Energy Consumption for Nutrient Removal from High-Nitrate and High-Phosphorus Wastewater in Aerobic and Anaerobic Bioelectrochemical Reactors

Abstract

An attempt was undertaken to determine indicators of energy consumption in bio-electro reactors (BERs) i.e., an aerobic rotating electrobiological disc contactor (REBDC) and an anaerobic sequencing batch biofilm reactor (SBBR), during contaminant removal from soilless tomato cultivation wastewater having a specific composition, i.e., high nitrate and phosphorus concentrations and low COD. Because of this specificity, the energy consumption during the treatment process was characterized by a cumulative indicator for simultaneous removal of phosphorus and nitrates—EEI_NUTRIENTSrem (electric energy consumption per unit of removed nutrient load, expressed as kWh/kg_NUTRIENTSrem). Four values of direct current density were tested: 0.63, 1.25, 2.5, and 5.0 A/m². The indicator values were compared at a hydraulic retention time (HRT) of 24 h. The study demonstrated that the values of electric energy consumption per unit of removed nutrient load determined in the anaerobic SBBR ranged from 30 to 464 kWh/kg _NUTRIENTSrem and were lower than the values obtained in the aerobic REBCD, i.e., 80–1380 kWh/kg _NUTRIENTSrem.

1. Introduction

Wastewater treatment plants are estimated to be responsible for approximately 1% of the annual electric power consumption in Europe [1]. In turn, electric energy accounts for 38–52% of the total energy consumption by a wastewater treatment plant. The other components of energy consumption include manual energy, mechanical energy, and chemical energy [2,3]. The mean contribution of biotreatment processes in the total electric power consumption accounts for 83% [4]. The numbers indicate the importance of employing technologies with possibly the lowest demand for electric energy in wastewater treatment. This is not just a question of low energy consumption, as it goes hand in hand with reducing the carbon footprint of the wastewater treatment plant. This is particularly important in the case of industrial and other wastewater having a specific composition requiring treatment technologies other than those involving activated sludge.

The currently used indicators of energy consumption in wastewater treatment plants are not easy to interpret and prevent a reliable comparison of different systems. The literature provides the following indicators of energy consumption per unit of removed pollutant load—electric energy indicators (EEIs) [1]:

• EEIQ—indicator of electric energy consumption per unit of volume of influent wastewater processed, expressed in kWh/m³,

• EEIL_rem—indicator of electric energy consumption per unit of removed load (L), expressed in kWh/kgL_rem,

• EEIPE—indicator of electric energy consumption per year and per population equivalent (PE) served, expressed in kWh/PE·year.

The EEIQ is useful exclusively to compare installations and facilities treating wastewater of similar quality. In turn, the EEIPE indicator enables comparing any treatment technologies designed for household, municipal, or industrial wastewater, and is commonly used regarding organic compounds (most often including BOD). It also allows determining energy consumption consumed for nitrogen (TN) and phosphorus (TP) removal; however, in this case it is necessary to compute the population equivalent based on unit loads of nitrogen and phosphorus. Unfortunately, the EEIPE takes no account of the wastewater treatment plant efficiency, consequently making it impossible to compute the costs of energy consumed to remove loads of individual pollutants. It only provides information about the annual electric energy consumption per 1PE for an instillation that serves a specified PE.

In the case of biotreatment processes, it is recommended to compute energy costs on the basis of the indicators referring to the mass unit of the removed contaminant load (EEIL_rem). The EEIL_rem is the most reliable indicator, which allows determining the energy consumption per unit of removed load referring to the measurable effect of the treatment process, i.e., to the load of any contaminant removed in the biological process, including organic compounds expressed as BOD, COD, and TOC or nutrients including TN and TP (EEIBOD_rem, EEICOD_rem, EEITOC_rem, EEITN_rem, and EEITP_rem, respectively). Energy consumption per unit of removed loads of organic and biogenic compounds is the main cause of energy consumption in wastewater treatment plants.

Ample studies have presented data concerning the electric power consumption of the treatment process in wastewater treatment plants with activated sludge tanks, expressed using the EEIPE indicator determined for organic compounds [5,6,7,8,9]. The analysis of the cited data indicates that the EEIPE (PE based on BOD) computed for most European wastewater treatment plants exceeds 20 kWh/PE·year, which is the value claimed by Svardal and Kroiss [10] as the maximal permissible one considering emission rates from the treatment plant. Generally, a the facility has a higher unit energy consumption expressed by the EEIPE indicator [9]. However, studies lack data on the values of this indicator computed for energy consumption during nitrogen and phosphorus removal [11,12,13].

In the case of the EEIL_rem indicator, its values have been broadly presented for organic compounds expressed as COD and BOD in activated sludge technology [14,15,16], but sparsely reported for nutrient removal. One of the few examples in this case is the research conducted by Zylka [11,12], providing values of EEITN_rem and EEITP_rem indicators for loads of total nitrogen and total phosphorus removed from dairy wastewater in the biological process in SBR-type reactors. They reached 52.90 kWh/kgTN_rem and 141.26 kWh/kgLTP_rem, respectively. The mean EEITN_rem value achieved by Zylka [11] was over threefold higher than that reported by Panepinto et al. [6], which amounted to 14.66 kWh/kgTN_rem and was computed for an entire wastewater treatment plant with a capacity of 615,000 m³/day. They were also much higher than the values reported by Henze et al. [17] (who stated that, in activated sludge processes with nitrification–denitrification, energy consumption was 2.3 kWh/kg N removed) and by Capodaglio et al. [18]. The latter studies presented data concerning ANITA^TMMOX (MBBR system) technology energy requirements of 1.6–1.9 kWh/kg N removed. Chiavola et al. [16] reported only values of the indicator referring to the removed load of ammonia nitrogen (5.5–24.5 kWh/kgN-NH₃rem). To the best of our knowledge, no values of EEITP_rem can be found in the available literature.

The majority of the aforementioned values concerned treatment plants operating based on the activated sludge technology and treating municipal sewage. In turn, sparse and old data are available on biofilm reactors, considered to be less energy-consuming than those with suspended biomass [19]. Data are especially lacking on the bioelectrochemical reactors that employ electrobiological and electrochemical processes. Few articles on this subject confirm this statement, i.e., the article by Wang et al. concerning phosphorus removal during electrochemically mediated precipitation [20] and the article by Yang et al. on nitrate-contaminated wastewater treatment in a biofilm electrochemical reactor [21].

The coupled use of electric current flow and bio-denitrification has been exploited in a new type of reactor called a bio-electro reactor (BER). The flow of electric current induces water electrolysis, which in turn leads to “hydrogen production”. Simultaneously, the surface of a BER cathode allows the site to immobilize denitrifying microorganisms, which then enables direct access to hydrogen produced in situ [22]. Investigations conducted by Kłodowska et al. [23,24] showed that the electric current flow was the main factor determining the number and composition of denitrifying bacteria.

Our earlier studies [25,26,27] demonstrated that wastewater rich in nitrogen (mainly as nitrate) and phosphorus but poor in organic compounds (such as drainage from soilless tomato cultivation) could be effectively treated in biofilm bioelectrochemical reactors, such as a rotating electrobiological disc contactor (REBDC) and a sequencing batch biofilm reactor (SBBR) [28,29]. Research conducted by Mielcarek et al. (2019) [30] showed that wastewater from greenhouses with soilless plant cultivation was characterized by a COD from 37.35 to 78.12 mg O₂/L, TN from 270.00 to 614.89 mg N/L, and TP from 35.44 to 104.00 mg P/L. Mentioned reactors may be used especially at greenhouses located in agglomerations and operating on the basis of the method of vertical, soilless plant cultivation in closed farms (vertical farms).

