M. Medina and U. Neis Institute of Wastewater Management and Water Protection, University of Technology Hamburg Harburg, FSP 1-02, D-21071 Hamburg, Germany (E-mail:
[email protected]) Abstract Algal incorporation into the biomass is important in an innovative wastewater treatment that exploits the symbiosis between bacterial activated sludge and microalgae (Chlorella vulgaris sp. Hamburg). It allows a good and easy algae separation by means of clarification. The effect of process parameters food to microorganisms ratio (F/M) and hydraulic retention time (HRT) on the process performance, evaluated by settleability, microalgae incorporation to biomass and nutrient removal, was studied. HRT hinted at a significant influence in the growth rate of algae, while F/M turned out to be important for stability when algae are incorporated into the biomass. This parameter also affects the total nitrogen removal of the treatment. Stable flocs with incorporated algae and supernatants with low free swimming algae concentrations were obtained at high HRT and low F/M values. Keywords Chlorella vulgaris; F/M (food to microorganism) ratio; hydraulic retention time; symbiotic algal bacterial wastewater treatment; symbiotic biomass
Introduction
The project “Wastewater treatment with symbiotic algae-bacteria-biomass” introduced a symbiosis between microalgae (Chlorella vulgaris sp Hamburg) and activated sludge bacteria to improve the performance of waste stabilization ponds. The results reported good removal of organic matter, nutrients and pathogens; high content of oxygen in the system with no need of aeration and effective separation of algae by sedimentation due to their incorporation into the biomass flocs (Gutzeit et al., 2005). The successful and easy algae removal constitutes one of the main advantages of the proposed treatment, since secondary contamination of the pond effluent due to algae is avoided. In this way, the floc formation turns to be a key step for the success of the proposed innovation. Although a crucial step in the activated sludge process, bioflocculation still remains a not fully understood aspect. The floc structure has been characterized as a net of exocellular polymeric substances (EPS) where bacteria, bacteria colonies, inorganic particles are enmeshed (Nielsen, 2002). Bivalent cations should play a relevant role in the net linking (Urbain et al., 1993; Sobeck and Higgins, 2002). The effect of some operation parameters on the biomass characteristics has also been explored. Low organic loads lead to more stable flocs (Chao and Keinath, 1979). Barbusinski and Koscielniak (1995) observed that an increase in the organic load caused an increase in the floc size and Liao et al. (2001) concluded that long sludge ages (or sludge retention time, SRT) yield flocs with higher stability than those from short SRTs. From the standpoint of process parameters relevant for algae, one of the first studies of symbiotic algae bacteria applications for wastewater determined that the hydraulic retention time (HRT) and the organic load were among the main parameters for symbiotic algal growth in wastewater (Oswald et al., 1953). Algae growth is favoured by short doi: 10.2166/wst.2007.351
Water Science & Technology Vol 55 No 11 pp 165–171 Q IWA Publishing 2007
Symbiotic algal bacterial wastewater treatment: effect of food to microorganism ratio and hydraulic retention time on the process performance
165
M. Medina and U. Neis
