Citric acid application for denitrification process support in biofilm reactor

Chemosphere 171 (2017) 512e519

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Citric acid application for denitrification process support in biofilm reactor Artur Mielcarek a, Joanna Rodziewicz a, *, Wojciech Janczukowicz a, Dorota Dabrowska b, Slawomir Ciesielski b, Arthur Thornton c, Joanna Struk-Sokołowska d a

University of Warmia and Mazury in Olsztyn, Department of Environment Engineering, Warszawska St. 117a, Olsztyn 10-719, Poland University of Warmia and Mazury in Olsztyn, Department of Environmental Biotechnology, Słoneczna St. 45g, Olsztyn 10-709, Poland Atkins, Woodcote Grove, Ashley Road, Epsom KT18, United Kingdom d Bialystok University of Technology, Department of Technology in Engineering and Environmental Protection, Wiejska St. 45a, Białystok 15-351, Poland b c

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


The suitability of citric acid for aiding the denitrification in AnSBBR was proved.
The kinetic parameters were connected with the microbial communities structure.
Microorganisms from the Trichococcus genus developed at first (above 40%).
Propionibacterium and Agrobacterium genera dominated in the second phase.
The biomass yield coefficient was ten times lower than in suspended biomass reactors.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 September 2016 Received in revised form 23 November 2016 Accepted 20 December 2016 Available online 22 December 2016

The study demonstrated that citric acid, as an organic carbon source, can improve denitrification in Anaerobic Sequencing Batch Biofilm Reactor (AnSBBR). The consumption rate of the organic substrate and the denitrification rate were lower during the period of the reactor's acclimatization (cycles 1e60; 71.5 mgCOD L1 h1 and 17.81 mgN L1 h1, respectively) than under the steady state conditions (cycles 61e180; 143.8 mgCOD L1 h1 and 24.38 mgN L1 h1). The biomass yield coefficient reached 1 0.04 ± 0.02 mgTSS$ mgCOD1 re (0.22 ± 0.09 mgTSS mgNre ). Observations revealed the diversified microbiological ecology of the denitrifying bacteria. Citric acid was used mainly by bacteria representing the Trichoccocus genus, which represented above 40% of the sample during the first phase of the process (cycles 1e60). In the second phase (cycles 61e180) the microorganisms the genera that consumed the acetate and formate, as the result of citric acid decomposition were Propionibacterium (5.74%), Agrobacterium (5.23%), Flavobacterium (1.32%), Sphaerotilus (1.35%), Erysipelothrix (1.08%). © 2016 Elsevier Ltd. All rights reserved.

Handling Editor: A Adalberto Noyola Keywords: Citric acid Biofilm reactor External carbon source Denitrification Metagenomic analyses

* Corresponding author. E-mail addresses: [email protected] (A. Mielcarek), [email protected] (J. Rodziewicz), [email protected] (W. Janczukowicz), dorota.dabrowska@ uwm.edu.pl (D. Dabrowska), [email protected] (S. Ciesielski), [email protected] (A. Thornton), [email protected] (J. Struk-Sokołowska). http://dx.doi.org/10.1016/j.chemosphere.2016.12.099 0045-6535/© 2016 Elsevier Ltd. All rights reserved.

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

1. Introduction The availability of an organic substrate is essential in the processes of nitrogen and phosphorus removal (BNR). In technological systems for wastewater treatment it is in fact the shortage of organic compounds which often limits the effectiveness of BNR. The problem of an insufficient amount of organic compounds may be solved by the application of an external source of organic carbon (Rodziewicz et al., 2015). The source may consist of simple organic compounds, such as methanol, ethanol, as well as waste products, such as glycerol or wastewater characterized by an adequate C:(N:P) ratio (Janczukowicz et al., 2012; Mielcarek et al., 2013). However, there is still the search for new organic compounds, which may prove to be more economically competitive. Promising substrates include citric acid, the positive effect of which has already been proved in relation to enhanced biological phosphorus removal (EBPR) (Kargi et al., 2005; Mielcarek et al., 2015a). Citric acid is one of the acids most commonly used in the industry. In 2007 its production exceeded 1.7 million tonnes and is still growing (Singh Dhillon et al., 2011). 75% of global production is used in the food industry, 10% in the pharmaceutical industry, with the remaining 15% used for other purposes (in the food, plastic, steel, tanning, photographic, ceramics and many other industries, including the application as an acid cleaning agent (Soccol et al., 2006)). Despite its widespread use, the search for new areas of application for this acid is being still conducted. Citric acid is easily available as well as conveniently transported and stored. The market price of technical citric acid is close to that of acetic acid, and it is lower compared to such substrates as glucose, propanol or ethanol (Mielcarek et al., 2015a). Sequencing Batch Biofilm Reactors (SBBR), combines the advantages of conventional Sequencing Batch Reactors (SBR), providing both the resilience of the biofilm system with the flexibility in both the flow and loading of pollutants (Mielcarek et al., 2015b). Recent studies on a SBBR indicated that this solution could be used for effective denitrification (Mielcarek et al., 2016, 2015b). Simultaneously they are characterized by the lowest possible biomass yield coefficient (Yobs). This is particularly important, as the utilization costs of generated biomass could account for up to 60% of the total cost of wastewater treatment (Low and Chase, 1999). From the microbiological perspective, the organic substrate introduced into a biological reactor may exert a significant influence on the development of microbial communities and on their enzymatic activity, and there by on the kinetics and effectiveness of the wastewater treatment. On this basis the kinetics of pollutant removal have been related to the investigation of the composition of the microbial communities (Hallin et al., 2006). The aim of this study was to investigate if adding citric acid as an external source of carbon has the potential to improve denitrification. The scope of the study included the determination of the kinetics of the organic substrate used and the denitrification, both during reactor activation and under steady conditions. . Moreover, the microorganisms which developed in the biofilm as a result of the application of citric acid were characterized, and the influence of the presence of selected bacteria groups on wastewater treatment kinetics was defined. 2. Materials and methods The Anaerobic Sequencing Batch Biofilm Reactor (AnSBBR used in the studies was described by Mielcarek et al. (2015b)). The synthetic wastewater was prepared using KH2PO4 and NaNO3, MgSO4$7H2O (0.308 g L1), KCl (0.021 g L1), CaCl2 (0.021 g L1), enriched broth (0.02 g L1) and tap water (pH 6.99; N-NO3