However, the implementation of this solution additionally requires the estimation of the energy costs of wastewater treatment.

BERs with discs similar to RBCs are characterized by easy construction, simple process control, and low operating and maintenance cost [19]. Among their advantages, the level of greenhouse gas (GHG) emissions should also be taken into account. Wastewater treatment plants (WWTPs) with BERs, like other biological wastewater treatment devices, are a source of anthropogenic GHG emissions to the atmosphere [31]. GHGs from WWTPs can be emitted directly as methane (CH₄), nitrous oxide (N₂O), and carbon dioxide (CO₂) or indirectly as carbon dioxide (CO₂) of fossil origin (due to energy and materials used in treatment processes). Estimations revealed that BERs with discs require moving the shaft and discs using only about 25% of the energy consumption of an activated sludge system and 70–80% of a packed-bed reactor [31]. This means that indirect carbon dioxide emissions connected with energy consumption related to the movement of the shaft and discs for BERs are significantly lower than in municipal WWTPs that typically use activated sludge process technology.

When considering direct GHG emissions from BER with discs treating soilless plant cultivation wastewater, the specificity of this wastewater should be taken into account. It is characterized by a very low concentration of organic compounds (COD below 80.0 12 mg O₂/L [30]. Moreover, the complexity of the processes taking place in BER with discs (biological and physicochemical removal of nitrate, electrocoagulation of phosphorus) has an impact on lower (compared to activated sludge technology) direct GHG emissions. The key advantage of this solution is that, in addition to electrochemical reduction and heterotrophic denitrification, nitrates are removed in the hydrogenotrophic denitrification process using gaseous hydrogen as a donor of electrons and inorganic carbon (carbon dioxide) as a source of energy.

According to Wang et al. [31], biological treatment technologies generate onsite GHG emissions during wastewater treatment processes, but physical, chemical, and physicochemical treatment technologies do not. It is safe to assume that the emissions of GHG, including CO₂ emissions, from BERs with discs treating wastewater from soilless plant cultivation are much lower than from biological treatment technologies [31]. According to this paper, on-site carbon dioxide emission, for municipal wastewater, was 560 g CO₂/kg COD removed for activated sludge technology, 436 g CO₂/kg COD removed for the rotating biological contactor (RBC), and 381 CO₂/kg COD removed for anaerobic biological treatment. Cited emission factors [31] regarded CO₂ emissions as the sum of fossil CO₂ and biogenic CO₂ emission. Bearing in mind that the COD of the treated wastewater was 260 COD mg O₂/L, twofold lower than in the facilities described in [31], it can be assumed that the carbon dioxide emission in the aerobic reactor based on RBC (REBDC) was 218 g CO₂/kg COD removed, whereas that in the anaerobic reactor (SBBR) was about 190 CO₂/kg COD removed.

The electric current flow in both reactors ensures conditions for the simultaneous removal of phosphorus and nitrates. This occurs mainly because of the electrocoagulation process with the use of aluminum or iron anodes (phosphorus) and the processes of hydrogenotrophic denitrification (using gaseous hydrogen as an electron donor) and electrochemical reduction (nitrates) [27].

The effectiveness of hydrogenotrophic denitrification can be established on the basis of the following indicators [32,33]:

• -. Current efficiency (CE) is usually employed to express the denitrification performance of bioelectrochemical reactors:

(1)$\mathrm{CE}=\frac{\left(C_{{\mathrm{NO}}_{3}in}-C_{{\mathrm{NO}}_{3}eff}\right)·5-C_{{\mathrm{NO}}_2}·3}{14·I}·26.8·\frac{Q}{1000}·100 [\%],$

where C_NO3in is the nitrate concentration in the influent (mg N/L), C_NO3eff is the nitrate concentration in the effluent (mg N/L), C_NO2 is the nitrite concentration in the effluent (mg N/L), Q is the flow rate (mL/h), and I is the current applied in the reactor (mA).

• -. Specific removal rate of denitrification (SD):

(2)$\mathrm{SD}=\frac{Q \left(C_{\mathrm{N}\mathit{in}}-C_{\mathrm{N}eff}\right)}{U I} \left[\mathrm{mg} \mathrm{N}/\left(\mathrm{kWh}\right)\right],$

where C_Nin is the total nitrogen concentration in the influent (mg N/L), C_Neff is the total nitrogen concentration in the effluent (mg N/L), Q is the flow rate (L/h), I is the current applied in the reactor (mA), and U is the cell voltage (V).

Current efficiency expressed as a percentage does not present the energy cost of nitrogen removal in BER [28]. The specific removal rate of nitrogen indicates the size of the load removed using 1 kWh, which hampers the direct comparison of results obtained with literature data. The removal of nitrates and phosphorus in bioelectrochemical reactors is largely due to the flow of electric current. In the case of wastewater containing mainly nitrogen and phosphorus compounds, the energy costs of its treatment process can be established using either the indicator of electric energy consumption per unit of removed load (EEIL_rem) computed separately for phosphorus and for nitrogen or a cumulative indicator demonstrating energy consumption during nutrient removal [34]:

(3)${\mathrm{EEIL}}_{\mathrm{rem}}=\frac{U·I·t}{\left(C_{\mathrm{X}in}-C_{\mathrm{X}eff}\right) V}(\mathrm{kWh}/\mathrm{kg} \mathrm{TP} \mathrm{removed}; \mathrm{kg} \mathrm{TNremoved}; \mathrm{kg} \mathrm{nutrients} \left(\mathrm{TP}+\mathrm{TN}\right) \mathrm{removed}),$

where U is the cell voltage (V), I is the current applied in the reactor (A), t is the hydraulic retention time (h), C_Xin is the contaminant concentration in the influent (g TP/L; g TN/L; g (TP + TN)/L), C_Xeff is the contaminant concentration in the effluent (g TP/L; g TN/L; g (TP + TN)/L), and V is the treated wastewater volume (L).

The indicators computed separately for each pollutant type take no account of energy costs incurred during the concomitant removal of other pollutants. Therefore, the cumulative indicator is more representative because the electric energy is consumed simultaneously in both processes (electrobiological and electrochemical) and because its value denotes energy costs incurred in the entire treatment process of a sum of selected contaminants, e.g., biogenic amines. A similar opinion was expressed by Longo et al. [35], who recommended using a broader key performance indicator (KPI) providing information about all loads of removed contaminants. In turn, it is a common practice observed in the literature to present values of energy consumption per unit of removed load of only one type of contaminant, usually COD [36,37,38], nitrogen compounds [16,39], or phosphorus compounds [28]. In contrast, there are no data on the energy consumption per unit of the nutrient (nitrogen plus phosphorus) load removed in the bioelectrochemical reactors.

Considering the above findings, a study was undertaken to determine electric energy consumption in two different bioelectrochemical biofilm reactors treating wastewater with specific composition, e.g., high nitrate and phosphorus concentrations and low COD (soilless tomato cultivation wastewater), i.e., in an aerobic REBDC and an anaerobic SBBR, and comparing the obtained values with data for other BERs. It also allowed answering whether the aerobic or the anaerobic reactor variant was more energy-saving. The study’s goal was achieved by determining values of the electric energy consumption indicators related to the mass unit of the removed pollutant load (EEIL_rem) for nitrogen compounds (TN), phosphorus (TP), and their sum (TN + TP)—EEI_NUTRIENTSrem.