HRTs and high loading because under such conditions they are in their logarithmic growth phase. Under longer HRTs algae reach their steady growth phase, where they still multiply, although slower, and daughter cells are smaller. Gutzeit et al. (2005) obtained better results for total nitrogen removal with a symbiotic wastewater treatment when HRT was increased; on the other hand, Gutzeit (2006) reported that no stable biomass could be gotten when this parameter was too low (, 24 h). The success of the proposed treatment can be easily assessed by i) the settleability of the biomass, determined by the sludge volume index (SVI), ii) the nutrient removal efficiency and iii) the algae concentration in the biomass and supernatant after sedimentation, measured as chlorophyll a (Chl a) concentrations. The concept of this process aims towards abundance of algae in the biomass and less free swimming algae. This study was intended to evaluate the influence of the process variables food to microorganism (F/M) ratio and hydraulic retention time (HRT) on the process quality criteria. Materials and methods
Figure 1 shows the laboratory set up. 1 L of activated sludge from the sludge treatment plant of Seevetal (100,000 pp) was mixed with 4 L of Chlorella vulgaris sp. Hamburg culture in steady growth phase (Institute of Botany, University of Hamburg), in 0.03 m3 reactors which were constantly stirred, and illuminated 12 hours/day (2,000 mmol s21 m22). They were operated as sequence batch reactors (SBR) and fed with synthetic wastewater, as referred to in Gutzeit et al. (2005). Conditions of the performed experimental runs are summarized in Table 1. Samples were taken twice a week during 4 to 6 weeks after incubation, which was the time estimated by Gutzeit (2006) for the biomass to reach stability. SVI (sludge volume index) was selected as the criterion to assess the settleability of symbiotic sludge and chlorophyll a (Chl a) concentration in the supernatant and in the biomass to evaluate effectiveness of algae incorporation into biomass. Removal efficiency of total organic carbon (TOC) and total nitrogen (TN) were determined to assess treatment performance. Chlorophyll a (Chl a) was measured according to the German Standard 38412, part 16 (DIN, 1985), using extraction with ethanol (90%). Total organic carbon (TOC) was measured to filtered samples (membrane 0.45 mm) with an infrared absorption detector (Multi N/C 3000, Analytik Jena, Jena, Germany) and total nitrogen (TN) was determined with the same apparatus, using an electrochemical detector. Suspended solids (SS), volatile suspended solids (VSS) and Sludge volume index (SVI) were measured according to the Standard Methods (1985).
166
Figure 1 Schema of the laboratory set up
Table 1 Experimental arrays for evaluating the effect of food to microorganisms ratio (F/M) and hydraulic retention time (HRT) on process performance Effect of F/M (HRT 5 4 d)
Effect of HRT (F/M 5 0.17 d21)
Lower F/M ¼ 0.15 d21 Higher F/M ¼ 0.56 d21
Low HRT ¼ 2.7 d High HRT ¼ 4.0 d
Results Effect of food to microorganism ratio
Table 2 summarizes the results of the performance of the symbiotic wastewater treatment system for the variation of F/M ratios at constant hydraulic retention time (HRT) and presents the outcome of the one-way ANOVA performed with these data. The effect of F/M variations on the free swimming algae concentration (expressed as Chl a in the supernatant) was very significant, as indicated by the high F value, while concentration of algae attached to the flocs was not affected. The low F/M ratio yielded the best results, with a significantly lower content of free swimming algae and higher algae incorporation into the biomass. Settleability of symbiotic biomass, evaluated by means of sludge volume index (SVI), was not affected significantly by variation of F/M. Since all obtained values were below the threshold value of 150 ml gSS21, the performance of the system was satisfactory according to this criterion. Total organic carbon (TOC) removal efficiency was not significantly influenced by F/M ratios, while total nitrogen (TN) removal efficiency evidenced significant differences. The larger TN removal efficiency average was observed at the lower F/M (0.15 d21).
M. Medina and U. Neis
Experimental data were evaluated with analysis of variance ANOVA and Pearson’s correlation coefficients ( p ¼ 95%).