513

0.50 mg L1; N-NH4 0.10 mg L1; P-PO4 0.20 mg L1; K 6.50 mg L1; Ca 88.74 mg L1; Mg 16.26 mg L1; Na 1.42 mg L1; Cl 10.20 mg L1; S 0.30 mg L1). The parameters of the synthetic wastewater used in the studies were typical for nitrified municipal wastewater (pH ¼ 7.59; COD ¼ 20.0 mg L1; total nitrogen ¼ 80.00 mgN L1; nitrate nitrogen ¼ 80.00 mgN L1; total phosphorus ¼ 14.02 mgP L1; orthophosphate ¼ 14.00 mgP L1). Citric acid (monohydrate 99.4% p.a.) was the only source of organic carbon. According to Henze (1991) the concentration ratio of organic compounds to nitrogen should be from 5 to 10 mgCOD$mg1NNO3. The organic compounds loading was 3.98$104 mgCOD$m2$d1 (1$106 mgCOD$m3$d1). Hydraulic retention time in the reactor was 12 h (6 h e mixing, 6 h e aeration). The lack of a sedimentation phase resulted from the replacement of the whole treated wastewater together with the exfoliated biofilm. The studies lasted 180 cycles and were divided into two stages. In the first stage (cycles 1e60), the reactor was activated in order to form the biofilm on the packing. In the second stage (cycles 61e180), having established a distinct stable biofilm on the media, the investigations were carried out to determine’ the rate of the use of citric acid, the denitrification kinetics and changes in the communities of microorganisms inhabiting the media. In the initial period of the AnSBBR operation (cycles 0e60), 50% of the treated wastewater volume was exchanged, with the aim of creating favorable conditions for the colonization of the media surface by the microorganisms introduced with the inoculum. The oxygen level in the reactor was below 0.1 mgO2$L1. The AnSBBR was operated at a temperature of 20e22  C. Activated sludge from the mixing chamber of the “Łyna” Municipal Sewage Treatment Plant in Olsztyn constituted the inoculum. The kinetics of the utilization of organic compounds and denitrification were monitored after 100, 140 and 180 operation cycles. 2.1. Physicochemical analyses In the study were measured pH value and temperature using a CP-105 waterproof pH-meter (Elmetron, Poland); dissolved oxygen using Oxi 330i/set (WTW, Germany), total nitrogen using Total Organic Carbon Analyzer TOC-L CPH/CPN with TNM-L device (Shimadzu Corporation, Japan) for determination of total nitrogen with the “oxidative combustion-chemiluminescence” method. Chemical oxygen demand (COD), ammonia nitrogen, nitrites, nitrates, orthophosphate and dry mass according to APHA (1992) were also determined for reactor effluent. The biomass growth during one cycle was determined by filtration of the whole volume of the treated wastewater through a medium filter, followed by determination of the dry mass retained on the filter according to APHA (1992). The measurements were carried out three times per week following the cycle, after which no physicochemical analyses were made. 2.2. Sampling for metagenomic analyses The samples for metagenomic analysis were collected from the inoculum (at the beginning of the research) (AMW), and on the 100th (AMC1) (at the moment when the total nitrogen concentration below 2 mgN L1 was obtained at the outflow of the reactor) and the 180th (during the stable operation of the reactor) (AMC2) cycles (Fig. 1A). 2.3. DNA extraction Genomic DNA was extracted from 0.2 g of semidry biofilm samples collected at 100 and 180 cycle of the process and from activated sludge used as inoculum. DNA extracts were prepared in