2. Materials and Methods

The research was carried out in two stages using research models of two BERs, REBDC and SBBR, under laboratory conditions, at a temperature of about 20.0 ± 1 °C. In both reactors, sodium acetate was added to wastewater as a source of carbon to ensure the growth of the biofilm biomass [27,40]. The results of the first stage served to limit the scope of the second one with regard to the current density and the hydraulic retention time.

2.1. Rotating Electrobiological Disc Contactor (REBDC)—Stage 1

Technological assumptions were as follows: disc mounted on a horizontal shaft, disc submergence 40%, and aerobic conditions. The specificity of REBDC is that, in the reactor flow, tank there are aerobic conditions, while, in the biofilm attached to the discs, its profile contains not only oxygen but also transitional and anaerobic layers.

The effect of hydraulic retention time (HRT) and electric current density (J) on energy consumption (E) during removal processes of phosphorus and nitrogen in REBDC was analyzed. Nutrient removal was mainly due to the processes of hydrogenotrophic denitrification, electrochemical reduction of nitrates, and electrocoagulation of phosphorus compounds [27].

The reactor used in the study was a rotating biological contactor in which stainless-steel discs with immobilized biomass served as a cathode, whereas aluminum electrode mounted in the flow tank of the contactor served as an anode (Figure 1). Both electrodes were connected to a laboratory power supply (HANTEK PPS2116A, Qingdao, China) which served as a source of direct electric current and allowed maintaining its desired intensity.

The experiment was carried out in four single-stage rotating contactors with technical parameters provided in Table 1.

The research assumed four HRT values—4 h, 8 h, 12 h, and 24 h, and the following direct electric current densities were used for each HRT: 0.63 A/m², 2.50 A/m², 5.00 A/m², and 10.00 A/m². The study was conducted with synthetic wastewater, the composition of which was similar to that of wastewater from soilless cultivation of tomatoes [27]. The composition of wastewater used in the study is presented in Table 2 [27].

2.2. Sequencing Batch Biofilm Reactor (SBBR)—Stage 2

Technological assumptions were as follows: discs mounted on a vertical shaft, disc submergence 100%, and anaerobic conditions. The specificity of SBBR is that, in the reactor tank, there are anaerobic conditions, while, in the biofilm attached to the discs, its profile contains only anaerobic layers.

The study was performed in an SBBR with an aluminum electrode (anode) embedded inside. Rotating stainless-steel discs mounted on a vertical shaft were used as the cathode. The electrodes were supplied with direct current from the laboratory power supply (Rhode&Schwarz HMP 4040, Munich, Germany). The experiment was carried out in four reactors with technical parameters provided in Table 1. A single BEBR is shown in Figure 2 [20], whereas treated wastewater characteristics are presented in Table 2.

The current density range was limited to four values: 0.63 A/m², 1.25 A/m², 2.5 A/m², and 5.0 A/m², in line with previous REBDC research results (which showed that the temperature of the wastewater rose in the reactor to 50 °C during the run with the highest current density of 10.0 A/m² [24]). HRT range was also limited to one value, because REBDC research results showed that the highest efficiencies of nitrogen and phosphorus removal were obtained for the hydraulic retention time of 24 h.

2.3. Analytical Procedures

The following parameters of the inflow and outflow from both reactors were measured: pH value and electrical conductivity (EC), using a CP-105 waterproof pH-meter (Elmetron, Zabrze, Poland); total nitrogen (TN), using a TNM-L device (Shimadzu Corporation, Kyoto, Japan; oxidative combustion—chemiluminescence method); total phosphorus (TP; HACH Lange LCK 348–350 method), using a DR5000 HACH Lange spectrophotometer (Düsseldorf, Germany); chemical oxygen demand (COD; titrimetric method), using a Gerhardt KI 16 (Königswinter, Germany) laboratory heater; ammonia nitrogen (spectrophotometric method); nitrates (spectrophotometric method), using a VWR UV-3100PC Spectrophotometer (Shanghai, China).

3. Results

The electric power consumption in aerobic REBDC, computed for the removed load of nitrogen (EEITN_rem) ranged from 118 at the lowest current density tested (0.63 A/m²) and the shortest HRT (4 h) to 2170 kWh/kg TN_rem (Figure 3) at the highest current density (10.00 A/m²) and the longest HRT (24 h).

It is worth noting that our previous research [27] demonstrated the highest nitrogen removal effectiveness (68.6%) at the highest electric current density and the longest hydraulic retention time. At all current densities tested, the effectiveness of nitrogen removal increased along with extended hydraulic retention time of wastewater in the REBDC. At the current density of 0.63 A/m², the EEITN_rem value increased successively from 118 to 398 kWh/kg TN_rem with HRT extension from 4 to 24 h. At the current density of 1.25 A/m² and the same HRTs, the EEITN_rem values were higher and reached 198 and 678 kWh/kg TN_rem, respectively. The successive increase in the current density to 2.50 A/m² caused a higher energy consumption, i.e., from 290 to 962 kWh/kg TN_rem. Similar observations were made for current densities of 5.00 A/m² and 10.00 A/m², i.e., HRT extension from 4 to 24 h caused the EEITN_rem value to increase from 503 to 1945 kWh/kg TN_rem and from 786 to 2170 kWh/kg TN_rem, respectively. The results of technological experiments [27] demonstrated that HRT extension at a stable current density ensured both a technological effect, i.e., increased effectiveness of nitrogen removal, and a direct energetic effect, i.e., increased energy demand per unit of removed contaminant load.

Similar dependencies were observed for energy consumption during the removal of phosphorus compounds. The electric energy consumption in the REBDC computed for the removed load of phosphorus (EEITP_rem) ranged from 104 kWh/kg TP_rem at the lowest tested current density (0.63 A/m²) and the shortest HRT (4 h) to 9456 kWh/kg TP_rem (Figure 3) at the highest current density (10.00 A/m²) and the longest HRT (24 h). The above values were noticeably higher than those computed for nitrogen removal. Our previous study [27] demonstrated phosphorus removal effectiveness of 81.0% at the lowest electric current density and the shortest HRT, which peaked t 99.8% at the highest current density and the longest HRT. At all current densities tested, the effectiveness of phosphorus removal increased along with extended HRT of wastewater in the REBDC. At the current density of 0.63 A/m², the EEITP_rem values ranged from 104 to 611 kWh/kg TP_rem for HRT extended from 4 to 24 h. At the current density of 1.25 A/m² and the same HRTs, the EEITP_rem values were higher and reached 198 and 1184 kWh/kg TP_rem, respectively. The successive increase in the current density to 2.50 A/m² caused a higher energy consumption, i.e., from 386 to 2360 kWh/kg TP_rem. At current densities of 5.00 A/m² and 10.00 A/m², HRT extension in the reactor from 4 to 24 h caused the EEITP_rem values to increase successively from 764 to 4739 kWh/kg TP_rem and from 1504 to 9456 kWh/kg TP_rem, respectively. The results of a technological experiment [27] demonstrated that, in the case of phosphorus removal, the extension of HRT at a given current density yielded not only a technological effect seen as an increase in phosphorus removal effectiveness but also an energetic effect, i.e., increased energy demand per unit of removed contaminant load. It is worth emphasizing that the dephosphatation effectiveness exceeded 90% even at the lowest current density and HRT of 4 h. At the current density of 2.50 A/m², the phosphorus removal effectiveness exceeded 97% for all HRTs tested, whereas, at 10.00 A/m², it exceeded 99% [27].