Effect of HRT
Table 3 shows the obtained results for the variation of hydraulic retention times (HRT) at constant food to microorganism (F/M) ratio and the outcome of the analysis of variance of these data. A significant effect of HRT on SVI was observed. The average SVI is higher for the system with longer HRT. Likewise, variations in HRT significantly affected the distribution of algae between supernatant and biomass. Chl a concentration in the supernatant Table 2 Obtained results of symbiotic biomass wastewater treatment performance for different F/M ratios and outcome from the one way ANOVA (HRT ¼ 4 d) F/M d21
0.15
0.56
Variance F, p < 0.05
Criteria
SVI mL g SS21 Chl a in the supernatant mg Chl a L21 Chl a in the biomass mg Chl a g SS21 TOC removal efficiency % TN removal efficiency %
91.2 ^ 35.2 (27) 0.26 ^ 0.18 (23) 10.32 ^ 4.72 (19) 85.31 ^ 12.99 (11) 65.96 ^ 11.96 (11)
Quantities are: mean values ^ SD (number of observations) (a): degrees of freedom between treatments, within treatments Marked coefficients * are significant for p ¼ 95%
117.7 ^ 39.8 (39) 1.70 ^ 0.76 (32) 8.69 ^ 2.04 (28) 78.57 ^ 14.62 (10) 40.82 ^ 11.22 (10)
F
(a)
1.56 33.56* 0.43 1.12 20.92*
1,64 1,53 1,45 1,19 1,19
167
Table 3 Obtained results of symbiotic biomass wastewater treatment performance for different HRT and outcome from the one way ANOVA (F/M ¼ 0.17 d 21) HRT, d
2.7
4.0
Variance F, p < 0.05
Criteria
M. Medina and U. Neis
SVI g SS21 Chl a in the supernatant mg Chl a L21 Chl a in the biomass mg Chl a g SS21 TOC removal efficiency % TN removal efficiency %
F
67.7 ^ 26.2 (11) 2.51 ^ 1.44 (17) 5.84 ^ 3.14 (18) 86.50 ^ 11.45 (12) 33.43 ^ 6.59 (11)
92.5 ^ 37.4 (23) 1.25 ^ 0.77 (16) 3.12 ^ 1.74 (17) 83.89 ^ 12.72 (10) 43.76 ^ 12.15 (10)
5.00* 4.49* 4.92* 1.73 1.20
(a)
1, 1, 1, 1, 1,
32 31 33 20 19
Quantities are: mean values ^ SD (number of observations) (a): degrees of freedom between, within treatments Marked coefficients * are significant for p ¼ 95%
as well as Chl a concentration in the biomass were lower for longer HRT. The other evaluated properties (efficiencies of TOC and TN removal) did not show any effect of HRT. Figure 2 summarizes the obtained results for Chl a content and TN and TOC removal efficiencies. Figure 2(a) shows the average algae distribution in the system. The most favourable conditions, according to the concept developed with this innovation (abundance of algae in the biomass and less free swimming algae), were obtained with the lower F/M ratio. Although no significant effect was observed for the algae concentration in the
168
Figure 2 Summary of resultant values. a) Chl a in the supernatant (mg Chl a ml supernatant-1) and Chl a in the sludge (mg Chl a g oSS-1) b) TOC, TN removal efficiencies
Discussion
Chao and Keinath (1979) found that long SRTs (corresponding to low F/Ms) yielded SVI values below 150 ml gSS21 and made bioflocculation more efficient. The authors attributed this to a high concentration of exocellular polymeric substances (EPS), which they assumed conferred stability on the floc structure. Liao et al. (2001) regarded differences for SVI at different SRTs as not large enough to be considered significant and remarked on the difficulty to establish a general relationship between SRT and SVI. In another paper, these authors proposed that one of the main factors determining floc stability is the physical arrangement of EPS in the floc net and not their concentration, since EPS molecules are organized in a less compact, more open matrix at short SRTs (or high F/Ms) than at high ones (Liao et al., 2002). The results of this study may agree more with the findings of Liao’s group, since the SVI did not change significantly (or significance of change was slight) by the introduced variations of F/M. Moreover, no significant variation of measured EPS concentrations in this study was observed for the introduced F/M changes (F1,41 ¼ 0.46, p . 0.05, data not shown), which supports the assumption of a more compact arrangement of EPS at low F/M and discards the consideration of EPS concentration influence on the algae distribution in the system. Results showed that algae growth in the biomass was significantly influenced by HRT changes; significantly higher Chl a concentrations in the biomass were reached for larger HRT, while no significant change was noticed for F/M variations. Nevertheless, Chl a concentrations in the biomass remained higher under F/M variations than under different HRTs. Oswald et al. (1953) found in their research of symbiosis in oxidation ponds that short HRTs were most favourable for algal growth because these conditions resulted in the logarithmic growth phase. This could explain the high Chl a concentrations found in the biomass and in the supernatant for shorter applied HRT. The higher concentrations of Chl a in