514

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

triplicates and pooled before PCR amplification. DNA was purified using FastDNA Spin Kit for soil (MP Biomedicals, USA) as per the instructions of the manufacturer. Qubit 2.0 Fluorometer (Invitrogen, USA) was used to obtain accurate DNA quantification. The purified DNA was suspended in 100 mL of deionized, DNAase free water and stored at 20  C. 2.4. Polymerase chain reaction and Denaturing Gradient Gel Electrophoresis Bacterial 16S rRNA gene fragments were amplified with these primers: F968-GC (50 -CGC CCG GGG CGC GCC CCG GGCGGG GCG GGG GCA CGG GGG GAA CGC GAA GAA CCT TAC-30 ) and R-1401 (50 CGG TGT GTA CAA GAC CC-30 ), as described by Nübel et al. (1996). PCR was performed in an Eppendorf Mastercycler® gradient thermalcycler (Eppendorf). 50 ng of extracted DNA were used for the PCR mixture, which contained 0.5 mM of each primer, 100 mM of deoxynucleoside triphosphate (Promega, Wisconsin, U.S.A.), 0.6 U of Hypernova DNA polymerase (DNA-Gdansk, Poland), 3 mL of reaction buffer (100 mM Tris-HCl, 500 mM KCl, 1.5% Triton X-100), 1.5 mM MgCl2 and sterile water for a final volume of 30 mL. The temperature program for DNA amplification was as follows: 94 C for 5 min; 35 cycles of denaturation at 94 C for 45 s, annealing at 58 C for 45 s, extension at 72 C for 45 s, and a single final elongation at 72 C for 10 min. The PCR products were checked via

electrophoresis on 1.0% agarose gels stained with ethidium bromide. PCR products with a GC clamp were resolved in a 6% polyacrylamide gel (37.5:1 acrylamide:bisacrylamide) with a gradient ranging from 30 to 60% of denaturating agent (formamide and urea). Electrophoresis was performed for 15 h at 60 V in 1 TAE buffer (2 M Tris base, 2 M Acetic acid, 0.05 M EDTA) using a Dcode system (Bio-Rad Laboratories Inc., Hercules, Canada). The DNA mixture resolved in gel was visualized by staining with 1:10.000 SybrGold (Invitrogen) for 20 min followed by UV transillumination. Images were recorded and analyzed with KODAK 1D 3.6 Image Analysis Software (Eastman Kodak Company, USA). Based on the band intensity on the gel tracks, measured by the peak heights of the densitometric curves, a Shannon index of general diversity H (Shannon and Weaver, 1963) was calculated. The H P value was calculated using the following equation: H ¼  (ni/N) *log(ni/N); where ni is the height of the peak and N the sum of all peak heights of the densitometric curve. 2.5. Library preparation and Illumina sequencing The microbial communities of the three analyzed samples were identified by amplifying and sequence analysis of the V3-V4 region of 16S rRNA gene from the metagenome. 16S rRNA gene fragment was amplified using Illumina recommended PCR primers. These primers were created by adding Illumina adapter overhang

Fig. 1. A e changes in the COD and total nitrogen concentration during the studies (AMW, AMC1, AMC2 e the biofilm sample collection for the metagenomic analysis); B e kinetics of changes in the organic compounds and total nitrogen concentration on the 100th cycle of the AnSBBR operation; C e changes in nitrites concentration and pH on the 100th cycle of the studies, D e kinetics of changes in the organic compounds and total nitrogen concentration on the 180th cycle of the AnSBBR operation; E  changes in nitrites concentration and pH on the 180th cycle of the studies.

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

nucleotide sequences to the PCR primers given by Klindworth et al. (2013). PCR amplification was performed accordingly to Illumina protocol using Platinum® Taq DNA Polymerase High Fidelity (ThermoFisher Scientific). Amplicons were indexed using Nextera® XT Index Kit accordingly to producer's instruction. DNA was sequenced on an Illumina MiSeq instrument using 2  250 pairedend mode. For the sequencing Miseq reagent kit v3 (Illumina, San Diego, USA) was used. Sequencing and quality control followed the manufacturer's recommendation. 2.6. Bioinformatic analyses The sequencing results were recorded as FASTQ files and uploaded to the MetaGenome Rapid Annotation Subsystems Technology (MG-RAST) server for analysis (Meyer et al., 2008). Each file underwent quality control (QC), which included quality filtering (removing sequences with 5 ambiguous base pairs), length filtering (removing sequences with a length 2 standard deviations from the mean). The automated pipeline provided by MG-RAST was also used to obtain taxonomic classification using the BLAT program referencing the M5RNA database. All identifications were made using a maximum e-value of 1e-5, a minimum identity cutoff of 90%, and a minimum alignment length of 50 bp. A 100% stacked column chart comparing the relative abundances of each order in the two groups was generated using Microsoft Excel. Rarefaction plots were created to estimate the limits of detection of the sequencing efforts. 2.7. Accession numbers The Illumina metagenomic datasets are available at MG-RAST under accession numbers 4613174.3 (AMW), 4613170.3 (AMC1) and 4613171.3 (AMC2). 3. Results The studies demonstrated that citric acid aids the denitrification process. Simultaneously, the biomass growth in AnSBBR was lower by an order of magnitude compared to the activated sludge process. During the studies, the kinetics of the consumption of organic compounds and nitrogen removal changed as did the composition of the microbial community.