Because the electric current flow through a reactor contributes to the simultaneous removal of nitrogen and phosphorus compounds, energy consumption was computed per mass unit of the removed load of both types of nutrients. The technological effect of REBDC operation consisting of the removal of both nutrients, entailing electric energy consumption. Figure 4 presents the values of the energy consumption indicator computed for the nutrient removal (EEI_NUTRIENTSrem) at varying densities of electric current and hydraulic retention times. At the highest current density and the longest HRT, they did not exceed 2000 kWh/kg _NUTRIENTSrem (1999 kWh/kg _NUTRIENTSrem), whereas, at the lowest current density, they ranged from 80 to 330 kWh/kg _{NUTRIENTSrem.} At the subsequent current densities tested, the EEI_NUTRIENTSrem values fell within the following ranges: 130–560 kWh/kg _NUTRIENTSrem, 210–840 kWh/kg _NUTRIENTSrem, 370–1380 kWh/kg _NUTRIENTSrem, and 600–1990 kWh/kg _NUTRIENTSrem.

In the anaerobic sequencing batch biofilm reactor (SBBR), the values of the energy consumption in the denitrification process (EEITN_rem) ranged from 49 to 144.5 kWh/kg TN_rem at current densities ranging from 0.63 to 5.00 A/m² and HRT 24 h (Figure 5). The EEITN_rem indicator value noted at the lowest current density (0.63 A/m²) was eightfold higher than that determined in the REBDC reactor (at the same technological conditions: J range from 0.63 to 5.00 A/m²; HRT = 24 h), whereas, at the highest current density tested, the value was more than 13-fold higher (144.5 and 1945 kWh/kg TN_rem, respectively). The denitrification efficiency reached 36.5% at the lowest current density and 21.9% at the highest one.

In the case of phosphorus, the EEITP_rem values computed in the SBBR reactor (Figure 5) ranged from 82 to 787.0 kWh/kg TP_rem (in J range from 0.63 to 5.00 A/m²; HRT = 24 h) and were several times lower than those determined in the REBDC reactor (611 and 4739 kWh/kg TP_rem, respectively). Here, the dephosphatation effectiveness exceeded 90% at both electric current densities tested (91.2% and 95.8% at 0.63 A/m² and 5.00 A/m², respectively).

Likewise, in the REBDC reactor, the electric current flow through the SBBR reactor caused the simultaneous removal of nitrogen via hydrogenotrophic denitrification and electrochemical reduction, and of phosphorus in the electrocoagulation process. The electric energy is consumed in all these processes simultaneously. In the SBBR, the values of the energy consumption during nutrient removal (Figure 5) ranged from 30 to 464 kWh/kg _NUTRIENTSrem, whereas, in the REBDC, they reached 330 and 1380 kWh/kg _NUTRIENTSrem (Figure 4).

The above results indicate that the electric current was substantially more effectively utilized and that the energy costs per unit of nutrient load removed from drainage generated during soilless tomato cultivation were lower in the anaerobic SBBR reactor.

4. Discussion

The electric current flow through both bioelectrochemical reactors treating wastewater from soilless tomato cultivation ensured conditions for the simultaneous dephosphatation, denitrification, and electrochemical nitrate reduction [27,41,42,43].

In all processes, the direct or indirect roles were ascribed to electrons resulting from electric current flow.

During the investigation, at current densities ranging from 0.63 to 5.00 A/m² and HRT 24 h, in the anaerobic SBBR, the values of EEITN_rem ranged from 49 to 144.5 kWh/kg TN_rem, whereas EEITN_rem indicators for aerobic REBDC were between 398 and 1945 kWh/kg TN_rem.

In the case of BER type aerobic reactors, the literature shows EEITN_rem values of 3.8 [44], 12–14.2 [45], 70.0 [46], and 200.0 [45] kWh/kg TN_rem. These values are lower than those obtained in our study. However, the aforementioned studies were conducted with reactors of different design, intended for the removal of nitrates from water or wastewater with a lower nitrate concentration and with a low or null concentration of phosphorus compounds. In some of the BERs, palm fiber and woodchips were used as carbon sources [21,44], which necessitated additional loads of nitrates and phosphorus to be removed during sorption processes. The highest of the cited values were determined in the aerobic multi-cathode BER treating water free of phosphorus compounds and having a low concentration of nitrates—20 mg N_NO3/L, i.e., 70 kWh/kg TN_rem [46], as well as in the bioelectrochemical systems treating synthetic wastewater with a nitrate concentration of 205 mg N_NO3/L and also free of phosphorus compounds, i.e., 200.0 kWh/kg TN_rem [45].

According to the literature data, the energy consumption in anaerobic reactors may reach even 20.0 kWh/kg TN_rem [47], 70 kWh/kg TN_rem [48], and 440 kWh/kg TN_rem [32]. These values fell within the same range and were comparable to those determined in our study, i.e., 49 to 144.5 kWh/kg TN_rem.

In the case of phosphorus, the EEITP_rem values computed in the anaerobic SBBR (Figure 5) ranged from 82 to 787.0 kWh/kg TP_rem and were several times lower than those determined in aerobic REBDC reactor (611 and 4739 kWh/kg TP_rem, respectively). The EEITP_rem values determined for the aerobic REBDC are comparable with those reported by Wang [20] in bioelectrochemical systems and ranging from 2.2 to 2050 kWh/kg TP_rem. In turn, the values noted in the anaerobic SBBR are similar to those determined by Cid et al. [49] during electrochemical removal of phosphorus from onsite toilet wastewater when the energy consumption reached 4399 kWh/kg TP_rem.

In BER treating high-nitrogen and high-phosphorus wastewater, these pollutants are removed most of all due to electrocoagulation, hydrogenotrophic denitrification, and electrochemical nitrate reduction. All these processes proceed simultaneously but their rate and effectiveness vary. The electric energy is consumed concomitantly in all processes that lead to the removal of a nutrient load, i.e., the load of nitrogen and phosphorus. This means that the energy costs of the treatment process in BER should be determined using a cumulative energy consumption indicator denoting energy expenditures for the removal of a sum of selected pollutants, like nutrients in the present study. Similar conclusions were formulated by Longo et al. [35] and Sabia et al. [50], who recommended that this indicator should provide information about all removed pollutant loads. In the case of drainage derived from soilless tomato cultivation in a greenhouse (characterized by high-nitrate and high-phosphorus concentration and very low COD value), it is absolutely justified.

Experiments were carried out with two types of BER reactors—REBDC and SBBR, with electric current densities ranging from 0.63 to 5.0 A/m², and hydraulic retention time of 24 h. In both reactors, sodium acetate was fed as an external carbon source. Limiting the experimental assumptions to one HRT (24 h) and four electric current densities (0.63 to 5.0 A/m²) was driven by the conclusions from our previous study [24].