the biomass in the experiments varying F/M – where HRT was kept at the higher value – might be the result of the steady growth phase of algae in the system determined by the longer HRT, since in this phase the cell size reduces. This may facilitate the incorporation of more cells into the symbiotic biomass net. On the other hand, concentration of free swimming algae showed to be significantly affected by both tested variables. It was significantly higher for shorter HRT and low for higher F/M ratios. The significant effect of F/M on this parameter – higher F/M values yielded higher concentrations of free swimming algae – points to the factors affecting the equilibrium between algae growth and incorporation into the biomass rates. As mentioned above, Chl a concentration in the biomass remained high and was not affected by F/M ratio changes, indicating stabilization. The increase of free swimming algae concentration for larger F/M value may be due to factors present in the outer layers of the stable flocs under such conditions – according to the floc model suggested by Liao et al. (2001). The model differentiates inner and outer EPS layers. Microalgae located in outer layers may have more availability of nutrients and light, which favours their growth. Furthermore, these outer layers may be less
M. Medina and U. Neis
biomass for the tests with varying F/M ranges, it can be recognized that the algae concentration in the biomass for the chosen load values is higher than that observed with different HRTs. On the other hand, free swimming algae showed more sensitivity to the applied changes of both variables. The effect on algae in the supernatant was larger for the F/M variation than for the different HRTs, according to the variance values in Tables 2 and 3. Figure 2(b) illustrates the effects on removal efficiencies. While TOC removal efficiencies were not significantly different for any of the applied variations, changes of F/M affected TN removal efficiencies more than the variation of HRT.
169
M. Medina and U. Neis Figure 3 Correlation of algae in the biomass and TN removal efficiency
hydrophobic, which could constitute a barrier for algae incorporation into the biomass (Medina and Neis, 2006). Total nitrogen (TN) removal was highly influenced by F/M variations. The wastewater treatment using symbiotic biomass was more effective at lower F/M ratio. Considering the current results, it could be suggested that TN removal improves when more algae are incorporated into the biomass and fewer swim freely. Since the free algae concentration was the other variable affected by the F/M variations, TN removal and algae concentration in the supernatant were correlated. The obtained Pearson’s coefficient (r ¼ 0.02 (21), p ¼ 95%) indicated that both variables are not related. On the contrary, the correlation between TN removal and algae in the biomass turned out to be very significant (r ¼ 0.67 (18), p ¼ 99%) and is illustrated in Figure 3. Gutzeit (2006) pointed out the relevance of algae concentration in the biomass for TN removal when it was observed that under longer light cycles total nitrogen concentration in the effluent was up to 60% lower than under short light periods, when algae were not massively incorporated into the flocs. The relevance of the algae incorporated in the biomass for this nutrient removal could be attributed to the higher activity of immobilized algae over free swimming algae, as shown by Lau et al. (1998) when comparing growth and nutrient removal by Chlorella vulgaris either swimming freely or being immobilized in alginate and carrageenan. The more efficient nitrogen removal by immobilized algae was suggested to be the result of compensation of light limitation in the gel matrix and better availability of nutrients there. Such interpretations could be applied if algae are assumed to be “immobilized” in an EPS matrix of activated biomass. Eventual occurrence of algae growth-promoting bacteria in the biomass population could also favour activity of attached algae. Conclusions
170
Algae distribution in the symbiotic algal bacterial wastewater treatment was shown to be affected by the parameters related to retention times. The hydraulic retention time tends to regulate the growth rate of algae. Lower HRT was shown to increase the amount of free swimming algae as well as their concentration in the biomass. The cell retention time, associated with the food-to-microorganism ratio F/M, seems to have a direct stabilizing action on the algae incorporated into the biomass. Low F/M ratios (high SRTs) favour this stability. High F/M ratios promote free swimming algae, probably due to the conditions in the outer layer of flocs. F/M also seemed to affect the total nitrogen removal of the system: better removal values were attained for lower F/M.