may be distinguished on the 100th cycle. The first period was characterized by a low consumption of organic compounds (0e3 h) and a low rate of total nitrogen removal (0e4 h) (Fig. 1B). Then, accumulation of nitrites occurred. Their maximum concentration was observed in the 4th hour of the cycle, amounting to 19.42 mgN L1 (Fig. 1C). In the second period, increases in the degradation rate of the organic substrate and the denitrification rate occurred. After depletion of the oxidized forms of nitrogen, the degradation rate of citric acid decreased to 1.9 mgCOD$L1$h1 (Fig. 1B). In the later period of the studies (cycles 101e180), a gradual increase occurred in the removal rate of organic compounds. The average effluent COD between cycles 101e140 was 101.5 ± 15.6 mgCOD$L1, while between cycles 141e180 51.5 ± 10.5 mgCOD$L1. Simultaneously, the total nitrogen concentration remained below 2.00 mgN L1 (Fig. 1A). Between cycles 141e180, the average total phosphorus effluent concentration was 12.23 ± 0.18 mgP L1. The rate of organic compounds consumption (Fig. 1D) was 149.6 and 143.8 mgCOD$L1$h1, while the denitrification rate was 21.42 and 24.38 mgN L1 h1 (140th and 180th cycles respectively). The maximum concentration of nitrites was observed much earlier than in the 100th cycle(Fig. 1E). The nitrates concentration was 11.37 mgN L1e in the 180th cycle. Moreover, in the initial hours of the cycle, no slowdown of pollutant removal was observed as was the case in the 100th cycle (Fig. 1B and D). During the studies, the ratio of consumed CODre/Nre also increased. At the total reduction of nitrates in the 100th cycle, the CODre/Nre ratio was 5.18, while under steady conditions (cycles 141e180)5.73 ± 0.11. The biomass growth (biomass yield coefficient e Yobs), was 0.04 ± 0.02 mgTSS mgCOD1 re , (0.22 ± 0.09 mgTSS mgN1 re ). 3.2. Microbial communities’ composition 3.2.1. PCR-DGGE Denitrifying organisms were introduced to the AnSBBR as an inoculum from an activated sludge municipal wastewater treatment plant (WWTW). Analysis using PCR-DGGE showed that the bacterial community makeup of the inoculum (AMW) was different from those of biofilm from AnSBBR. Significant difference was also

3.1. Physicochemical studies In the 95th operation cycle, total nitrogen concentration below 2.00 mgN L1 was observed for the first time in the effluent, and it remained at that level throughout the experiment. The concentration of organic compounds amounted to 103.42 mgO2$L1 in the 95th cycle (Fig. 1A). In this period, the biofilm thickness was 1e12 mm and did not change to the end of the studies. A thinner film occurred at the edges of the discs, while a thicker film was found at the center of the disk. The color of the film ranged from milky to light brown, exhibiting a slight diversification in the cross-section (lighter at the liquid-film boundary, darker in the deeper layers), which probably resulted from its different “age” in the individual layers (external layers e “the youngest biofilm”) (Fig. 2). The introduction of citric acid decreased the pH of the treated wastewater to the value of 4.59 at the beginning of the cycle. The consumption of the organic substrate and an increase in the alkalinity resulting from denitrification led to an increase in the pH to approx. 8.10 at the end of the cycle. Two periods of the consumption of the organic substrate and in the course of denitrification

515

Fig. 2. AnSBBR media on the 100th operation cycle.

516

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

registered for samples of biofilm (Fig. 3). The values of Shannon index of diversity measured for inoculum (AMW), AMC1, and AMC2 were 2.67, 2.16 and 2.35, respectively. 3.2.2. Metagenomic analysis In order to analyze the composition of the microbial communities, the phylogenetic profiling of two samples taken during development of the biofilm with the control sample representing activated sludge from WWTP were compared. The statistics of analyzed DNA sequences obtained by Illumina approach are summarized in Table 1. Taxonomic affiliation was assigned to contigs with rRNA genes based on comparison with M5RNA database. Alpha diversity for samples AMW, AMC1 and AMC2 were 80.88, 11.70 and 22.43, respectively. Rare fraction curves for each of the samples were asymptotic suggesting that the majority of bacterial taxons were revealed. Both alpha diversity and refraction curves were examined using MG-RAST (Table 1, Supp. 1). Bacteria constitute the dominant domain (94.5e98.3 of the total community) in all samples. At order level, 109 orders were found in AMW, 69 orders in AMC1, and 83 orders in AMC2 sample (Supp. 2). The proportions of the 6 major orders are presented in Fig. 4. Contigs affiliated to Lactobacillales and Actinomycetales constituted the majority of the communities. Lactobacillales predominated in