At the current density range common for both reactors, i.e., from 0.63 to 5.0 A/m², the EEI_NUTRIENTSrem ranged from 370 to 1380 kWh/kg _NUTRIENTSrem in an aerobic biofilm biomass reactor (REBCD), whereas in an anaerobic biofilm biomass reactor (SBBR), they were significantly lower and ranged from 30 to 464 kWh/kg _NUTRIENTSrem. These differences in energy consumption per removed nutrient load can be due to various oxygen conditions (in SBBR—anaerobic in the reactor and in the biofilm; in REBCD—aerobic in wastewater and external biofilm layer, and anaerobic in the deepest biofilm layers) [51], a different microbiological structure of the biofilm [21,52], various contributions of hydrogenotrophic and heterotrophic denitrification, electrochemical nitrate reduction, electrocoagulation, and precipitation of calcium and magnesium phosphate under the influence of pH changes [27,29], and consequently different effectiveness of nitrogen and phosphorus removal.

Comparing the results of the present study with findings from other experiments is difficult because of the lack of exact literature data on the electric energy consumption per unit of removed nutrients in BER treating wastewater having high concentrations of nutrients. Therefore, to enable such a comparison, the values of the EEI_NUTRIENTSrem indicator were computed for results from our previous studies. In the first article [21], which presents the results of a study on the effect of electrical direct current density (0.053–0.210 A/m²), with citric acid as an external source of carbon and C/N_NO3 ratio of 0.5, on the effectiveness of nitrogen and phosphorus removal from synthetic wastewater with physicochemical parameters typical of municipal sewage subjected to biotreatment in the highly efficient system for organic compound removal ensuring an efficient course of the nitrification process, the EEIL_rem values ranged from 12.4 to 30.68 kWh/kg _NUTRIENTSrem. These values are several times lower than those computed in the present study (maximally 1.99 kWh/kg _NUTRIENTSrem for REBDC and 0.46 kWh/kg _NUTRIENTSrem for SBBR). However, it needs to be emphasized that Kłodowska et al. [24] investigated nutrient removal from municipal sewage pretreated with mechanical and biological methods, having a nitrate content of 50.6 mg N_NO3/L and phosphorus content of 5.16 mg P/L. These are severely lower values than those in the present study.

The second article [29] presented the results from the research performed in bioelectrochemical batch reactors treating wastewater from soilless tomato cultivation with TN of 502 mg/L and TP of 81 mg/L. It is very important to emphasize that the above-cited authors used an alternating current (AC) at densities of 4.4, 8.8, and 13.2 A/m². The EEIL_rem values determined for their study ranged from 197 to 870 kWh/kg_NUTRIENTSrem and are lower than those determined in REBCD and SBBR in the present study. However, it needs to be remembered that Bryszewski et al. [29] used an alternating current with higher densities in their reactor.

According to Liu et al. [53], the processes ongoing in the BER result in the increased pH values along with increasing densities of electric current. The higher pH values can be due to the increased concentration of OH⁻ ions produced upon water electrolysis or to the increased wastewater alkalinity upon anode corrosion at higher intensities of the electric current. This increase occurred faster in the anaerobic reactor, i.e., in SBBR. The above findings coupled with a hallmark feature of drainage water, i.e., high concentrations of calcium and magnesium ions [27], mean that dephosphatation could have also been due to the precipitation of calcium and magnesium phosphate upon pH change. Previous investigations by Jóźwiak et al. [54] and Saxena and Bassi [55] indicated that TP removal effectiveness caused by precipitation with calcium and magnesium ions may exceed 97% at alkaline pH. Therefore, the removal of phosphorus was rather due to the precipitation of calcium and magnesium phosphates induced by pH changes than to the electrocoagulation process. It should also be remembered that, in the SBBR reactor, anaerobic conditions occurred in wastewater and the entire biofilm, while it occurred in the aerobic REBCD reactor only in the deepest biofilm layers. This means that the anaerobic SBBR reactor offered more beneficial conditions for heterotrophic denitrification resulting from feeding an external carbon source [29]. Due to the above, heterotrophic denitrification contributed to nitrate removal, along with hydrogenotrophic denitrification and electrochemical nitrate reduction.

The results obtained for both reactors were higher than those from Zylka et al.’s study [11], who reported the mean values of EEILTN_rem and EEILTP_rem indicators per loads of total nitrogen and total phosphorus removed from dairy wastewater in the biological process (activated sludge) in the SBR type reactors, i.e., 52.90 kWh/kgLTN_rem and 141.26 kWh/kgLTP_rem. The mean value of the EEILTN_rem indicator obtained in this study differs significantly from that reported by Panepinto et al. [6] and reached 14.66 kWh/kgLTN_rem for the entire treatment plant with the capacity of 615,000 m³/day. It needs to be noted, however, that the cited experiments were conducted in technical installations with suspended biomass designed to treat various types of wastewaters (dairy wastewater and municipal sewage) of incomparable quality and having various capacities, which significantly affected the energy consumption per unit of removed pollutants [56]. The scale of the installation also strongly determines the values of technological parameters and electric energy consumption, whereas the results achieved on a laboratory scale are usually higher than those obtained in the technical scale. This difference increases with wastewater treatment plant capacity [57].

According to literature data published for aerobic BERs, nitrogen removal required energy consumption (EEITN_rem) of even 200.0 kWh/kg TN_rem [45], whereas in anaerobic reactors, energy consumption reached up to 440 kWh/kg TN_rem [32]. In the case of phosphorus compounds, the energy consumption (EEITP_rem) reached up to 2050 kWh/kg TP_rem in aerobic reactors [20], compared to 4399 kWh/kg TP_rem in anaerobic reactors [49].

Given the fact that, in the above-cited studies investigating energy consumption in nitrogen removal process, the treated wastewater was free of phosphorus compounds [45] or the EEITP_rem was computed with no account taken of the energy consumed for nitrogen removal [49], the values of the energy consumption needed for nutrient removal EEI_NUTRIENTSrem ranging from 30 to 464 kWh/kg _NUTRIENTSrem in anaerobic SBBR and from 330 to 1380 kWh/kg _NUTRIENTSrem in aerobic REBDC do not diverge from the literature data. According to the Regulation of the Polish Minister of Climate and Environment [58], the production of 1 kWh of electricity generates 721 g CO₂. This means that indirect carbon dioxide emissions during soilless tomato cultivation wastewater treatment were in the range of 21.63 to 125.744 kg/kg _NUTRIENTSrem in SBBR and in 237.9 kg to 994.98 kg/kg _NUTRIENTSrem in REBDC. In both reactors, heterotrophic denitrification takes place (due to the use of sodium acetate as an external source of organic carbon), resulting in direct emission of 3.93 g CO₂ per gN removed. Because the COD of the treated wastewater was much lower (260 COD mg O₂/L, after adding sodium acetate) compared to municipal wastewater, direct emissions resulting from heterotrophic denitrification in BER would be lower than in other biological treatment technologies.

The results indicate how qualitative characteristics of wastewater, biomass type, reactor type, installation scale, the characteristic of media applied in the reactor, the ratio of the media area to the reactor volume, current densities, current intensities, materials the electrodes are made of, type and load of external carbon source, and many other factors can affect the values of electric power consumption per unit of pollutants removed in electrochemical and bioelectrochemical processes.

5. Conclusions

This study aimed to determine the electric power consumption per unit of nutrients removed from wastewater generated during soilless tomato cultivation, characterized by a high concentration of nitrates and phosphorus, in two types of reactors—an aerobic rotating electrobiological disc contactor and an anaerobic sequencing batch biofilm reactor. Both bioelectrochemical reactors operated on the basis of electric current flow and sodium acetate fed as an external carbon source. Because the flow of electric current through both reactors induced the simultaneous removal of nitrogen and phosphorus and because of specific wastewater composition, the electric energy consumption in the treatment process was established on the basis of a cumulative indicator EEI_NUTRIENTSrem providing information about energy consumption per the sum of removed loads of total nitrogen and total phosphorus. This indicator is more reliable in determining energy consumption than the indicators computed per load of one nutrient type, i.e., nitrogen or phosphorus.

