage and thus an increase of the size of particles contained in its volume. Algae, forming colonies, increased the diameters of the “crystallisation nuclei,” and then an increase was noted in the content of particles with diameters below 225 μm. The largest particle diameters in treated sewage in tank A1 were noted after 6 weeks from the start of the experiment (percentile d(0.9) ≈ 446 μm). At that point in time, approximately 55% of particles had diameters larger than those in the raw treated sewage on the day when the experiment was established. The visible effect of appearance of larger flocs was distinct turbidity of the treated sewage. There was also an abundant increase of algal biomass (Fig. 3).

Fig. 3
figure 3

Research stand on the day of start of experiment (1), after 6 weeks (2) and on the day at the end of experiment (3) (from the left: tanks A1, A2, A3, and A4) (photo by T. Garbowski)

Full size image

On the final day of the experiment (after 8 weeks), a decrease of diameters of particles present in the treated sewage was observed. The effect of this phenomenon was increased clarity of the treated sewage. Contrary to the other three tanks where algae formed thick scum on the surface of the treated sewage, in tank A1, as a result of circulation, most of the algal biomass was retained in the pine bark filling. On the final day of the experiment, the diameters of particles remaining in the volume of the treated sewage were similar to values observed after the initial 2 weeks. This means that the pine bark barrier is effective in retaining particles with diameters above 8 μm. In tank A1, the increase of the size of flocs began later than in the other tanks and lasted longer (until week 6).

In tank A2 (Fig. 4), a different dynamic of changes in suspension particle diameters was observed than in the case of tank A1. In this tank, from the initial 2 weeks, an increase of particle diameters took place and lasted until week 4 of the experiment (percentile d(0.9) ≈ 421 μm). At that time, approximately 80% of particles were characterised by diameters larger than in the raw treated sewage. After the initial 2 weeks, a majority of flocs had diameters above 325 μm. In week 6 of the experiment, the diameters of most of the flocs did not exceed 279 μm. In the final week of the experiment, only the smallest flocs remained in the sewage, with diameters of < 36 μm, and the treated sewage was relatively clear, with a visible scum of algae on the sewage surface and sludge on the bottom of the tank. The coefficient of determination for the applied nonlinear estimation is R = 0.997–1.000.

Fig. 4
figure 4

Cumulative volume frequency of particles of a given diameter in tank A2

Full size image

The treated sewage in tank A3 (Fig. 5) was characterised by an increase in floc size from week 2 to week 4, but the increase was not as pronounced as in tank A2. The difference in the value of percentile d(0.9) on the first day of the experiment and after 4 weeks amounts to approximately 192 μm. After 6 weeks, a general decrease in floc size was observed, until week 8, when the diameter of most flocs did not exceed 37 μm. In the case of data from tank A3, the coefficient of determination for the applied nonlinear estimation is R = 0.997–1.000. With the passage of time, the treated sewage in tank A3 became clear as a result of the formation of algal biomass scum on the surface, as well as sedimentation of flocs on the bottom of the tank.

Fig. 5
figure 5

Cumulative volume frequency of particles of a given diameter in tank A3

Full size image

In tank A4 (Fig. 6), only a slight increase of floc size was observed after 2 weeks (percentile d(0.9) ≈ 179 μm), in relation to the raw treated sewage (percentile d(0.9) ≈ 99 μm), in spite of abundant growth of biomass. A high level of stability of the formed aggregates was observable, as the particle sizes were characterised by low variation (percentile d(0.9) ≈ 159–179 μm), until the final day of the experiment on which the floc diameter decreased significantly (percentile d(0.9) ≈ 25 μm) (Fig. 7). Until week 8 of the experiment, over 80% of the flocs had diameters smaller than those in the treated sewage on the day of the start of the experiment. The coefficient of determination for the data from that tank is R = 0.995–1.000. After the termination of the experiment, the treated sewage was also characterised by a high clarity, which is illustrated by the graph of particle size distribution, and also by the abundant bloom of algae which formed scum on the surface of the treated sewage and sludge on the bottom of the tank. Tank A4 was enriched with the highest dose of nitrogen in the form of ammonium nitrate and calcium nitrate. It should be noted that in tanks A2, A3, and A4, the stability of the flocs increased with increase in the concentration of nitrogen, thanks to which their diameters did not vary much between weeks 2 and 6. In those tanks, the pine bark barrier was poorer in algal biomass than in the tank with forced circulation of sewage.