AMC1 (42.9% of the total community), whereas in inoculum (AMW) and AMC2 this order was detected at much lower level (4.7 and 4.6 respectively). Except Lactobacillales, five others major orders were distinguished (Fig. 4). Contigs affiliated to Lactobacillales and Actinomycetales constituted the majority of the communities. The contigs number of the second order - Actinomycetales, was highest in sample of inoculum (AMW) where it was counted at the level of 22.3%. In sample AMC1 Actinomycetales constituted 3.9% of the total communities and in sample AMC2 it abundance increased to 8.9%. A minor fraction of the total communities was represented by the contigs affiliated to orders: Rhizobiales (5.9e0.55%), Flavobacteriales (4.7e1.7%), Sphingobacteriales (3.8e0.22%), and Burkholderiales (4.0e0.9%). Moreover the proportions of abundance between major orders were most equal in sample AMC2 (Fig. 4). Bacterial diversity and the proportion of the organisms were analyzed mores specifically at the genus level (Fig. 5). The sample of inoculum (AMW), characterized by the highest indices of biodiversity, was predominated by two genera: Arthrobacter (3.9%) and Trichococcus (2.9%). The relative proportions of other major genera in this sample were as follows: Terrimonas (1.2%), Clostridium (1.02%), Enterococcus (0.86%), and Micrococcus (0.53%). In the sample AMC1 number of genera with abundance above 1% was higher than in AMW sample. The sample of biofilm taken at 50 day (AMC1) was predominated by Trichococcus (Lactobacillales), which represented almost 43% of the total community. Less abundant were genera: Flavobacterium (3.63%), Nesterenkonia (1.39%), Kocuria (1.38%), Comamonas (1.27%) and Alistipes (1.01%). Whereas sample AMC2 was predominated by Propionibacterium (5.74%), Agrobacterium (5.23%), Trichococcus (4.37%), Flavobacterium (1.32%), Sphaerotilus (1.35%) and Erysipelothrix (1.08%). In the both samples of biofilm numerous genera possibly, involved in nitrate reduction were recorded, among others Trichococcus, Propionibacterium, Comamonas, Agrobacterium, Flavobacterium, Acidovorax, Chryseobacterium, Micrococcus, Microbacterium, Rhodobacter, Thauera, Azonexus and Sinorhizobium. However the number of these genera member cells was different in AMC1 and AMC2 samples. The read numbers of Trichococcus, Flavobacterium, Comamonas, Chryseobacterium, Micrococcus and Rhodobacter was lower in AMC2 than in AMC1 sample. The highest increase of cell number during process was noted for genera Azospira, Sinorhizobium, Propionibacterium and Arthrobacter (Fig. 5). 4. Discussion

Fig. 3. The difference in bacterial diversity of analyzed samples obtained by Denaturing Gradient Gel Electrophoresis (DGGE).

The diversification of microbial communities reflected differences in both the effectiveness and kinetics of pollutant removal. In the initial stage of the studies e cycles 60e100 e when the whole volume of the treated wastewater was exchanged, and a distinct biofilm was already formed on the discs, a gradual decrease in the pollutant concentration in the treated wastewater occurred. A high proportion of bacteria from the Lactobacillaces genus, and particularly those from the Trichococcus genus, was a characteristic feature of the studied biofilm samples in this period. Importantly, in the sample collected from the 100th cycle, a share of these bacteria exceeding 40% of all contigs was found. This may be explained by the ability of the bacteria from the Trichococcus genus to use citric acid, which had been proved earlier by Stams et al. (2009). In their paper, it was ascertained that Trichococcus strain R210 is able to decompose citric acid to acetate and formate. Most probably, in the presented work the presence of more easily absorbable acetate and formate contributed to the growth of other bacteria, leading in consequence to a decrease in Trichococcus abundance after 180 cycles of the process. pH of the treated wastewater is an important factor, resulting from both the occurrence and the activity of microorganisms. It is

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

517

Table 1 Metagenomic sequences statistic derived from MG-RAST platform (QF - quality filtering). Sample

Total reads

Reads (post QF)

Mean sequence length (post QF)

Mean GC percent (post QF)

a diversity

AMW AMC1 AMC2

284,097 353,077 308,323

260,504 332,684 282,780

313 ± 158 bp 353 ± 162 bp 317 ± 160 bp

55 ± 4% 53 ± 4% 54 ± 4%

80.88 11.70 22.43

Fig. 4. Order level affiliations assigned to contigs with 16S rRNA genes in analyzed samples. Only orders with abundance higher than 4.5% are given.