The study demonstrated that the values of electric power consumption per unit of removed nutrient load determined in the reactor with anaerobic biofilm biomass ranged from 30 to 464 kWh/kg _NUTRIENTSrem and were lower than the values obtained in the reactor with aerobic biofilm biomass, i.e., 330–1380 kWh/kg _NUTRIENTSrem. This difference increased negligibly with electric current density increase. The values of energy consumption indicators determined in both reactors were comparable to those reported in the literature. However, it needs to be emphasized that these differences could be due to various factors, including qualitative characteristics of wastewater, biomass type, reactor type, installation scale, current densities, current intensities, materials the electrodes are made of, and the type and load of external carbon source.

Author Contributions

Conceptualization, W.J.; methodology, J.R. and A.M.; validation, K.B. and K.K.; formal analysis, W.J., J.R. and A.M.; investigation, A.M., K.K. and K.B.; resources, W.J. and K.B.; data curation, J.R. and K.B.; writing—original draft preparation, J.R., W.J., A.M. and K.B.; writing—review and editing, J.R. and W.J.; visualization, J.R. and K.B.; supervision, A.M.; project administration, J.R. and A.M.; funding acquisition, J.R., A.M. and K.B. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures and Tables

Figure 1. Scheme of unit experimental model: 1—laboratory DC power supply, 2—discs (cathode), 3—anode, 4—tank, and 5—electric engine.

Figure 2. Scheme of SBBR experimental model [23]: 1—reactor, 2—DC power supply, 3—outflow of treated effluent, 4—cathode (biofilm-coated stainless-steel discs), 5—aluminum anode.

Figure 3. Electric energy consumption per unit of removed TN and TP load in aerobic REBDC.

Figure 4. Electric energy consumption per unit of removed nutrients load in an aerobic REBDC.

Figure 5. Electric energy consumption per unit of removed nutrient load in an anaerobic SBBR.

Table 1

Technical parameters of reactors.

Parameter	REBDC	SBBR
Number of discs	8	5
The diameter of a single disc [cm]	22	12
Disc thickness [mm]	3.0	2.0
The distance between the discs [cm]	1.0	1.5
The total surface of the disc [m2]	0.56	0.113
Submergence [%]	40	100
The volume of flow chamber [L]	2.0	2.0
Rotational speed [rpm]	10	14

Table 2

The composition of wastewater treated in reactors.

Parameters	Value Mean	Value Minimum	ValueMaximum	Standard Deviation
COD [mg O2/L]COD [mg O2/L] *	45260	37218	57318	640
Total nitrogen [mg N/L]	486	469	504	12
Nitrate [mg N/L]	471	457	490	15
Ammonia nitrogen [mg N/L]	18	17	21	2
Nitrite [mg N/L]	0.092	0.014	0.244	0.011
Total phosphorus [mg P/L]	77.5	62	89	7.5
pH	6.19	5.85	6.40	0.11
Electrolytic conductivity [mS/cm]	6.0	5.7	6.3	0.2

* Wastewater with sodium acetate.

References

1. Foladori, P.; Vaccari, M.; Vitali, F. Energy audit in small wastewater treatment plants: Methodology, energy consumption indicators, and lessons learned. Water Sci. Technol.; 2015; 72, pp. 1007-1015. [DOI: https://dx.doi.org/10.2166/wst.2015.306]

2. Sánchez, F.M.; Filgueira, A.; Seijo, M.A.; Muñoz, M.E.; Muñoz, E. Energy audit model for a waste water treatment plant. Proceedings of the 2009 International Conference on Clean Electrical Power; Capri, Italy, 9–11 June 2009; pp. 550-554.

3. Singh, P.; Carliell-Marquet, C.; Kansal, A. Energy pattern analysis of a wastewater treatment plant. Appl. Water Sci.; 2012; 2, pp. 221-226. [DOI: https://dx.doi.org/10.1007/s13201-012-0040-7]

4. NYSERDA—New York State Energy Research and Development Authority. Municipal Wastewater Treatment Plant Energy Evaluation Summary Report; New York State Energy Research and Development Authority: Albany, NY, USA, 2006.

5. de Haas, D.; Dancey, M. Wastewater treatment energy efficiency. A review with current Australian perspectives. Water; 2015; 11, pp. 53-58.

6. Panepinto, D.; Fiore, S.; Zappone, M.; Genon, G.; Meucci, L. Evaluation of the energy efficiency of a large wastewater treatment plant in Italy. Appl. Energy; 2015; 161, pp. 404-411. [DOI: https://dx.doi.org/10.1016/j.apenergy.2015.10.027]

7. Marner, S.T.; Schröter, D.; Jardin, N. Towards energy neutrality by optimising the activated sludge process of the WWTP Bochum-Ölbachtal. Water Sci. Technol.; 2016; 73, pp. 3057-3063. [DOI: https://dx.doi.org/10.2166/wst.2016.142]

8. Haslinger, J.; Lindtner, S.; Krampe, J. Operating costs and energy demand of wastewater treatment plants in Austria: Benchmarking results of the last 10 years. Water Sci. Technol.; 2016; 74, pp. 2620-2626. [DOI: https://dx.doi.org/10.2166/wst.2016.390]

9. Capodaglio, A.G.; Olsson, G. Energy Issues in Sustainable urban wastewater management: Use, demand reduction and recovery in the urban cycle. Sustainability; 2020; 12, 266. [DOI: https://dx.doi.org/10.3390/su12010266]

10. Svardal, K.; Kroiss, H. Energy requirements for wastewater treatment. Wat. Sci. Technol.; 2011; 64, pp. 1355-1361. [DOI: https://dx.doi.org/10.2166/wst.2011.221]

11. Zylka, R.; Karolinczak, B.; Dabrowski, W. Structure and indicators of electric energy consumption in dairy wastewater treatment plant. Sci. Total Environ.; 2021; 782, 146599. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.146599]

12. Zylka, R.; Dabrowski, W.; Malinowski, P.; Karolinczak, B. Modeling of Electric Energy Consumption during Dairy Wastewater Treatment Plant Operation. Energies; 2020; 13, 3769. [DOI: https://dx.doi.org/10.3390/en13153769]

13. Dabrowski, W.; Zylka, R.; Malinowski, P. Evaluation of energy consumption during aerobic sewage sludge treatment in dairy wastewater treatment plant. Environ. Res.; 2017; 153, pp. 135-139. [DOI: https://dx.doi.org/10.1016/j.envres.2016.12.001] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27951462]

14. Svenskt Vattens—Swedish Water & Wastewater Association. Svenskt Vattens Undersökning VASS Reningsverk 2015—Nyckeltal Från Första Året. 2016; Available online: https://www.svensktvatten.se (accessed on 16 June 2022).