Fig. 6
figure 6

Cumulative volume frequency of particles of a given diameter in tank A4

Full size image
Fig. 7
figure 7

Percentile d(0.9) for the particles present in sewage in tanks A1, A2, A3, and A4

Full size image

The obtained values of the R coefficient (0.995–1.000) are very high in all four tanks. This leads to the conclusion that the applied Avrami equation describes with high accuracy (over 99%) the process of increasing the size of flocs in wastewater. This means that the increase in the size of flocs is significantly similar to the crystallisation process.

3.2 Microscopic Analysis of Phytoplankton

As a result of the experiment, 22 taxa of cyanobacteria and algae were identified. Due to the differences in the level of nitrogen, each of the glass tanks was characterised by a specific phycophlora (Table 1).

Table 1 Quantitative contributions of cyanobacteria and algae in phytoplankton in studied tanks (A1–A4)
Full size table

Before the start of the experiment, the treated sewage was characterised by an absence of developed flora of microalgae. In the treated sewage, only the presence of algae from the genus Oedogonium sp. was noted and individuals from the species Gomphonema pseudoaugur Lange-Bertalot. After 4 weeks of the experiment, in tank A1, a poor phlora was formed by filamentous and epiphyte forms. The dominant species were Microthamnion kuetzingianum Nägeli and Tribonema aequale Pascher, with a smaller participation of Ulothrix tenerrima Kützing which were overgrown by Characium sp. Also, a sporadic presence in the phytoplankton was noted of a representative of filamentous cyanobacteria Planctolyngbya limnetica (Lemm.) Komárková-Legnerová et Cronberg. After 8 weeks of the experiment, in the sewage in tank A1, the quantitative structure of phytoplankton was distinctly modified. The dominant role was taken over by filamentous green algae Ulothrix tenerrima with Tribonema aequale, the volume frequency of which remained unchanged, whereas Microthamnion kuetzingianum considerably decreased its quantitative participation (to sporadic frequency). The appearance of new filamentous species was observed, such as Klebsormidium sp. and Mougeotia sp. Filamentous forms were still overgrown by epiphyte Characium sp. Also, the presence of a species belonging to diatoms of Fragilaria sp. was noted.

The treated sewage in tank A2 was characterised by the greatest species diversity, both in the initial phase (week 4) and in the final phase (week 8) of the experiment. The dominant role in the flora of that tank was played by filamentous forms—Microthamnion kuetzingianum, Ulothrix tenerrima, Tribonema minus, and diatom Gomphonema pseudoaugur. Filaments of the algae, as in the case of sewage in tank A1, were overgrown by Characium sp. In the final phase of the experiment, in the sewage in tank A2, the number of species increased (15 taxa). The quantitative frequency of filamentous green algae species remained unchanged, but surprisingly within the group of Xantophyceae, the domination of Tribonema aequale was noted (that did not occur at all in the initial phase of the experiment). Tribonema minus was observed sporadically. Among the filamentous forms of green algae, there appeared additionally Klebsormidium sp. Noteworthy is the appearance of a taxon of Petalomonas sp. belonging to euglenids. No occurrence of Gomphonema pseudoaugur was noted in the final phase of the experiment, while there appeared Fragilaria sp., at a negligible quantitative frequency.

In the case of sewage in tank A3, after 4 weeks of the experiment, no distinct dominants were observed, and only Microthamnion kuetzingianum and Ulothrix tenerrima occurred at somewhat greater numbers. After 8 weeks of the experiment, the number of taxa increased considerably (nearly double), as well as their quantitative frequency. The presence of Microspora quadrata Hazen was noted at increased levels, while in the initial phase of the experiment, it was not observed at all. The dominant role was assumed by filamentous green algae, among which Ulothrix tenerrima attained the greatest numbers. On the filamentous forms, a fairly high presence of epiphyte green algae was observed. There also appeared coccoid forms of green algae.

The sparse flora in tank A4 was formed by species of filamentous green algae and Tribonema aequale (dominant species) that already in the initial phase of the experiment occurred in fairly large numbers. After 8 weeks of the experiment, the quantitative frequency of filamentous green algae increased. The dominant role in the assemblages was taken over by Microthamnion kuetzingianum, and also the presence of Klebsormidium sp. was observed. There appeared additional species belonging to the group Ochrophyta.

4 Discussion

In the development of algae, five main stages are distinguished: the lag phase, the exponential phase, the declining relative growth phase, the stationary phase, and the death phase (Bawiec et al. 2018), whereas Kuśnierz and Łomotowski (2015) identify three growth phases of algae, relating to the crystallisation of mineral particles, and namely the nucleation phase, the growth phase, and the stationary phase. The study presented here was aimed at verifying the correlation between the run of the life cycles of algae analysed with the microscope method and the change in the particle size distribution of treated sewage determined with the use of laser diffraction. Identified correlations of this type could largely contribute to a more efficient observation of the dynamics of growth of algae, both in natural conditions and in cultivation conditions.