also of importance in the denitrification process. It is assumed that the optimum range of pH is 7.0e8.0. For pH below 4.0, the accumulation of more nitrites occurs, which exerts a toxic effect on the majority of enzymes. The denitrification practically ceases in acidic and basic environments e for pH below 4.0 and above 9.5. In the research of Thomsen et al. (1994), in the case of activated sludge at a pH of 5.5, nitrites started to prevail (92%) when nitrates became exhausted. In the following phases of the wastewater treatment the reduction of nitrites occurred sequentially. In the studies carried out, a low pH at the beginning of the cycle was most probably a cause of nitrites accumulation. Their maximum concentration was observed in the 4th hour during investigations of kinetics in the 100th cycle. The consumption rate of organic compounds was also connected with the accumulation of nitrites. The rate increased (from the 3rd h of the cycle) slightly earlier than when the denitrification rate increased (from the 4th h of the cycle). Most

probably, this was a result of an increase in pH of the treated wastewater as a consequence of the decomposition of a part of the introduced citric acid. However, it should be emphasized that in spite of the low pH and significant changes in its value during the cycle, a stable, almost 100% effectiveness of nitrates reduction was achieved. These results from the buffering properties of the extracellular polysaccharides forming the matrix of the biofilm, in which the microorganisms are embedded (Sheng et al., 2010). The accumulation of nitrites was also found at the end of the studies (180th cycle). However, the maximum concentration was observed as early as in the 1st hour of the cycle, and organic compounds were consumed steadily in the presence of nitrates. Most probably, this was an effect of the change and adaptation of the composition of microbial communities to the applied substrate, and conditions in the reactor. While analyzing the microbiological composition, it may be concluded that in the initial period, citric acid favors bacteria from the Trichococcus genus, the metabolites of which enable the growth of such bacteria as Propionibacterium, Agrobacterium, Flavobacterium, Sphaerotilus and Erysipelothrix, having a similar share in the microbiological composition. The more uniform microbiological structure in the mature biofilm could additionally be determined by the phenomenon of quorum sensing (QS). Owing to the synthesis of signal molecules by microorganisms, they are capable of intercellular signaling regarding cell count of a given population in a biofilm. The following processes may be regulated in response to a signal that may be recognized by the same species or also by other species: differentiation of cells, production of enzymes and toxins, biosynthesis of secondary metabolites, virulence, transfer of plasmids, DNA replications, sporulation or bioluminescence. Hence, communication between cells in a biofilm allows treating the biofilm as some kind of a primitive organism with a significantly better developed regulation system

Fig. 5. The abundance of the major genera in analyzed samples. Only orders with abundance higher than 0.25% are shown.

518

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519

compared to single cells (Daniels et al., 2004; de Kievit, 2009; Jain et al., 2007). Investigating the influence of acetate and methanol on the microbial communities of activated sludge, Hagman et al. (2008) found a higher denitrification rate while using a mixture of substrates, in comparison to pure compounds. Moreover, other bacteria were responsible for the consumption of methanol, and another ones e for the consumption of acetate. This, as a consequence of using a mixture, allowed for obtaining more complex microbial communities. Dionisi et al. (2004) used acetate, glucose, glutamic acid, ethanol, and their mixture as the external carbon source. They also ascertained that the denitrification rate is higher when a mixture is used, in comparison to that obtained with a single substrate. Moreover, depending on the applied source of organic carbon, metabolic processes run via various pathways. Most probably, interactions between various groups of microorganisms in the experiments carried out allowed for a more effective use of citric acid in the denitrification process even in spite of the significant decrease in pH of the treated wastewater, and the limited accumulation of nitrites. The lack of electron acceptors in the form of nitrates or nitrites at the end of the cycle, both during the activation and the stable operation, limited the consumption of organic compounds. Despite achieving almost 100% denitrification and the occurrence of microorganisms participating in this process as early as in the 100th cycle of the studies, reaching a complete stabilization of the reactor's operation continued to approximately the 140th cycle of the studies. There were no significant changes in the kinetics and effectiveness of the consumption of organic compounds or denitrification later in the process. The COD value at the level of 51.5 ± 10.5 mgCOD L1, total nitrogen below 2 mgN L1 and total phosphorus of 12.23 ± 0.18 mgP L1 in the applied conditions of the reactor's operation (anaerobic conditions, HRT ¼ 12 h) constituted the minimum values which might be obtained in the treated wastewater. Phosphorus was used for the synthesis of new biomass and that's why P removal was low (12.6%). Citric acid fed to the reactor as a source of organic carbon process was also used for the denitrifies biomass growth and maintenance of the biomass. The percentage of carbon utilized in this way depends on the species composition of microorganisms, the chemical composition of the carbon source and other variables. The studies carried out indicated that with an excess of the introduced citric acid, the amount of carbon utilized in growth or maintenance increased over time. Moreover, this increase corresponded to the increase in the rate of consumption of organic compounds, and simultaneously to the increase in the denitrification rate. This phenomenon could be a consequence of the better adapted structure of microorganisms which were able to use citric acid in the denitrification together, but lost more energy than the microorganisms occurring in the 100th cycle, when Trichococcus prevailed. From the point of view of the operation of a wastewater treatment plant, the magnitude of the biomass growth, generated by the introduction of an additional load of organic compounds in the form of an external carbon source, is a very important factor. While analyzing the literature data in relation to the activated sludge, where various substrates were used as the external source of organic carbon for aiding the denitrification, the biomass growth in the presented studies was lower by an order of magnitude (Janczukowicz et al., 2013; Randall Clifford et al., 1992). In order to understand and anticipate the effect of a particular organic substrate for the effectiveness of wastewater treatment, the selective influence of organic substances on the growth of microorganisms should be taken into account. Based on the experiments carried out, it may possible that the biofilm in an AnSBBR utilizes the carbon facilitating the reduction of the pH of the treated