15. Wróblewski, J.; Heidrich, Z. Energy consumption of wastewater treatment plants. Pt. I, Analysis and assessment of literature data. Gaz Woda Tech. Sanit.; 2017; 8, pp. 325-329. (In Polish)

16. Chiavola, A.; Romano, R.; Bongirolami, S.; Giulioli, S. Optimization of energy consumption in the biological reactor of a wastewater treatment plant by means of Oxy Fuzzy and ORP control. Water Air Soil. Pollut.; 2017; 228, 277. [DOI: https://dx.doi.org/10.1007/s11270-017-3462-x]

17. Henze, M.; Harremoes, P.; La Cour Jansen, J.; Arvin, E. Wastewater Treatment Biological and Chemical Processes; 2nd ed. Springer: Berlin/Heidelberg, Germany, 1996.

18. Capodaglio, A.G.; Hlavinek, P.; Raboni, M. Advances in wastewater nitrogen removal by biological processes: State of the art review. Rev. Ambiente Agua; 2016; 11, pp. 250-267. [DOI: https://dx.doi.org/10.4136/ambi-agua.1772]

19. Cortez, S.; Teixeira, P.; Oliveira, R.; Mota, M. Rotating biological contactors: A review on main factors affecting performance. Rev. Environ. Sci. Biotechnol.; 2008; 7, pp. 155-172. [DOI: https://dx.doi.org/10.1007/s11157-008-9127-x]

20. Wang., Y.; Kuntke, P.; Saakes, M.; van der Weijden, R.D.; Buisman, C.J.; Lei, Y. Electrochemically mediated precipitation of phosphate minerals for phosphorus removal and recovery: Progress and perspective. Water Res.; 2022; 209, 117891. [DOI: https://dx.doi.org/10.1016/j.watres.2021.117891] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34875541]

21. Yang, C.; Liu, T.; Chen, N.; Tong, S.; Deng, Y.; Xue, L.; Hu, W.; Feng, C. Performance and mechanism of a novel woodchip embedded biofilm electrochemical reactor (WBER) for nitrate-contaminated wastewater treatment. Chemosphere; 2021; 276, 130250. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2021.130250]

22. Watanabe, T.; Motoyama, H.; Kuroda, M. Denitrification and neutralization treatment by direct feeding of an acidic wastewater containing copper ion and high-strength nitrate to a bio-electrochemical reactor process. Water Res.; 2001; 35, pp. 4102-4110. [DOI: https://dx.doi.org/10.1016/S0043-1354(01)00158-0]

23. Kłodowska, I.; Rodziewicz, J.; Janczukowicz, W.; Gotkowska-Płachta, A.; Cydzik-Kwiatkowska, A. Hydrogenotrophic denitrification process efficiency and the number of denitrifying bacteria (MPN) in the sequencing batch biofilm reactor (SBBR) with platinum and carbon anodes. J. Environ. Sci. Health Tox. Hazard. Subst. Environ. Eng.; 2016; 51, pp. 389-392. [DOI: https://dx.doi.org/10.1080/10934529.2015.1120516]

24. Kłodowska, I.; Rodziewicz, J.; Janczukowicz, W.; Cydzik-Kwiatkowska, A.; Rusanowska, P. Influence of carbon source on the efficiency of nitrogen removal and denitrifying bacteria in biofilm from bioelectrochemical SBBRs. Water; 2018; 10, 393. [DOI: https://dx.doi.org/10.3390/w10040393]

25. Krzemieniewski, M.; Rodziewicz, J. Nitrogen compounds removal in a rotating electrobiological contactor. Environ. Eng. Sci.; 2005; 22, pp. 816-822. [DOI: https://dx.doi.org/10.1089/ees.2005.22.816]

26. Rodziewicz, J.; Filipkowska, U.; Dziadkiewicz, E. Electrolytically-aided denitrification on a rotating biological contactor. Environ. Technol.; 2011; 32, pp. 93-102. [DOI: https://dx.doi.org/10.1080/09593330.2010.487642] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21473272]

27. Rodziewicz, J.; Mielcarek, A.; Janczukowicz, W.; Jóźwiak, T.; Struk–Sokołowska, J.; Bryszewski, K. The share of electrochemical reduction, hydrogenotrophic and heterotrophic denitrification in nitrogen removal in rotating electrobiological contactor (REBC) treating wastewater from soilless cultivation systems. Sci. Total Environ.; 2019; 683, pp. 21-28. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2019.05.239] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31129328]

28. Rodziewicz, J.; Mielcarek, A.; Janczukowicz, W.; Bryszewski, K. Electric power consumption and current efficiency of electrochemical and electrobiological rotating disk contactors removing nutrients from wastewater generated in soil-less plant cultivation systems. Water; 2020; 12, 213. [DOI: https://dx.doi.org/10.3390/w12010213]

29. Bryszewski, K.Ł.; Rodziewicz, J.; Mielcarek, A.; Janczukowicz, W.; Jóźwiakowski, K. Investigation on the improved electrochemical and bio-electrochemical treatment processes of soilless cultivation drainage (SCD). Sci. Total Environ.; 2021; 783, 146846. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.146846]

30. Mielcarek, A.; Rodziewicz, J.; Janczukowicz, W.; Dobrowolski, A. Analysis of wastewater generated in greenhouse soilless tomato cultivation in central Europe. Water; 2019; 11, 2538. [DOI: https://dx.doi.org/10.3390/w11122538]

31. Wang, D.; Ye, W.; Wu, G.; Li, R.; Guan, Y.; Zhang, W.; Wang, J.; Shan, Y.; Hubacek, K. Greenhouse gas emissions from municipal wastewater treatment facilities in China from 2006 to 2019. Sci. Data; 2022; 9, 317.Available online: www.nature.com/scientificdata (accessed on 2 August 2022). [DOI: https://dx.doi.org/10.1038/s41597-022-01439-7]

32. Zhou, M.; Fu, W.; Gu, H.; Lei, L. Nitrate removal from groundwater by a novel three-dimensional electrode biofilm reactor. Electrochim. Acta; 2007; 52, pp. 6052-6059. [DOI: https://dx.doi.org/10.1016/j.electacta.2007.03.064]

33. Rodziewicz, J. Removal of Nitrogen and Phosphorus Compounds from Wastewater Originating from Soil-Less Cultivation of Plants in a Rotating Electrobiological Contactor; Disserations and Monographs Series, 202 UWM Publishing House: Olsztyn, Poland, 2017; (In Polish)

34. Hooshmandfar, A.; Ayati, B.; Khodadadi Darban, A. Optimization of material and energy consumption for removal of Acid Red 14 by simultaneous electrocoagulation and electroflotation. Water Sci. Technol.; 2016; 73, pp. 192-202. [DOI: https://dx.doi.org/10.2166/wst.2015.477]

35. Longo, S.; d’Antoni, B.M.; Bongards, M.; Chaparro, A.; Cronrath, A.; Fatone, F.; Lema, J.M.; Mauricio-Iglesias, M.; Soares, A.; Hospido, A. Monitoring and diagnosis of energy consumption in wastewater treatment plants. A state of the art and proposals for improvement. Appl. Energy; 2016; 179, pp. 1251-1268. [DOI: https://dx.doi.org/10.1016/j.apenergy.2016.07.043]

36. Butler, E.; Hung, Y.-T.; Yeh, R.Y.-L.; Al Ahmad, M.S. Electrocoagulation in Wastewater Treatment. Water; 2011; 3, pp. 495-525. [DOI: https://dx.doi.org/10.3390/w3020495]

37. Gerek, E.E.; Yilmaz, S.; Koparal, S.; Gerek, O.N. Combined energy and removal efficiency of electrochemical wastewater treatment for leather industry. J. Water Process. Eng.; 2019; 30, 100382. [DOI: https://dx.doi.org/10.1016/j.jwpe.2017.03.007]