After 2 weeks from the start of the experiment, a decrease of floc size was observed in tank A1, probably caused by the forced circulation of the sewage through the pine bark filling in that tank. As a result of that circulation, the largest flocs, originally contained in the treated sewage, were retained on the surface of the bark filling. This caused a reduction of the presence of crystallisation nuclei in the volume of the sewage, which in consequence delayed the process of agglomeration of particles in relation to the other tanks. From the second week, the exponential phase was observable, during which larger aggregates were formed as a result of biomass flocculation with mineral particles (Fettweis et al. 2014). The rate of growth of algae decreased after week 4 (declining relative growth phase). In the course of growth, algae formed aggregates which were retained on the pine bark, which prevented the formation of scum on sewage surface and sludge on the bottom of the tank.

The difference in the dynamics of the growth of algae between tank A1 and the other three tanks is an extension of the aggregate growth phase, lasting until week 6 (in tanks A2, A3, and A4, that process lasted until week 4). This may be related to the difference in nitrogen concentration in the particular glass tanks. Algae growing in an environment where the concentration of nutrients is the factor limiting their growth excrete larger amounts of extracellular polymeric substances (Bawiec et al. 2018). Those substances increase the resistance of the formed flocs to shear stress generated as a result of liquid flow (Fettweis et al. 2014; Tran et al. 2013; Jarvis et al. 2005). Shear stress causes the breakup of flocs at their weakest points until they become strong enough to resist the stress acting on them (Jarvis et al. 2005). Thanks to the secretion of the extracellular polymers, aggregates in tank A1 could grow longer and attain larger sizes than those in the other tanks, in spite of the hydrodynamic stress. After reaching the stationary phase, flocs were retained on the pine bark filling, which resulted in a notable decrease of diameter of particles remaining in the volume of the treated sewage. This indicates that the size and shape of flocs has a significant effect on their properties and determines the method of their removal (Zhenhua et al. 2001; Kuśnierz and Wiercik 2016; Huang et al. 2010).

Microscopic analysis conducted after 4 and 8 weeks of the experiment revealed that in all glass tanks in the study, the flocs were built primarily of entangled filamentous species such as Microthamnion kuetzingianum, Ulothrix tenerrima, Microspora qudrata, or Tribonema aequale. Those are species with a very broad ecological spectrum (Hancock 1973; Messyasz et al. 2015) and they largely contributed to the change in the diameters of the flocs. After the termination of the experiment, those species increased their volume frequency, and in individual glass tanks, there also took place certain changes in the species composition. Filamentous species, observed in the plankton over the entire duration of the experiment, were overgrown by epiphyte species which increased their surface area and, most probably, also facilitated the formation of larger aggregates.

In the 4th week of the experiment, in tanks A2, A3, and A4, an increase was observed in the quantitative frequency of Ulothrix tenerrima, and also an increase of floc diameter was noted. In tank A1, that species appeared only in the final phase of the experiment and instantly took over the dominant role in the assemblage. With a suitable level of supply of nutrients, Ulothrix tenerrima forms abundant and long tufts settling down or adhering to the substrate (Wang et al. 2013), which caused the formation of sludge on the bottom of those tanks. The abundant growth of that species could have been caused by nitrogen supplied to tanks A2–A4. The same phenomenon was observed also for the genus Tribonema, the frequency of which increased distinctly after 8 weeks of the experiment in all the tanks. The analysis of the dynamics of growth of algae on the basis of changes in particle size distribution of sewage is in conformance with the analysis of the life cycles of algae with the use of the microscopic method.

In tanks A2, A3, and A4, one can observe a distinct effect of the processes of sedimentation and flotation of flocs formed during the growth of algal biomass in the clarification of the sewage. As a result of the presence of primary crystallisation nuclei, the process of nucleation was shorter, and the aggregate growth phase was visible already in the second week of the experiment. The stationary phase was also reached sooner (week 4). In those tanks, due to the additional source of nitrogen, the growth of algae was more intensive, and due to the absence of shear stress and high concentration of biomass, flocs could form spatially developed structures. Flocs with larger sizes and a high degree of branching have more adherence points, thanks to which gas bubbles adhere to them more easily, bringing them up to the surface (Fettweis et al. 2014). The flocculation of a part of the biomass with mineral particles causes an increase of density and weight of flocs, due to which they undergo sedimentation (Yuheng et al. 2011).