wastewater, aiding the denitrification, without the need to adjust the pH value. It may be possible to use sources of organic carbon which have not been used hitherto, such as wastewater from the agricultural and food industries, containing mineral acids originating from installation cleaning. This study indicates that to improve the kinetics and effectiveness of treatment using biofilm reactors there is a value in understanding the influence on microbial communities brought about by dosing citric acid, or other substrates in biofilm reactors. In order to achieve stable operation of the AnSBBR denitrification process the periodic application of a mixture of the citric acid and acetic acid, being a metabolite of citric acid, should be considered. This procedure should aid faster growth of additional groups of bacteria. 5. Conclusion Citric acid turned out to be a very efficient source of organic carbon in the denitrification process occurring in an AnSBBR. The rate of citric acid consumption and the denitrification rate were lower during the reactor's activation period than under steady conditions. Using citric acid as the source of organic carbon, microorganisms from the Trichococcus genus develop at first (above 40%), enabling, in the later period, a better growth of microorganisms from Propionibacterium (5.74%), Agrobacterium (5.23%), Flavobacterium (1.32%), Sphaerotilus (1.35%), Erysipelothrix (1.08%) genera with a 4.37% share of Trichococcus. Acknowledgments The project was funded by the National Science Centre, Poland (the decision no DEC-2012/07/N/ST8/03201). Supported by the Foundation for Polish Science (FNP) - beneficiary is Artur Mielcarek. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.chemosphere.2016.12.099. References APHA-AWWA-WEF, 1992. Standard Methods for the Examination of Water and Wastewater, eighteenth ed. (Washington, USA). Daniels, R., Vanderleyden, J., Michiels, J., 2004. Quorum sensing and swarming migration in bacteria. FEMS Microbiol. Rev. 28, 261e289. http://dx.doi.org/ 10.1016/j.femsre.2003.09.004. de Kievit, T.R., 2009. Quorum sensing in Pseudomonas aeruginosa biofilms. Environ. Microbiol. 11, 279e288. http://dx.doi.org/10.1111/j.1462-2920.2008.01792.x. Dionisi, D., Renzi, V., Majone, M., Beccari, M., Ramadori, R., 2004. Storage of substrate mixtures by activated sludges under dynamic conditions in anoxic or aerobic environments. Water Res. 38, 2196e2206. http://dx.doi.org/10.1016/ j.watres.2004.01.018. Hagman, M., Nielsen, J.L., Nielsen, P.H., Jansen, J., La, C., 2008. Mixed carbon sources for nitrate reduction in activated sludge-identification of bacteria and process activity studies. Water Res. 42, 1539e1546. http://dx.doi.org/10.1016/ j.watres.2007.10.034. Hallin, S., Throback, I.N., Dicksved, J., Pell, M., 2006. Metabolic profiles and genetic diversity of denitrifying communities in activated sludge after addition of methanol or ethanol. Appl. Environ. Microbiol. 72, 5445e5452. http:// dx.doi.org/10.1128/AEM.00809-06. Henze, M., 1991. Capabilities of biological nitrogen removal processes from wastewater. Water Sci. Technol. 23, 669e679. Jain, A., Gupta, Y., Agrawal, R., Jain, S., Khare, P., 2007. Biofilmsda microbial life perspective: a critical review. Crit. Rev. Ther. Drug Carr. Syst. 24, 393e443. http://dx.doi.org/10.1615/CritRevTherDrugCarrierSyst.v24.i5.10. Janczukowicz, W., Rodziewicz, J., Czaplicka, K., Kłodowska, I., Mielcarek, A., 2013. The effect of volatile fatty acids (VFAs) on nutrient removal in SBR with biomass adapted to dairy wastewater. J. Environ. Sci. Heal. Part A 48, 809e816. http:// dx.doi.org/10.1080/10934529.2013.744658.