The consideration of these effects is relevant for the design of facilities using this technology. Hydraulic retention times should be long enough to ensure that algae would reach their steady growth phase, when good algae incorporation to biomass could take place. F/M ratio should be low or cell retention times values should be long enough to allow the formation of a stable floc where algae tend to remain attached. Total nitrogen removal will be favoured as well.
This research was possible thanks to a scholarship from the IPSWat Program of the German Federal Ministry of Education and Research. Special thanks to Dr. Dieter Lorch, Institute of Botany, University of Hamburg, for generous provision of Chlorella vulgaris cultures.
M. Medina and U. Neis
Acknowledgements
References Barbusinski, K. and Koscielniak, H. (1995). Influence of substrate loading intensity on floc size in activated sludge process. Wat. Res., 29, 1703– 1710. Chao, A.C. and Keinath, T.M. (1979). Influence of process loading intensity on sludge clarification and thickening characteristics. Wat. Res., 13, 1213– 1223. Deutsches Institut fu¨r Normung (DIN) (1985). Deutsche Norm 38 412, Teil 16. Bestimmung des Chlorophylla-Gehalts von Oberfla¨chenwasser Beuth Verlag GmbH, Berlin. Gutzeit, G. (2006). Entwicklung und Modellierung eines neuartigen Abwasserreinigungsverfahrens mit symbiotischer Algen-Bakterien-Biomasse Hamburger Berichte zur Siedlungswasserwirtschaft. Technische Universita¨t Hamburg-Harburg. Gutzeit, G., Lorch, D., Weber, A., Engels, M. and Neis, U. (2005). Bioflocculent algal-bacterial biomass improves low-cost wastewater treatment. Wat. Sci. Tech., 52(12), 9– 18. Lau, P.S., Tam, N.F.Y. and Wong, Y.-S. (1998). Effect of carrageenan immobilization on the physiological activities of Chlorella vulgaris. Bioresource Technology, 63, 115 – 121. Liao, B., Allen, D.G., Droppo, I.G., Leppard, G.G. and Liss, N. (2001). Surface properties of sludge and their role in bioflocculation and settleability. Wat. Res., 35, 339 – 350. Liao, B., Allen, D.G., Droppo, I.G., Leppard, G.G. and Liss, N. (2002). Interparticle interactions affecting the stability of sludge flocs. J. Colloid and Interface Science, 249, 372 – 380. Medina, M. and Neis, U. (2006). Symbiotic Algal Bacterial Wastewater Treatment: Characteristics of symbiotic biomass and their relation to the process performance. In Young Researchers 2006, Stuetz, R. and Teik-Thye, L. (eds), Water and Environmental Management Series, IWA Publishing, London, pp. 33 – 42. Nielsen, P.H. (2002). Activated sludge – The floc. In Encyclopedia of Environmental Microbiology, Bitton, G. (ed.), Wiley, New York, a.o., pp. 54– 61. Oswald, W.J., Gotaas, H.B., Ludwig, H.F. and Lynch, V. (1953). Algae symbiosis in oxidation ponds II: Growth characteristics of Chlorella pyrenoidosa cultured in sewage. Sewage and Industrial Wastes, 25, 26 –37. Sobeck, D.C. and Higgins, M.J. (2002). Examination of three theories for mechanisms of cation-induced bioflocculation. Wat. Res., 36, 527 – 538. Standard Methods for the Examination of Water and Wastewater (1985). 16th edn, American Public Health Association/American Water Works Association/Water Environment Federation, Washington DC, USA. Urbain, V., Block, J.C. and Manem, J. (1993). Bioflocculation in activated sludge. An analytic approach. Wat. Res., 27, 829 – 838.
171