In the course of the experiment, between weeks 4 and 8, growth of algae from the genus Klebsormidium sp. was observed. Even in tanks in which at the start of the experiment no presence of algae from that genus was noted, in the final phase, they were present. This may suggest that at the start of the experiment, spore forms were present in the tanks, and with the addition of nutrients, those species started to grow intensively. The genus Klebsormidium is one of the most widespread species of filamentous green algae in the world, occurring in a broad range of freshwater habitats. If it is present in phytoplankton, it is always characterised by a large quantitative domination (Rindi et al. Rindi and Guiry 2008). In the tanks, it was present in the form of dense silky mats formed by many tangled filaments.

The increase of the mass of algae, the rate of their growth, and the manner and rate of forming large assemblages depend largely in the individual preferences in nitrogen requirements of individual species. Nitrogen introduced into the tanks accelerated the growth of algae which began to form colonies with ever larger diameters, dispersed in the volume of the sewage, and over time forming scum on the sewage surface and sludge on tank bottom, thus clarifying the sewage (Fettweis et al. 2014). This resulted in, faster than in tank A1, decrease of the diameter of particles remaining in the volume of the sewage (already in week 6) until the final day of the experiment. The decrease of variation in floc diameter with increase of nitrogen concentration resulted from the increasing density of biomass and increased number of intermolecular bonds increasing the strength of the aggregates formed (Jarvis et al. 2005). Numerous studies indicate that the dynamics of consumption of nutrients and their concentration affect changes of the structure of biocoenosis in sewage (Kuśnierz and Łomotowski 2015; Barbusiński and Kościelniak 1995). In the case of this experiment, changes of nitrogen concentration affected the species composition of flocs, their stability, and size in the individual tanks.

It is known from the literature that as a result of variation in the stoichiometry of algae, their requirements for nutrients and the rate of their absorption depend on the species and sometimes are highly divergent (Pedersen and Borum 1997). An important factor affecting the removal of nitrogen and phosphorus is the N/P ratio. It is an individual feature and characteristic for particular taxa (Arbib et al. 2013; Leonardos and Geider 2004; Perini and Bracken 2014).

Algae from the genus Klebsormidium occurring in the plankton of the tanks in this study have a high capacity for the removal of nitrogen and phosphorus from sewage. In the conditions of adequate access to nutrients, they form large aggregates created by the compaction of fragments of filaments into dense assemblages (Rindi and Guiry 2008). Laboratory studies demonstrated that this genus is suitable for the removal of nutrients from sewage, and the best effects are achieved with a relatively balanced N/P ratio (10/7) (Liu and Vyverman 2015). Thus, it is a genus which contributes to the purification of waters. The genus Tribonema prefers contaminated waters, and therefore it grows well in water reservoirs contaminated with sewage. Algae from that genus have the ability of forming small assemblages and very high flotation activity (Starmach 1968; Richter and Matuła 2012; Wang et al. 2013), which is very important in relation with its application in the process of removal of nutrients from the solution. Ulothrix tenerrima forms tufts floating within water, which facilitates its identification and would surely significantly facilitate its separation from plankton. Microthamnion kuetzingianum forms strongly branched filaments and occurs in groups, often on various substrates in stagnant waters (Starmach 1972). The ability of those species for filamentous organisation and forming tufts with unlimited availability of nutrients can be a positive aspect in the processes of sewage treatment. This takes place on the principle of separation of large colonies and aggregates of algae. Recently, researchers concentrate to an increasing extent on the application of filamentous forms of algae with high capacity for attaching to porous surfaces, flotation, and self-aggregation capacity (Olguin 2003; Roberts et al. 2013; Wang et al. 2013). Such a method of separation of algae may largely reduce the costs of their removal from the solution.

5 Conclusions

The study demonstrated that with increase in the size of flocs in treated sewage, there appeared algae (e.g., Klebsormidium, Tribonema, Ulothrix tenerrima, Microthamnion kuetzingianum) with spatially developed structure and the ability to form large aggregates undergoing flotation and sedimentation. Results obtained with the method of laser diffraction are in agreement with microscopic observation of flocs. Therefore, laser diffraction applied in the analysis of floc size can be a fast and inexpensive method of monitoring of the dynamics of the growth of algae in both natural and cultivation conditions.

Important parameters in the estimation of properties of flocs are not only their size and spatial structure but also their composition. Therefore, knowledge of the taxonomy of algae, of the way in which they form larger structures, the possibility and scope of nutrient uptake by individual species of algae, is necessary for the evaluation of their effect on the quality of treated sewage and also for the selection of a suitable method of separation of the formed biomass. For this reason, microscopic analysis should be, next to laser diffraction, a method for the estimation of the dynamics of changes in the growth of algae in sewage. It supplements the method of laser granulometer application and helps orient the research onto the most effective way of sewage treatment and the separation of algae.