A. Mielcarek et al. / Chemosphere 171 (2017) 512e519 Janczukowicz, W., Rodziewicz, J., Thornton, A., Czaplicka, K., 2012. Effect of fermented wastewaters from butter production on phosphates removal in a sequencing batch reactor. Bioresour. Technol. 120, 34e39. http://dx.doi.org/ 10.1016/j.biortech.2012.06.020. Kargi, F., Uygur, A., Bas¸kaya, H.S., 2005. Phosphate uptake and release rates with different carbon sources in biological nutrient removal using a SBR. J. Environ. Manag. 76, 71e75. http://dx.doi.org/10.1016/j.jenvman.2005.01.010. € ckner, F.O., Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., Glo 2013. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 41, e1. http://dx.doi.org/10.1093/nar/gks808. Low, E.W., Chase, H.A., 1999. Reducing production of excess biomass during wastewater treatment. Water Res. 33, 1119e1132. http://dx.doi.org/10.1016/ S0043-1354(98)00325-X. Meyer, F., Paarmann, D., D'Souza, M., Olson, R., Glass, E., Kubal, M., Paczian, T., Rodriguez, A., Stevens, R., Wilke, A., Wilkening, J., Edwards, R., 2008. The metagenomics RAST server e a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinforma. 9, 386. http:// dx.doi.org/10.1186/1471-2105-9-386.  Mielcarek, A., Janczukowicz, W., Ostrowska, K., Jo zwiak, T., Kłodowska, I.,  ski, M., 2013. Biodegradability evaluation of wastewaters Rodziewicz, J., Zielin from malt and beer production. J. Inst. Brew. 119, 242e250. http://dx.doi.org/ 10.1002/jib.92. Mielcarek, A., Rodziewicz, J., Janczukowicz, W., Dulski, T., Ciesielski, S., Thornton, A., 2016. Denitrification aided by waste beer in anaerobic sequencing batch biofilm reactor (AnSBBR). Ecol. Eng. 95, 384e389. http://dx.doi.org/10.1016/ j.ecoleng.2016.06.083. Mielcarek, A., Rodziewicz, J., Janczukowicz, W., Thornton, A., 2015a. The feasibility of citric acid as external carbon source for biological phosphorus removal in a sequencing batch biofilm reactor (SBBR). Biochem. Eng. J. 93, 102e107. http:// dx.doi.org/10.1016/j.bej.2014.10.001.  Mielcarek, A., Rodziewicz, J., Janczukowicz, W., Thornton, A.J., Jo zwiak, T., Szymczyk, P., 2015b. Effect of the C: N:P ratio on the denitrifying

519

dephosphatation in a sequencing batch biofilm reactor (SBBR). J. Environ. Sci. 38, 119e125. http://dx.doi.org/10.1016/j.jes.2015.05.018. Nübel, U., Engelen, B., Felske, A., Snaidr, J., Wieshuber, A., Amann, R.I., Ludwig, W., Backhaus, H., 1996. Sequence heterogeneities of genes encoding 16S rRNAs in Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. J. Bacteriol. 178, 5636e5643 doi:0021-9193/96/$04.00þ0. Randall Clifford, W., Bernard, James L., Stensel, H.D., 1992. Design of activated sludge biological nutrient removal plants. In: Design and Retrofit of Wastewater Treatment Plants for Biological Nutrient Removal, Water Quality Management Library - Volume 5. Technometric Publishing Company, Inc., Lancaster, USA, pp. 97e184. Rodziewicz, J., Janczukowicz, W., Mielcarek, A., Filipkowska, U., Kłodowska, I., Ostrowska, K., Duchniewicz, S., 2015. Anaerobic rotating disc batch reactor nutrient removal process enhanced by volatile fatty acid addition. Environ. Technol. 36, 953e958. http://dx.doi.org/10.1080/09593330.2014.969328. Shannon, C., Weaver, W., 1963. The Mathematical Theory of Communication. University of Illinois Press, Urbana. Sheng, G.-P., Yu, H.-Q., Li, X.-Y., 2010. Extracellular polymeric substances (EPS) of microbial aggregates in biological wastewater treatment systems: a review. Biotechnol. Adv. 28, 882e894. http://dx.doi.org/10.1016/ j.biotechadv.2010.08.001. Singh Dhillon, G., Kaur Brar, S., Verma, M., Tyagi, R.D., 2011. Recent advances in citric acid bio-production and recovery. Food Bioprocess Technol. 4, 505e529. http:// dx.doi.org/10.1007/s11947-010-0399-0. Soccol, C., Vandenberghe, L., Rodrigues, C., Pandey, A., 2006. New perspectives for citric acid production and application. Food Technol. Biotechnol. 44, 141e149. Stams, A.J.M., Huisman, J., Garcia Encina, P.A., Muyzer, G., 2009. Citric acid wastewater as electron donor for biological sulfate reduction. Appl. Microbiol. Biotechnol. 83, 957e963. http://dx.doi.org/10.1007/s00253-009-1995-7. Thomsen, J.K., Geest, T., Cox, R.P., 1994. Mass spectrometric studies of the effect of pH on the accumulation of intermediates in denitrification by paracoccus denitrificans. Appl. Environ. Microbiol. 60, 536e541 doi:0099-2240/94/ $04.00þ0.