38. Fernandes, A.; Spranger, P.; Fonseca, A.D.; Pacheco, M.J.; Ciriaco, L.; Lopes, A. Effect of electrochemical treatments on the biodegradability of sanitary landfill leachates. Appl. Catal. B; 2014; 144, pp. 514-520. [DOI: https://dx.doi.org/10.1016/j.apcatb.2013.07.054]

39. Karabulut, B.Y.; Atasoy, A.D.; Can, O.T.; Yesilnacar, M.I. Electrocoagulation for nitrate removal in groundwater of intensive agricultural region: A case study of Harran plain, Turkey. Environ. Earth Sci.; 2021; 80, 190. [DOI: https://dx.doi.org/10.1007/s12665-021-09488-8]

40. Rodziewicz, J.; Janczukowicz, W.; Mielcarek, A.; Filipkowska, U.; Kłodowska, I.; Ostrowska, K.; Duchniewicz, S. Anaerobic rotating disc batch reactor (ARDBR) nutrient removal process enhanced by volatile fatty acid addition. Environ. Technol.; 2015; 36, pp. 953-958. [DOI: https://dx.doi.org/10.1080/09593330.2014.969328] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25252632]

41. Kabdaşli, I.; Arslan-Alaton, I.; Ölmez-Hanci, T.; Tünay, O. Electrocoagulation applications for industrial wastewaters: A critical review. Environ. Technol. Rev.; 2012; 1, pp. 2-45. [DOI: https://dx.doi.org/10.1080/21622515.2012.715390]

42. Lee, K.C.; Rittmann, B.E. Applying a novel autohydrogenotrophic hollow-fiber membrane biofilm reactor for denitrification of drinking water. Water Res.; 2002; 36, pp. 2040-2052. [DOI: https://dx.doi.org/10.1016/S0043-1354(01)00425-0]

43. Li, M.; Feng, C.P.; Zhang, Z.N.; Lei, X.H.; Chen, R.Z.; Yang, Y.N.; Sugiura, N. Simultaneous reduction of nitrate and oxidation by-products using electrochemical method. J. Hazard. Mater.; 2009; 171, pp. 724-730. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2009.06.066] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19608341]

44. Yang, C.; Feng, C.; Chen, N.; Deng, Y.; Hu, W.; Xue, L. Development of a novel palm fiber biofilm electrode reactor (PBER) for nitrate-contaminated wastewater treatment: Performance and mechanism. Environ. Sci. Water Res. Technol.; 2020; 6, pp. 839-850. [DOI: https://dx.doi.org/10.1039/C9EW00919A]

45. Pous, N.; Koch, C.; Vila-Rovira, A.; Balaguer, M.D.; Colprim, J.; Muhlenberg, J.; Muller, S.; Harnisch, F.; Piug, S. Monitoring and engineering reactor microbiomes of denitrifying bioelectrochemical systems. RSC Adv.; 2015; 5, pp. 68326-68333. [DOI: https://dx.doi.org/10.1039/C5RA12113B]

46. Sakakibara, Y.; Nakayama, T. A novel multi-electrode system for electrolytic and biological water treatments: Electric charge transfer and application to electric charge transfer and application to denitrification. Water Res.; 2001; 35, pp. 768-778. [DOI: https://dx.doi.org/10.1016/S0043-1354(00)00327-4]

47. Tong, S.; Zhang, B.; Feng, C.; Zhao, Y.; Chen, N.; Hao, C.; Pu, J.; Zhao, L. Characteristics of heterotrophic/biofilm-electrode autotrophic denitrification for nitrate removal from groundwater. Bioresour. Technol.; 2013; 148, pp. 121-127. [DOI: https://dx.doi.org/10.1016/j.biortech.2013.08.146]

48. Prosnansky, M.; Sakakibara, Y.; Kuroda, M. High-rate denitrification and SS rejection by biofilm electrode reactor (BER) combined with microfiltration. Water Res.; 2002; 36, pp. 4801-4810. [DOI: https://dx.doi.org/10.1016/S0043-1354(02)00206-3]

49. Cid, C.M.A.; Jasper, J.T.; Hoffmann, M.R. Phosphate recovery from human waste via the formation of hydroxyapatite during electrochemical wastewater treatment. ACS Sustain. Chem. Eng.; 2018; 6, pp. 3135-3142. [DOI: https://dx.doi.org/10.1021/acssuschemeng.7b03155]

50. Sabia, G.; Petta, L.; Avolio, F.; Caporossi, E. Energy saving in wastewater treatment plants: A methodology based on common key performance indicators for the evaluation of plant energy performance, classification and benchmarking. Energy Convers. Manag.; 2020; 220, 113067. [DOI: https://dx.doi.org/10.1016/j.enconman.2020.113067]

51. Waqas, S.; Bila, M.R. A Review on Rotating Biological Contactors. Indones. J. Sci.; 2019; 4, pp. 241-256. [DOI: https://dx.doi.org/10.17509/ijost.v4i2.18181]

52. Hassard, F.; Biddle, J.; Cartmell, E.; Jefferson, B.; Tyrrel, S.; Stephenson, T. Rotating biological contactors for wastewater treatment—A review. Process Saf. Environ.; 2015; 94, pp. 285-306. [DOI: https://dx.doi.org/10.1016/j.psep.2014.07.003]

53. Liu, H.; Tong, S.; Chen, N.; Liu, Y.; Feng, C.; Hu, Q. Effect of electro-stimulation on activity of heterotrophic denitrifying bacteria and denitrification performance. Bioresour. Technol.; 2015; 196, pp. 123-125. [DOI: https://dx.doi.org/10.1016/j.biortech.2015.07.076]

54. Jóźwiak, T.; Mielcarek, A.; Janczukowicz, W.; Rodziewicz, J.; Majkowska-Gadomska, J. Chojnowska, M. Hydrogel chitosan sorbent application for nutrient removal from soilless plant cultivation wastewater. Environ. Sci. Pollut. Res.; 2018; 25, pp. 18484-18497. [DOI: https://dx.doi.org/10.1007/s11356-018-2078-z]

55. Saxena, P.; Bassi, A. Removal of nutrients from hydroponic greenhouse effluent by alkali precipitation and algae cultivation method. J. Chem. Technol. Biotechnol.; 2013; 88, pp. 858-863. [DOI: https://dx.doi.org/10.1002/jctb.3912]

56. Guerrini, A.; Romano, G.; Indipendenza, A. Energy efficiency drivers in wastewater treatment plants: A double bootstrap DEA analysis. Sustainability; 2017; 9, 1126. [DOI: https://dx.doi.org/10.3390/su9071126]

57. Lu, Q.; Zhou, W.; Min, M.; Ma, X.; Ma, Y.; Chen, P.; Zheng, H.; Doan, Y.T.T.; Liu, H.; Chen, C. et al. Mitigating ammonia nitrogen deficiency in dairy wastewaters for algae cultivation. Bioresour. Technol.; 2016; 201, pp. 33-40. [DOI: https://dx.doi.org/10.1016/j.biortech.2015.11.029] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26623940]

58. Regulation of the Minister of Climate and Environment of 15 December 2021 on the Greenhouse Gas Emission Index for Electricity in 2022. Dz. U. (J. Laws). 2021; 2348.Available online: https://isap.sejm.gov.pl/isap.nsf/DocDetails.xsp?id=WDU20210002348 (accessed on 2 August 2022). (In Polish)